diff --git "a/sanskrit book json data.json" "b/sanskrit book json data.json"
new file mode 100644--- /dev/null
+++ "b/sanskrit book json data.json"
@@ -0,0 +1,2026 @@
+[
+    {
+        "page": 1,
+        "content": "BIBLIOTHEK\nINPOGERMAMSCHER GRAMMATIKEN\nBEARBEITET VON\nF.BUCHELER, H,HUBSOHMANN,A,LESKIEN, G,MEYER,\nE.SIEVERS,H,WEBER, W,D,WHITNEY,E,WINDISOH,\nBAND II.\nASANSKRIT GRAMMAR, INCLUDING BOTH THE CLASSICAL LANGUAGE, AND\nTHEOLDER DIALECTS, OFVEDA ANDBRAHMANA\nBYWILLIAM DWIGHT WHITNEY.\nLEIPZIG,\nDRUCK UNDVERLAG VONBREITKOPF UNDHARTEL.\n1879.\n"
+    },
+    {
+        "page": 2,
+        "content": "PREFACE.\nItwas inJune, 1875. asIchanced tobeforaday or\ntwo inLeipzig,that Iwas unexpectedlyinvited toprepare\ntheSanskrit grammarfortheIndo-Europeanseriesprojected\nbyMessrs. Breitkopf andHartel. After some consideration,\nandconsultation with friends,Iacceptedthetask, andhave\nsince devoted toitwhat time could bespared fromregular\nduties,after thesatisfaction ofengagementsearlier formed.\nIfthedelay seems along one,itwasnevertheless unavoid-\nable;and Iwouldgladly,intheinterest ofthework itself\nhavemade itstilllonger.Inevery suchcase,itisnecess-\narytomake acompromise between measurably satisfyinga\npresent pressing need, anddoing thesubjectfullerjustice\natthecostofmore time; and itseemed asifthecall for\naSanskrit grammar onasomewhat differentplanfrom those\nalreadyinuse excellent assome ofthese inmany respects\nare- -was urgent enoughtorecommend aspeedy com-\npletionofthework begun.\nTheobjects hadespeciallyinview inthepreparation\nofthisgrammar have been thefollowing:\nTomake apresentationofthe facts ofthelanguage\nprimarilyastheyshow themselves inuseintheliterature,\nandonly secondarilyasthey arelaiddown bythenative\ngrammarians. The earliest European grammars werebythe\nnecessityofthecasechiefly founded ontheir nativeprede-\ncessors;andatraditional method wasthus established which\nhasbeen perhaps somewhat tooclosely adheredto,atthe\nexpenseofclearness andofproportion,aswell asofscien-\ntific truth. Accordingly, myattention hasnotbeen directed\ntoward aprofounder studyofthegrammaticalscience ofthe\nHindu schools :their teachingsIhave been contented totake\n"
+    },
+    {
+        "page": 3,
+        "content": "viPREFACE.\nasalready reportedtoWestern learners intheexisting\nWestern grammars.\nToinclude also inthepresentationtheforms andcon-\nstructions oftheolder language,asexhibited intheVeda\nandtheBrahmana. Grassmann's excellent Index-Vocabulary\ntotheRig-Veda, andmyownmanuscript onetotheAtharva-\nVeda (whichIhopesoon tobeable tomakepublic), gave\nmeinfull detail thegreatmass ofVedic material;andthis,\nwithsome assistance frompupilsandfriends,Ihavesong]it\ntocomplete,asfarasthecircumstancespermitted, from the\nother Vedic textsandfrom thevarious works oftheBrah-\nmanaperiod,bothprintedandmanuscript.\nTotreat thelanguage throughout asanaccentedone,\nomitting nothingofwhat isknownrespecting thenature of\ntheSanskrit accent,itschangesincombination andinflection,\nandthetone ofindividual words- -being, inallthis, ne-\ncessarily dependent especially upon thematerialpresented\nbytheolder accentuated texts.\nTocast allstatements; classifications, andsoon,intoa\nform consistent with theteachiogs oflinguistic science. In\ndoing this,ithasbeennecessarytodiscard afew ofthe\nlong-used andfamiliar divisions andterms ofSanskrit gram-\nmar--forexample, the classification andnomenclature of\n\"specialtenses\" and\"general tenses\" (whichissoindefen-\nsible thatonecanonlywonder atitshaving maintained itself\nsolong),theorderandterminologyoftheconjugation-classes,\ntheseparation intreatment ofthefacts ofinternal andex-\nternal euphonic combination, andthelike. Butcarehasbeen\ntaken tofacilitate the transition from theoldtothenew;\nandthechanges,itisbelieved, willcommend themselves\ntounqualified acceptance. Ithasbeensought also tohelp\nanappreciation ofthecharacter ofthelanguage byputting\nitsfacts asfaraspossible intoastatistical form. Inthis\nrespect thenative grammar isespecially deficient andmis-\nleading.\nRegard hasbeenconstantly had tothepractical needs\nofthelearner ofthelanguage, and ithasbeenattempted,\nbyduearrangement andbytheuseofdifferent sizes of\n"
+    },
+    {
+        "page": 4,
+        "content": "PREFACE. vii\ntype,tomake thework asusable byonewhoseobject\nitistoacquireaknowledgeoftheclassical Sanskrit alone\nasthose are inwhich the earlier forms arenotincluded.\nThecustom oftransliteratingallSanskrit words intoEuro-\npean characters, which hasbecome usual inEuropean Sans-\nkritgrammars, is,asamatter ofcourse, retained through-\nout; and. because ofthedifficultyofsetting even asmall\nSanskrit typewithanythingbutalarge European,itis\npracticedalone inthesmaller sizes.\nWhile thetreatment ofthe facts ofthelanguagehas\nthusbeenmade ahistorical one, within the limits ofthe\nlanguage itself,Ihave notventured tomake itcomparative,\nbybringingintheanalogous forms andprocessesofother\nrelated languages. Todothis,inaddition toallthatwas\nattempted beside, would have extended thework, both in\ncontent and intime ofpreparation,farbeyondthelimits\nassignedtoit.And, having decided toleave outthis ele-\nment, Ihave done soconsistently throughout. Explanations\noftheoriginofforms have alsobeen avoided, forthesame\nreason and forothers, which hardlycall forstatement.\nAgrammarisnecessarilyingreat partfounded onits\npredecessors, and itwould beinvain toattempt anacknowl-\nedgmentindetail ofalltheaidreceived from other schol-\nars. Ihavehad athand always especiallytheveryschol-\narlyand reliable briefsummaryofKielhorn, the fulland\nexcellent work ofMonier Williams, thesmaller grammarof\nBopp (awonder oflearning andmethod forthetimewhen\nitAvasprepared ,andthevolumes ofBenfey and Mtiller.\nAsregards thematerial ofthelanguage, nootheraid,of\ncourse, hasbeen atallcomparable with thegreatPeters-\nburg lexicon ofBohtlingk andRoth, theexistence ofwhich\ngives byitself anew character toallinvestigationsofthe\nSanskritlanguage. What Ihave notfound there orinthe\nspecialcollections made bymyselforbyothers forme,I\nhave called below \"notquotable*'-aprovisional designa-\ntion^ necessarilyliable tocorrection indetail bytheresults\noffurther researches. Forwhat concerns theverb,itsforms\nand their classification anduses,Ihave had, asevery one\n"
+    },
+    {
+        "page": 5,
+        "content": "viiiPREFACE .\nmust have, byfarthemost aidfromDelbruck. inhisAlt-\nindisches Verb urnand hisvarious syntacticalcontribu-\ntions. Former pupilsofmyown. Prof. Avery and Dr.\nEdgren. have alsohelpedme. inconnection with thissub-\njectandwith others, inawayandmeasure that calls for\npublic acknowledgment.Inrespecttotheimportant matter\nofthedeclension intheearliest language.Ihavemade great\nuseoftheelaborate paperintheJourn. Am. Or.Soc.(print-\ningcontemporaneouslywith thiswork, andused byme\nalmost, butnotquite,totheend ofthesubject) bymy\nformerpupilProf.Lanman; mytreatment ofitisfounded\nonhis.Mymanifold obligationstomyown teacher. Prof.\nWeber ofBerlin, also requiretobementioned :among other\nthings,Iowe tohim theuseofhiscopiesofcertain un-\npublishedtexts oftheBrahmanaperiod,nototherwise access-\nible tome;andhewaskindenoughtolookthrough with\nmemywork initsinchoate condition, favoring mewith\nvaluablesuggestions. For this lastfavor Ihave likewise to\nthank Prof.Delbruck who, moreover, hastaken thetrouble\ntoglance over foralikepurposethegreater part ofthe\nproof-sheets ofthegrammar, astheycame from thepress.\nToDr.L.Schroder isduewhatever useIhave been able\ntomake(unfortunately avery imperfect one)oftheimport-\nantMatriayani-Sanhita.\nOfthedeficiencies ofmywork Iam. Ithink, notless\nfullyaware thananycritic ofit.even theseverest,islikely\ntobe.Should itbefound toanswer itsintended purpose\nwellenoughtocome toanother edition, myendeavor will\nbetoimprove andcomplete it;and Ishall begratefulfor\nany corrections orsuggestions which mayaidmeinmak-\ningitamore efficienthelptothestudyoftheSanskrit\nlanguage and literature.\nGOTH A,July1879.\nW.D.W.\n"
+    },
+    {
+        "page": 6,
+        "content": "INTRODUCTION.\nBRIEF ACCOUNT OFTHEINDIAN LITERATURE.\nItseems desirable togivehere such asketch ofthe\nhistoryofIndian literature asshall show the relation to\noneanother ofthe differentperiods andforms ofthelan-\nguagetreated inthefollowing grammar,and theposition\noftheworks therequoted.\nThename \"Sanskrit\" (samskrta, 1087 d,'adorned, elab-\norated,perfected'), which ispopularly appliedtothewhole\nancient andsacredlanguageofIndia, belongs moreproperly\nonlytothat dialect which, regulatedandestablished bythe\nlabors ofthenativegrammarians,hasledforthe lasttwo\nthousandyearsormore anartificiallife,likethat ofthe\nLatinduringmost ofthesameperiodinEurope,asthe\nwritten andspoken means ofcommunication ofthelearned\nandpriestlycaste\n;andwhich even atthepresent dayfills\nthat office. Itisthusdistinguished,ontheonehand, from\nthe later and derived dialects asthePrakrit, forms of\nlanguage which have datable monuments from asearlyas\nthethirdcenturybefore Christ, andwhich arerepresented\nbyinscriptions and coins, bythespeechoftheuneducated\ncharacters intheSanskrit dramas(see below), andbya\nlimited literature;thePali, aPrakritic dialect which became\nthesacredlanguageofBuddhism inFarther India, and is\n"
+    },
+    {
+        "page": 7,
+        "content": "x INTRODUCTION.\nstill inservice there assuch;andyetlaterandmore altered\ntongues formingthetransition tothelanguagesofModern\nIndia. And, ontheother hand, itisdistinguished,but\nverymuch lesssharplyandwidely,from theolder dialects\norforms ofspeech presentedinthecanonical literature,\ntheVeda andBrahmana.\nThisfact,ofthefixation bylearned treatment ofan\nauthorized mode ofexpression,which should thenceforth be\nusedaccordingtorule intheintercourse oftheeducated,\nisthecardinal oneinIndianlinguistic history; and asthe\nnativegrammaticalliterature hasdetermined theform of\nthelanguage,soithas also toalargeextent determined\nthegrammaticaltreatment ofthelanguage byEuropean\nscholars.\nMuch inthehistoryofthelearned movement isstill\nobscure, andopinionsareatvariance even astopointsof\nprime consequence. Onlytheconcluding works inthedevel-\nopmentofthegrammatical science havebeenpreservedto\nus;andthough theyareevidentlytheperfectedfruits ofa\nlongseries oflearnedlabors, therecords ofthelatter are\nlostbeyond recovery. The timeandtheplaceofthecre-\nation ofSanskrit areunknown;and astoitsoccasion, we\nhaveonlyourinferences andconjecturestorelyupon.It\nseems, however, altogether likelythatthegrammaticalsense\noftheancient Hindus wasawakened ingreat measure by\ntheirstudyofthetraditional sacred texts, andbytheircom-\nparison ofitsdifferentlanguagewith that ofcontemporary\nuse. Itiscertain that thegrammatical studyofthose texts\n(gakhas, lit'ly 'branches'), phonetic and other, waszealously\nandeffectively followed intheBrahmanic schools;this is\nattestedbyourpossession ofanumber ofphonetico-gram-\nmaticaltreatises, firatigaJchyas (prati $ahham, 'belongingto\neach several text1\n),onehavingforsubject eachprincipal\nVedictext,andnotingallitspeculiarities ofform; these,\nbothbythedepth and exactness oftheirown researches\nandbythenumber ofauthorities whichthey quote, speak\nplainlyofalivelyscientificactivity continuedduringalong\ntime. Whatpart, ontheother hand, thenotice ofdiffer-\n"
+    },
+    {
+        "page": 8,
+        "content": "xii'INTRODUCTION.\nlimits oftheartificialityofthisprocessisnotyetknown.\nThe attention ofspecialstudents oftheHindu grammar\nandthesubjectissointricate and difficult that thenumber\nisexceedinglysmall ofthosewhohave mastered itsuffi-\ncientlytohave acompetent opinion onsuchgeneral matters)\nhasbeen hitherto mainlydirected toward determining what\ntheSanskrit accordingtoPaninireallyis,towardexplaining\nthelanguagefrom thegrammar. Arid, naturally enough,\ninIndia,orwherever elsetheleading objectistolearn to\nspeakandwrite thelanguage correctlythatis,asauthor-\nizedbythegrammariansthat isthepropercourse to\npursue. This, however,isnottheway reallytounderstand\nthelanguage. The timemust soon come, orithascome\nalready, when theendeavor shall beinstead toexplainthe\ngrammarfrom thelanguage;totest inalldetails, sofar\nasshall befoundpossible,thereason ofPanini' srules\n(whichcontain notalittle thatseemsproblematical,oreven\nsometimesperverse);todetermine what andhowmuch\ngenuine usage hehadeverywhereasfoundation, andwhat\ntraces maybeleftintheliterature ofusages possessingan\ninherentlyauthorized character, thoughunratified byhim.\nBytheterm \"classical'' or\"later\"language, then, as\nconstantlyused below inthegrammar,ismeant thelan-\nguageofthoseliterary monuments which arewritten incon-\nformitywith therules ofthenative grammar:virtually,the\nwholeproperSanskrit literature. Foralthough partsofthis\naredoubtless earlier than Panini,itisimpossibletotell\njustwhatparts,orhow fartheyhaveescapedintheirstyle\nthelevelling influence ofthegrammar. The whole, too,\nmaybecalled sofaranartificial literature asitiswritten\ninaphonetic form(seegrammar, 103)which never can\nhavebeen atrulyvernacular andlivingone.Nearlyallof\nitismetrical :notpoetic worksonly,butnarratives, histories\n(sofarasanything deservingthatname canbesaid toexist),\nand scientific treatises ofevery variety,aredone intoverse;\naprose andaprose literature(exceptinthecommentaries)\nhardlyhasanexistence. Oflinguistic historythere isnext\ntonothinginitall;butonlyahistoryofstyle,and this\n"
+    },
+    {
+        "page": 9,
+        "content": "INTRODUCTION. xiii\nforthemostpartshowingagradual depravation, anincrease\nofartificiality andanintensification ofcertain more unde-\nsirable features ofthelanguage such astheuseofpas-\nsive constructions andofparticiplesinstead ofverbs, and\nthesubstitution ofcompoundsforsentences.\nThisbeingthecondition ofthelaterliterature,itisof\nsomuch thehigher consequencethat there isanearlier\nliterature,towhich thesuspicionofartificialitydoes not\nattach, orattaches atleastonlyinaminimaldegree, which\nhasatrulyvernacular character, andabounds inproseas\nwell asverse.\nThe results oftheveryearliestliterary productiveness\noftheIndianpeoplearethehymnswith which, whenthey\nhadonlycrossed thethreshold ofthecountry, andwhen\ntheirgeographicalhorizon was still limited tothe river-\nbasin oftheIndus with itstributaries, they praisedtheir\ngods,the deified powersofnature, andaccompanied the\nrites oftheircomparatively simple worship. Atwhatperiod\nthese weremade andsungcannot bedetermined withany\napproachtoaccuracy:itmayhavebeen asearlyas2000\nB.C.They were longhanded down byoraltradition, pre-\nserved bythe care, andincreased bytheadditions and\nimitations, ofsucceeding generations;themass wasever\ngrowing, and, with thechangeofhabits and beliefs and\nreligious practices, wasbecoming variously applied sung\ninchosen extracts, mixed with other material intoliturgies,\nadapted withmore orlessofdistortion tohelptheneeds\nofaceremonial which wascomingtobeofimmense elab-\noration andintricacy. And, atsome time inthecourse\nofthishistory,there wasmade forpreservationagreatcol-\nlection ofthehymn-material, mainlyitsoldest andmost\ngenuine part,totheextent ofover athousand hymns and\ntenthousandverses, arranged accordingtotraditional author-\nshipand tosubject andlengthofhymn:this collection is\ntheRig-Veda, -Veda ofverses(re)orhymns'. Other col-\nlections weremade also outofthesamegeneralmass of\ntraditional material :doubtlesslater, althoughthe inter-\nrelations ofthisperiodareasyettoounclear toallow of\n"
+    },
+    {
+        "page": 10,
+        "content": "xiv INTRODUCTION.\nourspeakingwith entire confidence astoanythingconcern-\ningthem. Thus, theSama- Veda. 'Veda ofchants(saman}\\\ncontaining onlyabout asixth asmuch, itsversesnearlyall\nfound intheRig-Vedaalso, butappearingherewithnume-\nrous differences ofreading;these werepassages puttogether\nforchantingatthesoma-sacrifices. Again,collections called\nbythecomprehensivename ofYajur-Veda, 'Veda ofsac-\nrificial formulas (yajusV:these contained notverses alone,\nbutalsonumerous prose utterances, mingledwith theformer,\nintheorder inwhichtheywerepractically employedin\ntheceremonies; theywerestrictly liturgicalcollections. Of\nthese, there areinexistence several texts, which have their\nmutual differences: theVajasaneyi-Samhita (intwoslightly\ndiscordant versions, Madhyandina andKanvd,sometimes\nalso called theWhiteYajur-Veda;and thevarious and\nconsiderably differingtexts oftheBlack Yajur-Veda. namely\ntheTaittirlya-Samhita,theMaitrayam-Samhita,and the\nKathaka (thetwo lastnotyetpublished). Finally,another\nhistorical collection, liketheRig-Veda,butmade upmainly\noflaterand lessacceptedmaterial, andcalled (amongother\nless current names) theAtharva-Veda, 'Veda oftheAth-\narvans(alegendary priestly family)';itissomewhat more\nthan half asbulkyastheRig-Veda, andcontains acertain\namount ofmaterialcorrespondingtothat ofthe latter, and\nalso anumber ofbriefprose passages. Tothis last col-\nlection isvery generallyrefused intheorthodox literature\nthename ofVeda; butforusitisthemostinterestingof\nall. after theRig-Veda, because itcontains thelargest\namount ofhymn-material (ormantra, asitiscalled, in\ndistinction from theprose brahmana], and inalanguage\nwhich, though distinctlylessantique than that oftheother,\nisneverthelesstrulyVedic. Two versions ofitareextant,\noneofthemonlyinasingle knownmanuscript.Anotinsignificant bodyoflikematerial, andofvarious\nperiod (although doubtless inthemainbelongingtothe\nlatest time ofVedicproductiveness, and inpart perhaps\ntheimitative work ofayetmoremoderntime),isscattered\nthrough thetexts tobelater described, theBrahmanas and\n"
+    },
+    {
+        "page": 11,
+        "content": "INTRODUCTION. xv\ntheSutras. Toassemble and siftandcompareitisnow\noneofthepressingneeds ofVedicstudy.\nThefundamental divisions oftheVedic literature here\nmentioned allhave had their various schools ofsectaries,\neach ofthese with atext ofitsown. showing some differ-\nences from those oftheother schools :butthose mentioned\nabove are allthat arenowknown tobeinexistence; and\nthechance ofthediscoveryofothers grows every year\nsmaller.\nThe labor oftheschools intheconservation oftheir\nsacred textswasextraordinary, andhasbeen crowned with\nsuch success that thetext ofeach school, whatever may\nbe itsdifferences from those ofother schools, isvirtually\nwithout variousreadings, preservedwith allitspeculiarities\nofdialect, and itssmallest andmostexceptionaltraits of\nphonetic form, pure andunobscured. Itisnottheplace\nhere todescribe themeans bywhich, inaddition tothe\nreligiouscare ofthe sectaries, thisaccuracy wassecured:\nforms oftext, lists ofpeculiarities and treatises upon them,\nand soon.When thiskind ofcarebeganinthecase of\neachtext, andwhat oforiginalcharacter mayhave been\neffaced beforeit,orlostinspiteofit,cannot betold. But\nitiscertain that theVedic recordsfurnish, onthewhole,\nawonderfullyaccurate andtrustworthy pictureofaform of\nancient Indianlanguage (aswell asancient Indian beliefs\nandinstitutions) which was anatural andundistorted one,\nandwhichgoesback agoodwaybehind theclassical San-\nskrit. Itsdifferences from thelatter thefollowingtreatise\nendeavors toshow indetail.\nAlong with theverses and sacrificial formulas and\nphrasesinthetexts oftheBlackYajur-Vedaaregiven\nlong prose sections,inwhich theceremonies aredescribed,\ntheirmeaning andthereason ofthedetails andtheaccom-\npanying utterances arediscussed andexplained,illustrative\nlegendsarereportedorfabricated, andvariousspeculations,\netymological andother, areindulgedin.Such matter comes\ntobecalled brahmana(apparently 'relatingtothebrahman\norworship';.IntheWhite Yajur-Veda.itisseparated into\n"
+    },
+    {
+        "page": 12,
+        "content": "xvi' INTRODUCTION.\nawork by itself, beside thesamhitci ortext ofverses and\nformulas, and iscalled theCatapatha-Brahmana, 'Brahmana\nofahundredways'.Other similar collections arefound, be-\nlongingtovarious other schools ofVedicstudy, andthey\nbear thecommon name ofBrahmana, with thename ofthe\nschool, orsome other distinctivetitle, prefixed. Thus, the\nAitareyaandKamhitaki- Brahmanas, belongingtothe\nschools oftheRig-Veda,thePancavinqa andShadvin$a-\nBrahmanas andother minor works,totheSama-Veda; the\nGopatha-Brahmana,totheAtharva-Veda;andaJaimini-\nBrahmana, totheSama-Veda, hasjust (Burnell) been dis-\ncovered inIndia; theTaittirlya-Brahmanaisacollection\nofmingledmantra andbrahmana, like thesamhita ofthe\nsame name, butsupplementaryand later. These works are\nlikewise regardedascanonical bytheschools, andarelearn-\nedbytheir sectaries with thesame extreme carewhich is\ndevoted tothesamhitas, and their condition oftextual\npreservationisofakindred excellence. Toacertain\nextent, there isamong them thepossessionofcommon\nmaterial: afact thebearingsofwhich arenotyetfully\nunderstood.\nNotwithstandingtheinanityofnosmallpartoftheir\ncontents, theBrahmanas areofahighorder ofinterest in\ntheirbearings onthehistoryofIndian institutions; and\nphilologically theyarenot lessimportant,sincetheyre-\npresentaform oflanguageinmostrespects intermediate\nbetween theclassical andthat oftheVedas, and offerspe-\ncimens onalargescale ofaprose style,andofonewhich\nisinthemain anatural andfreely developed one the\noldest andmostprimitive Indo-European prose.\nBeside theBrahmanas aresometimes found laterap-\npendices,ofasimilarcharacter, called Aranyakas ('forest-\nsections'): astheAitareya-Aranyaka, Taittirtya-Aranyaka,\nBrhad-Aranyaka, and soon.Andfromsome ofthese, or\neven from theBrahmanas, areextracted theearliest Upa-\nnishads('sittings,lectures onsacredsubjects') which,\nhowr\never, arecontinued andadded todown toacompara-\ntively modern time. TheUpanishadsareoneofthe lines\n"
+    },
+    {
+        "page": 13,
+        "content": "INTRODUCTION. xvii\nbywhich theBrahmana literaturepassesover into thelater\ntheologicalliterature.\nAnother line oftransition isshown intheSutras\n(-lines,\nrules'). Theworks thusnamed areanalogous with the\nBrahmanas inthatthey belongtotheschools ofVedic\nstudyand arenamed from them, andthattheydealwith\nthereligiousceremonies:treating them, however, inthe\nwayofprescription,notofdogmatic explanation. They,\ntoo, contain some mantra orhymn-material, notfound to\noccur elsewhere. Inpart ($rautaorkalpa-sutras\\ theytake\nupthegreatsacrificial ceremonies, with which theBrah-\nmanas have todo;inpart (grhya-sutras), theyteach the\nminor duties ofapious householder; insome cases(sa-\nmayacarika-sutras) they laydown thegeneral obligationsof\nonewhose life isinaccordance withprescribed duty. And\noutofthelasttwo, orespeciallythe last,come bynatural\ndevelopmentthelaw-books(dharma-$astras), which make\naconspicuous figureinthelater literature :theoldest and\nmost noted ofthem beingthat called bythename of\nManu(anoutgrowth,itisbelieved, oftheManava Vedic\nschool);towhich areadded thatofYajnavalkya. andmany\nothers.\nRespectingthechronologyofthisdevelopment,orthe\ndate ofanyclass ofwritings,stillmore ofanyindividual\nwork, the less that issaidthebetter. Alldatesgivenin\nIndianliterary historyarepinssetuptobebowled down\nagain. Every important work hasundergonesomany more\norlesstransforming changesbeforereachingtheform in\nwhich itcomes tous,that thequestionoforiginal con-\nstruction iscomplicatedwith that offinal redaction. Itis\nsowith thelaw-book ofManu, justmentioned, which has\nwell-founded claims tobeing regardedasoneofthevery\noldest works oftheproperSanskrit literature,ifnotthe\noldest(itisvariously assigned,toperiodsfrom sixcenturies\nbefore Christ tosoon afterChrist).Itisso,again,ina\nstillmorestriking degree,with thegreat legendary epicof\ntheMahdbharata. Theground-workofthis isdoubtless of\nvery early date; but ithasserved asatext intowhich\nb\n"
+    },
+    {
+        "page": 14,
+        "content": "xviiiINTRODUCTION.\nmaterials ofvarious character andperiod havebeeninwoven,\nuntil ithasbecome aheterogeneous mass, akind ofcyclo-\npediaforthewarrior-caste, hard toseparateinto itscon-\nstituent parts.ThestoryofNala, and thephilosophical\npoem Bhagavad-Glta,aretwo ofthemost noted ofits\nepisodes.TheRamayana,theother most famousepic,\nisawork ofanother kind: thoughalsoworked overand\nmore orless altered initstransmission toourtime,itis\ntheproduction,inthemain, ofasingleauthor(Valmiki);\nand itisgenerallybelieved tobeinpart allegorical,re-\npresentingtheintroduction ofAryanculture anddominion\nintoSouthern India. Byitssidestand anumber ofminor\nepics,ofvarious authorship andperiod,astheRaghuvah$a\n(ascribedtothedramatistKalidasa),theMaghakavya,the\nBhattikavya (the last, writtenchieflywith thegrammatical\nintent ofillustrating byuse asmanyaspossibleofthe\nnumerous formations which, through taught bythegram-\nmarians,findnoplaceintheliterature).\nThePurdnas. alargeclass ofworksmostlyofimmense\nextent, arebestmentioned inconnection with theepics.\nTheyarepseudo-historical andpropheticincharacter, of\nmodern date, andofverysmall value. Realhistoryfinds\nnoplaceinSanskritliterature, nor isthere anyconscious\nhistorical element inanyoftheworkscomposingit.\nLyric poetryisrepresented bymany works, some of\nwhich, astheMeghaduta andGitagovinda, areofnomean\norder ofmerit.\nThedrama isastillmorenoteworthy andimportant\nbranch. The first indications ofdramatical inclination and\ncapacity onthepartoftheHindus areseen incertain\nhymnsoftheVeda, where amythologicalorlegendary\nsituation isconceiveddramatically, and setforth inthe\nform ofadialogue well-knownexamplesarethedialogue\nofSarama andthePanis, that ofYama and hissister Yami,\nthat ofVasishtha andtherivers, that ofAgni andtheother\ngods butthere arenoextant intermediaries between these\nandthestandard drama. Thebeginningsofthelatter date\nfrom aperiod when inactual lifethehigher andeducated\n"
+    },
+    {
+        "page": 15,
+        "content": "INTRODUCTION. xix\ncharacters used Sanskrit, andthelower anduneducated used\nthepopulardialects derived fromit,thePrakrits; and their\ndialoguereflects this condition ofthings. Then, however,\nlearning (nottocall itpedantry)intervened, andstereotyped\nthenew element; aPrakrit grammar grew upbeside the\nSanskrit grammar, accordingtotherules ofwhich Prakrit\ncould hemadeindefinitely onasubstrate ofSanskrit; and\nnone oftheexisting dramas need todatefrom thetime of\nvernacular useofPrakrit, while most orallofthem are\nundoubtedly much later. Amongthedramatic authors,\nKalidasa isincomparablythe chief, and hisCakuntala as\ndistinctlyhismasterpiece.His datehasbeen amatter of\nmuchinquiry andcontroversy;itisdoubtless some cen-\nturies later than our era.Theonlyother work deserving\ntobementionedalongwith Kalidasa' sistheMrchakafl of\nCudraka, also ofquestionable period,butbelieved tobe\ntheoldest oftheextant dramas.\nApartlydramatic character belongsalso tothefable,\ninwhich animals arerepresentedasactingandspeaking.\nThemost noted works inthisdepartmentarethePanca-\ntantra, whichthroughPersian andSemitic versions hasmade\nitswayallover theworld, and contributes aconsiderable\nquotatothe fable-literature ofevery European language,\nand, partly founded onit,thecomparativelyrecent and\npopular Hitopade$a ('salutary instruction').\nTwo oftheleading departmentsofSanskrit scientific\nliterature, thelegal andthegrammatical,havebeen already\nsufficiently noticed; ofthoseremaining,themost important\nbyfar isthephilosophical. Thebeginningsofphilosophic-\nalspeculationareseenalreadyinsome ofthelaterhymns\noftheVeda, moreabundantlyintheBrahmanas andAran-\nyakas, andthenespeciallyintheUpanishads.Theevo-\nlution and historic' relation ofthesystemsofphilosophy,\nandtheageoftheir text-books, arematters onwhichmuch\nobscuritystill rests. There aresixsystemsofprimary rank,\nandreckoned asorthodox, although really standinginno\naccordance with approved religiousdoctrines. Allofthem\nseek thesame end, theemancipationofthesoulfrom the\nb*\n"
+    },
+    {
+        "page": 16,
+        "content": "xx INTRODUCTION.\nnecessityofcontiiiuingitsexistence inasuccession of\nbodies, and itsunification with theAll-soul; butthey\ndiffer inregardtothemeans bywhichtheyseek toattain\nthisend.\nTheastronomical science oftheHindus isareflection\nofthat ofGreece, and itsliterature isofrecent date; but\nasmathematicians, inarithmetic andgeometry ,theyhave\nshown more independence.Their medical science, although\nitsbeginnings goback even totheVeda, intheuseof\nmedicinalplantswithaccompanyingincantations, isoflittle\naccount, and itsproperliterature bynomeans ancient.\n"
+    },
+    {
+        "page": 17,
+        "content": "CONTENTS.\nChap.\nPREFACE\nINTRODUCTION\nI.ALPHABET\nII.SYSTEM OFSOUNDS :PRONUNCIATION .\nVowels, 8;Consonants, 11;Quantity, 26;Accent, 27.\nHI.RULES OFEUPHONIC COMBINATION . .\nIntroductory, 33;Principles, 36;Rules ofVowel Com-\nbination, 41;Permitted Finals, 46;Deaspiration, 50;\nSurd andSonant Assimilation,51\n;Combinations of\nFinal sandr,53;Conversion ofstos,57\n;Con-\nversion ofnton,60;Conversion ofDental Mutes to\nLinguals and Palatals, 62;Combinations ofFinal n,\n63;Combinations ofFinal m,65;thePalatal Mutes\nand Sibilant, andh,66;theLingual Sibilant, 71;\nExtension andAbbreviation, 72;Strengthening and\nWeakening Processes, 73;Guna and Vrddhi, 74;\nVowel-lengthening, 76;Vowel-lightening, 77\n;Nasal\nIncrement, 78; Reduplication, 79.\nIV.DECLENSION\nGender, Number, Case, 80:Uses oftheCases, 81;\nEndings ofDeclension, 92; Variation ofStem, 95;\nAccent inDeclension, 97.\nV.NOUNS ANDADJECTIVES\nClassification etc.,99;Declension I.,Stems ina,100;\nDeclensionII.,Stems in iandu,104; Declension\nIII., Stems inLong Vowels(a, I,u):A.Root-words\netc., Ill; Stems inDiphthongs, 116; B.Derivative\nStems etc.117; Declension IV., Stems inrorar,\n123; Declension V.,Stems inConsonants, 127;\nA.Root-stemsetc., 129; B.Derivative Stems inas,\nis,us,138; C.Derivative Stems inan,140; D.,\ninin,145; E.,inantorat,146; F.Perfect Par-\nticiples invans, 152; G.Comparativesinyas,155;\nComparison,156.Page.\nV\nix\n18\n832\n3379\n8098\n99159\n"
+    },
+    {
+        "page": 18,
+        "content": "xxiiCONTENTS.\nChap.\nVI.NUMERALS 160167\nCardinals, 160; Ordinals etc., 166.\nVIE.PRONOUNS 168 181\nPersonal, 168\n;Demonstrative, 171;Interrogative,\n176; Relative, 177;Emphatic, 179;Nouns used pro-\nnominally, 179; Pronominal Derivatives, Possessives\n179; Adjectives declined pronominally, 181.\nVIII. CONJUGATION 182 206\nVoice, Tense, Mode, Number, Person, 182; Verbal\nAdjectives andNouns, 185; Secondary Conjugation,\n185; Personal Endings, 186;Subjunctive Mode, 191;\nOptative, 193;Imperative, 195;Uses oftheModes,\n196;Participles, 201;Augment, 201;Reduplication,\n202; Accent oftheVerb, 203.\nIX.THEPRESENT-SYSTEM . ;.[,,,.T.~^J-rt^>207 255\nGeneral, 207; Conjugations andConjugation Classes,\n208;I.Root-class (second orad-class), 211;II.Re-\nduplicating Class (thirdorftu-class), 221;III.Nasal\nClass (seventhorrwd/i-class), 229; IV.Nuandw-Classes\n(fifthandeighth,orsuandtan-classes), 232; V.Na-\nClass (ninth orfcrz-class), 238; VI.a-Class(first or\n6ftu-classj, 241;VII.Accented d-Class(sixth ortud-\nclass), 245; VIII. Fa-Class(fourth ordiu-elass), 248;\nIX.Accented yd-Class orPassiveConjugation, 252;\nUses ofthePresent andImperfect, 254.\nX.THEPERFECT-SYSTEM 255 270\nPerfect Tense, 255; PerfectParticiple, 266; Modes\nofthePerfect, 267; Pluperfect, 269; Uses ofthe\nPerfect, 270.\nXI.THEAORIST-SYSTEMS 271 299\nClassification, 271;I.Simple Aorist: 1.Root-aorist,\n273\n4;Passive Aorist 3dsing., 277; 2.thea-Aorist,\n278; II. 3.Reduplicated Aorist, 281; III. Sibilant\nAorist, 285; 4.thes-aorist, 286; 5.the is-Aorist,\n290; 6.the>-aorist, 293;7.thesa-Aorist, 294;\nPrecative, 296; Uses oftheAorist, 298.\nXII.THEFUTURE-SYSTEMS 299 307\nI.Thes-future, 300; Modes ofthe5-future, 302;\nParticiples ofthes-future, 302; Preterit ofthe s-\nfuture:Conditional, 303; II.ThePeriphrastic Future,\n303\n;Uses oftheFutures andConditional, 305.\n"
+    },
+    {
+        "page": 19,
+        "content": "CONTENTS. xxiii\nChap. Page.\nXIII. VERBAL ADJECTIVES ANDNOUNS :PARTICI-\nPLES, INFINITIVES, GERUNDS 307 321\nPassiveParticipleintaornd,307;Past Active Par-\nticipleintavant, 310; Future PassiveParticiples:\nGerundives, 310; Infinitives, 313; Uses oftheInfini-\ntives, 315; Gerunds, 319; Adverbial Gerund inam,\n321.\nXIV. DERIVATIVE ORSECONDARY CONJUGATION .321 347\nI.Passive, 322;II.Intensive, 323; Present-System,\n325; Perfect.Aorist, Future, etc.,329; III.Desider-\native, 331; Present-System, 334; Perfect, Aorist,\nFuture, etc.,335; IV.Causative, 337; Present-System,\n339; Perfect, Aorist, Future, etc.,340; V.Denomi-\nnative, 343.\nXV.PERIPHRASTIC ANDCOMPOUND CONJUGATION 347 357\nThe Periphrastic Perfect, 347; Participial Periphras-\nticPhrases, 349;Composition withPrepositional\nPrefixes, 350; Other Verbal Compounds, 355.\nXVI. INDECLINABLES 357 370\nAdverbs, 358;Prepositions, 366;Conjunctions, 369;\nInterjections, 369.\nXVH. DERIVATION OFDECLINABLE STEMS . . .370 424\nA.Primary Derivatives, 373; B.Secondary Deriva-\ntives, 403.\nXVIII. FORMATION OFCOMPOUND STEMS.... 424 456\nClassification, 425;I.Copulative Compounds, 428;\nII.Determinative Compounds, 431;A.Dependent\nCompounds, 432; B.Descriptive Compounds, 437;\nIII.Secondary Adjective Compounds, 443; A.Pos-\nsessive Compounds, 443; B.Compounds withGoverned\nFinal Member, 452; Adjective CompoundsasNouns\nandasAdverbs, 453; Anomalous Compounds, 455;\nStem-finals altered inComposition, 455;Irregular\nConstruction withCompounds, 456.\nAPPENDIX 457 460\nA.Examples ofVarines Sanskrit Type457;B.Exam-\npleofAccentuated Text, 459.\nSANSKRIT INDEX 461 475\nGENERAL INDEX 476485\n"
+    },
+    {
+        "page": 20,
+        "content": "XXIVABBREVIATIONS.\nABBREVIATIONS.\nAB.Aitareya-Brahmana.\nAPr.Atharva-Prati^akhya.\nAV.Atharva-Veda.\nBB. Bohtlingk andRoth (Petersburg\nLexicon).\n9or ak.Qakuntala.\n$B.$atapatha-Brahmana.\nQGS. Qankhayana-Grihya-Sutra.\nGB.Gopatha-Brahmana.\nH.Hitopade$a.\nK.Kathaka.\nKB.Kaushitaki-Brahmana.\nKSS. Katha-Sarit-Sagara.\nM.Mann.\nMBh. Mahabharata.\nMegh. Meghadiita.MS. Maitrayani-Sauhita.\nPB.Pancavin^a-Brahmana.\nR.Ramayana.\nRagh. Raghuvan^a.\nRPr.Rigveda-Prati^akhya.\nRV.Rig-Veda.\nSB.Shadvin^a-Brahmaua.\nSV.Sama-Veda.\nTA. Taittiriya-Aranyaka.\nTB.TaittirTya-Brahmana.\nTPr.Taittiriya-Prati^akbya.\nTribh. Tribhashyaratna.\nTS.TaittirTya-Sanhita.\nV.Veda.\nVPr.Vajasaneyi-Prati^akhyj\nVS.Vajasaneyi-Sanhita.\n"
+    },
+    {
+        "page": 21,
+        "content": "CHAPTER I.\nALPHABET.\n1.THE natives ofIndia write their ancient andsacred\nlanguageinavarietyofalphabets generally,ineach\npartofthecountry,inthesamealphabet whichtheyuse\nfortheirown vernacular. Themode ofwriting, however,\nwhich isemployed throughouttheheart ofAryan India, or\ninHindustanproper,isalone adopted byEuropeanscholars :\nitiscalled thedevanagari.\nThisname isofdoubtful origin and value. Amorecomprehensive name\nisnagarl (perhaps,'ofthecity'); anddeva-nagarl is'nagarl ofthegods,'\nor'oftheBrahmans.'\n2.Much that relates tothehistory oftheIndianalphabetsisstill\nobscure. The earliest written monuments ofknown date inthecountry are\ntheinscriptions containing the edicts ofA^okaorPiyadasi, ofabout the\nmiddle ofthethird century B.C.They areintwo different systems of\ncharacters,ofwhich oneshows distinct signsofderivation from aSemitic\nsource, while theother isalsoprobably, though much less evidently, ofthe\nsame origin (Burnell). From thelatter, theLath, orSouthern Acoka cha-\nracter(ofGirnar), come thelater Indian alphabets, both those ofthenorthern\nAryan languages, and those ofthesouthern Dravidian languages. The\nnagari, devanagari, Bengali, Guzerati, andothers,arevarieties ofitsnorthern\nderivatives; andwith them arerelated some ofthealphabets ofpeoples\noutside ofIndia asinTibet andFarther India whohaveadopted Hindu\nculture orreligion.\nThere isreason tobelieve that writing was firstemployedinIndia for\npractical purposesforcorrespondence andbusiness and thelike and\nonlybydegrees came tobeappliedalso toliterary use. Theliterature,to\nagreat extent, andthemore fully inproportiontoitsclaimed sanctity and\nauthority, ignoresallwritten record, andassumes tobekeptinexistence by\noral tradition alone.\n3.Ofthedevanagari itself there areminor varieties, depending on\ndifferences oflocality orofperiod,asalso ofindividual hand(seeexamples\nWhitney, Grammar. 1\n"
+    },
+    {
+        "page": 22,
+        "content": "I.ALPHABET.\ninWeber's catalogueoftheBerlin Sanskrit MSS., inRajendralala Mitra's\nnotices ofMSS. inIndian libraries, inthepublished fac-similes ofin-\nscriptions, and soon);andthese areinsome measure reflected inthetype\npreparedforprinting,both inIndia and inEurope. But astudent who\nmakes himself familiar with one style ofprintedcharacters willhave little\ndifficulty with theothers, and willsoonlearn, bypractice,toreadthemanu-\nscripts. Afewspecimensoftypesother than those used inthiswork will\nbegiveninanAppendix.\nOnaccount ofthe difficulty ofcombining them with thesmaller sizes\nofourRoman and Italictype,thedevanagari characters willbeused below\nonly inconnection with the first orlargest size. And, inaccordance with\nthelaudable usageofrecent grammars, they will, wherever given, bealso\ntransliterated initalic letters; while the latter alone willbeused inthe\nother sizes.\n4.Thestudent maybeadvised totrytofamiliarize himself\nfrom the start with thedevanagarl mode ofwriting. Atthesame\ntime,itisnotnecessarythatheshould dosountil, having\nlearned theprincipal paradigms, hecomes tobegin reading and\nanalysing andparsing;andmanywill find thelatter themore\npractical, andintheendequallyormoreeffective, way.\n5.The characters ofthedevanagarl alphabet, andthe\nEuropeanletters which willbeused intransliterating them,\nareasfollows :\nshort.long.\nVowels :\nsimplepalatal\nlabial\nlingual\ndental\ndiph-(palatal\nthongsIlabial\nVisarga\nAnusvara$l\n3u\n^rd*\nFTI\n^e\n37ou\nf\n1}\nai\nau\n,norm(see 73)\nMutessurd\nguttural\npalatal\nlingual\ndental\nlabial qpsurd asp.\nis1^kh\n233TCh\n2S%th\n335Tth\nssCfiphsonant\n19*Tg\n24Sfj\n293\"d\n34$d\n39STbson. asp.\noqgh\n53Tjhnasal\n263Tn\ndh 31HIn\ndh se^n\nbh 4iITm\n"
+    },
+    {
+        "page": 23,
+        "content": "palatal\n"
+    },
+    {
+        "page": 24,
+        "content": "4I.ALPHABET.\nalphabeticscheme above areusedonlywhen thevowel\nforms asyllable byitself, orisnotcombined with apre-\ncedingconsonant: thatis,when itisinitial, orpreceded\nbyanother vowel. Incombination with aconsonant, other\nmodes ofrepresentationareused.\nB.Ifmore consonants than oneprecedethevowel,\nformingwith itasingle syllable,their characters must be\ncombined into asingle compoundcharacter.\nOrdinary Hindu usagedoes notdivide thewords ofasen-\ntence, anymore than thesyllablesofaword;afinalconsonant\niscombined intoonesyllablewith the initial vowel orconso-\nnant ofthenext following word.\n10.Under A, itistobenoticed that themodes of\nindicatingavowel combined with aprecedingconsonant\nareasfollows:\na.The short^ahasnowrittensignatall:thecon-\nsonant-signitselfimpliesafollowingTa,unless some other\nvowel-signisattached toit(orelsethevirama:11).Thus\ntheconsonant-signsasgivenabove inthealphabetic scheme\narereallythesignsofthesyllables ka,kha, etc. etc.(toha].\nb.The-long ^Taiswrittenbyaperpendicularstroke\nafter theconsonant: thus, 5fiTka, EfTdha,^Tha.\nc.Short^iandlong ^e,byasimilarstroke, which\nforshort iisplaced before theconsonant and forlongIis\nplacedafterit,and ineither case isconnected with the\nconsonant byahook above theupperline:thus, f^R ki,\n3ft ki',Prbhi,Htbhi\n;ftm', jftnl.\nThehookabove, turning tothe left ortotheright,ishistorically the\nessentialpart ofthecharacter, having beenoriginally thewhole ofit;the\nhooks were only laterprolonged, soastoreach allthewaydown beside\ntheconsonant. IntheMSS., they almost never have thehorizontal stroke\ndrawn across themabove, though this isadded inalltheprintedforms of\nthecharacters*.\n*Thus, originally dfiW,ofjit;intheMSS., jcfj, effj ;inprint,\n"
+    },
+    {
+        "page": 25,
+        "content": "12] WRITING OFVOWELS. 5\nd.Thew-sounds, short andlong,arewritten byhooks\nattached tothelower end oftheconsonant-sign: thus,Sfj\nku, 3\\ku;Idu,Idu.Onaccount ofthenecessities of\n6\\ O SX\ncombination, duandduaresomewhatdisguised: thus, If.\n^;andtheforms with^rand^Thare stillmore irre-\ngular:thus,\"^\\ru,%ru;<^Thu,f^1hu.\ne.The r-vowels, short andlong,arewritten byasub-\njoined hook, singleordouble,opening toward theright:\nthus, ^\\kr, Sfjkf ;dr,^dr.Inthe/j-sign,thehooks\nareusuallyattached tothemiddle: thus,^hr,^hr.\nAstothecombination ofrwith preceding r,seebelow, 14.\nf.The/-vowel iswritten with areduced form ofits\nfull initial character: thus,efikl:thecorresponding long\nhasnorealoccurrence(23),butwould bewritten with a\nsimilar reducedsign.\ng.Thediphthongsarewritten bystrokes, singleor\ndouble, above theupper line, combined, for Jtoand ETF\naUjwith the#-signafter theconsonant :thus. 3ftke,fi\nkai;^tko,\nInsome devanagari MSS.(asintheBengali alphabet),thesingle stroke\nabove,orone ofthedouble ones,isreplaced byasign like thea-sign\nbefore theconsonant :thus, (off fee, \\fi\\ feat, fofil feo, |cftl feaw.\n11.Aconsonant-sign, however,iscapableofbeing\nmade tosignifytheconsonant-sound alone, without an\nadded vowel, byhavingwritten beneath itastroke called\nthevirama('rest, stop'): thus,fik^<d,^h.\nSince, aswas pointed outabove, theHindus write thewords ofa\nsentence continuously,likeoneword(9,end),thevirama isingeneral called\nforonlywhen afinalconsonant occurs before apause. But itisalsooccasion-\nallyresorted tobyscribes,orinprint,inorder toavoid anawkward or\ndifficult combination ofconsonant-signs; and itisused freely inpublished\ntexts which fortheconvenience ofbeginners have their words printed sepa-\nrately.\n12.Under B,itistobenoticed that theconsonant\ncombinations areforthemostpartnot atalldifficult to\n"
+    },
+    {
+        "page": 26,
+        "content": "6I.ALPHABET. [12\nmake ortorecogniseforonewho isfamiliar with the\nsimple signs.The characteristicpartofaconsonant-sign\nthat istobeadded toanother istaken(totheexclusion\nofthehorizontal oroftheperpendicular framing-line,or\nofboth), andtheyareputtogether accordingtoconveni-\nence,either sidebyside, oroneabove theother: insome\ncombinations either arrangementisallowed. Theconsonant\nthat istobepronouncedfirst issetbefore theother inthe\none order, andabove itintheother order.\nExamplesoftheside-by-side arrangementare :ITfgga,\nSTjja, C?jpya, ZTJnma, f^Ttiha,^Tbhya, F^Iska, STUma,\nf3Ttka.\nExamplesoftheabove-and-belowarrangement are :\n^TMa, sfcca,^nja,^dda,Hpta,^tna.\n13.Insomecases, however, there ismore orless\nabbreviation ordisguiseoftheindependent form ofacon-\nsonant-signincombination.\nThus, of3\\kin\"^7kid,^\\Ida;and in^HTknaetc.;\nofrTtinfFtta;\nof^din\"%dga,^dna, etc.;\nof*Tmand ZTy,whenfollowing other consonants :\nthus,^Tkya,3Rkrna,^nma, 2Tnya,^dma, t%dya,^T\nhma, ^Thya.\"5Tchya,^dhya ;\nof5Tp,whichgenerally becomes $Twhen followed\nbyaconsonant:thus, 31pea,Wpraa, ^j$va, V3J$ya.The\nsamechangeisusualwhen avowel-signisadded below:\nthus, 5Tpw, 5Tcr.O (.\nOthercombinations, ofnotquite obviousvalue, are\nSTnna, ^T//a,^ddha,^dbha,^sta, Tgstha; andthe\ncompounds of^h:as^r7m,^A^a.\nInacase ortwo,notrace oftheconstituent letters is\nrecognisable:thus,^ksa,^jfia.\n14.Thesemivowel^r,inmaking combinations with\n"
+    },
+    {
+        "page": 27,
+        "content": "16]COMBINATIONS OFCONSONANTS. 7\nother consonants,istreated inawholly peculiar manner,\nanalogouswith that ofthevowels. Ifpronouncedbefore\nanother consonant (orconsonant-combination),itiswritten\nwith ahook above, openingtotheright (likethesubjoined\nsignofr:10e):thus. Rrka,^rsa(fP rtsna).Ifpro-\nnounced after another consonant (aloneorincombination),\nitiswritten with aslantingstroke below: thus, Elgra,\nPfpra,Tsra(and CETgrya,Tsrva); and, with modifica-\ntions ofthepreceding consonant-signlikethose noted above,\n\"5Ttra, \"5Tpr, ?Tdra.\nWhen}Tristobecombined with afollowing5fJr,it\nisthevowel which iswritten infull,with itsinitial char-\nacter, and theconsonant insubordination toit:thus,\nftrr.\n15.Further combinations, ofthree,orfour,oreven\nfiveconsonant-signs,aremade accordingtothesame rules.\nExamplesare:\nofthree consonants, f[ttva, 1STddhya, $3dvya,^\ndrya,SETdhrya,t3psva, 5JT9^ya, ^Jstya,^1hvya;\noffour consonants, ^Jktrya, ^Tnksya, ^Jstrya,\nrF?I tsmya;\noffiveconsonants, fF^U rtsnya.\nThe manuscripts, andthetype-fontsaswell,differ from oneanother\nmore intheir managementofconsonant combinations than inany other\nrespect,often having peculiarities which oneneeds alittle practicetounder-\nstand. Itisquiteuseless togiveinagrammar thewhole series ofpossible\ncombinations (manyofthem excessively rare) which areprovidedforinany\ngiven type-font,oreven inall.There isnothing which due familiarity\nwith thesimple signs andwith theabove rules ofcombination willnotenable\nthestudent toanalyse andexplain.\n16.Asigncalled theavagraha ('separator') namely,\nvl--isused inthemanuscripts,sometimes inthemanner\nofahyphen, sometimes asamark ofhiatus, sometimes to\nmark theelision ofinitial 51aafter final^eorsqjo(135).\nInprintedtexts,especially European,itisordinarilylimited\n"
+    },
+    {
+        "page": 28,
+        "content": "I.ALPHABET. [16\ntotheuselastmentioned: thus. ?t^^f^telbruvan, HT\nso'bravit, for teabruvan, soabramt.\nThe signisused tomark anomission ofsomething.\nInsome texts, ithasalso thevalue ofahyphen.\nSignsofpunctuationareIand II.\n17.Thenumeralfiguresare\n11,^2,\\3,94,H5,|6,b7.TT8.$9,0.\nIncombination, toexpress largernumbers,theyare\nused inpreciselythesamewaywithEuropean digits:thus,\n^H25,^0630, <(000 1000, ^T7b 1879.\n18.TheHindu grammarianscallthedifferent sounds, and\nthecharacters representing them, byakara('maker')added to\nthesound ofthe letter,ifavowel, ortotheletter followed by\na,if-aconsonant. Thus, thesound orcharacter aiscalled\nakara; kiskakara: and soon.Butthekara isalso omitted,\nand a,ka,etc. areused alone. Ther,however, isnever called\nrakara, butonlyraorrepha ('snarl':theonly exampleofa\nspecific name foranalphabetic element ofitsclass). The ami-\nsvara andmsargaarealsoknown bythese names alone.\nCHAPTER II.\nSYSTEM OFSOUNDS; PRONUNCIATION.\nI.Vowels.\n19.THE a,i,and^^-vowels. TheSanskrit hasthese\nthree earliest andmost universal vowels ofIndo-European\nlanguage,inboth short andlongform %aand TTa,\n^tand^I.3uand3Tu.Theyaretobepronouncedin\nthe\"Continental\" or\"Italian\" manner asinfarorfather,\npinandpique, pulland rule.\n20.The aistheopenest vowel, anutterance from theex-\n"
+    },
+    {
+        "page": 29,
+        "content": "24] VOWELS. 9\npandedthroat;itstands, therefore., innorelation ofkindred\nwith anyoftheclasses ofconsonantal sounds. The ianduare\nclose vowels, made withmarkedapproachofthearticulating\norganstooneanother :iispalatal, andshades through yinto\nthepalatal andguttural consonant-classes;uissimilarly related,\nthrough v,tothelabialclass, asinvolvinginitsutterance a\nnarrowing androundingofthelips.\nThePaninean scheme (commentarytoPanini's grammar,i.1.9)classes a\nasguttural, butapparently only inorder togive that series aswell asthe\nrest avowel: noone ofthePraticakhyas putsaintoone class with ketc.\nAllthese authorities concur incalling the iandw-vowelsrespectively palatal\nand labial.\n21.The short aisnotpronouncedinIndia with the full\nopennessofa,asitscorresponding short, butusuallyasthe\n\"neutral vowel\"(Englishso-called \"short\",ofbut, son, blood,\netc.). Thispeculiarity appears very early, being acknowledged\nbyPanini andbytwo ofthePraticakhyas (APr.i.36;VPr. i.\n72),which call theutterance samvrta, 'coveredup,dimmed'.\nItis,however, ofcourse notoriginal ;and itisjustly wont to\nbeignored byWestern scholars(exceptthosewhohave studied\ninIndia).\n22.Thea-vowels aretheprevailing vowel-sounds ofthe\nlanguage, being about twice asfrequentasalltheothers(in-\ncluding diphthongs)takentogether. The -vowels, again,are\nabout twice asnumerous asthew-vowels. And, ineachpair,\ntheshort vowel ismore than twice(2y2to3times)ascommon\nasthelong.\nFormore precise estimates offrequency,ofthese and oftheother\nalphabetic elements, and fortheway inwhich they were obtained, see\nbelow, 75.\n23.The rand/-vowels. Tothethree simplevo-\nwelsalready mentioned theSanskrit addstwo others, the\nr-vowels and the/-vowel, both ofthemplainly generated\nbytheabbreviation ofsyllables containinga^\"ror^T/\nalongwith another vowel :the TRrcoming (almost always:\nsee237, 241-3) fromT|\"arorfra,theFT/from FTal.\nSome oftheHindu grammarians add tothealphabetalso alongI;\nbut this isonly forthesake ofanartificial symmetry, since thesound does\nnotoccur inasingle genuine word inthelanguage.\n24.Thevowel :fjrissimplyasmooth oruntrilled\nr-sound, assumingavocalic office insyllable-making\n"
+    },
+    {
+        "page": 30,
+        "content": "10II.SYSTEM OFSOUNDS. [24\nas,byalike abbreviation,ithasdone also incertain Sla-\nvonic languages.Thevowel FT Iisan-sound similarly\nuttered liketheEnglish/-vowel insuchwords asable,\nangle,addle.\nThemodern Hindus pronouncethese vowels as ri, ri,li\n;orevenIri), having longlost thehabit and thefacilityof\ngivingavowel value tothepurerand^-sounds. Their example\niswidelyfollowed byEuropeanscholars;andhence also the\n(distortingandquite objectionable)transliterationsri,n,li.\nThere isno-real difficultyinacquiringandpractisingthetrue\nutterance.\nSome ofthegrammarians (seeAPr. i.37,note) attempttodefine more\nnearly theway inwhich,inthese vowels, areal ror^-element iscombined\nwith somethingelse.\n25.Like their corresponding semivowels, randI,these\nvowels belong respectivelyinthegeneral lingual anddental class-\nes\n;theeuphonicinfluence ofrandf(180)shows thisclearly.\nTheyaresoranked inthePaninean scheme;butthePraticakhyas\ningeneral strangelyclassthem with thejihvcimuliya sounds, our\n\"gutturals\".\n26.The short risfound inevery varietyofword andof\nposition, and isnotrare, being justabout asfrequentaslongu.\nLong fisverymuch more unusual, occurring onlyincertain\npluralcases ofnoun-stems inr(374, 378).The /ismetwith\nonlyinsome oftheforms and derivatives ofasinglenotvery\ncommon verbal root(kip).\n27.Thediphthongs. Ofthefourdiphthongs, two,\nthe^eand ETTo,areingreat part original Indo-European\nsounds. IntheSanskrit, they wear theaspectofbeing\nproductsoftheincrement orstrengtheningof^iand3u\nrespectively; andtheyarecalled thecorresponding guna-\nvowels tothelatter(seebelow, 235). The other two,^ai\nand tau,arebytheprevalent andpreferable opinionheld\ntobeofpeculiar Sanskritgrowth (thereisnocertain trace\nofthem tobefound even intheZend); theyare also in\ngeneralresults ofanother andhigherincrement of^iand\n3u,towhichtheyarecalled thecorrespondingvrddhi-\nvowels(below, 235). But allarelikewise sometimesgene-\n"
+    },
+    {
+        "page": 31,
+        "content": "32 DIPHTHONGS. 1j\nrated byeuphoniccombination(127); and TTo,especially,\niscommon asresult ofthealteration ofafinal *3Ras175).\n28.The^eand 3TFoare,both inIndia and inEu-\nrope, usually pronouncedastheyaretransliterated that\nis,aslonge(English \"long \",oreinthey]ando-sounds,\nwithoutdiphthongalcharacter.\nSuchthey apparently already were totheauthors ofthe\nPraticakhyas, which, while ranking them asdiphthongs\\san-\ndfa/afaara), giverulesrespectingtheirpronunciationinamanner\nimplying them tobevirtually unitary sounds. But their euphonic\ntreatment(131-4) clearly shows them tohave been still atthe\nperiod when theeuphonic laws established themselves, asthey\nofcourse were attheirorigin,realdiphthongs,ai(a-f- *)and\nau\\a-\\- tf).From them, onthesame evidence, theheavier or\nvrddhi diphthongs weredistinguished bythelengthoftheir a-\nelement, asai(a-{- i]andau(a-\\-u).\nThe recognisable distinctness ofthetwoelements inthevrddhi-diph-\nthongs isnoticed bythePraticakhyas (seeAPr. i.40,note); buttherelation\nofthose elements iseither defined asequal,ortheaismade oflessquan-\ntitythan the iand u.\n29.Thelighterorywwa-diphthongsaremuch more frequent\n(6or7times)than theheavier orvrddhi- diphthongs, andthe\neand aithan the oandau(ahalfmore).Bothpairsare\nsomewhat more than half ascommon asthesimpleiandu-\nvowels .\n30.Thegeneral name given bytheHindu grammarianstothevowels\nissuara, 'tone'; thesimple vowels arecalled samanaksara, 'homogeneous\nsyllable', andthediphthongsarecalled sandhyaksara,icombination-syllable'.\nThepositionoftheorgans intheir utterance isdefined tobeoneofopenness,\norofnon-closure.\nAstoquantity and accent, seebelow. 76 ff.;80 ff.\nII.Consonants.\n31.TheHindu name for'consonant' isvyan/ana,'mani-\nfester'. The consonants aredivided bythegrammariansinto\nsparca,icontact' or'mute', antahstha, 'intermediate' orisemivowel',\nandusman, 'spirant'. Theywillhere betaken upanddescribed\ninthis order.\n32.Mutes. Themutes, sparca,aresocalled asinvolving\nacomplete closure orcontact\\sparca], andnotanapproximation\n"
+    },
+    {
+        "page": 32,
+        "content": "12II.SYSTEM OFSOUNDS. [32\nonly,ofthemouth-organs bywhich theyareproduced. Theyare\ndivided into fiveclasses orseries(varga], accordingtotheorgans\nandpartsoforgans bywhich thecontact ismade;andeach\nseries iscomposedoffivemembers, differing accordingtothe\naccompanimentsofthecontact.\n33.The fivemute-series arecalledrespectively guttural,\npalatal, lingual (orcerebral), dental, and labial; andthey\narearrangedintheorder asjustmentioned, beginningwith\nthecontact made furthest back inthemouth, comingfor-\nward frompointtopoint,andendingwith thefrontmost\ncontact.\n34.Ineach series there aretwosurdmembers, two\nsonant, andonenasal (whichisalsosonant):forexample,\ninthelabial series,tj^pandVf^ph, 3^band^bh,and ?Tm.\nThemembers arebytheHindu grammarians calledrespectively 'first',\n'second', 'third', 'fourth1\n,and 'last' or'fifth'.\nThesurd consonants areknown asaghosa, 'toneless', andthesonants\nasghosavant, 'having tone'\n;andthedescriptionsofthegrammarians arein\naccordance with these terms. All alike recognise adifference oftone, and\nnotinanymanner adifference offorce, whether ofcontact orofexpulsion,\nasseparating thetwogreat classes inquestion. That thedifference depends\nonvivara, 'opening', orsarhvara, 'closure'(oftheglottis),isalsorecognised\nbythem.\n35.The firstandthirdmembers ofeach series arethe\nordinary correspondingsurdandsonant mutes ofEuropean\nlanguages: thus,^kand\\g, t^tand d,qpand^b.\n36.Nor isthecharacter ofthenasalanymore doubtful.\nWhatq^misto^pand3[^,or^ntot^tand <d,that\nisalsoeach other nasal toitsown series ofmutes :asonant\nexpulsion intoandthrough thenose, while themouth-\norgansareinthemute-contact.\nTheHindu grammarians give distinctly this definition. The nasal\n(anunasika, 'passing through thenose') sounds aredeclared tobeformed by\nmouth andnosetogether; ortheirnasality (anunasikya)tobegiven them\nbyunclosure ofthenose.\n37.Thesecond andfourth ofeach series areaspirates:\nthus, beside thesurdmute^kwehave thecorresponding\n"
+    },
+    {
+        "page": 33,
+        "content": "38 ASPIRATE MUTES. 1$\nsurdaspirate ^kh,andbeside thesonantTf.^,thecorres-\npondingsonantaspirate %gh.Ofthese, theprecisechar-\nacter ismore obscure and difficult.\nThat theaspirates,allofthem, arerealmutes orcontact sounds, and\nnotfricatives(likeEuropeanthandphandch, etc.),isbeyond question.\nItisalsonotdoubtful inwhatway thesurd M,forexample,differs\nfrom theunaspiratedt:such aspiratesarefound inmany Asiatic languages,\nandeven insome European:they involve theslipping-outofanaudible bit\nofflatusoraspiration between thebreach ofmute-closure andthefollowing\nsound, whatever itmay be.They areaccurately enough represented bythe\nthetc., withwhich,inimitation oftheLatin treatment ofthesimilar ancient\nGreekaspirates, weareaccustomed towrite them.\nThe sonant aspirates aregenerally understood and described asmade\ninasimilar way, with aperceptibleft-sound after thebreach ofsonant\nmute-closure. But there areinsuperabletheoretical difficulties intheway\nofacceptingthisexplanation ;andsome ofthebest phonetic observers(as\nA.J.Ellis) deny thatthemodern Hindu pronunciationisofsuch acharacter,\nanddefine theelement following themute asa\"glottal buzz\", rather,oran\nemphasizedutterance ofthebeginning ofthesuceeding sound. Thequestion\nisone ofgreat difficulty, anduponittheopinionsofthehighest authorities\nare stillmuch atvariance. Sonant aspiratesare still inuse inIndia,in\nthepronunciationofthevernacular aswell asofthelearned languages.\nBythe Pratic.akhyas, theaspiratesofboth classes arecalled sosman:\nwhich might mean either 'accompanied byarush ofbreath' (taking usman\ninitsmore etymological sense), or'accompanied byaspirant' (below, 59).\nAndsome authorities define thesurd aspiratesasmade bythecombination\nofeach surd non-aspirate with itsown corresponding surd spirant; andthe\nsonantaspirates,ofeach sonantnon-aspirate with thesonant spirant, the\nft-sound(below, 65). But thiswould make thetwo classes ofaspirates of\nquite diverse character, andwould alsomake ththesame asts,thasts,ch\nascf which isinanymeasureplausible only ofthe last. Panini hasno\nname foraspirates ;thescheme given inhiscomment(toi.1.9)attributes\ntothem mahaprana, 'great expiration1\n,and tothenon-aspirates alpaprana^\n'smallexpiration'.\nItisusualamong Europeanscholars topronounceboth\nclasses ofaspiratesasthecorresponding non-aspirateswith\nafollowingh:forexample,2TthnearlyasinEnglishboat-\nhook,mphasinhaphazard,Tdhasinmadhouse, and so\non.This is(aswehave seen above) confessedlyaccurate\nonlyasregardsthesurdaspirates.\n38.Thesonantaspirates are(intheopinionofmost),or\natleastrepresent, original Indo-European sounds, while thesurd\n"
+    },
+    {
+        "page": 34,
+        "content": "14II.SYSTEM OFSOUNDS. i38\naspiratesaregenerally regardedasaspecialIndian development.\nTheformer aremore than twice ascommon asthelatter. The\nunaspirated (non- nasal)mutes areverymuch more frequent\n(5times)than theaspirates (forthespecial frequencyofbhand\noriginal gh,see50and66);andamong them thesurds aremore\nnumerous (2Y2times)than thesonants. The nasals(chieflyn\nandm)arenearlyasfrequentasthesurd non-aspirates.\nWetakeupnow theseveral mute-series.\n39.Guttural series:^k,T^kh, JT^,\\gh,3\"n.\nThese aretheordinary Europeankand^-sounds,with their\ncorresponding aspiratesandnasal(the last, likeEnglish ng\ninsinging}.\nThe gutturalsaredefined bythePratic.akhyas asmade bycontact of\nthebase ofthetongue with thebase ofthejaw, andthey arecalled, from\ntheformer organ, jihvamullya, 'tongue-root sounds'. ThePaninean scheme\ndescribes them simplyasmade inthethroat(kantha). From theeuphonic\ninfluence ofakonafollowings(below, 180), wemay perhapsinfer that\nintheir utterance thetongue waswelldrawn back into thehinder mouth.\n40.The kisbyfarthecommonest oftheguttural series,\noccurring considerably more often than alltheother four taken\ntogether.The nasal, exceptasstandingbefore oneoftheothers\nofthe series,isfound onlyasfinal(afterthe loss ofafol-\nlowing k),and inaverysmall number ofwords.\n41.TheSanskritgutturalseriesrepresents onlyaminority\nofIndo-European gutturals;these lasthave suffered more and\nmore general corruption thananyother class ofconsonants. By\nprocessesofalteration which areprovedtohavebeguninthe\nIndo-European period,since thesame words exhibit connected\nchangesalso inotherlanguages ofthefamily, thepalatal mutes,\nthepalatalsibilantc,and theaspiration h,have come from\ngutturals. Seethese various sounds below.\n42.Palatal series :r(c,^ch, s?Fj,37jh,3Tn.This\nwhole series isderivative, being generated bythecorruptionof\noriginal gutturals. The ccomes from anoriginalk asdoes\nalso, byanotherdegreeofalteration, thepalatalsibilant c(see\nbelow, 64).The/,inlikemanner, comes from ag;butthe\nSanskrit jincludes initself twodegrees ofalteration, onecor-\nrespondingtothealteration ofktoc,theother tothat ofktog\n(seebelow, 219 :intheZend, these twodegreesareheld dis-\ntinctly apart). The cissomewhat morecommon than thej\n(aboutasfour tothree). Theaspirate chisverymuch less fre-\nquent (atenth ofc),andcomes from theoriginal groupsk.\nThesonantaspirate jhisexcessively rare(occurringbutonce\n"
+    },
+    {
+        "page": 35,
+        "content": "45] PALATAL ANDLINGUAL MUTES. 15\nintheVedic texts, andnothalf-a-dozen times intheBrahma-\nnas);where found, itiseither onomatopoeticorofanomalous\nornotIndo-European originiintheso-called rootujh,itcomes\nfromjandh}.The nasal, n,never occurs except immediately\nbefore or,inasmall number ofwords, also after(201)\noneoftheothers ofthe series.\n43.Hence, intheeuphonic processesofthelanguage,the\ntreatment ofthepalatalsisinmany respects peculiar. Insome\nsituations, theoriginal unaltered guttural shows itself or,as\nitappearsfrom thepointofview oftheSanskrit, thepalatal\nreverts toitsoriginal guttural. Nopalatalever occurs asafinal.\nThejisdifferently treated, accordingasitrepresentstheone\northeother degreeofalteration. And candj(except artificially,\ninthealgebraicrules ofthegrammarians)donotinterchange,\nascorrespondingsurdandsonant.\n44.Thepalatalmutes arebyEuropeanscholars, asby\nthemodern Hindus also, pronouncedwith thecompound\nsounds ofEnglishchandj(inchurch andjudge}.\nTheir description bythe oldHindu grammarians, however, gives them\nanot less absolutely simplecharacter than belongs tothe other mutes.\nThey arecalled talavya, 'palatal1\n,and declared tobeformed against the\npalate bythemiddle ofthetongue. They seem tohave been, then,\nbrought forward inthemouth from theguttural point, andmade against the\nhard palateatapoint not farfrom thelingual one (below, 45), butwith\ntheupperflatsurface ofthetongue instead ofitspoint. Such sounds, in\nalllanguages, pass easily into the(English)chand ./-sounds. The value\nofthechasmaking theprecedingvowel \"long byposition\" (227), and its\nfrequent origination from t-f-p(203),lead tothesuspicion that it.atleast,\nmayhave had this character from thebeginning:compare 37,above.\n45.Lingual series:t,3\"th,Zd,?odh,HTn.The\nlingual mutes arebyallthenative authorities denned as\nuttered with thetipofthetongueturned upanddrawn\nback intothedome ofthepalate (somewhat astheusual\nEnglish smooth rispronounced). Theyarecalled bythe\ngrammarians murdhanya, literally'head-sounds,capitals,\ncephalics'; which term isinmany European grammars\nrendered by'cerebrals' .Inpractice, among EuropeanSans-\nkritists, noattemptismade todistinguish them from the\ndentals : tispronouncedlike rTt,Jdlike\ne[\"d,and so\nw.ith the rest.\n"
+    },
+    {
+        "page": 36,
+        "content": "16II.SYSTEM OFSOUNDS. [46\n46.The lingualsareanother non-originalseries ofsounds,\ncoming mainly from thephoneticalteration ofthenextseries,\nthedentals, but also inpart occurringinwords thathave no\ntraceable Indo-European connection, andareperhapsderived from\ntheaboriginal languagesofIndia. Thetendencytolingualization\nisapositiveoneinthehistoryofthelanguage:dentalseasily\npassinto linguals under theinfluence ofcontiguousorneighbor-\ninglingual sounds, butnotthecontrary;and allthesounds\noftheclassbecome markedly more frequentinthelater litera-\nture. The conditions oftheir ordinaryoccurrence arebriefly\nthese :a.scomes froms,much more rarely fromq,j,ks,in\neuphoniccircumstances stated below(180, 218, etc.);b.a\ndental mute followingsisassimilated toit,becoming lingual\n(t,th, n);c.nisoften changedtonafter alingual vowel or\nsemivowel orsibilant inthesameword(189 etc.) ;d.dh,which\nisofveryrare occurrence, comes from assimilation ofadental\nafter s(198 a)orh(222);e. tanddcomeoccasionally by\nsubstitution forsome other sound which isnotallowed tostand\nasfinal(142, 145).Whenoriginatedinthese ways,thelingual\nletters mayberegardedasnormal;inanyother cases oftheir\noccurrence, theyareeither productsofabnormalcorruption,or\nsignsofthenon-Indo-Europeancharacter ofthewords inwhich\nthey appear.\nInacertain number ofpassages numerically examined (below, 75),\ntheabnormal occurences oflingual mutes were less than half ofthewhole\nnumber (74out of159), andmost ofthem(43)were ofn: allwere found\nmore frequentinthelater passages. IntheRig-Veda, only 15words have\nanabnormal t;only 6,such ath;only 1,such adh;about 20(including\n9roots, nearly allofwhich havederivatives) show anabnormald,besides\n9thathavend;and30(including1root)show an.\nTaken alltogether, thelingualsarebyfartherarest class\nofmutes(about iy2percent, ofthealphabet) hardlyhalf\nasfrequent even asthepalatals.\n47.Dental series:<^t,Z^tk, <[d,V^dk,^n.These\narecalledbytheHindus alsodantya, 'dental', and are\ndescribed asformed attheteeth(orattheroots ofthe\nteeth), bythetipofthetongue. Theyarepracticallythe\nequivalents ofourEuropean t,d,n.\nButthemodern Hindus aresaid topronouncetheir dentals with the\ntipofthetongue thrust wellforward against theupper teeth, sothatthese\nsounds getaslight tinge ofthequality belongingtotheEnglish andModern\nGreek tft-sounds. Theabsence ofthatquality intheEuropean (especially\n"
+    },
+    {
+        "page": 37,
+        "content": "52]LABIAL MUTES; SEMIVOWELS 17\ntheEnglish) dentals isdoubtless thereason why tothe earofaHindu the\nlatterappearmore analogous with hislinguals, andhe isapttousethe\nlingualsinwriting European words.\n48.Thedentals areoneofthethreeIndo-European original\nmute-classes. Intheir occurrence inSanskrittheyarejustabout\nasfrequentasalltheother four classes takentogether.\n49.Labial series: qpCfiph^b.^bh, 3}m.X^-X**X -X^^X\nThese sounds arecalledosthya,'labial1\n,bytheHindu gram-\nmarians also. They are, ofcourse, theequivalents ofour\np,b,m.\n50.Thenumerical relations ofthe labials arealittlepe-\nculiar. Owingtotheabsence(oralmost entireabsence)ofbin\nIndo-European,theSanskrit balso isgreatly exceeded infre-\nquency bybh,which isthemostcommon ofallthesonant\naspirates,asphistheleastcommon ofthesurd. Thenasalm(not-\nwithstandingitsfrequent euphonic mutations when final: 212ff.)\noccurs justabout asoften asalltheother fourmembers ofthe\nseries together.\n51.Semivowels: fy,^\"r,s$ I,3v.Thename given tothis\nclass ofsounds bytheHindu grammariansisantahstha, \"standing between'\neither from their character asutterances intermediate between vowel and\nconsonant, or(more probably) from thecircumstance oftheir being placed\nbetween themutes and spirants inthearrangement oftheconsonants.\nThesemivowels areclearlyakinwith theseveral mute series\nintheirphysical character, andtheyare classifiedalong with\nthose series though notwithout some discordances ofview\n-bytheHindu grammarians. Theyaresaid tobeproduced\nwith theorgans \"slightlyincontact\"(isatsprsta),or\"inimperfect\ncontact\"(duhsprsta].\n52.The^risclearly shown byitsinfluence inthe\neuphonic processesofthelanguagetobealingual sound,\noronemade with thethetipofthetongue turned upinto\nthedome ofthepalate.Itthus resembles theEnglish\nsmoothr,and,likethis,, seems tohavebeen untrilled.\nThePaninean scheme reckons rasalingual. None ofthePraticakhyas,\nhowever, does so;norarethey entirely consistent with oneanother inits\ndescription. Forthemostpart, they define itasmade at'the roots ofthe\nteeth'. Thiswould giveitapositionlikethat ofthevibrated r;butnoau-\nthority hints atavibration asbelongingtoit.\nWhitney, Grammar. 2\n"
+    },
+    {
+        "page": 38,
+        "content": "1II.SYSTEM OFSOUNDS.[52\nInpointoffrequency,rstands very highonthe listof\nconsonants ;itisabout equalwith v,n,m,andy,andonly\nexceeded byt.\n53.The TIisasound ofdentalposition,and isso\ndefined andclassed byallthenative authorities.\nThe peculiarcharacter ofanZ-sound,asinvolving expulsion atthe\nsides ofthetongue along with contact atitstip,isnotnoticed byanyHindu\nphonetist.\nItisadisputed questionwhether rand Iwere distinguished from one\nanother inIndo-European speech ;intheSanskrit, atany rate, they arevery\nwidely interchangeable, both inroots andinsuffixes: there ishardly aroot\ncontaining an Iwhich does notshow alsoforms with r;words written with\ntheone letter arefound inother texts,orinother partsofthesametexts,\nwritten with theother. Inthelater periodsofthelanguage they aremore\nseparated,andthe Ibecomes decidedly more frequent, though always much\nrarer than ther(onlyas1to7or8or10).\n54.Some oftheVedic texts have another -sound, written\nwith aslightlydifferent character(itisgivenattheendofthe\nalphabet, 5),which issubstituted foralingual d(asalso the\nsame followed byhforadh]when occurring between twovowels.\nItis,then, doubtless alingual I,onemade bybreach(atthe\nsides ofthetongue)ofthelingualinstead ofthedental mute-\nclosure.\n55.The <ETyinSanskrit, asinotherlanguages gene-\nrally,stands intheclosestrelationshipwith thevowel^i\n(shortorlong);thetwoexchangewith oneanother in\ncases innumerable.\nAnd intheVeda(asthemetre shows) an iisvery often toberead\nwhere, inconformity with the rules ofthelater Sanskrit euphony, ayis\nwritten. Thus, the final i-vowel ofaword remains ibefore aninitial vowel;\nthat ofastem maintains itself unchanged before anending; andanending\nofderivation asya,tya has iinstead ofy.Such cases willbecon-\nsidered inmore detaillater,asthey arise. Theconstancy ofthephenomenon\nincertain words and classes ofwords shows that thiswasnomerely optional\ninterchange. Very probably, the Sanskrit yhadeverywhere more ofan\ni-character thanbelongstothecorresponding European sound.\n56.Theyisbyitsphysical character apalatal utterance;\nand itisclassed asapalatal semivowelbytheHinduphonetists.\nItisoneofthemostcommon ofSanskrit sounds.\n57.The ofvispronounced asEnglishorFrench v\n(German w)bythemodern Hindusexcept whenpre-\n"
+    },
+    {
+        "page": 39,
+        "content": "60] SEMIVOWELS; SPIRANTS. 19\nceded byaconsonant inthesamesyllable,inwhich case\nithasthesound ofEnglish w;andEuropeanscholars follow\nthesamepractice (withorwithout thesameexception).\nByitswhole treatment intheeuphonyofthelanguage,\nhowever, thevstands related toanw-vowelpreciselyasyto\nan/-vowel. Itis,then, avonly accordingtotheoriginal\nRoman value ofthat letter that istosay, az^-sound inthe\nEnglish sense :though (aswas stated above forthey]itmay\nwellhave been lessmarkedly separated from uthanEnglish w,\nmore likeFrench ouinouietc. But, astheoriginal whasin\nmost European languages been changedtov(English),soalso\ninIndia, and thatfrom avery earlytime :thePaninean scheme\nandtwo ofthePraticakhyas (VPr. andTPr.) distinctly define\nthesound asmade between theupperteeth andthelowerlip\nwhich, ofcourse, identifies itwith theordinary modern v-sound.\nAsamatter ofpractice,theusualpronunciation need notbe\nseriously objectedto\n;yetthestudent should not fail tonote\nthat therules ofSanskrit euphony andthename of\"semivowel\"\nhave noapplication excepttoazp-sound intheEnglish sense :\nav-sound(German w)isnosemivowel, butaspirant, standing\nonthesame articulatestage with theEnglish M-sounds and\nthe/.\n58.Visclassed asalabial semivowelbytheHindu\nphoneticalauthorities. Ithasasomewhatgreater frequency\nthan they.\nIntheVeda, under thesame circumstances asthey(above, 55),visto\nberead asw.\n59.Spirants. Unjler thenameusman(literally 'heat,\nsteam, flatm\\ which isusually and wellrepresented by\n'spirant', some oftheHindu authorities include allthe\nremaining sounds ofthealphabet;othersapplytheterm\nonlytothethree sibilants andtheaspiration towhich\nitwillhere alsoberestricted.\nTheterm isnotfound inthePaninean scheme; bydifferent treatises\ntheguttural andlabialbreathings, these andthevisarga, orallthese andanu-\nsvara, arecalled usman (seeAPr. i.31note). Theorgans ofutterance are\ndescribed asbeing inthepositionofthemute-series towhich each spirant\nbelongs respectively, butunclosed, orunclosed inthemiddle.\n60.TheHs.Ofthethreesibilants, orsurdspirants,\nthis istheoneofplainest and leastquestioned character:\n2*\n"
+    },
+    {
+        "page": 40,
+        "content": "20II.SYSTEM OFSOUNDS.[60\nitistheordinary Europeans--ahissexpelled between\nthetongue and theroof ofthemouthdirectly behind the\nupperfront teeth.\nItis,then, dental, as itisclassed byalltheHindu\nauthorities. Itistheoneprimitive Indo-European sibilant.\nNotwithstandingthegreatlosses which itsuffers inSanskrit\neuphony, byconversion totheother sibilants, tor,tovisarga,\netc.,itisstillveryhighamongtheconsonants intheorder of\nfrequency,orconsiderably morecommon than both theother\ntwo sibilants together.\n61.The Ef*.Astothecharacter ofthissibilant, also,\nthere isnogroundforrealquestion:itistheoneproduced\ninthelingual position,orwith thetipofthetonguere-\nverted intothedome ofthepalate.Itis,then, akind of\ns^-sound; andbyEuropeanSanskritists itispronounced\nasanordinarysh(French ch,Germansch\\noattempt\nbeing made(anymore than inthecase oftheotherlingual\nsounds :45)togiveititsproper lingual quality.\nItslingual character isshown byitswhole euphonic influence,\nand itisdescribed andclassed aslingual byalltheHindu author-\nities(theAPr. adds,i.23, that thetongueinitsutterance\nistrough-shaped).Initsaudiblequality,itisasA-sound\nrather than as-sound; and,inthe considerablevarietyof\nsibilant-utterance, even inthesamecommunity,itmaycoincide\nwithsomepeople'ssh.Yetthegeneral andnormal shispalatal\n(seebelow, 63);andtherefore thesign s,marked inaccordance\nwith theotherlingual letters,istheonlyunexceptionable trans-\nliteration fortheHindu character.\nInmodernpronunciation inIndia,sismuch confounded with kh;and\ntheMSS. areapttoexchange thecharacters. Later grammatical treatises,\ntoo,take note oftherelationship (seeWeber'sPratijfia, p.84).\n62.This sibilant(aswasnoticed above, 46,and willbe\nmoreparticularly explained below, 180fF.)isnooriginal sound,\nbutaproduct ofthelingualization ofsunder certain euphonic\nconditions. Theexceptionsareextremely few(9outof145\nnoted occurrences :75),andofapurely sporadiccharacter. The\nRig-Veda has(apart fromysah,182) only twelve words which\nshow asunder other conditions.\nThe final sofaroot has insome cases attained amore independent\n"
+    },
+    {
+        "page": 41,
+        "content": "65 SPIRANTS. 21\nvalue, and does notrevert toswhen theeuphonicconditions areremoved,\nbutshows anomalous forms (225.\n63.The 5Tg>This sibilant isbyallthenative author-\nities classed and described aspalatal,nor isthereany-\nthinginitshistoryoritseuphonictreatment tocastdoubt\non itscharacter assuch. Itis,then, made with the flat\nofthetongue againsttheforwardpartofthepalatalarch\nthat istosay,itistheusual andnormal sA-sound. By\nEuropeanscholars itisvariously pronounced moreoften,\nperhaps,assthan assh.\nThetwo s/i-sounds, sandp,aremade inthesame part ofthemouth\n(thesprobably rather furtherback), butwith adifferent partofthetongue ;\nandthey aredoubtless notmore unlike than,forexample, thetwot-sounds,\nwritten tand t;and itwould benotlesspropertopronounce them both as\noneshthan topronounce thelinguals and dentals alike. Toneglect the\ndifference ofsand fismuch less tobeapproved. Thevery nearrelationship\nofsand fisattested bytheir euphonic treatment, which istoaconsiderable\nextent thesame, andbytheir notinfrequentconfusion bythewriters\nofmanuscripts.\n64.Aswasmentioned above(41),ther,likec,comes\nfrom thecorruptionofanoriginal &-sound, byloss ofmute-\ncontact aswell asforward shift ofthearticulating point.In\nvirtue ofthis derivation,itsometimes(thoughlessoften thanc)\n\"reverts\" tok thatis,theoriginalkappearsinstead of it\n;\nwhile, ontheother hand, asas/j-sound,itistoacertain\nextent convertible to s.Inpointoffrequency,itslightly\nexceeds the latter.\n65.Theremaining spirant,^hyisordinarily pronounced\nliketheusual Europeansurdaspirationh.\nThis isnot,however,itstrue character. Itisdefined byallthenative\nauthorities asnotasurdelement, butasonant(orelseanutterance inter-\nmediate between thetwo) ;and itswhole value intheeuphonyofthelanguage\nisthat ofasonant: butwhat isitsprecise value isvery hard tosay. The\nPaninean scheme ranks itasguttural,asitdoes alsoa:thismeans nothing.\nThePratic.akhyas bringitintonorelation with thegutturalclass :oneofthem\nquotes theopinionofsome authorities that \"ithasthesame positionwith\nthebeginning ofthefollowing vowel\" (TPr. ii.47) which sofaridenti-\nfies itwith our h.There isnothing initseuphonicinfluence tomark it\nasretaining any trace ofgutturally articulated character. Bysome ofthe\nnative phonetistsitisidentified with theaspirationofthesonant aspirates\n"
+    },
+    {
+        "page": 42,
+        "content": "22II.SYSTEM OFSOUNDS.[65\nwith theelement bywhich,forexample, ghdiffers from g.Thisview\nissupported bythederivation ofhfrom theaspirates (next paragraph), by\nthat ofl+hfrom dh(54), andbythetreatment ofinitial hafter afinal\nmute (163).\n66.The h,asalready noticed,isnotanoriginal sound,\nbutcomes innearlyallcases from anoldergTi(forthefew\ninstances ofitsderivation from dhand bh,seebelow, 223).It\nisavastly more frequent sound than theunchanged gh(namely,\nas7to1):more frequent, indeed, than any oftheguttural\nmutes exceptk. Itappears,likej(219),toinclude initself\ntwo stagesofcorruptionofgh:onecorrespondingwith that of\nktoc,theother with that ofktoc;seebelow, 223, forthe\nroots belongingtothetwo classesrespectively.Like theother\nsounds ofguttural derivation,itsometimes exhibits \"reversion\"\ntoitsoriginal.\n67.The :h.orvisarga (visarjamya,asitisuniformly\ncalled bythePrati^akhyas andbyPanini, probablyas'be-\nlongingtotheend' ofaword), appearstobemerelyasurd\nbreathing,afinal A-sound(intheEuropeansense ofh),\nuttered inthearticulating positionoftheprecedingvowel.\nOnePraticakhya (TPr.ii.48)gives just this lastdescriptionofit. It\nisbyvarious authorities classed withft,orwith handa:allofthem are\nalike sounds inwhose utterance themouth-organs have nodefinite shaping\naction.\n68.Thevisargaisnotoriginal, butalways onlyasubsti-\ntute forfinal sorr,neither ofwhich isallowed tomaintain\nitself unchanged.Itisacomparatively recent member ofthe\nalphabetic system;theother euphonic changes offinal sand r\nhave notpassed through visargaasanintermediatestage. And\ntheHindu authorities areconsiderably discordant with onean-\nother astohow farhisanecessary substitute, andhow fara\npermitted one, alternative with asibilant, before afollowing\ninitial surd.\n69.Before asurdguttural orlabial, respectively, some of\nthenative authoritiespermit, while others require, conversion of\nfinal sorrinto theso-calledjihvamutiya andupadhmariiya spi-\nrants. Itmaybefairly questioned, perhaps, whether these two\nsounds arenotpure grammatical abstractions, devised(likethe\nlong /-vowel :23)inorder toround outthealphabettogreater\nsymmetry. Atany rate, neither printed texts normanuscripts\n(exceptintherarest andmostsporadic cases) makeanyaccount\nofthem. Whatever individual charactertheymayhavemust be,\n"
+    },
+    {
+        "page": 43,
+        "content": "71]ANUSVABA. 23\nitwould seem,inthedirection ofthe(German)chand^sounds.\nWhen written atall,theyarewont tobetransliterated by%\nand(p.\n70.The -anusvara, norw\n7isanasal soundlacking\nthat closure oftheorgans which isrequiredtomake a\nnasal mute(36);initsutterance there isnasal resonance\nalongwithsomedegreeofopennessofthemouth.\n71.There isdiscordance ofopinion bothamong theHindu phonetists\nandtheirmodern European successorsrespectingthe real character ofthis\nelement :hence alittle detail isnecessary here with regardtoitsoccurrence\nand their views ofit.\nCertain nasals inSanskrit areofservilecharacter, always tobeassi-\nmilated toafollowing consonant,ofwhatever character thatmay be. Such\narefinalminsentence-combination(213), thepenultimate nasal ofaroot,\nand anasal ofincrement(255) ingeneral.Ifone ofthese nasals stands\nbefore acontact-letter ormute,itbecomes anasal mute correspondingtothe\nlatter thatis,anasal utterance inthesame positionofthemouth-organs\nwhich gives thesucceeding mute.If,ontheother hand, thefollowing con-\nsonant does notinvolve acontact (being asemivowel orspirant), thenasal\nelement isalso without- contact :itisanasal utterance withunclosed mouth-\norgans. Thequestion is,now, whether this nasal utterance becomes merely\nanasal infection ofthepreceding vowel, turning itinto anasal vowel(as\ninFrench on,en,un, etc., byreason ofasimilar loss ofanasal mute);or\nwhether itisanelement ofmore individualcharacter, having place between\nthevowel andtheconsonant; or,once more, whether itissometimes theone\nthing andsometimes theother. TheopinionsofthePraticakhyas andPanini\narebriefly asfollows :\nTheAtharva-Pratic.akhya holds that theresult iseverywhere anasalized\nvowel, except when normisassimilated toafollowingI,inthatcase,the\nnormbecomes anasal I:thatis,thenasal utterance ismade inthe\n^-position, andhasaperceptiblei-character.\nThe other Praticakhyas teach asimilar conversion into anasal counter-\nparttothesemivowel, ornasalsemivowel,before yand Iandv(notbefore\nralso). Inmost oftheother cases where theAtharva-Pratic.akhya acknow-\nledges anasal vowelnamely, before randthespirantstheothers\nteach theintervention after thevowel ofadistinct nasal element,called the\nanusvara, 'after-tone'.\nOfthenature ofthis nasal afterpiecetothevowel nointelligiblyclear\naccount isgiven. Itissaid(RPr.)tobeeither vowel orconsonant;itis\ndeclared (RPr., VPr.)tobemade with thenosealone,or(TPr.)tobenasal\nlikethenasal mutes;itisheld bysome (RPr.)tobethesonant tone of\nthenasal mutes;initsformation,asinthat ofvowel andspirant,there is\n(RPr.) nocontact. Astoitsquantity,seefarther on.\nThere are,however,certain cases and classes ofcases where these other\n"
+    },
+    {
+        "page": 44,
+        "content": "24II.SYSTEM OFSOUNDS.[71\nauthorities alsoacknowledgeanasal vowel.So,especially, wherever afinal\nnistreated (208;asifitwere ns(itshistorically olderform); and alsoin\nasmall number ofspecified words. They alsomeiitiou thedoctrine ofnasal\nvowel instead ofanusvara asheldbysome (and TPr. isuncertain andincon-\nsistent initschoice between theoneandtheother).\nInPanini, finally,theprevailingdoctrine isthat ofanusvaraevery-\nwhere; and itiseven allowed inmany cases where thePratic.akhyas pre-\nscribe only anasal mute. But anasal semivowel isalsoallowed instead be-\nfore asemivowel, and anasal vowel isallowed inthecases (mentioned above)\nwhere some ofthePratic,akhyas requireitbyexception.\nItisevidently afairquestion whether thisdiscordance anduncertainty\noftheHinduphonetistsisowingtoareal difference ofutterance indiffer-\nentclasses ofcases andindifferentlocalities,orwhether toadifferent scho-\nlastic analysis ofwhat isreally everywhere thesame utterance. Ifanu-\nsvara isanasal element following thevowel,itcannot well* beanything\nbuteither aprolongation ofthesame vowel-sound withnasality added, ora\nnasalised bitofneutral-vowel sound(inthelattercase, however, thealtering\ninfluence ofan iorw-vowel onafollowingsought tobeprevented, which\nisnotthecase: see183).\n72.The assimilated nasal element, whether viewed as\nnasalized vowel, nasal semivowel, orindependent anusvara, has\nthevalue ofsomething added, inmakingaheavy syllable, or\nlength byposition (79).\nThePraticakhyas (VPr., RPr.) give determinations ofthequantity of\ntheanusvara combining with ashort andwith alongvowelrespectively to\nmake along syllable.\n73.Two differentsigns,1and-,arefound intheMSS.,\nindicating thenasal sound here treated of.Usually theyare\nwritten above thesyllable, andtheretheyseem mostnaturally\ntoimplyanasal affection ofthevowel ofthesyllable, anasal\n(anunasika)vowel. Hence some texts(Sama andYajur Vedas),whentheymean arealanusvara, bring one ofthesignsdown\ninto theordinary consonant-place ;buttheusageisnotgeneral.Asbetween thetwosigns, someMSS.employ,ortend toemploy,the -where anasalized(anunasika) vowel istoberecognized,andelsewhere the1;and this distinction isconsistently observed\ninmany European printed texts; and theformer iscalled the\nanunasikasign: but itisvery doubtful whether thetwo arenot\noriginally andproperly equivalent.\nItis^averycommon custom ofthemanuscriptstowrite\ntheanusvara-signforanynasalfollowing thevowel ofasyllable,either before another consonant orasfinal(notbefore avowel),without anyreference towhether itistobepronounced asnasal\nmute, nasalsemivowel, oranusvara. Someprinted texts follow\nthisslovenly andundesirable'habit;butmost write anasalmute\n"
+    },
+    {
+        "page": 45,
+        "content": "751 TABLE OFALPHABETIC SOUNDS. 25\nSon.\nSurd\nSon.\nSurdwhenever itistobepronounced excepting where itisan\nassimilated m(213).\nItisconvenient also intransliteration todistinguishthe\nassimilated mbyaspecial sign, m,from theanusvara ofmore\nindependent origin, n;and thismethod willbefollowed inthe\npresent work.\n74.This isthewhole systemofsoundsrecognised bythe\nwritten character; forcertain transitional sounds, more orless\nwidely recognised inthetheories oftheHinduphonetists,see\nbelow, 230.\n75.Thewhole spoken alphabet, then,maybearranged\ninthefollowing manner, soastoshow, sofarasispossible\ninasingle scheme, therelations andimportantclassifications\nofitsvarious members :\nVowels\nSemivowels\nNasals\nAnusvara\nAspiration\nVisarga\nSibilants\n111asp.\n1-27\nbunasp.\n.46\nphasp.\n"
+    },
+    {
+        "page": 46,
+        "content": "2(jII.SYSTEM OFSOUNDS. [75\nThe figuressetunder the characters givetheaverage\npercentageoffrequencyofeach sound, found bycounting the\nnumber oftimes which itoccurred inanaggregateof10,OOC\nsounds ofcontinuous text, in10different passages,^of1,000\nsounds each,selected from different epochsofthe literature :\nnamely,twofrom theRig-Veda, onefrom theAtharva-Veda,\ntwofrom different Brahmanas, andoneeachfromManu, Bha-\ngavad-Gita, Qakuntala, Hitopade9a,andVasavadatta*.\nIII.Quantityofsounds and syllables.\n76.TheHindu grammarianstake thepainstodefine\nthequantityofaconsonant (withoutdistinction among\nconsonants ofdifferent classes)ashalfthatofashort vowel.\n77.Theyalso define thequantityofalong (dirgha)\nvowel ordiphthongastwice thatofashortvowel making\nnodistinction inthisrespectbetween thegunaand the\nprefab-diphthongs.\n78.Besides these twovowel-quantities,theHindus\nacknowledgeathird,called pluta (literally 'swimming'),\norprotracted, andhavingthree moras,orthree times the\nquantityofashort vowel. Aprotractedvowel ismarked\nbyafollowing figure3:thus, 5TT$a3.\nTheprotractedvowels arepracticallyofrareoccurrence(in\nRV., three cases; inAV., fifteen; intheBrahman aliterature,\nrather morefrequent). Theyareused incases ofquestioning,\nespeciallyofabalancing between two alternatives, and also of\ncallingtoadistance orurgently. Theprotractionisofthe last\nsyllableinawrord, orinawholephrase;and theprotracted\nsyllablehasusuallytheacute tone, inaddition toany other\naccent thewordmayhave\n;sometimes ittakes also anusvara, or\nismade nasal.\nExamples are: adhdh svidasi3d updri svid asl3t(RV.),'wasit,forsooth,\nbelow? wasit,forsooth, above?' iddmbhUydS id$3m Hi(AV.), 'saying,isthis\nmore,oristhat?' dgndSi pdtmvdSh s6mam piba (TS.),'ohAgni! thou with\nthyspouse!drink thesoma'.\nAdiphthongisprotracted byprolongationofits first ora-element:\nthus,etoa3i, otoa3u.\n*See J.A.0.S.,vol.X.\n"
+    },
+    {
+        "page": 47,
+        "content": "82]QUANTITY. 37\nThe signofprotractionisalsosometimes written astheresult ofac-\ncentual combination, when so-called kampa occurs: seebelow, 90b.\n79.For metricalpurposes, syllables (notvowels)are\ndistinguished bythegrammariansas'heavy' (guru)or'light'\n(laghu). Asyllableisheavyifitsvowel islong,orshort\nandfollowed bymore than oneconsonant(\"long bypo-\nsition\"). Anusvara andvisargacount asfullconsonants in\nmakingaheavy syllable. The lastsyllableofapada (pri-\nmarydivision ofaverse)isreckoned aseither heavyor\nlight.\nThe distinction interms between thedifference oflong and short in\nvowel-sound andthat ofheavy and light insyllable-construction isvaluable,\nandshould beretained..\nIV.Accent.\n80.Thephenomenaofaccentare,bytheHindu gram-\nmarians ofallages alike, described andtreated asdepend-\ningonavariation oftone orpitch;ofanydifference of\nstress involved, theymake noaccount.\n81.Theprimarytones(svara)oraccent-pitchesare\ntwo :ahigher (udatta, 'raised'),oracute;and alower\n(anudatta, 'notraised'),orgrave. Athird(calledsvarita :\naterm ofdoubtfulmeaning),isalwaysofsecondary origin,\nbeing (when notenclitic :seebelow, 85)theresult ofactual\ncombination ofanacute vowel andafollowing grave vowel\nintoonesyllable.Itisalsouniformlydefined ascompound\ninpitch, aunion ofhigher andlower tone within the\nlimits ofasingle syllable.Itisthus identical inphysical\ncharacter with theGreek andLatin circumflex, andfully\nentitled tobecalled bythesame name.\n82.Strictly, therefore, there isbutone distinction oftone\nintheSanskrit accentual system:theaccentedsyllableisraised\nintone above theunaccented\n;while then further, incertain\ncases ofthefusion ofanaccented andanunaccented element\n"
+    },
+    {
+        "page": 48,
+        "content": "2II.SYSTEM OFSOUNDS.[82\nintoone syllable,thatsyllableretains thecompoundedtone of\nboth elements.\n83.The svarita orcircumflex isonly rarely found ona\npure longvowel ordiphthong,butalmost always onasyllable\ninwhich avowel, short orlong,ispreceded byayorvre-\npresentinganoriginallyacute torw-vowel.\nIntransliteration, inthiswork, theudatta oracute willbe\nmarked with theordinary signofacute, andthesvarita orcir-\ncumflex(asbeingadownward slide ofthevoiceforward)with\nwhat isusuallycalled thegraveaccent :thus, d,acute,yaor\nva,circumflex.\n84.ThePraticakhyas distinguish andname separatelythecircumflexed\ntones arising bydifferent processesofcombination :thus, thecircumflex is\ncalled\na.Ksaipra ('quick'), when anacute iorw-vowel(shortorlong)iscon-\nverted intoyorvbefore adissimilar vowel ofgrave tone :thus, vyhpta\nfrom vi-apta, apsvantdrfrom apsuantdr.\nb.Jatya ('native')ornitya ('own'), when thesame combination lies\nfurther back, inthemake-upofastem orform, and soisconstant, or\nbelongs toaword inallcircumstances ofitsoccurrence: thus, kva(from fcwa),\nsvhr(stiar), nybk (nfak), budhnya (budhnfa), kanyh (fcanla), nadyas (nadf-as),\ntanvh(tanU-a).\nThewords ofboth these classes areintheVeda, inthegreat majority\nofcases,toberead with restoration oftheacute vowel asaseparate syllable:\nthus, apsu antdr, suar, nadias,etc..Insometexts, part ofthem are\nwrittencorrespondingly:thus, suvar, tanuva, budhnfya.\nc.Praflista, when theacute andgrave vowels areofsuch character that\nthey arefused into along vowel ordiphthong (128):thus divi 'va(RV.\nandAV.), fromdM iva; shdgata (TS.), from su-udyata; nai'vhJ\nfriiyat\n(B.), fromndevdapniyat.\nd.Abhinihita, when aninitial grave aisabsorbed byafinal acute e\nor6(135): thus, te'bruvan, from teabruvan;t>'bravit, from soabravlt.\n85.Butfurther, theHindugrammarians agreeinde-\nclaringthe(naturally grave) syllable following anacute,\nwhether inthesame orinanother word,tobesvarita or\ncircumflex--unless, indeed, itbeitself followed byan\nacute orcircumflex; inwhich case itretains itsgravetone.\nThis iscalledbyEuropean scholars theenclitic ordepend-\nentcircumflex.\nThus, intenaand teca,thesyllable naandword caare\nregarded andmarked ascircumflex :but intena teand teca\nsvarthey aregrave.\n"
+    },
+    {
+        "page": 49,
+        "content": "87]METHODS OFWRITING ACCENT. 29\nThis seems tomean that thevoice, which isborne upatthehigher\npitchtotheend oftheacutesyllable, does notordinarily droptograve pitch\nbyaninstantaneous movement, butdescends byamore orlessperceptible\nslide inthecourse ofthefollowing syllable. NoHindu authority suggests\nthetheory ofamiddle orintermediate tone forthe enclitic, anymore than\nfortheindependent circumflex. Forthemostpart,thetwo areidentified\nwith oneanother,intreatment anddesignation. The enclitic circumflex is-\nlikewise divided into anumber ofsub-varieties, with different names: they\nareoftoo little consequencetobeworthreporting.\n86.The essential difference ofthetwokinds ofcircum-\nflex isshownclearly enough bythese facts :a.theindependent\ncircumflex takes theplaceoftheacute astheproper accent of\naword, while the enclitic isthemereshadowfollowing anacute,.\nandfollowingitinanother wordpreciselyasinthesameword;\nb.theindependentcircumflex maintains itscharacter in all\nsituations, while the enclitic before afollowing circumflex or\nacute loses itscircumflex character, andbecomesgrave ;more-\nover, c.inmanyofthesystemsofmarking accent(below, 88),\nthetwo arequite differentlyindicated.\n87.Theaccentuation ismarked inmanuscripts onlyofthe\nolder literature :namely,inthedifferent Vedic texts, intwoof\ntheBrahmanas(TaittirlyaandQatapatha),and intheTaittiriya-\nAranyaka. There areanumber ofmethods ofwriting accent,\nmore orlessdifferent from oneanother\n;theonefound inMSS .of\ntheRig-Veda, which ismost widely known, andofwhich most of\ntheothers areonly slight modifications,isasfollows :theacute\nsyllableisleftunmarked\n;thecircumflex, whether independent\norenclitic, hasabriefperpendicularstroke above; andthegrave\nnext preceding anacute or(independent)circumflex hasabrief\nhorizontal stroke below. Thus\nstcfjtri juhdti; rp^T tanvh; WTkva.\nTheintroductory gravestroke below, however, cannot begivenifanacute\nsyllable isinitial, whence anunmarked syllable atthebeginning ofaword\nistobeunderstood asacute;andhencealso,ifseveral grave syllables precede\nanacute atthebeginning ofasentence, theymust allalike have thegrave\nsign. Thus,\n^:mdrah; ftte;cflf^fHkarisyasi ;rT%TfTTtuvijatd.\nAllthegrave syllables, however, which follow amarked circumflex are left\nunmarked, until theoccurrence ofanother accented syllable causes theone\nwhich precedesittotake thepreparatory stroke below. Thus,\nsudfcikasamdrk ;\njV__jbutH^lfctiH^J JNIH sudfftkasamdrg gdvam\n"
+    },
+    {
+        "page": 50,
+        "content": "30II.SYSTEM OFSOUNDS.[88\n88.The other methods itisnotworth while toattempttosetforth.\nTheymaybefound illustrated inthedifferent texts, andexplained bythe\neditors ofthem. Inpart,their peculiaritiesconsist inother forms orplaces\ngiventothegrave andcircumflex signs.'Insome methods, theacute isitself\nmarked, byaslight stroke above. Inseveral,thteindependent circumflex is\ndistinguishedfrom the enclitic. Themost peculiar systems arethescanty\nandimperfectone oftheQatapatha-Brahmana,with asingle sign, written\nbelow;andthehighlyintricate oneoftheSama-Veda,with adozen different\nsigns,written above.\n89.Inthiswork, aseverything giveninthedevanagarichar-\nacters isalso givenintransliteration,itwill ingeneral be\nunnecessarytomark theaccent exceptinthetransliterated form\n;\nwhere, however, thecase isotherwise, there willbeadoptedthe\nmethod* ofmarking onlytheaccentedsyllables,theacute\nandtheindependentcircumflex :thelatter bytheusual svarita-\nsign,theformer byasmall u(for udatta]above thesyllable:\nthus,\n*^3\\ indra, Mi4dgne,\nThese being given, everything elsewhich theHindu theory recognises\nasdependent onandaccompanying them can readily beunderstood asim-\nplied.\n90.Thetheory oftheSanskrit accent, ashere given (aconsistent and\nintelligible body ofphenomena), hasbeen overlaid bytheHindu theorists,\nespeciallyofthePraticakhyas, with anumber ofadded features, ofamuch\nmore questionable character. Thus :\na.Theunmarked grave syllables, following acircumflex(eitheratthe\nendofasentence,ortillthenear approach ofanotheracute),aredeclared\ntohave thesame high tonewith the(alsounmarked)acute. They arecalled\npracayaorpracita ('accumulated': because liable tooccur inanindefinite\nseries ofsuccessivesyllables).\nb.Thecircumflex, whetherindependentorenclitic,isdeclared tobegin\nonahigher pitch thanacute, and todescend toacutepitchinordinary cases :\nthe'concluding instant ofitbeing brought down tograve pitch, however,in\nthe case ofanindependent circumflex which isimmediately followed by\nanother ascent ofthevoice tohigher pitch (inacute orindependentcir-\ncumflex).\nThis lastcase,ofanindependent circumflex followed byacute orcir-\ncumflex, receivespeculiar written treatment. IntheRig-Veda method, a\nfignre 1or3issetafter thecircumflexedvowel, according asitisshort or\nlong, andthesigns ofaccent arethusapplied:\n*Introduced byBohtlingk, andused inthePetersburg lexicon andelsewhere.\n"
+    },
+    {
+        "page": 51,
+        "content": "93]ACCENT. 31\n:apsv alntdh from apsu antdh;\nraybS 'vdnih from rayo avdnih .\nThe other methods, more orlessakin withthis, need notbegiven.\nInthe scholastic utterance ofsuch asyllable ismade apeculiar quaver\norroulade ofthevoice, which iscalled kampa orvikampana.\nC.Panini gives theambiguous name ofeka$ruti ('monotone')tothepra-\ncita syllables, andsays nothingoftheupliftingofthecircumflex toahigher\nplane: heteaches, however,adepression below thegrave pitchforthemark-\nedgrave syllable before acute orcircumflex, callingitsannatara(otherwise\nanudattatara).\n91.Thesystem ofaccentuation asmarked intheVedic texts hasassum-\nedinthe traditional recitation oftheBrahmanic schools apeculiar and\nartificial form,inwhich thedesignated syllables, grave and circumflex\n(equally,theenclitic andtheindependent circumflex), have acquired acon-\nspicuous value, while theundesignated, theacute, hassunk into insigni-\nficance *.\n92.TheSanskrit accent taughtinthenative grammars and\nrepresented bytheaccentuated texts isessentiallyasystemof\nword-accentonly. Nogeneral attemptismade(anymore than\nintheGreeksystem)todefine ormark asentence-accent, the\neffect oftheemphasis andmodulation ofthesentence inmo-\ndifyingtheindependentaccent ofindividual words. Theonly\napproachtoitisseen inthetreatment ofvocatives andpersonal\nverb-forms.\nAvocative isusually without accentexceptatthebeginning\nofasentence :forfurther details, seethechapter onDeclension.\nApersonal verb-form isusually accentless inanindepend-\nentclause, except whenstandingatthebeginningoftheclause :\nforfurther details, seethechapter onConjugation.\n93.Certain other words also are, usuallyoralways, without\naccent.\na.Theparticles ca,vd,u,sma, iva, cid, svid, fta,arealways without\naccent.\nb.Thesame istrue ofcertain pronouns andpronominal stems :md,me,\nndu, na<, tvd, te,i?am, vas, ena-, tva-.\nc.The cases ofthepronominal stem aaresometimes accented andsome-\ntimes accentless.\nAnaccentless word isnotallowed tostand atthebegin-\nningofasentence :alsonotofapadaorprimarydivision of\naverse\n;apada is,inallmattersrelatingtoaccentuation, treat-\nedlikeanindependentsentence.\n*Hang, Wedischer Accent, inAbh. d.Bayr. Akad.,vol.XIII, 1874.\n"
+    },
+    {
+        "page": 52,
+        "content": "32II.SYSTEM OFSOUNDS.\n94.Some words havemore than asingleaccentedsyllable.\nSuch are :\na.Dual collective compounds:asfndravdrunau.\nb.Afewother compounds,inwhich eachmember irregularly retains\nitsownaccent :astdnundpat, vdnaspdti, brhaspdti.Inarare case ortwo,\nalso their further compounds,asbrhaspdtipramitta.\nC.Infinitive datives intavdf: asetavaf,\nd.Aword naturally barytone, buthaving itsfinal syllable protracted:\nseeabove, 78.\ne.The particle vdvd (intheBrahmanas).\n95.OntheplaceoftheaccentedsyllableinaSans-\nkritword there isnorestriction whateverdepending upon\neither thenumber orthequantityoftheprecedingor\nfollowing syllables.The accent rests where therules of\ninflection orderivation orcomposition place it,without\nregardtoanythingelse.\nThus, indre, agnau, indrena, agnina, agriintim, bahucyuta,\ndnapacyuta, parjdnyajinvita,abhimatisahd}dnabhimlatavarna, abhicas-\nticatana, hiranyavacimattama.\n96.Since theaccent ismarkedonlyintheolder litera-\nture, and thestatements ofthegrammarians, with the\ndeduced rules ofaccentuation, arefarfrombeingsufficient\ntosettle allcases, theplaceofthe stress ofvoice fora\nconsiderablepartofthevocabularyisundetermined. Hence\nitisageneralhabit withEuropeanscholars topronounce\nSanskrit wordsaccordingtotherules oftheLatin accent.\n97.Inthiswork, theaccent ofeachword andform will\ningeneral bemarked, sofarasthere isauthority determiningits\nplace andcharacter. Wherespecific words andforms arequoted,\ntheywillonly besofaraccentuated astheyarefound with\naccent inaccentuated texts.\n"
+    },
+    {
+        "page": 53,
+        "content": "103] 33\nCHAPTER III.\nRULES OFEUPHONIC COMBINATION.\nIntroductory.\n98.THE individual elementscomposingalanguage as\nactually used are itswords. These are inpart uninflected\nvocables(indeclinables, particles);inthemain, theyarein-\nflected forms.\n99.The inflected forms areanalysable into inflective en-\ndings, ofdeclension orofconjugation, and inflected stems to\nwhich thoseendingsareadded.\n100. The inflected stems, again, areforthemostpart\nasare also inparttheuninflected wordsanalysableinto\nderivativeendingsorsuffixes, androots, towhich, eitherdirectly\northrough more primary stems, those endingsareadded.\nBut, notafewstems andparticles areirreducible toroots;and,onthe\notherhand, roots areoften used directly asinflectedstems,indeclension as\nwell asinconjugation.\n101. The roots are, inthecondition ofthelanguageas\nitliesbefore us,theultimate attainable elements;toagreat\nextent notactually ultimate, but,whereotherwise, theresult of\nprocesses ofdevelopmenttooirregular andobscure tobemade\nthesubjectoftreatment inagrammar.\n102. The formativeprocesses bywhich both inflectional\nforms and derivative stems aremade, bytheaddition ofendings\ntobases and toroots, aremoreregular andtransparent inSan-\nskrit than inanyother Indo-European language, andthegram-\nmaticalanalysisofwords into their component elements is\ncorrespondingly complete. Hence itbecame themethod ofthe\nnative grammarians, andhascontinued tobethatoftheirEuro-\npean successors, toteach thelanguage bypresentingtheendings\nandstems androots intheir analysed forms, andlaying down\nthewaysinwhich these are tobecombined togethertomake\nwords. And hence astatement oftheeuphonicrules which\ngovern thecombination ofelementsoccupiesinSanskrit grammar\namoreprominent andimportant placethan inother grammars.\n103. Moreover, theformation ofcompound words, bythe\nputting together oftwo ormore stems,isaprocessofvery\nexceptional frequencyinSanskrit;and thiskind ofcombination\nalsohas itsowneuphonicrules. And once more, intheform\nWhitney, Grammar. 3\n"
+    },
+    {
+        "page": 54,
+        "content": "34III-EUPHONIC COMBINATION.[103\ninwhich thelanguageishanded down tousbythe litera-\nture, thewords composingasentence orparagraphareadapted\ntoandcombined with each other bynearlythesame ruleswhich\ngovernthemakingofcompounds,sothat itisimpossibleto\ntakeapartandunderstand thesimplestsentence inSanskrit\nwithout understandingthose rules. Hence also agreatly added\ndegreeofpractical importance belongingtothesubjectof\neuphoniccombination.\nThis euphonic interdependenceofthewords ofasentence, which is\nunknown toanyother languageinanything like thesame degree, isshown\ntobeatleast inconsiderable measure artificial, implying anerection into\nnecessary and invariable rules ofwhat intheliving language were only\noptional practices, bytheevidence oftheolder dialect oftheVedas andthe\nyounger Prakritic dialects, inboth ofwhich these rules(especially asregards\nhiatus: 113) arevery often violated.\n104.Wehave, therefore, inthe firstplacetoconsider the\neuphonic principles andlawswhich govern thecombination of\ntheelements ofwords(andtheelements ofthesentence);and\nthen afterward totakeupthesubjectofinflection, under the\ntwoheads ofdeclension andconjugation:towhich willsucceed\nsome account oftheclasses ofuninnected words.\n105. The formation ofconjugatioixal stems(tenseand\nmode-stems, etc.)willbetaught,asisusual, inconnection\nwith theprocesses ofconjugationalinflection\n;thatofuninflected\nwords, inconnection with thevarious classes ofthose words.\nButthegeneral subjectofderivation, ortheformation ofde-\nclinablestems, willbetakenupbyitself later forabriefpre-\nsentation;and itwillbefollowed byanaccount oftheformation\nofcompound stems.\nAlthough, namely, thegeneral plan ofthis series ofgrammars excludes\nthesubject ofderivation, yet, because ofthecomparative simplicity and\nregularity oftheprincipal processes ofderivation inSanskrit, andtheimport-\nance tothestudent ofaccustoming himself from thebeginning totrace those\nprocesses, inconnection with theanalysis ofderived forms, back totheroot,\nanexception willbemade inregard tothesubject inthepresent work.\n106.Weassume,then, forthepurposesofthepresent\nchapter, theexistence ofthematerial ofthelanguageina\ngrammatically analysed condition, intheform ofroots, stems,\nandendings.\n107.What istobetaken astheproper form ofaroot or\nstem isnotinallcases clear. Verymanyofboth classes showmapart oftheir derivatives astronger and inapartaweaker\nform(260). Thisis,inmostcases, theonly difficulty affecting\n"
+    },
+    {
+        "page": 55,
+        "content": "108]INTRODUCTORY. 35\nstems whether, forexample, weshall speakofderivatives in\nmat orinmant, ofcomparatives inyasorinyam,ofaperfect\nparticipleinvatorinvaiis orinus.TheHindu grammarians\nusually givetheweaker form asthenormal one,andderive the\nother from itbyastrengthening change ;someEuropean author-\nities adopttheoneform andsome theother :thequestionisan\nunessential one, givingrise tonopractical difficulty.\n108.Asregardstheroots, thedifficultyisgreater, partly\nbecause complicatedwith otherquestions, arising frompractices\noftheHindu grammarians, which have beenmore orlesswidely\nfollowed bytheir Europeansuccessors. Thus :\na.More than half ofthewhole number ofroots given bytheHindu\nauthorities (which areover2000) have never been found actually used in\ntheliterature; andalthough some ofthese may yetcome tolight, ormay\nhave existed without finding their way intoany ofthepreserved literary\ndocuments,itiscertain thatmost arefictitious, made inpartfortheex-\nplanation ofwords claimed tobetheirderivatives, andinpartforother and\nperhaps unexplainable reasons. Ofthe roots unauthenticated bytraceable\nusenoaccount willbemade inthisgrammar or,ifatallconsidered,\ntheywillbecarefully distinguished from theauthenticated.\nb.Those roots ofwhich theinitial nand sareregularly converted to\nnand *after certainprefixesarebytheHindu grammarians given asbe-\nginning withnands:noEuropean authority follows thisexample.\nc.Anumber ofroots endinginawhich isirregularly treated inthe\ninflection ofthepresent-systemarewritten intheHindu lists withdiph-\nthongseoratoro;and so,after thisexample, bymany Western scholars.\nHere theywillberegarded asa-roots :compare below, 251. The oofsuch\nroots, especially,ispurely arbitrary ;noforms made from theroot justify it.\nd.Theroots showing interchangeably r,ar,and irand Irorurandur\nforms arewritten bytheHindus withr,orwithf,orwith both. Here also\nthefisarbitrary and indefensible. Asbetween randar,even thelatest\nEuropean authorities areatvariance, and itmaybeleft tofurther research to\nsettle whether theoneortheother isalone worthy tobeaccepted. Here (mainly\nasamatter ofconvenience :compare below, 237)ther-forms willbeused.\ne.Intheother cases ofroots showing astronger and aweakerform,\nchoice isingreat measure amatter ofminor consequence unless further\nresearch andthesettlement ofpending phonetic questions shall show that\ntheone ortheother isdecidedly thetruer andmoreoriginal. From the\npointofview oftheSanskrit alone, thequestionisoften impossibleto\ndetermine.\nf.TheHindusclassify assimpleroots anumber ofderived stems :\nreduplicated ones,asdidhi, jagr, daridra;present-stems,asurnu;and\ndenominative stems, asavadhir, kumar, sabhaj, mantr, santv, arth, andthe\nlike. These areinEuropean works generally reduced totheir true value.\ng.But itisimpossible todraw any definite linebetween these cases\n"
+    },
+    {
+        "page": 56,
+        "content": "36III.EUPHONIC COMBINATION.[108\nandothers inwhich root-forms evidently ofsecondary origin have attained a\ndegreeofindependentvalue inthelanguage which almost orquite entitles\nthem torank asindividual roots. Even theweak andstrong forms ofthe\nsame root asvadandvand,citandcint,mahandmahh mayhave\nsuch adifference ofusethatthey count astwo;oradifference ofinflection\ncombined with adifference ofmeaning inaroothasthesame effect as\ninvrvrnoti andvrvrnlte, inhajahati andhajihite; oranevident present-\nstem becomes aseparateroot asjinvandpinv. Not afew roots occur\ninmore orless clearly related groups, themembers ofwhich areofvarious\ndegreesofindependence. Thus, aconsiderable class ofrootsshow anadded a;\nandsuch asmna anddhma arereckoned onlyasside-forms ofmanand\ndham; whileJra,pra, pya, psa,and others, presumably made inthesame\nmanner, figureasseparate from their probable originals. Many final con-\nsonants ofroots have thevalue of\"root-determinatives\",orelements of\nobscure orunknown origin added tosimpler forms. Aclass ofderivative\nroots show signs ofreduplication,ascaks, jaks, dudh; orofadesiderative\ndevelopment,asbhaks andbhiks, ?rus, afes, naks. Yetanother classseem\n)ocontain aprepositionfused with aroot,asvyac, ap }andthe later ujh\nandvyas.\nWith most ofsuch cases itisnotthepart ofaSanskrit grammar, but\nrather ofageneral Indo-European comparative grammar, todealaccording\ntotheir historical character. Wemust becontent toacceptasroots what\nelements seem tohave onthewhole that value intheexisting condition of\nthelanguage; with due recognition ofauthorized differences ofopinion on\nmany points,aswell asofthe fact that further knowledgewill setmany\nthings now doubtful inaclearerlight.\nPrinciplesofEuphonic Combination.\n109.The rules ofcombination areinsomerespects\ndifferent, accordingasthey apply\na.totheinternalmake-upofaword, bytheaddition\nofderivative and inflectionalendingstoroots andstems;\nb.tothemore externalputting togetherofstems to\nmakecompound stems, andtheyetlooser andmore acci-\ndental collocation ofwords inthesentence.\nHencetheyareusuallydivided into rules ofinternal\ncombination(orsandhi, -putting together'), and rules of\nexternal combination.\n110. Inboth classes ofcases, however, thegeneral princi-\nples ofcombination arethesame and likewise, toagreat\nextent, thespecific rules. The differences dependinparton\n"
+    },
+    {
+        "page": 57,
+        "content": "113]GENERAL PRINCIPLES. 37\ntheoccurrence ornon-occurrence ofcertain combinations inthe\none class ortheother;inpart, onthedifference oftreatment of\nthesame sound asfinal ofaroot orofanending,theformer\nbeing much more persistentthan the latter\n;inpart, onthe\noccurrence inexternal combination ofcertain changes which\nareapparently phoneticbutreallyhistorical;and,most frequent\nandconspicuousofall,onthefact that(157)vowels andse-\nmivowels andnasals exercise asonantizinginfluence inexternal\ncombination, butnotininternal. Hence, toavoid unnecessary\nrepetitionaswell astheseparationofwhatreally belongsto-\ngether,therules forbothkinds ofcombination willbegiven below\ninconnection with one another.\n111. Moreover, before case-endings beginningwith bhand\n*(namely, bhis, bhyas, bhyam, su},thetreatment ofthe finals\nofstems isingeneralthesame asinthecombinations ofwords\n(pada)withoneanother whence those endingsaresometimes\ncalledjt?ae?a-endings, andthecases theyform areknown aspada-\ncases. And withsome ofthesuffixes ofderivation thesame is\nthecase.\nThe importanceofthis distinction issomewhat exaggerated bythe\nordinary statement ofit.Infact, dhistheonly sonant mute initial ofan\nending occurring inconjugation,asbhindeclension;and the difference of\ntheir treatment isinpartowing totheonecoming into collision usually\nwith thefinal ofarootandtheother ofanending, andinpart tothefact\nthat dh,asadental,ismore assimilable topalatals andlinguals than bh.\nAmore marked andproblematicdistinction ismade between suand the\nverbal endings si,sva, etc., especiallyafterpalatal sounds and s.\n112. Theleading rules ofinternal combination arethose which areof\nhighest andmost immediate importancetoabeginner inthelanguage, since\nhis first task istomaster theprincipal paradigms ofinflection; therules of\nexternal combination may better beleftuntouched until hecomes todealing\nwith words insentences, ortotranslating. Then, however, they areindis-\npensable, since theproper form ofthewords thatcompose thesentence is\nnottobedetermined without them.\nThegeneral principlesofcombination underlyingthe\neuphonic rules, anddeterminingtheir classification, maybestated\nasfollows :\n113.Hiatus. Ahiatus isavoided.\nThere arebuttwoorthree words inthelanguagewhich\nintheiracceptedwritten form exhibit successive vowels forming\ndifferentsyllables:theyaretitau,'sieve'(perhapsfortitasu,\nBE,.) andprdilga, 'wagon-pole' (forprayuga ?};and,inRV.,\nsuuti. For thenotinfrequentinstances ofcompositionand\nsentence combination where therecent loss ofsoryorv\nleaves apermanent hiatus, seebelow, 132ff., 175b,177.\n"
+    },
+    {
+        "page": 58,
+        "content": "38III.EUPHONIC COMBINATION.[113\nIthasbeen already noticed that intheVeda, asthemetrical form of\nthehymns plainly shows, there isnoavoidance ofhiatus,either asbetween\nthestem-finals andendings ofwords, between themembers ofcompounds,\norbetween thewords composingasentence. Incases innumerable, ay\nandv(especiallyafter twoconsonants,oralong vowel andconsonant) are\ntoberead as iandu.But also alongvowel issometimes toberesolved into\ntwo syllables oftenest, aintoa-a :thisresolution issometimes historical,\nbutordinarily purelymetrical. For details, seebelow. Itiswith regard\ntothehiatus thattherules ofthegrammatically regulated classical Sanskrit\naremost demonstrably andconspicuouslydifferent from themore living usages\nofthesacred dialect.\n114.Deaspiration. Anaspirate mute isliable to\nlose itsaspiration, beingallowed tostand unchanged only\nbefore avowel orsemivowel ornasal.\n115. Assimilation. Thegreat bodyofeuphonic\nchangesinSanskrit,aselsewhere,fallsunder thegeneral\nhead ofassimilation which takesplace both between\nsounds which aresonearlyalike thatthedifference between\nthem istooinsignificanttobeworthpreserving,andbetween\nthose which aresodiverse astobepractically incompat-\nible.\n116. Inpart,assimilation involves theconversion of\nonesound toanother ofthesameseries, without changeof\narticulating position ;inpart,itinvolves achangeofposition,\nortransfer toanother series.\n117. Ofchanges within theseries, themost frequent and\nimportant occur intheadaptationofsurdandsonant sounds to\noneanother;but thenasals and Ihave also incertain cases\ntheirspecial assimilative influence. Thus:\na.Inthetwo classes ofnon-nasal mutes andspirants, surdandsonant\narewholly incompatible ;nosurd ofeither class caneitherprecedeorfollow\nasonant ofeither.\nAmute, surd orsonant,isassimilated bybeing changedtoitscorres-\npondent oftheother kind;ofthespirants, thesurd sistheonlyonehaving\nasonantcorrespondent, namely r,towhich itisconvertible inexternal\ncombination.\nb.The nasals aremorefreely combinable: anasalmay eitherprecede\norfollow amute ofeitherkind, orthesonantspirant h;itmay also follow\nasurdspirant (sibilant); nonasal, however, everprecedesasibilant inthe\n"
+    },
+    {
+        "page": 59,
+        "content": "121]GENERAL PRINCIPLES. 39\ninterior ofaword (itischanged instead toanusvara); and inexternal com-\nbination their concurrence isusually avoided byinsertion ofasurdmute.\nc.Asemivowel has still lesssonantizing influence;and avowel least\nofall*\n:both arefreely preceded and followed bysounds ofevery other\nclass,intheinterior ofaword.\nBefore asibilant, however,isfound, ofthesemivowels, only randvery\nrarelyI.Moreover,inexternal combination, risoften changed toitssurd\ncorrspondent5.\nBut\nd.Incomposition andsentence-collocation,initial vowels andsemivowels\nandnasals alsorequiretheprecedingfinal tobesonant. And\ne.Before anasal andi,the assimilativeprocessissometimes carried\nfurther, bytheconversion ofafinalmute toanasal or Irespectively.\n118. Ofconversions involvingachangeofarticulateposi-\ntion, themostimportantarethose ofdental sounds tolingual,\nand, less often, topalatal. Thus :\na.Thedental sandnare\"very frequently converted toandnbythe\nassimilating influence ofcontiguous orneighboring lingual sounds :the,\neven bysounds namely,iandw-vowels andk which have themselves\nnolingual character.\nb.Anon-nasal dental mute is(withafewexceptionsinexternal\ncombination) made lingual when itcomes into collision with alingual sound.\nc.Thedental mutes andsibilant aremadepalatal byacontiguous palatal.\nBut also:\nd.Am(not radical) isassimilated toafollowing consonant, ofwhat-\never kind.\n119.Theeuphoniccombinations ofthepalatal mutes, the\npalatal sibilant, andtheaspiration,asbeing sounds derived by\nphoneticalteration from more original gutturals (42 ff.),are\nmadepeculiar andcomplicated bytwocircumstances :their rever-\nsion toagutturalform(ortheappearanceoftheunaltered\ngutturalinstead ofthem^\n;andthedifferent treatment of/andh\naccordingasthey representoneoranother degreeofalteration\ntheonetending,likec,more totheguttural reversion, the\nothershowing,likec,amore sibilant andlingualcharacter.\n120. Thelingualsibilants,also ofderivative character\n(fromdental s}:shows asradical final apeculiarandproblematic\nmode ofcombination.\n121.Extension andabbreviation ofconso-\nnant-groups. The native grammariansallow orrequire\ncertain extensions, byduplicationorinsertion,ofgroupsof\nconsonants. And, ontheother hand,abbreviation ofcer-\n*Inconformity with general phonetic law: seeSievers, Lautphysiologie, p.140.\n"
+    },
+    {
+        "page": 60,
+        "content": "40III.EUPHONIC COMBINATION.[121\ntain other groupsisallowed, andfound often practisedin\nthemanuscripts.\n122.Permitted Finals. Thepermittedoccurrence\nofconsonants attheend ofaword isquite narrowlyre-\nstricted. Ingeneral, onlyoneconsonant isallowed after\nthelastvowel; and thatmust beneither theaspiration,\nnorasibilant, norasemivowel (save rarely5J^,noran\naspiratemute, norasonant mute ifnotnasal, norapalatal.\n123.Increment andDecrement. Besides these\nmore orlessregular changes accompanyingthecombination\nofthepartsthatmake upwords, there isanother class of\nasomewhat different character, notconsistinginthemutual\nadaptationsoftheparts,butinstrengtheningorweakening\nchangesofthepartsthemselves.\n124. Itisimpossibletocarry through aperfectly systematic\narrangement ofthedetailed rules ofeuphonic combination, be-\ncause the different varieties ofeuphonic change more orless\noverlap and intersect oneanother. The order followed below\nwillbeasfollows :\n1.Rules ofvowel combination, fortheavoidance ofhiatus.\n2.Rules astopermittedfinals(sincethese underlie the\nfarther treatment offinal consonants inexternalcombination).\n3.Rules forloss ofaspirationofanaspiratemute.\n4.Rules ofsurdandsonant assimilation, including those\nforfinal sand r.\n5.Rules fortheconversion ofdental sounds tolingual\nandpalatal.\n6.Rules forthechanges offinal nasals, including those in\nwhich aformer finalfollowing thenasal reappearsincombi-\nnation.\n7.Rulesregarding thespecial changesofthederivative\nsounds thepalatal mutes and sibilant, theaspiration, and\nthelingual sibilant.\n8.Rules astoextension and abbreviation ofconsonant\ngroups.\n9.Rules forstrengthening andweakening processes.\nEverywhere, rules formoresporadic andless classifiable cases\nwillbegiven inthemostpractically convenient connection;and\ntheIndex willrender whathelpisneeded toward finding them.\n"
+    },
+    {
+        "page": 61,
+        "content": "127]VOWEL COMBINATION. 41\nRules ofVowel Combination.\n125.Theconcurrence oftwovowels, orofvowel and\ndiphthong,withoutintervening consonant,isforbidden by\ntheeuphonyofthelater orclassicallanguage.Itisavoided,\naccordingtothecircumstances ofthecase, either byfusion\nofthetwoconcurrent sounds into one.bythereduction of\noneofthem toasemivowel, orbydevelopmentofasemi-\nvowel between them.\nThe texts oftheolder orVedic dialect arewritten according tothe\neuphonicrules ofthelater, although inthem(aswaspointed outabove, 113)\nthehiatus isreally ofvery frequent occurrence. Hence they arenottobe\nread aswritten, butwith almost constant reversal oftheprocessesofvowel\ncombination which theyhave artificially undergone.\nTherules ofvowel combination, asregards both theresult-\ningsound and itsaccent, arenearlythesame ininternal and\ninexternal sandhi.\n126.Two similarsimplevowels, short orlong,coalesce\nandform thecorresponding longvowel: thus, twoa-yowels\n(eitherorboth ofthem short orlong) form 5TTa;two^-vow-\nels, ^l\\twow-vowels. T37w; and, theoretically, twor-vow-\nelsform^f,but itisquestionable whether thecase ever\npracticallyoccurs. Examplesare:\n:saca'prajah (ca-f-aprajah);\nail\\a(ati-\\-iva) ;\nsuktam (su-uktam);\ncga\"sit(raja -\\-asit) ;\niadhiqvarah (adhi-~i$varah).\nAstheabove examples indicate,itwillbethepractice everywhere in\nthiswork,intransliteration(butnotinthedevanagari text), toseparate\nindependent words;and ifaninitial vowel ofafollowing word hascoalesced\nwith afinal ofthepreceding,this willbeindicated byanapostrophe\nsingleifthe initial vowel betheshorter, double ifitbethelonger,ofthe\ntwodifferent initials which inevery case ofcombination yield thesame result.\n127.Ana-vowel combines with afollowing*-vowel to\ne/withanw-vowel, toRo;with ^Jr,toSTTar;with\n"
+    },
+    {
+        "page": 62,
+        "content": "42*\nIII.EUPHONIC COMBINATION.|127\n^Ttheoretically',to3^al;with^eor^ai,to^a*/with\n5JT or5tau,to3T#M.Examplesare:\nI?RTrajendra (raja-indra);\n\\\\hitopadegah (hita-upadec,ah).;\nmaharsih (maha-rsih);\nva(sa}cvaj;\nrajaigvaryam (raja-aigvaryam);\ndivaukasah (diva-okasah);\njvarausadham (jvara-ausadham).\nInsome oftheVedic texts, thevowel riswritten unchangedafter the\na-vowel, which,iflong,isshortened: thus,maharsih instead ofmaharsiti.\nThetwovowels, however,areusually pronouncedasone syllable.\nWhen successive words like indra aihiaretobecombined,the first\ncombination,toindra,ismadefirst, andtheresult isindre\"\n'/u(notindrai\"\n'hi,from indra e'hi).\n128.Asregardstheaccent ofthese vowel combinations,itis\ntobenoticed that, a.asamatter ofcourse, theunion ofacute\nwith acuteyields acute, andthatofgravewithgrave yields grave:\nthatofcircumflex with circumflex cannot occur;b.acircumflex\nwith followingacuteyields acute,the final grave element ofthe\nformer beingraised toacutepitch ;agravewith following acute\ndoes thesame, asnoupwardslide ofthevoice onasyllableis\nacknowledgedinthelanguage;but, c.when theformer ofthe\nfused elements isacute andthelattergrave, wemight expect\ntheresulting syllabletobeingeneral circumflex, torepresent\nboth theoriginal tones. Panini infact allows this accent in\neverysuch case; andinasingle accentuated Brahmana text(QB.),\nthecircumflex isregularlywritten. Butthelanguage shows, on\nthewhole, anindispositiontoallow thecircumflex toreston\neitherlongvowel ordiphthongasitssole basis, andtheacute\nelement issuffered toraise theother toitsown level ofpitch,\nmakingthewholesyllable acute. Theonly exceptiontothis,\ninmost ofthetexts,isthecombination ofiandt,which be-\ncomes I:thus, div\\'va,from divi iva; intheTaittiriyatexts\nalone such acase follows thegeneral rule, while uand u,in-\nstead, make ^:thus, shdgata fromsii-udgata.\n>129.Thee-vowels, thew-vowels, and ftr,before a\ndissimilar vowel oradiphthong,areregularlyconverted\neach into itsowncorresponding semivowel, ETyorcfvor\n!fr.Examplesare :\n"
+    },
+    {
+        "page": 63,
+        "content": "132]VOWEL COMBINATION. 43\nityaha(iti-\\-aha) ;\nmadhv iva(madku -f-wa);\nduhitrarthe(duhitr-arthe);\nstryasya (strl-\\-asya);\n^tfvadhvdi(vadhu-ai).\nBut ininternal combination(neverinexternal)the iand\nw-vowels arenotseldom changedinstead toiyanduv and\nthisespeciallyinmonosyllables,oraftertwoconsonants, where\notherwise agroupofconsonants difficult ofpronunciation would\nbetheresult. The cases willbenoticed below, inexplaining\ninflected forms.\nAradical a-vowel isconverted intoyeven before iinper-\nfect tense-inflection :soninyima (nirit -|-ima).\n130.Asregardstheaccent here, asinthepreceding\ncase(128),theonly combinationrequiringnotice isthat ofan\nacute iorw-vowel with afollowing grave:theresult iscircum-\nflex\n;andsuch cases ofcircumflex aremanytimes more frequent\nthan anyand allothers. Examplesare :\nvyusti (vt-usti); ER^ffcT abhyarcati ;\nfrnadyau (nadi-au);\nsvuta(su-istaj; H*W tanvas(tanti-as).\nOfasimilar combination ofacute rwith following grave, only asingle\ncasehasbeen noted inaccented texts :namely, vijndtr etdt(B. xiv. 6.8n\n):\ntheaccentuation isinaccordance with therules for iand u.\n131.Ofadiphthong,thefinal iorw-element ischang-\nedtoitscorrespondingsemivowel. ^yorof0,before any\nvowel ordiphthong:thus,^e(reallyai:28)becomes %may,\nand EJTo(that is,au)becomes 5(Sfav;^aibecomes SHITay,\nandfaubecomes 35Rav.\n^Nochangeofaccent, ofcourse, occurs here; eachoriginal\nsyllable retains itssyllabic identity, andhence also itsown tone.\nExamples canbegiven only forinternalcombination, since inexternal\ncombination there arefurther changes: seethenext paragraph. Thus,^naya (ne-a); ^TFTnaya (nai-a);\nH^bhava(bko-a); >TRbhava(bhau-a).\n^132. Inexternalcombination, wehave theimportant\nadditional rule that thesemivowelresultingfrom thecon-\nversion ofthe final element ofadiphthongisingeneral\n"
+    },
+    {
+        "page": 64,
+        "content": "44III.EUPHONIC COMBINATION.[132\ndropped;and theresultinghiatus isleftwithout further\nchange.\n133. That istosay.afinal^e(byfarthemost fre-\nquent case) becomes simply3fabefore aninitial vowel\n(except %a:see135, below), andboth then remain un-\nchanged;andafinal^ai,inlikemanner, becomes(every-\nwhere) a.Thus,\nttaagatah (te-\\-agatah);\nnagaraiha(nagare -f-7^/\nadadat (tasmai -f-adadat);\nuktam(striyai-{~uktam).\nThe later grammariansallow thei/insuch combinations tobeeither\nretained ordropped;buttheuniform practiceoftheMSS.,ofevery age,\ninaccordance with thestrict requirementoftheVedic grammars (Pratigakhyas),\nistoomit thesemivowel andleave thehiatus.\nThe persistence ofthehiatus caused bythisomission isaplainindi-\ncation ofthecomparativelyrecent loss oftheintervening consonantal sound.\nInstances ofthecombination oftheremaining finaland initial arenotunknown,\nbutthey areofsporadicallyrare occurrence.\n/134. The diphthongo(exceptasphoneticalteration of\nfinal as:see175a)isanunusual final, appearing onlyinthe\nstemgo(356),inthevoc.sing,ofw-stems, inwords ofwhich\nthefinal aiscombined with theparticle u,asatho, andina\nfewinterjections. Inthe lasttwo classes itisuncombinable\n(below, 138);thevocatives sometimes retain thevandsometimes\nlose it(the practices ofdifferent texts aretoodifferent tobe\nbriefly stated); go(incomposition only)does notlose itsfinal\nelement, butremains gavorgo.Afinal asbecomes a,with\nfollowing hiatus, before anyvowel save a(forwhich, seethe\nnextparagraph).\nThe ofvofSTRavfrom aftauisusuallyretained :thus,\nfiic<ef taveva(tau-\\-eva) ;\n3H\"i(e(-*ii4l ubhav indragm (ubhau -\\-indragm).\nInsometexts, however,itislost before anw-vowel, theaalone\nremaining, withhiatus;inatleast onetext(Kathaka),itisdroppedbefore\nevery vowel. The latergrammarians allow ittobeeither retained ordropped.\n135. After final^eor^o.aninitial ?Jadisappears.\nTheresulting accent isasiftheawere notdropped,but\nrather absorbed into thepreceding diphthong, havingitstone\n"
+    },
+    {
+        "page": 65,
+        "content": "137JVOWEL COMBINATION. 45\nduly representedinthecombination.If,namely,the eorois\ngrave orcircumflex andtheaacute, theformer becomes acute;\nifthe eoroisacute andtheagrave, theformer becomes cir-\ncumflex, asusuallyinthefusion ofanacute and agraveele-\nment. Ifboth areacute orbothgrave, nochange,ofcourse,\nisseen intheresult. Examplesare :\nte'bruvan(teabruvan);\nso'bravit (sdh abravit);\nthihsitavyd 'gnih (hinsitavydh agnih);\nyaa>indro 'bravit (yddindrah dbramt);\nyddrajanyo'bramt (yddrajanydh dbravit).\nAstotheuse oftheavagraha sign inthecase ofsuchanelision,see\nabove, 16. Intransliteration, thereversedapostrophe,orrough breathing,\nwillbeused inthis \\vork torepresentit.\nThis elision orabsorptionofinitial aafter final eoro,which inthe\nlater languageistheinvariablerule,isintheVeda only anoccasional\noccurrence; and there isnoclose accordance with regard toitbetween the\nwritten andthespoken form oftheVedic texts. IntheAtharvan, forex-\nample, theaisomitted inwriting inabout onethird ofthecases, but is\ntobeomitted inreadinginlessthan one fifth(includinganumber inwhich\nthewritten text preserves it).SeeAPr. iii.54,note.\nTotherules ofvowel combination, asabove stated, there\narecertainexceptions. Some ofthemore isolated ofthese will\nbenoticed where theycome upintheprocessesofinflection\netc.;afewrequire mention here.\n136. Ininternal combination :\na.Theaugmentamakes with the initial vowel ofaroot\nthecombinationsat,au,ar(vrddhi-vowels),instead ofe,o,ar\n(ywraz-vowels),asrequired by127.\nb.The final oofastrengthened stem (238 b)becomes avbefore the\nsuffix ya(originally za);\nC.The final vowel ofastem isoften dropped when asecondary suffix\nisadded.\nFortheweakening and loss ofradical vowels, and forcertain insertions,\nseebelow, 249ff.,2578.\n137. Inexternal combination :\na.The final aoroofapreposition,with initial rofa\nroot,makes arinstead ofar.\nb.The final aofaprepositionbefore roots beginning with eorois\nusually omitted.\nC.Afinal aincomposition maybecut offbefore otuand ostka.\nd.The form uhfrom yvahsometimes makes theheavier(vrddhi)\n"
+    },
+    {
+        "page": 66,
+        "content": "4(jIII.EUPHONIC COMBINATION.\ndiphthongalcombination with aprecedinga;thus, prauha. praudhn, akmuhini\n(from pra-uha etc.).\n138. Certain final vowels, moreover, areuncombinable\n(pragrhya) ,ormaintain themselves unchangedbefore any\nfollowingvowel. Thus,\na.Thevowels/,uand easdual endings,both ofdeclen-\nsional and ofconjugationalforms. Thus,landliu asate imau;\ngiriarohatam.\nb.Thepronounami,(nom. pi.);andtheVedic pronom-\ninalforms asme, yitsme,tve.\nc.Afinal omade bycombination ofafinalo-vowel with theparticle u:\nthus, atho, mo, no.\nd.Afinal IoruofaVedic locative case.\ne.Aprotractedfinalvowel(78).\nf.The final, oronly, vowel ofaninterjection, asaho, he,a, i,u.\nPermitted Finals.\n139.Thesounds allowed tooccur asfinals inSanskrit\nwords, standing bythemselves(notineuphoniccombination\nwith something following),areclosely limited, and those\nwhich wouldetymologically come tooccupysuch aposition\nareoftenvariously altered,ingeneralaccordance with their\ntreatment inother circumstances, oraresometimes omitted\naltogether.\nThe variety ofconsonants thatwould evercome attheend ofeither\naninflected form oraderivative stem inthelanguageisverysmall :namely,\ninforms, onlyt(ordj,n,m,s;inderivativestems, only t,d,n,r,s(and,\ninafewrarewords, j).Butalmost allconsonants occur asfinals ofroots;\nandevery root isliable tobefound, alone oraslastmember ofacompound,\ninthecharacter ofadeclined stem.\n140. Allthevowel sounds, bothsimpleanddiphthongal,\nmaybesounded attheendofaword.\nButneitherfnor Iever actually occurs; and risvery rare(only as\nneuter sing, ofastem inroror,orasfinal ofsuch astem incomposition).\nThus, indra, civdya, dkari, nadk, d&tu, camu, janayitfj dyne,\ncivQyai, vnyo, agnau.\n141.Ofthenon-nasalmutes, onlythe first ineach\nseries, thenon-aspirate surd,isallowed; theothers surd\n"
+    },
+    {
+        "page": 67,
+        "content": "145 PERMITTED FINALS. 47\naspirate,andboth sonants whenever theywould etymo-\nlogicallyoccur, areconverted into this.\nThus, aynimdfforagnimdth, su/iftforsuhfd,vlrut forvlnid/t.\nInafew roots, when their final sonantaspirate)thus\nloses itsaspiration,theoriginalsonantaspirationofthe\ninitialreappears: compare cjk.below. 147.\nTims, daghbecomes dhak, budhbecomes Wiut, andsoon.\nThe roots exhibiting thischange arestated below, 155.\nThere wassome question among theHindu grammariansastowhether\nthe finalmute istobeestimated asofsurd orofsonantquality;butthe\ngreat weight ofauthority, and theinvariable practice oftheMSS., favor\nthesurd.\n142.Thepalatals, however, form here(asoften else-\nwhere) anexceptiontotherules fortheother mutes. No\npalatalisallowed asfinal. The 3Tcreverts toitsoriginal\nefik:thus, oTTfi vak, Jbjc^j^e^anhomuk. The $5\"ch(extremely*X *S O^s. L\nrare :perhaps onlyintheroot5f^prach) becomes T,t:thus,\n\"SnTTprat. The sf/either reverts toitsoriginal gutturalor\nbecomes,inaccordance with itstreatment inother combi-\nnations(219): thus, pTERI bhisdk,virat.Thesfi/^ doesnotoccur,\nbut isbythenative grammariansdeclared convertible to t.\n143.Ofthenasals, the *Tmand^nareextremely com-\nmon, especiallytheformer(ITmandHsareofallfinal\nconsonants themostfrequent);the HInisallowed, but is\nquiterare;3\"nisfound(remainingafter theloss ofafol-\nlowing ^1k)inaverysmall number ofwords; 31n\nnever occurs.\nBut thefinalmofaroot ischangedton(compare212.\nbelow):thus, dganfromgam,dnanfrom nam.\n144.Ofthesemivowels, theFTIalone isanadmitted\nfinal, and itisveryrare.The^ris(likeitsnearest surd\ncorrespondent, Hs:145)changedasfinal tovisarga. Of\nETyand ofvthere isnooccurrence.x17\n-v.\n145.Ofthesibilants, nonemaystand unaltered atthe\nendofaword. TheRs(which ofallfinal consonants\n-x \\\n"
+    },
+    {
+        "page": 68,
+        "content": "48III.EUPHONIC COMBINATION.\n[\nwould otherwise bethecommonest) islike^\"rchangedtoa\nbreathing,thevisarga.The$T^?either reverts toitsoriginal\n5R,or,insome roots,ischangedto^t(inaccordance\nwith itschangesininflection andderivation;seebelow, 218).\nThe *Tsislikewise changedtoJ.\nThechange of to tisofvery rare occurrence:seebelow, 226.\nFinal radical sissaidbythegrammarianstobechanged tot:thus,\ndhvat from dhvas :butnoexampleofthe conversion appears tooccur :\nsee168.\n146.Thecompound^{ksisprescribedtobetreated as\nsimple*T (notbecomingsfjkby150, below). Butthecase\nisarareone.and itsactual treatment intheolder language\nirregular.\nIntheonlyRV. caseswhere thekshasaquasi-radical character namely\nandk from anafcs, anddmyak fromymyakstheconversion istok.Also\nofforms ofthes-aorist(seethis aorist below) wehave adhak, asrafc, araife,\netc. for(adhdks-t etc.); but also aprat, dbhrat, ayat (forapraks-t etc.).\nAndAV. hasintwo cases srah(i.e.sras), apparentlyforasraks-s, from\nysrj (wronglyreferred byBR. toj/srcms).\nThenumeral a, 'six',isperhaps better toberegarded assafes, with\nitskstreated ass,according totheaccepted rule.\n147.Theaspiration ^hisnotallowed tomaintain\nitself, but(likej^/and $Fg]either reverts toitsoriginal\nguttural form, appearingas3Tk,orischangedtoZt\nboth inaccordance with itstreatment ininflection; seebe-\nlow, 222. And, also asininflection, theoriginalsonant\naspirationofafewroots(givenat155) reappears when their\nfinal thusbecomesunaspirated. Where the^hisfrom\noriginalV^dh(223),itbecomes rTt.\n148.Thevisarga andanusvara arenowhere etymolog-\nical finals;theformer isonlythesubstitute foranoriginal\nfinalH^5or^r;the latter occurs asfinalonlysofaras\nsome latergrammarians allow ittobesubstituted for *Tm-\n149.Apart from thevowels, then, theusual finals,\nnearlyintheorder oftheirfrequency,are :#,qm,^n,N\n"
+    },
+    {
+        "page": 69,
+        "content": "152]PERMITTED FINALS. 49\nrft,^k,^p,t;those ofonly sporadicoccurrence are\n3\"n,^l,tITn,-and,bypermitted substitution,-m.\n150. Ingeneral, onlyoneconsonant, ofwhatever kind,\nisallowed tostand attheendofaword;iftwo ormore\nwouldetymologicallyoccur there, thelast isdropped, and\nagainthelast,and soon,tillonlyoneremains.\nThus, tudants becomes tudant, andthistudan;udanc-s becomes\nudank, and thisudan; andachdntst(s-aor.,3dsing.,ofVchand]\nisinlikemanner reduced toachan.\nBut anon-nasal mute,ifradical andnotsuffixal,isre-\ntained after r:thus, urkfromurj,vdrkfrom)A>r/,dmart from\nymrj,suhurt from suhdrd. The case isnotacommon one.\nFor relics offormer doublefinals, preserved bythelater language under\nthedisguise ofapparent euphonic combinations, seebelow, 207 ff.\n151. Anomalous conversions ofafinalmute toone ofanother class are\noccasionally met with. Examples are:\na.Offinal ttofc,inafewwords that have assumed aspecial value\nasparticles: thus, jytik, tdjdk (beside tajdt], prthak, rdhak;butalsonow\nandthen inaverbal form,assavisak(AV.andVS. Kan.), avisyak (Parask.);\nand inthefeminines inkm(asasiknl besideasita). SV. hasonce\nprks'iforprtsu.\nb.Ofkorjtof,inanisolated exampleortwo,assamydt and dsrt\n(TS., K.).\nc.InTaittiriya texts, ofthefinal ofanustubh andtristubh toaguttural:\nas,anustuk ca,tristiigbhis (Weber, Ind. St.,xiii.109ff.).\nd.Ofalabial toadental :inkakud forandbeside kaktibh;insaihsfdbhis\n(TS.)fromysrp,andinadbMs, adbhyds, fromaporap(chap. V).Excepting\nthefirst,these look like cases ofdissimilation\n;yetexamplesofthecom-\nbination bbh arenotunknown inthe older language: thus, kakubbhyam,\ntristubbhis, anustub bhi.\n152. For alltheprocessesofexternal combination that\nistosay,incomposition and sentence-collocation astem-\nfinal orword-final isingeneraltoberegardedashaving,not\nitsetymological form, butthatgivenitbytherules astoper-\nmitted finals. Fromthis, however, are tobeexceptedthe s\nand r:thevarious transformations ofthese sounds have nothing\ntodowith thevisargatowhich asfinals before apause they\nhave come doubtless atacomparativelyrecent periodof\nphonetic historytobereduced. Words willeverywherein\nthiswork bewritten with final sorrinstead ofh;and the\nrules ofcombination willbestated asforthetwomore original\nsounds, andnotforthevisarga.\nWhitney, Grammar. 4\n"
+    },
+    {
+        "page": 70,
+        "content": "50III.EUPHONIC COMBINATION.[153\nDeaspiration.\n153.Anaspiratemute ischangedtoitscorresponding\nnon-aspiratebefore another non-nasal mute orbefore asib-\nilant;itstands imaltered onlybefore avowel orsemi-\nvowel ornasal.\nSuch acasecanonly arise ininternal combination, since theprocesses\nofexternal combination presupposethereduction oftheaspirate toanon-\naspiratesurd (152).\nPractically, also, therules astochangesofaspirates concern almost only\nthesonant aspirates,since thesurd, being oflater development and rarer\noccurrence,arehardly ever found insituations that call fortheirapplication.\n154. Hence,ifsuch amute istobedoubled,itis\ndoubled byprefixingitsowncorresponding non-aspirate.\nButintheMSS., bothVedic andlater, anaspirate mute isnotseldom\nfound written double especially,ifitbeone ofrare occurrence :for\nexample (BV.), akhkhali, jdjhjhatl.\n155. Inafewroots, when afinal sonantaspirate (f\ngh,Udh,*Tbh;also^A,asrepresentinganoriginal5Jgh)\nthus loses itsaspiration,the initial sonant consonant(JJg\nor^\"dorSfV)becomesaspirate.\nThat istosay, theoriginal initial aspirateofsuch roots isrestored,\nwhen itspresence does notinterfere with theeuphonic law,ofcomparatively\nrecentorigin, which(inSanskrit asinGreek) forbids aroot toboth begin\nandendwith anaspirate.\nThe roots which show thispeculiar changeare :\ningh dagh;\ninh(fororiginal gh} dah, dih,duh, druh, drhh, guh; andgrah (in\nthelater desiderativejighrksa)]\nindhbandh, badh, budh;\ninbh dabh(but only inthelater desiderativedhipsa,forwhich the\nolderlanguage hasdipsa).\nThesamechange appears when thelaw astofinals causes theloss of\ntheaspiration attheendoftheroot: seeabove, 141.\nButfromdah, duh,andguharefound intheVeda alsoforms without\ntherestored initial aspirate: thus, daksnt; aduksat, duduksaetc.; juguksa.\nThesameanalogyisfollowed bydadh, theabbreviated substitute ofthe\npresent-stem dadha, fromydha,insome oftheforms ofconjugation; thus,\ndhatthas fromdadh-+-thas, adhatta fromadadh+ta,etc.\n"
+    },
+    {
+        "page": 71,
+        "content": "160]ASSIMILATION. 51\nSurd andSonant Assimilation.\n156.Under thishead, there isespecially oneverymark-\nedandimportantdifference between theinternal combi-\nnations ofaroot orstem with suffixes andendings, and\ntheexternal combinations ofstemwith stem incomposition\nandofword withword insentence-making: namely,\n157. Ininternal combination,the initial vowel orse-\nmivowel ornasal ofanendingofderivation orinflection\nexercises .noalteringinfluence uponafinalconsonant ofthe\nroot orstem towhich itisadded.\nTothisrule there areafewexceptions only:thus, arevertedpalatal\nsometimes before anasal (216. 4,5) ;dbefore theparticiple-suffix na(161) 5\nafinal consonant before maya(161).\nInexternal combination, ontheother hand, aninitial\nsonant ofwhateverclass, even avowel orsemivowel or\nnasal, requirestheconversion ofafinal surd tosonant.\nIthasbeen pointed outabove (152) that intherules ofexternal com-\nbination onlyadmittedfinals, along with sandr,need betaken account of,\nallothers being regardedasreduced tothese before combining with initials.\n158. Final vowels, nasals, and ^T/arenowhere lia-\nbletochangeintheprocessesofsurdandsonant assimi-\nlation.\nTher,however, hasacorrespondingsurd ins,towhich it\nischangedinexternal combination under circumstances thatfavor\nasurd utterance.\n159.With theexceptionsabovestated, the collision\nofsurdandsonant sounds isavoided incombinations\nand, regularly andusually, byassimilatingthefinal tothe\nfollowinginitial.\nThus, ininternal combination :dtsi, dtti, atthds, attd(yad\n-\\-sietc.); qagdhi, cagdhvdm (}/cak -\\-dhi etc.);inexternal\ncombination, dbhudaydm,jyog jiva, sadaqitdyah,tristubdpi;\ndiggaja, sadahd, brhddbhdnu, abjd.\n160.If,however, thefinal sonantaspirateofaroot is\nfollowed by rTtor5Tthofanending,theassimilation isin\n4*\n"
+    },
+    {
+        "page": 72,
+        "content": "52III.EUPHONIC COMBINATION.[160\ntheother direction: thecombination ismade sonant, and\ntheaspirationofthefinal(lostaccordingto153, above)is\ntransferred tothe initial oftheending.\nThus, ghwith torthbecomes gdh;dhwith thesame be-\ncomes ddh,asbaddhd (ybadh -f-ta),runddhds(^rundh -f-thusor\ntas);bhwith thesame becomes bdh, aslaMhd(yiabh-\\-ta),\nlabdhvh, (Vldbh -\\-foa).\nMoreover, h,asrepresenting original gh,istreated inthe\nsamemanner :thus, dugdhd, dogdhum from duh andcompare\nrudhd andlldhd from ruhandUhetc., 222.\nInthiscombination,asthesonant aspirationisnot lostbuttransferred,\ntherestoration ofthe initial aspiration (155) does nottakeplace.\nIndadh fromydha (155, end),themore normal method isfollowed;the\ndhismade surd, andtheinitial aspirated:thus, dhatthas, dhattas. AndRV.\nhasdhaktam instead ofdagdham fromydagh.\n161. Before anasal inexternal combination, afinal\nmutemaybesimply made sonant, oritmaybestill fur-\nther assimilated, being changedtothenasal ofitsown class.\nThus,either Iddndmas ort&nndmas, vagmeorvhnme,\nbddmahan orbanmahan, tristub nundm ortristum nundm.\nInpractice,theconversion into anasal isalmost invariably made in\ntheMSB., as,indeed,itisbythePraticakhyas required andnotpermitted\nmerely. Even bythegeneral grammariansitisrequiredinthecompound\nsdnnavati, andbefore matra, and thesuffix maya (really anoun incom-\nposition): thus, vanmdya, mrnmdya.\nEven ininternal combination,final dofarootbecomes nbefore the\nparticiple-suffix na :thus, bhinnd, sannd, tunnd.\n162. BeforeI,afinal tisnotmerely made sonant, but\nfully assimilated, becomingI:thus, tdllabhate, uttuptam.\n163. Before^h(thecase occursonlyinexternal com-\nbination),afinalmute ismade sonant; andthen thehmay\neither remainunchangedorbeconverted into thesonant\naspirate correspondingwith theformer: thus, eitherrrf^\ntadhiorrrf%taddhi.\nInpractice, the latter method isalmost invariably followed; andthe\ngrammarians ofthePrati$akhya periodarenearly unanimous inrequiringit.\nThephonetic difference between thetwo isvery slight.\nExamplesare:vagghutdh, sdddhota(sat-\\-hota),tod-\ndhita(tat-\\-hita),anustub bhi.\n"
+    },
+    {
+        "page": 73,
+        "content": "169]FINAL sAND r. 53\nCombinations offinalsandr.\n164.Theeuphonic changesofHsand^\"rmaybest\nbeconsidered together, because ofthepracticalrelation of\nthetwo sounds,incomposition andsentence-collocation,\nascorrespondingsurdandsonant :inahost ofcasesHs\nbecomes^Trinsituationsrequiringorfavoringtheoccur-\nrence ofasonant;and,lessoften, ^~rbecomes Hswhere\nasurd isrequired.\nIninternal combination, thetwo arefarlessexchangeable\nwith oneanother :and this class ofcasesmay bestbetaken up\nfirst.\n165. Final rradical orquasi-radical (that is,notbelonging\ntoanendingofderivation)remains unchanged before bothsurd\nandsonant sounds, andeven before suindeclension :thus, pi-\nparsi, caturthd, cattirsu, pursu.\n166. Final radical sremains before asurd ingeneral, and\nusuallybefores,asincassi, cassva, aclssu(butthe last isalso\nwritten ackhsu:172):itishoweverdroppedindsi(forassi :\nyas-\\-si).Before asonant(that is, bh)indeclension,itis\ntreated asinexternal combination :thus, acirbhis. Before aso-\nnant(that is,dh)inconjugation,itisdropped:thus, cadhi,\nadhvdm(butcases like the latter may bebyabbreviation[232]\nforaddhvam):inedhi(as-\\-dhi),thepreceding vowel isanoma-\nlouslyaltered.\n167. Inaveryfewcases,final radical sbefore sischangedtot(per-\nhapsbydissimilation):they are,fromyvas,thefuture vatsydmi and aorist\ndvatsam; fromyghas,thedesiderative stem jighatsa.\n168. Accordingtothegrammarians, the final sofcertain otherroots,\nused asnoun-stems, becomes tattheend oftheword, and before bhand\nsu:thus, dhvat, dhvadbhis, sradbhyas, sratsu. Butgenuine examplesofsuch\nchange donotappeartohave beenmetwith inuse.\nSporadiccases ofalike conversion arefound intheVeda :namely,\nmadbhts andmadbhyds from ma's;usddbhis from usds;svdtavadbhyas from\nsvdtavas;svdvadbhis etc. (notfound inuse), from svdvas. Butthereality\noftheconversion here isopentograve doubt;itrather seems thesubsti-\ntution ofaf-stem foras-stem. Thesame istrue ofthechangeofvans\ntovatinthedeclension ofperfect participles (chap. V.).\nInthecompounds ducchuna (dus-funa) andpdrucchepa (parus-cepa),the\nfinal softhe firstmember istreated asifat(203).\nFor tasapparent endingofthe3dsing,ins-verbs,seechap. VIII.\n169.Asthefinal consonant ofderivative stems andofin-\nflected forms, both ofdeclension andofconjugation,sisextreme-\n"
+    },
+    {
+        "page": 74,
+        "content": "54III.EUPHONIC COMBINATION. [169\nlyfrequent;and itschangesform asubjectoffirst-rate im-\nportanceinSanskrit euphony.Ther,ontheother hand,is\nquite rare, beingfound onlyincertain forms ofr-stems and\ninafewparticles.\nTheeuphonictreatment ofsand ryielding preciselythesame result\nafter allvowels except aand a,there arecertain forms with regard towhich\nitisuncertain whether they end insorr,and opinionsdifferrespecting\nthem. Such areus(orur}ofthegen.-abl. sing,ofr-stems, andus(orur}\nofthe3dplur.ofverbs.\n170. a.TheHs,asalreadynoticed(145), becomes\nvisargabefore apause.\nb.Itisretained unchanged only-when followed by\nrTtor%thjthesurdmutes ofitsown class.\nc.Before thepalatalandlingualsurdmutes ^cand\n^ch,Itand ~&th itisassimilated, becomingthesibilant\nofeither classrespectively, namelySTQorEfs.\nd.Before thegutturaland labial surdmutes 3\\k\nand IrM,Qpand Cfiphitisalsotheoreticallyassimil-\nated, becoming respectivelythejihvamuliyaandupadhma-\nniya spirants (60);but inpracticethesebreathingsare\nunknown, andtheconversion istovisarga.\nExamplesare: tob.tataste,caksus te;toc.tataq ca,ta-\nsyag chaya;tod.nalah kamam, purusahkhanati;yacah prapa,\nvrksahphalavan.\n171.The first three ofthese rules arealmost universal;to\nthe lastthere arenumerousexceptions,thesibilant beingre-\ntained(or,by180, converted intos),especiallyincompounds ;\nbut also, intheVeda, even insentence combination.\nIntheVeda, theretention ofthe sibilant incompoundsisthegeneral\nrule, theexceptionstowhich aredetailed intheVedic grammars.\nInthelaterlanguage,theretention ismainly determined bytheintimacy\northeantiquity andfrequency ofthecombination. Thus,the final sibilant\nofaprepositionorawordfilling the office ofaprepositionbefore averbal\nroot iswont tobepreserved; andthat ofastem before aderivative ofykr,\nbeforepati, before kalpa andkama, and soon.Examplesarenamaskara,\nvacaspati, ayuskama, payaskalpa.\nTheVedic retention ofthesibilant insentence-collocation isdetailed in\nfull inthePraticakhyas. The chief classes ofcases are: a.the final ofa\npreposition oritslikebefore averbal form;b.ofagenitive before agovern-\ningnoun: asdivas putrdh, iddspade ;c.ofanablative before part: as\n"
+    },
+    {
+        "page": 75,
+        "content": "175]COMBINATIONS OFFINAL s. 55\nhimdvatas part; d.ofother less classifiable cases: asdyaus pitd,trisputvd,\nydspdtih, paridhfs pdtati,etc.\n172. Before aninitial sibilant STc,^s,HsHsis\n^JT> ^.7^ ^\neither assimilated, becomingthesame sibilant, oritis\nchangedintovisarga.\nThenative grammarians areinsomemeasure atvariance(seeAPr. ii.40,\nnote)astowhich ofthese changes should bemade, and inpartthey allow\neither atpleasure. Theusage oftheMSS. isalsodiscordant;theconversion\ntovisargaistheprevalent practice, though the sibilant isalso notinfre-\nquentlyfound written. Europeaneditors generally write visarga ;butthe\nlater dictionaries and glossaries make thealphabetic placeofaword the\nsame asifthesibilant were read instead.\nExamplesare :manuh svayamormanussvayam ;indrah curah\norindracqurah.\n173. There areoneortwoexceptionstothese rules:\na.Ifthe initial sibilant has asurdmute afterit,the final smaybe\ndropped altogether andbysome authorities isrequiredtobesodropped.\nThus, vayavastha orv&yavah stha,' catustanam orcatuhstanam.\nWith regardtothispoint theusage ofthe different MSS. and editions\nisgreatlyatvariance.\nb.Before(s,the sisallowed tobecome visarga, instead ofbeing\nretained.\n174. Before asonant, either vowel orconsonant(ex-\ncept^r:see179),Hsischangedtothesonant^r\nunless, indeed,itbepreceded by[aor TTa.\nExamplesare :devapatir iva,crir iva;manurgacchati, tanur\napsu ;fayoradrstakamah;sarvair gunaih agnermanve.\nForafewcases likeduda$a, duna$a, seebelow, 199.\nTheendings 3^qasand^THas(bothofwhich areex-\ntremely common)follow rules oftheir own, asfollows :\n175. a.Final 5R7as,before anysonant consonant and\nbefore short^a,ischangedtomo andthe 5faafter\nitislost.\nThe resulting accentuation, andthe fact thattheloss ofaisonly oc-\ncasional intheolder languageoftheVeda, havebeen pointed outabove, 135.\nExamplesare :nalonama, brahmanyo vedavit; hantavyo'smi.\nb.Final 3^fasbefore anyother vowel than^aloses\nitsHs,becoming simple ^a;and thehiatus thus occa-\nsioned remains.\n"
+    },
+    {
+        "page": 76,
+        "content": "5(>III.EUPHONIC COMBINATION.[175\nThat istosay, asistreated asanoriginal o,orane,would betreated\ninthesame situation: see132 4.\nExamplesare :vrhadacva uvaca, aditya iva, ndmaiikti.\n176. Exceptionstotherules astofinal asare:\na.Thepronounssasandeshas(also syasintheVeda)lose\ntheir sbefore anyconsonant :thus, sadadarca, eshapurusah ;but\nsada tusah, so'bravit. Theexclamation Ihos loses itssbefore\nallvowels and allsonant consonants.\nb.IntheVeda, andmore rarely inthelater language, therule forthe\nmaintenance ofthehiatus issometimes violated, and theremaining con-\ntiguous vowels arecombined intoone :forexample,se'dague,se'morn,\nsawJsadhih (forsaidague, saimam, saosadhih).\nc.Afewinstances arefound (almostallVedic) ofsapparently changed\ntorafter a,asafter other vowels :butinnearly every case there istobe\nassumed, rather, astem inarbeside that inas,evidences oftheformer\nbeing sometimes found inthekindred languages:thus, informs ofudhas\nanddhas (seechap. V.);inamnas (nooccurrence);inbhiivas (second ofthe\ntrio bhus, bhuvas, svar), exceptinitsoldest occurrences; inavds (once, in\nRV.);inusds(voc., andinusarbudh};invddhar andvadhary (RV.);in\nvanargu, dnarvi$, vasarhdn, sabardugha, andone ortwoother more doubtful\nwords; and inaseries ofwords inasingle passage ofTS.andK., viz.\njinvdr, ugrdr, bhimdr, tvesdr, frutdr, bhutdr, and(K.only) putdr.\nInaharpdti (VS.), andvanarsdd andvanarsdd(RV.), weseethesame\nchange even before asurd consonant.\nd.Final asisoncechanged tooinRV. before asurdconsonant: thus,\nadopito.\n177. Final 5TTCTasbeforeanysonant, whether vowel or\nconsonant, loses its *Ts,becoming simple5TTa;andthe\nhiatus thus occasioned remains.\nThemaintenance ofthehiatus inthese cases,asinthat ofoand e\nand ai(above, 1334), seems toindicate arecent loss oftheintermediate\nsound. Opinions aredivided astowhat thisshould have been. Some of\nthenative grammarians assimilate thecase ofastothat ofai,assuming the\nconversion toayinboth alike butprobably only asamatter offormal\nconvenience inrule-making.\n178. Final^\"r,ingeneral, shows thesameformwhich\nHswould show under thesame conditions. But\n-\\\na.Originalfinalr,after aora,maintains itself unchang-\nedbefore asonant :thus, punar eti,pratar-jitjdhardkmna, dkar\njyotih.\nb.Also before asurd,rispreservedinafewVedic compounds:thus,\nsuarcanas, svarcaksas, yvarpati, svarsd, svbrsati; dhursdd, dhurs&h; purpati.\n"
+    },
+    {
+        "page": 77,
+        "content": "181]CONVERSION OFsTOs. 57\nvdrkaryd, dfirpada, punartta. Insome ofthese, therisoptionally retained\neven inthelater language.\nc.Ontheother hand,rislost, likes,inone ortwoVedic cases:\naksd fnduh, ahaevd.\n179.Adouble risnowhere admitted: ifsuchwould occur,\neither byretention ofanoriginalrorbyconversion of*tor,\none risomitted, and thepreceding vowel,ifshort,ismade\nlongbycompensation.\nInsome Vedictexts, however (Yajur-Veda), arbecomes obefore ini-\ntialr;thus, svbrohdva.\nConversion of3U to\nv1180.The dental sibilant Hsischangedtothelingual\nET5,ifimmediately preceded byanyvowel save Efaand\nma,orbyfikor^\"r unless theHsbefinal, orfol-\nlowed by^~r.\nThe assimilating influence ofthepreceding lingual vowels andsemi-\nvowel isobvious enough; that of fcandtheother vowels appears tobedue\ntoasomewhat retracted position ofthetongueinthemouth during their\nutterance, causingitstiptoreach theroof ofthemouth more easily ata\npointfurther back than thedental one.\nThegeneral Hindu grammar prescribesthesame change after aIalso;\nbutthePratic.akhyas give nosuch rule, andphonetic considerations, the I\nbeingadental sound, aredecidedly against it.Actual cases ofthecom-\nbination,ifthey occur atall,areexcessively rare.\nThevowels that cause thealteration ofstosmaybecalled\nforbrevity's sake \"alterant\" vowels.\nAsaconsequence ofthis rule,\n181. Inthe interior ofaSanskrit word, thedental sis\nnotusually found after anyvowel save aanda,but, instead\nofit,thelinguals.But\na.Afollowingrpreventstheconversion :thus, usra, tisras,\ntamisra. And itisbutseldom made intheforms and derivatives\nofaroot containing anr-element(whetherrorr),whatever\nthepositionofthatelement :thus, sisarti, sisrtam, sarisrpd, tistire,\nparisrut. To this rule there areafewexceptions,asvisfard,\nnistrta, vispardhas, gdvisthira,etc. Inajusranthe final sofa\nroot ispreserved even immediatelybefore r.\nThis dissimilating influence ofafollowing r,ascomparedwith thein-\nvariable assimilating influence ofapreceding r,ispeculiarandproblematical.\nb.Therecurrence ofsinsuccessive syllablesissometimes avoided by\n"
+    },
+    {
+        "page": 78,
+        "content": "58III.EUPHONIC COMBINATION.[181\nleadingtheformer sunchanged: thus, sisaksi,but sisaktiydsislsthds, but\nydsislmahi. Similarly,incertain desiderative formations: seebel'-w, 184c.\nC.Other cases aresporadic:RV. hastheforms sisice and sisicus(but\nsisicatus),andthestems rbisa, kistd, bfsa, busd, bfsaya; asingle rootpis,\nwith itsderivative pesuka,isfound once in(JR.;forpuns andthe roots-\nninsand hins, seebelow, 183.\n182.Ontheother hand(aswaspointedoutabove, 62),\ntheoccurrence ofsinSanskrit words isnearlylimited tocases\nfallingunder this rule :others arerather sporadic anomalies\nexceptwhere sistheproductofcorksbefore adental, asin\ndrastum, caste, tvastar :see218, 221). Thus, wefind:\na.Four roots, kas, Zas, bhas, bha--, ofwhich the last iscommon and\nisfound asearly astheBrahmanas.\nb.Further, inRV.,dsa, kavdsa, casala, casa, jdldsa, pdsyh, baskdya,\nvdsat(forvaksat?), kdsthd(forkaksta, Fick); and, byanomalous alteration\noforiginal s,-sdh(turdsdh, etc.), dsadha, upasttit, andprobably apdsthd and\nasthivdnt. Such cases growmorecommon later.\nThenumeralsas,asalready noted,ismore probablysafes.\n183. The nasalization ofthealterant vowelor,inother\nwords,itsbeing followed byanusvara does notpreventits\nalteringeffect uponthe sibilant :thus, havihsi, paruhsi. And\nthealteration takesplaceinthe initial sofanendingafter the\nfinal sofabase, whether the latter beregardedasalsochanged\ntosorasconverted intovisarga:thus,havissu orhavihsu, pa-\nrussu orparuhsu.\nButthesofpuns (chap. V.)remains unchanged, apparently onaccount of\ntheretained sense ofitsvalue aspums ,also that ofyhihs,because ofits\nvalue ashins (hinasti etc.); ynins (RV. only)ismore questionable (perhaps\nnims, from nam).\n184. Theprincipalcases ofalteration ofsininternal com-\nbination are :\na.Inendings, inflectional orderivative, beginning with s\nsu; si,se,sva;sofsibilant-aorist, future, anddesiderative\n;\nsuffixes sna, snu, sya,etc. after afinal alterant vowel or\nconsonant ofroot orstem, oraunion-vowel :thus, juhosi, $ese,\nanaisam, bhavisyami, cucruse, desna, jisnu, viksu, akarsam.\nb.The final sofastem before anendingorsuffix :thus,\nhavisa, Jwvisas, etc., from havis;caksusmant, coctska, manusa,\nmanusya.\nRoots having afinal sibilant(except p)after analterant vowel are\nwith theexception offictitious onesandpis, nins, hins regarded asend-\ningins,nots;andconcerning thetreatment ofthis sincombination,see\nbelow, 225 ff.\n"
+    },
+    {
+        "page": 79,
+        "content": "188]CONVERSION OFsTOs. 59\nc.The initial sofaroot after areduplication:thus,si-\nsyade, susvapa, sisasati, coskuyate,sanisvanat.\nExceptedisingeneral aninitial radical sinadesiderative stem, when\nthedesiderative-sign \"becomes s:thus,sisanisati fromysan,sisanksati from\nysanj.\n185. Butthesame changeoccurs also, onaconsiderable\nscale,inexternal combination, especiallyincomposition:thus,\nBoth inverbal forms and inderivatives, the final ior\nMofaprepositionorother likeprefix ordinarily lingualizesthe\ninitial softheroot towhich itisprefixed ;since suchcombi-\nnations areboth ofgreat frequency and ofpeculiar intimacy,\nanalogouswith those ofroot orstem and affix :thus, abhisttc,\npratisthh, ntsikta, visita; anusvadhdm, suseka.\nThe principal exceptionsareinaccordance with theprinciples already\nlaiddown :namely, when theroot contains anr-element, andwhen arecur-\nrence ofthesibilant would takeplace. Butthere arealsoothers, ofamore\nirregular character; and tliecomplete account ofthetreatment ofinitial\nradical safter aprefix would beamatter ofgreat detail, andnotworth\ngivinghere.\nInafew cases, the initial,usually altered after acertainprefix,\nretains thealtered sibilant even after aninterposed aofaugment orreduplic-\nation :thus, abhy astham, pary asasvajat, vyasahanta, nyasadama, abhy\nasincan, vyastabhnat; vitasthe,vitasthire.\nMuch more anomalous istheoccasional alteration ofinitial radical s\nafter ana-element ofaprefix. Such cases areavastambh(againstnistambh\nandprati stambh)arid(accordingtothegrammarians)avasvan.\n186. Inother compounds,the final alterant vowel ofthe\nfirstmember notinfrequently (especiallyintheVeda) lingualizes\nthe initial softhesecond :forexample, yudhisthira, pitrsvasr,\ngosthd, agnistomd, annstubh, trisandhi, divisdd, paramesthin, abhi-\nsend, pitrsdd, purustutd.\nAveryfewcases occur ofthesame alteration after ana-element:thus,\nsavyasthd, apasthd, upastut; also}/sa/i,when itsfinal, by146, becomes t:\nthus, satrdsdt(butsatrasdham).\n187. The final softhe firstmember ofacompoundoften\nbecomes safter analterant vowel :thus, the sofaprepositional\nprefix,asnissidhvan, dustdraffordusstdra), aviskrta; and, regu-\nlarly,asretained instead ofbeing converted tovisargabefore\nalabial orguttural mute(171) ,ashavispa, jyotiskft ;tapuspa,\n188. Once more, intheVeda, thesamealteration, both ofaninitial\nand ofafinals,isnotinfrequent even between thewords composinga\nsentence. The cases aredetailed inthePrati^akhya belonging toeachtext,\nandareofvery various character. Thus,\na.The initials,especially ofparticles:asusu,Mma,kdmusvft;\n"
+    },
+    {
+        "page": 80,
+        "content": "$0III.EUPHONIC COMBINATION.[188\nalso ofpronouns:ashisdh; ofverb-forms, especially fromyas:as\nhfsthd,divf stha; andinother scattering cases: asustuhi, nusthirdm,\ntrisadhdsthd.\nb.Afinal s,oftenest before pronouns (especially toneless ones): as\n(ignis tva, niste,lyus te,cucistvdm, sddhis tdva; butalso inothercases,\nandwherever afinal sispreserved,instead ofbeing turned intovisarga,\nbefore agutturalorlabial (171):asirisputvd, dyus krnotu, vdstospdtih,\ndyaus pits.\nConversion ofnton.\n189.The dental nasal^n,whenimmediately followed\nbyavowel orby^norJTmorTfyorcfv,isturned in-\ntothelingualHInifprecededinthesame wordbythe\nlingualsibilant orsemivowel orvowels that istosay,\nby*Ts,~$r.OYftrortyf:andthis, notonlyifthe\nalteringletter stands immediatelybefore the nasal, but at\nwhatever distance from thelatter itmaybefound:unless,\nindeed, there intervene(aconsonant movingthefront of\nthetongue:namely)apalatal (exceptTy) ,alingual,ora\ndental.\nWemay thus figure toourselves therationale oftheprocess:inthe\nmarkedproclivityofthelanguage toward lingual utterance, especially ofthe\nnasal, thetipofthetongue, when once reverted into theloose lingual position\nbytheutterance ofanon-contact lingual element, tends tohang there and\nmake itsnext nasal contact inthat position: anddoesso,unless thepro-\nclivity issatisfied bytheutterance ofalingual mute, ortheorganisthrown\noutofadjustment bytheutterance ofanelement which causes ittoassume\nadifferent posture. This isnotthecasewith thegutturals orlabials, which\ndonotmove thefrontpart ofthetongue (and,astheinfluence ofkon\nfollowingsshows,theguttural position favors thesuccession ofalingual):\nandtheyistooweakly palatal tointerfere with thealteration (asitsnext\nrelative,thei-vowel, itselflingualises as).\nThis isaruleofconstantapplication;and(aswaspointed\noutabove)thegreat majorityofoccurrences ofninthelanguage\naretheresult ofit.\n190.The rulehasforceespecially\na.Whensuffixes,ofinflection orderivation,areadded toroots or\nstemscontaining oneofthealtering sounds :thus, rudrena, rudrdndm, vdrine,\nv&rinl, vdrini, datfni, hdrani, dveshani, krindmi, prntiti, ksubhand, ghrnd,\nkdrna, vrknd, rugnd, drdvina, isdni, purand, reknas, cdksana, cfkirsamana,\nkrpamana.\n"
+    },
+    {
+        "page": 81,
+        "content": "194]CONVERSION OFnTOn, 61\nb.When the finalnofaroot orstem comes tobefollowed, ininflec-\ntion orderivation, bysuchsounds asallow ittofeeltheeffect ofapreceding\naltering cause: thus, from}/ran, rdnanti, rdnyati, rdrana, ardnisus; from\nbrahman, brdhmand, brdhmani, brdhmand, brahmanya, brdhmanvant.\n191. This rule(likethat forthechangeofsto siapplies\nstrictly and especially when thenasal andthecause ofitsalter-\nation both liewithin the limits ofthesameintegral word;\nbut(alsolike theother)itisextended, within certainlimits,\ntocompoundwords and even, intheVeda, tocontiguous\nwords inthesentence.\n192.Especially,aprepositionorsimilarprefixtoaroot,\nifitcontain rorend ineuphonicrfor s(174), very often\nlingualizesthenofaroot orofitsderived stems and forms.\nThus:\na.The initial nofaroot isusually and regularly soaltered, inall\nforms and derivatives, after para, pari, pra, nir(for nis), antar, dur(for\ndus):thus,para naya,panmyate, prdnudasva;pardnutti, parindma, pranava,\nnirnfj, durndfa.\nRoots suffering thischange arewritten with initial ninthenative root-\nlists. The only exceptionsofimportance arenrt, nabh, nand(very rare),\nandnaywhen itspbecomes *(asinprdnasta],\nb.The finalnofaroot islingualized insome oftheforms ofanarid\nhan: thus, prd 'nitz, prdnd, prdhanyate, prahdnana.\nc.The class-signs nuandnaarealtered after the roots hiandmi:\nthus, pari hinomi, prdminanti(butthelatter notintheVeda).\nd.The 1stsing. impv. ending ani issometimes altered :thus, prd\nbhavdni.\ne.Derivatives bysuffixes containing nsometimes havenbyinfluence\nofapreposition: thus, prayana.\nf.Thenoftheprepositionniissometimes altered,likethe initial of\naroot,after another preposition:thus, pranipdta, pranidhi.\n193. Incompound words, analtering cause inonemember sometimes\nlingualizesanotthenext following member either itsinitial orfinaln,\norninitsinflectional orderivative ending. The exercise ofthealtering\ninfluence canbeseen todependinpartupon thecloseness orfrequencyof\nthecompound, oritsintegration bybeing made thebase ofaderivative.\nExamplesare :grdmani, trindman, urunasd\n,-vrtrahdnam etc. (butvrtraghnd\ntc. :195), nrmdnas, drughand; pravdhana, nrpdna, purydna, pitrydna;\nsvargena, durgdni, usrdydmne, tryangdnam.\n194. Finally, intheVeda,an(usually initial)islingualized evenby\nanaltering sound inanother word. Thetoneless pronounsnasandena- are\noftenest thus affected :thus, pdri nas. prai 'nan, {ndra enam;butalsothe\nparticle nd,'like': thus, vdrnd;andafewother cases, asvdrndma, punar\nnayamasi, agner dvena.\n"
+    },
+    {
+        "page": 82,
+        "content": "62HI.EUPHONIC COMBINATION.[195\n195.Theimmediate combination ofanwith apreceding guttural or\nlabial seems insome cases tohinder theconversion ton:thus, vrtraghnS\netc.ksubhnati, trpnoti (butinVeda trpnu).\nConversion ofdental mutes tolinguals and palatals.\n196.When adental mute comes incontact with, a\nlingualorpalatalmute orsibilant, thedental isusually\nassimilated, becoming lingualorpalatal respectively.\nThe cases arethefollowing:\n197.Adental surdmute ornasal, orthedentalsibilant,\nwhen immediately preceded byas,iseverywhere converted into\nthecorresponding lingual.\nUnder thisrule,thecombinations at,sth,andsnareverycommon;aa\nisrarely sowritten, thevisarga being putinstead oftheformer sibilant:\nthus, jydtihsuinstead ofjy6tissu.\nThose cases inwhich final abecomes tbefore su(226 b)donot, of\ncourse,fallunder this rule.\n198. Intheother(comparatively infrequent)cases where\nadental ispreceded byalingualininternal combination, the\ndental(exceptofsuloc.pi.)becomeslingual. Thus :\na.Adhafter dfrom aofaroot orstem (226 b):asddviddhvam etc.\nb.Only averyfew other instances occur: itteand ditto, from}/id;\nsaddhd(also saddhd andsodha], andsanndm (aaa -(-nam :anomalousgen.\npi.ofaaa) ;trnna(ytrd-\\-na). Asmall number ofwords follow thesame\nrule inexternal combination: seebelow, 199.\nBut tadhi (Vedic: ytad -\\-dhi] shows loss ofthe final lingualafter assi-\nmilation ofthedental, andcompensatory lengthening.\nSome ofthecases ofabnormal occurrence ofdareexplained inasimi-\nlarway, asresults ofalingualized andafterward omitted sibilant before d :\nthus nlda fromnisda, ypldfrompisd, j/mrdfrommrsd (Zend marezhdd).\nForwordsexhibiting alikechange incomposition, seebelow, 199b.\n199. Inexternalcombination,\na.Afinal tisdirected tobeassimilated toaninitial lingual mute :\nthus, tat-tlka, taddayate, tat-thalirii, taddhaukate :but thecase never\noccurs intheolder language, andvery rarely inthe later. For finalnbe-\nfore alingual,see205.\nb.An initial dental after afinallingual usually remains\nunchanged ;and suoftheloc.pi.follows thesame rule :thus,\nsatsuyratsu.\nExceptions are: afewcompounds withaaa,'six': namely, sdnnavati,\nsannabhi (and one ortwoothers notquotable from theliterature).\n"
+    },
+    {
+        "page": 83,
+        "content": "204]DENTAL MUTES TOLINGUALS ANDPALATALS. 63\nInafewcompounds, moreover, there appearsalingualized dental, with\ncompensatory lengthening,after alostlingual sibilant oritsrepresentative:\nnamely,incertain Vedic compounds with dus :duddbha, duddf, dudhi, du~\nndfa, dundfa; and, inthelanguage ofevery period, certain compounds of\nsas, with change ofitsvowel toanalterant quality (asinvodhum and\nsodhum: 224b:sddafa, sodhd(alsosaddhd andsaddhd], sodant.\nC.Between final tand initials,theinsertion ofatispermitted\nor,accordingtosome authorities, required: thus, sdtsahdsrah orsaltsahdsrah.\n200. The cases ofassimilation ofadental toacontiguous\npalataloccur almost onlyinexternal combination, and before\naninitialpalatal. There isbutonecase ofinternal combina-\ntion, namely:\n201. A.^ncomingtofollow apalatal mute ininter-\nnalcombination isitselfmadepalatal:\nThus, yacnh, (theonlyinstance afterc),yajna, jajne, ajnata,\nrkjna, rhjni.\n202.Anfinal rTtbefore aninitialpalatal mute isas-\nsimilated toit,becoming ^cbefore rfcor^ch,and sT/\nbefore sf/(^f\\jhdoes notoccur).sr^yAfinal ^Tnisassimilated before 5fj,becoming 3Tn.\nAllthegrammarians,ofevery period, require this assimilation ofn;\nbut itismore often neglected, oronly sporadically made, intheMSS.\nFornbefore asurd palatal, seebelow, 208.\n203. Before thepalatalsibilant 5Tp,both rTtand^n\nareassimilated, becoming respectivelyrjcand oTn;and\nthen thefollowingST$may be,and inpractice almost\nalways is,converted to^ch.\nSome authorities regard theconversion ofptochaseverywhere oblig-\natory, others asonly optional ;someexcept, peremptorilyoroptionally, a\nffollowed byamute. Andsome require thesame conversion after every\nmute savem,reading alsovtpat chutudrl, dnat chuci, anustup charadi, puk\nchuci. TheMSS. generally writec/,instead ofccA, asresult ofthecom-\nbination oftand p.\nCombinations offinal n.\n204. Final radical nisassimilated ininternal combination\nloafollowing sibilant, becominganusvara.\nThus, vdnsi, vdhsva, vdhsat, mahsydte, tfghahsati.\nAccording tothegrammarians,itistreated before bhandsuindeclen-\n"
+    },
+    {
+        "page": 84,
+        "content": "4 HI.EUPHONIC COMBINATION.[204\nsion asinexternal combination. Butthecases areextremely rare, andRV.\nhasrdnsu andvdhsu (theonlyVedic examples).\nFinal nofaderivative suffix isregularly andusually dropped before a\nconsonant ininflection andcompositionincomposition, even before a\nvowel; and aradical noccasionally follows thesame rule.\nForassimilation ofntoapreceding palatal,see201.\nTheremainingcases arethose ofexternal combination.\n205. The assimilation ofninexternal combination toa\nfollowing sonant palatal and thepalatalsibilant chave been\nalreadytreated(2O2, 2O3).\nThenisalso declared tobeassimilated (becoming n)be-\nfore asonantlingual (d,dh, n),but the casehardly ever\noccurs.\n206.Anisalso assimilated toafollowinginitialI,be-\ncoming (likem :213c)anasal I.\nTheMSS. ingeneral attempttowrite thecombination inaccordance\nwith this rule.\n207. Before thelingual anddental sibilants,sands,final\nnremains unchanged;butatmayalsobeinserted between\nthenasal andthesibilant :thus, thnsdtorthnt sdt;mahhn sdn\normahhnt sdn.\nAccording tomost ofthegrammarians ofthePraticakhyas (not RPr.),\ntheinsertion ofthe tinsuch cases isanecessary one. IntheMSS. itis\nvery frequently made, butnotuniformly.Itisprobablyapurely phonetic\nphenomenon,atransition-sound toease thedouble change ofsonant tosurd\nand nasal tonon-nasal utterance althoughthenotinfrequent cases in\nwhich finalnstands fororiginal nt(asbharan, abharan, agnimari) mayhave\naided toestablish itasarule. Itsanalogy with theconversion ofnpinto\nnch(203)ispalpable.\n208. Before thesurdpalatal, lingual, and dental mutes,\nthere isinserted after finalnasibilant ofeach ofthose class-\nesrespectively, before which thenbecomes anusvara :thus,\nhqc;ncch;nst;nsth;hst;nsth.\nThisrule, which inthe classical language hasestablished itself inthe\nform here given, asaphonetic rule ofunvarying application, really involves\nahistoric survival. The large majorityofcases offinalninthelanguage\n(notfarfrom threequarters: seeAPr. ii.26,note)arefororiginal ns;and\ntheretention ofthe sibilant insuch, cases, when once itshistorical ground\nhadbeenforgotten, wasextended byanalogy toallothers.\nPractically, theruleapplies only tonbefore candt,since cases in-\nvolving theother initials occur either notatall,oronly withextreme rarity\n(theVeda does notpresent anexampleofany ofthem).IntheVeda, the\ninsertion isnotalways made, andthedifferent texts have with regard toit\ndifferentusages, which arefully explainedintheir Praticakhyas;ingeneral,\n"
+    },
+    {
+        "page": 85,
+        "content": "213]COMBINATIONS OFFINAL nANDm. 65\nitislessfrequentintheolder texts. When the pdoes notappear between\nnandc,thenisassimilated, becoming n(asbefore j:202).\n209. Thesame retention oforiginalfinal safter anasal,\nandconsequenttreatment of(apparent)final an, in,un,fnas\niftheywere ans, Ins, uns, fns(longnasalized vowel with finals},\nshows itself also inother Vedic forms ofcombination, which,\nforthesake ofunity, maybebrieflystated heretogether:\na.Final anbecomes an(nasalized a)before afollowing vowel :that is\ntosay, ans, with nasal vowel, istreated likeas,with pure vowel(177):\nthus, devdn e'hd,upabaddhdn ihd,mahdn asi. This isanextremely com-\nmon case, especially inRV. Once ortwice, the sappears ashbefore p:\nthus, svdtavdnh pdyuh.\nb.Inlikemanner,sistreated after nasalz,u,fasitwould beafter\nthose vowels when pure, becomingrbefore asonant sound(174), and(much\nmorerarely)hbefore asurd(170j:thus, rafminr iva,sununr yuvanyuhr ut,\nnfnr abhi; nfnh patram.\n210. The nasalsn,n,n,occurringasfinals after ashort\nvowel, aredoubled before anyinitial vowel :\"\nthus, pratydnn ud\nesi,udydnn aditydh.\nThis isalso toberegardedasahistoricalsurvival,thesecond nasal\nbeing anassimilation ofanoriginal consonant following the first. Itis\nalways written intheMSS., although theVedic metre seems toshow* that\ntheduplication wassometimes omitted.\n211. The nasal nandnbefore asibilant areallowed to\ninsertrespectivelykand t asn(207)inserts t:thus, praty-\ndnksomah.\nCombinations offinalm.\n212. Final radical JTm,ininternal combination,isas-\nsimilated toafollowing mute orspirantinthelatter case,\nbecoming anusvara ;intheformer, becomingthenasal of\nthesame class with themute.\nBeforemorv(aswhen final: 143),itischangedton;thus, from\n|/gram, dganma, aganmahi, ganvahi, jaganvans (which appeartobetheonly\nquotable cases). According tothegrammarians, thesame change ismade in\nthe inflection ofroot-stems, before bhand su :thus, prafdnbhis, prafdnsu\n(from pra$am: pra-f- j/fam). Noderived noun-stem ends inm.\nQB.hasonce theanomalous kdmvant, from theparticle kdm.\n213. Final fminexternal combination isaservile\nsound, beingassimilated toanyfollowingconsonant.Thus :\n*Kuhn, inhisBeitrage etc.,iii.125.\nWhitney, Grammar.\n"
+    },
+    {
+        "page": 86,
+        "content": "(J5III-EUPHONIC COMBINATION.[213\na.Itremains unchanged onlybefore avowel oralabial\nmute.\nBut also, byananomalous exception, before roftheroot rajinsamrtij\nand itsderivatives samrdjnl andsamrajya.\nb.Before amute ofanyother class thanlabial,itbecomes\nthenasal ofthat class.\nc.Before thesemivowelsy, I,vitbecomes, accordingto\ntheHindu grammarians,anasal semivowel, thenasal counter-\npartofeachrespectively (see 71).\nd.Beforer,asibilant, orh,itbecomes anusvara(see 71).\nTheMSS. and theeditions ingeneral make noattempttodistinguish\nthenasal tones arising from theassimilation ofmbefore afollowing semivowel\nfrom that before aspirant.\ne.But ifhbeimmediately followed byanother consonant (which canonly\nbeanasal orsemivowel), themisallowed tobeassimilated tothatfollowing\nconsonant.\nThis isbecause thehhasnoposition ofthemouth-organs peculiarto\nitself, but isuttered inthepositionofthenext sound. ThePraticakhyas\ndonottakeanynotice ofthecase.\nIthasbeenpointedoutabove(73)that theassimilated\nmisgenerally representedintexts bytheanusvara-sign, and\nthat inthiswork itistransliterated bym(insteadofanasal\nmute orw).Also, thatthegeneral grammariansallowmtobe\npronouncedbeforeanyandevery consonant asanusvara.\nThepalatal mutes and sibilant, and h.\n214. These sounds show insome situations areversion to\ntheoriginal gutturals from which theyarederived. The treat-\nment ofjand/j,also,isdifferent, accordingasthey represent\ntheoneortheother oftwo different degreesofalteration from\ntheiroriginals.\n215. Thepalatals andharetheleast stable ofalphabetic\nsounds, undergoing,invirtue oftheir derivative character,alter-\nation inmany cases where other similar sounds areretained.\n216. Thus, inderivation, even before vowels, semivowels,\nandnasals, reversion toguttural form isbynomeans rare. The\ncases arethefollowing:\n1.Before aofsuffixa,finalobecomes kinankd, arkd, pakd, vakd,\nparka, markd, vfka, prdtika etc.,rc/ca, seka, moka, rokd, fo'fca, tokd, mrokd,\nvraskd: final jbecomes gintyagd, bhdga, bhagd, yaga, bhangd, sanga,\nvarga, marga, mrgd, sarga, vega, bhoga, yugd, yoga, roga: final hbecomes\nghinaghd, maghd, arghd, dlrghd (anddrdghiyas, drdghistha) ,meghd, dogha.\ndrtigha, mdgha; andindughana.\nInanother series ofderivatives witha,the altered sound appears:\nexamples areajd, yaja, pucd, foca, vrajd, vevijd, yuja, urjd, doha.\n"
+    },
+    {
+        "page": 87,
+        "content": "217]COMBINATIONS OFFINAL PALATALS. 67\nBefore thesuffixes asandana,theguttural only rarely appears:namely,\nindnkas, okas, rofcas, ptf/cas, bhdrgas; andinrogana.\n2.Before ani-vowel, thealtered sound appears (exceptinthedoubtful\nabhogi):e.g.o/f, tuji, rtici, fdci, vivid, rocisnu.\n3.Before w,theguttural reappears,asarule(thecases arefew): thus,\nanku, vanku, reku, bhrgu.\n4.Before n,theexamplesofreversion arefew, exceptofj(becoming g)\nbefore theparticipial ending na :thus, reknas, vagnu (with the final also\nmadesonant) ;andparticiples bhagnd, rugnd, etc.; andapparently prgna from\nYprc.\n5.Beforem(ofwa,man, min),theguttural generally appears: thus,\nrukmd, tigmd, fagmd (with sonant change) ;vdkman, sdkman, rukmdn; rgmtn\nandvagmm (with sonant change): butdjman, ojman, bhujmdn.\n6.Beforer/,thealtered sound isused :thus, pacya, yajya, yajyu, yujya,\nbhujyu. Such cases asbhogya andyogya aredoubtless secondary derivatives\nfrombhoga andyoga.\n7.Beforer,thecases arefew,andtheusage apparently divided :thus,\nfukrd, ugrd, mrgra;butvdjra andpajrd(?).\n8.Before v(ofthesuffixes va,van, vin, etc., andparticipial vans]\nthegutturalisregularly preserved:thus, rkvd, pakvd ;fkvan, rikvan, fukvan,\nmrgvan, yugvan; vagvfa (withfurther sonant change); ririkvahs, rurukvdhs,\nCUfukvdns; fupukvand, pupukvani: also before theunion-vowel iinokivdns\n(RV., once). Anexceptionisydjvan.\nThe reversion ofhinderivation isexhibited only before the suffix a\n(andintheparticiple dughana, RV.). The final jwhich isanalogous with\nf(219) shows much lessproclivitytoreversion than thatwhichcorresponds\nwith c.\n9.Alikereversion shows itself also tosome extent inconjugational\nstem-formation and inflection. Thus, the initial radical becomes guttural\nafter thereduplicationinthepresentorperfectordesiderative orintensive\nstems oftherootsci,cit, ji, hi,han;andhanbecomes ghnontheelision of\na.TheRV.hasvivakmi fromyvac. And before ran etc. of3dpi.mid.we\nhave gforradical jinasrgran, asrgram, asasrgram (allinRV.).\n217. Final rfcofaroot orstem,iffollowed inin-\nternal combination byanyother sound than avowel or\nsemivowel ornasal, reverts toitsoriginal gutturalvalue,\nandshowseverywherethesame form which a5fikwould\nshow inthesame situation.\nThus, vakti, uvdktha, vaksi, vaksykmi, vagdhi; vagbhis,vaksu;\nulita, ukthd, vaktdr.\nAnd, asfinal cbecomes k(above, 142),thesame ruleapplies\nalso tocinexternal combination :thus, vakca,vhg api,vanme.\n"
+    },
+    {
+        "page": 88,
+        "content": "6III.EUPHONIC COMBINATION.[217\nExamplesofcremaining unchangedininflection are :ucydte,\nriricre, vaci,mumucmdhe.\n218. Final 5T?reverts toitsoriginal3\\&,ininternal\ncombination, onlybefore trie\"Q^sofaverbal stem orending\n(whence, by180, ^fks]-,before rTtand 5Tih,iteverywhere\nbecomes Efs(whence, by197,^szand\"$\"s%before VIdh,\nHbh,andHswoftheloc.pi.,aswhen final(145),itre-\ngularlybecomes thelingualmute(tor3d).\nThus, dviksata, veksykmi ;vdsti, vista, didestu; dididdhi,\nvidbhis.\nButafewroots exhibit thereversion offinal ctokbefore\nbhandsu,and alsowhen final(145): theyare die, drc, sprc,\nandoptionallynac(always,inV.);andvifhas inV.always\nviksu,loc.pl. ;but vit, vidbhis,etc. Examplesarediksamcita,\ndrgbhis, hrdisprk,ndk.\nExamplesofcremaining unchanged before vowels etc. are:\nvici, vivicyas, avicran, acnomi, vacmi, ucmdsi.\n\\$remains irregularly unchanged before pinthecompound vifpdti.\n219.Final sf/isinone setofwords treated like rfc,sT -v7\nandinanother setlike STf.\nThus, fromyvj':dyukthas, dyukta, yunkte, yukli, yoktra,\nyoksyami, yuksu ;yungdhi, dyugdhvam, yugbhis.\nAgain, frommrjetc.: dmrksat, sraksy&mi; mtirsti, mrstd, srsti,\nrastrd;mrddhi, mrddhvdm, radbhis, ratsu, rat.\nTotheformer oryuj-cl&ss belong (asshown bytheir quotable forms)\nabout twenty roots and radical stems: namely, bhaj, saj, tyaj (riot V.), raj\n'color', svaj, majj, nij, tij,vij,iandIbhuj, yuj, ruj, vrj, anj, bhanj, fify ;\nurj, sraj, bhisdj, dsrj ; also, stems formed with the suffixes ajandy\n;383. 5),astrsndj, vanij:andrtvij, though containing therootyaj.\nTothelatter ormrj-class belong onlyabout onethird asmany:namely,\nyaj, bhrajj, vraj, raj, bhraj, mrj, srj.\nA.considerable number ofj-roots arenotplacedincircumstances toex-\nhibit thedistinction; butsuch roots areinpart assignabletoone orthe\nother class ontheevidence ofthe related languages*. The distinction\nappears, namely, onlywhen thejoccurs asfinal,orisfollowed,either in\ninflection orinderivation, byadental mute(t,th,dh), or,innoun-inflec-\ntion, bybhorau. Inderivation (above, 216) wefind agsometimes from\nthemrj-class: thus, mdrga (inapamarga) andsarga; andbefore rofVedic\n3dpi.mid. endings, asrgran, asrgram, asasrgram (beside saarjrire) while\nfrom theywj-class occur only yuyujre, ayujran, bubhujrire, with j.\n*SeeH&bschmann, inKZ. xxiii. 384 ff.\n"
+    },
+    {
+        "page": 89,
+        "content": "222]COMBINATIONS OFFINALch, Jcs,h. 69\n220. Final chfallsunder therules ofcombination almost\nonlyintherootprac/i,inwhich itistreated asifitwere c\n(andpracisperhapsitsmoreoriginal form):thus, praksy&mi,\nprstd;and also thederivative pracnd. As final and innoun-\ninflection(beforebhandsu],itisdirected tobechangedtothe\nlingual mute.\nMurtd iscalled theparticipleofmurcfc, and agerund murtva isgiven\ntothesame root. They (with mUrti] must doubtless come from asimpler\nform ofthe root.\nOfjhthere isnooccurrence :thegrammarians declare it\ntobetreated like c.\n221. Thecompoundksisnotinfrequentasfinal ofaroot\n(generallyofdemonstrably secondary origin),orofatense-stem\n(s-aorist:seebelow, chap. XI.);and, inthenotveryfrequentcases\nofitsinternal combination,itistreated asifasingle sound,\nfollowingtherules forc:thus, cdkse(caks -j-se),cdksva;caste,\ndcasta, dsrastam, dsrsta, tvdstar. Astoitstreatment when final,\nsee146.We aretaught bythegrammarianstomake such\nforms asgordt, gorddbhis, gordtsu (from gordks];andweactually\nhavesdt, sadbhts, satsu from saks orsas(146, end).\nInthesingle anomalous root vrapc,thecompound fcissaid tofollow\ntherule forsimple f:thus, vraksydti, dvraksam; vrdstum, vrasta. Its c.\nreverts to feinvraska. Itsparticipleisvrknd;ithasaVedic gerund vrktin.\n222.The roots infinal^A,likethose in^j,fallinto\ntwo classes, exhibitingasimilardiversityoftreatment, ap-\npearinginthesame kinds ofcombination.\nIntheone class, asduh,wehave areversion ofh(asofc]\ntoaguttural form, and itstreatment asifitwere still its\noriginal gh: thus, ddhuksam, dhoksy&mi; dugdham, dugdhd;ddhok.\nd/iuk, dhugbhis,dhuksu.\nIntheother class, asruhand sah,wehave agutturalrever-\nsion(asofc)onlybefore 5inverb-formation andderivation :thus,\ndruksat, roksyftmi, saksiyd,saksdni. Asfinal, inexternal combi-\nnation, andinnoun-inflection before bhand su,theh(like c]\nbecomes alingual mute :thus, turasnt, prtanasad ayodhydh,\nturasndbhis, turashtsu. Butbefore adental mute(ttth,dh)in\nverb-inflection and inderivation,itseuphoniceffect ispeculiarly\ncomplicated:itturns thedental into alingual (aswouldc];but\nitalsomakes itsonant andaspirate (aswould dh: see160);\nand further,itdisappears itself, and thepreceding vowel,if\nshort,islengthened bywayofcompensation:thus, from ruh\nwith tacomes rudhd, from lehwith ticomes Udhi, fromguh\nwith tarcomesgudhdr, frommehwith turncomes medhum, from\n"
+    },
+    {
+        "page": 90,
+        "content": "70 HI.EUPHONIC COMBINATION.[222\nlihwith tasorthascomes lid/ids, from lihwith d/warn comes\nlidhvum, etc.\nThis isasifwehad toassume astransition sound asonantaspirate\nlingualsibilant zh,with theeuphoniceffects ofalingual and ofasonant\naspirate (160),itself disappearing under thelawoftheexisting language\nwhich admits nosonant sibilant.\n223. The roots ofthetwo classes, asshownbytheirforms\nfound inuse, are :\nofthe first ordw/j-class :dah, dih, duh, druh, muh, snih\n(andthe final ofumih issimilarly treated);\nofthesecond orrttA-claSB :yah, sah,mih, rihorli/i,guh,\nruh, drh/i, trnh, brJi, bank, sprh(?).\nButmuh forms also(notinRV.) theparticiple mudha andagent-noun\nmudhdr, aswell asmugdhd andmugdhdr ;anddruhandsnih areallowed by\nthegrammarianstodolikewise: such forms asdrudha and sriidha, however,\ndonotappeartohave beenmetwith inuse.\nFrom roots oftherwft-class wefind also intheVeda theforms gartaruk,\nnom. sing., andpranadhrk anddadhfk; andhence puruspfk (theonlyoccur-\nrence) does notprove ysprhtobeofthe dwft-class.\nAnumber ofother ft-roots arenotproved bytheir occurring forms to\nbelongtoeither class;they, too, arewithmore orless confidence assigned\ntotheone ortheother bycomparison with therelated languages*.\nInderivation, before thesuffixa,wehave (216.1) meghd anddlrghd\nfrom roots ofthe rwft-class. Before therverb-endings, wehaveexamples\nonlyfromdw/t, with h:thus, duduhre etc.\nThe rootnahcomes fromoriginal dhinstead ofgh,and itsreversion is\naccordingly toadental instead ofaguttural mute: thus, natsydmi, naddhd,\nupanddbhis, upanad-yuga.\n224. Irregularities ofcombination are :\na.Thevowel risnotlengthened tocompensateforthe loss ofthe\nfe-element :thus, drdhd, trdhd, brdhd(theonly cases;andintheVeda their\nfirst syllable hasmetrical value asheavy orlong).\nb.Theroots vahandsahchange their vowel to-oinstead oflengthening\nit:thus, vodhdm, vodhdm, vodhdr, sddhum. Butfrom sahintheolder\nlanguage forms withaaremorefrequent: thus, sadhd, dsadha(also later),\nsadhar. The roottrhhchanges thevowel ofitsclass-sign nainto einstead\noflengthening it:thus, trnedhi, trnedhu, atrnet(thegrammarians teach also\ntrnehmi and trneksi: ifsuch forms areactually inuse,theymust havebeen\nmade byfalseanalogy with theothers).\nThese anomalousvowel-changes seem tostand inconnection with the\nfact that thecases showing them aretheonly ones where other than an\n*SeeHubschmann, asabove.\n"
+    },
+    {
+        "page": 91,
+        "content": "226 COMBINATIONS OFFINAL s. 71\nalterant vowel (180) comes before thelingualisedsibilantrepresentative of\nthe h.Compare sodafa etc.\nApparently bydissimilation, the final ofvahintheanomalous compound\nanadvah ischanged todinstead ofd:see404.\nThelingual sibilant s.\n225. Since thelingual sibilant, initsusual andnormal\noccurrences,is(182)theproductoflingualizationofsafter\ncertain alterant sounds, wemight expectfinal radicals,when\n(inrarecases)itcomes tostand where ascannot maintain it-\nself, torevert toitsoriginal, andbetreated asaswould be\ntreated under thesame circumstances. That, however, isonly\ntrue inaveryfew instances.\nNamely, intheprefix dus (evidently identical withydus);insajus\n(adverbially used case-form fromyjus);in(RV.)vives anddvives, fromyvis;\ninaiyes (RV.), fromyis;andinatfs, from ciassecondary form ofy$as.\nAllthese, exceptthe first two, aremore orlessopentoquestion.\n226. Ingeneral,finallingualTsistreated inthesame\nmanner aspalatal5T$.Thus :\na.Before tand thitremains unchanged, and the latter\nareassimilated :e.g. dvistas, dvisthas, dvestum.\nThis isacommon andperfectly natural combination.\nb.Before dh, bh,and su,asalso inexternal combination\n(145),itbecomes alingual mute;anddhismadelingualafter\nit :thus, pinddhi, viddhi, vivid dhi, dvidflkvam, dvidbhis, dvitsu.\nThesame holds good ofthealtered sofatense-sign:thus, dstoddhvam\n(from astos-dhvam}.\nTheconversion of*to*(ord)asfinalandbefore &/tandsuisparallel\nwith thelikeconversion ofp,and ofjandhinthemrjandruh classes of\nroots, andperhaps with theoccasional change ofstot(167 8).Itisavery\ninfrequent case, occurring (save asitmaybeassumed inthecase ofacts)only\nonce inRV.andonce inAV. (-dvit and-prut), although those texts have\nmore than40roots with final s;intheBrahmanas, moreover, hasbeen noticed\nfurther only -flit. From pins, RV. hastheirregular form pinak (2dand\n3dsing.,forpinas-s andpinas-t}.\nc.Before sininternal combination(exceptsuofloc.pi.)\nitbecomes k:thus, dveksi, dveksy&mi,adviksam.\nThis changeisofanomalous phonetic character, and difficult ofex-\nplanation. Itisalsopractically ofvery rare occurrence. The onlyRV.\nexamples (apart from pinak, above) areviveksi, fromyvis,andthe desid.\nstem ririksa fromyris ;AV.hasonly dviksat anddviksata, andthedesid. stem\nfifliksa fromy^lis. Other examples arequotable fromyykra (B. etc.), fis\n(QB.), andpis(Khand. Up.); andthey arebytheHindu grammarians pre-\nscribed tobeformed from about half-a-dozen other roots.\n"
+    },
+    {
+        "page": 92,
+        "content": "72III.EUPHONIC COMBINATION.[227\nExtension andAbbreviation.\n227.Asageneral rule, chisnotallowed bythegramma-\nrians tostand inthatform after avowel, but istobedoubled,\nbecomingcch(whichtheMSS. sometimes writechcfy.\nThevarious authorities disagree with oneanother indetail astothis\nduplication. According toPanini, chisdoubled within aword after either\nalong orashort vowel; and, asInitial, necessarily after ashort and after\nthe particles dandmd,and optionally everywhere after along. InRV.,\ninitial chisdoubled after alongvowel of <Jonly, and certain special cases\nafter ashort vowel areexcepted.Fortherequired usage intheother Vedic\ntexts,seetheir several Pratoakhyas. TheKathaka writes fororiginal ch\n(notchfrom combination oftandnwith c:203) after avowel every-\nwhere cch.\nOpinionsare still atvariance astohow farthisduplication hasan\netymological ground, andhow far itisonlyanacknowledgment ofthe fact\nthatchmakes aheavy syllable even after ashort vowel (makes \"position\": 79).\n228. Afterr,anyconsonant(saveaspirant before avowel)\nisbythegrammarianseither allowed orrequiredtobedoubled\n(anaspirate, byprefixingthecorresponding non-aspirate:154).\nSome oftheauthorities include, along withr,also horIor-u,ormore\nthan oneofthem, inthis rule.\nAdoubled consonant after risverycommon inMSS. andinscriptions,\nasalso innative text-editions andinthe earlier editions prepared byEuro-\npean scholars inlater ones, theduplicationisuniversally omitted.\nThe firstconsonant ofagroup whether interior, orinitial after\navowel ofapreceding word isbythegrammarians either allowed or\nrequiredtobedoubled.\nThis duplicationisallowed byPanini andrequired bythePraticakhyas\ninboth, with mention ofauthorities whodenyitaltogether. Forcertain\nexceptions,seethePraticakhyas:themeaningofthewhole matter istoo\nobscure tojustify thegiving ofdetails here.\n230. Other cases ofextension ofconsonant-groups, required\nbysome ofthegrammatical authorities, arethefollowing:\na.Between anon-nasal andanasal mute, theinsertion ofso-called\nyamas ('twins'),ornasalcounterparts,istaught bythePrati^akhyas (and\nassumed inPanini'scommentary): seeAPr. i.99,note.\nb.Between handafollowing nasal mute thePraticaktiyas teach the\ninsertion ofanasal sound called nasikya:seeAPr. i.100, note.\nc.Between rand afollowing consonant thePraticakhyas teach the\ninsertion ofasvardbhakti or'vowel-fragment': seeAPr. i.101 2,note.\nSome authorities assume this insertion only before aspirant;theothers\nregard itastwice aslong before aspirant asbefore any other consonant\n"
+    },
+    {
+        "page": 93,
+        "content": "234]ABBREVIATION OFCONSONANT-GROUPS. 73\nnamely, ahalf oraquarter mora before theformer, aquarter oran\neighth before the latter. One(VPr.) admits itafter Iaswell asr.Itis\nvariously described asafragment ofthevowel aorofr(or I).\nTheRPr. putsasvarabhakti alsobetween asonant consonant and a\nfollowing mute orspirant; andAPr. introduces anelement called sphotana\n('distinguished) between aguttural andapreceding mute ofanother class.\nForoneortwoother cases ofyetmore doubtful value, seethePrati^akhyas.\n231. After anasal, theformer oftwonon-nasal mutes\nmaybedropped, whether homogeneous onlywith thenasal, orwith\nboth :thus, yundhiforyungdhi, yundhvdmforyungdhvdm, antdm for\nanktdm, chintkm forchintt&m, bhinthd forbhintthd, indhe forinddhe.\nThis abbreviation, allowed byPanini,isrequired byAPr. (the other\nPraticakhyas take nonotice ofit).Itistheusualpractice oftheMSS.,\nthough thefullgroupisalso often written.\n232. Ingeneral, adouble mute(includinganaspirate\nwhich isdoubled bytheprefixionofanon-aspirate)incombi-\nnation withanyother consonant isbythemanuscriptswritten\nassimple.\nThat istosay, theordinary usage oftheMSS. makes nodifference\nbetween those groupsinwhich aphonetic duplicationisallowed bytherules\ngiven above (228, 229) andthose inwhich theduplicationisetymological.\nAsevery tvafter avowel may alsobeproperlywrittenttv, sodattvff and\ntattvd may be,andalmost invariably are, written asdatvd and tatvd. As\nkartana isalsoproperly kdrttana, sokarttika (fromkrtti]iswritten askartika.\nSoininflection, wehave always, forexample, majfid etc., notmajjnd, from.\nmajjdn. Even incomposition andsentence-collocation thesame abbreviations\naremade:thus, hrdyotd foxhrddyotd; chindty asyaforchindtty asya. Hence\nitisimpossible todetermine bytheevidence ofwritten usage whether we\nshould regard adhvam oraddhvam (from yas),ddvidhvam orddviddhvam\n(from ydvis)asthetrueform ofasecond person plural.\n233.Among occasional omissions ofanetymologically justified member\nofaconsonant- group,isofimportance enoughtobehere noticed* that\nAsissometimes lost(perhapsafterassimilation)between\ntwosurdmutes :thus,\na.The initial softhe roots sthaand stabh after thepreposition ut:\nthus, utthdtum forutsthdtum, uttabhnoti forutstabhnoti.\nb.Thetense-sign8ofthes-aorist(chap. XI.)after afinal consonant\nofaroot before the initial consonant ofanending: thus, achantta (and for\nthis, by231, achanta]forachantsta, fapta forfapsta, taptam fortapstam.\nc.Other instances areonly sporadic:thus, thecompoundrkthd (re-{-stha:\nPB.); thecollocations tasmat tute(forstute) andpuroruk tuta- (forstuta-: K.).\nStrengthening andWeakeningProcesses.\n234. Under thishead, wetakeupfirst thechangesthat\naffect vowels, andthen those that affect consonants adding,\n"
+    },
+    {
+        "page": 94,
+        "content": "74 HI.EUPHONIC COMBINATION.[234\nforconvenience's sake, ineachcase, abrief notice ofthevowel\nandconsonant elements that have come tobear theapparent\noffice ofconnectives.\nGuna andVrddhi.\n^235.The so-called gunaandvrddhi-changesarethe\nmostregularandfrequentofvowel-changes, beingofcon-\nstant occurrence both ininflection and inderivation.\nAguna-voweldiffers from thecorresponding simple\nvowel byaprefixed ^-element, which iscombined with the\nother accordingtotheusual rules; avrddhi-vowel, bythe\nfurtherprefixionofatotheffwia-vowel. Thus, of5ior\n|thecorresponding gunais(a-\\-i=) ^e;thecorrespond-\ningvrddhi is(a-f-e==) ^7at.Butinallgunating processes\n5faremains unchanged or,asitissometimesexpressed,\n51aisitsownguna;ma,ofcourse, remainsunchanged\nforbothguna andvrddhi.\n{236.The series ofcorresponding degreesisthen as\nfollows :\nsimple vowel aait uu r I\nGuna aa e o ar al\nVrddhi a ai au ar\nThere isnowhere anyoccurrence offinasituation toundergo either\nguna orvrddTu-change ;nordoes I(26) ever suffer changetovrddhi. Theo-\nretically, fwould have thesame changes asr;andthevrddhi ofIwould\nbeal.\n237. The historical relations ofthemembers ofeach vowel-series are\nstillmatters ofsome difference ofopinion. From thespecial point ofview\noftheSanskrit, thesimple vowels wear theaspect ofbeing ingeneral the\noriginal orfundamentalones, and the others ofbeing products oftheir\nincrement orstrengthening, intwo several degrees sothat therules of\nformation directa, i,u,r,Itoberaised toguna orvrddhirespectively,\nunderspecified conditions. But rissoclearly seen tocome byabbreviation\norweakening from anearlier ar(orra)thatmany European grammarians\nprefer totreat thepuna-form astheoriginal andtheother asthederivative.\nThus, forexample:instead ofassuming certain roots tobebhrandvrdh,\nandmaking fromthem bharati andvardhati, andbhrta andvrddha, bythe\nsame rules which from bhuandnlandfrom budhand citform bhavati and\nnayati, bodhati andcetati, bhuta andmta, buddha and citta theyassume\n"
+    },
+    {
+        "page": 95,
+        "content": "241\n1GUNA ANDVRDDHI. 75\nbharandvardh tobetheroots, andgive therules offormation forthem in\nreverse. Both methods have theiradvantages, and thequestion between\nthem isoneofminor consequence, which may fairly besettled byconsidera-\ntions ofconvenience. Mainly from such considerations, ther-form isgen-\nerally assumed inthiswork, though' notwithout constant recognition of\ntheother.\n238.The^ma-incrementisanIndo-European phenomenon,\nand isinmanycases seen tooccur inconnection withanaccent\nontheincreasedsyllable.Itisfound :\na.Inroot-syllables:either ininflection, asdvfati from\nydvis,dohmi fromyduh;orinderivation, asdvesa, dohas,\ndvestum, dogdhum.\nb.Informative elements :eitherconjugational class-signs,\nastanomi from tanu; orsuffixes ofderivation, infurther deri-\nvation orininflection, asagnaye fromagni,bhandvas from bhcinu,\npitdramfrompitf, hantavyd from hdntit.\n239. TheWfflUttt-incrementisspecificallyIndian(being\nunshared, save inafewdoubtful cases, even bytheZend),and\nitsoccurrence ismoreinfrequent andirregular.Itisfound :\na.Inrootandsuffix-syllables,instead ofguna:thus, stauti\nfromystu, sdkhayamfrom sdkhi, dnaisam from)/m,dkarsam and\nMrdyati andkarya fromykr (orkar),dataram from datr.\nb.Especially often, ininitialsyllablesinsecondaryderi-\nvation :thus, manasd from mdnas, vaidyutd fromvidyut, bhaumd\nfrom bhumiy parthiva fromprthivl.\nBut\n240. The^ma-incrementdoes notusuallytakeplaceina\nheavy syllable endingwith aconsonant :that istosay,the\nrulesprescribing gunainprocessesofderivation and inflection\ndonotapplytoashort vowel which is\"long byposition\", nor\ntoalongvowel unless itbefinal :thus, cetati fromj/ceV,but\nnindati fromynind; ndyati from)/m,butjivati fromyjlv.\nThevrddhi-increment isnotliable tothis restriction.\nExceptions to-the rule areoccasionally met with :thus, dideva from\nydw;hedas fromyhid;ohate from}/uh.\nAnd afewcases occur ofprolongation instead ofincrement: thus, dusdyati\nfrom ydus, guhati fromyguh.\nThechangesofr(more original arorra)aresovarious\nastocallforfurtherdescription.\n241. Theincrements ofraresometimes raand ra,instead\nofarandar :namely, especially, where bysuch reversal adif-\nficult combination ofconsonants isavoided :thus, fromydrc,\ndraksydmi andddraksam;but alsoprthu andpratk, prch and\nprach, krpaanddkrapista.\n"
+    },
+    {
+        "page": 96,
+        "content": "I\n7(5ill.EUPHONIC COMBINATION.|242\n242. Inanumber ofroots(aboutadozen quotable ones)\nendinginr(formore original ar),therexchanges both with ar,\nand,more irregularly,inapartoftheforms, with ir oralso\nwith ur(especiallyafter alabial, inpr>mT>v?>sporadicallyin\nothers'!:which irand ur,again,areliable toprolongationinto\nirand ur. Thus, forexample, from tr(or tar}.,wehave tarafi,\ntitarti, tatara, atarisam, byregular processes;but alsott'rati,\nfiryati, tirtva, -tirya, firna, andeven(V.)turyama, tuturyat,tar-\nturana. The treatment ofsuch roots has tobedescribed in\nspeakingofeach formation.\nForthepurposeofartificially indicating this peculiarity oftreatment,\nsuch roots arebytheHindu grammarians written with long f,orwith both r\nandf:nofactually appears anywhere among their forms.\nThe(quotable) f-rootsare3kr 'strew', igr 'praise', Igr 'swallow', ijr\n'wear out', *r,Iff'crush'*.\nThe(quotable)randf-rootsarer,idr'burst', Ipr 'fill',2mr'die',\n2w 'choose', str, hvr.\n243. Inafew cases, rcomes from the contraction ofother syllables\nthan arandra:thus, intrtaandtrtlya from ri;inprnu from ru;intrca,\neven from rir(orryr).\nVowel-lengthening.\n244. Vowel-lengthening hasregard especiallytoiandu,\nsince thelengtheningofaisinpart (exceptwhere inevident\nanalogy with that ofiandu)indistinguishablefrom itsincre-\nment, and rismade long onlyincertainpluralcases ofstems\ninr(orar :chap. V.).Lengtheningisamuch moreirregular and\nsporadic change than increment, and itscases will ingeneral\nbeleft tobepointed out inconnection with theprocessesof\ninflection andderivation :afewonlywillbementioned here.\n245. a.Final radical ianduareespeciallyliable toprolongation\nbefore y:asinpassive andgerund and soon.\nb.Final radical irandur(fromr-roots :242)areliable toprolongation\nbefore allconsonants except those ofpersonal endings:namely, before yand\ntvaandna: and indeclension before bhand(392).Radical isandus\nhave thesameprolongation indeclension.\n246. Compensatory lengthening, orabsorption byavowel ofthetime\nofalostfollowing consonant, isbynomeans common. Certain instances of\nithavebeen pointed outabove (179, 198b,199b, 222). Perhaps such cases\naspita forpitars anddhanl fordhanins (chap. V.)aretobeclassed here.\n247. The final vowel ofaformer member ofacompoundisoftenmade\nlong, especially intheVeda. Prolongationsoffinala,andbefore v.aremost\n*Numbers prefixed toroots denote theorder inwhich, there being more than\noneofthesame form, therootintended isgiven inthePetersburg lexicon.\n"
+    },
+    {
+        "page": 97,
+        "content": "250]VOWEL-LENGTHENING ANDLIGHTENING. 11\nfrequent;but cases arefound ofevery variety. Examplesaredevavi,\nvayundvfd, praWt, rtdvasu, mdrdvant, sadandsdd, fatamagha, vifvanara,\nekdda$a; aplju, parlndh, virudh, tuvlmaghd, tvfsimant, fdktlvant; vasuju,\nanurudh, sumdya, puruvdsu.\n248. IntheVeda, the finalvowel ofaword generally a,much less\noften iandu isinalargenumber ofcases prolonged. Usually the\nprolongation takes place where itisfavored bythemetre, butsometimes\neven where themetre opposes thechange (for details, seeAPr. iii.16note,\nBenfey, Abh. Gott. Ges. xix. xxi.[1874 6],and tlievarious Prati^akhyas).\nWords ofwhich the finals arethus treated are:\na.Particles: namely, dthd, ddhd, evd~, utd, ghd, fca,ihd, ivd, cd,smd,\nangd, ktld, dtrd, ydtrd, tdtrd, ktitrd, anydtrd, ubhaydtrd, adyd, dchd, dpd,\nprd ;yddi, nahi, abhi;u, tti,mi, sG,maksti.\nb.Case-forms :especiallyinstr.sing. ,asend, tend, t/ena, svend, aud\nothers; rarely gen. sing., asasya, harindsyd. Cases besides these arefew:\nsostmd(voc.);tanvi(loc.), anduruand(not rarely) puru.\nc.Verb-forms endingina,ingreatnumber andvariety:thus (nearly\nintheorder oftheir comparative frequency), 2dsing. impv. act., aspibd,\n57/a,gamayd; 2dpi.act.intaand tha, assthd, attd, bibhrtd, jayatd,\nfrn'ufa, anadatd, nayathd, jlvayathd (andoneintana: avistand); 1stpi.\nact. inma,asvidmd, risdmd, rdhydmd, ruhemd, vanuydmd, cakrmd, mar-\nmrjmd; 2dsing.impv. mid. insua,asyuksvd, idisvd, dadhisvd, vahasvd;\n1stand3dsing. perf. act.,asvedd, vive^d, jagrabhd; 2dsing. perf.\nact., vetthd; 2dpi.perf. act., ancya,cwfera. Ofverb-forms ending ini,\nonly the2dsing. impv. act., as$rudhi, prnuhi, dldihi, jahi.\nTothesemaybeadded thegerund inya,asabhigUryd, dcyd.\nVowel-lightening.\n249. The alteration ofshort atoaniorw-vowel inthe\nformative processesofthelanguage, exceptinrorarroots(as\nexplained above),isasporadic phenomenon only.\n250. Butthelighteningofalongaespeciallytoan/-vowel\n(asalso itsloss),isafrequent process:noother vowel isso\nunstable.\na.Oftheclass-sign nd(ofthefcn-class ofverbs :chap. IX.), thedisin\n\"weak\" forms changedtoi,and before vowel-endings dropped altogether.\nThe finaldofone ortworoots istreated inthesamemanner :thusma, hd.\nAnd from some roots, aand Iori-forms sointerchange that itisdifficult\ntoclassify them ortodetermine thetrue character oftheroot.\nb.Radical aisweakened tothesemblance oftheunion-vowel iin\ncertain verbal forms :asperfect dadima fromyddetc.;aorist adhithds from\nydhdetc.;present jahimas fromj/ftaetc.\nC.Radical aisshortened tothesemblance ofstem-a inanumber of\n"
+    },
+    {
+        "page": 98,
+        "content": "7$III.EUPHONIC COMBINATION.[250\nreduplicated forms,astistha, piba, dada,etc. :seechap.IX.\n;also inafew\naorists, asdhvam, dkhyam,etc. :seechap. XI.\nd.Radical asometimes becomes e,especiallybefore y:as,stheyasam, dey<i.\n251. Certain a-roots, because oftheir peculiar exchanges with *and\ni-forms, especiallyinformingthepresent stem,aregiven bytheHindu\ngrammariansasroots endingineoraioro.Thus, fromQdha 'suck' (dhe)\ncome thepresent dhdyatiandparticipleandgerund dhltd,dhitva;theother\nforms aremade from dha,asdadhus, adhat, dhasyati, dhdtave, dhdpayati.\nFrom 2$ra 'sing' (gai) come thepresent g&yati,the participle andgerund\ngltdandgltvti, andpassive glyate,andtheother forms from ga.From 3da\n'cut' (do)come thepresent dydti and participleditd ordma, and theother\nforms from da.The iiregularities ofthese roots willbetreatedbelow, under\nthevarious formations.\n252. Byaprocessofabbreviation essentially akin with that ofaror\nrator,theva(usually initial) ofanumber ofroots becomesw,andtheya\nofamuch smaller number becomesi,incertain verbal forms andderivatives.\nThus, from vaccome uvdca, ucydsam, uktva, ukta, uktf, ukthd,etc.;from\nyajcomeiydja, ijydsam, istvd, istd, tsti,etc. Seebelow, under thevarious\nformations.\nTothischangeisgenerally given byEuropean grammarians thename\nofsamprasarana, byadaptation ofaterm used inthenative grammar.\n253.Ashorta,ofroot orending,isnotinfrequentlylostbetween\nconsonants inaweakened syllable:thus, inverb-forms, ghndnti, dpaptam,\njagmus, jajnus, djnata ;innoun-forms, rajne, rajni.\n254.Union-vowels. Allthesimple vowels come toassume incer-\ntain cases theaspect ofunion-vowels, orinsertions between root orstem\nandending ofinflection orofderivation.\nThat character belongs oi'tenest toi,which isverywidely used :a.before\nthe sofaorist andfuture anddesiderative stems, asindjlvisam, jlvisydmi,\njijlvisami; b.intense-inflection, especially perfect,asjijivima; occasionally\nalso present,asdniti, roditi;c.inderivation, asjivitd, khdnitum, janitf,\nrocisnu,etc. etc.\nLongIisusedsometimes instead ofshort: thus, agrahlsam, grahlsydmi;\nbravlti, vavadlti; tarltr,savitr;itisalsooften introduced before sand tof\nthe2dand3dsing, ofverbs:thus, cms,dslt.\nFor detailsrespecting these, andthemore irregular andsporadic occur-\nrences ofuanda-vowels inthesame character, seebelow.\nNasal Increment.\n255. Both inroots and inendings,adistinction ofstrong-\nerand\\yeaker forms isveryoftenmade bythepresenceor\nabsence ofanasal element, mute oranusvara, before afollow-\ningconsonant. Ingeneral, thestronger form isdoubtless the\nmoreoriginal; but inthepresent condition ofthelanguage,\n"
+    },
+    {
+        "page": 99,
+        "content": "260]NASAL INCREMENT. 79\nthenasal hascome ingreat measure toseem, and tosome\nextent also tobeused, asanactually strengthening element,\nintroduced under certain conditions informative and inflective\nprocesses.\nExamples are,ofroots: acandanc, grath and yranth, vidandvind,\ndac,anddan? ,srasandmms,drhanddrhh :ofendings, bhdrantam and\nbhdrata, mdnasi andmdnahsi.\n256.Aflrial n,whether ofstem orofroot,isless stable than any\nother consonant, where aweaker form iscalled for :thus, from rajan we\nhave rajaand rajabhis, and incomposition raja; from dhanin,dharii and\ndhanibhis anddhani;fromyhanwehave hathd andhatdtetc.Afinal rad-\nicalm(mdoes notoccur asfinal ofastem)issometimes treated inthe\nsame way: thus, fromygam, yaht, yatdm, yata, ydti.\n257. Inserted n.Ontheother hand, thenasal nhascome tobe\nused withgreat and,inthelater history ofthelanguage, with increas-\ning frequencyasaunion-consonant,inserted between vowels :thus,\nfromayni, aynina andayninam ;from mddhu, mddhunas, mddhurii, md-\ndhuni; from fivd. fivena, $ivani, fivanam.\n258. Insertedy.After long ,ayisnotvery infrequently found\nasapparentlyamere union-consonant before another vowel: thus, inderi-\nvation, yaym, svadhayin, dhayas. sthayuka;ininflection, ddhayi, faydyati,\nandperhaps vivtiyas and gayati.\nReduplication.\n259. Reduplicationofaroot(originatingdoubtless inits\ncomplete repetition)hascome tobeamethod ofradical incre-\nment orstrengtheninginvarious formative processes:namely,\na.inpresent-stem formation :asdddamt, bibhdrmi;\nb.inaorist-stem formation: asadidharam, dcucyavam ;\nc.inperfect-stem formation, almostuniversally:astathna,\ndadhau, cakara;\nd.inintensive anddesiderative-stem formation, through-\nout :asjanghanti, johaviti, marmrjyate, pipasati, jighahsati ;\ne.intheformation ofderivative noun-stems :aspapri,\ncdrcara, sasa/ii, cikitu, malimlucd.\nRules forthetreatment ofthereduplicationinthese several cases will\nbegiven intheproper connection below.\n260. As,byreason ofthestrengthening andweakening\nchanges indicated above, thesame root orstem notseldom\nexhibits, intheprocessesofinflection and derivation,varieties\nofstronger andweaker form, thedistinction anddescriptionof\nthese varieties forms animportant partofthesubjectshereafter\ntobetreated.\n"
+    },
+    {
+        "page": 100,
+        "content": "80 [261\nCHAPTER IV.\nDECLENSION.\n261.THEgeneral subjectofdeclension includes nouns, ad-\njectives,andpronouns,allofwhich are inflected inessentially\nthesame manner. Butwhile thecorrespondence ofnouns and\nadjectivesissoclose thattheycannot wellbeseparatedintreat-\nment, thepronouns,which exhibit many peculiarities,willbe\nbest dealt with inaseparate chapter ;andthewords designat-\ningnumber, ornumerals, alsoform aclasspeculiar enoughto\nrequiretobepresented bythemselves.\n262. Declensional forms showprimarilycaseandnum-\nber; buttheyalso indicate gender since, thoughthe\ndistinctions ofgenderaremadepartlyinthestemitself,\ntheyalsoappear,tonoinconsiderable extent, inthechang-\nesofinflection.\n263.Gender. Thegendersarethree, namely mascu-\nline, feminine, andneuter, asintheother older Indo-Euro-\npean languages;andtheyfollow ingeneralthesame laws\nofdistributionas,forexample,inGreek andLatin.\nThe onlywords which show nosign ofgender-distinction aretheper-\nsonal pronounsofthefirstandsecondperson (along with thenumerals above\n'four': chap. VI.).\n264.Number. Thenumbers arethreesingular,\ndual, andplural.\nAfewwords areused only intheplural:asdaras, 'wife', Spas,\n'water'; thenumeraldva, 'two',isdual only; and, asinother languages,\nmany wordsare,hythenature oftheiruse, found tooccur only inthe\nsingular.\n265.Astotheuses ofthenumbers,itneeds onlytobe\nremarked that thedual isfound without theaddition ofthe\nnumeraldva, 'two', wherever thedualityoftheobjects spoken\nofisathing wellunderstood :thus, agvinau,'thetwo(horse-\nmen) Acvins'; indrasya hdri, 'Indra's(two) bays'; devadattasya\nfastau, 'Devadatta's twohands'; butdevadattasyadvavacvau stah,\n'Devadatta hastwo horses'.\n"
+    },
+    {
+        "page": 101,
+        "content": "270]USESOFTHENOMINATIVE ANDACCUSATIVE. 81\n266.Case. Thecases are(includingthevocative) eight:\nnominative, accusative, instrumental, dative, ablative, gen-\nitive, locative, andvocative.\nThe order inwhich they arehere mentioned isthat established for\nthem bytheHindu grammarians, andaccepted from these byWestern schol-\nars.TheHindu names ofthecases arefounded onthis order :thenomi-\nnative iscalled prathama, 'first', theaccusative dvitlya, 'second', thegen-\nitive sasthi,'sixth'(sc.vibhakti, 'division',i.e.'case'),etc.Theobject sought\ninthearrangementissimplytosetnext tooneanother those cases which are\ntoagreater orlessextent, inoneoranother number, identical inform;and,\nputtingthenominativefirst,asleading case, there isnoother order by\nwhich that object could beattained. The vocative isnotconsidered and\nnamed bythenative grammarians asacase like therest; inthiswork,it\nwillbegiven inthesingular (where alone itiseverdistinguished from the\nnominative otherwise than byaccent)attheend oftheseries ofcases.\nAbrief compendiousstatement oftheuses ofthecases\nisgiveninthefollowing paragraphs:\n267.Uses oftheNominative. Thenominative isthe\ncase ofthesubjectofthesentence, and ofanywordqualifying\nthesubject,whetherattributively,inapposition,oraspredi-\ncate.\n268. Assomewhatpeculiar constructions may bementioned apredicate\nnominative withmanye etc.,'think one's self tobe',andwith bruveetc., 'call\none's self: thus,somam manyate papivdn (RV.), 'hethinks hehasbeen drink-\ning sorna'; sdmanyeta puranavit (AV.),'hemay regard himself aswise in\nancient things'; fndro brahmano bruvdnah(TS.),'Indracalling himself(pre-\ntendingtobe)aBrahman'; andwith rupdrh kr:thus,krsno rupdrh krtvd\n(TS.), 'taking onablack form' (making shapeforhimself asone that is\nblack).\n269.Uses oftheAccusative. Theaccusative isespe-\nciallythecase ofthedirectobjectofatransitive verb, andof\nanyword qualifyingthatobject,asattribute orappositiveor\nobjective predicate. The construction oftheverb isshared, of\ncourse, byitsparticiples and infinitives;but also, inSanskrit,\nbyanumber ofother derivatives, having amore orlessparti-\ncipialorinfinitival character, andeven sometimes bynouns and\nadjectives. Afewprepositionsareaccompanied bytheaccusative.\nAsmore indirectobject,theaccusative isconstruedespecially\nwith verbs ofapproach andaddress. Itisfound usedyetmore\nadverbiallyasadjunctofplaceortime ormanner;andahost\nofadverbs areaccusative cases inform. Two accusatives are\noften found asobjectsofthesame verb.\n270. The use oftheaccusative asdirect object ofatransitive verb\nWhitney, Grammar. 6\n"
+    },
+    {
+        "page": 102,
+        "content": "g2IV.DECLENSION.[270\nand ofitsinfinitives andparticiples hardly needs illustration :anexample\nortwoare:agnim Ide,'Ipraise Agni'; ndmo bhdrantah, 'bringing homage';\nbh&yo d&tum arhasi, 'thou shouldst givemore'. Ofpredicatewords quali-\nfyingtheobject,anexampleistdm ugrdrh krnomi tarnbrdhmdnam (RV.j,\n'him Imake formidable, him apriest'.\n271. Ofverbal derivatives havingsofaraparticipialcharacter that\nthey share theconstruction oftheverb, thevarietyisconsiderable: thus,\na.Derivatives inufrom desiderative stems have wholly thecharacter\nofpresent participles:thus, damayantim abhipsavah (MBh.), 'desiringtowin\nDamayanti': rajdnamdidrksuh (MBh.), 'desiring toseetheking.\nb.So-called primaryderivatives ininhave thesame character :thus,\nmam kaminl(AV.), 'loving me';enam abhibhdsinl (MBh.), 'addressing him'.\nC.Derivatives inoka,inthelater language:as,bhavantam abhivddakah\n(MBh.), 'intendingtosalute you'.\nd.Nouns intar,very frequentlyintheolder language, and asperi-\nphrasticfuture forms (chap. XII.)inthelater: thus, hdntd y6vrtrdm sdnito\n*tavajam datamaghdni (RV.),'who slayeth thedragon, winneth booty, bestow-\nethlargesses'.\ne.The root itself,inthe older language, used with thevalue ofa\npresent participleattheend ofacompound: thus,yamyajndm paribhur\ndsi(RV.), 'whatoffering thou surroundest (protectest)'.\nf.Thederivative in ifrom the(especiallythereduplicated) root, inthe\nolder language: thus, babhrir vdjram papih somam dadir gdh (RV.), 'bearing\nthethunderbolt, drinking thesoma, bestowing kine'.\ng.Derivatives inuka, very frequentlyintheBrahmana language:thus,\nvatsahc caghdtuko wrkah (AV.),'and thewolf destroyshis calves'; veduko\nvdso bhavati(TS.),'hewins agarment'.\nh.Other cases aremore sporadic:thus, derivatives ina,asfndro drdhd\nciddrujdh (RV.), 'Indra breaks upevenwhat isfast'; inatnu, asvldu cid\ndrujatnubhih (RV.),'with thebreakers ofwhatever isstrong'; inana,as\ntarnnivarane(MBh.),'inrestraining him'.\n272. Examples ofanaccu ative with anordinary noun oradjective are\nonly occasional: anuvrata issoused: thus, damayantim anuvratah (MBh.),\n'devoted toDamayanti': andkama, asmdm kdmena (AV.), 'through lov-\ningme'.\n273. The direct construction ofcases withprepositionsiscomparatively\nrestricted inSanskrit(seethesubject ofPrepositions, chap. XVI.). With the\naccusative isoftenest foundprati, 'opposite to,inreferenceto', etc.; alsoanu,\n'after, inthecourseof;antar orantara, 'between'; rarely ati, 'across'; abhi,\n'against, to';and others. Case-forms which have assumed aprepositional\nvalue arealso often usedwith theaccusative :asantarena, uttarena, daksin-\nena, avarena, urdhvam, rte.\n274. The accusative isvery often found also asobject ofverbs which\nintherelated languages arenottransitive.\na.Itstandsespecially asthegoalofmotion, withverbs ofgoing, bring-\n"
+    },
+    {
+        "page": 103,
+        "content": "278]USES OFTHEACCUSATIVE. 83\ning, sending, andthelike :thus, vidarbhdn agaman (MBh.), 'they went to\nVidarbha'; divam yayuh (MBh.), 'they went toheaven'; vanagulmdn dhd-\nvantah(MBh.), 'running towoods andbushes'; apodwam lidvahanti(AV.),\n'they carry upwaters tothesky'; devd~n yaje (AV.),'Imakeofferingtothe\ngods'.\nWith verbs meaning 'go',this isanextremely common construction;\nandtheuseofsuch averb with anabstract noun makes peculiar phrases\nof'becoming': thus, samatdmeti,'hegoes toequality' (i.e.'becomesequal');\nsagached badhyatdm mama (MBh.),'heshall become liable tobeslain by\nme'; sapancatvam dgatah (H.),'hewasresolved into the fiveelements'(un-\nderwentdissolution, died).\nb.Verbs ofspeaking follow thesame rule :thus, tarnabravit, 'hesaid\ntohim'; prdkrocad uccdir ndisadham (MBh.),'she cried outloudlytotheNisha-\ndhan'; ydstvo:vd~ca(AV.),'who spoketotheeJ\n.\n275. The cognate accusative, oraccusative oftheimplied object, is,of\ncourse, found also inSanskrit :as,?apatham ?epe,'heswore anoath'.\n276. Asayetmore adverbial adjuncttoaverb, theaccusative isused\ntodenote space traversed :asyojana$atam gantum (MBh.),'togoahundred\nleagues'; much more often, duration oftime: astisro ratrir dlksitdhsydt\n(TS.),'lethimbeconsecrated threenights'; sdsamvatsardm urdhvbltisthat\n(AV.),'hestood uprightforayear'; gatvd trinahordtrdn (MBh.), 'having\ntravelled three complete days'. Butthepoint ofplaceortime also isoccasio-\nnally found represented bytheaccusative (instead ofthelocative).\nThesame case isused adverbially toexpress manner oraccompanying\ncircumstance; andmany adverbs have theaccusative form(seeAdverbs,\nchap. XVI.): this isespecially true ofcompound words, even tosuch an\nextent astoforma class ofadverbial compounds (below: chap. XVIII.).\n277. The accusative isfreely used along with other cases asobjects of\nthesame verb, wherever thesense admits. Andwhen itisusable with a\nverb intwo different constructions, theverbmay taketwoaccusatives, one\nineach construction. Thus, especially,verbs ofhaving recourse, appealing,\nasking:as,tvdm vayarh $aranam gatdh (MBh.),'wehave resorted totbee\nforsuccor'; ap6ydcdmi bhesajdm (RV.),'Iaskthewaters formedicine';\nandverbs ofsending, bringing, imparting, saying:as,gurutvarh nararh\nnayanti (H.), 'conduct aman torespectability';tarnidam abravit (MBh.),\n;shesaid this toher'; other lessusual casesare,forexample, tdmvisdm\nevdJdhok(AV.), 'hemilked from herpoison'; jitvd rdjyarh nalam (MBh.),\n'having won thekingdom from(i.e.bybeating)Nala'.\nAcausativeverb, naturally, takes adouble accusative :thus, tdmgrham\nprave?ayati, 'hemakes herenter thehouse'.\n278.Uses oftheInstrumental. Theinstrumental is\noriginallytheSvith'-case :itdenotesadjacency, accompaniment,\nassociationpassingover into theexpressionofmeans and\ninstrument(bythesame transfer ofmeaning whichappearsin\nourprepositions withandby}.\n6*\n"
+    },
+    {
+        "page": 104,
+        "content": "4IV.DECLENSION.[218\nNearlyalltheuses ofthecase arereadily deducible from thisfunda-\nmental meaning, andshow nothing anomalous ordifficult.\n279. Theinstrumental isoften used tosignify accompaniment:thus,\nagnirdevebhir dgamat (RV.), 'mayAgni come hither along with the gods';\ndvdparena sahdyenakvaydsyasi (MBh.), 'whither wiltthougo,withDvapara\nforcompanion?' kathayanndisadhena,(MBh.), 'talking with theNishadhan'.\nButtherelation ofsimple accompanimentismore often helpedtoplainer\nexpression byprepositions.\n280. Theinstrumental ofmeans orinstrument oragent isyetmore\nfrequent:thus, bhadrdm kdrnebhih prnuyama (RV.), 'maywehearwith our\nearswhat ispropitious'; fastrena nidhanam(MBh.),'death bythesword';\nkecitpadbhydrhhatd gajdih (MBh.), 'some were slain bytheelephants with\ntheir feet'. And this passes easily over into theexpression ofoccasion or\nreason(forwhich theablative ismorefrequent):thus, krpayd, 'through pity';\ntena satyena, 'byreason ofthat truth'.\n281. Ofspecial applications, thefollowing maybenoticed:\na.Accordance, equality, likeness, and the like :thus, samdrnjyotih\nsuryena (AV.),'abrightness equal with thesun'; yeshdrh napddarajasd\ntulyah (MBh.),'tothedust ofwhose feet Iamnotequal'.\nb.Price(bywhich obtained): thus, dafdbhir krindti dhentibhih(RV.), 'he\nbuys with tenkine'; gavdm sahasrenavikrinite,'he sells forathousand\ncattle'.\nc.Medium, alsospaceordistance orroad, traversed; andhence alsotime\npassed through:thus,udnd nandvam anayanta (RV.), 'they led[him]asit\nwere ashipthrough thewater';e'hdydtam pathfbhir devaydnaih (RV.), 'come\nhither bygod-travelled paths'; jagmur vihdyasd (MBh.), 'theywent offthrough\nthe air': vidarbhdn ydtum ichdmy ekdhnd(MBh.),'Iwish togotoVidarbha\ninthecourse ofoneday'; kiyatd kdlena pradhdnatdm labhante(H.), 'inhow\nlongtime dothey obtain chief rank?' But theexpression of'within the\nspace of, fortime,isalsosometimes made bytheaccusative orlocative.\nd.The part ofthebody on(orby)which anythingisborne isusually\nexpressed bytheinstrumental :as,kukkurah skandheno 'hyate (H.),'adogis\ncarried ontheshoulder'; and this construction isextended tosuch cases as\ntulayd krtam(H.), 'puton(i.e.soastobecarried by)abalance'.\n282. The construction ofapassive verb(orparticiple) with an\ninstrumental oftheagent iscommon from theearliestperiod, andbecomes\ndecidedly more solater, thepassive participle with instrumental taking to\nnosmall extent theplace ofanactive verb with itssubject. Thus, yamena\ndattdh(RV.), 'given byTama'; ffibhir idyah (RV.),'tobepraised bysages';\nvyddhena jdlarh vistlrnam(H.), 'bythehunter anet[was] spread'; teno\nJ\nktam, 'byhim[itwas] said';mayd gantavyam (H.),= 'Ishallgo'.\nThesecond object ofacausative verb issometimes putintheinstru-\nmental instead oftheaccusative :as,paksibhih pinddn khddayati, 'hecauses\nthecakes tobeeaten bythe birds'.\n283. Many instrumental constructions aresuch ascallintranslation for\n"
+    },
+    {
+        "page": 105,
+        "content": "287]USES OFTHEINSTRUMENTAL ANDDATIVE. 85\notherprepositionsthan 'with' or'by'; yetthetrue instrumental relation is\nusually tobetraced, especiallyiftheetymological sense ofthewords be\ncarefully considered.\nMore anomalously, however, theinstrumental isused interchangeably\nwith the ablative with words signifying separation:thus,vatsafrvfyutah\n(RV.), 'separated from their calves'; md~'hamatmdna viradhisi (AV.),'let\nmenotbesevered from thebreath oflife'; sataya vyayujyata (MBh.j,'he\nwaspartedfrom her'(compare English parted with}.\n284. Theprepositions taking theinstrumental arethose signifying\n'with' andthelike :thus, saha, with theadverbial words containing saasan\nelement, assakam, sardham, saratham; and, ingeneral, awordcompounded\nwithsa,sam, saha takes aninstrumental asitsregular and natural com-\nplement.\nBut also thepreposition uma, 'without' (along with thecompounds of\nvispokenofinthepreceding paragraph),takes sometimes theinstrumental.\n285.Uses oftheDative. The dative isthecase of\ntheindirectobjectofthattoward orinthedirection ofor\ninorder toorforwhich anythingisorisdone(eitherintrans-\nitivelyortoadirectobject).\nInmore physical connections, theuses ofthedativeapproach those of\ntheaccusative(themore proper 'to'-case), andthetwoaresometimes inter-\nchangeable ;butthegeneral value ofthedative asthe 'for'-case isalmost\neverywhere distinctly tobetraced.\n286. Thus,thedative isused with\na.Words signifying give, share out, assign, andthelike :thus, yond\ndddati sdkhye (RV.), 'who gives not toafriend'; ydcha 'smai fdrma (RV.),\n'bestow uponhimprotection'.\nb.Words signifying show, announce, declare,aridthelike :thus, avtr\nebhyo abhavat suryah (RV.),'thesunwasmanifested tothem'; rtuparnam\nbhlmaya pratyavedayan (MBh.), 'they announcedRituparnatoBhima'; tebhyah\npratijnaya (MBh.), 'having promisedtothem'.\nc.Words signifying give attention, have aregard orfeeling, andthe\nlike; thus, nive$aya memo dadhuh(MBh.), 'they settheir minds upon en-\ncamping'; mate *vaputrebhyo mrda (AV.), 'begracious asamother toher\nsons'; kimasmdbhyamhrnlse(RV.), 'why artthou angry atus?'\nd.Words signifying inclination, obeisance, andthelike :thus,mdhyam\nnamantam pradfyac cdtasrah (RV.),'letthefour quarters bowthemselves to\nme'; devebhyo namaskrtya (MBh.), 'having paidhomage tothegods'.\ne.Words signifying hurling orcasting: asycna dudSfe dsyasi (AV.),\n'with which thou hurlest attheimpious'.\nInsome ofthese constructions thegenitive and locative arealsoused :\nseebelow.\n287. Initsmore distinctive sense, assignifying for,forthebenefitof,\nwith reference to,and thelike, thedative isused freely, and inagreat\nvariety ofconstructions. And thisusepasses over into that ofthedative of\n"
+    },
+    {
+        "page": 106,
+        "content": "ggIV.DECLENSION.[287\nend orpurpose,which isextremely common. Thus, isurh krnvana dsanaya\n(AV.), 'making anarrow forhurling'; grhndmitesaubhagatvdya hdstam(RV.),\n'Itake thyhand inorder tohappiness'; rdstrdya mdhyam badhyatdm sa-\npdtnebhyah pardbhuve (AV.),'be itbound oninorder toroyalty forme, in\norder todestruction formyenemies'. Such adative ismuch used predicat-\nively (andoftenest with thecopula omitted),inthesense of'makesfor',\nltends toward'; also 'isintended for,and so'must';or'isliableto',and so\n'can'. Thus, upadeco murkhdndm prakopdya nacdntaye (H.), 'good counsel\n[tends]totheexasperation,nottheconciliation, offools';sacatasydh sarh-\ntosdyana'bhavat (H.),'and hewasnottohersatisfaction'; sugopd asina\nddbhdya (RV.),thou artagoodherdsman, notone forcheating' (i.e.notto\nbecheated).\nThese uses ofthedative areintheolder language especially illustrated\nbythedative infinitives, forwhich seechap. XIII.\n288. The dative isnotused withprepositions.\n289.Uses oftheAblative. Theablative isthe'from'-\ncase,inthevarious senses ofthat preposition:itisused to\nexpress removal, separation, distinction, issue, andthelike.\n290. Theablative isusedwhere expulsion, removal, distinction, release,\nprotection,andother kindred relations areexpressed:thus,tesedhanti pathd\nvrkam(AV.), 'they drive away thewolffrom thepath';mdprdgdma pathdh\n(RV.), 'maywenotgoaway from thepath': areasmdd astu hetih(AV.),'far\nfrom usbeyour missile'; pdtdm novfkdt (RV.),'save usfrom thewolf.\n291. The ablative isusedwhere procedure orissue fromsomething as\nfrom asource orstarting-pointissignified:thus, cukrd krsnddajanista\n(RV.),'thebright onehasbeen born from theblackone'; lobhdt krodhah\nprabhavati (MBh.), 'passionarises fromgreed'; vatdt teprdndm avidam(AV.),\n'Ihavewon thy life-breath from thewind'; yeprdcyd diedabhiddsanty\nasmdn(AV.),'who attack usfrom theeasternquarter';tacchrutvd sakhlgan-\ndt(MBh.), 'having heard thatfrom thetroop offriends'; vdyur antariksdd\nabhdsata (MBh.),'thewind spoke from thesky'.\nHence also, procedureasfrom acause oroccasion issignified bythe\nablative :this isespecially frequent inthelater language, andintechnical\nphraseologyisastanding construction;itborders oninstrumental construc-\ntions. Thus, vdjrasya ciisndd daddra (RV.), 'from(byreasonof)thefury\nofthethunderbolt heburst asunder; yasya dandabhaydt sarve dharmam\nanurudhyanti (MBh.), 'from fear ofwhose rod allareconstant toduty'; afca-\nramicritatvdd ekarasya (Tribh.), 'because econtains anelement ofa'.\nPerhaps byafurther modification ofthisconstruction (theeffect following\nthe cause), theablative also intechnical language signifies 'after':thus,\nsakarat, 'after s\\\n292. One ortwospecial applications ofthe ablative construction are\ntobenoticed:\na.The ablative with words implying fear(terrifiedrecoil from): thus,\n"
+    },
+    {
+        "page": 107,
+        "content": "294=]USES OFTHEABLATIVE. 87\ntdsyd jdtdyah sdrvam abibhet (AV.), 'everything was afraid ofher ather\nbirth'; ydsmdd rejanta krstdyah (RV.),'atwhom mortals tremble1\n;yusmdd\nbhiyd (RV.), 'throughfear ofyou'.\nb.The ablative ofcomparison (distinction from): thus, prdririce divd\nmdrah prthivydh (RV.),'Indra isgreater than theheaven andthe earth'.\nWith acomparative (orother word used inakindred way), theablative is\ntheregular andalmost constant construction :thus, svaddh svddiyah (RV.),\n'sweeter thanthesweet';kirhtasmdd duhkhataram (MBh.), 'what ismorepainful\nthan that?\" mitrdd anyah (H.), 'any other than afriend'; ndisadhad anyam\n(MBh.), 'another than theNishadhan';samatto ma/tan,'heisgreater than 1';\ntadanydtratvdn nfdadhmasi (AV.),'we setthisdown elsewhere (away)\nfrom thee'; purva vfyvasmad bhtivanat (RV.),'earlier than allbeings'.\nOccasionally, apartitive genitiveisused with thecomparative (aswith\nthesuperlative);oraninstrumental(ofholding together thethings compared).\n293. The ablative isused with avariety ofprepositions andwords\nsharing aprepositional character; but allthese have rather anadverbial value,\nasstrengthening ordenning the 'from'-relation, thananyproper governing\nforce. Thus :\na.IntheVeda,ddhiandpdri aremuch used asdirecting andstrength-\nening adjuncts with theablative :as,jatohimdvatas pdri (AV.),'born from\ntheHimalaya (forth)'; samudrdd ddhi jajnise (AV.), 'thou artbom from\ntheocean'; cdrantam pdri tastMsah (RV.), 'moving forth from thatwhich\nstands fast'.\nb.Alsopurd (and purds], inthesense of'forward from1\n,andhence\n'before':as,purd jdrasah (RV.),'before oldage':andhencealso, withwords\nofprotection andthelike, 'from': asfafamdndh purd niddh(RV.), 'setting\ninsecurity from ill-will'.\nc.Alsod,inthesense of'hitherfrom,alltheway from':as,dm&lad\ndnufusyatu (AV.),'let itdrycompletely upfrom the root';tdsmdd dnadyb\nndma siha(AV.),'since that time yearecalled rivers'. Butusually, and\nespeciallyinthelaterlanguage,themeasurement ofinterval implied in d,\nisreversed indirection, andtheconstruction means 'alltheway to.until' :\nas,a'sydyajndsyo'drca/i(VS.;,'until theend ofthissacrifice'; dsodayat\n(M.),'tillthesixteenthyear';apradandt ((,.),'until hermarriage'.\nd.Otherprepositional constructions offer little subject forremark: vina\ntakes theablative aswell asinstrumental (284);rteintheolder language\nusually hastheablative,inthelater theaccusative.\n294.Uses oftheGenitive. Theproper value ofthe\ngenitiveisadjectival ;itbelongstoandqualifies anoun, de-\nsignating something relatingtothe latter inamanner which\nthenature ofthecase, ortheconnection, defines morenearly.\nOthergenitive constructions, withadjective orverb orprepo-\nsition, appeartoarise outofthisbyamore orlessdistinctly\ntraceable connection .\nTheuseofthegenitive hasbecome much extended, espe-\n"
+    },
+    {
+        "page": 108,
+        "content": "88IV.DECLENSION.[294\nciallyinthelater language, byattribution ofanoun-character\ntotheadjective,andbypregnantverbal construction, sothat it\noften bears theaspectofbeingasubstitute forother cases as\ndative, instrumental, locative.\n295. The genitiveinitsnormal adjective construction isclassifiable\ninto theusual varieties,as :genitive ofpossession orappurtenance, including\nthecomplementofimpliedrelation this is,aselsewhere, thecommonest\nofall;theso-called partitive genitive;thesubjective andobjective genitives;\nand soon. Genitives ofappellation ('cityofRome'), material, andcharac-\nteristic ('man ofhonor'), donot occur. Examples are: indrasya vajrah,\n'India's thunderbolt'; pitaputranam,'father ofsons'; putrdh pituh, 'son of\nthefather' ;pituh kamah putrasya,'the father's love oftheson';kenah,\n'which ofus'; fatarh dasmam,'ahundred female slaves'.\nTheexpressionofpossessionetc. onthepartofpronounsismade\nalmost entirely bythegenitivecase, andnotbyaderived possessive ad-\njective (516).\n296. The genitiveisdependent onanadjective:\na.Inpart, byaconstruction similar tothat ofverbs which take a\ngenitive object: thus, abhijna rajadharmanam. 'understanding theduties of\naking'.\nb.Ingreat part, byatransfer ofthepossessive genitive fromnoun to\nadjective, theadjective being treated asifithadnoun-value: thus, tasya\nsamah oranurupah orsadrfah, 'resembling him'(i.e.hislike); tasya priya,\n'dear tohim'(hisdearone); tasya 'viditam, 'unknown tohim'(hisunknown\nthing); havya? carsariindm (RV.),'tobesacrificed tobymortals'(their object\nofsacrifice); Ipsito naranarinam(MBh.), 'desired ofmenandwomen'(their\nobject ofdesire); yasya kasya prasutah (H.),'ofwhomsoever born'(his son).\nc.The so-calledpartitive genitive with asuperlative, oranother word\nofsimilarvalue,isamatter ofcourse:thus, frestham viranam,\"best of\nheroes'; vlriidhaihvlryavatl (AV.),'ofplants themighty (mightiest)one'.\nd.Adjectives meaning 'capable', 'worthy', 'full', andafew others, take\nthegenitive byamoreoriginal andproper right.\n297. The genitive asobject ofaverb is:\na.Apossessive genitive oftherecipient, bypregnant construction, with\nverbssignifying give, impart, communicate, andthe like:thus, varan pra-\ndaya 'sya (MBh.), 'having bestowedgifts upon him' (made them hisby\nbestowal); rajfto niveditam(H.),'itwasmadeknown totheking' (madehis\nbyknowledge).\nThisconstruction, bywhich thegenitive becomes substitute foradative\norlocative, abounds inthe laterlanguage, and isextended sometimes to\nproblematic and difficult cases.\nb.A(inmostcases, probably) partitive genitive,asalesscomplete or\nless absolute object than anaccusative :thus, with verbs meaning partake\n(eat, drink, etc.), aspiba sutdsya (AV.), 'drink(of)theSoma'; mddhvah\npayaya (RV.), 'cause todrink thesweetdraught'; with verbs meaning\n"
+    },
+    {
+        "page": 109,
+        "content": "301]USES OFTHEGENITIVE. 89\nimpart (ofthething imparted) etc., asd'idata noamftasya (RV.), 'bestow\nupon usimmortality'; with verbs meaning enjoy, besatisfied orfilled\nwith: asmdtsy dndhasah (RV.), 'dothou enjoy the juice';--with verbs\nmeaning toperceive, note, carefor,regard with feelingsofvarious kinds :\nas,vdsisthasyastuvatd indro a$rot (RV.), Indra listened toVasishtha whowas\npraising him'; ydthamama smdrat(AV.),'that hemaythink ofme'; tasya\ncukopa (MBh.), 'hewasangry athim'; bibhlmas tava(MBh.),'weareafraid\nofthee'.\nc.Agenitive ofmore doubtful character, with verbs meaning torule\norhaveauthority, astvdm i$isevdsunam(RV.), 'thou art lord ofgood\nthings'; ydtha 'hamesam virdjani (AV.),'that Imay rule over them';\nwith verbs meaning throw at,injure, asydstadsyat (AV.), 'whoever hurl-\nedatthee'; andwithsome others.\n298.Agenitive initsusual possessive sense isoften found aspredi-\ncate, and notseldom with thecopula omitted :thus, ydthd'somama keva-\nlah(AV.), 'that thou mayest bewholly mine'; sarvah sampattayas tasya\nsamtustam yasya manasam(H.),'allgood fortunes arehiswhohasacon-\ntented mind'.\n299. Theprepositional constructions ofthegenitive areforthemost\npartwith suchprepositions asarereally noun-cases, andhave thegovern-\nment ofsuch :thus, agre, arthe, tote, andthelike\n;also with other prepo-\nsitional words which, inthegeneral looseness ofuseofthegenitive, have\nbecome assimilated tothese. Afewmore realprepositionstake thegen-\nitive :eitherusually, likeupdri, 'above',oroccasionally,likeadhds, antdr, dtL\nAgenitive isused intheolder language with certain adverbs oftime :\nthus, sakfd dhnah(RV.), 'once aday1\n;trihsamvatsarasya,'thrice ayear';\nidanim dhnah(RV.),'atthistime ofthe day'.\n300. Thegenitiveisused adverbially hardlyatall; afewgenitives of\ntime occur intheolder language: asaktos, 'bynight', vastos, 'byday'.\nAgenitive ofaccompanying circumstance, with aqualifying word,is\nsometimes usedabsolutely, instead ofalocative (303 b);but thisconstruction\nisunknown inthe earlier language, and rare inthe later. Itissaidby\nthegrammarians toconvey animplication ofcontempt: thus, vaidarbhyaJi\npreksamanayah panakalam amanyata (MBh.), 'hethoughtitatime forstak-\ning atplay, while theVidarbhan waslooking on'(or,inspiteofit) ;but it\nisfound without anysuch implication: thus, samkhyasyami phalany asya\npavyataste(MBh.),'Iwillcount itsfruits while youlookon1\n.\n301.Uses oftheLocative. The locative isproperly\nthe''irT-case, thecaseexpressingsituation orlocation;but its\nsphere ofusehasbeensomewhat extended, soastotouch and\noverlaptheboundaries ofother cases, forwhich itseems tobe\nasubstitute.\nUnimportant variations ofthesense of 'in' arethose of\n'amid' or'among', 'on,and 'at\\Ofcourse, also, situation in\ntime aswell asplaceisindicated bythecase;and itisapplied\n"
+    },
+    {
+        "page": 110,
+        "content": "90IV.DECLENSION.[301\ntoyetless physical relations, tosphereofaction andfeeling\nandknowledge,tostate ofthings,toaccompanyingcircumstance\n;\nandoutofthis lastgrowsthefrequentuseofthelocative as\nthecase absolute.\nMoreover, byapregnant construction, thelocative isused\ntodenote theplaceofrest orcessation ofaction ormotion\n('into'or'on to'instead of'in'or'on';German inwith accu-\nsative instead ofdative :compare Englishthere forthither],\n302. The locative ofsituation inspace hardly needs illustration. An\nexampleortwoare :ye,devd divtsthd (AV.),'which ofyougods areinheav-\nen';nadevesu nayaksesutddrk (MBh.),'notamong gods orYakshas issuch\naone'; pdrvatasya pfsthe (RV.),'ontheridge ofthemountain'; viddthe\nsantu devdh(RV.), 'may thegodsbeattheassembly'; dacame pade (MBh.),\n'atthetenthstep'.\nThe locative oftime indicates thepointoftime atwhich anything\ntakes place:thus, asydusdso vyitstdu (RV.),'attheshining forth ofthis\ndawn'; etasminn evakale(MBh.),'atjust thattime'; dvddace varshe (MBh.),\n'inthetenth year'. But theaccusative isoccasionally used inthis sense,\ninstead ofthelocative.\n303. The locative ofsphereorcondition orcircumstance isofvery\nfrequentuse: thus, made dhim ind.ro jaghdna (RV.),'infury Indra slew\nthe dragon'; mitrdsya sumatdu sydma (RV.), 'maywebeinthefavor of\nMitra';tevacane ratam(MBh.), 'delighted inthywords'.\na.This constructionis,ontheonehand, generalized intoanexpres-\nsion for'inthematter orcaseof, or'with reference to,respecting', and\ntakes inthelater languageavery wide range, touching upon genitive and\ndative constructions :thus,eJmdm bhaja grame devesu gosu (AV.),'begen-\nerous tohim inretainers,inhorses,incattle'; tdm itsakhitvd Imahe(RV.),\n'himwebeg forfriendship'; nydyolyammayd drsta dnayane tava(MBh.),\n'thismeans wasdevised bymefor(with referenceto)bringing theehither';\nsatltve kdranam striydh (H.),'thecause of(inthecaseof)awoman's chast-\nity';nacakto 'bhavan nivdrane (MBh.), 'hewasnotcapableofpreventing'.\nb.Ontheother hand, theexpression bythelocative ofacondition of\nthings inwhich anything takesplace,orofaconditioning oraccompanying\ncircumstance, passes over into awell-marked absoluteconstruction, which is\nknown even inthe earliest stage ofthelanguage, butbecomes more fre-\nquentlater.\nTransitional examples are :have tvdsura udite havemadhydndine divdh\n(RV.),'Icall tothee atthearisen sun(when thesunhasrisen),Icallat\nmidtime oftheday'; aparddhe krte'picanamekopah (MBh.), 'andeven\nincase ofanoffensecommitted, there isnoanger onmypart'.\nThenormal condition oftheabsolute construction iswith aparticiple\naccompanying thenoun :thus, stlrne barhisi samidhdne agndu (RV.), 'when\nthebarhis isstrewn andthe firekindled'; kalefubhe prdpte (MBh.),'apro-\npitious timehaving arrived'; avasanndydrh rdtrdv astdcalacuddvalambini can-\n"
+    },
+    {
+        "page": 111,
+        "content": "306]USES OFTHELOCATIVE. 91\ndramasi (H.),'thenight having drawn toaclose, andthemoon resting on\nthesummit ofthewestern mountain'.\nButthenounmaybewanting,ormaybereplaced byanadverbial sub-\nstitute(aseuam, tathd, iti): thus, asmdbhih samanujndte (MBh.), '[it]being\nfully assented tobyus';evam ukte kalind(MBh.),'itbeing thus spoken by\nKali';tatha 'nusthite(H.),'itbeing thusaccomplished'. Solikewise the\nparticiple maybewanting (acopulasati orthelikebeing tobesupplied):\nthus, dure bhaye,'thecause offearbeing remote'; while,ontheother hand,\ntheparticiplesati etc. issometimes redundantly added totheother parti-\nciple:thus, tatha krtesati,'itbeing thus done'.\nC.The locative isfrequently used adverbially orprepositionally:thus,\narthe orkrte,'inthematter of,forthesake of.\n304. Thepregnant construction bywhich thelocative comes toexpress\nthegoal orobject ofmotion oraction orfeeling exercised isnotuncommon\nfrom the earliest time. Itisbynomeans tobesharply distinguished from\ntheordinary construction\n;thetwo passintooneanother,with adoubtful\nterritory between. Itoccurs :\na.Especially with verbs,asofarriving, sending, placing, communicating,\nbestowing, andmany others, insituations where anaccusative oradative\n(oragenitive, 297a) might belookedfor,andexchangeable with them:\nthus, sdiddevesu gachati (RV.), 'that, truly, goes to(tobeamong)the\ngods' ;imam noyajndm amftesu dhehi(RV.),'set thisoffering ofoursamong\ntheimmortals'; yddsincdnti rdsam osadhisu (AV), 'who pourinthejuice\nintotheplants' (or,thejuice that isintheplants); maprayache \"$vare dhanam\n(H.),'donotofferwealth toalord'; papdta medinydm (MBh.),'hefell to(so\nastobeupon) theearth'; skandhe krtva(H.), 'putting ontheshoulder';\nsamfrutya purvam asmdsu (MBh.), 'having before promised us1\n.\nb.Often alsowithnouns and adjectivesinsimilar constructions(the\ninstances notalways easy toseparate from those ofthe locative meaning\n'with reference to': above, 303 a): thus, dayd sarvabhutesu, 'compassion\ntoward allcreatures';anurdgam ndisadhe (MBh.),'affection fortheNishadh-\nan';rdjdsamyagvrttah saddtvayi (MBh.),'theking always behaved properly\ntoward thee'.\n305. Theprepositions construed with thelocative stand toitonly in\ntherelation ofadverbial elements strengthening anddirecting itsmeaning.\nIntheVeda, such locative constructions aremost frequent withdand\nddhi: thus, mdrtyesv d,'among mortals'; prthivydm ddhy osadhih,'the\nplants upon the earth';tejomdyi dhdrayd'dhi(AV.),'establish glory in\nme': less often, upaanddpiareused inthesame way. Inallages of\nthelanguage, antdr, 'within, among',isconstrued with thelocative.\n306. Declensional forms aremade bytheaddition of\nendingstothestem,orbase ofinflection.\nThestemitself, however,inmanywords and classes\n"
+    },
+    {
+        "page": 112,
+        "content": "92IV.DECLENSION.[306\nofwords,isliable tovariation,especiallyasassuminga\nstrongerform insome cases andaweaker inothers.\nAnd between stemandendingaresometimes inserted\nconnectingelements (orwhat, intherecorded condition of\nthelanguage,have theaspectofbeing such).\nRespectingallthese points,the details oftreatment, asexhibited by\neach class ofwords orbysingle words,willbegiven inthefollowing chapters.\nHere, however,itisdesirable also togive abrief general view ofthem.\n307.Endings: Singular.Inthenominative, the\nusual masc. and fern, endingis*which, however,iswant-\ninginderivative aand -stems\n;itisalsoeuphonicallylost\n(150) byconsonant-stems. Neuters ingeneral havenoending,\nbutshow inthis case thebarestem\n;a-stems alone addm(as\nintheaccus. masc. IAmongthepronouns, am isafrequent\nmasc. and 1'em.nom. ending (and isfound even indu.andpi.);\nandneuters show aform ind.\nIntheaccusative, moram isthemasc. and fern,end-\ning ambeing added after aconsonant andrand after iand\nMintheradical division, andmelsewhere after vowels. The\nneuter accusative islike thenominative.\nTheinstrumental endingfor allgendersalike isa.\nWith final iandu-vowels, theaisvariously combined, and in\ntheolder languageitissometimes lostbycontraction with them.\nStems inamake thecase end inena(sometimeseriainV.),\nandthose inamake itend inaya; butinstances occur, inthe\nearly language,ofimmediate addition ofatoboth aand a.\nThedativeendingisingeneral e;andwith itlikewise\nthemodes ofcombination ofiandufinal arevarious(anddis-\nappearance bycontraction notunknown intheoldestlanguage).\nThe a-stems arequite irregularinthis case, makingitendin\nayaexceptedisthepronominal elementsnia,which combines\n(apparently]with etosmai. Inthepersonal pronounsisfound\nbhyam (orhyam}.Afullerending ai(like gen.-abl.asand loc.am: see\nbelow) belongstofeminine stemsonly.Itistaken(withinter-\nposed y]bythegreat class ofthose inderivative a:alsoby\nthose inderivativei,and (asreckoned inthelaterlanguage)\ninderivative u.And later itisallowed tobetaken byfemi-\nnine stems inradical iandM,andeven bythose iniandu:\nsuch have itinthe earliestlanguageinonlyrareandexcep-\ntional instances.\nTheablative has aspecial ending, d(or #),onlyin\n"
+    },
+    {
+        "page": 113,
+        "content": "308]CASE-ENDINGS. 93\na-stems, masc. and neut., theabeing lengthened before it\n(exceptinthepersonal pronouns of1stand2dperson; and\nthese have thesame ending inthepi.,andeven, intheold\nlanguage,inthedual). Everywhere else, theablative isident-\nicalwith thegenitive.\nThegenitive of-stems(andofonepronominal w-stem,\namu]adds sya. Elsewhere, theusualabl.-gen. endingisas:\nbut itsirregularities oftreatment incombination \"with astem-\nfinal areconsiderable. With iand w,itiseitherdirectly added\n(onlyintheoldlanguage), added withinterposed n,orfused toes\nand osrespectively. With r(orar)ityieldsus(orur :169, end'.\nThe fuller asistaken byfeminine stemspreciselyasai\njstaken inthedative :seeabove.\nThelocative endingisiinconsonant and randa-stems\n(fusingwith atoeinthelatter). The iandw-stems(unlessthe\nfinalvowel issaved byaninterposed n]make thecase end in\nau;buttheVeda hassome relics ortraces oftheolder forms\n(ay-i [?]and av-i] outofwhich thisappearstohave sprung.\nVedic locatives from /-stems end also inaand i.Thepronom-\ninalelement smamakes thelocative smin. Stems inaninthe\nolder language often losethe*,andusethebarestem aslocative.\nTheending am isthelocativecorrespondenttodat. aiand\nabl.-gen. as,and istaken under thesame circumstances :seeabove.\nThevocative(unless byaccent:314)isdistinguished\nfrom thenominativeonlyinthesingular, andnotquite always\nthere. Ina-stems,itistheunaltered stem;and soalso in\nmost consonant-stems :but neuters inanand inmay drop\nthen;and the oldestlanguagehassometimes avocative ins\nfrom stems inntand ns.Stems inrchangethis toar.In\nmasc. and fern, iandw-stems, thecaseendsrespectivelyineand\no;inneuters, inthesame oriniand u.Stems inachange\natoe;derivative 1anduareshortened;radical stems inlong\nvowels usethenominative form.\n308. Dual. Thedual has exceptsofarasthevoca-\ntive issometimesdistinguished fromnom. andaccus. byadif-\nference ofaccent: 314onlythree case-forms :onefornom.,\naccus., and voc.; one forinstr., dat., andabl.; andone for\ngen.and loc.(occasional confusion oftheuses ofthesecond and\nthird isseenearlier).\nBut thepronouns of1stand2dpersonintheolder lan-\nguage distinguishfivedual cases :see492.\nThemasc. and fern,endingfornom.-accu s.-voc. isin\nthe laterlanguage usually au; but instead ofthis theVeda\nhasprevailinglya.Stems inamake thecaseendine.Stems ini\nandu,masc. andfern., lengthen those vowels; andderivative iin\n"
+    },
+    {
+        "page": 114,
+        "content": "94IV.DECLENSION.[308\ntheVeda remains regularly unchanged, thoughlater itadds au,\nTheneuter endingisonly *,with final athiscombines toe.\nThe universal endingfortheinstr. -dat. -abl. isbhyam,\nbefore which final aismadelong. IntheVeda,itisoften to\nberead astwosyllables,bhiam.\nThe universal endingofgen.-loc.isos:before this, a\nandaalikebecome e(ai).\n309. Plural. Inthenominative, thegeneral masc.\nand fern, endingisas.The oldlanguage, however, oftenmakes\nthecase inasasinstead ofasfrom a-stems, andinafewexamples\nalsofrom a-stems. From derivative ?-stems,isinstead ofyas\nistheregular andusual Vedic form. Pronominal a-stems make\nthemasc. nom. ine.\nTheneuter ending (whichisaccusativealso)isingenerali;\nandbefore thisthe final ofastem isapttobestrengthened, by\nprolongationofavowel, orbyinsertion ofanasal, orbyboth.\nButintheVeda thehence resulting forms inani, mi,uniare\nfrequentlyabbreviated byloss ofthem,and sometimes by\nfurther shorteningoftheprecedingvowel.\nTheaccusative endingisalso asinconsonant-stems and\nintheradical division of iandw-stems(andintheoldlan-\nguage evenelsewhere). Stems inshort vowels lengthen those\nvowels andaddinthemasculine n(for ns,ofwhich abundant\ntracesremain),and inthefeminine s.Intheneuter, this case\nislike thenominative.\nIntheinstrumental, thecase-endingiseverywherebhis\nexceptina-stems, where inthelater languagethecase always\nends inais,butinthe earlier either inaisorthemore reg-\nular ebhis(abhisinthetwopersonal pronouns;andtheprono-\nminal stem a[501] makes ebhisonly).\nThedative andablative have inthepluralthesame\nform, with theending bhyas (inVeda oftenbhias], before which\nonly aisaltered, becominge.But thetwopersonal pronouns\ndistinguish thetwocases, havingfortheablative thesingular ending\n(asabovepointed out),andforthedative thepeculiar bhyam (almost\nnever inV.bhiam), whichthey extend also into thesingular.\nOfthegenitive, theuniversal endingisam; which\n(except optionallyafter radical Iand M,and inafew scatter-\ningVedicinstances)takes after final vowels aninserted conso-\nnant, sinthepronominal declension, nelsewhere;beforeri,a\nshort vowel islengthened ;befores,abecomes e.IntheVeda,\nitisvery frequentlytobepronouncedintwosyllables,asa-am.\nThelocativeendingissu,without anyexceptions, and\ntheonlychange before itisthat ofatoe.\n"
+    },
+    {
+        "page": 115,
+        "content": "311] CASE-ENDINGS; STRONG ANDWEAK STEM. 95\nThevocative, asinthedual, differs from thenomina-\ntiveonlybyitsaccent.\n310.Thenormal scheme ofendings,asrecognized by\nthenative grammarians (and convenientlytobeassumed as\nthebasis ofspecial descriptions),isthis :\n"
+    },
+    {
+        "page": 116,
+        "content": "95IV.DECLENSION.[311\nhaving endings beginningwith avowel(instr.toloc.sing.,\ngen.-loc.du., gen. pi.);the latter, with aconsonant(inst.-\ndat.-abl. du., instr., dat.-abl., and loc.pi.).\nThe class ofstrong cases, asabove denned, belongs\nonlytomasculine andfeminine stems. Inneuterinflection,\ntheonly strongcases arethenom.-acc.pi.;while,inthose\nstems thatmake adistinction ofweakest andmiddle form,\nthenom.-acc. du.belongtotheweakestclass, andthesame\ncases sing,tothemiddle :thus,forexample, compare SJHlRl\npratyafte-i,nom.-acc.pi.neut., and3rU ^iH^pralyanc-as>\nnom.pi.masc.; yrfWl prafic-1,nom.-acc. du. neut., and\nSfcfNrTCT pratw-os, gen.-loc.du.;HC^I pratyak,nom.-acc.\nsing, neut., andHrUN^^pratyag- bhis, instr.pi.\nEven inwords which exhibit novariation ofstem,itisoften con-\nvenient todistinguish thesame groups ofcases bythenames strong and\nweak and soon.\n312. Other variations concern chiefly the finalvowel ofastem, andmay\nbemainly left tobepointed outindetail below. Ofconsequence enough\ntomention here isonly thepuna-strengthening ofafinal ioru,which in\nthelater languageisalways made before asofnom.pi.and eofdat. sing.\ninrnasc. andfern.; intheVeda,itdoes notalways takeplace; nor isit\nforbidden indat.sing. neut.also; and itisseen sometimes inloc. sing.\nFinal rhasywraa-strengtheninginloc.sing.\n313.Insertions between Stem andEnding. After vowel-stems,\nanadded noften makes itsappearance before anending. Thisappendage\nisofleast questionable origin innom.-acc.pi.neut., where theinterchange\nintheoldlanguage oftheforms ofaandi-stems with those ofanand in-\nstems ispretty complete ;andthew-stems follow their analogy. Elsewhere,\nitismost widely andfirmly established inthegen. pi.,where inthegreat\nmass ofcases, andfrom theearliest period, theendingisvirtually nam after\navowel. Inthe iandM-stems ofthe later language,the instr.sing, of\nmasc. andneut. isseparated byitspresence from thefern., and itisinthe\nother weakest cases made ausual distinction ofneuter forms from mas-\nculine; buttheaspect ofthematter intheVeda isvery different; there\ntheappearance oftheniseverywhere sporadic ;theneuter shows nospecial\ninclination totakeit,and itisnotexcluded even from thefeminine. Inthe.\nending enafrom u-stems(later invariable,earlier predominating),itspresence\nappears tohave worked themost considerable transformation oforiginal shape.\nTheplace ofnbefore gen. pi.amistaken bysinpronominal aanda-stems.\nTheyafter abefore theendings di,as,anddm ismost probably an\ninsertion, such asismade elsewhere(258).\n"
+    },
+    {
+        "page": 117,
+        "content": "316]ACCENT INDECLENSION. 97\nAccent inDeclension.\n314. Asarulewithoutexception,thevocative,ifaccented\natall,isaccented onthe firstsyllable.\nAnd intheVeda(thecase isarareone), whenever asyllable written\nasone istobepronouncedastwobyrestoration ofasemivowel tovowel\nform, the firstelement only hasthevocative accent, and the syllable as\nwritten iscircumflex (by84a):thus, dyaiis (i.e.dfihis] whendissyllabic,\nbutdyatis when monosyllabic ;jykke when forjfake.\nButthevocative isaccentedonlywhen itstands atthe\nbeginningofasentence or,inverse, atthebeginningalso\nofametrical division orpada;elsewhere itisaccentless or\nenclitic :thus, dgneyamyajn&m paribhurdsi(RV.),'OAgni!\nwhatever offering thouprotectest1\n;butupatva'gnae'masi(RV.),\n'unto thee, Agni, wecome'.\nAword qualifyingavocative usually anadjective, butnotseldom alsoa\nnoun inthegenitive (very rarely inanyothercase) constitutes,sofaras\naccent isconcerned,aunity with it :thus, sdkhe vaso orvdsosakhe,'excellent\nfriend'; sUno sahasah orsdhasdh suno, 'ohsonofmight'; andsuditi suno\nsahaso didihi(RV.),'with excellent brightness, sonofmight, shine forth'.\nTwocoordinate vocatives, whether noun oradjective, have usually thesame\naccent;buttheVedic texts furnish notafewirregular exceptionstothis rule.\nForbrevity, thevocative dualandpluralwillbegiven intheparadigms\nbelow along with thenominative, without takingthetrouble tospecify in\neach instancethat,ifthe latter beaccented elsewhere than onthe first\nsyllable,theaccent ofthevocative isdifferent.\n315.Asregardstheother cases, rules forchangeofaccent\nindeclension have todoonly withmonosyllables andwithstems\nofmore thanonesyllable which areaccented onthefinal\n;forifa\nstem beaccented onthepenult, oranyothersyllablefurther\nback asinsdrpant, vari, bhagavant, sumdnas, sa/idsravaja\ntheaccent remains uponthatsyllable throughthewhole inflection\n(exceptinthevocative, asexplainedinthepreceding paragraph).\nThe only exceptions areafewnumeral stems :seebelow, chap. VI.\n316. Stems accented onthe final(including monosyllables)\naresubjecttovariation ofaccent indeclensionchieflyinvirtue\nofthefact thatsome oftheendings have, while others have\nnot, orhave inlessdegree, atendencytodraw theaccent for-\nwardupon themselves. Thus :\na.Theendings ofthenominative andaccusative singular anddualand\nofthenominative plural have notendency totake theaccent away from the\nstem, and aretherefore only accented when afinal vowel ofthestem and\nthevowel oftheending areblended together into asingle vowel ordiphthong.\nThus, from dattd come dattm't (=dattd-f-au)and dnttas(=datta-\\-as);\nbutfrom nadtcome nadydu (=nadi-\\-au) andnadytt* (==nadt-}-aa).\nWhitney, Grammar. 7\n"
+    },
+    {
+        "page": 118,
+        "content": "98IV.DECLENSION.[316\nb.Alltheother endings sometimes take theaccent; butthose beginning\nwith avowel dosomore readily than those beginning with aconsonant.\nThus, fromnanscome navti andnaubhis; frommahant, however, comemahaid\nbutmahddbhis.\nThegeneralrules ofaccent, then,maybethus stated :\n317. Inthedeclension ofmonosyllabic stems, theaccent\nfallsupontheendinginalltheweak cases(withoutdistinction\nofmiddle andweakest):thus, nava, naubhyhm, navam, nausu;\nvaci, vagbhis, vactim, vaksu.\nButsome monosyllabic stems retain theaccent throughout:thus, ptfbhis,\ngdvam, g6su. Forsuchcases,seebelow, 350,361 c,d,375,390,427.\n318. Ofpolysyllables endinginconsonants, onlyafew\nshift theaccent totheending, and that intheweakest(notthe\nmiddle)cases. Such are :\na.Present participlesinantorill:thus, fromtuddnt, tudatd and tudattis\nandtudatam; buttudddbhydm andtudatsu.\nb.Afew adjectives having theform ofsuchparticiples, asmahatd,\nbrhatds.\nc.Bases ofwhich theaccented final loses itssyllabic character bysyn-\ncopationofthevowel :thus, majjnd, murdhne, ddmnd*(frommajydn etc. :423 .\nOther sporadiccases willbenoticed under thedifferent declensions.\nCase forms used adverbially sometimes show achanged accent: see\nchap. XVI. (lllOff.).\n319. Ofpolysyllabic stems endinginaccented short vowels,\nthe final ofthestem retains theaccent ifitretains itssyllabic\nidentity:thus, dattena anddatthya from dattd;agnina andagndye\nfromagni; and alsodattebhyas, agnibhis, and soon. Otherwise,\ntheaccent isontheending:and that, whether the final and\ntheendingarecombined into one, asindattais, dhenau, agnln,\ndhenus, and soon\n;orwhether thefinal ischangedintoasemi-\nvowel before theending:thus, dhenvu, pitrh.\nButamofthegen. pi.from stems in/and tiand rmay, and inthe\nolder language always does, take theaccent, though separated bynfrom the\nstem: thus, agnlnnm, dheniindm, pitfndm.InRV., even derivative i-stems\nshow usually thesame shift :thus, bahvindm. Ofstems ina,onlynumerals\n(chap. VI.)follow this rule:thus, saptandm, dafdndm.\n320. Root-words inIanduasfinalmembers ofcompoundsretain the\naccentthroughout, notshiftingittoanyoftheendings. And intheolder\nlanguage there arepolysyllabic words inlongfinal vowels which follow in\nthisrespectasinothers theanalogy oftheroot-declension (below, 355ff.).\nApart fromthese, thetreatment ofstems inderivative long vowelsis,as\nregards accent, thesame asofthose inshort vowels save thatthetone\nisnotthrown forward upon theending ingen. plural.\n"
+    },
+    {
+        "page": 119,
+        "content": "323] 99\nCHAPTER V.\nNOUNS ANDADJECTIVES.\n321.THEaccordance ininflection ofsubstantive and\nadjectivestems issocompletethat thetwocannot besep-\narated intreatment from oneanother.\nTheymaybeclassified,forconvenience ofdescription,\nasfollows:\nI.Stems in fa;\nII.Stems in^iand3u;\nIII.Stems inETTa, *>and \"37u:namely,A.radical-\nstems(andafewothers inflected likethem); B.derivative\nstems;\nIV.Stems inftr(orq[or);\nV.Stems inconsonants.\nThere isnothing absolute inthis classification andarrangement ;itis\nmerely believed tobeopentoasfew objections asany other. Nogeneral\nagreement hasbeen reached among scholars astothenumber and order\nofSanskrit declensions. Thestems inaarehere treated firstbecause of\nthegreat predominanceofthe class.\n322. The division-line between substantive andadjective,\nalwaysanuncertain one inearly Indo-European language,is\neven more waveringinSanskrit than elsewhere. There are,\nhowever,inallthedeclensions asdivided above unless we\nexceptthestems inrorar- -words which aredistinctlyad-\njectives ;and, ingeneral, theyareinflectedpreciselylikenoun-\nstems ofthesame final :only,among consonant-stems, there are\ncertain sub-classes ofadjectivebases withpeculiaritiesofin-\nflection towhich there isamong nouns nothing corresponding.\nButthere are alsotwo considerable classes ofadjective-com-\npounds, requiring specialnotice;namely:\n323.Compound adjectives havingasfinalmember abare\nverbal root, with thevalue ofapresent participle:thus, su-dfc,\n'well-looking'; pra-budh, 'foreknowing ;a-druh, 'not hating'; veda-\nvid, 'Veda-knowing'; vrtm-han, 'Vritra-slaying'; upastha-sdd,\n'sittinginthelap'. Everyroot isliable tobeused inthis\n"
+    },
+    {
+        "page": 120,
+        "content": "'\n^00V.NOUNS ANDADJECTIVES.[323\nway,andsuchcompoundsarenotinfrequentinallages ofthe\nlanguage:seechapteronCompounds,below(XVIII.).\nThis class isoriginally and essentially only aspecialclass ofcompound\nadjectives,since intheearliest Veda thesimpleaswell asthecompounded\nrootwassometimes used adjectively. Butthecompoundedrootwasfrom the\nbeginning much more often soused, and the later themore exclusively,\nsothatpracticallytheclass isaseparateandimportantone.\n324.Compound adjectives havinganoun asfinalmember,\nbutobtaining anadjectivesense secondarily, byhavingtheidea\nof'possession' added, andbeinginflected asadjectivesinthe\nthree genders. Thus, prajakama,'desire ofprogeny', becomes\nanadjective meaning'desirous(i.e.having desire)ofprogeny';\nsabharya (sa-\\-b/iarya), 'havingone's wifealong'; and soon.\nInafewcases, also, the final noun issyntactically object ofthepre-\nceding member: thus, atimatra, 'immoderate' (atimatram, 'beyond measure');\nyavayaddvesas, 'driving away enemies'.\n325. Hence, under each declension, wehave tonotice how\naroot oranoun-stem ofthatdeclension isinflected when final\nmember ofanadjective compound.\nAstoaccent,itneedsonlytoberemarked that amono-\nsyllabic word ending acompoundloses thepeculiarityofmo-\nnosyllabic accentuation, anddoes notthrow thetonefoward upon\ntheending.\nDeclension I.\nSterns (masculine andneuter) ina.\n326. This declension contains themajorityofallthe\ndeclined stems ofthelanguage.\nItsendings deviate morewidelythan anyothers from\nthenormal.\n327.Endings: Singular. Thenom. masc. hasthenormal end-\ning.<t.\nThe ace. (masc. andneut.) addsm(notam); and thisform hasthe\noffice also ofnom, neuter.\nThe instr. changes atoenauniformly inthelater language; andeven\nintheoldest Vedic this isthepredominant ending (inRV., eight ninths of\nallcases).Itsfinal isinVedic verse notinfrequently made long (ena), where\nfavored bythemetre. But thenormal ending a thus, yajnd, suhdva,\nmahitvd(foryajnena etc.)isalsonotrare intheVeda.\nThe dat. hasaya (asifbyadding ayatoa),alike inallages ofthe\nlanguage.\nThe abl.has t(ormoreprobably d: itisimpossible from theevidence\n"
+    },
+    {
+        "page": 121,
+        "content": "329]DECLENSION I.,^-STEMS. ENDINGS. 101\noftheSanskrit totellwhich istheoriginal form oftheending), before which\naismade long:thisendingisfound innoother noun declension, butonly\ninthepersonal pronouns (ofallnumbers).\nThe gen.hassyaadded tothe final a;and thisendingisalsolimited\ntoa-stems (with thesingle exception ofthepronoun amusya:chap. VII.).Its\nfinal aisinonly three casesmade long intheVeda; and itsyisvocalized\n(asia) almost asrarely.\nThe loc.ends ine(asifbycombining thenormal endingiwith the\nfinal ofthestem), withoutexception.\nThe voc. isthebare stem.\n328. Dual. Thedual endingsingeneral arethenormal ones.\nThenom., ace.,arid voc.masc. end inthelater language always inau.\nIntheVeda, however, theusual endingissimple a(inRV.,inseven\neighthsofthe occurrences). Thesame cases intheneut. end ine,which\nappearstobetheresult offusion ofthestem-final with thenormal endingi.\nThe instr., dat., and abl.havebhyam (inonly one ortwoVedic in-\nstances resolved into6/u'am), with thestem-final lengthenedtoabefore it.\nThegen.and loc.have ayinserted after thestem-final before os(or\nasiftheahadbeen changedtoe}.Inone ortwo(doubtful) Vedic in-\nstances(asalso inthepronominal formsmosandyos),osissubstituted for\nthe final a.\n329. Plural. Thenom. masc. has inthelater language thenormal\nending ascombined with the final atoas.But intheVeda theending\nasas instead isfrequent (one third oftheoccurrences inRV., butonlyone\ntwenty-fifth inthepeculiar parts ofAV.).\nThe ace.masc. ends inan(forearlierans,ofwhich abundant traces\nare left intheVeda, arid, under thedisguise ofapparent euphonic com-\nbination, even inthelater language: seeabove, 208ff.).\nThenom. and ace. neut. have inthelater language\" always theending\nani(like thean-stems: see421;orelsewithnasinthegen. pi.before\nnormali).But intheVeda thisending alternates withsimple a(whichin\nRV. istoani asthree totwo,inpointoffrequency; inAV., asthree\ntofour).\nThe instr. ends later always inais;butintheVeda isfound abund-\nantly themore normal form ebhis(inRV., nearly asfrequentasaj's; in\nAV., onlyone fifth asfrequent).\nThe dat.and abl.have bhyasasending, with einstead ofthe finala\nbefore it(asintheVedic instr. ebhis, the loc.pi.,thegen.loc.du.[?],and\nthe instr. sing.). The resolution into ebhias isnotinfrequentintheVeda.\nThe gen. ends inanora, the finalabeing lengthened andhaving n\ninserted before thenormal ending. Theaoftheendingisnotseldom(in\nlessthan half theinstances)toberead astwo syllables, aam: opinions are\ndivided astowhether the resolution ishistorical ormetrical only.Avery\nsmall number(half-a-dozen)ofexamplesofsimple amasending instead of\nanam occur inRV.\n"
+    },
+    {
+        "page": 122,
+        "content": "102V.NOUNS ANDADJECTIVES.[329\nThe loc.ends inem that istosay,with thenormal ending, before\nwhich thestem-final ischanged toe(with consequent changeofstos:180).\nOfaccent inthis declension, nothing requirestobesaid\n;\nthesyllableaccented inthestem retains itsown accent through-\nout.\n330.Examplesofdeclension. Asexamplesof\ntheinflection ofa-stems maybetaken37Rkama, m., 'love';\n~%3devd, m.\n7'god'; EfTHT asya, n.,'mouth'.\nSingular:s\"\nN.\nA.kamas\nD.\nAb.\nL.\nV.\nDual\nN.A. V.\nI.D.Ab.\nG.L.kamena\ncftlHm\nkamaya\nkhmat\noJllHUJ\nkamasya\nx\nof-)|J-|\nkame\nkhmayos\nPlural :\nN.V.devds\n\\SF^\ndevdm\ndevena\n^TTT\ndevaya\ndevat\n*^\n<^e|Hl\ndevdsya\ndeve\n^\ndeva\nkamau devau\n^iT^^UH^TPTTT1?\nkamabhyam devabhyam\nkamasdevdyos\ndevasasydm\nasydm\nasyena\nasyhya\nasynt\n*%(H4t-tj\nasydsya\n-^\ndtijj^^j\nasye\nhsya\nasye\nasynbhyam\n^nHraiH^\nasydyos\n"
+    },
+    {
+        "page": 123,
+        "content": "334 DECLENSION I.,a-STEMs. 103\nA.\nkaman devfin asyani\nsRTifc^ ~|t^SCTHT\nkamais devais asyms\nD.Ab.\nkamebhyas devebhyas asyebhyus\nG.\n,r_ ,kamanam devanam asanam\nL. 37m\nkamesu devesu asyesu\nExamplesofthepeculiar Vedic forms are:\nSing.:instr. ravdthena, yajnd (such genitive forms asdfvasia arepurely\nsporadic).\nDu. :nom. etc.masc. devd; gen.-loc. pastybs (stem pastyh}.\nPI.: nom.-voc. masc^ devdsas; neut. yugd;instr. devebhis; gen. cara-\n(/tam, devanaam.\n331.Among nouns, there arenoirregularitiesinthisde-\nclension. Forirregular numeral bases ina(oran],seethe\nnextchapter. Fortheirregularitiesofpronominal stems ina,\nwhich aremore orlessfullyshared alsobyafewadjectivesof\npronominal kindred, seethechapter onPronouns.\nAdjectives.\n332. Original adjectivesinaareanexceedingly large\nclass, probablythemajorityofalladjectives.Thereis,however,\nnosuchthingasafeminine stem ina;forthefeminine, thea\nischangedtoa--oroften, thoughfarlessoften, toi\n;and its\ndeclension isthen like that ofsena ordevi(365). Anexample\nofthecompletedeclension ofanadjective a-stem inthethree\ngenderswillbegiven below(371).\n333. There arenoverbal roots endingina.Butaissome-\ntimes substituted forthe final aofaroot(and,morerarely,\nforfinalanoram),and itisthen inflected likeanordinary\nadjectiveina(seebelow, 354).\n334.Anoun endingina,when occurringasfinalmem-\nberofanadjective compound,isinflected likeanoriginal ad-\njectiveina,makingitsfeminine likewise inaor .\nOntheother hand, afeminine noun endinginderivative\nashortens itsfinal toatoform amasculine andneuter base.\n"
+    },
+    {
+        "page": 124,
+        "content": "104V.NOUNS ANDADJECTIVES.[335\nDeclension II.\nStems (ofallgenders) in*and u.\n335.The steins in^iand3uareinflected insoclose\naccordance with oneanother thattheycannot bedivided\nintotwoseparatedeclensions. Theyareofallthethree\ngenders,andtolerablynumerous those in^tmorenume-\nrous than those in3u,especiallyinthefeminine(thereare\nmore neuters in3uthan in^i).\nThe endingsofthis declension also differfrequently and\nwidely from thenormal, andtheVedicirregularitiesarenu-\nmerous.\n336.Endings: Singular. Thenom. masc. and fern, adds tothe\nstem thenormal endings.Thenom. and ace. neut. isthebare stem,\nwithout ending. IntheVeda, the finaluofafewneuters islengthened\n(248 b):thus, iwfi,puri.\nThe ace.masc. and fern, addsmtothestem. Vedic forms iniamand\nwarn, and, withn,inam andwnam, areexcessively rare, anddoubtful.\nThe instr. fern, inthelater languagetakes thenormal ending asimply,\nwhile themasc. andneut. insert nbeforeit,making inaanduna. But in\ntheVeda, forms inyaandva(oriaandua]arenotinfrequent inmasc.\nandneut. also;while ina isfound, very rarely, asafern, ending. Moreover,\nfern, yaisoften(intwothirds oftheoccurrences) contracted toI;and this\niseven sometimes shortened to i.Anadverbial instr. inuydfrom half-a-\ndozen stems inuoccurs.\nThe dat.masc. and fern, gunates the flnal ofthestem before theend-\ning e,making .ayeand ave. These aretheprevailing endings intheVeda\nlikewise; butthemore normalyeandvealsooccur; and the fern, hasin\nthiscase,asintheinstr., sometimes thecontracted form I.Inthelater\nlanguage, theneuter isrequired inthis, asinalltheother \"weakest\"cases,\ntoinsert nbefore thenormal ending:butintheVeda such forms areonly\nsporadic; andtheneut. dat.has alsotheformswe,ave, aye,liketheother\ngenders.\nThe abl.andgen.masc. and fern, haveregularly, both earlier andlater,\ntheendingswith gunated vowel before it:thus, a,os;andintheVeda,\ntheneut. forms thecases inthesame way; although wnas, required later,\nisalso notinfrequent (mas does notoccur). But thenormal forms yas (or\nias)andvas(oruas) are also frequent inboth masc. andneut. Asmasc.\nending, unas occurs twice inRV.\nThe loc.masc. and fern, has forregular ending inthelater language\n5ti,replacing bothfinals,iandu.And this isintheVeda alsothemost\nfrequent ending; but, besideit,thei-stems form (about half asoften in\nRV.) their loc.ina:thus, agna; and this isfound once even intheneut.\n"
+    },
+    {
+        "page": 125,
+        "content": "338]DECLENSION II., i-ANDM-STEMS. 105\nTheRV. has anumber ofexamples ofiriasc. and neut. locatives inavi\n(thenormal ending andtheugunated beforeit)from w-stems; and certain\ndoubtful traces ofacorresponding ayifrom i-steins. Half-a-dozen locatives\nin i(regarded bytheVedic grammarians asprayrhya oruncombinable :138d)\naremade from /-stems. The later language requires theneuter locatives to\nbemade mianduni;buttheformer never occurs intheoldesttexts, and\nthelatter only very rarely.\nThe latergrammar allows thedat., abl.-gen., and loc. fern, tobeformed\natwillwith thefuller fern, terminations oflong-vowel stems, namely ai,as,\nam. Such forms arequite rare intheolder language even from i-stems\n(less than40occurrences altogether inRV.\n;three times asmany inAV.);\nandfrom w-stems they arealmost unknown(liveinRV.andAV.).\nThe-voc. gunates the final ofthestem,inmasc. andfern., alike in\ntheearlier and inthe later language. Intheneut.,itislater allowed to\nbeofthesame form ortheunaltered stern :and thiswasprobably theusage\nintheolder time also;notinstances enough arequotabletodetermine the\nquestion (AV. hasuonce, andVS. oonce).\n337. Dual. The later andearlier language agree inmaking thenom.-\nacc.-voc. masc. and fern,bylengthening the final ofthestem. Thesame\ncases intheneuter(accordingtotherulegiven above) end later ininiand\nuni; butthese endings arenearly unknown intheVeda(as,indeed, the\ncases areofonly rare occurrence): AV. has inltwice (RV. perhaps once);\nVS.hasunionce; RV. hasulfromoneM-stem, andI,onceshortened toi,\nfrom one ortwoi-stems.\nTheunvarying ending ofinstr.-dat.-abl., inallgenders,isbhyam added\ntotheunchanged stem.\nThegen. -loc. ofallages adds ostothestem inmasc. and fern.;in\nneut., thelater language interposes,aselsewhere intheweakest cases, an;\nprobablyintheearlier Vedic theform would belike that oftheother gen-\nders; buttheonly occurrence noted isoneunos inAV.\n338. Plural. The nom.-voc. masc. and fern, adds thenormal end-\ningastothegunated stem-final, making ayasandavas. Theexceptions\nintheVeda areveryfew: oneword (ari) has iasinbothgenders, and a\nfewfeminines have Is(like i-stems); averyfeww-stems have uas. The\nneut. nom.-acc. ends later inInianduni(like anifroma:329); butthe\nVeda has Iand i(about equally frequent) much oftener thanmi; andu\nand(more usually) w,more than half asoften asuni.\nThe accus. masc. ends inInandtin, forolder Insanduns,ofwhich\nplain traces remain intheVeda innearly halftheinstances ofoccurrence,\nandeven notinfrequentlyinthe later language, intheguise ofphonetic\ncombination (208 ff.).The accus. fem. ends inisand us.Butboth masc.\nandfem. forms iniasanduasarefound sparinglyintheVeda.\nThe inst.ofallgenders adds bhis tothestem.\nThedat.-abl. ofallgenders addsbhyas (inV.,almost neverbhias]tothestem.\nThe gen. ofallgendersismade alike ininam andunam(ofwhich the\n"
+    },
+    {
+        "page": 126,
+        "content": "106V.NOUNS ANDADJECTIVES 338\naisnotseldom, intheVeda, toberesolved intoaam}. Stems withaccented\nfinal inthelater language may, andintheearlier always do,throw forward\ntheaccent upontheending.\nThe loc. ofallgenders adds su(assu:180) tothestem-final.\nTheaccent isinaccordance with thegeneral rulesalready\nlaiddown, and there arenoirregularities callingforspecial\nnotice.\n339.Examplesofdeclension. Asmodels of\n/-stems maybetaken 5|ftl agni, m., 'fire'; irf^ gdti, .,\n'gait'; snff vdri, n.,'water'.\nSingular: /\nN. jyiUH\nagnis\nagnim gdtim\nagnina gdtya wanna\nAb.G.agn&ye gdtaye,gdtyaivarine\nV.\nDual:\nN.A.V.agnes\n5^\nagnau\nqn\nagnevarinas\ngdtau, gdtyamm\ng&te van',\nagn\ngfn\nagnibhyam\nG.L.\nagnyosgdfi\nJiin\ngdtibhyam\n^IrJlH^\ngdtyosvarinl\nvtiribhyam\nPlural :\nagndyas gdtayas varmi\n"
+    },
+    {
+        "page": 127,
+        "content": "341]DECLENSION II., i-ANDW-STEMS. 107\nA.\nvarm\ni. rfiiftH TTWHH snfrPm\n,s\"^f . .\"^\nagnibhis gdtibhisvaribhis\nD.Ab.tlfijHJ^ rrfrRiq^ onf^HJ^\nagnibhyas gdtibhyas varibhyas\nG.\ntiiJl-nH^ JIrffaF[ ^{ImiH^\nagninhm gdtlnam varmam\nL.srftig JlfcFJ ^T3\nagnisu gdtisuvarisu\n340. Inorder tomark more plainlytheabsence inVedic languageof\nsome oftheforms which arecommonlater,alltheforms ofVedic occurrence\nareadded below, and intheorder oftheir frequency.\nSingular. Norn, agnts etc., asabove.\nAce.: masc.agnfin, yayiam, tmn/nam(?);fern,andneut. asabove.\nInstr. :masc. agnma, rayyd andurmid; fern, dcitti, utid, anuvrktf,\ndhasfna; neut. wanting.\nDat. :masc.agndye ;fern,tujaye, uti, prutyai ;neut. fucaye.\nGen.-abl. :masc. agues; fern, ddites, hetydsand6/mmias; neut. bhures.\nLoc. :masc. agnau, agnd, ajdyiC?)-,fern, dgatau, udita, dhdnasatayiC?),\nvedl,bhumyam; neut.apratd, saptdra$mau.\nVoc. :asabove(neut. wanting).\nDual: Nom.-acc.-voc.: masc. harl; fem.yuvatt;neut.tici,md/,/tdrmt(?).\nInstr.-dat.-abl. :asabove.\nGen.-loc. :masc. hdrios; fern, yuvatyos andjamios; neut. wanting.\nPlural. Nom.: masc. agndyas;fern, matdyas, bhumis; neut.{;wct,\nbhuri, bhurlni.\nAccus. :masc. agriin;fern,ksitis, ftica7/a(?).\nInstr., dat.-abl., andloc., asabove.\nGen. :masc. kavinam, rslnaam etc.\n341.Asmodels ofw-stems maybetaken ST5T?dtru,\nm.,'enemy'; Rdhenu, f.,'cow'; JTJmddhu, n.,'honey'.\nSingular:/\nN. ST^H ^H TOO N. O*x O\nfdtrus dhenus mddhu\nA. 5T^J7 ^T^T TOO\"s O*S O\ncdtrum dhenum mddhu\nTO^TT\nf o\ntruna dhenva mddhuna\n"
+    },
+    {
+        "page": 128,
+        "content": "108V.NOUNS ANDADJECTIVES.[341\ncatrave\nAb.G.\ncdtros\nqdtrau\nV.\nDual :\nN.A. V.STST\nI.D.Ab.\nG.L.\ncdtrvos\nPlural :\nN.V.\nA.vas\nD.Ab.cdtrubhis\ncdtrubhyas\nsr^nnTR\ncdtrunam\nL-dhendve,dhenvai\ndhenos,dhenvas\ndheno\ndhenu\ncdtrubhyam dhenubhyam\ndhenvos\ndhendvas\nex*>dhenus\ndhenubhis\ntpPTTH\ndhenubhyas\ndhenunkm\n00\ndhenusumddhune\nJT^RTT\nmddhunas\nmddhuni\n^\nmddhu,mddho\nmddhuni\no\nmddhubhyam\nmddhunos\nex\nmddhuni\nex.\nmddhuni\nmadhubhis\n^T^T^JHo%\nmddhubhyas\nex *x\nmddhunam\noo\nmddh\n342. Theforms ofVedic occurrence aregiven here forthew-stems in\nthesamemanner asforthei-stems above.\nSingular. Norn.: masc. and fern, asabove; neut. wrw,ww.\nAccus. :masc. ketum, dbhiruam, sucetunam^} ;fern,dhenum.\nInstr. :masc. ketiina, papva andkrdtud; fern,ddhenua andpanua, aput/a;\nneut. mddhuna, mddhva.\nDat. :masc. fecfave, pipve; fern, pdrave, isvai; neut. urduc, mddhune.\nAbl.-gen.:masc. manyos, pit-was, saramas; fern, sfndhos, isvds; ueut.\nmddhvas andrnddhuas, inddhos, mddhunas.\n"
+    },
+    {
+        "page": 129,
+        "content": "344]DECLENSION II.,i-ANDM-STEMS. 109\nLoo.: masc. purau, sunavi fern,sfndhiiu, rdjjvarn; neut. sanita,.si/mir/.\nadno, sflnuni.\nVoc. :asabove.\nDual. Nom.-acc.-voc. :masc. and fern, asabove; neut. wrvf, jdnunl.\nInstr.-dat.-abl. :asabove.\nGen.-loc. :asabove (but -vos or-uos).\nPlural. Nom. :masc. rbhtivas, mddhuas andnu'idhvas; fern, dhendvas,\nfutfikratvas ;neut. puruni, pnrn, puru.\nAccus. :masc. rfwn, pafvds ;fern, fsus, mddhvas.\nInstr., dat.-abl., andloc.,asabove;also gen. (but with theresolution\n-unanm inpart).\n343.Irregular declension. There arenoirregular\n?/-stems, andonlyaveryfew/-stems.\na.S&khij m., 'friend', hasforthe fivestrong cases apecu-\nliarly strengthened base(vriddhied), namely sakhay, which inthe\nnom.sing,isreduced tosdkha(without ending), and inthe\nother cases takes thenormal endings. The instr. and dat.sing,\nhave thenormal endings simply,without inserted norguna;\ntheabl.-gen. sing, adds us;and the loc.sing, adds au :the\nrest islikeagni. Thus :\nSing, safcfta, sdkhayam, stikhya, sdkhye, sdkhyus, sdkhyau, sdkhe.; Du.\nadkhayau, sakhibhyam, sdkhyos ;PI.sakhayas, stiktnn, etc. etc.\nTheVeda hasusually sdkhaya du., and often resolves theyto?,in\nsfikhia, sakhius, etc.Thecompoundsareusually declined likethesimple\nword, unless sakha besubstituted.\nb.Pad, m.,isdeclinedregularlyincomposition, andwhen it\nhasthemeaning 'lord, master';whenuncompounded andmean-\ning'husband',itisinflected like sdkhi intheinstr., abl.-gen.,\nand loc.sing., forming patya, petty e,patyus, pdtyau.There are\noccasional instances ofconfusion ofthetwo classes offorms.\nc.Jam', f.,'wife', hasthegen. sing, j&nyusintheVeda.\nd.Ari, 'eager, greedy, hostile', hasintheVeda arydsinpi.nom. and\naccus., masc. and fern.\ne.W, 'bird', hasinRV. thenom. ves(beside vfa).\nf.Thestems dksi, 'eye', dsthi, 'bone', dddhi, 'curds', andsdkthi, 'thigh',\naredefective, their forms exchanging with andcomplementing forms from\nbases inan(aksdn etc.): seethebases inan,below(431).\ng.Thestempathf, 'road',isused tomake uppart oftheinflection of\npdnthan:seebelow, 433.\nh.Kr6stu, m., 'jackal', lacks thestrong cases, forwhich thecorresponding\nforms ofkrostr aresubstituted.\nAdjectives.\n344. Original adjective stems iniarefew; those inuare\nmuch more numerous (manyderivative verb-stems forming a\n"
+    },
+    {
+        "page": 130,
+        "content": "110V.NOUNS ANDADJECTIVES.[344\nparticipial adjectiveinu}.Their inflection islikethat ofnouns,\nandhasbeen included intherules given above;thestem isin\ngeneralthesame inallthethree genders.Inthoseweakcases,\nhowever namely,the dat., abl.-gen., and loc.sing., and\nthegen.-loc. dual inwhich neuter nouns differ frommas-\nculines inthe later language byaninserted n(wehave seen\nabove that thisdifference does notexist intheVeda),theneuter\nadjectiveisallowed totake either form.\nButadjectivesinupreceded byoneconsonant sometimes\nform aderivative feminine stem byaddingi:thus, bahvi, urvi,\nprthvl,, vibhvi, andsoon.Morerarely,theuisprolongedtou\ntomake afeminine-stem, which isthen inflected like vadhu\n(below, 365).Someadjectivesform their feminine intwo of\nthese ways,oreven inallthethree :thus, blbhatsu andblbhatsu;\ntanu, tank, and tanvi.\n345. Roots endingin toru(or.r:380) regularly adda\ntwhen used asroot-words orasroot-finals ofcompounds ;and\nhence there arenoadjectivesoftheroot-class inthisdeclension.\nYet, intheVeda, afewwords ending inashort radical uaredeclined\nasifthiswere sufflxal: thus, dsmrtadhru, sustu; andtheAV. hasprtanajf\n(once). Roots inusometimes alsoshorten utou:thus, prdbhu, vibhu,etc.\n(354); go(361) becomes guincomposition; and reperhaps becomes ri\n(362);while roots inasometimes apparently weaken atoi(in-dhifromydha).\n346.Compound adjectives having nouns ofthisdeclension\nasfinalmember areinflected likeoriginal adjectivesofthesame\nendings.\nDeclension III.\nStems inlongvowels:a,1,u.\n347.The stemsendinginlong vowels fall intotwo\nwell-marked classes ordivisions: A.monosyllabicstems -\nmostlybare roots and theircompounds,with acompar-\natively small number ofothers inflected likethem; B.de-\nrivative feminine stems inEftaand^,with asmallnum-\nberin3Tuwhich inthelaterlanguagehavecome tobe\ninflected likethem. The latter division isbyfarthelarger\nandmoreimportant, since most feminineadjectives, and\nconsiderable classes offeminine nouns, endingin5TFaor\n^,belongtoit.\n"
+    },
+    {
+        "page": 131,
+        "content": "350] DECLENSION III,a-, 7-,ANDW-STEMS. 111\nA.Root-words, andthose inflected likethem.\n348.The inflection ofthese stems isbythenormal\nendings throughout,orinthemanner ofconsonant-stems\n(with^fam,not^m,'intheaccus.sing.); peculiarities\nlikethose oftheother vowel-declensions arewanting. The\nsimple wordsare, asnouns, withfewexceptions feminine;\nasadjectives (rarely),andinadjective compounds, theyare\nalike inmasculine andfeminine forms. They may,forcon-\nvenience ofdescription, bedivided intothefollowingsub-\nclasses :\n1.Root-words, ormonosyllables havingtheaspectofsuch.\nThose inaaresorare that itishardly possibletomake upa\nwhole scheme offorms inactual use;those in ianduaremore\nnumerous, but stillvery few.\n2.Compounds having suchwords, orother roots with long\nfinal vowels, aslastmember.\n3.Polysyllabic words, ofvarious origin and character, in-\ncludingintheVeda many which later aretransferred toother\ndeclensions.\n4.Asanappendixtothis classwemaymost conveniently\ndescribe thehalf-dozen bases, mostlyofirregular inflection,\nendingindiphthongs.\n349.Monosyllabic stems. Before theendings begin-\nning with vowels, final iischangedtoiyandutouv;while\nfinal aisdropped altogether, exceptinthestrongcases andin\nthe ace.pi.,which islike thenominative(accordingtothe\ngrammarians,aislosthere also :noinstances oftheoccurrence\nofsuch aformappeartobequotable).Stems in ianduare\ninthelater language allowed totakeoptionallythefuller end-\nings at, as,aminthesingular (dat., abl.-gen., loc.); butno\nsuch forms areevermetwith intheVeda (except Wiiyai [?],\nRV., once).Before amofgen. pi.,nmayormaynotbein-\nserted;intheVeda itisregularly inserted, with asingleex-\nception (dhiyhm, once).The vocative islike thenominative in\nthesingular aswell astheother numbers\n;butinstances ofits\noccurrence inuncompounded stems arenotfound intheVeda,\nandmust beextremelyrareeverywhere. The earlier Vedic dual\nendingisainstead ofau.\n350. Tothe iandw-stems, therules formonosyllabic\naccentapply:theaccent isthrown forward upontheendings\ninalltheweak cases excepttheaccus.pi.,which islike the\n"
+    },
+    {
+        "page": 132,
+        "content": "112 V.NOUNS ANDADJECTIVES.[350\nnom. Butthea-stems appear (theinstances areextremely few)\ntokeeptheaccent uponthestem throughout.\n351.Examples ofdeclension. Asmodels ofmo-\nnosyllabicinflection wemaytakesfT/#, f.,'progeny';sft\ndM,f..'thought'; and ?\\bhfy f.,'earth'.\nThe first ofthese isratherarbitrarily extended from thefour cases\nwhich actually occur: ofthe loc. sing, andgen. etc. du.,noVedic examples\nofa-stems arefound.\nSingular:\nN.\nA.jas\nam\nI>.it\nj*\nAb.o.\nsra^\nj&s\nL. |%\nV.\n.,\njas\nDual;\nN.A.V.\nI.D.Ab.au\nff\njabhyam\nG.L.\nJOS\nPlural :\nN.\nA.Jas\nsfTC\nfas,Jasdhiyau\nsffcEpw\ndhibhyhm,\ndhiyos\ndhiyas\ndhiyaso -\\\nbhuvam>TH\n/-\"^^T^dhls bhus\nirrgrf\ndhiyam\nf^HTT\ndhiya bhuvh\nRnr,fim H^,^\nc?Azy<?, dhiyai bhuve, bhuvai\n^,**. o -\\o -s\ndhiyas, dhiyas bhuvas, bhuvas\ndhiyi, dhiyam bhuvi, bhuvam\nbhuvau\nc\\ *\\\nf\nbhubhyam\n-X\nbhuvos\nbhuvas\nbhuvas\n"
+    },
+    {
+        "page": 133,
+        "content": "352) DECLENSION III,A,RADICAL-,e-,ANDW-STEMS. 113\nbhubhis\n1FC(TT\nC\\\"^\nbhubhi/dsI. sTTTR\n,~x x\njabhisdhibhis\nD.Ab. sfFTO tfr-TTR\nf\"** \"\\\njabhyas dhibhyds\nG. STHT^ isTFT?)\nI_-\\ >I-s\n., _ _,\njanam, Jam dhiyam, dhmam bhmam, bhunam\nL. arm uta wo c\\o\nbhumarmo\n352.Monosyllabic stems incomposition. When\nthenouns above described occur asfinalmember ofacompound,\norwhen anyroot inaor ioruisfound inalikeposition,\nthe inflection ofan-stem isasabove. But 1andw-stems\nfollow adivided usage:the final vowel before avowel-ending\niseither converted into ashort vowel andsemivowel(iyoruv,\nasabove) orinto asemivowel simply yorv).The accent is\nnowhere thrown forward upontheendings;andtherefore, when\niandubecome yandv,theresulting syllableiscircumflex.\nThus :\n-dhfyarn\n-dhiya\n-dhiye\n-dhfyas\n-dhfyiSingular:\nN.V.\nA.\nI.\nD.\nAb.G.\nL.\nDual:\nN.A.V. -dhtyau\nI.D.Ab.\nG.L. -dhfyo\nPlural:\nN.A.V.\nI.\nD.Ab.-dhs -bhus\n-dhyctm\n-dhya\n-dhye\n-dliyhs\n-dhyi\n-dhyaii-bhuvam\n-Ihuva\n-bhtive\n-bhuvas\n-bhtivi\n-bhuvau-bhvam\n-bhvh\n-bhvl\n-bhvtis\n-bhv*\n-bhvab,\n-dhibhyam -bhubhyam\n-bhuvos -bhvbs\n-bhtivas -bhvhs\n-bhabhis\n-bhubhyas\n-bhiivam\n.,,_-bhunam\n-bhusu,^-bhvam-dhfyas -dhyhs\n-dhibhis\n-dhlbhyas\n\\-dhiyam,\n-dhyam(-dhmam\nL. -dhim\nAstotheadmissibility ofthe fuller endings ai,as,andaminthe\nsingular (feminine), grammatical authorities aresomewhat atvariance; but\nthey arenever found intheVeda, andhave been omitted from theabove\nscheme asprobably unreal.\nIftwoconsomants precede the final Ioru,the dissyllabic forms, with\niyanditu,areregularly written;after oneconsonant, theusageisvarying.\nThegrammarians prescribe iyanduvwhen themonosyllabic stem hasmore\nWhitney, Grammar. 8\n"
+    },
+    {
+        "page": 134,
+        "content": "V-NOUNS ANDADJECTIVES.[352\nthecharacter ofanoun, andyandvwhen itismore purely averbal root\nwith participialvalue. Nosuch distinction, however,istobeseen inthe\nVeda where, moreover, thedifference ofthetwoforms isonly graphic,\nsince theyaaridvaforms andtherest arealways toberead asdissyllabic:\niaorlaanduaorua,and soon.\n353.Afewfurther Vedic irregularities orpeculiarities maybebriefly\nnoticed.\nOfthea-stems,theforms inas,am,a(du.)aresometimes toberead\nasdissyllables, aas,aam, aa.The dative ofthestem used asinfinitive is\ninat(asif<!-}-): thus, prakhyaf, pratimaf, paradat.\nIrregular transfer oftheaccent totheending incompoundsisseen in\nacase ortwo: thus, avadyabhiyd (RV.), ddhid(AV.).\n354. Butcompoundsoftheclass above described arenot\ninfrequently transferred toother modes ofinflection :thea\nshortened toaforamasculine stem, ordeclined like astem\nofthederivative a-class(below, 365)asfeminine\n;the iandu\nshortened to *and u,and inflected asofthesecond declension.\nThus, compound stems in-ga, -ja, -da, -stha, -bhu, andothers, are\nfound even intheVeda, andbecome frequent later; andsporadic cases from\nyetothers occur :forexample, yrtapdn, vayodhafs andratnadhebhis, dhanasais\n(allRV.); and, from Ianducompounds, vesa^rfs (TS.), dhrayas (RV.),\nganafrfbhis (RV.), rtanfbhyas (RV.) and senanfbhyas (VS.) andgramanibhis\n(TB.), supuna (AV.), citibhrdve (TS.). Stillmorenumerous arethefeminines\ninawhich have lost their root-declension :examples areprajd (ofwhich\nthefurthercompounds inparthaveroot-forms), svadhd, fraddhd, pratimd,\nand others.\n355.Polysyllabic Stems. Stems ofthis division of\nmore than onesyllable areveryrareindeed inthelaterlang-\nuage, andbynomeans common intheearlier. TheRig-Veda,\nhowever, presents anotinconsiderable bodyofthem\n;andas\nthe classnearly dies outlater, bythedisuse ofitsstems or\ntheir transfer toother modes ofdeclension,itmay bebest\ndescribed onaVedic basis.\na.Ofstems ina,masculines, half-a-dozen occur intheVeda: pdntha,\nmdntha, andrbhuksa areotherwise viewed bythelatergrammar: seebelow,\n433 4;ufdna (nom. pr.)hastheanomalous nom.sing, wpana (andloc. as\nwell asdat.ufdne); mahd, 'great1\n,isfound only inaccus.sing, andabund-\nantly incomposition: dta, 'frame', hasonly dtasu notderivable from dta.\nb.Ofstems ini,overseventy arefound intheVeda, nearly allfemi-\nnines, and allaccented onthe final. Half ofthefeminines areformed from\nmasculines withchange ofaccent:thus, kalyani (m.kalydna), purusi (m.\njn/rusa); others shownochange ofaccent: thus, yarn! (m.7/ama); others still\nhavenocorresponding masculines :thus, nadi, laksmi, surmi. Themasculines\nareabout teninnumber: forexample, rathi, pravi, stari, aAT,apathi.\n"
+    },
+    {
+        "page": 135,
+        "content": "356]DECLENSION III.A,RADICAL a-. 1-.ANDe^-STEMS. 115\nc.Ofstems inu.thenumber issmaller:these, too, arenearlyall\nfeminines. and allaccented onthe final. The majority ofthem arethe\nfeminine adjectivesinUtomasculines in<ioru(above, 344): thus. caranyU,\ncarisnu, jighatsu, madhu. Afew arenouns inw,with change ofaccent:\nthus, agru (dgru), prdaku (prdaku), pva^rU (fvtifura); orwithoutchange,as\nnrtu. And afewhave nocorresponding masculines: thus, tanft, radftd,\ncam'U. The masculines areonly two orthree; namely, prafU, krkadapTi,\nma/csw(?); andtheir forms areoftheutmost rarity.\n356. Themode ofdeclension ofthese words maybeil-\nlustrated bythefollowing examples:ratM, m., 'charioteer'; nadi,\nf.,'stream'; tann, f.,'body'.\nNooneoftheselected examples occurs inalltheforms :forms forwhich\nnoexampleatall isquotable areputinbrackets. Noloc. sing, from any\nI-stem occurs,todetermine what theformwould be.Thestem nadi isse-\nlected asexample partly inorder toemphasize thedifference between the\nearlier language andthelater inregardtothewords ofthis division :nadi\nislater themodel ofderivative inflection.\nSingular:\nN. rathis nadis tanUs\nA. rathiam nadiam tanuam\nI. rathfa nadia tanua\nD. raihie nadte tanue\nAb.G. rathfas nadfas tani'ias\nL . tanui\nV. rdthi(?) nadi tanu\nDual:\nN.A.V. rathm nadia tantia\nI.D.Ab. [rathibhyam] nadibhyam [tanubhyam]\nG.L. [rath-Cos] nadios tantios\nPlural :\nN.A. rathias nadfas tantias\nI. [rathibhis] nadibhis tanubhis\nD.Ab. [rathibhyas] nadibhyas tanUbhyas\nG. rathinam nadlnam tanunam\nL. [rathisu] nadisu tanusu.\nThe cases nadiam, fant/am,etc. arewritten above accordingto\ntheir true phonetic form,almost invariably belongingtothem intheVeda:\ninthewrittentext,ofcourse,thestem-final ismade asemivowel, andthe\nresulting syllableiscircumflexed :thus, nadyhm, tanvam, etc.;only,as\nusual, after twoconsonants theresolved formsiyanduvarewritten instead;\nand alsowhere thecombination yvwould otherwise result: thus, cakrfya,\n[agruvai,] andmitrayuvas. TheRV. really readsstarybmetc. twice, and\ntanvhs etc.fourtimes;andsuch contractions aremore often made inthe\nAV. Theending aofthenom.-acc.-voc. du. istheequivalentofthelater\ncm.Thenom.sing, insfrom z-stems isfound intheolder language about\nsixty times, from over thirty stems.\n8*\n"
+    },
+    {
+        "page": 136,
+        "content": "jj(3V.NOUNS ANDADJECTIVES.[357\n357. Irregularitiesofform, properlysocalled,arevery few inthis\ndivision: eamtZ asloc. sing, (insteadofcamv\"i} occurs afew times; and\nthere isanother doubtful case ortwoofthesame kind\n;thefinalUisregarded\naspragrhyaoruncombinable (138j;tanui islengthenedtotanv\\ inapassage\nortwo; -yuvasisonce ortwice abbreviated to-yUs.\n358. Theprocessoftransfer totheother form ofIandu-declension\n(below, 363ff.),which hasnearly extinguished thiscategory ofwords inthe\nlater language,has itsbeginnings intheVeda;butinRV. they areexcessive-\nlyscanty: namely, dutidm,loc.sing., once, andpuaprucfm, do., once, and\ndravitnud,instr. sing.,with two orthree other doubtful cases. Inthe\nAtharvan, wofind the ace. sing, kuh&m, tanUm, vadhUm; the instr. sing.\npalalidandone ortwo others; the dat. sing, vadhvdf, fvafruai, ayrtivai;\ntheabl.-gen. sing, punarbhuvas, prdakuds, fuafruds,and the loc. sing.\ntanuam (with anomalous accent). The accusatives pluralinIsandusare\nnowhere met with.\n359. Adjective compounds from these words arevery few; those which\noccur aredeclined likethesimple stems: thus, hiranyavcifis andsahdsrastarii,\ndtaptatanusandsdrvatanus,allnom. sing, masculine.\nStems ending indiphthongs.\n360. There arecertain monosyllabic stems endingindiph-\nthongs,which aretoofewandtoodiverse ininflection tomake\nadeclension of,andwhich maybemost appropriately disposed\nofhere,inconnection with thestems in iand u,withwhich\ntheyhavemostaffinity. Theyare :\nstems inan :nauandglau ;\nstems inai:rat;\nstems ino:goanddyo (ordyu, div}.\n361. a.Thestem nau, f.,'ship',isentirely regular, taking\nthenormalendings throughout, andfollowing therules for\nmonosyllabic accentuation(317) exceptthat theaccus.pi.\nissaid(itdoes notappeartooccur inaccentedtexts)tobe\nlike thenom. Thus: naus, navam, nava, nave, navds, navi;\nnavau, naubhyam,navos;novas, navas, naubhis, naub/iyds, navam,\nnausu.\nThestemglati, m., 'ball',isapparentlyinflected inthesame way; but\nfew ofitsforms havebeenmetwith inuse.\nb.Thestemrai,f.(orm.), 'wealth', might bebetter described\nasrawith aunion-consonant y(258) interposedbefore vowel\nendings, and isregularlyinflected assuch, withnormal endings\nandmonosyllabic accent. Thus :ras,rayam, raya,raye, rayas,\nrayi ;rayau, rabhyhm, rayos; rayas, rayas,rabhis, rabhyds, rayam,\nrasu. But intheVeda theaccus.pi.iseither rayasorrayas;\nforaccus.sing, andpi.are alsoused thebriefer forms ram\n"
+    },
+    {
+        "page": 137,
+        "content": "362]DECLENSION III.:DIPHTHONGAL STEMS. 117\n(which alone isofVedicoccurrence)andras;andthegen. sing,\nissometimes anomalously accentedrayas.\nc.Thestemgo,m.orf.,'bull' or'cow',ismuchmoreirreg-\nular. Inthestrong cases, exceptaccus.sing.,itisstrengthened\ntogau, forming (like nau) gaus, gavau, gavas.Inaccus. sing,\nandpi.ithas(like rat)thebrief forms ghmandgas.The\nabl.-gen. sing,isgos (asiffromgu).The rest isregularly\nmade fromgo,with thenormal endings, butwith accent always\nremaining irregularly uponthestem :thus, gdva, gave, gdvi ;\ngdvos ;gdvam; gobhyam, gobhis, gobhyas, gom.IntheVeda,\nanother form ofthegen. pi.isgonam;thenom. etc.du. is\n(asinallother suchcases)alsogava; andgam, gos,andghs\narenotinfrequentlytobepronouncedasdissyllables.\nd.Thestemdyo,f.(butinV.usually m.), 'sky, day',isyet\nmore anomalous, having beside itasimpler stemdyu,which\nbecomes divbefore avowel-ending. The native grammarians\ntreat thetwo asindependent words, but itismore convenient\ntoputthem together. Thestem dyoisinflectedpreciselylike\ngo,asabove described. Thecompletedeclension isasfollows\n(with forms notactually metwith inusebracketed):\nDual.\n[dwau] dyavau\n[dyubhyam dyobhyam]Singular.\n"
+    },
+    {
+        "page": 138,
+        "content": "118V.NOUNS ANDADJECTIVES.[352\nwhich have notbeen specifiedabove asbelongingtothe\nother orroot-word division;andalso,inthelaterlanguage,\nmost ofthe Iandw-stems oftheotherdivision, bytransfer\ntoamore predominantmode ofinflection. Thus :\na.The great mass ofderivative feminine a-stems, sub-\nstantive andadjective.\nThe inflection ofthese stems hasmaintained itself with little change\nthrough thewhole history ofthelanguage, being almostprecisely thesame\nintheVedas aslater.\nb.Thegreat mass ofderivative feminine e-stems.\nThis class iswithout exceptioninthelater language. Intheearlier,\nitsuffers theexception pointedoutabove (355 b]:thatfeminines made with\nchangeofaccent follow thismode ofdeclension onlywhen theaccent isnot\nontheT:thus, tdvisi, pdrumi, pdliknl, rtfhini.\nThei-stems ofthis division ingeneral areregarded asmade bycon-\ntraction ofanearlier ending inya. Their inflection hasbecome inthe later\nlanguage somewhat mixed with that oftheotherdivision, and sofardifferent\nfrom theVedic inflection:seebelow, 364, end.\nVery few derivative stems inIarerecognized bythegrammarians as\ndeclined liketheroot-division; theVedic words ofthat classare,ifretained\ninuse, transferred tothismode ofinflection.\nAvery small number ofmasculine z-stems(half-a-dozen) areinthe\nVeda declined asofthederivative division: they areafew rare proper\nnames, mdtali etc.;and r&stri andsin(only onecaseeach).\nc.Thew-stems arefewinnumber, andaretransfers from\ntheotherdivision, assimilated ininflection tothegreatclass of\nderivative i-stems(exceptthat theyretain theendingsofthe\nnom.sing.).\n363.Endings. Thepoints ofdistinction between thisandtheother\ndivision areasfollows :\nInnom. sing, theusual s-endingiswanting: exceptintheu-stems\nandaveryfewt-stems namely, laksmi, tarl,tantrl which have pre-\nserved theendingoftheother division.\nThe accus. sing, andpi.addsimplymand srespectively.\nThedat., abl.-gen., and loc. sing, take always the fuller endings ai,\nas,am;andthese areseparated from the final ofthea-stems byaninter-\nposed y.\nBefore theendings aofinstr. sing, and osofgen. -loc. du., the final\nofa-stems istreated asifchangedtoe;butintheVeda, the instr. ending\navery often(innearly half theoccurrences) blends with the final toa.\nTheyaofi-stems isinafewVedic examples contracted toi,andeven\nto i.Aloc. sing, inIoccurs afewtimes.\n"
+    },
+    {
+        "page": 139,
+        "content": "364]DECLENSION III.B,DERIVATIVE-,i-,ANDW-STEMS. 119\nInalltheweakest cases abovementioned, theaccent ofan Ioru-stem\nhaving acute final isthrown forward upon theending. Intheremaining\ncase ofthesameclass,thegen. pi.,anisalways interposed between stem\nandending, and theaccent remains upon theformer(inRV., however,it\nisusually thrown forward upon theending,asin iandw-stems).\nInvoc. sing.,finalabecomes e;final tanduareshortened.\nInnom.-acc.-voc. du.andnom.pi.appears inI(and u)-stems amarked\ndifference between the earlier and later language, thelatter borrowing the\nforms oftheother division. The du.ending duisunknown inRV., and\nvery rare inAV.;theVedic endingisi(acorresponding dual ofii-stems\ndoes notoccur). Theregularlaterpi.endingashasonly adoubtful ex-\nampleortwoinRV., andavery smallnumber inAV.;thecase there (and\nitisoneofvery frequent occurrence) adds ssimply; andthough yas-forms\noccur intheBrahmanas, along with zs-forms, both areused indifferently as\nnom. andaccus. Ofa-stems, thedu.nom. etc.ends ine,both earlier and\nlater;inpi.,ofcourse, s-forms areindistinguishable from ow-forms. The\nRV.hasafewexamplesofdsas foras.\nTheremaining cases call fornoremark.\n/\n364.Examples ofdeclension. Asmodels ofthe\ninflection ofderivative stemsendinginlong vowels, we\nmaytake?RTsena,f.;'army'; ^TT kanya, f., 'girl';\ndevi, f.,'goddess'; ofSJvadhu, f.,'woman'.\nSingular:\nN.\nA.\nI.\nD.\nAb.G.\nL.\nV.sena\nseriam\nsenaya\nH^iltT\nsenayai\nsenayas\nsenayam\nsenekanyhdevi\nkanyhm devim\nSR^^I^cETT\nkanyaya devya\n=fI*^JT7T 5^?J\nkanyhyai devyai\noh^lMIH rfctllH\nkanyhyas devytis\nkanyayam devyam\ndevivadhus\ncT^nr\nvadhum\nvadhvU\nvadhvai\nvadhvUs\nvadhvftm\nvadhu\n"
+    },
+    {
+        "page": 140,
+        "content": "120V.NOUNS ANDADJECTIVES. 364\nDual:\nN.A. V.\nsene kanye devyau }\nI.D.Ab.\nG.L.am kanyhbhyam dev^bhyam\nkanyayos devyos\nPlural :\nN.v.\nA.senas\nsenaskanyhs devyas\nkanyttsdevis\ncfl^tlhHH ^cjlMH\nkanyhbhisdevibhis\nD.Ab.\nG.\nL.\nsenasuvadhvau\nsr^RJFTc\\ -x_\nvadhubhyam\n5f5n^\nvadhvos\nvadhvas\nS^msx4v\nvadhus\nex\n,\nvadhubhis senabhis\nsenabhyas kanyabhyas devibhyas vadhubhyaso\n^l^l^m eft^tlMIH rfcjHlH eJM'SIH\nsenanam kanyhnamdevinam vadlmnam\nkanyhsudevisu vadhusu\nIntheYeda, vadhu isastern belongingtotheother division(like tanu,\nabove, 356).\n365. ExamplesofVedic forms are :\n1.a-stems: instr. sing,mamsd(this simplerform isespecially common\nfromstems intdandia);nom.pi.vafdsas (about twenty examples:Lanman,\np.362); accus. pi.aramgamdsas (acase ortwo). Half the 6fa/as-cases are\ntoberead asbhias; theamofgen. pi.isafewtimes toberesolved into\naam; andtheaandamofnom. and accus. sing, are, very rarely, tobe\ntreated inthesame manner.\n2.t-stems: instr. sing, pdrm, pdmi; loc.gauri ;nom. etc.du.devi;\nnom. pi.devis; gen. pi.bahvlndm. The final ofthestem istoberead as\navowel(not y)frequently, butnotinthemajority ofinstances :thus, devid,\ndevias, deviam, rddasios.\nThesporadic instances oftransfer between this division and thepre-\nceding have been already sufficiently noticed.\n3.Inthelanguage oftheBrahmanas, theabl.-gen. sing, endingasis\nalmostunknown, andinstead ofitisused the dat.ending di.Theassump-\ntion ofthesame substitution issuggested, butnotrequired, inafewRV.\n"
+    },
+    {
+        "page": 141,
+        "content": "368]DECLENSION III.B,DERIVATIVE-,i-3ANDW-STEMS. 121\npassages;and itisnecessary once inAY.(iv.5.6): svdpantv asydi jndtdyah,\n'letherrelativessleep'. Brahmana examples are:tdsydi di$dh (TS.), 'from\nthat direction'; striydi payah (AB.), 'woman'smilk'; dhenvaf vdetdd retail\n(TB.), 'that, forsooth,istheseed ofthecow'; jydyasl ydjydydi (AB.), 'superior\ntotheydjyd'.\n366. Thenounstrf, f.,'woman'(probably contracted fromsutrl, 'gene-\nratress'), follows amixed declension :thus,stri\n,striyamorsfrfm, striyd,\nstriyai, striyds, striydm, strf; strfydu, stribhydm, striyos ;strfyas, strfyasor\nstris, stribhfs, stnbhyds, strindm,strlsu(buttheaccusatives strim and stris\narenotfound intheolder language, andthevoc. stri isnotquotable). The\naccentuation isthat ofaroot-word; theforms(conspicuously thenom. sing.)\narethose oftheother division.\nAdjectives.\n367. a.Theoccurrence oforiginal adjectivesinlongfinal\nvowels, and ofcompounds havingasfinalmember astem of\nthe firstdivision, hasbeensufficientlytreated above, sofaras\nmasculine andfeminine forms areconcerned. Toform aneuter\nstem incomposition,therule ofthe later languageisthat the\nfinallongvowel beshortened;andthestem somade istobe\ninflected likeanadjectiveinioru(339, 341).\nSuch neuter forms arevery rare, and inthe older language almost\nunknown. Ofneuters from z-stems have been noted intheVeda only\nhariprfyam,ace. sing, (amasc.form), and suadhfas, gen. sing, (same as\nmasc. andfern.); fromii-stems, only afewexamples, andfrom stem-forms\nwhich might bemasc. and fern, also :thus, vibhti, subhti,etc.(nom. -ace.\nsing.: compare 354); supud andmayobhtivd,instr. sing.; andmayobhu,ace.\npi.(compare purti: 342); from a-stems occur only half-a-dozen examples\nofanom.sing, inas,likethemasc. and fern. form.\nb.Compounds having nouns ofthesecond division as\nfinalmember arecommon onlyfrom derivatives ina;andthese\nshorten the final toainbothmasculine andneuter :thus, from\na,'not', andpraja, 'progeny', come themasc. andneut. stem\napraja,fern,aprcy'a,'childless'. Such compoundswith nouns\nin ianduaresaid tobeinflected inmasc. and fern, like the\nsimple words(onlywith inanduninace.pi.masc.);butthe\nexamples given bythegrammariansare fictitious. Thestem\nstri isdirected tobeshortened to-stri for allgenders.\n368. Itisconvenient togiveacomplete paradigm,\nforallgenders,ofanadjective-stemin [a.Wetake for\nthepurpose efF7papa, 'evil', ofwhich thefeminine isusu-\nallymade inmainthelaterlanguage,butin^1inthe\nolder.\n"
+    },
+    {
+        "page": 142,
+        "content": "122 V.NOUNS ANDADJECTIVES.[368\nSingular:\nm.\nN.\nA.\nD.\nAb.\nG.\nL.\nV.\nN.A.V.papas papam\npapam\nTO^T\npapena\npapaya\nTOTH^\npapat\npapasya\nTO\nDual:\npapau pape\nI.D.Ab.\nG.L\npapayos\nPlural :\nN.\nA.\nD.Ab.papas papni\nHNM\npapan paphni\nHNH4H\ns\npapebhyaspapa\npapam\npapaya\npapayai\npapayas\npapayas\npapayam\npape\npapabhyam\nos\npapas\nMNIH\npapas\npapabhis\npapabhyasf.\nTOt\npapim\npapya\npapyai\npapyas\npapyas\nH(L(>J |i-|\npapyam\npapi\npapyau\npapibhyam\npapyos\npapibhis\nMNl^^UTT\n_ f-x\npaplbhyas\n"
+    },
+    {
+        "page": 143,
+        "content": "371] DECLENSION IV., /--STEMS. 123\nMMHIM\npapanam papanam paplnam\npapesu papasu papisu\nDeclension IV.\nStems inr(orar).\n369. This declension isacomparativelylimitedone,\nbeing almostentirely composedofderivative nouns formed\nwith the suffix rTtr(orcT|\"tar\\which makes masculine\nnominaagentis (used alsoparticipially), andafewnouns of\nrelationship.\nBut itincludes also afewnouns ofrelationshipnotmade\nwith that suffix: namely devf, m., svdsr andndnandr,f.;and,\nbesides these, nf,m.,sir(inV.),m.,nsf (inV.), f.,savyasthr,\nm.,andthefeminine numerals tisrand catasr(forwhich, see\nchap. VI.). Thefeminines intrareonlymatr, duhitr, andyhtr.\nTheinflection ofthese stems isquite closely analogous with\nthat ofstems iniandu(second declension);itspeculiarity,as\ncompared with them, consists mainlyinthetreatment ofthe\nstemitself, which hasadouble form, fuller inthestrong cases,\nbriefer intheweak ones.\n370.Forms oftheStem. Intheweak cases(excepting\ntheloc.sing.)thestem-final isr,which intheweakestcases,\norbefore avowel-ending,ischanged regularlytor(129).\nBut asregards thestrong cases, thestems ofthisdeclension\nfallintotwo classes :inoneofthem which isverymuch\nthelarger, containingallthenominaagentis, andalsothenouns\nofrelationship naptr andsvdsr, andtheirregular words sirand\nsavyasthr therisvriddhied, orbecomes ar; inthe other,\ncontaining most ofthenouns ofrelationship,with nfandusf,\ntherisgunated, orchangedtoar. Inboth classes, theloc.\nsing, hasarasstem-final.\n371. Endings. These areingeneral thenormal, but\nwith thefollowing exceptions:\nThenom. sing. (masc. andfern.) ends always ina(for original ars).\nThe voc. sing, ends inar.\nThe accus. sing, addsam tothe(strengthened) stem; theaccus.pi.\nhas(likeiandw-stems) nasmasc. ending and sasfern, ending, with the\nrlengthened before them.\n"
+    },
+    {
+        "page": 144,
+        "content": "124 v-NOUNS ANDADJECTIVES.[371\nThe abl.-gen. sing, changesrtowr(orus:169, end).\nThegen. pi. (asin iandw-stems) inserts nbefore am,andlengthens\nthestem-final before it.Buttherofnrmay alsoremain short.\nTheabove aretherules ofthelater language. The older presents cer-\ntain deviations from them. Thus :\nTheending innom. etc.du. is(asuniversally intheVeda) regularly a\ninstead ofdu(onlytencm-forms inRV.).\nThe iofloc. sing,islengthened toIinafewwords :thus, kartdri.\nInthegen. pi.,theRV. hasonce svdsram, without inserted n;and\nnardm instead ofnrndm isfrequent.\nOther irregularities ofnrarethesing. dat. narc, gen. ndras, and loc.\nndri. TheVeda writes always nrndm ingen. pi.,but itsrisinamajority\nofcases metrically long.\nThestem wsr, f.,'dawn', hasthevoc. sing, usar,thegen. sing, usrds;\nand the accus. pi.alsousrds, and loc. sing, usrdm (whichismetrically\ntrisyllabic: usram),asifinanalogy with Iandw-stems. Once occurs usri\ninloc. sing., but itistoberead asiftheregular trisyllabic form, usdri\n(fortheexchange ofsand\n,see181a).\nFrom stfcome only tdras(apparently) andstfbhis.\nInthegen.-loc. du., therisalmost always toberead asaseparate\nsyllable, r,before theending os:thus, pitrtis,etc.Onthecontrary, ndnandari\nisonce tobereadndnandri.\nForneuter forms, seebelow, 378.\n372.Accent. Theaccentuation followscloselytherules\nfor iand ?<-stems :ifonthe final ofthestem,itcontinues, as\nacute, onthecorresponding syllable throughout, exceptinthe\ngen. pi.,where itmaybe(andintheVeda always is)thrown\nforward upontheending ;where, intheweakest cases, fbecomes\nr,theending hastheaccent. Thetwomonosyllabic stems, nr\nandstf,donotshow themonosyllabic accent :thus(besidesthe\nformsalready given above), nfbhis,\n373.Examples ofdeclension. Asmodels ofthis\nmode ofinflection, wemaytakefrom the first class(with\n5TTJ\"arinthestrong forms)thestems^JrTdatr, m., 'giver,\nand^RT svdsr, f., 'sister'; from thesecond class(with ET^f\narinthestrong forms)thestemfarT^r, m.,'father'.\nSingular:\n^THT ^oFfT focTT\ndath svdsapith\nA.\ndataram svdsaram pitdram\n"
+    },
+    {
+        "page": 145,
+        "content": "373]DECLENSION IV., /--STEMS. 125\n1.\nD.\nAb.G.\nL.\nV.\nDual\nN.A. V.\nI.D.Ab.\nG.L.\nPlural :\nN.V.datrk\ndatre\ndatur\ndatdri\ndntar\ndatarau\ndatrbhyam\ndatros\n,\"s\ndataras\nA.\nD.Ab.\nG.\nL.datrn\n3JNPl\ndatrbhis\ndatrbhyas\ndafr'ntim\ndatrsu\nThefeminine stemsvdsra\nsvdsrbhyam\nsvdsrs\nsvdsrbhis\nt \"N,\nsvdsrbhyasfwr\npitrh\npitre\n^3T\npitur\npitdri\npitar\npitdrau\npitrbhyam\npitros\npitdras\npitfn\npitfbhis\npitrbhyas\npiffnhm\npitfsu\natr, 'mother',isinflectedpre-\nciselylike RT^r, exceptingthat itsaccusativepluralis\nmatfs.\n"
+    },
+    {
+        "page": 146,
+        "content": "126V.NOUNS ANDADJECTIVES.[374\nThe peculiarVedic forms have been sufficiently instanced above; the\nonly ones ofother than sporadicoccurrence being thenom. etc.du.datdra,\navdsara, pitdra,andthegen. pi.ofnr,nardm.\n374. Thestem krostf, m., 'jackal' (lit'ly 'howler'),substi-\ntutes inthemiddle cases thecorrespondingforms ofkrostu.\n375.Neuter forms. Thegrammarians prescribeacom-\npleteneuter declension also forbases intr,precisely accordant\nwith that ofvhri ormddhu(above, 339, 341). Thus, forex-\nample:\nSing. Du. Plur.\nN.A. dhatf dhatfni dhatfni\nI. dhatfna dhdtfbhyam dhdtfbhis\nG. dhatfnas dhatfnos dhdtfndm\nV. dhdtr,dhdtar dhdtrnldhdtfni.\nTheweakest cases, however(asof *andw-stems used ad-\njectively:344),areallowed also tobeformed like thecorre-\nspondingmasculine cases :thus, dhatrh etc.\nNosuch neuter forms chance tooccur intheVeda, butthey begin to\nappearintheBrahmanas, under influence ofthecommon tendency (compare\nGerm. Better, Retterin; Fr.menteur, menteuse) togive thenomen agentis a\nmore adjective character, makingitcorrespondingender with thenoun\nwhich it(appositively) qualifies. Thus, wehave inTB. bhartf andjanayitf,\nqualifying antdriksam; andbhartfni andjanayitfni, qualifying ndksatrdni;as,\ninM., grahitfni, qualifying indriydni.\nWhen afeminine noun istobequalifiedinlikemanner, theusual\nfeminine derivative inIisemployed: thus, inTB., bhartryas andbhartrydii,\njanayitryas andjanayitrydu, qualifying dpah and ahoratre;andsuch in-\nstances arenotuncommon.\nTheRV. shows thesame tendency very curiously once intheaccus.\npi.matfn, instead ofmatfs,inapposition with masculine nouns (RV. x.35.2).\nOther neuter forms inRV. aresthattir, gen. sing., dhmatdri,loc.sing. ;\nand forthenom.sing., instead of-tr, afewmore orlessdoubtful cases,\nsthatar, sthatur, dhartdri (Lanman, p.422).\nAdjectives.\n376. a.There arenooriginal adjectivesofthisdeclension :\nforthequasi-adjectival character ofthenouns composing it,see\nabove(378). Thefeminine stem ismade bythesuffix i:thus,\ndatrk, dhatri.\nb.Rootsendinginr(likethose iniandu:345)add\nattomake adeclinable stem, when occurringasfinalmember\nofacompound:thus, 'karmakrt(]/'kr\\ vajrabhft (ybfy), balihft\n(yfo). From somer-roots, also, aremade stems inirandur :\nseebelow, 383a,b.\n"
+    },
+    {
+        "page": 147,
+        "content": "379 DECLENSION V.,CONSONANT-STEMS. 127\nc.Nouns inrasfinals ofadjective compoundsarein-\nflected inthesame manner aswhensimple,inthemasculine\nandfeminine\n;intheneuter, theywould doubtless have the\npeculiarneuter endingsinnom.-acc.-voc. ofallnumbers.\nDeclension V.\nStems ending inConsonants.\n377. Allstems endinginconsonants mayproperlybe\nclassed together,asformingasingle comprehensivedeclen-\nsion; since, though some ofthem exhibitpeculiarities of\ninflection, these have todoalmostexclusivelywith thestem\nitself, andnotwith thedeclensionalendings.\n378. Inthis declension, masculines and feminiiies of\nthesame final areinflectedpreciselyalike;andneuters are\npeculiar (asusuallyintheotherdeclensions) onlyinthe\nnom.-acc.-voc. ofallnumbers.\nThemajorityofconsonantal stems, however, arenot\ninflected inthefeminine, butform aspecial feminine deriv-\native stem in^*(neverin5TTa),byaddingthatendingto\ntheweak form ofthemasculine.\nExceptionsareingeneralthestems ofdivisions AandB\nnamely,theradical stems etc .,andthose inasand isandus .\nForspecial cases, seebelow.\n379. Variations, asbetweenstrongerandweaker forms,\narevery general amongconsonantal stems :either oftwo\ndegrees (strong andweak),orofthree(strong, middle, and\nweakest):seeabove,311.\nThepeculiarneuter forms,accordingtotheusual rule\n(311),aremade intheplural from thestrong stem,insin-\ngular anddual from theweakor,when thegradation\nisthreefold,insingular from themiddle stem,indual\nfrom theweakest.\nAsinthecase ofstems endinginshort vowels(asyatu,\n)&rmi, mddhuni, datfni, etc.),anasal sometimes appearsinthe\n"
+    },
+    {
+        "page": 148,
+        "content": "128V.NOUNS ANDADJECTIVES.[379\nspecialneuter pluralcases which isfound nowhere else inin-\nflection. Thus, from thestems inas, is,us,thenom. etc.pi.\nin-ansi, -insi, -unsi areverycommon atevery period. Accord-\ningtothegrammarians,theradical stems etc.(division A)are\ntreated inthesame way;butexamplesofsuch neuters areof\nexcessive rarityintheolder language ;noVedic text offers one,\nandintheBrahmanas have been notedonly -hunti(AB.vii.2),\n-vrnti(PB.xvi.2.7 etal.),and-bhanji (KB.xxvii.7):itmay\nbequestionedwhether theyarenot lateanalogical formations.\n380.The endingsarethroughoutthosegiven above\n(310)asthe\"normal\".\nBythegeneral lawastofinals(150),the softhenom.sing,\nmasc. and fern, isalways lost; andirregularities oftreatment\nofthe final ofthestem inthis case arenotinfrequent.\nThe gen. and abl.sing,areneverdistinguishedinform\nfrom oneanother norare,byending,thenom. andaccus.pi.:\nbutthese sometimes differ instem-form, orinaccent, orinboth.\n381. Changeintheplace oftheaccent islimited tomon-\nosyllabic stems andtheparticiplesindnt(accented onthefinal).\nFor details, seebelow, under divisions AandE.\nAfewofthecompounds oftherootanc oracshow anirregular shift\nofaccent intheoldest language: seebelow, 410.\n382. Forconvenience andclearness ofpresentation,it\nwillbewell toseparate from thegeneral mass ofconson-\nantal steins certainspecialclasses which show kindredpe-\nculiarities ofinflection, andmaybebestdescribedtogether.\nThus:\nB.Derivative stems inas, is,us;\nC.Derivative stems inan(an,man, van);\nD.Derivative stems inin(in,min, vin);\nE.Derivative stems inant(ant, mant, vant);\n1?.Perfect activeparticiplesinvans;\nG.Comparativesinyas.\nThere remain, then, toconstitute division A,especially\nradical stems, orthose identical inform withroots, to-\ngether with acomparativelysmallnumber ofothers which\nareinflected like these.\nTheywillbetaken upintheorder thus indicated.\n"
+    },
+    {
+        "page": 149,
+        "content": "383]DECLENSION V.A,CONSONANTAL ROOT-STEMS etc. 129\nA.Root-stems, andthose inflected likethem.\n383.Thestems ofthis division maybeclassified as\nfollows :\na.Root-stems, havinginthem nodemonstrable element\nadded toaroot :thus, re, 'verse';gir, 'song' ;pad,'foot';\ndie, 'direction'; mdh(V.), 'great'.\nSuch stems, however,arenotalways precisely identical inform with\ntheroot :thus, vacfromyvac, srdjfromysrj,musfromymus, vriffrom\nyvrafc (?),t/5fromyvas 'shine'; andfrom roots infinal rcome stems\ninirandur :thus, gfr, a-ffr, stir; jur, ttir, dhtir, pur,mur, 3tur; and\npsur fromypsar.\nWith these may beranked thestems with reduplicated root, asciktt,\nyaviytidh, vdnivan, sasydd.\nWords ofthis division inuncompounded use aretolerably frequent in\ntheolder language:thus,inRV. arefound more than ahundred ofthem;\ninAV., about sixty; but inthe classical Sanskrit thepowerofusing any\nroot atwill inthiswayislost, andtheexamplesarecomparatively few.\nInallperiods, however, theadjective use asfinal ofacompoundisvery\ncommon (seebelow, 401).\nb.Stems made bytheaddition of ttoafinal shortvowel\nofaroot.\nNoproper root-stem ends inashort vowel, although there are(above,\n354) examplesoftransfer ofsuch tovowel-declensions;but ioruorr\nadds attomake adeclinable form: thus, -jft, -crtit,-Urt. Roots inr>\nhowever,ashasjustbeen seen,alsomake stems inirorur.\nAsregards thefrequency anduseofthese words, thesame istrue as\nwas stated above respecting root-stems. TheVeda offers examples ofnearly\nthirty such formations, afew ofthem(mtt, rtt, stut, hrtit, vrt,anddytitif\nthis istaken from dyu)inindependentuse. Ofroots inr,kr,dhr, dhvr,\n&Ar, vr,sr\n}spr )hr,hvraddthe t.The roots ga (orgam) andhan also\nmake -gdtand -hdtbyaddition ofthe ttoanabbreviated form ina(thus r\nadhvagdt, dyugdt, navagdt, andsamhdt}.\nAstotheinfinitive useofvarious cases oftheroot-noun inthese two-\nforms, seechap.XIII.\nC.Monosyllabic (also apparently reduplicated)stems notcer-\ntainlyconnectible with anyverbal root inthelanguage, but\nhavingtheaspectofroot-stems, ascontaining notraceable suffix :\nthus, tvdc, 'skin';path,'road';hfd,'heart';dp,'water';as,\n'mouth';kakubh andkakud, 'summit' .\nThirty orforty suchwords arefound intheolder language, andsome\nofthem continue inlater use, while others have been transferred toother\nmodes ofdeclension orhavebecome extinct.\nd.Stems more orlessclearly derivative, butmade with\nsuffixes ofrare oreven isolated occurrence. Thus :\nWhitney, Grammar. 9\n"
+    },
+    {
+        "page": 150,
+        "content": "130 v -NOUNS ANDADJECTIVES.[333\n1.derivatives (V.)from prepositions with thesuffix vat:arvavdt, avdt,\nudvdt, nivdt, paravdt, pravdt, samvdt;\n2.derivatives (V.)intat(perhaps abbreviated fromtati], inafew iso-\nlated forms :thus, upardtat, devdtat, vrkdtat, satydtat,sarvdtat;\n3.other derivatives int,preceded byvarious vowels :thus, vehdt, vahdt,\nsravdt, safcdt, vdghat; ndpat; tadft, divtt, yosit, rohtt, sartt, hartt; martit;\nydkrt, fdkrt; andthenumerals for'30,40,50', trihfdt etc.(475);\n4.stems inad: thus, drsdd, dhrsdd, bhasdd, vandd, cardd;\n5.stems inj,preceded byvarious vowels: thus, trsndj, dhrsdj, sandj ;\nutfj, vanfj, bhurfj, nintj(?); dsrj;\n6.afewstems ending inasibilant apparently formative :thus, jnds,\n-das, bhdsj mds,bhis;\n1.aremnant ofunclassifiablecases, such asvistdp, vtpaf, kdprth,\n,isfdh, prkstidh, raghdtf?).\n384.Gender. The root-stems areregularly feminine as\nnomen actionis, andmasculine asnomenagentis (whichisprobably\nonlyasubstantive useoftheiradjectivevalue :below, 400).\nBut thefemininenoun, without changingitsgender,isoften\nalsousedconcretely:e.g.,druhi.(ydruh,ebeinimical') means\n'harming, enmity', and also 'harmer, hater, enemy' thus\nbordering onthemasculine value. Andsome ofthefeminines\nhave acompletely concrete meaning. Throughthewhole divis-\nion, themasculines aremuch lessnumerous than thefeminines,\nandtheneuters rarest ofall.\nTheindependent neuter stems arehrd(also -hard), ddm, vdr, svbr,\nmds'flesh', ds'mouth', bhds, dos,andtheindeclinables fdmandy<5&:also\nthederivativesydkrt, fdkrt, dsrj.\n385.Strong andweak stem-forms. The distinc-\ntion ofthese two classes offorms isusually made either\nbythepresence orabsence ofanasal, orbyadifference\ninthequantityofthestem-vowel, aslongorshort; less\noften, byother methods.\n386.Anasalappearsinthestrong cases ofthefollow-\ningwords :\na.Compounds having asfinalmember therootacoranc: seebelow,\n407 ff.; b.ThestemJ/MJ, sometimes(V.): thus, nom. sing,ydn (for\nyunk), accus.yCinjam, du. ytinja (but also ytijam andyuja); c.The\nstem-drf, asfinal ofacompound (V.); butonly inthenom.sing, masc.,\nandnotalways:thus, anyadrn, Urn, kidfn, sadrn andpratisadrn:but also\nIdrk, tadfk, svardfk, etc.,- d.Forpathandpums, which substitute more\nextendedstems, and fordant, seebelow,3946.\n387. Thevowel aislengthened instrongcases asfollows :\n"
+    },
+    {
+        "page": 151,
+        "content": "390]DECLENSION V.A,CONSONANTAL ROOT-STEMS etc. 131\na.Oftheroots vac, sac, sap,nabh, pas, inafewinstances(V.),atthe\nend ofcompounds ; b.Ofthe roots vahandsah,butirregularly:see\nbelow, 403 5;c.Ofap'water'(see393) ;also initscompound rltyap ;\nd.Ofpad,'foot': inthecompounds ofthisword, inthelater language,\nthesame lengtheningismade inthemiddle cases also;and inRV.andAV.\nthenom. sing. neut. isboth -patand-pat, while -pddbhis and-patsu occur\nintheBrahmanas; e.Ofnas, 'nose';f.Sporadic cases (V.) are:\nyaj (?),voc. sing. ;pathds and -rapas, accus. pi. ;vdnivanas, nom.pi.The\nstrengthened forms bhdjandrajareconstant, throughallclasses ofcases.\n388. Other modes ofdifferentiation, byelision ofaor\ncontraction ofthesyllable containing it,appearinafewstems :\na.In-han: seebelow, 402; b.Inksam(V.), along with prolong-\nation ofa:thus, ksdmd du.,ksdmaspi.;ksamd instr.sing.,ksdmi loc.\nsing., ksmds abl.sing. ; c.Indvar, contracted(V.)todurinweak cases\n(butwithsome confusion ofthetwo classes); d.Insvar, which becomes\n(RY.)surinweak cases :later itisindeclinable.\n389.Theendingsareasstated above(380).\nRespectingtheir combination with the final ofthestem, as\nwell asthetreatment ofthe latterwhen itoccurs attheendof\ntheword,therules ofeuphonic combination(chap. III.)areto\nbeconsulted;they require much more constant and various\napplicationhere thananywhereelse indeclension.\nAttention maybecalled toafewexceptional cases ofcombination(V.):\nmadbhis andmddbhyds frommas 'month'\n;thewholly anomalous padbMs (RV.\nandVS.:AV.hasalways padbhis} frompad; andsardt andsarddbhyascor-\nrespondingtoanom.pi.sardghas (insteadofsardhas :222). Ddn isappar-\nently forddm, by143, end. Agnfdhisabbreviated from ayni-fdh.\nAccording tothegrammarians, neuter stems, unless theyendinanasal\norasemivowel, take innom. -ace. -voc.pi.astrengthening nasal before the\nfinal consonant. Butnosuch cases from neuter noun-stemsappear ever to\nhavebeenmetwith inuse;and asregards adjective stems ending inaroot,\nseeabove, 379.\n390. Monosyllabic stems have theregular accent ofsuch,\nthrowingthetone forward upontheendingsintheweak cases.\nButtheaccusativepluralhas itsnormal accentuation asa\nweak case,upontheending ;inonlyaminority (hardly more\nthan athird)ofthestems :namelyindatds, pathds, padds,\nnidds, apds, usds, jnasds, puthsds, masds, mahds;andsometimes\ninvacds, srucds, hrutds, sridhds, ksapds, vipds, durds, isds, dvisds,\ndruhds(besidev&casetc.).\nExceptional instances, inwhich aweak case hasthetoneonthestem,\noccur asfollows :sddd, nddbhyas, tana(also tana) and tdne, rantandransw,\nvdnsu, svdni, vfpas, ksdmi, surdandsuras (but sure), dhhas, andvdnas and\nbfhas (invdnaspdti, brhaspdti}. Ontheother hand,astrong case isaccented\n9*\n"
+    },
+    {
+        "page": 152,
+        "content": "132V.NOUNS ANDADJECTIVES. i390\nvak\nvacam\nvace\n'HIS\nvact\nDual:pad\npadam\npada\npade\npadds\npadimarut\nmarutam\nmarutaontheending inmahds, uom. pi.,andkasdm (AV.:perhapsafalsereading).\nAnd presd,instr. sing.,isaccented asifpreswere asimple stem, instead of\npra-fs. Vimrdhdh isofdoubtful character. Forthesometimes anomalous\naccentuation ofstems inacoranc, see410.\n391.Examplesofinflection. Asanexampleof\nnormal monosyllabic inflection, wemaytake thestem\ncHrTvac, f.,'voice'(from y^ff vac,with constantprolong-\nation);ofinflection withstrong andweak stem, q^~pad,\nm., 'foot'; ofpolysyllabic inflection, H^^marut, m.,'wind'\nor'wind-god';ofamonosyllabic root-stem incomposition,\nf^eJH trivrt, 'three-fold',intheneuter. Thus :\nSingular:\nN.V.\nA.\nD.\nAb.G.\nL.\nN.A.V.trwrt\ntrivrt\nmarute\nmarutitrivfta\ntrivfte\nf^^rlTT\ntrivftas\ntrivrti\ni__vacau\nI.D.Ab.\nvagbhytim\nG.L.\nN.V.\nA.vacos\nPlural :padau\"\nH^(H\npadbhyam\npadosmarutau trivrti\nmarudbhyam trivfdbhyam\nmarutostrivftos\nr\"^\nvacas padas\nvacds, vacas paddstrivnti\nmarutas trivfnti\n"
+    },
+    {
+        "page": 153,
+        "content": "395 DECLENSION V.A,CONSONANTAL HOOT-STEMS etc. 133\nvagbhis padbhis marudbhis trivfdbhis\nD.Ab.\nG.\nL.vagbhyds padbhyds mariidbhyas trivrdbhyas\npadam marutam trivrtam vacam\npatsu marutsuC\ntrivftsu\n392. Thestems inirand wr,anduandws,lengthentheir\nvowel (245b)when their final isfollowed byanother consonant,\nand also inthenom.sing, (wherethefollowingsislost):thus,\nfromyzV, f.,fsong', ^Ir(^ijj, giram, girhetc.\n;girau, g'irbhyUm,\ngiros ;giras, girbhis, glrbhyds, giram, girsu (165);and sopur,\npuram, purbhis, pursu; andafis, acisam, acisa, aclrbhis, aclhsuj\nand soon.\n393. Thestemdp, f.,(water', isinflected onlyinthe\nplural, andwith dissimilation ofitsfinal before bhtod(151 d):\nthus, Upas, apds, adbhis, adbhyds, aphvn, apsu.\nButRV. hasthesing, instr. apdand gen. apds. InAV. often, and\ninaninstance ortwo inRV., thenom. and accus.pi.forms areconfused\ninuse, dpas being employedasaccus., and(inaninstance ortwo) apds as\nnominative.\n394. Thestem pums, m., 'man',isvery irregular, sub-\nstituting pumcihsinthestrong cases, andlosingitss(necessarily)\nbefore initial bhofacase-ending, andlikewise(byanalogy with\nthis, orbyanabbreviation akin with thatnoticed at231)in\ntheloc.plural. Thevocative is(inaccordance with that ofthe\nsomewhat similarlyinflectedperfect participles:below, 462a)\npitmaninthe laterlanguage, butpumasinthe earlier. Thus :\npuman, pumansam, pumsa, pumse, pumsds, pumsi, puman; pum-\nansau, pumbhytim, pumsos ;pumansas, pumsds, pumbhis, pumbhyds,\npumshm, pumsu.\nTheaccentuation oftheweak forms,itwillbenoticed,isthat ofatrue\nmonosyllabic stem. The forms with &/i-endings nowhere occur intheolder\nlanguage, nordothey appeartohavebeen cited from the later. Astothe\nretention ofsunlingualized intheweakest cases (whence necessarily follows\nthat inthe loc.pi.), see183.\n395. Thestem path, m., 'road', isdefective indeclension,\nforming onlytheweakest cases, while thestrongaremade from\npdnthaorpdnthan, and themiddle frompathl:seeunder an-\nstems, below, 433.\n"
+    },
+    {
+        "page": 154,
+        "content": "134 v-NOUNS ANDADJECTIVES.[396\n396. Thestem ddnt, m., Hooth',isperhapsofparticipial\norigin, andhas,likeaparticiple,theforms ddntandddt,strong\nandweak :thus(V.), ddn, ddntam, data, etc.;datds ace.pi.\netc. But inthemiddle cases ithasthemonosyllabic andnot\ntheparticipialaccent :thus, dadbhis, dadbhyds.Innom.pi.\noccurs also -datas instead of-dantas. Bythegrammarians, the\nstrongcases ofthisword arerequiredtobemade from ddnta.\n397.Anumber ofother words ofthis division arede-\nfective, making partoftheir inflection from stems ofadiffer-\nentform.\nThus, hfd, n., 'heart', mdhs ormds, n., 'meat', mds, m., 'month',\nnds, f.,'nose', nip, f.,'night' (notfound inthe older language), pft, f.,\n'army',are saidbythegrammarians tolack thenom. ofallnumbers and\ntheaccus. sing, anddu.(the neuters, ofcourse, theace.pi. also), making\nthem respectivelyfrom hfdaya, mdnsd, mdsa, ndsikd, nifd, pftand. Butthe\nusage intheolder languageisnotentirely inaccordance with thisrequire-\nment :thus,wefindmds, 'flesh', accus. sing.; mds, 'month', nom. sing.;\nandndsa, 'nostrils',du.From pftoccurs only the loc.pi.prtsti and(RV.,\nonce) thesame casewith double ending, prtstisu.\n398.Ontheother hand, certain stems ofthis division,\nallowed bythegrammarians afullinflection, areused tofillup\nthedeficiencies oftho'pe ofanother form.\nThus, dsrj fn., 'blood', fdkrt, n., 'ordure', ydkrt, n., 'liver', do's, n.\n(also m.), 'fore-arm', have beside them defective stems inan: seebelow,\n432. Ofnone ofthem, however,isanything butthenom. -ace. sing, found\nintheolder language, andother cases later arebutvery scantily represented.\nOfds,n.,'mouth', andtid, 'water', only acase ortwo arefound, in\ntheolder language, beside asdn anddsya, anduddn andtidaka (432).\n399. Some ofthealternative stems mentioned above areinstances of\ntransition from theconsonant toavowel declension :thus, ddnta, masa.\nAnumber ofother similar cases occur, sporadicallyintheolder language,\nmorecommonly inthe later. Such arepdda, -mdda, -ddfa, bhrdjd, vistdpa,\ndvdra anddura, pura, dhura, -drya, ndsa, nidd, ksipd, ksapd, dfd, and\nperhapsafew others.\nAfewirregular stems will find amore proper place under thehead of\nAdjectives.\nAdjectives.\n400.Original adjectives havingtheroot-form arecompar-\nativelyrareeven intheoldest language.\nAbout adozen arequotable from theRV.,forthemost part only ina\nfewscattering cases. Butmah, 'great',iscommon inRV., thoughitdies\noutrapidly later. Itmakes aderivative feminine stem, mahi, which con-\ntinues inuse, asmeaning'earth' etc.\n"
+    },
+    {
+        "page": 155,
+        "content": "403]DECLENSION V.A,CONSONANTAL ROOT-STEMS etc. 135\n401. Butcompound adjectives, havingaroot asfinal\nmember, with thevalue ofapresent participle,areabundant\ninevery periodofthelanguage.\nPossessiveadjective compounds, also, ofthesame form,\narenotveryrare :examplesareyatdsruc,'with offered bowl';\ns&ryatvac,'sun-skinned';cdtuspad, 'four-footed;suhurd, 'kind-\nhearted, friendly'; rityap (i.e.riti-ap], 'having streamingwaters';\nsahdsradvar, 'furnished with athousand doors'.\nThe inflection ofsuchcompoundsislike that ofthesimple root-stems,\nmasculine andfeminine being throughout thesame, andtheneuter varying\nonly inthenom.-acc.-voc. ofallnumbers.\nOnly rarelyisaderivative feminine stem inIformed :intheolder\nlanguage, onlyfrom thecompounds with acorone(407 ff.),those with han\n(402), andthose with pad,asekapadi, dvipddl.\nIrregularitiesofinflection appearinthefollowing:\n402. The root han, 'slay',asfinal ofacompound,isin-\nflected somewhat like aderivative noun inan(below, 420ff.),\nbecoming hainthenom.sing., andlosingitsninthemiddle\ncases and itsaintheweakest cases(butonly optionallyinthe\nloc.sing.). Further, when thevowel islost, hincontract with\nfollowing nreverts toitsoriginal gh.Thus :\nSingular. Dual.\nA. vrtrahdnam\n"
+    },
+    {
+        "page": 156,
+        "content": "136 VI.NOUNS ANDADJECTIVES.[403\nable)issaid tobefurther irregularinmakingthenom.sing,\ninvasandthevocative invasorvas.\nInthe earlier language, only strong forms ofcompounds with vahhave\nbeen found tooccur: namely, -vdt, -vdham, -vdhdu or-vdha, and-vdhas.\nTS.hastheirregular nom. sing, pasthavdt.\n404. Ofvery irregularformation and inflection isone\ncommon compoundofvah,namely anadvdh(anas -\\-vah, 'burden-\nbearing'or'cart-drawing':i.e.'ox').Itsstem-form inthe\nstrongcases isanadvah, intheweakest anadu/i, and inthe\nmiddle anadud(perhaps bydissimilation fromanadud}. Moreover,\nitsnom. and voc.sing,aremade invanandvan(asiffrom\navant-stem}.Thus :\nSingular. Dual. Plural.\nN. anadvdn )3~-i- anadvdhas\n,\\anadvahau\nA.anadvaham) anaduhas\nI. anaduha\n|anadudbhis\nD. anaduhe {.anadudbhyam \\\n\\anadudbhyas\n_,'\nIanaduhas\nG. }t anaduham\n}anaduhos\nL. anaduhi| anadutsu\nV. dnadvan anadvahau dnadvahas\nAnadudbhyas (AV., once)istheonly middle case-form quotable from\ntheolderlanguage.\nThecorresponding feminine stem(ofvery infrequent occurrence)iseither\nanaduhi(QB.)oranadvahl(K.).\n405. The rootsah, 'overcome', hasintheVeda adouble irregularity-\nitssischangeableto*even after ana-vowel asalsoinitssingle oc-\ncurrence asanindependent adjective (RV.,tvdrhsdt)while itsometimes\nremains unchanged after an iorw-vowel; and itsaiseither prolongedor\nremains unchanged,inboth strong andweak cases. Thequotable forms are :\n-sdt, -sdham or-sdham or-sdham, -sdha, -sdhe or-sdhe, -sdhas or-sdhas\nor-sdhas; -sdha(du.);-sahas or-sdhas.\n406. Thecompound avaydj (j/yaj,'makeoffering'),'acertainpriest'or\n(BR.)'acertainsacrifice', issaid toform thenom. and voc. sing, avayas,\nand tomake itsmiddle cases from avayds.\nItsonly quotable form isavayds,f.(RV.andAV., each once).Ifthe\nstein isaderivative fromava-{-yyaj, 'conciliate', avayasisprobably from\nava-j-yyd, which hasthesame meaning.\n407.Compounds with ancorac. The root acor\nancmakes, incombination withprepositionsandother words,\naconsiderable class offamiliarly usedadjectives,ofquite irreg-\nularformation and inflection, insome ofwhich italmost loses\nitscharacter ofroot, andbecomes anendingofderivation.\nApart oftheseadjectives have only twostem-forms :a\nstronginanc(yielding an,from anks, innom.sing, masc.),\n"
+    },
+    {
+        "page": 157,
+        "content": "410]DECLENSION V.A,CONSONANTAL ROOT-STEMS etc. 137\nandaweak inac;othersdistinguish from themiddle inaca\nweakest stem inc,before which theaiscontracted with apre-\ncedingioruinto ioru.\nThefeminine ismade byaddingItothestem-form used\nintheweakest cases, and isaccented likethem.\n408.Asexamplesofinflection wemaytakeprhnc, 'forward,\neast';pratydnc, 'backward, west';visvanc, 'going apart1\n.\nSingular:\npratydn pratyak vfsvan vfsvak\npratydncam pratyak vfsvancam vfsvak\npratlcd vfsucd\npratice vfsuce\npratlcds visucas\npratlci vCsuciN.V.\n"
+    },
+    {
+        "page": 158,
+        "content": "138 v -NOUNS ANDADJECTIVES.[410\nthetoneandsometimes theother, without anyapparentreason forthe dif-\nference. Ifthecompoundisaccented onthe final syllable, theaccent is\nshifted inRV. totheendingintheweakest cases providedtheir stem shows\nthecontraction toiorit:thus, prdca, arvdca, adhardcas, butpratlcd, anucds,\nsamici. ButAV. and later texts usually keep theaccent upon thestem:\nthus, praticl, samici, anUcl (RV.haspraticlm once). Thechange ofaccent\ntotheendings, andeven inpolysyllabic stems,isagainstallusual analogy.\nB.Derivative stems inas, is,us.\n411.Thestems ofthisdivision areprevailingly neuter;\nbutthere arealso afewmasculines, andasinglefeminine.\n412. The stems in^fasarequite numerous, and\nmostly made with the suffix 5fHas(asmall number also\nwith clTTtasand^nas,andsome areobscure);theothers\narefew,andalmost allmade with the suffixes^jsand\n3H\nV413. Their inflection isalmostentirely regular. But\nmasculine andfeminine stems in^Haslengthenthevowel\noftheendinginnom.sing.; andthenom.-acc.-voc.pi.neut.\nmake thesameprolongation (of5faor^ior3u)before\ntheinserted nasal(anusvara).\n414.Examples ofdeclension. Asexamples we\nmaytakeJHTmdnas, n.,'mind'; ^H^H dngiras, m.,'An-\n*X\"v*S\ngiras'; ^TJ^T havis, n.,'libation'.\nSingular:\nN.\nmdnasdngirashavis\nmdnas dngirasamhavis\nmdnasadngirasahavisa\nmdnasedngirasehavise\nAb.G. H^HHJbH^-^HH ^fc^^^\nmdnasasdngirasashavisas\n"
+    },
+    {
+        "page": 159,
+        "content": "416] DECLENSION V.B,DERIVATIVE STEMS INas, is,us. 139\nL.\nV.\nDual:\nN.A.V.\nI.D.Ab.\nG.L.mdnasidngirasihavisi\nmdnasngrashdvis\nmdnasidngirasauhavisi\nmanobhyam angirobhyam havirbhyam\nmdnasosdngirasoshavisos\nPlural :\nN.A.V.\nmdnahsi\nI.\nD.Ab.\nG.\nL.manobhisdngirasas\ndngirobhis\nmdnobhyas dngirobhyas\nmdnasam rasamhavirbhis\nhavlrbhyas\nhavisam\no\nmdnahsudngirahsuhavihsu\nInlike manner, tftHH cdksus, n., 'eye', forms\nrO\"X'v3\ncdksusa, TJ^I^IH cdksurbhyam, ri^fq cdksunsi, and soon.O \"X 6\\\n415.Vedic irregularities, a.Themasc. and fern. du.ending d\ninstead ofdu isasusual elsewhere; b.The fern, usds, 'dawn', often\nprolongsitsaintheotherstrong cases(besides nom. sing.): thus, usdsam,\nusdsd, usasas. Ininstr.pi.occurs (RV., once) usddbhis instead ofusdbhis\n(only quotable example ofamiddlecase). From tofds isonce found (RV.)\ninlikemanner thedu. to$tisd; c.JanHs hasthenom. sing. masc. janUs,\nlikeanas-stern; d.From svdvas andsvdtavas occur inRV.nom. sing,\nmasc. invan; e.One ortwoapparentlycontracted forms thus, vedhdm\nforvedhdsam, andsurddhds forsurddhasas, nom.pi.aremet with.\n416. Thegrammarians regard ufdnas, m.,asregularstem-form ofthe\nproper name noticed above (355 a),butgiveittheirregular nom. u^dnd and\nthevoc.ucanas orufana oruyanan. Forms from theas-stem, even nom.,\naresometimes metwith inthelater literature.\n"
+    },
+    {
+        "page": 160,
+        "content": "140V.NOUNS ANDADJECTIVES.[41e\nAstoforms from as-stems todhan oraltar andudhan orudhar, see\nbelow, 430.\nAdjectives.\n417.Afewneuter nouns inaswith accent ontheradical\nsyllablehave corresponding adjectivesorappellativesinas,with\naccent ontheending: thus,forexample, apas, 'work', apds,\n'active'; tdras, 'quickness', tar&s, 'quick'; ydcas, 'beauty', yacds,\n'beauteous'. Afewother similaradjectivesastavds, 'mighty',\nvedhds, 'pious'arewithoutcorrespondingnouns.\nOriginal adjectivesinisdonotoccur. But inusarefound\nasmany adjectivesasnouns(abouttenofeachclass); and in\nseveral instances adjective andnoun stand sidebyside, without\ndifference ofaccent such asappearsinthestems inas: e.g.\ntdpus,'heat' and 'hot';vdpus, 'wonder' and 'wonderful' .\nV\"418. Adjective compounds having nouns ofthis division\nasfinalmember areverycommon :thus, sumdnas, 'favorably\nminded';dirghayus, 'long-lived' ;cukrdgocis, 'havingbrilliant\nbrightness'. Thestem-form isthesame for allgenders, and\neach genderisinflected intheusual manner, thestems inas\nmakingtheir nom.sing. masc. and fern, inas(like dngiras,\nabove). Thus, from sumdna$, thenom. andaccus. areasfollows :\nsingular. dual. plural,\nm. f. n. m. f. n. ra. f. n.\nN.sumdnas -nas )\nisumanasau -nasi sumanasas -nansi\nA.sumanasam -nas)\nandtheother cases (save thevocative) arealike inallgenders.\nFrom dirghdyus,inlikemanner:\nN. dirghdyus\nA.dlrghdyusam -yus\nI. dlrghdyusa dlrghdyurbhyam dlrghdyurbhis\netc. etc. etc.\n419. Thestemone/ids,'unrivalled' (definedasmeaning 'time' inthe\nlater language), forms thenom. sing. masc. and fern, aneha.\nC.Derivative stems inan.\n420.Thestems ofthis division arethosemade bythe\nthree suffixes EFTan,Rman. and^Tvan.togetherwith a\nfewofmorequestionable etymologywhich areinflected like\nthem.Theyaremasculine andneuteronly.\n\\421.Thestem hasatripleform. Inthestrongcases\nofthemasculine, thevowel oftheendingisprolongedto>.dirghdyusau -yusl dlrghdyusas -yunsi\n"
+    },
+    {
+        "page": 161,
+        "content": "424 DECLENSION V.C,DERIVATIVE STEMS INan. 141\nETT<i;intheweakest cases itisingeneralstruck out al-\ntogether;inthemiddlecases, orbefore acase-ending be-\nginningwith aconsonant, thefinal^nisdropped. The\n^nisalso lostinthenom.sing,ofbothgenders (leaving\n5TT asfinal inthemasculine,faintheneuter).\nThepeculiarcases oftheneuter follow theusual ana-\nlogy (311):thenom.-acc.-voc.pi.have thelengthening\nto5fTa,asstrong cases; thenom.-acc.-voc. du., asweakest\ncases, have thelossof5fa but thisonly optionally,not\nnecessarily.\nInthe loc.sing, also, theamaybeeitherrejectedorre-\ntained(comparethecorresponding usage with r-stems :373).And after themorvofman orvan,when these arepreceded\nbyanother consonant^ theaisalways retained, toavoid atoo\ngreat accumulation ofconsonants.\n422.Thevocative sing,isinmasculines thepurestem\n;\ninneuters, either this orlikethenominative. The restof\ntheinflectionrequires nodescription.\n423.Astoaccent,itneedsonlytoberemarked thatwhen,\nintheweakestcases, anacute dofthe suffix islost, thetone\nisthrown forward upontheending.\n424.Examples ofdeclension. Assuchmaybe\ntaken^TsR rajan, m.?'king'; ttirH^ atmdn, m., 'soul, self;\nS(H*Snaman, n.,'name'. Thus:\n\"^ '/'\nSingular:\nN.^TsTTtjirHi *rm\nraja atma nama\nrajanam atmanam nama\nrajna atmdna namna\nrajne atmdne namne\nraj'nas atmdnas namnas\n"
+    },
+    {
+        "page": 162,
+        "content": "142V.NOUNS ANDADJECTIVES.[424\nrajni, rajaniatmdniL.\nV.\nDual:\nN.A.V.\nrajanau\nI.D.Ab.{(sc-UM^\nrajabhyam\nG.L.fiifr^\nrajnos\nPlural:\nN.^isiMH^\nrajanas\nrajnas\nrajabhis\nD.Ab.\n{|sl^6|H^\nrajabhyas\nG.\nL.tman\natmhnau\natmhnas\natmdnas\natmdbhis\natmdbhyas\nJbllrHHIH\natmdnam\nMlrMH\natmdsunamni, namani\nnaman, nama\nnamnl, namani\natmdbhyam namabhyam\nMrH*TlH^\na^wiawos\nnamani\nSIHIM\nnamabhis\nHIH^UH^\nnamabhyas\nSIHH\nro/a?^\nTheweakest cases ofmurdhdn, m., ^head', would beac-\ncented murdhna, murdhne, murd/mos, murdhnds(ace. pi.),murdhnam,\netc.;and soinallsimilar cases(loc. sing., murdhni ormurdhdni].\n425.Vedic Irregularities, a.Here, aselsewhere, theending of\nthenom.-acc.-voc. du.masc. isusually ainstead ofaw.\nb.Thebriefer form (with ejected a)ofthe loc. sing., and oftheneut.\nnom. etc. du.,isalmost unknown totheolder language. RV. writes once\n$atadavni, but itistoberead fataddvani; and afewsimilar cases occur in\nAV. IntheBrahmanasalso, such forms asdhamani andsamani aremuch\nmorecommon than such asahniandlomni.\nc.Butthroughout bothVeda andBrahmana, anabbreviated form ofthe\nloc.sing., with theendingiomitted, oridentical with thestem, iscommon\n(inRV., ofconsiderably morefrequent occurrence than theregular form):\nthus, murdhdn, kdrman, ddhvan, beside murdhdni etc.'\n"
+    },
+    {
+        "page": 163,
+        "content": "429]DECLENSION V.C,DERIVATIVE STEMS INan. 143\nd.Inthenom.-acc. pi.neut., also, anabbreviated form iscommon,\nending inaor(twiceasoften) a,instead ofani :thus, brdhma andbrdhma,\nbeside brdhmani :compare thesimilar series ofendings from a-stems, 329.\ne.From afewstems inman ismade anabbreviated inst.sing., with\nlossofmaswell asofa:thus, mahina, prathina, varina, dana, prena, &/mna,\nformahimnd etc.Anddraghmd (RV., once)isperhaps (Grassmann)for\ndrdghmdnd.\nf.Other oftheweakest cases than the loc. sing, aresometimes found\nwith theaofthe suffix retained :thus,forexample, bhtimana, ddmane,\nydmanas, uksdnas(accus. pi.),etc. Inthe infinitive datives trdmane,\nvidmdne, ddvdne,etc. theaalways remains. Still more numerous are\ntheinstances inwhich the a,omitted inthewritten form ofthetext, is,\nasthemetre shows, toberestored inreading.\ng.The voc. sing, invas,which istheusual Vedic form from stems\ninvant (below, 453b),isfound alsofrom afew invan, perhaps bya\ntransfer totheuarU-declension: thus, rtauas, evayavas, khidvas(?),prataritvas,\nmdtarifvas,vibhavas.\nForwords ofwhich theaisnotmade longinthestrong cases,seethe\nnext paragraph, b.\n426.Afew steins donotmake theregular lengtheningof\nainthestrongcases(exceptthenom.sing.).Thus :\na.Thenames ofdivinities, pusdn andaryamdn: thus, pusa, pusdnam,\npuma,etc.\nb.IntheVeda, uksdn, 'bull'(butRV.uksdnam once); yosan,'maiden':\nvfsan, 'virile,bull' (butvfsanam andvfsanas arealsomet with); fmdn,\nabbreviation ofdtmdn;andtwoorthree other scattering forms :anarvdnam,\njemana. And inanumber ofadditional instances, theVedic metre seems\ntodemand awhere aiswritten.\n427. Thestemsqvdn, m., 'dog', andyuvan, 'young',have\nintheweakest cases thecontracted form cunandyhn (with\nretention oftheaccent);inthestrong andmiddle cases they\nareregular. Thus, cvti, cvftnam, cuna, cune, etc., cvdbhyam,\ncvabhis, etc.\nIndual, RV. hasonceyUnaforyuvana.\n428.Thestemmaghdvan, 'generous' (later, almost exclusively\naname ofIndra),iscontracted intheweakest cases tomaghon:\nthus, maghdva, maghdvanam, maghona, maghone,etc.\nTheRV. hasonce theweak formmaghdnas innom.pi.\nParallel with this isfound thestemmaghdvant (division E);andfrom\nthe latter alone intheolder language aremade themiddle cases: thus,\nrnaghdvadbhis, maghdvatsu,etc.(notmaghdvabhis etc.).\n429.Anumber ofan-stems aremore orless defective,\nmaking apartoftheir forms from other stems. Thus :\n"
+    },
+    {
+        "page": 164,
+        "content": "144 V.NOUNS ANDADJECTIVES.[430\n430. a.Thestem dhan, n.,'day',isinthelaterlanguage\nused onlyinthestrong andweakest cases, themiddle(with\nthenom. sing., which usually follows theiranalogy) coming\nfrom dhar orahas :namely, dharnom. -ace.sing., dhobhyam,\ndhobhis, etc.(PB.hasaharbhis};butdhnaetc., dhni ordhani,\ndhni ordhani, dhani(and,inV., aha).Incomposition, only\nahar orahas isused aspreceding member;asfinalmember,\nahar, ahas, ahan, orthederivatives aha, alma.\nInthe oldest language, themiddle casesdhabhis, dhabhyas, dhasu\nalso occur.\nb.Thestem Udhan, n., 'udder', exchanges inlikemanner, inthe old\nlanguage,with udhar andudhas, buthasbecome later anas-stern only\n(exceptinthe fern, udhnl ofadjective compounds): thus, udhar orudhas\n(soRV. :AV. only thelatter), udhnas, udhan orudhani, udhabhis, udhahsu.\n431. Theneuter stems aksdn, 'eye', asthdn, 'bone', dadhdn,\n'curd', sakthdn, 'thigh', form inthe later language only the\nweakest cases, aksna, asthne, dadhnds, sakthni orsakthdni, and\nsoon;therestoftheinflection ismade from stems ini,dksi\netc. :seeabove, 343 f.\nIntheolder language,other cases from thean-stems occur :thus,\naksdni\n.andaksdbhis;asthdni, asthdbhis, andasthdbhyas ,sakthdni.\n432. Theneuter stems asdn, 'blood', yakdn, 'liver', cakdnr\n'ordure', asdn, 'mouth', uddn, 'water', dosdn, 'fore-arm', yusdn,\n'broth', arerequiredtomake their nom.-acc.-voc. inallnumbers\nfrom theparallel stemsdsrj, ydkrt, qdkrt, asya, udaka(inolder\nlanguage udakd], dos,yusa, which arefullyinflected.\nEarlier occurs alsothedual dosdnl.\n433. Thestem pdnthan, m., 'road',isreckoned inthe\nlater languageasmakingthecompletesetofstrong cases, with\ntheirregularitythat thenom.-voc.sing, adds as.The corre-\nsponding middle cases aremade frompathi, andtheweakest\nfrompath. Thus :\nfrompdnthan pdnthas, pdnthanam; pdnthanau; pdnthanas;\nfrom pathi pathibhyam; pathibhis, pathlbhyas, pathlsu ;\nfrom path patha, pathe, pathds, pathi; pathos; pathds\n(accus.), patham.\nIntheoldest language (RV.), however, thestrong stem isonlypdnthd:\nthus, pdnthas, nom.sing. ;pdntham, ace. sing. ;pdnthas, nom.pi. ;and\neven inAV., pdnthanam andpdnthanas arerarecomparedwith theothers.\nFrom patht occur also thenom. pi.pathdyas andgen. pi.pathmdm. RV.\nhasoncepathds, ace.pi.,with long a.\n434. The stems mdnthan, m., 'stirring-stick', and rbhuksdn, m.,an\nepithetofIndra, aregiven bythegrammariansthesame inflection with\npdnthan; butonly afew cases have been found inuse. InV.occur from\n"
+    },
+    {
+        "page": 165,
+        "content": "439]DECLENSION V.C,DERIVATIVE STEMS INan. 145\ntheformer the ace.sing, mdntham, andgen. pi.mathlndm (like thecor-\nrespondingcases from pdnthan) ;from thelatter, thenom. sing, rbhuksds and\nvoc.pi.rbhuksas,like thecorresponding Vedic forms ofpdnthan; but also\nthe ace. sing, rbhuksdnam andnom.pi.rbhuksdnas, which are after quite\nanother model.\nAdjectives.\n435. Original adjective stems inanarealmostexclusively\nthose invan, asydjvan, 'sacrificing', sutvan, 'pressingthesoma',\njitvan, 'conquering'. Thestem ismasc. and neut.only (two\northreesporadiccases ofitsuse asfern, occur inRV.);the\ncorrespondingfern, stem ismade invari: thus, yajvari, jitvari.\n436.Adjective compounds havinganoun inanasfinal\nmember areinflected after themodel ofnoun-stems;andthe\nmasculine forms aresaid tobeallowed inusealso asfeminine;\nbutusuallyaspecialfeminine ismade byaddingItotheweakest\nform ofthemasculine stem :thus, durnamnl, somarajm.\n437. Butnouns inanoccuring asfinalmembers ofcompounds often\nlose then,orsubstitute astem inaforthat inan: thus, -raja, -adhva,\n-aha; thecorrespondingfeminine isina.And feminines ina,replacing\nan.areallowed tobewidely formed inthecompounds ofthis division.\nTheremainingdivisions oftheconsonantal declension are\nmade upofadjectivestems only.\nD.Derivative stems (adjective) inin.\n438.Thestems ofthis division arethose formed with\nthesuffixes^fin,jrprmin,andf^R vin.Theyaremascu-\nlineand neuteronly;thecorrespondingfemininebeing\nmade byadding ^i.\nThestems ininarenumerous, since almost anynoun ina\ninthelanguage mayform apossessivederivativeadjectivewith\nthis suffix: thus, bdla, 'strength', balin, m.n., balirii, f.,'pos-\nsessing strength, strong'. Stems invin,however, arevery few,\nandthose inmm stillfewer.\n439. Their inflection isquite regular, exceptthatthey\nlose their final^ninthemiddle cases(beforeaninitial\nconsonant oftheending), and alsointhenom.sing.,where\nthemasculinelengthensthe<ibywayofcompensation.\nThe voc.sing,isinthemasculine thebare stem; inthe\nneuter, either thisorthenominative.\nWhitney, Grammar. 10\n"
+    },
+    {
+        "page": 166,
+        "content": "146V.NOUNS ANDADJECTIVES. 439\nInallthese respects,itwillbenoticed, thew-declension\nagreeswith theaw-declension;itdiffers from thelatter onlyin\nnever losingthevowel oftheending.\n440.Exampleofinflection. Assuchmaybe\ntaken srf^FT balm, 'strong'.Thus:\nDual.x\nSingular,\nm. n.Plural,\nm. n.\nN.\nA.\nI.\nD.\nAb.\nG.balk bali\n(khlH GJM\nbalinam bait\nbalina\nbalinesrf^RT\nbalinau balini balinas bafini\nbalinasbalibhis\nbalibhy\nbalibhyas\nL.\nV.SIMM\nbalinibalinosbalinam\nsJM^j\nbalisu\nsjidHi <MHH!^M^H^\nb&lin b&lin, bali balinau balini balinas balini\nThe derived feminine stem inini isinflected, ofcourse,\nlikeanyother feminine inderivative i(365).\n441. There arenoirregularitiesintheinflection ofin-\nstems, ineither theearlier languageorthelater exceptthe\nusual* Vedic dual endinginainstead ofait.\nE.Derivative stems (adjective) inant (orat}.\n442. These stems fallintotwosub-divisions: 1.those\nmade bythe suffix IHant(or?IrTat\\being,with avery\nfewexceptions, activeparticiples, presentand future;\n2.those made bythepossessivesuffixes J7Hniant and\nSffi^vant(orT^rnatand cfrT vat}. Theyaremasculine and\nneuteronly; thecorrespondingfeminine ismade byad-\nding?I.\n"
+    },
+    {
+        "page": 167,
+        "content": "447]DECLENSION V.E,DERIVATIVE STEMS INant. 147\n1.Participles inantorat.\n443.Thestemhasingeneraladouble form, astronger\nand aweaker, ending respectivelyin5JfTantandEFT ^-\nTheformer istaken inthestrongcases ofthemasculine,\nwith, asusual, thenom.-acc.-voc.pi.neuter; thelatter is\ntaken byalltheremainingcases.\nBut,inaccordance with therule fortheformation ofthefeminine stem\n(below, 449),thefutureparticiples, and thepresent participlesofverbs of\nthetod-class oraccented a-class(752), andofverbs ofthead-class orroot-\nclass ending ina,arebythegrammarians allowed tomake thenom.-acc.-\nvoc. du.from either thestronger ortheweaker stem;andthepresent par-\nticiples from allother present-stems ending inaarerequired tomake the\nsame from thestrong stem.\n444. Those verbs, however, which inthe3dpi.pres.\nactive lose^noftheusual ending f^fnti(550)lose it\nalso inthepresent participle,andhave nodistinction of\nstrong andweak stein.\nSuch aretheverbs forming their present-stem byreduplication without\nadded a:namely, those ofthereduplicating or/m-class (655) andtheinten-\nsives (1012): thus, fromyhu, present-stem juhu, participle-stem jtihvat ;inten-\nsive-stemjohu,intensiveparticiple-stem jtihvat. Further, theparticiples of\nroots apparently containing acontractedreduplication:namely, cdksat, ddfat,\ndasat, fffsat, sdfcat] andtheaoristparticiple dhdksat. Vavrdhdnt(RV., once),\nwhich hasthennotwithstandingitsreduplication, comes, likethedesiderative\nparticiples (1032), from astem ina:compare vavrdhdnta, vavrdhdsva.\nEven these verbs areallowed bythegrammarians tomake thenom.-\nacc.-voc.pi.neut. inanti.\n445. The inflection ofthese stems isquite regular.\nThenom.sing. masc. comes toendinERanbytheregu-\nlar(150)lossofthetwo final consonants from theetymo-\nlogical formQff^ants.The vocative ofeachgenderislike\nthenominative.\n446. Steins accented onthe finalsyllable throw theaccent\nforward uponthecase-endingintheweakest cases (not inthe\nmiddlealso).\nInthedual neut.(asinthefeminine stem) from such participles,the\naccent isdntl ifthenisretained,ati ifitislost.\n447.Examples ofdeclension. Assuchmayserve\n10*\n"
+    },
+    {
+        "page": 168,
+        "content": "148V.NOUNS ANDADJECTIVES. 1447\nJ^bhdvant, 'being';5^addnt, 'eating'; sT^r^juhvat,\n'sacrificing'.Thus :\nN.\nA.\nI.\nD.\nAb.G.\nL.\nV.\nN.A.V.\nI.D.Ab.\nG.L.\nN.V.\nA.\nI.\nD.Ab.Singular:/\n^opT ^cfrT51^*T Ef^rT\nbhdvan bhdvat addn addt juhvat juhvat\nbhdvantam bhdvat addntamaddt juhvatamjujivat\nbhdvata adata juhvata\nbhdvate adate\nbhdvatas adatds\nHc(fri t^lci\nbhdvati adati\nbhdvan bhdvat\nDual:juhvate\njuhvatas\njuhvati\njuhvat\nbhdvantau bhdvanti addntau adati juhvatau juhvati\nbhdvadbhyam adddbhyam juhvadbhyam\njuhvatosbhdvatos\nPlural :\nbhdvantas bhdvanti addntas addntijuhvatas juhvati\nbhdvatas bhdvanti adatds addntijuhvatas juhvati\nbhdvadbhis adddbhisjuhvadbhis\nbhdvadbhyas adadbhyas juhvadbhyas\n"
+    },
+    {
+        "page": 169,
+        "content": "449] DECLENSION V.E,DERIVATIVE STEMS INant. 149\nG.\nbhdvatam adathmjuhvatam\nL.H^Tfg ^f f^rH\nbhdvatsu addtsujuhvatsu\nThefutureparticiple bhavisydnt mayform innom. etc.dual\nneuter eitherbhavisydntiorbhavisyatl; tuddnt, either tuddntl or\ntudati; yhnt (}/ya),eitheryantloryatk. Andjuhvat, innom.\netc.plural neuter, maymake alsojuhvanti (beside juhvati, as\ngivenintheparadigm above).\nBut these strong forms(aswell asbhdvantl, du.,and itslikefrom\npresent-stemsinunaccenteda)arequite contrary togeneral analogy, and of\nsomewhat doubtful character. NoVedic example ofthem isfound; norhave\nthey been noticed anywhere intheolder language. The cases concerned,\nindeed, would beeverywhere ofrare occurrence.\n448. TheVedic deviations from themodel asabove given arefew.\nThedual ending du isonly one sixth ascommon asa.Anomalous accent\nisseen inacase ortwo :acoddte andrathiraydtam. The only instance in\nV.ofnom. etc.pi.neut. issdnti, with lengtheneda:compare theforms\nin-manti and -vanti, below, 454 c.\nx449. Thefeminineparticiple-stem,asalready stated,\nismade byadding ^itoeither thestrongortheweak\nstem-form ofthemasc.-neut. The rules astowhich ofthe\ntwoforms shall betaken arethesame with thosegiven\nabove respectingthenom. etc.dual neuter; namely:\na.Participles from tense-stems endinginunaccented a\nadd itothestrong stem-form, ormake their feminine inanfi.\nSuch arethebhuorunaccented a-class and thedlvorj/a-classofpre-\nsent-stems(chap. IX.), andthe desideratives and causatives(chap. XIV.):\nthus, fromybhu (stem bhdva), bhdvantl; fromydlv (stem divyd), divyantl;\nfrom bubhusa andbhdvdya (desid. andcaus. of|/&/m),bubhusanti and\nbhdvdyantl.\nExceptionstothis rule are rare. RV. has tdksati and jaratl; Bopp\n(Gr., 530) quotesafewcases from theNala. TheAV. jlvanti, with irregular\naccent,isdoubtless toberegardedasaproper name.\nb.Participles from tense-stems inaccented dmayaddthe\nfeminine-signeither tothestrongortotheweak stem-form,\normaymake their feminines indnti orinatl(withaccent as\nnerenoted).\nSuch arethepresent-stemsofthetud oraccented d-class (chap. IX.),\nthes-futures(chap. XII.), and thedenominatives (chap. XIV.): thus, from\nytud (stem tudd], tuddntl ortudati;from bhavisyd (fut.ofybhu}, bhavisydnti\norbhavisyati; from devayd (denom.ofdcvd), devaydntl ordevayati.\n"
+    },
+    {
+        "page": 170,
+        "content": "150V.NOUNS ANDADJECTIVES.[449\nTheforms inanil from this class aretheprevailing ones. Nofuture\nfern, participleinati isquotable from theolder language. Frompres.-stems\nindarefound there rnjati and sincati(RV.),tudati andpinvati (AV.).\nFrom denominatives, devayati (RV.), durasyad andfatruyad (AV.).\nVerbs oftheadorroot-class (chap. IX.)ending inaaregiven bythe\ngrammariansthesame optionasregards thefeminine ofthepresent parti-\nciple:thus, from>/j/a, ydntiorydd. Theolder language affords noexample\noftheformer,sofarasnoted.\nc.From other tense-stems than thosealready specified\nthat istosay,from theremainingclasses ofpresent-stems and\nfrom theintensives thefeminine isformed inad(or,ifthe\nstem beotherwise accented thanonthefinal, inafi)only.\nThus, adad fromyad; jfihvatl fromyhu; yunjad fromyyuj; wnvad\nfromysu;kurvad fromykr ;krmad fromykri; dedifati from dedif (intens.\nofydif).\nExceptionsareoccasionally metwith inthelater language, asdvisantl\n(BL), rudantl andkurvanti(N.).AndAV. hasydntl once.\n450.Afewwords areparticipialinform andinflection,\nthoughnotinmeaning. Thus :\na.brhdnt(oftenwritten vrhdnt inthelaterlanguage), 'great':\nitisinflected like aparticiple (withbrhati and brhdnti indu.\nandpi.neut.).\nb.mah&nt, 'great' ;inflected like aparticiple, butwith the\nirregularitythat theaoftheendingislengthenedinthestrong\nforms: thus, mahun, mah&ntam; mahantau(neut. mahad); mah-\nUntas, mahanti: instr. mahata etc.\nc.pfsant, 'speckled', and(inVedaonly) rfifani, 'shining'.\nd.jdgat, 'movable, lively' (inthelater language, asneuter noun, 'world'),\nareduplicated formation fromygam, 'go';itsnom. etc.neut.pi.isallowed\nbythegrammarians tobeonly jdganti.\ne.rhdnt, 'small'(only once,inRV., rhate).\nAllthese form their feminine inationly:thus, brhati,\nmahad, prsafi and rucatl(contrarytotherule forparticiples),\njdgafi.\nForddnt, 'tooth', which isperhapsofparticipial origin, seeabove, 396.\n451. Thepronominal adjectives iyant andMyantarein-\nflected likeadjectives inmant and vant, having (452) iyanand\nkiyanasnom. masc.sing., iyafi andMyatiasnom. etc.du.\nneut. and asfemininestems, andiyanti andkiyantiasnom.\netc.plur. neut.\nButtheneut.pi.fyanti andthe loc.sing.(?) Mydtiarefound inRV\n2.Possess! vesinmant andvant.\n452.Theadjectives formed bythese two suffixes are\n"
+    },
+    {
+        "page": 171,
+        "content": "453]DECLENSION V.E,DERIVATIVE STEMS INant. 151\ninflectedprecisely alike, andvery nearlyliketheparticiples\ninfffl^ant.From thelattertheydifferonlybylengthening\nthe 51ainthenom.sing. masc.\nThe voc.sing,isinan,like that oftheparticiple (inthe\nlaterlanguage, namely:forthat oftheoldest, seebelow, 454b).\nThe neut. nom. etc. areinthedualonlyaft(or dfi],and in\nthepluralanti(or dnti).\nThefeminine isalways made from theweak stem :thus,\nmati, vatl(ormdti, vdti}.\nThe accent, however,isnever(asintheparticiple)thrown\nforward uponthecase-endingorthefeminineending.\n453.Toillustrate theinflection ofsuch stems itwill\nbesufficient togiveapartoftheforms ofH^MHTI paqumdnt,\n'possessing cattle', and ^PToftT Ihdgavant, 'fortunate, blessed'.\nThus:\nSingular:\nm. n. n.\nN.\nA.O\npacumanO\npacumdt bhdgavan bhdgavat\npacumdntam pacumdt bhagavantam bhdgavat\npacumdta\netc.bhdgavata\netc.\nV.\nN.A.V.\nN.V.\nA.pacuman pdcumat\nDual:bhdgavan bhdgavat\no o\npacumdntau pacumdfi\netc.bhdgavantau bhdgavati\netc.\nPlural :\npacumdntas pacumdnti\npacumdtas pacumdntibJidgavantas bhdgavanti\nbhdgavatas bhdgavanti\npacumddbhis\netc.bhdqavadbhis\netc.\n"
+    },
+    {
+        "page": 172,
+        "content": "152V.NOUNS ANDADJECTIVES.[454\n454.Vedic Irregularities,a.Indual masc. nom.etc., a(for\nau)isthegreatly prevailing ending.\nb.Invoc. sing, masc.,theendinginthe oldest language (RV.)is\nalmost always inasinstead ofan(asintheperfect participle:below, 462 a):\nthus, adrivas, harivas, bhanumas, havismas. Such vocatives inRV. occur\nmore than ahundred times, while notasingle unquestionableinstance of\noneinan istobefound. Intheother Vedic texts, vocatives inasare\nextremely rare (but bhagavas and itscontraction bhagos aremetwith, even\ninthelater language); and intheir reproductionofRV. passages theasis\nusually changedtoan.\nItwaspointed outabove (425 g)that theRV.makes thevoc. inas\nalsoapparentlyfrom afewan-stems.\nc.InRV., thenom. etc.pi.neut., intheonly two instances that\noccur, ends indnti instead ofanti: thus, ghrtdvanti, papumdnti. Nosuch\nforms have been noted elsewhere intheolder language: theSV. reads anti\ninitsversion ofthecorresponding passages,and afewexamplesofthesame\nending arequotable from theBrahmanas :thus, tavanti, etdvanti, ydvanti,\npravanti, rtumanti, yugmanti. Compare 448, 451.\nd.Inafew (eight orten)more orless doubtfulcases,aconfusion of\nstrong andweak forms ofstem ismade :they aretoopurely sporadicto\nrequire reporting. Thesame istrue ofacase ortwowhere amasculine\nform appearstobeused with afeminine noun (seeLanman).\n455. Thestem dryant, 'running, steed', hasthenom.sing.\ndrva, from drvan; and intheolder languagealso thevoc. arvan\nandaccus. drvanam.\n456. Besides theparticiple bhdvant, there isanother stem\nbhavant, frequently used inrespectfuladdress assubstitute for\nthepronounofthesecondperson (but construed, ofcourse,\nwith averb inthethirdperson),which isformed with thesuffix\nvant, and sodeclined, havinginthenom.sing, bhdvan; and\nthecontracted form bhos ofitsold-stylevocative bhavas isa\ncommon exclamation ofaddress :'you,sir !'Itsoriginisvariously\nexplained ;itismostprobably acontraction ofbhdgavant.\n457. Thepronominal adjectives tdvant, etdvant, ydvant, andtheVedic\nivant, mdvant, tvdvant, etc., areinflected likeordinary derivatives fromnouns.\nF.Perfect Participles invans.\n458.The activeparticiples oftheperfect tense-system\narequite peculiarasregards themodifications oftheir stem.\nInthestrong cases, including thenom. -ace. -voc.pi.neut.,\ntheform oftheir suffix isSfftTvans,which becomes, byregu-\nlarprocess (150), van inthenom.sing., andwhich is\n"
+    },
+    {
+        "page": 173,
+        "content": "461 DECLENSION V.E,PARTICIPLES INvans. 153\nshortened to3FTvaninthevoc.sing.Intheweakest cas-\nes,the suffix iscontracted into 3^Tus.Inthemiddle cas-x*\nes,includingthenom.-acc.-voc. neut.sing.,itischanged\ntoofrTvat.\nAunion-voweli,ifpresentinthestrong andmiddle cases,\ndisappearsintheweakest, before MS.\n459.Theforms asthus described aremasculine and\nneuter only;thecorrespondingfeminine ismade byadding\n|Itotheweakest form ofstem, endingthus in3^ft usl.\n460. The accent isalways uponthesuffix, whatever be\nitsform.\nr461.Examples ofinflection. Toshow thein-\nflection oftheseparticiples, wemaytake thestems fi^TH\nvidvahs, 'knowing' (whichhasirregularloss oftheusual\nreduplication andoftheperfect meaning) from}/T^\" vid;\nand crfTSJoTfa tasthivans, 'having stood', fromj/FSTTstha.\nSingular: r^\nm. n. m. n.m\nN.\nA.vidvan vidvdt\nRteiHH^ fen^\nvidv&hsam vidvdttasthivan tasthivdt\ntasthivUiisam tasthivdt\nvidusa\nD.\nAb.G.\nL.\nV.vtasthusa\nrlTOT\ntasthuse\nvidvan vidvattasthusi\nisn^\ntdsthivan tdsthivat\nDual:\nvidvahsau vidusl tasthivansau tasthusi\n"
+    },
+    {
+        "page": 174,
+        "content": "154NOUNS ANDADJECTIVES.[461\nI.D.Ab.\nvidvddbhyam tasthivddbhyam\nG.L.\nvidusos tasthusos\nPlnral :\nN.v.^\nvidvtinsas vidvfthsi tasthivansas tasthivaiisi\nA.f^fJTO^Mdift fl^H^crfr^lftr\nvidusas vidvfrhsi tasthusas tasthivnhsi\nI.\nvidvadbhis tasthivddbhis\nD.Ab.\nvidvddbhyas tasthivddbhyas\nG.fa^MI^ rT^IH^\nvidusam tasthusam\nL.\nm\nvidvdtsu tasthivdtsu\nThefeminine stems ofthese twoparticiplesare\nvidusl and riw$ tasthusl.o\nOther examplesofthedifferent stems are :\nfromyAr cakrvtihs, cakrvdt, cakrus, cakrusz;\nfrom|/n niriivfins, niriivdt, ninyus, ninyusi ;\nfrom|/MM babhuvUhs, babhuvdt, babhuvus, babhuvusl ;\nfrom|/tan tenivahs, tenivdt, tenus, tenusi.\n462. a.Intheoldest language (RV.), thevocative sing. masc.(like\nthat ofvantandmant-stems :above, 454b)hastheending vasinstead of\nvan.' thus,cikitvas (changed to-van inaparallel passageofAV.), titirvas,\ndidivas, nudhvas.\nb.Forms from themiddle stem, invat, areextremely rare earlier:\nonly three(tatanvdt andvavrtvo'J, neut.sing., andjagrvddbhis,instr.pi.),\narefound inRV., andnotoneinAV.And intheVeda theweakest stem\nandnotthemiddle one, aslater,ismade thebasis ofcomparison: thus,\nvidtistara, mldhtistama.\nc.Anexample ortwo oftheuse oftheweak stem-form forcases\nregularly made from thestrong arefound inRV. :they arecakrtisam, ace.\nsing., anddbibhyums, nom.pi. ;emuam, byitsaccent (unless anerror),is\nrather from aderivative stememusd :andQB.hasprostisam.Similar in-\nstances, especially fromvidvdhs, arenowandthenmetwith later(seeBR.,\nundervidvcihs).\n"
+    },
+    {
+        "page": 175,
+        "content": "464]DECLENSION V.E,COMPARATIVES INyas. 155\nd.TheAV. hasonce bhaktivdhsas\n,asifaparticipial form from anoun\n;\nbutK.andTB. give inthecorresponding passage bhaktivdnas ;cakhvdhsam\n(RV., once)isofdoubtfulcharacter; okivdhsa(RV., once) shows areversion\ntoguttural form ofthe final ofYUC,elsewhere unknown.\nG.Comparatives inyas.\n463.Thecomparative adjectives ofprimaryformation\n(below, 467)have adouble form ofstem formasculine and\nneuter :astronger, endinginTXfaj/ahs(usually\ninthestrong cases, andaweaker,inEJTTyets (or\nintheweak cases(there being nodistinction ofmiddle and\nweakest). The voc.sing.masc. ends inVF(^yan(butfor\ntheolder languageseebelow, 465a).\nThefeminine ismade byadding ^itotheweak masc.-\nneut. stem.\n464. Asmodels ofinflection,itwillbesufficient to\ngiveapartoftheforms of3TOT $reyas, 'better', and of\ngdriyas,'heavier'. Thus :\nSingular:\nN.\nA.creyan creyas gdriyan\ncreyansam creyasgariyas\nTlfklTT\ngariyas\ncreyasa\netc.gariyasa\netc.\nV.\nN.A.Y.\nN.V.creyan\nDual:\ngreyansaucreyas\nSj^nl\ncreyasigdriyan gariyas\ngdriyahsau gdrlyasi\netc.\nPlural :etc. etc. etc.\ncreyahsas creyansi gdriyansas gdriyahsi\n"
+    },
+    {
+        "page": 176,
+        "content": "V.NOUNS ANDADJECTIVES.[464\nftrfm\ncreyasas creyansi gariyasas gdriyansi\ns\nslcreyobhis gdriyobhis\netc. etc.\nThefeminine stems oftheseadjectivesare 44tin)greya\nandJ|(ldHl gdriyasl.\n465. a.TheVedic voc. masc.(asinthetwopreceding divisions :\n454b,462 a)isinyasinstead ofyan: thus, ojiyas, jyayas (RV.:noex-\nampleselsewhere havebeen noted).\nb.Noexampleofamiddle case occurs inRV. orAV.\nc.Inthe later language arefound afewapparent examplesofstrong\ncasesmade from theweaker stem-form: thus, kaniyasam, ace.masc., kariiyasau\ndu.They areperhapsrather tobeviewed astransition-forms toana-\ndeclension.\nComparison.\n466. Derivativeadjectivestems havingacomparative\nandsuperlative meaningoroften also(andmoreorigin-\nally)amerelyintensive value aremade eitherdirectly\nfrom roots(byprimary derivation),orfrom other derivative\norcompound stems(bysecondary derivation).\nThe subject ofcomparison belongs properlytothechapterofderivation;\nbut itstands insuch near relation toinflection that itis,inaccordance with\ntheusual custom ingrammars, conveniently and properly enough treated\nbriefly here.\n467.The suffixes ofprimaryderivation are^JH lyas\nforthecomparative and^?isfha forthesuperlative. The\nrootbefore them isaccented, andusually strengthened by\ngunating,ifcapableofitor,insome cases, bynasali-\nzation orprolongation. Theyaremuch more frequentlyand\nfreely used intheoldestlanguagethan later; intheclass-\nicalSanskrit, onlyalimited number ofsuch comparatives\nandsuperlativesareacceptedinuse\n;and these attach\nthemselves inmeaningforthemostparttootheradjectives\nfrom thesame root, which seem tobetheir corresponding\n"
+    },
+    {
+        "page": 177,
+        "content": "468]COMPARISON OFADJECTIVES. 157\npositives;but inpartalsotheyareartificially connected\nwith other words, unrelated withthem inderivation.\nThus, from,yksip, 'hurl', comeksepiyasandksepistha, which\nbelonginmeaningtoksiprd, 'quick' ;fromyvr, 'encompass',\ncome vdriyas and vdristha, which belongtouru, 'broad'\n;while,\nforexample, kdniyas andkdnistha areattached bythegrammarians\ntoyuvan, 'young',ordlpa,'small';andvdrslyas and vdrsistha\ntovrddtid 'old'.\n468.FromVeda andBrahmana together, rather more than\nahundred instances ofthisprimary formation inIyasand istha\n(inmanycases only oneofthepair actually occurring)areto\nbequoted. About half ofthese(inRV., thedecidedmajority)\nbelong,inmeaningasinform, tothebare root initsadjective\nvalue, asusedespeciallyattheendofcompounds, butsome-\ntimes alsoindependently:thus, fromy'tap, 'burn', comestdpist/ia,\n'excessively burning' ;from1/yaj, 'offer', comeydjiyas and\nydjistha,'better andbest(orvery well) sacrificing' ;fromyyudh,\n'fight', comesyodhtyas, 'fightingbetter'\n; inafewinstances,\nthesimpleroot isalsofound used ascorresponding positive:\nthus, ju, 'hasty, rapid',withjdviyas andjdvistha. Inalittle\nclass ofinstances(eight), theroot has apreposition prefixed,\nwhich then takes theaccent :thus, agamistha, 'especially coming\nhither';vicayistha,'bestclearing away' ; inacoupleofcases\n(dcramistha, dparavapistha, dstheyas),thenegative particleispre-\nfixed;--inasingle word(cdmbhavistha),anelement ofanother\nkind. Thewords ofthisformation often take anaccusative\nobject: thus, ndbhastdriyan (RV.), 'traversing rapidlythecloud';\nvrtrdm hdnisthah(RV.),'bestslayerofVritra'.\nButeven inthe oldest language appearsnotinfrequently\nthesame attachment inmeaningtoaderivativeadjectivewhich\n(aspointedoutabove)isusual inthelaterspeech.Besides the\nexamplesthatoccur also later, others aremetwith like vdristha,\n'choicest'(vdra, 'choice'), bdrhistha, 'greatest' (brhdnt, 'great'),\nosistha, 'quickest' (6sam, 'quickly'),and soon. Probably by\nanalogy with these,like formations areinafew casesmade\nfrom theapparentlyradicalsyllablesofwords which haveno\notherwise traceable root inthelanguage:thus, kradhiyasand\nkradhistha(K.)from krdhu, sthdviyas and sthdvistha from sthurd,\ncdciyas (RV.)from cdcvant, dmyas (AV.)and dnistha(TS.)from\nanil; and soon.Andyetagain,inafew exceptional cases,\nthesuffixesiyasand istha areappliedtostems which arethem-\nselves palpablyderivative :thus, acistha from acu(RV.:only\ncase), tiksniyas (AV.)from tiksnd, brdhmiyasandbrdhmistha(TS.)\nfrom brahman, dhdrmistha (TA.)from dhdrman, drddhistha (TB.:\n"
+    },
+    {
+        "page": 178,
+        "content": "158V.NOUNS ANDADJECTIVES.[468\ninstead ofddrhistha]from drdhd, rdghiyas (TS.)fromraghu.\nThese arebeginnings,notfollowed uplater, oftheextension\noftheformation tounlimited use.\nInndviyasornavyasand navistha, from ndva, 'new', and\ninsdnyasfrom sana, 'old1\n(allEV.), wehave also formations\nunconnected with verbal roots.\n469. Thestems inistha areinflected likeordinary adject-\nives ina,makingtheir feminines ina;those inlyashave a\npeculiar declension, which hasbeen described above(463 ff.).\n470. Ofpeculiaritiesand irregularitiesofformation, the\nfollowing maybenoticed.\nThe suffix lyas hasinafewinstances thebriefer formyas, generally\nasalternative with theother: thus, tdvlyas andtdvyas, ndviyas andn&vyas,\nvdsiyas andvdsyas, pdnlyas andpdnyas; and sofrom rabh andsah; sdnyas\noccurs alone. From bhucome bhuyas andbhuyistha,beside which RV. has\nalso bhdviyas.\nOfroots ina,the final blends with the initial ofthesuffix toe:thus,\nstheyas, dhestha, yestha;butsuch forms areintheVeda generally tobe\nresolved,asdhd'istha, ydutha. The rootjyaforms jytetha, butjydyas (like\nbhUyas).\nThetwo roots inz,prlandpn,form preyas andprestha and pret/as\nand frestha.\nFrom the root ofrjticome, without strengthening, rjlyas andr/i^fca ;\nbutintheolder language also,moreregularly, rdjiyas andrdjistha.\n471.The suffixes ofsecondaryderivation arerljtara\nand rFTtama. Theyareofalmost unrestrictedapplication,\nbeing added toadjectives ofevery form, simple andcom-\npound, endinginvowels orinconsonants and thisfrom\ntheearliestperiodofthelanguageuntil the latest. The\naccent oftheprimitive remains(withrareexceptions) un-\nchanged;andthatform ofstem isgenerallytaken which\nappears before aninitial consonant ofacase-ending (weak\normiddleform).\nExamples (ofolder aswell aslater occurrence) are :from\nvowel-stems, priydtara, vdhnitama, rathitara andrathitama(RV.),\ncarutara, potrtama; from consonant-stems, cdmtama, cdcvattama,\ntavdstara andtavdstama, tuvisfama, vdpustara, tapasvitara, bhdga-\nvattara, hiranyavafimattama ; from compounds, ratnadh&tama,\nabhibhutam, sukfttara, purbhittami, bhuridavattara, cucivratatama,\nstrlkamatama.\n"
+    },
+    {
+        "page": 179,
+        "content": "474]COMPARISON. 159\nBut intheVeda the finalnofastem isregularly retained:thus,\nmadfntara andmadfntama, vrsdntama;and ofaperfect participle theweakest\nstem istaken :thus, vidtistara, midhustama. Afeminine final Iisshortened :\nthus, devitama (RV.), tejasvinitama (K.).\nIntheolder language, thewords ofthisformation arenotmuch more\nfrequent than those oftheother:thus, inRV. thestems intaraandtama\naretothose inlyasand istha asthree totwo;inAV., only assixtofive:\nbutlater theformer winagreat preponderance.\n472. Thesecomparatives andsuperlatives areinflected like\nordinary adjectivesina,formingtheir feminine ina.\n473. That(especiallyintheVeda) some stems which are\nnouns rather thanadjectives form derivatives ofcomparisonis\nnatural enough, considering theuncertain nature ofthedivision-\nlinebetween substantive andadjective value. Thus, wehave\nmatftama, nrtama, maruttama, and others.\nThe suffixes taraandtama alsomake forms ofcomparison\nfrom some ofthepronominal roots, aska,ya,i(seebelow,\n520);andfrom certain oftheprepositions,asut;andthead-\nverbial accusative(older, neuter; later, feminine)ofacompar-\native intarafrom aprepositionisused tomake acorresponding\ncomparativetotheprepositionitself(below,11lie).\nTheHindu grammarians even allow the suffixes ofcomparisoninthe\nadverbial accusative feminine, taram andtamam, tobeappendedtocon-\njugational forms: thus, pacati, 'hecooks', pacatitaram, 'hecooks better':\nbutsuch arebarbarous combinations, having nowarrant intheearlier uses\nofthelanguage.\nThe suffixes ofsecondary comparison areoccasionally added tothose of\nprimary, forming double comparatives andsuperlatives:thus, gariyastara,\nfresthatama.\nTheuseoftama asordinal suffix isnoted below(487);\nwith thisvalue,itisaccented onthefinal, andmakes itsfemin-\nineini:thus, catatamd, m.and n.,catatamk, f.,'hundredth'.\n474.From afewwords, mostly prepositions, degreesof\ncomparisonaremade bythe briefer suffixes raandma: thus,\nddhara andadhamd, dpara andapamd, dvara andavamd, upara\nandupamd, dntara, dntama, paramd, madhyamd,caramd. And\nma isalsoused tomake ordinals(below, 487).\n"
+    },
+    {
+        "page": 180,
+        "content": "160 [475\nCHAPTER VI.\nNUMERALS.\n475. The simplecardinal numerals forthe first ten\nnumbers (whicharethefoundation ofthewholeclass),\nwith their derivatives,thetens, andwithsome ofthehigh-\nermembers ofthedecimal series, areasfollows:\neka\ndvd\ntri\ncatur\npdnca\n10astd\n^\nndva\n^1\nddcaddca'\n20\n30\n40vincati\ntrinc&t\ncatvarincdt\n50\npancacdt\n60\nsas^'\n70Httfri\nsaptati\nso\n90\n100100\n1000\n10,000\n100,000\n1,000,000catd\nH^f\nsahdsra\n^TrTo\nayuta\npr&yuta\n10,000,000\n10\n10\n10\n10'arbudd\nr\nmaharbuda\n*f\nJcharvd\nnikha\nThe accentsajrta and a*fd isthat belonging tothese words in all\naccentuatedtexts; accordingtothegrammarians, they aresdpta anddsta in\nthelater language. Seebelow, 483.\nThe series ofdecimal numbers maybecarried stillfurther;\nbutthere aregreat differences among the different authorities\n"
+    },
+    {
+        "page": 181,
+        "content": "476]ODDNUMBERS. 161\nwith regardtotheirnames\n;andthere ismore orlessofdiscord-\nance even fromayutaon.\nThus, intheTS.,wefindayuta, niyuta, prayuta, drbuda, nyhrbuda,\nsamudrd, mddhya, dnta, parardhd; K.reverses theorder ofniyuta and\nprayuta, and inserts badva after nyarbuda (reading nyarbudha): these are\nprobably theoldest recorded series.\nInmodern time,theonlynumbers inpracticalnseabove 'thousand' are\nlaksa('lac'orlakh1\n)and koti('crore'); andanIndian sum iswont tobe\npointed thus: 123,45,67,890,tosignify '123 crores, 45lakhs, 67thousands,\neight hundred and ninety'.\nAstothestem-forms pancan etc., seebelow, 484. Astotheform\nsaks instead ofsas, seeabove, 146 end. Thestem dvaappearsincom-\nposition andderivation also asdvaanddvi; catiir incompositionisaccented\ncdtur. The older form ofasta isasta :seebelow, 483. Forms in-fatand\n-fati forthetens areoccasionally interchanged.\nTheothernumbers areexpressed bythevarious composition\nandsyntactical combination ofthose givenabove. Thus :\n476. Theoddnumbers between theeven tens aremade\nbyprefixingthe(accented)unit totheten towhich itsvalue\nistobeadded: butwith various irregularities. Thus:\nekain'11'becomes eka,but iselsewhere unchanged;\ndvabecomes everywhere dva; butin'42' '72'andin'92' itisinter-\nchangeable withdvi,andin'82'dvialone isused;\nfor triissubstituted itsnom. pi.masc. trdyas;but triitself isalso\nallowed in'43' '73'andin'93', andin'83' trialone isused;\nsasbecomes soin'16',andmakes theinitial dofdafa lingual (199b);\nelsewhere itsfinalundergoes theregular conversion (226 b)to*ordorn;\nandin'96'thenofnavati isassimilated toit(199b);\nastabecomes [asta (483)in'18' '38', andhaseither form inthe\nsucceeding combinations. Thus :\nnekada^a\n"
+    },
+    {
+        "page": 182,
+        "content": "162VI.NUMERALS.[476\nTheforms made with dvdandtrayas aremore usual than those with\ndviandtri,which arehardly tobequoted from theolder literature(V.and\nBr.). The forms made with asta (instead ofasta) arealone found inthe\nolder literature (483), and areusual inthe later.\n477. Theabove arethenormal expressionsfortheodd\nnumbers. Butequivalentsubstitutes forthem arealsovariously\nmade. Thus :\na.Byuse oftheadjectives una, 'deficient', andadhika, 'redundant',\nincompositionwith lesser numbers which aretobesubtracted oradded, and\neither independently qualifyingor(more usually)incomposition with larger\nnumbers which aretobeincreased ordiminished bytheothers :thus, tryuna-\nsasti, 'sixty deficient bythree'(i.e.'57'); astadhikanavati, 'ninety increased\nbyeight' (i.e.'98'); ekddhikam patam,'ahundred increased byone'(i.e.\n'101'); panconam catam, '100 less 5'(i.e.'95'). Forthenines, especially,\nsuch substitutes asekonavihcati,'20 less1',or'19',arenotuncommon;\nand later theeka,T,isleftoff,andunavihcati etc.have thesame value.\nb.Acase-form ofeka, 'one',isconnected byna, 'not', with alarger\nnumber from which one istobededucted:thus, ekayd ndtrincdt(QB.\nPB.KB.),'not thirty byone'(i.e.'29'); ekasmdn ndpancacdt (inordinal),\n'49'(TS.); ekasyai (abl.fern.: 366.3) ndpancacdt, '49'(TS.); mostoften,\nekan (i.e.ekdt, irregular abl. forekasmdt) ndvincatf, '19';ekdnndcatdm,\n*99'. This lastform isadmitted also inthe later language:theothers are\nfound intheBrahmanas.\nc.Instances ofmultiplication byaprefixed number areoccasionally met\nwith:thus, trisaptd, 'thrice seven'; trinavd,'thrice nine'; tridacd,'thrice ten'.\nd.Ofcourse,thenumbers tobeadded together maybeexpressed by\nindependent words, with connecting'and' :thus, ndva canavatfy ca,orn&va\nnavatfm ca,'ninety andnine'; dvau cavihcatfc ca,'twoandtwenty'. But\ntheconnective isalso(atleast, intheolder language) notseldom omitted:\nthus, navatfr ndva, '99'; trincdtam trin, '33'; acltir astdti, '88'.\n478. Thesamemethods arealsovariously used forforming\ntheoddnumbers above 100. Thus:\na.Theadded number isprefixed totheother, andtakes theaccent:\nforexample, ekacatam, '101'; astdcatam, '108'; trincdchatam, '130'; asta-\nvihcaticatam, <148J\n;cdtuhsahasram(RV.:unless theaccent iswrong),'1004'.\nb.Or,thenumber tobeadded iscompounded with adhika, 'redundant',\nandthecompound iseither made toqualify theother number orisfurther\ncompounded with it:thus, pancddhikarh catam orpancddhikacatam,'105'.\nOfcourse, una, 'deficient'(asalso other words equivalenttouna or\nadhika}, maybeused inthesame way: thus,panconam catam,'95'.\nC.Syntactical combinations aremade atconvenience :forexample, ddca\ncatdmca,'110'; catam ekamca,401'.\n479. Another usual method(beginningintheBrahmanas)\nofforming theoddnumbers above 100 istoqualify thelarger\n"
+    },
+    {
+        "page": 183,
+        "content": "482]COMPOUND NUMBERS. 163\nnumber byanadjective derived from thesmaller, and identical\nwith the briefer ordinal(below, 487):thus, dvadacdm catdm,\n'112'(lit'ly,'ahundred ofa12-sort, orcharacterised by12\n;\ncatuccatvarihcdm catdm, '124'; satsastdm catdm, '166'.\n480. Tomultiply onenumber byanother, amongthehigher\northelower denominations, thesimplest and least ambiguous\nmethod istomake ofthemultiplied number adual orplural,\nqualified bytheother asanyordinary noun would be\n;and this\nmethod isacommon one inallagesofthelanguage. Forex-\nample:pdnca pancacdtas,five fifties('250');ndvanavatdyas,'nine\nnineties'('810');acitibhistisrbhis, 'with threeeighties' ('240');\npdnca gatani,'five hundreds'; trlnisahdsrani,'three thousands';\nsastini sahdsrani, '60,000'; daca casahasranyastau cacatani,\n'10,800': and,combined with addition, trim cathnitrdyastrincatam\nca,'333'; sahasre dvepanconam catam eva ca,'2095'.\nByapeculiar and wholly illogical construction, such acombination as\ntrinisastifatani, which ought tosignify '480'(3x100+60),isfrequently\nused intheBrahmanas tomean '360'(3x100+60);soalsoduecatustrinfe\nfate,'234'(not '268'); andother like cases.\n481. But thetwo factors, multiplier andmultiplied,are\nalso, andinlater usage moregenerally, combined into acom-\npound (accentedonthefinal); and this isthen treated asan\nadjective, qualifyingthenumbered noun; orelse itsneuter or\nfeminine(in i]singularisusedsubstantively:thus, dacacatas,\n'1000'; satcataih padatibhih (MBh.),'with 600 foot-soldiers';\ntrdyastrihqattricatah satsahasrtih(AV.), '6333'; dvicatdm ordmcatk,\n'200'; astadacacatl, '1800'.\nIntheusual absence ofaccentuation, there arises sometimes aquestion\nastohow acompound number shall beunderstood: whether asta?atam, for\nexample,isastdfatam, '108', orastacatdm, '800', andthe like.\n482.Inflection. The inflection ofthecardinal nu-\nmerals isinmany respects irregular. Gender isdistinguish-\nedonlybythe first four.\na.Eka,T,isdeclined after themanner of .apronominal\nadjective (like sdrva, below, 524);itspluralisused inthe\nsense of'some, certain ones'. Itsdual does notoccur.\nOccasional forms oftheordinarydeclension aremetwith :\nthus, eke(loc. sing.),ekat.\nInthelateliterature, eka isused inthesense of'acertain',\norevensometimes almost of'a',asanindefinite article. Thus,\nekovyaghrah (H.),'acertaintiger';ekasmin dine, 'onacertain\nday';hastedandam ekamadaya (H.), 'takingastick inhishand'.\n11*\n"
+    },
+    {
+        "page": 184,
+        "content": "164VLNUMERALS.[482\nb.Dva, '2',isdual only, and isentirelyregular^:thus,\nN.A.V. dvau(V. dva), m., dve,f.n.\n;I.D.Ab.dvubhyam;\nG.L.dvdyos.\nc.Tri, '3',isinmasc. andneut. nearly regular,likean\nordinarystem in*;buttheir genitiveisasiffromtraya(only\ninthelater language:theregular trmhm occurs once inRV.).\nForthefeminine ithasthepeculiarstemtisr,which isinflected\ningenerallikeanr-stem;butthenom. andaccus. arealike,\nandshow nostrengtheningofther;.andtherisnotprolonged\ninthegen. (exceptingintheVeda).Thus :\nm. n. f.\nN. trdyastrini tisrds\nA. trin trini tisrds\n1. tribhfs tisrbhis\nD.Ab. tribhyds tisfbhyas\nG. trayandm tisrndm\nL. tristi tisfsu.\nTheVeda hastheabbreviated neut. nom. andaccus. tri.Theaccent-\nuation tisrbhfs, tisrbhyds, tisrnam, and tisrsti issaid tobealsoallowed in\nthelater language.\nThestem tisroccurs incomposition intisrdhanvd(Br.),'abowalong\nwith three arrows'.\nd.Catur, '4',hascatv&r(themoreoriginal form)inthe\nstrongcases;inthefern, itsubstitutes thestem catasr, apparently\nakin withtisf,and inflected like it(butwithanomalous change\nofaccent,like that inthehigher numbers :seebelow, 483).\nThus:\nm. n. f.\nN. catvdras catvdri cdtasras\nA. cattiras catvdri cdtasras\n+, I. cattirbhis catasrbhis\nD.Ab.cattirbhyas catasfbhyas\nG. caturndm catasrndm\nL. cattirsu catasr su.\nTheuseofnbeforeamofthegen. masc. andneut. after afinal con-\nsonant ofthestem is(asinsas:below, 483) astriking irregularity. The\nmore regular gen. fern, catasfnamalso sometimes occurs. Inthe later\nlanguage, theaccentuation ofthefinal syllable instead ofthepenultisallow-\nedininst., dat.-abl., and loc.\n483. Thenumbers from '5'to'19'have nodistinction of\ngender, noranygeneric character. Theyareinflected, somewhat\nirregularly, asplurals, save inthenom. -ace., where theyhave\nnoproper plural form, butshow thebarestem instead. Ofsds\n(asofcatur\\ nam isthegen. ending, withmutual assimilation\n(198b)ofstem-final and initial ofthetermination. Asta(as\n"
+    },
+    {
+        "page": 185,
+        "content": "485]INFLECTION. 165\naccented intheolderlanguage)hasanalternative fuller form,\nasta, which isalmostexclusively used intheolder literature\n(V.andBr.),both ininflection and incomposition (butsome\ncompoundswith asta arefound asearlyastheAV.);itsnom.-\nacc. isastd(usuallater :found inRV. once, and inAV.),or\nasta(RV.),orastau(mostusual inRV.; also inAV., Br.,\nandlater).\nThe accent isinmany respects peculiar. Inalltheaccentedtexts,the\nstress ofvoice liesonthepenult before theendings this, bhyas, andsu,\nfrom thestems ina,whatever betheaccent ofthestem :thus, pancdbhis\nfrom pdnca, navdbhyas from ndva, dafdsu from ddfa, navadafdbhis from\nndvadafa, ekddafdbhyas from ekddafa, dvddafdsu fromdvadaya; according to\nthegrammarians,either thepenult orthe final isaccented inthese forms\ninthelater language. Inthegen. pi.,theaccent isontheending (asin\nthat ofi,u,andr-stems). The cases ofsas,andthosemade from thestem-\nform asta, have theaccent throughout upon theending.\nExamplesoftheinflection ofthese words areasfollows :\nN.A. pdncasat astati astd\nI. pancdbhissadbhis astabhfs astdbhis\nD.Ab. pancdbhyas sadbhyds astdbhyds astabhyas\nG. pancdndm sanndm astdndm\nL. pancdsusatsu astdsu astdsu.\nSapid (inthelaterlanguage sdpta, asdsta forastd) andndva andddfa,\nwith thecompoundsofddfa ('11' '19'),aredeclined likepanca, andwith\nthesame shift ofaccent(orwith alternative shift totheendings,aspointed\noutabove).\n484. TheHindu grammarians give tothestems for '5'and '7' '19'a\nfinaln:thus, pancan, saptan, astan, navan, dacan, andekddafan etc. This,\nhowever, hasnothing todowith thedemonstrably originalfinal nasal of'7',\n'9',and '10'(compare septem, nouem, decem; seven, nine, ten)',itisonly\nowing tothe factthat, starting from such astem-form, their inflection is\nmade toassume amore regular aspect, thenom.-acc. having theform ofa\nneut. sing, inan,andtheinstr., dat.-abl., and loc.that ofaneut. ormasc.\npi.inan:compare ndma, ndmabhis, ndmabhyas, ndmasu thegen. alone\nbeing likethat, rather, ofaa-stem :compare dafdndm with {ndranam and\nndmnam oratmdnam. Notrace whatever ofafinalnisfound anywhere in\nthelanguage,ininflection orderivation orcomposition, from anyofthese\nwords.\n485. The tens, vihqatiand trihc&t etc., with theircompounds,\naredeclinedregularly,asfeminine stems ofthesame endings,\nand inallnumbers.\nCatdandsahdsra aredeclinedregularly,asneuter(or, rarely,\ninthelater language,asmasculine)stems ofthesame final, in\nallnumbers.\n"
+    },
+    {
+        "page": 186,
+        "content": "VI.NUMERALS.[485\nThe like istrue ofthehigher numbers which have, in-\ndeed, nopropernumeral character, butareordinary nouns.\n486.Construction. Asregardstheir construction with\nthenouns enumerated bythem :\na.Thewords for '!'to'19' areusedadjectively, agreeing\nincase, and,ifthey distinguish gender,ingender also, with\nthenouns :thus, dacdbhir viraih, 'with tenheroes'\n;yedevadivy\nekadaca sthd(AV.),'what eleven godsofyouare inheaven';\npancdsu jdnesu, 'among the five tribes'; catasfbhir girbhth,'with,\nfoursongs'.\nb.Thenumerals above '19' areconstruedusuallyasnouns,\neither takingthenumbered noun asadependent genitive, or\nstandinginthesingularinapposition with it :thus, catam dasih\norcatam dasmam, 'ahundred slaves' or'ahundred ofslaves';\nmiwatyn Mrib/iis, 'with twenty bays'; sastyam cardtsu, 'in60au-\ntumns'; Catena, pacaih,'with ahundredfetters'; catam sahdsram\nayutam nydrbudam jaghhnacakrb ddsyunam (AV.),'themighty\n[Indra]slew ahundred, athousand, amyriad,ahundred mil-\nlion, ofdemons'.\nOccasionally theyareputintheplural,asifusedmore\nadjectively: thus, pancacadbhir vanaih, 'withfiftyarrows'.\nc.Intheolder language, thenumerals for '5'andupward\naresometimes used inthenom.-acc. form(orasifindeclinably)\nwith other cases also :thus, pdnca krstisu, 'among the five races';\nsaptd rsmam, 'ofseven bards'; sahdsram rsibhih, 'with athousand\nbards'; catampurbhih,'with ahundredstrongholds'.\n487.Ordinals. Oftheclasses ofderivative words\ncoming from theoriginalorcardinal numerals, theordin-\nalsarebyfarthemostimportant; andthemode oftheir\nformation maybestbeexplainedhere.\nSome ofthe first ordinals areirregularly made :thus,\neka, '!',forms noordinal; instead isusedprathamd (i.e.\npra-tama, 'foremost');adi israre intheBrahmanas, andadya\neven intheSutras;\nfrom dvd, '2',andtri, '3',comedvitiya andtrtiya (second-\narily, through dvitaandabbreviatedtrita}',\ncatur, '4', sds, '6',andsaptd, '?',take theendingtha :\nthus, caturthd, sastM, saptdtha; but for 'fourth' areused also\nturiya andturya, andsaptdtha belongstotheolder language\nonly: pancatha, for'fifth', isexcessively rare;\nthenumerals for '5'and '7'usually, andfor'8', '9', '10',\naddma,forming pamama, saptamd, astamd, navamd, daqamd;\n"
+    },
+    {
+        "page": 187,
+        "content": "489]NUMERAL DERIVATIVES. 167\nfor41th' to'19th', theforms areekadacd, dvadacd, and\nsoon(thesame with thecardinals, except changeofaccent);\nforthetensandintervening oddnumbers from '20'onward,\ntheordinal hasadouble form onemade byaddingthe full\n(superlative) ending tamd tothe cardinal :thus, vihcatitamd,\ntrincattamd, acititamd, etc.;theother, shorter,ina,with abbre-\nviation ofthecardinal: thus, vihqd, '20th'; trincd, '30th'; ca-\ntvarincd, '40th'; pancacd, '50th'; sastd, '60th'; saptatd, '70th';\nacitd, '80th'; navatd, '90th'; and solikewise ekavihcd, '21st',\ncatustrihcd, '34th'; astacatvarihcd, '48th'; dvapancacd, '52d';\nekasastd, '61st'; andekannavihfd and unavincd and ekonavtncd,\n'19th'; andsoon.Ofthese twoforms, thelatter andbriefer\nisbyfarthemore common, theother being notquotable from\ntheVeda, andextremely rarely from theBrahmanas. From '50th'\non,thebriefer form isallowed bythegrammarians onlytothe\noddnumbers, made upoftensand units\n;but itissometimes\nmetwith, even inthelater language, from thesimpleten.\nOfthehigher numbers, gatdandsahdsra form catatamd and\nsahasratamd;but their compoundshave alsothesimpler form :\nthus, ekagata,'101st'.\nOfthe ordinals, prathamd (and adya\\ dvitlya, trtiya, and\nturiya (with turya]form their feminine ina; alltherestmake\nitin i.\n488. Theordinals,asinother languages, have other than ordinal offices\ntofill;andinSanskrit especially they aregeneral adjectives tothecardinals,\nwith aconsiderable variety ofmeanings,asfractionals,assignifying 'composed\nofsomany parts' or'so-many-fold',or'containing somany', or(aswasseen\nabove, 479) 'having somany added'.\nInafractional sense, thegrammarians direct that their accent be\nshifted tothe firstsyllable:thus, dvtiiya,'half;trtlya,'third part' ;cdturtha,\n'quarter', andsoon.Butinaccented texts only trtlya, 'third', andttiriya,\n'quarter', arefound sotreated;for'half occurs only ardhd; andcaturthd,\npancamd, and soon,areaccented asintheir ordinal use.\n489. Other numeral derivatives thus,\nmultiplicative adverbs, asdvis, tris, catus, 'twice', thrice'.,\n'four times';\nadverbs with thesuffixes dhaand cas :forexample, ekadhh,\n'inoneway', catadhu, 'inahundredways'; ekacas, 'onebyone',\ncatacds, 'byhundreds';\ncollectives, asdvitaya,ordvayd,'apair', ddcatayaordacdt,\n'adecade'\nbelongrather tothedictionary,ortothechapterofderivation.\n"
+    },
+    {
+        "page": 188,
+        "content": "168 [490\nCHAPTER VII\nPRONOUNS.\n490.THEpronounsdiffer from thegreatmass ofnouns\nandadjectives chieflyinthattheycome byderivation from\nanother andaverylimited setofroots, theso-called'pro-\nnominal' or'demonstrative' roots. Buttheyhave alsomany\nandmarkedpeculiaritiesofinflection some ofwhich,\nhowever,findanalogiesalso inafewadjectives; andsuch\nadjectiveswill accordinglybedescribed attheendofthis\nchapter.\nPersonal Pronouns.\n491.Thepronounsofthe firstandsecondpersonsare\nthemostirregular andpeculiarofall,being made upof\nfragments coming from various roots andcombinations of\nroots. Theyhavenodistinction ofgender.\nTheir inflection inthelaterlanguageisasfollows :\nSingular:\n1stpers. 2dpers.\nN.^^ OT^\nahdm tvdm\nA.TF[^Jqj^T^^7\nmam,ma tvam, tva\nI- TITT (TOT\nmdya tvdya\nD.*-|c^|4-|, If ^jV^JTf, (^\nmdhyam, metubJiyam,te\nAb.3^ ^7^\nmat tvdt\n"
+    },
+    {
+        "page": 189,
+        "content": "492]\nG.\nL.\nDual:\nN.A.V.\nI.D.Ab.\nG.L.\nandA.D.G.\nPlural :\nN.\nA.\nI.\nD.\nAb.\nG.\nL.PERSONAL PRONOUNS.\nT rR,^169\nmama, me\nmdyi\navam\navabhyam\navdyos\nnan\nvaydm\nasman, nas\nasmabhis\nasmdbhyam,nas\nasmdt\nasmakam, nas\no\nasmasuyuvam\nETSTFTT\nyuvabhyam\nyuvdyos\nyuydm\no^\nyusman,vas\n^T^TTPTTTo,-\\\nytismabhis\n^JCrjT^JTI1\n,^{\nyiismdbhyam,vas\ntfa^j^\nyusmdt\nyusmakam,vas\n,0\nyusmasu\nThe briefer second forms foraccus., dat., andgen.,inall\nnumbers, areaccentless\n;andhence theyarenotallowed to\nstand atthebeginningofasentence, orelsewhere where any\nemphasisislaid.\nThe ablative mat isaccentless inone ortwoAV. passages (xi.4.26;\nxii.3.46).\n492.Forms oftheolder language.Alltheforms\n"
+    },
+    {
+        "page": 190,
+        "content": "170VI1-PRONOUNS.[492\ngivenabove arefound also intheolder language ;which, how-\never, hasalso others thatafterward disappear from use.\nThus,theVeda (RV.)hasafewtimes theinstr. sing, tva(likemanisd\nformanlsdya); further, the Ioc4sing, tve,the dat.pi.(less oftenloc,) asme,\nandthe loc.pi.yusme:the final eofthese forms isuncombinable(or\npragrhya: 138 b).The datives inbhyamareinRV. notseldom toberead\nasifinbhya,with loss ofthe final nasal; asmdkam andyusmdkam suffer\nthesame loss only inarare instance ortwo. The usual resolutions of\nsemivowel tovowel aremade, and areespecially frequentintheforms of\nthesecond person (tuam fortvdm, etc.).\nButtheduals, aboveall,wear avery differentaspectearlier. InVeda\nandBrahmana, thenominatives areavdm andyuvdm, andonly theaccusa-\ntivesavdm andyuvdm (butinRV. thedual forms of1stpers. chance not\ntooccur, unless invam[?J, once,fora-uam); the instr. inRV. iseither\nyuvdbhyam (notelsewhere found) oryuvabhyam; anabl.yuvdt appears once\ninRV., andavdt twice inTS.; thegen. -loc. isinRV.(only) yuv6s instead\nofyuvdyos. Thus wehave here adistinction(elsewhere unknown) offive\ndifferent dual cases byendings, inpart accordant with those oftheother\ntwonumbers.\n493. Peculiar endings. Theending am,appearing inthenom.\nsing, andpi.(andVedic du.) ofthese pronouns, willbefoundoften, though\nonly insing., among theother pronouns. Thebhyam (orhyam) ofdat. sing.\nandpi.ismetwith only here;itsrelationship with thebhyam, bhyas, bhis\noftheordinary declension ispalpable. The t(ord)oftheabl., though here\npreceded byashort vowel,isdoubtless thesame with that ofthea-declension\nofnouns andadjectives.That thenom., dat., and abl.endings should be\nthesame insing, andpi.(and inpartintheearlier du.also), only thestem\ntowhich they areaddedbeing different,isunparalleled elsewhere inthe\nlanguage. Theelement smaappearingintheplural forms willbefound\nfrequentintheinflection ofthesingular inother pronominal words :infact,\nthecompound stem asma which underlies theplural ofaham seems tobe\nthesame that furnishespart ofthesingular forms ofat/am (501), and its\nvalue of'we' tobeaspecialisation ofthemeaning 'thesepersons'. The\ngenitives singular, mama andtdva, have noanalogies elsewhere; thederi-\nvation fromthem oftheadjectives mamaka andtavaka(below, 516) suggests\nthepossibility oftheirbeing themselves stereotyped stems. The gen. pi.,\nasmakam andyusmdkam, arecertainly ofthischaracter :namely, neutersing,\ncase-forms oftheadjective stems asmdka andyusmaka 1other cases ofwhich\narefound intheVeda.\n494. Stem-forms. TotheHindu grammarians, the\nstems ofthepersonal pronouns aremadandasmad, andtvad\nandyusmad, because these areforms used toacertainextent,andallowed tobeindefinitely used, inderivation andcompo-sition(like tad, kad, etc.: seebelow, under theotherpronouns).Words arethusformed from them even intheolder language\n"
+    },
+    {
+        "page": 191,
+        "content": "4951 PERSONAL PRONOUNS. 171\nnamely, mdtkrta, mdtsakhi, asmdtsakhi, tvddyoni,mattds(AV.),\ntvdtpitr (TS.), yuvdddevatya (QB.);butmuch more numerous are\nthose thatshow theproper stem ina,orwith thealengthened\nto^a: thus,^mavant;asmatra, asmadruh, etc.; tvadatta, tvanid,\ntvnvasu, tvtihata, etc.; yusmftdatta, yusmesita, etc.; yuvddhita,\nyuvndatta, yuvUriita,etc.And thelaterlanguagealsohasafew\nwords made inthesameway,asmadrc.\nTheVedas have certain more irregular combinations, withcomplete\nforms :thus, tvamkama, mampapyd, mamasatyd, asmehiti, ahampurvd, aham-\nuttard, aharhyu, ahamsana.\nFrom thestems ofthegrammarians come also thederiv-\nativeadjectives madiya, tvadkya, asmadiya, yusmadlya, havinga\npossessivevalue :seebelow, 516.\nFor svaandsvaydm,seebelow, 513.\nDemonstrative Pronouns.\n495.Thesimplest demonstrative, rTta,which answers\nalsothepurposeofapersonal pronounofthethirdperson,\nmaybetaken asmodel ofamode ofdeclension usual in\nsomany pronouns andpronominal adjectivesthat itisfairly\ntobecalled thegeneral pronominaldeclension.\nBut this root hasalso thespecial irregularitythat inthe\nnom.sing. masc. and fern, ithassds(forwhosepeculiar euphonic\ntreatment see176a)andsa,instead ofidsand ta(compareGr.\no, 7],TO,andGoth,sa,so, thata}. Thus:\nSingular:\nm. n.\nN.\nA.sds\ntdmtat\ntdtsa\ntarn\ntena\ntdsmai\nAb.\ntdsmattdya\nrT^\"\ntdsyai\nrTHTI\ntdsyas\n"
+    },
+    {
+        "page": 192,
+        "content": "172VII.PRONOUNS.[495\n*\\\ntdsya tdsyas\nL.\ntdsmin tdsyam\nDual:\nN.A.V. ftt ft ft\ntau te te\nI.D.Ab.\nthbhyam thbhyam\nG.L.HU^\ntdyos\nPlural:\nN. ft ftlH\nte thni\nA.ftT^ftTH\nthn thni\nD.Ab.\ntebhyas thbhyai\ntasam\nL.ftj\ntesu thsu\nTheVedas show noother irregularities ofinflection than those which\nbelongtoallstems inaanda:namely, tendsometimes;usually tdfortali,\ndu.;often idfortdni, pi.neut.;usually tebhis fortats,instr.pi. ;andthe\nordinary resolutions. TheRV. hasonemore case-form from the root sa,\nnamely sdsmin (occurring nearly half asoften astdsmin).\n496. Thepeculiaritiesofthegeneral pronominal declension,\nitwillbenoticed, arethese :\nInthesingular, theuseoft(properly d)asending ofnom.-acc. neut.;\nthecombination with theroot ofanother element smainmasc. andneut.\ndat., abl., and loc.,and ofsyinfern, dat., abl.-gen., and loc.; andthe\nmasc. andneut. loc.ending in,which isrestricted tothisdeclension (except\nintheanomalous yddffmin, RV., once).\nThedual isprecisely that ofnoun-stems inaand d.\n"
+    },
+    {
+        "page": 193,
+        "content": "500]DEMONSTRATIVE PRONOUNS. J73\nIntheplural,theirregularities arelimited toUforidsinnom.masc.,\nandtheinsertion ofsinstead ofnbefore amofthegen., thestem-final\nbeing treated before itinthesamemanner asbefore suofthe loc.\n497. Thestem ofthispronounisbythegrammarians given\nastad; andfrom thatform come, infact, thederivative ad-\njective tadiya,withtattvd, tadvat, tanmaya; andnumerous com-\npounds, such astacchila, tajjna, tatkara, tadanantara, tanmatra,\netc. These compoundsarenotrareeven intheVeda :sotddanna,\ntadvid, tadvacd,etc. But derivatives from thetrue root taare\nalsomany: especially adverbs, astatas, tdtra, tatha, tada; the\nadjectivestavant andtdti;andthecompound tadfc etc.\n498. Though thedemonstrative root taisprevailinglyof\nthethirdperson,itisalsofreely used, both inthe earlier\nlanguage and inthelater, asqualifying thepronouns ofthe\nfirstandsecond person, giving emphasistothem :thus,so'Mm,\n'thisI',or 'Ihere'; sdtvdm, 'thou there';tevayam,'wehere';\nandsoon.\n499.Two other demonstrative stemsappeartocontain ta\nasanelement; and both, like thesimple ta,substitute sain\nthenom.sing. masc. and fern.\na.The one, tya,istolerably common(although only a\nthird ofitspossible formsoccur)inRV., butrare inAV., and\nalmost unknown later: itsnom.sing.,inthethreegenders,is\nsyds, sya, tydt, and itmakes theaccusativestydm, tyam, tydt,\nandgoes onthroughtheremaining cases inthesamemanner\nasta. IthasinRV. the instr. fern,tya (for tydya].\nb.The other istheusual demonstrative ofnearerposition,\n'this here', and isinfrequentusethroughallperiodsofthe\nlanguage.Itprefixesetothesimple root, formingthenomin-\nativesesds, esa,etdt and soonthroughthewhole inflection.\nThestemtyahasneither compounds nor derivatives. But\nfrom etaareformed both,inthesame manner asfrom the\nsimple ta,onlymuch lessnumerous :thus, etadda(QB.),etad-\nartha, etc., from theso-called stem etat;and etadfc and etavant\nfrom eta.\n\\500. There isadefective pronominal stem, ena,which is\naccentless, andhence used onlyinsituations where noemphasis\nfallsuponit. Itdoes notoccur elsewhere than intheaccusa-\ntive ofallnumbers, the instr.sing., and thegen.-loc. dual:\nthus,\nm. n. f.\nSing. A.enam enat enam\nI. enena enaya\n"
+    },
+    {
+        "page": 194,
+        "content": "174VII.PRONOUNS.[500\nDu. A. enau ene ene\nG.L. enayos enayos\nPI. A. enan enani enas\nTheRV. hasenosinstead ofenayos, andinoneortwoinstances accents\naform: thus, endm, endsC?).\nThisstem forms neither derivatives norcompounds.\n501.Two other demonstrative declensions aresoirreg-\nularly made upthatthey have tobegiveninfull. The\none, %|t|Haydm etc.,isused asamore indefinite demon-\nstrative,'this' or'that'; theother, ERTTasauetc., signifies\nespeciallytheremoterrelation, 'yon'or'yonder'.\nTheyareasfollows :\nSingular:\nm. n.\nN.\nA.aydm iddm\nimam iddm\nanena\nD.\nAb.\nG.\nL.\nN.A.\nI.D.Ab.\nG.L.asmat\nasyd\nasrrnn\nDual:\n^TT <JR\nimau ime\nabhyUmf.\niydm\nimam\nandya\nasyai\n^HJIH\nasyas\nEiHrra\nasyas\nSJUJIH\nasyam\nimem.\namum adds\no\namunaamum\n*HUI\namuya\nOJSamusmatamusyai\n'O*^\namusyas\namusya\namusmino_-s\namusyas\namusyam\ns\namubhyam\nanyos amuyos\n"
+    },
+    {
+        "page": 195,
+        "content": "5021 DEMONSTRATIVE PRONOUNS.175\nPlural :\nN.\nA.ime imani imas\nm imani masEPTR\namuni\n^^5FTFT\namun amuni\nebkis abhis amlbhis amubhis\nD.Ab.\nG.\nL.ebhyds\nesam\nesuabhyds\nasamf t^\namibhyas amubhyas\namisam amusam\nasu amusu\nThesame forms areused intheolderlanguage, withoutvariation, except\nthatimaoccurs forimau andimdni. andamu foramuni; amuya when used\nadverbially isaccented onthefinal, amuya; asau (with accent, ofcourse, on\nthefirst, dsau)isused also asvocative.\n502. Theformer ofthese twopronouns, aydm etc., plainly shows itself\ntobepieced together from anumber ofdefective stems. Themajority of\nforms come from theroota,with which, asintheordinary pronominal de-\nclension, sma(f.sy)iscombined inthesingular. Allthese forms from a\nhave thepeculiarity that intheir substantive usethey areeither accented,\nasintheparadigm,oraccentless(like enaandthesecond forms fromdhdm\nandtvdm}. Theremaining forms arealways accented. From andcome, with\nentireregularity, anena, andya, andyos. Thestrong cases indualandplural,\nand inpartinsingular, come not less regularly from astem imd. And\naydm, iydm, iddm areevidently tobereferred toasimple root i(iddm being\napparently adouble form:id,like tadetc.,withending am).\nTheVeda hasfrom therootaalsotheinstrumental endandayd (used\ningeneral adverbially), andthegen. loc.du.ay6s; from zma, imdsyaoccurs\nonce inRV. TheRV. hasinasmall number ofinstances the irregular\naccentuation dsmai, dsya,dbhis.\nInanalogy with theother pronouns,iddm isbythegram-\nmarians regardedasrepresentative stem ofthispronominalde-\nclension\n;and itisactually found sotreated inavery small\nnumber ofcompounds (idammdyaandicldmrupa areofBrahmana\nage).A.Sregardstheactual stems, anafurnishes nothingfurther;\nfrom imacomes onlytheadverb imdtha(RV., once);butaand\n*furnish anumber ofderivatives, mostlyadverbial :thus, for\nexample, dtas, dtra, dtha;itds, id(Vedic particle), ida, ihd, itara,\ntm(Vedic particle), idfc, perhapsevdandevdm, and others.\n"
+    },
+    {
+        "page": 196,
+        "content": "176VII.PRONOUNS.[503\n503. Theother pronoun,asati etc., hasamft for itsleading stem, which\ninthesingulartakes incombination,like theo-stems, theelement sma\n(f. at/),andwhich shifts toami inpartofthemasc. andneut.plural. In\npart, too,like anadjective w-stem,itlengthensitsfinal inthefeminine.\nThegen. sing, amtisyaistheonly exampleinthelanguage oftheending\nsyaadded toany other than ana-stem. Thenom.pi.ami isunique in\nform;its Iis(likethat ofadual) pragrhya,orexempt from combination\nwith afollowing vowel (138 b).Asati andadds arealsowithoutanalogies as\nregards their endings.\nThegrammarians,asusual, treat adds asrepresentative\nstem ofthedeclension, and itisfound inthis character inan\nextremelysmall number ofwords, asadomula; adomdyaisof\nBrahmana age. TheQB. has also asaunaman. Butmost of\nthederivatives, asofthecases, come fromamu:thus, amutas,\namutra, amutha, amurhi, amuvdt, amuka.\nInthe older language occurs the root tva(accentless), meaning 'one,\nmany aone';itisoftenest foundrepeated,as'one' and 'another'. Itfollows\ntheordinary pronominal declension.\nFragments ofanother demonstrative root ortwo aremet with :thus,\ndmas, 'he', occurs inaformula inAV.and inBrahmanas etc.;av6s as\ngen.-loc. dual isfound inRV.\n;theparticle upoints toaroot u.\nInterrogative Pronoun.\n504.The characteristicpartoftheinterrogative pro-\nnominal root isRk;ithasthethree forms^ka,faki,\n3\\ku;butthewhole declensional inflection isfrom 3\\ka.O i\nexceptingthenom. -ace.sing, neut., which isfromfaki,\nandhastheanomalous formfaqkim(notelsewhere known\ninthelanguage from aneuter/-stem). Thenom. and\naccus.sing., then, areasfollows:\nm. n. f.\nN -^H faft SfiT\nMs kim ka\nMm kim\nandtherestofthedeclension ispreciselylike that ofrT\nta(above, 495).\nTheVeda has itsusualvariations,Teaandkebhis forkdni and kais. It\nalsohas, along withkfm, thepronominally regular neuter kdd; andkdm\n"
+    },
+    {
+        "page": 197,
+        "content": "509]INTERROGATIVE PRONOUN. 177\n(orkam)isafrequent particle. Themasc. formkis,corresponding tokim,\noccurs asastereotyped case inthecombinations ndkis andmdkis.\n505. Thegrammarians treat kirnasrepresentative stem of\ntheinterrogative pronoun ;and itisinfact soused inanot\nlargenumber ofwords, ofwhich afewkimmdya, kimkard,\nkimkamya, kimdevata, and thepeculiar kimyti goback even\ntotheVeda andBrahmana. Incloseranalogy with theother\npronouns,theform kad, acouple oftimes intheVeda(katpayd.\nkddartha), andnotinfrequently later,isfound asfirstmember of\ncompounds. Then, from the real rootska, ki,kuaremade\nmanyderivatives\n;andfrom kiandku,especially thelatter, many\ncompounds:thus, kdti, katha, kathdm, kada, katard, katamd, kdrhi;\nkiyant, kulfc; kutas, kutra, kuha, kvd, kucard, kukarman, ku-\nmantrin,etc.\n506. Various forms ofthispronoun,askad, kim,andku\n(and, rarely, ko),atthebeginningofcompounds, havepassed\nfrom aninterrogative meaning, through anexclamatory,tothe\nvalue ofprefixes signifying anunusualquality either some-\nthing admirable, or,oftener, something contemptible. This use\nbeginsintheVeda, butbecomes much morecommon inlater time.\n507. Theinterrogative pronoun,asinother languages,\nturns readilyinitsindependent use also toanexclamatory\nmeaning. Moreover,itisbyvarious addedparticles converted\ntoanindefinite meaning:thus, byca,cand, cit, dpi, va,either\nalone orwith the relativeya(below, 511) prefixed: thus, kdc\ncand, 'any one'; ndko'pi,'notany one'; yaniktinicit,'what-\nsoever'.Occasionally,theinterrogative byitselfacquires asimi-\nlarvalue.\nRelative Pronoun.\n508.The root ofthe relative pronounisET2/#,which\nfrom theearliestperiodofthelanguagehas lost alltrace\nofthedemonstrative meaning originally (doubtless) belonging\ntoit,and isused asrelativeonly.\n509. Itisinflected with entireregularity accordingto\ntheusual pronominaldeclension:thus,\nSingular. Dual. Plural,\nm. n. f. m. n. f. m. n. f.\n5 ETTR\nyds yat ya\nyam ydt yamye yani yas\nyau yeye\nyan yam yas,\nWhitney, Grammar. 12\n"
+    },
+    {
+        "page": 198,
+        "content": "178VH-PRONOUNS.[509\nyena ydya\nD. ETFl UHT ynbhyam\n^\nydsmai ydsyai yebhyas yabhyas\netc. etc. etc. -etc. etc.\nTheVeda shows itsusual variations ofthese forms: ydforydtiand for\nydni,and yebhisforj/afs ,-yosforI/CM/OS alsooccurs once;yena, withpro-\nlonged final,isinRV. twice ascommon asyena. Resolutions occur '.in\nydbhias,andyesaam andydsaam.\n510. The use ofyatasrepresentative stembegins very\nearly:wehaveydtkamaintheVeda, andyatkarin, yaddevatyd\nintheBrahmana;later itgrows moregeneral. From theproper\nrootcome also aconsiderable series ofderivatives :ydtas, ydti,\nydtra, ydtha, ydda, yddi, yavant, yatard, yatamd ;andthecom-\npound yadrc.\n511.Thecombination ofyawith katomake anindefinite\npronounhasbeen noticed above(507).Itsownrepetition\nasydd-yat givesitsometimes alikemeaning, wonthrough\nthedistributive.\n512.One ortwomarkedpeculiaritiesintheSanskrit use\noftherelative maybeherebriefly noticed :\na.Avery decided preference forputting therelative clause before that\ntowhich itrelates: thus, yahsunvatdh sdkhd tdsmafndraya gayata (RV.),\n'who isthefriend ofthesoma-presser,tothatIndra sing ye' ;yamyajnam\nparibhurdsisd{ddevesu gachati (RV.), 'whatoffering thouprotectest,that\nintruth goes tothe gods'; yetrisaptdh pariydntibald tesdm dadhdtu me\n(A.V.), 'what thrice seven goabout, .their strength may heassign tome';\nasdu y6adhardd grhas tdtra santvardyyah (AV.),'what house isyonder in\nthedepth, there letthewitchesbe'; sahdydnmedstitena(TB.), 'along\nwith thatwhich ismine';hansdndm vacanam yattutanmam dahati(MBh.),\n'butwhat thewords oftheswanswere, thatburns me';sarvasya locanam\ncastram yasya nd'styandha evasah(H.), 'who does notpossess learning,\ntheeyeofeverything, blind indeed ishe'. The other arrangementiscom-\nparatively unusual.\nb.Afrequent conversion ofthesubject orobject ofaverbbyanadded\nrelative into asubstantive clause :thus,me'mdm prd\"pat pdtiruseyo vadhti\nyah (AV.), 'may there notreach him ahuman deadly weapon' (lit'ly, 'what\nissuch aweapon'); partnopdhi ydddhdnam(AV.), 'protectofuswhat\nwealth[there is]';apamdrgd 'pamdrstuksetriydm capdthaf caydh (AV.),\n'may thecleansing plant cleanse away thedisease andthecurse'; puskarena\nhrtarh rdjyam yaccd'nyad vasu kimcana(MBh.), 'byPushkara wastaken\naway thekingdom andwhatever otherproperty [there was]'.\n"
+    },
+    {
+        "page": 199,
+        "content": "516 PRONOMINAL DERIVATIVES. 179\nEmphatic Pronoun.\n513. The isolated and uninflected pronominal word\nt<^H svayam (fromtherootsva] signifies 'self,own self.\nByitsform itappearstobeanom.sing. 7and itisoften-\nestused asnominative, butalongwithwords ofallpersons\nandnumbers; andnotseldom itrepresents other cases also.\nSvayamisalso used asastem incomposition:thus,\nsvayamja, svayambhu.But sva itself(usually adjective:below,\n516)hasthesame value incomposition:andeven itsinflected\nforms are(intheolder language very rarely)used asreflexive\npronoun.\nNouns usedpronominally.\n514. Thenoun atmdn, 'soul',iswidely employed,inthe\nsingular,asreflexive pronounofallthreepersons.\nTheadjective bhavant,f.bhavafi, isused(asalready pointed\nout :456)inrespectfuladdress assubstitute forthepronounof\nthesecond person.Itsconstruction with theverb isinaccord-\nance with itstrue character, asaword ofthethirdperson.\nPronominal Derivatives.\n515.From pronominalrootsandstems, aswell asfrom\nthelargerclass ofroots andfrom noun-stems, areformed\nbytheordinarysuffixes ofadjectivederivation certain words\nand classes ofwords, which have thusthecharacter ofpro-\nnominal adjectives.\nSome ofthemore'importantofthesemay bebriefly no-\nticed here.\n516.Possessives. From therepresentative stems mad\netc.areformed theadjectives madiya, asmadiya, tvadiya, yusmadlya,\ntadtiya,andyadkya,which areused inapossessivesense :'relat-\ningtome, mine', and soon.\nOtherpossessivesaremamaM(alsomdmaka, RV.)andtavakd,\nfrom thegenitives mama and tava.\nAnanalogous derivative from thegenitive amusyaisamusyayanti (AV.\netc.), 'descendant ofsuch aone'.\nItwaspointed outabove (493)that the\"genitives\" asmdkam andyu-\nsmdkam arereally stereotypedcases ofpossessive adjectives.\n12*\n"
+    },
+    {
+        "page": 200,
+        "content": "180VII.PRONOUNS.[516\nCorrespondingtosvaydm '513)isthepossessive sva,mean-\ning'own', asrelatingtoallpersons andnumbers. TheRV.\nhasonce thecorresponding simple possessiveofthesecondper-\nson, tvd, 'thy'.\nFortheuseofsvaasreflexive pronoun,seeabove, 513, end.\nAllthese words form their feminines ina.\nOther derivatives ofalike value have noclaim tobementioned here.\nBut(excepting sva)thepossessivesaresorarely used astomake buta\nsmall figureinthelanguage, whichprefers generally toindicate thepossessive\nrelation bythegenitivecase ofthepronounitself.\n517.Bythe suffix vant areformed from thepronominal\nroots, with prolongationoftheir final vowels, theadjectives\nmavant, tv&vant, yusmnvant, yuvtivant, tavant, ettivant, ytivant, mean-\ning'ofmy sort, like me', etc. Ofthese, however, only the\nlast three areinuse inthe later language,inthesense of\n'tantus and'quantus. Theyare inflected like otheradjective\nstems invant,makingtheir feminines invatl(452if.).\nWords ofsimilar meaning from theroots tand Idareiyant\nandkiyant,inflected inthesame manner: seeabove, 451.\n518. Thepronominalroots show alikeprolongation of\nvowel incombination with therootdrc, 'see, look', and its\nderivatives drcaand(quite rarely) dfksa: thus, madrc, madrca;\nasmadrca:tvadrg etc.; yusmadfc etc.; tadfc etc.; etadfc, etadfca,\netadfksa (VS.); yadfc etc.; idfcandIddfcetc.Theymean 'ofmy\nsort, like orresembling me',andthelike, andthe last fiveare\nnotuncommon, with thesense ofltalijand'qualis1\n.Theforms\nindrcareunvaried forgender ;those indrga (anddrksa?]have\nfeminines in t.\n519.Fromto,ka,yacometdti, 'somany', Mti, 'how\nmany?' ydti,'asmany'. They have aquasi-numeral character,\nandareinflected(likethenumerals pdncaetc.: above, 483) only\nintheplural, andwith thebarestem asnom. andaccus.: thus,\nN.A.tdti;I.etc.tatibhis, t&tibhyas, tdtmam, tdtisu.'\n520.Fromya(inV.andBr.)andkacome thecompara-\ntivesandsuperlatives yatard andyatamd, andkatard andkatamd;\nandfromi,thecomparative itara. For their inflection, see\nbelow, 523.\n521. Derivatives with the suffix ka,sometimes conveying\nadiminutive oracontemptuous meaning,aremade from certain\nofthepronominal roots andstems(andmay, accordingtothe\ngrammarians, bemade from themall): thus, from fa,takam,\n"
+    },
+    {
+        "page": 201,
+        "content": "526 ADJECTIVES DECLINED PEG-NOMINALLY. 181\ntakdt, takas; from sa,saka :fromya,yakds, yaka, yake:from\nasau, asakau: from amu, amuka.\nForthenumerous andfrequently used adverbs formed frompronominal\nroots,seeAdverbs (below, chapter XVI.).\nAdjectives declinedpronominally.\n522.Anumber ofadjectives some ofthemcoming\nfrom pronominalroots, others more orlessanalogous with\npronounsinuse areinflected, inpartorwholly, accord-\ningtothepronominaldeclension(like cTta.495\ny.Thus :\n523. Thecomparatives andsuperlatives from pronominal\nroots namely, katard andkatamd, yatard andyatamd, anditara;\nalsoanyd, 'other', and itscomparative anyatard aredeclined\nlike tathroughout.Their feminine stems areina.\nButeven from these words forms made according totheadjective de-\nclension aresporadically metwith(e.g.itarayam, K.).\n524. Others words are soinflectedexceptinthenom.-\nacc.-voc. sing, neut., wherethey have theordinary adjective\nform am, instead ofthepronominalat(ad). Such aresdrva,\n'all', vtcva, 'all, every', eka, 'one'.\nThese, also,arenotwithout exception,atleast intheearlier language\n(e.g.vfyvaya, vfyvat, RV.;eke loc. sing., AV.).\n525. Yet other words follow thesame modelusually,or\ninsome oftheirsignifications,oroptionally;butinother senses,\norwithout known rule, lapseinto theadjectiveinflection.\nSuch arethecomparatives and superlatives fromprepositional stems:\nddhara andadhamd, dntara anddntama, dpara andapamd, dvara andavamd,\nuttara anduttamd, upara andupamd. Ofthese, pronominal forms arede-\ncidedly more numerous from thecomparatives than from thesuperlatives.\nFurther,thesuperlatives (without corresponding comparatives) paramd.\ncaramd, madhyamd; andalsoanyatama (whose positive andcomparative belong\ntotheclass firstmentioned: 523).\nFurther,thewords para, 'distant, other';purva, 'prior,east';ddksina,\n'right, south'; ubhdya (f.ubhdylorubhayi),'ofboth kinds orparties';and\ntheraresama (accentless), 'any orevery one1\n,simd, 'each, all',nema, 'the\none, half;andthepossessivesvd.\n526. Occasional forms ofthepronominal declension aremetwith from\nnumeral adjectives:e.g.prathamdsyas, trtiyasydm; andfrom other words\nhaving anindefinite numeral character: thus, dlpa,'few': ardhd, 'half;\nkevala, 'all'; dvtiaya,'ofthetwokinds' and others.\n"
+    },
+    {
+        "page": 202,
+        "content": "182 [527-\nCHAPTER VIII.\nCONJUGATION.\nf527.THEsubjectofconjugationorverbal inflection\ninvolves,asintheother languagesofthefamily,thedis-\ntinctions ofvoice, tense, mode, number, andperson.\nThen,besides thesimplerorordinary conjugationof\naverbal root,there arecertain more orlessfullydeveloped\nsecondaryorderivative conjugations.\n528.Voice. There are(asinGreek) twovoices, active\nandmiddle, distinguished byadifference inthepersonal\nendings.This distinction isapervadingone: there isno\nactive personalformwhich does nothave itscorresponding\nmiddle, and viceversa; and itisextended also inpartto\ntheparticiples (butnottotheinfinitive).\n529.Anactive form iscalled bytheHindu grammarians\nparasmai padavn,'aword foranother', and amiddle form is\ncalled atmane padam,'aword forone's self: theterms might be\nbestparaphrased by'transitive' and 'reflexive'. And thedistinc-\ntionthus expressedisdoubtless theoriginal foundation ofthe\ndifference ofactive andmiddle forms :intherecorded condition\nofthelanguage, however, the antithesis oftransitive andre-\nflexive meaningisinnosmall measure blurred, oreven alto-\ngether effaced.\n530.Some verbs areconjugatedinboth voices, others\ninoneonly; sometimes apartofthetenses areinflected\nonlyinonevoice, othersonlyintheother orinboth; of\naverbusuallyinflected inonevoicesporadic forms ofthe\nother occur; andsometimes thevoice differs accordingas\ntheverb iscompounded with certainprepositions.\n531.Themiddle forms outside thepresent-system (for\nwhich there isaspecial passive inflection: seebelow.768),\n"
+    },
+    {
+        "page": 203,
+        "content": "533]TENSES ANDMODES. 183\nandsometimes alsowithin thatsystem,areliahle tobeused\nlikewise inapassivesense.\n532.Tense. The tenses areasfollows: 1.apresent,\nwith 2.animperfect, closelyrelated with itinform, having\naprefixed augment;3.aperfect, made withreduplication\n(towhich intheVeda isadded,4.aso-calledpluperfect,\nmade from itwithprefixed augment);5.anaorist,ofthree\ndifferent formations :a.simple;b.reduplicated;c.sibilant;\n6.afuture, with 7.aconditional, anaugment-tense,stand-\ningtoitintherelation ofanimperfecttoapresent;and\n8.asecond,aperiphrastic,future(notfound intheVeda).\nThe tenses here distinguished (inaccordance withprevailing\nusage)asimperfect, perfect, pluperfect, andaorist receive those\nnames from their correspondenceinmode offormation with tenses\nsocalled inother languagesofthefamily, especiallyinGreek,\nandnot atallfrom differences oftime designated bythem. In\nnoperiodoftheSanskrit languageisthere anyexpressionof\nimperfectorpluperfecttime nor ofperfect time, exceptin\ntheolder language, where the\"aorist\" has thisvalue;later, im-\nperfect, perfect, andaorist(ofrareuse)aresomany undiscrim-\ninatedpasttenses orpreterits:seebelow, under thedifferent\ntenses.\n533.Mode. Inrespecttomode, thedifference be-\ntween the classical Sanskrit andtheolder languageofthe\nVeda and,inalessdegree,oftheBrahmanas is\nespecially great.\nIntheVeda, thepresent tense has, besides itsindicative\ninflection, asubjunctive,ofconsiderablevarietyofformation,\nanoptative, andanimperative (in2dand3dpersons).The\nsame three modes arefound, thoughofmuch lessfrequentoc-\ncurrence, asbelongingtotheperfect; andtheyaremade also\nfrom the aorists, beingofespecial frequency from thesimple\naorist. The future hasnomodes(anoccasional case ortwo\narepurely exceptional).\nIntheclassicalSanskrit, thepresentadds toitsindic-\nativeanoptative andanimperativeofwhichlast,more-\nover, the firstpersonsarearemnant oftheoldsubjunc-\n"
+    },
+    {
+        "page": 204,
+        "content": "184VIII. CONJUGATION.[533\ntive. And the aorist has alsoanoptative,ofsomewhat\npeculiar inflection, usuallycalled theprecative (orbene-\ndictive).\n534.Thepresent, perfect,andfuture tenses haveeach\nofthem, alike intheearlier and laterlanguage,apairof\nparticiples,active andmiddle, sharinginthevariouspe-\nculiarities ofthetense-formations;and intheVeda are\nfound suchparticiples belongingalso tothe aorist.\n535.Tense-systems. The tenses,then, with their\naccompanying modes andparticiples,fallinto certain well-\nmarkedgroupsorsystems:\nI.Thepresent-system, composedofthepres-\nenttense with itsmodes,itsparticiple, and itspret-\neritwhich wehave called theimperfect.\nII.Theperfect-system, composedoftheper-\nfecttense(with,intheVeda,itsmodes and itspreterit,\ntheso-calledpluperfect) and itsparticiple.\nIII.Theaorist-system,orsystems, simple,\nreduplicated, and sibilant, composedoftheaorist tense\nalongwith, inthelaterlanguage,its\"pnecative\" opta-\ntive(but.intheVeda, with itsvarious modes and its\nparticiple).\nIV.Thefuture-systems:a.theoldorsibi-\nlant future, with itsaccompanying preterit,thecondi-\ntional, and itsparticiple; and b.thenewperiphrastic\nfuture.\n536.Number andPerson. Theverb has, ofcourse,\nthesame three numbers with thenoun, namely singular,\ndual, andplural ;andineachnumber ithasthethreeper-\nsons, first, second, and third. Allofthese aremade in\nevery tense andmodeexceptthat the firstpersonsof\ntheimperative numbers aresupplied from thesubjunctive.\n"
+    },
+    {
+        "page": 205,
+        "content": "540\n1 VERBAL ADJECTIVES ANDNOUNS. 185\n537.Verbal adjectives andnouns: Partici-\nples. Theparticiples belongingtothetense-systems have\nbeenalready spokenofabove(534). There isbesides, com-\ningdirectly from theroot oftheverb, aparticiple, prevail-\ninglyofpastandpassive (orsometimesneuter) meaning.\nFuturepassive participles,orgerundives,ofseveral different\nformations, arealsomade.\n538.Infinitives. Intheolderlanguage,averycon-\nsiderablevarietyofderivative abstract nounsonlyina\nfewsporadic instanceshaving anythingtodowith thetense-\nsystems areused inaninfinitive orquasi -infinitive\nsense;most often inthedativecase,butsometimes also in\ntheaccusative, inthegenitive andablative, and(very rarely)\ninthelocative. Inthe classical Sanskrit, there remains a\nsingle infinitive, ofaccusative case-form,having nothing\ntodowith thetense-systems.\n539.Gerund. Aso-called gerund (orabsolutive)-\nbeing,like theinfinitive, astereotypedcase-form ofade-\nrivative noun isapartofthegeneral verb-systemin\nboth the earlier and laterlanguage, being especiallyfre-\nquentinthe latter. IntheVeda ithasasomewhat various\nform;inthelaterlanguage,ithasonlytwoforms, onefor\nsimple verbs, andtheother forcompound.Itsvalue isthat\nofanindeclinable activeparticiple,ofindeterminate but\nprevailingly pasttense-character.\nAsecond gerund,anadverbially used accusative inform,\nisfound, butonly rarely, both earlier and later.\n540.Secondary conjugations. Thesecondaryor\nderivativeconjugationsareasfollows: a.thepassive;b.the\nintensive; c.thedesiderative; d.thecausative. Inthese,\naconjugation-stem,instead ofthesimpleroot, underlies\nthewhole systemofinflection. Yetthere isclearlytobe\nseen inthem thecharacter ofapresent-system, expanded\n"
+    },
+    {
+        "page": 206,
+        "content": "L186VIII. CONJUGATION.[540\ninto amore orlesscomplete conjugation;andthepassive\nissopurelyapresent-systemthat itwillbedescribed in\nthechapterdevoted tothatpartofthe inflection ofthe\nverb.\nUnder thesamegeneralhead belongsthesubjectof\ndenominative conjugation,ortheconversion ofnoun and\nadjective-stemsintoconjugation-stems. Further, that of\ncompound conjugation,whether bytheprefixionofprepo-\nsitions toroots orbytheaddition ofauxiliaryverbs tonoun\nandadjective-stems. Andfinally,that ofperiphrasticcon-\njugation,orthelooser combination ofauxiliaries with verbal\nnouns andadjectives.\n541.The characteristic ofaproper (finiteorpersonal)\nverb-form isitspersonal ending. Bythis alone isdeter-\nmined itscharacter asregards number andperson and\ninpartalso asregards mode and tense. Butthedistinc-\ntions ofmode andtense aremainly made bytheformation\noftense andmode-stems,towhich, instead oftothepure\nroot, thepersonal endingsareappended.\nInthischapterwillbegiven ageneralaccount oftheper-\nsonalendings, and also oftheformation ofmode-stems from\ntense-stems, and ofthose elements intheformation oftense-\nstems theaugment andthereduplication which arefound\ninmore than onetense-system. Then, inthefollowing chapters,\neachtense-system will betaken upbyitself, andthemethods\nofformation ofitsstems, both tense-stems andmode-stems, and\ntheir combination with theendings,willbedescribed indetail.\nPersonalEndings.\n542. Theendingsofverbal inflectionare, aswaspointed\noutabove, differentthroughout intheactive andmiddle voices.\nTheyarealso, asinGreek, usuallyoftwosomewhat varying\nforms forthesamepersoninthesame voice :one fuller, called\nprimary ;theotherbriefer, calledsecondary. There arealso less\npervading differences, depending upon other conditions.\nAcondensed statement ofallthe varieties ofending foreach person\nandnumber here follows.\n"
+    },
+    {
+        "page": 207,
+        "content": "545]PERSONAL ENDINGS. 187\n543.Singular: First person. Theprimary ending in\ntheactive ismi.Thesubjunctive, however(later imperative),\nhas niinstead;and inthe oldest Veda this niissometimes\nwanting, and theperson ends ina(asiftheniofamwere\ndropped).Thesecondary endingism;and tothismanahas\ncome tobesopersistently prefixed, appearing always where the\ntense-stem does not itself end ina(vamforvarm orvaram in\nRV., once, isanisolatedanomaly),that itisnecessarytoreckon\namasending,alternate with m.But theperfecttense has\nneither minorm;itsendingissimply a(sometimesa:248c):\nor,from -roots, au.\nTheprimary middleending, accordingtotheanalogyof\ntheotherpersons, would beregularly me. Butnotense or\nmode,atanyperiodofthelanguage, shows anyrelic whatever\nofaminthisperson:theprimary ending, presentaswell as\nperfect, from a-stems and othersalike,ise;and toitcorre-\nsponds*assecondary ending, which blends with the final ofan\na-stem toe.Theoptative has,however, ainstead ofi;and\ninthesubjunctive (later imperative) appearsaifor*e.\n544.Second person. Intheactive, theprimary ending\nissi,which isshortened tosassecondary:astotheloss of\nthis safter afinal radical consonant, seebelow, 555. Butthe\nperfect and theimperativedesert hereentirelytheanalogyof\ntheother forms. Theperfect endingisinvariablytha(ortha,\n248c).Theimperativeisfarless regular. The fullest form\nofitsendingisdhi;which, however,ismore often reduced to\nhi;and inthegreat majorityofverbs(includingalla-stems,\natevery periodofthelanguage) noendingispresent,butthe\nbare stem stands aspersonalform. Inavery small class of\nverbs(722)ana istheending. TheVeda hasalsoanending\ntat;and this iseven usedsporadicallyinother personsofthe\nimperative (seebelow, 5701).\nInthemiddle voice, theprimary ending,bothpresentand\nperfect,isse.Thesecondarystands innoapparentrelation to\nthis, being thas; and intheimperativeisfound onlysva(or\nsva:248c),which intheVeda isnotseldom toberead asma.\nIntheolderlanguage,seissometimes strengthenedtosaiinthe\nsubjunctive.\n545.Third person. The active primary endingisti;\nthesecondary,t:astotheloss ofthelatter after afinal radical\nconsonant, seebelow, 555. But intheimperative appearsin-\nstead thepeculiar endingtu;and intheperfectnocharacteristic\nconsonant ispresent, andthethird personhasthesame ending\nasthe first.\n"
+    },
+    {
+        "page": 208,
+        "content": "188VIII. CONJUGATION.[545\nTheprimarymiddle endingiste,with taascorresponding\nsecondary.Intheolder language,teisoften strengthened to\ntaiinthesubjunctive.Intheperfect,themiddle thirdperson\nhas, like the active, thesame ending with thefirst, namelye\nsimply;andintheolder language,thethird person present also\noften loses thedistinctive partofitstermination, andcomes to\ncoincide inform with the first. Tothis eperhaps corresponds,\nassecondary,the toftheaorist 3dpers. passive (842 ff.).The\nimperativehas tarn(or,intheVeda, rarely am]for itsending.\n546.Dual: First person. Both inactive andinmiddle,\nthedual firstpersonisinall itsvarietiespreciselylike the\ncorresponding plural, onlywith substitution ofvforthemofthe\nlatter :thus, vas(novasihasbeen found tooccur), va, vahe,\nvahi, vahai. Theperson is,ofcourse, ofcomparativelyrare\nuse,andfrom theVeda noform inpas, even,isquotable.\n547.Second andThird persons. Intheactive, the\nprimary endingofthesecondpersonistkas, and that ofthe\nthird istas;and this relation ofthtotappearsalso inthe\nperfect, andruns throughthewhole series ofmiddleendings.\nTheperfect endingsareprimary,buthave uinstead ofaas\nvowel;andanahasbecome sopersistently prefixedthat their\nforms have tobereckoned asathus andaim. Thesecondary\nendingsexhibit nodefinable relation totheprimaryinthese\ntwopersons;theyare tarnand tarn;andtheyareused inthe\nimperativeaswell.\nInthemiddle, alonga which, however, with thefinal\naofa-stems becomes e hasbecomeprefixedtoalldual\nendingsofthesecond andthirdpersons,soastoform anin-\nseparable partofthem. Theprimary endings, present andper-\nfect, areatheandcite; thesecondary (and imperative)areatham\nandatam(or,with stem-finala,etheetc.).\nTheRig-Veda hasavery fewforms inaithe andaite, apparently from\netheand etewith subjunctive strengthening (they are alldetailed below :\nsee615, 701, 737, 752, 836, 1008, 1043).\n548.Plural: First person. The earliest form ofthe\nactiveendingismast, which inthe oldest languageismore\nfrequent than thebriefer mas(inRV., asfive toone; inAV.,\nhowever, only asthree tofour).Inthe classical Sanskrit, mas\nistheexclusiveprimary ending; butthesecondaryabbreviated\nmabelongsalso totheperfect andthesubjunctive (imperative).\nIntheVeda, maoften becomes ma(248 c),especiallyinthe\nperfect.\nTheprimary middleendingismake. This islightenedin\nthesecondary form tomahi; and,ontheother hand, itisregu-\n"
+    },
+    {
+        "page": 209,
+        "content": "550] PERSONAL ENDINGS.]89\nlarly (intheVeda, notinvariably) strengthenedtomahai inthe\nsubjunctive (imperative).\n549.Second person. The activeprimary endingistha .\nThesecondary,alsoimperative, endingista(intheVeda, taonly\nonce inimpv.j. But intheperfect anycharacteristic consonant is\nwanting, andtheendingissimplya.IntheVeda, thesyllable\nna, ofproblematic origin,isnotinfrequently added toboth\nforms oftheending, making ihana(rarely thana]andtana. The\nforms inwhich this occurs will bedetailed below, under the\ndifferent formations :theaddition isvery rarely madeexcepting\ntopersonsofthe firstgeneral conjugation.\nThemiddleprimary endingisdhve, whichbelongstothe\nperfectaswell asthepresent. Inthesubjunctiveoftheolder\nlanguageitissometimesstrengthenedtodhvai. Thesecondary\n(and imperative) endingisdhvam(inRV.,oncedhva];and\ndhvat isonce met with intheimperative (570).IntheVeda,\nthevofalltheseendingsissometimes resolved into u,andthe\nending becomesdissyllabic.\n550.Third person. The fullprimary endingisantiin\ntheactive, with ante ascorrespondingmiddle. Themiddle sec-\nondary endingisanta, towhich shouldcorrespondanactive\nant; but ofthe tonly altogether questionabletraces areleft,\nintheeuphonic treatment ofafinaln(207);theendingisan.\nIntheimperative,antuandantam take theplaceofantiand\nante. The initial aofallthese endingsislike that ofamin\nthe1stsing., disappearingafter the final aofatense- stem.\nMoreover, anti, antu, ante, antam, anta are allliable tobe\nweakenedbytheloss oftheir nasal, becomingatietc. Inthe\nactive, thisweakeningtakesplace onlyafterreduplicated stems\n(andafter afew roots which aretreated asifreduplicated:\n639ff.);inthemiddle,itoccurs after alltense-stems save those\nendingina.\nFurther, forthesecondaryactive ending anthere isasub-\nstitute tts(orur :169, end),which isused inthesame reduplicating\nverbs thatchangeanti toatietc., andwhich accordingly appears\nasaweaker correlative ofan.Thesame usisalsoused uni-\nversallyintheperfect,intheoptative (notinthesubjunctive),\ninthose forms oftheaorist whose stem does notendina,and\nintheimperfectofroots endingina,and afewothers(621).\nTheperfect middle hasinallperiodsofthelanguage the\npeculiar ending re,andtheoptativehastheallied ran, inthis\nperson.IntheVeda, avarietyofother endings containingar\nasdistinctive consonant aremetwith :namely,re(and ire]and\nrate inthepresent ;rata intheoptative (bothofpresentandof\n"
+    },
+    {
+        "page": 210,
+        "content": "190VIII. CONJUGATION, 550\naorist);rireintheperfect;ranta, ran,andram inaorists(and\ninanimperfectortwo);ramandratam intheimperative. The\nthree rate, ratam, and rata arefound even inthelaterlanguage\ninoneortwoverbs(629).\n551. Below aregiven,forconvenience, intabular form,\ntheschemes ofendingsasacceptedinthe classical orlater\nlanguage:namely,a.theregular primary endings, used inthe\npresentindicative and thefuture(andthesubjunctiveinpart);\nand b.theregular secondary endings, used intheimperfect,\ntheconditional, theaorist, theoptative (andthesubjunctive in\npart);and further, ofspecial schemes, c.theperfect endings\n(chiefly primary, especiallyinthemiddle);andd.theimperative\nendings (chiefly secondary). Totheso-calledimperative endings\nofthe first personisprefixedtheawhich ispractically apart\nofthem\n;though really containingthemode-signofthesub-\njunctive from whichtheyarederived.\n552. Further, apartoftheendingsaremarked with an\naccent, and apartare leftunaccented. The latter arethose\nwhich never, under anycircumstances, receive theaccent;the\nformer areaccented inconsiderable classes ofverbs, though by\nnomeans inall. Itwill benoticed that, ingeneral,theun-\naccentedendingsarethose ofthesingularactive\n;butthe2d\nsing, imperative hasanaccented ending ;and, ontheother\nhand, thewhole series of1stpersons imperative,active and\nmiddle, have unaccentedendings (thisbeingacharacteristic of\nthesubjunctive formation whichthey represent).\n553. Theschemes ofnormalendings, then, areasfollows :\na.Primary Endings.\n"
+    },
+    {
+        "page": 211,
+        "content": "557]PERSONAL ENDINGS. 191\nd.Imperative Endings.\n1ani dva dma ai avahdi dmahdi\n2dhi, 7if, tarn ta svd dthdm dhvdm\n3tu tdmdntu, dtu tdm dtdm dntam, dtdm\n554. Ingeneral, therule isfollowed thatanaccented ending,ifdis-\nsyllabic,isaccented onitsfirst syllable and theconstant union-vowels\nareregarded,inthisrespect,asintegral partsoftheendings. But the\n3dpi.ending ateofthepres.indie, middle has inRV. theaccent atein\nanumber ofverbs(see613, 685, 699, 718;; andanoccasional instance\nismetwith inother endings:thus, make(see719,735 .\n555. Thesecondary endings ofthesecond and third persons singular,\nasconsisting ofanadded consonant without vowel, should regularly (150)\nbelostwhenever theroot orstem towhich they aretobeadded itself ends\ninaconsonant. And this rule isingeneral followed; yetnotwithout ex-\nceptions. Thus :\na.Arootending inadental mute sometimes dropsthis finalmute\ninstead oftheadded sinthesecondperson ;and,ontheother hand, aroot\norstem ending inssometimes dropsthis sinstead oftheadded tinthe\nthird person ineithercase, establishing theordinary relation ofsand t\ninthesepersons,instead ofsands,ortand t.Asimilar loss ofany\nother final consonant before theendingisexceedingly rare. Forinstances,\nseebelow, 692.\nb.Again, aunion-vowel issometimes introduced before theending,\neither aorI:seebelow, 621, 631, 819,880.\nInafewisolated cases intheolderlanguage,this Iischanged toai :\nseebelow, 904b,1068.\n556. Thechangesofform which roots andstemsundergo\nintheir combinations with these endingswillbepointed outin\ndetail below, under thevarious formations. Heremaybesimply\nmentioned inadvance,asbyfarthemostimportant among them,\nadistinction ofstronger andweaker form ofstem inlargeclasses\nofverbs, standinginrelation with theaccent thestembeing\nofstronger formwhen theaccent fallsupon it,orbefore an\naccentlessending, and ofweaker formwhen theaccent ison\n.theending.\nOftheendings marked asaccented inthescheme, thetaof2dpi.is\nnotinfrequently intheVeda treated asunaccented, thetone resting onthe\nstem, which isstrengthened. Much lessoften, thetarn of2ddu. istreated\ninthesame way:other endings, only sporadically.\nSubjunctive Mode.\n557. Ofthesubjunctive mode(aswaspointedoutabove)\nonlyfragmentsare leftinthelater orclassical language:namely,\nintheso-called first persons imperative, and intheuse(58O)\n"
+    },
+    {
+        "page": 212,
+        "content": "192VIII. CONJUGATION.[557\noftheimperfectand aorist personswithout augmentaftermn\nprohibitive.Inthe oldestperiod, however,itwas averyfre-\nquent formation, beingthree orfour times ascommon asthe\noptativeintheRig-Veda, andnearlythesame intheAtharvan\n;\nbutalreadyintheBrahmanas itbecomes comparativelyrare.\nItsvarieties ofform areconsiderable, andsometimesperplexing.\n558. Initsmost normal andregular formation, aspecial\nmode-stem ismade forthesubjunctive byaddingtothetense-\nstem ana which combines with afinal aofthetense-stem\ntoa.The accent restsuponthetense-stem, whichaccordingly\nhasthestrong form. Thus, from thestrong present-stem dok\n(yduh)ismade thesubjunctive-stem doha; from juho (yhu),\njuhdva;fromyunaj C^yuj), yundja ;from bh&va(ybhu), bhdva;\nfrom tudd(ytud), tudti; fromucyd (pass., yvac), ucyti; and soon.\n559. Thestem thus formed isinflected ingeneral asan\na-stem would beinflected intheindicative, with constant accent,\nandaforabefore theendingsofthe firstperson (733)but\nwith thefollowing peculiaritiesastoendingetc. :\n560. Intheactive,the1st sing, has niasending: thus, ddhani,\nyundjdni,bhdvdni. But intheRig-Veda sometimes asimply:thus, at/a,\nbravd.\nIn1stdu., 1stpi.,and3dpi., theendings arealways thesecondary:\nthus, dohdva, dtihdma, d6han; bhdvdva, bhdvdma, bhdvdn.\nIn2dand3ddu.and2dpi., theendings arealways primary: thus,\nddhathas, ddhatas, dohatha;bhdvdthas, bhdvdtas, bhdvdtha.\nIn2dand3dsing., theendings areeither primaryorsecondary: thus,.\ndtihasi ordohas,d6hati orddhat; bhdvasi orbhdvds,bhdvdti orbhdvat.\nOccasionally, forms with double mode-sign a(byassimilation tothe\nmorenumerous subjunctives from tense-stems ina)aremetwith fromnon-\na-stems :thus,dsdtha from as;at/as, at/at, dydn from e\n561. Inthemiddle, forms with secondary instead ofprimary endings\narevery rare, being found only inthe3dpi.(where they aremore frequent\nthan theprimary), andinacase ortwo ofthe3dsing.\nThestriking peculiarity ofsubjunctive middle inflection isthefrequent\nstrengthening ofe'to diintheendings. This islessgeneral inthevery\nearliestlanguage than later. In1stsing.,dialone isfound asending, even\ninRV.; and in1stdu.also(ofrareoccurrence), only dvahdi ismetwith.\nIn1stpi.,dmahdiprevails inRV. andAV. (dmaheisfound afew times),\nand isaloneknown later. In2dsing.,sdifor sedoes notoccur inRV.,\nbut istheonly form inAV. andtheBrahmanas. In3dsing.,tdiforte\noccurs once inRV., and isthepredominant form inAV.,andtheonlyone\nlater. In2dpi.,dhvdi fordhve isfound inoneword inRV., and afew\ntimes intheBrahmanas. In3dpi.,nidi fornteistheBrahmana form (of\nfarfromfrequent occurrence);itoccurs neither inRV. norAV. Nosuch\n"
+    },
+    {
+        "page": 213,
+        "content": "564]SUBJUNCTIVE MODE. 193\ndual endingsasthaiandtai, forthe aridte,areanywhere found; butRV.\nhasinafewwords (nine: above, 527, end)aithe and aite, which appear\ntobealikesubjunctive strengthening ofetheand ete(although found inone\nindicative form, krnvdite}. Before theai-endings thepenultimate vowel is\nregularly long a;butantdi instead ofantdi istwo orthree times met with,\nandonce(TS.)atdi fordtdi.\n562. Thesubjunctive endings, then, incombination with\nthesubjunctive mode-sign, areasfollows :\nactive. middle.\n8. d. p. s. d.p.\n(dvahdi (dmahdi\n1am ova ama ai\n[avahe [amahe\n(asi (ase (adhve\n2i athas atha{aithe\n[as [asai [adhvai\n(ati (ate (ante, anta\n3 atas an aite <_\n[at [atai [antdi\nAnd, infurther combination with final aofatense-stem,\nthe initial aofalltheseendings becomes a:thus, forexample,\nin2dpers.:asioras,athas, atha, ase, adhve.\n563. Besides this proper subjunctive, withmode-sign, initstriple\nform with primary,with strengthened primary, andwith secondary end-\ningsthere isintheolder language another, without mode-sign andwith\nsecondary endings,orinallrespects coinciding with theforms ofanaugment-\ntense(imperfectoraorist) save fortheabsence oftheaugment. Subjunc-\ntives ofthis character arefrequent inRV., decidedly lesscommon inlater\nVedic, andvery little used intheBrahmanas except aftermdprohibitive\n(580)after which they stand alsointhelater language.\nThese forms aresometimes called\"imperfect subjunctive\", but the\nappellationisanevident misnomer:\"improper subjunctive\" ispreferable.\nSince (below, 587) theforms ofaugmented tenses are alsofreely used in\nanindicative sense without augment intheoldest Veda, thedistinction of\nthetwo classes ofuse isoften difficult tomake.\nAstotheuses ofthesubjunctive, seebelow, 572 ff.\nOptative Mode.\n564.Ashasbeenalready pointed out, theoptativeisof\ncomparativelyrare occurrence inthelanguageoftheVedas\n;\nbut itgains rapidlyinfrequency, andalreadyintheBrahmanas\ngreatly outnumbers thesubjunctive, which still later itcomes\nalmostentirelytoreplace.\nItsmode offormation isthesame inallperiodsofthe\nlanguage.\nWhitney, Grammar. 13\n"
+    },
+    {
+        "page": 214,
+        "content": "194VIII. CONJUGATION.[565\n565. The optative mode-signisintheactive voice adif-\nferent one, accordingasitisadded toatense-stem endingina,\norinsome other final. Inthelatter case,itisya,accented\n;\nthisyaisappendedtotheweaker form ofthetense-stem, and\ntakes theregularseries ofsecondary endings, with, in3dplur.,\nMSinstead ofan,andloss oftheabefore it.After ana-stem,\nitis,unaccented\n;this iblends with the final atoe(which\nthen isaccented ornotaccordingtotheaccent ofthea);and\nthe eismaintained unchangedbefore avowel-ending (am, us),\nbymeans ofaninterposed euphonic y.\nInthemiddle voice, themode-signis ithroughout, and\ntakes thesecondary endings,with ain1stsing., andranin\n3dpi.After ana-stem, therules astoitscombination toe,\ntheaccent ofthelatter, and itsretention before avowel-ending\nwithinterpositionofay,arethesame asintheactive. After\nanyother final, theweaker form ofstem istaken, andtheac-\ncent isontheending (exceptinone class ofverbs, where it\nfallsuponthetense-stem :see645);andthe I(aswhen com-\nbined toe)takes aninserted ybefore avowel-ending.\nItis,ofcourse, impossibletotellfrom theform whether ior1is\ncombined with the final ofana-stem toe;butnogood reasonappearsto\nexist forassuming i,rather than the Iwhich shows itself intheother class\nofstems inmiddle voice.\n566. Thecombined mode-sign andendingsoftheoptative,\nthen, areasfollows, intheir double form, fora-stems and for\nothers :\n\\a.fornon-a-stems.\n"
+    },
+    {
+        "page": 215,
+        "content": "570]OPTATIVE MODE. 195\nprecative from thesimple aorist, themiddle from thesibilant\naorist arepracticallyofrareoccurrence atevery period ofthe\nlanguage, andespeciallylater.\nThe inserted sruns inthe active through thewhole series ofpersons;\ninthemiddle,itisallowed only inthe2dand3dpersons sing, anddu.,\nandthe2dpi.,and isquotable from theolder literature only forthe2dand\n3dsing. Inthe2dsing, act., theprecative form, byreason ofthenec-\nessary loss oftheaddeds,isnotdistinguishable from thesimple optative;\ninthe3dsing, act., thesame isthe case inthe laterlanguage, which\n(above, 555) saves thepersonal endingtinstead oftheprecative-sign s;\nbut theRV.usually, and theother Vedic texts tosomeextent, have the\nproper ending yds (for yast).\nTheaccent isasinthesimple optative.\n568.Theprecative endings, then, acceptedinthe later\nlanguage (including,inbrackets, those which areidentical with\nthesimple optative),areasfollows :\nactive. middle,\ns. d.p.s. d.p.\n1yasam yasva yasma [iy&] [wdhi] [Imdhi]\n2[yas] yastam yastaisihas lyasthdm idhvdm\n3[ydt] ydstdm ydsusistd iydstdm [Iran]\nAstotheuses oftheoptative, seebelow, 572 ff.\nImperative Mode.\n569.Theimperativehasnomode-sign;itismadeby\naddingitsownendings directlytothetense-stem, justas\ntheother endingsareadded toform theindicative tenses.\nHence, in2dand3ddu.and2dpi.,itsforms areindistinguishable\nfrom those oftheaugment-pretexitfrom thesame stem with itsaugment\nomitted.\nTherules astotheuseofthedifferent endings especiallyin2dsing.,\nwhere thevariety isconsiderable willbegiven below,inconnection with\nthe -various tense-systems. Theending tat,however,hassomuch that is\npeculiarinitsusethat itcalls foralittle explanationhere.\n570.TheImperativeintat. This isnowhere afrequent form,\nthough found from the earliest periodofthelanguage, andallowed tobe\nmade inthelatest;inthefiveleading Vedic texts itisformed from nearly\nfifty verbs, andhaslessthan seventy occurrences. Itsusual value isthat\nofasecond person singular; but itoccurs as1stsing, once,inAV. (dvyu-\nsdmjdgrtdd ahdm,'letmewatch tillday-break');as3dsing., toward adozen\ntimes(e.g.punarmd\"vi?atdd rayih, TS.,'letwealth come againtome';\naydrh tydsya rdjdmurdhdnarh mpdtayatdt, $B.,'the king here shallmake\n13*\n"
+    },
+    {
+        "page": 216,
+        "content": "196VIII. CONJUGATION.[570\nhishead fly off'); and as2dpi.several times inTS.(e.g.apah...devesu\nnah sukrto brutdt, 'yewaters, announce ustothegods aswell-doers'), and\nmany times inaBrahmana passage (repeated inK.xvi.21.TB. iii.6.1,and\nAB. ii.6,7)inwhich, moreover, twoauthorities (K.andAB.) have once\nvarayadhvdtforvdrayatdt: noother occurrence ofdhvat hasbeen noted.\n571.Asregardsitsmeaning,thisform hasbeenshown\n(Delbriick)tohaveprevailinglyintheBrahmanas, andtraceably\nbutmuch lessdistinctlyintheVedic texts, aspecific tense-value\nadded toitsmode-value assignifying, namely, aninjunction\ntobecarried out atalater time than thepresent:itis(like\ntheLatin forms intoandtote]aposteriororfutureimperative.\nExamples are :ihaiJvdmatfsthantam abhyehiJ\ntibruhitdm Itinadgatdm\npratiprdbrutat (QB.), 'say toher\"come tomeasIstandjust here\", and\n[afterward]announce her tousashaving come'; yddurdhvds tfsthd drdvine\n'hddhattdt(RV.), 'when thou shalt standupright, [then] bestow riches here'\n(and similarly inmany cases); utkulam udvah.6 bhavo'dtihya prdti dhdvatdt\n(AV.),'beacarrier uptheascent; afterhaving carriedup,runbackagain';\nvdnaspdtir ddhi tvdsthdsyati tdsyavittdt(TS.),'the tree willascendthee,\ntake[then]note ofit'.\nAccordingtothegrammarians,theform intatmaybeusedwhen bene-\ndiction isintended.\nUses oftheModes.\n572.Ofthethree modes, theimperativeistheone\nmost distinct andlimited inoffice, andmostunchangedin\nusethroughout thewholehistoryofthelanguage.Itsigni-\nfiesacommand orinjunction anattemptattheexercise\nofthespeaker's willupon some one orsomething outside\nofhimself.\nThis, however(inSanskrit asinotherlanguages),isbynomeansalways ofthesame force;thecommand shades off\ninto ademand, anexhortation, anentreaty, anexpression of\nearnest desire. Theimperative alsosometimessignifies anas-\nsumption orconcession\n;andoccasionally, bypregnant construc-\ntion,itbecomes theexpression ofsomething conditional or\ncontingent; but itdoes notacquire anyregular useindepend-\nent-clause-making.\n573.Theoptative appearstohave asitsprimaryoffice\ntheexpression ofwish ordesire; inthe oldestlanguage,\nitsprevailing useinindependent clauses isthat towhich\nthename\"optative\" properly belongs.\n"
+    },
+    {
+        "page": 217,
+        "content": "575 USES OFTHEMODES. 197\nThe so-calledprecative forms(567,arerestricted tothis\nuse,butarenototherwisedistinguished from thesimple optatives.\n'Buttheexpressionofdesire, ontheonehand, passes nat-\nurally over into that ofrequest orentreaty,sothat theoptative\nbecomes asoftenedimperative ;and,ontheother hand,itcomes\ntosignify what isgenerally desirable orproper, what should or\noughttobe,and sobecomes themode ofprescription;or,yet\nagain,itisweakened intosignifying whatmayorcanbe,what\nislikelyorusual, and sobecomes atlastasoftened statement\nofwhat is.\nFurther, theoptativeindependent clauses, with relative\npronouns andconjunctions, becomes aregular means ofex-\npressionoftheconditional andcontingent,inawide andin-\ncreasing varietyofuses.\n574. Thesubjunctive,ashasbeenpointed out,be-\ncomesnearlyextinct atanearly periodinthehistoryofthe\nlanguage;there are leftofitinclassicalusage onlytwo\nrelics;theuseofitsfirstpersonsinan\"imperative\" sense,\nortosignifyanecessityorobligation restingonthespeak-\ner,oraperemptoryintention onhispart;andtheuseof\nitsotherpersons,with thenegative particlemmd,ina\nprohibitiveornegative imperativesense.\nAnd thegeneralvalue ofthesubjunctive from thebegin-\nningwaswhat these relics would seem toindicate :itsfunda-\nmental meaningisperhapsthat ofrequisition,lessperemptory\nthan theimperative, more sothan theoptative. Butthismeaning\nisliable tothesame modifications and transitions with that of\ntheoptative;andsubjunctiveandoptativerunclosely parallel\nwith oneanother inthe oldest languageintheir use inin-\ndependent clauses, and arehardly distinguishableindependent.\nAnd instead oftheir being (asinGreek)both maintained in\nuse, andendowed with nicer andmore distinctive values, the\nsubjunctive gradually disappears, andtheoptative assumes alone\nthe offices formerlyshared byboth.\n575.The difference, then, betweenimperativeandsub-\njunctive andoptative,intheir fundamental andmost char-\nacteristic uses,isone ofdegree:command, requisition,\nwish :andnosharplineofdivision exists between them;\ntheyaremore orlessexchangeablewithoneanother, and\ncombinable incoordinate clauses.\n"
+    },
+    {
+        "page": 218,
+        "content": "VIII. CONJUGATION.[575\nThus,inAV.,wehave inimpv.:catdmjlva carddah, 'do\nthou live ahundred autumns'; ubhau taujivatam jarddasti,'let\nthem both live toattain oldage';--insubj., adyd jivani,'let\nmelive this day';catdm jivati carddah, 'heshall live ahundred\nautumns';---inopt., jlvemacardddm catani, 'maywelivehund-\nreds ofautumns'; sdrvam hyur jivyasam (prec.j,'Iwould fain\nlive outmywhole term oflife'. Here themodes would be\ninterchangeablewith ahardly perceptible changeofmeaning.\nExamples, again,ofdifferent modes incoordinate construc-\ntion are :iydm agnenaripdtimvidesta .. .suvana putran mdhisl\nbhavatigatva pdtim subhdga mrajatu (AV.), 'maythiswoman,\nOAgni!findaspouse;givingbirth tosons sheshallbecome\nachieftainesg;havingattained aspouseletherrule inhappiness' ;\ngopaydnahsvastdye prabudhenahpunar dadah(TS.),'watch over\nusforour welfare, grant unto ustowakeagain'; syhn nah\nsunuh ...satesumatir bhutv asme(RV.), 'maythere betous\nason;letthat favor ofthine beours' .Itisnotveryseldom\nthecase thatversions ofthesamepassageindifferent textsshow\ndifferent modes asvarious readings.\nThereis,infact, nothinginthe earliest employmentof\nthese modes toprovethattheymightnot allbespecialized uses\nofformsoriginally equivalent having,forinstance, ageneral\nfuture meaning.\n576.Asexamplesofthe less characteristic use ofsub-\njunctive andoptativeinthe older language,inindependent\nclauses, may bequotedthefollowing:aghatagachanuttara\nyughni (RV.),'those later ageswill doubtless come'; ydd...nd\nmara itimdnyase (RV.),'ifthou thinkest \"Ishall notdie'\"; nd\ntanacanti nddabhati tdskarah(RV.), 'they donotbecome lost;\nnothief canharm them'; kdsmai devhyahavisa vidhema(RV.),\n'towhat godshallweoffer oblation?'agniria rayimacnavat .. .\ndive-dive(RV.), 'byAgni onemay gain wealthevery day';utai\n'nam brahmdnedadyattdtha syonacivasyat (AV.),'one should\ngive her, however, toaBrahman;inthat case she willbe\npropitious andfavorable'; dhar-ahar dadyat (QB.),'one should\ngive every day'.\n577. Theuses oftheoptativeinthelater languageareof\ntheutmostvariety, coveringthewhole fieldoccupied jointly by\nthetwomodes inearlier time.Afewexamples from asingle\ntext(MBh.)willbeenoughtoillustrate them :uchistam nai'va\nbhunjlyamnakuryam padadhavanam,'Iwillnoteatoftherem-\nnant ofthesacrifice, Iwillnotperform thefoot-lavation'\n;jnatln\nvrajet,'lethergotoherrelations'; nai'vam sakarhicitkuryat,\n'sheshould not actthus atany time'; katham vidyam nalam\nnrpam, 'how canIknow king Nala?'utsarge samcayah syattu\n"
+    },
+    {
+        "page": 219,
+        "content": "581]USES OFTHEMODES. 199\nvindeta'pisukharii kvacit, 'but incase ofherabandonment there\nmay beachance;shemayalso findhappiness somewhere';\nhitham vasowkarteyamnacabudhyeta mepriya,'how can Icut\noffthegarment andmybeloved notwake?'\n578. The later useofthe firstpersons subjunctiveasso-,\ncalled imperativeinvolves nochange ofconstruction from former\ntime, butonlyrestriction toasingle kind ofuse :thus, divyava,\nkletustwoplay';kirn karavcinite,'what shall Idoforthee?'\n579. Thesubjunctive withmi isintheoldest language\nalmost the sole form ofprohibitive expression, and isvery\ncommon. Thekind ofsubjunctive employedisthatwhich cor-\nrespondstotheaugmentless forms ofapasttense\n;and inthe\ngreat majorityofcases(fivesixths inRV., nine tenths inAV.)\nitistheaugmentlessaorist that ischosen. Thus :prdpatame\n'hdraiisthah(AV.), 'flyaway;donotstay here'; dvisdnc camd-\nhyam rddhyatu maca'Mm dvisate radham(AV.),'both letmy\nfoebesubjecttome,and letmenotbesubjecttomy foe';\nurvdcyam dbhayam jyotirindramanodirghaabhinacan tamisrah\n(RV.),'Iwould winbroad fearlesslight,OIndra;letnotthe\nlong darknesses come uponus';manaayuh prd mosih(RV.),\n'donotstealawayour life';malibher(impf.)namarisyasi (RV.),\n'donotfear; thou wiltnotdie';masmai 'tintsdkhm kuruthah\n(AV.),'donotmake friends ofthem'.\nOnly oneoptative (bhujema)isused prohibitively withmdinRV., and\nonly once(inaprobably corrupted passage)animperative; neither con-\nstruction isfound inAV.; andthecases inthelaterlanguagearerare.\n580. Thisverydefinite andpeculiar construction, ofan\naugmentless pasttense withma,haspreserveditself inuse,and\nisoccasionally metwith inthelater language: thus, samaqvasihi\nmacucah, 'becomforted, donotgrieve'; mabhaih, 'donot fear'\n(both MBh.).\nButtheuseoftheoptativewithnd, 'not', inaprohibitive\nsenseappears even(very rarely)intheVeda, andbecomes later\ntheprevalentconstruction;thus, ndrisyema kada cand(RV.),\n'maywesuffer noharm atany time'; ndca'tisrjenndjuhuyat\n(AV.),'and ifhedonotgrant permission,lethimnotsacrifice';\ntadutdtha ndkuryat (QB.),'buthemust notdothat so';na\ndivacayita (QGS.),;lethim notsleep byday'; natvam vidyur\njandh (MBh.),'letnotpeople know thee'. This inthe later\nlanguageisthecorrelative oftheprescriptive optative,andboth\nareextremely common\n;sothat inatext ofprescriptivechar-\nacter theoptative forms maycome tooutnumber theindicative\nandimperative together (asisthecase, forexample,inManu).\n581. Inalldependent constructions,itisstillharder even\n"
+    },
+    {
+        "page": 220,
+        "content": "200 VLLL CONJUGATION.[581\ninthe oldest languagetoestablish adistinction between sub-\njunctiveand optative:amethod ofuse ofeither isscarcely\ntobefound towhich theother does notfurnish apractical\nequivalent--and then, inthe later language,such uses are\nrepresented bytheoptativealone. Afewexampleswill be\nsufficient toillustrate this :\na.After relative pronouns andconjunctionsingeneral:\nyavyuchur yaccanundm vyuchhn (RV.),'which haveshone forth\n[hitherto],andwhich shall hereafter shine forth'; yoltojayata\nasmakam sdeko 'sat(TS.),'whoever shall beborn ofher, let\nhimbeoneofus';yovaithnvidyat pratydksamsdbrahma vedita\nsyat (AV.),'whoever shallknow them face toface, hemay pass\nforaknowing priest'; putranam...jatUnam jandyaccayhn(AV.),\n'ofsons born andwhom thou mayest bear'; ydsya...dtithir\ngrhhn agdchet (AV.),'towhosesoever house hemaycome as\nguest'; yatamdtha kamdyetatdthakuryat (QB.),'inwhatever way\nhemay choose, somayhedoit';ydrhihotaydjamanasya nhma\ngrhmyhttdrhibruyat (TS.j,'when thesacrificing priestshallname\nthename ofthe offerer, then hemay speak'; svarupam yada\ndrastum ichethah(MBh.),'when thou shalt desire toseethine\nown form'.\nb.Inmoredistinctlyconditional constructions :ydjama\ndev&nyddi cakndvama(RV.),'we will offer tothegodsifwe\nshall beable'; yddagnesyhmahdm tvdm tvdm vaghasyaahdm\nsyustesatyaiha\"cisah(RV.),'ifIwere thou, Agni,orifthou\nwertI,thywishes should berealized onthespot'; yodyam\natisdrpat pardsfan ndsdmucyatai vdriinasya rajnah (AV.), 'though\none steal faraway beyondthesky, heshall notescape king\nVaruna'; ydddnacvanupavdset ksodhukah syadyddacniyadrudrb\n'sya pacttn abhimanyeta(TS.),'ifheshould continue without\neating, hewould starve :ifheshould eat,Rudra would attack\nhis cattle'; prarthayed yadimam kaccid dandyahsamepuman\nbhavet(MBh.),'ifanyman soever should desire me,heshould\nsufferpunishment'. These and thelike constructions, with the\noptative,areverycommon intheBrahmanas and later.\nc.Infinal clauses :ydtha'ham catruho 'sani(AV.),'that\nImaybeaslayer ofmyenemies'; grnanti ydtha pibatho dndhah\n(RV.),'thatbeing praised with song yemaydrink thedraught';\nurauydthutdvacdrman mddema(RV.),'inorder thatwerejoice\ninthywideprotection'; iipajamta ydthe 'yampimar agdcJiet (QB.j,\n'contrive that shecome backagain'; krpam buryadyatha mayi\n(MBh.),'sothathemay takepityonme'. This isintheVeda\noneofthemostfrequent uses ofthesubjunctive;andinits\ncorrelativenegative form, withned, 'inorder that not' or'lest1\n,\nitcontinues notrare intheBrahmanas.\n"
+    },
+    {
+        "page": 221,
+        "content": "586] PARTICIPLES; AUGMENT. 201\nThe indicative isalsoused infinal clauses after yatha: thus, yatha\n'yarn nafyati tathd vidheyam (H.),'itmust besomanaged thatheperish'\n(and thus usually inH.).\nWith theconditional use ofsubjunctive andoptativeisfurther tobe\ncompared that oftheso-called \"conditional\" tense :seebelow, chap. XII.\n582.Nodistinction ofmeaning hasbeen established be-\ntween themodes ofthepresent-system andthose(intheolder\nlanguage)oftheperfect andaorist-sy stems.\nParticiples.\n583.Participles,active andmiddle, aremade from all\nthetense-stemsexcepttheperiphrastic future, and, in\nthelaterlanguage,theaorist(andaoristparticiplesarerare\nfrom thebeginning).\n584.Thegeneral participial endingsare ST^Tant(weak\nform 5fcTat;fern. Trftantl orSTrft atl: seeabove, 449)for\ntheactive, and 5TFTana(fern.t\\\\*\\\\ana)forthemiddle. But:\na.After atense-stem endingina,the activeparticipial\nsuffix isvirtually nt,oneofthetwo a'sbeinglost inthecom-\nbination ofstem-final and suffix.\nb.After atense-stem endingina,themiddleparticipial\nsuffix ismana instead ofana.\nc.Theperfecthas intheactive thepeculiarsuffix vans\n(weakest form us,middle form vat;fern, usi:see, forthein-\nflection ofthisparticiple, above, 458ff.).\nFor details, astoform ofstem etc., and forspecialex-\nceptions,seethefollowing chapters.\nAugment.\n585.Theaugmentisashorta,prefixedtoatense-stem\n-and,ifthelatter beginwith avowel, combiningwith\nthatvowelirregularlyintotheheavier orvrddhi diphthong\n(136 a).Itisalways (without anyexception)theaccented\nelement intheverbal form ofwhich itmakes apart.\nIntheVeda, theaugmentisinafewforms long a:thus, emap, auar,\navrni, avrnak, avidhyat, ayunak, ayukta, arinak, araik.\n586. Theaugmentisasignofpasttime. And anaug-\nment-preteritismade from each ofthetense-stems from which\n"
+    },
+    {
+        "page": 222,
+        "content": "202VIII. CONJUGATION.[586\nthesystemofconjugationisderived: namely,theimperfect,\nfrom thepresent-stem;thepluperfect (intheVedaonly),from\ntheperfect-stem;theconditional, from thefuture-stem;while\nintheaorist such apreteritstands without anycorresponding\npresentindicative.\n587. Intheolder language (mainlyintheVeda; theusage\nisarareoneintheBrahmana)theaugmentisoftenlost,and\ntheaugmentless forms have thesame value asifthey were\ncomplete ;or,rather more often, theyareused assubjunctives\n(above, 563).\nTheaccentuation oftheaugmentless forms isthroughout accordant with\nthat ofthecorresponding unaugmented tense that istosay,where such\natense exists (whichisnotthecasewith thevarieties ofsibilantaorist).\nReduplication.\n588.The derivation ofconjugational and declensional\nstems from roots byreduplication,either alone oralong\nwith other formative elements, hasbeenalready spokenof\n(259), and theformations inwhichreduplication appears\nhave beenspecified: they are, inprimary verb-inflection,\nthepresent (ofacertain class ofverbs),theperfect (of\nnearly all),andtheaorist(ofalarge number); andthein-\ntensive and desiderativesecondary conjugationscontain in\ntheir stems thesame element.\n589.Thegeneral principleofreduplicationisthepre-\nfixion toaroot ofapartofitselfrepeatedifitbegin\nwithconsonants, the initial consonant andthevowel;ifit\nbegin with avowel, thatvowel, either alone orwith afollow-\ningconsonant. The varieties ofdetail, however,arevery\nconsiderable.Thus, especiallyasregardsthevowel, which\ninpresent andperfect anddesiderative isregularlyshorter\nandlighterinthereduplication than intheroot-syllable,\ninaorist islonger, and inintensive isstrengthened. The\ndifferences asregards aninitial consonant areless, and\nchieflyconfined totheintensive; forthe others, certain\ngeneral rulesmaybeherestated,allfurther details being\n"
+    },
+    {
+        "page": 223,
+        "content": "592]REDUPLICATION. 203\nlefttobegiveninconnection with theaccount ofthesep-\narate formations.\n590.Theconsonant ofthereduplicating syllableisin\ngeneralthe firstconsonant oftheroot :thus,W^\"paprach\nfromysC prach;wS\\W\\J*$li8frmVfiPT^w; ^^bubudh\nfrom ysiTJ. But:'o~x\na.Anon-aspirateissubstituted inreduplicationforan\naspirate: thus, ^TJTdadha from|/ETT; f^RfoMrfromy>T6/*r.\nb.Apalatalissubstituted foragutturalorfor^h:\nthus, rJSfica^rfromy^\\kr;nrfl^cikhid from\nSQWjagrc&hfrom]/ERgrabh ;^\"^jahr from\nThe occasionalreversion, ontheother hand,ofapalatal intheradical\nsyllable toguttural form hasbeen noticed above(216.9).\nc.Oftwo initialconsonants, thesecond,ifitbea\nnon-nasal mutepreceded byasibilant,isrepeatedinstead\nofthefirst: thus, cTFSCT tastha fromyF2JTstha; rj^r^caskand\nformi/FfrJskand ;3Fppasprdh fromyFT^-T sprdh:--but\nTFFsasmr fromyF7.\nc. c\nAccent oftheVerb.\n591. The statements which have beenmade above, and\nthose which willbemade below, astotheaccent ofverbal\nforms, applytothose cases inwhich theverb isactually accented.\nBut, accordingtothegrammarians, andaccordingtothe\ninvariablepracticeinaccentuated texts, theverb isinthegreat\nmajorityofitsoccurrences unaccented ortoneless.\nThat istosay, ofcourse, theverb initsproper forms,itspersonalor\nso-called finite forms. The verbal nouns and adjectives, ortheinfinitives\nandparticiples,aresubjecttopreciselythesame laws ofaccent asother nouns\nandadjectives.\n592. Thegeneral rule, covering most ofthecases,isthis :\nTheverb inanindependentclause isunaccented, unless itstand\natthebeginningoftheclause oralso, inmetrical text, at\nthebeginningofapada.\nFortheaccent oftheverb, aswell asforthat ofthevocative case\n(above, 314), thebeginningofapada counts asthat ofasentence, what-\neverbethelogical connection ofthepada withwhat precedesit.\nExamplesoftheunaccented verb are :agnimidepurohitam,'Ipraise\n"
+    },
+    {
+        "page": 224,
+        "content": "204VIII. CONJUGATION.[592\nAgni, thehouse-priest';sdiddevesu gachati, 'that, truly, goes tothegod?';\nague supdyano bhava,'0Agni, beeasy ofaccess'; iddm indra prnuhi somapa,\n'hear this, Indra, sorna-drinker'; ndmas terudra krrtmas, 'homagetothee,\nRudra, weoffer'; ydjamdnasya pa?un pahi, 'protect thecattle ofthe sacrificer'.\nHence, there aretwoprincipalsituations inwhich theverb\nretains itsaccent :\n593. First, theverb isaccented when itstands atthe\nbeginningofaclause or,inverse, ofapada.\nExamples oftheverb accented atthehead ofthesentence are, inprose,\ncundhadhvam ddfvydya kdrmane, 'bepure forthedivineceremony'; apnotl\nJmdm lokdm, 'hewins this world'; inverse, where thehead ofthesentence\nisalsothat ofthepada, syame'dindrasya cdrmani, 'maywebeinIndra's\nprotection'; darcdya maydtudhandn, 'showmethesorcerers'; gdmad vdjebhir\nasdnahj 'may hecome withgood things tous'; inverse, where thehead\noftheclause iswithin thepada, tesam pahi crudhi hdvam, 'drink ofthem,\nhear ourcall'; sdstumdtd sdstu pitd sdstu $vdsdstuvifpdtih,'letthemother\nsleep,letthefathersleep,letthedog sleep,letthemastersleep'; vfyvakar-\nmanndmas tepdhy asmdn, 'Vicvakarman, homage tothee; protectus!'\nyuvam. ..rdjna uceduhitd prchevdmnara,'theking's daughter said toyou\n\"Ipray you, yemen'\"; vaydmtevdya indra viddhi sunahprdbharamahe,\n'we offerthee, Indra, strengthening; take note ofus1\n.\nExamplesoftheverb accented atthehead ofthepadawhen this isnot\nthehead ofthesentence are: dthd tedntamdndm vidydma sumatindm,'so\nmayweenjoy thymost intimatefavors'; dhdtd 'syd agrtivdi pdtim dddhdtu\npratikamyam, 'Dhatar bestow uponthisgirlahusband accordingtoherwish';\nydtudhdnasya somapa jahtprajdm, 'slay, Soma-drinker, theprogeny ofthe\nsorcerer'.\n594. Certainspecial cases under thishead areasfollows:\na.Asavocative forms nosyntactical partofthesentence towhich it\nisattached, but isonlyanexternal appendagetoit,averb following an\ninitialvocative,ormore than one, isaccented,asifitwere itself initial in\ntheclause orpada: thus, d?rutkarna fractal hdvam, *0thou oflistening ears,\nhear our call!' sitevdndamahetva,1QSita,wereverencethee'; vfyve deva\nvdsavo rdksateJmdm,'allyegods, yeVasus, protectthis man'; utd\"gay\ncakrtisam deva deva jivdyathd punah, 'likewise him, gods, whohascom-\nmittedcrime, yegods, yemake toliveagain'.\nb.Ifmore than oneverb follow aword orwords syntactically connected\nwiththemall,only the first loses itsaccent, theothers being treated asif\ntheywere initial verbs inseparate clauses, with thesame adjuncts under-\nstood:thus, tardnirfjjayatiksetiptisyati, 'successful heconquers, rules,\nthrives'; amitrdn ...pdrdca indra prdmrnd jahi ca,'our foes, Indra, drive\nfaraway andslay'; asmdbhyam jesi yotsi ca,'forusconquer andfight';\ndgnisomd havisahprdsthitasyavitlrhhdryatarh vrsana jusetham, '0Agni and\nSoma, oftheoblation setforthpartake, enjoy, yemighty ones, take plea-\nsure'.\n"
+    },
+    {
+        "page": 225,
+        "content": "595]ACCENT OFTHEVERB. 205\nc.Inlikemanner (butmuch lessoften^, anadjunct, assubject or\nobject, standing between twoverbs andlogically belonging toboth, isreckoned\ntothe first alone, and thesecond hasthe initial accent: thus, jaht prajdrh\nndyasva ca, 'slay theprogeny, andbring [it]hither'; crnotu nahsubhdgd\nbodhatu tmdnd, 'may theblessed onehearus,[andmay she]kindly regard [u>]1\n.\nd.Astocases inwhich asingle verb standing between twoadjuncts\nhastheinitial accentperhapsasbeing inthedivision ofthesentence reckoned\ntothesecond rather than thefirst,seebelow, 597.\n595. Second, theverb isaccented, whatever itsposition,\ninadependentclause.\na.Thedependencyofaclause isinthevery great majority ofcases\nconditioned bytherelative pronoun t/a,oroneofitsderivatives orcompounds.\nThus: yarnyajndm paribhur dsi, 'whatoffering thouprotectest'; 6teyanti\nyeaparT.su pacyan, 'they arecoming who shall behold herhereafter';solid\nydnmedstiUna, 'along with thatwhich ismine'; ydtra nahpurve pitdrah\npareytih,'whither ourfathers ofolddeparted'; adya muriya yddiydtudhano\ndsmi,'letmedieonthespot,ifIamasorcerer'; ydthdJ\nhdnyanupurvdm\nbhdvanti,'asdays follow oneanother inorder'; ydvad iddm bhtivanarh vfcvam\ndsti, 'how great thiswhole creationis';ydtkdmdstejuhumds tannoastu,\n'what desiring wesacrifice tothee,letthatbecomeours'; yatamds tftrpsdt,\n'whichever onedesires toenjoy'.\nThepresenceofarelative word inthesentence does not, ofcourse,\naccent theverb,unless this isreally thepredicateofadependent clause :thus,\ndpa tyetdydvo yathd yanti, 'theymake offlikethieves(asthievesdo}'; ydt\nsthdjdgac carejate, 'whatever[is]movable andimmovable trembles'; yathd-\nkdmarh nipadyate,;heliesdown athispleasure'.\nb.The particlecawhen itmeans 'if,andced(ca+id), 'it',givean\naccent totheverb :thus, brahmd ceddhdstam dgrahlt,'ifaBrahman has\ngrasped herhand';tvdrh casoma novdcojwdturh ndmardmahe,'ifthou,\nSoma,wiliest ustolive,weshall notdie';dcagdchdn mitrdm endda-\ndhdma,'ifhewillcome here,wewillmake friends with him'.\nc.There areaveryfewpassagesinwhich thelogical dependenceofa\nclause containing nosubordinating word appearstogivetheverb its\naccent :thus,sdmdcvaparndf cdranti nondro'smdkam indra rathino jayantu,\n'when ourmen, horse-winged, come into conflict,letthechariot-fighters of\nour side, Indra, winthe victory'. Rarely, too,animperativesofollowing\nanother imperative that itsaction mayseem aconsequenceofthe latter's is\naccented; thus, tuyam dgaliikdnvesu susdcd ptba, 'come hither quickly;\ndrink along with theKanvas'(i.e.inorder todrink).\nd.Afewother particles givetheverb anaccent, invirtue ofaslight\nsubordinating force belongingtothem :thus, especiallyhi(withitsnegation\nnahf), which initsfullest value means'for', butshades offfrom that into\namere asseverative sense;theverb orverbs connected with itarealways\naccented: thus,mtemuncantdm vimtico hisdnti,'letthem release him,for\nthey arereleasers'; ydcciddhf. ..andcastd ivasmdsi,'ifwe, forsooth,are\n"
+    },
+    {
+        "page": 226,
+        "content": "206VIII. CONJUGATION.[595\nas itwere unrenowned';alsoned (na-\\-id), meaning 'lest, that not':\nthus,nettva tdpati sUro artfsa,'that thesunmay notburn thee with his\nbeam'; virdjam nedvichindddniJ\nti,'saying tohimself, \"lest Icut offthe\nviraf- and theinterrogative kuvfd, 'whether?' thus, ukthebhih kuvid\nagdmat,'willhecome hither forourpraises?'\n596. But further, theverb ofapriorclause isnotinfre-\nquentlyaccented inantithetical construction.\nSometimes, therelation ofthetwo clauses isreadily capable ofbeing\nregarded asthat ofprotasis andapodosis ;butoften, also, such arelation is\nvery indistinct; andthecases ofantithesis shade offinto those ofordinary\ncoordination, thelinebetween them appearingtoberather arbitrarily drawn.\nInthemajority ofcases, theantithesis ismade distincter bythepres-\nence inthetwoclauses ofcorrelative words, especially anya am/a, eka\nefca,vava,caea:thus, prd-praJ\nnyeydnti pdry anyd asate, 'some go\nonandon,others sitabout'(asifitwere 'while somego' etc.);tidva\nsincadhvam upavaprnadhvam,'either pour out,orfillup'; sdrh ce'dhydsva\n'gneprdcavardhaye 'mam,'both dothou thyself becomekindled, Agni,\nanddothou increase this person'. But itisalsomade without suchhelp:\nthus, pro, 'jatah prajd jandyati part prdjata grhnati,'theunbornprogeny he\ngenerates, thebornheembraces'; dpayusmdd dkramm ndJsmdnupdvartate,\n'[though] shehasgoneaway from you, shedoes notcome tous';nd'ndhb\n'dhvarytir bhdvati ndyajndrh rdksansi ghnanti,'the priest does notbecome\nblind, thedemons donotdestroy the sacrifice'.\n597. \"Where theverbwould bethesame inthetwoantithetical clauses,\nitisnotinfrequently omitted inthesecond:thus,beside complete expres-\nsions likeurvi ca 'sivdsvl ca'si,'both thou artbroad andthou artgood',\noccur, muchoftener, incomplete ones like agnfr amusmih lokd dsidyamb\n'smfn, 'Agni wasinyonder world, Yama[was]inthis'; asthnd 'nydh prajdh\npratitfsthanti mansenaJ\nnydh, 'bybone some creatures stand firm, byflesh\nothers'; dvipac casdrvarh nordksa cdtuspad ydccanahsvdm, 'both protect\neverything ofours that isbiped, and alsowhatever that isquadruped\nbelongs tous'.\nExamples from theBrahmanas likethe first ofthose here given (with\nthesecond verbexpressed), and like thethird(incomposition with apre-\nposition), show that thisexplanation oftheverbal accent ispreferable tothe\noneformerly given namely, that theverb istoberegardedasunderstood\ninthe first clause and initial inthesecond.\n598. Inavery small number ofmore orless doubtful\ncases, theverbappearstobeaccented foremphasis.\nThus, beforecand,'inany wise'; inconnection with theasseverative\nparticles fd,dha, ktla, angd, evd, butsporadically; and soon.The detail\nandexamination ofthecases isnotworth while here*.\n*Thespecialities andirregularities oftheEV. asregards verbal accent arediscussed\nbyA.Mayr inSitzungsb. d.Wiener Akad. for1871; oftheAV.,byW.D.Whitney in\nJ.A.0.8.,vol. v.(andKuhn'sBeitrage, vol.i.) ;oftheTS.,byA.Weber inInd. Stud.,\n-vol. xiii.\n"
+    },
+    {
+        "page": 227,
+        "content": "601] 207\nCHAPTER IX.\nTHEPRESENT-SYSTEM.\n599.THEpresent-system, orsystem offormscoming\nfrom thepresent-stem,iscomposed (aswaspointed out\nabove)ofapresentindicativetense, along with asubjunc-\ntive(mostlylostintheclassicallanguage), anoptative, an\nimperative, and aparticiple, and also apast tense, an\naugment-preterit,towhich wegive (byanalogy with the\nGreek)thename ofimperfect.\nThese forms generally goinSanskrit grammar bythename of\"special\ntenses\", while theother tense-systems arestyled \"general tenses\" asif\ntheformer weremade from aspecial tense-stem ormodifiedroot, while the\nlatter came,allalike, from theroot itself. There isnoreason whysuch a\ndistinction andnomenclature should beretained; since, ontheonehand,\nthe\"special tenses\" come inone setofverbs directly from theroot, and,\nontheother hand,theother tense-systems aremostly made from stems\nand, inthecaseoftheaorist, from stems having avariety ofform comparable\nwith that ofpresent-stems.\n600.Practically,thepresent-systemisthemostpro-\nminent andimportant partofthewholeconjugation,since,\nfrom theearliestperiodofthelanguage,itsforms arevery\nmuch morefrequent than those ofalltheothersystems\ntogether.\nThus,intheVeda, theoccurrences ofpersonal forms ofthissystem are\ntothose ofallothers about asthree toone;intheAitareya Brahmana, as\nfive toone;intheHitopadega,assixtoone;intheQakuntala,aseightto\none;inManu, asthirty toone.\n601. And, asthere isalsogreat varietyinthemanner\ninwhich different roots form theirpresent-stem, this,as\nbeingtheirmostconspicuous difference,ismade thebasis\noftheirprincipalclassification;andaverb issaid tobeof\nthis orofthatconjugation,orclass, accordingtotheway\ninwhich itspresent-stemismade.\n"
+    },
+    {
+        "page": 228,
+        "content": "208 IX -PRESENT-SYSTEM.[602\n602. Inasmall minorityofverbs, thepresent-stemis\nidentical with theroot. Then there arebesides(excluding\nthepassiveand causative) eight more orlessdifferent ways\nofformingapresent-stemfrom theroot, eachwaybeing\nfollowed byalargerorsmaller number ofverbs. These\narethe\"classes\" or\"conjugation-classes\",aslaiddown by\nthenative Hindugrammarians. Theyarearranged bythe\nlatter inacertain whollyartificial andunsystematicorder\n(thegroundofwhich hasnever beenpointed out);andthey\narewont tobedesignatedinEuropean works accordingto\nthisorder,orelse,afterHinduexample, bytherootstand-\ningatthehead ofeach class intheHindu lists.Adiffer-\nentarrangement andnomenclature willbefollowed here,\nnamelyasbelow theclassesbeingdivided(asisusual\ninEuropean grammars)intotwomoregeneralclasses or\nconjugations, distinguishedfromoneanother bywider differ-\nences than those whichseparatethespecialclasses.\n603. The classes oftheFIRST CONJUGATION are as\nfollows :\nI.Theroot-class(second class, orad-class, of\ntheHindugrammarians);itspresent-stemiscoincident\nwith theroot itself: thus,^ad, 'eat';^,'go';EfT\nya, 'go';fef dvis, 'hate';3^duh,'milk'.\nII.Thereduplicating class(thirdorhu-\nclass);theroot isreduplicatedtoform thepresent-stem:\nthus,sp^\"juliu fromy^hu, 'sacrifice'; ^7dada from\nI/^T, 'give'; \\3p\\bibhr fromj/H, 'bear'.\nIII.Thenasal class(seventhorrudk-class) ;a\nnasal, extended tothesyllable ^nainstrong forms,is\ninserted before the final consonant oftheroot: thus,\n~^{jundh(or\"^TTTCJ runadh)fromi/\"^Urudh;TFj(^yunj\n(orW^yunaj]from>/?JsT yuj.\nIV. a.The^w-class(fifthorsw-class);thesyl-\n"
+    },
+    {
+        "page": 229,
+        "content": "606]CONJUGATION-CLASSES. 209\nlable *7nuisadded totheroot: thus,Rsunu from\nyH;3T3apnu from yJETR ap.\nb.Averysmall number'only half-a-dozen) of\nrootsending alreadyin^n,andalsooneverycommon\nandquite irregularlyinflected root not soending (off\nkr,'make'), add3ualone toform thepresent-stem. This\nistheeighthortow-class oftheHindugrammarians ;it\nmaybebestrankedbyusasasub-class, theu-class:\nthus,cHtanufrom i/rR tan.O -V\nV.Thena-class(ninthor^n-class);thesyllable\nJTTna(or,inweak forms,^m\\isadded totheroot :\nthus, stjlmi farina(orjtiluD krinl] fromi/sfft\" &n, 'buy';\nFrPTT stabhna(orFrPft stabhnl) from i/FcR sta&A, 'estab-\nlish'.\n604.These classes have incommon, astheirmost fund-\namentalcharacteristic,ashift ofaccent: thetone being\nnowupontheending, andnowupontheroot ortheclass-\nsign. Alongwith thisgoesavariation inthestemitself,\nwhich hasastrongerorfuller formwhen theaccent rests\nupon it,andaweaker orbriefer formwhen theaccent is\nontheending:these forms aretobedistinguishedasthe\nstrongstemandtheweak stemrespectively (inpart,both\nhavebeengiven above). The classes alsoform theiropta-\ntiveactive, their 2dsing, imperative,and their 3dpi.\nmiddle, inadifferent manner from theothers.\n605. Inthe classes oftheSECOND CONJUGATION, the\npresent-stemends in,andtheaccent hasafixedplace,\nremaining always uponthesamesyllableofthestem, and\nnever shifted totheendings. Also, theoptative,the2d\nsing, impv.,andthe3dpi.middle are(asjust stated)un-\nlikethose oftheotherconjugation.\nV606.The classes ofthisconjugationareasfollows:\nVI.Thea-class,orunaccented a-class (first\nWhitney, Grammar. 14\n"
+    },
+    {
+        "page": 230,
+        "content": "210IX.PRESENT-SYSTEM.[606\norJ/m-class);theaddedclass-signisasimply;andthe\nroot, which hastheaccent, isstrengthened byguna\nthroughout:thus.*Rbhdva from y*\\bhu, 'be';^Jnaya\nfromy^\\m, 'lead'; SJTEIbodha fromy^3j)udh,'wake';\n^vddafromy^ vad^ 'speak'.\nVII.Thea-class, oraccented a-class(sixthor\ntud-cl&ss);theaddedclass-signisa,asinthepreceding\nclass; but ithastheaccent, andtheunaccented root\nremains unstrengthened:thus,c^tuddfrom^ff^.'thrust';\nHsTsrjdfrom/TO srj.'letloose';Rsuvdfrom/H su,c t\"v*\" o c\\\n'givebirth'.\nVIII. The/#-class (fourthorcfoiJ-class);yaisadded\ntotheroot,which hastheaccent :thus,^oT d^vya from\nY^(div(more properly ^fadiv: see765); ^3fndhya\nfromy7^nah, 'bind'; ^F^TT krudhya fromy&m krudh,\n'beangry'.\nIX.Thepassive conjugationisalsoproperlya\npresent-system only, havingaclass-sign which isnot\nextended intotheother systems;thoughitdiffers mark-\nedlyfrom theremainingclasses inhavingaspecific\nmeaning,and inbeingformable inthemiddle voice\n(only) from alltransitive verbs. Itsinflection may\ntherefore best betreated next tothat ofthe?/a-class,\nwithwhich itismostnearlyconnected,differing from\nitasthea-class from thea-class. Itforms itsstem,\nnamely, byadding anaccented ydtotheroot :thus,\n^7Jadydfromy*$Z[ad;\"^EETrudhydfromi/^Trudh ;\nbudhyd from y'SFlbudh; H\"2^tudyd fromy\"^tud.\n607. TheHindu grammarians reckon atenth class orcur-\nclass, havingaclass-sign ayaadded toastrengthenedroot(thus,\ncor&ya fromycur), andaninflection like that oftheother a-\nstems. Since, however, thisstem isnotlimited tothepresent-\nsystem, butextends also into the restoftheconjugation\nwhile italsohas toagreat extent acausative value, andmay\n"
+    },
+    {
+        "page": 231,
+        "content": "611]CONJUGATION-CLASSES.\nbeformed inthatvalue from alargenumber ofroots itwill\nbebesttreated along with thederivativeconjugations (chap. XIV.).\n608.Asmall number ofroots addinthepresent-system\nach,orsubstitute achfortheir final consonant, andform a\nstem endinginchaorchd,which isthen inflected likeana-\nstem. This ishistorically, doubtless, atrueclass-sign, analogous\nwith the rest;but theverbs showingitaresofew, and in\nformationpartlysoirregular,thattheyarenotwell tobeput\ntogetherinto aclass, butmaybestbetreated asspecialcases\nfalling under theother classes.\nRoots adding charerandyu,which make thestems rchdandyucha.\nRoots substituting chfortheir final areis,us(orvas'shine'), gam,\nyam, which make thestemsichd, uchd, gdcha, ydcha.\nOfso-called roots ending inch,several aremore orless clearly stems,\nwhose usehasbeen extended from thepresent toother systems oftenses.\n609. Roots arenotwholly limited, even inthelater language,toone\nmode offormation oftheir present-stem, butaresometimes reckoned asbe-\nlonging totwo ormore different conjugation-classes. And such variety of\nformation isespecially frequent intheVeda, being exhibited byaconsider-\nableproportionoftheroots thereoccurring; already intheBrahmanas, how-\never,acondition isreached nearly agreeing inthisrespect with theclassical\nlanguage. The different present-formations sometimes have differences of\nmeaning ;yetnotmore important ones than areoften found belongingto\nthesame formation, norofakind toshow adifference ofvalue asoriginally\nbelongingtotheseparateclasses ofpresents.Ifanything ofthiskind isto\nbeestablished,itmust befrom thederivativeconjugations, which aresepar-\natedbynofixed linefrom thepresent-systems.\n610.Wetakeupnow the different classes, intheorder\ninwhichtheyhave been arranged above, todescribe more in\ndetail, andwith illustration, theformation oftheir present-\nsystems, and tonotice theirregularities belonging under each\nclass.\nI.Root-class (second, ad-c\\ass).\n\\611. Inthis class there isnoclass-sign;theroot itself\nisalsopresent-stem,and toitareaddeddirectlytheper-\nsonal endings--butcombined insubjunctiveandoptative\nwith therespective mode-signs,and intheimperfect taking\ntheaugment prefixedtotheroot.\nTheaccented endings (552) regularlytake theaccent except\nintheimperfect, where itfallsontheaugment andbefore\n14*\n"
+    },
+    {
+        "page": 232,
+        "content": "212 IX.PRESENT-SYSTEM.[611\nthem therootremains unchanged ;before theunaccentedendings,\ntheroot takes theguna- strengthening.\nItisonly inthe first three classes that theendings come immediately\nincontact with afinalconsonant oftheroot, andthat therules forconsonant\ncombination have tobenoted andapplied.\n1.Present Indicative.\n\"612. Theendingsaretheprimary (with f?Tatein3d\npi.mid.),added tothehare root.Theroottakes theaccent,\nandhasguna,ifcapableofit,inthethree persons sing.act.\nExamplesofinflection :a.root^e,'go': strong\nform ofroot-stem, ^e;weak form,^i.\nactive. f* middle.* P\ns. d. p.s.\nemi ivds imdsiyeivdhe intake\nithds ithd iseiyathe idhve\neti Ltds ydntiiteiyhte iydte\nb.root fir^T dvis, 'hate'; strong stem-form, \"^dves ;\nweak, fer dvis.\nForrules ofcombination forthe finals,see226.\ni&RH I^^IH I^Q-HH T^\"|&fe|t(. f^a-H1^\ndvesmi dvisvds dvismds dvise dvisvdhe dvismdhe\ndveksi dvisthds dvisthd dvikse dvisathe dviddhve\nlidvistds dvisdnti dviste dvisate dvisdte\nc.root~^^duht'milk':strongstem-form ^^doh; weak,\nForrules ofcombination forthe finalft,and fortheconversion ofthe\ninitial todft,see222, 155, 160.\ndohmi duhvds duhmds duhe duhvdhe duhmdhe\n*Used inthemiddle with thepreposition adhi, tosignify 'goover forone's self,\ni.e.'repeat, learn, read'.\n"
+    },
+    {
+        "page": 233,
+        "content": "616]I.ROOT-CLASS (SECOND, #G?-CLASS). 213\n2MUSIIT^H^I^M\" uif ^c^isj qrij\ndugdhds dugdhd dhukse duhhthedhugdhve\n^JMH^ .S^lTi 3\"^ ^c^(rt Js^H\ndogdhi dugdhds duhdntidugdhe duhhte duhdte\n613. Examples ofthe3dsing. mid. coincident informwith the1stsing,\narenotrare inthe older language (both V.andBr.): themost frequent\nexamples areice,duhe, vide, cdye: moresporadic arecite, bruve, huve.\nTheirregular accent ofthe3dpi.mid. isfound inRV. inrihate, duhate,\nExamples ofthesame personinreand rate alsooccur: thus(besides those\nmentioned below, 629 30,635), vidre, and, withauxiliary vowel, arhire\n(unless these aretoberanked, rather,asperfect forms withoutreduplica-\ntion:790 b).\n2.Present Subjunctive.\n614. Subjunctive forms ofthis class arenotuncommon in\ntheolder language, andnearlyallthose which theformation\nanywhere admits arequotable, fromVeda orfrom Brahmana.Acomplete paradigm, accordingly,isgiven below, with thefew\nforms notactually quotableforthis class enclosed inbrackets.\nWemay take asmodels, fortheactive therooti,'go',and for\nthemiddle therootas, 'sit', ofboth ofwhich numerous forms\naremetwith(althoughneither forthese nor foranyothers can\nthewhole series befound inactualuse).\nThemode-stems aredya (e\n"
+    },
+    {
+        "page": 234,
+        "content": "214 IX.PRESENT-SYSTEM.[616\ngiveninfullabove(566). Thestem-form istheunaccented\nandunstrengthenedroot.Thewhole formation issoregular\nthat asingle exampleofinflection willbeenough.\nactive. middle.\na. d. p.s. d.p.\n^\ndvisyam dvisy&va dvisy&ma dvisiyddvisivdhi dvisimdhi\n2\ndvisy&s dvisy&tam dvisy&tadvisiihksdvisiyatham dvisldhvdm\n3\ndvisyat dvisyatam dvisyusdvisitddvisiyatamdvisirdri\nSolikewise, fromj/i,iydm andiylyd; fromyduh, duhydm andduhiyd;\nand soon.\nTheRV. hasonce tana in2dpi.act.(insyatana).\n4.Present Imperative.\n\"^617. Theimperative adds,insecond andthirdpersons,\nitsown endings (with SfrTFT atum in3dpi.mid.) directly\ntotheroot-stem. Thestem isaccented andstrengthened\nin3dsing, act.; elsewhere, theaccent isontheending\nandtherootremainsunchanged. The firstpersons,socalled,\nofthe laterlanguagearefrom theoldsubjunctive, and\nhave itsstrengthened stem and accent; theyarerepeated\nherefromwheretheyweregiven above(614).Inthe2d\nsing, act., theendingisregularly (asinthetwofollowing\nclasses) ftjdhi iftherootendwith aconsonant, andf%hi\nifitendwith avowel. Asexamples wetakesome ofthe\nrootsalready used forthepurpose. r\nactive. middle.\ns- d. p. g. d. p.\ndyani dyava dyama asai asavahai asamahai\n^p ^rFT ^rT MlfH MIHIMM t4i^H\nihi itdm ltd assvd ashtham addhvdm\n3yr\\\netu itamydntu astam asatam asatam\n"
+    },
+    {
+        "page": 235,
+        "content": "620]I.ROOT-CLASS [SECOND, ad-CLASx. 215\ndohani doJiava dohama dohai dbhavahai dohdmahdi\njfirip~^jrcr g^r ^IMIH^ gnj\\\ndugdlii dugdhdm dugdhd dhuksvd duhhtham dhugdhvdm\n^\ndogdhu dugdhum duhdntudugdlihm duhktam duhdtdm\n618. The2dsing.act.endingtat isfound intheolder language ina\nfewverbs ofthis class: namely, vittdt, vitat, brutat. In3dsing, mid., two\northree verbs have intheolder language theending am :thus,duhdm(only\nRV.case), uidara, fayam; andin3dpi.mid.AV. hasduhrdm andduhratdm.\nTheuse oftana fortain2dpi.act. isquite frequent intheVeda: thus,\nitana, yatdna, hantana, etc.And instota, etana, sofana, wehaveexamples\ninthesame personofastrong (and accented) stem.\n5.Present Participle.\n619.The activeparticiple hastheending*3fi[jint(weak\nstem-form 5IH at]added totheunstrengthenedroot. Me-\nchanically,itmaybeformed from the3dpi.bydropping\nthefinal^i.Thus, fortheverbs inflected above, theactive\nparticiplesare EPFTydnt,p^H duhdnt, f^Nri dmsdnt. The\nfeminine stemendsusuallyin5^ ati :thus, ZTrft yati-,T^^ft\nduhati, f^Mrfl dvisati: but,from roots in\n,in5Ttf?ft anti\nor5flrft ati(449).\nThemiddleparticiplehastheending ETRand,added\ntotheunstrengthenedroot: thus, ^ETR iyand, 3^11 duhand^\ndmsand.\nButanumber oftheseparticiplesintheolder language\nhave adouble accent, either ontheendingorontheradical\nsyllable:thus, duhdnd andduhana(also dughana],vidand and\nvidana, stivand and siivana, stuvand andstdvdna thelasthaving\nalso astronger form oftherootwhen accented. The root as,\n'sit', forms theuniqueUsma(along with,intheVeda, dsand}.\n6.Imperfect.\n620. This tense adds thesecondary endingstotheroot\nasincreased byprefixionoftheaugment.Theroothasthe\n^w^a-strengthening (ifcapableofit)inthethree personsof\n"
+    },
+    {
+        "page": 236,
+        "content": "216 IX.PRESENT-SYSTEM.[620\nthesingular, althoughtheaccent isalways upontheaugment.\nExamplesofinflection are:\nactive. middle,\ns. d. p.s. d.p.\nat/amaiva aima Usi asvahi asmahi\nais aitam aita asthas Usatham addhvam\nait aitam ayanasta asatam mata\nand,from theroot\nddoham dduhva dduhma dduhi dduhvahi dduhmahi\n2\nddugdham ddugdha ddugdhas dduhatham ddhugdhvam\nddhok ddugdham dduhan ddugdha dduhatam dduhata\n621. Roots endinginamayinthelater language option-\nallytake usinstead ofanin3dpi.act.(theabeinglostbefore\nit) ;andintheolderthey always doso :thus, dyusfromVya,\ndpusfromYpa 'protect'. Thesame endingisalsoallowed and\nmetwith inthecase ofafew roots endinginconsonants :\nnamely vid,'know', caks, dvis, duh, mrj.\nTheending tana, 2dpi.act., isfound intheVeda indyatana, dsastana,\naftana.\nTosave thecharacteristic endings in2dand3dsing, act., therootad\ninserts a:thus, ddas, ddat; theroot asinserts 1:thus, dsis, dsit(see\nbelow, 636).\n622. The use ofthepersons ofthis tense, without augment,inthe\nolderlanguage, either inthesame sense aswithaugment, orassubjunctives,\nhasbeen noticed above(587). Augmentless imperfects ofthis class are\nrather uncommon intheVeda :thus, h&n, ves,2dsing. ;ftan, vet, staut,\nddn(?),3dsing.; bruvan, duhus, eafesus, 3dpi.; vasta, suta, 3dsing. mid.\n623. The first orroot-form ofaorist isidentical initsformation with\nthisimperfect: seebelow, 829 ff.\n624. IntheVeda(butalmost limited toRV.)arefound certain second\npersons singular, made byadding theendingsitothe(accented andstreng-\nthened) root, andhaving animperative value. There issome difference of\nview astotheir formalcharacter; butthemostacceptable opinion regards\nthem asisolated indicativepersonsofthisclass, usedimperatively. They\n"
+    },
+    {
+        "page": 237,
+        "content": "629]I.ROOT-CLASS (SECOND, a</-CLASs). 217\nare :ksesi(1fcai'rule'), jesi, jdsi (for jossi, fromyjus), ddrsi, dhaksi, naksi\n(2naf 'attain'), nesi, pdrsi (2pr 'setacross'), prdsi, bhaksi, mdtsi, mdsi\n(2ma'measure'), ydksi, ydrhsi, yasi, y6tsi, rdtsi, rdsi, vdksi,vesi(ivi'strive\nafter'), frfoj, safest, sdtsi, hosi.\nIrregularities oftheRoot-class.\n625. Itisimpossible (atleast atpresent)todetermine with\naccuracy howmanyoftheactually used roots ofthelanguage\nare inflected inthepresent-system accordingtothisclass, or\naccordingtoanyoftheother classes, because theolder language\nespecially, andthelater inlessdegree, hassporadicforms which\nareeither ofdoubtful classification ortooisolated todetermine\nthecharacter oftheroot towhichthey belong. The root-class\nmaybesaid, however, toinclude from seventytoninetyroots.\nAconsiderable number ofthempresent irregularitiesofinflection,\nabrief account ofwhich(notclaiming exhaustivecompleteness)\nisgiveninthefollowing paragraphs.\n626. The roots ofthe class endinginuhave intheir\nstrong forms thevrddhi instead oftheywwa-strengtheningbefore\nanending beginningwith aconsonant :thus, from\\stu, staumi,\ndstaut, andthelike; butdstavam, stdvani, etc.\nRoots found toexhibit this peculiarityinactual useareksnu, yu, sfcit,\nstu,snu(thesefiveintheearlierlanguage), nw, ru,su'impel', andhnu.\n627. The rootmrjalsohasthevrddhi-vowel initsstrong\nforms: thus, marjmi, dmarjam, dmart; and thesamestreng-\ntheningisallowed inweak forms before endings beginningwith\navowel :thus, marjantu, amarjan:but this isnotfound to\noccur intheolder language.\nIntheother tense-systems, also, andinderivation, mrjshows often\nthevrddhi instead ofthepuna-strengthening.\n628.Anumber ofroots accent theradicalsyllable through-\nout,both instrong and inweak forms :thus, allthose begin-\nningwith alongvowel, as,id, ir,iq;and also caks, taks, tra,\nnihs, vas 'clothe', cinj, cl,andsu'generate'.All these, except\ntaksand tra(andtraintheonlyVedicforms),areordinarily\nconjugatedinmiddle voice only. Forms with thesame irregular\naccent occurnowandthen intheVeda from other verbs :thus,\nmdtsva, fdhat. Middleparticiplessoaccented have been noticed\nabove(619).\n629. Oftheroots mentioned inthelastparagraph,cihas\nthe^wwa-strengthening throughout: thus, cdye,cese, qayiya,\ncdyana, and soon. Other irregularitiesinitsinflection(in\npart already noticed)arethe3dpi.personscerate (AV.etc.\n"
+    },
+    {
+        "page": 238,
+        "content": "218IX.PRESENT- SYSTEM.[629\nhave alsocere], ceratdm, dcerata {RV.has also dceran], the3d\nsing, pres/ cdye (R.)andimpv. cdydm. The isolated active form\ndcayatiscommon intheolder language.\n630. Ofthesame roots, Idand ifinsert aunion-vowel ibefore endings\nbeginningwiths,sv,th:thus, ifise, ifidhve, idisva (these three being the\nonly forms noted intheolder language); butRV. hasiksebeside Ifwe.The\n3dpi.ifire (onaccount ofitsaccent)isalsoapparently present rather than\nperfect.\n631. Theroots rud(notinVeda), svap, an,andcvasinsert\naunion-vowel ibefore alltheendings beginning with acon-\nsonant, exceptthe sand tof2dand3dsing, impf., where they\ninsert instead either aori:thus, svdpimi, cvdstsi, dniti, and\nmat orunit.And intheremaining forms, thelastthree are\nallowed toaccent either root orending:thus, svdpantu and\ncvdsantu(AV.),orsvapdntuetc.\nIntheolder language, yvam makes thesame insertions :thus, vamiti,\navamit; andother cases occasionally occur: thus, janisva, vasisva(yvas 'aim'),\nfnathihi,stanihi(allRV.). Ontheother hand, ansometimes makes forms\nfrom ana-stem: thus, anati (AV.); pplednant(QB.); opt. anet (AB.).\n632. The root bru(ofveryfrequent use)takes theunion-\nvowel iafter therootwhen strengthened, before the initial vowel\nofanending:thus, brdvimi, brdvisi, brdviti, dbravts, dbravit; but\nbrumds, bruyam, dbravam, dbruvan, etc.Specialoccasionalirreg-\nularities arebrumi, bravihi, abruvam, abruvan, bruydt, andsporadic\nforms from ana-stem. Thesubj.dual brdvaite hasbeen noticed\nabove(615)..<\n633. Some ofthe roots inuareallowed tobeinflected likebru:\nnamely, few, tu,ru,andstu;andanoccasional instance ismet with ofa\nform somade(inV.,only tavltinoted).\n634. The rootam(hardly found inthelater language)takes Iasunion-\nvowel :thus,arriisi(RV.),amlti andamit andamisva (TS). From|/pam\noccurcamwvo (VS.: TS. famisva) andfamidhvam (TB. etc.).\n635. Theirregularities ofyduhintheolder language havebeen already\ninpartnoted: the3dpi.indie, mid. duhate, duhre, andduhrdte; 3dsing,\nimpv. duhdm, pi.duhrSm andduhratam; impf.act.3dsing, dduhat (which\nisfound also inthe laterlanguage), 3dpi.aduhran (beside dduhan and\nduhtis); themid. pple dtighana; and(quite unexampled elsewhere) theopt.\nforms duhiydt andduhiydn (RV. only).\nSome oftheroots ofthis class areabbreviated orotherwise\nweakened intheirweak forms :thus\n(636.Theroot 5TFT as, 'be', loses itsvowel inweak forms\n(except whereprotected bycombination with theaugment).\n"
+    },
+    {
+        "page": 239,
+        "content": "637]I.ROOT-CLASS (SECOND, 219\nIts2dsing,indie, isgr dsi(instead ofassi);its2dsing,\nimpv.is^ftfedhi(irregularly fromasdhi). The insertion of\n|iin2dand3dsing. impf. hasbeen noticedalready\nabove.\nTheforms ofthisextremely common verbare, then, as\nfollows :\nIndicative,\nd.\ndsisvas\nsthds\ndsti stds\nImperative.p.\nT\nsmds\nW\nsthd\nH%\nsdntir\nOptative,\nd.\nsyam\nHTTT^\nsyassyava\nKllrtH^\nsyatam\nsyat syatam\nImperfect.syama\nnnrT\nsyatam\nsyus\ndsani dsava dsama asam asva asma\nedhi stdm aSIS astam asta\ndstu stam sdntu\nParticiple H?Tasit astam\n(fern. Hrft sailasan\nTheVedic subjunctive forms aretheusual ones, made uponthestem\ndaa. They areinfrequent use, andappear (asat especially) even inlate\nBrahmanas where thesubjunctiveisalmost lost. The resolution sidm etc.\n(opt.)iscommon inVedic verse. In2dand3dsing. impf.isafewtimes\nmetwith themore normal as(for as-s, ds-t}.\nMiddle forms fromyasare alsogiven bythegrammariansasallowed\nwith certainprepositions (vi+ati),butthey donotappeartohave beenmet\nwith inuse.Amiddle presentindicative iscompounded (in1stand2dper-\nsons) with thenomen agentisintr(tar]toform theperiphrasticfuture in\nthemiddle voice(see below, 942ff.).The 1stsing, indie, ishe;therest\nisintheusual relation ofmiddle toactive forms (in2dpers., e,dhve,\nsva,dhvam, with total loss oftheroot itself).\nTheonlyother tense ofthisverb inuse istheperfect,which isentirely\nregular initsinflection.\n637. The root han, 'strike, slay',istreated somewhat after\nthemanner ofnoun-stems inanindeclension (421):inweak\n"
+    },
+    {
+        "page": 240,
+        "content": "220IX-PRESENT-SYSTEM.[637\nforms,itloses itsnbefore aninitial consonant(except mandv)\nofanending, and itsabefore aninitial vowel andinthe\nlatter case itsA,incontact with the n,ischangedtogh(com-\npare 402). Thus, forexample:\nPresent Indicative. Imperfect.\ns. d. p.s. d.p.\n1hdnmi hanvds hanmds dhanam dhanva dhanma\n2hdnsi hathds hathd dhan dhatam dhata\n3hdnti hatds ghndnti dhan dhatam dghnan\nItsparticipleisghndnt (fern, g/matij.Its2dsing. impv.\nisjaM (byanomalous dissimilation, onthemodel ofredupli-\ncating forms).\nMiddle forms from this root arefrequent intheBrahmanas, andthose\nthat occur areformed accordingtothesame rules :thus, hate, hanmahe,\nghnate; ahata, aghnalam. aghnata (inAB., alsoahata)] ghnlta (but alsohanita).\n638. The root vac, 'wish',isintheweak formsregularly\nandusuallycontracted touc(asintheperfect:seechap. X.):\nthus, ucmdsi(V.:onceapparently abbreviated inRV. tocmasi),\nucdnti; pple ucdnt, ucand. Middle forms(exceptthepple)donot\noccur;nordotheweak forms oftheimperfect, which aregiven\nasaufva, austam, etc.\nRV. hasinlikemanner theparticiple usand from therootvas 'clothe'.\n639. The root cas,'command', shows some ofthepeculi-\narities ofareduplicated verb, lacking (646)thenbefore tin\nall3dperss. pi.and inthe activeparticiple. Apartofits\nactive forms namely,theweak forms having endings begin-\nning with consonants(includingtheoptative)come from a\nstemwithweakened vowel, fis(asdotheaorist, 854, andsome\nofthederivatives). Thus, forexample:\nPresent Indicative.Imperfect.\ns. d.p. s. d.p.\n1fasmi tfsvds fismds dfasam dfisva dfisma\n2fassi cisthds fisthd d?as dfistam dfista\n3fasti fistds fasati dfat dfistam dfdsus\nIn2dsing, impf., afdt issaid tobealsoallowed. Ifitisactually so\nused, the tmust bethesporadic conversion offinal radical stot(167);\nandthen itwould beopentoquestion whether the tof3dsing,isradical\noroftheending (according to555). Theoptativeisfisydmetc.The2d\nsing. impv.is$adhf (with total loss ofthes);andRV. hasthestrong\n2dpi.fastdna (with anomalousaccent). Butnopis-forms ofthepresent-\nsystem occur intheVeda.\nThemiddle inflection isregular, andtheaccent(apparently)\nalways upon theradicalsyllable (pple casana, RV.etc.).\nThe rootdap, 'worship', has inlikemanner (RV.)thepple ddfat (not\n"
+    },
+    {
+        "page": 241,
+        "content": "644]II.REDUPLICATING CLASS (THIRD, AW-CLASS). 221\n640. The double so-called rootjaks, 'eat, laugh',isanevident re-\nduplication ofghas and has. Ithastheabsence ofninact.3dperss. pi.\nandpple, andtheaccent onthe root beforevowel-endings, which belong\ntoreduplicated verbs; and italsotakes theunion-vowel iinthemanner of\nrud etc.(above, 631).A2dpers. impv. jagdhi from itoccurs.\n641. Otherobviously reduplicated verbs aretreated bythe\nnative grammarians asifsimple, andreferred tothisconjugation:\nsuch are theintensively reduplicated jagr, daridra, and vevl\n(chap. XIV.),didhi etc.(676), andcakas(677).\nII.Reduplicating Class (third, /m-class).\n642. This class forms itspresent-stem byprefixinga\nreduplicationtotheroot.\n\\643. a.Asregardstheconsonant of.thereduplication,\nthegeneral ruleswhich havealready beengiven above(590)\narefollowed.\nb.Alongvowel isshortened inthereduplicating syl-\nlable: thus,^Jdada fromi/^Tda; fspft bibhlfromy$bhl;\nst^juhu fromyj^hu.Thevowel fjrneverappearsinthe\nreduplication, but isreplaced by^i:thus. f^Tbibhr from\nyy{bhr;fqtfrfpiprc from yvftpro.\nForverbs inwhich aandaalso areirregularly representedinthere-\nduplication by i,seebelow, 660.\nc.Theonlyroot ofthis class with initial vowel isr(or\nar]\\ittakes asreduplication t,which isheldapart from the\nrootbyaninterposed y:thus, iyarandiyr(thelatter hasnot\nbeen found inactualuse).\n\\644.Thepresent-stemofthis class(asoftheother\nclassesbelongingtothe firstconjugation)has adouble\nform :astronger form, with gunatedroot-vowel;anda\nweaker form, without guna: thus, fromj/^\"hu,thetwoforms\nare^^Ij'uhoand3^juhu; fromj/~>ft bhi,theyarei^Hbibhe\nand fipft bibhl. And therule fortheir use isthesame as\nintheother classes ofthisconjugation:thestrongstem is\nfound before theunaccented endings (552),and theweak\nstem before theaccented.\n"
+    },
+    {
+        "page": 242,
+        "content": "222 IX-PRESENT-SYSTEM.[645\n645. Accordingtoalltheanalogiesofthe firstgeneral\nconjugation,weshould expecttofind theaccent upontheroot-\nsyllable when this isstrengthened. That isactuallythecase,\nhowever, onlyinasmall minorityoftheroots composingthe\nclass :namely,inhu, bhi(notest-forms intheolderlanguage),\nhri(notfound intheolderlanguage), madanddhan(both very\nrare), jan (noforms ofthis class found tooccur),ci'notice'\n(inV.),yu'separate' (inolder language only),and inbhrin\nthe later language (inV. itgoes with themajority:butRV.\nhas bibharti once\n;and this, thelateraccentuation,isfound also\nintheBrahmanas).Inalltherestapparently, byarecent\ntransfer itrests uponthereduplicating instead ofuponthe\nradical syllable. And inboth classes alike, theaccent isanom-\nalously thrown back uponthereduplication inthose weak\nforms ofwhich theending begins with avowel\n;while inthe\nother weak forms itisupontheending.\nApparently (the cases with written accent aretoofew todetermine the\npoint satisfactorily) themiddle optative endings, lyaetc. (566), arereckoned\nthroughoutasendings with initial vowel, andthrow back theaccent upon\nthereduplication.\n\\646.Theverbs ofthis class lose the ?ninthe3d\nK,\npl.endingsinactive aswell asmiddle, andintheimper-\nfecthave3^wsinstead of3Ran andbefore thisafinal\nradical vowel hasguna,\n1.Present Indicative.\n\"647.Thecombination ofstem andendingsisasin\ntheprecedingclass.\nExamples ofinflection: a.y\"^ hu, 'sacrifice':\nstrong stem-form, sl^T/ttAo/weak form, sC^juhu (orjuhu).\nactive. middle.\nsw_d.p.s. d.p.\n2\njuhoti juhutas juhvati juhute juhvate juhvate\nb.Root\n*j[bhr,'bear'(givenwithVedicaccentuation):\nstrong stem-form, fsPTfbibhar ;weak,^Rbibhr(orbibhr}.juhomi juhuvds juhumds juhve juhuvdhe juhumdhe\njuhosi juhuthds juhutha juhuse juhvathe juhudhve\n"
+    },
+    {
+        "page": 243,
+        "content": "650]II.REDUPLICATING CLASS (THIRD, AW-CLASS). 223\nN,\nbibharmi bibhrvds bibhrmds bibhre bibhrvdhe bibhrmdhc\nC. *\\ C. c. C.\nbibharsi bibhrthds bibhrthd bibhrse bibhrathe bibhrdhve\n-x\nbibharti bibhrtds bibhrati bibhrte bibhrate bibhrate\nTheuofhu(likethat oftheclass-signs nuandu:seebelow, 697)\nissaid tobeomissible before vandmoftheendings of1stdu.andpi.:\nthus, juhvds, juhvdhe. etc.\n2.Present Subjunctive.\n648. Itisnotpossible (atleast,atpresent)todraw adistinct line\nbetween those subjunctive forms ofthe older language which should be\nreckoned asbelongingtothepresent-system andthose which should beas-\nsigned totheperfect even, insome cases, tothereduplicated aorist and\nintensive. Here willbenoticed only those which most clearly belongtothis\nclass; themore doubtful cases willbetreated under theperfect-system.\nExcept infirst persons (which continue inuseas\"imperatives\" down tothe\nlaterlanguage), subjunctives from roots having unmistakably areduplicated\npresent-system areoffarfrom frequent occurrence.\n649. Thesubjunctive mode-stem isformed intheusual\nmanner, with themode-sign aandgunaoftheroot-vowel,if\nthis iscapableofsuchstrengthening. Theevidence ofthefew\naccented forms metwith indicates that theaccent islaid inac-\ncordance with that ofthestrongindicative forms :thus, from\nyhu,thestemwould bejuhdva:fromybhr,itwould bebibhara\n(butbibhdralater).Before themode-sign,final radical awould\nbe,inaccordance withanalogies elsewhere, dropped:thus ddda\nfromyda,dddha fromydha (alltheforms actually occurring\nwould bederivable from secondary roots,asdadanddadh}.\n650. Instead ofgivingatheoretically complete scheme of\ninflection,itwillbebetter tonote alltheexamples quotable\nfrom theolder language (accented when found sooccurring).\nThus, of1stpersons, wehave intheactive juhdvani, bibhardni, dadani,\ndadhdni, jahani; juhavdma, dddhama, jdhama;inthemiddle, dadhai,\nmimdij dadhdvahdi; juhavdmahdi, daddmahe, daddmahai, dadhdmahdi.\nOfotherpersons, wehave with primary endingsintheactive bibhardsi\n(with double mode-sign: 560, end), dddhathas, juhavatha (do.)andjuhavatha;\ninthemiddle, dddhase; dddhate, rdrate, dddhdtdi,daddtdi: with second-\naryendings, dddhas, vfvesas, juhavat, bibharat, yuydvat, dddhat, dadhdnat,\nbabhasat;dadhan^ yuyavan, juhavan.\n"
+    },
+    {
+        "page": 244,
+        "content": "224IX.PRESENT-SYSTEM.[651\n3.Present Optative.\n\\651.Toform thismode, theoptative endings given\nabove(566), asmadeupofmode-sign andpersonal endings,\nareadded totheunstrengthenedstem. The accent isas\nalreadystated(645). The inflection issoregular that itis\nunnecessarytogiveheremore than the firstpersonsofa\nsingleverb :thus,\nactive. middle.\ns. d.p.s. d.\njuhuyhm juhuy&va juhuy&ma juhvlya juhvwahi juhvimahi\netc. etc. etc. etc. etc. etc.\n4.Present Imperative.\n\\652.Theendings, andthemode oftheir combination\nwith theroot, havebeenalready given. In2dsing, act.,\ntheendingisf%hiafter avowel, but f?Idhi after acon-\nsonant :^hujhowever, forms sT^Rl juhudhi (apparently.\ninorder toavoid therecurrence of^hintwo successive\nsyllables): andotherexamplesoffadhi after avowel are\nfound intheVeda.\n{653.Exampleofinflection:\nactive. middle.\ns. d. . s. d. p.\njuJidvani juhavava juhdvama juhdvai juhdvavahai juhdvamahai\n2g^iu g^riH^ g^H spra ^lyiH^ 5^^\njuhudhi juhut&m juhuta juhusvd juhvatham juhudhvdm\n33^'5^niH^ g^g ^diH^ g^iniH^ sj^riiH^\njuhdtu juhutnm jtihvatu juhuthm juhvatam juhvatam\nThe other division ofthis class differ here, asinthein-\ndicative, intheaccentuation oftheir strong forms only:namely,\ninallthe firstpersons (borrowed subjunctives),and inthe3d\nsing,act.: thus(intheolderlanguage)Ubharani etc., Ubhartu,\nlibharai etc.\n654. Vedic irregularities ofinflection are: a.the occasional use of\nstrong forms in2dpersons: thus, yuyodhi, fifadhi (beside tifihf); yuyotam\n(beside yuyutdm}-, {yarta, dddata, dddhata anddddhatana(seebelow, 673),\n"
+    },
+    {
+        "page": 245,
+        "content": "6581 II.REDUPLICATING CLASS (THIRD, 225\npipartana, juh6ta andjuhdtana, yuyotana; b.theuseofdhiinstead ofhi\nafter avowel (only inthetwo instances just quoted) ;c.theending tana\nin2dpi.act.(intheinstances just quoted and inothers, asmamaltana,\njigatana, dhattana, etc.); d.theendingtatin2dsing, act., indattdt,\ndhattat, piprtat.\n5.Present Participle.\n655.Aselsewhere, the activeparticiple-stem maybe\nmademechanically from the3dpi.indie, bydropping ^i:\nthus,^^juAvat, fsffi^bibhrat.Ininflection,ithasnodis-\ntinction ofstrong andweak forms(444). The feminine\nstem ends inEfi-fj1\nati.Themiddleparticiplesareregularly\nmade:thus,sl^M Jukvana, I^WU blbhrana.\n6.Imperfect.\n656.Asalready pointed out, the3dpi.act.ofthis\nclass takes theending 3?T us,andafinal radical vowel has\ngunabefore it.Thestrongformsare,asinpres. indie.,\nthethreesing.act.persons.\nf*657.Example ofinflection:\nactive,\nd.middle.\nd.\ndjuhavam djuhuva ajuhuma djuhvi ajuhuvahi ajuhumahi\ndjuhos djuhutam djuhuta djuhuthas djuhvatham djiihudhvam\ndjuhot djuhutam djuhavus djufiuta djuhvatam djuhvata\nFrom }/Hbhr, the2dand3dsing.act. are\ndbibhar(forabibhar-s andabibhar-t]and soinallother\ncases where thestrongstem ends inaconsonant. The3d\npi.act. is^^^dbibharus;andfromy^\\ bhi,itistjf^tHH^\nabibhaym.\n658. Theusual Vedic irregularitiesin2dpi.act. strong forms, and\ntheending tana occur inthistense also :thus, adadata, ddadhata ;ddat-\ntana, djagantana, djahatana. TheRV. has alsoonce apiprataforapiprta\nin3dsing, mid., andabibhran forabibharus in3dpi.act.Examplesof\naugmentless forms arepi'?as, vives, jtgat; jthlta, ffylta, jihata.\nWhitney ,Grammar. 15\n"
+    },
+    {
+        "page": 246,
+        "content": "226 IX -PRESENT-SYSTEM.[659\nIrregularities oftheReduplicating Class.\n659. Itisstillmore difficult todetermine thepreciselimits\nofthis class than oftheroot-class, because oftheimpossibility\n(referredtoabove, under subjunctive:648)ofalways separating\nitsforms from those ofotherreduplicating conjugations and\npartsofconjugations.IntheRV., aboutfortyrootsmaybe\nconfidently assignedtoit;intheAV., lessthanthirty; many\nofthem haveirregularities (besidesthose intense-inflection\nalready pointed out).\n660. Besides theroots inrorar namely, r,ghr(usually\nwritten ghar\\tr,pr, Ihr, sr,prothefollowingroots having\naoraasradical vowel take tinstead ofainthereduplicating\nsyllable: ga'go',Spa 'rise',ma'measure', ma'bellow', ca,ha\n'remove'(mid.), vac, sac; vachasboth ianda;rahas *once\ninRV.: forstha,pa'drink', gkra, Jian, seebelow(670 4).\n661. Several roots ofthis class infinal achange theain\nweak forms to *(occasionallyeven toi),andthen dropitalto-\ngetherbefore endings beginning with avowel.\nThis isinclose analogy with thetreatment ofthevowel ofthe class-\nsign oftheno-class: below, 717.\nThese roots are :\n662. fa,act.andmid.:thus, fifati, fifimasi, fiflhi (also fifadhi:above,\n654), fifatu, afifat, tf$lte.\n663.ma'bellow', act., andma'measure', mid.(rarelyalsoact.): thus,\nmimati, mimanti, mimiyat ;mmuie, mimafc, amimlta; mimihi, mtmatu.\n664. ha'remove', mid.: thus, jtfiite, jihidhve, jtfiate; jihisva, jihatam;\ndjihlta, ajihata.\n665. ha'quit',act.(originallyidentical with theformer), may further\nshorten the Itoi:thus, jahati, jahita, jahltat (AV.); jahimas (AV.), jahitas\n(TB.), jahitam (TA.), ajahitam (TS.). Intheoptative,theradical vowel is\nlost altogether; thus, jahyam, jahyus (AV.). The2dsing. impv.isjahlhi\norjahihi.\nCompare with thistheforms indhifromydha (below, 669).\n666. ra'give', mid.: thus, ranaTivam, ranthas(impf. without augment):\nand, with iinreduplication,ririhi.\nInallthese verbs, theaccent isconstant onthereduplicating syllable.\n667.Thetwo roots daanddha(thecommonest ofthe\nclass)lose their radical vowelaltogetherintheweak forms,\nbeing shortened todadanddadh. In2dsing. impv. act., they\nformrespectively dehiand dhehi. Incombination with afol-\nlowingtorth,thefinal dhofdadh does notfollow thespecial\nrule ofcombination ofafinal sonantaspirate (becoming ddh\nwith the torth:160), but asalsobefore sanddhv the\n"
+    },
+    {
+        "page": 247,
+        "content": "671]II.REDUPLICATING CLASS (THIRD. 227\nmore general rules ofaspirate and ofsurd andsonant com-\nbination\n;and itslostaspirationisthrown backuponthe initial\noftheroot.\n668.The inflection ofydhais.then, asfollows:\nPresent Indicative,\nactive.\ns. d. p.\n1dddhdmi dadhvds dadhmds\n2dddhdsi dhatthds dhatthd\n3dddhati dhattds dddhatis.\ndadhe\n"
+    },
+    {
+        "page": 248,
+        "content": "228IX -PRESENT-SYSTEM.[671\ntisthami, pibami (with irregular sonantizingofthesecond p:\nlater often writtenpivami),andjiyhramiwhich then arein-\nflected notlikemimami, butlikebMvami, asiffrom thepresent-\nstems tistha, piba, jighra.\n672. IntheVeda,thereduplicatedroots daanddhaarealsosometimes\nturned into thea-stems ddda anddddha, orinflected asifrootsdadand\ndadh ofclass VI.;and single forms ofthesame character aremade from\nother roots: thus,mimanti (jAna 'bellow'), rdrate(j/>a 'give': 3dsing. mid.).\n673. IntheVeda, also,alikesecondary root, jighn,ismade fromyhan\n(withomission oftheradical vowel, andconversion, usual inthisroot,ofh\ntoghwhen incontact withn);andsome oftheforms ofsafe, fromy~sac,\nshow thesame conversion toana-stem, safca.\n674. InAB.(viii. 28),asimilar secondary form, jighy,isgiven toj/^/u:\nthus, jighy ati,jighy atu.\n675.Afewso-called roots ofthe first orroot-class aretheproducts of\nreduplication, more orlessobvious: thus, jaks (640), andprobably fas(from\ny~fas)andcaks (from j/fcaporalostrootfcas, 'see'). IntheVeda isfound\nalso sa$c, fromysac.\n676. Thegrammarians reckon(asalready noticed, 641) several roots of\nthemost evidently reduplicatecharacter assimple, andbelonging totheroot-\nclass. Some ofthese(jagr, daridra, vevl)areregular intensive stems, and\nwillbedescribed below under Intensives(chap. XIV.); didhi, 'shine', along\nwithVedic dldl 'shine' andplpl 'swell',aresometimes alsoclassed asinten-\nsives;buttheyhave nottheproper reduplication ofsuch, andmayperhaps\nbebest noticed here,asreduplicated present-stems withirregularly long\nreduplicating vowel.\nOfpres. indie, occurs intheolder language only didyati, 3dpi.,with\nthepples dZdyat anddidhyat, andmid. didye, didhye, dldhyatham, with the\npples didyana, didhyana, pipyana. The subj. stems arediddya, dldhaya,\nplpdya, andfromthem aremade forms with bothprimary (from dlddya] and\nsecondary endings (and the irregularly accented didayat anddldayat and\ndldhayan}. Noopt. occurs. Inimpv. wehave didiln (and didihi) andpipihi,\nandpipyatam, pipyatam, pipyata.Inimpf., adldes andpipes,ddldet and\nddldhet andaplpet (with augmentless forms), aplpema (with strong form of\nroot), andadidhayus and(irregular) apipyan.\nAfewforms from allthethree show transfer toana-inflection :thus,\ndldhaya andplpaya (impv.), dpipayat,etc.\nSimilar forms fromy~mi'bellow' areamimet andmimayat.\n677. Thestem cafcas (sometimes cafeap)isalsoregarded bythegram-\nmarians asaroot, andsuppliedassuch with tenses outside thepresent-\nsystem which, however, hardly occur ingenuine use. Itisnotknown\nintheolderlanguage.\n678. Theroot bhas, 'chew', loses itsradical vowel inweak forms, taking\ntheform baps: thus, bdbhasti, butbdpsati (3pi.), bdpsat (pple).\n679. The rootbhl, 'fear',isallowed bythegrammarians toshorten its\nvowel inweak forms :thus, bibhlmas orbibhimas, bibhlyam orbibhiyam.\n"
+    },
+    {
+        "page": 249,
+        "content": "684]II.REDUPLICATING CLASS (THIRD, 229\n680. Forms ofthis classfromyjan, 'give birth', withadded i thus,\njajnise, jajnidhve aregiven bythegrammarians, butdonotappearto\nhave been found inuse.\n681. The roots ciand cithave intheVeda reversion ofetokinthe\nroot-syllable after thereduplication: thus, cikeai, cikethe(anomalous, for\ncikydthe], cikitam, aciket, cikyat (pple); cikiddhi.\n682. The rootvyac has iinthereduplication (from they),and iscon-\ntracted tovieinweak forms :thus, viviktds, dviviktam. Sotheroothvar\n(ifitsforms are tobereckonedhere) hasuinreduplication, andcontracts\ntohur :thus, juhurthas.\nIII.Nasal Class (seventh, ra///-class).\n683.The roots ofthisclass allendinconsonants. And\ntheirclass-signisanasalprecedingthefinal consonant :in\ntheweak forms, anasalsimply, adaptedincharacter tothe\nconsonant\n;butinthestrongformsexpandedtothesyllable\n^ndjwhich hastheaccent.\nInafew oftheverbs oftheclass, thenasal extends also intoother\ntense-systems: they arean}, bhanj, hins :seebelow, 694.\n1.Present Indicative.\n684.Examples ofinflection: a.theroot Jsfyuj,\n'join': strong stem-form, 7H5Tyundj ;weak, ET^\"yunj.\nFortherules ofcombination offinalj,see219.\nactive.\nd.middle.\nd.\nyundjmi yunjvas yunjmds yunje yunjvdhe yunjmahe\n2yrf^i UC^H^ g^r nf\" i^w\\ jnr\nyundhsi yunkthds yunkthd yunhse yunjathe yungdhve\n3uni^Urh^rsing%\" ^iTig^1\n**yundkti yunktds yunjdnti yunkte yunjhte yunjdte\nb.theroot\"^Trudh, 'obstruct'; bases\nrundh.\nForrules ofcombination offinal dh.see153^ 160.and\nrunddhmi rundhvds rundhmds rundhe rundhvdhe rundhmahe\n"
+    },
+    {
+        "page": 250,
+        "content": "230IX-PRESENT-SYSTEM.[684\n2\nrundtsi runddhds runddhd runtse rundhathe runddhve\n3\nrundddhi runddhds rundhdnti runddhe rundhnte rundhdte\nInstead ofyunkthas, yungdhve, and the like(here and in\ntheimpv.andimpf.),itisallowed andmore usual(231)to\nwrite yunthas, yundhve, etc.; and, inlikemanner, rundhas, rundhe,\nforrunddhas, runddhe; and soinother like cases.\n685. Vedic irregularitiesofinflection are :a.theordinary use ofa\n3dsing. mid. likethe1stsing.,asvrnje; b.theaccent onteof3dpi.mid.\ninanjate, indhate, bhunjate.\n2.Present Subjunctive.\n686. Thestem ismade,asusual, byadding atothestrong\npresent-stem:thus, yundja,runddha. Below aregiven asif\nmade from~\\'yujalltheforms forwhichexamples havebeen\nnoted asactually occurringintheolderlanguage.\nactive. middle.\ns. d. p.s. d.p.\n1yunajani yundjava yundjama yunajai yundjamahai\n2yundjas yunajadhvai\n3yundjat yundjatas yundjan yundjate\n687. TheRV.hasonce anjatas, which isanomalous asbeing made\nfrom theweak tense-stem. Forms with double mode-sign aremetwith:\nthus, trnahan(AV.), yunajan (QB.);andtheonlyquotable example of3ddu.\nact. (besides anjatas)ishinasdtas(^B.). QB.has alsohinasavas as1stdu.\nact.: anelsewhere unexampled form.\n3.Present Optative.\n688.Theoptativeismade,aselsewhere, byaddingthe\ncompounded mode-endingstotheweak form ofpresent-\nstem. Thus :\nactive. middle.\ns. d. . s. d. .\nyunjyhm yunjy&va yunjyhma yunjiyd yunjivdhi yunjimdhi\netc. etc. etc. etc. etc. etc.\n4.Present Imperative.\n689. Inthis class(astheroots allendinconsonants)\ntheendingofthe2dsing.act. isalwaysftfdhi,\n"
+    },
+    {
+        "page": 251,
+        "content": "693]III.NASAL CLASS (SEVENTH, rudh-CLAss}. 231\nactive. middle.\ns. d.p. s. d. p.\niyteiiPs JHsTrau^sllHZpfffUHslN< UHsllH^\nyun&jani yundjava yundjama yundjai yundjavahai yimdjamahai\n2gfrr1^ J3T 3?^ y^iyiH^ H7^\nyungdhi yunktam yunktd yunksvd yunjatham yungdhvam\nyundktu yunktam yunjdntu yunktam yunjatam yunjdtam\n690. There isnooccurrence, sofarasnoted, oftheendingtat,inverbs\nofthis class. TheVeda has, asusual, sometimes strong forms, andsome-\ntimes theending tana,inthe2dpi.act.:thus, undtta, yundkta, anaktana,\npinastana.\n5.Present Participle.\nf691.Theparticiplesaremade inthis class asinthe\nprecedingones: thus, act.^c^yunjdnt(fern, y^rfl yunjati);\nmid. ET^FT yunjand (butRV.hasindhana).\n6.Imperfect.\n692.Theexampleoftheregularinflection ofthistense\nneeds nointroduction:\nactive. middle.\ns. d. p.s. d.p.\n**tt\\*\\\ndyuajam dyunjva dyunjma dyunji dyunjvahi dywymahi\ndyunak dyunktam dyunkta dyunkthas dyunjatham dyungdhvam\n3\ndyunak dyunktam dyunjan dyunkta dyunjatam ayunjata\nTheendingssand /arenecessarilylostinthenasal class\nthroughoutin2dand3dsing, act., unless saved(555)atthe\nexpenseofthe final radical consonant :which isacase ofvery\nrareoccurrence :intheolder language have been notedonly\nahinat(TB.),3dsing.,forahinas(\\/hins\\andonce inAV.\nalhanas, 2dsing.,forabhanak(ybhanj)',this last isacase of\ntheutmost rarity.\n693. TheVeda shows noirregularitiesinthis tense. Occurrences of\naugmentless forms arefound, especiallyin2dand3dsing, act., showing\nanaccent likethat ofthepresent:forexample, bhindt, prndk, vrnak, pindk,\nrinak.\n"
+    },
+    {
+        "page": 252,
+        "content": "232IX.PRESENT-SYSTEM.[694\nIrregularities oftheNasal Class.\n694. The roots which thusexpandapenultimate nasal in\nthestrongforms ofthepresent-systeminto asyllable ndare\nabout twenty-five: namely, tac,pro, ric, vie, anj, bhanj, vrj,\nbhuj, yuj,krt'spin', chrd, trd, chid, bhid, ud,rdh, idh, rudh,\nubh, ac(anacamahai, once, RV.), pis, cis, hihs,trh.Those here\nwritten with thenasal namely anj, bhanj, hihs--have that\naddition also intheother tense-systems. Two, rdhand ubh,\nmake present-systemsalso ofother classes with nasalclass-signs:\nthus, rdhnoti(cl.IV.),ubhnati(cl.V.).Several have a-stems\nwith penultimatenasal: thus, prncd, cihsd, trhhd, umbhd;and\noccasional a-forms, especiallyinthe later language, aremet\nwithfrom others :thus, bhunjet, chindeta, apinsat, arundhat(com-\nparethenasalized roots ofthed-class, below, 758).\n695. The root trhcombines trnah with U,tu,etc. intotrnedhi, trnedhu;\nand, accordingtothegrammarians, has alsosuch forms astrnehmi: see\nabove, 224b.\n696. The root hihs (byorigin apparently adesiderative fromyhari)\naccents irregularlytheroot syllableintheweak forms :thus, hfhsanti, hfhste\n(buthindsat etc.).\nIV.Nuand^-classes(fifth and eighth,suandto-classes).\n\\697. A.Thepresent-stemoftheww-class ismadeby\naddingtotheroot thesyllable ^nu,which then inthe\nstrongforms receives theaccent, and isstrengthenedto^fino.\nB.Thefew roots ofthew-class(about half-a-dozen)\nendin^n,with theexceptionofthelaterirregularSfikr\n(orkar]forwhich, seebelow, 714. Thetwoclasses,\nthen, areclosely correspondentinform;andtheyarewholly\naccordant ininflection.\nTheuofeitherclass-signisallowed tobedroppedbefore\nvandmofthe 1stdu.and 1stpi.endings, except when the\nroot(raw-class)ends inaconsonant; andtheubefore avowel-\nending becomes voruv,accordingasitispreceded byoneor\nbytwoconsonants(129).\n1.Present Indicative.\nr698.Examples ofinflection: A.raw-class; roof\nHsu,'pressout':strong form ofstem, RTsuno ;weak form,\nRsunu.\n"
+    },
+    {
+        "page": 253,
+        "content": "70O]IV.NU-ANDU-(FIFTH ANDEIGHTH, SU-ANDtan-) CLASSES. 233\nactive. middle.\ns- d.p. s. d. n.\noo _\nsunomi sunuvds sunumds sunve sunuvdhe\nH^lfo0^^ oo^ oo^U-^lS)\nsunosi sunuthas sunuthd sunuse sunv&the\n00^.\nsunoti sunutds sunvdnti sunute sunvhte\nTheforms sunvds, sunmds, sunvdhe, sunmdhe arealternative\nwith thosegivenhere for 1stdu.andpi.,and inpractice are\nmorecommon(noexamples ofthefuller forms havebeen noted\nfrom theolderlanguage). From}/ap, however(forexample),\nonlytheforms with ucanoccur :thus, apnuvds, apnumdhe ;and\nalsoonly apnuvdnti, apnuve, apnuvdte.\nB.w-class; rootcHtan,'stretch':strong form ofstem.\nrPTTtano ;weak, rT'Ttanu.\ntanomi tanvds tanmds tanve tanvdhe tanmdhe\netc. etc. etc. etc. etc. etc.\nThe inflection issopreciselylike thatgiven above that it\nisnotworth writing outinfull. The abbreviated forms in1st\ndu.andpi.arepresented here, instead ofthefuller, whichrarely\noccur(asnodouble consonant everprecedes).\n699.' Inthe older language, nostrong 2dperss. du. orpi.,andno\nt/iana-endings, chance tooccur(but they arenumerous intheimpv. and\nimpf.: seebelow). TheRV. has several cases oftheirregular accent in\n3dpi.mid.: thus, tanvate, manvate, sprnvate.\nInRV.occur also several 3dpll.mid. inirefrom present-stemsofthis\nclass :thus, invire, rnvire, pinvire, crnvire, sunvire, hinvire. Ofthese,\npinvire andhinvire might beperfectswithout reduplicationfrom thesecond-\naryroots pinv andhinv(below, 716). The2dsing. mid.(with passive value)\nprnvise (RV.)isofanomalous andquestionablecharacter.\n2.Present Subjunctive.\n700. The subjunctivemode-stem ismade intheusual\nmanner, byaddingatothegunated andaccented class-sign:\nthus, sundva, tandva. Inthefollowing scheme aregivenallthe\nforms ofwhich exampleshave beenmetwith inactual usein\ntheolder language from either division oftheclass :some of\nthem arequite numerously representedthere.\n"
+    },
+    {
+        "page": 254,
+        "content": "234IX.PRESENT-SYSTEM.[700\nactive.\ns. d. p.\n1sundvani sundvava sundvdma\n2sundvas sundvatha\n3sundvat sundvanmiddle.\ns. d.\nsundvai sundvavahai\nsundvase sundvdithe\n(sundvate\n\\sundvdtdisundvdmaJtdt\nsundvanta\n701. Ofthebriefer 1stsing, act.,RV. haskrnava andhinava. Forms\nwithdouble mode-sign occur(notinRV.): thus, krndvdt andkaravdt (AV.);\nafnavdtha (K.),krnavatha (VS.; but-vatha inKanva-text), karavatha(QB.).\nOntheother hand, apnavataiisfound once(inTS.). RV. hasinasingle\npassagekrnvdtte (insteadofkrndvdile):theonlyform indithe isapndvaithe.\n3.Present Optative.\n702.Thecombinedendings (566)areadded, asusual.\ntotheweak tense-stem: thus,\nactive.\nd.middle.\nd.\nsunuyl\netc.sunuyava\netc.sunuyama sunviydsunvivdhi sunvimdhi\netc. etc. etc. etc.\nFromyap,themiddleoptative would beapnuviyd and\nsoinother like cases.\n4.Present Imperative.\n703.The inflection oftheimperativeisingenerallike\nthat intheprecedingclasses. Asregardsthe2dsing, act.,\ntherule ofthelaterlanguageisthat theending f^Thiis\ntaken whenever theroot itself ends inaconsonant; other-\nwise, thetense-(ormode-) stem standsbyitself as2dper-\nson(fortheearlierusage,seebelow, 704). Exampleofin-\nflection is :\nactive.\nd.middle,\nd.\n*- -sunavani sundvava sundvama sundvai sundvavahai sundvamahai\nsunu sunutam sunutd sunusvd sunvaiham sunudhvdm\n3THTrf HHHIH'^^5^OO\nj*SS. ^~r -*** >*/>^ -\"XN/ --V ^-^ \"S\nsunotu sunutam sunvdntu sunutam sunvatam sunvdtam\n"
+    },
+    {
+        "page": 255,
+        "content": "707]IV.Nil-ANDU-(FIFTH ANDEIGHTH, SU-ANDtan-} CLASSES. 235\nFromyap,the2dsing.act.would beapnuhi; from >'\nag,\nacnuhi; fromydJirs,dhrsnuhi;and soon.Fromyap, too,would bemadeapnuvdntu, cipnuvatham, apnuvatam, apnuvatam.\n704. Inthe earliestlanguage, therule astotheomission ofhiafter\narootwith final vowel does nothold good: inRV., such forms asinuhi,\nkrnuhf, cinuhf, dhunuhi, crnuhf, sprnuhi, hinuhi, and tanuhi, sanuhi, are\nthrice asfrequent inuse asinu, prnu, sunn, tanu, and their like; inAV.,\nhowever, they arenotmore thanonethird asfrequent; andintheBrahmanas\nthey appear only sporadically ;even frnudhi (with dhi) occurs several times\ninRV. Theendingtat isfound inkrnutat andhinutat, andkurutat. The\nstrong stem-form isfound in2ddu. act. inhinotam; and in2dpi.act.in\nkrnota andkrnotana, crnota andcrnotana, sun6ta andsuntitana,hintita and\nhinotana, and tanota, karota. Theending tana occurs only intheforms\njust quoted.\n5.Present Participle.\n^705.Theendings 5FFTantandTRandareadded tothe\nweakform oftense stem :thus, from]/gsucome act.\nsunvdnt(fern. M^lrfl sunvati),mid. H-^fM sunmmd\\ from\ntan, rT^Trr tanvdnt(fern, ri^ril tanvatfy rT^FT tanvand. From\n/>,theyare5fio|ti ojowwt?a< andtUirH apnuvand.\n6.Imperfect.\n/\"706.Thecombination ofaugmentedstemandendings\nisaccordingtotherules alreadystated: thus,\nactive. middle.\ns. d.p.s. d. p\ndsunavam dsunuva dsunuma dsunvi dsumivahi dsunumahi\ndsunos dsunutam dsunuta dsumithas dsunvatham dsunudhvam\n3-^ MHHHIH^ yyH^ MH^ri ^H^lrilH^^H-^ri\ndsunot dsunutam dsunvan dsunuta dsunvatam dsunvata\nHere, aselsewhere, the briefer forms dsunva, dsunma,\ndsunvahi, dsunmahi areallowed, andmore usual, exceptfrom\nroots with final consonant,asdhrs; which makes,forexample,\nalways ddhrsnuma etc., and also ddhrsnuvan, ddhrsnuvi, ddhrmu-\nvatham, ddhrsnuvatam, ddhrsnuvata.\n707. Strong stem-forms andJana-ending arefound only inRV., in\nakrnota, akrnotana. Augmentlessforms with accent areminvan, hinvdn, rnutd.\n"
+    },
+    {
+        "page": 256,
+        "content": "236 1X -PRESENT-SYSTEM.[708\nIrregularities ofthenuandw-classes.\n708. Less than thirtyroots form theirpresent-systemin\nthemanner setforth above, bytheaddition oftheclass-sign\nnutotheroot :theyareaks, ac'attain', talcs, dabh, cak, sagh,\nap,dac, r,kr'make', vr(ur), str, spr, rdh, trp, dhrs,i'send1\n(orin:seebelow, 716),ksi'destroy', ci,dhi,mi'prop', hi,u,\ndu, cru, su,sku,prus, dhu: andofthese, several(astaks, sagh,\ndac, u,skii)have onlyisolated forms ofthis class.\n709. The roottrp, 'enjoy',issaidbythegrammarians toretain then\nofitsclass-sign unlingualizedinthelaterlanguage where, however, forms\nofconjugationofthis class hardly occur; while intheVeda theregular\nchangeismade: thus, trpnu.\n710. The root cru, 'hear',iscontracted tocrbefore the\nclass-sign, formingcrnoand crnu asstem. Itsforms.crnvise\nand crnvire have been noted above(699).\n711. Therootdhuinthelaterlanguage shortens itsvowel,\nmakingthestem-forms dhuno anddhunu(earlier dhuno, dhunu).\n712. The so-called rooturnu, treated bythenative grammarians as\ndissyllabic andbelongingtotheroot-class(I.),isproperlyapresent-stem of\nthisclass, with anomalous contraction, from the root vr(orvar}. Inthe\nVeda,ithasnoforms which arenotregularly made according tothenu-\nclass; butintheBrahmanalanguagearefound sometimes such forms as\nurnauti, asiffrom anw-root ofcl.I.(626); andthegrammarians make for\nitaperfect, aorist, future, etc. Its2dsing. impv.act. isurnu orurnuhi;\nitsimpf. ,aurnos, aurnot;itsopt. mid., urnuvita(K.)orurnvltd(TS.).\n713. The roots oftheother division, oroftheej-class,\nareextremely few :theyare tan,man, van, san;also ksan(not\ninV.: in(TB., and very rarely later),and kr'make'(inlate\nVedic andlater);andBR.assume inofthew-class instead ofi\noftheraw-class.\n714.Theextremely common root fikr(orkar\\ 'make',\nisinthelaterlanguageinflected inthepresent-systemex-\nclusively accordingtothew-class(beingtheonlyroot of\nthat class notendingin^n}.Ithastheirregularitythat in\nthestrong form ofstem it(aswell astheclass-sign)has\nthe^ww-strengthening, and that intheweak form itis\nchangedtokur, sothat thetwoforms ofstem are^T(Tkaro\nandgT^kuru. Theclass-sign 3uisalways droppedbe-\nfore^vandqmofthe 1stdu.and pi.,and alsobefore\nv -\\\nITyoftheopt.act.Thus :\n"
+    },
+    {
+        "page": 257,
+        "content": ""
+    },
+    {
+        "page": 258,
+        "content": ""
+    },
+    {
+        "page": 259,
+        "content": "741T VI.^4-CLASS (FIRST, $Aw-CLA88). 243\nsuch asbhdvanta (whichareverycommon) are, ofcourse, properly aug-\nmentless imperfects. TheBrahmanas(especially QB.) prefer the2dsing,\nact.inasiandthe3dinat.A3dpi.inantai(vartantai, KB.) hasbeen\nnoted once. RV. hasanexample, area, ofthebriefer 1stsing. act.\n3.Present Optative.\n738.Thescheme ofoptative endingsascombined with\nthefinal ofanez-stem wasgiveninfullabove(566.\nactive. middle.\ns. d. p. s. d. p.\n1\nbhdveyambhdveva bhdvemabhdveya bhdvevahi bhdvemahi\nbhdves bhdvetam bhdveta bhdvethasbhdveyatham bhdvedhvam\n3\nbhdvet bhdvetambhdveyusbhdvetabhdveyatam bhdveran\nTheRV. hasonce the3dpi.mid. bharerata(foroneother example,\nsee752).\n4.Present Imperative^\n739.Anexampleoftheimperativeinflection is :\nactive. middle.\ns. d. p.s. d.p.\nbhdvani bhdvava bhdvama bhdvai bhdvavahai bhdvamahai\n-*N ^^\n\"S \"V \"N.\nbhdva bhdvatam bhdvata bhdvasva bhdvetham bhdvadhvam\n3WT H^riIH^ ^ ^\nbhdvatu bhdvatam bhdvantu bhdvatam bhdvetam bJidvantam\n740. Theending tana in2dpi.act. isasrare inthiswhole conjuga-\ntion asisiftana inthepresent:theV.affords only bhajatana inthea-class\n(andnahyatanainthe7/a-class:760). Theendingtatof2dsing, act., on\ntheother hand, isnotrare; theRV. hasavatat, osatat, dahatdt, bhavatdt,\nyachatat, yacatat, rdksatat, vahatdt; towhich AV. adds jinvatat, dhavatat;\nandtheBrahmanas bring other examples.\n5.Present Participle.\n741.Theendings ^rTantandqHmana areadded to\nthepresent-stem,withloss, before theformer, ofthefinal\nstem-vowel: thus,act.HcfHbhdvant(fern.^T^tft Ihdvanti}-,\nmid.HCHM bhdvamana.\n16*\n"
+    },
+    {
+        "page": 260,
+        "content": "244 IX.PRESENT-SYSTEM.[742\n6.Imperfect.\n742.Anexampleoftheimperfectinflection is:\nactive. middle.\ns. d. p. s. d.p.\ndbhavam dbhavava dbhavama dbhave dbhavavahi dbhavamahi\nTOT^ 3WFT^^W^rTEPWHH^ EW'^IH^SW^EFT\ndbhavas dbhavatam dbhavata dbhavathas dbhavetham dbhavadhvam\ndbJiavat dbhavatam dbhavan dbhavata dbhavetam dbhavanta\n743. Noforms infana aremade inthis tense from any o-class. Ex-\namplesofaugmentless forms (which arenotuncommon) are: cydvam, dvas,\nddhas, bddhat, bhdrat, cdran, ndpan; badhathas, vdrdhata, fdcanta. Thesub-\njunctively used forms of2dand3dsing. act. aremore frequent than those\nofeither ofthemote proper subjunctive persons.\nIrregularities ofthe -class.\n744.Afarlarger number ofroots form theirpresent-\nsystem accordingtothea-class thanaccordingtoanyofthe\nother classes: intheRV., theyareabout twohundred andforty\n(nearlytwo fifths ofthewhole bodyofroots);intheAV. ,about\ntwohundred(nearlythesameproportion);totellprecisely,or\nvery nearly, howmany theyareinthe later languageisnot\npossible (ofthenumber \"about athousand\", asusually stated,\nthegreater partarefictitious :see108a).Among them areno\nrootsendinginlong d except afewwhich make ana-stem\n,insome anomalous way: below, 749a.\n745.Afewverbs haveirregular vowel-changesinforming\nthepresent-stem: thus,\na.uh'notice' has^wna-strengthening (against 240): thus, ohate.\nb.krp (orfcrop), 'lament', onthecontrary,remains unchanged:thus,\nkrpate.\nc.guh, 'hide', hasprolongation instead ofguna: thus, guhati.\nd.kram, 'stride', lengthens itsvowel inthe active, but notinthe\nmiddle: thus, kramati, krdmate; klam, 'tire',issaid toform klamatietc.,\nbuthardly occurs; cam with thepreposition a,'rinse themouth', forms\nacamati.\ne.Inthe later language arefound occasional forms ofthis class from\nmrj, 'wipe'; andtheyshow thesame vrddhi(insteadofguna} which belongs\ntotherootinitsmore properinflection(627): thus, marjasva.\nf.Thegrammarians give anumber ofroots inurv,which they declare\ntolengthen theuinthepresent-stem. Only three arefound in(quite\n"
+    },
+    {
+        "page": 261,
+        "content": "751]VI.A-CLASS (FIRST, b/lU-CLASS). 245\nlimited) use, and they show noforms anywhere with short u.Allappear\ntobeofsecondary formation from roots inrorar.The rootmurch or\nmurch, 'coagulate',haslikewise onlyuinquotable forms.\ng.Theonomatopoetic rootathiv, 'spew',iswritten bythegrammarians\nasrthiv. anddeclared tolengthen itsvowel inthepresent-system.\n746. The rootsdanc, 'bite', ranj, 'color', sanj, 'hang',\nsvanj, ''embrace'., ofwhich thenasal isinotherpartsofthe\nconjugation not constant, lose itinthepresent-system:thus,\nddcati etc.; sanjforms bothsajati andsajjati (probablyforsajyati,\norforsasjatifromsasajati];math ormanth hasmathati later.\nIngeneral,asthepresentofthis class isastrengthening form-\nation, aroot thathassuch anasal anywhere has ithere also.\n747. The rootsgam, 'go', andyam, 'furnish', make the\npresent-stems gacha andydcha: thus, gdchamietc.: see608.\n748. The root sad, 'sit', forms sida(conjecturedtobe\ncontracted from stsd forsisad):thus, sidami etc.\n749. Transfers tothis classfrom other classes arenotrare,\nashasbeen already pointed out above, both throughoutthe\npresent-systemand inoccasional forms. Themost noticeable\ncases arethefollowing:\na.The roots ina,stha, 'stand', pa, 'drink', andghra,\n'smell', form thepresent-stemstistha(tisthami etc.), pibaor(later)\npiva (pibami etc.),andjighra (jighrami etc.); and, intheVeda,\nda, 'give', and dha, 'place', form sometimes ddda and dddha,\nhan, 'slay', forms sometimesjighna,andhi,'impel',formsjighya\nallthesebytransfer from thereduplicatingclass :see671 4.\nb.Secondaryroot-forms like inv, jinv, pinv, fromsimpler\nroots oftheraw-class, areeither found alongsidetheiroriginals,\norhave crowded these outofuse: see716.\n750.Ontheother hand, therootdham ordhma, 'blow',\nforms itspresent-stemfrom themoreoriginal form oftheroot :\nthus, dhdmati etc.\nVII.Accented -class (sixth, ?W-class).\n'751.Thepresent-stemofthis class hastheaccent on\ntheclass-sign ^a,andtherootremains unstrengthened.In\nitswhole inflection,itfollows socloselythemodel ofthe\npreceding class,that togivetheparadigminfullwillbe\nunnecessary (onlyforthesubjunctive,alltheforms found\ntooccur willbeinstanced).\n"
+    },
+    {
+        "page": 262,
+        "content": "246IX.PRESENT-SYSTEM.[752\n752.Exampleofinflection :root fifST vi$, 'enter'; stem,\nvgd:\n1.Present Indicative.\nactive. middle.\ns. d. p. s. d. p.\nvicGmi victlvas vicamas vice vicavahe vicamahe\netc. etc. etc. etc. etc. etc.\n2.Present Subjunctive.\nivifdni vifdva vifdma vi$ai vi$dvahai vifdmahai\n{vifdsi.i, (vi?dse\n, vi(atha {,_ vicaithe\nvifas \\vi?a8ai\n(vicdtij_ ._i fvifate,\n3: , vi^atas vi$an < ,_ vifatte vicantai\n[vifat [vifatai\nAsingle exampleofthebriefer 1stsing. act. ismrksd. Theonlyforms\ninaithe and aiteareprnafthe andyuvafte.\n3.Present Optative.\nviffyamviceva vicemaviceyavicevahiviqemahi\netc. etc. etc. etc. etc. etc.\nTheRV. hastheending tana once intiretana 2dpi.act.,andrata in\njuserata 3dpi.mid.\n4.Present Imperative.\nThe firstpersons having beengiven above assubjunc-\ntives, thesecond areadded here:\n2f%5FfelrFT^felrT N^IHN^I&IIH^ feiy^mca vicdtam vicdta vicAsva vicetham vicddhvam\netc. etc. etc. etc. etc. etc.\n,Theendingtatisfound inRV.andAV. inmrdatat, vrhatdt, suvatat;\notherexamples arenotinfrequent intheBrahmana language:thus, khidatat,\nsrjatdt.\n5.Present Participle.\nThe activeparticipleisfek\\i\\^vi$dntfthemiddle is\ni^^WM vigdmana.\nThefeminine oftheactiveparticipleisusually made from thestrong\nstem-form:thus, vifdnti; butsometimes from theweak: thus, sincantl and\nsincati(RV. andAV.), tuddntl andtudati(AV.): seeabove, 449b.\n"
+    },
+    {
+        "page": 263,
+        "content": "756]VII.ACCENTED G-CLASS SIXTH, tud-CLA.ss). 247\n6.Imperfect.\ni^foSR^^fciN tjfoiiH ^rf^tl 4iief!MNf^ ytomJ^\ndvicam dvicava dvicama dvice dvicavahi dvicamahi\netc. etc. etc. etc. etc. etc.\nExamplesofaugmentless forms accented aresrjds, srjdt, tirdnta.\nThe a-aorist (846 ff.)isingeneral theequivalent,asregardsitsform,\nofanimperfectofthis class.\nIrregularities ofthea-class.\n753. Itisimpossibletodeterminecloselythelimits of\nthis class, partly because oftheoccurrence offorms unaccented,\norinunaccentuated texts, which might belongeither toitor\ntothepreceding class, partly because itsmodes andimperfect\nareaccordant inform with those ofthea-aorist(below, chap.\nXI.),and theirseparationisnotalways practicable, andpartly\nforother reasons. With considerable confidence maybereckoned\nasbelongingtoitabout seventy roots: namely, ksi,yu'join',\nru'roar', su(orsu)'stirup', dhu, hu,kr'strew', gr'swallow',\ntr,rikhorlikh, sic, ich, vij, khid, vid'find', vidh, ksip, lip, riph,\ndie, pic, ric, vie, is, tvis,mis,muc, uch, ubj, tuj, ruj,khud, tud,\nnud, rud, lup, ubh, cubh, gur, jur, tur, bhur, sphur, jus, prus,\nrus, cus, uks, vrcc(orvracc], rch,prch (orprach], rnj, srj, bhrjj\n(orhhrajf),mrd.prn, mrn, krt 'cut', crt, rd, trp, vnrc, sprc,rs\n'push',krs'plough', mrks, vrs, drh, vrhorbrh.Some even of\nthese have eitheronlyisolated orveryrareoccurrences ofa-forms.\nThe roots ich, uch,andrcharereckoned assubstitutes inthe\npresent-systemfor is'wish', vas 'shine', and r'goto'(608).\nPrnandmrnhavebeen noticed above(731)assecondaryroots\nfrompresent-stemsofthena-class(V.).\n754. Certainpeculiaritiesofthis \"body ofroots arevery noticeable: it\ncontains only one ortwo roots with long vowels, andnone with longinterior\nvowels;veryfewwith finalvowels;andnone with aasradical vowel, except\nasthisforms acombination withr,which isthen reduced inthepresent-\nsystem,asintheweak forms generally,tororsome oftheusual sub-\nstitutes ofr.\n755.The roots inianduanduchange those vowels into\niyanduvbefore theclass-sign:thus, ksiydti,suvdti(svainstead\nofsuva occurs inAV.; and theBrahmanas have forms inksya\nfromksi}.\n756. The three roots inrform thepresent-stems kird,\ngird, tird,andtheyaresometimes written askiretc.;andgur,\njur,turarereally onlyvarieties ofgr,jr, tr;andbhurand\nsphurareevidentlyrelated with other arorrroot-forms.\n"
+    },
+    {
+        "page": 264,
+        "content": "248 IX.PRESENT-SYSTEM.[757\n757.Two other roots which areused only inmiddleforms, and in\ncombination with theprepositiona(sometimesfurthercombined), make the\npresent-stems a-driya and a-priyd, and arereckoned asrorarroots:dr,\n'regard', andpr,'bebusy' (neitherisfound inV.).Itisaquestion whether\nthey aremore properlyreckoned tothis class oraspassives ;andthesame\nquestionarises astothestems mriyd anddhriyd, from theroots mr, 'die',\nanddhr,'hold': seebelow, 773.\n758. Althoughthepresent-stemofthis class shows in\ngeneralaweak form oftheroot, there arenevertheless anumber\nofroots belongingtoitwhich arestrengthened byapenultimate\nnasal. Thus, thestem muncd ismade fromymuc, 'release';\nsincd froml/sic, 'sprinkle';vindd fromyvid 'find'; krntd from\nykrt 'cut'; pined fromypig, 'adorn'; trmpd from}/trp, 'enjoy':\nlumpd fromyiup, 'break'; limpd fromyiip, 'smear'; andocca-\nsional forms ofthesame character aremetwith from afew\nothers, astundd fromytud, 'thrust'; umbha fromyttbh, 'hold';\nbrn/id fromybr/i 'strengthen'; drnhd(beside drnhd] fromydrh,\n'make firm'; cumbhd(beside qumbha] fromyculh,'shine'. TS.\nhas crnthati fromy$rath (insteadofcrathriati).\nVIII. Fa-class (fourth, <%'-class).\n(759.Thepresentstem ofthis class adds TJyatothe\naccented butunstrengthenedroot. Itsinflection isalsopre-\nciselylike that ofthe#-class\n7andmaybepresentedinthe\nsame abbreviated form asthat ofthea-class.\n^760. Example ofinflection: root^na^^'bind';\nstem ^T^Tndhya.\n1.Present Indicative.\nactive. middle.\nd. . s. d.\nndhyami ndhyavas ndhyamas ndhye ndhyavahe ndhyamahe\netc. etc. etc. etc. etc. etc.\n2.Present Subjunctive.\n1ndhydni ndhyama ndhyai ndhyavahai ndhydmahai\nfndhyasi ,,_,, _.\n2{ ndhyasai nahyadhvai\n[ndhyas\n3-1 , ndhyatas ndhyan ndhydtdi ndhydntdi\n\\ncthyat\nA3dpi.mid. inantai (jdyantdi) occurs once inTS.\n"
+    },
+    {
+        "page": 265,
+        "content": "76!] VHI. Ytf-CLASS (FOURTH. rfw/'-CLASS . 249\n3.Present Optative.\nndhyeyam ndhyeva ndhyema ndhyeya ndhyevahi ndhyemahi\netc - etc - etc. etc. etc. etc.\n4.Present Imperative.\n2\n^\nndhya ndhyatam ndhyata ndhyasva ndhyetkdm ndhyadhvam\netc. etc. etc. etc. etc. etc.\nOftheending tana,RV. hasoneexample, nahyatana; theendingtat\nisfound inasyatdt, chyatdt.\n5.Present Participle.\nThe activeparticipleis*1$lvX ndhyant (fern,^rft nd-\nhyantl)', themiddle is^RH ndhyamana.\n6.Imperfect\ndnahyam dmhyava dnahyama dnahye dnahydvahi dnahyamahi\netc. etc. etc. etc. etc. etc.\nExamples ofaugmentless forms showing theaccentbelonging tothe\npresent-system aregdyat, pdfyat, pdfyan, jdyathds.\nIrregularities oftheya-el&sa.\n761. The roots oftheya-class aremore than ahundred\ninnumber. Theymaybegrouped asfollows :\na.Roots signifying astate offeeling, oracondition ofmind orbody.\nThese arenearly half thewhole class. They are(alphabetically) asfollows :\nwe,'bepleased'; kup, 'beangry'; fcrp, 'belean'; klam, 'beweary'; krudh,\n'beangry'; fcsam,'bepatient'; ksudh,'behungry'; ksubh, 'beagitated1\n;grdh,\n'begreedy'; jas, 'beworn out1\n;tarn,'beexhausted1\n;tus,'besatisfied'; trp,\n'bepleased'; trs,'bethirsty'; tras, 'bealarmed'; dam, 'besubmissive1\n;dus,\n'bespoiled1\n;drp, 'becrazed'; druh, 'behostile'; pus,'beingood condition';\nbudh, 'beawake'; bhram,'beunsteady'; mad, 'beexcited1\n;man, 'beminded';\nmuh, 'beconfused'; mrs, 'beforgetful'; mrit,'bein illcondition'; med,\n'befat'; yas,'beardent'; yudh, 'behostile'; raj, 'becolored'; radh, 'besub-\nject'; ran,'behappy'; rddh,'besuccessful'; ris,'behurt'; rup,'beinpain';\nlubh,'belustful'; pam,'bequiet'; pwc, 'beinpain'; fus, 'bedry'; pram,\n'beweary'; ftar,'begratified'; hrs,'beexcited1\n;andwemay perhaps add\ndas,'bedeficient', andnap,'bemissing1\n.Some ofthese areofonly early-\nuse,some only oflater;andsome have only sporadic forms ofthisclass,\nmadeperhaps under theinfluence oftheanalogy oftheothers.\n"
+    },
+    {
+        "page": 266,
+        "content": "250 IX -PRESENT-SYSTEM.[761\nb.Roots which have amore orless distinctly passive sense, andwhich\nareinpartevident and inpartpresumable transfers from thepassiveor\ni/a-class, withchange ofaccent, andsometimes alsowithassumptionofactive\nendings.Itisnotpossibletodraw preciselythelimits ofthe division, or\ndetermine inallcaseswherepassiveform andmeaning pass into intransitive;\nbutthere areanumber ofclearcases, where intheolder language theaccent\nwavers andchanges, andtheothers aretobejudged bytheir analogy. Thus,\nmucforms mucyate once ortwice, beside usualmucydte,inRV.andAV.;\nandintheBrahmanas theformer istheregular accent :and similar changes\narefound inother verbs :thus, jiorjya,ksi'destroy', ha'leave', pac,dr\n'burst', chid, bhid. Cases closely analogous with these aremiyateetc.from\nymiorml, 'lessen'; ricyateetc.fromyric, 'leave'; viyateetc.fromj/w,\n'impregnate'; fiyante from y^ya, 'coagulate'; tfsyate etc.fromYfis, 'leave';\ndrhyasva fromydrhh,'makefirm'; puryate etc.fromj/pr, 'fill';andItipyate,\ntdpyate, tlryate, klfyyate, fdhyate, mayberanked along with them. Active\nforms areearlymade sporadically from some ofthesethus, drhya (RV.),\nksiyati andpuryati (TA.); anddlryati, klipyati, and other likecases,are\nfound later. TheAV. hasjlryati, 'growsold'(lateralsojiryate)', andQB.\nhasaprusyat, 'was sprinkled'. And from theearliest period jdyate etc.,'is\nborn',iseither alteredpassiveororiginal ya-lormation fromy'ja, servingas\ncomplementtoyjan, 'givebirth'.\nc.Asmall body ofroots areeither transitive, ornotintransitive ina\nway that clearly connects them with either oftheabove classes :thus,as\n'throw';is'send'; fra, 'save'; nah, 'bind'; pap, 'see'; vyadh, 'split'; slv, 'sew';\ndiu, 'play'; tur, 'overcome' (RV., once); tan,'thunder' (RV., once), r/,'press\non'; nri, 'dance'; pad, 'go'; t>ap, 'bleat'; dl,'hover'; n,'flow'; srlv, 'fail';\nylifj 'hang on'; 6/irap, 'fall'; sidh, 'succeed'; dip,'shine' (andperhaps das\nandnaparebetter classed here thanundera).\nd.Abody ofroots, ofvarious meaning, and ofsomewhat questionable\ncharacter andrelations, which arebythenative grammarians reckoned asending\nwith diphthongs:thus,\n1.Roots reckoned asending inaiandbelonging tothea-class :thus,\ngdyati fromygai. Asthese show abundantly (andinmost cases exclusively)\na-forms outside thepresent-system,there seems noreason whythey should\nnotberegarded asa-roots oftheya-c.la.ss. They are: ga'sing'; .gla,'be\nwearied ordisgusted': dhya, 'think'; pya, 'swell'; mia, 'wither'; ra, 'bark';\nva, 'droop'; fj/a, 'coagulate'; pra, 'cook'; stya,'becoagulated'; and, inone\nortwosporadic forms, fca, 'burn'; da'cleanse'; sta,'behidden'; spha, 'be\nfat'. Tra, 'save', was given intheprecedingdivision. Many ofthese are\nevident extensions ofsimpler roots with added a.With them may be\nmentionedtay, 'extend'(compare pass, tayate fromytan: 772), andc%,\n'beshyoranxious'(which connects itself with uses ofyd}.\n2.Roots reckoned asending ineandbelongingtothea-class: thus,\ndhdyati fromydhe. These, too,havea-forms, andsometimes t-forms, outside\nthepresent-system, andmust beregarded asa-roots,either withaweakened\ntoabefore the class-sign ofthisclass, orwith aweakened toIor iand\n"
+    },
+    {
+        "page": 267,
+        "content": "767]VIII. Ya-CLAss (FOURTH, div-CLAss). 251\ninflected accordingtothe a-class. They are :dha, 'suck'; va,'weave'; vya,\n'hide'; hva,'call'(oneoftheforms ofy7w);and alateexampleortwoare\nfound from ma, 'exchange'. With themmay bementioned day, 'share,\nsympathize, pity'; vyay,'be'wasted' (denom. ofvyaya?}-, cay,'visit with\nretribution' (probablyaform ofci),\n3.Roots artificially marked with afinal o(108c) andreckoned tothis\nclass, theradical vowel being declared dropped before theclass-sign:thus,\ndydti from do.They have, asshowing anaccented yd,norealrighttobe\nclassed here atall,butseem more accordant information with thepresent-\nstems svaandksya,noticed under thepreceding class(755). Outside these\npresent-systems, theyshow aandi-forms; andtheyaintheonlyRV. oc-\ncurrence, andinmost oftheAV. occurrences,isresolved into ia which\ninthetrue class-sign yaisthecase only invery rareandpurely sporadic\ninstances. They might, then, perhapsbebestviewed asa-roots withaweak-\nened toi,andinflected bythea-class, butwithout theusual conversion of\nitoiy(755). They are:da'cut', da'bind'; ca,'sharpen'; sa,'bind'; cha,\n'cut off'.\n762. Thej/a-classistheonlyonethus fardescribed which shows any\ntendency toward arestriction toacertain variety ofmeaning. Inthisten-\ndency,aswell asintheform ofitssign,itappearsrelated with theclass\nofdistinctly defined meaning which isnext tobetaken up thepassive,\nwitht/a-sign. Though very farfrombeing aswidely used asthelatter beside\nother present-systems,itisinnovery smallnumber ofcases anintransitive\nconjugation bytheside ofatransitive ofsome other class.\n763. The roots ofthis class endinginamlengthentheir\nvowel informingthepresent-stem:theyare tarn, dam, bhram\n(butbhramyatalsooccurs), cam, cram, klam(hardlyfound in\nuse),andksam(butksamyate also):forexample, camyati, crhmyati.\n764. Therootmadhasthesame lengthening: thus, madyati.\n765. The roots inw namely, div, siv, sriv(orcriv)\narewritten bythegrammarianswith iv,andasimilar lengthening\ninthepresent-systemisprescribedforthem.\nThey appeartobeproperly dmetc., since their vocalized final inother\nforms isalways u:div isbythisprovedtohave nothingtodowith the\nassumed root div, 'shine', which changestodyu(361 d):comparealsothe\ndesiderative stem jujyusa fromyjlv (1028h).\n766. From theroots jrand tr(alsowritten asjurand tirortur]come\nthestems jirya andtirya, andj&rya andtUrya (thelasttwoonly inRV.);\nfrom prcomes pUrya.\n767. The rootvyadhisabbreviated tovidh: thus, vfdhyati. Andany\nrootwhich inother forms hasapenultimatenasal loses ithere :thus, drhya\nfrom drnh ordrh; bhrdfya (also bhfcya) from bhrahf orbhraf; rajya from\nran? orraj.\n"
+    },
+    {
+        "page": 268,
+        "content": "252 IX.PRESENT-SYSTEM.[768\nIX.Accented //</-class: Passive conjugation.\n768.Acertain form ofpresent-stem,inflected with\nmiddle endings,isusedonlywith apassive meaning, and\nisformed from allroots forwhich there isoccasion tomake\napassive conjugation.Itssignisanaccented 7Jydadded\ntotheroot: thus, t^*yhanydfromj/^T han, ETTQJ apyd\nfromj/ETPT ap }JT^TgrhydfromJ/JT^n grh (orgrah}\\ and\n\"X C.(.\nsoon,without anyreference totheclassaccordingtowhich\ntheactive andmiddle forms aremade.\n\\769. Theform oftheroot towhich thepassive-signis\nadded is(sincetheaccent isonthesign)theweak one :thus,\napenultimatenasal isdropped, andanyabbreviation which is\nmade intheweak forms oftheperfect (794),inthe aorist\noptative (922 b),orbefore to,ofthepassive participle (954),is\nmade also inthepassive present-system:thus, ajyafromj/aw/,\nbadhyd fromybandh, ucyafromj/wzc, ijydfromyyaj.\n770.Ontheother hand, afinal vowel ofaroot isin\ngeneralliable tothesame changesasinotherpartsofthe\nverbal system where itisfollowed byy:thus.,\na. iandufinal arelengthened: thus,miya fromymi; suydfromysu;\nb.afinal isusually changed toi:thus, dlyd fromyda ;hiyd from\nyha:butjnayd fromyjna, khyayd fromykhyd;\nC.rfinal isingeneral changedtori:thus, kriyd fromykr-but if\npreceded bytwoconsonants (and also,itisclaimed, intherootr),ithas\ninstead thepwna-strengthening: thus, smart/a fromj/smr, staryd fromytr;\nandinthose roots which show achange ofrtoirandur(so-called f-\nverbs :see242),thatchangeismade herealso, andthevowel islengthened:\nthus, firyd fromy$r;puryd fromypr.\n\\771.The inflection ofthepassive-stemispreciselylike\nthat oftheother a-stems\n;itdiffersonlyinaccent from\nthat ofthe class lastgiven.Itmay beherepresented,\ntherefore,inthesame abbreviated form:\nExample ofinflection: root ^ikr,'make'; passive-\nstem T5FTET kriyd:\n1.Present Indicative.\nd. p.\nkriye kriyavahe kriy&mahe\netc. etc. etc.\n"
+    },
+    {
+        "page": 269,
+        "content": "773]IX.ACCENTEDya-CLASS (PASSIVE). 253\n2.Present Subjunctive.\nTheforms noticed asoccurringintheolderlanguage are\nalone here instanced :\na.p.\n1kriyai kriydmahai\n2kriyddhvai\ntkriydte\n3\n\\kriydtai\nThe3dpi.ending antai isfound once(ucyantai, K.).\n3.Present Optative.\nkriyeya kriyevahi kriyemahi\netc. etc. etc.\nNoforms ofthepassive optative chance tooccur inRV. orAV.; they\narefound, however, intheBrahmanas.\n4.Present Imperative.\nkriydsva hriyetham kriyddhvam\netc. etc. etc.\n5.Present Participle.\nThis ismade with thesuffix JTRmana: thus,\nkriydmana.\nInuse,thisparticipleiswell distinguished from theotherpassive par-\nticiple hyitsdistinctively present meaning: thus, krtd, 'done'; butkriydmana,\n'inprocess ofdoing',or'being done'.\n6.Imperfect.\ndkriye dkriyavahi dkriyamahi\netc. etc. etc.\nThepassive-signisnever resolved into iaintheVeda.\n772. The roots tanandkhanusually form theirpassives\nfromparallelroots ina:thus, taydte, khaydte (butalsotanydte,\nkhanydte).Thecorresponding form to|//aw, namely jhyate (above,\n761b),isapparentlyatransfer totheprecedingclass.\n773.Bytheir form, mriydie,'hedies', anddhriydte,'he\nmaintains himself,issteadfast', arepassivesfrom theroots mr,\n'die', and dAr, 'hold'; althoughneither isused inaproper\npassive sense, andmr isnottransitive exceptinthederivative\nformmrn(above, 731).With them are tobecomparedthe\nstemsa-driydanda-priyd (above, 757),which may possiblybe\n"
+    },
+    {
+        "page": 270,
+        "content": "254 IX.PRESENT-SYSTEM.[773\npeculiar adaptationsofmeaningofpassivesfrom therootspr,\n'fill', and dr, 'scatter .\n774. Instances are occasionally found inthe later languageofan\napparent assumptionofactive instead ofmiddle endings bypassive persons\nofthepresent-system. Probably, however, these arerather toberegarded\nasexamplesoftransfer tothet/a-class, such aswere considered above (761 b).\n775.Aswaspointedoutabove(607),theformation and\ninflection ofstems indya (thetenth orcwr-class oftheHindu\ngrammarians)willbetreated under thehead ofsecondary con-\njugation (chap. XIV.), along with theintensive anddesiderative\nformations, because, inallalike, thestem isnotapresent-stem\nmerely,buthasbeen extended also into othertense-systems.\nUses ofthePresent andImperfect.\n776. The uses ofthemode-forms ofthepresent-system\nhavebeen already brieflytreated inthepreceding chapter (572 ff.).\nThetense-uses ofthetwo indicative tenses, present andimper-\nfect,callhere foronlyaword ortwo ofexplanation.\n777. Thepresent has, besides itsstrictly present use, the\nsame side-uses which belongingeneraltothetense :namely,\ntheexpressionofhabitual action, offutureaction, and ofpast\ninlivelynarration.\na.Examplesoffuture meaning are:abruvan hrstd gachdmo vayam apt\n(MBh.), 'they saidwithgladness, \"wewillgotoo'\"; agnir atmabhavam prdddd\nyatra vdnchati ndisadhah (MBh.), 'Agni gave hisownpresence wherever the\nNishadhan should desire'.\nb.Examples ofpastmeaning are :tittard surddharah putra dsldddnuh\n$aye sahdvatsd nddhenuh (RV.), 'themother was over, thesonunder; there\nDanulies,like acowwith hercalf; prahasanti catdm kecidabhyasuyanti\ncd'pare akurvata daydmkecit (MBh.), 'some ridicule her, some revile her,\nsome pitied her'; tatoyasya vacandt tatrd 'valambitds tarnsarve tiraskurvanti\n(H.), 'thereupon they all fall toreproaching himbywhose advice theyhad\nalighted there'.\n778. Inconnection with certainparticles,thepresenthas\nrather moredefinitelythevalue ofapasttense. Thus:\na.Withpwrcf, 'formerly': thus, saptarsin uhasma vdipurd rkshdity\ndcaksate(QB.),'the sevensages, namely, areofoldcalled thebears'; tan-\nmdtram apicenmahyam nadaddti purd bhavdn (MBh.),'ifyouhave never\nbefore givenmeeven anatom\".\nb.With theasseverativeparticle sma.' thus, frdmena hasmavdttad\ndevdjayanti yddesdmjdyyam dsdrsayaf ca(QB.), 'for,intruth, both gods\nandsages werewont towinbypenance what was tobewon'; dvistah kalind\ndyute jlyate sma nalas tadd(MBh.), 'thenNala, being possessed byKali,\nwasbeaten inplay'.\n"
+    },
+    {
+        "page": 271,
+        "content": "781]USES OFTHEPRESENT ANDIMPERFECT. 255\nNoexample ofthis construction isfound ineither RV. orAV., or\nelsewhere inthemetricalparts oftheVeda. IntheBrahmanas, onlyhabitual\naction isexpressed by it(Delbruck). In allperiods ofthelanguage, the\nuseofsmawith averb aspure asseverativeparticle, with noeffect onthe\ntense-meaning,isverycommon; and theexampleslater arehardly tobe\ndistinguished from thepresent oflively narration ofwhich thewhole\nconstruction isdoubtless aform.\n779.Theimperfect hasremained unchangedinvaluethrough\nthewholehistory ofthelanguage:itisthetense ofnarration ;\nitexpresses simple past time, withoutanyotherimplication.\nCompare what issaid later(endofchap. X.andchap. XI.)astothe\nvalue oftheotherpast tenses, theperfect and aorist.\nCHAPTER X.\nTHEPERFECT-SYSTEM.\nL780.THEperfect-systeminthelaterlanguage,ashas\nbeen seen above(535),consistsonlyofanindicative tense\nandaparticipleboth ofthem inthetwovoices, active and\nmiddle.\nIntheoldest language,theperfecthas also itsmodes and\nitsaugment-preterit,orpluperfect,orisnot less full inits\napparatusofforms than isthepresent-system.\n^781.The formation oftheperfectisessentiallyalike\ninallverbs, differences among tbem beingofonlysubord-\ninate consequence,orhavingthecharacter ofirregularities.\nThe characteristics oftheformation arethese :\na.astemmadebyreduplicationoftheroot;\nb.adistinction betweenstronger andweaker forms of\nstem,theformerbeing used(asinpresentsoftheFirst\nconjugation)inthesingular active, the latter inallother\npersons;\nc.endingsinsomerespects peculiar,unlike those of\nthepresent;\nd.thefrequent use, especiallyinthe later language,\nofaunion-vowel *between stemandendings.\n"
+    },
+    {
+        "page": 272,
+        "content": "256 x -PERFECT-SYSTEM.[732\n\\782.Reduplication.Inrootsbeginningwith acon-\nsonant, thereduplication which forms theperfect-stemis\nofthesame character with thatwhich forms thepresent-\nstem ofthereduplicating conjugation-class (II.:see643)\nhutwith thisexception,that radical 5[aand 5TT ftand\nffr(or3^|\" ar)haveonly5fa,andnever^i,asvowel of\nthereduplicating syllable:thus, fromy^pr, 'fill', comes\nthepresent-stem fqqpip?, buttheperfect-stem qqpapr ;\nfromj/JTT, 'measure', comes thepresent-stem MHImima,\nbuttheperfect-stem r^fTmama; and soon.\nIrregularities ofroots with initial consonants willbegiven below, 784.\n^783.For rootsbeginningwith avowel, therules of\nreduplicationarethese :\na.Aroot with initial fabefore asinglefinal con-\nsonantrepeatsthe fa,which then fuses with theradical\nvowel to51Ta(throughoutthewholeinflection): thus,\nadfromy^ ad;andinlikemannerENs^a/,5TRan,\nas,ETFJah.\nThe root TRrforms likewise throughout Eff^\"ar(asif\nfromET^\" ar].\nb.Arootwith^ior3ubefore asinglefinal conso-\nnant follows thesameanalogy, exceptinthestrongforms\n(sing, act.), where thevowel oftheradicalsyllablehas\nguna, becoming ^eorEfTo;beforethis,thereduplicating\nvowel maintains itsindependent form, and isseparatedfrom\ntheradicalsyllable byitsownsemivowel: thus, fromy'^T\n35comes^Tisinweak forms, buttmiyesinstrong;from\n1/3*1 uc,inlikemanner, come^Tucand3^Trf uvoc.\n*N -\\ -V\nThe root5i,asingle vowel,fallsunder this rule,and\nforms^1lyand^ETiye.\nc.Roots whichbeginwith vowels longbynature orby\nposition donotingeneral make aperfect-system,butuse\ninstead aperiphrastic formation,inwhich theperfecttense\n"
+    },
+    {
+        "page": 273,
+        "content": "78' REDUPLICATION. 257\nofanauxiliaryverb isadded totheaccusative ofaverbal\nnoun(seebelow, chap. XV.: 1070ff.).\nTothis rule, however, yap (probably originally ap:1087fj\nconstitutes anexception, making theconstantperfect-stem ap\n(asiffrom ap:above.,a).\nForthepeculiar reduplication an,belonging tocertain roots with initial\nvowels, seebelow, 788.\n784.Anumber ofroots beginning with vaandending\nwith asingle consonant, which invarious oftheir verbal forms\nandderivatives abbreviate thevatou,do italso intheperfect,\nand aretreated like roots with initial u(above, 783b),except\nthattheyretain the fullform ofroot inthestrong persons of\nthesingularactive. Thus, fromyvaccome ucanduvac;from\n\\/vascome usanduvas;and soon.\nThe roots showingthisabbreviation arevac, vad, vap, vac,\ncas, vah;and va,'weave'(so-calledve:761d,2),issaid to\nfollow thesame rule.\nAsinglerootbeginningwithya,namely yaj,f\noffer', has\nthesame contraction, formingthestemsiyajandIf.\n785.Anumber ofroots having yaafter afirst initial con-\nsonant take t(fromthey]instead ofainthereduplicating\nsyllable:thus, fromyvyaccomesvivyac ;fromYpyacomespipya.\nThese roots arevyac, vyath, vyadh, vya, jya,pya; and,intheVeda,\nalso tyaj, with cyuanddyu, which have theroot-vowel u.\nAsingleroot with vaistreated inthesameway:namely\nsvap, which formssusvap.\nThese roots are forthemost part abbreviated intheweak forms :see\nbelow, 794.\n786.Aconsiderable number ofroots have intheVeda a\nlongvowel intheir reduplication.\nThus, ofroots reduplicatingwitha:kan, kip, gr'wake', grdh, tan, trp,\ntrs,dhr, dhrs, nam, mdh, mrj, mrp, ran, radh, rabh, vale, van, vat,vas\n'attack', vrj, vrt, vrdli, pad, sah, skambh. Some ofthese occur only in\nisolated cases;somehave alsoforms with short vowel. Most areVedic only ;but\ndadhdra iscommon also intheBrahmana language, and iseven found later.\nForjagr,see1020 below.\nOfroots reduplicatingwith I:theso-called roots (676)dldhl and dldl,\nwhich make theperfectfrom thesame stem with thepresent:thus, dldetha,\ndiddya ;dldhima, didhyus (also didhiyus, dldiyus). Butplplhaspipye, pipyus,\netc., with short i.InAV. occurs once jlhida.\nOfroots reduplicatingwithu:juand fit(or pt>5).\n787.Afew roots beginningwith the(derivative: 42) palatalmutes\nand aspiration show areversion tothemore original gutturalintheradical\nWhitney, Grammar. 17\n"
+    },
+    {
+        "page": 274,
+        "content": "258 x -PERFECT-SYSTEM.[737\nsyllable after thereduplication: thus,ydforms ciki; }/citformscikit; yji\nforms jigi; yhiforms jighi; yhanforms jaghan (andthesame reversions\nappearinother reduplicated forms oftheseroots). Aroot da, 'protect',is\nsaidbythegrammarianstoformdigi: butneither rootnorperfectisquotable.\n788.Asmallnumber ofroots with initial aorr(ar)show\ntheanomalous reduplication anintheperfect.\nThus,intheVeda:\nyanjoraj,which forms thepres. andkti(cl. HI.), hastheperfect\nanajeetc. (with anaja andanajyat);\n]/ap,'attain' (from which comes once anacamahai], hastheweak forms\nanapma etc. (with opt. anapt/am), andthestrong forms ananpa andanapa\nalong with theregular apa etc.;\nyrdh (from which comes once rnddhat) hasanrdhe;\ny<rcorarchasanrctis andanrce-\nyarhhas(inTS.) anrhus;\nanaha (RV., once) hasbeen referred toaroot ah,elsewhere unknown,\nandexplainedasofthisformation;butwith altogetherdoubtfulpropriety.\nThe later grammar, then,setsuptherule that roots beginning with a\nandending withmore than oneconsonant haveanastheir regular redupli-\ncation;andsuch perfects aretaught from roots likeafcs, ar;,andancorac;\nbuttheonly other quotable forms appear tobeanarchat (MBh.) andanarsat\n(TA.); which areaccordingly reckoned as\"pluperfects\".\n789. One ortwo individual cases ofirregularity arethe\nfollowing:\na.Theextremely common root bhu, 'be', hastheanoma-\nlousreduplication ba,formingthestem babhu; and, intheVeda,\nysuforms inlikemanner sasu.\nb.The rootbhr, 'bear', hasintheVeda theanomalous reduplication\nja(asalsoinintensive: 1002b): butRV. hasonce alsotheregularbabhre.\nc.The rootsthiv, 'spew', forms either tisthiv (B.etal.)ortisthlv.\nd.Vivakvan (RV., once)isdoubtlessparticipleofyvac,with irregular\nredublication(asinthepresent, 660).\n790. Absence ofreduplicationismetwith inthefollow-\ningcases :\na.The root vid'know' has,from theearliestperiodtothe\nlatest, aperfect withoutreduplication,butotherwiseregularly\nmade and inflected: thus, veda, vettha, etc., pplevidv&ns. It\nhas themeaning ofapresent. The root vid 'find' forms the\nregular viveda.\nb.Afew other apparently perfect forms lacking areduplicationare\nfound inRV.: they aretaksathus, yamdtus, skambhdthus andskambhus, nindima\n(forninidima?), dhise and dhire(?ydha),and vidre andarhireC?see613).\nAndAV. hascetatus. Theparticipial words dafvdns, midhvdns,sahvahs are\ncommon intheoldest language.\n"
+    },
+    {
+        "page": 275,
+        "content": "793\"| STRONG ANDWEAK STEM-FORMS. 259\nc.One ortwosporadic cases have been quoted from the later language:\nnamely, fahsus andpansire (MBh.).\n791. Forananomalous case ortwo ofreduplicated preposition, see\nbelow, 1087f.\n792.Strong andweak stem-forms. Inthethree\npersonsofthesingular active, theroot-syllableisaccented,\nand exhibitsusuallyastronger form than intherestofthe\ntense-inflection. The difference iseffectedpartly bystrength-\neningtheroot inthethreepersons referred to,partly by\nweakeningitintheothers, partly bydoingboth.\n/793.Asregardsthestrengthening:\na.Afinalvowel takes either thegunaorvrddhichange\nin1stsing, act.,gunain2d,andvrddhi in3d:thus, from\nytf bhl,1stfsftbibhe orf^Rbibhai ;2di^Rbibhe; 3d\nbibhai; from y%\\kr,1st^Fu|\"cakdr or*Jtil^ cakar, 2d\ncakdr, 3d^37T|\"cakdr.\nButtheuofybhuremains unchanged, andadds vbefore\navowel-ending:thus, babhuva etc.\nto.Medial %abefore asinglefinalconsonant follows\ntheanalogyofafinalvowel, and islengthenedorvriddhied\ninthe3d.sing.,andoptionallyinthefirst: thus, from\ntap,1stcfflWatdport\\mj,atap,2dt&V{tatdp,3d\ntatap,\nIntheVeda, however, theweaker ofthetwoforms allowed bythese\nrules inthe firstpersonisalmost exclusively inuse: thus, 1stonly bibhdya,\ntatdpa, 3dbibhdya, tat&pa. The only exceptionsnoticed arecakdra and\njagrdha (doubtful reading)inAV.\nc.Amedial short vowel hasinallthreepersonsalike\ntheguna- strengthening (wherethis ispossible:240):thus,\nfromj/gqjdruhcomesl^FJ dudroh; fromj/fipT^Vcomes\n\\d^\\ viveg; from y^fift krtcomesr^\\^cakdrt.\nd.These rules aresaidbythegrammarianstoapplyto\nthe2dsing, always when ithassimplethaasending ;ifithas\nitha (below, 797),theaccent isallowed tofallonanyoneof\nthesyllablesoftheword, andtheroot-syllableifunaccented has\nsometimes theweak form(namely,incontracted stems with efor\nmedial a:below, 794e;andincertain other verbs :thus, vivijithd.\n17*\n"
+    },
+    {
+        "page": 276,
+        "content": "260 X.PERFECT-SYSTEM.[793\nThe earlierlanguage, however, appearstoafford noexample ofa2d\nsing., whatever itsending, accented onany other than theradicalsyllable,\norfailingtoconform totherules ofstrengthening asgiven above(ina,b,c).\ne.Sporadicinstances ofastrengthening inother than the singular\npersonsarefound inRV.: thus, yuyopimd, vivefus. And therootspr,pr,\nanddr'tearjare saidbythegrammarians tohave thestrong stem inthe\nweak forms; andjr'decay' tobeallowed todothesame.\nf.The rootmrjhas(asinthepresent-system:627) vrddhi instead of\ngunainstrong forms :thus,mamdrja ;andyguh (also asinpresent:745 c)\nhasuinstead ofo.\n794.Asregardstheweakeninginweak forms :\na.Ithasbeen seen above(783 b)that roots beginning\nwith iorufusereduplicating and radicalsyllable togetherto i\noruintheweak forms; and(784)that rootscontracting va\nandyatouoriinthereduplication do italso intheroot in\nweak forms, thetwoelements here alsocoalescingtouor i.\nb.Afew roots having yaandvaafter afirst initial consonant, and\nreduplicatingfrom thesemivowel(785), contract theyaandvato iandu:\nthus,vivic fromyvyac,vividh from yvyadh, susup fromysvap.The ex-\ntended rootsjya, pyd, vyd, $ya, hvashow asimilar apparent contraction,\nmaking their weak forms from thesimpler rootsji,pi,m,CM,7m,while hva\nmust andfvdmay gettheir strong forms alsofrom thesame(anditis\nquestionable whether from theothers strong formsoccur).\nC.The rootgrabh orgrah (ifitbewrittenthus)contracts togrh,making\nthethree forms ofstem jagrdh (1stand2dsing, act.), jagrah (3d), and\njagrh; butprach (ifitbesowritten) remains unchanged throughout.\nd.Anumber ofroots having medial abetween single con-\nsonants dropthatvowel. These are, inthelater language, gam,\nJthan, Jan, han, ghas: theyform theweak stemsjctgm, jakhn,\njajn, jaghn (compare 637), jaks (compare 640):butRV. has\noncejajanus.\nIntheoldlanguage arefound inlikemanner mamndthe andmamnate\nfrom}/man,-vavne fromyvan; tatne, tatnise, tatnire from ytan (beside\ntatane, andtate,asiffromyta); paptima andpaptus andpaptivdns from\nypat (beside pet-forms; below, e);saccima and safcus, sacce andsafdre\nfromysac.\ne.Roots ingeneral having medial abefore asinglefinal\nconsonant, andbeginningalso with asingle consonant that is\nrepeated unchangedinthereduplicationthatis,notanas-\npirate,aguttural mute, orh contract their rootandredupli-\ncation togetherinto onesyllable, havingeasitsvowel :thus,\nysadforms theweak stem sed,ypacforms pec,yyamforms\nyem; and soon.\nCertain roots nothaving theform here defined aredeclared bythe\ngrammarians toundergothesame contraction most ofthem optionally;\n"
+    },
+    {
+        "page": 277,
+        "content": "797] STEM-FORMS; ENDINGS. 261\nandexamples ofthem allareofrareoccurrence(ofoneonly, bhaj, quotable\nfrom the olderlanguage). They areasfollows: raj (occursinMBh.) and\nradh(radh?), notwithstanding their long vowel; phan, phal, bhaj (occurs in\nRV.etc.), though their initial ischanged inreduplication; trap, tras(occurs\ninMBh.), syam, svan, though they begin withmore than oneconsonant;\ndambh (forming debhfrom theweakerdabh), thoughitends withmore than\none; andbhram(occursinKSS.), bhraj, granth, ^ranth, svanj,inspiteof\nmore reasons than one tothecontrary.\nThis contraction isallowed also in2dsing. act.when theendingis\nithd: thus, tenitha beside tatantha (but noexamples arequotable from the\nolderlanguage).\nThe roots papanddad(from da:672)are said toreject thecon-\ntraction :butnoperfect forms ofeitherappeartohavebeenmetwith inuse.\nFromytr (or tar)occurs terus(R.); and jerus fromyjrisauthorized\nbythegrammarians both against thegeneral analogy ofroots inr.\nf.Roots endinginalose their abefore allendings be-\nginningwith avowel, including those thatassume theunion-\nvowel i(796)unless inthe latter case itbepreferableto\nregardthe iasaweakened form ofthe a.\n795.Endings, andtheir union with thestem.\nThegeneralscheme ofendingsoftheperfectindicative has\nbeenalready given (553); and ithasalsobeenpointedout\n(543)that roots endingin TTahave^auin1stand3d\nsing,active.\n(796. Those oftheendingswhichbeginwith acon-\nsonant--namely %tha,^va, JJmainactive; Hse,3sc\nvahe,R%make, 9dhve,\\reinmiddle arevery often,\nandinthelater language usually, joinedtothebasewith\nthehelpofaninterposedunion-vowel^i.\nTheunion-vowel iisfound widely used alsoinother parts ofthegeneral\nverbal system: namely,inthe sibilant aorist,thefutures, andtheverbal\nnouns andadjectives (asalso inother classes ofderivative stems).Inthe\nlater language,acertain degree ofcorrespondenceisseenamong thedifferent\nparts ofthesame verb,asregards their useornon-use oftheconnective;\nbut this correspondenceisnot soclose that generalrules respectingitcan\nbegiven withadvantage ;and itwillbebest totreat each formation byitself.\nTheperfectisthetense inwhich theuseof thasestab-\nlished itself most widely andfirmlyinthelater language.\n797.Themost importantrules astotheuseof^*in\nthelater languageareasfollows:\n"
+    },
+    {
+        "page": 278,
+        "content": "262 X.PERFECT-SYSTEM.[797\na.The\"^reof3dpi.mid. has italways.\nb.The otherconsonant-endings, except5Jihaof2d\nsing, act., take itinnearlyallverbs.\nBut itisrejected throughout byeight verbs namely Ar\n'make', Ihr'bear', sr'go',vr'choose', dru'run', cru'hear',\nstu'praise',sru'flow'; and itisallowably (not usually) rejected\nbysome others, ingeneral accordance with their usageinother\nformations.\nc.In2dsing, act.,itisrejected notonlybytheeight\nverbsjust given,butalsobymany others, endinginvow-\nelsorinconsonants, which inother formations haveno\n^i;but itisalsotaken bymanyverbs whichrejectitin\nother formations;and itisoptionalinmany verbs,in-\ncludingthose in%Ra(ofwhich the 5TT islostwhen the\nendingis^2fitha],andmost ofthose in^',3,and3u.\nThe rules ofthegrammarians, especiallyasregards theuseofihaor\nitha,runoutinto infinite detail, and \"2-tre notwholly consistent with one\nanother; and, astheforms arebynomeansfrequent,itisnotpossibleat\npresenttocriticise thestatements made, and totellhow farthey arefounded\nonthefacts ofusage.\nWith thisi,afinal radical ior iisnotcombined, but\nchangedintoyoriy.Theuof~\\/bhubecomes uvthroughout\nbefore avowel.\n798. Intheolder language,theusageisinpart quite\notherwise .Thus :\na.IntheRV., theunion-vowel iistaken byroots ending incon-\nsonants providedthelastsyllable ofthestem isaheavy one,butnotother-\nwise: thus, dsitha, uvocitha, viveditha, buttatdntha andvivydktha; ucimd,\npaptima, sedima, yuyopimdj butjaganma andyuyujma; ucise, jajnise,\nsasahise, but vivitse anddadrkse;bubhujmdhe and$a$admdhe etc.(noex-\namplesofivahe orimahe chance tooccur, noranyofeither idhve ordhve}\\\nijire, jajnire, yetire, tataksire, but caklpre, vividre, duduhre, pasprdhre,\ntatasre (and soon :twenty-two forms). TheonlyexceptioninRV. isvettha\nfromyvid,without i(inBr.,also attha fromyah: below, 801a).The\nother Vedic texts present nothing inconsistent with this rule, but inthe\nBrahmanas 3dpi.forms inirearemade after light syllables also :thus,\nsasrjire, bubudhire.\nb.Inroots ending with avowel,the early usage ismore nearly like\nthe later. Thus: forroots inatherule isthesame(exceptthatno2dsing,\ninitha ismetwith),asdadhimd, dadhise, dadhidhve, dadhire (theonly per-\nsons with iquotable fromRV.andAV.);roots inrappearalso tofollow\n"
+    },
+    {
+        "page": 279,
+        "content": "800]EXAMPLES OFINFLECTION. 263\nthelater rule :ascakrse, paprse, vavrse, vavrmdhe, butdadhrise andjabhrise,\nandin3dpi.mid. both cakrire and dadhrire; ybhuhasboth babhutha\n(usually) andbabhuvitha, butonlybabhuvimd (AV.). But there arefound\nagainst thelater rules, susuma, cicyuse, juhure, andjuhure, without i:the\ninstances aretoofew tofound aruleupon.\n799. Theending rire of3dpi.mid. isfound inRV. insixforms :\nnamely, cikitrire, jagrbhrire, dadrire, bubhujrire, vividrire, sasrjrire ;towhich\nSV.adds duduhrire.\n800.Examplesofinflection. Bywayofillustra-\ntion oftherulesgiven above, maybegiveninfulltheper-\nfectindicative inflection ofthefollowingverbs :\na.Asexampleofthenormal inflection ofarootwith\nfinal consonant, wetaketheroot^TJbudh,'know': itsstrong\nform ofperfect-stemis^t^T bubodh; weak form, SRTJ bu-\nactive. middle.\nd. . s. d.\nbubodha bubudhiva bubudhimd bubudhe bubudhivdhe btibudhim&he\n2^^^ ^^ ^_ ^^\nbubodhitha bubudhdthus bubtidhd bubudhise bubudhathe bubudhidhve\ny, .\nbubodha bubudhdtus bubudhiis bubudhe bubudhate bubudhire\nThe asserted variety ofpossibleaccent in2dsing.act.(above, 793d)\nneeds tobenoted both inthisandintheremaining paradigms.\nb.Asexampleofthenormal inflection ofarootwith\nfinal iorw-vowel, wemaytake theroot ^ftm,'lead': its\nforms ofstem areHTO nindyorpRTTT ninay, andft^ ninl.\nnindy a,ninhya ninyivd ninyimd ninye ninyivdhe ninyimdhe\n2i^r,PnRj^in-dy^SRI PiP^5PFOT ftfefij\nnine fha,nindyitha ninydthus ninyd ninyise ninyhthe ninyidhve\n3PRUT PKIHUPRR f^F& Pi^icJ Mp^l(\nt o\"v.\"s., L '\nninaya ninydtus ninyus ninye ninyate ninyire\nTheroot kriwould make inweak forms cikriyivd, cikriydtus,\ncikriyus, etc.;ybhumakes babhnva, babhutha(V.)orbabhuvitha,\n"
+    },
+    {
+        "page": 280,
+        "content": "264 X.PERFECT-SYSTEM.[800\nbabhuvivd, babkuvtts; bab/mve, babhuvire, etc.; other roots inuor\nuchangethis touvbefore the initial vowel ofanending.\nc.Asexampleoftheinflection ofarootendingin5TT\n,wemaytake^Tda, 'give':itsforms ofstem are^Tdada\nand([Jdad(or dadi):seeabove, 794 f.\ndadau dadivd dadima, dade dadivdhe dadimdhe\n3$m,ffi ^m^^^f^^\ndadatha, dadithd daddthus dadd dadise dadhthe dadidhve\ndadau daddtus dadus dade dadate dadire\nTheRV. hasoncepaprdforpaprcrf* (and Bafts' forjahati?).\nd.Asexampleofarootwith medial ^ashowingfu-\nsionofrootandreduplication, resultinginmedial^e,in\ntheweak forms(794 e),wemaytakerRtan,;stretch': its\nforms ofstem arenTrH tatdn orH'HH tatan, andrR^ew.\n,rTrTR\ntatdna, tatnna tenivd tenimd tene tenivdhe tenimdhe\ntatdntha, tenithd tendthus tend tenise tenathe tenidhve\n3rTrTHHHHH^ ^5^^^^^T\ntatana tendtus tenus tene tenate tenire\nTherootJan,with theothers which expel medial ainweak\nforms(794 d),makesjajdnthaorjajnithd, jajnivd, jajnus ;jajne,\njajnimdhe, jajnire ;and soon.\ne.Asexampleofarootwith initial 3fvacontracted\nto3uinthereduplication,andcontracted with theredu-\nplicationto3Tuinweak forms(784),wemaytake 3frlvac,\n'speak':itsforms ofstem are3^uvdc or331^ uvac, and\nV^NC|\nuvdca,uvaca ucivd ucimd uce ucivdhe ucimdhe\n2\nuvdktha, uvdcitha ucdthus ucd ucise ucathe ucidhve\n"
+    },
+    {
+        "page": 281,
+        "content": "801]IRBEGULARITIES.265\nuvaca ucdtus ucus uce ucate ucire\nInlikemanner, ^yajformsiydja oriyaja, iydsthaoriydjitha ;\nye, yise, and soon\n;j/wc hasnvoca and uvocitha inthestrong\nforms, and alltherest like vac.\nf.Ofthefour roots inftrmentioned at797b,thefirst\npersonsaremade asfollows:\ncakdra, cakara cakrvd cakrmd cakre cakrvdhe cakrmdhe\nThe2dsing.act. iscakdrtha; the3dpi.mid. iscakrire,\nOftheroots inftringeneral,the firstpersons are\nmade asfollows :\ndadhdra,dadharadadhrivd dadhrimd dadhre dadhrivdhe dadhrimdhe\n801.Afewmiscellaneousirregularitiescall still for\nnotice :\na.The root ah, 'speak', occursonlyintheperfectindic-\native, andonlyinthe3dpersonsofallnumbers and inthe\n2dsing,anddu., inactive(andin2dsing,thehisirregularly\nchangedto tbefore theending): thus, attha, nha; ahathus,\nahatus; ahus(inV.,only nhaandahus aremetwith).\nb.From yva 'weave', the3dpi.act.uvus occurs inRV., andnoother\nperfect form appearstohave beenmetwith inuse. Itisallowed bythe\ngrammarians tobeinflected regularly asva;and also asvay (thepresent-\nstem isvdya:761d,2),with contraction ofvatouinweak forms;and\nfurther,intheweakforms,assimple w.\nc.The rootvya, 'hide', hasinRV. theperfect-forms vivyathus and\nvivye, andnoothers appeartohave beenmetwith inuse; thegrammarians\nrequire thestrong forms tobemade from vyay, andtheweak from vi.\nd.The rooti,'go', forms inRV.andAV. the2dsing.act. iydtha\nbeside theregular iyetha.\ne.TheAV. hasonce vavrdhete(for -dhdte), and once jaharus (for\njdhrus}:both areperhapsfalse readings.\nf.Persons oftheperfect from theer-forms ofroots inchangeable f\n(242)aretitirus and tistire (bothRV.); andtheyhave corresponding participles.\ng.The bastard rooturnu (712)issaidbythegrammarianstomake\ntheperfect-stem urnunu.\nh.The roots majj andnapare said toinsert anasal inthe2dsing.\n"
+    },
+    {
+        "page": 282,
+        "content": "266 X.PERFECT-SYSTEM.[801\nactive, when theendingissimple tha: thus, mamanktha, nanahstha (also\nmamajjitha andne?itha).\ni.Theanomalous ajagrabhaisarh (AB. vi.35)seems aformation onthe\nperfect-stem (butperhapsforajigrabhisan, desid.?).\nPerfect Participle.\n802 .TheendingoftheactiveparticipleisSfftTvans(that\nistosay,inthestrongforms :itiscontracted to3^usinthe\nweakest, andreplaced bycffiTvatinthemiddle forms :see\nabove, 458if.).Itisadded totheweak form oftheperfect\nstem asshown, forexample,inthedualandpluralof\ntheactive inflection ofthegiven verb; and, mechanically,\ntheweakestparticiple-stemisidentical with the3dpi.\nactive. Thus, sHMlU bubudhvans, PHl^l^ nimvans, rloNfH\ncakrvans.\n\\803. Iftheweak form oftheperfect stem ismonosyl-\nlabic, theendingtakes theunion-vowel Si(which, however,\ndisappearsintheweakestcases): thus, (Tf^lH tenivans,\nucivans, sii^JH jajnivans, ^ii^cjtHadivans(from\nad:783a),and soon;^f^TCTdadivans and itslike,\nfrom roots in5TTa,istobereckoned intheone class or\ntheother accordingasweview its^iasweakened root-\nvowel orasunion-vowel(794f).\nButparticiplesofwhich theperfect-stemismonosyllabic byabsence of\nthereduplication donottake theunion-vowel: thus, vidvdns,and,inV.,\ndafvahs, midhvans, sahvans;andRV.hasalsodadvans (AV. dadivans and\nonce dadavdhs) from ydd (ordad :672) andAV. has vifivdns and\nvarjivahs (innegativefern,dvarjusi).\n804. Other Vedic irregularities calling fornotice arefew. The long\nvowel ofthereduplication (786) appearsintheparticiple asintheindicative :\nthus, vavrdhvdhs, sasahvdhs, jujuvdhs. RV.andAV. have sasavdns from\nysanorsa.RV.makes theparticipial forms ofytrortarfrom different\nmodifications ofthe root:thus, titirvdhs, but tatarusas. Respecting the\noccasional exchanges ofstrong andweak stem ininflection, seeabove, 462 c.\n805. From rootsgam andhan theVeda makes thestrong stems\njaganvdhs (astothen,see212) andjaghanvdhs; thelater language allows\neither these orthemore regular jagmivdhs andjaghnivdhs (the weakest\nstem-forms being everywhere jagmus andjaghntis).\n"
+    },
+    {
+        "page": 283,
+        "content": "810] PARTICIPLE; MODES.267\n806. From threeroots, vid'find', uip,anddry,thelater language allows\nstrong participle-stems tobemade with theunion-vowel, aswell asinthe\nregular manner without it:thus, vivifivdna orvivifvdns. PB. hasonce\ncichidivahs.\n807.Theendingofthemiddleparticipleisand. It\nisadded totheweak form ofperfect-stem, asthisappears\ninthemiddle inflection :thus, 5RTJH bubudhand,\nninyana,^R dadand, ^F? tenand, sT^H jajnand,\nucdnd.\nIntheVeda,thelong reduplicating vowel isshown bymany middle\nparticiples:thus, vavrdhand, vdvasdnd, dddrhdnd, fucuvand, etc.RV.has\n?afayand fromy$l (with irregular guna, asinthepresent-system: 629);\ntistirand from]/sr,- andonce, withmana, sasrmdnd fromj/sr.\nModes ofthePerfect.\n808. Modes oftheperfect belong onlytotheVedic lan-\nguage, and areevenrarely found outside oftheRig-Veda.\nTodraw thelinesurely anddistinctly between these andthemode-forms\nfrom otherreduplicated tense-stems thepresent-stem ofclassII., the\nreduplicated aorist, and theintensive isnotpossible, since nocriterion\nofform exists which does not insome casesfail, and since thegeneral\nequivalence ofmodal forms from allstems(582), andthecommon useof\ntheperfect asapresentintheVeda(823), deprive usofacriterion of\nmeaning. There canbenoreasonable doubt, however, that aconsiderable\nbody offorms aretobereckoned here :optativeslikeanacydm andbabhuyds\nandbabhuydt, imperativeslikebabhutu, subjunctives likejabhdrat, show such\ndistinctive characteristics oftheperfect formation thatbytheir analogy other\nsimilar words areconfidently classed asbelongingtotheperfect.\n809. Thenormal method ofmaking such forms would\nappeartobeasfollows :from areduplicated perfect-stem,as\n(forexample) mumuc, animperative would bemade bysimply\nappending,asusual, theimperative endings;thederived sub-\njunctive mode-stem would bemumoca(accentedafter theanalogy\nofthestrong forms oftheperfect indicative),andwould take\neitherprimaryorsecondary endings ;and theoptative mode-\nstems would bevnumucyhintheactive, andmumuci(accenton\npersonal endings)inthemiddle.\nAnd thegreat majorityoftheforms inquestion (about\nthreequarters)aremade inthese ways. Thus :\n810. Examplesoftheregular subjunctiveformation are :\na.with secondary endings, active: 2dsing., paprdthas, mamdhas,\npiprdyas, bubodhas; 3dsing., jabhdrat, sasdhat, paspdr$at, piprdyat, ciketat;\n"
+    },
+    {
+        "page": 284,
+        "content": "268 X.PERFECT-SYSTEM.[810\n1stpi.,tatdndma, (u^dvdma; 3dpi.,tatanan, paprdthan (other persons do\nnotoccur). This isthelargest class ofcases.\nb.with primary endings, active: hereseem tobelong only dadhdrshati\nandvavdrtati: compare theformation with different accentbelow, 81la.\nc.ofmiddle forms occur only the3dsing, tatdpate, papamafe, yuyojate,\njujosate (SV.; RV.hasjujosate)-, andthe3dpi.tatdnanta (andperhaps two\northree others:below, 811b,end).\n811. Butnotafewsubjunctives ofother formation occur; thus:\na.with strengthened root-syllable, asabove, butwith*accent onthe\nreduplication (asinthemajority ofpresent-formsofthereduplicatingclass :\nabove, 645). Here theforms with primary endings, active, preponderate,\nandarenotvery rare: forexample, jtijosasi, jtijosati, jujosathas, jtijosatha\n(other persons donotoccur). With secondary endings, jtijosas, jtijosat, and\njujosan aretheforms that belong most distinctly here(since ddda$as and\nstisudas etc. areperhapsratheraorists). And there isnomiddle form but\njtijosate (RV.: seeabove, 81Oc).\nb.withunstrengthened root-syllable occur asmall body offorms, which\nareapparentlyalsoaccented onthereduplication (accented examples arefound\nonly in3dpi.mid.): thus, active, forexample, mumucas;vavrtat, vividat,\nptipuuai; theonly middle forms aredadhrsate, vdvrdhate, 3dsing.; and\ncdkramanta, dddhrsanta, rurucanta(with dadabhanta, paprathanta, mdmahanta,\njuhuranta, which might alsobelong elsewhere: 81Oc).\nc.accented ontheending arevdvrdhdnta andcakrpdnta (whichare\nrather tobecalled augmentless pluperfects).\nAstoforms with double mode-sign, ortransfers toana-conjugation,\nseebelow, 815.\n812. Examplesoftheregular optativeformation are:\na.inactive: 1stsing., dna$ydm, jagamydm, paprcydm, riricydm; 2d\nsing., vavrtydSj vivifyds, fUfruyas, babhuyds; 3dsing., jagamydt, vavrtydt,\ntutujydt, babhuydt; 2ddu.,jagamydtam, pucruydtam; 1stpi.,sasahyama,\nvavrtyama, fu^uydma; 3dpi., tatanyus, vavrtyus. The forms arequite\nnumerous.\nb.inmiddle, theforms arefew:namely,1stsing., vavrtlya; 2dsing.,\nvawrdhithas, caksamlthds;3dsing., jagrasita, vavrtita, mdmrjlta, pufuclta;\n1stpi.,vawrtimahi. And sdsahlsthds and ririsista appeartofurnish examples\nofprecative optative forms.\nThere isnoirregular mode offormation ofperfect optatives.Individual\nirregularitiesareshown bycertain forms :thus, cakriyds, paplydt, fu$ruyas\nandwruydtam,with treatment ofthe final asbefore thepassive-sign yd\n(770); anajydt with shortinitial; jak&ydtisanomalous; ririses istheonly\nform thatshows aunion-vowel a.\n813. Ofregular imperative forms, only avery small number aretobe\nquoted:namely, active, cikiddhi, mumugdM, pucugdhf, andpiprihi; mumoktu\nandbabhutu; mumuktam andvavrktam; jujustana andvavrttana (unless we\n"
+    },
+    {
+        "page": 285,
+        "content": "818\n1 MODES; PLUPERFECT. 269\nare toaddmamaddhf, mamattu, mamdttana); middle, vavrtsva and\nvavrddhvam.\n814. Asirregular imperatives may bereckoned several which show a\nunion- vowela,orhave been transferred toana-conjugation. Such are, in\ntheactive, mum6catam andjujosatam (2ddu.), andmumticata (2d pi.};in\nthemiddle, piprdyasva (only onefound withaccent), andmdmahasva,vdvrdh-\nasva, vdvrsasva(2d sing.), andmdmahantdm(3d pi.: probablytobeac-\ncented -dsva and-dntam).\n815. Such imperatives asthese, taken inconnection with some ofthe\nsubjunctives given above (and afew ofthe\"pluperfect\" forms:below, 820),\nsuggestasplausible theassumption ofadoublepresent-stem, withredupli-\ncation aridadded a(with which thedesiderative stems would becomparable:\nbelow, chap. XIV.):forexample, jujosa fromYJUS,fromwhich would come\njujosasietc.andjujosate (81la)asindicative, jujosas etc. assubjunctively\nusedaugmentless imperfect, andjujosatamasimperative. Most oftheforms\ngiven above assubjunctives withprimary ending lack amarked andconstant\nsubjunctive character, andwould pass fairly well asindicatives. And it\nappears tolerably certain thatfrom one root atleast, vrdh, such adouble\nstem istoberecognized ;from vdvrdha come readily vdvrdhate, vdvrdhdnta,\nandfrom italone cancome regularly vdvrdhasva, vdvrdhete(above, 801e),\nandvdvrdhdti(once, RV.) and, yetmore, theparticiple vavrdhdnt (once,\nRV.: anisolatedcase): yeteven herewehave alsovdvrdhltha's,notvdvrdh-\nethds. Toassume doublepresent-stems, however, inallthecases would be\nhighly implausible ;itisbetter torecognize theformation asonebegun,\nbutnotcarried out.\nOnly oneother subjunctive with double mode-sign namely, paprcdsi\nisfound tosetbeside vdvrdhdti.\n816. Forms ofdifferent model arenotveryseldom made from thesame\nroot: forexample, fromj/mwc,thesubjunctives mumticas, mumocati, and\nmumucas;fromydhrs,dadhdrsati anddadhrsate,fromj/pn,theimperatives\npiprlM andpiprdyasva.\nPluperfect.\n817. Ofanaugment-preterit from theperfect-stem,towhich\nthename ofpluperfectisgiven onthegroundofitsformation\n(thoughnotofitsmeaning),theVeda presentsafewexamples ;\nandoneortwoforms ofthe later language (mentioned above,\n788, end)have alsobeen referred toit.\nThere issomethingofthesame difficulty indistinguishingthepluperfect\nastheperfect-modes from kindred reduplicatedformations. Between itand\ntheaorist, however,adifference ofmeaning helpstomake aseparation.\n818. Thenormal pluperfectshould show astrong stem inthesingular\nactive,arid aweak oneelsewhere thus,mumoc andmumuc with\naugment prefixed and secondary endings added (usin3dpi. act.,atain\n3dpi.mid.).\n"
+    },
+    {
+        "page": 286,
+        "content": "270 X.PERFECT-SYSTEM.[818\nOfforms made according tothismodel, wehave, intheactive: 1stsing.,\najagrabham andacacaksam (which, by itsform, might beaorist: 860);\n2dsing, djayan; 3dsing., ajagan andaciket; 2ddu.,amumuktam; 2dpi.\najaganta, andajagantana andajabhartana (astrong form, assooften inthis\nperson: 556); 3dpi.(perhaps), amamandus andamamadus .Tothesemay\nbeadded theaugmentless ciketam andcakaram. Inthemiddle, the3dpi.\nacakriran andajagmiran (with Iran instead ofata), and theaugmentless\n2dsing, jugurthas andsusupthds,arethemost regular forms tobefound.\n819. Several forms from roots ending inconsonants save theendings\nin2dand3dsing. act.byinserting anI(555 b):thus, dbubhojis, avive&s;\narireclt, djagrabhlt (avavarit andavava^ltdm areratherintensives); andthe\naugmentless jfhinszs (accent?) anddadharslt belong with them.\n820.Afewforms show astem ending ina:they are,intheactive:\n3dsing., asasvajat, acakrat; inthemiddle: 3dsing., dpiprata; 2ddu.,\ndpasprdhetham ;3dpi.,atitvisanta (which byitsform might beaorist),\nddadrhanta; andcakradat, cakrpdnta, vdvrdhdnta, juhuranta, would perhaps\nbebest classified here asaugmentless forms (compare 811, above).\nUses ofthePerfect.\n821. Inthelater language,theperfectissimplyapreterit\norpast tense, equivalenttotheimperfect, andinterchangeable\nwith it.Exceptascoming from afewoften used verbs(espe-\nciallyahaanduvaca),itismuch morerarely employedthan the\nimperfect.\n822. IntheBrahmana language, very nearly thesame thing istrue.\nInmost Brahmanas, theimperfectistheusual tense ofnarration, andthe\nperfect only occasional;intheQatapatha Brahmana, theperfectismuch\nmore widely used.\n823. IntheVeda, thecase isvery different. The perfectisused as\npasttense innarration, butonly rarely ;sometimes also ithasatrue \"perfect1'\nsense,orsignifiesacompleted past;butoftenest ithasavalue notdistin-\nguishableinpointoftime from thepresent.Itisthus theequivalent of\nimperfect, aorist, andpresent; and itoccurs coordinated withthem all.\nExamplesare :ofperfectwithpresent, ndfrdmyanti ndvfmuncanty\netevdyondpaptuh (RV.), 'they weary notnorstop, they flylikebirds';se\n'durdjd ksayati carsanindm ardnndnemth pdri tdbabhuva (RV.), 'hein\ntruth rules king ofmen;heembraces themall,asthewheel thespokes';\nofperfectwithaorist, tiporuruce yuvatfr ndy6sd...dbhud agnih samtdhe\nmdnusanam dkar jyotir bddhamdnd tdmdhsi(RV.),'she iscomebeaminglike\nayoung maiden;Agni hasappearedforthekindlingofmortals;shehath\nmade light, driving away thedarkness';ofperfect withimperfect, dhann\ndhim dnvapds tatarda (RV.),'heslew thedragon, andpenetratedtothe\nwaters'. This lastcombination isofconstant occurrence inthelaterlanguage.\n"
+    },
+    {
+        "page": 287,
+        "content": "824] 271\nCHAPTER XI.\nTHEAORIST SYSTEMS.\n<^824.UNDER thename ofaorist areincluded(aswas\npointedoutabove, 532) threequite distinct formations,\neach ofwhich has itssub-varieties :namely,\nI.ASIMPLE-AORIST(equivalenttotheGreek \"second\naorist\"), analogousinallrespectsastoformand inflection\nwith theimperfect.Ithastwo varieties: 1.theroot-\naorist, with atense-stem identical with theroot(corres-\npondingtoanimperfectoftheroot-class, I.);2.the\na-aorist, with atense-stemendingin f#,orwithunion-\nvowel 5fabefore theendings (correspondingtoanimper-\nfectofthea-class, VII.).\nII.3.AREDUPLICATED AORIST, perhapsinoriginident-\nicalwith animperfectofthereduplicatingclass(II.),but\nhaving come tobeseparatedfrom itbymarkedpeculiarities\nofform. Itusuallyhasaunion-vowel 51abefore theend-\nings,orisinflected likeanimperfectofoneofthe^-classes;\nbutafewforms occur intheVeda without such vowel.\nIII.ASIBILANT-AORIST(correspondingtotheGreek \"first\naorist\"), havingfor itstense-signaHsadded totheroot,\neitherdirectlyorwith apreceding auxiliary ^*/itsendings\nareusually added immediatelytothetense-sign,butina\nsmallnumber ofroots with aunion-vowel 5[a;averyfew\nroots alsoareincreased byHsforitsformation;andaccord-\ningtothese differences itfalls into four varieties :namely,\nA.without union-vowel ^abefore endings:4.s-aorist,\nwithHsalone added totheroot;5.as-aorist,thesame\nwithinterposed ^i;6.s^s-aorist, thesame asthepre-\ncedingwithHsadded attheend oftheroot; B.with\nunion-vowel 5!\"a,7.sa-aorist.\n"
+    },
+    {
+        "page": 288,
+        "content": "272 XI.AORIST-SYSTEMS.\nj825\n825, Allthese varieties areboundtogether andmade\ninto asingle complex system bycertaincorrespondences of\nform andmeaning. Thus,inregardtoform, theyare all\nalike,intheindicative, augment-preteritstowhich there\ndoes not exist anycorresponding present;inregardto\nmeaning, althoughinthe later orclassicallanguage they\naresimply preterits, exchangeable withimperfects andper-\nfects, theyallalike have intheolderlanguagethegeneral\nvalue ofacompleted pastor\"perfect\", translatable by'have\ndone' andthelike.\n826. The aorist-systemisaformation ofvery infrequent occurrence in\nthe classical Sanskrit(itsforms arefound, forexample, only twenty-one\ntimes intheNala, eight intheHitopadeca, seven inManu, sixeach inthe\nBhagavad-Gita andQakuntala), and itpossesses noparticiple, noranymodes\n(exceptingintheprohibitive use ofitsaugmentless forms: see580; and\nthe so-called precative:see921ff.);intheolder language, ontheother\nhand,itisquite common, andhasthewhole variety ofmodes belongingto\nthepresent, andsometimesparticiples.Itsdescription, accordingly, must be\ngiven mainly asthat ofapartoftheolder language, with duenotice ofits\nrestriction inlater use.\n827. IntheRV., nearly half the rootsoccurring show aorist forms,\nofone oranother class;intheAV., rather lessthan onethird; andinthe\nother texts oftheolder language comparatively few aorists occur which are\nnotfound inthese two.\nMore than fifty roots, inRV.andAV. together, make aorist forms of\nmore than one class(nottakingintoaccount thereduplicatedor\"causative\"\naorist); butnolawappears tounderlie this variety; ofany relation such as\nistaught bythegrammarians, between active ofone class andmiddle of\nanother ascorrelative, there isnotrace discoverable.\nExamples are :ofclasses 1and4,adham anddhasus fromydha, ayuji\nandayuksatafromyyuj ; of1and5,agrabham andagrdbhxsma from\nygrabh,mrsthas andmarsisthas fromymrs ; of1and2,artaandarat\nfromyr ;of2and4,avidam and avitsi fromyvid 'find', anijam and\nanaikslt fromj/ny;of2and 5,sanema andasanisam fromysan;of\n2and7,aruham andaruksat from yruh;of4and5,amatsus and\namadisus fromymad;of4and 6,hasmahi and hdsisus fromyha;\nof1and2and 4,atnata andatanat andatan fromytan;of1and4\nand5,abudhran andabhutsi andbodhisat from ybudh,astar and strslya and\nastaris fromystr.Often thesecond, orsecond and third, class isrepre-\nsented byonly anisolated form ortwo.\n"
+    },
+    {
+        "page": 289,
+        "content": "831]SIMPLE AORIST: 1.ROOT-AORIST. 273\nI.Simple Aorist.\n828. Thisis,ofthethreeprincipaldivisions ofaorist,\ntheone leastremoved from theanalogyofformsalready ex-\nplained;itislikeanimperfect,oftheroot-class orofthed-\nclass, without acorresponding present indicative, butwith(more\norlessfragmentarily)alltheotherparts which gotomake up\nacomplete present-system.\n1.Root-aorist.\n829. This formation isinthelaterlanguage limited to\nafewroots in TTaandtherootHbhu,and isallowed to\nbemade intheactiveonly,themiddleusinginstead the\ns-aorist(4),orthe is-aorist(5).\nThe roots inETFatake 3^Tusas3dpi.ending, and, as\nusual,lose their TTabefore it;*Fbhu(asintheperfect:\n793a)retains itsvowel unchanged throughout, inserting ^\nvafter itbefore theendings FTamandRanof1stsing,\nand3dpi.Thus:\nddam ddava ddama dbhuvam dbhuva dbhuma\n-X\nddas ddatam ddata dbhus dbhutam dbhuta\n(Tx\nddat ddatam ddus dbhut dbhutam dbhuvan\nFortheclassicalSanskrit,this isthewholestory.\n830. IntheVeda,, these same roots aredecidedlythemost\nfrequent andconspicuous representativesoftheformation :es-\npeciallytherootsffd,da,dha,pa'drink', stha, bhu; while spo-\nradic forms aremade fromJnd, pra, sa,ha.Astotheirmiddle\nforms, seebelow, 834a.\nInstead ofabhuvam, RV. hastwice dbhuvam.\n831. But aorists ofthesame class arealsomade from a\nnumber ofroots inr,andafewiniandu(shortorlong)\nwith, asrequired bytheanalogyofthetense withanimperfect\noftheroot-class, ywwa-strengtheninginthethree personsofthe\nsingular.\nThus(intheactive), fromj/?rw, afravam and afrot; from}/?n, acres\nandofret;fromykr 'make', akaram andakar(forakars and afcar-<); from\nWhitney ,Grammar. 18\n"
+    },
+    {
+        "page": 290,
+        "content": "274 XI.AORIST-SYSTEMS.[831\nvr'enclose', dvar (585); and soadar, aaiar, aspar. Dual andplural forms\naremuch lessfrequent thansingular; but forthemost partthey alsoshow\nanirregular strengtheningoftheroot-vowel: thus(including augmentless\nformsjj akarma andakarta, vartam, spartam, ahema andahetana, bhema,\nftoma, afravan; regular areonly avran, afcrara, ahyan, afriyan, andanltdm.\n832. Further, from afew roots with medial(orinitial)\nvowel capableofywraz-strengthening, andhavingingeneral that\nstrengthening onlyinthesingular.\nThus,abhet(2dand3dsing.) fromybhid; amok (3dsing.) fromj/mwc ,\navart fromyvrt ;vark fromyvrj (AV.hasonceavrk) ;adarfam from}/drp,\nandadrfan, avrjan, afvitan. Butchedma, withguna, fromychid.\n833. Again, from alarger number ofroots with aasra-\ndical vowel :\nOfthese, gam (withnformwhen final orfollowed bym:143, 212)\nisofdecidedly most frequent occurrence, andshows thegreatest variety of\nforms :thus, agamam, agan (2dand3dsing.), aganma, aganta (strong form),\nagman. The other cases areakran fromykram;atan fromytan;askan\nfromyskand ;dsrat fromytrans(?VS.);dAafc unddaghma fromydagh ,\nana (585) andanastam from}/nap (?) ;afcsan(foragh-san,likeapman) from\nyghas;andthe3dpll.inw,dfcramws, kramus, ayamus, yamus, abddhus,\ndabhus, nrtus (impf.?):mandus andtaksus areperhaps rather tobereckoned\nasperfect forms withoutreduplication (790 b).\n834. Sofaronlyactive forms havebeen considered. In\nthemiddle, aconsiderablepartoftheforms aresuch asare\nheldbythegrammarians (881)tobelongtothes-aorist, with\nomission ofthe s:they doubtless belong, however, mostlyor\naltogether,here. Thus :\na.From roots endinginvowels, wehave adhithas andadhita; adita\nandadimahi (andadlmahi fromydd 'cut'); aathithas and asthita anddsthiran,\nforms ofa-roots (arddhvamisdoubtless forarasdhvam); ofr-roots, afcri,\nakrthds, akrta, akratam, akrata; avri, avrthas, avrta; drta (with augmentless\narta), drata; mrthds, amrta; dhrthds; drihds; astrta; asprta; gurta; of i\nanduroots, theonly examplesaredhvi (?AV., once) andacidhvam. The\nabsence ofanyanalogies whatever fortheomission ofasinsuchforms, and\ntheoccurrence ofavriandakriandakrata, show that their reference tothe\n-s-aorist iswithout sufficient reason.\nb.Asregardsrootsending inconsonants,thecase ismore questionable,\nsince loss ofsafter afinal consonant before thdsand ta(and,ofcourse,\ndhvam) would beinmany oases required byeuphonicrule(233). Wefind,\nhowever,such unmistakable middle inflection ofthe root-aorist asayuji,\nayukthds, ayukta, ayujmahi, ayugdhvam, dyujran;dstaanddfata; apadi\n(1st sing.) andapadmahi andapadran; amanmahi; aganmahi andagmata;\natnata; ajani (1st sing.) andajnata (3dpi.); fromygamaremade agathds\nandagata, andfrom>/wan, amafa,with treatment ofthefinal likethat of\nhan inpresent inflection(637). Theending ran isespecially frequentin\n"
+    },
+    {
+        "page": 291,
+        "content": "837]SIMPLE AORIST: 1.ROOT-AORIST. 275\n3dpi.,being taken byanumber ofverbs which have noother person of\nthis aorist: thus, agrbhran, dsrgran, adr^ran, abudhran, dvrtran, ajuaran,\nakrpran, asprdhran, avasran, dvifran; andram isfound beside ran in\nddr$ram, dbudhram, dsrgram.\nc.From roots ofwhich the finalwould combine with stofes,itseems\nmore probable that aorist-forms showing7c(insteadofa)before theending\nbelong totheroot-aorist: such areamukthas(andamugdhvam], aprkthds and\naprkta, abhakta, avrkta, asakthas andasakta, rikthas, vikthas and vikta :\ndspasta, asrsta, mrsthas would bethesame ineither case.\nd.Thereremain,ascases ofdoubtful belonging: amatta, arabdha,\nasrpta, atapthas, chitthas, patthas, andnutthas.\nModes oftheRoot-aorist.\n835. Insubjunctive use, forms identical with theaugmentless indic-\native ofthis aorist aremuch more frequent than themoreproper subjunc-\ntives. Those towhich nocorresponding form withaugment occurs have\nbeen given above;theothers itisunnecessary toreport indetail.\n836. Oftruesubjunctives,theforms with primary endings arequite\nfew. Intheactive, ganiistheonlyexampleof1stsing, (astobhuvani,\nseebelow);of3dsing, occursthati, and ddtianddhdti, which arealmost\nindicative inuse;ofdualpersons, sthdthas, darfathas, fravathas andprd-\nvatas. Inthemiddle, 3dsing, idhate(?anomalousaccent), 2ddu.dhethe\nanddhaithe, and 1stpi.dhamahe.\nForms with secondary endings are,intheactive, ddrfam; tdrdas, pdrcas,\nydmas ;karat, garat, daghat, yamat^ yodhat, frdvat, spdrat, sdghat ;ddrpan,\ngaran,yaman. Nomiddle forms areclassifiable with confidence here.\nThe series bhuvam, bhuvas, bhuvat, bhuvan, and bhuvani (compare\nabhuvam :830, end), andthe isolatedfriivat,areofdoubtfulbelongings ;\nwith adifferent accent, they would seem tobeofthenext class; here,a\n^Una-strengthening would bemore regular (but note theabsence ofguna in\nthe aorist indicative andtheperfect ofybhu).\n837. Asregards theoptative, wehave todistinguish between pure\noptatives andoptatives with sinserted before theendings,orprecatives.\nPureoptative forms, both active andmiddle, aremade from anumber\nofroots. From roots inaoccur intheactive,with changeofatoebefore\nthey,deyam, dheyam anddheyus, stheyama, andjneyas (which mightalso\nbeprecative);inthemiddle, only simahi anddhimahi (which might beaug-\nmentlesspreterit,asadhimahi alsooccurs once, andadhltam once :butydha\nshows nowhere else conversion ofitsatolong 1],From bhu, bhuyas and\nbhuyat (possible precatives), andbhuyama. From roots inr,themiddle\nformsarlta, muriya,vurlta. From roots ending inconsonants, afyam and\nacydma anda?yus act., andaflya andafimdhi mid. (this optativeisespe-\ncially common intheolder texts) ;vrjydm ;mrdhyas, sahyas, gamyas and\ngamyat (possible precatives), andsahyama; indhlya, gmiya, ruciya, vaslya;\nidhimahi, nafimahi, nasimahi, prclmahi, mudlmahi, yamimahi. Andrdhyam\netc.andrdhimdhi belong perhaps here,instead oftoapresent-system (cl. I.).\n18*\n"
+    },
+    {
+        "page": 292,
+        "content": "276 XI-AORIST-SYSTEMS.[838\n838. Precative optative forms ofthis aorist active areinthe later\nlanguage allowed tobemade from every verb(922). InRV., however,\nthey donotoccur from asingle rootwhich does notshow alsoother aorist\nforms ofthesame class. They are, indeed,ofvery limited use:thus,in\n3dsing.act. (endingin-ydsfor-ydst), we findavyds, cifyas, daghyds,\nbhuyds, yamyds, yuyds, vrjyds, ?ruyds, sahyds (RV. hasno3dsing, in-yat};\nandbesides these andthe2dsing,inyas(giveninthepreceding paragraph),\nwhich might beofeither formation,occur intheactive onlybhuydsam (beside\nbhuyama) andkriyasma, each once. Inthemiddle, RV.furnishes thethree\nforms grabhlsta, padistd, muclsta; nothing additional isfound inanyother\ntext. From theAV.on,theactive precative forms aremore frequentthan\nthepure optative (which arenotwholly unknown, however); they arenowhere\ncommon, exceptingasmade fromybhu;andfrom noother root isanything\nlike acompleteseries ofpersons quotable (only bhuyasva andbhuydstdm being\nwanting; andthese persons having norepresentative from any root). In-\ncluding thecases already given, theyhave been noticed asmade from about\ntwenty roots,asfollows: fruyasam etc., kriyasam etc., priydsam, bhriydsam,\nsaghydsam, bhrdjydsam, udydsametc.(j/vad), rdhydsam etc.,radhydsam etc.,\nbadhydsam etc., trpydsma, jivydsam, pusydsametc.AV.hasoncebhuydstha,\nwith primary ending, but itisdoubtless afalse reading (TB. has-stain\nthecorresponding passage). TS.hasdldydsam, from thequasi-root dldl(676).\n839. Imperative forms oftheroot-aorist arenotrare intheearly lang-\nuage. Inthemiddle, indeed, almost only the2dsing, occurs :itisaccented\neitherregularly, ontheending,asfcrsua, dhisvd, yuksvd, orontheroot,as\nmdtsva, ydksva, vdnsva, rdsva, sdksva;mdsva isnotfound with accent;the\n2dpi.isrepresented bykrdhvam, vodhvam. Intheactive,allthepersons\n(2dand3d)arefound inuse; examplesare:2dsing., krdhi, prudhf, gadhi,\nyandhf, gahi, mafti, mogdhi; 3dsing., dato, a*u,s6tu;2ddu., ddtam,\njitam, frwtam, bhutdm, sprtdm, gatam, riktdm; 3ddu., onlygantam, vodhSm;\n2d-p\\.,'gatd, bhutd, prwta, fcrta, gata, dhatana; 3dpi.,only dhdntu, fru-\nvantu. These arethemost regular forms;butirregularities both astoaccent\nandstrengthening arenotinfrequent. Thus, strong forms in2ddu.andpi.\narevarktam, vartam; kdrta, gdnta, yanta, vartta, heta, yr6ta, stita; and,\nwithtana, kdrtana, aanfana, t/antana,sotana. Much more irregular are\nyodhi (insteadofyuddhf) fromyyudh,andbodM from bothybudhandybhu\n(instead ofbuddhi andbhudht). Asingle form(3dsing.) intatisfound,\nnamely fastat.\nAsto2dpersons singular insifrom thesimple rootused inanim-\nperative sense,seeabove, 624.\nParticiples oftheRoot-aorist.\n840. Inthe oldest language,oftheE-V., arefound a\nnumber ofparticipleswhich must bereckoned asbelongingto\nthisformation.\nIntheactive, they areextremely few :namely ferdnt, sthdnt, andbhiddnt,\nandprobablyrdhdnt.\n"
+    },
+    {
+        "page": 293,
+        "content": "844]SIMPLE AORIST: PASSIVE AORIST. 277\nInthemiddle, they aremuch more numerous :examplesarearana,\nidhand, krand, jusana, drparad, prcand, bhiyand, vrand, sprdhand, hiydnd.\nSuch formations areextremely rareinthelaterVeda andintheBrahmana.\n841 .Alltogether,therootsexhibitingintheolderlanguage\nforms which arewith fairprobabilitytobereckoned totheroot-\naorist-systemarevery nearlyahundred(more thanninety);about\neightyofthemmake such forms intheRV.\nPassive Aorist third person singular.\n842.Amiddle thirdperson singular, ofpeculiar formation\nandprevailingly passive meaning,ismade frommany verbs in\ntheolderlanguage, andhasbecome arecognized part ofthe\npassive conjugation, being, accordingtothegrammarians, tobe\nsubstitutedalwaysfortheregular thirdpersonofanyaorist\nmiddle that isused inapassivesense.\n843. Thispersonisformed byadding ^itotheroot,\nwhich takes also theaugment, and isusually strengthened.\nTheendingibelongs elsewhere only tothe firstperson ;and this third\nperson apparently stands inthesame relation toafirst in iasdo,inthe\nmiddlevoice,theregular 3dsing, perfect, and thefrequent Vedic 3dsing,\npresent, which areidentical inform with theirrespectivefirstpersons. That\nafuller ending hasbeen lost offisextremely improbable; andhence,asan\naorist formation from thesimple root,this ismost properly treated here, in\nconnection with theordinary root-aorist.\n844. Before theending ^i.afinal vowel, andusually\nalsoamedial %abefore asingle consonant, have thevrddhi-\nstrengthening;other medial vowels have theguna-strengtli-\neningifcapableofit(240);after final TT#isadded Ty>\nExamples (allofthem quotable from theolder language) are :from roots\nending ina,djnayi, ddhayi, dpayi ;inother vowels, a'prai/i, dstavi, dhavi,\na/carz,astern,from roots with mediali,M,r,acefz, dchedi, apesi, dbodhi,\ndmofi, dyoji, ddarfi, asarji, varhi; from roots with medial astrengthened,\nagami, dpadi, at/ami, avaci, vapi, dsddi(theseare allthe earlier cases):\nwithaunchanged, only djani (andRV. hasoncejdni), and,inheavy syl-\nlables, dmyaksi, vandi, fansi; with medial a.dbhraji, dradhi; from roots\nwith initial vowel,ardhi (only case).\nAccordingtothegrammarians,certain roots inam,andybadh,retain\ntheaunchanged: thus, adami, abadhi; and there arenoted besides, from\nroots sometimes showing anasal, arambhi, arandhi, ajambhi, abhanji or\nabhaji, alambhi(always,with prepositions)oralabhi: QB.hasasanji.\nAugmentless forms,asinallother likecases,aremetwith,with either\nindicative orsubjunctive value: examples (besidesthetwo orthree already\n"
+    },
+    {
+        "page": 294,
+        "content": "278 XI.AORIST-SYSTEMS. 844\ngiven)are :dhdyi, frdvi, bhari, red, vedi, rod, jdni, pddi, sddi,ardhi. The\naccent, whenpresent,isalways ontheroot- syllable.\n845. These forms aremade inRV.from forty roots, and alltheother\nearlier texts combined add only about fifteen tothenumber; inthelater\nlanguage they are(likeallthekinds ofaorist) very rare.When theycome\nfrom roots ofneuter meaning,likegam, pad, sad, bhraj, sanj, etc., they\nhave(like theso-calledpassive participleinta:952)avalueequivalent to\nthat ofother middle forms\n;inacase ortwo(RV.vii.73.3[?];VS.xxviii.15)\nthey appear even tobeused transitively.\n2.The a-aorist.\n846. This aorist isinthelaterlanguageallowed tobe\nmade from alargenumber ofroots (near ahundred).Itis\nmade inbothvoices, but israre inthemiddle, most ofthe\nroots formingtheir middle accordingtothe s-class(4)or\nthe/5-class(5).\nItsclosestanalogyiswith theimperfectofthea-class\n(VII.);itsinflection isthesame with thatinallparticulars;\nand ittakes ingeneralaweak form ofroot save the\nroots inftr(threeorfouronly),which have theguna-\nstrengthening.\nAsexampleofinflection maybetaken theroot\nsic, 'pour': thus,\nactive.\nd.middle.\nd.\ndsicam dsicava dsicama dsice dsicavahi dsicamahi\ndsicas dsicatam dsicata dsicathas dsicetham dsicadhvam\ndsicat dsicatam dsican dsicata dsicetam dsicanta\n847. The a-aorist makes intheRV. asmall figure beside theroot-\naorist, being represented bylessthan halfthe latter's number ofroots. It\nbecomes, however, morecommon later(itistheonlyform ofaorist which is\nmade frommore verbs inAV.than inRV.); andinVedaandBrahmana together\nabout seventy roots exhibit theformation more orless fully. Ofthese alarge\nnumber(fully half)areofthetype ofthe roots which make their present-\nsystem according tothed-class(VII.), having avowel capableofpwia-strength-\nening before afinal consonant (753): thus,withi,chid, bhid, nij, pis,\n"
+    },
+    {
+        "page": 295,
+        "content": "849] SIMPLE AORIST: 2.-AORIST. 279\nris,vid, ifis(fds], 2pi, fvit, sic,sridh; withM,krudh, ksudh, dm, dyut,\ndruh, pus, budh, bhuj, muc, mus, yuj, rudh, muh, ruh, pwc,- withr,rdh,\nfer,grdh, trp, trs, trh,drp, dYp,mrs, vrj, vrt, vrdh, srp.Asmall number\nend invowels:thus, r, fcr, gr,sr(which have the gruna-strengthening\nthroughout),hi(?ahyat once inAV.), stu(?stuvatdm, 3dsing. impv. mid.,\nonce inAV.); and several ina,apparent transfers from theroot-class by\ntheweakening oftheir atoa:thus, Mi/a, hvd, vyd, fvd,and(inRV. only)\ndaanddha.Afewhave apenultimate nasal inthepresent andelsewhere,\nwhich inthis aorist islost:thus, bhrah?, srahs, krand, manth, randh. Of\nless classifiable character areferaw, gam, tarn, fam, fram, tan, van, san,\nsad, radh, dabh, sap, ap, IT,das, fafc. Theanomalous astham istheaorist\nofas'throw'. The rootspat, nap,vacform thetense-stemspapta, ncpa,\nvoca, ofwhich the first ispalpably andtheother twoareprobablytheresult\nofreduplication ;butthelanguage has lostthesense oftheir being such,\nandmakes otherreduplicatedaorists(3)from thesame roots(seebelow, 854).\n848. The inflection ofthis aorist isingeneral soregular that itwill\nbesufficient togive only examples ofitsVedic forms. Wemay take as\nmodel avidam, fromyvid 'find', ofwhich thevarious persons andmodes are\nmore frequent and infuller variety than those ofanyother verb. Only the\nforms actually quotable areinstanced; those ofwhich theexamples found\narefrom other verbs than vidarebracketed. Thus :\nactive. middle.\ns. d.p.s. d. p.\n1avidam dvidava dvidama dvide [dvidavahi] dviddmahi\n2dvidas[avidata] [dvidathds]\n3dvidat dvidan [avidata] [avidetdm] dvidanta\nThemiddle forms arerare intheearlier language,asinthelater: we\nhave dhveetc., dkhye etc., dvide(?)and avidanta, avocathds andavocdvahi\n(andaviddmahe GB.andasicdmahe KB. aredoubtless tobeamended tomahi).\nAugmentless forms, with indicative orsubjunctive value, arenotin-\nfrequent. Examples, showing accent onthetense-sign, according tothe\ngeneral analogiesoftheformation, areruhdm, srpas, bhujdt, viddt, aratdm,\nvocata, fakan; vidata (3d sing.), ardmahi, fisdmahi, viddnta, budhdnta,\nmrsanta (forexceptions asregards accent, seebelow, 853).\nModes ofthea-aorist.\n849. Thesubjunctive forms ofthis aorist arefew; those which occur\nareinstanced below, inthemethod which wasfollowed fortheindicative :\n1[vidava] viddma [vidamahe]\n2<,_, viddthas viddtha\n[vidas\n3viddt [viddtdi ?]\nTheending ihana isfound once,inrisdthana. Ofmiddle forms occur\nonly tfsdtdi (AV.: butdoubtless misreadingfor(fsydtdi) andfisdmahe (AV,,\nforRV.fisdmahi).\n"
+    },
+    {
+        "page": 296,
+        "content": "280 XI.AORIST-SYSTEMS.[850\n850. The optativesarefew inthe oldest language, butbecome more\nfrequent, and intheBrahman asarenot rare. Examples are :inactive,\nvideyam, saneyam ;vides, games ;garnet, vocet;gametam ;pafeema, sanema;\nvareta; inmiddle, (only) videya; gamemahi, vanemahi.\nAsingle middle precative form occurs, namely videsta (AV., once);it\nissoisolated thathowmuchmaybeinferred from itisvery questionable.\n851.Acompleteseries ofactive imperative forms aremade fromysad\n(including sadatana, 2dpi.),andthemiddle sadantam. Other imperatives\narevery rare: namely, sdra, ruha; ruhdtam, viddtam; khydta ;and, inmiddle,\nstuvatam(?).\nParticiples ofthefl-aorist.\n852. The activeparticiples trpdnt, rudhant, vrdhdnt, fisdnt, fucdnt,\nhuvdnt, savant, and(inparticipial compounds, 1309) kftant-, guhant-,\nvidant-(allRV.)aretobeassigned withplausibilitytothis aorist.\nLikewise themiddleparticiples vrdhand, sridhdnd, huvand; andperhaps\nalsovipand andcubhand, although nopersonal forms corresponding tothem\noccur.\nIrregularities ofthe -aorist.\n853.Afewirregularities andpeculiarities maybenoticed here.\nThe roots inr,which (847) show astrengtheninglike that ofthe\npresent ofthesixth orunaccented a-class, have also theaccent onthe\nradicalsyllable, like that class: thus, fromj/r,dranta (augmentless 3dpi.),\nsdrat andsdra. The rootsad follows thesame rule: thus, sddatam; and\nfromysanarefound sdnas and sdnat andsanema,beside saneyam and\nsanema. Itisquestionable whether these arenottrueanalogues ofthesixth-\nclass (unaccented a-class) present-system. Ontheother hand, ruhat(beside\nruhdm, ruhdva, ruhdtam} and rfsat(onlyaccented form) areanomalies. From\nyvac,theoptativeisaccented voceyam, vocema, voceyus ;elsewhere the\naccent isontheroot-syllable:thus, vdce, v6cat, vdcati, vdcanta.\n854. Thestem vochasinVedic usewell-nigh assumed thevalue of\naroot;itsforms arevery various and offrequent use, inRV. especiallyfar\noutnumbering inoccurrences allother forms fromyvac.Besides those already\ngiven, wefindvoca(1st sing, impv.) andvocdti, vocavahai; voces, voceya,\nvocemahi; vocatdt (2d sing.), vocatu, vocatam, vocata.\nOfthestem neya fromj/nap only ne^at occurs.\nThe root pas (asinsome ofitspresent forms :639)isweakened topz,\nandmakes afisam.\n855. Isolated forms which have more orless completely\ntheaspectofindicative presentsaremade fromsome roots be-\nside theaorist-systemsofthe firsttwo classes. Itmust beleft\nformaturer research todetermine how fartheymayberelics\noforiginal presents,andhow farrecent productions, made in\nthewayofconversion oftheaorist-stem toaroot invalue.\n"
+    },
+    {
+        "page": 297,
+        "content": "858! 3.REDUPLICATED AORIST.281\nII.(3)Reduplicated Aorist.\n856.Thereduplicated aorist isdifferent from theother\nforms ofaorist inthat ithascome tobeattached inalmost\nallcases tothederivative(causative etc.) conjugationin\nTCJdya,astheaorist ofthatconjugation, and istherefore\nliable tobemade from allroots which have such aconju-\ngation, beside the aorist oraorists whichbelongtotheir\nprimary conjugation. Since, however, theconnection of\nthetwo isnoformal one(theaoristbeing madedirectly\nfrom theroot, andnotfrom thecausativestem), butrather\namatter ofestablished association, owingtokinshipof\nmeaning,theformation and inflection ofthiskind ofaor-\nistisbest treated here, alongwith theothers.\n857. Itscharacteristic isareduplicationoftheradical\nsyllable, bywhich itisassimilated, ontheonehand, to\ntheimperfectofthereduplicatingclass(II.), and, onthe\nother hand,totheso-calledpluperfect. Buttheaorist re-\nduplicationhastaken onaquite peculiarcharacter, with\nfew tracas lefteven intheVeda ofadifferent condition\nwhich mayhaveprecededthis.\n858.Asregards, indeed, theconsonant oftheredupli-\ncation,itfollows thegeneralrulesalready given (590).And\nthequalityofthereduplicatedvowel isingeneralasinthe\nformationsalreadytreated: itneedsonlytobenoted that\nana-vowel and r(orar)areusually (forexceptions,see\nbelow, 860) repeated byan^-vowel asthey are,toa\nconsiderable extent,inthereduplicated presentalso(660).\nBut inregardtoquantity,this aorist aims alwaysat\nestablishingadiversitybetween thereduplicatingandradi-\ncalsyllables, makingtheoneheavyand theotherlight.\nAnd thepreferenceisverymarkedlyforaheavy redupli-\ncation andalightroot-syllablewhich relation isbrought\nabout wherever theconditions allow. Thus :\n"
+    },
+    {
+        "page": 298,
+        "content": "282 XI.AORIST-SYSTEMS.[859\n859. Iftheroot isalight syllable (havingashort\nvowel followed byasingle consonant),thereduplicationis\nmade heavy.\nAnd this, usually bylengtheningthereduplicating vowel,\nwith iforradical aorror/(inthesingleroot containing that\nvowel):thus, arlnsam, adudusam, ajljanam, avivrdham, aciklpam.\nThegreat majorityofaorists areofthisform.\nIf,however, theroot begins withtwoconsonants, sothat\nthereduplicating syllablewillbeheavy whatever thequantityof\nitsvowel,thevowel remains short :thus, aciksipam, acukrudham,\natitrasam. apisprcam.\n860. Iftheroot isaheavy syllable (havingalong\nvowel, orashort before twoconsonants),thevowel ofthe\nreduplicationisshort :and,inthiscase 5faor (Ta,and\nfTr(ifitoccurs),arereduplicated byfa.\nThus adidlksam, abubhusam, adadaksam, adadhavam, atatahsam.\nAnd, intherare cases inwhich arootboth begins andends\nwithtwoconsonants, bothsyllablesarenecessarily heavy,not-\nwithstandingtheshort vowel intheformer :thus, apapraccham,\nacaskandam(neither, apparently, found inuse).\nThese aorists arenotdistinguishableinform from theso-called pluper-\nfects (817ff.).\n861. Inorder, however, tobring about thefavored relation\nofheavy reduplication andlightradicalsyllable,aheavyroot is\nsometimes madelight:either byshorteningitsvowel, asin\navivacam fromy'vac,aslsadham from~\\/sad/i, adidipam (K.and\nlater: RV. hasdidlpas]fromydip,abibhisam fromybhis, asu-\nsucam from]/sue;orbydroppingapenultimate nasal, asin\nacikradam from]/krand, asisyadam from~|/syand.\nInthose cases inwhich(1047)anaorist isformed directly\nfrom acausal stem inap,theaisabbreviated toi:thus, atisthip-\nametc., ajijnipat, jihipas, ajijipata (butVS.ajijapata];butfrom\ncrapcomesacicrapama\n862. Examplesofthisaorist from roots with initial vowel arevery rare;\ntheolder language hasonlyamamat(oramamat) from]/am,andarpipam\n(augmentless) from thecausative stem arpof]/rinwhich latter theroot\nisexcessively abbreviated. Thegrammarians give other similar formations,\nasarcicam from]/arc, aubjijam from]/a&/, arjiham from|/arfe,aiciksam\nfromyiks,ardidham fromyrdh. Compare the similar reduplicationinde-\nsiderative stems: 1029 b.\n863. Ofspecial irregularities maybementioned :\n"
+    },
+    {
+        "page": 299,
+        "content": "867]3.REDUPLICATED AORIST. 283\nFrom ydyutismade adidyutam (taking itsreduplicating vowel from the\nyinstead oftheu);yplu makes apiplavam (QB. etc.).\nSome verbs with radical aorrarebythegrammarians allowed tore-\nduplicatewith either i(1}ora,oreven with aonly. Others areallowed\neither toretain orshorten along root-vowel. Details areunnecessary, the\nwhole formation being sorare, andtheforms instanced having never been\nmetwith inuse.\nAstoapaptam, avocam, andanepam,seeabove, 847.\n864,The inflection ofthereduplicated aorist islike\nthat ofanimperfectofthesecond general conjugation:that\nistosay,ithastheunion-vowel ^abefore theendings,\nwith allthepeculiarities wrhich thepresenceofthatvowel\nconditions. Thus, fromy^ftjan, 'givebirth':\nactive. middle.\ns. d.p. d.p.\nMsJlsMIH' MsflsH M\ndjijanam djijanava djljanama djyane djijanavahl djijanamahi\n^ ^\ndjljanas djyanatam djyanata djtjanathas djijanetham dfyanadhvam\nfclsilswr^ ^siWlrilH^ WlsH*^*tellstlrlWlsHcilH^fctsflswri\najijanat ajijanatam djijanan djyanata dfljanetam djljananta\n865. Themiddle forms arequiterareintheolder language\n(the3dpi.isdecidedlythemostcommon ofthem, beingmade\nfrom eleven roots;the3d s.fromseven);but allarequotable\nexcept1stand2ddu. and ofthe 1stdu.noactive example\noccurs.\nAtitape appearstobeonce used (RV.)as3dsing., withpassivesense.\n866.Afinal rhasthe^wwa-strengtheningbefore theend-\nings:thus, apiparam, atitaras, adidharat, avivaran. Ofsimilar\nstrengthened forms from iandw-roots arefound apiprayan (TS.),\nabibhayanta (RV.), apiplavam (QB.), acucyavat (K.);ofunstrength-\nened, acucruvat(GB.). Few roots endinginother vowels' than r\nmake this aorist :seebelow, 868.\n867. Forms oftheinflection without union-vowel areoccasionally met\nwith: namely, from roots ending inconsonants, sfsvap (2dsing., augmentless)\nfromysvap,andapipnat fromy$nath;from roots inrorar,didhar (2dsing.),\nandajigar (2dand3dsing.). Of3dpi.inusarefound almost only aform\nortwofrom iand w-roots, withguna before theending: thus, apipraj/ws,\nacucyavus, apwpravws, asusavus(?AB.); but also ablbhajus (QB.). And the\n3dpi.mid. avavrtran andavavrtranta andasasrgram seem tobelong here\nrather than tothepluperfect.\n"
+    },
+    {
+        "page": 300,
+        "content": "284 XI.AORIST-SYSTEMS.[868\n868. Inthe later language,afew roots aresaidbythe\ngrammarianstomake this aorist asapartoftheirprimary con-\njugation:theyarecriandcvi,druand sru,kam, anddha 'suck'\n(cvianddhaoptionally).\nIntheolder language arefound fromy$ri a$i$ret andafifrayus (noticed\ninthepreceding paragraph);from ydru,adudrot andadudruvat (TB.: not\nused asaorist); fromj/sra,asusrot and(augmentless)susros and susrot. Of\nforms analogous with these occur anumber from roots inuoru:thus,\nanunot andnunot fromynu; yuyot fromyyu 'repel'; dudhot fromydhu;\napupot from]/pu,-tutos and tutot fromytu; andoneortwofrom roots in i\nori:thus,siset fromysi (orso) 'bind'; apipres (with apiprayan,noticed\nabove) fromyprl (andthe\"imperfects\" from dldhletc.,676, areofcorres-\nponding form). Andfromycyuaremade, with union-vowelz,acucyavit\nandacucyavltana. None ofthese formspossessanecessarily causative ora\ndecidedly aoristic value, and itisvery doubtful whether they should not all\nbeassigned totheperfect-system.\nModes oftheReduplicated Aorist.\n869.Asinotherpreterit formations, theaugmentlessin-\ndicativepersonsofthis aorist areusedsubjunctively,andthey\nareverymuch more frequent than truesubjunctives.\nOfthe latter arefound only riradha (1st sing.); titapasi; ciklpati and\nsigadhati,and pisprfati (asifcorrespondingtoanindicative apisprk,like\napifnat);andperhapsthe1stsing. mid. $a$vacaf.\nTheaugmentless indicative forms areaccented ingeneral ontheredu-\nplication:thus, didharas. nma^as; jijanat, piparat; jijanan;also sisvap ;\nbut, ontheother hand, wehave fifrdthas and $i$ndthat; anddudrdvat,\npufrdvat, tustdvat(which perhaps belong rather totheperfect: compare 810).\nAccordingtothenative grammarians, theaccent rests either onthe radical\nsyllable orontheonethat follows it.\n870. Optative forms areeven rarer. The leastquestionablecase isthe\nmiddle \"precative\" ririsista(ririsista hasbeen ranked above with sasahlsta,\nasaperfect: 812b).Cucyuvimahi andcucyavwata belong either here orto\ntheperfect-system.\n871. Ofimperatives, wehave theindubitable forms pupurantu and\nCifrathantu. Andjigrtdm andjigrtd, anddidhrtam anddidhrtd(allRV.only),\naredoubtless tobereferredhither,ascorrespondingtotheindicatives (without\nunion-vowel) ajigar andadldhar :their shortreduplicatingvowel and their\naccent assimilate themclosely tothereduplicated imperfects (cl. II.),with\nwhich weareprobablytoregard this aorist asultimately related.\n872.Noparticipleisfoundbelongingtothereduplicatedaorist.\n873. Thenumber ofroots from which this aorist ismet\nwith intheearlier languageisaboutninety.Inthe classical\nSanskrit itisveryunusual;inthewhole series oflater texts\nmentioned above(826)itoccursonly once.\n"
+    },
+    {
+        "page": 301,
+        "content": "877] SIBILANT AOEIST. 285\nIII.Sibilant-Aorist.\n874.Thecommontense-signofallthevarieties ofthis\naorist isaHs(convertible to^s)which isadded tothe\nroot informingthetense-stem.\nThis sibilant hasnoanalogues among the class-signs ofthepresent-\nsystem ;but itistobecompared with thatwhich appears (and likewise with\norwithout thesame union-voweli)inthestems ofthefuture tense- system\n(chap. XII.) and ofthedesiderative conjugation (chap. XIV.).\nTotherootthus increased theaugmentisprefixed and\nthesecondary endingsareadded.\n875. Inthecase ofafewroots, the sibilant tense-\nstem(always endingin^ks]isfurther increasedbyan fa,\nandtheinflection isnearlylike that ofanimperfectofthe\nsecondgeneral conjugation.\n876. Inthevastmajorityofcases, thesibilant isthe\nfinal ofthetense-stem, andtheinflection islike thatofan\nimperfectofthe firstgeneral conjugation.\nAnd these, again,fallintotwonearly equal andstrongly\nmarked classes, accordingasthe sibilant isadded imme-\ndiatelytothefinal oftheroot, orwith anauxiliaryvowel\n^i,makingthetense-sign^w.Finally,before this\"^ts\ntheroot isinaverysmallnumber ofcases increased bya\nHs,makingthewhole additionffpST^**.\n877.Wehave, then,thefollowingclassification forthe\nvarieties ofsibilant-aorist :\nA.With endingsaddeddirectlytothesibilant:\n4.withHssimplyafter theroot: s-aorist;\n5.with^ibefore the \"Rs:*s-aorist;\n6.thesame, with 3fsatendofroot: m-aorist.\nB.With Efaadded tothesibilant before theendings:\n7.with sibilant and 5fa:Sfl-aorist.\nAsregardsthe distinction between thefourth and fifth forms,itmay\nbesaid inageneral way that those roots incline totake theauxiliaryiin\ntheaorist which take italsoinother formations;but itisimpossibletolay\n"
+    },
+    {
+        "page": 302,
+        "content": "286 XI.AORIST-SYSTEMS.[877\ndown anystrict rules astothisaccordance, andalso, considering the rarity\nofaorist-forms inthelaterlanguage, practically quite useless toattempt the\ntask. Seebelow, 903.\n4.Thes-aorist.\n878.The tense-stem ofthis aorist ismade byadding\n^stotheaugmentedroot, ofwhich alsothevowel isusu-\nallystrengthened.\n879.Thegeneralrules astothestrengtheningofthe\nroot-vowel arethese :\na.Afinalvowel(including ftr)hasthevrddhi-ch&nge\nintheactive, and(excepting ftr)gunainthemiddle :thus,\nfromy^iactive stem ERST anais, middle stem^faanes;\nfromyy^fru, wmm agratis- and 5P2Tfa agros ;from|/5Rkr,\nfaiFSfakars and *33n^ akrs.^ c-s\nb.Amedial vowel hasthevrddhi-ch&ugeintheactive,\nandremains unaltered inthemiddle :thus, fromy'^>*^ chand,\nactive stemiirH achants, middle stemtj^rHachants\n;from\naraiks andtii^rl ariks; from]/\"^I rudh,\nand5T^fHaruts; fromy'flsT srj]JblHW asraks\nand 5TH^T asr^s.\nC\"S\n880.Theendingsaretheusualsecondary ones, with\n3TTus(notRow)in3dpi.act.,and^{ata(not 3^tT wte)\nin3dpi.mid.\nButbefore Hsand cT^of2dand3dsing.act. isin\nthelaterlanguage alwaysinserted an^2,makingtheend-\nings^Tisand^rTIt.\nThis insertion isunknown inthe earliest language (oftheRV.):see\nbelow, 888.\n881. Beforeendings beginningwith torth,thetense-\nsignsis(233b)omitted after thefinal consonant ofaroot\nunless thisber,ornorm(convertedtoanusvara}.\nThesame omission isofcourse made before dhvam,after either vowel\norconsonant; andtheending becomes dhvam, providedthesibilant,ifretained,\nwould have been s:thus, astodhvam andavrdhvam (besideastosata and\navrsata), butaradhvam (beside arasata}. These three aretheonly test-cases\nfortheform oftheending which have been noted intheolder language,\n"
+    },
+    {
+        "page": 303,
+        "content": "884] SIBILANT AORIST: 4.S-AORIST. 287\nexcept drdhvam(ydr 'regard': ^B., once), which istodrthas (2dsing.) as\navrdhvam andavrsata toavriandavrthas.\nAccording tothegrammarians, theomission ofsbefore tand thtakes\nplacealso after ashort vowel(the case canoccur only inthe2dand3dsing,\nmid.);butwehave seen above (834 a)that this istobeviewed rather as\nasubstitution inthose persons oftheforms oftheroot-aorist. Neither in\ntheearlier norinthelaterlanguage, however, doesanyexample occur ofan\naorist-form with sretained after ashort vowel before these endings.\n882.Asexamplesofthe inflection ofthisvarietyof\nsibilant aorist wemaytake theroots\"^Jrudh, 'obstruct',\nand^rii,'lead'. 'Thus:\nactive. middle.\ns. d. p. s. d.p.\ndrautsam drautsva drautsma drutsi drutsvahi drutsmahi\ndrautsis drauttam drautta drutthas drutsatham druddhvam\ndrautsit drauttam drautsus drutta drutsatam drutsata\ndnaisam, dnaisva dnaisma dnesi dnesvahi dnesmahi\ndnaisis dnaistam dnaista dnesthas dnesatham dnedhvam\n^RM IHIH\ndnaisit dnaistam dnaisus dnesta dnesatam dnesata\n883. The omission ofsintheactive forms isacase ofvery rare oc-\ncurrence. Intheolder language havebeen noted onlyachantta (RV.; written\nachanta, by231), abhakta(AB.), and theaugmentless taptam and fapta\n(repeatedlyintheBrahmanas). Themiddle forms with omission arein-\ndistinguishable from thecorresponding ones oftheroot-aorist; andwhether,\nintheolder language, agiven form istobeassignedtotheoneaorist orto\ntheother isaquestiondifficult tosettle. Above (834bd)were givenall\ntheolder forms ofbelongings thus questionable.\n884. Certain roots inaweaken theainmiddle inflection\ntoi(asalso intheroot-aorist: above, 834a):these aresaid to\nbestha, da(da 'give', andda'cut'and'share'),anddha (dha\n'put', anddha'suck');intheolder language have been noted\nonlyddisi and adisata(fromda'give')and asthisata. TA. has\naglsata from]/y 'sing'.\nThemiddle inflection oftheaorist ofydawould be,then,\n"
+    },
+    {
+        "page": 304,
+        "content": "288 XI.AORIST-SYSTEMS.[884\naccordingtothegrammarians ;ddisi, ddithas, ddita;ddisvahi,\nddisatham, ddisatam; ddismahi, ddidhvam, ddisata.\n885. Roots ending inchangeabler(so-called roots inf:242) aresaid\nbythegrammarianstoconvert thisvowel toirinmiddle forms :thus, astlrsi,\nastlrsthds, etc. (from ystr): hardly anysuch forms, however, havebeenfound\nintheolder language (only akirsata, PB.; and, onthecontrary, astrsi occurs\nonce, AB.).\n886. The s-aorist ismade intheolderlanguage from some-\nwhat over ahundred roots(inRV., from about seventy;in\nAV., from aboutfifty,ofwhich fifteen areadditional tothose\ninRV.;andtheother textsaddabout twenty more notcount-\ninginanycase those ofwhich theforms maybefrom theroot-\naorist).Ithasthere certainpeculiaritiesofstem-formation andin-\nflection, andalsothefullseries ofmodes ofwhich theoptative\nmiddle isretained also inthelater languageas\"precative\".\n887. Irregularitiesofstem-formation are:\na.The roots hu,dhu,andnuhaveuinstead ofointhemiddle :thus,\nahusata, adhusata, anusi andanusatam andanusata;ydhur (ordhurv) makes\nadhursata.\nb.Fromymadoccurs amatsus (RV., once), withunstrengthened vowel.\nc.Fromygamoccurs agasmahi, apparently foragansmahi (compare\nmany a,below, 895).\n888.Theprincipal peculiarityoftheolder languageinre-\ngardtoinflection isthefrequent absence of ~iintheendingsof\n2dand3dsing, act., andtheconsequentloss oftheconsonant-\nending, andsometimes ofroot-finals(150). Theforms without\naretheonly onesfound inRV.andK.,andtheyoutnumber\ntheothers inAV. andTS.; intheBrahmanastheygrowrarer\n(only one, adrak, occurs inGB.andQB.; PB.hasnone).\n889. Iftherootends inavowel, only theconsonant oftheendingis\nnecessarily lost: thus, aprds (forboth aprds-s and apras-t] fromyprd:and\ninlikemanner ahas fromyhd; ajais (for ajdis-t]fromyji;andinlike\nmanner acdis fromyd,andndis (augmentless) fromym;andydus (for\naydus-t) fromyyu.\nBat(asinother likecases :555)in3dsing, theendingtissometimes\npreservedattheexpense ofthetense-sign; andwehaveajdit (beside ajais\nandajdmt) fromyji ;andinlikemanneracdit, afrdit, ahdit, ndit :noex-\namples have been noted except from roots in iand I.\n890. Iftheroot(ineither itssimpleorstrengthened form) ends ina\nconsonant, thetense-signislostwith theending. Thus, abhdr (forabhdrs-t :\nbeside abharsam, abhdrstdm] fromybhr;other likecases areahar, and(from\nroots inar)aksar, atsar, asvar,hvar. Further, drdik (585, end: forardiks-t)\nfromyric;like cases areafvdit fromycvit,and (from roots with medial)\nadydut fromydyut,ardut fromyrudh,andmduk fromymuc. Further,\nfrom roots ending inthepalatals andft,aprdk from]/prc,asrdk from|/sr;,\n"
+    },
+    {
+        "page": 305,
+        "content": "894]SIBILANT AORIST :4.S-AORIST. 289\nabhdk fromybhaj,adrdk from}/drf,adhdk fromydah; but, with adifferent\nchangeofthefinal, ayat fromyyaj, apra/ fromj/prc/z,andavatfrom|/uafe ,\nand (above, 146)srasappearstostand twice inAV. forsras-t fromysrj:\nRV. hasalsotwice ayasfromyyaj. Further, from roots ending inanasal,\natan fromj/taw,khan fromykhan, aydn andanan fromyyyamandnam\n(143).\nIf,again, therootends inadoubleconsonant, thelatter ofthetwo is\nlost along with tense-sign andending: thus, achan(forachdnts-t; beside\nachantta andachdntsus] fromychand;andother like cases areakrdn, asfcan,\nandasydn.\nArelic ofthis peculiarityoftheolder inflection hasbeen preservedto\nthe later language inthe2dsing, bhdis, fromybhi.\n891. Irregularities ofinflection donotoccur further: avesam fromym\nistoogreat ananomaly tobeaccepted.\nModes ofthes-aorist.\n892. The indicative forms without augmentareused ina\nsubjunctive sense, especiallyafter mtiprohibitive,and arenot\nuncommon. Exampleswith accent, however, areextremelyrare\n;\nthere hasbeen noted only vdnsi, middle;judgingfromthis,\nthetone would befound ontheradicalsyllable. According\ntotheHindu grammarians,itmay belaidoneither root or\nending.\n893. Proper subjunctive forms arenotrare inRV., but\naremarkedlylesscommon inthe later Vedic texts, and very\nseldom metwith intheBrahmanas. Theyareregularly made\nwith^wm-strengtheningoftheradical vowel, inboth active and\nmiddle, andwith accent ontheroot.\nTheforms with primary endings are :inactive, stosdni;darsasi;nesati,\nparsati, pdsati, matsati, yosati, vaksati, saksati; dtisathas, dhdsathas, pdrsathas,\nvaksathas, varsathas;pasatas, yaihsatas, yaksatas,vaksatas;dhdsatha, nesatha,\npdrsatha, mdtsatha; inmiddle, narhsdi, mahsdi; mdhsase;krarhsate, trdsate,\ndarsate, mdhsate, yaksate, rdsate, vahsate, sdksate, hdsate; trdsdthe(not\ntrdsdithe,asweshould ratherexpect); ndmsahte, mdhsante: and, with the\nfuller ending in3dsing., mdsdtdi.\nTheforms with secondary endings are(active only): jesas, vdksas; ddrsat,\nnesat, pdksat, ydksat, yosat, vdhsat, vdksat, vesat, sdtsat, chantsat,etc.\n(14others); yaksatdm; stosdma; parsan, yamsan, yosan, rasan, vaksan, cesan,\nfrdsan. Ofthese, yaksat andvaksat arefound notrarely intheBrahmanas;\nany others, hardly more thansporadically.\n894. Ofirregularities aretobenoted thefollowing:\na.Theforms dfksase andprksase (2dsing, mid.)lackthe#wna-strength-\nening.\nb.Jesam, stosam, andyosam (AV. yusam,withuforoasinanusata\nWhitney, Grammar. 19\n"
+    },
+    {
+        "page": 306,
+        "content": "290 XI.AORIST-SYSTEMS.[894\netc.) appeartobe first persons formed under government oftheanalogy of\nthesecond and third unless they arerelics ofastate ofthings anterior\ntothevrddhi- strengthening: inwhich casejesma istobecompared withthem\n(weshould expect jdisma orjesdma}.\nc.From roots indaremade afewforms ofproblematic character :\nnamely, yesam (onlycase inRV.), khyesam, jnesam, gesam andgesma, desma,\nsthesam and sthesus. Their value isoptative. The analogy ofjesam and\njesma suggeststhepossibility oftheir derivation from i-forms ofthea-roots;\northe sibilant might beofaprecative character(thus, yd-i-s-am).That\nthey aretobereckoned tothe is-aorist appears highly improbable.\nd.TheRV. hasafew difficult first persons middle inse,which are\nperhaps bestnoted here. They are: 1.from thesimple root, krse,hise(and\nohise?),stuse;2.from present-stems, arcase, rnjase, yajase, gdyise, grnue\nandpunlse. They have thevalue ofindicativepresent. Compare below, 897.\n895. Optative forms ofthis aorist aremade inthemiddle only, and\nthey have in2dand3dsing, always theprecativesbefore theendings.\nThose found tooccur intheolder language are :disiya, dhisiya, bhakslyd,\nmasiya (formahsiya], muksiya, rdsiya, sdksiya, strislya; mansisthds; darslsta,\nbhaksista, manslsta, mrkslsta; bhaksimahi, dhukstmdhi, mahslmdhi, vahsimdhi,\nsaksimdhi; mahslrata. PB.hasbhuksisiydj which should belong toasz's-aorist.\nTheRV.form trdsltham(fortrdslydthdm ortrdsdthdm)isanisolated anomaly.\nThis optative makes apartoftheaccepted \"precative\"ofthe later\nlanguage: seebelow, 921 ff.\n896. Imperative persons from this aorist areextremely rare :wefind\nonly the2dsing. act.nesa andparsa (both from a-stems, andshowing\nrather, therefore, atreatment oftheaorist-s'tem a,saroot), andthe3dsing,\nmid. rdsatdm andpi.rdsantdm(ofwhich thesamemaybesaid).\nParticiples ofthes-Aorist.\n897. Activeparticiples areddksat ordhdksat, andsdksat (both RV.).\nIfrnjase (above, 894 d)istobereckoned asans-aorist form, rnjasdnd\nisans-aoristparticiple ;and ofakindred character, apparently,arearfasdndj\ntihasdna, jrayasdnd, dhiyasdnd, mandasdnd, yamasdnd, rabhasdnd, vrdhasdnd,\nsahasdnd, cavasdnd,allinRV.; withnamasdnd, bhiydsdna,inAV.\n5.The ^5-aorist.\n898.The tense-stem ofthis aorist adds thegeneral\ntense-sign Hsbyhelpofaprefixed auxiliary vowel^,\nmaking ^Tis,totheroot, which isusually strengthened,\nandwhich hastheaugment.\n899.The rules astothestrengtheningoftheroot are\nasfollows:\n"
+    },
+    {
+        "page": 307,
+        "content": "902]SIBILANT AORIST: 0.^61-AORIST. 291\na.Afinalvowel hasvrddhi intheactive, andgunain\nthemiddle:thus, ^Mlf^N apavis andSOT^R\" apams fromyq\npu;WH^^itaris.act., fromyc[#V^lilUlN p<W, mid.,\nfromy5TTf.\nb.Aninterior vowel hasguna,ifcapableofit,inboth\nvoices: thus, ti^fem ale$is,act.andmid., from yf^5T lig;\n~x r*~ . r^r\"^\ni(llT(qarocts fromy\"^tf-rwc; Njq&j avarsts from\nbut CfeiV|G|fc| ajlvis fromysffojw.\nc.Medial faissometimeslengthenedintheactive;\nbut itmoreusuallyremains unchangedinboth voices.\nThe roots intheolder language which show thelengthening arekan,\nsan, ran, stem, tans, vraj, vad,mad, car, tsar, hvar, jval, das.From svan\nandsahoccur forms ofboth kinds.\n900. Ofexceptions maybenoted:yrnr?has(aselsewhere :627) vrddhi\ninstead ofguna: thus, amarjisam; ystrhasastaris, andj/frhasaparit (also\nafaraitinAV.), withguna inactive.\nThe rootgrabhorgrah has(asinfuture etc., below, 935d,956)long\niinstead of ibefore the sibilant :thus, agrabhisma, agrahlsta, agrabhlsata.\nThe roots inchangeabler(so-calledroots inf:242), andyvraresaidby\nthegrammarians todothesame optionally; butnoforms with longIfrom\nsuch roots arefound intheolder language.\n901.Theendingsareasintheprecedingformation\n(3Husand^rTatain3dpi.).But in2dand3dsing.,\nthecombination is-sand is-tarefrom the earliestperiod\nofthelanguagecontracted into\"$3isand^rT^.\nThe2dpi.mid. should endalwaysinidhvam(oriddhvam,\nfrom is-dhvam :232);and this isinfact theform intheonly\nexamples quotable from theolder language, namely ajanidhvam\nandartidhvam andaindhidhvam;butthegrammarians giverules\nbywhich thelingual dhisoptional only, and that after ipre-\nceded byg,r, I,v,h.\n902.Asexamplesoftheinflection ofthe ^f\ns-aorist may\nbetaken theroots^pu,'purify', and^Uf^budh,'awake1\n.\nThus :\nactive. middle.\ns. d. p.s. d. p.\ndpavisam dpavisva dpavisma dpavisi dpavisvahi dpavismahi\n19*\n"
+    },
+    {
+        "page": 308,
+        "content": "292 XI.AOBIST-SYSTEMS.[902\ndpavis dpavistam dpavista dpavisthas dpavi&atham dpavidhvam\napamt dpavistam dpavisiis dpavista dpavisatam dpavisata\ndbodhisam dbodhisva dbodhisma dbodhisi dbodhisvahi dbodhismahi\netc. etc. etc. etc. etc. etc.\n903. Thenumber ofroots from which forms ofthisaor-\nisthave been noted intheolder languageisabout ahundred\nandtwenty (inRV.,abouteighty;inAV.more thanthirty,of\nwhich adozen areadditional tothose inRV.\n;intheother texts,\nnearly thirty more). Amongthese arenoroots ina\n;butother-\nwise theyareofevery varietyofform(rarestinfinal iandi).\nActive andmiddle personsarefreely made;butvery sparingly\nfrom thesame root(onlyabout fifteen roots have both active\nandmiddle forms, and ofthese apart only exceptionallyinthe\nonevoice ortheother).\nNoruleappearstogovernthechoice ofusage between the\nisandthes-aorist;and innosmallnumber ofcases(morethan\nafifth ofall),thesame rootshows forms ofboth classes.\n904. Irregularities oftheolder language aretobenoticed asfollows :\na.The contracted formsaferarram, agrabhlm, andavadhtm(with aug-\nmentless vddhim] arefound in1stsing.act.\nb.For afarlt occurs inAV. a$arait; also(inapartoftheMSS.)\nfarais forpans:agrahaisamisfound inAB.(alsothemonstrous form\najagrabhaisarh:see801i).\nc.Fromyvadisfound vadisma (once, AB.),with short root-vowel. AV.\nhasnudisthas, without guna.\nd.Theforms atarima (RV.) andavddiran(AV.), though they lack the\nsibilant,areperhapstobereferred tothis aorist.\ne.Ajayit,with short iintheending, occurs inTS.\nModes ofthe/s-aorist.\n905. Asusual, augmentless indicative forms ofthis aorist aremore\ncommon than proper subjunctives. Examples, ofallthepersons found to\noccur (and includingalltheaccented words), are, intheactive: pdrmsam,\nvddhim; mathis, vddhls, ydvls, sdvls; dvlt, jurvit, mdthit, vddhlt, vef.lt;\nmardhistam, dosistam, hihsistam; avistdm, jdnistam, bddhistdm; framima,\nvadisma; vadhista aridvadhistana, mathistana, hihsista; hvdrisus, grahisus ;\ninthemiddle :rddhisi;jdnisthds, marsisthas, vyathisthds ;krdmista, janista,\npavista, prdthista, mdndista; vyathismahi. The accent isontheroot-syllable\n(tarisus, AV. once,isprobablyanerror).\n"
+    },
+    {
+        "page": 309,
+        "content": "911]SIBILANT AORIST: 5.^S-AORIST. 293\n906. Ofsubjunctiveforms withprimary endings occur only the1stsing,\nact.davisani, and the1stpi.mid.(with unstrengthened e)yacisamahe and\nsanisamahe.\nForms withsecondary endings arealmost limited to2dand3dsing. act.\nThere arefound :avisas, kanisas, tarisas, raksisas, vddhisas, vddisas, vesisas,\nfahsisas ;karisat, jambhisat, josisat, tarisat, nfndisat, pdrisat, Mdhisat,\nmdrdhisat, yacisat, yodhisat, raksisat, vanisat, vyathisat, fahsisat, sanisat,\nsavisat. They aremade,itwillbenoticed, with entireregularity, byadding\natothetense-stem inisbefore theendings. Theonly other persons found\ntooccur arethe3dpi. act. sanisan andmid. sdnisanta, which are also\nregular. Bhavisat (AB. once)isasolitary example ofaform with double\nmode-sign. Theradical syllable always hastheaccent, and itsvowel usually\naccords with that oftheindicative :butwehave san- inthesubjunctive\nagainst asanisam(astocay-andran-,seebelow, 908).\n907. Themiddleoptativeofthis aorist alsoforms apartoftheac-\ncepted \"precative\" ofthelater language.Itisvery rare atallperiods, being\nmade inRV.from onlyfiveroots, and inAV.from two ofthesame and\nfrom three additional ones(sixoftheeight have otherw-forms); andthe\nremaining textsadd,sofarasnoticed, only four other roots. Alltheforms\nfound tooccur are asfollows:janislya, indhislya, edhislyd, rucislya and\nrocisiya, gmisiya; modisisthas;janisista, vanislsta;sahisivahi- edhislmdhi,\njanisimahi, tarislmahi, mandisimahi, vandisimdh^ vardhisimdhi, sahislmahi\nandsahislmahi. The accent isontheending, and thiswould leadusto\nexpectaweak form ofrootthroughout ;buttheusage inthisrespect appears\ntobevarious, andthecases aretoofew toallow ofsetting upany rule.\n908. Ofimperative forms, wehavefromyavaseries :namely, aviddhf,\navistu, avistdm, avitd(ifthis,asseemsprobable,stands anomalously for\navistd) andavistdna :two ofthese are ofunmistakably imperative form.\nOther forms occur only in2ddu.and2dpi.,and areaccordingly such as\nmight alsobesubjunctives used imperatively (whichisfurther madeprobable\nfortwo ofthem bytheir accentuation ontheroot-syllable): they arekram-\nistam, gamistam, cayistam (against acayisam], tSristam, yodhistam, vadhistam,\npnathistam; rdnistana (against aranisus), pnathistana.\n909.Nowords havingaparticipial endingafter isarefound\nanywheretooccur.\n910. This istheonlyaorist ofwhich forms aremade in\nthesecondary anddenominative conjugations:seebelow, chap.\nXIV. (1019, 1035, 1048, 1068).\n6.The sss-aorist.\n911. Accordingtothegrammarians,thisaorist ismade\nfrom roots in5(Ta(including ftmi'establish', ftmi'di-\nminish', and ^ft ll'cling',which substitute forms ina),and\n"
+    },
+    {
+        "page": 310,
+        "content": "294 XI.AORIST-SYSTEMS.\nfrom^t^nam, FTj/#m,andJT^ram,and isusedonlyinthe\nactive; thecorresponding middlebeingofthes-form(4).\nItsinflection ispreciselylike that ofthe^s-aorist\n;itisun-\nnecessary, then,togivemore than itsfirstpersons, which\nwemayformfrom theroots ETTyci, 'go',and^nam, 'bend'.\nThus:\ns. d.p.s. d.p.\ndyasisam dyasisva dyasisma dnamsisam dnamsisva dnamsisma\netc. etc. etc. etc. etc. etc.\n912. The szs-aorist isproperly only asub-form ofthe zs-aorist, having\nthetense-sign andendings ofthelatter added toaform ofroot increased\nbyanadded s.Itisofextreme rarity intheolderlanguage, being made\ninRV.onlyfrom theroots gaandyd,andinAV.onlyfromfta,andprobably\nfrompyaandvan(seebelow, 914): theremaining texts addjna (TB.) and\ndhd(QB.);theforms adrdszt (K.)and ahvdslt (GB.) might\"beeither from\nthis orfrom the s-aorist.\nTheparticiple hdsamdna and causative hdsayanti (RV.) show thathas\nhadassumed, even atavery early period, thevalue ofasecondary root\nbeside hdforother forms than the aorist.\n913. Thewhole series ofquotable indicative forms isasfollows:\naydsisam, adhdsisam; agdsls ;agdslt, aydsit (and adrdslt and ahvdslt?);\naydsistdm; ajndsisma; ajndsista, aydsista; agdsisus, aydsisus (dksisus is\nfromyaks}.\nForms without augment arethese: hdsisam;hdsls;hdslt, ydslt; hdsistam;\nhdsistdm; hdsista; hdsisus, gdsisus. The accent would doubtless beupon\ntheroot-syllable.\n914. Ofproper subjunctives arefound two, gdsisat andydsisat (both RV.).\nOptativesarenot less rare :namely, ydsislsthds andpydsisimahi (for\nwhich theAV.manuscripts readpydfisimahi,altered intheedition topydyis-);\nand doubtless vahfisiya (AV., twice)istobecorrected tovohsisiya, and\nbelongs here.\nThe accent ofydsistdm (like avistdm, 908) shows ittobeatrueim-\nperative form; andydsista (RV., once)isdoubtless thesame, withanomalous\nIfor i.\n915. Middle forms ofthisaorist,itwillbenoticed, occur from the\noptative only; but, considering thegreat rarity ofthewhole formation, we\narehardly justified inconcluding that intheancient languagethemiddle\npersons in-sisi, -sisthds, etc., were notallowable,likethose in-isi, -isthds,\nandtheothers ofthei-aorist.\n7.The sa-aorist.\n916. Inthe laterlanguage,theroots allowed toform\n"
+    },
+    {
+        "page": 311,
+        "content": "920]SIBILANT AORIST: 7.sa-AomsT. 295\nthis aorist endinS^p, E[^,or^h--allofthem sounds\nwhich incombination with thetense-sign make ftks;and\ntheyhave^i,3u,orftrasradical vowel.\nThey are :dip,rif,Zip,uip, A;i'p, femp, mp, mrp, #prp ,tvis, dvis, flis,\nvis, krs; dih, mih, lih,guh, duh, ruh, trh, wft,strh(Kielhorn). Some of\nthemmay,orwith certain meanings must, take aorists ofother forms. And a\nfewareallowed todrop bothtense-sign andunion-vowel aincertain persons\nofthemiddle :thatis,theymaymake instead forms oftheroot-aorist(1).\n917.Asthetense-stem ends in5j7the inflection is\ninthemain like that ofanimperfectofthesecondgeneral\nconjugation. But(accordingtothegrammarians:theforms\nunfortunately have notbeen found intheolderlanguage)\nthe 1stsing.mid. ends in^iinstead of^e,and the2d\nand3ddu.mid. inETTSfFT atham and*JIHH atam,asinimper-\nfects oftheotherconjugation. Both active andmiddle in-\nflection isadmitted. The root isthroughout unstrength-\nened.\n918.Asexampleofinflection wemaytake theroot\n$pt dip, 'pointout'. Thus :\nactive. middle.\ns. d. p.s. d. p.\nddiksam ddiksava adiksama ddiksi ddiksavahi ddiksamahi\nSff^TrT2\nddiksas ddiksatam ddiksata ddiksathas ddiksatham ddiksadhvam\nddiksat ddiksatam ddiksan ddiksata ddiksatam ddiksanta\n919. Intheearlier language,theforms ofthesa-aorist arehardly more\nthansporadic, and arewithmuch probabilitytoberegarded astransfers of\nthes-aorist toaninflection after themanner ofana-stem. They aremade\ninRV.from eight roots;inAV.,from twoofthese andfrom twoothers;\nand theremaining texts add sixmore, making sixteen inall.Aslater,\nallhave ioruorrasroot-vowel, and afinal consonant which combines\nwith stoks;butthere areinthe listalsotwoending inj(unless theforms\nascribed tomrjand vrjbeforced under mrpandvrh).Alltheexamples\nnoted aregiven below.\n920. Inthe indicative, wefind, inthe active: avrksam; adruksas,\naruksas, asprksas; adiksat, amiksat, aviksat, dkruksat, aghuksat, aduksat and\n"
+    },
+    {
+        "page": 312,
+        "content": "296 XI.AORIST-SYSTEMS.[920\nddhuksat, druksat, akrksat, dmrksat, dsprksat; aghuksatdm; aruksama, amr-\nksdma, avrksdma; ddhuksan, apiksan (j/pis), aruksan;inthemiddle, only\nakrksathas(]/Ars), adhuksata, andamrksanta.\nForms without augment (notrue subjunctives occur) are, intheactive :\ndrksam, mrksam; duksas, ruksas, mrksas; dviksat;mrksata; dhuksdn and\nduksan; inthemiddle, dviksata, duksata anddhuksata, dhuksdnta.\nThere arenooptative forms.\nImperative are :intheactive, mrksatam;inthemiddle, dhuksdsva.\nThefewaccented forms without augment which occur have thetoneon\nthetense-sign sd,inanalogy with thea-aorist(2)andtheimperfectofthe\na-class(VII.):asingle exceptionisdhtiksata, whichprobably needs emend-\nation todhuksata.\nTheaspirationofinitial dandp,after lossoftheaspirated qualityof\ntheroot-final(155),isseen informs from the roots duhandguh, butnot\nfrom druh(onlyasingle case, AB.); RV., however, hasalsoaduksat and\nduksas, duksan, duksata.\nPrecative.\n921.Astheso-called precativeisallowed bythegram-\nmarians tobemade inthelater language fromevery root, and\ninanindependent way,without reference tothemode ofform-\nation ofthe aorist from thesame root,itisdesirable toput\ntogetherhere abrief statement oftherules givenfor it.\n922.Theprecativeactive ismadej byadding|theactive\nprecative endings (above, 568) directlytotheroot. But :\na.Offinalroot-vowels(asbefore thepassive-sign yd:770),\nianduarelengthened;risusually changedtori,but toir\nandurinthose roots which elsewhere show irandurforms\n(so-called f-roots:242),and toarinrandsmr;aischanged\ntoeintheroots da,dha, stha,pa'drink', go,'sing', andafew\nothers, inpart optionally.\nb.The root ingeneral assumes itsweakest form :apenulti-\nmate nasal islost, asinbadhyasam fromybandh;theroots\nwhich areabbreviated intheweakpersonsoftheperfect (794)\nhave thesame abbreviation here, asinucyasam, ijyasam, vidhya-\nsam, supyasam, grhyasam; yeasformscisyasam (compare 639,\n854):andsoon.\nInthe older language also, ashasbeen seen above(838), precative\noptative forms aremade inthe active only forthe root-aorist, and ina\nmanner accordant with thathere described.\n923.Theprecativemiddle ismade byaddingthemiddle\nprecative endings (above, 568)totherootincreased byHs\n"
+    },
+    {
+        "page": 313,
+        "content": "925]PRECATIVE. 297\nor\"$&ts--that is,tothetense-stem ofans-aorist orof\naneis-aorist(butwithoutaugment).\nThe root isstrengthened accordingtotherules that\napplyinformingthemiddle-stem ofthe sandofthe is-\naoristsrespectively:ingeneral, namely,afinal vowel is\ngunatedinboth formations;butamedial vowel, onlybe-\nfore^Tis.\nOther minor rules itisunnecessarytorepeathere.\nIntheolder language,ashasbeen pointed outindetail above, prec-\native optative forms ofthemiddle voice areoftenest made from thes-aorist\n(895) andthew-aorist(907); but alsofrom the root-aorist(838), thea-\naorist(850), thereduplicatedaorist(870), and the w-aorist(914); and\neven from theperfect (812 b).\n924.Asexampleofinflection, wemaytake theroot\nHbku, 'be',which issaid(nomiddle aorist orprecativeis\nmade from itintheolderlanguage)toform itsmiddle on\nthes-stem .Thus :\nactive. middle.\ns. d. p. s. d. p.\nHtllt-H\nc\\_ ,\"sc\\_ , <r\\_ f _ ._ . .\nbhuyasam bhuyasva bhuyasma bhavisiyd bhaviswdhi bhavisimdhi\n_._ f _ , f_\nbhuyas bhuyastam bhuyastabhavisisthasbhavisiyasthambhavisidhvdm\nHUIHIH H^llHHcx\nt\"^ SNs_r\"^\n._ L\nbhuyat bhuyastam bhuyasusbhavislstdbhavisiyastambhavisirdn\nAccordingtothegrammarians, thedental orlingual character ofthe\ninitial ofthemiddle ending dhvam depends uponhow the aorist tense-sign\nispreceded:inthes-form,itisdhvam ifthe siispreceded byanyother\nvowel than aord;inthe z's-form,itmaybeoptionally dhvam ifthe isis\npreceded byy,r,I\n,v,orh.Thisseems wholly irrational :thetrue question\nis,whether theprecativesistoberegarded asreally presentin2dpi.mid.,\nasinallother 2dpersonsofboth voices :ifso which isaltogether prob-\nable, but intheabsence ofquotable forms from theolder language cannot\nbepronouncedcertain theendingisnecessarily andalways dhvam.\n925. Theprecativeisaform ofrare occurrence inthe\nclassical language. Ineach ofthetexts already more thanonce\nreferred to(Manu, Nala, Bhagavad-GIta, Qakuntala, Hitopadeca)\n"
+    },
+    {
+        "page": 314,
+        "content": "298XI.AORIST-SYSTEMS.[025\nitoccurs onceandnomore. Itsvalue, asalreadystated(573),\nispurely optative:thus, sarvarastresv idam vacobruyasta (Nala),\n'[Ibegthatyou] speakthese words inallkingdoms'.\nUses oftheAorist.\n926. The uses ofthe aorist mode-forms(ashasbeen\nalready pointedout :582) appeartoaccord ingeneral with those\nofthemode-forms ofthepresent-system. Thepredilection of\ntheearlier language,continued sparinglyinthelater, forthe\naugmentlessforms inprohibitive expressionafter mU,was suffi-\ncientlystated and illustrated above(579).\nThetense-value ofthe aorist indicative hasalsobeenmore than once\nreferredto,and calls only forsomewhat more ofdetail and for illustra-\ntion here.\n927 .The aorist ofthelaterlanguageissimplyapret-\nerit, equivalenttotheimperfect andperfect,andfrequently\ncoordinated with them.\nThus, tatah sagardabham lagudena tddaydmdsa; tendJsdupancatvam\nagamat (H.), 'thereupon hebeat thedonkey with astick;andhereof the\nlatter died'; tatah savidarbhdn agamat punah ,tarn tubandhujanah sama-\npujayat (MBh.), 'thereuponshewent back toVidarbha; andherkindredpaid\nherreverence'; pritimdn abhut: uvdca catJnam (MBh.),'hewas filled with\naffection, and said tohinr.\n928. The aorist oftheolder language hasthevalue ofa\nproper \"perfect\":thatis,itsignifies something pastwhich is\nviewed ascompletedwith reference tothepresent ;and itrequires\naccordinglytoberendered byourtensemade with theauxiliary\nhave. Ingeneral,itindicates what hasjusttakenplace ;and\noftenest something which thespeakerhasexperienced.\nExamplesfrom theVeda are :pdrl 'megdm anesata pdry agn'fm ahrsata,\ndevesv akrata frdvah kdimdn d~dadharsati(RV.),'these herehave ledabout\nacow, theyhave carried around thefire, theyhave done honor tothegods\nwho shall venture anything against themVyamdichdma mdnasd sb'yam\na'gat (RV.), 'hewhom we(formerly, impf.) sought withourmind has(now,\naor.) come'; yene'ndro havtsdkrtvy dbhavad dyumny uttamah, idam tadakri\ndevd asapatndhkilo,'bhuvam(RV.),'that libation bywhich India, making it,\nbecame(impf.)ofhighest glory,Ihavenowmade, yegods;Ihavebecome\nfreefrom enemies'.\nExamples from theBrahmanalanguage are :sdha'smih jydguvdsa ...\ntdtohagandharvah sdm udire:jydgvdiydm urvdfl manusyesv avatslt(QB.),\n'shelived withhim along time. Then theGandharvas said tooneanother,\n\"thisUrvac.i, forsooth, hath dwelt along timeamong mortals'\"; tasyaha\ndantah pedire: tarnhoJvaca: apatsata vdasyadantdh (AB.),'histeeth fell\n"
+    },
+    {
+        "page": 315,
+        "content": "928]USES OFTHEAORIST. 299\nout.He said tohim: \"his teeth truly have fallenout'\"; indrasya vrtrdm\njaghnusa indriydrh viryam prthivim dnuvyhrchat taddshadhayovlrudho\nlbhavan sdprajapatim upa'dhdvad vrtrdm mejaghnusa indriydm viryam\nprthivim dnuvyarat taddshadhayo vlrudholbhuvann iti(TS.),'ofIndra,\nwhen hehad slain Vritra,the force andmight went away intotheearth,\nandbecame theherbs and plants:heran toPrajapati, saying:\"my force\nandmight,after slaying Vritra,have gone away into theearth, andhave\nbecome theherbs andplants'\"; svaydm enamabhyudetya bruydd vrdtya kvh\n'vatslh (AV., inprose passage), 'going uptohim inperson,lethim say:\n\"Vratya, where hastthou abode\"?' yddiddnlm dvauvivddamandv eydtam ahdm\nadar^am ahdm a^rdusam itiydevdbruydd ahdm adar?am ititdsmd evd\nyrdddadhydma (QB.),'ifnow twoshould come disputing with oneanother,\n[the one] saying\"Ihave seen\", [the other]\"Ihaveheard\", weshould believe\ntheonewho said \"Ihave seen'\".\n929. This distinction ofthe aorist from theimperfect andperfect as\ntenses ofnarration isverycommon intheBrahmanas, and isclosely observed :\nneglect ofitisvery rare, and istoberegarded aseither due tocorruption\noftext orindicative ofalateorigin.\nIntheVedic hymns,thesame distinction isprevalent, but isboth less\nclear and less strictly maintained :many passages would admit aninter-\npretation implying either sense; andevident aorist-forms (especiallyofthe\nsimple aorist: Delbruck)aresometimes usednarratively,while imperfect-\nforms arealso occasionally employedintheaorist sense.\n930. Theboundary between thatwhich hasjustbeen andthatwhich\nnow isisoccasionally overstepped, andtheaorist becomes nearly theequiv-\nalent ofapresent. Notvery rarely, intheVeda,itisconvenient torender\ntheformer asifitwere the latter; and intheBrahmana thesame istrue\nespecially oftheaorist a/car.\nCHAPTER XII.\nTHEFUTURE-SYSTEMS.\n931.THE verb hastwo futures,ofverydifferent age\nandcharacter. Theonehasfortense-signasibilant follow-\nedbyITyet,and isaninheritance from thetime ofIndo-\nEuropean unity. Theother isaperiphrastic formation, made\nbyappending anauxiliaryverb toaderivative noun of\n"
+    },
+    {
+        "page": 316,
+        "content": "300 XIl!FUTURE-SYSTEMS.[931\nagency,and itisarecent addition totheverb-system;its\nbeginnings onlyaremetwith intheearliest language. The\nformer maybecalled thes-future(ortheoldfuture, or\nsimplythefuture,; the latter maybedistinguishedasthe\nperiphrasticfuture.\nI.The s-future.\n932.Thetense-signofthisfuture isthesyllable HT\nsyd,added totheroot eitherdirectlyorbyanauxiliary\nvowel^i(inthe latter casebecoming ^ET isyd). The root\nhasthe^wwa-strengthening. Thus, from|/^Tdaisformed\nthefuture tense-stem ^THT dasyd; from]/^i,thestem^T\nesyd ;fromy~^> d*uh, thestem J7??Jdhoksyd; fromyHbhu,\nthestemHf^W Wiavisyd;fromy'fJEJ rdh, thestem ^[MCVU\nardhisyd;andsoon.\nButfromyjwthestem isjivisyd,fromyuks itisuksisyd,\nand soon(240).\n933. This tense-stem isthen inflectedpreciselylikea\npresent-stem endingin fa(second general conjugation).\nWemaytakea#models ofinflection thefuture of]/^Tda,\n'give',andthat of]/5Rkr,'make'. Thus:\nactive,\nd.middle,\nd.\n^TRT\ndasyhmi dasyhvas dasytimas dasye dasytivahe dasyamahe\ndasyasi dasydthas ddsydtha dasydse dasyetfie dasyddhv\nddsydti dasydtas ddsydnti ddsydte ddsyete dasydnte\nkarisyfimi karisyhvas karisyamas karisye karisyavahe karisyamahe\netc. etc. etc. etc. etc. etc.\n934.With regardtotheuseornon-use oftheauxiliary\nvowel ibefore thesibilant, there isadegreeofgeneralaccord-\nance between this tense andtheother future andthedesidera-\n"
+    },
+    {
+        "page": 317,
+        "content": "935] THE^-FUTURE. 301\nlive\n;but itisbynomeans absolute, norareanydefinite rules\ntobelaiddown withregard toit(andsomuch theless, because\noftheinfrequencyofthetwo latter formations inactualuse):\nbetween thisandtheaorist(s-aorist ontheone side, or/s-aorist\nontheother), anycorrespondenceisstill less traceable. Prac-\ntically,itisnecessarytolearn, asamatter ofusage, howany\ngivenrootmakes these variouspartsofitsconjugational system.\n935. Below isadded astatement oftheusage,asregards theauxiliary\nvowel,oftheroots observed toform thes-future intheolder language (more\nthan ahundred andfifty:thecollection isbelieved tobetolerably complete)\nforthemostpart, intheform ofaspecificationoftheroots which add\nthetense-sign directly tothe root; inbrackets arefurther mentioned the\nother roots which according tothegrammariansalso refuse theauxiliary\nvowel.\na.Ofroots ending invowels, thegreat majority (excepting those inr)\ntake no i.Thus,allina(numerous, andunnecessary tospecify);allin i\ni,ksi, ci,ji except pri[and cvi] ;allinIkri, ni, bhi,ml,vU\nexceptci[and dl];allinucyu, dru, plu,cm,druexceptsu'press',\nandstu,which follow either method,asstosyami andstavisyami [and except\nksu, ksnu, nu,yu<ru,snu].But allinr(numerous, andunnecessary to\nspecify) take i[andthose inchangeable r,orso-calledf-roots (242) are\nsaid totake either iorI;noI-forms, however, arefound inthe older\nlanguage] ;andlikewise those inu namely bhu, dhu.\nb.Ofroots endinginmutes, two thirds addsyadirectly. Thus, of\nroots ink,fafc; inc(allbutyac:namely), muc, ric, vac, vrace, sic\n[and pac, vie];inch,prach (only case);inj(allbutvraj:namely),\nbhaj, majj (manksya), mrj(marksya), yaj, yuj, vrj, srj[also tyaj, bhrajj,\nbhanj, ran/, sari;, svanj, nij, vij, ruj, bhuj];int,krtandwt[krt, crt,nrt\noptionally];ind,ad,pad, cad, sad, skand, syand, chid, bhid,vid'find',\nnud[also had, khid, svid, ksud, tad,and chrdand trdoptionally]:only\nobservedexceptions, vad,and vid 'know';indh,bandh, radh, budh,\nyudh,rudh[also vyadh, sadh, sidh, krudh, ksudh, c.udh, vrdh\\:onlyobserved\nexceptions, rdhandgrdh , inn,tanandman (butmanforms sometimes\nmanisya);inp,tap, vap, ap,gup, drp, srp, kip [also cap, ksip, Up,\nlap]: svap forms both svapsya andsvapisya;inbh,yabhandlabh[also\nrabh]:noexceptionsobserved;inm,nam, yam, ram :kram follows\neither method.\nc.Ofroots ending insemivowels,all(theyarevery few) take the\nauxiliary.The roots vaorvi(\"ve\") 'weave' andhva orhu(\"hve\")'call'\ntake theforms vayaridhvay,asbefore theaoftheir present-stem:thus,\nvayisya, hvayisya.\nd.Ofroots ending inspirants,theminority (aboutathird)arewithout\ntheauxiliary vowel. They are :roots inc\n,vie,drc(draksya), mrc(mraksya)\n[also danp, dip, ric, Mp,kruc, rue, sprf] ;ins,cis,krs(kraksya) [also\ntvis, dvis, pis, vis, clis, tus, dus, pus, pua] ;in8,vas 'dwell' (vatsya:\n"
+    },
+    {
+        "page": 318,
+        "content": "XII.FUTURE-SYSTEMS.[935\n167) [also yhas]-inh,dah, vah, mi/*, dwft, ruh[also naft, dih, lih]:\nexceptions areyrah (grahlsya) andmuh.\nIntheolder language, adecided, though notalarge, majority ofsimple\nroots addthesyawithout auxiliary *;intheclassical Sanskrit, doubtless the\ncontraryisthecase,as iisgenerally taken byany root oflate origin and\nderivative character asit isalso uniformly insecondary conjugation\n(chap. XIV.).\n936. Astheroot isstrengthenedinformingthestem of\nthisfuture,- so,ofaroot that hasastrong andaweak form,\nthestrong form isused: thus, fromynapornanc ismade\nnanksya ;from]/sras orsrans, sransisya, andthe like.\nSpecial irregularities informing stems from certain roots have been\nnoted inconnection with those roots above(seemajj, mrj, vayandhvay,\ngrah).\nTheQB.hasonce themonstrous form a?nuvisyamahe, made uponthe\npresent-stem apnw (el.IV.) ofj/ap.\n937. This future iscomparativelyrare inthe oldest languagein\np;irt, apparently, because theuses ofafuture aretoalarge extent answered\nbysubjunctive forms butbecomes more andmorecommon later. Thus,\ntheRV. hasonlyseventeen occurrences ofpersonal forms, from nine different\nroots (with participles from sixadditionalroots) ;theAV. has fiftyoccurrences,\nfrom twenty-five roots (with participles from seven more);theTS.hasoc-\ncurrences(personal forms andparticiples together) from over sixty roots;and\n(ashasbeen noticed above) forms fremmore than ahundred andfifty roots\narequotable from theolder texts.\nModes ofthes-future.\n938. Mode-forms ofthefuture occur only sporadically. The soleVedic\nexampleiskarisyds, 2dsing. subj.act.(inRV., once ortwice); GB.has\nesyamahai, tansyamahai, sthasy amahai,all1stpi.subj. mid.; and bhavi-\nsyadhvam, vetsyadhvam, savisyadhvam, 2dpi.impv. mid., arequoted (Bopp)\nfromMBh.\nParticiples ofthes-future.\n939.Participlesaremade from thefuture- stempre-\nciselyasfrom apresent-stemin5fa:namely, byadding\nintheactive theending rTnt,inthemiddle theendingqH\nmana; theaccent remains uponthestem. Thus, from the\nverbs instanced above, ^iHJtt dasydntand^IKIHM dasyd-\nmana, Sfjf^H karisydnt and^i^MHim karisydmana.\nAccordingtothegrammarians, thefeminine oftheactiveparticipleis\nmade either inanil orinati;butonly theformer hasbeen noted asoccur-\n"
+    },
+    {
+        "page": 319,
+        "content": "942]CONDITIONAL. 303\nring intheolder language, andthelatter,ifmetwith atall,isvery rare:\nseeabove, 449.\nInRV. occurs oncesusyanti, from|/su,with anomalous accentuation.\nPreterit ofthes-future: Conditional.\n940.From thefuture-stem ismade anaugment-preterit,\nbyprefixingtheaugment andaddingthesecondary endings.\ninpreciselythesamemanner asanimperfect from apres-\nent-stem inEfa.Thispreteritiscalled theconditional.\nItstands related tothefuture,inform andmeaning, astheFrench\nconditional aurais tothefuture aurai, orastheEnglish would have towill\nhave nearly astheGerman wiirde haben towerde fiaben.\nThus, from therootsalreadyinstanced :\nactive. middle.\ns. d. p.s. d.p.\nddasyam ddasyava ddasyama ddasye ddasyavahi ddasyamahi\n2\nddasyas ddasyatam ddasyata ddasyathds ddasyetham ddasyadhvam\ns^(Kir *(^IHIriIH\nddasyat ddasyatam ddasyan ddasyata ddasyetam ddasyanta\ndkarisyam dkarisyava dkarisyama dkarisye dkarisyavahi dkansyamahi\netc. etc. etc. etc. etc. etc.\n941. Theconditional istherarest ofalltheforms oftheSanskrit verb.\nTheRV. hasbutasingle example, dbharisyat,'was goingtocarry off,and\nnone oftheVedic texts furnishes another. IntheBrahmanas itishardly\nmorecommon(fifteen occurrences, ofwhich several areinrepetitionsofthe\nsamepassage) except inQB->where itismetwithmore than fifty times.\nNordoesit,like thefuture, become more frequent later: notanexample\noccurs inNala, Bhagavad-Gita,orHitopadeca ;only oneinManu;andtwo\ninQakuntala.\nII.The Periphrastic Future.\n942. This formation containsonlyasingleindicative\ntense, active andmiddle, without modes, orparticiple,or\npreterit.\nItsconsists inaderivative nomenagentis. havingthe\n"
+    },
+    {
+        "page": 320,
+        "content": "304 XII.FUTURE-SYSTEMS.[942\nvalue ofafuture activeparticiple, andused, either with\norwithout anaccompanying auxiliary,inthe office ofa\nverbal tense with futuremeaning.\n943.Thenoun isformed bythesuffix cTtr(orcT|~\ntar}-,and this(asinitsother than verbal uses: seechap.\nXVII.)isadded totheroot eitherdirectlyorwith apreced-\ningauxiliary x^owel ^i,theroot itselfbeing strengthened\nbyguna,buttheaccentresting onthesuffix: thus, ^TcT\ndatrfromy^TTda; SficTkartr fromy^fikr;^rf^rT bhavitr\nfrom yv(bhu.\nc^\nAsregardsthepresenceorabsence ofthevowelz,theusageissaidby\nthegrammarians tobegenerally thesame asinthes-future from thesame\nroot(above, 935). Themost important exceptionisthat theroots inrtake\nnoi:thus, kartr(against karisya); roots hanandgamshow thesame dif-\nference;whilei?rt,vrdh, andsyand have ihere, though notinthes-future.\nThefewforms which occur intheolder language agree with these statements.\n944. Inthethirdpersonsofbothvoices, thenom.\nmasc. ofthenoun,inthethreenumbersrespectively (373),\nisused withoutauxiliary:thus, Hf^FTT bhavita,'heorsheor\nitwill be';HftHI|f bhavitarau,'both will be';H&cU(Hbha-\nvitdras, 'theywill be'. Intheotherpersons,the firstand\nsecondpersons presentofy'ERfas'be'(636)areused as\nauxiliary;andtheyarecombined, inallnumbers, with the\nsingular nom. masc. ofthenoun. Asanindependent verb,\n33TTashasnomiddle forms;butforthisauxiliaryusemiddle\npersonshavebeenmade byanalogy, ^Tliebeingused in\n1stsing.\nThus, fromy^J da, 'give':\nactive,\nd.middle,\nd.\ndatasmi datasvas datasmas datahe datasvahe datasmahe\ndatasi datasthas datastha datase datasathe datadhve\n^Trn^\ndata datarau daturas data datarau dataras\n"
+    },
+    {
+        "page": 321,
+        "content": "948] PERIPHRASTIC FUTURE. 305\nVery rarely, other persons than thethird areusedwithout theauxiliary\nverb:thus, aham drasta,'Ishall see'(MBh.); tvam bhavita (MBh. Megh.),\n'thou shaltbe';andexamples arenotunknown oftheauxiliary inthe3d\nperson: thus, vakta 'sti(MBh.), 'hewillspeak'; and oftheuseindualand\npluraloftheproper number-form with the auxiliary:thus, kartarau svah\n(MBh.), 'wetwo shall do'.\n945. The accent inthese combinations, asinalltheor-\ndinarycases ofcollocation ofaverb with apreceding predicate\nnoun oradjective (592),isonthenounitself; and, unlike all\nthetrue verbal forms, thecombination retains itsaccentevery-\nwhere even inanindependentclause :thus, tdrhi vriatinastro\nbhavitUsmi(QB.),'then Ishallbeoutofdanger' (where lhavisyami,\nifused, would beaccentless). Whether inadependent clause\ntheauxiliaryverbwould takeanaccent(595), andwhether,if\nso,attheexpenseoftheaccent ofthenoun(asinthecase of\napreposition compoundedwith averb-form :1083), weare\nwithout themeans ofdetermining.\n940. IntheVeda, thenomina agentis intrortar,likevarious other\nderivative nouns (271), butwith especial frequency, areused inparticipial\nconstruction, governing theaccusative iftheycome from roots whose verbal\nforms doso. Often, also, they areused predicatively, with orwithout ac-\ncompanying copula; yetwithout anyimplicationoftime; they arenotthe\nbeginnings, butonly theforerunners, ofanewtense-formation. Thetense-\nusebegins, butrather sparingly,intheBrahmanas (from which over thirty\noccurrences arequotable), andgrows morecommon later, though theperi-\nphrastic future isnowhere sofrequentasthes-future.\n947. Middle forms areextremely fewintheolder language. TS.has\nonce prayoktdse, which seems tobe1stsing, (the usual endingeadded to\ntheabbreviated root ');butTA.(i.11)hasonce thelater form yastdhe;\nfayitdseinQB.isclearly 2dsing.; TB.hasonce yastdsmahe,1stpi.\nUses oftheFutures andConditional.\n948.Asthes-future isthecommoner, soalso itisthe\nonemoreindefinitely used. Itexpressesingeneral what isgo-\ningtotakeplaceatsome time tocome but often, asin\nother languages, adding ontheonehand animplicationofwill\norintention, orontheother hand that ofpromiseorthreat-\nening.\nAfewexamplesare: varsisydty aisdmah parjdnyovrstiman bhavisyati\n(B.),'itisgoingtorain; Parjanya isgoing toberich inrain this year';\nydstanndvedaMmredkarisyati (RV.), 'whoever does notknow that, what\nwillhedowith verse?' dva{vaydm agni dhasyamahadtha yuydrhkfrh\nkarisyatha (B.), 'wearegoingtobuild thetwo fires; thenwhat willyou\ndo?' tarn fndro 'bhyddudrava hanisydn (QB.),'him Indra ran at,intending\ntoslay'; yddy evdkarisyathasakdih devafr yajnfyaso bhavisyatha (RV.),'if\nWhitney, Grammar. 20\n"
+    },
+    {
+        "page": 322,
+        "content": "306 XII.FUTURE-SYSTEMS.[943\nyewilldothus, yeshall beworthy ofthe sacrifice along with the gods';\nddntds tecatsyanti (AV.), 'thy teeth will fallout';namarisyasi mdbibheh\n(AV.), 'thou shalt notdie; benotafraid'; bruhi kvaydsyasi (MBh.),'tellus;\nwhere areyougoingtogo?' yadimampratydkhydsyasi visam dsthdsye (MBh.),\n'ifyou shall reject me, Iwill resort topoison'. Asinother languages, the\ntense isalsosometimes used fortheexpression ofaconjecture: thus: ko\n'yamdevo gandharvo vdbhavisyati (MBh.), 'who isthis? heisdoubtless a\ngod,oraGandharva'.\n949. Theperiphrasticfuture isdefined bythegrammarians\nasexpressing somethingtobedone atadefinite time tocome.\nAndthis,though butfaintlytraceable inlater use,isadistinct\ncharacteristic oftheformation inthelanguage where itfirst\nmakes itsappearance (Delbrtick).Itisespeciallyoften used\nalong with$vds,'tomorrow'.\nAfewexamplesare :yatardn vdimefvah kamitdras tejetdras (K.),\n'whichever ofthetwoparties these shall choose tomorrow, they willconquer';\nprdtar yastdsmahe (TB.), 'we shall sacrifice tomorrow morning'; ityahe vah\npaktdsmi (QB.),'onsuchandsuch aday Iwillcook foryou'; tanmaekdm\nrdtrim dnte fayitdse jdtdute'yamtdrhi putrd bhavita(QB.),'then youshall\nliewithmeone night, and atthat time this sonofyours willbeborn'.\nInother cases, this definiteness oftime iswanting, butanemphasis, asof\nspecial certainty, seems perhaps tobelong totheform; thus, bibhrhf md\npdrayisydmitve'U:kdsmdn mdpdrayisyasi 'tydughd imdh sdrvdh prajd\nnirvodhd, tdtas tvdpdrayitdsmiJ\nti(QB.), 'support meand Iwillsave you,\nsaid it.From what willyousaveme? said he.Aflood isgoing tocarry\noffallthese creatures; from that Iwillsave you, saidit';paridevaydm cakrire\nmahac chokabhayam prdptdsmas (GB.), 'they setupalamentation :\"we are\ngoingtomeet with great pain and dread\"'; yajelyaksi yastdhe ca(TA.),\n'Isacrifice, Ihave sacrificed, and Ishall sacrifice'. Inyetother cases, in\ntheolder language even, and yetmore inthelater,thisfuture appears to\nbeequivalenttotheother :thus, prajdydm enam vijndtdsmo yadividvdn vd\njuhoty avidvdn vd(AB.), 'we shallknow him inhischildren whether heis\nonethat sacrifices with knowledge orwithout knowledge'; vaktdsmo vdidam\ndevebhydh (AB.),'we shall tell this tothegods'; yadisvdrtho mama'pi\nbhavitd tataevam svdrtham karisydmi (MBh.),'iflatermyown affair shall\ncome up,then Iwillattend tomyownaffair'; katham tubhavitdsy ekaHi\ntvdm nrpa yocimi (MBh.),'buthow willyougetalong alone?that, king,\nisthecause ofmygrief about you'.\n950. The conditional would seem tobemostoriginally and\nproperlyused tosignifythatsomething 'was goingto'bedone.\nAnd thisvalue ithasinitsonlyVedic ocurrence, andoccasion-\nallyelsewhere. Butusuallyithasthesenseordinarilycalled\n\"conditional\"; andinthegreat majorityofitsoccurrences itis\nfound(likethesubjunctive andtheoptative, when used with\nthesamevalue)inboth clauses ofaconditional sentence.\n"
+    },
+    {
+        "page": 323,
+        "content": "952]USES OFTHECONDITIONAL. 307\nThus, yovrtraya sinam dtra'bharisyat prdtdm jdnitri vidUsa uvdca\n(RV.)i 'him, whowasgoing here tocarry offVritra'swealth,hismotherpro-\nclaimed totheknowing one'; fatdyum gamakarisyam (AB.),'Iwasgoing to\nmake (shonld have made) thecow live ahundred years' (inother versions\nofthesame storyisadded theotherclause, inwhich theconditional hasa\nvalue more removed from itsoriginal:thus,inGB.,'ifyou, villain, hadnot\nstopped [prdgrahlsyah] mymouth'); tdtaevdJ\nsyabhaydmv\\'ydya kdsmdd\ndhydbhesyad dvitiydd vdfbhaydm bhavati(QB.), 'thereupon hisfeardeparted ;\nforofwhom washetobeafraid? occasion offear arises from asecond\nperson'; titpapdta dram tanmene yddvdsah paryddhdsyata (QB.),'heleaped\nup ;hethoughtitlong thatheshould putonagarment'; sdtadevdnd\n'vindat prajdpatir ydtrd 'hosyat (MS.), 'Prajapati, verily, didnotthen find\nwhere hewas to(should) sacrifice'; evam cennd'vaksyo murdhd tevyapatisyat\n(GB.),'ifyoushould notspeak thus, yourhead wouldfly off'; adydddhdi\n^tdvad evd'bhavisyad ydvatyo hai'vd'greprajdh srstdstdvatyo hdi 'v<T\n'bhavisyan ndprd 'janisyanta (QB.),'ifhehadbeen only somuch, there\nwould have been only somany living creatures aswere created atfirst;they\nwould have hadnoprogeny';kirnvd'bhavisyad arunas tamasdm vibhettd\ntarn cetsahasrakirano dhuri nd'karisyat (Qak.),'would theDawn, forsooth,\nbethescatterer ofthedarkness,ifthethousand-rayed onedidnotsether\nonthefront ofhischariot?'\nCHAPTER XIII.\nVERBAL ADJECTIVES ANDNOUNS: PARTICIPLES,\nINFINITIVES, GERUNDS.\n951. THOSE verbaladjectives,orparticiples, which are\nmade from, tense-stems, and soconstitute apartofthevarious\ntense-systems, have been alreadytreated. Itremains todescribe\ncertain others, which, being madedirectly from theroot itself,\nbelongtotheverbal systemasawhole, andnottoanypartic-\nularpartofit.\nThe infinitive(withafewsporadic exceptionsintheolder\nlanguage)alsocomes inallcases from therootdirectly, and\nnotfrom anyofthederived tense-stems.\nThesame istrue oftheso-called gerunds,orindeclinable\nparticiples.\nPassive Participleintdornd.\n952.Bytheaccented suffix rTtd--or,inacompar-\nativelysmall number ofverbs, *fnd isformed averbal\n20*\n"
+    },
+    {
+        "page": 324,
+        "content": "308 XIII. VERBAL ADJECTIVES ANDNOUNS.[952\nadjective which, when coming from transitiveverbs, quali-\nfiesanythingashaving endured theactionexpressed by\ntheverb :thus, ^ffdattd, 'given'; 3WTukta, 'spoken'.Hence\nitisusuallycalled thepassive participle; or,todistinguish\nitfrom theparticiple belongingtothepassive present-sys-\ntem(771),thepast passive participle.\nWhen made from anintransitive orneuter verb, the\nsameparticiple,asinotherlanguages,hasnopassive but\nonly anindefinitepastsense: thus,JHgatd, 'gone'; >7rT\nbhutd, 'been'; ^faci patitd,'fallen'.\n953. Ingeneral,thisparticipleismade byadding rf\ntdtothebare verbal root, with observation oftheordinary\nrules ofeuphoniccombination.\nSome roots, however, require theprefixion oftheauxiliary\nvowel itothe suffix. For these, and fortheverbs thatadd\nndinstead oftd,seebelow, 956, 957.\nAstotheaccent when theroot ispreceded byapreposi-\ntion, see1085 a.\n954.Therootbefore rTtdhasusuallyitsweakest form,\nifthere isanywhereintheverbal systemadistinction of\nweak andstrongforms. Thus :\na.Apenultimatenasal isdropped:e.g.aktafromyanj,\nbadclM from]/bandh, srastd from]/srans orsras.\nto.Roots which intheweak forms oftheperfectareabbre-\nviated(794)suffer thesame abbreviation here :thus, uktdfrom\nyvac,udhd fromyvah, istdfromyyaj, suptd fromysvap,\nviddhd fromyvyadh, prstd from]/prach.\nC.Final aisweakened to iingitdfromygo,'sing', pita\nfrom}/pa 'drink', dhitd fromydha 'suck', sphitd,vitafrom\nyvya, fitafromyjya,cltdfromy'fya;-and itisweakened\ntoiinsthitd, hitdfromydha'put' (withdhalsochangedtoh:\nbutdhitd isfound also incompoundsinV.),ditdfromyda'cut'\nandyda'bind'; sitd, mitdfromyma'measure', citd(orcatd),\nchitt(orchatd}.\nd.Afinalmornislost after aingatd, natd, yatd,ratd\n(fromygam etc.);hatd, matd, ksatd, tatd, vatd(fromyhan etc.).\ne.More isolated cases are: utdfrom yav,utdfromyva 'weave', fistd\nfromy^as,murtd referred toymurch, syuta fromysiv, dyutd fromydw\n'play', mutd fromymlv,dhautd fromydhav'cleanse1\n(RV. has alsodhutd}.\n"
+    },
+    {
+        "page": 325,
+        "content": "957] PASSIVE PARTICIPLE INtaORna. 309\n955. Ofmoreirregular character arethefollowing:\na.Anumber ofroots ending inamretain thenasalandlengthen\ntheradical vowel(asinothers oftheir verbalforms):thus, kantd,\nkrantd, tantd, canto,,crania, fromy'krametc.;yd/ivan 'becov-\nered' forms inlikemanner dhvanta.\nb.Three roots inanmake theparticiple fromparallelroots\nina:thus, jatd, khatd, said, fromyjanetc.\nC.The rootda'give' forms datta(fromthederivative form\ndad],butdata also isfound incompositioninV.Thecontracted\ntta(asiffordata, with theradical vowellost)iswidely found\nincomposition, especiallywithprepositions (1087 e),but also\nwith other elements :thus, devdtta(RV.); punartta (PB.vi.5.12);\nand, accordingtothegrammarians, sutfa. The rootjaks (deriv-\native ofghas: 675)foimsjagdM,asifhomjagh (once apparently\nabbreviated incompositiontogdhainTS. :thus, agdhad}; ysvad\nmakes inVeda svattd(beside svaditd).\n956.The suffix with^i,orintheform^itd,isreg-\nularly used with thederivative verb-stems insecondarycon-\njugation (chap. XIV.),also often with roots ofaderivative\ncharacter(asflf^fjinv,f^T hins), andnotinfrequently with\noriginalroots(as^pat,r^car, rpgjmanth, sftpi).\nInRV.andAV., theparticiplesinitafrom simple roots aremore than\nasixth ofthewhole number. Among them, uditd(}/vad)istheonly case\nofabbreviation ofvatou.From frath comes prthita (once). Jahita from\nyha (bysubstitution ofthepresent-stem asshown injahami)isanisolated\nirregularity, payita shows thesame strengthening which appears inthe\npresent-system (629).\nAfew roots form theparticipleeither with orwithout theauxiliaryi:\nthus, guptd and gupitd, drptd anddrpitd, dhrstd anddhrsitd, mattd and\ninaditd, vittd(also vinnd) and viditd.\nThe rootgrabh orgrah has, aselsewhere, longI:thus, grbhitd, grhltd.\n957.The suffix ^nd(alwayswithoutauxiliary ^i)is\ntaken instead ofrTtdbyanumber ofroots. Thus:\na.Certain roots ina,and in iandw-vowels :namely, pyand orfind\nfromj/fya, frdnd from|/pra, jlnd (beside jitd) fromyjyaorjl,fund from\nycvaorfvi,hand andhind fromyha,dind fromyda'bind' and'cut',\nkslnd (beside ksitd) fromyksi 'destroy', plnd fromypyaorpi,vllnd from\nyvll,Unafromyil,dund fromydu, dyundfromydlvordev 'lament', lund\nfrom yiuandsome others.\nb.The roots invariable r(so-called f-roots: 242), which before the\nsuffix becomes Irorur :thus, klrnd, glrnd, jlrnd, tlrnd, dlrnd, plrnd, stlrnd\n(beside strtd); purnd, murnd; andjurnd.\n"
+    },
+    {
+        "page": 326,
+        "content": "310 XIII. VERBAL ADJECTIVES ANDNOUNS.[957\nAfew participial forms intdfrom such roots aremetwith intheolder\nlanguage: thus, gurtd, purtd, flrtd.\nc.Afewroots ending inj(which becomes gbefore thesuffix: 216. 4):\nthus, bhagnd fromybhaj, bhugnd fromybhuj, magnd fromymajj, rugnd\nfromyruj. Also, one ortwo others thatshow aguttural before thena:\nthus, lagnd fromyiag,vrknd from]/urapc,aknd fromj/ac.\nd.Anumber ofroots, some ofthem verycommon ones, ind(which\nbecomes nbefore thesuffix :161, end): thus, channd, chinnd, bhinnd, vinnd\n(beside vittdandviditd), skannd, syannd, svinnd, tunnd, pannd, sannd (beside\nsattd, which alone isfound inV.);andhlannd (according tothegrammarians)\nfrom yhlad;alsotrnnd fromytrdandchrnnd fromychrd, which show an\nirregular lingualization ofthenasal;and others.\n958.Thegrammarians reckon asparticiplesofthisforma-\ntion afewmiscellaneous derivativeadjectives, coming from roots\nwhich donotmake aregular participle:such areksama, 'burnt',\nkrqa, 'emaciated', pakvd, 'ripe', phulla, 'expanded', cuska, 'dry'.\nPast Active Participle intavant.\n959.From thepast passive participleismade, by\naddingthepossessivesuffix SftTvant, asecondary derivative\nhavingthemeaning and construction ofaperfectactive\nparticiple:forexample, cTrTRriIM tatkrtdvan, 'having done\nthat'. Itsinflection islike that ofother derivatives made\nwith this suffix(452 if.);itsfeminine ends incfcftvati; its\naccent remains ontheparticiple.\n960. Derivative words ofthisformation arefound inRV., butwithout\nanything like aparticipial value. TheAV. hasasingle example, withpar-\nticipial meaning: afitdvaty dtithau,'one's guest having eaten'(loc. abs.).\nIntheBrahmanas also itisextremely rare. Inthelaterlanguage, however,\nitcomes tobequite common. And there itisalmost always used predic-\natively, and generally without copula expressed,orwith thevalue ofa\npersonal verb-form intheperfect tense(likethe derivative intainthe\nfuture: 942ff.). Forexample: mamnakafdd drstavan, 'noonehasseen\nme'; sanakularh vyapadiiavan,'hedestroyed theichneumon'; or,withcopula,\nmahat krchram praptavaty asi, 'thou hast fallen upon great misery'. Although\noriginally and properly made onlyfrom transitive verbs (with anobject,to\nwhich theparticiple intastands inthe relation ofanobjective orfactitive\npredicate),itisfinally found alsofrom intransitives :thus, cutena sam$ritavatl\n(Qak.),'hasbecome united with themango-tree'; gatavatl (ib.),'shehasgone'.\nFuture Passive Participles: Gerundives.\n961. Certain derivativeadjectives (forthemostpart\n"
+    },
+    {
+        "page": 327,
+        "content": "9631 GERUNDIVES . 311\nmore orlessclearly secondary derivatives) haveacquiredin\nthelanguageavalue asqualifying something which isto,\norwhichought to,suffer theactionexpressed bytheroot\nfromwhichtheycome; andtheyareallowed tobemade\nfromeveryverb. Hencethey are, likemoreproper par-\nticiples, usuallytreated asapartofthegeneral verbalsys-\ntem, and called futurepassive participles,orgerundives\n(liketheLatin forms inndus,towhichthey correspondin\nmeaning).\n962.The suffixes bywhich suchgerundivesareregu-\nlarlyandordinarily made arethree :namely TJ?/, rTcET tct-\nvya,and EFffal aniya.\nDerivatives inyahaving thisvalue aremade inallperiodsofthe\nlanguage, from theearliest down; theother twoareofmore modern origin,\nbeing entirely wantingintheoldest Veda(RV.), andhardly known inthe\nlater. Other derivatives ofasimilarcharacter, which afterward disappear\nfrom use, arefound intheVeda.\n963. The suffixyainitsgerundive usehasnothingto\ndistinguishitfrom thesame suffix asemployedtomake adjec-\ntivesandnouns ofother character(seebelow, chap.XVIII.:1213).\nAnd itexhibits also thesamevarietyinthetreatment ofthe\nroot.\nThe original value ofthesuffix isio,and assuch ithastoberead in\nthevery great majority ofitsVedic occurrences. Hence theconversion ofe\nandotoayandavbefore it(seebelow).\nThus :a.Final abecomes ebefore thesuffix :deya, khyeya,\nmet/a (probablydk-ia etc., with euphonic yinterposed):butRV.\nhasonce-jnaya.b.Theother vowels either remain unchanged,\norhave thegunaorthevrddhistrengthening;and eusually and\noalwaysaretreated before theyaastheywould bebefore a\nvowel :thus, jdyya, bhdyya, layya ;ndvya, bhdvya, hdvya, bhavyd ;\nvarya: and, inthelaterlanguage, riiya, Jet/a, dhuya (suchcases\narewanting earlier).Inafewinstances, ashort vowel adds t\nbefore the suffix :thus, itya, mitya, cnitya, stutya, krtya (the\nonlyVedicexamples).c.Medial aremains unchanged oris\nlengthened: thus, ddbhya, vdndya, sddya; madya, vacya.\nd.Mediali,u,andr-vowels areunchangedorhave theguna-\nstrengthening:thus,L\nidya, guhya,, dhrsya ;dvesya, y6dhya, mdrjya.\nTheRV.hasabout forty examplesofthisgerundive,andtheAV. adds\nhalf asmany more. Except inbhavid(once),theaccent inRV. isalways\n"
+    },
+    {
+        "page": 328,
+        "content": "312 XIII. VERBAL ADJECTIVES ANDNOUNS.[963\nontheroot;AV. hasseveral cases ofaccent onthe iofthesuffix (hence\nwrittenadya, a$ya, -vyadhya, -dharsyh}. According tothegrammarians, the\naccent isontheroot orelsetheendingiscircumflexed :always theformer,\niftheyafollow avowel.\n964. The suffixtavyaisasecondary adjectivederivative\nfrom the infinitival noun intu(below, 972), made byadding\nthesuffixya(properlyiatwhence theaccentya],before which\nthefinalu,asusual, has^w^a-strengthening, and isresolved\ninto av.\nHence, both asregardstheform taken bytherootand\ntheuseoromission ofanauxiliary vowel ibefore thetavya,\ntherules arethesame asfortheformation ofthe infinitive\n(below, 968).\nNoexampleofthis formation isfound inRV., andinAV. occur only\ntwo,janitavya andhihsitavya. IntheBrahmana languageitbeginstobenot\nrare, and ismade both from thesimple rootandthederived conjugational\nstems(next chapter);inthe classical languageitisstillmore frequent.\nAccordingtothegrammarians, theaccent oftheword iseither circumflex\nonthe final oracute onthepenult: thus, kartavyaorkartdvya; inthe\naccentuatedtexts,itisalways theformer.\nAstotheimpersonal use ofthis gerundive, seebelow, under Passive\n(999).\n965.The suffixaniyaisinlikemanner theproductofsec-\nondary derivation, made byaddingtheadjectivesuffixlya(1215)\ntoanomen actionis formed bythecommon suffix ana.\nItfollows, then, asregardsitsmode offormation, therules\nforthesuffix ana(below, chap. XVIII.:1150).\nThis derivative also isunknown inRV., andinAV. isfound only in\nupajivaniya andamantraniya (inboth ofwhich, moreover,itsdistinct\ngerundive value admits ofquestion). IntheBrahmanas (wherelessthan a\ndozen examples ofithavebeennoted), andinthelater language,itismuch\nlesscommon than thegerundiveintavya.Itsaccent, asinallthederiv-\natives with thesuffixlya,isonthepenult: thus, karaniya.\n966. Other formations ofkindred value arefound intheVeda asfollows :\na.Gerundives intuaortva,apparently made from theinfinitival noun\nintuwith theadded suffix a(1209). They arekdrtua(intwooccurrences\nkartva], jdntua, jetua, ndmtua, v&ktua, s6tua, sndtua, hdntua, hetua; and,\nwith auxiliaryi(or 5),jdnitva, sdnitva, bhdvltva.\nb.Gerundives inenia orenya (compare 1217): they areIdenia, carenia,\ndryenia, bhusenya, yudhenia, vdrenia; with oneexample from anapparent\naorist-stem, yamsenya, and three orfourfrom secondary verb-stems (see\nbelow, 1038).\nC.Gerundives indyia (once dyya: compare 1218): they aredaksayia,\npandyia, viddyia, fravdyia, hnavayia ;with afewfrom causative secondary con-\njugation-stems (below, chap. XIV.):and stuseyiaisofclose kindred withthem.\n"
+    },
+    {
+        "page": 329,
+        "content": "970] INFINITIVES. 313\nd.Afewadjectives inelima,assacelima, bhidelima (apparently not\nfound inuse) arereckoned asgerundives bythegrammarians.\n967. The division-line betweenparticipial andordinary\nadjectivesislessstrictly drawn inSanskrit than intheother\nIndo-European languages. Thus, adjectivesinu,aswillbeseen\nlater(chap. XVII. :1178), fromsecondary conjugational stems,\nhaveparticipial value\n;andintheBrahmanas(withanexample\nortwoinAV.)isfoundwidely andcommonly used aparticipial\nadjective formed with thesuffix uka(ibid., 1180.\nInfinitives.\n968.The laterlanguagehasasingleinfinitive, which\nistheaccusative case ofaverbal noun formed bythesuf-\nfix rTtu,added totherootusually directly,butoften also\nwith aidofthepreceding auxiliary vowel^i.Theform\noftheinfinitiveending, therefore,istj^^tumor^s^itum.\nTheroothastheyema-strengthening, and isaccented. Thus,\nforexample, \"^^etumfrom|/^ifSficFTkdrtum fromj/sfi\nTvr;r\\[^y\\^cdritumfromy^TJ\"car; Hi^lrlH bhdvitum from\nThe rules astotheuseoromission oftheauxiliaryiare\nthesame asthose thatapplytotheformation oftheperiphras-\nticfuture-noun intrortar(943).\nThesame form, inalike use, isfound alsointheolder language, back\ntoitsearliest recorded period 5but itisthere onlyone ofawhole body of\nrelated formations, anaccount ofwhich.isinbrief asfollows :\n969. IntheVeda andBrahmana, anumber ofverbal nouns,\nnominaactionis, invarious oftheir cases, areused inconstruc-\ntions which assimilate them totheinfinitive ofother languages\nalthough, were itnotforthese other laterandmore devel-\noped andpronounced infinitives, theconstructions inquestion\nmight pass asordinary case-constructions ofasomewhat pecu-\nliarkind.\n970. Thenouns thususedinfinitivelyarethefollowing:\na.Theroot-noun, without derivative suffix,issoused in\nitsaccusative inam,itsdative ineor(from a-roots) at, its\ngenitive and ablative inas,and itslocative in t.\nb.The verbal noun intuissoused initsaccusative in\nturn, itsdative intave ortavai, and itsablative and genitive\nintos.\n"
+    },
+    {
+        "page": 330,
+        "content": "314 XIII. VERBAL ADJECTIVES ANDNOUNS.[970\nOfother nouns, only single cases, generally datives, arereckoned as\nused with infinitive value; thus:\nC.From theverbal noun inas,thedative inase;andalso,\ninanextremely small number ofinstances, adative inse(or\nse\\from anoun formed with ssimply.\nd.From nouns inmanand van, datives inmaneandvane.\ne.From nouns inti,datives intaye,or(fromoneortwo\nverbs)intyai.\nf.From nouns ini,datives indye.\ng.From nouns indhiandsi,datives indhyai andsyai.\nh.Afew infinitives insaniareperhapslocatives fromnouns\ninanadded toaroot increased bys.\ni.From asingle root, dhr, aremadeinfinitively usedforms\nintdri, ofwhich thegrammaticalcharacter isquestionable.\nAmongallthese, the(formswhich have best righttospecial treatment\nasinfinitives, onaccount ofbeingofpeculiar formation, orfrom suffixes not\nfound inotheruses,orboth,arethose inse,sani, tari, dhyai, and tavdi.\nExcept thevarious cases ofthederivative intu,and oftheroot-noun,\nthese infinitives arealmost wholly unknown outside theRig-Veda.\nOther suffixes andforms than those noticed above might beadded;for\nitisimpossibletodraw anyfixed linebetween theuses classed asinfinitive\nandtheordinary case-uses;andtheso-called infinitives arefound coordinated\ninthesame sentence withcommon nouns, andeven withcompound nouns.\nMorespecialrules astothevarious formations areasfollows:\n971. Theroot-noun used asinfinitive hasthesameform, andthesame\naccent, bothwhen simple andwhen combined withprepositions,asinits\nother uses. Inthevery great majority ofinstances,itismade from roots\nending inaconsonant; but alsofrom afewind(khyd, da,dhd, pa?, ma,\nyd],from two orthree in iandu-vowels(hi,ml, bhu), andfrom one ortwo\ninchangeable r,which takes thezr-form(tir, stir).\nThe roots indform theaccus. indm(pratidhdm, AV.), thedat. indi,\nthe abl. inas(understanding avasd before dasforavasds andnotavasdt\ninRV. iii.53.20),and thelocative ine(only twoexamples,ofwhich one\nisbetter understood asdative).\n972. The infinitive noun intuismade freely from roots ofevery form.\nThe root takes theywna-strengthening,ifcapableofit,andoften adds the\nauxiliary vowel ibefore thesuffix (according totherule already stated, 968).\nThe root isaccented, unless thenoun becombined with apreposition,in\nwhich case the latter hastheaccent instead :thus, kdrtum, etave, hdntos,\nbutnfkartum, nfretave, nfrhantos.\nThe dative intavdi isintworespects anomalous :inhaving theheavy\nfeminine ending dialong with astrengthened u;and intaking adouble\naccent, oneontheroot orontheprefixed preposition, andtheother onthe\nending di:thus, etavdf, hdntavdf, dtyetavdi, dpabhartavdf.\nTherootgrahmakes (asinother kindred formations) grdhltu ,andlongi\nisshown alsoby?dritu, stdrltu, hdvltu (andcompare bhdvitva, 966 a).\n"
+    },
+    {
+        "page": 331,
+        "content": "980]INFINITIVES. 315\n973. The infinitive inase ismade inRV.from about twenty-five roots;\ninAV.and later there havebeen noted noother examplesofit.Innear\nthree quarters ofthecases, theaccent isonthesuffix :thus, rnjdse, jlvdse,\nbhiydse, tujdse ;theexceptions arecdksase; dhdyase (with yinserted before\nthesuffix); anddyase, bhdrase, spdrase, hdrase(with ^Una-strengthening of\ntheroot). Strengthening oftheroot isalsoshown byjavdse, dohdse, bhojdse,\n$obhdse. Inpusydseisseen, apparently,thepresent-stem instead oftheroot.\nTheendingseisextremely rare, being found only injiseandperhaps\nstate, andone ortwo stillmore doubtful cases.\n974. Infinitives inmane aremade from onlyfive roots: thus, trdmane,\ndamane, dhdrmane, bhdrmane, and(withdifferentaccent) vidmdne. From\nyda comes davdne; turvdne maycome directly fromytr,orthrough the\nsecondary root turv;dhurvane israther fromydhurvthan fromydhvr.\n975. The infinitives intaye areistdye (j/j's), pltdye (ypa 'drink'), vitdye,\nsdtdye.Intydi, theonlyexamples noted areitydf (RV.) andsadhyai (AB.).\nWithayeareformedtujdye, drfdye, mahdye, yudhdye, sandye.\n976. Theending dhyai is,more than any other, irregular andvarious\ninitstreatment. Ithasalways anabefore it\n;and inthemajority ofcases\nitisaccented uponthisa,andadded toaweak form ofroot: thus, fwc-\nddhyai, prnddhyai, dhiyddhyai, huvddhyai. Buttheform ofroot isthestrong\none inafew cases :namely, $ayddhyai, stavddhyai, tarddhyai, jarddhyai,\nmandddhydi, vandddhyai. Inhalf-a-dozen forms, again,the root hasthe\naccent :namely, ksdradhyai, gdmadhyai, ydjadhyai (but once ortwice also\nyajddhyai], vdhadhyai, sdhadhyai, bhdradhyai.Inasingle instance, ptbadhyai,\nthesuffix isadded distinctly toapresent-stem ;and inone, vdvrdhddhydi,\ntoaperfect stem. Finally, inanumber ofinstances(ten),this infinitive\nismade from acausative stem inay:thus, madayddhyai, risayddhyai,etc.\nThis infinitive isbynomeans rare inRV., being made inthirty-five\ndifferent forms (with seventy-two occurrences). But itishardly known out-\nside oftheRV.; theAV. has itbutonce(inapassage found alsoinRV.);\nand inthebranches oftheYajur-Veda buttwo orthree examples have been\nnoticed (one ofthem TS. falsely reads gdmadhye)-,intheBrahmana language\nitappearstobeentirely wanting.\n977.Anexampleortwoaremetwith ofaninfinitive insyai: thus,\nrohisyai (TS.), avyathisyai (K.).\n978. Theinfinitives insani are: -bhusdni fromybhu; fusdni fromy?u\nor?va; nesdni fromynl;saksdni fromysah; parsdni from}/pr,tarlsdni\nfromytr;andgrnudni and-strnlsdni fromyygrand str the lastcon-\ntaining evident present tense-signs (comparethe1stsing, grnise, 894 d).\n979. Theonly infinitive intari isdhartdri (withitscompound vidhartdri),\nfromydhr.\nUses oftheInfinitives.\n980. Theuses oftheso-called infinitives areforthemost\npart closely accordant with those ofthecorrespondingcases from,\nother abstract nouns. Thus :\n"
+    },
+    {
+        "page": 332,
+        "content": "316 XIII. VERBAL ADJECTIVES ANDNOUNS.[981\n981. The accusative, which ismadeonly from theroot-\nnoun andthenoun intu,isused asobjectofaverb.\nEspecially, offorms from therootspafc,'beable', andarh,'beworthy,\nhave theright orthepower'. Thus, fakema tvdsamidham (RV.), 'maywe\naccomplish thykindling'; mdpafccm pratidhdm {sum(AV.), 'may they notbe\nable tofitthearrow tothestring'; mdno vdimdm sadydh pdrydptumarhati\nmdnah pdribhavitum (TS.)>'themind, forsooth, can atonce attain andsurpass\nher'; kohyetdsyd'rhatigtihyarh ndma grdhltum (QB.),'forwho isworthy to\ntake hissecret name?' IntheVeda, theconstruction with these verbs is\nonlyoneamong others;intheBrahmana,itbecomes thegreatly prevalent\none (three quarters ormore ofallthecases).\nFurther, ofverbs ofmotion(next most frequent case): thus,ddksindni\nh6tum eti(TS.), 'hegoes tosacrifice things pertainingtosacrificial gifts';\nfndram pratfram emydyuh (RV.),'IgotoIndra for(i.e.beseech ofhim)\nthelengthening outoflife'; ofydhr, 'persist in,undertake': as,sdiddrh\njdtdh $drvam evdddgdhum dadhre(QB.), 'he, assoon asborn, begantoburn\nthisuniverse'; ofverbs meaning 'desire, hope, notice, know', and the\nlike:as,pdfdn vicrtarh vettha sdrvdn (AV.), 'thou knowest how toloosen\nallbonds'; tdsmdd agnfth nd\"driyeta pdrihantum (QB.),'therefore oneshould\nnotbecareful tosmother thefire'; and ofothers,\n982. Ofthe infinitive datives, thefundamental andusual\nsense isthatexpressed by'for, inorder to,forthepurposeof.\nExamples are\n;:vfyvam jlvdrh cardsebodhdyanti (RV.), 'awakening every\nliving creature tomotion'; tdntipaydta ptbadhydi (RV.), 'come todrink\nthem'; ndt'tdm tedevdadadur dttave(AV.), 'thegods didnotgiveher to\nthee foreating'; prat\"dyudhdye ddsyum tndrah (RV.), 'Indra went forward\ntofight thedemon'; cdksur nodhehivikhydt. (RV.), 'give ussight forlooking\nabroad'.\nSome peculiar constructions, however, grow outofthisuseofthein-\nfinitive dative. Thus:\na.Thenoun which islogically thesubject ortheobject oftheaction\nexpressed bythe infinitive isfrequently putbeside itinthedative(bya\nconstruction which isinpartaperfectly simple one, butwhich isstretched\nbeyond itsnatural boundaries byakind ofattraction): thus, cakdra surydya\npdnthdm dnvetavd u(RV.),'hemade atrack forthesun tofollow (made\nforthesunatrack forhisfollowing)'; pfplie pfnge rdksobhyovinfkse (RV.),\n'hewhets hishorns topierce thedemons'; rudrdya dhdnur dtanomi brah-\nmadvtse fdrave hdntavd u(RV.),'1stretch thebow forRudra, thatwith his\narrow hemay slay the6raftma-hater'; asmdbhyam dr$dye surydya punar\nddtdm dsum, 'may they grantlifeagain, thatwemay seethesun'.\nb.An infinitive with|/fcr, 'make',isused nearly inthesense ofa\ncausative verb :thus, prd 'ndhdrh frondm cdksasa etave krthah(RV.), 'ye\nmake theblind andlame toseeandgo'; agnfih samfdhe cakdrtha(RV.),\n;thou hastmade the fire tobekindled'. Ofsimilar, character isanoccasional\n"
+    },
+    {
+        "page": 333,
+        "content": "9841 USES OFTHEINFINITIVES. 317\nconstruction with another verb:as,yddimupmdsi kdrtave karat tat(RV.),\n'what wewish tobedone, mayhedothat'.\nc.Adative infinitive isnotseldom used asapredicate, sometimes with,\nbutmore usually without, acopula expressed:thus, agnfo ivandpratidhfse\nbhavati(TS.),'likefire, he isnot toberesisted'; mahimd teanyena nd\n9amnd?e (VS.), 'thy greatness isnottobeattained byanother'; ndldm indro\nnikartave ndfakrdh pdri?aktave (RV.), 'Indra isnot tobeputdown, the\nmighty one isnottobeoverpowered1\n.\nd.Sometimes aninfinitive soused without acopula haspretty clearly\nthevalue ofanimperative:thus, tydmeyafdsa...aufijd huvddhyai [asti]\n(RV.), 'these glorious ones shall thesonofUc.ij invoke forme'; suktebhir\nvah ...indra nvagni dvase huvddhyai [stah] (RV.), 'with yourhymns shall\nyecallnowonIndra andAgni foraid'; vandddhya agnim ndmobhih [asmi]\n(RV.),'letmegreet Agni with homage'; asmdkasaf casurdyo vfyva dfds\ntarlsdni(RV.), 'and letour sacriflcers cross allregions'. The infinitives in\ndhyai and sani (whichlatter isinall itsuses accordant withdatives)are\nthose inwhich theimperative value ismost distinctly toberecognized.\ne.Inthe B.(with only asporadic case ortwoelsewhere) thedative\nintavai isfrequently used with averb signifying 'speak' (bru, vac, aft),to\nexpress theorderingofanything tobedone :thus, tdsmad 6sadhlnam evd\nmulany ucchettavaf bruydt, 'therefore lethim direct theroots oftheplants\ntobecutup(speakinorder tocutting up)'.\n983.The ablative infinitive which, like theaccusative,\nismade onlyfrom theroot-noun and that intu isfound\nespecially with theprepositions ft,'until', andpurh,, 'before'.\nThus, dtdmitos (TS. etc.),'until exhaustion'; purd vacdh prdvaditos\n(TS.),'before utterance ofthe voice'. IntheBrahmanalanguage, this is\nthewell-nigh exclusive construction oftheablative; intheVeda,thelatter\nisused also afterrte, 'without', and after several verbs, astraandpa,\nI/M,bhi.\nIntwo orthree instances, byanattraction similar tothat illustrated\nabove forthedative (982 a),anoun dependent onthis infinitive isputin\ntheablative beside it :thus, pura vagbhyah sampravaditoh (PB.),'before the\nutterance together ofthe voices'; trddhvam kartdd avapddah (RV.), 'save us\nfrom falling down into thepit'.\n984. Thegenitiveinfinitive(havingthesame form asthe\nablative)isincommon useintheBrahmana languageasdepend-\nentonicvard, lord, master', employed adjectivelyinthesense\nof'capable'or'likely'or'exposedto'.\nExamplesare :td[devdtah] ifvard enam praddhah (TS.), 'they arelikely\ntoburnhim up'; dtha havdlyvarb '^nfrh citvd Mrhcid damritdm apattor vf\nvahvdlitoh(QB.),'sointruth heisliable,afterpilingthefire,tomeet with\nsome mishap orother,ortostagger'; Ifvararh vairathantaram udgatuq, caksuh\npramathitoh (PB.),'therathantara isliable toknock outtheeye ofthe\nchanter'.\n"
+    },
+    {
+        "page": 334,
+        "content": "318 XIII. VERBAL ADJECTIVES ANDNOUNS.[934\nThe dative isonce used in B.instead ofthegenitive (inifvarati jd-\nnayitavaf); and,inthe later language, sometimes the accusative inturn.\nOccasionally themasc.sing.nom. Ifvarah isused, without regardtothegen-\nderornumber oftheword which itqualifies:thus, tdsye \"fvardh prajd\npdplyaslbhdvitoh(QB.),'hisprogenyisliable todeteriorate1\n.And inafew\ninstances theword Ifvara isomitted, andthegenitive hasthesame value\nwithout it:thus, dvemadhyandinamabhipratyetos (AB.), 'twomaybeadded\ntothenoonlibation';tdto diksitdh pamano bhdvitoh(B.), 'then theconse-\ncrated isliable togetthe itch'.\nThis construction with ifvara, which istheonly one forthegenitive\ninfinitive intheBrahmana,isunknown intheVeda, where thegenitiveis\nfound inavery small number ofexamples withmadhyd, andwith the root\nIf:thus,madhyd kdrtoh (RV.),'inthemidst ofaction'; iferay6 ddtoh(RV.),\n'he ismaster ofthegiving ofwealth'.\n985. Unless theinfinitives insaniand tariarelocative inform(their\nuses arethose ofdatives), thelocative infinitive issorare, andhassolittle\nthat ispeculiar initsuse, that itishardly worth making anyaccount of.\nAnexampleisusdso budhf(RV.),'attheawakening ofthedawn'.\n986. IntheVeda, thedative infinitive forms areverymuch\nmore numerous than theaccusative(inRV., their occurrences\naretwelve times asmany;inAV., more than threetimes);and\ntheaccusative inturn israre(onlyfour forms inRV., only\neightinAV.).IntheBrahmanas, theaccusative hasrisen to\nmuchgreater comparative frequency (itsforms arenearly twice\nasmanyasthose ofthedative);buttheablative-genitive, which\nisrare intheVeda, hasalsocome tofullequalitywith it.The\ncomplete disappearanceintheclassical languageofallexcept-\ningtheaccusative inturn isamatter fornosmallsurprise.\n987. The later infinitive inturn isoftenest used incon-\nstructionscorrespondingtothose oftheearlier accusative :thus,\nnavaspam acakat sod/ium, 'hecould notrestrain histears'; tarn\ndrastum arhasi, 'thou oughtesttoseeit';praptum ichanti, 'they\ndesire toobtain'; samkhyatum arabdham, 'having beguntocount'.\nButalso, notinfrequently,inthose oftheother cases. So,\nespecially,ofthedative :thus, avasthatum sthanantaramcintaya,\n'devise anotherplacetostay in';tvam anvestum iha\"gatah,'he\nhascome hither toseek forthee'; butlikewise ofthegeni-\ntive :thus, samarthogantum, 'capableofgoing'; samdhatum ic-\nvarah, 'able tomend'. Even aconstruction asnominative is\nnotunknown :thus, yuktam tasyamayasamaqvasayitumbha-\nryain (MBh.),'itisproperforme tocomfort his wife'; na\nnaptaram svayam nyayyam captum evam(R.),'itisnot suitable\nthus tocurse one'sown grandson'.\n988. Inthe later language, asintheearlier,theinfinitive incertain\nconnections haswhatwelookuponasapassivevalue. Thus, kartum arabdhah,\n"
+    },
+    {
+        "page": 335,
+        "content": "991]GERUNDS. 319\n'begun tobemade': frotum nayujyate,'itisnot fittobeheard(forhear-\ning)'. This isespecially frequent along with thepassive forms ofy$ak: thus,\ntyaktum na$akyate,'itcannot beabandoned'; fakyaviha\"netum, 'they two\ncanbebrought hither'; nacavibhutayah ?akyam avaptum urjitah,'nor are\nmighty successes athing capableofbeing attained'.\nGerunds.\n989.The so-calledgerundisastereotypedcase(doubt-\nlessinstrumental) ofaverbal noun, usedgenerally, butin\nthelater languagenotexclusively,aslogical adjuncttothe\nsubjectofaclause, denotinganaccompanyingor(usually)\naprecedingaction tothatsignified bytheverb oftheclause.\nIthasthus thevirtual value ofanindeclinableparticiple,\npresentorpast, qualifyingtheactor whose action itde-\nscribes :\nThus,forexample:grutvai'vaca'bruvan, 'andhearing (or\nhaving heard) they spoke'; tebhyah pratijnaya'thai 'tanparipa-\npracha, 'having given them hispromise,hethen questioned\nthem' .\n990.Thegerundismade inthelaterlanguage byone\nofthetwo suffixes ^Ttvaand Ty<i.theformerbeing used\nwith asimple root, thelatter withone that iscompound-\nedwith aprepositional prefix or,rarely, with anele-\nment ofanother kind,asadverb ornoun.\nExceptions tothisdistribution ofusesbetween thetwosuffixes arevery\nrare :examplesofsimpleroots withyaarearcya, grhya, usya (yvas 'dwell') ;\nofcompoundedroots with tva areanudhyatva, apatyaktva, pratyarpayitvd\n(AV.: only case noticed intheVeda: TA.has-rocayitva). Thegerund in\ntvd, however, may have thenegative particle prefixedtoit:thus, akrtva,\nanlrayitva.\nOfcompoundsofthegerund inyawith other elements than theusual\nverbalprefixes, RV. haspunarddya, karnagfhya, padagfhya, hastagrhya,\nararhkrtya, akkhallkftya, mithaspfdhya ;AV. hasfurther namaskrtya.\n991.The suffix ^Ttvahastheaccent. Itisusually\naddeddirectlytotheroot, butsometimes withinterposition\noftheauxiliary vowel^iwithregardtowhich, aswell\nastotheform oftheroot beforeit,thisformation closely\nagreeswith that oftheparticipleinrTta(above, 952if.).\n"
+    },
+    {
+        "page": 336,
+        "content": "320 XIII. VERBAL ADJECTIVES ANDNOUNS.[991\nWhen iisused,thedisposition totake aweak form ofroot isless\nmarked. Roots which have nainstead oftaasparticipialsuffixusually\nreject the i.\n992.The suffix 7Jyaisaddeddirectlytotheroot,\nwhich isaccented, buthas itsweak form.Arootending\ninashort vowel takes rTtyainstead ofTJya:thus, ftfrET\n-jitya, TOT -krtya.\nRoots inamandanwhose passive participle ends inata(954 d)form\nthis!gerund also inatya: thus, -gatya, -hatya. Butsuch am-roots are\nallowed inthe later language topreserve their nasal inthegerund: thus,\n-gamya (nosuch form occurs intheVeda). Final changeablerbecomes Ir\norur :thus, -glrya, -purya. Final aremains unaltered :thus, -gay a,-sthaya ;\nandmi'establish' andmi'diminish' take theformma;II'cling' isallowed\ntodothesame.\n993. The older language hasthesame twogerund forma-\ntions, having thesamedistinction, andused inthesameway.\na.InRV., however,the final ofyaisinthegreat majority ofin-\nstances(fully twothirds) long (asiftheinstrumental ending ofaderivative\nnoun in iorti).InAV., longaappears only once, inaRV.passage.\nb.Instead oftvdalone, theVeda hasthree forms ofthe suffix, namely\ntvd, tvdya, and tvi. Ofthesethree, tvi isdecidedly thecommonest inRV.\n(thirty-five occurrences, against twenty-one oftoa); but itisunknown in\nAV., andvery rare elsewhere intheolder language; tvayaisfound nine\ntimes inRV.(only once outside thetenthBook), twice inAV., andbut\nfewtimes elsewhere. The historical relation ofthethree forms isobscure.\nc.Two other gerund suffixes, tvanam andtvinam, arementioned by\nthegrammarians asofVedic use,buttheyhavenowhere been found actually\ntooccur.\n994. Theuseofthisgerund, through notchanginginits\ncharacter, becomes much morefrequent, andeven excessive, in\nthelaterlanguage.\nThus, intheNalaandBhagavad-Gita, which have only onetenth as\nmany verb-forms asRV., there aremore than three times asmany examples\nofthegerund asinthe latter.\nEarly examples are :vdjrena hatvd nfrapdh sasarja (RV.), 'striking^vith\nhisthunderbolt, bepouredforth thewaters'; strfyam drstvdya kitavdm tatapa\n(RV.),'thegamblerisdistressed when hesees awoman'; pitvi sdmasya\nvavrdhe(RV.), 'having drunk ofthesoma, hewaxedstrong'. Inthe older\nlanguage almost withoutexception, andinthelaterusually,itexpresses an\naction orcondition belonging tothesubject ofthesentence; but itisinsome\ntexts more loosely construed :thus, tatah cabdad abhijnaya savyaghrena\nhatah(H.), 'thereupon hewasslain bythetiger, thelatter having recognized\nhimbyhisnoise'; kirnnumesydd idam krtvct(MBh.), 'what,Iwonder,\n"
+    },
+    {
+        "page": 337,
+        "content": "996]GERUNDS. 321\nwould happen tome,having done this?'sucintya coJktarhsuvicdrya yatkrtam\n(H. ,-what issaid after maturethought, anddone after full deliberation'.\nAdverbial Gerund inam.\n995.The accusative ofaderivative nomen actionis ina,used\nadverbially, assumes sometimes avalue and construction so\naccordant with that oftheusual gerundthat itcannot wellbe\ncalled byadifferent name.\nNoexample ofapeculiar gerundial construction with such aform occurs\neither inRV. orAV., although afewadverbial accusatives areprobably to\nbeclassed asrepresentingthis formation :thus, abhydkrdmam, pratdnkam,\npranodam, nilayam, dbhiskdndam. Thegerundisfoundespecially inthe\nBrahmanas (much oftenest inQB.),andsparinglylater. Inthe classical\nlanguageitisquiterare.\nAfinalvowel hasfrdd/u'-strengthening before thesuffix;final aaddsy;\namedial vowel hasguna; butmedial aisusually lengthened. The accent\nisontheradical syllable.\nExamplesare:kamam vaimdny dngdni vyatyasam fete (QB.),'helies\nchanging thepositionofthese limbs atpleasure'; uttardm-uttardrh fdkhdrh\nsamdldmbharh rohet(QB.),'hewouldclimb, taking hold ofahigher andever\nahigher limb'; apansu mahdnagdm iva'bhisarhsaram didrksitdrah(QB.),\n'hereafter, running together as itwere about agreat snake, they willwish\ntoseehim'; ndmdny dsdm etdnindmagrdham (QB.),'withseparate naming\nofthese their names'; ydviparydsam avaguhati (QB.),'whoever buries it\nupside down'. Asinthese examples,theform isalmost always acompound\none. Inthelater language,ititsaid tobeusedmost oftenrepeated:thus,\npdyam-pdyam vrajati,'hegoes afterdrinking repeatedly'; prathamam bhojarh\nvrajati, 'havingfirst eaten, hegoes'; bdhutkseparh krandUum pravrttd (Qak.),\n'sheproceeded tocry,throwing upherarms(with arm-tossing)'.\nCHAPTER XIV.\nDERIVATIVE ORSECONDARY CONJUGATION.\n996.SECONDARYconjugationsarethose inwhich awhole\nsystemofforms, like thatalreadydescribed asmade from\nthesimple root,ismade, withgreaterorlesscompleteness,\nfrom aderivativeconjugation-stem;and isalsousuallycon-\nnected with acertain definite modification oftheoriginal\nradical sense.\nWhitney, Grammar. 21\n"
+    },
+    {
+        "page": 338,
+        "content": "322 XIV. SECONDARY CONJUGATION.[996\nWehave seen, indeed, that thetense-systems arealso forthemostpart\nmade from derivative-stems;andeventhat, insome cases, suchstems assume\ntheappearanceandvalue ofroots, and aremade thebasis ofacomplete\ncoiijugational system. Nor isthere any distinct division-line tobedrawn\nbetween tense-systems andderivative conjugations thelatter arepresent-\nsystems which have beenexpanded intoconjugations bytheaddition ofother\ntenses, and ofparticiples, infinitives, and soon. Intheearliestlanguage,\ntheir forms outside ofthepresent-system are still quite rare, hardly more\nthansporadic;andeven later they are with theexceptionofone ortwo\nformations which attain acomparative frequency much lesscommon than\nthecorresponding forms ofprimary conjugation.\n997. Thesecondary conjugationsare: I.Passive;\nII.Intensive;III.Desiderative;IV.Causative;V.Denom-\ninative.\nThe passiveisclassed here rather asamatter ofconvenience and of\ngeneral usage than because itisofthesame kind with theothers.\nI.Passive.\n998.Thepassive conjugationhasbeenalreadyinthe\nmain described. Thus, wehave seen that:\na.Ithasaspecial present-system,thestem ofwhich\nispresent only, andnotmade thebasis ofanyofthere-\nmainingforms :thisstem isformed with theaccented class-\nsignITyd*-,and ittakes themiddleendings.Itistreated\nabove, 768 ff.\nb.Intheother tenses, themiddle forms areused also\ninapassivesense. But:\nc.There isaspecial passive 3dsing,oftheaorist,\nendingin^i:itistreated above, 842 ff.And:\nd.Accordingtothegrammarians,there may beformed\nfromsome verbs, forpassive use, aspecial stem fortheaorist\nandthetwofuturesystems, coincidinginform with thepecu-\nliar3dsing,aorist. ,\nThus: fromyda (aor.3dsing, adayi],beside ddasi, dasye, datdhe, also\nadayisi, dayisye, ddyitdhe. The permissiontomake this double formation\nextends to.allroots ending invowels, and tograh, drp,and han. The\nduplicate forms have notbeen noticed intheolder language, andthey are,\natthebest, extremely rare inthe later.\nAstotheprescribed passive inflection oftheperiphrastic perfect,see\nbelow, 1072.\n"
+    },
+    {
+        "page": 339,
+        "content": "1001] PASSIVE; INTENSIVE. 323\ne.Besides theparticiple from thepresenttense-stem\n(771. 5),thepassivehasapast participleinrTta(952).or^\nna(957), and futureparticiples,orgerundives,ofvarious\nformation(961 ff.).madedirectly from theroot.\n999. Thepassive construction, with thelogical subject intheinstru-\nmentalcase,isafrequent andfavorite one, especially inthelaterlanguage:\nthus, evam uktva tena sarvesam bandhanani chittani(H.), 'thus saying, he\ncutthebonds ofthem all'.And, extremely frequently, animpersonal passive\ninthethirdpersonisused ;and itmay (asinotherlanguages) beformed\nfrom intransitive aswell astransitive verbs :thus, cruyatam,'let itbeheard'\n(i.e.hearye!), iha\"gamyatam,'comehither'; sarvair jalam adayo 'ddiyatam\n(H.),'let allflyup,takingthenetwith them'; tacchrutva jaradgaveno\n'ktam, 'hearing that, Jaradgava said'; katharhjlvitavyam, 'how isone tolive?'\nyavadane.ua munina sthatavyam (H.),'aslong asthatsageshall exist'. The\npredicate totheinstrumental subject ofsuch aconstructionis,ofcourse,\nalso intheinstrumental :thus, adhuna tavaJnucarena maya sarvatha bhavi-\ntavyam (H.), 'henceforth Ishall always bethycompanion';tenatvaya ya-\nvajjlvarh sukhinabhavitavyam (H.), 'with thatthou shalt behappy aslong as\nthou livest'. Thegerundiveiscommon inthis construction, andnotseldom\nithasapurelyfuture sense.\nII.Intensive.\n1000. The intensive (sometimesalso called frequent-\native)isthatone ofthesecondary conjugations which is\nleast removed from theanalogyofformationsalreadyde-\nscribed. Itis,likethepresent-systemofthesecond con-\njugation-class (642 ff.),theinflection ofareduplicatedstem,\nbutofonethat ispeculiarinhavingastrengthenedredu-\nplication.Itisdecidedlylessextended beyondthelimits\nofapresent-systemthananyother ofthe derivative con-\njugations.\nThe intensiveconjugation signifiestherepetitionor\nintensification oftheaction expressed bytheprimarycon-\njugationofaroot.\n1001. Accordingtothegrammarians,theintensive con-\njugation maybeformed from nearlyalltheroots inthe\nlanguagetheexceptions being:roots ofmore than one\n21*\n"
+    },
+    {
+        "page": 340,
+        "content": "324 XIV. SECONDARY CONJUGATION.[1001\nsyllable,thoseconjugated only causatively (below, 1056\\\nandingeneralthosebeginning with avowel.\nInfact, however, intensives inthe later language areextremely rare,\nsorare that itishard totellprecisely what value istobegiven tothe\nrules ofthenative grammar respecting them. Nor arethey atallcommon\nearlier, except (comparatively)intheRV.,which contains about sixsevenths\nofthewhole number(rather over ahundred) quotable fromVeda andBrah-\nmana-texts (AV.has lessthan half asmany asRV., andmany ofthese in\nRV. passages).\nHence,inthedescriptiontobegiven below,the actualaspectofthe\nformation,asexhibited inthe olderlanguage,willbehad primarily and\nespeciallyinview; andtheexampleswillbeofforms found there inuse.\n1002. Thestrong intensivereduplicationismade inthree\ndifferent ways:\na.Thereduplicating syllable is,aselsewhere, composedof\nasingle consonant withfollowing vowel, and, sofarasthe\nconsonant isconcerned, follows therules forpresent andper-\nfectreduplication (590);butthevowel isaheavy one, radic-\nalaand r(orar]being reduplicated with a,an/-vowel by e,\nandanw-vowel byo.\nExamplesare :vavad, babadh, cacvas, rarandh; dadr, dadhr;cekit, tetij,\nneni, vevli; cocuc, popruth, cosku, johu.\nb.Thereduplicating syllablehasafinal consonant, taken\nfrom theendoftheroot. With anexceptionortwo, thiscon-\nsonant iseither r(oritssubstituteI)oranasal.\nExamplesare: carcar, calcal, sarsr, marmrj, jarhrs ;cankram, janghan,\ntahstan, danda$ (ydah?orda(), janjabh (yjambhorjabh), tantas(ytans\nortas),nannam(ynam).\nOnly roots having aorrasvowel make thisform ofreduplication, but\nwith such roots itismorecommon than either oftheother forms.\nIrregular formations ofthis class are :with afinal other than rornin\nthereduplication, badbadh; with afinal nasal inthereduplication which is\nnotfound intheroot,jangah (RV.), janjap (B.: and the later language\nhasdandah); with ananomalous initial consonant inreduplication, jarbhur\nfromybhr (comparetheVedicperfect jabhdra, 789b) ;with various treatment\nofanrorar-element, dardar anddardir, carkar andcarkir,tartar andtartur,\ncarcar and carcur, jargur andjalgul andgalgul.\nThe root ristheonlyonewith vowel initial forming anintensive stem\nintheolder language:itmakes theirregular alar orair.\nQ.Thereduplicationisdissyllabic, an/-vowel being added\nafter afinal consonant ofthereduplicating syllable.This/-vowel\nisintheolderlanguageshort before adouble consonant, and\nlong before asingle.\nExamples are:ganiyam (but ganigmatam}, varivrt, vamvah, caniskad,\n"
+    },
+    {
+        "page": 341,
+        "content": "1005]INTENSIVE. 325\nsanisvan; navlnu, davidyut (and theparticiples ddvidhvat buttdvituat).Asingle exception astothequantity ofthe iisdavidhdva.\nThismethod ofreduplicationisfollowed intheolder language byover\ntwenty roots.Thus,ofroots having final orpenultimate n(once m],andn\ninthereduplicating syllable, pan, phan, san, suem, han;gam; krand, fcand,\nskand, syand; ofroots havingfinal ormedialr,and rinthereduplicating\nsyllable, kr'make', tr,bhr, vr,mrd, vrj,vrt; further, ofrootsassuming\ninthereduplication annotfound intheroot, onlyvah(QB.:thegrammarians\nallow alsokas, pat, pad}-, finally, ofroots having uoruasradicalvowel,\nwith, avbefore thei-vowel, tu,dhu, nu,dyut.\nInthisclass, thegeneral rules astotheform ofthereduplicating con-\nsonant(590) areviolated inthecase ofghanlghan and bharlbhr, and of\nganlgam, karlkr(buttheregular carlkr alsooccurs), kanikrand, andkaniskand\n(butalsocaniskandoccurs).\nThereversion tomore original guttural form after thereduplication in\ncekit, andjanghan andghanlghan,isinaccordance with what takes place\nelsewhere(216.9).\n1003. Thesame root isallowed toform itsintensive stem\ninmore than oneway.\nThus,intheolder language, dddranddardr;dadhr anddardhr;cdcal\nandcarcar (and carcur);tartar(and tartur] andtantr;janghan andghanlghan ;\nvarvrt and varlvrt; jarbhur and bharibhr; dodhu anddavldhu; nonu and\nnavlnu; bdbadh andbadbadh.\n1004. Themodel ofnormal intensive inflection isthe\npresent-systemofthereduplicating conjugation-class (II.);\nand this isindeed toaconsiderable extent followed, in\nrespecttoendings, strengtheningofstem, and accent. But\ndeviations from themodel arenotrare;andtheforms are\ningeneraloftooinfrequent occurrence toallow ofsatis-\nfactoryclassification andexplanation.\nThemostmarkedirregularityisthefrequent insertion\nofan^*between thestemandending. Accordingtothe\ngrammarians,this isallowed inallthestrongforms before\nanending beginningwith aconsonant; and before the\n*ieafinalvowel hast/wwa-strengthening,butamedial one\nremains unchanged.\nPresent System.\n1005.We willtakeupthepartsofthepresent-systemin\ntheir order, givingfirstwhat isrecognizedasregularinthe\n"
+    },
+    {
+        "page": 342,
+        "content": "326 XIV. SECONDARY CONJUGATION.[1005\nlater language, andthenshowing how theformationappearsin\ntheearlier texts. Asmost grammarians donotallow amiddle\ninflection, andmiddle forms arefeweven intheVeda, noat-\ntemptwillbemade tosetupaparadigmforthemiddle voice.\n1006. Asexampleofinflection maybetaken theroot\nf^ vidjofwhich theintensive stem iscjf^ vevid, or,in\nstrong forms, ifi^veved.\nNeither from this norany other root aremore than afewscattering\nforms actually quotable.\n1.Present Indicative.\ns. d. .\nvevedmi, vevidlmi vevidvds vevidmds\n9\nvevetsi, v6vidlsi vevitthds vevitthd\n3^fo|Tfr,^|cj<{Jiri cfNTlHil<=^irj\nvevetti, vevidlti vevittds vevidati\nFromy~^ ku,thesingular forms withauxiliary vowel\nwould besuc^ojiJZj jdhavlmi, sii^cum johavlsi, sii^<Jiiri jo-\nhavlti.\n1007. Theforms found intheolder language agreeingeneralwith\ntheparadigm. Examples are: 1stsing., carkarmi, vevesmi; 2dsing., alarsi,\nddrdarsi; 3dsing., dlarti, veveti, nenekti, janghanti, kdnikrantti, ganlgamti;\n3ddu., jarbhrtds ;1stpi.,nonumas; 3dpi.,ndnadati, bharibhrati, vdrvrtati,\nddvidyutati, nenijati; and, with the auxiliary vowel, j6havimi, cdka&mi;\ncdka$iti, nonavlti, dardariti, jarbhuriti. Nostemwith dissyllabic reduplication\ntakes theauxiliaryiinany ofitsforms. AV. hasjdgrdti, with irregular\naccent.\nAsingle dualform with Iandstrong stem occurs: namely,tartarithas.\nThemiddle forms found tooccur are: 1stsing., jdguve, nenije; 3dsing.,\nnenikte, sarsrte; and, withirregular accent, tetikte, dediste; with irregular\nunion-vowel, ndnnate; with endingeinstead ofte,jdngahe, jdguve, yoyuve,\nbdbadhe, and (with irregular accent) badbadhe; 3ddu., sarsrate;3dpi.,\ndedicate.\n2.Present Subjunctive.\n1008. Subjunctive forms with primary endings areextremely rare:\nthere havebeen noticed onlyjanghdndni, jdgardsi (AV.); and, inthemiddle,\ntantasatte (3ddu.).\nForms with secondary endings aremore frequent: thus, 2dsing., jan-\n"
+    },
+    {
+        "page": 343,
+        "content": "1011]INTENSIVE. 327\nghanas, jalgulas ;3dsing., jdgarat, cdrkrsat, jdnghanat, bdrbrhat, mdrmrjat,\nmdrmr?at, parpharat, dardirat, caniskadat, davidyutat, sanisvanat\n,1stdu.;\njanghanava; 1stpi.,carkirama, vevidama; 3dpi.,pdpatan, tfcucan, carkiran;\nand, with doublemode-sign, cdkafdn (AV.). Besides these, rardnas and\nrardnat, cdkdnas andcakdnat andcakdnama, which their accent assimilates\nrather toperfect mode-forms with long reduplication,likemamdhas and\nsasdhat etc.(810a).Ofthemiddle arefound only3dpersons plural: thus,\njdnghananta, jarhrsanta, marmrjanta, nonuvanta, fOfucanta ;andcakdnanta\n(andcakanantaonce).\n3.Present Optative.\n1009. Thismode would show theunstrengthened stem,\nwith theusualendings (566), accented. Thus:\nd._p.\nvevidyftm vevidy&va vevidyfima\netc. etc. etc.\nTheoptativeisrepresented byonly anexampleortwo intheolder\nlanguage: thus, active, vevisyat (AV.), jagryat (AB.); RV.hasonlycakanyat\n(pft.?); middle, nenijita (K.).\n4.Present Imperative.\n1010. Theregular forms oftheimperative, including\ntheusualsubjunctivefirstpersons, would beasfollows :\nvevidani vevidava vevidama\n25!=l!icjftrlH^^f^ff\nveviddhi .vevittdm vevitta\nvevettu, veviditu vevittam vevidatu\n1011. Older imperative forms are less rare thanoptative. The first\npersons have been given above (janghdnani, theonlyaccented example, does\nnotcorrespond with themodel, but isinconformity with thesubjunctive of\nthereduplicating present) ;theproper imperatives are :2dsing. ,dadrhf,\ndardrhi, carkrdhi, jagrhi, nenigdhi, and raranddhf, rarandhf, cakandhi,\nvavandhf; theending tat isfound incarkrtat andjagrtat; andthelatter\n(aswaspointedoutabove, 570)isused inAV. as first person sing. ;\nbarbrhi shows anelsewhere unparalleledloss ofhbefore theending hi;\n3dsing., vevestu, dardartu, marmarttu, and rarantu; 2ddu.,jagrtam;\n3ddu.,jagrtam; 2dpi., jagrtd, andraranta; cankramata (RV., once) has\nananomalous union-vowel; 3dpi.,only theanomalous cakantu (RV., once) r\napparentlyforcakanatu. Inthemiddle voice isfound only neniksva(QB.).\n"
+    },
+    {
+        "page": 344,
+        "content": "328 XIV.SECONDARY CONJUGATION.[1011\nOfimperative forms with auxiliary I,RV. hasnone; AV. hasvavaditu\nandjohavltu, andsuch aresometimes found intheBrahmanas\n;AV. has\nalso, against rule, tanstanihi andjanghanihi.\n5.Present Participle.\n1012. The intensiveparticiples, both active andmiddle,\narecomparatively common intheolderlanguage. Theyare\nformed and inflected likethose ofthereduplicating present,\nandhave theaccent onthereduplicating syllable.\nExamples are:active, cdkafat, ndnadat, cekitat, memyat, ftffweat, roruvat,\nddrdrat, mdrmrjat, jdnghanat, ndnnamat, pdriiphanat, kdnikradat, ddvidyutat ;\nmiddle, bdbadhana, memyana, cekitana, y6yuvana, rdrucana, jdrbhurana,\nsarsrana, ndnnamana, ddndafana. Nomiddleparticiple shows thedissyllabic\nreduplication.\n1013. Onaccount oftheiraccent, rarahand, raraksand, andjdhrsand\n(beside jdrhrsana) areprobablytoberegarded asperfect participles, although\nnootherperfect forms with heavy reduplication from thesame roots occur.\nThe inferenceis,however, rendered uncertain bytheunmistakably intensive\nbadbadhand andmarmrjand (beside mdrmrjana}.\nTheRV. hasonce jdnghnatas, gen. sing., with root-vowel cast out;\nkdnikratappears tobeused once forkdnikradat;ifcakdt istobereferred\ntoyka (Grassmann),itistheonlyexample ofanintensive from arootin\na,and itsaccent isanomalous. Marmrcantas (AB.)isprobably afalse\nreading.\n6.Imperfect.\n1014. Theimperfectisregularlyinflected asfollows :\ndvevidam dvevidva dvevidma\n23^rTSl^f^te^ ^ijfori^5^j%fT\ndvevet, dvevidis dvevittam dvevitta\n3\ndvevet, dvevidlt dvevittam dvevidus\n1015. Theimperfect forms found intheearlier texts arenotnumerous.\nThey are, including those from which theaugmentisomitted, asfollows:\ninactive,1stsing., acaka^am, dedi^am; 2dsing., ajagar, adardar, ddrdar,\ncakdn; 3dsing., adardar, adardhar, avarivar, dardar, kdniskan, ddvidyot,\nndvinot, andcakdn andrardn; 2ddu.,adardrtam; 1stpi.,marmrjma;\n3dpi.,araranws, anannamus, adardirus, acarkrsus, djohavus, anonavus :\nand, with auxiliary z,in3dsing., dvavafit, dvavarit, dyoyavit, droravlt;\ndjohavit; and, irregularly, in3ddu., avavafltam. Themiddle forms are\n"
+    },
+    {
+        "page": 345,
+        "content": "1018]INTENSIVE. 329\nextremely few: namely, 3dsing., ddedista, dnannata(withloss ofthe final\nradical inaweak form ofroot); 3dpi.marmrjata, andavavafanta (which,\nifitbelongs here, shows atransfer toana-stem).\n1016. Derivative Middle Inflection. Fromevery\nintensive stem..asabovedescribed, maybeformed inthe\npresent-systemafurther derivativeconjugationwhich is\nformallyidentical with apassive, beingmadebytheaccented\nsignTyd,alongwith middleendings only.Ithas not,\nhowever, apassive value, but isinmeaning anduse in-\ndistinguishable from thesimpler conjugation.\nAfinalvowel before thisyaistreated asbefore thepassive-\nsignya(770).\nThe inflection ispreciselylike that ofanyother stemend-\ninginainthemiddle voice :thus, fromymrj,intensive stem\nmarmrj,ismade thepresentindicativemarmrjye, marmrjydse,\nmarmrjydte, etc.; optative marmrjyeya, marmrjyethas, marmrjyeta,\netc.; imperative marmrjydsva, marmrjydtam, etc.; participle\nmarmrjydmana ;imperfect dmarmrjye, dmarmrjyathas, dmarmrjyata,\netc.; subjunctive forms donotoccur.\n1017. This kind ofintensive inflection issaid tobe\nmuch more usual than theother inthelater language;in\nthe earlier,itiscomparativelyrare.\nInRV., 7/6-forms aremade from eight roots,five ofwhich have also\nforms ofthesimpler conjugation;theAV. adds onemore; theother earlier\ntexts (sofarasobserved) only twelve more, andhalf ofthem have likewise\nforms ofthesimpler conjugation. Thus: fromj/mr/, marmrjydte etc., and\nmarimrjytta ;fromytr, tarturyante; fromyear, carcurydmana ;fromynl,\nnenlyeran etc.; fromyvi, veviyate; fromyrih, rerihydte etc.; fromyvij,\nvevijydte; fromysku, coskuydse etc.; fromydif, dedicate; fromj/fcaf,\ncakacydte ;fromyvad, vavadydmana; from}/nam, nannamyadhvam ;from\nyvah, vanivahyetaetc. (with lengthened root-vowel, elsewhere unknown);\nfromykrand, kanikradydmana ;fromyvrt, varivartydmana (QB.:should be\nvarivrty-);fromymrf, amarimrfyanta (QB.?thetext reads amarimrt syanta) ;\nfromyyup, yoyupydnte etc.; fromynud, anonudyanta; fromyvli, avevllyanta;\nfromyjabh, janjabhydte etc.; fromyjap, janjapydmana.\nPerfect.\n1018. Thegrammariansareatvariance astowhether\naperfect maybeformed directly from theintensive stem,\n"
+    },
+    {
+        "page": 346,
+        "content": "330 XIV. SECONDARY CONJUGATION.[1018\norwhetheronlyaperiphrastic perfect (below, 1070if.)isto\nbeadmitted.\nIntheolder language, noexampleofanintensive periphrastic perfect\nhascome tolight.Afewunmistakable perfect forms aremade from the\nintensively reduplicatedrootinRV. :namely, davidhava andndnava, 3dsing.,\nandnonuvus, 3dpi.,andTS.hasonce dodrdva. Tothese may beadded\njdgara 1stsing, andjdgara 3dsing.: butastothese, seebelow, 1020.\nAorist, Future, etc.\n1019. Astotheremaining partsofafullverbal con-\njugation, also, thegrammariansarenotagreed (occurrences\nofsuch forms, apparently, beingtoorare toafford even\nthem anybasis forrules) ;ingeneral,itisallowed totreat\ntheintensive stem further asarootinfilling upthescheme\nofforms, using alwaystheauxiliary vowel^iwhere itis\neverused inthesimple conjugation.\nThus, fromyvidtintensive stemvevid, would bemade the\naorist avevidisam withprecative vevidyasam,thefuturesvevidisyami\nand veviditasmi, theparticiples vevidita, veviditavya, etc., thein-\nfinitive veviditum, andthegerundsvevidiiva and-vevidya. And,\nwhere theintensiveconjugationisthe derivative middle one,\ntheaorist andfutures would take thecorrespondingmiddle form.\nOfallthis,intheancient language,there ishardly atrace. TheRV.\nhascdrkrse, 3dsing, mid.,ofaformation like hiseand stuse (894 d),and\nthegerundives vitantasayya, andmarmrjenya andvdvrdhenya ,andQB.has\ntheparticiple vanivahitd, andthe infinitivededlyitavdf. Astojdgarisydnt\nandjdgaritd, seethenext paragraph.\n1020. There aresystemsofinflection ofcertain roots, the\nintensive character ofwhich isquestioned orquestionable. Thus :\nThe rootgr(orgar]'wake' hasfrom the firstnopresent-systemsave\nonewith intensivereduplication; and itsintensive stem, jdgr, begins early\ntoassume thevalue ofaroot,andform acompleter conjugation;while by\nthegrammariansthisstem isreckoned asifsimple andbelongingtotheroot-\nclass(I.),and isinflected throughout accordingly. Those ofitsforms which\noccur intheolder language have been given along with theother intensives\nabove. They are, forthepresent-system, thesame with those acknowledged\nasregular later. The older perfectisliketheother intensive perfects found\ninRV.: namely, jdgara etc., with theparticiple jdgrvdns; and afuture,\njagarisydnt, andapassive participle jdgaritd aremetwith intheBrahmanas.\nThe oldaorist(RV.)istheusual reduplicatedorso-called causative aorist:\nthus, ajigar. Thegrammarians giveitinthelater language aperfectwith\n"
+    },
+    {
+        "page": 347,
+        "content": "1026]INTENSIVE. 331\nadditionalreduplication, jajdgdra etc., anzs-aorist, ajagarisam, withprecative\njdgarydsam, andeverything else that isneeded tomake upacomplete con-\njugation.\n1021. Thestemirajya (active only), 'regulate', fromwhich anumber\nofforms aremade inRV., hasbeen viewed asanintensive fromyrajor\nrj.Itlacks, however, anyanalogy with theintensive formation. Thesame\nistrue ofiradh, 'propitiate' (only iradhanta andirddhyai, apparentlyfor\niradhqdhyai}.\nThemiddle stemTj/a,notinfrequent intheoldest language,isusually\ncalled anintensive ofyi 'go', butwith very doubtfulpropriety,asithas\nnoanalogy ofform with any intensives. The isolated 1stpi.imafte, com-\nmon inRV.,isalso ofquestionable character.\n1022. The rootU'totter', with constant intensivereduplication, leli,\nisquite irregular ininflection andaccent: thus, pres., leldyati andlelayate\npples leldydntl andleldyatas (gen. sing.) andlelayamana, impf. alelayat,\nalelet andalellyata.\n1023. TheRV.anomalous form dart(ordard), 2dand3dsing, from\nydvordar,is[doubtfully referred totheintensive, asifabbreviated from\ndardar. RV.hasonce avarlvus(or-vur) where thesense requiresaform\nfromyvrt,asavarivrtus. Theform rarandta (RV., once) seems corrupt.\n1024.Amarked intensive orfrequentative meaningisnot\nalways easilytobetraced intheforms classed asintensive;and\ninsome ofthem itisquiteeffaced. Thus, therootscit,mj,\nvisusetheir intensivepresent-systemasifitwere anordinary\nconjugation-class ;nor isitotherwise with rand/i, ran,kan(of\nwhich, asnoticed above, theforms admit ofbeingreferred to\ntheperfect-system),andwithgr(jagr). Thegrammariansreckon\ntheinflection ofnijand visasbelongingtothereduplicating\npresent-system (II.),withirregularly strengthened reduplication;\nandtheytreat inthesamewayvieandvij;jagr,aswehave\nseen, they account asimpleroot.\nAlso daridra, intensive ofydra 'run',ismade bythegrammariansa\nsimple root,and furnished with acompletesetofconjugational forms :as\ndadaridrau; adaridraslt,etc. etc. Itdoes notoccur intheolder language.\nThe so-called root vevi 'flutter' isapure intensive.\n1025. Itisallowed bythegrammarianstomake from theintensive\nstem alsoapassive, desiderative, causative, and soon :thus,from vevid,\npass, vevidye;desid. vevidisdmi;caus. veviddydmi;desid. ofcausative,ve-\nvidayisami. Butsuch formations arenotfound intheolder language, and,\nifthey occur atall,areexcessively rare inthe later.\nIII.Desideraiive.\n1026.Bythedesiderative conjugationissignifiedade-\nsire fortheaction orcondition denoted bythesimpleroot :\n"
+    },
+    {
+        "page": 348,
+        "content": "332 XIV. SECONDARY CONJUGATION.[1026\nthus, sHMpibami,'Idrink', desid. nHlHl1\npipasami,'I\nwish todrink'; ^SRwfivami,'Ilive',desid. ist\nvisami,(Idesire tolive'. Such aconjugationisallowed to\nbeformed fromanysimpleroot inthelanguage, and also\nfromanycausative stem.\nThe desiderativeconjugation, althoughitsforms outside\nthepresent-systemareextremelyrare intheoldest language,is\nearlier andmorefully expandedinto awhole verbal system\nthan theintensive. Itsforms arealso ofincreasing frequency:\nmuch fewer than theintensives inRV., more numerous inthe\nBrahmanas and later\n;notone third ofthewhole number of\nroots(about ninety)noted ashaving adesiderativeconjugation\ninVeda andBrahmana have such inRV.\n1027. The desiderative stem isformed from thesimple\nrootbytheaddition oftwo characteristics: a.areduplica-\ntion,whichalwayshastheaccent; b.anappended Hsa\nwhich, however(likethetense-signsofaorist andfuture),\nsometimes takes before ittheauxiliary vowel Se,becoming\n^isa.\n1028. The rootingeneralremainsunchanged;butwith\nthefollowing exceptions:\na.Afinal *oruislengthened before sa:thus, ciksisa, cikisa,\njigisa ;cucrusa, juhusa.\nb.Afinal rbecomes irorurbefore sa:thus, cikirsa,\nsislrsa, jihirsa; bulhursa, tustursa(theonlyexamples noted from\ntheoldertexts).\nc.Beforeisa, thesame finalsnecessarily, andapenulti-\nmate ioruorroptionally, have theywm-strengthening (no\nexamplesarequotable from theoldertexts).\nMorespecial exceptionsare :\nd.Afew roots inaweaken thisvowel toIoreven i:thus, piplsa\n(beside pipasa)fromypa 'drink', jihlsa (AV.) fromyha'remove'(jihlte: 664);\ndidhisa(beside dhitsa) fromydha.\ne.Afew roots inanoramlengthen thevowel :thus, jigansa (beside\njiyamisa) fromygam ;jighahsa fromyhan ;mlmahsa from]/man,andytan\nissaid tomake titahsa.\nf.Reversion toguttural form ofaninitial after the reduplicationis\nseen incikisa fromyd,cikitsa fromycit, jiglsa fromyji,jighahsa from\nyhan;andyhiissaid tomakejighlsa.\n"
+    },
+    {
+        "page": 349,
+        "content": "1030]DESIDERATIVE. 333\ng.The roots vanandsanmake vivasa andsisasa, from theroot-forms\nvaand sa.\nh.The rootjivformsjujyusa (B.: jijlvisa, VS.); andtheother roots\ninlv(oriv:765)arerequired tomake thesame change beforesa,and to\nhave guna before isa: thus, susyusaorsisevisa. Svap formsswsupser.\nDhurv forms dudhursa.\ni.Initial sisusuallyleftunchanged to*after thereduplication when\nthedesiderative signhas s(184 c):thus, sisanksa (B.: ysanj),andsusyusa\nandsisanisa, according tothegrammarians.\nk.Further may bementioned asprescribed bythegrammarians:\nninanksa(orninafisa) fromj/nap, 'perish' ;mimanksa fromymajj ;mimar-\njisa (ormimrksa) fromJ/mr/.\n1029. Theconsonant ofthereduplicationfollows the\ngeneralrules(590);thevowel is^iiftheroothasana-\nvowel, orftr,orani-vowel;itis3uiftheroothasan\nu-vowel. But:\na.Afew roots have along vowel inthereduplicating syllable:thus,\nblbhatsa fromybadhorbadh; mlmansa fromyman;andtutursa (RV.)i'rom\nytur.\nb.Fromj/apismade(in B).atffisa (with amode ofreduplication\nlike that followed sometimes inthereduplicatingaorist: 862). Thegram-\nmarians give other cases ofthesame kind: thus, arjihisa fromyarh,ici-\nksisa fromyiks,undidisa fromyund,ardidhisa fromy-rdh.Inthe older\nlanguage, apistheonly rootwith initial vowel which forms adesiderative\nstem, except apandrdh,which have abbreviated stems :seethenext paragraph.\nC.RV. hasthestems fnaksa andfyaksa, regardedasdesideratives\nfromyynaf'attain' andyaj,with mutilated reduplication.\n1030.Anumber ofroots, including some ofverycom-\nmonuse,form anabbreviated stemapparently byacon-\ntraction ofreduplicationand roottogetherintoonesyllable:\nthus, 37~ipsafrom|/^TC ap;rTTditsa fromy^Jda.\nSuch abbreviated stems arefound intheolder languageasfollows :\ndhitsa (beside didhisa) fromydhd;ditsa(beside didasa) fromyda ;dipsa from\nydabh; fiksafrom]/fafc;slksafromysah:these arefound inRV.: inAV. are\nadded ipsafromyap(RV. hasapsa once), andIrtsa fromyrdh:theother texts\nfurnish lipsa(3.)orllpsa (TB.) fromyiabh, ripsa (GB.)fromyrabh, pitsa\n(93.)fromypad,anddhiksa(QB.)fromydih (Or,rather, dah). Thegram-\nmarians prescribe dhipsa ordhlpsa fromydabh,instead ofdipsa; they form\npitsa fromypataswell aspad ;andthey add ritsa from yradh, jnlpsa\n(beside jijnapayisa] from thecausative quasi-root jnap (below, 1042e), and\nmitsa fromyymaardmlandmi: this lastcould beonlyananomalous\n"
+    },
+    {
+        "page": 350,
+        "content": "334 XIV. SECONDARY CONJUGATION. 1030\nformation, made after theanalogy oftheothers. Alsomoksa isreckoned as\nadesiderative stem from)/mwc (itisdenominative, rather).\n1031. Theuseoftheauxiliary vowel^iisquiterare\nintheearly language ,butmorecommon later\n;and itis\nallowed orprescribed bythegrammariansinmanystems\nwhich have notyetbeen found inactual use.\nItisdeclared tofollow ingeneral, ithough notwithout\nexceptions, necessaryoroptional,theanalogyofthefutures\n(934, 943).\nNoexample oftheuseof iisfound inRV., andonlyoneeach inAV.\n(pipatisa), VS.(jijlvisa), andTS.(jigomisa). The other examples noted in\nthe early texts areafifisa, cikramisa, jigrahlsa (withIfori,aselsewhere\ninthisroot), cicarisa, jijanisa, didlksisa, bibadhisa, riradhisa, vividisa, jihih-\nsisa: most ofthem arefound only inQB.Stems alsowithout theauxiliary\nvowel aremade from roots gram, jiv,badh, vid,\n1032. Inflection: Present-System. Thedesider-\native stem isconjugatedinthepresent-systemwithper-\nfectregularity,likeother a-stems, inboth voices,inallthe\nmodes(including,intheolderlanguage,thesubjunctive),\nandwithparticiples andimperfect.Itwillbesufficient to\ngive here the firstpersons only.Wemaytake asactive\nmodel<%ipsa,'seek toobtain', from|/5TFJ dp;asmiddle\nicifrtsl titiksa, 'endure', from]/fer tij\\'besharp' (seebelow,\n1040).\n1.Present Indicative.\nactive. middle,\nd.\npsam\netc.ipsavas\netc.ipsamas\netc.titikse\netc.titiksavahe\netc.titiksamahe\netc.\n2.Present Subjunctive.\nIpsani Ipsava\netc. etc.Ipsamatitiksai titiksavahai titiksamahai\netc. etc. etc. etc.\n3.Present Optative.\nipseyam ipseva\netc. etc.ipsema titikseyatitiksevahi titiksemahi\netc. etc. etc. etc.\n"
+    },
+    {
+        "page": 351,
+        "content": "1035]DESIDERATIVE. 335\n4.Present Imperative.\nfcilritf^ J\nipsa Ipsatam ipsatatitiksasva titiksetham titiksadhvam\netc. etc. etc. etc. etc. etc.\n5.Present Participle.\n^Hti ipsant (f.^Htilfysanfi] (rilrl^tHIUl titiksamana.\n6.Imperfect.\naipsam aipsava aipsamadtitikse dtitiksavahi dtitiksamahi\netc. etc. etc. etc. etc. etc.\nThere arealmost noirregularities ofinflection tobereported from the\nolder language. No1stpi.inmost, or2dpi.inihana ortana, orimpv.\nintat,ismetwith. Thequotable subjunctive forms arethose insani,sat\nandsat,san,and santa.\nBut the fern, pplesisasati (instead ofsisasanti)occurs once ortwice\nintheolder texts.\n1033. Desiderative forms outside thepresent-systemare\nextremelyrare intheoldest language. TheRV. hasonly per-\nfectforms from astem mimiks thus, mimiksdthus, mimiksdtus,\nmimiksus; mimikse, mimiksire along with thepresent forms\nmimiksati, mimiksa etc., mimiksant(pple): theyshow thatmimiks\normiks hastaken onthecharacter ofanindependentroot. In\nAV. arefound two aorist forms,irtsls and acikitsis, and a\nparticipleortwofrom mimansa(seebelow, 1037, 1039)all\nofthem from stems which have lost their distinct desiderative\nmeaning, andcome tobearanindependentvalue. Theforms\nnoted from theother earlier texts willbegiveninfullbelow.\nInthelaterlanguage,thecomplete systemofverbal\nforms isallowed tobemade inthedesiderativeconjugation,\nthedesiderative stem,less itsfinal vowel, beingtreated as\naroot. Thus:\n1034. Perfect. The desiderativeperfectistheperi-\nphrastic (1070 if.).\nThus, ipsamcakara etc.; titiksam cakre etc. Such forms\naremade inQB.from)/)/kram, dhurv, badh, ruh.\nApparent perfect forms oftheordinary kindmade frommimiks inBV.\nhave been noticed inthepreceding paragraph. AndAB.(viii. 21)hasonce\ndidasitha,'thou hast desired togive'.\n1035. Aorist. The aorist isofthew-form(5):thus.\naipsisam 7tffriffiiji^ dtitiksisi.\n"
+    },
+    {
+        "page": 352,
+        "content": "336 XIV. SECONDARY CONJUGATION.[1035\nTheAV. has acikitsis, and irtsis(augmentless, withmaprohibitive:\n579). TB.hasaipsit; andQB. afrtsit,dcikirsls andajighahsis, andaml-\nmahsisthas.\nAprecativeisalsoallowedthus, tpsyasam, titiksisiya,\nbut itprobablynever occurs.\n1036. Futures. The futures aremade with theauxil-\niaryvowel^i:thus,^H^UIM Ipsisyami and\nasmi; icifrti^Ti titiksisye and(rlirlRifli^titiksitahe.\nTheQB.hastitiksisyate and didrksitdras.\n1037.Verbal Nouns andAdjectives. These too\naremade with theauxiliary vowel^i,inallcases where\nthatvowel isever taken.\nInthe older language have been rioted :participleinta,mimahsitd\n(AV., GB.), jijyusita (AB.), fUfrusitd anddhlksita(QB.); gerundive intavya,\nllpsitavya (AB.), didhyasitavya (QB.); gerund intva,mimahsitva(K.).\n1038. Adesiderative adjectiveinu forexample, dipsu, bibhatsii,\nsisasu isoffrequent occurrence, andhasthemeaning andconstruction of\napresent participle. Anabstract noun ina forexample, jiglsd is\nalso ausual appendagetothedesiderative conjugation. Adjectives inenya\n(having agerundive character :966 b)areoccasionally metwith from the\nearliest time :thus, didrksenya (RV.), cuprusenya (TS.), ninlsenya (PB.); also,\nwith irregular reduplication (apparently) paprksenya (RV.). RV. has also\ndidhisdyya (966 c).\n1039.Derivative orTertiary Conjugations. A\npassiveisallowed tobemade, byaddingthepassive-sign\n7Jydtothedesiderative root(orstemwithout finala):thus,\nsCHTcT ipsydte,'itisdesired tobeobtained'; andacaus-\native byaddinginlikemanner thecausative-sign 3BRJdya\n(1041): thus, ^HtillH ipsdyami,'Icause todesire obtain-\nment'.\nTheonly trace ofsuch formations noticed intheolder languageisthe\nparticiple mimansydmana (apparentlytoberead instead ofmimahsdmana,\nAV. ix.6.24).\nForthedesiderativeconjugation formed oncausative stems,\nwhich isfound asearlyastheBrahmanas, seebelow, 1052 b.\n1040. Some stems which aredesiderative inform have lost\nthepeculiarityofdesiderative meaning, andassumed thevalue\nofindependentroots: examplesarecikits, 'cure', juyups, 'despise',\ntitties, 'endure', bibhats, 'abhor', mimans, ''ponder'.Doubtless\n"
+    },
+    {
+        "page": 353,
+        "content": "10421 CAUSATIVE. 337\nsome oftheapparentroots inthelanguage with sibilant final\nareakin with thedesideratives inorigin.\nIV.Causative.\n1041 .Inthelaterlanguageisallowed tobemade from\neveryroot acompletecausativeconjugation. The basis of\nthis isacausative stem, formed byappendingthecausative-\nsign 5HTdyatothe,usually strengthened,root.\nButbynomeans allconjugation-stems formed bythe\nsign 5RJdyaareofcausative value; and thegrammarians\nregard them asaconjugation-class,thetenth orcwr-class,\naccordingtowhich rootsmaybeinflected asaccordingto\ntheotherclasses, and either alone oralongwith others.\nInRV., theproportion without causative value isfully one third. The\nformation isamore obviously denominative onethanany oftheother con-\njugation-classes,anintermediate between them andtheproper denominatives.\nAcausative meaning hasestablished itself inconnection with theformation,\nandbecome predominant, though notexclusive. Anumber ofroots oflate\nappearanceandprobably derivative character areincluded intheclass, and\nsome palpable denominatives, which lackonly theusual denominative accent\n(below, 1055).\nThe causative formation isofmuch more frequent use,andmore de-\ncidedly expandedinto afull conjugation, than either theintensive orthe\ndesiderative. Itismade frommore than twohundred and fifty roots inthe\nearly language (inRV.,from about onehundred andfifty);butintheoldest,\nitsforms outside thepresent-systemare(apart from theattached reduplica-\ntedaorist: 1046) exceedingly few.\n1042. Thetreatment oftheroot before thecausative-\nsign 5HIctyaisasfollows:\na.Medial orinitiali,u,r,Ihave the#ww#-strengthening\n(ifcapableofit):thus, vedaya fromj/Wd, codayafromycud,\ntarpayafrom~\\/trp;andkalpaya from}/klp (only example).\nBut afew roots lack thestrengthening: these are, intheolder lang-\nuage,cit(citaya andcetaya), vip(vipaya and vepaya), is,iland,ri*\n(rimyaand resaya), tuj, tur, dyut (dyutaya anddyotaya), mrd, sprit; and\nyrabli makes inRV. grbhaya. Dus andguh lengthen thevowel instead.\nMrj sometimes has vrddhi,asinother forms :thus, marjaya (beside\nmar/at/a).\nb.Afinalvowel hastherrdfZ/-strengthening:thus, cyavaya,\nbhavaya, dharaya, saraya.\nWhitney, Grammar. 22\n"
+    },
+    {
+        "page": 354,
+        "content": "338 XIV. SECONDARY CONJUGATION.[1042\nButnoroot in iorIhasvrddhi inthe older language (unless payaya\n[d,\"below]comes from pirather thanpa) as,indeed, regular causatives\nfrom such roots arehardly quotable:onlyRV.hasksayaya fromyksi'dwell';\nforafew alternatively permitted forms, seebelow, e.\nAfew roots have (generally inthe older language only)aform also\nwith puna-strengthening:thus, dru, yu'wardoff', fru,pu,jr'decay\",dr\n'burst', sr,hr;vr'choose' makes varaya later(itisnotfound inV.orBr.).\nc.Amedial orinitial ainalight syllableissometimes\nlengthened,andsometimes remains unchanged:thus, bhajaya,\nsvapaya, adaya; janaya, crathaya, anaya.\nThe roots intheolder language which keeptheir short aarean, ;'an,\npan, svan, dhan, ran, stan, dhvan, gam(gdmaya once inRV.),*am>dam,\nnam, prath, pratfe, fnath, vyath, svad, nad, das, dhvas, mah, nabh, tvar,\nsvar. Some have both forms: namely, pat, chad, mad, ram,cam; rajhas\nrajaya (AV., once) and ranjaya. The roots which lengthen thevowel are\ndecidedly themore numerous. Ifanasal istaken inany ofthestrong\nforms ofaroot,itusually appears inthecausative stem: thus, randhaya,\nlambhaya, rambhaya, skandaya.\nd.Most roots infinala,andtheroot r,addpbefore the\nconjugation-sign: thus, dapaya, dhapaya, sthapaya; arpaya.\nSuch stems aremade inthe older language from theroots khyd, gla,\nghrd, jnd,da'give',drd'run', dha'put'anddha'suck',ma'measure', mid,vd\n'blow', andva'tire', stha, sna,ha'remove' andha'leave'. From ;naand\nsnaarefound inAV.and later theshortened forms jnapaya andsnapaya,\nandfrom fraonlyfrapaya (notinRV.). Also glaforms inthelater language\nglapaya.\nStems from a-roots showing nopare, earlier, payaya fromypa'drink'\n{or pi),pyayayafromypyaorpyay ;sayaya fromysa (or si); also, later,\npayayafrom }/f, hvayaya fromyhva; and further, from roots cha, va\n'weave', andvya, according tothegrammarians.\ne.Thesamepistaken alsobyafew iand I-roots, with various ac-\ncompanying irregularities: thus, ksepaya fromyksi'dwell' (RV., beside\nksayaya} ;ksapaya (AV.)andksapaya andksayaya fromyksi 'destroy' ;\njapaya (VS. andlater) fromyji; lapaya (TB. andlater) fromyil; adhya-\npayafromadhi-\\-yi; smapaya (beside smayaya, which does notoccur)\nfrom ysmi; hrepayafromyhri; and, accordingtothegrammarians.\nrepaya fromyrl, vlepaya fromyvll, krapaya fromykri, bhapaya (beside\nbhayayaand bhlsaya) fromybhl,andcdpaya (beside cdyaya) fromyd.\nMoreover, yruhforms later ropaya (earlier rohaya), andykniiorknuyis\nsaid toform knopaya.\nf.More anomalous cases, inwhich the so-called causative ispalpably\nthedenominative ofaderived noun, are:pdlaya fromypa 'protect'; prlnaya\nfromyprl; llnaya (accordingtogrammarians) fromyil;dhunaya (notcau-\nsative insense) fromydhu; bhlsaya fromybhl; ghdtayafromyhan; sphdvaya\n{accordingtogrammarians) fromysphdorsphdy.\n"
+    },
+    {
+        "page": 355,
+        "content": "1043] CAUSATIVE. 339\n1043. Inflection: Present-System. Thecausa-\ntivestem isinflected inthepresent-system preciselylike\nother stems in fa:itwillbesufficient togivehere ingen-\neralthe firstpersonsofthedifferent formations,takingas\nmodel thestemUT^I dhardya, fromj/Jdhr. Thus :\n1.Present Indicative.\nactive. middle.\nd.p.s. d.\ndhardyami dhardyavas dhardyamas dhardye dhardyavahe dhardyamahe\netc. etc. etc. etc. etc. etc.\nThe 1stpll.act. inmasi greatly outnumber(as10to1)those inmas\ninbothRV.andAV.Noexample occurs of2dpi.act.inthana, norof\n3dsing. mid. ineforate.\n2.Present Subjunctive.\nForthesubjunctive may beinstanced alltheforms noted\nasoccurringintheolder language:\n1dhdrdyani dhdrdydva dhdrdydma dhdrdydi dhdrayavahdi\n(dhdrdydsi (dhdrdyddhve\n2 I_,_dharayathas dharayatha dharayase\n(dharayas (dhdrdyadhvdi\n\\dhdrdydti _(dhdrdydte\n3 { dharayatas dharayan ],,_ , ..dharayaite\n(dharayat(dharayatai\nOnly onedual mid. form inaite occurs:maddyaite (RV.). Theonly\nRV.mid. form inai,exceptin1stdu.,ismadayadhvai. Theprimary end-\nings in2dand3dsing.act. aremorecommon than thesecondary.\n3.Present Optative.\ndhardyeyam dhardyeva dhardyema dhardyeya dhardyevahi dhurdyemahi\netc. etc. etc. etc. etc. etc.\nOptativeforms arevery rare intheoldest language (four inRV., two\ninAV.); theybecome morecommon intheBrahmanas. AB.hasoncekamayita.\n4.Present Imperative.\ndhardya dhardyatam dhurdyata dhardyasva dhardyetham dhardyadhvam\netc. etc. etc. etc. etc. etc.\nImperative personswith theendingtatoccur: dharayatat (AV.)is2d\nsing.; gamayatatandcyavayatat (K. etc.), andvarayatat (TB.)areused as\n2dpi.; varayadhvcit (K. etc.)is2dpi.,andtheonly noted example (see\nabove, 570).\n22*\n"
+    },
+    {
+        "page": 356,
+        "content": "I\n340 XIV. SECONDARY CONJUGATION.[1043\n5.Present Participle.\ndhardyant (f. Ttft-yanfi) ^|(t(HIUI dhardyamana.\n6.Imperfect.\nddharayam ddharayava ddhurayama ddharaye ddharayavahi ddharayamahi\netc. etc. etc. etc. etc. etc.\nFor afewforms inIsand Itwhich perhaps belong totheimperfect,\nseebelow.\n1044. Aswasabove pointed out, theformations from thecausative\nstem inayaoutside thepresent-system areintheoldest language very\nlimited. InRV. arefound twoforms ofthefuture insyami, andtenin-\nfinitives indhydi ;alsoone ortwo derivative nouns intr(bodhayitr, coday-\nitri),five inisnu, seven initnu, and afew ina(atiparayd, nidharayd,\nvacaminkhayd, vi$vamejaya). InAV., alsotwos-future forms andfour ger-\nunds intva;andafewderivativenoun-stems, from oneofwhich ismade\naperiphrastic perfect (gamayam cakdra}. IntheBrahmanas, verbal deriva-\ntiveforms become more numerous andvarious,aswillbenoted indetail\nbelow.\n1045. Perfect. Theacceptedcausativeperfectisthe\nperiphrastic (1070),thederivative noun in5TFa,inaccusa-\ntiveform,towhich theauxiliaryisadded, being formed\nfrom thecausative stem: thus,\nTT5[Ert cjsfil^ dharayamcakaraSTTfETttT^ITdharayhmcakre.\nOfthisperfect noexampleoccurs inRV. orSV. orVS., onlyone\ngamaydm cakdra inAV., andbuttwoorthree inallthevarious texts\noftheBlack Yajur-Veda, and these notinthemantra-parts ofthetext.\nThey are alsobynomeans frequent intheBrahmanas, exceptinQli-\n(where they abound :chiefly, perhaps,forthereason that thiswork uses\ninconsiderable part theperfect instead ofimperfect asitsnarrativetense).\n1046. Aorist. The aorist ofthecausativeconjugation\nisthereduplicated, which ingeneralhasnothingtodowith\nthecausative stem, but ismadedirectly from theroot.\nIthasbeen already fully described(above, 856ff.).\nItsassociation with thecausative isdoubtless founded on\nanoriginalintensive character belongingtoitasareduplicated\nform, and isamatter ofgradual growth:intheVeda itis\nmade from aconsiderable number ofroots(inRV., more than\nathird ofitsinstances; inAV., about afifth) which have no\ncausative stem inaya.\n"
+    },
+    {
+        "page": 357,
+        "content": "1050]CAUSATIVE. 341\nThe causative aorist ofyJdhr. then, isasfollows :\nddidharam ddidharava ddldharama ddidhare ddldharavahi adidharamahi\netc. etc. etc. etc. etc. etc.\n1047. Inafewcases, where theroot hasassumed apecu-\nliarform before thecausativesignasbytheaddition ofap\nors(above,1042dif.)thereduplicatedaorist ismade from\nthisform instead offrom thesimpleroot :thus, atisthipam from\nsthap (stem sthapaya]forystha.Aorist-forms ofthis character\nfrom quasi-rootsinaparemade fromsthap, jnap, hap, jap,and\ncrap (above, 861);theonlyother example from theolder language\nisbibhisas etc.from bhis for~\\/bhi.\n1048. Asmall number ofVedic forms havingIsand Itafteryofthe\ncausal stem areapparently sporadic attemptsatmaking anfc-aorist :thus,\nvyathayls (AV.), dhvanayit (RV.; TS.hasinstead theanomalousdhvanayit},\nandailayit (AV.). Thetwoformer areaugmentless forms, used withma\nprohibitive (comparethedenom.unayls, RV., alsowithma).\n1049.Aprecativeisofcourse allowed bythegram-\nmarians tobemade forthecausativeconjugation:inthe\nmiddle, from thecausative stem with theauxiliarySisub-\nstituted for itsfinal^a;intheactive, from theform of\ntheroot asstrengthenedinthecausative stem, butwithout\nthecausativesign: thus,\nLntllHH dharyasametc.E-H^lUcfltl dharayisiyaetc.\nThis formation isdoubtless toberegardedaspurelyfictitious.\n1050. Futures. Both futures, with theconditional,\naremade from thecausative stem, with theauxiliary ^t,\nwhich takes theplaceofitsfinal fa.Thus:\nxS'-Future.\ndharayisyhmietc.^|^|6jo.6| dharayisyeetc.\nConditional.\nMadharayisyametc.fc||^|kj adharayisyeetc.\nPeriphrastic Future.\ndharayit&smietc.M(KJcn< dharayithheetc.\nThe 5-futureparticiplesaremade regularly:thus, dharayi-\nsyant (fern, -ydnfior-yatl), dharayisydmana.\n"
+    },
+    {
+        "page": 358,
+        "content": "342 XIV. SECONDARY CONJUGATION.[1050\nIthasbeen mentioned above thatRV. andAV. contain only two\nexamples each ofthes-future, andnone oftheperiphrastic. Theformer\nbegin toappearintheBrahmanas more numerously, but stillsparingly,\nwithparticiples, andconditional(only adharayisyat, QB.);ofthelatter, QB.\naffords twoexamples (parayitdsmi andjanayitdsi).\n1051. Verbal Nouns andAdjectives. These are\ninpartformed from thecausative stem inthesamemanner\nasthefutures;inpart namely,thepassive participlein\ncTtaandthegerundive andgerundinUya(and theroot-\ninfinitive) from thecausatively strengthened root-form.\nTheauxiliary ^iistaken inevery formation which ever\nadmits thatvowel.\nThus, offormationspermittedinthe laterlanguage (but\ntheexamples taken from theearlier):\nparticipleinta:irita, vasitd, cravitd;\ngerundiveintavya: tarpayitavya, kalpayitavya, gamayitavya,\nlhaksavitavya ;<7 e7 f\ngerundiveinya:sthapya, yajya ;\ninfinitive inturn:josayitum, dharayitum, janayitum, parayittim;\ngerundintva:kalpayitva, sadayitvn, -arpayitvti (AV.:see\n990), -rocayitva (TA.), crapayitva (AB.);\ngerundinya:-gharya, -padya, -vasya, -sadya, -sthapya ;\ngerundinam:-sthapam (QB.).\nFurther, offormations found only intheolder language:\nroot-infinitive, accusative: -sthapam (PB.);\ninfinitive intavai :jdnayitavaf, tdrpayitavat, pdyayitavai, -kalpayitavaf,\n-Ccotayitavaf (all$B.);\ninfinitive indhyai: isayddhyai, irayddhyai, tahsayddhyai, na$ayddhyai,\nmandayddhyai, madayddhyai. risayddhyai, vartayddhyai, vajayddhyai,\nsyandayddhyai (allRV.);\ngerundive inayya:panaydyya, sprhaydyya, trayaydyya (ytra:?).\nAllthese,itwillbenoticed, follow thesame rule astoaccent with\nthesimilar formations from thesimple root, showing notrace ofthespecial\naccent ofthecausative stem.\n1052. Derivative orTertiary Conjugations.\nFrom thecausative stemmaybemade apassive andade-\nsiderativeconjugation. Thus :\na.Thepassive-stemisformed byaddingtheusualpas-\nsive-signTydtothecausatively strengthened root, thecaus-\native-sign being dropped: thus, STRTrT dharydte.\n"
+    },
+    {
+        "page": 359,
+        "content": "1055] CAUSATIVE; DENOMINATIVE. 343\nSuchpassives arenotfound IntheVeda, butthey aremetwith here\nand there throughout theBrahmana language:examples arejnapyd- (TS.),\nsadya- (K.),padya- (AB.), vadya- (TB.), sthapya- (GB.),and soon.\nb.The desiderative stem ismade byreduplication and\naddition ofthesign^isa,ofwhich theinitial vowelreplaces\nthefinal ofthecausative stem :thus, (^MI^fllMTH didharayisati.\nThese, too, arefound here and there intheBrahmanas and later:\nexamples arepipayayisa (K.), bibhavayisa andcikalpayisa andlulobhayisa\n(AB.), didrapayisa andriradhayisa andapipayisa (QB.), and soon.\nAstocausatives made from thedesiderative stem, seeabove, 1039.\nV.Denominative.\n1053.Adenominativeconjugationisonethathasfor\nitsbasis anoun-stem.\nItisaviewnowprevailinglyheld thatmost ofthepresent-\nsystemsoftheSanskrit verb, along with other formations anal-\nogous with apresent-system,areintheir ultimateorigin denom-\ninative\n;and thatmany apparentroots areofthesame character.\nThedenominatives which aresocalled differ from theseonlyin\nthat theiroriginisrecent andundisguised.\n1054. Thegrammariansteach thatanynoun-stem in\nthelanguage maybeconverted, without other addition than\nthat ofan fa(asunion-vowel enablingittobeinflected\naccordingtothesecondgeneral conjugation)intoapresent-\nstem, andconjugatedassuch.\nAsexamplesofwhat isallowed inthisway,aregiven\nkrswati, 'acts like Krishna'; malati,'islike agarland (mala)';\nkavayati, \"playsthepoet (kam)'; bhavati,'islike theearth(bhuf;\ncrayati, 'resembles Qri (goddess)'; pitarati,'acts thefather';\nrajanati,'iskingly'. Butsuch formations are atthebest of\nextremerarityinactual use. TheRV. hasafew isolated and\ndoubtfulexamples,the clearest ofwhich isbMsd&ti,'heheals',\nfrombhisaj, 'physician' ;itismade like aform oftheroot-class;\nabhisnak seems tobe itsimperfect accordingtothenasal class.\nAndpatyate,'herules', appearstobeadenominative ofpdti,\n'master'. Otherpossiblecases are(Delbriick)isanas etc., krpd-\nnanta, tarusema etc., vanusanta, bhurafanta,vdnanvati. None of\ntheotherVeda orBrahmana texts hasanythingadditional ofthe\nsame character.\n1055. Ingeneral,thebase ofdenominative conjugation\n"
+    },
+    {
+        "page": 360,
+        "content": "I\n344 XIV. SECONDARY CONJUGATION.[1055\nismade from thenoun-stem bymeans oftheconjugation-\nsignETya-which hastheaccent.\nThe identity ofthisyawith theyaoftheso-called causativeconjugation,\nasmaking with the final aofthenoun-stem thecausative-sign aya,is\naltogether probable. What relation itsustains totheyaofthe t/a-class\n(VIII.),ofthepassive, and ofthederivative intensive stem,ismore open\ntoquestion.\n1056. Intermediate between thedenominative andcausative\nconjugationsstands aclass ofverbs, plainly denominative in\norigin, buthavingthecausative accent.Examples, beginning\ntoappearatthe earliest periodofthelanguage,aremantrdyate\n(from mantra, yman-\\-tra]andkirtdyati (from kirti,ykr 'praise').\nThese, along with likeforms from roots which have noother\npresent-system (though theymaymakescattering forms outside\nthatsystem from therootdirectly),orwhich have thisbeside\notherpresent-systemswithout causative meaning,arereckoned\nbythegrammariansasaseparate conjugation-class (above, 607).\n1057. Denominatives areformed atevery periodinthe\nhistoryofthelanguage,from theearliest down.\nTheyaremost frequentinRV., which contains over a\nhundred, ofallvarieties\n;AV. hasonlyhalf asmany (and per-\nsonal forms from hardlyathird asmany:from the rest, present\nparticiples,orderivativenouns); AB., lessthan twenty; QB.,\nhardly more than adozen; and soon. Inthelater language\ningeneral, theyare farfrom numerous;andmost ofthose\nwhich occur are\"cwr-class\" verbs.\n1058. Thedenominative meaning is,asinother lang-\nuages,ofthegreatest variety;some ofthemost frequent\nforms ofitare: 'belike, actas,playthepartof, 'regard\nortreat as','cause tobe.makeinto','use.makeapplica-\ntionof, 'desire, wish for, crave' thatwhich issigni-\nfiedbythenoun-stem.\nThemodes oftreatment ofthestem-final arealso various;\nandthegrammarians make acertain more orless definite as-\nsignment ofthevarieties ofmeaningtothevarieties ofform;\nbut this allotment findsonlyadubious supportintheusages\nofthewords asmetwith even inthelater language,and still\nless inthe earlier. Hence theformal classification, according\ntothe final ofthenoun-stem, and thewayinwhich this is\ntreated before thedenominativesign yd,willbethebestone\ntofollow.\n"
+    },
+    {
+        "page": 361,
+        "content": "10631 DENOMINATIVE. 345\n1059.From stems ina.a.The final aofanoun-\nstem oftenest remains unchanged:thus, amitraydti, 'playsthe\nenemy,ishostile'; devaydti, 'cultivates thegods,ispious'.\nb.Final aislengthened: thus, aghaydti, 'plans mischief;\npriyaydte,'holds dear'; acvaydti,'seeks forhorses'; acanaydti,\n'desires food'.\nIntheVeda, forms ofthesame verb with short andlong abefore ya\nsometimes exchange with oneanother.\nc.Itischangedto~i,orrarely ;thus, adhvariydti, 'per-\nforms thesacrifice'; tavislydti,'ismighty'; piitriydtiorputriy&ti,\n'desires ason'; mahsiydti,'craves flesh'.\nDenominatives ofthisform show aspecial proclivity toward themean-\ning 'desire'.\nd. Itisdropped (afternorr}:thus, turanydti,'israpid';\nadhvarydti, 'performs the sacrifice'.\ne.Other modes oftreatment aresporadic:thus, theaddition\nofs,asinstanasyati,'seeks thebreast'; thechange ofatoe,\nasinvarcydti, 'plays thewooer' .\n1060.From stems ina.Final ausually remains, as\ningopaydti, 'playstheherdsman, protects'; prtanayati, 'fights';\nbut itissometimes treated intheother methods ofana-stem :\nthus, prtanyati, 'fights'.\n1061.From stems ine,i,andtt,u.Such stems are\n(especiallythose inu,u]much lesscommon. They showreg-\nularlyiandubefore ya: thus, aratiydti (also -tiy-}, 'plotsin-\njury'; jamydti (also -my-),'seeks awife'; sakhiydti,'desires\nfriendship'; catruydti,'acts the foe'; rjuydti,'isstraight';\nvasuydti,'desires wealth'; asuydti, 'grumbles,isdiscontent': with\nshort u,gatuydti,'sets inmotion'.\nMorerarely,iistreated asa(orelse isgunated, with loss ofay}:\nthus, dhunaydti, 'comessnorting'. Sometimes, astoa(above, e),asibilant\nisadded: thus, avisydti,'isvehement'; urusydti,'saves'. From dhl,RV.\nmakesdhiyaydte.\n1062.From other vowel-stems, a.Final rissaid\ntobechangedtori:thus, pitriydti,'isfatherly': noexamplein\nusehasbeen noted.\nb.Thediphthongs,inthefewcases thatoccur, have their\nfinalelement changedtoasemivowel :thus, gavydti,'seeks cattle,\ngoes a-raiding'.\n1063.From consonant-stems. Afinal consonant\nusually remains before ya:thus, bhisajydti, 'playsthephysician,\ncures'; uksanydti,'acts likeabull'; apasydti,'isactive'; namasydti,\n'pays reverence'; sumanasydte,'isfavorably disposed'; tarmydti,\n'fights'.\n"
+    },
+    {
+        "page": 362,
+        "content": "346 XIV. SECONDARY CONJUGATION.[1063\nBut afinalnissaid tobesometimes dropped,andthepreceding vowel\ntreated asafinal :thus, rajaydteorrajlydti,'iskingly', fromro/an: vrsaydte\nfrom vrsan istheonlyexample quotable from theolder language. Sporadic\ncases occur ofother final consonants similarly treated: thus, ojaydte from\nojas ; while, ontheother hand, ana-vowel isoccasionally added tosuch\naconsonant before ya:thus, isaydti from10,satvanayati from satvan.\n1064. Byfarthelargestclass ofconsonantal stems are\nthose showingasbefore theya;and, ashasbeen seen above,\nasibilant issometimes, byanalogy, added toafinal vowel,\nmakingthedenominative-sign virtually syaoreven, with a\nalsoadded after aniorw-vowel, asya; and thiscomes tobe\nrecognisedinthelaterlanguageasanindependent sign, forming\ndenominatives that expressdesire :thus, madhusyatiorma-\ndhvasyati, longsforhoney'; ksirasyati,'craves milk'.\n1065. Thegrammarians reckon asaspecial class ofdenominatives in\nkamya what arereally onlyordinary onesmade from acompound noun-stem\nhaving kama asitsfinalmember :thus, rathakamyati, 'longs forthechariot'\n(K.: onlyexample noted from theolder language); putrakamyati,'desires a\nson' coming from thepossessive compounds rathakama, putrakama. And\nsatyapayati,'declares true' (from satya],isanexample ofyetanother form-\nation declared tooccur.\n1066. a.Anumber ofdenominative stems occur intheVeda forwhich\nnocorresponding noun-stems arefound, although for allornearlyallofthem\nrelated wordsappear: thus, ankuya, stabhuya, isudhya; dhisanya, risanya,\nruvanya, huvanya, isanya; ratharya, fratharya, saparya; irasya, dapasya,\nmakhasya, panasya, sacasya. Those inanya, especially, look likethebegin-\nnings ofanewconjugation-class.\nb.Havingstillmore thataspect, however, areaVedic group ofstems\ninaya,which ingeneral have allied themselves topresent-systems ofthe\nna-class(V.), andarefound alongside theforms ofthat class: thus, grbhaydti\nbeside grbhndti. Ofsuch. RV. has grbhaya, mathaya, prusaya, musaya,\n$rathaya, skabhaya, stabhaya. Afew others have nona-class companions:\nthus, damaya, camaya, tudaya (AV.); andpanaya, nafaya, vrsaya (]/ur\n'rain'), vasaya (yvas 'clothe'), andperhaps afaya (j/ap 'attain').\n1067. Thedenominative stems inRV.andAV.with causative accent-\nuation are :RV. ankhdya, arthdya, isdya (also isayd)^ urjdya, rtdya, krpdya,\nmantrdya, mrgdya, vavrdya, vajdya (also vajayd), vlldya, susvdya (also\nsusvayd); AV. adds kirtdya, dhupdya, paldya, virdya, sabhagdya.\nThe accent ofdnniya andhdstaya (RV.)iswholly anomalous.\n1068. Inflection. The denominative stems arein-\nflected withregularityliketheother stemsendingin51a\nthroughoutthepresent-system.Forms outside ofthatsys-\n"
+    },
+    {
+        "page": 363,
+        "content": "1070]DENOMINATIVE. 347\nternexcept from thestems which arereckoned tothe\ncausative orcwr-class, andwhich follow inallrespectsthe\nrules forthat class areoftheutmostrarity.\nInRV. occurs noform notbelonging tothepresent-system,unless(as\nseems mostlikely) unayls (withmdprohibitive)istoberegarded as2d\nsing, oftheto-aorist. Unquestionable examplesofthis aorist areasuyit (B.),\npapayista (TS. iii.2.83\n:pi., withmdprohibitive) and avrsayisata (VS.).\nTheform dsaparyait (AV.xiv. 2.20),with aifor z(555 b),might beaorist;\nbut, asthemetre shows,isprobablyacorrupt reading; amanasyait, certainly\nimperfect, appears tooccur inTB.(ii.3.S3\n).B.hasthefuturegopayisyati,\nandTS.theparticiples kanduyisydnt andkanduyitd. From roots assimilated\ntothecausatives occur intheolder language mantrayam asa(AB., GB.),\nmantraydm cakratus andcakre(QB.), mantritd(QB., TA.), -mantrya (TB.),\nandone ortwoother likeforms. Thegerundival adjectives saparyenya and\natasayya also aremet with.\nCHAPTER XV.\nPERIPHRASTIC ANDCOMPOUND CONJUGATION.\n1069. ONEperiphrastic formation, theperiphrastic\nfuture, hasbeenalreadydescribed (942 fT.),since ithas\nbecome inthelaterlanguageanecessary partofeveryverb-\nalconjugation,andsince, thoughstillremainingessen-\ntially periphrastic,ithasbeen sofused initspartsand al-\ntered inconstruction astoassume inconsiderable measure\nthesemblance ofanintegraltense-formation.\nByfarthemostimportantother formation ofthe\nclass is\nThePeriphrastic Perfect.\n1070. This(thoughalmost unknown intheVeda, and\ncoming only graduallyinto use intheBrahmaias)isa\ntensewidely made andfrequentlyused inthe classical\nSanskrit.\n"
+    },
+    {
+        "page": 364,
+        "content": "348 XV.PERIPHRASTIC ANDCOMPOUND CONJUGATION.[1070\nItismade byprefixingtheaccusative ofaderivative\nnoun-stem in TTa(accented)totheperfecttense ofan\nauxiliaryverb :namely,of]/5fjkr'make', more often of\ny^fc^as'be1\n,andvery rarelyof]/>Tbhu,'be'.\nIntheolder language,asispointed outbelow,krisused asauxiliary\nalmostalone, andbhunot atall.Even inMBh., bhuhardly ever occurs\n(Holtzmann).\n1071. Theperiphrastic perfectoccurs asfollows:\na.Itistheperfectofthederivativeconjugations:in-\ntensive, desiderative, causative, anddenominative;thenoun\nin^Tabeing made from thepresent-stem which isthe\ngeneralbasis ofeachconjugation:thus, fromj/SfT budh,\nintensive emsthiiH bobudhdm. desiderative SfarHM bubhutsdm.o *s oo *s\ncausative SHMUIH bodhayam; denominative H^IUIH mantray-\nam.\nTheformation from causative stems, andfrom those denominatives which\nareassimilated tocausatives,isbyfarthemost frequent.\nb.Most rootsbeginningwith avowel inaheavy syl-\nlable(longbynature orlongbyposition) make thisperfect\nonly, andnotthesimple one :thus,^IHH asam fromyETITT\nas,f^Fl^iksamfromyjjfi^iks; ~S^\\^ubjamfrom y3&!R ubj.\nExceptedaretherootsapandanch, and those beginning\nwith abefore twoconsonants(and taking anasreduplication:\n788).\nc.The roots(that is,stems reckoned bythegrammarians\nasroots)ofmore than onesyllable have theirperfectofthis\nformation :thus, T^CJU^J^ cakasam.\nButurnu (712)issaid toform urnonava only, andjagr anddaridra\n(1020, 1024) tohave aperfectofeither formation.\nd.Afewotherscatteringroots :namely, ay,day,and kas\n,\nandoptionallyvidandus,andafewroots ofthereduplicating\nclass, bhi, bhr, hu,and hri. Allthesemake thederivative noun\nfrom theirpresent-stem:thus, dayUm, vidhm, os&m, bibhayhm,\nJuhav&m, bibhartim, jihrayhm (thesewithgunaofthe finalvowel\nbefore thea).\nAnoccasional exampleismetwithfrom other roots :thus,naydm from\nnl(pres.-stem naj/a); hvayam fromyhvd (pres.-stem hvaya).\n1072. Theperiphrastic perfectofthemiddle voice is\n"
+    },
+    {
+        "page": 365,
+        "content": "1074]PERIPHRASTIC PERFECT. 349\nmadeonlywith themiddle inflection ofy^\\kr;that of\ntheactive, withanyoneofthethree auxiliaries. Forpas-\nsive use, theauxiliaries 3R7asand^bhu arealsoallowedx c\\\ntotake amiddle inflection.\nItisunnecessarytogive aparadigmofthisformation, as\ntheinflection oftheauxiliaries isthesame asintheirindepend-\nentuse(forthatofykr,see800f);ofybhu,see800b;ofy'cw,\nsee636).\nThe connection ofthenoun and auxiliary isnot soclose that other\nwords arenotsometimes allowed tocome between them :thus, tdmpatayam\nprathamam asa, 'him he firstmade tofall';prabhrancaydm yonaghusam\ncakarn, 'whomade Naghusha fallheadlong' (both Raghuvanc.a).\n1073. Theabove isanaccount oftheperiphrastic forma-\ntion with aderivative noun inamas itappearsinthelater\nlanguage ;earlier,itsaspectisquitedifferent :namely,asthat\nofamoregeneral, butquite infrequent, combination ofsuch a\nnoun with various forms oftheroot kr.Thus :\na.Offorms with the perfect ofthe auxiliary occurs only asingle\nexample inthewhole body ofVedic texts(metrical): namely, gamaydm\ncakara (AV. xviii.). InthebrahmanapartsoftheBlack Yajus texts are\nfound vidam cakara(TS., K.,MS.) andviddm cakrma(K.),andydjaydm\ncakara(K.). IntheBrahmanas, examples from causative etc.stems inaya\nbegin toprevail overothers, andinB.they arerather frequent. Examples\nfrom desiderative stems have been noted only fromQB.: they arecikramisdm,\nruruksam, dudhursdm, blbhatsdm. From simpleroots having thesame form-\nation inthe later language, occur viddm(TB., QB., GB.), asdm(QB., GB.),\niksdm (B., GB.), edhdm (8.), juhavdm (AB., TB., B.),bibhaydm ($B.);\nand alsolaydm (nilaydm) fromyil (CB.).\nb.Forms with theaorist oftheauxiliary areintheoldest Brahmanas\nasnumerous asthose with theperfect. Thus, with akar occurramayam\n(K.),janayam andsadayam andsvadayam andsthapayam (MS.); andwith\nakran, viddm (TS., MS., TB.). With the aorist optative orprecative has\nbeen noted onlypavaydrh kriyat (MS.).\nC.Like combinations with other tenses areexcessively rare, butnot\nentirely unknown: so,juhavam karoti(Qankh. Qr. Su.).\nd.With anyother auxiliary thanykr appears onlymantraydm asa\n(AB., GB.; in^!B.thesame noun iscombined withykrinmantrayarh\ncakratus andmantraydm cakre).\nAstheexamples show, thenoun(asinthecase ofthe\nperiphrasticfuture :945)has itsindependentaccent.\nParticipial PeriphrasticPhrases.\n1074. Combinations ofparticipleswithauxiliary verbs,\nofcondition ormotion, forming phraseswhich haveanoffice\n"
+    },
+    {
+        "page": 366,
+        "content": "350 XV.PERIPHRASTIC ANDCOMPOUND CONJUGATION.[1074\nanalogouswith that ofverb-tenses, arenotunknown inany\nperiodofthelanguage.\nTheyoccur even intheVeda, butarefarmorecommon\nandconspicuousintheBrahmanas, andbecomeagainoflittle\naccount inthelaterlanguage.\n1075. Examplesofthevarious formations areasfollows .\na.A(usually present) participle with thetenses oftheverb i.'go'.\nThis isthecombination, onthewhole, ofwidest andmost frequent occur-\nrence. Thus :yatha sucya vasah sarhdadhad iyadevam evai 'tabhir yajnasya\nchidram sarhdadhad eti(AB.), 'just asonewould mend[habitually]agarment\nwith aneedle, sowith these onemends anydefect ofthesacrifice'; agnir va\nidarh vdifvdnaro dahann ait(PB.), 'Agni Vaicvanara kept burning this\ncreation';te'surah pdrajitd ydnto dydvaprthivi updfrayan (TB.),'those\nAsuras, getting beaten,took refuge withheaven and earth';te(\nsyagrhdh\npafdva upamurydmana lyuh (B.), 'the animals, hisfamily, would becon-\ntinually destroyed'.\nb.Thesame with theverbcar, 'go(continuallyorhabitually)', signifying\nstillmore distinctly than theprecedingacontinued orhabitual action. Thus :\nagndv agnfy carati prdvistah (AV.), 'Agniisconstantly presentinthefire';\nadandyam dandena ghnantaf caranti(PB.), 'theymake apractice ofbeating\nwith arodwhat isundeserving ofpunishment'.\nc.The\"same with theverbsas, 'sit', andstha, 'stand', with alike\nmeaning. Thus, juhvata asate(K.), 'they continuesacrificing';telpakramya\nprativavadato 'tisthan (AB.), 'they, having gone off,keptvehemently refusing'.\nInthelater language,stha istheverb oftenest used, withpredicates,of\nvarious kind, tomake averbal phrase ofcontinuance.\nd.Aparticiplewith asand bhu,'be'. The participleisoftenest a\nfuture one; asonlyisused inthe optative, bhuusually inother forms.\nThus: yahpurvam ariijanah syat (AB.), 'whoever may nothavemade sacri-\nficebefore'; samavad evayajne kurvana asan(GB.), 'they didthesame thing\natthesacrifice'; parikridanta asan (MS.), 'theywere playing about'; itarame\nkena devatd upaptd bhavisyanti (AB.), 'wherewith shall theother deities be\nwonbyme?' ydtra suptvd punar nh'vadrasydn bhdvati(QB.), 'when,after\nsleeping, heisnotgoingtofallasleep again'; havyamhivaksyanbhavati\n(AB.),'forheisintending tocany the sacrifice'; ddsyant syat (K.), 'may be\ngoingtogive'; ytna vdhanena syantsydnt sydt (QB.),'with what vehicle he\nmaybeabout todrive'.\nCompositionwithPrepositionalPrefixes.\n1076. Alltheforms, personal andother, ofverbal con-\njugationofboth primaryandsecondary conjugation,\nandeven tosome extent ofdenominative(sofarasthe\n"
+    },
+    {
+        "page": 367,
+        "content": "PREFIXES.\ndenominative stems havebecome assimilated invalue to\nsimple roots)--occurvery frequentlyincombination with\ncertain words ofdirection, elements ofanadverbial character\n(seethenextchapter),theso-calledprepositions, according\ntotheoriginal useofthat term, ortheverbalprefixes.\nPractically, inthe laterlanguage,itisasifacompounded rootwere\nformed, outofrootandprefix,from which then thewhole conjugation (with\nmany derivatives :below, chap. XVII.)ismade, just asfrom thesimple\nroot. Yet, even there (andstillmore intheolder language: 1081), the\ncombination issoloose, andthemembers retain somuch oftheir independent\nvalue, that inmost dictionaries (that ofMonier Williams isanexception)\ntheconjugationofeach rootwithprefixesistreated under thesimple root,\nandnotinthealphabeticorder oftheprefix. Derivative words, however,\narebyuniversal agreement givenintheir independent alphabetic place,like\nsimplewords.\n1077. Those verbalprefixes which have value assuch\nthroughoutthewholehistoryofthelanguagearegiven\nbelow, inalphabetic order, with theirfundamental meanings:\ndtij 'across, beyond, past, over,toexcess';\nddhi, 'above, over, on,onto';\nami, 'after, along, toward';\nctntdr, 'between, among, within';\ndpa, 'away, forth, off';\ndpi, 'unto, closeuponoron';\nabhi, 'to,unto, against' (oftenwithimpliedvio-\nlence);\nRdva, 'down, off';\n5TT,'to,unto, at';\n3^\"ud, 'up,upforth orout';\n3qupa, 'to,unto, toward';\nftni,'down; in,into';\nFKT m's, 'out, forth';\n^^\\pdra,'toadistance, away, forth';\ncrf^pdri,'round about, around';\n\"3prdj 'forward, onward, forth, fore';\n"
+    },
+    {
+        "page": 368,
+        "content": "352 XV.PERIPHRASTIC ANDCOMPOUND CONJUGATION. 1077\nprdti,'inreversed direction, back tooragainst,\nagainst,inreturn';\n|cfm',-apart, asunder, away, out';\n?PTsdm, 'along, with, together'.\na.Some ofthese, ofcourse, areusedmuch more widely andfrequently\nthan others. Inorder offrequency intheolder language (asestimated by\nthenumber ofroots withwhich they arefound used inRV.andAV.), they\nstand asfollows: pra, a,vi,sam, abhi, ni, ltd,pari, arm, upa, prati, ava,\nnis, ati,apa, para, adhi, api, antar. Aptisofvery limited useasprefixin\nthelater language, having become aconjunction, 'too,also'.\nb.Themeanings given areonly theleading ones. Incombination with\ntheroots they undergo much modification, both literal and figurative yet\nseldom insuch away that thestepsoftransition from thefundamental\nsense arenoteasy totrace. Sometimes, indeed, thevalue ofaroot ishardly\nperceptibly modified bytheaddition oftheprefix. Anintensive force isnot\ninfrequently given bypari, vi,andsam.\n1078. Prefixes essentially akin with theabove, butmore\ndistinctly adverbial, andofmore restricted use, arethese:\ndcha(oracha\\ 'to,unto':tolerably frequentinRV. (used\nwith overtwenty roots),butalready veryrare inAV.(onlytwo\nroots), andentirelylostinthelater language ;\navis, 'forth tosight,inview': used onlywith theroots\nbhu, as,andkr;\ntiros, 'through, crossways;outofsight':hardly used except\nwith'kr,dha, bhu(inRV., with three orfourothers);\npurds,'infront, forward': usedwith onlyhalf-a-dozen roots,\nespecially kr,dha, i;\npradus,'forth toview':onlywith bhu, as,kr.\nAfewothers,asbahia, 'outside', vina, 'without', saksat,'inview', are\nstill lessremoved from ordinary adverbs.\n1079. Of stillmore limited use,and ofnoun rather than\nadverb-value are :\ntrad (orprat/i?), only with dha(inRV., once alsowithkr}:?raddha,\n'believe, credit';\nhin, onlywith kr(and obsolete intheclassical language): hinkr, 'make\nthesound hing, low,murmur'.\nAnd beside these stand yetmore fortuitous combinations: seebelow,\n1091.\n1080. More than oneprefix maybesetbefore thesame\nroot. Combinations oftwo arequiteusual;ofthree, much\nlesscommon; ofmore thanthree,rare. Their order isin\ngeneraldeterminedonlybytherequirementsofthemeaning,\n"
+    },
+    {
+        "page": 369,
+        "content": "1083]VERBAL ^PREFIXES. 353\neach addedprefix bringingafurther modification tothe\ncombination before which itisset.But 5TTaisnever allowed\ninthelaterlanguage, andonlyextremely rarelyintheolder,\ntobeputinfront ofanyoftheothers.\n1081. InclassicalSanskrit, theprefix alwaysstands\nimmediatelybefore theverbal form.\nIntheolder language, however, ofbothVeda andBrah-\nmana,itspositionisquitefree :itmaybeseparated from the\nverbbyanother word orwords, andmayeven(muchlessoften)\ncome after theform towhich itbelongs ;itmayalso stand\nalone, qualifyingaverb that isunderstood, orconjointlywith\nanotherprefix onethat isexpressed.\nThus, sddevdn e'hdvaksyati (RV.)f'heshallbring thegods hither';\npro,no,dyunsitdrisat (AV.), 'may helengthen outourlives';tdvdyatam\ntipadravdt(RV.), 'doyetwocome hither quickly1\n;gdmad vdjebhir dsdnah\n(RV.), 'mayhecome withgifts hither tous';pdrimdm pdrimeprajdm pdri\nnahpdhiydddhdnam(AV.), 'protect me,myprogeny, andwhat wealth we\nown'; ydtah sadyd dcapara caydnti (AV.), 'from whence every daythey\nadvance andretire'; vyahdrh sdrvena pdpmdnd [avrtam] vfydksmena sdm\ndyusd (AV. ),'Ihaveseparated from allill-luck, fromdisease, [Ihave joined\nmyself]with life'.\nThe separationofthe prefix from averbal noun oradjectiveisvery\nmuch more difficult, and ofquite rare occurrence.\n1082. Asregardstheaccent ofverb-forms compounded\nwithprefixes, onlythe case needs tobeconsidered inwhich\ntheprefixstands(asalwaysinthelaterlanguage) immediately\nbefore theverb\n;otherwise, verb andprefixaretreated astwo\nentirely independentwords.\n1083.Apersonal verbal form, ashasbeen seen above\n(592),isordinarily unaccented :before such aform, theprefix\nhas itsown accent;or,iftwo ormore precede thesame form,\ntheone nearest the latter issoaccented, andtheothers lose\ntheir accent.\nIf,however, theverb-form isaccented, theprefixorpre-\nfixes lose their accent.\nThatis,inevery case, theverb alongwith itsnormally\nsituatedprefixorprefixessofarconstitutes aunitythat the\nwhole combination isallowed totake butasingleaccent.\nExamplesare: pareJ\nhinari ptinare'hiksiprdm (AV.), 'goaway,\nwoman;come again quickly'; dthdJstam vipdretana (RV.), 'then scatter ye\naway toyour home'; samdcinusvd 'nusamprdyahi (AV.), 'gather together, go\nforth together after'; yddgrhdn upodatti (AV.), when hegoesup tothe\nWhitney, Grammar. 23\n"
+    },
+    {
+        "page": 370,
+        "content": "'\n354 XV.PERIPHRASTIC ANDCOMPOUND CONJUGATION.[1083\nhouse1\n;evdcatvdrh sarama ajagdntha (RV.), 'now that you, Saraina, have\nthuscome hither1\n;yend\"vistitah pravivefithd 'pah (RV.), 'envelopedinwhich\nthou didst enter thewaters'.\n1084.Aprefix,however, notseldom has amore inde-\npendent value, asageneraladverb ofdirection, orasapreposition\n(intheusual modern sense ofthatterm), belongingtoand\ngoverninganoun ;insuch case,itisnotdrawn intoform\npartofaverbal compound,buthas itsown accent. Thetwo\nkinds ofuseshade intooneanother, and arenot divisible by\nanydistinct andfixed line.\n1085. Incombination with thenon-personal partsofthe\nverb-system,withparticiples, infinitives, andgerunds,thegeneral\nrule isthat theprefixloses itsaccent,finfavor oftheother\nmember ofthecompound. But theprefixinstead hassometimes\ntheaccent :namely, when combined\na.with thepassive participleintaorna :thus, pdreta,\n'gone forth'; antdrhita, 'concealed'; dvapanna, 'fallen'; sdmpurna,\n'complete'.\nb.with theinfinitive intu(972),inall itscases :thus.,\nsdmhartum, 'collect'; dpidhatave,\"tocoverup'; dvagantos,'of\ndescending';thedoubly accented dative intavai retains itsfinal\naccent, butthrows theother backupontheprefix:thus, dnvetavai,\n'forfollowing'; dpabhartavai,'forcarryingoff.\n1086. The closeness ofcombination between therootand\ntheprefixisindicated notonly bytheirunityofaccent, but\nalsobytheeuphonicrules(e.g.185, 192),which allow the\nmutualadaptationsofthetwo tobemade tosome extent asif\ntheywerepartsofaunitary word.\n1087 .Afewspecial irregularitiescall fornotice :\na.Inthelater language, opi, adhi, andava, inconnection with certain\nroots, sometimes lose their initial vowel: namely, apiwith nahanddha,\nadhi with stha, avawith gah. IntheVeda, ontheother hand,isisina\nfewcases found instead ofniswithykr,\nb.The finalvowel ofaprefix, especially ani,is(oftenestintheolder\nlanguage) sometimeslengthened, especiallyinderivative words:thus, pratlkara,\nnivrt, parihara, virtidh, adhlvasd, dpivrta, abhlvartd anurudh;pravrs, upavasu.\nIntheVeda, the initial ofarm issometimes lengthened after negative\nan :thus, ananuda.\nC.Incombination withyi'go',theprefixes para, part, andprasometimes\nchange their rto i.\nInthiswayisformed akind ofderivative stem palay, 'flee',inflected\naccordingtothea-class,inmiddle voice, which isnotuncommon from the\nBrahmanas down, andhassolosttheconsciousness ofitsorigin that ittakes\n"
+    },
+    {
+        "page": 371,
+        "content": "1090]VERBAL PREFIXES. 355\ntheaugment prefixed:thus, apalayisthas ;itmakes theperiphrastic perfect\npalayam cakre. Thestempalyay, similarly inflected, seems tooccur only in\nB.;andplayhasbeen found nowhere except inMS.\nd.Therootkr'make1sometimes assumes(orretains from amoreoriginal\ncondition) aninitial safter theprefixes sam, pari, andupa: thus, samskurute,\nsamaskurvan, sarhskrta, pariskrnvanti, pariskrta, upaskrta. And]/fcr'scatter'\nissaidbythegrammarians toadd sinthesame manner, under certain\ncircumstances, afterupaand prati.\ne.Thepassive participleoftheroots da'give' andda'cut' hasoften\ntheabbreviated form tta after aprefixofwhich the -finalvowel,ifi,is\nlengthened (compare thesimilar contraction with otherelements, above, 955 c,\nandthat ofthederivative inti,below, chap. XVII.). Thus, inAV. arefound\nfromda'give', vydtta, pdritta, dpratitta:inBrahmanas, from thesame, atta,\npratta, apatta ;from da'cut', dvatta, nirdvatta, samdvatta.\nf.TheAB. hasonceniniyoja (insteadofni-yuyoja] from ni-\\~yyuj,\nandudaprapatat; and inMBh. arefound afew cases like vivyasa from\nvi-\\~yas (where, however, anintentional playonthewordmaybeassumed:\nDelbriick); andanvasarhcarat (insteadofanusamacarat).Such unifications of\nprefix androot, with treatment oftheresult after themanner ofasimple\nroot,areextremely rare.\nSome hold, however,that certain oftheapparent roots ofthelanguage\nareresults ofthis unification:thus, dpfrom a-\\-ap, vyac from in-j-ac, tyaj\nfromati-(-aj,etc. (seeWeber, Ind. Stud., xiii.61).\ng.The loss ofthe initial sofsthaandstambh after theprefix udhas\nbeen noticed above (233 a).\nAlso(137 a,b),certainpeculiaritiesofcombination ofaprefix with the\ninitial vowel ofaroot.\n1088. Astothemore general adverbial uses ofthepre-\nfixes, and theirprepositional uses, seethenextchapter.\n1089. The adverbial prefixes sw, 'well', anddus, 'ill', aresaid tobe\nsometimes combined with verbal forms\n;butnoexamples ofsuch combination\narequotable from accentuated texts.\nAstotheaddition ofthecomparative andsuperlativesuffixes taram and\ntamam toverbs, seeabove, 473.\nOther Verbal Compounds.\n1090. Ithasbeen seen above thatsome ofthepreposi-\ntionalprefixesareemployedincombination withonlyvery small\nclasses ofroots, namely those whose meaning makes them best\nfitted forauxiliary andperiphrasticuses such askr,'make',\nbhuandas, 'be', dha, 'put', i,'go' and that the first three\nofthese arewidely used incombination with aderivative in\namtomake aperiphrastic conjugation.Such roots have also\nbeen, from the earliest periodofthelanguage,butwith increas-\ningfrequency, used insomewhat analogouscombinations with\n23*\n"
+    },
+    {
+        "page": 372,
+        "content": "356 XV.PERIPHRASTIC ANDCOMPOUND CONJUGATION.[1090\nother elements, substantive andadjectiveaswell asadverbial\n;\nand thishasbecome, inpart, developed finallyinto aregular\nand indefinitelyextensible method ofincreasingtheresources of\nverbal expression.\n1091. Most analogous with hin-\\~ykr (1079)areafewother onomato-\npoetic compoundsintheVeda:akkhallkftya (RV.), 'making acrackling sound',\njanjanabhdvant (RV.), 'flimmering', andalalabhdvant(RV.), 'making merry';\nand(inAB.) bababakurvan, 'crackling'. AndAV. hasmasmasd withykr\n(TS.andVS., masmasd),'crush'.\nFurther, combinations ofykrwith terms used atthesacrifice,asvdsat,\nsvdha, svadhd, svaga:at first phrases only, andnoun-compounds, but\nbecoming-verbal combinations inwhich theprefixed word istreated like a\nprepositional prefix: thus, svagakardti (B.: butsvadhdkartiti, TA.); and\nother prefixesaresetbefore them,asanuvasatkuryat.\n1092. a.Thenoun namas, 'obeisance, homage', inastillmore purely\nnoun-value, becomes combined withykr:intheVeda, onlywith thegerund,\ninnamaskrtya (beside hastagfhya andkarnagfhya: above, 990.).\nb.Asolitary combination withyi, 'go',isshown bytheaccusative\ndstam, 'home'; which, appearing only inordinary phrasesinRV.,isinAV.\ncompounded with the participlesinastamydnt, astamesydnt, dstamita\n(with accent like that ofordinary compounds with aprefix) andinthe\nBrahmanas and the later languageistreated quitelike aprefix:thus,\nastameti(QB.).\nc.Other ordinary accusative forms ofadjectivesincombination with\nverbal derivatives ofkrandbhuarefound here andthere intheolder lan-\nguage:thus, prtamkftya andnagnamkrtya (TS.); nagnambhdvuka, pama-\nnambhdvuka,etc.(TS.etal.).\n1093. Intheearly butnotintheearliest language, anoun-\nstem thuscompoundedwith krorbhu, inverbal nouns and\nordinary derivatives, andthen also inverbal forms, beginsto\nassume aconstant endingi(ofdoubtfulorigin).\nThere isnoinstance ofthis inRV., unless the Iofakkhallkftya (above,\n1091)istobesoexplained.InAV.,besides theobscure vatikrta and\nvaiikard,isfound only phallkdrana. IntheBrahmanalanguage, examples\nbegintooccur occasionally cyetl andmithum inTS.; these andphali,\nkruri, udvasl inTB.; the firstthree, with suphali, ekl, sin,brahmam, and\ndaridri inQB.;and soon.The accent ofthecombination isingeneral\naccordance with theaccent ofcompounds with theusual prefixes; and ifthe\nprefixed stem takes thetone,this restsupon the final i.Sometimes amere\ncollocation takes place: thus, mithum bhdvantis(TS.), phali kriydmananam\n(TB.). vajri bhutvd(TA.). The Iisvariously treated :now asanuncombinable\nfinal, asinfyeti akuruta andmithuni abhavan(TS.); now asliable tothe\nordinary conversions,asinmithuny enaya syam andsvyakurvata (QB.).\nTheexamplesinaccentuatedtexts,andespeciallythose inwhich the\n"
+    },
+    {
+        "page": 373,
+        "content": "1096]NOUN ANDADJECTIVE-COMPOUNDS. 357\nverb isentitled tothetone, aretoofew tofurnish more than afragmentary\nillustration oftheformation.\nOut ofsuch beginnings hasgrown inthelater language thefollowing\nrule :\n1094. Anynoun oradjective stem isliable tobecom-\npounded with verbal forms orderivatives oftheroots]/5fi\nkrand^bhu(and,itissaid, ofETHasalso;butsuch ca-\nses,ifthey occur, areatleastextremely rare),intheman-\nnerofaverbalprefix.Ifthefinal ofthestem beanaor\ni-vowel,itischangedto^l\\ifanw-vowel,itischanged\nto^37u.\nItisprescribedalso that afinal rbecomen,and that as\nandanbechangedtot;butnogenuine examples appeartobe\nquotable.\nExamplesare :stambhibhavati, 'becomes apost'; ekacittlbhuya, 'becoming\nofonemind'; upaharikarosi, 'thou makest anoffering'; nakhapraharajarjankrta,\n'torn topieces with blows ofthe claws'; fithilibhavanti, 'become loose';\nkundallkrta, 'ring-shaped.'\n1095. Ofalltheforms which constitute orareattached to\ntheverbalsystem,thepassive participleistheonemostclosely\nassimilated initstreatment asacombinable element toanor-\ndinary adjective. Next toitcome thegerund andthegerund-\nives. Combinations ofthekind here treated ofareespecially\ncommon withpassive participles andgerunds.\nCHAPTER XVI.\nINDECLINABLES.\n1096. THE indeclinable words are lessdistinctlydivi-\ndedintoseparate partsofspeechinSanskrit than isusual\nelsewhere inIndo-European language especially owing\ntothefact that theclass ofprepositions hardlyhasareal\nexistence, but isrepresented bycertain adverbial words\nwhich aretoagreaterorlessextent usedprepositionally.\nThey will, however, bebrieflydescribed here under the\nusual heads.\n"
+    },
+    {
+        "page": 374,
+        "content": "358 XVI. INDECLINABLES.[1097\nAdverbs.\n1097.Adverbs bySuffix. Classes ofadverbs, some-\ntimes ofconsiderable' extent, areformed bytheaddition\nofadverb-makingsuffixesespeciallytopronominalroots or\nstems, butalso tonoun andadjectivestems.\nThere isnoultimate difference between these suffixes andthecase-\nendings indeclension;andtheadverbs ofthis division sometimes areused\ninthemanner ofcases.\n1098. With the suffix tasaremade adverbs having an\nablative sense, andnotrarelyalsoanablative construction. Such\naremade :\na.From pronominal roots, indtas, itds, tdtas, ydtas, ktitas, amutas,\nsvatas(notfoundearlier); andfrom thepronominal stems in tord(494) of\nthepersonal pronouns;thus, mattds (onlyexample inV.), tvattas, asmattas,\nyusmattas.\nb.From noun and adjective stems ofevery class, since the earliest\nperiod,butmore freely later :thus, mukhatds, agratds, rbhutds, hrttas,\nCirsatds, nastds, yajustas, pardtas, anydtas, sarvdtas, daksinatds, abhipatds\n(once,inRV., from acase-form:patsutds}.\nc.From afewprepositions:thus, abhttas, parttas, dntitas.\nExamplesofablative construction are :tdtah sasthdt(AV.), 'from that\nsixth';tdtojydyan (AV.), 'older than they'; kuta? ciddefad agatya (H.),\n'arriving fromsome regionorother'.\nButthedistinctive ablative meaning isnotinfrequently effaced, andthe\nadverb hasamore locative value :thus, agratds,'infront';asmatsamlpatas,\n'inourpresence'; dharmatas,'inaccordance withduty'; chagatas (H.), 'with\nreference tothegoat1\n.\n1099. With the suffix tra(inV.oftentra)aremade\nadverbs having alocative sense, andoccasionallyalso alocative\nconstruction.\nThese adverbs areformed from pronominal roots, namely dtra, tdtra,\nydtra, ktitra, amtitra, asmatrd, satrd; and alsofrom noun and adjective\nstems,asanydtra, vi$vdtra, samandtra, martyatrd, daksinatrd, devatrd,\npurutrd, bahutrd.\nThewords in(accented)trdareVedic only, except satrd, 'altogether' (of\nwhich satram isalsogivenasanalternative form).\nExamplesofquasi-locative orlocative construction are :hdsta ddaksinatrd\n(RV.),'intheright hand';tatra 'ntare(H.),'inthatinterval'; prabhutvamtatra\nyujyate (H.), 'sovereigntyissuited tohim'; ekatrapuruse (MBh.),'inasingle man'.\nAsthe locative case isused also toexpressthegoalofmotion (304),\nsotheadverbs intrahave sometimes anaccusative aswell asalocative\nvalue :thus, tatra gacha, 'gothere orthither'; pathddevatrd ydnan (RV.),\n'roads that gotothegods'.\n"
+    },
+    {
+        "page": 375,
+        "content": "1103]ADVERBS BYDERIVATION 359\n1100. One ortwo other suffixes oflocalityare :\na.ha,inzM, 'here', k-'tha, 'where?1andtheVedic vipvdlia (alsovifvdhd,\nvifvcihd), 'always' (compare below,\",! 104, end).\nb.tat,which isadded towords having already alocal ordirective\nvalue: thus,toadverbial accusatives, asprdktdt, tidaktdt; toadverbial abla-\ntives,asdrattdt, uttarattdt, pardkattdt; and toprepositional adverbs, as\nadhdstdt, avdstdt, pardstdt, purdstdt, bahistdt. Apparently byanalogy with\nthese last, thesuffix hasoccasionally theform stdt :thus, updristat.\nC.hi,inuttarahi(QB.)anddaksinahi (nooccurrence).\n1101. Bythesuffix thaaremade adverbs ofmanner, es-\npecially from pronominalroots orstems.\nThus, tdthd, ydthd;katha and ittha(bythe side ofwhich standkathdm\nanditthdm}', andtherareimdthd andamuthd. And dtha (V.oftendthd],'so\nthen',doubtless belongs with them. Further, from afewadjective andnoun\nstems :as,vifvdthd, sarvdthd, anydthd, ubhaydthd, itardtha, yatamdthd,\nurdhvdthd, rtutha, ndmdthd(once, AV.).\nYatha becomes usually toneless inV.,when used inthesense ofiva\nafter anoun forming thesubject ofcomparison: thus, tdydvo yathd (RV.),\n'like thieves'.\n1102. One ortwoother suffixes ofmanner are:\na.ti,initi, 'thus', verycommonly used, from theearliest\nperiod,asparticleofquotation, followingthewords quoted.\nExamplesare :brahmajdye ''yamiticeddvocan(RV.),'iftheyhave said\n\"this isaBrahman's wife\"';tdrhdevd abruvan vrdtya kimnutisthasi 'ti\n(AV.),'thegodssaid tohim: \"Vratya, why doyoustand?\"1\nOften, the iti\nisusedmore pregnantly:thus, yahfraddddhdti sdnti devd iti(AV.), 'whoever\nhasfaith that thegods exist'; tarnvydghram munir musiko 'yamitipa?yati\n(H.),'the sage looks uponthat tigerasbeing really amouse'; yuyarh kim\nitisidatha(H.), 'why (lit'ly, alleging what reason) doyousit?' Orthe iti\nmarks anonomatopoeia,orindicates agesture:thus,bahfs teastu bal iti\n(AV.),'let itcome outofyouwith a\"splash\"'; itydgre krsatydthe 'ti(^B.),\n'heploughsfirst this way, then this way'.Aword made byitilogically\npredicatetoanobjectisusually nominative: thus, svargo lokd itiyarn vddanti\n(AV.),'what they call\"theheavenly world\"'; vidarbharajatanaydm damayanti\n'tividdhimam(MBh.), 'knowmefortheVidarbha-king's daughter, Damayanti\nbyname'; butajnam bdlamitydhuh (M.), 'theycallanignorant man achild'.\nWith thesuffix ofitiistobecomparedthat oftdti etc.(519).\nb.vainiva(toneless), 'like, as',andevd(inV.often evd\\earlier\n'thus',later aparticle emphasizingthepreceding word;for'thus' isused\nlater therelated evam, which hardly occurs inRV.,andinAV. onlywith\nyvid:as,evdm vidvdn, 'knowing thus'.\nInlaterVedic (AV. etc.) ivamore often counts foronly asingle syllable.\n1103. a.Bythesuffix daaremade adverbs oftime, but\nalmost onlyfrom pronominalroots.\n"
+    },
+    {
+        "page": 376,
+        "content": "360 XVI. INDECLINABLES.[1103\nThus, tadd, yadd, kadd, idd(only inV.); and sdda, beside which is\nfound earlier sddam. Besidesthese, inthe older language, onlysarvadd;\nlater afewothers,asanyada, ekada, nityadd.\nb.Bytheperhaps related danlm aremade iddnlm, taddnlm, vifvaddriim.\nC.With rhiaremade, frompronominal roots, tdrhi, etdrhi, ydrhi, kdrhi,\namtirhi.\nd.The suffix di,found only inyddi, 'if,isperhaps related withda,\ninform asinmeaning.\n1104. Bythesuffix dhaareformed adverbsespecially from\nnumerals, signifying '-fold, times, ways',etc.\nThus, ekadhd, dvidhd (also dvfdha anddvedhd), trfdhd(intheoldlan-\nguage usually tredhd), saddhd(also sodhd andsaddhd), dvada^adhd, sahasradhd,\nand soon. Also, naturally, from words having aquasi-numeral character :\nthus, tatidha, bahudha, purudha, vifvddhd, facvadha, etdvaddha. Inavery\nfewcases,alsofrom general noun andadjective stems :thus, mitradhd(AV.),\npriyadhd (TS.), rjudhd (TB.), paristubdhd (PB.).\nTheparticle ddha orddha,aVedic equivalent ofdtha, probably belongs\nhere (purudhd andvifvddha, with shortenedfinal, occur afewtimes inRV.);\nalsoaddhd,'intruth'; andperhaps sahd, 'with', which hasanequivalent\nsadha- inseveral Vedic compounds. And theother adverbs inha(1100 a)\nmaybeoflike origin.\n1105. From afewnumerals aremade multiplicative adverbs with s:\nnamely, dvfs, frfs,and cattir(probably,forcaturs).\nThecorresponding word for'once', safcft,isacompoundrather than a\nderivative; andthesame character belongsstillmore evidently topancakftvas,\nnavakftvas, aparimitakrtvas, etc., thoughkrtandkrtvas areregarded bythe\nnative grammariansassuffixes (AV. hasdd$a krtvas andsapid krtvas).\n1106.Bythesuffix cdsaremade, especially fromnumeral\norquantitative stems, adverbs ofquantityormeasure ormanner,\ngenerally useddistributively.\nThus, efcafas, 'onebyone', patafds, 'byhundreds', riufds, 'season by\nseason', pacchas,'foot byfoot', aksara^ds, 'syllable bysyllable', ganafds,'in\ncrowds', stambafds, 'bybunches', pamppds,'limb bylimb', tdvacchds,'in\nsuchandsuchnumber orquantity': and, inamore general way, sarva$dsi\n'wholly', mukhyafas, 'principally', krchra$as, 'stingily'.\n1107. Bythesuffix vataremade withgreat freedom, in\nevery periodofthelanguage,adverbssignifying'after theman-\nnerof,like',etc.\nThus, angirasvdt,'likeAngiras', manusvdt(RV.),'asManudid',jamad-\nagnivdt,'after themanner ofJamadagni'; purvavdt orpratnavdtorpurdnavdt,\n'asofold', kakatallyavat,'after thefashion ofthecrow andthepalm-fruit'.\nThis isreally theadverbially used accusative (withadverbial shift of\naccent :below, 1111e)ofthesuffix vant (next chapter), which intheVeda\nmakes certain adjective compoundsofasimilar meaning:thus, tvavant,'like\nthee', mdvant,'ofmy sort',etc.\n"
+    },
+    {
+        "page": 377,
+        "content": "1111]CASE-FORMS ASADVERBS. 361\n1108. Bythesuffix sataremade adverbs signifying 'into thecondition\nof,which areused along with verbs ofbecoming and ofmaking.\nThus, agnisat -f-)//cr, 'reduce tofire,burnup';bhasmasat-f-ybhu,'turn\ntoashes'; atmasatkrta, 'made one's own'.\nThese derivatives areunknown intheearlier language, andnotcommon\ninthe later. The sofsat isnotliable toconversion into s.Theconnection\nwith thefollowing verb isnot soclose astorequire theuseofthegerund\ninya:thus, bhasmasatkrtva(not-krtya: above, 990).\n1109. Suffixes,notofnoun-derivation orinflection, maybetraced with\nmore orlessplausibilityinafew other adverbs. Thus, forexample,in\npratdr, 'early', andsanutdr, 'away'; indaksintt,'with right hand', andcikitvtt,\n'with consideration'; innwndm, 'now', andnandndm, 'variously'. Butthe\ncases areinthemain toorareanddoubtful tobeworth notice here.\nTheadverbs ofthis division arealmost never usedprepo-\nsitionally. Those ofthenext division, however, areinmany\ninstances soused.\n1110.Case-forms used asAdverbs. Alargenum-\nber ofadverbs aremore orlessevidentlycases inform,\nmade from stems which arenototherwise inuse. Also\ni\nmanycases ofknown stems, pronominalornoun oradject-\nive, areused withanadverbial value,, being distinguished\nfrompropercases bysome difference ofapplication, which\nissometimes accompanied byanirregularityofform.\n1111. Theaccusative isthecasemostfrequently and\nwidely usedadverbially. Thus:\na.Ofpronominalstems: as,t/ad, 'if,when, that', etc.; tad, 'then'etc.;\nfcfm, 'why, whether', etc.; iddm, 'now, here'; adds, 'yonder'; and soon.\nOflike value, apparently,arethe(mostly Vedic) particles kdd,kdmand\nkam(?), id,cid(commonatevery period), smdd andsumdd, imandslm\n(bysome regardedasstillpossessing pronoun-value),-kirn.\nCompoundswith fdareced, 'if, ncd, 'lest', svid, kuvfd;withcid,\nkucid;with -kim, ndklm andmakim, andaklm.\nb.Ofnoun-stems :as, noffto, 'byname'; stikham, 'happily'; kdmam,\n'atwill,ifyouplease'; n'iktam, 'bynight'; raftas, 'secretly'; osdm, 'quickly'\n(V.);and soon.\nC.Ofadjective stems, ingreat numbers :as,satydm, 'truly'; cirdm,\n'long'; nttyam, 'constantly'; bh&yas, 'more, again'; and soon.\n1.The neuter singularisthecasecommonly employed inthisway;\nand itisformed andused adverbially from alargeclass ofcompound stems\nwhich donotoccur inadjective use (the so-called avt/at/i&/iaua-compounds:\nbelow, chap. XVIII.).\n"
+    },
+    {
+        "page": 378,
+        "content": "362 XVI. INDECLINABLES.[1111\n2.Butthefeminine singularalso issometimes used; especially inthe\nso-called adverbial endings ofcomparison, taram andtamam, which are\nattached toparticles,andeven, asitisclaimed(473),toverb-forms :thus,\npratardm, pratamdm, uccaistaram, fanaistaram, jyoktamdm.\nIntheoldest language (RV.andAV.),theneuter instead ofthefeminine\nform ofthese suffixes isalmost alone inuse: see1119.\nd.Many adverts ofobscure form orconnection aretobeexplained with\nprobabilityasaccusatives ofobsolete noun oradjective stems:examples are\ntusnim,'insilence'; saydm,'atevening'; dlakam, 'vainly'; sakdm, 'together,\nwith(prep.)'; dram oratom,'sufficient'(inthelaterlanguage used withykr\ninthemanner ofaprefix); prayas, 'usually'; Isdt, 'somewhat'; amnds, 'un-\nexpectedly'; bahfs, 'outside'; mtthu andmithds, mtihu andmuhus, jdtu,\nand soon.Madrtketc., andninfk(inRV.),areperhaps contracted forms\nofadjectives having j/oeoranc astheir final(407 ff.);andvisundk and\nvfthak, with prthak andfdhak, maybeofthesame character. Thepresence\nofother roots asfinalmembers isalsoprobableforu^ddhak, anusdk and\ndyusdk, anusthu and susthu, yugapdt,etc. Comparealsotheforms inam\nbeside those in5,above, 1099, 1101, 1103a.\ne.In(Vedic) dravdt, 'quickly',istobeseen achange ofaccent for\ntheadverbial use(pple drdvant, 'running'); anddrahydt, 'stoutly' (RV., once),\nmay beanother example. Thecomparative andsuperlative suffixes(above, c)\nshow alikechange ;and itisalso toberecognized inthederivatives with\nvdt(1107).\n1112. Theinstrumental isalsoveryoften used with\nadverbial value :generallyinthesingular, butsometimes also\nintheplural. Thus :\na.Ofpronominalstems :as,endandayd, kdyd, and, sdna, amd,\namuyd.\nb.Ofnoun-stems :as,ksanena, 'instantly'; afesena, 'completely'; vifesena,\n'especially'; dtva, 'byday'; distyd, 'fortunately'; sdhasd, 'suddenly'; aktubhis,\n'bynight'; and soon.\nC.Ofadjectives, both neuter(not distinguishable frommasculine) and\nfeminine :as,ddksinena,'tothesouth'; uttarena,'tothe north'; antarena,\n'within'; cirena, 'long'; fdnais and fdnakais, 'slowly'; uccafs^'onhigh';\nparacafs, 'afar'; tdvisibhis, 'mightily'; and soon.\nd.More doubtful cases, mostly from theolderlanguage, maybein-\nstanced asfollows :tirafcdta, devdta, bahtita, andsasvdrta(allRV.), hom-\nonymous instrumental fromnouns inta; dvitd, tadttna, irmd, mrsd, vftha,\nsdca, asthd(?), mudha(not V.),adhund (Br.andlater).\ne.Adverbially used instrumental are(inthe older language), oftener\nthan anyother case, distinguished from normal instrumentals bydifferences\nofform:thus, especially, byanirregular accent:as,amdanddfva(given\nabove); perhaps guha; apakd, asayd, kuhaydf?); naktayd, svapnayd, samand\nadatrayd, rtayd, ubhayd, sumnaydf?); daksind, madhyd ;nicd~, pracd, uccd,\npaced, tirafcd; inafeww-stems, byayinserted before theending,\n"
+    },
+    {
+        "page": 379,
+        "content": "11181 CASE-FORMS ASADVERBS. 363\nwhich isaccented :thus, amuyd (given above), acuyd, sadhuyd, raghuyd,\ndhrsnuya, anusthuyd, mithuyd ;andurviyd (forurvyd) andvfyvya (properly\nvfcvayd] aremore slightly irregular.\n1113. Thedative hasonly veryseldom anadverbial use.\nExamplesare (from the later language only): ciraya, 'long'; arthaya,\n'forthesake of;ahnaya, 'presently'.\n1114. Theablative isnotinfrequently usedadverbially.\nThus :\na.Ofpronominal stems:as,kdsmat, 'why?' akasmat, 'casually, un-\nexpectedly'; dt, tdt, ydt (V.: normal forms, instead ofthepronominal\nasmat etc.).\nb.Ofnoun-stems:as,asdt, 'near'; ardt, 'afar'; balat, 'forcibly'; kutu-\nhalat, 'emulously'; sakafat,'onthepart of.\nC.Oftenest,ofadjective stems :as,durdt, 'afar'; nlcdt, 'below'; pafcdt,\n'behind'; saksdt, 'plainly, actually' ;samantat, 'completely'; acirat,'notlong'.\nd.Inafewinstances, adverbially used ablatives likewise show achanged\naccent intheearly language:thus, apakdt, 'fromafar'; amdt,'from near\nby1\n;sandt,'from ofold'(but instr.sdna); uttardt, 'from thenorth'; adhardt,\n'below'.\n1115. Thegenitiveisalmost never usedadverbially.\nInthe older language occurakttis, 'by night', andvdstos, 'by day';\nlater, cirasya, 'long'.\n1116. Thelocative issometimes used with adverbial\nvalue .Thus :\nFrom noun and adjective stems:tike, 'near';areanddure, 'afar';\nabhisvare, 'behind'; astamlke,'athome'; rt\", 'without'(prep.); dgre,'in\nfront'; sthane, 'suitably'; sapadi, 'immediately'; -arthe and -krte (commonin\ncomposition),'forthesakeof\";aparisu,'inafter time'.\n1117. Even anominative form appearstobestereotypedintoanad-\nverbial value in(Vedic) fefs, interrogative particle, and itscompounds ndkis\nandmdkis, negative particles.\n1118. Verbal Prefixes andkindred words. The\nverbalprefixes,described inthepreceding chapter (1076 ff.),\nareproperly adverbs, havingaspecialoffice andmode of\nuseinconnection with verbal roots and theirmoreimme-\ndiate derivatives.\nTheir occasional looser connection with theverb hasbeen\nnoticed above(1084).Inthevalue ofgeneral adverbs, how-\never, they only very rarelyoccur(exceptasdpihasmainly\nchangeditsoffice fromprefixtoadverb orconjunctioninthe\nlaterlanguage);buttheirprepositionaluses aremuch more fre-\nquent andimportant:seebelow, 1125.\n"
+    },
+    {
+        "page": 380,
+        "content": "364 XVI. INDECLINABLES.[1118\nIncompositionwith nouns, they (like other adverbial elements) notin-\nfrequently have anadjective value: seebelow, chap. XVIII.\n1110. Several oftheprefixes (asnoticedabove, 473 4)formcom-\nparative andsuperlative adjectives, bythesuffixes taraandtama, orraand\nma :thus, uttara anduttamd;ddhara andadhamd, dpara andapamd, dvara\nandavamd, tipara andupamd ;andprathamdisdoubtless ofthesame char-\nacter; also, dntara anddntama. Andaccusatives ofsuch derivative adjectives\n(forthemostpartnototherwise found inuse)have thevalue ofcomparatives,\nand rarely superlatives,tothe prefixes themselves:thus, sdrhfitam cit\nsarhtardm sdrh fifadhi (AV.), 'whatever isquickened, dothou still further\nquicken';vitardrh v{kramasva(RV.),'stride outyetmore widely'; prd tdrh\nnaya pratardm vdsyodcha(RV.),'leadhimforward still further toward ad-\nvantage'; udenam uttardrh naya (AV.),'leadhimupstill higher'.\nBesides thoseinstanced, arefound also nitardm, avatardm, paratardm,\nparastardm. IntheBrahmanas and later(above, 11lie), thefeminine\naccusative isused instead: thus, pratitardm, sarhtardm, nitardm, uttardm,\npratardm andpratamdm (and sarhtardm, RV., once).\n1120. Kindred inorigin andcharacter with theverbal pre-\nfixes, andused likethemexceptincomposition with verbs, are\nafewother adverbs: thus, avds, 'down'; adhds, 'below'; paras,\n'faroff\"; pura, 'before'; antara(apparently, antdr-\\-a), 'among';\ndnti, 'near'; updri,'above': andsahd(already mentioned, 1104),\n'along, with', andsdca, 'together, with', maybenoticed withthem.\nVina, 'without', and visu-, 'apart', appeartoberelated with vi.\n1121. Inseparable Prefixes. Asmall number of\nadverbialprefixesarefoundonlyincombination with other\nelements. Thus :\na.Thenegative prefixaoran anbefore vowels, abefore\nconsonants.\nItiscombinedespeciallywithinnumerable nouns andadjectives ;much\nmore rarely, withadverbs,asaktitra anddpunar (RV.), dnadhas(TB.), akas-\nmat, asakrt; and, accordingtothegrammarians, sometimes alsowithpronouns\n(asas, anesas), andwith verbs(apacati,'does notcook'), butnosuchcom-\nbinations appeartobequotable.\nTheindependent negative adverbs, ndandmd, areonly inrareand\nexceptionalinstances used incomposition: seebelow, 1122b.\nb.The comitativeprefix sa,used instead ofthepreposition\nsdm, andexchangeably with sahd, before nouns andadjectives.\nc.Theprefixofdispraise dus, 'ill, badly' (identical with\nydus: 225).\nItiscombined inthesame manner asaoran.Ofcombinations with\naverbal form,atleast asingle example appearstobequotable^ dupcarati\n(B.), 'behaves ill'(BR.).\n"
+    },
+    {
+        "page": 381,
+        "content": "1122]ADVERBS. 365\nd.Thecorresponding laudatory prefix *M, 'well',isingen-\neral soclosely accordant initsusewith theprecedingthat itis\nbestmentioned here, throughitoccurs notrarelyasaninde-\npendent particleintheoldest language (inRV.\n,more thantwo\nhundred times;inthepeculiar partsofAV.\n,onlyfourteen\ntimes], andevenoccasionallylater.\nNocombination ofsuwith averbal form appears tobequotable from\nanyaccentuated text (though theworthless pada-text ofAV. xix. reads\nsu-dpdyatiat49.10). K.hasnasuvijnayeteandnavaisuviduh(orsu-\nvijnayete ?).\ne.Theexclamatory andusually depreciative prefixed forms oftheinter-\nrogative pronoun (506) aremost analogous with theinseparable prefixes.\n1122. Miscellaneous Adverbs. Other words of\nadverbial character andoffice, notclearlyreferable toany\noftheclasses hitherto treated, maybementioned asfollows:\na.Asseverativeparticles (inpart, Vediconly): thus, angd,\nhdnta, kila, khdlu,tu(rareinolderlanguage), vai, vavd(in\nBrahmana language only), hi,hind, u,aha, ha,aha,samaha, sma,\nbhala.\nOfthese, hdnta isaword ofassent;hihaswon alsoanillative meaning,\nandaccents theverb withwhich itstands inconnection (595 d);smasome-\ntimes gives apastmeaning toapresenttense (778 b);uisoften combined\nwith the finalaofotherparticles: thus, dtho, no,md, uto, upo,pro; but\nalsowith that ofverb-forms,asdatto,vidmti. The final othus produced\nispragrhyaoruncombinable (138c).\nParticles ofkindredvalue, already mentioned above, arefd,kdm orkam,\ncid,jdtu, evd.\nSome oftheasseverativeparticlesaremuch used inthelater artificial\npoetry with apurely expletive value,asdevices tohelpmake outthemetre\n(padapurana, 'verse-fillers');soespecially ha, hi,tu,sma.\nb .Negative particlesare :nd,signifying simple negation ;\nma, signifying prohibition.\nAstotheconstruction oftheverb withmd, seeabove, 579 80.\nIntheVeda, nu(ornu :248a)has alsosometimes anegative meaning.\nFortheVedic ndofcomparison,seebelow, d.\nInnahf, nd iscombined with M,both elements retaining their full\nmeaning;alsowith fdinned. 'lest'. Itisperhaps present innanti and\ncand, butnotinhind (RV., once). Ingeneral, neither ndnormd isused\nincompositiontomake negative compounds, but, instead, theinseparable\nnegative prefixaoran(1121 a):exceptions aretheVedic particles ndkis\nandmdkis, ndkim andmakim;alsonaciram andmaciram, and afewothers.\nc.Interrogative particlesareonlythose already given:kdd,\nUrn, kuvid, svid, nanu, ofwhich thelastintroduces anobjection\norexpostulation.\n"
+    },
+    {
+        "page": 382,
+        "content": "306 XVI. INDECLINABLES.[1122\nd.Ofparticlesofcomparison have been mentioned the\ntoneless iva,andyatha (alsotoneless when used inthesame\nway).Offrequentoccurrence intheoldest languageisalso nd,\nhaving (withoutloss ofaccent)thesameposition andvalue as\nthepreceding.\nExamplesare :rsidvfsa {sum ndsrjata dvfsum(RV.),'let loose your\nenmity likeanarrow attheenemy ofthesinger'; vdyo ndvrksdm(AY.),\n<asbirds tothe tree'; gaur6 ndtrsitdh piba (RV.),'drink like athirsty\nbuffalo'. This use isgenerally explainedasbeing amodification oradaptation\nofthenegative one: thus, '[although,tobesure]not[precisely]athirsty\nbuffalo'; and soon.\ne.Ofparticlesofplace, besides those already mentioned,\nmaybenoticed kva(inV.,alwaystobereadMa).\nf.Particles oftime are :nu,'now'(also nu :nundm was\nmentioned above, 1109) adydandsadyds andsadivas(RV., once),\n'today,atonce'(allheld tocontain theelement divordyu),\nhyds, 'yesterday', qvds, 'tomorrow', jyok (alsorelated withdyu),\n'long'; punar, 'again'.\ng.Ofparticlesofmanner, besides those already mentioned,\nmaybenoticed nuna, 'variously' (fornanandm,itsderivative,\nsee1109);sasvdr(RV.), 'secretly'.\nIntheabove classifications areincluded alltheVedic adverbialwords,\nandmost ofthose ofthelater language:fortherest,seethedictionaries.\nPrepositions.\n1123. Thereis,asalready stated, noproperclass of\nprepositions (inthemodern sense ofthatterm), nobodyof\nwords havingfortheir exclusive office the\"government\" of\nnouns. Butmanyoftheadverbial words indicated above\nareused with nouns inawaywhichapproximates them\ntothemorefully developed prepositionsofotherlanguages.\nIfoneandanother ofsuchwords asvina,rte occurs almostsolely\ninprepositional use,this ismerely fortuitous, andofnoconsequence.\n1124. Words areusedprepositionally along with allthe\nnoun-cases exceptingthedative. But ingeneraltheir office is\ndirectiveonly, determining moredefinitely,orstrengthening,the\npropercase-use ofthenoun. Sometimes, however, thecase-\nuse isnoteasytotrace, andthenoun then seems tobemore\nimmediately \"governed\" bytheprepositionthatis,tohave\nitscase-form morearbitrarily determined byitsassociation with\nthe latter. This isoftenest true oftheaccusative;and ofthe\n"
+    },
+    {
+        "page": 383,
+        "content": "1128]PREPOSITIONS. 367\ngenitive,which, has, here aselsewhere(294),suffered anexten-\nsion ofitsnormalsphereofuse.\n1125. Theadverbs byderivative form(1097 ff.)have least\nofaprepositionalvalue(exceptionsareespeciallyafewmade\nwith thesuffix tas :1098).\nMost oftheverbalprefixes (exceptionsareud, ni,para,\npra: andavaand niarealmostsuch)have theirprepositional\norquasi-prepositionaluses with cases;butmuch morewidelyin\ntheolder time than inthelater :inthe classicallanguagethe\nusageismainlyrestricted toprati, anu,and a.\nMost ofthedirective words akin with themore proper pre-\nfixes areusedprepositionally:some ofthem assaha, vina,\nupari, antara, purafreely,earlier and later.\nThecase-forms used adverbiallyareinmanyinstances used\nprepositionallyalso :oftenest, aswas tobeexpected,with the\ngenitive; butfrequently, andfromanearly time, with theaccus-\native;morerarely with other cases.\nWe will takeupnow thecases forabrief exposition, beginning with\nthose that areleast freely used.\n1126. TheLocative. This case isleast ofallused with words that\ncanclaim thename ofpreposition. Ofdirectives, antdr and itslater deriv-\nativeantard, meaning 'within, in',areoftenest added toit,and inthe\nclassical languageaswell asearlier. Offrequent Vedic usewith itarea\nandadhi(illustrated above, 305); aptandupa aremuch rarer:thus,\nydapdm dpivrate [sdnti] (RV.), 'who areinthedomain ofthewaters';\namur ydUpasurye [sdnti] (RV.),'who areupyonder inthesun'; sdca,\n'along with',isnot rare inRV., butalmost entirely unknown later:thus,\npitrdh sdcasail, 'staying with herparents'.\n1127. TheInstrumental. The directives used with this case are\nalmost only those which contain theassociative pronominal root sa :assaha\n(most frequent), sakam, sardham, samam, samaya, saratham- and, inthe\nVeda,theprefix sam :as,tesumatlbhih sdm pdtnlbhir ndvrsano nasimahi\n(RV.), ''maywebeunited with thy favors asmenwith their spouses'. By\nsubstitution oftheinstrumental fortheablative ofseparation (283), vina,\n'without'(notVedic), takes sometimes theinstrumental;andso,intheVeda,\navas, 'down', andparas, 'beyond', withwhich theablative isalso,andmuch\nmore normally, construed. And adhi, inRV.,isusedwith theinstrumental\nsnuna andsnubhis, where thelocative would beexpected.\n1128. TheAblative. Intheprepositional constructions oftheab-\nlative(aswaspointed outandpartlyillustratedabove, 293),theablative\nvalue ofthecase, andthemerely directive value oftheaddedparticle, are\nforthemostpart clearly tobetraced. Many oftheverbal prefixes aremore\norlessfrequently joined intheolder language with thiscase:oftenest, adhi\nandpari; moresporadically, anu, apa, ava, prati, andtheseparatives nis\nand vi.Thechange ofmeaningoftheablative witha,'hither', bywhich\n"
+    },
+    {
+        "page": 384,
+        "content": "368 XVI. INDECLINABLES.[1128\nitcomes tofillthe office ofitsopposite,theaccusative, was sufficiently ex-\nplained above (293 c).Ofdirective words akin with theprefixes, many\nasbahis, puras, avas, adhas, paras, pwra, -uma, andtzras,'outofknowledge\nof accompanythis casebyaperfectly regular construction. Also the\ncase-forms arvak, prak, pafcdt, urdhvam, purvam, param, parena, prabhrti;\nandrte, 'without', ofwhich thenatural construction with anablative is-\npredominantearlier. Antikam, 'near',issaid totake theablative aswell as\nitsmore normal companionthegenitive.\n1129. TheAccusative. Many oftheverbalprefixes andrelated words\ntake anaccompanying accusative. Most naturally (sincetheaccusative i&\nessentially the'to'-case), those thatexpressamotion oraction toward any-\nthing:asabhi, prati, arm, upa, a,atiandadhi inthesense of'over onto*\nor'across, beyond', tiras, 'through', antar andantard when meaning 'between',\npan, 'around'. Examples are:ydhpradfyo abhf sUryovicdste (AV.), 'what\nquartersthesunlooks abroad unto'; dbodhy agnfli prdty dyatim usdsam (RV.),\n'Agni hasbeen awakened tomeet theadvancing dawn'; gached kaddcit\nsvajanam prati (MBh.),'shemight gosomewhither toherownpeople'; imam\npraksydmi nrpatim prati (MBh.), 'him Iwillaskwith reference totheking';\nmama cittdm anu cittebhir e'<a(AV.),'follow aftermymind with your\nminds';eJ\nhydnah(AV.),'come hither tous'; tipanae'hyarvdn(RV.),\n'come hither untous';yddevomdrtydnati(AV.),'thegodwho isbeyond mortals';\nadhisthdya vdrcasd'dhyanydn (AV.), 'excelling above others inglory'. Also\nabhitas andparitas, which have alikevalue with thesimple abhiandpari ,-\nandupari :'above' (oftener withgenitive). Less accordant with ordinary-\naccusative constructions istheuse ofthis case withadhas, paras, puras,\nvind, beside other cases which seem more suited tothemeaning ofthose\nparticles. And thesamemay besaid ofmost oftheadverbial case-forms\nwith which theaccusative isused. Thus,anumber ofinstrumentals of\nsituation ordirection: asyeivarena \"ditydm, yepdrendJJ\nditydm (TB.),\n'those who arebelow thesun,those who arebeyond thesun'; dntarena\nytinim (QB.),'within thewomb';tehiJdam antarena sarvam(AB.),'for all\nthisuniverse isbetween them';fittarena gdrhapatyam (B.),'tothenorth of\nthehouseholder's fire'; ddksinena vedim(^)B.),tothesouth ofthe sacrificial\nhearth'; daksinena vrksavdtikdm(Qak.),'totheright oftheorchard'; nikasd,\n'near to'. Similarly, urdhvam andpurvam have anaccusative object aswell\nasanablative; and thesame istrue later ofrte.Abhimukham, 'toward',\nhas amore natural righttoconstruction with this case; andsamdyd (later\nsamayd], 'through between',isanalogous with antard and tiras.\n1130. TheGenitive. Thewords which areaccompanied bythegenitive\naremostly case-forms ofnouns, orofadjectives usedsubstantively, retaining\nenough ofthenoun-character totake this case astheir natural adjunct.\nSuch arethelocatives agre,'infront of,abhydce, 'near', arthe andkrte,\n'forthesake of,nimitte and hetdu, 'byreason of,madhye,'inthemidst\nof;andother cases,asartham andarthdya, antikam andabhimukham (which\nhave alsootherconstructions), kdrandt, sakdcdt, hetos. Andreally, although\n"
+    },
+    {
+        "page": 385,
+        "content": "1134 CONJUNCTIONS.\nless directly and obviously, ofthesame character areother adjective cases\n(someofthem showing otherconstructions, already noticed):asadharena,\nuttarena and uttarat, daksinena anddaksinat, pafcat, urdhvam, anantaram,\nsamaksam,saksat. More questionable, and illustrations rather ofthegeneral\nlooseness ofuseofthegenitive, are itsconstructions (almost wholly unknown\nintheoldest language) withmore proper words ofdirection :thus, with the\nderivative paritas, paratas, andantitas, and parastat and purastat (these\nfound intheBrahmana language:as,sarhvatsarasya parastat,'after ayear';\nsuktasya purastat,'before thehymn' [AB.]);withanti, adhas, auas, puras ;\nwith upari,'above' (common later); andwith antar.\nConjunctions.\n1131. Theconjunctions, also, asadistinct class of\nwords, arealmostwanting.\nThecombination ofclauses isinSanskrit ingeneralofa\nvery simplecharacter\n;much ofwhat inother Indo-European\nlanguagesiseffected bysubordinating conjunctionsishereman-\naged bymeans ofcompositionofwords, bytheuse ofthe\ngerunds (994),andofiti(1102 a),and soon.\n1132. The relative derivative adverbs, already given\n(1098 if.),may properlyberegardedasconjunctions;and\nafewotherparticlesofkindred value, ascedandned(1111a.\n1133.Purelyofconjunctivevalue are tfc#, 'and', and\nSfT vcij'or'(both toneless, andnever havingthe firstplace\ninasentence orclause).\nOfcopulative value, along with ca, isintheolder lan-\nguage especiallyuta(lateritbecomes aparticleofmore inde-\nfiniteuse);andapi, tatas, tatha, him ca,with otherparticles\nandcombinations ofparticles,areused often asconnectives of\nclauses.\nAdversative istu, 'but'(rareintheolderlanguage);alsor\nless strongly, u(toneless).\nOfillative value ishi,'for'(originally, andingreat partat\nevery period,asseverativeonly): compare above, 1122 a.\nToca(aswell astoitscompound ced}belongs occasionally themeaning 'ifV\nItisneedless toenter into detail with regard tothose useswhich may\nbenot lessproperly,ormoreproperly,called conjunctive than adverbial, of\ntheparticles already given, under thehead ofAdverbs.\nInterjections.\n1134. The utterances which maybeclassed asinter-\njections are, asinother languages,inpart voice-gestures,\nWhitney, Grammar. 24\n"
+    },
+    {
+        "page": 386,
+        "content": "370 XVI. INDECLINABLES.[1134\ninpart onomatopoeias,andinpartmutilations andcorrup-\ntions ofotherpartsofspeech.\n1135. a.Oftheclass ofvoice-gestures are, forexample:\na,ha,haha, ahaha, he,hai(AV.), ayi, aye, haye (RV.), aho,\nbat(RV.),bata(RV.)orvata,and(probably)hiruk andhuruk(RV.).\nb.Onomatopoeticorimitative utterances are, forexample\n(intheolderlanguage); cicca, 'whiz'(ofanarrow :RV.);kikira\n(palpitation: RV.);balandphat (phds ?)OTpMl, 'splash' (AV.);\nbhuk, 'bow-wow'(AV.); cdl, 'pat' (AV.); as,Ms, as,andhas\n(PB.);and seethewordsalready quotedincompositionwith\ntheroots krand bhu, above, 1091.\nc.Nouns andadjectives which haveassumed aninter]ectional\ncharacter are, forexample:bhos(forthevocative bhavas, 456) ;\nareorre(voc.ofari, 'enemy'); dhik,'alas!'(maybemere\nvoice-gesture, butperhapsrelated withydih); kastam, 'woe is\nme!'distya,'thank heaven!'svasti, 'hail!' susthu, sadhu, 'good,\nexcellent!' None ofthese areVedic ininterjectional use.\nCHAPTER XVII.\nDERIVATION OFDECLINABLE STEMS.\n1136. THE formation from roots ofconjugable stems\nnamely, tense-stems, mode-stems, andstems ofsecondary con-\njugation (not essentiallydifferent from oneanother, nor, itis\nbelieved, ultimately from theformation ofdeclinedstems)\nwasmostconvenientlytreated above,inthechapters devoted to\ntheverb. Likewise theformation ofadverbs byderivation(not\nessentiallydifferent fromcase-formation),inthechapter devoted\ntoparticles. And theformation ofthose declinable stems\nnamely,ofcomparison, andofinfinitives andparticiples which\nattach themselves mostcloselytothesystemsofinflection, has\nalsobeenmore orlessfully exhibited. But theextensive and\nintricatesubjectoftheformation ofthegreat bodyofdeclin-\nable stems wasbest reserved foraspecial chapter.\nOfcourse, only abrief andcompendious exhibition ofthesubject can\nbeattempted within thehere necessary limits: noexhaustive tracing outof\ntheformative elements ofevery period;stillless, acomplete statement of\nthevaried uses ofeach element; least ofall,adiscussion oforigins: but\nenough tohelp thestudent inthatanalysis ofwords which must form apart\n"
+    },
+    {
+        "page": 387,
+        "content": "1140] PRIMARY ANDSECONDARY SUFFIXES. 371\nofhislabor from theoutset, giving ageneral outline ofthefield, andpre-\nparingformore penetrating investigation.\nThe material from accentedtexts, andespecially theVedic material,\nwillbehadespeciallyinview(nothing that isVedic being intentionally left\nunconsidered); andtheexamples given willbe,sofarasispossible, words\nfound insuch texts with their accent marked. Noword notthus vouched\nforwillbeaccented unless thefact isspecifically pointed out.\n1137. The roots themselves, both verbal andpronom-\ninal, areused intheir bare form, orwithoutanyadded\nsuffix, asdeclinable stems.\nAstothisuseofverbal roots, seebelow, 1147.\nThepronominal roots, so-called, areessentially declinable;\nandhence, intheir further treatment inderivation, theyarein\naccordance with other declinable stems, andnotwith verbal\nroots.\n1138. Apart fromthis, every such stem ismade bya\nsuffix. And these suffixes fallintotwogeneralclasses:\nA.Primary suffixes, orthose which are'addeddirectly\ntoroots;\nB.Secondary suffixes, orthose which areadded tode-\nrivative stems(alsotopronominal roots, asjustpointed out,\nandsometimes toparticles).\nThe. division ofprimarysuffixes nearly correspondstothe krt(more\nregular) andunadi(less regular)suffixes oftheHindu grammarians ;the\nsecondary,totheir taddhita-snfftxes.\n1139. But thisdistinction, thoughoneofhigh value,\ntheoretically andpractically,isnotabsolute. Thus:\na.Suffixes come tohave theaspect andtheuseofprimary\nwhichreallycontain asecondary element that istosay,the\nearliest words exhibiting them weremade byaddition ofsecond-\narysuffixes towords alreadyderivative.\nSundry examples ofthiswillbepointed outbelow: thus,thegerundival\nsuffixes, tavya, anlya, etc., thesuffixes ukaand afca, tra,andothers. This\noriginisprobableformore cases than admit ofdemonstration;and itis\npossibleforothers which show nodistinct signs ofcomposition.\nb.Less often, asuffix ofprimaryusepassesover inpart\nintosecondary, throughthemedium ofusewith denominative\n\"roots\" orotherwise: examplesareyu,iman, lyasand istha.\n1140. Moreover, primarysuffixes areadded notonly\n24*\n"
+    },
+    {
+        "page": 388,
+        "content": "372 XVII. DERIVATION.[1140\ntooriginal roots, but, generallywithequal freedom, toele-\nments which havecome towear inthelanguagetheaspect\nofsuch, hybeing made thebasis ofprimary conjugation\nandeven,toacertain extent,tothebases ofsecondary\nconjugation,theconjugation-stems,andthebases oftense-\ninflection, thetense-stems.\na.Themost conspicuous examplesofthis aretheparticiples, present\nandfuture andperfect, which aremade alike from tense andconjugation-\nstems ofevery form. The infinitives (968 ff.)attach themselves only insporadic\ninstances totense-stems, andeven from conjugation-stems aremade but\nsparingly earlier; andthesame istrue ofthegerundives.\nb.General adjectives andnouns aresomewhat widely made from con-\njugation-stems, especially from thebase ofcausative conjugation:seebelow\nthesuffixesa,a,as,anz, u, ti,tr,tnu, snu.\nc.From tense-stems, theexamplesare farfewer, butnotunknown:\nthus, from present-stems,occasional derivatives ina(1148. 3b), a(1149),\nu(1178e,g,h),ta(1176e),tu(1161c),tra(1185d), ru(1192), vin(or\nin:1232, end); from stems inasapparentlyofaoristic character(besides\ninfinitives andgerundives),occasional derivatives ina(1148. 3b),ana\n(1150.2b), am(1159b), an(1160a), ana(1175),as(1151.1b),istha\n(1184), u(1178e),us(1154),tr(1182c),in(1183).\n1141. Theprimarysuffixes areadded also toroots as\ncompoundedwith theverbalprefixes.\nWhatever, namely, may have beenoriginally andstrictly\nthemode ofproductionofthederivatives withprefixes,itis\nthroughout therecorded lifeofthelanguageasiftherootand\nitsprefixorprefixesconstituted aunity, from which aderiva-\ntive isformed inthesame manner asfrom thesimple root, with,\nthat modification oftheradical meaning whichappearsalso in\ntheproperverbal forms ascompounded with thesameprefixes.\nNot derivatives ofevery kind arethusmade; but, inthe\nmain, those classes which have most oftheverbal force, or\nwhich aremost akin invalue with infinitives andparticiples.\nTheoccurrence ofsuch derivatives withprefixes, and their accent,will\nbenoted under each suffix below. They arechiefly (innearly theorder of\ntheir comparative frequency),besides root-stems, those ina,inana, into',\nintarandtra,andinin,ya,vanandman,iandu,as,andafewothers.\n1142. The suffixes ofboth classes aresometimes joined\ntotheirprimitives byapreceding union-vowel that isto\nsay,byonewhich wears thataspect, and, inouruncertainty as\ntoitsrealorigin, maymost conveniently andsafely becalled\nbythatname. The linebetween these vowels andsuch asde-\n"
+    },
+    {
+        "page": 389,
+        "content": "1144]PRIMARY DERIVATIVES. 373\nserve toberanked asoforganicsuffixal character cannot be\nsharply drawn.\nEach ofthetwogreat classes willnow betaken upby\nitself, formoreparticular consideration.\nA.PrimaryDerivatives.\n1143.Form ofroot. Theform ofroot towhich a\nprimarysuffix isadded isliable tomore orless variation.\nThus:\na.Byfarthemost frequentisastrengthening change, by\ngunaorvnfr/^z-increment. The former may occur under all\ncircumstances(except,ofcourse, where ywmz-changeisingeneral\nforbidden: 235.240): thus, vedafromyvid,modafromymud,\nvdrdha from]/vrdh :ayanafromyi,sdvana fromysu,sdrana\nfromysr;and soon. But the latter isonly allowed under\nsuch circumstances asleavelong aastheresulting vowel :that\nis,with non-final #,andwith afinal iorw-vowel andr,before\navowel: thus, nadd fromynad, grabhd fromygrbhorgrabh,\nvahdfromyvah, nayafromym,bhavd fromybhu,hardfrom\nykr;such strengtheningaswould make vaida andmauda is\nentirely unknown inprimaryderivation.\nNogeneral rules canbegiven astotheoccurrence ofstrengtheningin\nderivation :itmust beleft tobepointed outforeach suffix.\nb.Other vowel changes aremore sporadic, and willbenoticed indetail\nbelow: thus, occasional abbreviation ofaroot, asinuktifromyvac ;change\noffinal rtoirandur;theloss offinala,oritsconversion toani-vowel;\nand soon.\nc.Thereversion ofafinalpalatalorhtoaguttural hasbeen already\nnoticed(216). Afinalnormisoccasionally lost,asinformations already\nconsidered.\nd.After ashort final vowel issometimes added at:namely, where a\nroot isused asstem without suffix (1147), and before afollowing yorv\nofvan(1169),vara andvan(1171), yuonce(1165 a),andya(1213).\nThepresenceoftbefore these suffixes indicates anoriginal secondary\nderivation from derivatives in tiand tu.\ne.The root issometimesreduplicated:rarelyintheusewithout suffix :\noftenest beforez,a,u;inonly afewcases each, alsobefore other suffixes\n(ana, vana, vanandvarl, vani, vi, vit, ani, in,aw, tnw, ia, ti,tr,tra,\nukaandIfca,ma, ya}.\n1144. Accent. Nogenerallaws governingtheplaceof\nthe accent are toberecognized;each suffix must inthis\nrespect beconsidered byitself.\nInconnection with avery few suffixes istoberecognized acertain\n"
+    },
+    {
+        "page": 390,
+        "content": "374 XVII. PKIMARY DERIVATION.[1144\ndegree oftendency toaccent theroot incase ofanomen actionis orinfini-\ntivalderivative, andtheending inthecase ofanomen agentis orparticipial\nderivative: seethesuffixesa,ana, as,an,andman, below, where theex-\namplesareconsidered. Differences ofaccent inwords made bythesame\nsuffix are also occasionally connected with differences ofgender:seethe\nsuffixes asandman.\n1145. Meaning. Asregardstheirsignification, thepri-\nmaryderivatives fallinto twogreat classes, theoneindicating\ntheaction expressed bytheverbalroot, theother theperson or\nthinginwhich theactionappears,theagentoractor the\nlatter, eithersubstantivelyoradjectively. Theone class isab-\nstract, infinitival; theother isconcrete, participial.Allother\nmeanings maybeviewed asmodifications orspecializationsof\nthese two.\nEven thewords indicating recipience ofaction, thepassive participles,\nare, astheir use also asneuter orreflexive shows, only notably modified\nwords ofagency. Thegerundives are, aswaspointed outabove (961),sec-\nondary derivatives, originally indicating only 'concerned with the action'.\n1146. Butthese twoclasses, intheprocessesofformation,\narenotheldsharply apart. There ishardlyasuffix bywhich\naction-nouns areformed which does notalsomake agent-nouns\noradjectives; althoughthere arenotafewbywhich aremade\nonlythe latter. Intreating them indetail below, wewill first\ntakeupthe suffixes bywhich derivatives ofboth classes are\nmade, andthen thoseforming only agent-nouns.\nTofollow analphabetical arrangement, although insome respects more\npractically convenient, seems tooartificial, and destructive ofallnatural\nconnection. TheIndex will givewhat helpisneeded toward finding any\nparticular suffix which issought.\n1147.Stems without suffix: Root-words. These\nwords and their useshave beenalready pretty fullyconsid-\neredabove(323, 348ff.,383 ff..400, 401).\na.Theyareusedespecially (inthelater language,almost\nsolely)asfinals ofcompounds, andhave both fundamental val-\nues, asaction-nouns(frequentlyasinfinitives :971),and as\nagent-nouns andadjectives (often governing anaccusative :27le).\nAsaction-nouns, theyarechiefly feminines (384:inmanyin-\nstances, however, theydonotoccur insituations thatdetermine\nthegender).\nb.Inasmall number ofwords, mostlyofrare occurrence,\nthereduplicatedroot isused without suffix.\nTheVedic cases are: with simple reduplication, sasydd, dfe/t,sasrut\n(irregular, forsusrut], didyd anddidytit, juhu, andperhaps pffM; with in-\n"
+    },
+    {
+        "page": 391,
+        "content": "1148]STEMS INa. 375\ntensive reduplication, -nenf, jogU, yamytidh, vdnlvan (withtheintensive\ninstead oftheusual radicalaccent). Inddridra isseen atransfer tothe\na-declension. AsusQ isprobablytobeunderstood asacompound,asu-sfi.\nc.Iftherootend inashort vowel, atisregularlyand\nusually added(383 b).\nExamples, andtheexceptions found tooccur, have been givenatthe\nplace just quoted.\nd.Words ofthisform incombination with verbal prefixes\nareverynumerous. Theaccent rests(asincombination ofthe\nsame with other preceding elements)ontheroot-stem.\ne.Inasingle instance, crtitkarna (RV.),'oflistening ears',astem of\nthis class occurs aspriormember ofacompound.\n1148. 51a.With thesuffix^aismade avery large\nandheterogeneous bodyofderivatives,ofvarious meaning\nandshowingvarious treatment oftheroot :^^-strengthen-\ning,vrddM-strengthening,retention unchanged, andredu-\nplication.\nIngood part, theyareclassifiable under thetwousual gen-\neralheads;butinpart theyhavebeen individualized intomore\nspecialsenses.\n1.With^wmj-strengtheningoftheroot(wherethat ispos-\nsible :235, 240).These arethegreat majority, being more\nthan twice asnumerous asallothers together.\na.Many nomina actionis :as,prdma, 'weariness', grdha, 'seizure', aya,\n'movement', veda, 'knowledge', hdva, 'cair, krodha, 'wrath', jo'sa, 'enjoyment*,\ntdra, 'crossing', sdrga, 'emission'.\nb.Many nomina agentis:as,ksamd, 'patient', svajd, 'constrictor', hinvd,\n'impelling', jivd, 'living', meghd, 'cloud', coda, 'inciting', plavd, 'boat', sard,\n'brook', sarpd, 'serpent', area, 'radiant'.\nc.Oftheexamples here given, those under aaccent theradical syllable,\nandthose underbtheending. And this isinamajority ofcases the fact\nasregards thetwoclasses ofderivatives;sothat, taken inconnection with\nkindred facts astoothersuffixes,itindicates such a.difference ofaccent as\nageneral tendency ofthelanguage. Afewsporadic instances aremetwith\nofthesame form having theoneortheother value accordingtoitsaccent:\nthus, esa, 'haste', esd, 'hasting'; fdsa, 'order', fasd, 'orderef (other examples\narefeama, fafea, pofea): compareasimilar difference with other derivatives\nas,ana, an,man;. But exceptions arenumerousthus,forexample,\njayd, javd, smara, action-nouns; frdva, mdgha, Mnsa, agent-nouns and\nthesubjectcalls foramuch wider anddeeper investigation than ithasyet\nreceived, before theaccentuation referred tocanbesetupasalawofthe\nlanguage inderivation.\n2.With ^^'-strengtheningoftheroot butonlywhere\n"
+    },
+    {
+        "page": 392,
+        "content": "376 XVII. PRIMARY DERIVATION.[1148\naistheresultingradical vowel :thatis,ofmediala,andof\nfinal r(most often),uoru,ior i(rare'.\na.Examples ofaction-nouns are:kama, 'love', bhagd, 'share', nada,\n'noise', davd, 'fire', tard, 'crossing'. Hardly anyforms ofclear derivatior\nandmeaning arequotable with accent ontheroot-syllable.\nb.Examples ofagent-nouns are:grabhd, 'seizing', vahd, 'carrying',\nnayd, 'leading', jard,'lover1\n.\n3.With unstrengthened root, theexamples arequite few:thus, fcrpd,\n'lean', yuga, 'yoke', sruvd, 'spoon', priyd, 'dear', vrd, 'troop', cued, 'bright'.\na.Anumber ofwords ofthisclass, especially asoccurring incom-\nposition,areprobably results ofthetransfer ofroot-stems tothea-declension :\nthus, -raja, -ghusa, -sphura.\nb.AfewVedic examples aremade from conjugational stems :thus,\nfrom the causative stem, kupaya andtdnaya andmrgraj/a(?), -Inkhaya,\n-ejaya, -dhdrayd, -parayd; from present-stems ofthenit-class(IV.), orsec-\nondary roots made fromthem(716), hinvd, -mud, -jinvd, -pinvd, -sinvd,\n-sunvd-j from others, -prnd, -mrnd, -pacyd, -mam/a, -dasya, -jurya, -ksudhya,\n-trsya, -jighrd, -piba; fromaorist-stems(?), yesd, nesa(innesatama, RV., once).\n4.Withreduplicated root,thederivatives intheolder language area\nclass ofsome extent. They aremade inafew instances with thesimple\nreduplication: thus, cacard, 'movable'(?),cicayd, 'inciting', sasrd, 'flowing',\nuaura, 'hiding'; much more often, with theintensivereduplication:thus,\nreriM, 'licking', vevijd, 'hasting', -roruda, 'weeping', -canfcrama, 'striding',\ncaracard and calacald, 'movable', kanikradd, 'noisy', panispadd, 'quivering',\nvarlvrtd, 'rolling'.\nOnly asingle exampleofanaction-noun hasbeen noticed, namely\nfipndtha,'attack', with accent onthe root. Other examples ofaccent else-\nwhere thanonthesuffix areonly thestems(ofdoubtful meaning orderivation)\ncarcaraandgargara.\n5.Derivatives with this suffix from roots ascompounded\nwith theverbalprefixesarequite common, inallthemodes of\nformation(ineach,inproportiontothefrequencyofindepend-\nentwords):constituting,infact, considerably thelargest body\nofderivative stems withprefixes. Theyareofboth classes as\ntomeaning. The accentis,with fewexceptions, ontheend-\ning and that, witlftmt anyreference tothevalue ofthestem\nasaction-noun oragent-noun.\nExamplesare :sarhgamd, 'assembly', nimesd, 'wink', abhidrohd, 'enmity',\nanukard, 'assistance', wdand, 'inspiration', pratyacravd, 'response'; paricara,\n'wandering', sarhjayd, 'victorious', vibodhd, 'wakeful', atiyajd, 'over-pious1\n,\nudard, 'inciting, elevated', uttudd, 'rousing', samgird, 'swallowing', adardird,\n'crushing', adhicankramd, 'climbing'.\nThe only definable class ofexceptionsinregard toaccent appearsto\nbethat oftheadverbial gerundsinam(above, 995). which areaccented\nonthe root-syllable. Avery few other stems have thesame tone: forex-\n"
+    },
+    {
+        "page": 393,
+        "content": "1150] STEMS INa,a.ana. 377\nample, utpata, 'portent'; apresa, 'plague'. Afewothers, mostly agent-nouns,\nhave theaccent ontheprefix:forexample, vybsa (i.e.vf-osa), 'burning',\nprdtivefa, 'neighbor', dbhaga, 'sharing'; but alsosdrhkufa, 'appearance'.\nFortheremaining compounds ofthesederivatives, with theinseparable\nprefixes andwith other elements, seethenext chapter.Itmaybemerely-\nmentioned here thatsuchcompounds arenumerous, andthat thea-derivative\nhasoften anactiveparticipial value, and isfrequently preceded byacase-\nform, oftenest theaccusative.\nMany words inthelanguage appeartoendwith asuffix a,while yet\nthey arereferable tonorootwhich canbeotherwise demonstrated assuch.\n1149. 5TTa.This suffix has. forthemostpart,the\nsubordinate office ofmaking correspondingfeminines(espe-\ncially adjective]tomasculines in%a inwhich use, of\ncourse, itmight beopentoquestion whether weshould\nnotregarditrather asanelement ofsecondarycharacter.\nItiscomparativelylittleemployed independently.\nThere are, however, aconsiderable number offeminine\nnouns inainthelanguagetowhich nocorresponding mascu-\nlines exist\n;andmanyofthem aredistinctlytraceable toroots.\nThe latter areespecially action-nouns, accented onthe suffix :\nandsuch derivatives inuform adistinctclass, being made some-\ntimes fromsimple roots, butespecially, and atwill, from\nconjugational stems, desiderative, causative, anddenominative.\nExamples from simple roots are :ifa, 'lordship', Ifcndd, 'play', jara,\n'oldage:\n,nindd, 'reproach'.\nExamples from desiderative stems arevery fewintheoldest language,\nbutcommon later: thus, jigisd andbhiksd (RV.); thelatter and virtsd (AV.);\nblbhatsd (VS.);inBrahmana language, pipasa, bubhuksa, mimahsa, $iksa or\nCiksd,etc.\nFrom causative stems,theonlyVedic exampleisgamayd (comparethe\nderivatives in-at/a, above, 1148. 3b);buttheformation becomes common\nlater,inthemaking-upoftheperiphrastic perfect (above, 1070ff.). The\nderivatives from present-stems thus, vidd, edhd, juhavd,etc. made\nforasimilar office, have been mentioned inconnection with this perfect\n(1073).\nFrom denominative stems,theolder instances aremore numerous :for\nexample, afvayd, sukratuyd, apasyd, urusyd, asuyd, afanayd, jwanasya.\nTheir yisnotresolved into i.\n1150.Rana.With this suffix(aswith 5[a)areform-\nedmanyderivatives, ofboth theprincipalclasses ofmean-\ning,andwith notinfrequent specializations.The roothas\n"
+    },
+    {
+        "page": 394,
+        "content": "378 XVII. PRIMARY DERIVATION.[1150\nusually <7^#-strengthening,butsometimes vrddhi instead;\nandinafewcases itremainsunstrengthened.Derivatives\nofthisformation arefrequent from roots withprefixes, and\nalso incompositionwith other elements.\nThenormal andgreatly prevalentaccent isupontheroot-\nsyllable, without regardtothedifference ofmeaning:butcases\noccur ofaccented final, andafewofaccentedpenult (whichlast\narepossiblytobereferred toanother suffix ofthesame form\nbutofadifferentorigin).\nTheaction-nouns areingeneraloftheneuter gender. The\nfeminine ofadjectivesismade either inaorin 1(for details,\nseebelow). And afewfeminine action-nouns inanaandam\noccur, which aretoberanked asbelongingtothis suffix.\n1.With strengthened and accented root-syllable. Under\nthisheadfall, asaboveindicated,thegreat mass offorms.\na.With^Una-strengthening:examples ofaction-nouns aresddana, 'seat',\nrdksana, 'protection', ddna, 'giving', cdyana, 'collection', vedana, 'property',\nhdvana, 'call', bhtijana, 'enjoyment', kdrana, 'deed', vdrdhana, 'increase';\nofagent-nouns, tdpana, 'burning', cetana, 'visible', co'dana, 'impelling'.\nThe derivatives dUsana and-mdrjana have thesame irregular strength-\nening whichappearsintheir present-systems (627, 1042 a),and other\nformations from thesame roots.\nb.With weZd/u'-strengthening (onlyinsuch circumstances thataremains\nasvowel oftheradical syllable): examples (allthat have been noted inthe\nolder language) are-cdtana, ndfana, mddana, -vdcana, -vdsana, -vdhana, sdd-\nana, -spd?ana, svddana, -dyana, -ydvana, -srdvana, -pdrana, vdrana.\nc.From roots withprefixes, the derivatives ofthisformation arevery\nnumerous, being exceeded infrequency onlybythose made with thesuffix\na(above, 1148.5).' Afewexamplesare: akrdmana, 'striding on',udydna,\n'upgoirig', nidhdna, 'receptacle', prdnana, 'expiration', vimocana, 'release' and\n'releasing', samgrdmana, 'assembly' and'assembler', adhivikdrtana, 'cutting\noff',avaprabhrdh$ana, 'falling away down'.\nForother compounds ofthese derivatives, showing thesame accent (and\nthesame femininestem), seethenext chapter (below, 1272).\nAfewexceptions occur: vicaksand, upariyayand, and thefeminines\npramandani andnirdahani.\nd.The adjectives ofthis formation, simpleorcompound, make their\nfeminine usually in1:thus, cddanl, pfyani, spdranl, jdmbham ;prajndnl,\nprtiksani, sarhgrdham, abhisdvanl, vidhdrani(cetariiisofdoubtful meaning:\nbelow, 2a).\nAn adjective compound, however, having anoun inana asfinal\nmember, makes itsfeminine ina:thus, supasarpand,'ofeasy approach',\nsddvidhana, 'ofsextuple order', anapavacand,'not tobeordered away'.\n2.Themore irregular formations maybeclassed asfollows :\n"
+    },
+    {
+        "page": 395,
+        "content": "1151] STEMS INana, as. 379\na.With accent onthe final: asmall number ofagent-nouns and ad-\njectives,askarand, 'active'(against kdrana, {'act'), tvarand, 'hasting',\nrocand, 'shining', fcrofemd, 'yelling', svapand, 'sleepy', ksayand, 'habitable'.\nThese, unlike themore regular class, make their feminines ina:thus,\ntvarana, spandana,etc.And averyfewfeminine action-nouns have thesame\nform: jarand, dyotand (?RV. i.123.4),fvetand (andcompare kapand, rafand).\nBeside thesemaybementioned afewfeminines inani,ofmore orless\ndoubtful character:arsant,cetarii(tocetana, taparil (totdpana), vrjaril (with\nvrjdna), hdyani (with hdyand), rajanl,tedaril.\nb.With accent onthepenult:avery small number ofadjectives:as\ndohdna, 'milking', mandna, 'considerate', manddna, 'rejoicing', saksdna,\n'overcoming', andperhaps vaksdnd, 'carrying' (withaoristics);and astill\nsmaller number ofneuter action-nouns: dansdna, 'great deed', vrjdna,'en-\nclosure, town', krpdna, 'misery' (against krpand, 'miserable'), with the\nmasculinekirdna,'dust'.\nThe only noticed example ofafeminine isina:turdna. And afew\nfeminine nouns have thesame form :asjardnd, bhanddnd, kanydna,\nvaksdna, etc.(andcompare theanomalous masc. name ufdna: 355a).\nc.Without strengthening ofthe root aremade avery smallnumber of\nderivatives :namely (besides those already noted, krpdna andkrpand, vrjdna\nandvrjaril, kirdna, turdna), bhurana, -stivana andbhuvana (compare the\nperfectsofthesame roots, sasuva andbabhuva, without theusual strength-\nening:789 a):kfyana, pffana, vrsana aredoubtful.\nd.Afewisolated casesmaybeherementioned :jagarand andpdlpulana\nfrom reduplicated root-forms;sdmana apparently from the prefix sam;\nanhurand apparently fromahhurd; yosana, 'woman' (beside ydsan, y6sa, etc.),\nandpftana, 'fight',aretheonlyfeminines with accented root-syllable.\n1151.EfH^s. Bythis suffix aremade(usuallywith\n<7W^a-strengtheningoftheroot-vowel) especiallyalargeclass\nofneuter nouns, mostlyabstract(action-nouns),butsome-\ntimes assumingaconcrete value; andalso,intheolder\nlanguage,afewagent-nouns andadjectives, andaconsid-\nerable number ofinfinitives.\nThe accent inwords ofthe first class isontheroot, and\ninthesecond ontheending ;and inafewinstances words of\nthetwo classes havingthesame form aredistinguished bytheir\naccent; theinfinitives have forthemostparttheaccent onthe\nsuffix .\n1.a.Examplesofthe firstandprincipalclass are:avas,\n'aid, favor', tdpas, 'warmth, prdyas, 'pleasure', tejas, 'splendor',\ngrdvas, 'fame', ddhas, 'milking', karas, 'deed', prathas, 'breadth',\ncetasandmdnas, 'mind', cdkms, 'eye', sdras, 'pond', vdcas, 'speech'.\nb.Afewwords ofthis class are ofirregularformation :thus, without\n"
+    },
+    {
+        "page": 396,
+        "content": "380 XVII. PRIMARY DERIVATION.[1151\nstrengthening oftheroot, juvas, 'quickness' (beside jdvas], uras, 'breast\\\nmfdhas, 'contempt'; andpfras, 'head',istobecompared; with vrddhi-\nstrengthening, -uacas, vdsas, vdhas; -svadas, -hayas; perhaps with an\naoristics,hesas,'missile'.\nc.After final aofaroot isusuallyinserted ybefore the\nsuffix(258):thus, dhhyas, ghyas.\nButthere areinthe oldest language apparent remains ofaformation\ninwhich aswasadded directly toradical a:thus, bhasand-das(oftento\nbepronounced astwosyllables), jrias, mas; and -jasand-dhas and -das,\nfrom theroots jaanddhaandda(Benfey, Abh. Gott. Ges.xxiii., 1878).\n2.a.The instances inwhich anagent-nounisdifferentiated byits\naccent from anaction-noun are :dpas, 'work', and apds, 'active'; ydfas,\n'beauty', andj/apds, 'beauteous'; fdras, 'quickness', and tards(SV., once),\n'quick'; dtivas, 'worship', and duvds, 'lively' (?);mdhas, 'greatness', and\nmahds, 'great'; between rdksas, n.,and raksds, m.,bothmeaning 'demon',\nandbetweentydjas, n.,'abandonment'(?),and tyajds, m.,'descendant'(?),\ntheantithesis ismuch less clear.\nb.Adjectives inaswithout correspondingabstracts are:tavds, 'strong',\nvedhds, 'pious', probably ahands, 'heady'; and afewother words ofisolated\noccurence,asuepas, dhvards. From adenominative stem ismademrgayds,\n'wild animal' (RV. }once).\nc.Butthere are also avery few cases ofabstract nouns, notneuter,\naccented ontheending: thus, jards,'oldage', bhiyds, 'fear'; anddoubtless\nalsohavds, 'call', and tvesds. 'impulse'. The feminineusds, 'dawn', might\nbelong either here orunder the lastpreceding head.\nd.Apparently containing asuffix asarethenoun upas, 'lap', andcertain\nproper names :dngiras, nodhds, bhalands, arcanands. Thefeminineapsards,\n'nymph',isofdoubtful derivation.\nThe irregular formation ofsome ofthewords ofthis division willbe\nnoticed, without special remark.\n3.The infinitives made bythesuffix ashavebeenexplain-\nedabove(973):theyshow various treatment oftheroot, and\nvarious accent(whichlastmay perhaps mark adifference of\ngender,like thatbetween saJias andjards}.\n4.Theformation ofderivatives inasfrom rootscompounded with pre-\nfixes isvery restrictedif,indeed,itistobeadmitted atall.Noinfini-\ntive inasoccurs with aprefix; norany action-noun; and theadjective\ncombinations areinsome instances evidently, andinmost othersapparently,\npossessive compounds ofthenoun with theprefix used adjectively: themost\nprobable exceptionsare-nybkas and wfspardhas. Asintheseexamples,the\naccent isalways ontheprefix.\nInconnection withthis, themostcommon andimportant\nsuffix endingins,maybebest treated theothers, kindred in\noffice andpossiblyalso inorigin, which end inthesame sib-\nilant.\n"
+    },
+    {
+        "page": 397,
+        "content": "1155 STEMS INas,tasETC.,is,us, 381\n1152. cTR tas,^Hnas,W$sas.With these suffixes are\nmade anextremelysmall number ofaction-nouns. Thus :\na.With tasaremaderetas, 'seed', andsrotas,'stream'.\nb.With nas aremade dpnas, 'acquisition', amas, 'wave', -bharnas,\n'offering', reknas, 'riches'; and indrdvinas, 'wealth', andpdrlnas, 'fulness',\nisapparently tobeseen thesamesuffix, with prefixed elements having the\npresent value ofunion-vowels. Probably thesame istrue ofddmunas,\n'house-friend'.\nC.With sas isperhaps made vdpsas, 'beauty' (?);and tdrusas may be\nmentioned with it(rather tarus-a?}.\n1153.^FTis.With thesuffix isisformed asmallnum-\nber(aboutadozen)ofnouns.\nTheyareinpartnouns ofaction, butmost areused con-\ncretely. The radicalsyllablehasthe$wm-strengthening, and\ntheaccent isonthesuffix(exceptinjyotis, 'light', vyathis (?),\nand amis, 'rawmeat'). Examplesare :arcis, rods, and cods,\n'light', hams, 'oblation', vartis, /track'.\nTheforms tuvis- andsurabhis- appear inorganically fortuviand surablii\ninafewcompoundsorderivatives.\n1154.3Hus.With this suffix aremade afewwords,K\nofvariousmeaning, root-form, and accent.\nThey arewords signifying both action andagent. Afewhave both\nmeanings, without difference ofaccent:thus, tapus,'heat' and'hot'; drus,\n'wound' and'sore'; cdksus, 'brightness' and'seeing, eye'; vdpus, 'wonderful'\nand 'wonder'. Thenouns aremostly neuter,and accented ontheroot-\nsyllable; exceptions are: inregard toaccent, jantis, 'birth'; inregard to\ngender, mdnus, 'man', andnahus, proper name. Ofadjectives, areaccented\nontheending jaytis, vanus,vidus (whichalone shows anunstrengthened\nroot), and daksus, 'burning' (which appearstoattach itself totheaorist-\nstem).\n1155.^*.With this suffix areformed aconsiderable\nbodyofderivatives,ofallgenders: adjectivesandmascu-\nlineagent-nouns,feminineabstracts, and afew neuters.\nThey show avarious form oftheroot:strong, weak, and\nreduplicated. Their accent isalso various.\nMany words inihavemeanings muchspecialized;andmany\n(including most oftheneuters)arehardlytobeconnected with\nanyroot elsewhere demonstrable.\n1.Thefeminine action-nouns areofvery various form: thus, with\nweak root-form, ruci, 'brightness', tvtii, 'sheen'; krst, 'ploughing', rtrtf,\n"
+    },
+    {
+        "page": 398,
+        "content": "382 XVII. PRIMARY DERIVATION.[1155\n'dance', yudhf, 'fighting; with ptma-strengthening (where possible), ropi,\n'pain', rdndhi, 'subjection', poof,'heat:\n,vanfandsanf, 'gain'; with vrddhi-\nstrengthening, grdhi, 'seizure', dkraji, 'course', ajf, 'race'; fromydus comes\ndusi(compare dusayati, 1042 a).Thevariety ofaccent, which seems reducible\ntonorule,isillustrated bytheexamples given.\nThefew infinitively used words ofthis formation(above, 975) have a\nweak root-form, with accent ontheending.\n2.Theadjectives andmasculine agent-nouns exhibit thesame variety. Thus :\na.With unstrengthened root: fuci, 'bright', bhrmi, 'lively' (ybhram],\ngrbhi, 'container'.\nb.With strengthened root (orroot incapable ofgruna-change):dri,\n'enemy', mdhi, 'great', arcf, 'beam', praniK, 'knot', kridi, 'playing'; and\nsome words, ofobscure connections, show anapparent vrdd/u-increment :\nthus, drapf, 'mantle', raft' 'heap', pani, 'hand', etc.\nc.Withreduplicatedroot. This isintheolder language aconsiderable\nclass, ofquite various form. Thus: withweak orabbreviatedroot, cdfen,\njdghri (yghar), pdpri, sdsri, -mamrz, babhrf, vawrf, jdgmi, jdghni, sdsni,\nsusvi, -fifvi; and,with displacementoffinal a(oritsweakening tothe\nsemblance ofthesuffix), dadf, papf, yayf (withacase ortwofromyayi],\n-jey'fii, dddhi;from thewr-form ofroots inchangeable r,jdguri, tdturi,\npdpuri; with simple reduplication, yuyudhi, vfvici; with strengthened\nreduplication, -cacali, tdtrpi, dddhrsi, vdvahi, sasaM, tutuji andtutujt, yUyuvi,\nyuyudhi; andjarbhdri. And karkart.'lute', anddundubhi, 'drum', have the\naspect ofbelongingtothesameclass, butareprobably onomatopoetic. The\naccent,itwillbenoticed,ismost often onthereduplication,butnotseldom\nelsewhere(only once ontheroot).\nItwas noticed above (271 f)that thesereduplicatedderivatives in i\nespeciallyoften takeanobject intheaccusative, like apresent participle.\nd.Formations in ifrom therootcompounded with prefixes arenot at\nallnumerous. They areaccented usually onthe suffix. Examples are:\nabhidrohf, ayaji, vyanavt, nijaghnf, paradadf, visasaht; butalsoajdni, amtiri,\nvfvavri.\nAscompounded with other preceding words, theadjectives oragent-\nnouns in iarenotrare, andareregularly accented ontheroot :seethenext\nchapter, 1276.\ne.From ydha comes aderivative-dhi, forming many masculine com-\npounds,with thevalue both ofanabstract and aconcrete: thus, with pre-\nfixes, antardhf, uddhi, nidhi, paridhf, etc. Opinionsaredivided astowhether\nitistoberegardedasformed with thesuffixi,displacing theradicala,or\nwithweakening ofato i.\n3.Neuter nouns in iarefew,and ofobscure derivation: examples are\nafcsi, 'eye', dsthi, 'bone', dddhi, 'curds',etc.\n1156. i.The suffixt,like ETT (above, 1149), has\ningeneralthe office ofmakingafeminine from amasculine\nstem. Thus :\n"
+    },
+    {
+        "page": 399,
+        "content": "1157]STEMS IN\n',iti. 383\nEspecially, from consonant-stems ingeneral: from stems inr(or\nar):376a;from some stems inu:344; frommany stems in\n(for details, seethevarious suffixes ending ina): stems in ivery rarely\nhave corresponding feminines in I:anexampleiskrmi(masc. kfmi),\nIntheoldest language,afewmasculines also aremade infinal I:they\nhave been noticed, and their inflectiondescribed, above (355 ff.).\n1157. fH ti.This suffix forms alargeclass offre-\nquentlyused feminine nouns ofaction; and also afew\nagent-nouns (masculine) andadjectives. The root has in\ngeneralthesame form asbefore thesuffix rTto,ofthepass-\niveparticiple (952 ff.)that istosay,aweak, andoften\naweakened orabbreviated, form.\nThe accent ought,inanalogy with that oftheparticiple,\ntorestalways uponthesuffix;and itdoes sointhe(small)\nmajorityofwords quotable from accented texts;buttheex-\nceptionsarenumerous. Inthefewwords used asinfinitives\n(975),theaccent isonthesuffixonly.\n1.a.Examplesofthenormal formation are:rait, 'gift', itf\n:'progress',\nritf, 'flow', stuff, 'praise', bhutf, 'prosperity', bhrtf, 'bearing', pezfctf, 'power',\nkirtf, 'fame', purtf, 'bestowal', maff, 'thought', pitf, 'drink'(|/pa; pple pita],\ndhautf,'stream' (ydhav;pple dhauta); and, with accentedroot, gdti,\n'motion', fdnti, 'repose', dfti,'division'(yda; pple ditd], trpti, 'satisfaction',\ntsti, 'offering' (yyaj:pple zsfd), -ukti, 'speech' (yvac:pple wfefd), vfddhi,\n'increase'.\nb.The roots which form theirparticipleinita(956) donothave tho\nialso before ti:thus, only gupt i,-drpti (butAV. hasoncetiditi, beside\nuditd, fromj/uad). Afewroots having theirparticipleinnainstead ofta\n(957), form theabstract noun alsoinni(below, 1158). Andfrom theroots\ntanandran occur tdnti andrdnfi, beside themore regulartatiand rati;\nalso dhanti(once, VS.) beside dhati. From roots having theform da,the\nderivative incompositionissometimes -tti(forddti with loss ofradical\nvowel: compare theparticiple-form -tfa, above, 955c): thus, niravatti(K.),\nvdsutti, bhdyatti, maghdtti (allRV.).\nc.Afewderivatives aremade fromreduplicated roots;their accent is\nvarious :thus, carfcrff,didhiti and -diditi.jfyarti, andperhapstheproper\nname yaydti.\nd.Derivatives from roots withprefixesarenumerous, andhave(asin\nthecase oftheparticiplesinfa,and theaction-nouns intu)theaccent on\ntheprefix:examplesarednumati, abhlti, dhuti, nfrrti, vykpti, sdmgati. The\nonly exceptions noticed aredsaktt and asuti.\nInother combinations than withprefixes,theaccentuation isingeneral\nthesame :seethenext chapter.\n2.The adjectives andagent-nouns which, asmasculines,aretobe\n"
+    },
+    {
+        "page": 400,
+        "content": "384 XVII. PRIMARY DERIVATION.[1157\nconnected with these instead of\\\\ith thefeminine abstracts areveryfew :\nthus, pUti, 'putrid', vdsti, 'eager', dhuti, 'shaker', jnatf, 'relative', pattf,\n'footman'; andafewothers,ofmore orlessdubious character. Theaccent\nisvarious,asintheother class.\n3.Afewwords show thesame suffix tipreceded byavowel,inwhich\nnoorganiccharacter seemsrecognizable, andwhich may therefore passfora\n\"union-vowel\". One, uditi,hasbeen quoted above\n;another with iissaniti\n(RV., once); and snihitl andsnehili, notwithstanding their long final,maybe\nmentioned with them. With atiaremade afew, variously accented :thus,\ntheaction-nounsanhatf, drfatf,vasatf'nest', ramdti, vratdti, amdti and\ndmati, -dhrajati; andtheagent-words aratf, khalatf, vrkdti, rdmati. With\nIti, rjiti and ddbhiti. IntheBrahmanas appearoccasional derivatives from\nconjugational stems,likejdnayati (TS.) andagamayiti (K.xxviii.6).The\nfeminineyuvati, 'young (adj.), maiden',isofisolated character.\n4.Insome ofthewords instanced inthe lastparagraph,tiisperhaps\nappliedasasecondary suffix. Akindred character belongs toitinthe\nnumeral derivatives frompronominal roots, kdti, tdti, ydti, andfromnumerals,\nasvihcatf, sastf, etc.; with panktf (from pdfica);andinaddhatf, from the\nparticle addha.\n1158. ftni.This suffixagreesingeneralinitsuses\nandintheform ofitsderivatives with thepreceding;but\nitmakes amuch smaller number ofwords, among which\nthefeminine abstracts areaminority.\n1.Aswas noticed above (1157. Ib),afewverbs (endinginvowels)\nmaking their passive participleinnainstead oftamake their action-noun\ninniinstead ofti.From theVeda arequotable only -jyani, 'injury', and\njurnf, 'heat'; later occurglani, jirni, and others. Certain other feminine\nnouns ofconcrete meaning occur: thus, jdni, 'woman',etc.\n2.Examples ofwords oftheother class are: vdhni, 'carrying', fwrni,\n'hasty', bhurni, 'excited', prenf, 'loving', vrsnf andvfmi,'virile'.\nInprem, y6ni, meni, freni, fromisseen astrengthening oftheroot,\nsuch asdoes notappear among thederivatives in ti.\nDerivatives innifrom roots with prefixes donotappeartooccur.\nInthewords ending inani, theahasprobably thesame value with\nthat ofati(above, 1157.3); butanihasgained amore independent status,\nandmaybebest treated asaseparatesuffix.\n1159. ofa ani.Thewords made bythis suffix have\nthesame double value with thosemade bythepreceding\nones. Their accent isvarious. Thus:\na.Feminineaction-nouns, sometimes with concreted meaning:as,isani,\n'impulse', pardm, 'injury', dyotani, 'brightness', ksipanf, 'blow', apam,\n'missile', variant, 'track', djani (a-djani:theonlyexample withprefix), 'goad'.\n"
+    },
+    {
+        "page": 401,
+        "content": "1161]STEMS INti,ni,ani, an,tu. 385\nb.Adjectives and other agent-words aresuch ascardni, 'movable',\ncaksani, 'enlightener', vaksdni,;\nstrerigthener'. From areduplicated root-\nform comes -paptani. From desiderative stems aremaderuruksdni, sisasani,\nand(with prefix) a-pufuksdni. And asmall number ofwords appear to\nattach themselves toans-aorist stem: thus, parsdni, saksdni, carsant.\n3.Itisquestionable whether the infinitives insdni (978) are tobe\nputhere, asaccusatives ofaformation inam, orunder thenextsuffix, as\nlocatives ofaformation inan,from rootsandstems increased byanaoristic s.\n1160. ERan.Notmany wordsappeartobemade with\nasuffix ofthisform, andofthese stillfewer areplainlyto\nbeconnected with roots. Certain rareneuters(along with\nthedoubtfulinfinitives)arenouns ofaction; the rest are\nmasculine andneuteragent-nouns. The accent isvarious.\na.The infinitives which admit ofbeing referred tothissuffix,aslocative\ncases,arethose inam,ofwhich the sibilant appearstobethe final ofa\ntense-stem. They are allgiven above(978).\nb.The other action-nouns inanaremahdn, 'greatness', mahhdn(?),\n'liberality', rajdn, 'authority' (RV., once: compare rdjan; theaccent-relation\nisthereverse oftheusualone), andgdmbhan, 'depth' (VS., once).\nc.Examples ofagent-nouns are :masc. tdksan, 'carpenter', r<f?an, 'king',\nvfsan, virile, bull', wfesdn, 'ox'; neut. wddn,'water1\n,cdksan, 'eye': with\nprefixes, pratidivan, 'antagonistatplay' (dtidivan, AV.,isperhaps abad\nreading), vibhvdn, 'superior'.\nd.Afewstems inan,running parallel with those inother suffixes\nand filling outtheir declension, were mentioned above (429 ff.).\n1161. rTtu.The great mass ofthewords ofthisform-\nation arethe infinitives accusatives inthe later lan-\nguage,intheearlier likewise datives andablative-genitives:\nseeabove, 970b,972. Butafewarealsoused independ-\nently,asaction-nouns orwith concreted meaning;andan\nextremelysmall number,ofsomewhat questionablecharac-\nter,appeartohave thevalue ofagent-words. Theyareof\nallgenders,butchieflymasculine. The roothastheguna-\nstrengthening.\nThe infinitive words areaccented ontheradicalsyllable\nwhensimple, andmost oftheothers have thesame accent;but\nafewhave thetoneontheending.\na.Examplesare: oftheregular formation, masc., gdntu, 'way', dhdtu,\n'element', rndntu, 'counsel';fern, vdrtu, 'morning'; neut. vdstu, 'abode';\nWhitney, Grammar. 25\n"
+    },
+    {
+        "page": 402,
+        "content": "386 XVII. PRIMARY DERIVATION.[1161\nwith accent ontheending, jantu, 'being', gatti, 'way' and'song', hetit,\n'cause', fcetoJ,'banner'(allmasc.); withunstrengthened root, rtu, 'season',\npita, 'drink', sfttu, 'birth'; with vrddhi-strengthening, vdstu(above).\nb.The infinitives intuhave (972) often theunion-vowel ibefore the\nsuffix, and this inafewcases islengthened toI.Inother useoccur also\n-tdritu and-dhdrltu(both withdus); turphdritu seems ofthesame formation,\nbut isobscure.\nThe infinitives, whenmade from roots withprefixes, have(aswaspointed\noutabove, 972) theaccent ontheprefix. Butthesamewords, when used\n(not infinitivally)infurther combination (with suanddws), retain theradical\naccent which belongstothesimple word :thus, duratyetu, durniyGmtu,\ndusparihdntu, suprattu.\nc.Inafewinstances, thesuffix tuappearstobeadded toatense- or\nconjugation-stem ina :thus,edhatti andvahatu;tanyatu andtapyatu ,-and\nsisasdtu. Theaccent ofthe last isparalleled onlybythat ofjivdtu, 'life',\nwhich isfurther exceptionalinshowing alonga;itisused sometimes in\nthemanner ofaninfinitive.\n1162. ^nu.This suffix forms acomparativelysmall\nbodyofwords\n,generally masculine, andhaving both the\nabstract andtheconcrete value.\nThe accent isusually ontheending, and therootun-\nstrengthened.\nExamplesare :bhanu, 'light' (later 'sun'), vagnti, 'sound', sunu, 'son',\nddnu(with irregular accent), m.'demon', n.'drop, dew'; dhenii, f.,'cow';\ngrdhnu, 'hasty', dhrsnti,'bold'.\nThis also(like <u),appears sometimes with aprefixed a:thus, krandanu\nand nadanti, 'roaring', nabhanu (and -nti, f.), 'fountain', vibhanjanti (only\ninstance withprefix), 'breaking topieces'; andperhaps krfdnu belongs here.\n1163. 51ilia.Thewords made with this suffix areal-\nmost withoutexception action-nouns (though some have\nassumed aconcretevalue). Theyareofallgenders. The\nroot isofaweak(oreven weakened) form, andtheaccent\nusually onthe suffix.\na.Examplesare :masc.bhrthd, 'offering', -ferffta, 'making', -itha,\n'going'; neut. ukthd, 'saying', nithd, 'song', tlrthd, 'ford';fern, (with a)gatha,\n'song', nitha, 'way'. Radical aisweakened toiin-githd and -plthd. Final\nmornislostin-gathd andhdtha(assometimes intheverbal inflection of\nthesame roots: 637,834 b).\nAfewexamplesofcombination withprefixes occur, with accent onthe\nfinal:thus, nirrthd, 'destruction', sarhgathd, 'union',etc.\nb. Stillmorecommon intheolder languageisaform ofthis suffix to\nwhich hasbecomeprefixedana,which isprobablyofthematicorigin, though\n"
+    },
+    {
+        "page": 403,
+        "content": "1168]STEMS INnu. tha. thu,yu 1ma,mi,man. 387\nbecome aunion-vowel. Thus :masc. cardtha, 'mobility', yajdtha, 'offering',\nravdtha, 'cry1\n,capdtha, 'curse', stavdtha, 'praise';neut. ucatha, 'speech',\nviddtha, 'ordering'. Beforethis,aroothassometimes guna: thus, faydtha,\n'couch', tvesdtha, 'vehemence'. With aprefix, theaccent isthrown forward\nuponthe final :thus, avasathd, 'abode', pravasathd, 'absence'; prandtha,\n'breath',istreated asifpran were anintegral root.\nc.Isolated combinations ofthawith otherpreceding vowels occur:thus,\nvdrutha, 'protection', withanother doubtful case ortwo;andmatutha (yman ?).\n1164. 51thu.This suffix hasan [aattached toit(likeO' V\n2Ttha, above), and. intheveryfew derivatives which it\nmakes, appears onlyas5fSTdthu.\nThe only Vedic examplesareejdthu, 'quaking', vepdthu, 'trembling',\nstandthu, 'roaring'. Later cases arenanddthu (TS.), vamathu, cvayathu,etc.\n1165. Tyu.With this suffix aremade averyfewnouns,\nboth ofagent andofaction, withunstrengthenedrootand\nvarious accent. Thus :\na.Abstracts (masc.) aremanyv, 'wrath', mrtyti,'death' (withtadded\ntotheshort final oftheroot).\nb.Adjectivesetc. arebhujyu, 'pliable', yundhyti, 'pure'; ydjyu, 'pious',\nsdhyu, 'strong', ddsyu, 'enemy', andone ortwomore.\nForother derivatives ending inyu,seethesuffix,below, 1178g,h.\n1166. 13ma.Theaction-nouns made bythis suffix are\nalmost allmasculine;andtheyareofvarious root-form and\naccent, asarealso theagent-nounsandadjectives.\na.Examplesoftheformer class are :ajmd, 'course', gharmd, 'heat';\nema, 'progress', bhdma, 'brightness', sdrma, 'flow', s(dma, 'song ofpraise'.\nb.Examplesofthe latter class are :tigmd, 'sharp', bhimd, 'terrible',\nfagmd, 'mighty'; idhmd, 'fuel', yudhmd, 'warrior'. Asingle instance from a\nreduplicatedroot istutumd, 'powerful'.\n1167. FTmi.Avery small number ofnouns, masculine\nandfeminine, formed with mi,maybeconveniently noticed here.\nThus, fromr-roots, urmf, 'wave', -kurmi, 'action', surrm, f., 'tube';\nfrom others, jamf (?),'relation', bhUmi orbh&mi, f., 'earth', laksmi, 'sign';\nalsoprobably rafmf, 'line, ray'.\n1168.Rman. The derivatives made with this suffix\ns.\narealmostonly action-nouns .Thegreat majorityofthem\nareneuter, andaccented ontheroot-syllable;amuch small-\nernumber aremasculine, andaccented onthe suffix. The\nfewagent-words are,ifnouns, masculine, andhave the\n25*\n"
+    },
+    {
+        "page": 404,
+        "content": "388 XVII. PRIMARY DERIVATION.[1168\nlatter accent :inseveralinstances, aneuter andamasculine,\noftheoneandtheother value andaccent, stand sideby\nside. The roothasingeneral the^^-strengthening.\n1.a.Examples ofregularly formed neuters are:kdrman, 'action',\njdnman, 'birth', ndman, 'name', vdrtman, 'track', vefman, 'dwelling', homan,\n'sacrifice', -dytitman, 'splendor'.\nb.Examples ofmasculine abstracts are: ojmdn, 'strength', jemdn,\n'conquest', svddmdn, 'sweetness'.\nc.Corresponding neuter action-nouns andmasculine agent-nouns are :\nbrahman, 'worship', andbrahman, 'priest'; ddman, 'gift', anddaman, 'giver';\ndhdrman, 'rule', anddharmdn, 'orderer'; sddman, 'seat', andsadmdn,'sitter'.\nVery fewother agent-nouns occur; andall,except brahman, areofrare\noccurrence.\nOntheother hand, varsman andsvadman(andvariman) have thedif-\nference ofgender andaccent without acorresponding difference ofmeaning.\nThenoun dfman, 'stone', though masculine,isaccented ontheradical\nsyllable; andtwo orthree other questionable cases ofthesame kind occur.\nThe derivatives inman used asinfinitives(974) have forthemostpart\ntheaccent ofneuters: theonly exceptionisvidmdne.\nd.Afewwords, ofeitherclass, have anirregular root-form :thus,\nbhuman'earth', syuman; bhumdn'abundance', slmdn, bhujmdn, vidmdn; and\nkdrsman, bhdrman, fdkman.\ne.Derivatives inmanfrom roots withprefixes arenotnumerous. They\nareusually accented ontheprefix, whether action-nouns oradjectives:thus,\nprdbharman, 'forthbringing', prdydman, 'departure'; dnuvartman, 'following\nafter' :theexceptions, vijaman, prativartmdn, visarmdn, areperhaps ofpos-\nsessive formation.\n2.Thesamesuffix, though only with itsabstract-making\nvalue, hasinanumber ofcases before itaunion-vowel,ior\nI;andimdn comes tobeused asasecondary suffix, forming\nabstract nouns(masculine)from acertain number ofadjectives.\na.Theneuters iniman andIman areallprimary formations, belonging\ntothe older language: thus, jdniman andvariman (beside variman, asno-\nticed above); andddrlman, dhdrlman, pdriman (andpdreman, SV., once),\nbhdrlman, variman, sdrlman, stdrlman, sdvlman, andhdvlman. Those in\nIman arehardly metwith outside theRig-Veda.\nb.The masculines inimdn areintheoldest language lessfrequent\nthan theneuters just described :they arejarimdn, prathimdn, mahimdn,\nvariman (besidetheequivalent variman andvariman), varsimdn(beside the\nequivalent varsman andvarsman), harimdn,anddraghimdn (VS.) beside dragh-\nmdn(RV.). Some ofthese, aswell asofthe derivatives insimple man,\nattach themselves inmeaning, orinformalso,toadjectives,towhich they\nseem theaccompanyingabstracts :comparethesimilar treatment oftheprimary\ncomparatives andsuperlatives (above, 468): such arepapmdn (topapa,\n"
+    },
+    {
+        "page": 405,
+        "content": "1170]STEMS INman. van. vana, vani, vanu. 389\npdpiyas, etc.); drdghmdn etc.(todirghd, drdghlyas, etc.); vdriman etc.(to\nurn, vdrlyas, etc.); prdthiman (toprthu, prdthistha) ;harimdn(toMrz orftan'ta);\nvdrsman etc.(tovdrslyas etc.); svddman etc.(tosuadw, svddlyas, etc.).\nThen intheBrahmanalanguage arefound further examples: thus, dhumri-\nman (TS., K.),dradhiman(K.:todrdhd, drddhlyas, etc.), iaruniman(K.),\nlohitiman(KB.); and still later such aslaghiman; while ksepiman (toksiprd,\nkseptyas, etc.)andkrsniman, and soon,areallowed bythegrammarians.\n1169.5R tfaft.Bythis suffix aremade almostonly\nagent-words, adjectives andnouns, the lattermostly mas-\nculines. The root isunstrengthened, and toashort final\nYowel isadded arTtbefore the suffix. The accent isal-\nmostalways ontheroot, both inthesimple words andin\ntheir compounds.\nThe insertion oftisanindication that thewords ofthisform areorig-\ninally made bytheaddition ofantoderivatives inuandtu;yetvanhas\nthepresent value ofanintegral suffix inthelanguage, andmust betreated\nassuch.\n1.a.Examplesoftheusual formation are: masc. ydjvan, 'offering',\ndrtihvan, 'harming', fdkvan, 'capable', -rikvan, 'leaving', -jttvan, 'conquering',\nsfitvan, 'pressing', kftvan, 'active', -gdtvan (like -gat, -gatya), 'going', sdtvan\n(j/san), 'warrior', drvan(only example with strengthened root), 'courser';\nneut. pdrvan, 'joint', dhdnvan, 'bow'.\nb.Examples from roots withprefixes (whicharenotrare)are: atitvan,\n'excelling', upahdsvan, 'reviler', sambhrtvan, 'collecting'; andprobablyvivas-\nvan, 'shining': abhfsatvan isacompound with governing preposition (1310).\nForthecompounds with other elements, which, exceptinspecial cases,\nhave thesame accent,seebelow, 1277.\nC.Thestem musivdn, 'robber' (RV., once),istheonly onewith a\nunion-vowel, and isperhapsbetter regardedasasecondary derivative of\nwhich afewaremade with this suffix: seebelow, 1234.\nd.From areduplicated rootaremade rdravan andcikitvdn (and possibly\nvivdsvan}.\n2.Thenumber ofaction-nouns made, with thesuffix van isextremely\nsmall:namely, davdn, 'giving', andturvdn, 'overcoming', both used asin-\nfinitives (974), andbhurvdn, 'unrest'(?):likewise dhurvan, 'injury',alsoused\nasinfinitive (unlessthis isratherdhurv-an).\nThe feminines correspondingtoadjectivesinvan arenot\nmade(apparently) directly from this suffix, butfrom vara, and\nendinvari; seebelow, 1171 b.\n1170. SRvana.3Hvani.cRvanu. Theveryfewwords\nmade with these suffixes maybestbenoticed here, incon-\n"
+    },
+    {
+        "page": 406,
+        "content": "390 XVII. PRIMARY DERIVATION.[1170\nnection with3Rvan(ofwhich theothers areprobablysec-\nondary extensions).\na.With vana aremadevagvand, 'talkative', satvand,'warrior'(beside\nsdtvan, above); and,from areduplicated root,wukvand, 'shining'.\nb.With vani aremade from simple rootsturvdni, 'excelling', and\nbhurvdni,'restless' (compare turvdn andbhurvdn, just above); and, from\nreduplicated roots, fUfukvdni, 'shining', dadhrsvdni, 'daring', tuturvdni,\n'striving after', andjugurvdni, 'praising': arharisvdni isobscure.\nc.With vanu ismade only vagvanti, 'tone,noise'.\n1171.cTfvara. With this suffix aremade afewderiv-\natives, ofallgenders, havingforthemostpartthevalue\nofagent-nounsandadjectives.\nMuch more common arethefeminine stems inof^t\nvari, which, from theearliestperiod,serve asthecorres-\npondingfeminines tothemasculine stems in3Ffvan.\na.Afewmasculine adjectives invaraoccur, formally accordant (except\ninaccent) with thefeminines: thus, itvard, 'going', -advard, 'eating',\n-sadvard, 'sitting', ifvard, 'ruler, lord'; andwith them doubtless belongs\nvidvald, 'knowing' (withIforr).\nb.Thefeminines invan accord intreatment oftherootandinaccent\nwith themasculines invan towhich they correspond:thus, ydjvari, -jttvari,\nsftvari, -$ivari, -yavari, and soon(about twenty-five such formations in\nRV.); from areduplicated root, -fifvan.\n6.Avery small number ofneuters occur, with accent ontheroot:\nthus, kdrvara, 'deed', gdhvara, 'thicket'; andafeminine ortwo, with accent\nonthepenult:urvdra, 'field', andurvdrl,'tow'(bothofdoubtful etymology).\nWetakeupnow thesuffixes bywhich aremadeonlystems\nhavingthevalue ofagent-nouns andadjectives;beginning with\nabrief mention oftheparticipial endings, which ingeneral have\nbeen already sufficientlytreated.\n1172.%R[^ant(orETcT^).The office ofthissuffix, in\nmaking present andfutureparticiples active, hasbeenfully\nexplained above, inconnection with thevarious tense-stems\nandconjugation-stems (chaps.VIII.XIV.),incombination\nwithwhich alone itisemployed (not directlywith theroot,\nunless this isalsoused astense-stem).\nWith thesame oraformallyidentical suffix aremade from\npronominalrootslyant andkiyant (517). Andddvayant,'not\ndouble-tongued' (RV.,once), appearstocontain asimilar form-\n"
+    },
+    {
+        "page": 407,
+        "content": "1176]STEMS INvara, ant, vans, mana, ana,ta. 391\nation from thenumeral dm unless wearetoassume ade-\nnominative verb-stem asintermediate.\nHeremayalsobebestmentioned thewords made with\ntheso-called suffix anta(fern,anta oranti], being evident trans-\nfers(Prakritic)ofstems inanttothea-declension.\nExtremely fewsuch words occur intheoldest language:namely, pdnta,\n'draught' (RV.), vasantd, 'spring' (RV.:beside italsohemantd, 'winter');\nvefantd orve$anti, 'tank'(AV.; QB. ve$dnta], jivanti,acertain healing\nplant (RV.j; andprobably theproper name tarantd(RV.).Afewothers are\ninstanced asadmitted later :thus, nandanta andnandayanta, gadayanta,\njayanta, jaranta, prananta, etc.: allaresaid tobeaccented onthe final.\nAdifferent extension ofthesame suffix isexhibited intheproper\nnames dhvasdnti andpucantf (RV.),withwhich maybementioned purusdnti.\n1173. 3\"favans(orcfHvas).Forthe(perfect active) par-\nticiples made with thissuffix, seeabove, chaps.X.and\nXIV., and458 ff.\nAfewwords ofirregular and questionable formation were noticed at\n462, above. Also, apparent transfers toaform usorusa.\nThe oldest language (RV.) has avery fewwords invas,ofdoubtful\nrelations: fbhvas, 'seizing', and ffkvas,'skilful' (besidewords invaand\nvan], andperhapskhidvas(ykhad). The neuter abstract vdrivas, 'breadth,\nroom' (belongingtourti, 'broad',inthesamemanner with variyas andvarimdn],\nisquiteisolated.\nTheunique tatan'usti (RV., once)ispossiblytobedivided tatanus-ti,\nandconnected with this suffix.\n1174. JTHmana. Theparticiples havingthisending\nare, ashasbeen seen, presentand futureonly,andhave\nthemiddle,, orthederivedpassive,value belongingingen-\neral tothestems towhich thesuffix isattached.\n1175.^Hana.Theparticiples endinginTRanaare\nofmiddle andpassive value,like thosejust noticed, and\neitherpresent, perfect,or(partlywith theform$THsana:\nabove, 897)aorist.\nAfewother words ending inthesamemanner intheoldlanguage may\nbementioned here. TheRV. hastheadjectives vdsavana, 'well-endowed'r\nandurdhvasand, 'uplifted', evidently made onthemodel ofparticipialstems.\nAlso theproper names dpnavana, pfthavana, andcydvanaandcydvatana.\nPdr$ana, 'abyss',isdoubtful.\n1176. rTta.The useofthis suffix informing parti-\nciples directly from theroot, orfrom aconjugational (not\n"
+    },
+    {
+        "page": 408,
+        "content": "392 XVII. PRIMARY DERIVATION.[1176\natense) stem, wasexplained above, chap.XIII. Thepar-\nticiplesthusmade areinpart intransitive, butingreat part\npassiveinvalue(likethosemade bythetwoprecedingsuf-\nfixes, butinmuchlarger measure, andmoredecidedly:.\nFor thefrequentuseoftheconnecting vowel with the\nsuffix, makingitsform,ita,seealsochap.XIII.\na.Afew general adjectives,ornouns with concrete meaning, are\nadaptations ofthisparticiple. Examples are:trsta, 'rough', citd, 'cold1\n,\ndrdhd(fordrdhd: 224a), 'firm'; dutd, 'messenger', sutd, 'charioteer'; rtd,\n'right', ghrtd, 'ghee', jatd, 'kind', dyutd, 'gambling', nrttd, 'dance', fivitd,\n'life', caritd, 'behavior.'. Theadjective tigitd (RV.), 'sharp ,shows anomalous\nreversion ofpalataltoguttural before the i(216). Vavdta, 'dear',isasingle\nexample from areduplicated root.\nb.Doubtless after theexample andmodel ofparticiples from denomin-\native stems(ofwhich, however, noinstances arequotable from theYeda),\nderivatives initaareinthe later language made directly fromnoun and\nadjective-stems, having themeaning of'endowed with,affected by,made to\nbe',andthe like(comparethesimilar English formation ined, ashorned,\nbarefooted, bluecoated). Examplesare rathita, 'furnished with achariot', etc.\nc.Afewwords endinginfaareaccented ontheradical syllable, and\ntheir relation totheparticipialderivatives isvery doubtful: such aredsta,\n'home', mdrta, 'mortal', vdta, 'wind'; andwith them may bementioned\ngdrta (?),nafcta, 'night', hdsta,'hand'.\nd.Several adjectives denoting color endinita,butarehardly connect-\niblewith roots ofkindred meaning: thus, palitd, 'gray', dsita, 'black',rtihita\nandlohita, 'red', hdrita, 'green'; akin withthem areeta, 'variegated', cyetd,\n'white'.\nThefeminines ofthese stems areinpart irregular:thus, enlandcyenl;\nrohinl andlohirii, andhdrinl (but thecorresponding masc. hdrina also occurs);\nanddsikni, pdliknl, and hdriknl,\ne.Asmall number ofadjectivesintheolder language endinginata\narenot tobeseparated from theparticipial words infa,althoughtheir\nspecific meaningisinpart gerundive. They are: pacatd, 'cooked', dar$atd\nand pafyata, 'seen, tobeseen, worth seeing'; and soyajatd, haryatd,\nbharatd. The yofpafyata and haryatd indicates pretty plainlythat thea\nalso isthat ofapresent tense-stem. Rajatd, 'silvery',isofmore obscure\nrelation toyraj'color'.\n1177. ^na(and^Tina,3^una}.Theuseofthesuffix\n^nainforming from certain rootsparticiples equivalentto\nthose incTta,either alongside thelatter orinstead ofthem,\nwasexplained above, chap.XIII.(957).\na.With thesame suffix aremade anumber ofgeneral adjectives, and\n"
+    },
+    {
+        "page": 409,
+        "content": "1178] STEMS INta,na, u. 393\nofnouns ofvarious gender (fern,inna).Theaccent isonthesuffix oron\ntheroot.Afewexamples are :umd, 'hot', puna, 'fortunate', dpna, 'ravenous',\nfvitna, 'white'; masc. prapnd, 'question', yajna, 'offering', ghrnd, 'heat',\nuarna, 'color', svdpna, 'sleep'; neut. parnd, 'wing', rdtna, 'jewel' (?);fern,\ntrena, 'thirst', yacnd, 'supplication'. Butmany ofthestems ending inna\narenotreadily connectible with roots. Anantithesis ofaccent isseen in\nfedrna, 'ear', and fcarnd, 'eared'.\nb.The very fewwords endingininaareperhaps related with these\n(perhapsrather with those inana): thus, amind, 'violent', vrjind, 'crooked',\nddksina, 'right', andone ortwoothers ofquestionable etymology.\nc.Ofthewords inwna, fewareclearly referable toroots: thus, fcart/na,\n'action', dharuna, 'bearing', -cetuna, 'showing'; drjwna, 'white', tdruna,\n'young', vdruna, 'Varuna'. Inmeaning andinaccent they vary like thede-\nrivatives inana.\nThese are alltheproper participial endings ofthelanguage.\nThegerundives,laterandearlier, areinsogreat partevident\nsecondary formations, thattheywillbenoticed farther on,un-\nderthehead ofsecondary derivation.\nWe willtakeupnow theother suffixes forming agent-nouns\nandadjectives, beginning with those which havemore orless\naparticipialvalue.\n1178. 3u.With this suffix aremade aconsiderable\nbodyofderivatives, ofveryvarious characteradjectives,\nandagent-nounsofallgenders,with different treatment of\nthe root, andwith different accent. Itisespecially used\nwith certainconjugational stems, desiderative(particularly\nlater) anddenominative(mainly earlier), making adjectives\nwith thevalue ofpresent participles;and insuch use it\nwins inparttheaspectofasecondarysuffix.\nThe roothasoftenest aweak(orweakened)form;but it\nissometimes vriddhied;least often (when capableofguna],it\nhas the^wna-strengtheningallwithout anyapparent con-\nnection with either accent ormeaningorgender. After final\nradical aisusually addded y(258)before the suffix. Afew\nderivatives aremade from thereduplicatedroot.\nMany words ending inuarenotreadily, ornotatall,connectible with\nroots;exampleswillbegiven only ofthose thathaveanobvious etymology.\na.Examples' ofordinary adjectives are :wrti, 'wide', rju, 'straight',\nprtJiu, 'broad', mrdu, 'soft', sadhu, 'good', svadti, 'sweet', tdpu, 'hot', vdsu,\n'good'; jayu, 'conquering', darti, 'bursting'; pat/tJ, 'lying', reku, 'empty';\ndhayu, 'thirsty', payti, 'protecting'. Final aappearstobelostbefore the\nsuffix in-sthu (susthu, anusthu).\n"
+    },
+    {
+        "page": 410,
+        "content": "394 XVII. PRIMARY DERIVATION.[1178\nb.Examplesofnouns are :masc. ah$u, 'ray', ripu, 'deceiver', vayii,\n'wind-god', dsu, 'life', mdnu, 'man, Manu';fern, isu(also masc.), 'arrow',\nsfndhu (also masc.), 'river'; tanu, 'body'.\nc.Derivatives from reduplicatedroots are:cikitu, jigyu,sisnu. -tatnu\n(unlessthis ismade withnuortnu],ydyuoryayu andyfyu (withfinal a\nlost), pfpru (proper name), -didhayu;and babhru, -raru (araru), malimlu(?)\nhave theaspectofbeing similar formations.\nd.Afew derivatives aremade from roots withprefixes, with various\naccentuation :forexample, upayu, 'on-coming', pramayu, 'goingtodestruction',\nvikUndu,acertain disease, a&Tifpw,'rein(director)', sdmvasu, 'dwellingto-\ngether'.\ne.From tense-stems, apparently,aremadetanyu, 'thundering', and\n(withaoristics)ddksu anddhdksu(allRV.J.\nf.Participial adjectives inufrom desiderative \"roots\"(stemswith loss\noftheir finala)are sufficiently numerous intheancient language (RV.has\nmore than adozen ofthem, AV. notquitesomany) toshow that theform-\nation wasalready aregular one,extensible atwill\n;andlater such adjectives\nmaybemade from every desiderative. Examples (older)are: ditsu, dipsu,\ncikitsti, titiksu, piplsu, mumuksu, fifliksu; and, withprefix, abhidipsu.\nThese adjectives, both earlier andlater, may take anobject inthe\naccusative (27la).\ng.Afew similar adjectives aremade intheolder language from caus-\natives :thus, dharayu ('persistent'), bhajayu, bhdvayu, mahhayu, mandayu,\ncramayii; andmrgayu from thecaus.-denom. mrgdya.\nh.Much more numerous, however, aresuch formations from themore\nproper denominatives, especiallyinthe oldest language (RV. hastoward\neightyofthem; AV. only aquarter asmany, including sixoreight\nwhich arenotfound inRV.; andthey are still rarer intheBrahmanas).In\namajority ofcases, personalverbal forms from thesame denominative stem\nareinuse :thus, forexample,toaghayu, aratlyu, rjuyu, caranyu, memasyw,\nsanisyii, urusyu, saparyu; inothers, only thepresent participleinydnt,or\nthe abstract noun inyd(1149),ornothing atall.Afew aremade upon\ndenominative stems from pronouns:thus, tvayu (beside tvaydnt andtvayd},\nyuvayuoryuvayfi, asmayu, svayu, and themore anomalousahamyti, and\nkirhyu. Especially where noother denominative forms accompanythead-\njective,thishasoften theaspect ofbeing made directly from thenoun with\nthesuffix yu,either with ameaning of'seeking ordesiring', orwith amore\ngeneral adjective sense: thus, yavayti, 'seeking grain', varahayu, 'boar-hunting',\nstanasyti, 'desiring thebreast'; urnayti, 'woolen', yuvanyti, 'youthful',\nbhimayti,'terrible'. And sothe\"secondary suffix yu\"wins adegree ofstanding\nandapplicationasoneforming derivative adjectives (asinaharhyu andkimyii,\nabove, and doubtless someothers, even oftheRV.words). Inthree RV.\ncases, the final asofanoun-stem iseven changed toobefore it:namely,\nahhoyti, duvoyti (andduvoyd besideduvasyd], dskrdhoyu.\nNone ofthewords inyushow intheVeda resolution into iu.\n"
+    },
+    {
+        "page": 411,
+        "content": "1181]STEMS INw,u,uka^ aka. 395\n1179. \"37u.The longuisavastlyrarer suffix than\ntheother long simple vowels, alreadydescribed(1149,1155 .\nItmakes asmall number offemininescorrespondingto\nmasculines inu,averyfewindependent feminines, and\ntwo orthreeveryrare masculines :astoallwhich, see\nabove, 355 c.\n1180. 3Sfiuka.With this suffix aremade derivatives\nhavingthemeaning and construction (271g)ofapresent\nparticiple. The root isstrengthened, andhastheaccent.\nThe derivatives inukaarehardly known intheVeda; buttheybecome\nfrequent intheBrahmanas, ofwhose language they areamarked character-\nistic\n;andthey arefound occasionally inthe later language. Inallprob-\nability, they areoriginally andproperly obtained byadding thesecondary\nsuffix fca(1222)toaderivative inu;buttheyhave gained fully thechar-\nacter ofprimary formations, and inonly aninstance ortwo isthere found\ninactual useanw-word from which they should bemade.\nThe root isonly sofarstrengthened that theradical syllableisaheavy\n(79) one; and ithasthe accent, whether thederivative ismade from a\nsimpleroot orfrom onewithprefix.\na.Examples, from theBrahmanalanguage,are :vdduka, ndfuka,\nupakrdmuka, prapdduka, upasthdyuka (258), vydyuka, veduka, bhdvuka,\nkstidhuka, haruka, vdrsuka, samdrdhuka, ddn?uka, alambuka, yiksuka (GB.:\nRV. has fiksu], pramayuka (SB. haspramayu}.\nb.Exceptionsasregards root-form are:nirmdrguka (with urdd/u'-strength-\nening,asisusual with this root :627) ,-kasuka. AV. accents sdrhkasuka\n((^B.has sarhkdsuka) and vikasuka :RV. hassdnukd (whichisitsonly\nexampleoftheformation,ifitbeone;AV. has alsoghdtuka fromytian,\nanddpramdyuka) ;vasukd (TS.etal.)isprobablyofanother character.\nAfandyuka (PB. etal.)istheonlyexample noticed from aconjugation-stein.\nc.Aformation inuka(asuffix oflikeorigin, perhaps, with uka}may\nbementioned here :namely, from reduplicated roots, jdgaruka, 'wakeful',\ndandafwfca, 'biting', ydyajuka, 'sacrificing much', vdvaduka(later), 'talkative';\nsalaluka isquestionable.\n1181. fiaka. Here, asinthepreceding case,we\nseem tohave asuffix made bysecondaryaddition of 3fika\ntoaderivative in fa;but ithas, forthesame reason as\ntheother,arighttobementioned here. Itsfreeusein\nthemanner ofaprimarysuffix isofstill later date than\nthat ofuka;ithasveryfewexamplesintheolderlanguage.\n"
+    },
+    {
+        "page": 412,
+        "content": "396 XVII. PRIMARY DERIVATION.[1181\na.InRV. isfound(besides pavakd, which has adifferent accent, and\nwhich, asthemetre shows,isusually tobepronounced pavaka] only sdyaka,\n'missile'; AV. adds piyaka andvddhaka, andVS. abhikr<S$aka. Butinthe\nlater language such derivatives arecommon, usually withraisingofthe\nroot-syllable bystrengthening toheavy quantity:thus, nayaka, ddyaka (258),\npdcaka, grdhaka, dravaka, bodhaka;but alsojanaka, khanaka. They are\ndeclared bythegrammarians tohave theaccent onthe radical syllable.\nThey often occur incopulative composition with gerundives ofthesame\nroot: thus, bhakayabhaksaka,'eatable andeater', vacyavacaka, 'designated\nand designation', and soon.\nThat the derivatives inakasometimes take anaccusative object was\npointed outabove (27Ic).\nThe corresponding feminine ismade sometimes inaka orinaki,but\nmore usually inika: thus, ndyikd (with ndyakd), pacika, bodhika, dravika;\ncompare secondary aka, below, 1222 d.\nb.Derivatives inaka aresaid tobemade from afew roots :thus,\njalpdka, bhiksdka; butthey arenotfound intheVeda (unless in\"pavaka\":\nseeabove), andappeartobevery rare atevery period. With dku ismade\ninRV.mrdaydku, from thecausative stem :pfdaku andtheproper name\nfksvaku areofobscure connection.\nDerivatives inikaand Ika willbetreated below, inconnection with\nthose inka(1186).\n1182. cTtr(orrT|\" tar}.The derivatives made bythis\nsuffix, asregardsboth theirmode offormation and their\nuses, have been thesubjectofremark more than once\nabove(see369ff.,942if.).Agent-nounsareformed with it\natevery periodofthelanguage;these intheoldestlang-\nuageareveryfrequently usedparticipially, governinganob-\njectintheaccusative(271 d);latertheyenter intocombin-\nation withanauxiliaryverb, and,assumingafuture mean-\ning,make aperiphrasticfuture tense(942).\nTheircorresponding feminine isintri.\na.The roothasuniformlythe^tma-strengthening. Aunion-\nvowel i(very rarely, one ofanothercharacter)isoften taken :\nasregardsitspresenceorabsence intheperiphrasticfuture\nforms, seeabove(943).\n\"Without grwna-changeisonly ustr, 'plough-ox' (noproper agent-noun:\napparently uks-tf:compare thenouns ofrelationshipfurtheron).The root\ngrah has,asusual,I thus, -grahitf ;and thesame appearsin-taritf\nand-maritf. Anu-vowel istaken instead bytdrutr and tarutf, dhdnutr, and\nsdnutr; long invarutr;strengthenedtooinman6tr andmanotf. From are-\nduplicated rootcomes vdvdtr.\n"
+    },
+    {
+        "page": 413,
+        "content": "1183]STEMS INaka, tr,in. 397\nb.The accent, intheolder language,issometimes on\nthesuffix andsometimes ontheroot;or,from roots combined\nwithprefixes, sometimes onthe suffix andsometimes onthe\nprefix.\nIngeneral, theaccent ontheroot orprefix accompanies theparticipial\nuseoftheword; butthere areexceptionstothis: inaveryfewinstances\n(four), aword with accented suffix hasanaccusative object; verymuch\nmore often, accent onthe rootappears along with ordinary noun value.\nThe accent,aswell astheform,ofmano'tr isanisolated irregularity.\nExamples are:jetadhdnani, 'winning treasures'; yuydm mdrtarhfrtitarah,\n'yelisten toamortal'; but, onthe other hand, yarhtd vdsunividhate,\n'bestowing good things onthe pious'; and jetajdndndm, 'conqueror of\npeoples'.\nc.Theformation ofthese nouns intrfromconjugation-stems, regular\nandfrequent inthelater language, and notvery rareintheBrahmanas,is\nmetwithbutonce ortwice intheVeda(bodhayitf andcodayitri, RV.). In\nnestr,acertainpriest (RV. andlater),isapparently seen theaoristic s.\nd.Thewords ofrelationship which, inwhatever way, have gained the\naspect ofderivatives intr,arepitf, mat/, bhrdtr, ydtr, duhitf, ndptr, jdmatr.\nOfthese, only mdtf andydtrareinaccordance with theordinary rules of\ntheformation intr.\ne.Instead oftrisfound turinone ortwoRV.examples: j/amttfr,\nsthattir.\nf.Apparently formed byasuffix r(orar)areusr, savyasthr, ndnandr,\ndew,the lasttwobeing words ofrelationship. Forother words ending in\nr,see369.\n1183. ?Tin.This isanother suffix which hasassumed\naprimary aspectanduse, whileyetevidentlyidentical in\nrealcharacter with thefrequent secondarysuffix ofthesame\nformdenoting possession (below, 1230).\nHow far ithadgained aprimary value intheearly language,isnot\neasy todetermine. Most ofthewords ininoccurring inRV.andAV. are\nexplainable aspossessives ;inmany theother value ispossible, andinafew\nitisdistinctly suggested:thus, kevaladfa, bhadravadin, nitodtn, dfdrdisin,\ndnamin, vivyadMn;with aoristic8,saksfn; and,withreduplication, niyayin.\nAstheexamples indicate, composition, both with prefixes andwith other\nelements,isfrequent; and,inallcasesalike, theaccent isonthe suffix.\nLater, theprimary employmentisunquestionable, andexamples ofit,\nchiefly incomposition,arefrequent. The radical syllableisusually strength-\nened,amedial abeing sometimes lengthened and sometimes remaining\nunchanged. Thus, satyavddin, 'truth-speaking', abhibhasin, 'addressing',\nmano/mrm, 'soul-winning'. labhdvin has established itself aprevailingly\nfuture meaning: 'about tobe'.\n"
+    },
+    {
+        "page": 414,
+        "content": "398 XVII. PRIMARY DERIVATION.[1183\nTheuse ofanaccusative object with words ininwasnoticed above\n(271 b).\n1184.^TJH lyasandJ$ istha. These, which, fromforming\nintensive adjectives correspondingtotheadjectiveofroot-form,\nhavecome tobeused, within somewhat narrow limits, assuf-\nfixes ofadjective comparison, havebeenalready sufficientlytreat-\nedabove, under thehead ofcomparison (466 470).\nItmaybefurther noticed thatjyestha has intheolder language (only\ntwoorthree times inRV.) theaccent alsoonthefinal, jyestha ;andthat\npdrsisthaismade from asecondary from ofroot, with aoristic sadded.\nWhen thecomparativesuffix hastheabbreviated form yas(470) ,itsy\nisnever toberead intheVeda as i.\nNoother suffixes make derivativeshaving participialvalue\notherwise than inrareandsporadiccases;those that remain,\ntherefore, willbetaken upmainlyintheorder oftheir fre-\nquency andimportance.\n1185.^tra. With, this suffix areformed averyfew\nadjectives, and aconsiderable number ofnouns, mostly\nneuter, and oftenhavingaspecialized meaning,assigni-\nfyingthemeans orinstrument oftheactionexpressed by\ntheroot. The latter hasusuallythe^^-strengthening,but\nsometimes remains unchanged. Theaccent isvarious, but\nmore often ontheradicalsyllable.\nHere,asincertain other cases above, wehave probablyasuffix origin-\nally secondary, made byadding atotheprimarytrortar(1182); but its\nuse islike that ofaprimary suffix.\n,a.Examples ofneuter nouns are :gdtra, 'limb', pdttra, 'wing', pdtra,\n'cup', ytiktra, 'bond', vdstra, 'garment', prcftra, 'ear'; astra, 'missile', stotrd,\n'song ofpraise', potrd, 'vessel';ofmore general meaning, ddttra, 'gift',\nksetra, 'field', m<atra, 'urine', hotrd, 'sacrifice'. Thewords accented onthe\nfinal have often anabstract meaning: thus, fesatra, 'authority', rastrd,\n'kingdom', fasfra, 'doctrine', sattrd. 'sacrificial session'(also jnatrd, 'know-\nledge').\nb.Masculines are: ddhstra, 'tusk', mantra, 'prayer', attra(oratrd:232,\n'devourer', ustra, 'buffalo, camel', and afew ofquestionable etymology, as\nrm'fra, 'friend', putra, 'son', vrtrd,'foe'. Mitrd and vrtrd aresometimes\nneuters even intheVeda, and mitra comes later toberegularly ofthat\ngender.\nC.Feminines(intra) are :astra, 'goad', matra, 'measure', ftdfra,'sac-\nrifice' (beside hotrd), danstrd(later,forddhstra}; nastrd, 'destroyer'.\nd.Notseldom,a\"union- vowel\" appears before thesuffix; but this is\nnotusually theequivalent oftheunion-vowel used with tr(above, 1182a).\n"
+    },
+    {
+        "page": 415,
+        "content": "1188] STEMS INiyas, istha, tra, ka,ya,ra. 399\nForthewords initrahave theaccent oni:thus, arttra, 'impelling, oar',\nkhanttra, 'shovel', pavftra, 'sieve'; janttra, 'birth-place', santtra, 'gift',etc.:\nthecombination lirahasalmost won thecharacter ofanindependentsuffix.\nThepreceding vowel isalsosometimes a(sometimes apparently ofthepres-\nent-stem):thus, ydjatra, 'venerable', krntdtra, 'shred', gayatrd (L-trl), 'song',\npdtatra, 'wing'; butalsodmatra, 'violent', -krtatrd, 'cutting-place', vddhatra,\n'deadly weapon', andvaratrd, 'strap'. Tarutra, 'overcoming', corresponds\ntotarutr.\nThewords stillused asadjectives intraaremostly such ashave union-\nvowels before the suffix. Asingle example from areduplicatedroot is\njohutra, 'crying out'.\ne.Aword ortwo in triand trumay beaddedhere,asperhapsof\nkindred formation with those intra :thus, dttri, 'devouring', arcdtri, 'beaming';\nfdtru (fdttru: 232), 'enemy'.\n1186. 3fika.The suffix^haisofverycommon usein\nsecondaryderivation(below, 1222); whether itisdirectly\nadded toroots isalmostquestionable:atany rate, extreme-\nlyfewprimaryderivatives aremade with it.\nThewords which have most distinctly theaspectofbeing made from\nroots arefuska, 'dry', floka (j/fru, 'hear'), 'noise, report', etc.,and-sphaka,\n'teeming'; and stuka, 'flake', andstokd, 'drop', seem tobelong together to\naroot stu. Other words inkaareofobscure connections,forthemostpart.\nButkaenters, initsvalue assecondary, into thecompositionofcertain\nsuffixes reckoned asprimary:seeakaanduka (above, 1180, 1181).\nAfewwords inwhich ikaand Ikaseem added toaroot, though they\nare really ofakindred formation with thepreceding, maybemost conve-\nniently noticed here :thus, vrfdka (jAmifc), 'scorpion';dnlka(?),'face', dffifca,\n'aspect', mrdlkd, 'grace', vrdhlkd, 'increased, a$arlka and ufpcwifca, 'gripes',\n-rjlka, 'beaming'; and, fromreduplicated root, parpharika, 'scattering' (?).\nCompare secondary suffix ka(below, 1222).\n1187. ETyet\"Itisaltogether probablethat apartofthe\nderivatives made with this suffix arenotlessentitled toberank-\nedasprimary thanmanyofthose which areabove soreckoned.\nSuch, however, arewith somuch doubt anddifficultytobe\nseparated from thegreat mass ofsecondaryderivatives made with\nthesame suffix that itispreferredtotreatthem alltogetherunder\nthehead ofsecondary formation(below, 121013).\n1188.^ra.With this suffix aremade aconsiderable\nnumber ofadjectives,almostalwayswithweak root-form,\nandusually with accent onthe suffix. Also,afewwords\nused asnouns, ofvariousgender.\n"
+    },
+    {
+        "page": 416,
+        "content": "400 XVII. PRIMARY DERIVATION.[1188\nInsome cases, the suffix isfound with apreceding vowel,\nhavingtheaspectofaunion-vowel.\na.Examples ofadjectives inraofobvious derivation are :ksiprd, 'quick',\nchidrd, 'split', turd, 'strong', bhadrd, 'pleasing', paferd, 'mighty', pwfera, 'bright',\nhihsrd, 'injuriousJ\n; with accent ontheroot, grdhra, 'greedy', dhira,'wise'\n(secondary?), vfpra, 'inspired'.\nb.From roots with prefixes conie onlyanexample ortwo :thus, niczra,\n'attentive', nimrgra, 'joiningon'.\nc.Nouns inraare,forexample:masc. vird, 'man', vajra, 'thunderbolt',\npura, 'hero'; neut. ksird, 'milk', riprd, 'defilement';fern,dhdra, 'stream',\nstira, 'intoxicating drink'.\nThe forms ofthis suffix with preceding vowel may bestbeconsidered\nhere, although some ofthem have nearly gained thevalue ofindependent\nendings. Thus :\nd.With araaremade afewrarewords :theadjectives dravard,'run-\nning', patard, 'flying'., (with prefix) w/ocard, 'suiting'; and the neuters\ngambhdra, 'depth', tdsara, 'shuttle', sdnara, 'gain': bharvard andvasard are\ndoubtless ofsecondary formation\n;andthesame thing maybeplausibly con-\njectured ofothers.\ne.With iraaremade afewwords, some ofwhich areincommon use :\nthus, cy'ird, 'quick', isird, 'lively', madird, 'pleasing', dhvasird, 'stirring up',\nbadhird, 'deaf; perhaps sthdvira, 'firm'; andsthird, 'hard', andsphird, 'fat',\nwith displacementoffinal radical a;alsosarird,'wave'(usually salila), and\none ortwoother words ofobscure derivation.\nWith iraaremade gabhird orgambhird, 'profound', andpdumi, 'mighty';\nandperhaps pdnra, 'body'.\nf.With ura aremade afewwords, ofwhich thesecondary character\nisstillmoreprobable: thus, ahhurd (an/m-ra?), 'narrow',dsura(dsw-ra?),\n'living', vithurd, 'tottering', yddura, 'embracing'.\nWithwra, apparently,ismadesthurd,'stout' (compare sthdvira).\n1189. ^Tla.This suffix isonlyanother form ofthe\npreceding, exchangingwith itincertain words,inothers\nprevalentlyorsolely usedfrom their firstappearance.\nConspicuous examplesoftheinterchangearecukld, sthuld,\n-micla, cithila,salild.\nExamplesofthemore independent use are: paid, 'protecting',dnila\n(oram'to), 'wind', trpdla, 'joyous';later capala andtarala\"(saidtobeaccented\nonthefinal), andharsula(thesame). Many words ending inlaareofobscure\netymology.\n1190. ^va.Very fewwords ofclear derivation are\nmade with this suffix toofew tobeworthclassifying.\nTheyareofvarious meaningandaccent, andgenerally show\naweak root-form.\n"
+    },
+    {
+        "page": 417,
+        "content": "1195]STEMS INra. la,va,ri.ru, vi,snu, sna. 401\nExamplesare :rfcrd, 'praising', pakvd, 'ripe', fjfcud, 'artful', ranvd,\n'joyful', urd/iud, 'lofty': vdkva, 'twisting'; urud, 'stall', sruvd, 'spoon'; eva,\n'quick, course', dfua,'horse'.\nThewords invaexhibit only insporadic cases resolution oftheending\ninto ua.\n1191.f^\"ri.With this suffix areformed, directlyor\nwithpreceding u,asmall number ofderivatives.\nThus, forexample:fubhrf, 'beautiful', bhuri, 'abundant'; and,withun,\njdsuri, 'exhausted', ddpuri,4pious:\n,sdhuri, 'mighty'; angfiri (oranguli], 'finger'.\n1192.\"^ru.This suffix makes afewadjectives and\nneuter nouns, eitherdirectlyorwith aprecedingvowel.\nThus: dharu.-sucking', bhiru, 'timid', cdru, 'pleasant'; with preceding\na-vowel :patdru, 'flying', vandaru, 'praising', piyaru, 'scoffing', and(from\ncausative stem, with Iforr]patayalu, 'flying', sprhaydlit, 'desiring' (late);\nwith preceding e,maderu, 'rejoicing', saneru, 'obtaining', andperu (of\ndoubtful meaning).\n1193. fcfm.Bythis suffix aremade:\nTwo orthree derivatives fromreduplicatedroots :jdgrvi, 'awake', dddhrvi,\n'sustaining', didivi, 'shining'; and avery few other words;ghrsvi, 'lively',\ndhruvf,'firm'(andperhaps jvvri, 'wornout',forjirvi: BR.).\nHeremay bementioned cikitvti(RV., once), apparently made with a\nsuffix vitfrom areduplicated root-form.\n1194. Tsnu.With thissuffix, with orwithout aunion-\nvowel,aremade afewadjectivederivatives fromroots, but\nstillmore from causative stems.\na.From simpleroots:direct, jianti, 'victorious', danksnti, 'biting',\nbhusnu, 'thriving', ni-satsm'i, 'sitting down', sthasnu, 'fixed'; with union-\nvoweli,camnw, 'wandering', rocisnu, 'shining', gamisnti (TB.), 'going',\n-marisnu, 'mortal', pra-janisnti, 'generating'.\nb.From causative stems: forexample, cyavayisnti (AV.), 'settingin:\nmotion', parayisnu, 'rescuing', posayimu, 'causingtothrive', abhi-focayisnti.\n'attacking withheat', pra-janayisnu (K.), 'generating'.\nc.Itseems notunlikely that the softhis suffix isoriginally that ofa\nstem,towhich nu isadded. Such acharacter isstillapparentinkravisnv\n'craving raw flesh(ferain*)'; and also invadhasnu, 'murderous', andvrdh-\nasnu(?),'thriving'.\n1195.^sna. Extremelyfewwords have thisending.\nItisseen intlk&na, 'sharp', andperhapsinflaksnd, and-ruksna; and\nindesnd (usually trisyllabic: ddisna), 'gift'. Unless inthelast,itisnot\nfound preceded byi;but ithas(like snu, above) abefore itinvadhasnd,\n'deadly weapon', fcardsna,'fore-arm'.\nWhitney, Grammar. 26\n"
+    },
+    {
+        "page": 418,
+        "content": "402 XVII. PRIMARY DERIVATION.[1196\n1196. eTtnu.This suffix isusednearlyinthesameway\nwith^snu(above, 1194).\na.Asused with simple roots, the tisgenerally capableofbeing con-\nsidered theadscititious tafter ashortroot-final,towhich nuisthen added :\nthus, fcrfniJ, 'active', hatnti, 'deadly', -tatnu, 'stretching'; and,fromreduplic-\nated roots, jigatnu, 'hasting', andjighatnu, 'harming'; but alsodartnti,\n'bursting'. Also, with union-vowel, drauifntJ, 'running'.\nb.\"With causative stems :forexample, dravayitnu, 'hasting', madayitnu,\n'intoxicating', stanayitnti, 'thunder', -amayitnu, 'sickening'.\nC.With preceding a,inpiyatnu, 'scoffing', a-rujatnu, 'breaking into';\nandkavatnu, 'miserly' (obscure derivation).\n1197. Hsa.Thewords endinginsuffixal Hsa,with or\nwithout preceding union-vowel, areaheterogeneous group, and\ninconsiderable partofobscure derivation. Afewexamplesare :\na.With sasimply: jesd, 'winning' (aoristic ?),ruksd, 'shining'; titsa,\nn.,'fountain'; bhisd, f.,'fear'(rather from thesecondary rootbhis}.\nb.With precedingi-vowel: tavisd(f.tavisl], 'strong', mahisd(f.ma/im),\n'mighty', 6/iama(?), 'seeking booty'; manlsd, f.,'devotion'.\nC.With preceding w-vowel :arusd(f.artm), 'red', apt/sa, 'ravenous',\ntarasa, 'overcomer', ptirusa andmdnusa(-us-a?), 'man'; plyusa, 'biestings'.\n1198. 5RH asi.Afewwords intheoldest languageare\nmade with asuffix havingthisform(perhaps made bythe\naddition ofitoas}.\nThus, atasf, 'vagabond', dharnasf, 'firm', sanasi, 'winning'; anddhasf,\nm., 'drink', f.,'station'.\n1199. P7ablia.Afewnames ofanimals, forthemost\npartofobscure derivation, show thisending.\nThus, vrsabhd and rsabhd, 'bull', farabhd,acertain fabulous animal,\nferabha,acertain snake, gardabhd andrasabhd,'ass'. AV. hastheadjective\nsthulabhd, equivalenttosthuld.\n1200. a.\n35fr^a#,^V, 3rT^M^,Wf rt.Allthewords with\nthese endings were mentioned above(383d). They have trace-\nable root-connection onlyinpart:those inatareprobablyre-\nlated totheparticiplesinant.\nb.5T^ad.Thewords inadarealsogiven above(ibid.).\nc. (s{ajj SsT if,3s{ uf.Thewords with these endings were\ngivenatthesame place (tobeadded isbhisdj, 'healer', ofwhich\ntheetymologyisdisputed):trsndj appearstobeasecondary de-\nrivative, from trma,'thirst'.\n1201.Anumber ofother primarysuffixes are either set\nupbythegrammariansandsupported withexamplesofquestion-\nable value, ordoubtfullydeducible from isolated words traceable\ntoknown roots, orfrom words ofobscure connection.\n"
+    },
+    {
+        "page": 419,
+        "content": "1204]STEMS INtnu. sa, asi.abha. at,ETC. 403\nAfewsuchmaybementioned here :anda inkaranda andvdranda and\ncertain unquotable words(prakritized a-forms from thepresent participle);\neraandorainunquotable words, andelima(above, 966d:perhapsafurther\nderivative with secondary imafromera);mara (maorman with secondary\nraadded) inghasmara etc.; sara\n(inmatsara, fcara inptiskara andother\nobscure words, painpiispa and anumber ofother obscure words; and soon.\nB.SecondaryDerivatives.\n1202. Words ofsecondaryderivation aremade bythe\naddition offurther suffixes tostemsalready endinginevi-\ndent suffixes.\nBut also, aspointed outabove(1137 8),topronominal\nroots, and toverbalprefixes andafewother indeclinable words.\n1203. Changes ofthestem. Thestem towhich\nthesuffix isadded isliable tocertainchangesofform.\na.Before asuffix beginning with avowel orwithy(which\ninthisrespectistreated asifitwerei],final aand/-vowels\nareregularlylost altogether, while afinalw-vowel hastheguna-\nstrengthening andbecomes av;randoandau(allofrareoccur-\nrence)aretreated inaccordance with usualeuphonicrule.\nAnw-vowel alsosometimes remains unstrengthened.\nb.Afinal nisvariously treated, being sometimes retained\nandsometimes lost, even along with apreceding a;andsome-\ntimes anaislost, while thenremains :thus, vrsanvant, vrsana,\nvrsa, vrsatva, vrsnya, from vrsan. Ofastem endinginant, the\nweak form, inat,isregularly taken :thus, vaivasvata(vivasvant).\nOther alterations ofafinal aresporadic only.\n1204. Themostfrequent changeinsecondaryderivation\nisthevrddhi-sti'eiigtheningofaninitialsyllableofthestem\ntowhich asuffix isadded.\nThestrengthened syllable maybeofanycharacter :radic-\nal,ofaprefix,orofthe firstmember ofacompound:thus,\nacmna(aqvin), saumyd (soma), parthiva (prthivi),amitrd (amitra),\nsamrajya (samraj), saukrtya (sukrtd), maifravarund(mitravdruna),\nauccaihcravasd (uccaihcravas). As totheaccompanying accent,\nseethenextparagraph.\na.Ifastem begins with aconsonant followed byyorv,thesemi-\nvowel isusually vriddhied,asifitwere ioru,and theresultingaiorau\nhasyorvfurther added before thesucceedingvowel.\nThis ismost frequent where theyorvbelongstoaprefixasnz,\nuijSu altered before afollowinginitial vowel: thus, naiyayika from\n26*\n"
+    },
+    {
+        "page": 420,
+        "content": "404 XVII. SECONDARY DERIVATION.[1204\nnyaya (asifniydyaj, vaiya?vd from vyfyva (asifvfyacva], sauva^vya from\nsvdfva (asifsuvafva); but itoccurs also inothercases,assauvard from\nsvdra, fauva frompt>an, against svayambhuva (svayambhu), and soon.\nAV. hasirregularly kaverakd from kuvera(asiffrom kvera, without the\neuphonic yinserted).\nb.This strengthening takes place especially, andvery often,before the\nsuffixes aandya ;also regularly beforei,ayana (with ayani andkayani],\neya(with ineya], and laterlya;before thecompound okaandika,and later\naki;and, insingle sporadic examples, beforeno,ena,ma, ra,andfrua(?):\nseethese various suffixes below.\nC.Inafewexceptional cases, bothmembers ofacompound word have\nthe initial-ufddTw-strengthening:thus,forexample, saumapausnd (VS.:\nsomapusdn), kaurupdncala (QB., GB.: kurupancald), caturvaidya (caturveda),\ndihalaukika(ihaloka). Again, thesecond member ofacompound instead of\nthe first isoccasionally thusstrengthened:forexample, $atd?arada (RV., AV.),\npancafdradiya (TB. etc.), 'somardudra(TS.), trisahasri, da$asahasra, purva-\nvaraika(not quotable), caturadhyaya, gurulaghava.\nThe0tma-strengthening (except ofafinal tt-vowel: above, 1203 a)is\nnowhere anaccompanimentofsecondary derivation :foranapparent exception\nortwo, see1209h,i.\n1205. Accent. Thederivatives with initial vrddhi-strength-\nening always have their accent oneither the first orthe last\nsyllable. Andusually,itislaid, asbetween these two situa-\ntions, insuch awayastobefurthert removed from theaccent\noftheprimitive; yet,notrarely,itismerely drawn downupon\nthesuffix from the final ofthelatter; much less often,itre-\nmains upon aninitialsyllable without change. Onlyinthecase\nofoneortwo suffixes isthedistinction between initial and final\naccent connected withanydifference inthemeaning anduseof\nthederivatives(seebelow, suffixeya: 1216).\nNoother general rules astoaccent canbegiven. Usually\nthesuffix takes thetone, orelse thisremains where itwas in\ntheprimitive;quite rarely,itisthrown back tothe initialsyl-\nlable(asinderivation with initialvrddhi);andinasinglecase\n(ta:1237),itisdrawn down tothesyllable precedingthesuffix.\n1206. Meaning. Thegreat mass ofsecondarysuffixes\nareadjective-making:theyform from nounsadjectivesindicat-\ningappurtenanceorrelation, ofthemost indefinite andvaried\ncharacter. But, asamatter ofcourse, this indefiniteness often\nundergoes specialization:so,particularly,intodesignationofpro-\ncedure ordescent, sothat distinctivepatronymic andmetronym-\nicandgentile words arethe result\n;or,again,into thede-\nsignation ofpossession. Moreover, while themasculines and\nfeminines ofsuchadjectivesareemployedasappellatives,the\nneuter isalsowidely used asanabstract, denotingthequality\n"
+    },
+    {
+        "page": 421,
+        "content": "1208! STEMS INa. 405\nexpressed attributively bytheadjective:andneuter abstracts are\nwith thesame suffixes made fromadjectives. There arealso\nspecialsuffixes'very few)bywhich abstracts aremadedirectly,\nfromadjectiveornoun.\nAfew suffixes make nochangeinthepartofspeechof\ntheprimitive,buteither changeitsdegree (diminution andcom-\nparison),ormake other modifications, orleave itsmeaning not\nsensiblyaltered.\n1207. The suffixes willbetaken upbelow inthefollow-\ningorder. First, thegeneral adjective-making suffixes, begin-\nning with those ofmost frequent use(a,yaand itsconnections,\n,#a); then, those ofspecific possessive value(in.vantandmant,\nand theirconnections);then, theabstract-making ones(taand\ntva,andtheirconnections); then, thesuffixes ofcomparison etc.;\nandfinally, those bywhich derivatives aremade only oralmost\nonlyfromparticles.\n1208. 5fa.With thissuffix aremade avery largeclass\nofderivatives, fromnouns orfromadjectives havinganoun-\nvalue. Such derivatives areprimarily andespecially adject-\nives, denoting 'havingarelation orconnection(ofthemost\nvariouskind)with'that]denoted bythemoreprimitiveword.\nButtheyarealsofreelyusedsubstantively;themasculine\nandfeminine asappellatives,theneuter, especiallyandfre-\nquently,asabstract. Oftentheyhave apatronymicorgent-\nilevalue.\nTheregular andgreatly prevailingformation isthatwhich\nisaccompaniedwithvrdtf^i-strengtheningofthe firstsyllableof\ntheprimitive word, simpleorcompound. Examplesofthis\nformation are :\na.Fromprimitives ending inconsonants :with theusual shift ofaccent,\nayasd,'ofmetal'(dyas], manasd, 'relatingtothemind'(mdnas), saumanasd,\n'friendliness' (sumdnas], brahmand, 'priest' (brdhmari), haimavatd, 'from the\nHimalaya' (himdvant), angirasa,'oftheAngiras family' (dngiras}; hastina,\n'elephantine' (hastfn], mdruta, 'pertainingtotheMaruts'(martit);with\naccent thrown forward from the finaluponthe suffix, farada, 'autumnal',\nvairajd, 'relating totheviraf, pausnd, 'belongingtoPushan'; gairiksitd,\n'son ofGirikshit': with accent unchanged, mdnusa, 'descendant ofManus'.\nThe suffix isadded (asabove instanced)tothemiddle stem-form of\nstems invant;itisadded totheweakest inmdghona nndvartraghna; the\nending inremains unchanged; anusually does thesame, butsometimes loses\n"
+    },
+    {
+        "page": 422,
+        "content": "406 XVII. SECONDARY DERIVATION.[1208\nitsa,asinpdusnd, trdivrmd, dd^ardjnd ;andsometimes itsn,asinbrdhmd,\nduksd, bdrhatsdma.\nb.From primitivesinr:'jattra, 'victorious'(jetf orjetr, 'conqueror'),\ntvastrd, 'relating toTvashtar', savitrd, 'descendant ofthesun'(savitf).\nc.From primitives inu:usually with puna-strengthening ofthew,as\nvasavd, 'relatingtotheVasus', artavd, 'concerning theseasons'(f*w), ddnavd,\n'child ofDanu'(cZanw), saindhavd,'from theIndus'(sfndhu}; butsome-\ntimes without,asmadhva,'full ofsweets' (mddhu\\ pdr^vd,'side'(pdr$u,\n'rib'), paidvd, 'belongingtoPedu', tdnva,'ofthebody' (tanU}.\nd.From primitives in iand*,which vowels aresupplanted bythe\nadded suffix: pdrthiva, 'earthly' (prthivi), [sdrasvatd,'oftheSarasvati',\naindragnd, 'belongingtoIndra and Agni' (indrdgni); pdnkta,'five-fold'\n(panfcif), nairrtd, 'belongingtoNfrrtC: butdvayd, 'barrenness'(iffroma-vl}.\ne.From primitivesina,which inlikemannerdisappears: yamund,\n'oftheYamuna', saraghd, 'honey' etc.(sardgha, 'bee'), kanind, 'natural child'\n(kanina, 'girl').\nf.Alargenumber (nearly asmany asalltheresttogether) fromprim-\nitives ina,ofwhich the final isreplaced bythesuffix :forexample, with\ntheusual shift ofaccent, amitrd, 'inimical' (amitra, 'enemy'), vdrund,'of\nVaruna', vaifvadevd, 'belonging toallthegods' (vifvddeva), ndirhastd, 'hand-\nlessness' (ntrhasla), vaiya^vd, 'descendant ofVyacva'; gdrdabha, 'asinine'\n(garddbhd), dafva,'divine'(devd), mddhyandina,'meridional' (madhydndina],\npautra, 'grandchild' (putrd, 'son'), satibhaga, 'good fortune'(subhdgd), vddhry-\nafva,'ofVadhryac,va's race'; with unchanged accent (comparatively few),\nvasantd, 'vernal' (vasantd, 'spring'), maitrd, 'Mitra's', atithigvd,'ofAtithigva's\nrace', dafooddsa, 'Divodasa's'.\nThederivatives ofthisform aresometimes regarded asmade byinternal\nchange, without added suffix. Considering, however, that other final vowels\naresupplanted bythissuffix, thatadisappearsasstem-final also before\nvarious other suffixes ofsecondary derivation, andthatnoexamplesofderi-\nvation without suffix arequotable from primitivesofanyother final than a,\nitseems fartooviolent toassume here adeviation from thewhole course of\nIndo-European word-making.\n1209.Thederivatives madebyadding5fawithout vrddhi-\nchangeofthe initialsyllablearenotnumerous, andarein\nconsiderablepart^ doubtless, ofinorganic make, results of\nthetransfer toana-declension ofwords ofother finals.\na.Anumber ofexamples ofstems inamade bytransfer were noticed\nabove (399). The cases ofsuch transition occur most frequentlyincom-\nposition: thus, further, apa- (forapordp, 'water'), -rca, -nara, -patha,\n-gava, -diva; from stems inan,-adhva, -astha, -aha, -vrsa, but also-a/ma\nand-vrsna andvrsana; from stems ini,-angula and -rdtra; from the\nweakest forms ofanc-stems (407) wcca, nicd, pardcd.\nb.Also occurring especiallyincomposition, yetlikewise assimple words\n"
+    },
+    {
+        "page": 423,
+        "content": "1210]STEMS INa\n1ya. 407\noften enoughtohave anindependent aspect,arederivatives inafromnorms\ninas(rarely is,us}: thus, forexample, tamasd, rajasd, payasd, brahmavar-\ncasd, sarvavedasd, devdinasd, parusd, tryayusd.\nc.Similar derivatives from adjectives ininarereckoned bythegram-\nmarians asmade with the suffix ina: thus, malina, 'polluted', frngina,\n'horned'(notquotable). TheonlyVedic instance noted isparamesthtna (AV.).\nd.From emc-stems (407) aremade afewnouns inka: thus, dnuka,\ndpaka, upaka, prdfika, parakd,etc.\ne.From stems inr,nestrd, potrd, prafastrd, from titles ofpriests.\nf.Other scattering cases are :savidyutd, avyusd, virudha, kdkuda,\nkakubhd.\ng.TheVedic gerundives intva(tua) havebeen already (966 a)pointed\noutasmade byaddition ofatoabstract noun-stems intu.\nh.Trayd anddvayd come with grwna-strengthening fromnumeral stems;\nndva, 'new', perhapsinlikemanner fromnti,'now'; anddntara from antdr(?).\ni.Bhesaid, 'medicine',isfrombhisdj, 'healer', with ^wna-change; and\nperhaps devd, 'heavenly, divine, god', inlikemanner from div(compare\navayd, above, 1208d).\nOne ortwo cases havebeen noticed above, inwhich theaddition ofa\ntoanother suffix hasmade aseemingly independent suffix.\n1210. Tyet\"With this suffix aremade avery large\nclass ofwords, hoth intheoldlanguageand later.\nThe derivatives inyaexhibit agreat andperplexing variety ofform,\nconnection, and application; and therelations ofthe suffix toothers con-\ntaining at/a-element iya, lya, eya, ayya, eyya, enyaarealsoinpart\nobscure and difficult. Inthegreat majority ofinstances intheoldest lan-\nguage, theyawhen itfollows aconsonant isdissyllabic inmetrical value,\noristoberead asia.Thus,inRV., 266words (excluding compounds)\nhaveia,and only75haveyaalways ;46are tobereadnowwith iaand\nnowwithya,butmany ofthese haveyaonly inisolated cases. Asmight\nbeexpected,thevalue iaismore frequentafter aheavy syllable:thus, in\nRV., there are188examplesofiaand27ofyaafter such asyllable, and\n78ofiaand96ofyaafter alight syllable (thecircumfiexed yathat is\ntosay, ia being,aswillbepointed outbelow, more liable tothereso-\nlution thanyaoryd).Itmust beleft forfurther researches todecide whether\nintheyaarenotincluded more than onesuffix, \"with different accent, and\ndifferent quantityofthei-element; orwith anaadded toafinal iofthe\nprimitive.Itisalsomatter forquestion whether there isaprimaryaswell\nasasecondary suffix ya ;thesuffix atleast comes tobeused asifprimary,\nintheformation ofgerundives:but itisquite impossibletoseparatethe\nderivatives intotwosuch classes, and ithasseemed preferabletherefore to\ntreatthem alltogether here.\nThe derivatives made withyamaybefirstdivided intothose\nwhich doandthose which donotshow anaccompanyingvrddhi-\nincrement ofthe initialsyllable.\n"
+    },
+    {
+        "page": 424,
+        "content": "408 XVII. SECONDARY DERIVATION.[1211\n1211. Derivatives in7Jyawith initialvrddhi-stiengthen-\ningfollowquite closely,informandmeaning,theanalogy\nofthose in fa(above, 1208). They are,however, decidedly\nlesscommon than thelatter(inVeda, about three fifths asmany).\nExamples are :with theusual shift ofaccent, ddivya,'divine'(devd),\npdlitya, 'grayness' fpalitd), grdtvya,'cervical' (grivd), drtvijya, 'priestlyof-\nfice' (rMj), gdrhapatya, 'householder's' (grhdpati), janardjya, 'kingship'\n(janaTaj), sarhgrdmajitya, 'victory inbattle' (sarhgrdmajti), sduvafvya, 'wealth\ninhorses' (svdpva), dvpadrastrya,'witness' (upadrastr); ddityd, 'Aditya' (dditi),\nsdumyd, 'relating toso-met1\n,dtithyd, 'hospitality' (dtithi), prdjdpatyd, 'belonging\ntoPrajapati', vdimanasyd,'mindlessness' (vimanas), sdhadevya, 'descendant\nofSahadeva'; with accent thrown forward from the finalupon theending,\nIdukyd,'oftheworld' (lokd), kdvyd,'oftheKavi-race', drtvyd, 'descendant\nofRitu', vayavyd, 'belongingtothewind'(vayu), rdivatyd, 'wealth' (revdnt):\nwithunchanged accent (very few), ddhipatya, 'lordship' (ddhipati), frdfsthya,\n'excellence' (frestha), vdfyya, 'belongingtothethird class' (vff, 'people'),\npatiihsya,'manliness' (ptims).\nTheAV. hasonce ndirbddhya,with circumflexed final4ifnotanerror,\nitisdoubtless made through ndirbddha; vdisnavydii (VS.i.12)appears tobe\ndual fern, ofvdisnavi.\n1212. Derivatives in Tyawithout initialvrddhi-stiQiigth-\neningareusually adjectives, much lessoften (neuter, or,\nin TTya,feminine)abstract nouns. Theyaremade from\nevery varietyofprimitive,and areverynumerous(inVeda,\nthree orfour times asmanyasthepreceding class).\nThegeneral mass ofthese words maybebestdivided accord-\ningtotheir accent, into :a.Wordsretainingtheaccent ofthe\nprimitive;b.Words with retracted accent;c.Words with acute\nyd(id);d.Words with circumflexedya(ia). Finally maybe\nconsidered thewords, gerundives andaction-nouns, which have\ntheaspectofprimaryderivatives.\na.Examplesofderivatives inyaretaining theaccent oftheir primitives\nare:d$vyct, 'equine' (dfva), dngya,'ofthelimbs' (dnga), mtikhya,'foremost'\n(mukha, 'mouth'), dvya,'ovine' (dm), gdvya,'bovine' (g6), vfyya,'ofthe\npeople' (vfy), dtiryct,'ofthedoor' (dur), ndrya, 'manly' (nf), vfsnya,'virile'\n(vrsan), svardjya, 'autocracy' (svardj), suvirya,'wealth inretainers' (suvTra),\nvifvddevya,'ofallthegods' (vi$vddeva), mayurctfepya, 'peacock-tailed',\nInthe lastword, andinafewothers, theyaappearstobeused(like\nka,1222 C,2;ya,1212 d,5)asasuffix simply helpingtomake apossessive\ncompound:sosuhdstya (besidetheequivalent suhdsta), mddhuhastya, ddfa-\nmasya, mifrddhdnya.\n"
+    },
+    {
+        "page": 425,
+        "content": "1212] STEMS INya. 409\nb.Examples with retraction oftheaccent tothe first syllable (asin\nderivation withvrddhi-increment] are: kdnthya,,'guttural' (kanthd), skdndhya,\n'humeral' (skandhd), vrdtya,'ofaceremony' (vratd), meghya,'intheclouds'\n(meghd), pftrya,'oftheFathers'(pitr), prdtijanya, 'adverse'(pratijand).\nHiranyaya,'ofgold' (hfranya),isanomalous both indrawing theaccent\nforward, andinretaining the final aoftheprimitive.\nc.Examples with acute accent onthe suffix are:divyd, 'heavenly1\n(div), satyd, 'true'(sdnt), vyaghryd, 'tigrine' (vydghrd), kavyd, 'wise' (kavf);\ngramyd,'ofthevillage' (grdma), somyd, 'relating tothesoma,bhumyd,\n'earthly' (bh&mi), sakhyd, friendship' (sdkhi), jaspatyd, 'headshipofafamily'\n(jdspati).\nd.Ofderivatives endingincircumflexed ya(whichintheVeda are\nconsiderably more numerous than allthethree preceding classestogether),\nexamplesareasfollows :\n1.From consonant-stems:vi$yh,'ofthe clan' (RV.: vfp), hrdya, 'ofthe\nheart'(hid), vidyutya,'ofthelightning' (vidytit), rdjanya,'oftheroyal class'\n(rdjan), dosanyh,'ofthearm' (dosdn), firsanya,'ofthe head' (flrsdnj,\nkarmanya,'active' (kdrman), dhanvanya,'oftheplain' (dhdnvan), namasya,\n'reverend'(ndmas), tvacasytt,'cuticular' (tvdcas), barhisya,'ofbarhts1\n,ayusyb,\n'givinglife'(dyus), bhasadytt,'ofthebuttocks' (bhasdd), prdcyh, 'eastern'\n(prdnc),etc. Ofexceptional formation isaryamya'intimate'(aryamdn).\n2.From w-stems :hanavyci,'ofthejaws' (hdnu), vayavyh, 'belonging\ntoVayu', pa?avya, 'relatingtocattle' (pw&), isavya, 'relatingtoarrows' (isu);\nfaravyh, f.,'arrow'(fdru, do.); and there may beaddednauya, 'navigable'\n(especiallyinfern., naut/3, 'navigable stream': nat/, 'boat').\nUnder thishead belong, aswaspointed outabove(chap. XIV., 964).\ntheso-called gerundives intavya (later tdvya],asmade bytheaddition of\nyhtotheinfinitive noun intu.They arewholly wanting intheoldest lan-\nguage, andhardly found inlater Vedic, althoughstill later tavya wins the\nvalue ofaprimary suffix, andmakes numerous derivatives. TheRV. has\nprdfavyh,'tobepartakenof(pra -\\-Yaf),without anycorresponding noun\nprdfu; and alsourjavyb,'rich innourishment' (urj), without anyintermedi-\nateurju.\n3.From iandz-stems hardly anyexamplesaretobequoted. VS.has\ndundubhyafrom dundubhi.\n4.From a-stems :svargyh, 'heavenly' (svargd), devatyh, 'relatingtoa\ndeity' (devdtd), prapathyft, 'guiding' (prapatha), budhnyfr, 'fundamental'\n(budhnd), jayhanyh,'hindmost' (jaghdna), varunyti, 'Varuna's', viryh, 'might'\n(vir&), udaryh,'abdominal'(uddra), utsyh,'ofthefountain' (6tsa); andfrom\na-stems, urvaryct,'ofcultivated land' (urvdrd), svahya, 'relatingtotheex-\nclamation svdha1\n.\nThecircumflexed yhismore generally resolved (into fa)than theother\nforms ofthesuffix :thus,inRV. itisnever toberead asyaafter aheavy\nsyllable ending with aconsonant;andeven after alight one itbecomes fa\ninmore than three quartersoftheexamples.\n5.There areafewcases inwhich yhappearstobeused tohelpmake\n"
+    },
+    {
+        "page": 426,
+        "content": "410 XVII. SECONDARY DERIVATION.[1212\nacompoundwithgoverning preposition (next chapter, 1310): thus, apikaksyft,\n'about thearm-pit', upapaksya, 'upon thesides', udapya, 'upstream'; arid\nperhaps upatrnyh, 'lying inthegrass' (occurs only invoc.). But,with other\naccent, dnvdntrya, 'through theentrails', tipamasya,'oneach month'; abhi-\nnabhyd, 'uptothe clouds', ddhigartya,'onthechariot floor'.\n1213. The derivatives inTJyaastowhich itmaybe\nquestioned whethertheyarenot, aleast inpart, primary\nderivatives from thebeginning,areespeciallythegerund-\nives, togetherwith action-nouns coincident with these in\nform;inthelaterlanguage,thegerundive-formation (above,\n963)comes tobepracticallyaprimaryone.\na.InRV. occur about 40instances ofgerundivesinya,oftolerably\naccordant form :therootusually unstrengthened (but cetya, bhdvya, -hdvya,\nmdrjya, y6dhya;also-mddya, -vdcya, bhavyd);theaccent onthe radical\nsyllable when theword issimple,orcompounded withprepositions: thus,\nprafdsya, upasddya. vihdvya (but usually onthe final after thenegative pre-\nfix :thus, anapyd, anapavrjyd) exceptionsareonlybhavyd andthedoubt-\nfulakayya ;theyaresolved into iainthevery great majority ofoccurrences;\nafinal short vowel followed byt(in-ttya, -krtya, -tr&tya, -stutya, andthe\nreduplicated carkrtya, beside carkrti :notinndvya and-havya), andachang-\nedtoe(in-deya only).Ifregarded assecondary, theymight bemade with\nya,inaccordance with other formations bythissuffix, inpartfrom theroot-\nnoun,asanukft-ya,inpartfrom derivatives ina,asbhavyd (from bhava).\nb.TheAV. hasasomewhat smaller number (about 25)ofwords ofa\nlikeformation;but also aconsiderable group (fifteen)ofderivatives inya\nwith thesame value :thus,forexample, adya, 'eatable', karya,tobedone',\nsamapya,'tobeobtained', atitarya,'tobeoverpassed', nivibharya,'tobe\ncarried intheapron', prathamavasya,'tobefirst worn'. These seem more\nmarkedly ofsecondary origin:and especially such forms asparivargya,'to\nbeavoided', avimokyd,'not tobegottenridof,where theguttural rever-\nsion clearly indicates primitivesingaand fca(216.6).\nc.Throughout theolder language areofcommon occurrence neuter ab-\nstract nouns ofthesamemake with theformer ofthese classes. They are\nrarely found exceptincomposition (inAV., only cttya andsteya assimple),\nand areoften used inthe dative,after themanner ofadative infinitive.\nExamples are :brahmajyeya, vasudeya, bhagadheya, purvapeya, fataseya,\nabhibh&ya, devah&ya,\\ mantracrtitya, karmakftya, vrtraturya, hotrvurya,ahi-\nhdtya, sattrasddya, firsabhidya, brahmacdrya, nrsdhya. Ofexceptional form\narertddya (y~vad)andsaha$eyya (]/p);ofexceptional accent, sadhdstutya.\nAndAV. hasoneexample, ranya, with circumflexed final.\nd.Closely akin with these,inmeaning anduse,isasmall class offem-\ninines inyd: thus, krtyd, vidyd, ityd, mustihatyd, devayajyd, andafew\nothers.\n"
+    },
+    {
+        "page": 427,
+        "content": "1216 STEMS INya, iya, lya, eya.411\ne.There remain, ofcourse, aconsiderable number ofless classifiable\nwords, bothnouns andadjectives, ofwhich afewfrom theolder language\nmaybementioned, without discussion oftheir relations: thus, s&rya (with\nfern.suryfj}\n,ajya, pusya, ndbhya ;yujya, gfdhya, frya, arydanddrya, mdrya,\nmadhya.\nThe suffixesapparently mostneatlyakin withyamaybest\nbenext takenup.\n1214. <ftiya.This suffix isvirtuallyidentical with\nthepreceding, beingbutanother written form ofthesame\nthing.Itisusedonlyafter twoconsonants, where the\ndirect addition ofyawould create acombination ofdifficult\nutterance.\nIthasthesamevarietyofaccent withya.Thus :\na.With accentlya(=faorya):forexample, abhrfya,'from theclouds'\n(abhrd), ksatrtya, 'having authority' (ksatrd), yajntya, 'reverend'(yajnd),ho-\ntrfya,'libational' (lititra).\nb.With accentiyd(=yd):forexample, agriyd, 'foremost' (dgra),in-\ndriyd,'Indra's'(later,'sense': fndra), ksetriyd,'ofthe field' (ksetra).\nC.With accent ontheprimitive:$r6triya,'learned' (frotra),\n1215. ??I ~iya-This suffix also isapparently byoriginaya\n(m)ofwhich the firstelement hasmaintained itslong quantity\nbytheinterpositionofaeuphonic y.Itisaccented always on\nthe I.\na.InRV.occur,ofgeneral adjectives, only arjikiya andgrhamedhiya,\nandexamplesinthelaterVedic areveryfew: e.g.parvatiya,'mountain-\nous' (AV., beside RV.parvatyh}.IntheBrahmanas,anumber ofadject-\nivesfrom phrases (first words ofverses andthelike) aremade iniya: thus,\nkaya$ubhiya, svaduskiliya ,apohisthiya,etc.\nb.Itwaspointedoutabove (965)that derivative adjectives inlya\nfrom action-nouns inanabegin inlaterVeda andinBrahmana tobeused\ngerundivally, and arearecognized formation asgerundivesinthe classical\nlanguage.\nc.Derivatives inlyawith initial vrddhi aresometimes made inthe\nlater language:e.g.svasrlya, parvatlya.\nd.Thepronominal possessives madiyaetc.(516) donotoccur either\ninVeda orBrahmana; buttheordinalsdvitiyaetc.(487:with fractional\ntrtlya andturlya: 488),arefound from theearliest period.\n1216.^TTeya.With thissuffix, alongwith vrddhi-\\n-\ncrement ofaninitialsyllable,aremadeadjectives,often\nhavingapatronymicormetronymicvalue. Their neuter\nissometimes used asabstract noun.\n"
+    },
+    {
+        "page": 428,
+        "content": "412 XVII. SECONDARY DERIVATION.[1216\nThe accent rests usually onthefinal inadjectivesofdescent,\nandonthe firstsyllableinothers.\na.Thus, arseya, 'descendant ofasage' (rsi), sarameyd,'ofSarama's\nrace', catavaneyd, 'Qatavani's descendant', rathajiteyd, 'son ofRathajit'; ds-\nnet/a,'oftheblood' (asdn), vdsteya,'ofthebladder' (vastf), pauruseya, 'com-\ningfrom man'(purusa), etc.\nAmore than usualproportion ofderivatives ineyacome from primi-\ntives ini;andpossiblythesuffix firstgained itsform byaddition ofyato\nagunated i,though afterward used independently.\nThegerundiveetc. derivatives inya(above, 1213) from a-roots endin\neya; and, besides such, RV. etc.have sabheya from sabhd, anddidrkseya,\n'worth seeing', apparently from the desiderative noun didrksd, after their\nanalogy.\nb.Derivatives intheso-called suffix ineyd assaubhagineyd, kalyan-\nineyd (neither quotable) aredoubtless made upon proximate derivatives in\n-ini(fern.).\nC.Ineyya (i.e.eyia) end, besides theneuter abstractsahafeyya (above.\n1213c),theadjectiveofgerundival meaning stuseyya (withaoristic sadded\ntotheroot), andfapatheyya, 'curse-bringing' (or'accursed'), from capdtha.\n1217.^Tenya.This suffix isdoubtlesssecondaryin\norigin, made bytheaddition ofTJyatoderivatives inana-\nsuffix; but,likeothers oflikeorigin,itisappliedinsome\nmeasureindependently, chieflyintheolderlanguage, where\nithasnearlythevalue ofthe lateraniya (above,1215b),\nasmaking gerundival adjectives.\nTheyofthis suffix isalmost always toberead asvowel, andtheaccent\nis(exceptinvdrenya) onthee:thus, -enia.\nThegerundives have been allgiven above, under thedifferent conjuga-\ntions towhich they attach themselves (1019, 1038, 1068); except Idenya,\n'praiseworthy1\n,andvdrenya, 'desirable', they areofonly isolated occurrence\n(once ortwiceeach). TheRV.hasalsotwonon-gerundival adjectives, virenya,\n'manly' (vird), andkirtenya, 'famous'(kirtij, andTS. hasanabhicastenyd\n(abhicasti); vijenya (RV.)isaword ofdoubtful connections.\n1218.5TW <*>yya- With this suffix aremade gerundival\nadjectives,almost only inRV. They have been noticed above\n(966 c).Theendingiseverywheretoberead ashyia.\nAfew adjectives without gerundival value, andneuterabstracts,also\noccur :thus, bahupdyya, 'protecting many', nrpdyya, 'men-guarding'; kunda-\npdyya andpurumdyya, proper names; purvapdyya,'first drink', mahaydyya,\n'enjoyment'; andrasdyya, 'nervous', anduttamdyya, 'summit', contain no\nverbal root. Aldyyaisdoubtful;alsoakayya, which itsaccent refers toa\ndifferent formation, along with prahdyya (AV.: }//'), 'messenger', andpra-\nvayya (AV.),ofdoubtful value.\n"
+    },
+    {
+        "page": 429,
+        "content": "1222]STEMS INeya :enya, ayya, ayana, ayt, i,ka. 413\n1219. mVF\\ ayana. IntheBrahmanas andlater, patro-\nnymics made bythis suffix arenot rare. They come from\nstems in5fa,andhavet'n&^-strengtheningofthe first\nsyllable, andaccent onthe final.\nInRV., theonlyexample ofthisformation iskanvayana (voc.: kdnva};\nAV. hasinmetrical parts daksayand and the fern,ramayam ;andamus-\nyayana, 'son ofso-and-so'(516),initsprose. TheRV.name uksanydyana\nisofadifferent make, elsewhere unknown.\n1220. tJltfi ayi.Only one ortwowords aremade with\nthissuffix, namely agnhyl (agnij, 'Agni's wife', andvrsakapayt,\n'wife ofVrishakapi'.\nThey seem tobefeminines ofaderivative inamade with vrddhi-\nincrement ofthe final ioftheprimitive.\n1221. 5t.Derivatives made with this suffix arepatro-\nnymicsfromnouns ina.The accent rests onthe initial\nsyllable, which hasthe#rcfo^-strengthening.\na.InRV. arefound half-a-dozen patronymics ini:forexample, dg-\nnivefi, patirukutsi, prdtardani, sdmvarani; AV. hasbutone, prdhradi; in\ntheBrahmanas they aremorecommon: thus,inAB.,sauyavasi, janarhtapi,\narum, janafei,etc.Asingle word ofother valuesarathi,'charioteer'\n(sardtham)isfound fromRV.down.\nb.Thewords made with theso-called suffix aki asvaiyasaki,'des-\ncendant ofVyasa',aredoubtlessproperlyderivatives in ifrom others inka\noroka.\nThat thesecondary suffix ika isprobably made byaddition ofkatoa\nderivative in iispointed outbelow (1222 e.\nc.RV. hastdpusi, apparently from tdpus with asecondaryiadded.\n1222. Sfika.This isdoubtlessoriginallyone ofthe\nclass ofsuffixes forming adjectives ofappurtenance. And\nthatvalue itstillhas inactual use; yetonlyinasmall\nminorityofoccurrences. Ithasbeen, ontheonehand,\nspecializedintoanelementformingdiminutives;and, on\ntheother hand, andmuch morewidely,attenuated intoan\nelement without definable value, added toagreatmany\nnouns andadjectivestomake others ofthesame meaning\nthis lastis,even intheVeda, and stillmore inthelater\nlanguage,itschief office.\nHence, kaeasilyassociates itself with thefinals ofderiv-\n"
+    },
+    {
+        "page": 430,
+        "content": "414 XVII. SECONDARY DERIVATION.[1222\natives towhich itisattached, andcomes toseem along with\nthem anintegral suffix, and isfurther used assuch. Ofthis\noriginaredoubtless, aswasseen above(1180, 1181),theso-\ncalled primarysuffixes ukaandaka;andlikewise thesecondary\nsuffix ika(below, e).\nTheaccent ofderivatives inkavariesapparentlywith-\noutrule, save that thewords mostplainlyofdiminutive char-\nacter have thetoneusually onthe suffix.\na.Examples (fromtheolderlanguage)ofwords inwhich thesuffix has\nanadjective-making value are :dntaka (dnta), 'end-making',bdlhika (bdlhi),\n'ofBalkh', andika(andd), 'egg-bearing', sudka (suci), 'stinging', urvarukd,\n'fruit ofthegourd' (urvaru), paryayikd (paryayd), 'strophic'; from numerals,\nekakd, dvakd, trikd, dstaka; trtiyaka,'ofthethirdday'; from pronoun-stems,\nasmdka, 'ours', yusmdka, 'yours', mdmaka,'mine'(516;;fromprepositions,\ndntika, 'near', dnwfca, 'following' (later, ddhika, utka, dvaka)-, and,with accent\nretracted tothe initial syllable (besides dstaka andtftiyaka, already given),\nrupaka (rupd),'withform', bdbhruka(babhrii, 'brown'), acertain lizard.\nb.Ofwords inwhich adiminutive meaningismore orlessprobable:\nafvakd, 'nag', kaninaka andkumarakd, 'boy', kanmakd orkaninika, 'girl',\npadakd,'little foot', putrakd,'little son', rajakd, 'princeling', fakuntakd,\n'birdling'. Sometimes acontemptuous meaning isconveyed bysuch adimin-\nutive: forformations with thisvalue from pronominal stems, seeabove,\n521;another exampleisanyakd (RV.).\nThediminutives inkahave thegenderoftheirprimitives.\nc.The derivatives inkawithunchanged meaning aremade from prim-\nitives ofevery variety ofform, simple andcompound, andhave thesame\nvariety ofaccent astheadjective derivatives (with which they areatbottom\nidentical). Thus :\n1.From simple nouns andadjectives: dstaka, 'home', ndsika, 'nostril',\nmdksikd, 'fly', avikd, 'ewe', isukd, 'arrow', durakd, 'distant', sarvakd, 'all',\ndhenuka (dhenu), 'cow', ndgnaka (nagnd), 'naked', bdddhaka (baddhd), 'cap-\ntive', vamrakd, 'ant', arbhakd, 'small', pifwfcd, 'young', amyaska, 'finer',\nejatkd, 'trembling', patayimukd, 'flying'.\nSuch derivatives inthe later language areinnumerable: from almost\nanygiven noun oradjective maybemade anequivalent, ending inkaorka\n(accordingtothegender).\n2.From compound primitives: svalpakd, 'very small', vfmanyuka,\n'removing wrath', viksinatkd, 'destroying', pravartamanakd, 'moving forward',\nviksinakd, 'destroyed'.\nIntheBrahmanas andlater,feaisoften added toapossessive adjective\ncompound ^1307), sometimes redundantly, butusually inorder toobtain a\nmore manageable stem forinflection :thus, anaksika, 'eyeless', atvdkka,\n'skinless', aretdska, 'withoutseed', vyasthaka, 'boneless', safiraska, 'along\nwith thehead', ekagayatrika, 'containing asingle gayatrl-versQ\\ grhltd-\nvasativarika, 'onewho hastaken yesterday's water', sapatriika,'with his\nspouse', abhinavavayaska,'ofyouthful age', angusthamatraka,'ofthumb size'.\n"
+    },
+    {
+        "page": 431,
+        "content": "1223] STEMS INka,ana, ma, ena. ETC. 415\nd.Thevowel bywhich thekaispreceded hasoften anirregular char-\nacter; andespecially, afeminine inika iscommon beside amasculine in\naka(asisthecasewith theso-called primary aka :above, 1181).\nInRV. arefound beside oneanother only iyattakd andiyattikd ;but\nAV. hasseveralexamples, andtheybecome much more numerous later.\ne.Two suffixes made upofkaand apreceding vowelnamely, aka\nand ika aregiven bythegrammarians asindependent secondary suffixes,\nrequiring initialvrddhi-stiengihemng oftheprimitive. Both ofthem are\ndoubtless inreality made byaddition ofkatoafinal iora,though com-\ningtobeusedindependently.\n1.Ofvrddhi- derivatives inakanoexamples have been noted from the\nolderlanguage (unless mamaka, 'mine', istobesoregarded); andthey are\nnotcommon inthelater: thus, aumaka(not quotable), 'flaxen', mdnojnaka\n(not quotable), 'attractiveness', ramaniyaka, 'delightfulness'.\n2.Ofwddfa'-derivatives inika, theVeda furnishes avery fewcases :\nvasantika, 'vernal', vdrsika,'oftherainy season', haimantika, 'wintry' (none\nofthem inRV.); AV. has kairatikd,'oftheKiratas', apparentfern, toa\nmasc. kairataka, which isnotfound tilllater. Examples from amore re-\ncentperiod (when theybecome abundant) are :vaidika, 'relatingtotheVedasJ\n,\ndharmika, 'religious', ahnika, 'daily', vainayika, 'well-behaved', dauvarika,\n'doorkeeper', naiyayika,'versed intheNyaya'.\n1223. Severalsuffixes, mostlyofrareoccurrence andques-\ntionable character, contain a^Tnasconsonantal element, andmay\nbegrouped together here.\na.With ana aremade, apparently,apairofderivatives inRV.from\nprimitives inu:thus, bhfgavana, vdsavana.\nb.With anl(whichisperhaps thecorresponding feminine)aremade a\nnumber ofwife-names :thus, indrani, varunani(these,with ufindrani, pu-\nrukutsani, mudgaldnl, arefound inRV.), rudranl, matulani, 'maternal un-\ncle'swife', ksatriyanl,'wife ofakshatriyd1\n(not quotable).\nC.Thefeminines inmandknifrom masculine stems intahave been\nalready noticed above (1176d). Frompdti, 'master, husband', thefeminine\nispdtni, both asindependent word, 'spouse', and asfinal ofanadjective\ncompound:thus, devdpatnl, 'having agodforhusband', sfadhupatnl, 'having\ntheIndus asmaster'. And thefeminine ofparusd, 'rough',isintheolder\nlanguage sometimes pdrusrii.\nd.With Inaaremade afull series ofadjective derivatives from the\nwords with final anc(407 ff.) Ithey areaccented usually uponthepenult,\nbutsometimes onthe final;andthesame word hassometimes both accents :\nforexample, apacina, nlcina, pradna, arvadna andarvacmd, pratlcma and\npraticmd, samicmd. Besidesthese,anumber ofother adjectives,earlier and\nlater :examplesaresarhvatsanna, 'yearly', pravrsina,'oftherainy season',\nvi<;vajanina,'of allpeople', jnatakulina,'ofknown family', adhvamna,\n'traveller' (ddhvan, 'way'), a^vlna, 'day's journeyonhorseback'(dfva,'horse' .\ne.With ena ismade samidhend, fromsamfdh, with initial strengthening.\n"
+    },
+    {
+        "page": 432,
+        "content": "416 XVII. SECONDARY DERIVATION.[1223\nf.Afewwords inma,having possessive meaning asmalina, ?rngina,\n(said tobeaccented onthefinal)aredoubtless transfers ofadjectives in\nm(1230)tothea-declension;compare 1209c.\ng.The adjectives made withsimple nafallunder another head (below,\n1245d, except thealmost isolatedstrat'na, 'woman's', fromstri, with initial\nstrengthening (itscorrelative, paumsna, occurslate).\n1224. Certain suffixescontaininga*Tmmaybesimilarly\ngrouped.\na.With imaaremade asmall number ofadjectives fromnouns intra:\nthus, khanttrima, 'made bydigging', krtrima, 'artificial', putrfma, 'purified'.\nAlso agrimd, 'foremost', (RV.:themahaving perhaps anordinalvalue).\nto.The uses ofsimple mainforming superlatives (474) and ordinals\n(487)have been already noticed, andthewords thusmadespecified; adimd\nisfurther given bythegrammarians, but isnotquotable until averyrecent\nperiod.\nC.Afewneuter abstracts endinmna: thus, dyumnd, 'brightness',\nnrmnd, 'manliness'; and,fromparticles, nimnd, 'depth', andsumnd, 'welfare'.\nThe suffix comesperhaps fromman with anadded a.\nd.Forthewords showing afinal min, seebelow, 1231.\n1225. rjTTmaya.With this suffix areformedadjectives\nsignifying 'made orcomposedorconsistingof.\nThe accent isalways onthepenult. Before them,the final ofthe\nprimitive isingeneral treated asinexternal combination :thus, mrnmdya,\n'made ofearth'(mrd), vanmdya, consistingofutterance'(vac), tejomdya,\n'made upofbrightness', adomdya,'ofthenature ofwhat isyonder' (QB.);\nbut intheVeda arefound manasmdya andnabhasmdya. with ayasmdya,\nwhich isallowed inthelater language alongside ayomdya. RV. hassumdya,\n'ofgood make'(sti, 'well'), andfczmmdya, 'made ofwhat?'\nThe suffix mayaisperhaps byorigin aderivative noun(yVna)incom-\nposition, but ithas atany rate the fullvalue ofasuffix from theearliest\nperiod ofthelanguage.\n1226.^ra.Afewderivativeadjectivesaremade with\nthis suffix. Accent andtreatment oftheprimitiveareva-\nrious.\na.Withsimple addition ofraaremade, forexample:pahsurd, 'dusty',\n-frira (also -(lila)inafrird, 'ugly', dhumrd, 'dusky' (dhumd, 'smoke'),\nmadhura(late), 'sweet'.\nInanexample ortwo,thereappearstobeaccompanyinginitial strength-\nening: thus, dgnldhra,'ofthe fire-kindler'(agnidh), fankurd, 'stake-like'\nb.With aninorganic vowel before theendingaremade,forexample,\nmedhira, 'wise', rathird,'inachariot; karmdray'smith'; dantura(late),\n'tusked'; andothers ofobscure connections.\n"
+    },
+    {
+        "page": 433,
+        "content": "1230]STEMS INma,may a,ra, la,va, fa,in. 417\nC.Theuse ofminforming afewwords ofcomparative meaning was\nnoticed above (474, andthewords somade were given.\n1227. Tl<*This andtheprecedingsuffix arereally\nbuttwoforms ofthesame. Insome wordstheyexchange\nwith oneanother, and 5Tlaisusually,butnotalways,the\nlaterform inuse.\nExamples are: bahuld, 'abundant', madhuld(latermadhura] andmadhula,\n'sweet', bhlmala, 'fearful', jiuald, 'lively', axilla (and aprird), 'wretched;\nwith a,vacala, 'talkative'(late); with?, phenila, 'foamy' (late: phena)', with\nu,vatula and vatula, 'windy' (late: vdta); andmatula, 'maternal uncle',\nisasomewhatirregular formation frommair,'mother'.\nThe later language hasafewadjectivesinlu,askrpalu and dayalu,\n'compassionate'.\n1228. 3Tva.Asmall number ofadjectives have this\nending (accented, added toanunalteredprimitive).\na.Thus, arnavd, 'billowy', kepavd, 'hairy'; rasnavd, 'girded'; anjivd,\n'slippery', fantivd, 'tranquillizing', fraddhivd,'credible'.\nb.There areavery few adjectivesinvalaandvaya which maybe\nnoticed here; thus, krslvald, 'peasant' (krsi, 'ploughing'), fikhavala anddan-\ntavala(late); druvdya, 'wooden dish'.\nc.With vyaaremade two orthree words fromnames ofrelationship:\nthus, pttrvya, 'paternal uncle', bhrdtrvya, 'nephew, enemy'.\n1229. $\\$a.Averyfewadjectives appeartobemade\nbyanadded endingofthisform.\nThus, rorna$d orlomafd, 'hairy', etafa, 'hasting', babhlu^d orbabhrufd,\n'brownish', yuva$a, 'youthful'.\nManyoftheadjectivederivatives alreadytreated havesome-\ntimes apossessive value, thegeneral meaningof'being con-\ncerned with, havingrelation to'being specializedinto that of\n'beinginpossession of.Butthere arealso afewdistinctively\npossessivesuffixes\n;andsome ofthese, onaccount oftheun-\nlimited freedom offorming themandthefrequencyoftheir occur-\nrence, arevery conspicuous partsofthegeneral systemofderi-\nvation. These willbenext considered.\n1230.^Tin.Possessiveadjectivesofthisending may\nbeformed almostunlimitedlyfrom stems in^aor TTa>,\nand aresometimes (butvery rarely) made from stems with\nother finals.\nAfinalvowel disappearsbefore the suffix. The accent is\nWhitney, Grammar. 27\n"
+    },
+    {
+        "page": 434,
+        "content": "418 XVII. SECONDARY DERIVATION.[1230\nalwaysin.Astotheinflection oftheseadjectivesseeabove,\n438 if.Theyaretobecounted byhundreds intheolder lan-\nguage, and areequallyormore numerous inthe later.\na.Examples from a-stems are :afvfn, 'possessing horses', dhanfn,\n'v.ealthy', paksfn, 'winged', baltn, 'strong', bhagfn, 'fortunate', vajrfn, 'wield-\ningthethunderbolt', cikhandfn, 'crested', hastfn, 'possessing hands', sodatfn,\n'ofsixteen', gardabhanadfn, 'having anass's voice', brahmavarcastn,'ofemin-\nentsanctity', sadhudevfn, 'having luck atplay', kucidarthfn, 'having errands\neverywhither'; froma-stems, manlsfn, 'wise', fikhfn, 'crested', rtayfn, 'pious'.\nb.Theexamples from other stems areonly sporadic: thus, from i-stems,\nabhimaMn, 'plotting against', arcm, 'gleaming', wrmfn, 'billowy', kJtadfn,\n'spangled',andperhaps atitMn. 'wandering'; insurabhtntara, thenasal is\nrather aeuphonic insertion, tobreak thesuccession oflight syllables ;\nfrom stems inan,varmfn, 'cuirassed'; from stems inas,retfn,'rich in\nseed', probably varctn,n.pr.:andfavastn and sahasin (unlessthese come\nthrough stem-forms in-so],\nC.Itwasnoticed above (1183) that derivatives ofthisform often have\navalue equivalent tothat ofpresent participles; andthat thesuffix isused,\nboth earlier andlater, inthemanner ofaprimary suffix, making words of\nsuch value directly from roots. The originally secondary character ofthe\nwhole formation isshown, ontheonehand, bythefrequent use inthesame\nmanner ofwords bearing anunmistakably secondary form, asprapntn, garbMn,\njurntn, dhumtn, paripanthfn, pravepanfn, matsarfn; and, ontheother hand,\nbytheoccurrence ofrevertedpalatals (216j before thein,which could only\nbeasinreplaced a:thus, arkin, -bhangfn, -sangfn, and -roktn...4.\nd.Inthree orfoursporadic cases,ay,preceded bya,appearsbefore\nin:thus, svadhaytn (VS.: TB.-vfn; theyishere evidently aninsertion:\n258), srkdy-Cn, dhanvaytn, andatatayin (VS.: TS. -vin), which isfound also\ninthelater language. RV. hasonlymaraytn (value?) andrtayfn (apparently\nthrough artayd [1149]from thedenominative rtay). Toassume forthese\nasuffixyinisquite needless.\n1231. frpfmin.With this suffix aremade anextremely\nsmall number ofpossessive adjectives.\nInthe oldlanguage, thewords inmmhave theaspectofderivatives\nininfrom nouns inma, although intwo orthree cases imfn, rgmfn,\nkrudhmin inRV.,vagmfn inQB. nosuchnouns arefound inactual use\nbeside them. Inthelaterlanguage, mm isused asindependentelement in\naveryfewwords :thus, gomin, 'possessing cattle', svamin (Sutras andlater),\n'master, lord' (wo, 'own').\n1232. fiR vin.Theadjectives made with this suffix are\nalsonot atallnumerous. They have thesame meanings\nwith those in^in.\nTheRV.hastenadjectivesinvin(ayudhvfn, so-called,isdoubtless a\n"
+    },
+    {
+        "page": 435,
+        "content": "STEMS INm,min, vin, vant. 419\ngerund intvl :a-yudh-tvi] ;theybecome rather morecommon later. Though\nforthemmaybesuspected asimilarorigin tothose inyinandmm(above),\nsigns ofitaremuch lessclearly traceable.\nThe majority have vinadded after as: thus, namasvfn, 'reverential',\ntapasvin. 'heated', tejasvfn, 'brilliant', yafasvfn, 'beautiful'; andfatasvfn,\n'possessing hundreds'(RV.), hasaninserteds,byanalogy withthem. Others\nhave a(sometimes, bylengthening):thus, medhavfn, mayavfn, astravtn,\n'obedient tothegoad', dvayavfn, 'double-minded', ubhayavin, 'possessing of\nboth kinds'. Morerarely, vfn isadded after another consonant thans;thus,\nvagvfn. dhrsadvtn. The doubtful word vya$nuvtn (VS., once: TB.vydfniya)\nappears toaddtheending (orin,with euphonic v)toapresent tense-stem.\n1233. effivant. Verynumerouspossessive adjectivesare\nmade bythis suffix, from noun-stems ofevery form, both\nintheearlierlanguage andinthe later.\nThe accentgenerally remains upontheprimitive, without\nchange;butanaccentresting onastem-final,ifthisbeany-\nthing butaora,isinthemajorityofcases thrown forward\nuponthe suffix.\nAfinalvowel oftenesta,very rarely u isinmany words lengthened\nintheolder language (247) before thisending,asincomposition. Nouns\ninanretain then.\na.Examples ofthenormal formation are :with unchanged accent,\nkefavant, 'hairy', putrdvant, 'havingason', prajdnanavant, 'procreative',\npunddrikavant,'rich inlotuses', Mranyavant,'rich ingold', apupdvant,\n'having cakes'; prajdvant, 'having progeny', urnavant, Svooly', ddksindvant,\n'rich insacrificialgifts'; sdkhivant, 'having friends', saptarsivant, 'accompanied\nbytheseven sages'; fdcwant, 'powerful', tdvisivant, 'vehement', pdtriivant,\n'withspouse', dhivant, 'devoted', dydvaprthivivant (94),'with heaven and\nearth'; visnuvant, 'accompanied byVishnu'; awftvant,'hither turned', afirvant,\n'mixed withmilk', svhrvant, 'splendid', farddvant,'full ofyears', pdyasvant,\n'rich', tdmasvant, 'dark', brdhmanvant, 'accompaniedwithworship', romanvant,\n'hairy'; butalsoItimavant andvrtrahavant (PB.); with accent onthesuffix,\nagnivdnt, 'having fire', rayivdnt, 'wealthy', nrvdnt, 'manly', padvdnt, 'having\nfeet', nasvdnt, 'with nose', asanvdnt, 'having amouth', ?irsanvdnt, 'headed'.\nb.With final stem-vowel lengthened:forexample, d^vuvant (beside\nd^vavant), 'possessing horses', sutdvant, 'having soma expressed', vrsnyavant,\n'ofvirile force' (about thirty such cases occur inV.); cdktwant, 'mighty',\nsvddhitwant, 'having axes', ghrnlvant, 'hot'; visuvdnt, 'dividing' (vfau, 'apart').\nC.Afewspecial irregularities are :mdrasvant, with sadded tothe\nprimitive ;krfandvant, withanomalous accent(iffrom fcr?ana, 'pearl') ;and,\nfromparticles, antdrvant, 'pregnant', andvisuvdnt (just quoted).\nd.Bythesame suffix aremade from pronominalroots andstems ad-\njectives inwhich isshown another specialized meaning,that of'liketo,\nresembling'. They were given above (517:tobeadded areivant andkivant).\n27*\n"
+    },
+    {
+        "page": 436,
+        "content": "420 XVII. SECONDAEY DEEIVATION.[1233\nAnd derivatives from nouns inthe older language areperhaps occasionally\ntobeunderstood inthesame way: e.g.fndrasvant,'like Indra'.\nIthas alsobeen pointed out(1107) that theadverb ofcomparisonin\nvat isdoubtless tobeunderstood asaccusative neuter ofaderivative of\nthis class.\ne.Invivdsvant orvfvasvant, 'shining' (also proper name),isseen a\nside-formation tovivdsvan (1169), having theaspectofaprimary derivative.\nf.Forthe derivatives invatfromprepositions, which appeartohave\nnothing todowith thissuffix,seebelow, 1245f.\nNone ofthe suffixes beginning with vshow intheVeda resolution of\nVtOM.\n1234.SRvan.Thesecondaryderivatives inthis suffix be-\nlongtotheolder language, and areasmall number, ofwhich\nextremely fewhavemore thananoccurrence ortwo.\nThey have theaspectofbeing produced under thejoint influence of\nprimary vanandsecondary vant. Afinal short vowel isusually lengthened\nbefore the suffix. Theaccent isvarious, butoftenest onthepenultofthe\nstem. Thefeminine(like that ofthederivatives inprimary van: 1169, end)\nisinvarl.\nTheVedic examplesare :from a-stems, rnavdn orrnavdn, rtdvan (andf.\n-van), fghavan, dhitdvan, satydvan. sumnavdri, andmaghdvan; froma-stems,\nsunftavari, svadhdvan (andf.-van); fromz-stems, amatlvdn, arativdn,\nfrusfivdn, muswdn, and krslvan (only inthefurther derivative karslvana) ;\nfrom consonant-stems, dtharvan, samddvan, sdhovan (badAV. variant toRV.\nsahdvan). Somewhat anomalous aresahavan andsahasavan, tndhanvan(for\nfndhanavan?), andsanttvan(forsdnitivan?). The only words ofmore than\nsporadicoccurrence arertavan, maghdvan, dtharvan.\n1235. JTrTmant. This isatwin-suffix tocTrT vant(above,\n1233);their derivatives have thesame value, and areto\nsome extentexchangeablewith oneanother. Butpossess-\nives iJTrTinant aremuch lessfrequent (intheolder lan-\nguage, about athird asmany),and areonly very rarely\nmade from a-stems.\nIftheaccent oftheprimitive word isonthefinal,itisin\nthegreat majorityofinstances(three quarters)thrown forward\nupontheadded suffix;otherwise,itmaintains itsplace un-\nchanged. Afinalvowel before thesuffix isinonlyafewcas-\nesmade long. Examplesare :\na.With theaccent oftheprimitive unchanged: kdnvamant, andyd-\nvamantj'rich inbarley' (thesealone from a-stems, and theformer only\noccurring once), dvimant, 'possessing sheep', afdnimant, 'bearing thethunder-\nbolt', 6sadhimant,'rich inherbs', vd&mant, 'carrying anaxe', vdsumant,\n"
+    },
+    {
+        "page": 437,
+        "content": "1238] STEMS INvan,mant, ta, tati, tat. 421\n'possessing good things', mddhumant,'rich insweets', tvdstrmant, 'accom-\npanied byTvashtar', hotrmant, 'provided withpriests', dyusmant, 'long-lived',\njyotismant,'full ofbrightness'; ulkusimant, 'accompanied withmeteors',\npilumantf?), prasumant, 'having young shoots', gro'mant,'rich inkine', ka-\nkudmant, 'humped', vidytinmant (with irregular assimilation oft:VS. has\nalsokakunmant], 'gleaming', virukmant, 'shining', havfsmant,'with libations'.\nb.With theaccent thrown forward upon theending: agnimdnt, 'having\nfire', isudhimdnt,'with aquiver', pa$umdnt, 'possessing cattle', vayumdnt,\n'withwind', pitrmdnt (AV. pitrmant h'accompanied bytheFathers', matrmdnt,\n'having amother'; nolong final vowels arefound before the suffix inthis\ndivision, andonly once aconsonant, indasmdt(RV.. once).\nC.Protraction ofafinal vowel isseen inMslmant, dhrdjlmant, Mrl-\nmant;injyotislmantisirregularly inserted an I(after theanalogy oftdvisi-\nmant).\n1236. Ithasbeen seenabove(especiallyinconnection with\nthesuffixes aandya]that theneuter ofaderivativeadjective\nisfrequently used asanabstract noun. There are,however,\ntwo suffixes which have inthelater language thespecificoffice\nofmakingabstract nouns fromadjectives andnouns\n;andthese\narefound also, moresparingly used, inthe oldest language,\neach havingthere one ortwo otherevidentlyrelated suffixes\nbeside it.\nForderivatives ofthesame value made with thesuffix iman, seeabove,\n1168.2.\n1237. cTTta.With this suffix aremade feminine abs-\ntract nouns, denoting'thequalityofbeingsoand so',from\nbothadjectivesandnouns.\nTheform oftheprimitiveisunchanged, andtheaccent is\nuniformly onthesyllable precedingthe suffix.\nExamples (fromtheolder language)are: devdta, 'divinity', virdta, 'man-\nliness', purusdta, 'humannature', bandhtita, 'relationship', vastita, 'wealth';\nnagndta, 'nakedness', suvlrdta, 'wealth inretainers', anapatydta,'lack of\ndescendants', agtita, 'povertyincattle', abrahmdta,'lack ofdevotion', apra-\njdsta,'absence ofprogeny';also doubtless sunfta (from sunara), although the\nword isafewtimes used asanadjective (like famtati andsatyatati:see\nnext paragraph). InRV. isfound avirata, with exceptional accent.\n1238 ecilfri tali, rTTrT tat.These suffixes areVediconly,\nandthelatter islimited toRV. Theirrelationshiptothepre-\ncedingisevident, butopinionsareatvariance astoitsnature.\nThe accent isasinthederivatives with ta.\nThequotable examplesintati are: arishtdtati, 'uninjuredness', ayaks-\nmdtati, 'freedom fromdisease', grbhltdtati,'the being seized', jyesthdtati,\n'supremacy', devdtati, 'divinity', vastitati, 'wealth', fdrhtati, 'good-fortune',\n"
+    },
+    {
+        "page": 438,
+        "content": "422 XVII. SECONDARY DERIVATION.[1238\nsarvdtati, 'completeness'; and,withexceptional accent, dstatati, 'home', and\nddksatati,'cleverness'. Twowords intatiareused adjectively (inorganically,\nbyapposition?): fdmtdti (RV., twice; andAV. xix. 44.1,inMSS.), and\nsatyatdti (RV., once:voc.).\nThewords intat(apparently made byabbreviation fromtati) occur only\ninacase ortwo: theywere allmentioned above (383d. 2).\n1239. pTtva.With this suffix aremade neuter nouns,\nofthesame value asthefeminines incHta(above, 1237).\nTheneuter abstracts intvaareintheolder language con-\nsiderably morecommon than thefeminines inta,although them-\nselves alsonotverynumerous. The accent iswithoutexception\nonthe suffix.\nExamples (fromtheolder language)are :amrtatvd, 'immortality', devatvd,\n'divinity', subhagatvd, 'good-fortune', ahamuttaratva, 'struggleforprecedency',\nfucitvd, 'purity', patitvd, 'husbandship', taranitvd, 'energy', dirghayutvd,\n'long life', $atrutvd, 'enmity', bhratrtvd, 'brotherhood', vrsatvd, 'virility',sat-\nmatvd, 'soulfulness', maghavattvd, 'liberality', raksastvd, 'sorcery'. Inana-\ngdstvd andaprajastvd, there isalengtheningofthe finalsyllable oftheprim-\nitive; and insduprajdstvd (AV., once)thisappears tobeaccompanied by\ninitial vrddhi (sdubhagatvd isdoubtless fromsdubhaga, notsubhdga). Invas-\nativaritvd (TS.) there isshortening offinal feminine Ibefore the suffix.\nInisitatvdta (RV., once), 'incitedness', andpurusatvdtd (RV., twice),\n'humanquality', appears tobeacombination ofthetwoequivalent suffixes\ntvaand ta.\nThevoftva istoberead inVeda asuonlyonce (raksastud).\n1240. ppftvana. The derivatives made with this suffix are,\nlike those intva,neuter abstracts. They occur onlyinRV.,\nand, exceptinasingle instance(murtiatvand) ,have beside them\nequivalent derivatives intva.The accent isonthe final, and\nthe tva isnever resolved into tua.\nThewords are :kavitvand, janitvand, patitvand, martiatvand, mahitvand,\nvasutvand, vrsatvand, sakhitvand.\n1241.Afew suffixes make nochangeinthecharacter as\npartofspeechoftheprimitivetowhichtheyareadded, but\neither aremerely formalappendages, leavingthevalue ofthe\nword what itwas before, ormake achangeofdegree,orin-\ntroduce some other modification ofmeaning.\n1242. The suffixes ofcomparison andordinal suffixes\nhave forthemostpartbeen treatedalready,andneedonly\nareference here.\na.rT^taraandrFTtama aretheusual secondarysuffixes\nofadjective comparison:respectingtheir useassuch, seeabove,\n"
+    },
+    {
+        "page": 439,
+        "content": "1245! STEMS INtea, tvana, tara, tama,ETC. ETC. 423\n471 3;respectingtheuseoftama asordinal etc.suffix, see\n487;respectingthat oftheir accusatives asadverbial suffixes to\nprepositions etc., see1111 c.\nInvrtrntdra RV., once: perhapsanerror) theaccent isanomalous; in\nmrdaydttama,itisdrawn forward tothe final oftheparticiple,asoften in\ncomposition (1309 ;$a?vattamd (RV.) hastheordinal accent; sarhvatsara-\ntamd (B.)isanordinal; dfvatara(RV., once: anerror?)isanordinary\nadjective,'oftheday';ina$vatard, 'mule', anddhenustari, 'cow losing her\nmilk',theapplicationofthesuffix ispeculiar and obscure;inrathamtara,\nname ofacertain saman,itisthesame.\nb.^raand *Tma, like taraandtama, have acomparative\nandsuperlativevalue;andthe latter ofthem forms ordinals :\nseeabove, 474, 487.\nc. STtha, like tama andma,forms ordinals from afew\nnumerals :see487.\nd.\\^\\ tithaforms words ofanordinal character from bahu\n(bahutitha, 'many-eth')and tavant(withloss ofatinthecom-\nbination :tavatitha, 'so-many-eth');and,itissaid, from other\nwords meaning'anumber orcollection'(gana, puga, samgha).\n1243. Ofdiminutive suffixes there arenone inSanskrit\nwithclearly developed meaning and use. The occasional em-\nploymentofka,inasomewhat indistinct way,tomake dimin-\nutives, hasbeen noticed above(1222).\n1244. Oftheordinary adjective-making suffixes, given above,\nsome occasionally makeadjectives fromadjectives,withslightor\nimperceptiblemodification ofvalue. Theonly oneused toany\nconsiderable extent inthiswayiska: astowhich, see1222.\n1245.Afew suffixes areused tomake derivatives from\ncertain limited andspecialclasses ofwords, asnumerals and\nparticles. Thus :\na.cUTtayamakes afewadjectives, meaning'ofsomany\ndivisions orkinds'(usedintheneuter ascollectives),from nu-\nmerals :thus, dvitaya, tritaya, cdtustaya (AV.), saptdtaya (QB.),\nastataya (AB.), ddcataya (RV.).\nb.fTTtyamakes afewadjectivesfromparticles:thus,\nnitya, 'own', nistya, 'strange', amatya, 'companion', dpatya (as\nneut.noun), 'descendants', avistya, 'evident', sdnutya, 'distant',\ntatratya (late), 'belonging there'; and, accordingtothegramma-\nrians, ihatyaandkvatya.\nTheyoftyaisinRV.always toberead as iafter aheavy syllable.\nc.rTtaforms dvitd andtritd, alsomuhurtd, 'moment', and\napparently avatd, 'well(forwater)'.\n"
+    },
+    {
+        "page": 440,
+        "content": "424 XVII. SECONDARY DERIVATION.[1245\nd.With Tnaaremadepurand, 'ancient', visuna, 'various',\nandperhaps samand,'like'.\ne.With rpTtana or^tnaaremadeadjectives from ad-\nverbs, especiallyoftime :thus, pratnd, 'ancient', nutana ornutria,\n'present', sanatdna orsandtna, 'lasting', pratastdna, 'early', diva-\ntana(late),'oftheday', gvastana (late),'ofthemorrow'\n;PB.has\nalso tvattana, 'belongingtothee'. Inthelatelanguagethesuf-\nfixisused sometimes with anadjectiveoftime :thus, ciratna.\nf.^rTvatmakes fromparticlesof.direction thefeminine\nnouns mentioned above(383d. 1).\ng.^Ikata, properlyanoun incomposition,isreckoned\nbythegrammarians asasuffix inutkata, nikata, prakata,vikata\n(RV., once, voc.), andsamkata(allsaid tobeaccented onthe\nfinal).\nh.Occasional derivatives made with theordinarysuffixes\nofprimary andsecondaryderivation from numerals andparticles\nhave been noted above: thus, seeana(1150. 2di,ti(1157.4),\nant(1172),u(I178h), ayya (1218),ka(1222aj, mna(1224),\nmaya (1225),vant(1233 c).\nCHAPTER XVIII.\nFORMATION OFCOMPOUND STEMS.\n1246. THEfrequentcombination ofdeclinable stems\nwith oneanother toformcompoundswhich then aretreated\nasifsimple,inrespecttoaccent, inflection, andconstruc-\ntion,isaconspicuousfeature ofthelanguage,from its\nearliestperiod.\nThereis,however, amarked difference between theearlier\nandthelaterlanguageasregardsthelength andintricacyofthe\ncombinations allowed. InVeda andBrahmana,itisquiterare\nthatmore thantwostems arecompounded together except\nthat tosomemuch used andfamiliar compound,astoaninte-\ngralword, afurther element issometimes added. Butthelater\ntheperiod, and, especially,themore elaborate thestyle,the\nmore acumbrous and difficult aggregateofelements, abnegat-\n"
+    },
+    {
+        "page": 441,
+        "content": "1247] CLASSES OFCOMPOUNDS. 425\ningtheadvantagesofaninflectivelanguage,takes theplace of\ntheduesyntactical union offormed words into sentences.\n1247. Sanskritcompoundsfall into threeprincipal\nclasses :\nI.Copulativeoraggregative compounds,ofwhich\nthemembers aresyntacticallycoordinate :ajoining together\ninto one ofwords which inanuncompounded condition\nwould beconnectedbytheconjunction'and'.\nExamplesare :indravdrunau, 'Indra andVaruna', satyanrte,\n'truth and falsehood', krtakrtdm, 'done andundone', devagan-\ndharvamanusoragaraksasas ,'gods andGandharvas andmen and\nserpents anddemons'.\nThemembers ofsuch acompound may obviously beofanynumber, two\normore than two. Nocompound ofanyother class cancontain more than\ntwomembers ofwhich, however, either orbothmay becompound, or\ndecompound (below, 1248).\nII.Determinativecompounds,ofwhich theformer\nmember issyntactically dependent onthelatter, asitsde-\nterminingorqualifying adjunct: being either,1.anoun\nlimitingitinacase-relation, or,2.anadjectiveoradverb\ndescribingit.And, accordingasitistheoneortheother,\nmaybedistinguishedthetwosub-classes :A.Dependent\ncompounds; andB.Descriptive compounds;their differ-\nence isnotanabsolute one.\nExamplesare :ofdependent compounds, amitrasenft, 'army\nofenemies', padodaka, 'water forthefeet', ayur'dh, 'life-giving',\nhdstakrta, 'made with thehands'; ofdescriptive compounds, ma-\nharsi, 'great sage', priyasakhi,'dear friend', amltra, 'enemy',su-\nkrta, 'well done'.\nThese two classes areofprimaryvalue\n;theyhave under-\ngone nounifyingmodification intheprocessofcomposition;\ntheir character aspartsofspeechisdetermined bytheir final\nmember, andtheyarecapableofbeingresolved intoequivalent\nphrases bygivingtheproper independent formandformal means\nofconnection toeachmember. That isnotthecasewith the\nremaining class, which accordinglyismore fundamentallydistinct\nfromthem than theyarefrom oneanother.\nIII .Secondary adjective compounds,thevalue of\n"
+    },
+    {
+        "page": 442,
+        "content": "426 XVIII. COMPOSITION.[1247\nwhich isnotgiven byasimpleresolution into their com-\nponent parts,butwhich, though havingasfinalmember a\nnoun, arethemselvesadjectives. These, again,areoftwo\nsub-classes: A.Possessive compounds, which arenoun-\ncompoundsofthepreceding class, with theideaof'possess-\ning' added, turning them from nouns intoadjectives;\nB.Compoundsinwhich thesecond member isanounsyn-\ntactically dependent onthe first:namely,1.Prepositional\ncompounds,ofagoverning preposition andfollowing noun;\n2.Participial compounds (only Vedic), ofapresent par-\nticiple and itsfollowing object.\nThe sub-class B. iscomparatively quite small, and itssecond division\n(participial compounds)ishardly metwith even inthelater Vedic.\nExamplesare :vtrasena, 'possessingahero-army', prqfakama,\n'having desire ofprogeny', tigmdqrnya, 'sharphorned', hdritasraj,\n'wearing green garlands'; atimatrd, 'excessive'; yavaydddvesas,\n'driving away enemies'.\nThe adjective compounds are,likesimple adjectives,liable tobeused,\nespeciallyintheneuter,asabstract and collective nouns, and intheaccus-\native asadverbs; andoutofthese uses havegrown apparent specialclasses\nofcompounds, reckoned andnamed assuch bytheHindu grammarians. The\nrelation indetail ofthe classification given above tothatpresented inthe\nnative grammar, andwidely adopted from thelatter bytheEuropean gram-\nmars,willbemade clear aswegoontotreat theclasses indetail.\n1248.Acompound may,likeasingle word, become a\nmember inanother compound, and this inyetanother\nand soon,without definite limit. Theanalysisofany\ncompound,ofwhateverlength (unlessitbeacopulative),\nmust bemade byasuccession ofbisections.\nThus,thedependent compound purvajanmakrta, 'done inaprevious\nexistence',isfirst divisible into krtaand thedescriptive purvajanman, then\nthis into itselements; thedependent sakalaniticastratattvajfia, 'knowing the\nessence ofallbooks ofbehavior', has first theroot-stem jna (for |//na),\n'knowing', separated from therest, which isagain dependent ;then this is\ndivided intotattva, 'essence', andtheremainder, which isdescriptive; this,\nagain, divides into sakala, 'all', and nlti^astra, 'books ofbehavior',ofwhich\nthe latter isadependent compound andtheformer apossessive (saandkala,\n'having itsparts together').\n"
+    },
+    {
+        "page": 443,
+        "content": "1251]FORM OFPRIORMEMBER OFCOMPOUND. 427\n1249. The final ofastem iscombined with the initial of\nanother stem incomposition accordingtothegeneralrules for\nexternal combination :theyhavebeengiven, with theirexceptions,\ninchap. III., above.\na.Ifastem hasadistinction ofstrong andweak forms,\nitregularlyenters intocompositionaspriormember initsweak\nform; or,ifithasatripledistinction(311),initsmiddle form.\nThatis,especially, stems inrorar,atorant, acorawe, etc.,show\nincomposition theforms inr,at,ac, etc.;while those inanand inusually\nlose their finaln,andarecombined asifaarid iwere their properfinals.\nb.Occasionally, astem isused aspriormember ofacompound,which\ndoes notappear,ornotinthatform,asanindependent word :examples\naremahd, 'great' (apparently usedindependentlyinV.inaccusative), tuvi,\n'mighty' (V.), dvi,'two'.\nC.Notinfrequently,the finalmember ofacompound assumes aspecial\nform: seebelow, 1315.\n1250. But acase-form inthepriormember ofacompound\nisby\"nomeans rare, from theearliest periodofthelanguage.\nThus :\na.Quite often,anaccusative, especiallybefore aroot-stem,oraderiv-\native inaofequivalent meaning:forexample, patamgd, 'going byflight',\ndhanamjayd, 'winning wealth', abhayarhkard, 'causing absence ofdanger',\npustimbhard, 'bringing prosperity', vdcammkhayd, 'inciting the voice'; but\nalsosometimes before words ofother form, ashrdamsdni, 'heart-winning',\ndfvamisti, 'horse-desiring', (ubharhydvan, 'goinginsplendor', subhdgamkdrana,\n'making happy'.\nb.Much morerarely, aninstrumental :forexample, girdvrdh, 'increasing\nbypraise', vdcdstena, 'stealing byincantation', krdtvdmagha, 'gladly bestowing',\nbhdsdketu, 'bright withlight', vidmandpas,'active with wisdom'.\nC.Notseldom,alocative;and this also especially with aroot-stem or\na-derivative :forexample, agregd, 'goingatthehead', divikstt, 'dwelling in\nthesky', vanesdh, 'prevailinginthewood', angesthd, 'existing inthelimbs',\nprostheyayd, 'lying onacouch', sutekara,'active with thesoma', divtcara,\n'moving inthesky'; dre^atru, 'having enemies farremoved', sumnddpi, 'near\ninfavor', mdderaghu, 'hasting inexcitement'; apsujd, 'born inthewaters',\nhrtsvds, 'hurlingathearts'.\nd.Leastoften,agenitive: thus, rdydskdma-,'desirous ofwealth'. But\ntheolder language hasafewexamplesoftheputting togetherofagenitive\nwith itsgoverning noun, bothmembers ofthecombination keeping'their own\naccent: seebelow, 1267d.\ne.Excepting occasionally, where case-forms areused, there isnodesig-\nnation ofplural meaning byplural form;butafeminine word, where clear-\nness requires it,sometimes keepsitsdistinctive feminine form.\n1251. The accent ofcompoundsisvery various, andliable\ntoconsiderable irregularity even within the limits ofthesame\n"
+    },
+    {
+        "page": 444,
+        "content": "428 XVIII. COMPOSITION.'[1251\nformation\n;and itmust beleft tobepointed out in,detail below.\nAllpossiblevarieties arefound tooccur. Thus:\na.Eachmember ofthecompound retains itsownseparate accent. This\nisthemost anomalous andinfrequent method. Itappearsincertain Vedic\ncopulative compounds chiefly composedofthenames ofdivinities(so-called\ndevatd-dvandvas :1255ff.),and inasmall number ofaggregations partly\ncontainingagenitive case-form aspriormember (1267 d).\nb.The accent ofthecompoundisthat ofitsprior member. This is\nespecially thecase inthegreatclass ofpossessive compounds ;but also in\ndeterminatives having theparticipleintaorno,asfinalmember,inthose\nbeginning with thenegative aoran,and inother lessnumerous andim-\nportant classes.\nc.The accent ofthecompoundisthat ofthe finalmember. This is\nnotonsolarge ascale thecase asthepreceding; but itisnevertheless\nquite common, being found inmany compounds having averbal noun or\nadjective asfinalmember,incompounds beginning with thenumerals dvi\nand triortheprefixessuanddus,andelsewhere innotinfrequent exceptions.\nd.Thecompound takes anaccent ofitsown, independentofthat of\neither ofitsconstituents, onits finalsyllable (notalways, ofcourse,tobe\ndistinguished from thepreceding case). This method islargely followed:\nespecially, bytheregular copulatives, andbythegreat mass ofdependent\nanddescriptive noun-compounds, bythose havingaroot-stem asfinal, by-\nmostpossessives beginning with thenegative prefix; andbyothers.\ne.Thecompoundhasanaccent which isaltered from that ofone of\nitsmembers. This iseverywhere anexceptional and sporadically occurring\ncase, and theinstances ofit,noted below under eachformation,donot\nrequiretobeassembled here. Examplesare :medhdsati (medha), tildmifra\n(tfla), khddihasta (khad{), yavaydddvesas (yavdyant); cakadhuma (dhumd),\namfta (mrta), suvira (vlrd), tuvigriva (grlvd). Afewwords asvfyva,\npUrva, andsometimes sdrva take usually achanged accent aspriormembers\nofcompounds.\nI.Copulative Compounds.\n1252.Two ormore nouns much less oftenadject-\nives, and,inaninstance ortwo, adverbs havingacoord-\ninate construction,asifconnected by'and',aresometimes\ncombined intocompounds.\nThis istheclass towhich theHindu grammarians givethe\nname ofdvandva, 'pair, couple';advandva ofadjectives, however,\nisnotrecognized bythem.\n1253. The noun-copulatives fall,asregardstheir in-\nflective form, intotwo classes :\n"
+    },
+    {
+        "page": 445,
+        "content": "1255]COPULATIVE COMPOUNDS. 429\na.Thecompound hasthegender anddeclension ofits\nfinalmember, and isinnumber adual oraplural,accord-\ningtoitslogical value,asdenoting two ormore thantwo\nindividualthings.\nExamplesare :prdnapdnau, 'inspiration andexpiration',\nvrihiyavau,'riceandbarley', rksame, Averse and chant', kapoto-\nlukau, 'dove and owl', candradityau, 'moon andsun', hastyacvau,\n'theelephant and horse'; ajdvdyas, 'goats andsheep', devdsuras,\n'thegods anddemons', atharvdngirdsas,'theAtharvans andAn-\ngirases', sambddhatandryds,'anxieties andfatigues', hasty aqvas,\n'elephants andhorses'; ofmore than twomembers(noexamples\nquotable from theolderlanguage), caydsanabhogds, 'lying, sitting,\nandeating', brdhmanaksatriyavitqudras,'aBrahman, Kshatriya,\nVaicya, andQudra', rogacokaparitdpabandhanavyasandni, 'disease,\npain, grief, captivity, and misfortune'.\nb.Thecompound, withoutregardtothenumber de-\nnoted, ortothegenderofitsconstituents, becomes aneuter\nsingularcollective.\nExamplesare :istapurtdm,'what isoffered andbestowed',\nahordtrdm, 'adayandnight', krtdkrtdm, 'thedoneandundone',\nbhutabhavydm, 'past andfuture', kecagmacru,'hair andbeard',\nahinakulam, 'snake andichneumon', yukdmaksikamatkunam,'lice,\nflies, andbugs'.\n1254. a.Astem inraspriormember sometimes takes itsnominative\nform,ina:thus, pitdputrau,'father andson', hotapotarau, 'theinvoker and\npurifier (priests)'.\nb.Aconsonant-stem asfinalmember sometimes adds anatomake a\nneuter collective: thus, chattropdnaham, 'anumbrella and ashoe'.\nc.Thegrammarians give rules astotheorder oftheelements composing\nacopulative compound: thus, that amoreimportant,abriefer, avowel-\ninitial member should stand first;andthatoneending inashould beplaced\nlast. Violations ofthemall,however, occur.\n1255. Intheoldest language (BV.), copulative compounds\nsuch asappearlater arequite rare, the class being chieflyre-\npresented bydual combinations ofthenames ofdivinities and\notherpersonages, andofpersonifiednaturalobjects.\na.Inthese combinations, eachname hasregularly and\nusuallythedual form, and itsownaccent;but, intheveryrare\ninstances(only3occurrences outofmore than300)inwhich\nother cases than thenom.-acc.-voc. areformed, thefinalmember\nonlyisinflected.\n"
+    },
+    {
+        "page": 446,
+        "content": "430 XVIII. COMPOSITION.[1255\nExamplesare :fndrdstfmd, fndrdwfsnu, fndrdbrhaspdtl, agnlsomau, tur-\nvdfdyddu, dydvaprthivi,usdsdndktd, sUrydmdsd. The only pluralisindra-\nmarutas(voc.). Thecases ofother thannominative form aremitrdvdrundbhydm\nandmitravdrunayos, andindravdrunayos (each onceonly).\nFrom dydvaprthiviismade thevery peculiar genitive divdsprthivyts\n(4times :AV. hasdydvdprthivibhydm anddydvdprthivyos],\nb.Inonecompound, parjdnyavdta, the firstmember does nothave the\ndual ending along with thedouble accent(indrandsatyd, voc.,isdoubtful as\ntoaccent). Inseveral, thedouble accent iswanting, while yetthedouble\ndesignation ofnumber ispresent; thus, indrapusnds (beside fndrdpusand],\nsomdpusdbhydm .(somdpusand occurs only asvoc.), vdtdparjanyd, surydcandra-\nmdsd, and indrdgrii (with indrdgnfbhydm andindrdgnytis]:somdrudrdti is\naccented only inQB.And inone, indravdyu, form and accent areboth\naccordant with theusages ofthelater language.\nc.Ofother copulatives,likethose madelater,theRV. hastheplural\najdvdyas,theduals rksdme, satydnrte, sdfandnafane ;also theneuter col-\nlective istdpurtdm, andthesubstantively used neuter ofacopulative adjective,\nnllalohitdm. Further, theneuterplurals ahordtrdni, 'nycthemera1\n,and\nukthdrka, 'praises andsongs',ofwhich the finalmembers asindependent\nwords arenotneuter. Noone ofthese words hasmore than asingle oc-\ncurrence.\n1256. Inthelater Vedic(AV.),theusageismuch more\nnearly accordant with that ofthe classicallanguage, save that\ntheclass ofneuter singularcollectives isalmostwanting.\na.Thewords withdouble dualform areonly asmall minority (aquarter,\ninstead ofthree quarters,asinRV.); and half ofthem have only asingle\naccent, onthe final: thus, besides those inRV., bhavdrudrdu, bhavdfarvdu;\nugndvisnu, voc.,isofanomalous form. Thewhole number ofcopulativesis\nmore than double that inRV.\nb.The only proper neuter collectives, composedoftwo nouns, are\nfcepafmaprw, 'hairand beard', dnjandbhyanjandm,'salve andointment7\n,and\nkacipupabarhandm, 'mat and covering', unified because ofthevirtual unity\nofthetwo objects specified. Neuter singulars, used inasimilar collective\nway, ofadjective compounds, are(besides those inRV.): krtdkrtdm, 'what\nisdoneandundone' (instead of'what isdoneandwhat isundone'), cittdkutdm,\n'thought anddesire', bhadrapdpdm, 'good andevil', bhutabhavydm, 'past\nand future'.\n1257.Copulative compounds composedofadjectives\nwhich retain theiradjective character aremade inthesame\nmanner, butareincomparison veryrare.\nExamplesare: cuklakrsna, light and dark', sthalajaudaka,\n'terrestrial andaquatic', useddistributively ;andvrttapma, 'round\nandplump', cantanukula,'tranquil andpropitious', hrsitasragra-\njohina, 'wearingfresh garlands and freefromdust',nisekadi-\n"
+    },
+    {
+        "page": 447,
+        "content": "1262]COPULATIVE COMPOUNDS. 431\ncmacananta, 'beginning withconception andending with burial',\nused cumulatively.\nIntheVeda, theonlyexamples noted arethecumulative riilalohitd and\nistapurtd etc., used intheneut. sing, ascollectives(aspointed outabove),\nwith tamradhumrd, 'darktawny'; and the distributivedakainasavy.j, 'right\nand left', saptamastamd, 'seventh andeighth', andbhadrapdpd, 'good and\nbad' (beside thecorresponding neut.collective). Such combinations assat-\nydnrte,'truth andfalsehood', priyapriydni, 'things agreeable anddisagreeable',\nwhere eachcomponentisusedsubstantively, are,ofcourse, nottobeseparated\nfrom theordinary noun-compounds.\n1258. Inaccentuatedtexts, thecopulative compounds have\nuniformlytheaccent(acute) onthe final ofthestem.\nExceptionsareonly acase ortwo inAV., where doubtless thereading\nisfalse :thus, vdtaparjanyh (once:beside-nydyos), devamanusyhs (once),\nbrahmardjanyabhydin (once:part oftheMSS.).\n1259. Anexample ortwoaremetwith ofadverbialcopulatives: thus,\ndhardivi, 'daybyday', sdydmprdtar,'atevening andinthemorning'. They\nhave theaccent oftheirpriormember.\n1260. Repeated words. Inallagesofthelanguage,\nnouns andadjectives andparticlesarenotinfrequently repeated,\ntogiveanintensive, oradistributive, orarepetitional meaning.\nThough these arenotproperly copulative compounds,there isnobetter\nconnection inwhich tonotice them than here. They are, astheolder lan-\nguage shows,asort ofcompound, ofwhich thepriormember has itsown\nindependent accent, and theother iswithout accent. Thus :jahy esdrh\nvdrarh-varam, 'slay ofthem each best man'; dive-dive ordydvi-dyavi, -from\nday today'; prd-pra yajndpatim tira, 'make themaster ofthe sacrifice live\nonandon';bhuyo-bhuydh yvdh-fvah, 'further andfurther, tomorrow andagain\ntomorrow'; ekaydi-kayd,'with ineach case one'.\nSuch combinations arecalled bythenative grammarians dmredita,'re-\npeated1\n.\n1261. Finally maybenoticed inpassing thecompound numerals, eka-\ndafa, '11', dvdvihfati, '22', trtfpata, '103', cdtuhsahasram, '1004', and soon\n(476 if.),asaspecial andprimitiveclass ofcopulatives. They areaccented\nonthepriormember.\nII.Determinative Compounds.\n1262.Anoun oradjectiveisoften combined into a\ncompoundwith apreceding determiningorqualifying word\nanoun, oradjective,oradverb. Such acompoundis\nconvenientlycalled determinative.\n"
+    },
+    {
+        "page": 448,
+        "content": "432 XVIII. COMPOSITION.[1263\n1263. This istheclass ofcompounds which isofmost\ngeneralandfrequent occurrence inallbranches ofIndo-\nEuropean language.Itstwoprincipaldivisions havebeen\nalready pointed out: thus, A.Dependent compounds,in\nwhich thepriormember isasubstantive word (nounorpro-\nnoun orsubstantively usedadjective), standingtotheother\nmember inthe relation ofacase dependent011it;and\nB.Descriptive compounds,inwhich thepriormember is\nanadjective,orother word havingthevalue ofanadject-\nive,qualifyinganoun; orelseanadverb oritsequivalent,\nqualifyinganadjective.Each ofthese divisions then falls\nintotwo sub-divisions, accordingasthefinalmember, and\ntherefore thewhole compound,isanoun oranadjective.\nThewhole class ofdeterminatives iscalled bytheHindu\ngrammarians tatpurusa (theterm isaspecimenofthe class,\nmeaning'hisman');and theseconddivision, thedescriptives,\nhasthespecial name ofkarmadharaya (ofobscureapplication:\nthe literal sense issomethinglike'office-bearing').After their\nexample,thetwo divisions are inEuropean usage generally\nknown bythese twonamesrespectively.\nA.Dependent Compounds.\n1264.Dependent Noun-compounds.Inthis di-\nvision, thecase-relation ofthepriormember totheother\nmaybeofanykind; but,inaccordance with theusual re-\nlations ofonenoun toanother,itisoftenestgenitive, and\nleast often accusative.\nExamplesare :ofgenitive relation, devasenh,, 'army ofgods',\nyamadutd,'Yama's messenger', Jwaloka,'theworld oftheliving',\nindradhanus, 'Indra's bow', brahmagavi,'theBrahman's cow',\nvisagiri, 'poison-mount', mitralabha, 'acquisitionoffriends', mur-\nMacatani, 'hundreds offools', virasenasuta, 'Virasena'sson', ra-\njendra,'chief ofkings', asmatputras,'our sons', tadvacas,'his\nwords'; ofdative, padodaka,'water forthe feet', masanicaya,\n'accumulation foramonth'; ofinstrumental, atmasadrcya,'like-\nnesswith self'j dhanyartha,'wealth acquired bygrain', dharma-\npatrii,'lawfulspouse', pitrbandhu, 'paternalrelation'\n;ofablative,\napsarahsambhava, 'descent from anymph', madviyoga, 'separation\n"
+    },
+    {
+        "page": 449,
+        "content": "1267]DEPENDENT COMPOUNDS. 433\nfrom me', caurabhaya,'fear ofathief: oflocative, jalakrida,\n'sportinthewater', gramavcisa, 'abode inthevillage';ofac-\ncusative, nagaragamana, 'goingtothecity'.\n1265. Dependent Adjective-compounds.In\nthis division, onlyaverysmallproportionofthecompounds\nhave anordinary adjectiveasfinalmember; butusuallya\nparticipleoraderivative ofagencywith thevalue ofapar-\nticiple. Thepriormember stands inanycase-relation which\nispossibleintheindependentconstruction ofsuch words.\nExamplesare :oflocative relation, sthalipakva, 'cooked ina\npot', acvakovida, 'knowinginhorses', vayahsama,'alike inage',\nyudhisthira,'steadfast inbattle', tanucubhra, 'beautiful inbody';\nofinstrumental, matrsadrqa,'like hismother';ofdative, gohita,\n'goodforcattle'; ofablative, bhavadanya,'other thanyou',ta-\nrangacancalatara, 'more mobile than thewaves'; ofgenitive,\nbharatacrestha, 'best oftheBharatas',dvijottama, 'foremost ofBrah-\nmans' : withparticipial words, inaccusativerelation, vedavid,\n'Veda-knowing', annadd, 'food-eating', tanupUna, 'body-protect-\ning', saiyavadln, 'truth-speaking', pattragata, 'committed topaper'\n(lit'ly 'gonetoaleaf);ininstrumental, madhupu, 'cleansing\nwith honey', svaydmkrta, 'self-made', Indragupta, 'protected by\nIndra', vidyahma,'deserted by(=destituteof)knowledge' ; in\nlocative, hrdayavidh, 'piercingintheheart', rtvij, 'sacrificingin\ndue season', divicara, 'movinginthesky';inablative, rajya-\nbhrasta,'fallen from thekingdom', vrkabhita, 'afraid ofawolf;\nindative, caranagata,'come forrefuge'.\n1266.We takeupnowsome oftheprincipal groupsof\ncompounds falling under these two heads, inorder tonotice\ntheirspecialitiesofformation and use, their relative frequency,\ntheir accentuation, and soon.\n1267. Compounds havingasfinalmember ordinary nouns\n(such, namely,asdonotdistinctlyexhibit thecharacter ofverbal\nnouns, ofaction oragency)arequite common. Theyareregu-\nlarly andusuallyaccented onthefinalsyllable, without reference\ntotheaccent ofeither constituent. Examples were given above\n(1264).\na.Aprincipal exception with regard toaccent ispdti, 'master, lord',\ncompoundswithwhich usually retain theaccent ofthepriormember: thus,\nprajdpati, vdsupati, dtithipati, gdpati (comparetheverbal nouns inti,below,\n1274). But inafewwords pdti retains itsown accent: thus, vifpdti,\npa<;updti, etc.; andthemore generalrule isfollowed inapsarapati and\nvrajapatt (inAV.).\nWhitney, Grammar. 28\n"
+    },
+    {
+        "page": 450,
+        "content": "434 XVIII. COMPOSITION.[1267\nb.Other exceptionsaresporadic only:forexample, janardjan, deva-\nvdrman, hiranyatejas, prtanahdva, fakadhuma (but dhumd)', vacdstena.\nc.Theappearanceofacase-form insuchcompoundsisrare :examples\narevacdstena, ucca{h$rava$, uccafirghosa, durebhas(thethree lastinpossessive\napplication).\nd.Anumber ofcompoundsareaccented onbothmembers :thus,\nfdcipdti, sddaspdti, brhaspdti, vdnaspdti, rdthaspdti, jdspdti (also jdspati),\nndrdfdnsa, tdnundpdt (tanuasindependent word),\n1268. Thecompounds having anordinary adjectiveasfinal\nmember are(asalready noticed) comparatively few.\nSofarascanbegathered from thescanty examples occurring inthe\nolder language, they retain theaccent ofthepriormember: thus, gdvisthira\n(AV. gavtsthira), mdderaghu, yajnddhira, sdmavipra, tildmifra (but MYa); but\nkrstapacyd, 'ripeningincultivated soil'.\n1269. Theadjective dependent compounds havingasfinal\nmember thebare rootor,ifitend inashort vowel, gener-\nallywith anadded t areverynumerous inallperiodsof\nthelanguage,ashasbeenalready repeatedlynoticed(thus, 383if.,\n1147). Theyareaccented ontheroot.\na.Inaveryfewinstances, theaccent ofwords having apparentlythis\noriginisotherwise laid :thus, dnsatra, anarvfc, svdwrj.\nb.Before afinal root-stemappearsnotvery seldom acase-form :for\nexample, patarhgd, giravfdh, dhiyajtir, aksnayadruh, hrdispr?, divispf?, vanesdh,\ndivisdd, angesthd, hrtsvds, prtsutur, apsujd.\nC.Theroot-stem hassometimes amiddle orpassive value: forex-\nample, manoyuj, 'yoked (yoking themselves) bythewill', hrdayavfdh, 'pierced\ntotheheart'.\n1270. Compounds made with verbal derivatives ina,both\nofaction and ofagency,arenumerous, and take theaccent\nusually ontheir finalsyllable (asinthecaseofcompounds with\nverbalprefixes:1148.5).\nExamples are :hastagrabhd, 'hand-grasping', devavandd, 'god-praising',\nhaviradd, 'devouring the offering', bhuvanacyavd, 'shaking theworld', vra-\ntyabruvd, 'calling one's selfavratya1\n;aksaparajayd,'failure atplay', vasatkard,\n'utterance ofvashaf, goposd, 'prosperityincattle', angajvard, 'paininthe\nlimbs'.\na.Inafew instances, theaccent is(asincompoundswith ordinary\nadjectives: above, 1268)that ofthe prior member: thus, martidvrdha,\nsutekara (andothermore questionable words). And dugha, 'milking, yielding',\nissoaccented asfinal: thus, madhudugha, kamadtigha.\nb.Case-forms areespecially frequent inthepriormembers ofcompounds\nwithadjective derivatives ina,showing ^wna-strengtheningoftheroot: thus,\nforexample, abhayarhkard, yudhimgamd, dhanamjayd, purarhdard, vi^vam-\nbhard, divakard, talpefayd,divistambhd,\n"
+    },
+    {
+        "page": 451,
+        "content": "1273]DEPENDENT COMPOUNDS. 435\n1271. Compounds with verbal nouns andadjectivesinana\nareverynumerous, andhave theaccent always ontheradical\nsyllable (asinthecase ofcompoundswith verbalprefixes:\n1150. 1o).\nExamplesare :kefavdrdhana, 'hair-increasing', ayuspratdrana,'life-\nlengthening', tanupdna, 'body-protecting', devahedana, 'hatred ofthegods',\npurhsuvana, 'giving birth tomales'.\na.Avery fewapparent exceptionsasregards accent arereally cases\nwhere thederivative haslost itsverbal character :thus, yamasadand, 'Yama's\nrealm', achddvidhana, 'means ofprotection'.\nb.Anaccusative-form issometimes found before aderivative inana:\nthus, saruparhkdrana, ayaksmamkdrana, subhagarhkdrana, vanarhkdrana.\n1272. a.The action-nouns inya(1213)arenotinfre-\nquentincompositionasfinalmember, and retain theirown\nproperaccent(asincombination withprefixes).\nSufficient examples were given above (1213).\nb.Thesame istrue oftheequivalent feminines inyd:seeabove,\n1213d.\nc.Thegerundives inya(1213) hardly occur intheolder language in\ncombination with other elements thanprefixes. Thetworiivibharyh and\nprathamavasyb (thelatter adescriptive) have theaccent oftheindependent\nwords ofthesame form; balavijnayd anddfvabudhya (?)areinconsistent with\nthese andwith oneanother.\n1273. Compounds made with thepassive participleinfa\nornahave theaccent oftheirpriormember(asdothecombi-\nnations ofthesame words withprefixes: 1085a).\nExamplesare: hdstakrta, 'made with thehand', virdjata,'born ofa\nhero', ghosabuddha, 'awakened bynoise', prajapatisrsta, 'created byPrajapati';\nand,ofparticiples combined withprefixes, fndraprasuta, 'incited byIndra',\nbrhaspdtipranutta, 'driven away byBrihaspati', ulkdbhihata, 'struck bya\nthunderbolt'.\na.AV. hastheanomalous apsusamfita.\nb.Anumber ofexceptions occur,inwhich the final syllable ofthe\ncompound hastheaccent :forexample, agnitaptd, indrotd, pitrvittd, ratha-\nkrltd, kaviprafastd.\nc.One ortwospecial usages may benoticed. Theparticiple gata,\n'gone to',asfinal ofacompound,isused inaloose way inthelater lan-\nguage toexpressrelation ofvarious kinds :thus, jagatigata, 'existing inthe\nworld', tvadgata, 'belonging tothee', sakhlgata, 'relatingtoafriend',etc.\nTheparticiple bhuta, 'been, become',isused incomposition with anoun as\nhardly more than agrammatical device togiveitanadjective form :thus,\nidarh tamobhutam,'this creation, being darkness (existing inthecondition\nofdarkness)';tarnratnabhutarh lokasya, 'her, being thepearl oftheworld';\nand soon.\n28*\n"
+    },
+    {
+        "page": 452,
+        "content": "436 XVIII. COMPOSITION.[1273\nThe other participles donotoccur asfinals ofcompounds except inrare\nandsporadiccases: thus, gdutamabruvand, 'calling himself Gautama'.\n1274. Compounds with derivatives in tihave(likecom-\nbinations with theprefixes: 1157.Id)theaccent oftheprior\nmember.\nExamplesare :dhdnasati, 'winning ofwealth', stimapiti, 'soma-drinking',\ndevdhuti, 'invocation ofthegods', ndmaiikti, 'utterance ofhomage', havyddati,\n'presentationofofferings'.\nInnemddhiti, medhdsdti, vanddhiti(allRV.), theaccent oftheprior\nmember ischanged from penulttofinal.\nWhere theverbal character ofthederivative islost,thegeneral rule of\nfinal accent (1267)isfollowed: thus, devahetf, 'weapon ofthegods', deva-\nsumatf, 'favor ofthegods'.\nInsarvajyanf,'loss ofeverything', theaccent isthat ofcompounds with\nordinary nouns.\n1275. Compoundswith aderivative ininasfinalmember\nhave(asinallothercases)theaccent onthe in.\nThus, ukthafahstn, 'psalm-singing', vratacarm, 'vow-performing', rsabha-\ndaytn, 'bullock-giving', satyavadfn, 'truth-speaking', fronipratodfn, 'thigh-\npounding'.\n1276. There isagroupofcompounds with derivatives in\n,havingtheaccent onthepenultorradicalsyllable.\nThus, pathirdksi. 'road-protecting', havirmdthi, 'sacrifice-disturbing',\natmadusi, 'soul-harming', pathisddi, 'sitting inthepath', sahobhdri, 'strength-\nbearing', vasuvdni, 'winning good things', dhanasdni, 'gaining wealth'; and,\nfrom reduplicated root, urucdkri, 'making room'.\nCompounds with -sdni and-vdni areespecially frequent inVeda and\nBrahmana. Asindependent words, nouns,these areaccented sanfandvani.\nCombinations with prefixes donotoccur insufficient numbers toestablish a\ndistinctrule, buttheyappeartoheoftenest accented onthe suffix. (1155. 2d).\nFromyhanaremade incomposition -ghni and-ghrii, with accent on\ntheending:thus, sahasraghni, ahighni, fvaghni; -dhifromydha (1155. 2e)\nhastheaccent initsnumerous compounds:thus, isudhf, garbhadhf, puchadM:\n6sadhi andptiramdhi areprobahly ofother derivation.\n1277. Compounds with derivatives invanhave(likecom-\nbinations withprefixes: 1169.Ib)theaccent ofthe finalmem-\nber namely ,ontheradicalsyllable.\nThus, somapdvan, 'soma-drinking', baladdvan, 'strength-giving', papa-\nfcrtuan, 'evil-doing', bahusuvan, 'much-yielding', talpafivan, 'lying ona\ncouch', drusddvan, 'sitting onatree', agretvari, f.,'goingatthehead'.\na.Theaccent ofthe(somewhat obscure) words matarfyvan andmata-\nribhvan isanomalous.\nb.Thefewcompoundswith finalmanappeartofollow thesame rule\n"
+    },
+    {
+        "page": 453,
+        "content": "12801 DESCRIPTIVE COMPOUNDS. 437\nasthose with van :thus, svaduksddman, 'sharing out sweets', at;uheman,\n'steed-impelling'.\n1278. Compounds with other derivatives, ofrare orsporadic occurrence,\nmaybebriefly noticed :thus,inw,rastradipsu, devaplyti, govindti, vanargu (?j:\ncompare 1178d;innuortnu, lokakrtnu, surupakrtnti:compare 1196;\nintr,nrpatr, mandhatf, haskartr(vasudhdtaras, AV.,isdoubtless afalse read-\ning). The derivatives inasareofinfrequent occurrence incomposition (as\nincombination withprefixes: above, 1151.4), andappeartobetreated as\nordinary nouns: thus, yajnavacds (but hiranyatejas, AV.).\nB.Descriptive Compounds.\n1279. Inthis division oftheclass ofdeterminatives,\nthepriormember stands totheother innodistinct case-\nrelation, butqualifiesitadjectivelyoradverbially,accord-\ningasit(thefinalmember isnoun oradjective.\nExamplesare :nilotpala,'bluelotus', sarvagima,'allgood\nquality', priyasakha,'dear friend', maharsi, 'great sage', rajata-\npatrd,'silvercup'; d/nata, 'unknown', sukrta, 'well done', duskrt,\n'ill-doing', purustutd, 'muchpraised', punarnava, 'renewed'.\na.The priormember isnotalways anadjective before anoun, and\nanadverb before anadjective ;other partsofspeech aresometimes used ad-\njectively andadverbially inthatposition.\nb.Theboundary between descriptive anddependent compoundsisnot\nanabsoluteone; incertain cases itisopentoquestion,forinstance, whether\naprior noun, oradjective with noun-value,isusedmore inacase-relation,\noradverbially.\nc.Moreover, where the finalmember isaderivative having bothnoun\nandadjective value,itisnotseldom doubtful whether anadjective com-\npoundistoberegardedasdescriptive, made with final adjective,orpos-\nsessive, made with, final noun. Sometimes theaccent oftheword determines\nitscharacter inthisrespect, butnotalways.\nAsatisfactorily simple andperspicuousclassification ofthedescriptive\ncompoundsisnotpracticable ;wecannot hold apart throughout thecom-\npounds ofnoun and ofadjective value, butmay better group both together,\nasthey appear with prefixed elements ofvarious kinds.\n1280. Thesimplestcase isthat inwhich anoun as\nfinalmember ispreceded byaqualifying adjectiveasprior\nmember.\nInthiscombination, bothnoun and adjective may beof\nanykind. The accent is(asinthecorrespondingclass ofde-\npendent noun-compounds:1267)onthefinalsyllable.\nThus, ajnatayaksma,'unknown disease', mahadhand, 'great\n"
+    },
+    {
+        "page": 454,
+        "content": "438 XVIII. COMPOSITION.[1280\nwealth', krmacakuni, 'blackbird', daksinagni, 'southernfire',\nuruksiti, 'wide abode', adharahanu, 'lowerjaw', ttarajand, 'other\nfolks', sarvatmctn, 'wholesoul', ekavird, 'solehero', saptarsi,\n'seven sages', ekonavihcati, 'ascore diminished byone', jagrat-\nsvapnd, 'waking sleep', yavayatsakhd, 'defendingfriend'.\na.There aresome exceptionsasregards accent. Especially, compounds\nwith vfyva (incomposition, accentedvifvd), which itself retains theaccent:\nthus, v'ifvddevas,'allthegods', vifvdmanusa, 'every man'. Forwords inti,\nseebelow, 1287d. Sporadiccases aremadhydndina, vrsdkapi, both of\nwhich show anirregular shift oftone inthepriormember;andafewothers.\nb.Instead ofanadjective, thepriormember isinafew\ncases anoun usedappositionally,orwith aquasi-adjective value.\nThus, rajayaksmd, 'king-disease', brahmarsi, 'priest-sage', ra/arsi,\n'king-sage', rajadanta, 'king-tooth', devajana, 'god-folk', duhitr-\njana, 'daughter-person', qamilata, 'creeper named cami\\ Jayacabda,\n'theword \"conquer\"', ujjhitacabda,'theword \"deserted\"'.\nThis groupisofconsequence, inasmuch asinpossessive applicationit\nisgreatly extended, andforms anumerous class ofappositional compounds:\nseebelow, 1302.\nC.Thiswhole subdivision,ofnouns withpreceding qualifying adjectives,\nisnotuncommon; but itisgreatly (inAV., forexample, more than five\ntimes) exceeded infrequency bythesub-class ofpossessivesofthesame\nform: seebelow, 1298.\n1281. The adverbial words which aremostfreely and\ncommonlyused aspriormembers ofcompounds, qualifying\nthefinalmember, aretheverbalprefixes andthewords of\ndirection related with them, and theinseparable prefixes,\naoran, su,dus,etc.(1121). These arecombined notonly\nwithadjectives,butalso,inquasi-adjectival value, with\nnouns;andthetwo classes ofcombinations will bestbe\ntreatedtogether.\n1282.Verbal adjectives andnouns withpreced-\ningadverbs. Asthelargest andmost importantclass under\nthisheadmight properly enough beregardedthederivatives with\nprecedingverbalprefixes. These, however, have been here\nreckoned rather asderivatives from roots combined withprefixes\n(1141),andhave been treated under thehead ofderivation, in\nthepreceding chapter.Intaking uptheothers, wewillbegin\nwith theparticiples.\n1283. Theparticiples belongingtothetense-systems\nthose inant(or at],mana, ana, vans areonly very rarely\n"
+    },
+    {
+        "page": 455,
+        "content": "1285]DESCRIPTIVE COMPOUNDS. 439\ncompoundedwithanyother adverbial element than thenegativea\noran,which then takes theaccent.\nExamplesare :dnadant, ddadat, dnapnant, dsravant, dlubhyarit, ddasyant,\ndditsant, ddevayant ;dmanyamdna, dhihsana, dchidyamana; ddadivdns,dbi-\nbhwdns; and, with verbal prefixes, dnapasphurant, dnagamisyant, dvirddh-\nayant, dvicacalat, dpratimanyuyamana.\na.Exceptionsinregard toaccent arevery few: arundhati, ajdranti,\nacoddnt (RV., once: doubtless afalse reading;thesimple participleiscvdant};\nAV. hasanipddyamdnaforRV.dnipadyamdna (and thepublishedtexthas\nasarhydnt,with apartoftheMSS.).\nb.Ofother compoundsthan with thenegative prefix have been noted\nintheVeda -punardiyamana (indpunard-) and stividvahs. Inalaldbhdvant\nandjanjanabhdvant (RV.),asinastamydntandastamesydnt (AV.), wehave\nparticiplesofacompound conjugation (1091),inwhich,ashasbeen pointed\nout,theaccent isasincombinations with theverbalprefixes.\n1284. Thepassive (orpast) participleintaornaismuch\nmore variously compounded, and ingeneral (asinthecase of\ntheverbal prefixes:1085a)theprecedingadverbial element has\ntheaccent.\nThus, with thenegative aoran(byfarthemostcommoncase):dkrta,\nddabdha, arista, dnadhrsta, dparajita, dsamkhyata, dnabhydrudha ; with\nSM, stijata, suhuta, stisarhfita, svararhkrta; withdus, dtiecarita, dtirdhita\nanddurhita; with other adverbial words, ddhsujuta, ndvajata, sdnafruta,\nsvaydmkrta, trtpratisthita :dramkrta andkakajakrta areratherparticiples of\nacompound conjugation.\na.Exceptionsinregard toaccent are :withaoran,anapasta, apra-\nyastd, and, with theaccent oftheparticipleretracted totheroot, amrfa,\nadfsta, acftta, aytita 'myriad', aturta (beside dturta], asurta(?beside sGrfa);\nwith su(nearlyhalf asnumerous astheregular cases), subhutd, suktd,\nsupracastd, svdkta, sukrtd andsujatd (beside sukrta andsujatd), and afew\nothers; with dus(quiteasnumerous astheregular cases), duritd, duruktd,\nduskrtd, durbhutd; withsa,sajatd; with other adverbs, amotd, aristutd,\ntuvijatd, pracinopavitd, andthecompounds withpwm, purujatd, puruprajatd,\npuruprafastd, purustutd,etc.Theproper name asadhd stands beside dsadha;\nandAV. hasabhinnd forRV. dbhinna.\n1285. Thegerundivesoccur almostonlyincombination\nwith thenegative prefix, andhaveusuallytheaccent onthe\nfinalsyllable.\nExamplesare :anapyd, anindyd, asahyd, ayodhyd, amokyd ,advisenyd ;\nahnavayyd; and, along with verbalprefixes,thecases areasamkhyeyd, a-\npramrsyd, anapavrjyd, anatyudyd, anadhrsyd, avimokyd, ananukrtyd (the\naccent ofthesimple word being sarhkhyeya etc.).\na.Exceptions inregardtoaccent are :dnedya, ddabhya, dgohya, djosya,\ndyabhya. Thetwoanavadharsyh andanativyddhya (bothAV.) belong tothe\n"
+    },
+    {
+        "page": 456,
+        "content": "440 XVIII. COMPOSITION.[1285\ni/a-division (1213 b)ofgerundives, andhave retained theaccent ofthe\nsimpleword. Anddghnya andaghnydoccur together.\nb.The onlycompounds ofthese words with other adverbial elements\ninV.aresuyabhya (accentedlike itstwindyabhya) andprathamavasyh\n(which retains the finalcircumflex), andperhaps ekavadyd.\nG.Theneuter nouns ofthesame form(1213 C:except sadhdstutya]\nretain theirown accent after anadverbialpriormember:thus, pwrvapdyya,\npurvapeya, amutrabhuya ;andsaha^eyya. And thenegatived gerundives\ninstanced above arecapable ofbeing viewed aspossessive compounds with\nsuch nouns.\nSome oftheother verbal derivatives which have rules of\ntheirown as.toaccent etc.maybenext noticed.\n1286. The root-stem(pure root, orwith tadded after a\nshort finalvowel :1147)isveryoften combined with apreced-\ningadverbial word, ofvarious kinds;and inthecombination\nitretains theaccent.\nExamples are :with inseparable prefixes, adruh,'notharming', asw, 'not\ngiving birth', arfie,'notshining'; sukft, 'well-doing', sucrtit, 'hearing well';\nduskft, 'ill-doing', dudd? (199 b), 'impious'; sayuj, 'joining together',\nsamdd, 'conflict'; sahajd,'borntogether', sahavdh, 'carrying together' ;with\notheradverbs, amajur, 'growing oldathome', wpcm'sprp, 'touching upward',\npunarbhu, 'appearing again', pratarytij, 'harnessedearly', sadyahkn, 'bought\nthesameday', sakarhvrdh, 'growing uptogether', sadamdf, 'ever-binding',\nvisuvft, 'turningtobothsides', vrthasdh, 'easily overcoming'; with ad-\njectives used adverbially, uruvydc, 'wide-spreading', prathamajd, 'first-born',\nraghusydd, 'swift-moving', navasu, 'newly giving birth', ekajd, 'only born',\npwferopfp, 'brightly adorned', dvijd,'twice born', trivft, 'triple', svardj,'self-\nruling'; with nouns used adverbially, pam&ftCt, 'beneficent', suryavvtt,\n'shining likethesun', ifanakft, 'actingaslord', svayambhtt, 'self-existent';\nand, with accusative case-form, patamgd, 'going byflight'.\na.When, however,aroot-stem isalready incomposition, whether with\naverbal prefixoranelement ofother character, thefurther added negative\nitself takes theaccent(asincase ofanordinary adjective: below, 1288 a):\nthus, forexample, dnavrt, 'notturning back', dvidvis,'notshowing hostility',\ndduskrt,'notill-doing', dna^vada, 'not giving ahorse', dpa?uhan, 'not\nslayingcattle' (andgds would beanexception,ifitcontained}/ga:which is\nvery unlikely). Similar combinations with suseem toretain theradical ac-\ncent :thus, suprattir, svabhu, svaytij:svavrjisanunsupported exception.\nb.Afewother exceptions occur, mostly ofdoubtful character,assadh-\ndstha, ddhrigu, and thewords having anc asfinalmember (407ff.: ifthis\nelement isnot,afterall,asuffix).\n1287. Other verbal derivatives, requiringtobetreated\napart from thegeneral bodyofadjectives,arefewand ofmi-\nnorimportance. Thus :\n"
+    },
+    {
+        "page": 457,
+        "content": "1288]DESCRIPTIVE COMPOUNDS. 441\n.a.The derivatives inaareingreat partofdoubtful character, because\nofthepossibilityoftheir being used -with substantive value tomake apos-\nsessive compound. The least ambiguous, probably,arethederivatives from\npresent-stems (1148. 3b),which have theaccent onthesuffix:thus, asunvd,\napa$yd, aksudhyd, avidasyd, anamrnd, sadaprnd, punarmanyd ;andwith\nthem belong such cases asatrpd, avrdhd, ararhgamd, urukramd, evavadd,\nsatrasahd, punahsard, purahsard; and thenouns myambhavd, sahacard,\npratahsavd, mithoyodhd. Differently accented, ontheother hand, although\napparentlyofthesameformation,aresuch asdnapasphura, dnavahvara\n(compare thecompounds noticed at1286a),saddvrdha, sUbharva, nyagrodha,\npurodafa, sadhamada, and others. Words like addbha, durhdna, sukdra,\nareprobably possessives.\nb.Thederivatives invankeepingeneral theaccent ofthefinalmember,\nonthe root (compare 1169. Ib,1277): thus, apupatvan, 'swift-flying',\npuroyavan, 'going infront', sukrtvan, 'well-doing'; andsutdrman andsuvdhman\nareprobablytobeclassed with them. Butthenegative prefix hastheaccent\neven before these :thus, dyajvan, dravan, dprayutvan ;andsatydmadvan\n(ifitbenotpossessive) hastheaccent ofitsprior member.\nc.Afewwords in iseem tohave(asindependent compounds:1276)\ntheaccent ontheradical syllable:thus, durgfbhi, rjuvdni, tuvisvdni.\nd.The derivatives intiarevariously treated :thenegative prefix has\nalways theaccent before them: asdcitti, dbhuti, dnahuti; with suand dus,\nthecompoundisaccented now ontheprefix andnowonthefinal, andin\nsome words onboth(suniti andsuniti,dustuti anddustutf};with other ele-\nments,theaccent oftheprefix prevails: thus, sdhuti, sadhdstuti, pwtihiti,\npurvdplti, purvydstuti.\ne.The derivatives ininhave,asingeneral,theaccent onthesuffix:\nthus, purvasfn, bdhucarfn, sadhudevtn, savasfn, kevaladtn. But, with the\nnegative prefix, tmamm, dvitarin.\nf.Other combinations aretoovarious intreatment, orarerepresented\nbytoofewexamples inaccentuated texts, tojustify the setting upofrules\nrespecting them.\n1288. Oftheremaining combinations,thosemade with the\ninseparable prefixes form insome measure aclassbythemselves.\na.Thenegative prefixaoran,when itdirectly negatives\ntheword towhich itisadded, hasavery decided tendencyto\ntake theaccent.\nWehave seenabove (1283)that itdoes soeven inthecase ofpresent\nandperfect andfutureparticiples, although these incombination with averbal\nprefix retain theirown accent (1085); and alsointhecase ofaroot-stem,\nifthisbealready compounded withanother element (1286 a).And thesame\nistrue ofitsother combinations.\nThus, with various adjective words :dtandra, ddabhra, ddafuri, dnrju,\nddevayu, dtrsnaj, dtavyans, anamm, ddvayavin, dpracetas, dnapatyavant,\ndnupadasvant, dpramayuka, dmamri, dprajajni, dvidldhayu, dnagnidagdha,\n"
+    },
+    {
+        "page": 458,
+        "content": "442 XVIII. COMPOSITION.[1288\ndkdmakar$ana, dpafcdddaghvan. Further, with nouns, dpati. dkumdra,\ndbrdhmana, dvidyd, d?raddhd, dvrdtya.\nBut there areanumber ofexceptions,inwhich theaccent isonthe\nfinal syllable, without regardtotheoriginal accentuation ofthe finalmember:\nthus,forexample, acitrd, a$rird, aviprd, ayajniya, andsmdkd, asthurf, andpu,\najarayu, andmayitnu; andinamftra, 'enemy', and avira, 'unmanly', there\nisaretraction oftheaccent from the final syllable ofthe finalmember to\nitspenult.\nb.Theprefixessuand clushave thistendencyinamuch\nlessdegree, and theircompounds arevery variously accented,\nnowontheprefix, nowonthefinalsyllable, nowontheaccented\nsyllableofthe finalmember.\nThus, forexample, subhadra, stivipra, supakva, subrdhmana.stibhisaj ;\nsutirthd, suvasand, susdratht, supdfd, sucitrd; sitfeva, suhdtr :suvira islike\navira; durmitrd, dusvapnya; andduchund (168), with irregular retraction\nofaccent (fund).\nC.Thecompounds with saaretoofew tofurnish occasion forseparate\nmention;andthose with theinterrogative prefixinitsvarious forms arealso\nextremely rare intheVeda :examples arekucard, katpayd, kdbandha, kunan-\nnamd, kumdrd, kuyava, kusdva.\n1289. Theverbalprefixesaresometimes used inageneral\nadverbial way, qualifyingafollowing adjectiveornoun.\na.Examplesofsuchcombinations arenotnumerous intheVeda. Their\naccentuation isvarious, though thetone rests oftenest onthepreposition.\nThus, ddhipati, 'over-lord', dparupa, 'mis-form', prdti?atru, 'opposing foe',\nprdpada,'fore partoffoot', prdnapdt, 'great-grandchild', vtpakva, 'quite\ndone', sdmpriya, 'mutually dear'; upajfhvikd, 'side tongue' (with retraction\noftheaccent ofjihva)] antarde$d, 'intermediatedirection', pradfv, 'forward\nheaven', prapitdmahd (alsoprdpitdmaha], 'great-grandfather', pratijand, 'oppo-\nnent', vyadhvd, 'midway'. These compounds aremore frequent withpossess-\nivevalue(below, 1305).\nb.This useoftheverbalprefixesismorecommonlater, andsome of\nthem have aregular value insuchcompounds. Thus,atidenotes excess,\nasinatidura, 'very far', atibhaya, 'exceeding fear', dtipurusa (QB.),'chief\nman'; adhi, superiority,asinadhidanta, 'upper tooth', adhistrl,'chief woman';\nasignifies 'somewhat',asinakutila, 'somewhatcrooked', dnila, 'bluish';\nupadenotes something accessory orsecondary, asinupapurdna,'additional\nPurana'; pari, excess,asinparidurbala, 'very weak'; prati, opposition,as\ninpratipaksa, 'opposing side', pratipustaka, 'copy'; vi,variation orexcess,\nasinvidura, 'very far', vipdndu, 'greyish', viksudra, 'respectively small';\nsam, completeness,asinsampakva, 'quite ripe'.\n1290. Other compoundswith adverbialprior members arequite irreg-\nularly accented.\nThus, thecompounds with puru, onthe final(comparetheparticiples\n"
+    },
+    {
+        "page": 459,
+        "content": "12931 SECONDARY\nwith puru, 1284 a;; as,purudasmd, purupriyd, punifcandrd; those with\npunar, ontheprior member,aspunarnava, punarvasu ;those with satds t\nsatlnd, satyd,thesame, assatomahant, satindmanyu, satydmugra;afew\ncombinations ofnouns intrandanawith adverbs akin with theprefixes,\nonthe finalsyllable, aspuraetf, purahsthatf, uparifayand, prdtahsavand ;\nandmiscellaneous cases aremithdavadyapa, hdrifcandra, dlpa$ayu, sadhvaryd,\nandyachresthd andyavdchresthd.\n1291. Theadjective purvaisinthe later languagefre-\nquently used asfinalmember ofacompoundinwhich itslogical\nvalue isthat ofanadverbqualifyingtheothermember(which\nissaid toretain itsownaccent). Thus, drstapurva, 'previously\nseen', parmitapurva, 'already married', somapitapurva, 'having\nformerly drunk soma', stripurva, 'formerlyawoman'.\nIII.Secondary Adjective Compounds.\n1292.Acompound havinganoun asitsfinalmember\nveryoften winssecondarilythevalue ofanadjective, being\ninflected inthethreegenderstoagreewith thenounwhich\nitqualifies, andused inalltheconstructions ofanadjective.\nThis class ofcompounds,aswaspointedoutabove\n(1247. III.),falls into thetwo divisions ofA.Possessives,\nhavingtheiradjectivecharactergiven them byaddition of\ntheidea of'possessing'; andB.those inwhich thefinal\nmember issyntactically dependent onorgoverned bythe\npriormember.\nA.Possessive Compounds.\n1293. Thepossessivesarenoun-compoundsofthepre-\nceding class, determinatives, ofallitsvarious subdivisions,\ntowhich isgiven anadjective inflection, andwhich take\nonanadjective meaningofakind which ismost conve-\nniently andaccuratelydefined byadding 'having'or'pos-\nsessing'tothemeaningofthedeterminative.\nThus: thedependent suryatejds,'sun's brightness', becomes\nthepossessive suryatejas, 'possessingthebrightnessofthesun';\nya/nakamd, 'desire ofsacrifice', becomes yajndkama, 'having de-\nsire ofsacrifice'; thedescriptive brhadratha, 'great chariot', be-\n"
+    },
+    {
+        "page": 460,
+        "content": "444 XVIII. COMPOSITION.[1293\ncomes thepossessive brhddratha, 'having great chariots', dhasta,\n'not hand', becomes ahastd, 'handless', durgandhi,'illsavor',\nbecomes durgdndhi,'of illsavor'; and soon.\na.Compoundsofthecopulativeclass donotappeartoassume apos-\nsessive value unless(ashasbeen suggested with someplausibility) we\nare toregard those ofsingular collective form asneuters ofsecondary ad-\njectives.\nb.Thename given bythenative grammarianstothepossessive com-\npoundsisbahuvrihi: theword isanexampleoftheclass, meaning 'possess-\ningmuch rice'.\nc.Thename\"relative\",instead ofpossessive, sometimes appliedto\nthisclass,isevidently not tobeapproved ;since, though themeaning of\nsuch acompound (asofany attributive word)iseasily cast into arelative\nform,itsessential character lies inthepossessive verbwhich hasnevertheless\ntobeadded,orinthepossessive case oftherelative which must beused:\nthus, brhddratha, 'whohas agreat chariot',or'whose isagreat chariot'.\n1294. That anoun, simpleorcompound, should beadded toanother\nnoun,inanappositive way, with avalue virtually attributive, andthatsuch\nnouns should occasionally gainbyfrequent association andapplication an\nadjective formalso,isnatural enough, andcommon inmany languages ;the\npeculiarity oftheSanskrit formation lies intwo things. First,thatsuch\nuseshould have become aperfectly regular andindefinitely extensible one\ninthe case ofcompounded words,sothatanycompoundwith noun-final\nmaybeturned without alteration intoanadjective, while toasimple noun\nmust beadded anadjective-making suffix inorder toadaptittoadjective\nuse :forexample, thatwhile hastamustbecome hastin andbdhumustbecome\nbahumant, hiranyahasta andmahabahu change fromnoun toadjective value\nwith noadded ending. And second, that therelation ofthequalified noun\ntothecompound should havecome tobesogenerally that ofpossession, not\noflikeness, norofappurtenance, norofanyother relation which isasnat-\nurally involved insuch aconstruction: thatwemay only say,forexample,\nmahabahuh purusah, 'manwith great arms', andnotalsomahabahur manih,\n'jewel foragreat arm', ormahabahavah cakhah,'branches like great arms'.\nThereis,however,intheolder language alittlegroupofderivative\nadjectives involving the relation ofappurtenance rather than ofpossession:\nthus, vifvdnara,'oforfor allmen, belonging toall';and sovifvdkrsti, vif-\nvdcarsani; also vi?vdfarada,'ofevery autumn', vipathd,'forbadroads',\ndvirajd, '[battle]oftwokings'. Andsome oftheso-called duigrw-compounds, as\ndvigu itself intheadjective sense 'worth two cows', dvinau, 'boughtfortwo\nships' (1312), maybecompared withthem(comparealsothewords inana,\n1296, end).Itisatleast opentoquestion whether these arenotsurvivals\nofanapplication independent ofthepossessive, and originally ofanequal\nrank withit,rather than instances ofapossessivemodified toitsopposite.\n1295. Thepossessive compoundisdistinguished from its\nsubstrate, thedeterminative, generally byadifference ofaccent.\n"
+    },
+    {
+        "page": 461,
+        "content": "1298]POSSESSIVE COMPOUNDS. 445\nThis difference isnotofthesame nature inallthedivisions of\nthe class\n;butoftenest, thepossessive hasasacompound the\nnatural accent ofitspriormember(asinmost oftheexamples given\nabove).\n1296.Possessivelyuseddependent compounds,orpos-\nsessivedependents,areverymuch lesscommon than those\ncorrespondingtotheother division ofdeterminatives.\nFurther examplesare :mayuraroman, 'havingtheplumesof\npeacocks', agnitejas, 'havingthebrightnessoffire', jnatimukJm ,\n'wearingtheaspectofrelatives', pdtikama, 'desiringahusband',\nhastipada, 'having anelephant'sfeet'.\nTheaccentis,asintheexamples given, regularly that ofthe prior\nmember, andexceptions arefew arid ofdoubtful character. Afewcompounds\nwith derivatives inanahave theaccent ofthefinalmember: thus, indrapdna,\n'serving asdrink forIndra', devasddana, 'serving asseat forthegods',\nrayisthdna, 'being source ofwealth'; butthey contain noimplication of\npossession, and areperhapsincharacter,asinaccent (1271), dependent.\nAlso afew inas, asnrcafcsas, 'men-beholding', nrvdhas, 'men-bearing',\nksetrasddhas, 'field-prospering',areprobablytobejudged inthesame way.\n1297.Possessively useddescriptive compounds,orpos-\nsessivedescriptiyes,areextremely numerous and ofevery\nvarietyofcharacter; andsome kinds ofcombination which\narerare inproper descriptiveuseareverycommon aspos-\nsessives.\nTheywillbetaken upbelow inorder, accordingtothe\ncharacter oftheprior member whether thenoun-final be\npreceded byaqualifying adjective,ornoun, oradverb.\n1298. Possessive compoundsinwhich anoun ispreceded\nbyaqualifying ordinary adjectiveare(aspointed outabove,\n1280c)verymuch morecommon thandescriptivesofthesame\nform.\na.They regularly andusually have theaccent oftheirprior\nmember: thus, anydrupa,'ofother form', ugrabahu, 'having pow-\nerful arms', jlvdputra, 'having living sons', dirghdcmacru, 'long-\nbearded', Irhdchravas, 'ofgreat renown', bhurimula, 'many-rooted',\nmah&vadha, 'bearingagreat weapon', vicvdrupa, 'havingallforms',\npukrdvarna,'ofbright color', qivalhimarcana,'ofpropitious touch',\nsatydsamdha,'oftruepromises', sdrvanga, 'whole-limbed', svdya-\ncas,'having ownglory', hdritasraj, 'wearing yellow garlands'.\nb.Exceptions, however, inregard toaccent arenotrare(aseventh or\neighth ofthewhole number, perhaps). Thus, theaccent issometimes that\n"
+    },
+    {
+        "page": 462,
+        "content": "446 XVIII. COMPOSITION.[1298\nofthe finalmember; especiallywith derivatives inas,astuvirddhas, puru-\npefas, prthupdksas,andothers inwhich(asabove, 1296, end)adeterminative\ncharacter maybesuspected:thus, urujrdyas beside urujrf, uruvydcas beside\nuruvydc, and soon;butalsowith those ofotherfinal,asrjuhdsta, fitikdksa,\nkrsnakdrna, citradrpika/ tuvifusma, rjukrdtu, prthupdrpu, puruvdrtman, raghu-\nydman, vldupdtman. Inavery fewcases, theaccent isretracted from the\nfinal tothe first syllable ofthesecond member :thus, ahhubheda, tuvignva,\nyuruvlra, purm-tipa, fitibdhu (also ?itibahti). The largest class isthat of\ncompounds which take theaccent upontheir final syllable (inpart,ofcourse,\nnotdistinguishable from that ofthose which retain theaccent ofthe final\nmember):forexample, bahvannd, riilanakhd, puruputrd, vifvangd, svapatf,\ntuviprati, pr^niparni f.,darfafaprf, putirajju, asitajnti, prthugmdn, bahuprajds.\nc.The adjective vfyva,'all1\n,aspriormember ofacompound (and also\ninderivation), changesitsaccent regularly tovifvd; sdrva, 'whole, all',does\nthesame inafew cases.\n1299. Possessive compounds with aparticiple preceding\nandqualifyingthefinalnoun-member arenumerous, although\nsuch acompoundwith simple descriptive value isalmost un-\nknown.\nThe accentis,with fewexceptions,that oftheprior\nmember.\na.Theparticipleisoftenest the passive one, intaorna. Thus,\nchinndpaksa,'with severed wing', dhrtdrdstra, 'offirmly heldroyalty',hatd-\nwiafr, 'whose mother isslain', iddhdgni, 'whose fire iskindled', uttandhasta,\n'with outstretched hand', prdyatadaksina, 'having presentedsacrificialgifts';\nand, withprefixed negative, dristavlra, 'whose men areunharmed', dtapta-\ntanu,'ofunburned substance', dnabhimlatavarna,'ofuntarnished color'.\nExceptionsinregardtoaccent arevery few: there have been noticed\nonly paryastaksd, vyastakeft L,achinnaparnd.\nb.Afewexamples occur ofapresent participleinthesame situation.\nInabout halftheinstances,itgivesitsownaccent tothecompound:thus,\ndyutddyaman, dhrsddvarna etc., pucddratha, riifadvatsa etc., bhrdjajjanman\netc., samyddvira, standyadama, sadhadisti;intheothers, theaccent isdrawn\nforward tothe final syllable oftheparticiple (asinthecompounds with\ngoverning participle:below, 1309): thus, dravatpanietc. (dravdt alsooccurs\nasadverb), rapfddudhan, svanddratha, arcdddhuma, bhandddisti, krandddisti.\nWith these lastagrees inform jarddasti, 'attaining oldage, long-lived'; but\nitsmake-up,inview ofitsmeaning,isaromalous.\nc.TheRV. hastwocompounds with theperfect middleparticipleas\npriormember: thus, yuyujandsapti,'with harnessed coursers' (perhapsrather\n^having harnessed theircoursers'), anddadrfandpavi (with regular accent,\ninstead ofdddrfana, aselsewhere irregularly inthisparticiple),'with con-\nspicuous wheel-rims'.\nd.Ofanearly participialcharacter istheprior element in?rtitkarna\n(RV.),'oflistening ear'; andwith this areperhaps accordant didyagni and\n(RV., each once).\n"
+    },
+    {
+        "page": 463,
+        "content": "13021 POSSESSIVE COMPOUNDS. 447\n1300. Possessive compounds havinganumeral asprior\nmember areverycommon, and forthemostpartfollow the\nsame rule ofaccent which isfollowed bycompoundswith other\nadjectives: exceptedareingeneralthose beginningwith dviand\ntri,which accent the finalmember.\na.Examples with other numerals than dviand triare :ekacakra, eka-\nCirsan, ekapad, cdturanga, cdtuspaksa, pdncanguri, pdncaudana, sdda$va,\nsdtpad, saptdjihva, saptdmatr, astapad, astdputra, ndvapad, ndvadvara,\nddfafakha, ddfdfirsan, dvadafara, trih$ddara, fatdparvan, fatddant, sahds-\nranaman, sahdsramula.\nb.Exceptionsinregardtoaccent arebutfew,andhave thetoneon\nthe final syllable, whatever may bethat belonging originallytothe final\nmember; they aremostly stems infinal a,used bysubstitution forothers\ninan, i,oraconsonant: thus, caturaksd etc.(aksdn ordksi: 431), sadahd\netc. (ahanordhar :430 a),dafavrsd etc. (vfsan), ekaratrd etc. (rdtri or\nrdtrl), ekarcd etc. (re); butalsoafewothers, assadyogd, astayogd, $atarghd,\nsahasrarghd, ekapard (?).\nC.Thecompoundswith dviand triforthemost parthave theaccent\noftheir finalmember :thus,forexample, dvijdnmcm, dvidhara, dvibdndhu,\ndvivartanf, dvipdd, tritdntu, trinabhi, trifdka, trivdrutha, tricakrd, triflrsdn,\ntripdd. Anumber ofwords, however,follow thegeneral analogy, andaccent\nthenumeral: thus,forexample, dvfpaksa, dvfyavas, dvyhsya, trisandhi,\ntry&ra, tryhfir, andsometimes dvCpad andtrfpad inAV. Asintheother\nnumeral compounds,asubstituted stem inaisapttotake theaccent on\nthefinal :thus, dvivrsd andtrivrsd, dvirajd, dviratrd, tryayusd,tridivd;and\nafew ofother character with trifollow thesame rule :thus, trikafd, trinakd,\ntribandhu, tryudhdn, tribarMs,etc.\nd.Theneuter,oralso thefeminine, ofnumeral compoundsisoften\nused substantively, with acollective orabstract value, andtheaccent isthen\nregularly onthefinal syllable: seebelow, 1312.\n1301. Possessive compounds havingaspriormember anoun\nwhich hasaquasi-adjectivevalue inqualifyingthefinalmem-\nberarevery frequent, andshow certainspecialitiesofusage.\nLeast peculiarisanoun ofmaterial aspriormember(hardlytobe\nreckoned asregular possessive dependents, because therelation ofmaterial\nisnotexpressed byacase): thus, hfranyahasta, 'gold-handed', Mranyasraj,\n'with golden garlands', dyahsthuna, 'having brazensupports', rajatdnabhi,\n'ofsilver navel'.\n1302.Especially common istheuse ofanoun asprior\nmember toqualifytheotherappositionally,orbywayofequiv-\nalence(theoccasional occurrence ofdeterminatives ofthischar-\nacter hasbeen noticed above, 1280b).These may conveniently\nbecalled appositional possessives.Their accent isthat\noftheprior member, liketheordinary possessive descriptives.\na.Examples are:d$vaparna, 'horse-winged', or'having horses aswings'\n"
+    },
+    {
+        "page": 464,
+        "content": "448 XVIII. COMPOSITION[1302\nsaid ofachariot), bhumigrha, 'having theearth ashouse', indrasakhi, 'having\nIndra forfriend', agnthotr, 'having Agniaspriest', gandharvapatnl, 'having\naGandharva forspouse', furdputra, 'having hero-sons', jardmrtyu, 'having\noldageasmode ofdeath, livingtilloldage', agnfvasas, 'fire-clad', tadania,\n'ending withthat', caracaksus, 'using spies foreyes', visnufarmanaman, 'named\nVishnuQarman'; and, withpronoun instead ofnoun, tvdduta, 'having thee as\nmessenger', tddapas, 'having this forwork'.\nExceptionsinregard toaccent occurhere, asinthemore regular de-\nscriptive formation :thus, agnijihvd, wsanapud, dhumafikhd, pavinasd, etc.\nb.Notinfrequently,asubstantively used adjectiveisthe finalmember\ninsuch acompound: thus, fndrajyestha, 'having Indra aschief,mdnahsastha,\n'havingthemind assixth', somacrestha, 'ofwhich soma isbest', ekapard,\n'ofwhich theace ishighest' (?),dsthibhuyas, 'having bone asthelarger part,\nchiefly ofbone', abhirupabhuyistha, 'chiefly composed ofworthy persons',\ndacavara, 'having ten asthelowest number', cintapara, 'having meditation\nashighest object oroccupation, devoted tomeditation', nihfvasaparama, 'much\naddicted tosighing'.\nc.Certain words areofespecial frequencyinthecompounds here de-\nscribed, andhave inpartwon apeculiar application.\n1.Thus, with adioradika oradya, 'first', aremade compounds sig-\nnifying thepersonorthing designated along with others, such apersonor\nthingetcetera. Forexample, devd indradayah,'the gods having Indra as\nfirst', that is,'thegodsIndraetc.', maricyadin mwmn, 'Marici andtheother\nsages', svdyambhuvddydh saptdiJ\ntemanavah, 'those seven Manus, Svayambhuva\netc.', agnistomadhikdn makhan, 'the sacrifices Agnishtoma and soon'. Or\nthequalified noun isomitted, asinannapanendhanadini, 'food, drink, fuel,\netc.', danadharmadikarh caratu bhavan,'letyour honorpractise liberality,\nreligious rites, andthe like'. Theparticles evam and itiarealsosometimes\nused bysubstitution asprior members :thus, evamddi vacanam, 'words to\nthisandthelikeeffect'; ato'ftam bravimi kartavyah sarhcayo nityam ityadi,\n'hence Isay\"accumulation isever tobemade\" etc.'\nUsed inmuch thesame way, but lessoften,isprabhrti, 'beginning':\nthus, vifvavasuprabhrtibhir gandharvaih ,'with theGandharvas Vi^vavasu etc.';\nespecially adverbially, inmeasurements ofspace and time, astatprabhrti,\n'thenceforward'.\n2.Words likepurva, purvaka, purahsara, purogama, meaning 'foregoer,\npredecessor', areemployed inasimilar manner, andespecially adverbially,\ntodenote accompaniment.\n3.Thenoun mafra, 'measure', stands asfinal ofacompound which is\nused adjectively orinthesubstantive neuter tosignify alimit that isnot\nexceeded, and obtains thus thevirtual value of'mere, only': thus, jala-\nmatrena vartayan, 'living bywater only' (lit'ly, 'bythatwhich haswater\nforitsmeasure orlimit'), garbhacyutimatrena, 'bymerely issuing from the\nwomb', pranaydtrikamdtrah sydt,'lethim beonepossessing what does not\nexceed thepreservationoflife'; uktamdtre tuvacane,'but thewords being\nmerely uttered'.\n"
+    },
+    {
+        "page": 465,
+        "content": "1304]\n4.Thenoun artha, 'object, purpose',isused attheendofacompound,\noftenest inthesubstantive neuter,tosignify'forthesakeoforthelike :\nthus, yajnasiddhyartham,'inorder totheaccomplishment ofthesacrifice',\ndamayantyartham,'forDamayanti's sake'.\n5.Other examplesareabha, kalpa, inthesense of'like, approaching':\nthus, hemabha, 'gold-like', mrtakalpa, 'nearly dead', pratipannakalpa, 'almost\naccomplished'; vidha,inthesense of'kind, sort': thus, tvadvidha,'ofthy\nsort', purusavidha,'ofhuman kind'; prat/a, inthesense of'mostly, often',\nand the like: thus, duhkhapraya,'full ofpain', trnapraya, 'aboundingin\ngrass', nirgamanapraya,'often going out'; antara(insubstantiveneuter),\ninthesense of'other': thus, devantara, 'another region' (lit'ly,'that which\nhasadifference ofregion'), janmantarani,'other existences', fdkhdntare,'in\nanother text'.\n1303. Inappositional possessive compounds, thesecond member,ifit\ndesignates apartofthebody, sometimeslogically signifies thatpart towhich\nwhat isdesignated bythepriormemberbelongs, thatonorinwhich itis.\nThus, ghrtdprstha, 'butter-backed', mddhujihva, 'honey- tongued', niskd-\ngrlva andmanigriva, 'necklace-necked', pdtrahasta, 'vessel-handed', vdjra-\nbahu, 'lightning-armed', dsrnmukha, 'blood-faced', kildlodhan, 'mead-uddered',\nvdjajathara, 'sacrifice-bellied'; with irregular accent, dhumaksif.,'smoke-\neyed', afrumukhi f.,'tear-faced'; andkhadihasta, 'ring-handed' (khadf). In\nthelater language, suchcompounds arenotinfrequent withwords meaning\n'hand':thus, fcutrapani, 'having asword inthehand', lagudahasta, 'carry-\ningastaff'.\n1304. Ofpossessive compounds having anadverbial ele-\nment asprior member, themost numerous byfararethose\nmade with theinseparable prefixes. Their accent isvarious.\nThus:\na.Incompounds with thenegative prefix aoran(inwhich thelatter\nlogically negatives theimported idea ofpossession), theaccent isprevailingly\nonthe final syllable, without regardtotheoriginal accent ofthefinalmember.\nForexample:anantd, 'having noend', abald, 'notpossessing strength', arathd,\n'withoutchariot', afraddJid, 'faithless', amant, 'without ornament', a?atru,\n'without afoe',avarmdn,'notcuirassed', addnt, 'toothless', apad, 'footless',\natejds,'withoutbrightness', anarambhand,'not tobegotten holdof,apra-\ntimand, 'incomparable', aduchund, 'bringing noharm'.\nBut anumber ofexamples (few inproportiontothose already instan-\nced)have theprefix accented(likethesimple descriptives:1288a):thus,\ndksiti, 'indestructible', dgu, 'kineless', dgopd, 'withoutshepherd', djlvana,\n'lifeless', dnapi,'withoutfriends', dfi?vt f.,'without young', dmrtyu, 'death-\nless', dbrahman, 'withoutpriest', dvyacas,'withoutextension', dhavis,\n'without oblation', and afew others :AV. hasdprajas, butQB. aprajds.\nAveryfewhave theaccent onthepenult:namely, apesas, ajdni, andavira\n(with retraction,from mra); andAV. hasabhrdtr, butRV. abhratf.\nb.Incompounds with theprefixesofpraise anddispraise, suanddtis,\nWhitney, Grammar. 29\n"
+    },
+    {
+        "page": 466,
+        "content": "450 XVIII. COMPOSITION.[1304\ntheaccent isinthegreat majority ofcases that ofthe finalmember:thus,\nsukdlpa,'ofeasymake1\n,subhdga,'wellportioned', sundksatra, 'ofpropitious\nstar', suputrd, 'having excellent sons', sugopd, 'well-shepherded', suklrtf,'of\ngood fame', sugdndhi, 'fragrant', subahu, 'well-armed', suydrhtu,'ofeasy\ncontrol', sukrdtu,'ofgood capacity', suhdrd, 'good- hearted', susrdj,'well-\ngarlanded', suvdrman, 'well-cuirassed', suvdsas, 'well-clad', suprdnlti,'well\nguiding': durbhdga, 'ill-portioned', durdrpika,'ofevilaspect', durdhdra, 'hard\ntorestrain', durgdndhi, 'ill-savored', duradhi,'ofevildesigns', durdhdrtu,\n'hard torestrain', dustdritu, 'hard toexcel', duratyetu, 'hard tocross', dur-\ndhur, 'ill-yoked', durndman, 'ill-named', durvdsas,'ill-clad'.\nThere are,however, anotinconsiderable number ofinstances inwhich\ntheaccent ofthese compoundsisuponthefinal syllable:thus, su$iprd,'well-\nlipped', svapatyd,'ofgood progeny', susarhkdfd,'ofgood aspect', svangurf,\n'well-fingered', svisu, 'having good arrows', supivds, 'well fatted'; andcom-\npoundswith derivatives inana,assuvijndnd,'ofeasy discernment', supa-\nsarpand,'ofeasy approach', duQcyavand, 'hard toshake'; andAV. hassu-\nphald andsubandhu against RV.suphdla andsubdndhu. Like avira, suvira\nshows retraction ofaccent. Only tMrapir hasthetone ontheprefix.\nOnthewhole,thedistinction byaccent ofpossessive from determina-\ntive isless clearly shown inthewords made with suanddusthan inany\nother body ofcompounds.\nc.The associative prefixsaor(less often) sahd istreated likeanad-\njective element, and itself takes theaccent inapossessive compound:thus;\nsdkratu, 'ofjoint will', sandman,'oflikename', sdrupa, 'ofsimilarform',\nsdyoni, 'having acommonorigin', sdvacas,'ofassenting words', sdtoka, 'ha-\nving progeny along,with one'sprogeny', sdbrdhmana, 'along with theBrah-\nmans', sdmula,'with theroot', sdntardefa,'with theintermediatedirections';\nsahdgopa,'with theshepherd', sahdvatsa, 'accompanied byone's young', sahd-\npatnl, 'having herhusband withher', sahdpurusa, 'along with ourmen'.\nInRV. (saveinadoubtful case ortwo), only saha insuchcompounds\ngives themeaningof'having along with one, accompanied by'; and. since\nsaha governs theinstrumental, thewords beginning with itmight beofthe\nprepositionalclass (below, 1310). But inAV.both saandsahahave this\nvalue (asillustrated byexamples given above); and inthelater language,\nthecombinations with saaremuch themore numerous.\nThere areafewexceptions,inwhich theaccent isthat ofthe final\nmember: thus, sajosa, sajosas, sadffa, saprdthas, sabddhas; andAV.shows\ntheaccent onthe final syllable insangd (QB.sdnga] andthesubstantivized\n(1312) savidyutd.\nd.Possessive compounds with theexclamatory prefixes kaetc. aretoo\nfew intheolder language tofurnish ground foranyrule astoaccent: kd-\nbandha isperhaps anexampleofsuch.\n1305. Possessive compoundsinwhich averbalprefixis\nused asprior member withadjective value, qualifyinganoun\nasfinalmember, arefound even intheoldest language, and\n"
+    },
+    {
+        "page": 467,
+        "content": "13081 POSSESSIVE COMPOUNDS. 451\narerather morecommon later (comparethedescriptive com-\npounds, above, 1289: andtheprepositional,below. 1310. They\nusuallyhave theaccent oftheprefix.\nMostcommon arethosemade withpra, vi,andsam: thus, forexample,\nprdmahas, 'having exceeding might', prd?ravas, 'widely famed'; vfyriva,'of\nwry neck', vyhnga, 'having limbs away orgone, limbless', vfjani, 'wifeless',\nviparva andvfparus, 'jointless', vyadhvan.'ofwideways', vfinanas, both 'of\nwide mind' and 'mindless', vivacas, 'ofdiscordant speech'; sampatrii, 'having\none's husband along', sdmmanas,'ofaccordant mind', sdrhsahasra, 'accompa-\nnied byathousand', sdmokas,'ofjoint abode'. Examplesofothers are :\ndtyurmi, 'surging over', ddhivastra, 'having agarment on',ddhyardha,'with\nahalfover', ddhyaksa, 'overseer', dpodaka, 'without water', abhirupa,'of\nadapted character', dvatoka,'that hasaborted', dmanas,'offavorable mind',\nudojas,'ofexalted power, nimanyu,'ofassuaged fury', nfrmaya,'freefrom\nguile', nirhasta, 'handless'.\nInacomparativelysmall number ofcases, theaccent isotherwise; and\ngenerally onthe final :thus, avakefd, upamanyu, vi$aphd, vifikhd (AV.m-\ncikha], vikarnd, sammatr, etc,;inaninstance ortwo, that ofthefinalmem-\nber; thus, samfffvari, 'having acommon young'.\n1306. Possessive compounds with anordinaryadverb as\npriormember arealsofound inevery periodofthelanguage.\nThey usually have theaccent which belongstotheadverb asin-\ndependent word.\nExamplesare :dntyuti, 'bringing near,help', avodeva, 'calling down the\ngods', itduti, 'helping onthisside', ihdcitta,'with mind directed hither',\ndaksinatdskaparda, 'wearing thebraid ontheright side', ndnadharman,'of\nvariouscharacter', purudhdpratlka,'ofmanifoldaspect', vi?vdtomukha, 'with\nfaces onallsides', sadydiiti,'ofimmediateaid', vfsurupa,'ofvarious form',\nsmddudhan, 'with udder'.\nAninstance ortwo ofirregular accent aremetwith :thus, purorathd,\n'whose chariot isforemost'.\n1307. Itwaspointed outinthepreceding chapter (1222\nc.2)that the indifferent suffix kaisoften added toapure\npossessive compound,tohelptheconversion ofthecompounded\nstem intoanadjective;especially, where the final ofthestem\nislessusual ormanageableinadjectiveinflection.\nAlso, thecompound possessive stemoccasionallytakes further\napossessive-making suffix, in(1230).\nThe frequent changes which areundergone bythe final ofastem\noccurring attheend ofacompoundwillbenoticed further on(1315).\n1308. Thepossessive compoundsarenotalways used in\nthe laterlanguage with thesimplevalue ofqualifying adject-\nive;oftentheyhave apregnantsense, andbecome theequiva-\nlents ofdependent clauses;orthe'having' which isimpliedin\n29*\n"
+    },
+    {
+        "page": 468,
+        "content": "452 XVIII. COMPOSITION.[1308\nthem obtains virtuallythevalue ofour'\nhaving'assignofpast\ntime .\nThus, forexample, praptayauvan'i ,'possessing attainedadolescence',\ni.e.'havingarrived atadolescence'; anadhiyutacastra, 'with unstudiedbooks',\ni.e.'who hasneglected study1\n;krtaprayatna, 'possessing performed effort',\ni.e.'onwhom effort isexpended'; anguliyakadar?anavasana, 'having the\nsight ofthering astermination',i.e.'destined toendonsight ofthering';\nuddhrtavisada$aly(ih, 'having anextracted despair-arrow',i.e.'when Ishall\nhave extracted thebarb ofdespair'; frutavistarah kriyatam,'lethimbemade\nwith heard details',i.e.'lethimbeinformed ofthe details'.\nB.Compounds withGoverned Final Member.\n1309. Participial Compounds. Thisgroupofcom-\npounds,inwhich thepriormember isapresent participle and\nthefinalmember itsobject,isasmall one(toward thirty exam-\nples),and exclusively Vedic indeed, almost limited tothe\noldest Vedic(oftheRig-Veda).\nThe accent isonthe finalsyllableoftheparticiple, what-\nevermayhave been the latter' saccent asanindependent word.\nExamplesare :vidddvasu, 'winning good things', ksayddvira, 'governing\n(ksdyant) heroes', tardddvesas, 'overcoming (tdrant) foes', abharddvasu, 'bring-\ninggood things', codaydnmati, 'inciting (coddyant) devotion', mandaydtsakha ,\n'rejoicing friends', dharaydtkavi, 'sustaining sages', mahhayddrayi, 'bestow-\ningwealth'.\nInsadddyoni, 'sitting inthelap' (sadat quite anomalously forsldat or\nsadat), andsprhayddvarna, 'emulous ofcolor', the case-relation ofthe final\nmember isother than accusative. Inpataydn mandaydtsakham (RV.i.4.7),\npataydt,with accent changed accordingly, represents pataydtsakham, the final\nmember being understood from thefollowing word. Viddda?va istobein-\nferred from itsderivative vdfdadafvi. Ofthisformation appear tobejam-\nddagni, pratddvasu (prathdd?), andtrasddasyu (fortrasdddasyu?}.Itwas\nnoticed above (1299c)thatyuyujandsaptiiscapable ofbeing understood as\naunique compound oflike character, with aperfect instead ofpresent par-\nticiple ;sddhadisti, onaccount ofitsaccent,isprobably possessive.\n1310. Prepositional Compounds. Bythisnamemay\nbeconvenientlycalled those combinations inwhich theprior\nmember isaparticle havingtrueprepositional value, and the\nfinalmember isanoun governed byit.\nSuch combinations, though fewinnumber ascompared with\nother classes ofcompounds,arenot rare, either intheearlier\nlanguageorthe later.\nTheir accent issovarious thatnorule canbesetup\nrespectingit.\n"
+    },
+    {
+        "page": 469,
+        "content": "1312]PREPOSITIONAL COMPOUNDS. 453\na.Examples are:dtyavi, 'passing through thewool1\n,atiratrd, 'over-\nnight', atimatrd, 'exceeding measure'; ddhiratha, 'lying onthechariot', adhi-\nyavd, 'belonging tothecow'; adhaspadd, 'under the feet', adhoaksd, 'below\ntheaxle'; dnuputha, 'following the road', dnuvrata, 'according todirection'\n(butperhaps possessive?}, anupurvd, 'followingtheonepreceding, one after\nanother', anusatyd,'inaccordance withtruth.', anukula, 'down stream1\n,etc.;\ndntaspatha (with anomalously changed accent ofantdr], 'within theway',\nantardavd, 'within the flame'(?),antarhastd,'inthehand'; dntigrha, 'near\nthehouse'; dpavrata,'out ofdue course' (possessive?); apiprana, 'accom-\npanying thebreath', dpivrata, 'concerned with theceremony', apifarvard,\n'bordering onthenight', apikarnd, 'next theear'; dbhijnti, 'reaching tothe\nknee', abhtvira and abhfsatvan, 'overcoming heroes'; apathi,'ontheroad',\nadevdj 'going tothe gods', ajarasd, 'reaching oldage', advadafd, 'upto\ntwelve'; upakaksd, 'reaching tothearm-pits', upottamd, next tolast, penult-\nimate'; updribudhna, 'above thebottom', updrimartya, 'rising abovemortals';\ntirojand, 'beyond people'; nihsald,'out ofthe house'; paripdd, ('about the\nfeet') 'snare', parihastd, 'about thehand, bracelet'; paroksa,'out ofsight',\npartimatra, beyond measure', parogavyutf, 'beyond thefields', parahsahasrd\n(pdrahsahasra, QB.),'above athousand'; pratidosd, 'toward evening', prati-\nlomd, 'against thegrain', pratikula, 'upstream', pratydksa, 'before theeyes';\nbahihparidhi, 'outside the enclosure'; vfpathi, 'outside theroad1\n;samaksd,\n'close totheeyes,insight'.\nb.Compoundsofthis character areinthe later language especially\ncommon withadhi :thus, adhyatma, 'relating tothesoul orself, adhiyajfia,\n'relatingtothesacrifice',etc.\nc.Asuffixal aissometimes added toafinal consonant, asinupanasd,\n'onthewagon', dvyusd,'until daybreak'. Inafewinstances, thesuffix ya\nistaken(seeabove, 1212 d.5);and inoneword thesuffix in:thus, pari-\npanthvn, 'besetting thepath'.\nd.The prepositional compoundsareespeciallyliable toadverbial use :\nseebelow, 1313a.\nAdjective CompoundsasNouns andasAdverbs.\n1311. Compound adjectives,likesimple ones, arefreely\nusedsubstantivelyasabstracts andcollectives, especiallyinthe\nneuter, less often inthefeminine;andtheyarealsomuch used\nadverbially, especiallyintheaccusative neuter.\nThematter isentitled tospecial notice only because certain forms of\ncombination havebecome ofspecial frequencyinthese uses, andbecause the\nHindu grammarians havemade outofthem distinct classes ofcompounds,\nwithseparate names. There isnothing intheVeda which byitself would\ncall forparticular remark under thishead.\n1312. The substantively usedcompounds havinganumera\nasprior member, along with, inpart,theadjective compounds\n"
+    },
+    {
+        "page": 470,
+        "content": "454 XVIII. COMPOSITION.[1312\nthemselves, aretreated bytheHindus asaseparate class, called\ndvigu.\nThename isasample oftheclass, andmeans 'oftwocows', said tobe\nused inthesense of'worth two cows'; asalsopancagu, 'bought forfive\ncows', dvinau, 'worth twoships', pdncakapala, 'made infivecups', and soon.\nVedic examples ofnumeral abstracts andcollectives are: dvirajd, '[combat]\noftwokings', triyugd,'threeages', triyojand, 'space ofthreeleagues', tridivd,\n'the triple heaven', pancayojand, 'space offiveleagues', sadahd,'sixdays'\ntime', dafdnguld,'tenfingers' breadth'; and, with suffixya,sahasrahnyd,\n'thousand days' journey'. Others, notnumeral, butessentially ofthesame\ncharacter, are, forexample: anamitrd, 'freedom fromenemies', nikilbisa,\n'freedom from guilt', savidyutd, 'thunderstorm', vfhrdaya, 'heartlessness', and\nsdhrdaya, 'heartiness', mtdivd, 'prosperity byday', sumrgd andsufakund,\n'prosperitywith beasts and birds'. Feminines oflikeusearenotquotable\nfromRV. orAV.;later occur such astri$ati,'three hundred'(481), trilokt,\n'thethree worlds', pancamuli, 'aggregate offive roots1\n.\nAstheexamples show, theaccent ofwords thus used isvarious;but\nitismore prevailinglyonthe final syllable than intheadjective compounds\nintheir ordinary use.\n1313. Thoseadverbially used accusatives ofsecondary ad-\njective compoundswhich have anindeclinable orparticle as\n.priormember arereckoned bytheHindu grammarians asasepa-\nrate class ofcompounds, and called bythenameavyayibhava.\nThis term isaderivative from thecompound verb(1094) made upof\navyaya,'uninflected', andybhii,andmeans 'conversion toanindeclinable'.\na.\"The prepositional compounds (1310) areespecially frequent inthis\nuse: thus,forexample, anusvadhdm, 'byone'sownwill', abhipurvdm, 'in\nsuccession', advadacdm, 'uptotwelve', pratidosdm,'atevening', samaksdm,\n'insight'.Instances given bythegrammarians are :adhihari, 'upon Hari',\nuparajam,'under theking', upanadam orupanadi, 'near theriver', pratyagni,\n'toward thefire', pratinifam, 'every night', nirmaksikam,'free from flies'.\nb.Alarge andimportant class ismade upofwords having arelative\nadverb, especially yatha,aspriormember. Thus, forexample, yathava?dm,\n'asone chooses'(vdfa, 'will'), yathakrtdm,'asdone[before], according to\ncustom', yathanamd, 'byname', yathabhagd, 'according toseveralportion',\nyathangdandyathaparu,'limb bylimb', yatrakdmam,'whither onewill',\nyavanmatrdm,'insome measure', yavajjivd,'aslongasonelives', yavat-\nsdbandhu, 'according tothenumber ofrelations'.\nThese compoundsarenotcommon intheoldlanguage; RV.haswith\nyatha only four ofthem, AV. only ten;andnosuchcompoundisused ad-\njectively except yachresthd RV., yavachresthd AV.,'asgood aspossible'.\nB.hasyathakarfn, yathacarfn, ydthdkama, ydthdkratuasadjectives (followed\nineach casebyacorrelative tdtha). Theadjective useinthelater language\nalso isquite rare ascomparedwith theadverbial.\nOther cases than theaccusative occasionally occur:thus, instrumental,\n"
+    },
+    {
+        "page": 471,
+        "content": "1315]ANOMALOUS COMPOUNDS. 455\nasyathdsarhkhyena ,yathdfaktyd, yathepsaydj yathdpratigundis ,andablative,\nasyathducitydt.\nc.Other adverbial compoundsofequivalent character occur earlier, and\narecommon later: forexample, rtekarmdm, 'according tosacrificial use',\nndndrathdm,'ondifferentchariots', 'ubhayadytis,'two days insuccession';\neitrapadakramam,'with wonderful progress', pradanapurvam^'with accom-\npanimentofagift';etc.\nAnomalous Compounds.\n1314. Asinevery language, compoundsarenowandthen\nmetwithwhich areofanomalous character, asexhibiting com-\nbinations ofelements notusually puttogether,ornotaftersuch\namethod, orforsuch apurpose. Some ofthese, especiallyof\nthose occurringintheoldlanguage, maywellbenoticed here.\na.Compounds having aparticleasfinalmember :as,aprati, 'having no\nequal', tuvipratf, 'mightily opposing', dtathd, 'refusing', vitatha, 'false', ya-\nthdtathd,'as itreally is', stisaha, 'prosperityincompanionship',etc.\nb.Agglomerations oftwoormore elements outofphrases:thus, aham-\npurvd, 'eager tobefirst', ahamuttard, 'contest forpreeminence', mamasatyd,\n'contest forpossession', itihdsd, 'legend' (itiha.\"so, 'thus, indeed,itwas'),\nnaghamdrd andnaghdrisd, 'not, surely, dying orcomingtoharm', kuvttsa,\n'some unknown person', tadfdartha, 'having justthat asaim', kucidarthfn,\n'having errands inevery direction', kdcitkard, 'doingallsorts ofthings', ku-\nhacidwCd, 'wherever found', yadbhavisya, 'What- will-be',etc.\nC.Agglomerationsinwhich thepriormember retains asyntactic form :\nas,anyonya andparaspara, 'one another'.\nd.Aggregations with thenatural order inverted :as,pitdmahd and\ntatarnahd, 'grandfather'.\ne.Aggregations ofparticles were pointed outabove (111 la);also(1122b)\ncases inwhich ndandmdareused incomposition.\nf.Inthelater language,aprepositionisoccasionally precededincom-\nposition byacasedependent onit:thus, dantantah, 'between the teeth',\ntadbahis, 'outside ofit',satyavind,'without truth'.\nStem-finals altered inComposition.\n1315. Inallclasses ofcompounds,certain changesoffinal\nareliable toappearintheconcluding member;generally, they\nhave the effect oftransferringthecompoundtothea-declension.\nThus:\na.Astem inanoften dropsitsfinal consonant: examplesareakaa,\nadhva, arva, astha, aha, taksa, brahma, murdha, raja, Zoma, Dfa, fva,\nsaktha, sama.\nb.An iorIischangedtoa:examplesareangula, anjala, a$ra,\nfeftara, nac/a, na&Aa, &ftuma, ratra, sakha, hala.\n"
+    },
+    {
+        "page": 472,
+        "content": "456 XVIII. COMPOSITION.[1315 6]\nC.Anaisadded after afinalconsonant, andsometimes after anw-vowel\noradiphthong:examplesarerca, tvaca;uda, pada, yarada ;apa ;dhura,\npura; ahna, apmana, udhna, rajfia; anasa, ayasa, ayusa, urasa, enasa,\ntamasa, manasa, yajusa, rajasa, rahasa, varcasa, vedasa, creyasa, sarasa;\nbhruva, diva, gava, gava. nava.\nIrregular Construction withCompounds.\n1316. Inthelooseness ofunlimited and fortuitous combi-\nnation, inthelaterlanguage,itisnotrare thataword incom-\npositionhasanindependent word inthesentence depending upon\nitalone, rather thanuponthecompoundofwhich itforms a\npart.\nAfewexamplesare: cittapramathirii baladevanam apt(MBh.),'agirl\ndisturbing theminds even ofthe gods'; jyotisam madhyacari (EL), 'moving\ninthemidst ofthestars'; bhagavata krtasamskare 'smin(.),'inhim, whose\nsacraments have been performed bytheblessed one'; syandanedattadrstih\n(9-),'with hiseyefixed onthe chariot'.\nSimilar evidences ofthelooseness ofcompositionare:nadrstapurva\n'thavafruta,'not seen norheard ofbefore'; darupatram camrnmayam,'a\nwooden andanearthern vessel'.\n"
+    },
+    {
+        "page": 473,
+        "content": "APPENDIX.\nA.Thefollowingtext isgiven (aspromised above,\n3)inorder toillustrate byanexamplethevarietyofSan-\nskrittypeinuse. Itisgiventwice over;and atrans-\nliteration intoEuropeanletters follows. The text isafable\nextracted from the firstbook oftheHitopadeca.\nTheHunter, Deer, Boar, andJackal.\n*\nI\nTFT: i\ntI Tra5TT\n'Irfrnifa fW\nMmrt I\n^Irf3\niHtI*f Iffff:\nv^prr\n"
+    },
+    {
+        "page": 474,
+        "content": "458 APPENDIX.\n:ia?^rpr^\noTT^rl\"\nItfcf:I\nffcpsf\nwidfa wimfa\nr\nf^Trdf\nasitkalyanakatakavastavyobhairavo namavyadhah.sacai'kada\nmansalubdhah sandhanuradaya vindhyatavimadhyam gatah.tatra\ntenamrgaekevyapaditah. mrgam adaya gachatatenaghorakrtih\nsukaro drstah. tatas tenamrgam bhumau nidhayasukarah carena\nhatah. sukarenaJ\npyagatya pralayaghanaghoragarjanamkrtva sa\nvyadhomuskadece hatac chinnadruma ivapapata. yatah:\njalam agnimvisam castramksudvyadhl patanam gireh,\nnimittam kimcidasadyadehlpranair mmucyate.\n"
+    },
+    {
+        "page": 475,
+        "content": "Ai'PENDIX. 459\natrantaredlrgharavo nama jambukah panbhramannahararthi\ntanmrtanmrgavyadhasukaran apacyat. alokya 'cintayadasau :a/to\nbhagyam. mahad bhojyam samupasthitam. athava :\nacintitani duhkhaniyathai \\a\"yantidehinam.\nsukhany apttathamanye daivam aira^tiricyate.\nbhavatu; esam mansairmasatrayamsamadhikam bhojanam me\nbhavisyati.tatahprathamabubhuksayamtavad imani svaduni mansani\nvihaya kodandatanilagnam snayubandhamkhadami'tyuktva tatha\n'karot. tatac chinnesnayubandhe drutamutpatitena, dhanusa hrdi\nbhinnah sadirgharavah pancatvam gatah.ato'ham bravlmi :\nkartavyah samcayo nityaiii kartavyona'tisamcayah;\natisathcayadosena dhanusa jambnkohatah.\nB.Thefollowingtext isgiveninorder toillustrate\nbyasufficientexampletheusualmethod ofmarking accent,\nasdescribed above(87, 90).IntheMSS., theaccent-signs\narealmostinvariably added inred ink. The text isa\nhymnextracted from thetenth orlastbook oftheRig-\nVeda;itisregarded bythe tradition asuttered byVac,\n'voice'(i.e.theWord orLogos).\nHymn X.125)from theKig-Veda.\nHIHHI^IH\nII II\nrT l T^EIJ:MlHf^VyNJ HifNUldrflH n$u\nHmm\nTtrT3qfeTOT f&H^f R1\nift II u\n"
+    },
+    {
+        "page": 476,
+        "content": "460 APPENDIX.\nZffFPJHHHM 4iuTlfH rt^<IIUI rPJKrf^FJT*^IIH\n3 I\ns5TT^^T II^II\n\"5Jtcj^HUUM fM^llfH IIbu\nrflHiHIH^T IITTn r\nahdm rudrebhir vdsubhic caramy ahdmadityairutavicv&devaih,\nahdm mitravaruno'Wiabibharmy ahamindragni ahdm acvino'bhh .1.\nahdmsomam ahandsambibharmy ahdm tvdstaram utdpusdnam bhdgam,\nahdm dadhami drdvinam havismatesupravye ydjamanayasunvate. 2.\nahdm rastrl samgdmam vdsunam cikitusi prathama yajniyanam,\nturnmadevavyddadhuhpurutra bhuristhatram bhury avecdyanflm.3.\nmayasodnnam attiyovipdcyati yahpraniti yd1mcrnoty uktdm,\namantdvo mU,m tdupaksiyanticrudhigruta qraddhivdmtevadami. 4.\nahdm evdsvaydmiddm vadami justam devebhir utdmanusebhih,\nyamkamdye tam-tam ugram krnomi tdmbrahmanam tarnfsim tdm\nsumedham. 5.\nahdmrudraya dhdnur atanomi brahmadvise cdrave hdntava u,\nahdmjdnaya samddam krnomy ahdmdyavaprthivlavivega.6.\nahdm suvepitdram asyamurdhdn mdmayonir apsvdntdh samudre,\ntdto vitisthe bhuvana ''nuvlcvo 'ta'mumdyamvarsmdno'pasprcami.1.\nahdm cvdvata ivaprdvamy ardbhamana bhuvanani vlcva,\nparodivapara enaprthivyai'tavati mahina sdm babhuva. 8.\n"
+    },
+    {
+        "page": 477,
+        "content": "SANSKRIT INDEX.\nThe references inboth Indexes are toparagraphs. Inthis one,many\nabbreviations areused;but itisbelieved thatthey willbefoundself-explain-\ning. Forexample, \"pron.\"ispronunciation; \"euph.\" points outanything\nrelating tophonetic form oreuphonic combination; \"pres.\",topresent-system ;\n\"int.\" isintensive\n;\"des.\" isdesiderative;and soon.Aprefixed hyphen\ndenotes asuffix; oneappended,aprefix.\na,pron. etc., 1922;combination\nwith following vowel, 126 7\n;lossof\ninitial after eando,135,175a;re-\nsulting accent, 135;notliable to\n(/una, 235;lightened toiorw,249;\nlost inweakenedsyllable, 253.\n-a,prmy, 1148; scdry, 1208 9; a-\nstems, dcln,32634; from rdcl\na-st., 333,354; incompsn, 1270,\n1283ff.,1287 a.\na-oran-, negative, 1121 a;incompsn,\n1288 a,1304 a.\n-oka, prmy, 1181; scdry, 1222e.\n-afcj,see1221 b.\nyaks, 108g;pres., 708; pf.,788.\naksara,8.\nak'sdn, dksi, 343f,431.\naghosa, 34.\nj/ac'or anc, pf.,788; pple, 957c;\nstems ending with, 407 10.\n-a;,219,383d.5, 1200 c.\n}/anV, see ac.\nyanjora?,euph., 219; pres., 694,\n687; pf.,788.\n-anda, 1201.\n-at,'383d. 3,1200 aandsee-ane.\n-ata.see1176e.\n-ati,see1157.3.\n-atw, see1161c.\n-atnu, see1196 c.\n-atra, see1185 d.\n-atha, see1163 b.\n-athu, see1164.i/ad, irnpf., 621.\n-ad,383d.4, 1200 b.\nadhi, loss ofinitial, 1087 a.\nadhika, inoddnumbers, 477 a,478b.\nj/an, euph., 192b; pres., 631.\n-an, 1160.\nan-,see a-.\n-ana, 1150; stems incompsn, 1271.\nanadvdh, dcln, 224b,404.\n-ana, 1150.\n-am', 1159.\n-am, 1150.\n-ami/a, 962, 965, 1215 b.\n-ana, see1162.\nanudatta, 81.\nanudattatara, 90c.\nanwnasz'fea, 36,73.\nanuvrata, withaccus., 272.\nanustubh, euph., 151 c.\nammjara, pron. etc.,703;trans-\nliteration, 73.\nanehas, dcln, 419.\n-an* or-a*,ofpples, 584,1172; their\ndcln, 443 ff.\n-anta, 1172.\nantahstha, 31,51.\nantara,incompsn, 1302 c.5.\n-anti, see1172.\naw/a, dcln, 523.\naporap,dcln, 151d,393.\napi,loss ofinitial, 1087 a.\n-abha, 1199.\na6ftim/'ta-circumflex, 84d.\n"
+    },
+    {
+        "page": 478,
+        "content": "462 SANSKRIT INDEX.\nI/am, pres., 634; aor., 862.\n-am,infln. in,970a,971\n;gerund, 995.\namnas, euph., 176 c.\n-aye,infln.in,970f,975.\n-am, see1188 d.\nart, dcln, 343d.\n-am, see1192.\n</arj, pf.,788.\nyarth, so-called, 108f,1067.\n(irtha, incompsn, 1302 c.4.\naryaman, dcln, 426 a.\ndrvan, arvant, 455.\n|/ar/i, pres., 613; pf.,788.\n-ala,see1189.\nnlpaprana, 37.\nyav, aor., 838, 908; pple, 954e.\n<tva,loss ofinitial, 1087 a.\nnvagraha, 16.\nj/auad/ur, so-called, 108 f.\navaydj, avayd,406.\navyaylbhava, 1313.\nj/a? 'attain', pres., 694,708;pf.,788;\naor., 834b,8379; fut., 936.\ni/af 'enjoy', pres., 727; des.,1029 b,\n1031.\nyas 'be', pres., 636,621;inperiphr.\nconjn, 10702, 1073d;inppial\nperiphr. phrases, 1075d\n;incmpd\nconjn, 1094.\nyas'throw1\n,pres., 761c;aor., 847.\n-as,1151;dcln ofstemsin,411 ff.\n-as,infln. in,970a,971.\nasdn. dsrj, 398, 432.\n-as?,' 1198.\ndsrj, euph., 219:andsee asdn.\n-ase,infln.in,970c,973.\nasthun, dsthi, 343 f.431.\n-asna, see1195.\nyah 'say', pf.,801 a.\nyah'connect'(?),788.\ndhan, dhar, dhas, 430 a.\na,pron. etc., 1920,22;combina-\ntion offinal,1267; vrddhi ofa,\n235 ff.\n;lightenedtoIori,250\n,-toa,\n250 c;inpres. ,6616, 761 d.2,3;\ninaor.,884;inpple,954 c;indes.,\n1028 d.\nff,withablative, 293 c.\n-a,1149.\na-stems, dcln, 347 ff.\n-afca, see1181 b.\n-dku, see1181b.\n-atu, see1161.\natman, used reflexively, 514.\natmane padam, 529.\nadi,adika, adya, incompsn,, 1302 c.1.-ana, 1223 a;inpples, 584, 1175.\n-am, see1223 b.\ndnundsikya, 36.\nyap,108g,1087f;pres., 708,727;\npf.,783c;aor.,847; des., 1030.\nabha, incompsn, 1302 c.5.\namredita. 1260.\n-ayana, 1219.\n-ayl, 1220.\n-ayya, 966 c,1218.\n-ara, see1226 b.\n-dru, see1192.\n-ala,see 1227.\n-alu, see1192,1227.\nyas, pres., 619.628;inppial peripli.\nphrases, 1075 c.\nff,asaw, asya, '398, 432.\ni,pron. etc.,1920, 22;iand y,\n55;combinations offinal, 126,129;\nfromya,252,784,769,922b, 954b.\ni,union-vowel, 254;inpres. v630,631\n;\ninpf.,7968, 803\n;inaor.,876 7\n;\ninfut.,9345, 943;inpple, 956;\nininfln., 968;indes., 1031.\ni-stems, dcln,335 46;from rdcl i-st.,\n354; incompsn, 1276, 1287 c.\nyi'go', pres., 612note; pf.,783b,\n801d; int.(?),1021; caus., 1042e;\ninppial periphr. phrases,1075 ;in\ncompd conjn, 1092 b.\nyi(in, inv) 'send', 708, 716.\n-i,prmy, 1155; scdry,1221.\n-ika, prmy, 1186; scdry, 1222e.\n-Oca, fern, to-afca, 1181 a,1222 d.\nyich, 608, 753.\n-i;,219,383d. 5,1200 c.\n-it,383d. 3,1200a; advbl, 1109.\n-ita,1176b,d.\nfti,uses of,1102 a.\n-itu,see1161 b.\n-itnu, see1196.\n-itra, see1185d.\nyidhorindh, pres., 694;aor.,8367.\nyin (orinv), 708, 713, 716,749b.\n-in,1183, 1230;m-stems^doln,438 ff.;\nincompsn, 1275, 1287 e.\n-ma, see1177b, 1209c,1223 f.\ninaksa. 1029 c.\n-ineya, see1216 b.\nyinv, see in.\n-ima, 1224 a.\n-iman, see1168. 2.\n-iya, 1214.\niyaksa, 1029 c.\nfyan't, dcln, 451.\nir-stems, dcln, 392.\n"
+    },
+    {
+        "page": 479,
+        "content": "SANSKRIT INDEX. 403\n-ira,see1188e,1226 b.\nirajya, iradha, 1021.\n-i7a,see1189, 1227.\niva,euph., 1102b.\nyis 'desire', pres., 608, 753.\nyis 'send', euph., 225; pres., 727,\n761 c.\n-isa,see1197b.\n-i'stka, 467 70.1184.\n-is'nu,1194.\n-is,1153: w-stems, dcln, 411 ff.\ni\",pron. etc.,1920,22;combinations\noffinal, 126, 129; circumflexed,\n128c;uncombinaole indual, 138a;\ninlocative sing., 138d;Iasfinal\nofstem inverbalcouipsn, 1093 4.\nz,union-vowel, 254\n;ininflection, 555b\n;\nofpres.,6324; ofimpf., 621,6314 :ofs-aor., 880,s888 90;of\nint.,1004 ff.;zfor?,900,935a,972,\n/-stems, dcln, 347 68.\n-z,1156; toibefore addedsfx,471,\n1239.\n-zfca,see1186.\nyid, pres., 628, 630.\n-It'i,see1157.3.\n-Itu,see1161 b.\n-ma, 1223d.\n-zman, see1168.2.\ni\"ya,conj.-stem, 1021.\n-iya, 1215.\n-iyas,46770, 1184; stemsin,dcln,\n463 ff.\nj/ir, pres., 628; aor., 847.\n-zra,see1188e.\n1/ip, pres., 628, 630.\nicvara, with gen. infln., 984.\n-isa,see1197.\nM,pron. etc., 1920,22;uandv,\n57;combinations offinal, 126,129;\nfrom va,252, 784, 769, 922b,\n954b,956.\nM-stems, dcln, 335 46;from rdcl\nu-st., 354.\n]/M, pres., 708.\n-u,1178.\n-uka, 1180.\nyuks 'sprinkle', pres., 753.\n)/MC, pres., 761 a.\nyuch, 608, 753.\nyujh, 42,108g.\nMnadi-sufflxes, 1138.\n-*,383d.3,1200 a.\n-utra, see1185d.\nyudorund, pres., 694.nrf,udaka, udan, 398, 432.\nudatta, 81.\n-wna, see1177 c.\nyubj, pres., 753.\nyubhoiumbh, pres., 694,727,753,758.\nttr-stems, dcln, 392.\n-wra, see1188f, 1226 b.\n-uri, 1191.\n-tila, see1189, 1227.\nwranas, wpana, dcln, 355a,416.\n|/w, pres., 608, 753, 727.\n-usa, see1197.\nusds, euph., 168; dcln, 415 b.\n-wsj, see1221 c.\nusnih, euph., 223.\n-MS,1154; Ms-stems, dcln, 411 ff.\nMS/,371.\nu,pron. etc.,1920,22;combniations\noffinal, 126, 129; oircumflexed,\n128c;uncombinable indual, 138a;\ninlocative, 138 c.\nM-stems, dcln, 347 68.\n-u,1179.\n-uka, see1180.\n-utha, see1163c.\nudhan, udhar, udhas, 430b.\nwna,inoddnumbers, 477 a,478 b.\n-ura,see1188f.\nurj, euph., 219.\nyurnu, so-called, 108f,712\n;pf. ,801g,\n1071 c.\n-usa, see1197.\nusman, 31,59.\nyuh 'notice', euph., 240, 745a.\nf,pron. etc.,236;transliteration,\n24;questionofrorarinroots\nandstems, 108d,237; retention\nafter afinal vowel, 127; combina-\ntions offinal, 126, 129\n;impedes\nchange ofprecedingsto*,181a\n;\nchanges succeeding nton,189 ff.;\ngunaandvrddhi incrementsof,235ff .\n;\nirregular changes, 241,243\n;variable\nfinal rofroots(so-called f),242.\nr-roots, root-nouns from, 383 a,b.\nr-stems, dcln,36976.\nr,variable(so-called f),rootsin,242,\n245b;their passive, 770c;aor.,885,\n900;prec., 922a;fut.,935a;pple,\n957b;root-infin., 971;gerund inya,\n992.\nyr,euph., 242;pres., 608,753,643c,\n708,716; passive, 770c;pf.,783a;\naor., 834 a,837, 847, 862; int.,\n1002 b.\n"
+    },
+    {
+        "page": 480,
+        "content": "464 SANSKRIT INDEX.\n-r,see1182 f.\nytcorarc, pf.,788; aor.894d,897.\ny'rch. 608,753; pf.,788.\n-r},383d. 5,1200 c.\nT/rnJ orrj'stretchout', pres., 753,\n'761c;'aor., 894d, 897.\nyrnv,716.\n-ri,'383d. 3,1200 a.\nrtvij, euph.,219.\ny-rd, pres., 753.\ny'rdh, pres. ,694,708,761b;pf.,788;\n'aor.,8378, 840,847; des., 1030.\nrbhuksdn, dcln, 434.\nVra 'push', pres., 753; pf.,788.\nrhdnt, dcln, 450 a.\nf,pron. and occurrence, 236;as\nallegedfinal ofroots, 108d,242\n(andseer,variable); changes suc-\nceeding nton,189 if.\nI,pron. andoccurrence, 236;its\nywna-increment, 236.\nI23end.\ne,pron. etc.,279;combinations of\nfinal, 1313,135; uncombinable\nindual, 138a;guna of tandI,\n235ff.;asallegedfinal ofroots,\n251, 761d.2.\ne,iufin.in,970 a,971.\neka, dcln, 482a;used asarticle,\n482a;inmaking 9's,477b.\nefcapmfi, 90c.\n-ena, 1223 e.\n-enya, 966b,1217.\n-eya, 1216.\n-eyt/a, 1216 c.\n-era, see1201.\n-cru, see1192.\n-elima, 966d,1201.\nesds, euph., 176 a.\nai,pron. etc.,279;combinations\noffinal,1313;vrddhi of iandi,\n235ff.;asallegedfinal ofroots,\n251, 761d.1;for Iininflection,\n555b;foreinsubj. endings, 561.\no,pron. etc., 279;combinations\noffinal,1312, 1345;before\nsuffixya,136b;uncombinable,\n138c;forfinalcw,175a;ar,179\nend;guna ofuandu,235ff.;as\nallegedfinal ofroots, 251,761d.3.\notuaridostha, euph., 137 c.-ore,see1201.\nosthya, 49.\nau,pron. etc.,279;combinations\noffinal,1312, 134; vrddhi of\nuand u,235 ff.\n/,pron. etc., 679;makes heavy\nsyllable, 79; occurrence asfinal,\n148,170a;forthelabial andgut-\nturalspirants, 170d; from finals,\n145, 170a\n?172; r,144, 178;\nallows change ofstos,183.\nnorm,pron. etc., 703;makes\nheavy syllable, 79; occurrence as\nfinal, 148; allows change of*tos,\n183; occurrence, 204, 212,213d.\nfe,pron. etc.,3940;relation to<,\n42;tof,64;stosafter, 180ff.;\nadded tofinal n,211; fromc,by\nreversion, 214ff.;asfinal, and in\ninternal combination. 142, 217;\nfromf,do.,145,218; froms,226 c.\n-ka,prmy, 1186; scdry, 1222.\n-fcatfl, see1245g.\nyk'an, pf.,786; aor., 899c;int.,\n1008ff.\n}/fcam, aor., 868; pple, 955 a.\nkampa, 90b.\nkdmvant, euph., 212.\n-feara, 1201.\nkarmadharaya, 1263.\nkalpa incompsn, 1302c. 5.\nykas, int.,1002 c.\nyka,int.(?),1013.\nfeama, with accus., 272.\nfeara,insound-names, 18.\nykay, int., 1007, 1017.\nkfyant, dcln, 451.\nykir,756.\nyku, pres., 633.\nykup, pres., 761 a.\nykumar, so-called, 108 f.\nykus, pres., 727.\nV/;r'<make', pres., 708,7135; pf.\n797b;aor.831,834a-40, 847,\n894d; int., 1002c;prefixes s,\n1087d;inperiphr. conjn, 1070\n3;incompd conjn, 1091 4;\nspecial constructions, 268.\nykr 'praise', int., 1002 b.\nykr 'scatter', 242; pres., 753, 756;\naor.,885; prefixes s,1087 d.\nykrt 'cut', pres., 753,758; aor.,847.\nykrt -spin', pres., 694.\n"
+    },
+    {
+        "page": 481,
+        "content": "SANSKRIT INDEX. 465\n-kvt,see1105.\nfcrj-suffixes, 1138.\n-icrtvas, see1105.\ny'krp, pies., 745b;aor., 834b.\n|/ferf ;pres., 761 a.\nfcf?&, 958.\nj/fcrs, eupli., 226c;pres., 753; aor.,\n916, 920; int., 1008.\nyklp, 26; pf.,786.\n-L-ni,_see1176d.\nyknuorknuy, cans., 1042e.\ni/fcrand orkrad, aor., 847,890; int.,\n1002c, 1017.\nykram, pres., 745d;aor., 833, 847,\n904a;pple, 955a;int., 1002b;\ndes., 1031.\nykri, pres., 727; cans., 1042 e.\nykrudh, pres., 761a;aor., 847.\nykruf, aor., 916, 920.\nkrostu, krostf, 343h,374.\nyklam, pres.', 745d,761a,763.\nykli<; pres., 727,761b; aor.,. 916.\nks,combinations of,146, 221.\nyksan, pres., 713; pple, 954d.\nyk'sam, pres., 76la,763.\nkstim, dcln, 388b.\nyksar, aor., 890.\nyksa, pres., 761d.l.\nksama, 958.\nyksi 'dwell', pres.. 753,755; caus.,\n1042 e.\nyksi 'destroy', pres., 708,727,761b;\npple, 957a;caus., 1042 e.\nyksip, pres., 753.\nyk'tudh, pres., 761a;aor., 847.\nyksubh, pres., 761 a.\nfesaipra-circumflex ,84a.\nyksnu, pres.,626.\nfcft,pron. etc.,39; relation to,61.\nykhan, pass., 772; pf.,794d;aor.,\n890; pple, 955b.\nykhid, pres., 753.\nykhud, pres., 753.\nykhya, aor.,847,894c;caus., 1042d.\ng,pron. etc., 39; relation toj,42;\nfrom.7byreversion, 214 ff.\nyata,incompsn, 1273 c.\nygam, pres., 608, 747; aor., 833,\n834b, 837-9, 847, 887c; pf.,\n794d,805; fut.,943; pple, 954d;\nint., 1002c; des., 1028e, 1031;\nroot-noun, 383b.\nyga 'go', pres., 660;aor.,830,836ff.,\n894 c.\nyga 'sing', 251; pres., 761d.1;\nWhitney, Grammar.aor., 884, 894d,912; pple,\n954 c.\nygahorgah, int.,1002 b.\n|/y/r,756.\nyga, int., 1007.\nguna, 27,235 ff.\nyy'up, des., 1040.\nygur, pres., 753,756; aor., 834a.\nyguh, euph., 155, 223,240; pres.,\n745c; pf.,793f.;aor., 852, 916,\n920; caus., 1042.\nygr 'praise', euph., 242; pres., 727;\naor.,894 d.\nygr'swallow', euph., 242; pres., 727,\n753, 756; aor., 836, 847; int.,\n1002b.\nygr (orjagr] -wake', 1020; pf.,786;\naor., 867, 871.\nygrdh, pres., 761a;pf. ,786;aor.,847.\ngfd,euph., 134; dcln, 361c,e.\nygrathorgranth, pres., 727, 730;\npf.,794 e.\nygrabhorgrah, euph., 155; pres.,\n727, 729, 732,1066b;pf.,794c,\n801i; aor.,834b,838,900,904a,b;\nfut.,935d;pple, 956; inttn., 972;\npass., 998d;des., 1031.\nygla, pres., 761d.l; caus., 1042d.\nglau, dcln, 361 a.\ngh,pron. etc., 39;hderived from,\n66;fromh,byreversion, 214 ff.\nyghas, euph., 167; jaks from, 640;\npf.,794d;aor., 833.'\nghosavant, 34.\nygkra, pres., 671,749a;caus., 1042d.\nn,pron. etc.,39;occurrence asfinal,\n143,386b,c,407; duplicationas\nfinal, 210; adds fcbeforesibilant,\n211.\nc,pron. etc.,424;asfinal, 142;\nfrom tbefore apalatal, 202 3;\nntonbeforeit,208end;internal\ncombinationsof,217; reversion to\nfc,216ff.;inpres., 681; pf.,787;\nint:,1002end; des.,10281\nycakdsorcakdf, so-called, 677.\nycaks,108g;pres., 444, 621, 628,\n675.\n|/cam, pres., 745d.\ni/caj/, pres.,761d.2.\nyear, aor.,899c;int.,1002b, 1017;\ndes., 1031; inppial periphr.\nphrases, 1075 b.\nycal, int.,1002 b,1003.\n30\n"
+    },
+    {
+        "page": 482,
+        "content": "466 SANSKRIT INDEX.\n,pres., 761 d.1.\n'gather', reversion ofctofc,\n216.9, 681, 787, 1028f; pres.,\n708; aor., 889; caus., 1042e.\nyd 'notice', pres., 645; aor.,834 a.\nVett,108g;reversion ofctofc,216. 9,\n681. 787, 1002c,1028f; int.,\n1002 a,1024; des., 1040.\nydnt,108g.\nycrt, pres., 753.\nycyu, pf.,785; aor.(?),868, 870.\neft,pron. etc.,424;asfinal, 142;\nfrom fafter torn,203;after\nother mutes, 203end\n;ininternal\ncombination, 220;duplication be-\ntweenvowels, 227.\nychand, aor., 883, 890.\nycha, pres., 761d.3;pf.,806; pple,\n954c;caus., 1042 d.\nVchid, pres., 694,761b;aor., 832,\n834d,847.\nychrd, pres., 694; pple, 957d.\nj,pron. etc.,424;asfinal, 142;\nininternal combination, 219;nto\nnbeforeit,202; from tbefore\nsonant palatal, 202\n;reversion tog,\n215ff.; inpf., 787;indes.,1028f.;\nbefore naofpple, 957 c.\nyjaks,socalled, 108g ;pres., 640;\npple, 955 c.\njagat, dcln, 450d.\nyjan, pres., 631, 645, 680, 761b,\n772; pf.,794d;aor.,834b;pple,\n955b; des., 1031.\njam, dcln, 343 c.\nyjap, int.,1002b,1017.\nyjabhorjambh, int., 1002b, 1017.\nyjas, pres., 761 a.\nyjagr,so-called, 108f,1020.\njafya-circumflex, 84b.\nyji'conquer', reversion ofjtog,216.\n9;inpf.,787; indes., 1028f;\naor.,839,889, 894b, 904e; caus.,\n1042e; caus.aor., 1047, 861.\nyiiorjl'injure' seejyd.\nyjinv,108g,716,749b.\njilivamuliya, 39,69.\nyjw, aor.,838; des., 1028h, 1031.\nyjur, pres., 753, 756, 766.\nyjus, pres., 753;aor.,834b;insajus,\n225.\nyju, pres., 727; pf.,786.\ni/jr 'decay', euph., 242; pres., 761b,\n756, 766; pf.,793e, 794e.\nw,pres.. 727,730; aor.,830,837,894 c\n;912; caus.. 1042 d;cau.\naor., 1047, 861; caus.des., 1030.\nyjya (oTji.ji), pres., 727,761b;pf.,\n785,794b;pple, 954c,957a.\nyjval, aor.,899 c.\njh,pron. andoccurrence, 42; asfinal,\n142;ininternalcombination, 220.\nn,pron. etc., 42;fromnafter apal-\natal, 201; beforej,202; ?,203;\nc,208end.\nt,pron. etc.,456;from afinal pal-\natal, 142; f,145,218; ,145;\n/t,147\n;adds tbefores,199c;added\ntofinalnbeforesibilant, 211;from\njininternal combination. 219;ch,\n220; ks,221; h,222; s,226 b.\nth,pron. etc.45 6.\nd,pron. etc.,456;Iusedfor,54;\nfromdwithpreceding sibilant, 198\nend, 199b.\ndfe,pron. etc., 456;lhused for,\n54;from dhwith preceding sibi-\nlant, 199a;from hwith following\ntorth,222.\ndhvam fordhvam, 881, 901, 924.\nn,pron,etc,456;asfinal, 143\n;\nchange ofnto,189 95\n;fromn\nwith preceding sibilant, 199b;\ndoubled asfinal,210\n;adds tbefore\nasibilant, 211.\ni,pron. etc.,478;from final radi-\ncals,145,639; do.ininternal combn.\n1678;with preceding sonant\naspirate, 160;assim. tofollowing I,\n162; added after fbefore's,199c;\nafternbefore sor*,207\n;topal-\natalbeforepalatal, 202;beforep,203.\n-t,added after short finalvowel ofroot,\n345,376b,383b, 1143d, 1147c,1196a.\n-to,ofpple,9526, 1176\n;ta-stems\nincompsn, 1273, 1284\n;scdry,1245c.\nytansortea, aor.,899c;int.,1002 b.\nytaks, pres., 628, 708.\nytac, pres., 694.\nytad, euph., 198b.\ntatpurusa, 1263.\ntaddhita-suffaes,1\\38.\nytan 'stretch', pres., 713; pass., 772;\npf.,786,794d;aor., 833,834b\n;\n"
+    },
+    {
+        "page": 483,
+        "content": "SANSKRIT INDEX. 467\n847, 890; pple, 9f>id; des.,\n1028 e.\nytan 'thunder, pres., 761 c.\n-tana, 1245 e.\nytap, pres., 761b;aor.,834d,883.\nytam, pres., 761a,763; aor., 847;\npple, 955 a.\n-tama, 4713, 487, 1242 a.\n-tamam and-tamam, llllc, 1119.\n-taya, 1245 a.\n-taye,infln.in,970 e,975.\n-far,see 1109.\n-tara,4713, 1242 a.\n-taramand taram,llllc, 1119. .\n-tart,infln. in,970i, 979.\n-tavant,pple_in,95960.\n-tave and-fauai, infin. in,970b,972.\n-tavya, 962, 964,1212 d.2.\nytas, see tans,\n-tas, 1152; advbl, 1098.\n-fa,1237.\n-tat, inapv. formsin,570 1.\n-tat, 1238; advbl, HOOb.\n-tati, 1238.\nytay, pres., 761d.1.\ntalavya, 44.\n-ti,1157; fi-stems incompsn, 1274,\n1287 d; scdry, 1157.4; advbl,\n1102 a.\nytij, euph., 219; int.,1002a; des.,\n1040.\n-titha, 1242d.\nytir, 756, 766.\nytu, pres. ,633;aor.,868;int.,1002c.\n-tu,1161, 970b,972.\nytuj, pres., 753.\nytud, pres., 753, 758.\n-turn,infln. in,968,970b,972,9878.\nytur, pres., 753, 756, 761c,766;\ndes., 1029 a.\n-tur, 1182e.\nytus, pres., 76la.\nj/fr,'euph., 242; pres., 715,753,756,\n761b,766; pf.,794e,804; aor.,\n904d;int., 1002 b,c,1017.\n-fr,943,1182; fr-stems, dcln, 369ff.\ntrca, euph., 243.\ntrta, trtiya, euph., 243.\nytrd, pres., 694; aor., 836; pple,\n198b,957d.\nytrp, pres., 708,709,753,758,761a;\npf.,786; aor., 838, 847.\n]/<r*,pres.,761a; pf.,786; aor.,847.\nytrh, euph., 223,224b;pres., 694\n5;aor., 847, 916.\n-tos,infln. in,970b,972.\nttafordata, 955c,1087 e.ttifordati, 1157. 1b.\n-tna, 1245 e.\n-tnu, 1196.\n-tya,for-ya,992; scdry, 1245 b.\nytyaj, euph., 219; pf.,785.\n-tyai, infln. in,970e,975.\n-tra, 1185;ortra,.advbl, 1099.\nytrap, pf.,794 e.\n-tram, advbl, 1099.\nytras, pres., 761a;pf.,794e.\nytrd, 108g;pres., 628,761 c\n;aor.,\n893, 895.\n-tra,see -tra.\n-tri,see1185 e.\ntristubh, euph., 151 c.\n-tri,376a,1182.\n-trutsee1185 e.\n-tva, gerundival, 966 a,1209 g;scdry,\n1239.\n-tvata, 1239.\n-tvana, 1240.\n~tva,9901, 993.\n-tvanam, 993 c.\n-tvaya, 993b.\nytvis, pres., 753; aor., 916.\n-tvl,'993b.\n-tvinam, 993 c.\nytsar, aor., 890,899 c.\nth,pron. etc.,478;withpreceding\nsonantaspirate.160.\n-tha, 1163; ordinal, 487;ortha,\nadvbl, 1101.\n-tham, advbl, 1101.\n-tha,see tha.\n-thu, 1164.\nd,pron. etc.,47 8.\nydagh, euph., 155, 160end; aor.,8338.\nydad, 672; pf.,794 e.\nydadh, 672; euph., 155,160end.\ndadlidn, dddhi, 343f.,431.\ndan, euph.. 389.\nddnt, dcln, 396.\ndantya,47.\nydabh, euph., 155; pres., 708; pf.,\n794e;aor., 833,847; des., 1030.\nydam, pres., 761a,763.\n-dam, advbl, 1103 a.\nyday, pres., 76Id.2.\n,so-called, 108f,1024.\nordafif, pres., 746;int.,1002b.\n,pres., 761a;aor., 847,899 c.\nydah, euph., 155, 223; aor., 890,\n897,444; int.,1002b; des., 1030.\nydd 'give', pres.,6679, 672,749a;\n30*\n"
+    },
+    {
+        "page": 484,
+        "content": "468 SANSKRIT INDEX.\npf.,803; aor., 830,834a,8369,\n847,884, 894c;pple,955c,1087e,\n1157b; des., 1030, 1034; caus.,\n1042 d.\nydd 'cut', 251; pres., 76id.3; aor.,\n834a,884;pple,954c,955c,957a,\n1087 e,1157 b.\nyda 'bind', 251\n;pres., 761 d.3\n;pple,\n954c,957 a.\nyda 'cleanse', pres., 761d.1.\nyda 'protect', pf.,787.\n-da, advbl,1103a.\n-danlm, advbl, 1103b.\nydac, pres., 444,639,708; pf.,790b.\nydas, pres., 444.\n-di,advbl, 1103d.\ndtv, dcln, 361 d.\ni/dip, euph., 218; pres., 753; aor.,\n916,920; int., 1007, 1017.\nydih, euph., 155, 223; aor.,916;\ndes., 1030.\nydi, pres., 761 c.\nydik+, des., 1031.\nydidi, so-called, 676; pf.,786; aor.,\n838.\nydldhl, so-called, 108f,676; pf.,786.\nydlp, pres., 761c;aor., 861.\nydiv 'play', euph., 240; pres., 761c,\n765; pple, 954 e.\nydlvordev'lament', pple, 957 a.\nydu, pres., 708; pple, 957 a.\nduchuna, euph., 168.\nydudh, 108g.\nydus, euph., 240, 1150. 1a,1155. 1;\npres.,761a;aor.,847; caus., 1042a.\ndus-, 225,1121c;incompsn, 1288 b,\n1304b.\nyduh, euph., 155,223; pres., 621,\n635; aor., 916, 920.\nydr 'burst', euph., 242; pres., 727,\n761b; pf.,793e; aor.,831;int.,\n1002a,b,1023.\nydr 'regard', pres., 757, 773; aor.,\n834 a,881.\nydrp, pres., 761a;aor., 847.\ndrt, euph., 218; pf.,806; aor.,\n832,834b,836, 847,890; pass.,\n998d;root-noun, dcln, 386 c.\ndrp, cZrfo, drksa, withpron. -steins, '\n518.'\nydrhordrhh, euph., 155,223; pres.,\n753, 758, 761 b,767.\ndevanagari, 1.\ndosdn, dts, 398, 432.\ndytianddyd, dcln, 361d,e.\nydyut, pf.,785; aor., 847, 863,890;\nin*., 1002 c.ydra, aor., 912; int., 1024; cans.,\n1042 d.\nydru, pf. ,797b;aor.,868\n;int.\nydruh, euph., 155,223; pres.,\naor., 847, 920.\nydru, pres., 727.\ndvandva, 1252.\ndvdr, dcln, 388 c.\ndvigu, 1312.\nydvis, euph., 226b, c;pres.\naor., 916, 920.,1018.\n761a;\n621;\ndh,pron. etc.,478;from torth\nafter sonantaspirate, 160.\n-dha, see -dha.\nydhan, pres., 645.\nydhamordhma, 750.\nydha 'put', pres.,6679, 672,749a;\naor., 830,8349, 847,884; pple,\n954c; des., 1028d, 1030; caus.,\n10_42d.\nydha 'suck', 251; pres., 761d.2;\naor., 868, 884, 912; pple, 954c;\ncaus., 1042 d.\n-dha_or-dha, advbl, 1104.\nydhav, pple, 954 e.\nydhi (ordhinv), 708, 716.\ndhi,final ofcompds, 1155.2e, 1276.\nydhu, pres., 708,711,753; aor.868,\n887a;int., 1002c,1003; caus.,\n1042 f.\nydhurv, aor.,887a;des., 1028h.\nydhr, pres., 757,773; pf.,786; aor.,\n834a, 867,871; int.,1002a, 1003.\nydhrs, pres., 708; pf.,786.\nydkmaordham, 108g,750.\nydhya, pres., 761d.l.\n-dhyai,infin. in,970g,976.\nydhvan, pple, 955 a.\nydhvas, euph., 168.\nn,pron. etc.,478;asfinal, 143;\nforfinal rdclm,143,212; change\nton,189 95;tonafterandbefore\npalatals, 201 3,208end; combin-\nations asroot-final, 204;loss as\n^tern-final, 204;assim. topalatals\nandlinguals, 205;toI,206;before\nsibilants, 207;treated asm,208\n9;duplicationoffinal, 210;in-\nstability asfinal, 256,1203b;used\nasunion-cons., 257, 313,482d;\nquestionoffinal ofpancan etc.,484.\nna,comparative, 1122 d.\n-na,ofpples, 952, 957,1177; euph.,\n161end; scdry, 1223g, 1245d.\nynaks, 108g.\n"
+    },
+    {
+        "page": 485,
+        "content": "SANSKRIT INDEX. 469\nynad, int.,1007.\nj/nand, euph., 192 a.\nynabh, euph., 192 a.\nynam, pf.,786; aor.,890, 911;pple,\n954d; int., 1002b, 1017.\n-nom, advbl, 1109.\ni/naf, 'lose', euph. 192a; pres.,\n761a;aor.,847,854; des.,1028k.\nynac 'attain', euph., 218; pf.,801h;\naor., 833,837; des., 1029 c.\nj/ncw, aor., 837.\nnds, dcln, 387e,397.\n-ncu, 1152.\nynah, euph., 223; pres., 761 c.\nnagarl,1.\nnasikya, 230b.\nnf,euph., 192f.\n-ni,1158.\nynins, euph., 183; pres., 628.\nVnij, euph., 219; aor., 847; int.,\n1007, 1024.\nmfa/a-circumflex, 84b.\nnfyand nipa, 397.\nnis, loss ofinitialof,1087 a.\nl/m, aor., 831,889,896; int.,1002a,\n1017; periphr. pf.,1071 d.\nnlddj euph., 198b.\nynu, pres., 626; aor., 868,887a;\nint.,1002c,1003.\n-nti, 1162.\nynud, pres., 753; aor.,834d,904c;\nint., 1017.\nnr,dcln, 371.\nynrt, euph., 192a; pres., 761c;\n_aor.,833.\nnau, dcln, 361 a.\np,pron. etc.,49 50.\n-p,caus.-sign, 1042d,e;aor.from\nsuch caus.,1047.\n-pa, 1201.\npdkvd, 958.\nypac, pres., 761 b.\nj/pat, pf.,794d; aor., 847;int..\n1002c;des., 1030, 1031.\npdti,dcln, 343b;indpndt compsn,\n1267 a.\npath, pathf, pdnthan, dcln,343g,433.\nypad, pres.,761c; aor., 834b,d,\n838;int., 1002c;des., 1030.\npad, dcln, 387d,389.\npada,111.\nypan, int., 1002 c.\npdnthan, see path,\nparasmai padam, 529.\np'iruchepa, euph.,168.\nflay, quasi-root, 1087 c.palyay, quasi-root, 1087 c.\nypac, pres., 761 c.\nypa 'drink', pres., 671,749a;aor.,\n830; pple, 954c;des., 1028d;\ncaus., 1042 d.\nypa 'protect', cans., 1042f.\nypa 'rise', pres., 660.\npada, 79,93.\npadapurana, 1122a.\nypinsoxpis,euph., 226b,c;pres.,\n694; aor:, 847, 920.\nypinv,108g,716,749b.\nY&, pres., 753, 758.\nypis, euph., 181 c.\ny_pld,euph., 198b.\nptpi, conj.-stem, 676, 786.\npurhs, pitmans, euph., 183; dcln, 394.\npura, pres.inpastsensewith, 778 a.\nypus, pres., 727, 761a;aor., 838,\n847.\nypu, pres., 727, 728;aor.(?),868,\npurva,incompsn, 1251e,1291\n;\n_1302c.2.\npusdn,dcln. 426 a.\nypr 'fill', euph., 242; pres., 727,731,\n761b,766; pf.,793e;aor., 838.\nypr 'cross', pf.,793e;aor., 896.\nypr'bebusy', pres., 757, 773.\nyp'rc, pres., 694; aor.,834c,8367,\n890.\nyprn, 731, 753.\npft,pftana, 397.\nprsant, dcln, 450 c.\nypyaorpi,108g;pres., 761d.1;\npf., 785,794b; aor., 912, 914;\npple, 957a;caus., 1042 d.\npracayaorpracita accent, 90a.\nyprachorprch, euph., 220; pres.,\n753; pf.,794c;aor.890.\nprabhrti incompsn, 1302 c.1.\npraptota-circumflex, 84c.\n>/p_ra,\"l08g;aor., 830, 889.\npraya,incompsn, 1302 c.5.\n|/pn, pres., 727;aor.(?),868; cans.,\n1042 f.\nypruth, int.,1002 a.\nypr*t, euph., 226b;pres., 708, 727,\n732, 1066b,753, 761b.\nplay, quasi-root, 1087 c.\nyplu, aor., 863.\npluta, 78.\nypsa,108 g.\nph,pron. etc.,49 50.\nyphan, pf.,794e; int., 1002c.\nyphar, int., 1008.\n"
+    },
+    {
+        "page": 486,
+        "content": "470 SANSKRIT INDEX.\nyphal, pf.,794 e.\npJiuM, 958.\n6,pron. etc.,-49 50.\nybahhorbah, euph., 223.\nybadhorvadh, aor., 838, 904a;\nint.,1002 a;des., 1029a,1040.\nybandh, euph., 155; pres., 727, 730.\nbahuvrihi, 1293 b.\nybadh, euph., 155; aor., 833; int.,\n1002a,b; des., 1031.\nybudh, euph., 155; pres.,761a;\naor.,834b,839, 847.\nybrh, euph., 223; pres., 753,758;\nint., 1008, 1011.\nbrhdnt, dcln,450 a.\n66ft, occurrence, 151d.\nybru, pres., 632; peculiarconstruc-\ntion, 268.\n6ft,pron. etc.,4950.\nybhaks, 108g.\nVbhaj', euph., 219; pf.,794e; aor.,\n834c,883, 890.\nybhanj, euph., 219; pres., 694, 692.\nbhdvant, 456, 514.\nybhas, pres., 678.\nybhas, euph., 182 a.\nybhiks, I08g.\nybhid, pres., 694, 761b;aor., 832,\n840, 847.\nbhisdj, euph., 219.\nybhi, pres., 645, 679; aor., 831,\n890; cans., 1042e,f; periphr. pf.,\n1071d,1073 a.\nybhis,1042f;aor., 861, 1047.\nybhuj 'enjoy', euph., 219; aor., 847,\n895\n]/6ftw/'bend', euph,., 219; pres., 694.\nybhur, pres., 753, 756.\nbhtivas, euph., 176 c.\nybhu, pf.,789a,793a;aor.. 829,\n830,8369, 924;inperiphr. conjn,\n1070 72;inppial periphr. phrases,\n1075d;incompd conjn, 10914.\nbhuta incompsn, 1273 c.\nybhr, pres., 645; pf.,789b,797b;\naor., 838, 890; int., 1002b,c;\nperiphr. pf.,1071 d.\n6ftos, 176a,456.\nybhrajj, euph., 219; pres., 753.\nybhram, pres., 761a,763; pf.,794e.\nybhraf or6ftranf, pres., 761 c.767;\naor., 847.\nybhraj, euph., 219; pf.,794e;aor.,\n838.ybhn. pres., 727.\nm,pron. etc.,4950;asfinal, 143;\nasfinalradcl, 143, 212,_256;in\nextrnl combn, 213\n;before raj,213 a.\n-ma, prmy, 1166; scdry, 474, 487,\n1224 b.1242 b.\nymanh, 108g;pf.,786.\nmaghdvan, dcln, 428.\nymajj, euph., 219; pf.,801h; fat.,\n935b; des., 1028k.\nymathormanth, pres., 727, 730,\n732,1066b,746; aor., 847.\nI/mad, pres., 645, 761a,764; aor.,\n834d,839,887b,899 c.\nyman, pres., 713,761a;pf.,794d;\naor., 834b;pple, 954d; des.,\n1028e, 1029a,1040; special con-\nstruction, 268.\n-man, 1168; man-stems incompsn,\n1277.\n-mane, infln.in,970d,974.\n-mant, 1235.\nymantr, so-called, 108f,1067, 1073d.\nmdnthan, dcln, 434.\n-maya, 161, 1225.\n-mara, 1201.\nj/maft,108g.\nmaft, maftl, 400.\nmahdnt, dcln, 450b.\nmaft<f, 355 a.\nmahaprana, 37.\nj/ma 'measure', pres., 660,663; aor.,\n839; pple, 954c; des., 1030;\ncaus., 1042 d.\n|/maormi'bellow', pres., 660, 663,\n672,676end.\nymd 'exchange', pres., 761d.2.\nmdns, mansd(and mds], 397.\nmatra incompsn, 1302 c.3.\n-mana, 584b,1174.\nmda, euph., 168; dcln, 389, 397:\nandseemdns.\nymi 'establish', pres., 192c,708;\naor., 911; gerd, 992; des., 1030.\nymiorml'diminish', pres., 727,\n731, 761b;aor.,911; gerd, 992;\nint., 1012; des., 1030.\n-mi, 1167.\nymite, 1033.\nmitrd, 1185b.\n-min, 1231.\nymi, seemaandmi.\nymis, pres., 753.\nj/mz'ft, euph., 223; pf.,790b:aor.,\n916, 920.\n,pple, 954 e.\nc,pres., 753, 758, 761b;aor.,\n"
+    },
+    {
+        "page": 487,
+        "content": "SANSKRIT INDEX. 471\n832,834c,8389, 847,890 des,\n1030.\nymud, aor., 837.\nj/roua, pres., 727, 732,1066b; aor.,\n847.\nymuh, euph., 223; pres., 761a;\naor., 847.\nymurch,220\n;pres., 745f\n;pple, 954e.\nmurdhanya,45.\nymr 'destroy', pres., 727, 731.\nymr 'die', euph., 242; pres., 757,\n773; aor.,834a,837.\nymrks, pres., 753.\nymrjl euph., 219; pres., 621, 627,\n745e;pf.,786,793f;aor., 900,\n920; fut., 935b; int., 1002b,\n1017; des., 1028k; caus., 1042.\nymrd, euph., 198b; pres., 727, 753.\nymrn, 731, 753.\nymrd, int., 1002c, 1011.\nymrdh, aor., 837.\nl/mrp, pres., 753; pf.,786; aor.,\n916, 920; int., 1008, 1017.\nymrs, pres., 761a;aor.,834c,847.\nymed, pres., 761 a.\n-mna, 1224 c.\nymna,108g.\nymrit, pres.,761 a.\nymla, pres.,761d.1\n;caus., 1042d.\ny,pron. etc., 51,556;relation to\ni-vowels, 55;nasaly,71,213c;\nyasunion-consonant, 258,313,844,\n1112e,1151. Ic,1230d, 995;re-\nsolved toi,55,113; yofsfxtrea-\ntedas,1203.\n-ya (orya}ofgerund, 990,992 3;\nofgerundive, 9623, 1213.\n-ya,prmy, 1187, 1213; i/a-stemsin\ncompsn, 1272; scdry,121012.\nyakdn, ydkrt, 398, 432.\nyyaj, euph., 219, 784; pf.,784;\naor.,839,890,894d; des.(?),1029c.\nydtha, accent, 1101; incompsn,\n1313 b.\nyyam, pres., 608, 747; aor., 833,\n8369, 890, 911; pple, 954d.\nyama, 230 a.\nyyas, pres., 761 a.\n-yasfor-lyas, 470.\nyya, aor.,894 c,912.\n-ya,1213d.\n-yin,see1230 d.\nyyu 'join', pres., 626, 753.\nVyu:\nrepel', pres., 608, 645; aor.,\n838, 868, 889,894b.\n-yu,1165; see1178 g,h.,euph., 219; pres., 694; aor.,\n834b, 839,847; root-noun, '386b.\nyyudh, pres., 761a; aor., 836, 839.\nyyup, int., 1017.\nytivan, dcln, 427.\nyusa, yusdn, 432.\nr,pron. etc., 512;randI,53;\nrand sassonant and surd, 117a,\n164; final, 144, 169; combination\nastlnalrdcl, 165;asother, 173;\navoidance ofdouble, 179;sorr\nasfinal ofcertainforms, 169end;\nfrom safter a,176c;sto*after,\n1808; but notbefore, 'l81a;\nchanges succeeding nton, 189ff.,\nduplication ofconsonantafter, 228;\nsvarabhaktiafter, 230 c.\n-ra,prmy, 1188; scdry, 474, 1226,\n1242b.\nVraj, euph., 219; pres., 746, 761a,\n767.\nyradhorrandh, pres., 761a; pf.,\n786; aor.,847; int.,1002a; des.,\n1030.\nyran, pres., 761a;pf.,786; aor.,\n899c;int.,1008 ff.\nyrdbh, pf.,786; aor.,834d;des.,\n1030.\nyYam, pres., 727; aor., 911; pple,\n954d.\ni/m 'give', pres., 660, 666, 672;\naor., 839, 896.\nrd'bark', pres., 761 d.1.\n,euph., 213a, 219; pf.,794e.\n,pres., 761a;pf.,794e;aor.,\n838,847; des., 1031.\nyriorrl,pres., 727, 761c;caus.,\n1042 e.\n-ri,1191.\nyriteh, pres., 753.\nyric, pres., 694, 761b;aor., 834c,\n839, 890.\nyriph, pres.,753.\nyrif, pres., 753; aor., 916.\nyris, euph., 226c; pres., 761a;aor.,\n847.\n|/rift, euph., 223; int., 1017.\nyri,see ri.\nyru, pres., 626, 633, 753.\n-ru,1192.\n1/ruc, aor., 837.\nj/ruj, euph., 219; pres.,753.\nyrud, pres., 631, 753.\nyrudh, pres., 694; aor., 847, 890.\nyrup, pres., 761 a.\n,aor.,916.\n"
+    },
+    {
+        "page": 488,
+        "content": "472 SANSKRIT INDEX,\nrufant, dcln, 450 c.\nyrus, pres., 753.\nyruh, euph., 223; aor., 847, 916,\n920; caus., 1042 e.\nrepha,18.\nrat, dcln, 361b,e.\n-rfci, advbl, 1103 c.\nI,pron. etc., 51,53;Iandr,53;\nnasali,71;asfinal, 144\n;assim.\nto,117; of*,162; ofn,206; ofm,\n213c;asserted stoaafter, 180;\nduplicationofconsonantafter, 228\n;\nsvarabhakti after, 230 c.\n-la,pimy, 1189; scdry, 1227.\nyiag, pple, 957c.\nyiabh, des., 1030.\nyiikh, pres., 753.\nyiip, pres., 753, 758.\n>/%, aor., 916.\nyiih, euph., 223; aor., 916.\nyil 'cling', aor., 911; pple, 957a;\ngerund, 992; caus., 1042e, f;\nperiphr. pf. ,1073a.\nyil 'totter', int., 1022.\nyiup, pres., 753, 758, 761b.\nyiubh, pres., 761a.\nyiu, pres., 727; pple, 957a.\nZ,pron. etc., 54.\nv,pron. etc., 51,578;relation to\nw-vowels, 57;nasalv,71,213c;\nresolved tow, 58,113; duplication\nofconsonants after, 228.\n-va,prmy, 1190; scdry, 1228; advbl,\n1102b.\nyvak, pf.,786.\nyvac, euph., 216.9; pres., 660; pf.,\n784,789d;aor., 847, 854.\n-vat, advbl, 1107, 1233d; scdry.\n383d.1,1245 f.\nyvad,108g;pf.,784; aor., 838,\n899c,904c,d; pple,- 956; int.,\n1002a, 1017.\nyvadh, seebadh.\ny-oan, pres., 713; pf.,786,794d;\naor., 839, 847, 912, 914; pple,\n954d; int., 1011; des., 1028g.\n-van, prmy, 1169; scdry, 1234; van-\nstems incompsn, 1277, 1287 b.\n-vana, -vani, -vanu, 1170.\n-vane,infin.in,970d,974.\n-vant, 517, 959, 1233.\n.pf.,784.,pres., 631.\nvam (from w),543.\n-vam, advbl,' 1102b.\n-vaya, 1228 b.\n-vara, 1171.\n-van, fern, tovan, 1169, 1171, 1234.\nvarga, 32.\n-vala, 1228 b.\nyvaf, pres., 638,660; pf.,784, 786.\nyvas 'shine', pres., 608; pf.,784;\naor.,834b.\nyvas 'dwell', euph., 167; pf.,784.\nyvas 'clothe', pres., 628,638.\nyvas 'attack', pres., 631; pf.,786.\n-vas, 1173: andseevans,\nyvah, euph., 137d,223.224b;pf.,\n784; aor., 839,890; int.,1002c,\n1017;atendofcompds, 403.\nyva 'blow', caus., 1042 d.\nVva'droop', pres., 761d.l; caus.,\n1042 d.\nyvaorvi'weave', pres., 761d.2;\npf.,784,801b;fut.,935c;pple,\n954e;caus., 1042 d.\n-vans (or-vas),ofpples, 584, 802\n6,1173; vans-stems, dcln,458 ff.\nyva?, pres., 761c;aor., 861.\nvf,dcln, 343 e.\n-vi,1193.\nvikampana, 90b.\njAric, pres., 694; int., 1024.\nyvij, euph., 219; pres., 753; aor.,\n834c;int., 1017, 1024.\n-vitjsee1193.\nyvid 'know', pres.613,618,621; pf.,\n790a; int., 1006; des., 1031;\nperiphr. pf. etc., 1073.\nyvid 'find', pres., 753, 758; pf.,\n806; aor., 847.\nyvidh, pres., 753.\nvidha,incompsn, 1302c.5.\n-vin, 1232.\nyvip, aor., 852.\nvirama, 11.\nyvif, euph., 218; pres., 753; pf.,\n803, 806; aor.,834b, 916, 920.\nvifva, dcln,524;incompsn, 1251 e,\n1281 a,1298 c.\nyvis, euph., 225,226c;aor.,916;\nint., 1007, 1024.\nvisarga, visarjaniya,67.\nyvi 'impregnate', pres., 761b;int.,\n1007.\nyvi 'flutter', int., 1017, 1024.\nyvr 'cover', 108g; pres. (urnu), 708,\n712, 727; aor., 831,834a, 839,\n900; int., 1002 c.\n"
+    },
+    {
+        "page": 489,
+        "content": "SANSKRIT INDEX. 473\nyvr 'choose', euph., 108g,242; pres.,\n727; pf.,797b;aor.837.\nyvrj, euph., 219; pres., 694; pf.,\n786,803; aor.,832, 834c,8379,\n847, 920; int., 1002 c.\nyvrt, pf.,786; aor.,832,834b, 839,\n847; fut.,943; int.,1002c, 1003,\n1017.\nvrtrd, 1185b.\nvrddhi, 27,235 ff.\nyvrdh, pf.,786; aor.,847; fut., 943.\nyvrs, pres., 753.\nyv'rh, pres., 753; aor., 916, 920.\nvac, quasi-root, 854.\n-vya, 1228c.\nyvyac, 108g, 1087f\n;pres., 682; pf.,\n785,794b.\nvyanjana, 31.\nyvyath, pf.,785.\nyvyadh, pres., 761c,767; pf.,785,\n794b;pple, 954b.\nyvyay, pres., 761d.2.\n}/vyas, 108g.\nyvydorm,pres., 761d.2;pf.,785,\n794b,801 c\n;aor.,847\n;pple,954 c\n;\ncaus., 1042 d.\nyvraj, euph., 219; aor.,899 c.\nyvroff,euph., 221\n;pres., 753;pple,\n957c.\nyvli, pres., 727; pple, 957a;int.,\n1002a,1017; caus., 1042e.\nC,pron. etc.,59,63 4,119; asfinal,\n145; inintrnl combn, 218;with\nprecedingtorn,203.\n-fa,1229.\nI/pans, pf.,790 c.\n}/pafc, pres., 708; aor.,847; des.,1030.\ntakdn, cdkrt, 398, 432.\nytad, pf.,786.\nytap, aor., 883.\nj/pam, pres., 634, 761a,763;aor..\n847; pple,955 a.\nytafj pf.,794e.\nj/pas, aor., 839.\n-fas, advbl, 1106.\nytfi, pres., 660, 662,761d.3\n;pple,\n954c;caus., 1042 d.\n]/pas (orpis), pres., 444, 639,675;\naor., 847, 854; pple, 954e;ayfy\nfrom, 225.\nytinj, euph., 219; pres., 628.\nyti? 'leave', euph., 226c;pres., 694,\n761b;aor., 847.\nytiSjsee pas.\n>/pi', pres.,6289; pf.,807; pple.\n956.}/puf, pres., 761a;aor., 847; int.,\n1002 a.\npres., 753, 758; aor.. 852.\n*,pres., 753, 761 a.\ntiiska, 958.\nycr 'crush', euph., 242; pfcs., 727,\n731; pf.,793e;aor., 900,904 b.\nyfcand, int., 1002 c.\n]/pcam, pres., 727.\nytnath, pres., 631; aor., 867.\nj/p7/aorp*,pres.,761b,d.l;pf.,794b;\npple, 954c,957 a.\nytroth, pres., 727, 732,1066b,758;\npf.,794e;pple, 956.\n|/pram, pres., 761a,763; aor.,847;\npple, 955 a.\nj/pra, pres., 761d.l;pple,957a;caus.,\n1042 d;caus.aor., 861, 1047.\nj/pri, aor., 831, 868, 889.\nj/pn, pres., 727.\n|/pm, euph., 243; pres., 708, 710;\npf.,797b;aor., 831,8369.\ny^rus, 108g.\nyflis, euph. ,226b,c,761c;aor.,916.\nfvdn, dcln, 427.\ny^vas, pres., 631; int., 1002 a.\nycvaoi tvi, pf.,786,794b;aor.,847,\n868; pple, 957 a.\nytvit, aor., 832, 847, 890.\na,pron. etc.,59,612,46,120,182;\nasfinal, 145;schanged to,1808;\nrecurrence avoided, 181b,184 c,\n1028i;asroot final, '182 a,184b,\n225 6;changes succeding nton,\n189 ff.;assim. ofdental after, 197;\nfromp,218.\n-sani(or-sani}, infin. in,970h,978, '\n1159c,1160a.\nBOS,euph., 146end, 199b.\n-s'e(or-M),infin. in,970c,973.\nysthlv, pres., 745g;pf.,789 c.\n-syai (or-syai),infin.in,970 g,977.\ns,pron. etc., 59,60;sand rassurd\nandsonant, 117a,164;sorras\nfinal ofcertain forms, 169end\n;as\nfinal, 145,169,170a,639;combina-\ntions offinal rdcls,145,1668,\n833;ofother,1707;finalas,175\n6\n;as,177\n;stos,1808;tadds\ntbefore, 199c;finalnadds (retains)\ns,208 9;slostbetween mutes,\n233;ins-aor., 834, 881,883; a\nbeforeamofgen. pi.,313, 496;\ninaor.,874ff.;infut., 931ff.;in\ndesid., 1027 ff.\n"
+    },
+    {
+        "page": 490,
+        "content": "474 SANSKRIT INDEX.\n-s,advbl, 1105.\n-sa,1197.\naa~,H21b;incompsn, 1288c,1304c.\nsamwrta a,21.\nsdkhi, dcln, 343 a.\nsakthdn^dkthi, 343f,431.\nj/sa^/t, pres., 708; aor., 836, 838.\nI/sac, pres., 660;pf.,794d;safefrom,\n673, 675.\ni/saj orsanj, euph., 219; pres., 746;\naor.,834c; des., 10281.\nysad, pres., 748; aor., 847, 853.\nysan, pies., 713; pf.,804; aor.,\n847,899c;pple,955b ;int.,1002 c;\ndes.,1028 g,i.\n-sani,inlin.in,see -sani.\nsamdhi, 109.\nsarhdhyaksara, 28,30.\nsannatara, 90c.\nysap, aor., 847.\nysabhag, so-called, 108f,1067.\nsamanaksara, 30.\nsamprasarana, 252.\nsamraj etc.,213 a.\n-sara, 1201.\nsardghorsardd, 389.\nsdrva, dcln, 524;incompsn, 1251e,\n1298 c.\nyaafc, 444, 673, 675.\nsas,euph., 176 a.\n-sas, 1152.\nT/O/I, euph., 186,223,224b;pf.,786,\n790b\n;aor.,837 9,897,899c\n;des.,\n1030;atend ofcmpds, 405.\ni/saorsi'bind', pres., 727,761d.3;\naor., 830, 837,868; pple, 954c;\ncans., 1042d.\n-sat, advbl, 1108.\nyaadh, aor., 861.\n-sana, ppial words in,897, 1175.\nysantv, so-called, 108 f.\nysi,see sa.\nysic, pres., 753,758; aor., 847.\n|/szd/i, pres.,761c.\nyslv, pres., 761c,765; pple, 954e;\ndes., 1028h,i.\nysu 'press out', pres., 708; aor.,839.\nysu 'imper, pres., 626, 753, 755.\nsu-,1121d;incompsn, 1288b, 1304b.\nysu, pres., 628; pf.,789a; fut.,939.\ny$uc, aor., 861.\nysr, pf.,797b, 807; aor.,847; int.,\n1002 b.\nysrj, euph., 216.9, 219; pres., 753;\naor.,834b,c,890.\nysrp, euph., 151d;aor.,834d,847.\n-86, infln.in,see -se.sosman, 37.\nsk]original ofch,42.\nyskand, aor., 833,890; int., 1002c.\nyskabhorskambh, pres., 727, 730,\n732,1066b;pf.,786.\nysku, pres.,626,708 ;int.\n,1002a, 1017.\nystan, pres., 631; aor.,899c; int.,\n1002 b.\nystabhorstambh, euph., 233a;pres.,\n727, 730, 732,1066 b.\nysta, pres.. 761d.l.\n-stat for-tat, advbl, 1100 b.\nystu, pres., 626,633; pf.,797b; aor.,\n847,894b,d.\nystr, euph., 242; pres., 708, 727;\nfpf.,807; aor., 831,834a,885, 900.\nsir,dcln, 371.\nystrh, aor., 916.\nystya, pres., 761d.l.\nstri, dcln, 366,367b.\nystha, euph., 233a;pres., 671,749a\n;\naor.,830,834a,836-40, 884,894c;\npple,954c; caus.,1042d; caus.aor.,\n861,1047;inppial periphr. phrases,\n1075c.\n-ma, 1195.\nysna, caus., 1042 d.\nysnih, euph., 223.\nysnu, pres., 626.\n-snu, 1194.\nsparfa,31,32.\nyspay. aor.,834 c.\nypr, 'pres., 708; aor., 831,834aff.\nysprdh, aor.,834b.\nl/sjorp, euph., 218; pres.. 753; aor.,\n916, 920.\nysprh, euph., 223.\nyspha, pres., 761d.l; pple, 954c;\ncaus., 1042 f.\nysphur, pres., 753, 756.\nsphotana, 230 c.\nsma, pres.inpast sense with,778b.\nysmi, caus., 1042 e.\ni/syawd, aor., 890; fut., 943; int.,\n1002 c.\nysyam, pf.,794 e.\nsyds, euph., 176 a.\n-syai,infln.in,see-syai.\nsrdj, euph., 219.\ni/sras orsrans, euph., 168,833; aor.,\n833, 847.\n1/aridfc, aor., 847.\nysriv, pres.,761c,765.\nysru, pf.,797b;aor., 868.\nVsvaj. euph., 219; pres., 746; pf.,\n794 e.\nsvdtavas, euph., 168,415 d.\n"
+    },
+    {
+        "page": 491,
+        "content": "SANSKRIT INDEX. 475\nysvad, pple, 955 c.\nj/sran, pf.,794e;aor.,899c;int.,\n1002 c.\nysoap, pres., 631; pf., 785,794b;\naor., 867; des., 1028h.\nysvar, aor., 890.\nsvar, dcln, 388d.\nsuara, 30,81.\nsvarabhakti, 230 c.\nsvarita, 81.\n,euph., 168,415d.\nA,pron. etc., 59,656,119; as\nfinal, 147;compensating aspiration\nofinitial, 147,155; withfollowing tor\nth,160; withpreceding mute, 163;mbefore liandanothercons., 213e;\nreversion togh,214ff.,222; inIn-\nflection, 402,637; inpf., 787;in\nint.. 1002end;indes., 1028f;\ninternal combn, 2224;dupli-\ncation ofacons, after, 228;loss\nbeforehi,1011.\n-ha, advbl, 1100a,1104 end.\nyhan, euph., 192b, 216.9, 402, 637,\n787; pres., 637, 673,749a; pf.,\n794d,805; fut.,943; pass., 998d;\npple. 954d; int., 1002b, c;des.,\n1028e, f;cans., 1042 f;root-noun,\n383b,402.\nyhar, pres., 761 a.\nyhas, jaks from, 640.yha 'move', 108g;pres., 660, 664;\ndes., 1028 d;caus., 1042 d.\nyha 'leave, 108g; pres., 665,761b\n;\naor., 830, 889, 912; pple, 956,\n957a;cans., 1042d; caus. aor.,\n861, 1047.\nyhi,euph., 192c,216.9, 674,749a,\n787; pres., 708,716; aor., 831,\n839, 847,889,894d;des.,1028 f.\nhf,595d,1122a.\n-hi, advbl, llOOc.\nyhins, euph., 183; pres., 687, 692,\n694, 696; des., 1031.\nyhinv, 716.\nyhld, euph., 240; pf.,786.\nyhii, pres., 645,647bend, 652; aor.,\n831; periphr. pf. etc., 1071 d,\n1073a,c.\n]//m,see hva.\nyhr 'seize', aor., 890.\nyhr'beangry', pres., 727.\nhfd, hfdaya,397.\nyhr?, 'pres., 761a;int., 1002b.\nyhnu, pres., 626.\nyhvaorhu, pres., 761d.2,753; pf.,\n794b; aor.,834a, 847,852,887a,\n912; fut.,935c;int.,1002a;caus.,\n1042d;periphr. pf.,1071 d.\nyhri, pres., 645; caus., 1042e.\nyhru, pres., 727.\nyhvrorhvar, euph., 242; pres.,\n682; aor., 890,899 c.\n"
+    },
+    {
+        "page": 492,
+        "content": "GENERAL INDEX.\na-aorist (simple aorist, 2j,846 54:in\nthelaterlanguage, 846 ;rootsforming\nitintheolder language, 847; in-\nflection, 848;modes,84951;par-\nticiples, 852;irregularities, 8534.\na-class(first, &7m-class) ofverbs,73450: formation ofstem, 734;\ninflection, 735 43;roots ofthe\nclass, 744; irregularities, 745 50.\na-class oraccented a-class(sixth,\ntad-class) ofverbs,7518: forma-\ntion ofstem, 751; inflection, 752;\nroots oftheclass, 753 4;irregu-\nlarities, 7558.\nablative case, usesof,28993; ab-\nlative ofcomparison, 292b; with\nprepositions, 293, 1128; used ad-\nverbially, 1114; abl.infinitive, 983;\nabl.useofadverbs in*s,1098.\nabsolute useofgenitive. 300; oflo-\ncative, 303b.\nabstract nouns, secondary derivation\nof,1206,123640.\naccent, general, 8097 :itsvarieties,\n806;modes ofdesignating, 879\n;\nillustration ofRV. method, p.459;\nover-retinements ofHindu theory,\n90;modern delivery ofancient\naccented texts, 91;nosentence\naccent, 92;accentless words, 93;\nwords doubly accented, 94,12556,\n1267 d;freedom ofplaceofaccent\n95; changes ofaccent invowel\ncombination, 128, 130, 135;\naccent indeclension, 314 20;of\nvocative, 314; change ofaccent in\nmonosyllabicetc. declension, 316\n8;innumeral, 482 3;offrac-\ntionals, 488;ofcase-forms used as\nadverbs, 1111 er1112e, 1114d;dif-\nferent accent ofaction-nouns and\nagent-nouns, 1144;ofdeterminative\nandpossessive compounds, 1295;\naccent ofpersonal endings, 5524\n;ofpersonal verb-forms inthe\nsentence, 591 8\n;ofperiphrastic\nformations, 945,1073\n;ofcompound-\nedverb-forms, 1082 5; accent\n'inprimary derivation, 1144;insec-\nondary, 1205;incomposition,\n1251.\naccusative case, usesof,269 77 :\nwith verbs, 270 4;with nouns\nand adjectives, 271 2;withpre-\npositions, 273,1129; cognate, 275;\nadverbial, 276,1111;double, 277;\naccus.infinitive, 981,986 8;ge-\nrund, 995; accus. aspriormember\nofcompound, 1250 a.\naction-nouns andagent-nouns, chief\nclasses ofprimary derivatives. 1145\ng\nactive voice,inverbs, 528 9.\nacute (udatta) accent, 81.\nad-class ofverbs see root-class.\nadjective,itsdistinction from noun,\n322; formation ofcompound adj..\n3235, 1292 ff.\n;inflection ofadj. :\n321 465;comparison, 466 74 :\nadj.pronominally inflected, 522 6.\nadjective compounds, secondary, 1247,\n1292 1310; ofother than possess-\nive value, 1294; adj. copulative\ncompounds, 1257.\nadverbs, 10971122: adv. byderi-\nvation, 1097 1109; case-forms\nused asadv., 1110 17;adverbial\ncompounds, 1313;verbal prefixes\netc.asadv., 1118 20;inseparable\nprefixes. 1121; other miscellaneous\nadv.,1122\n;adv.usedprepositionally.\n1123ff.;adv.copulative compounds,\n1259.\nagent-nouns see action-nouns.\naggregative compounds seecopu-\nlative compounds.\nalphabets used forwriting Sanskrit,\n1;older Indian, 2;theDevana-\n"
+    },
+    {
+        "page": 493,
+        "content": "GENERAL INDEX. 477\ngan alph.,i17; characters and\ntransliteration, 5;arrangement, 7;\ntheory ofuse,89;vowel-writing,\n10;consonantcombinations, 12\n15;othersigns, 11,16;numeral\nfigures, 17\n;names ofcharacters,\n18; signs and transliteration of\nanusvara, 73.\nalphabet, spoken seesystem of\nsounds.\nalterant vowels, changing followings\nto*,180.\nanalysis oflanguage into itselements,\n98106; anal, ofcompound words,\n1248.\nantithetical construction, itsinfluence\nonaccent ofverb,5967.\nanusvara,itspronunciation etc.,70\n2;signs andtransliteration, 73:\nseealso n,m.\naorist tense, 532;itsuses,92630;\ninprohibitive expression, 57980;\naor. system, 535,824930:\nclassification offorms ofaor., 824;\ncharacter andoccurrence, 8257;\nsimple aor., 824,828: 1.root-aor.,\n82941; passiveaor.3dsing.,\n8495; 2.a-aor.,84654;3.re-\nduplicatedorcausativeaor.,856\n73; sibilant-aor., 874920: 4.s-\naor.,878 97;5.ia-aor.,898910;\n6.sis-aor., 911 15;7.sa-aor.,\n916 20;aor.optative orprecative\noflater language, 921 5;aor. in\nsecondary conjugation, 1019, 1035,\n1046, 1068; periphrastic aor.,\n1073 b;aor.stem inderivation,\n1140c.\nappositional possessive compounds,\n1302.\narorrinrootandstem forms, 108d,\n237.\narticle, indefinite, represented laterby\neka,482 a.\naspirate mutes, phonetic character etc.\nof,378;theirdeaspiration, 114,\n1535; restoration oflostaspira-\ntionto,141, 147, 155; derivation\nofhfrom, 66;sonant aspirate with\nfollowing t,th,160: andsee\nthedifferent letters.\naspiration (ft),itspronunciation etc.,\n59,656: andsee h.\nasseverativeparticles, 1122 a.\nassimilation ineuphonic combination,\n11520; with orwithout change\nofarticulateposition, 116; surdandsonant, 117, 156 64; nasal,\n117e, 198b,199b;I,117e,206;\ndental tolingual andpalatal, 118;\nothercases,11820.\naugment, 585 7;itsomission, 587;\nuses ofaugmentless preterit per-\nsons, 563,587;withmaprohibit-\nive,57980.\navyaylbhava compounds, 1313.\nbahuvrihi compounds seeposses-\nsivecompounds,\nbenedictive see precative.\n6/zw-class ofverbs see a-class.\ncardinal numerals, 475;their combi-\nnations, 47681\n;inflection, 482\n5;construction, 486; derivatives,4879.\ncase-endings see endingsofde-\nclension.\ncase-forms used asadverbs, 1110 17;\nchange ofaccent insuch, 1111e,\n1112e, 1114d;their prepositional\nuses, 1125; case-forms incompo-\nsition, 1250.\ncases, 266;their order ofarrange-\nment, 266; uses,267305:\nandseethedifferent cases.\ncausative conjugation, 540, 607, 775,\n856 ff.?104152; relation toso-\ncalled tenth orcw-class, 607,1041;\ntodenominative, 1041, 1056;for-\nmation ofstem, 1041 2;inflection,\npresent-system, 1043; perfect,\n1045;attached reduplicated aorist,\n10467, 856ff.;other aorist forms,\n1048-9; futureetc.,1050; verbal\nnouns andadjectives, 1051;deriva-\ntive ortertiary conjugations from\ncaus. stem. 1052; declinable stems\nfromdo.,1140b.\ncerebral mutes see lingual.\ncircumflex (svarita) accent, 81 6,\n90b; independent, 814;itsva-\nrieties, 84; enclitic, 85;their dif-\nference, 86; designation, 8790;\noccurrence fromvowel combinations,\n128c,130, 135.\nclauses, simplicityofcombinationof,\n1131; dependent clauses, modein,\n581, 950; accent ofverbin,595.\ncollective singular form ofcopulative\ncompounds, 1253b;inVeda, 1255c,\n1256 b.\ncombination ofelements, 1024;\n"
+    },
+    {
+        "page": 494,
+        "content": "478 GENERAL INDEX.\neuphonicrulesfor,109 !260;dis-\ntinction ofinternal arid external,\n109 12; general arrangement of\nrules,124.\ncomparisonofadjectives etc., 466\n74; primary,inlyas andistha,\n467 70,1184; secondary,iniara\nandtama,4713, 1242;inraand\nma,474\n;inflection ofcomparatives\ninyas,4635;cornp.ofnouns,\npronouns, prepositions, 473 4,520,\n1119; ofverbs, 473, llllc.2;\nparticlesofcornp., 1101, 1102b,\n1107, 1122d.\ncompensatory vowel-lengthening, 246.\ncompositionofstems seecompound\nstems.\ncompound conjugation, 540,1076 95 :\nroots with verbal prefixes and like\nelements, 1076 89;accent ofcomp.\nforms, 1082 6;irregularities, 1087;\nroots withinseparable prefixes,\n1089, 1121;withnounandadjective\nstems,10905.\ncompound stems,formationof,103,\n12461316: difference ofearlier\nandlaterlanguageastocomposition,\n1246 :classification ofcompounds,\n1247; their analysis, 1248; rules\nofphonetic combination, 1249;case-\nforms asprior member, 1250; ac-\ncent, 1251; copulative comp., 1252\n61;determinative :dependent,\n126278; descriptive, 127991;\nsecondary adjective:possessive, 1292\n1308; participial, 1309; prepo-\nsitional, 1310; adjective comp.as\nnouns and asadverbs, 1311 3;\nanomalous comp., 1314; stem- finals\naltered incomp., 1315; irregular\nconstruction ofcomp., 1316.\nconditional tense, 532, 940 1;its\nuses, 950; conditional uses ofop-\ntative andsubjunctive, 581 b.\nconjugation,verbal inflection, 527\n1095; general, 52798: voice,\n528 31;tenses and theiruses,\n532,776-9, 8213, 92630,\n948 50;modes and their uses,\n533,55782, 9215; tense-sys-\ntems, 535; present-system, 599\n779;perfect- system, 780823;\naorist-systems, 824 930;future-\nsystems, 931 50;number and\nperson, 536;personal endings, 541\n56;verbal adjectives andnouns,\n5379,'951 95;secondary con-jugations, 540,9961068; peri-\nphrastic andcompound conjugation,\n540,106995.\nconjunctions, 1131 3.\nconsonants, pronunciation etc., 31\n75: mutes, 3250; semivowel-.\n518;spirants, 5966; visarya.\nandanusvara etc.,6773;quan-\ntity, 76; cons, allowed asfinal.-,\n122,139 ff.\n;occurringatend of\nstems andendings, 139: --and\nseethedifferent classesand\nletters.\nconsonant combinations,extension and\nabbreviationof,121,22733.\nconsonant-groups, how written inde-\nvanagari, 9,12 5.\nconsonantal stems, declension of,377\n465; their classification, 382.\ncontemptuous prefix, 506,1121e; do.\nsuffix, 521,1222 b.\ncopulative compounds, 1247, 1252\n61; ofnouns,12536; adjectives,\n1257; adverbs, 1259; numerals,\n1261; copulativesinlaterlanguage,\n12534; inRig-Veda, 1255; in\nAtharva-Veda, 1256;accent, 1258;\nquestion ofpossessives from cop.,\n1293 a.\ncur-class ofverbs, 607, 1041, 1056:\nandsee causative conjugation.\ndative case, usesof,285 8;dat.\ninfinitive, 982,986;dat.used ad-\nverbially, 1113.\ndeaspiratiori ofaspirate mutes, 114,\n153 5;consequent re-aspiration\nofinitial, 141, 147, 155.\ndeclension, ingeneral, 261320:\ngender, 263;number, 264 5;case,\n266; uses ofthecases,267305;\nendings ofdecl.,30610;varia-\ntion ofstemandinsertions, 311 3\n;\naccent, 314 20; decl. ofnouns\nandadjectives, 321 465 :1.a-stems,\n32634;II. iandw-stems, 335\n46;III.a,I,andu(and diph-\nthongal) stems, 34768; IV. r-\nstems, 369 76;V.consonant-\nstems, 377465 :A.root-stems\netc.,383410; B.derivative stems\ninas, is,us,411 9;C.inan,\n42037; D.in in,43841;\nE.inant,442 57; F.invans,\n45862; G.inyas,4635;\ndecl. ofnumerals, 482 5;of\n"
+    },
+    {
+        "page": 495,
+        "content": "GENERAL INDEX. 470\npronouns,491 521\n;ofadjectives\ninflected pronominally. 522 6.\ndeclinable steins, derivation ofsee\nderivation.\ndecompound compounds andtheir ana-\nlysis, 1248.\ndemonstrative pronouns. 495 503.\ndenominative conjugation, 540,1053\n68 :formation without sign,1054;\nwith sign j/a,from stems ofvarious\nfinal, 1055 63; relation tocausa-\ntive, 1056, 106T; with signs sya,\nfcarra/a, 1064 5\n;with aya, beside\nna-class verbsetc., 1066; inflec-\ntion, 1068;declinable stems from\ndenom. stem, 1149, 1178h, 1180b.\ndental series ofmutes(t, tft,d,dh,n),\npronunciation etc., 478;dent,\ncharacter ofI,25;ofI,51,53; of\ns,60;assimilation ofdent, topa-\nlatals andlinguals. 118,196203,\n205\n;dent, sibilant andnasal con-\nverted tolingual, 180 95 :and\nseethedifferent letters.\ndependent clause, accent ofverbin,\n595.\ndependent compounds, 1247, 1263,\n126478; noun, 1264; adjective,\n1265;their varieties, 126678:\nwith ordinary noun oradjectiveas\nfinalmember, 1267 8\n;with root-\nnoun, 1269; derivative ina,1270;\nana, 1271; ya,1272; participle\nintaorna,1273; ti,1274; in,\n1275; i,1276; van,man, etc.,\n1277 8;dep. comp.inpossessive\nuse, 1296.\nderivation ofadverbs, 10971109;of\ndeclinable stems, 11361245: in\ngeneral, 113642; primary, 1143\n1201; secondary, 1202-45.\ndescent, adjectives andnouns indicat-\ning,1206.\ndescriptive compounds, 1247, 1263,127991;ofordinary adjective with\nnoun, 1280;ofappositional noun\nwith noun, 1280 b;withparticiple\nasfinal member, 12834; with\ngerundive, 1285; with root-stem,\n1286;with other verbal derivatives,\n1287; with inseparable prefixas\nprior member, 1288; with verbal\nprefix etc.,1289; with other ad-\nverbial words, 1290; withpurva\ninreversedposition, 1291;descr.\ncomp. inpossessive use,1297 if.\ndesiderative conjugation, 540,102640;meaning,102(5. 1040: forma-\ntion ofstern, 1027 9;abbreviated\nstems. 1030 :useofunion-vowel/,\n1031\n;inflection, present-system,\n1032; other forms,10337; deri-\nvative ortertiary conjugations from\ndesid. stem, 1039\n;desid. from cau-\nsative stem, 1052 b;declinable stems\nfrom desid. stem, 1140b, 1149,\n1159b, 1161c,1178 f.\ndeterminative compounds, 1247, 1262\n91; dependent. 126478; de-\nscriptive, 1279 91;inpossessive\nadjective use,1293 ft1\n.\ndevata-dvandva compounds, 1251a,\n1255.\ndiminutives, secondary derivationof,\n1206, 1222, 1243.\ndiphthongs (e, ai, o,au) tpronuncia-\ntion etc.,27-30;protraction of,78;\neuphoniccombination asfinals, 131\n5:andseethedifferent\nletters.\ndiphthongal stems, declension of.360\ndiv-class ofverbs seeya-class.\ndouble stems, present, 815; aorist,\n894d,897.\ndoubling ofaspirate mutes, 154;of\nafinal nasal, 210;ofeft,227; of\nfirstconsonant ofagroup, 229;of\naconsonant after r(and h, I,v],\n228.\ndualnumber,itsuse,265;itsforms,\n308;inpersonal pronoun,492.\ndual finalse,I,uuncombinable, 138 a.\ndvandva compounds see copula-\ntive.\ndvigu compounds,1312.\neighthclass ofverbs see w-class.\nemphasis,accent ofverb for,598.\nemphatic pronoun,513.\nendings,ofinflection andderivation,\n99,100;ofdeclension, 30610;\nofsingular, 307; dual, 308; plural,\n309; normal scheme, 310; end. of\na-stems, 327 9;ofiandw-stems,\n3368;ofradical a, *,w-stems,\n349;ofderivative do.,363; ofr-\nstems, 371;ofpersonal pronouns,\n492 3;ofgeneral pronominalde-\nclension, 496;end. ofconjuga-\ntion, 541ff.;of1stsing., 543; 2d,\n544; 3d,545; of1stdu., 546;\n2dand3d,547; of1stpi.,54b;\n2d,549; 3d,550; normal schemes,\n"
+    },
+    {
+        "page": 496,
+        "content": "480 GENERAL INDEX.\n553, accent, 552 4;irregularities,\n555; end. ofsubjunctive combined\nwith mode-sign, 5602;ofopta-\ntive,566;ofprecative, 568\n;tatof\nimperative, 570 i;end. ofde-\nrivation see suffixes.\neuphonic combination ofelements, 102\n4\n;rules respecting it,109 226.\nexclamatory prefix from interrogative\npronoun, 506, 1121e.\nfifth class ofverbs see\nfinals, permitted, 122,139 52;final\nconsonants ofstems andendings,\n139.\nfinalclauses, modes usedin,581 c.\nfirst class ofverbs see a-class.\nfirst general conjugation ofverbs,its\ncharacteristics, 604.\nforms, stronger andweaker,ofroots\nandstems, 107 8: andsee\nvariation ofstem.\nfourth class ofverbs see ya-\nclass.\nfractional use ofordinals, 488.\nfrequentative conjugation seein-\ntensive.\nfuture passive participles see ge-\nrundives.\nfuture tenses, 532;their uses,948\n9;fut.systems, 535,93150;s-\nfuture and conditional, 932 41;\nperiphrastic future, 9427.\ngender indeclension, 262 3.\ngeneral and special tenses, 599.\ngenitive case, usesof,294 300 :with\nprepositions, 299, 1130; gen. ab-\nsolute, 300;loss ofaccent ofgen.\nwith vocative, 314; gen. infinitive,\n984; gen. used adverbially, 1115;\naspriormember ofcompound, 1250d.\ngerunds, 539, 989 95; their use,\n989, 994; ger.intva,990, 991,\n993;_in ya(orya),990,9923;\nintvaya andtvi,993b\n;intvanam\nandtvlnam, 993c;inam,995.\ngerundives,orfuture passive partici-\nples,9616, 12128; ger.inya,\n9623, 1213;intavya, 962, 964,\n1212 d.2;inaniya, 962, 965,\n1215b;intva,966a,_1209g;in\nenya, 966b,1217;inai/j/a, 966c,\n1218; inelima, 966d,1201; ger.\nincomposition, 1285.\ngrave (anudatta) accent,81.\ngana-strengthening, character andoc-currenceof,235 43.andpassim;\ninprimary derivation, 1143a;inse-\ncondary, 1203 a,1204 end.\nguttural series ofmutes(k, kh, g,\ngh,n),pronunciation etc.,3941\n;\nasserted gutt. character ofa,20;\nofft,65\n;palatals from original\ngutt.,413; fandhdo., 64,66;\nreversion ofpalatals etc.togutt. form,\n43,64,142, 145, 147, 214ff. :\nandseethedifferent letters.\nheavy andlight syllables, 79.\nhiatus, avoidanceof,113,12538;\nnotavoided inVeda, 113, 125;its\noccurrence asresult ofeuphonic\nprocesses, 1324, 175b,177.\nftw-class ofverbs seereduplicat-\ning class.\nimperative mode, 533, 569,572 ff.;\nscheme ofitsendings, 553;its1st\npersonsoldsubjunctive, 533, 574,\n578;impv.form intatand itsuses,5701;Vedic 2dsing, insi,624;\nimpv. useofinfinitives, 982d.\nimperfect tense, 532;itsuse, 779.\nimperfect time, nodesignation of,532.\nincrement anddecrement ofelements,\n123,234 ff.\nindeclinables, 98,10961135: ad-\nverbs, 1097 1122; prepositions,\n112330; conjunctions, 11313;\ninterjections, 1134 5;derivative\nstems fromindeclinables, 1138,\n1245.\nindefinite use ofinterrogative andre-\nlative pronouns, 507, 511.\ninfinitives, 538,96888; later, 968;\nearlier, 96979; uses,9808.\ninseparable prefixes, 1121;inde-\nscriptive composition, 1283ff., 1288;\ninpossessive, 1304.\ninsertions between stem andending\nindeclension, 313.\ninstrumental case, usesof,278 84\n;\nofseparation, 283;with preposi-\ntions, 284, 1127; gerundial, 989;\nused adverbially, 1112;asprior\nmember ofcompound, 1250 b.\nintensive (orfrequentative) conjuga-\ntion, 540,1000 25; character and\noccurence, 10001\n;reduplication,10023;inflection, present-system,\n100417; derivative middle in-\nflection, 1016 7;forms outside\npresent-system, 1018 9,1025;\n"
+    },
+    {
+        "page": 497,
+        "content": "GENERAL INDEX. 481\ndoubtful iutens. formations, 1020\n4;derivative ortertiary conju-\ngations from intens. stem, 1025.\ninterjections, 11345;their final\nvowel uncombinable, 138 f.\ninternal and external combination,\ndistinctionof,10912.\ninternal change, questionofderiva-\ntion by,1208 end.\ninterrogative particles, 1122c.\ninterrogative pronoun, 504 7;its\nindefinite use, 507; exclamatory\nprefix fromit,506, 1121e.\n*.?-aorist, 898 910: formation ofstem, '\n898 900; inflection, 9012;roots\nmaking it,903; irregularities, 904;\nmodes, 905 8;from secondary con-\njugations, 1019, 1035, 1048, 1068.\njihvamuliya-sipiTa.nt, 69,170d.\nkarmadharaya compounds seede-\nscriptive compounds,\nfcn-class ofverbs see na-class.\nlabial series ofmutes(p,ph, b,bh,\nm), pronunciation etc., 4950;\nlab. character ofw,u,20; ofv,\n51,5758: andseethedif-\nferent letters.\nlengthening ofvowels information\nandinflection, 244 6;offinal\nvowel incomposition, 247; inthe\nsentence inVeda, 248.\nlight andheavy syllables, 79.\nlightening ofaoratoan iorw-\nvowel, 2498.\nlingual series ofmutes(t, ife,d,dft,\nnj,pronunciation etc.,456;non-\noriginality, 46;ling, character of\nr,25; ofr,512; ling. I,Send,\n54;ling, character of,61;assimi-\nlation ofdentals toling., 118,\n196ff.;lingualization ofsand n,18095: andseethedif-\nferent letters.\nlocativecase, usesof,301 5;loc.\nabsolute, 303; ofgoal ofaction,\n304; withprepositions, 305,1126;\nusedadverbially. 1116;loc. infini-\ntive,985\n;loc.use ofadverbs in\ntra,1099; loc. aspriormember of\ncompound, 1250 c.\nlongand shortquantity, 76 9.\nmanner, particles of,11012, 1122g.\nmiddle stem-foim indeclension, 311.\nWhitney, Grammar.middle voice, 528 30;itsuse as\npassive, 531.\nmode inverbalinflection, 533; sub-\njunctive, 557 63; optative, 564\n8;imperative, 569 71;uses of\nthemodes, 57282.\nmultiplicative numeral adverbs, 489,\n11046.\nmutes, seriesof,their pronunciation\netc.,3250:classification, 328;\nguttural series, 3941; palatal,\n424;lingual,456; dental, 47\n8;labial, 4950\n;mutesper-\nmitted asfinals, 1413;anomalous\nconversions from one series toan-\nother, 151: andseethedif-\nferent series.\nna-class (ninth, fcn-class)ofverbs,71732: formation ofstem, 717;\nInflection, 71826; roots ofthe\nclass, 727; irregularities, 728 32\n;\naccompanying denominative inat/a,\n732.1066 b.\nnasal assimilation, 117, 161, 198b,\n199b.\nnasal class(seventh, rwd/t-class)of\nverbs, 683 96 :formation ofstem,\n683; inflection, 68493; roots of\ntheclass, 694;irregularities, 695\n-6.\nnasal increment instrong forms, 255.\nnasal mutes (n,n,n,n,mj, 34,36;\ntheir occurrence asfinals, 143\n;\nduplication, 210;assimilation of\npreceding mute, 161,198b, 199b;\nnasal spirantoranusvara, 703\n;\nnasal semivowels, 71,206,213c;\nnasal vowels,71.\nnegative particles, 1122b; neg. pre-\nfix,1121 a.\nninth class ofverbs see na-class.\nnominative case, usesof,267; nom.\nuseofinfinitive, 987; nom. form\nasparticle,1117.\nnoun and adjective,distinction of,\n322;inflection ofnouns --see\ndeclension,\nnw-class(fifth, sw-class)ofverbs,\n697716: formation ofstem, 697;\ninflection, 698707;roots ofthe\nclass 708; irregularities, 70912,\n716.\nnumber indeclension, 264 5;in\nconjugation,536.\nnumerals, 475 89 :simple cardinals,\n475\n;their combinations forodd\n31\n"
+    },
+    {
+        "page": 498,
+        "content": "482 GENERAL INDEX.\nnumbers, 47681\n;inflection, 48'2\n5;construction, 486; ordinals,\n4878;other num.derivatives,\n489;num. figures, 17;possessive\ncompoundswithnum., 1300; num.\nordvigu compounds, 1312.\nomission, sign indicating, 16.\nonomatopoetic words, 1091, 1135b.\noptative mode, 533,5648;itsfor-\nmation, 564 5;scheme ofend-\nings combined with mode-sign,\n566; precative, 567,9215;\nscheme ofprec. endings, 568; uses\nofopt.,57382.\nordinal numeraladjectives, 487 8.\npada-endingsindeclension, 111.\npalatalseries ofmutes(c,c/t,j,jh,n),\npronunciation etc.,424;derived\nfrom original gutturals, \"42; rever-\nsion toguttural form, 214ff.;eu-\nphonic combinations, 1189,214\n20;treatment asfinals, 142; pal.\ncharacter of f.I,20;ofy,51,56;\noff,634:andseethedif-\nferent letters.\nparticipial compounds, 1247, 1309.\nparticiples, 534,537,5834, 11727;\nofpresent-systems,619 etc. etc.\n;\nofperfect, 8027;ofaorist, 840,\n852, 872, 897, 909;offuture,\n939; passive part.,9508, 1176\n7;active,intavant, 95960;\nfuture passive, 961 7;ofsecond-\naryconjugations, 10123, 1019,\n1032, 1037, 1043, 1050-1, 1068;\npart.inpossessive composition, 1299;\ninflection ofpart,inant,443\n9;invans, 458 62; part.-\nphrases, periphrastic, 1074 5.\npassive conjugation, 531, 540, 998;\npresent-system (7/a-class), 76874;\naorist 3dsing. ,842 5\n;periphrastic\nperfect,1072\n;participle intaor\nna,9527, 11767; future par-\nticiples,961 6(and seegerund-\nives); pass,useofinfinitive, 988;\npass,ofsecondary conjugations,\n1025, 1039, 1052a;pass, con-\nstructions, 282, 999.\nperfect tense, 532; scheme of its\nendings, 553; uses,8213;\nperf.-system, 535,780823: for-\nmation ofstem, 78194; redu-\nplication, 78291;strong andweak\nstem-forms, 792 4\n;endings andtheir combination withstein, 795ft\".;\nunion-voweli,7968;inflection,\n800; irregularities, 801; participle,8027; itsinflection, 45862;\nmodes, 80816; pluperfect, 817\n20; periphrastic perf., 1070\n73.\nperfect time, expressed byso-called\naorist, 532, 928.\nperiphrastic conjugation, 1069 75;\nfuture, 931,9427;itsuses, 949;\nperfect, 1070-3, 1018, 1034, 1045,\n1068;aorist andprecative, 1073b;\npresent. 1073c;periph, participial\nphrases. 1074 5.\npersoninverbalinflection, 536.\npersonal endings see endings of\nconjugation.\npersonal pronouns, 491 4\n;nouns\nused assuch, 514.\nplace, particles of,1099, 1100, 1122e.\npluperfect tense, 532,81720\n;plup.\ntime, nodesignation of,532.\nposition, length ofsyllable by,79.\npossessive adjectives, 1206,123035;\npronominal, 516.\npossessive compounds, 1247, 1293\n1308; poss. dependents, 1296;\nposs. descriptives, 1297 ff.;with\nordinary adjective aspriormember,\n1298; withparticiple, 1299;with\nnumeral, 1300\n;with appositive\nnoun,13013; withadverb, 1304\n6;addedsuffixes, 1307; preg-\nnant use, 1308.\nprecative optative, 533\n;itsforma-\ntion, 567;scheme ofendings, 568';\nprec.inlater language, 9215;\nuse, 573.\nprepositions, 1123 30; words used\nassuch, 11235;cases construed\nwith them,112630; prep,in\ncomposition with roots seever-\nbalprefixes.\nprepositional compounds, 1247, 1310.\npresent tense, 532;itsuses, 777 8;\npres.-system, 535,599779:\nprominenceaspartofverb-system,\n600;varieties ofform and their\nclassification, 601ff.;conjugations\nandconjugation-classes, 602 10;\nfirstconjugation:I.root-class, 611\n41\n;II.reduplicating class, 642\n82;III. nasalclass,68396;\nIV.nuand w-class, 697716;\nV.na-class, 717 32;second con-\njugation:VI. a-class, 734 50;\n"
+    },
+    {
+        "page": 499,
+        "content": "GENERAL INDEX. 4S3\nVll.accenteda-class, 7518;VIII.\nt/a-class, 75967;IX.ya-class, or\npassive conjugation, 76874;uses\noftenses, 776 9\n;ofmodes, 572\n81\n; pres. steins, derivatives\nfrom, 1140c.\nprimary andsecondary personal end-\nings, 542ff.;normal schemes, 553.\nprimary derivation, 1138 1201 :re-\nlation tosecondary, 1139\n;fromwhat\nmade, 11401; union-vowels,\n1142; form ofroot, 1143; accent,\n1144; meaning, 11456;prim,\nsuffixes andthe derivatives made\nwith them, 11481201.\nprohibitive expression, 574,579 80.\npronominal roots, 490\n;their char-\nacter, ininflection and derivation,\n11378; adverbs fromthem, 1097ff.\npronouns, 490 521 :personal, 491\n4\n;demonstrative, 495 503\n;\ninterrogative, 504 7\n;relative, 508\n12;emphatic, 513; nouns used\npronominally, 514; pron. derivative\nadjectives, 515 21\n;adjectives de-\nclined pronominally, 522 6.\npronunciation see system of\nsounds.\nprotracted (pluta) quantity, 78; protr.\nfinal vowel uncombinable, 138 e.\npunctuation, signs of,indevanagarl, 16.\nquantityofconsonants and vowels,768\n;ofsyllables, 79.\nradical stems see root-stems.\nreduplicated (orcausative) aorist, 824,85673. 1046; formation ofstem,\n85763; inflection, 8647; use\ninprimary conjugation, 868;in\ncausative, 1046; modes, 86971.\nreduplicatingclass(third, /m-class) of\nverbs, 64282; reduplication and\naccent, 6426; inflection, 647\n58; irregularities, 65982.\nreduplication, occurrence of,259\n;\ngeneral rules forforming, 588 90\n;\npresent red., 643,660ff.;perfect,\n78291; aorist, 857 63;intensive,\n1002; desiderative, 1029;inderi-\nvation, 1143e;anomalous, 1087 f.\nrelationship, nounsof,inr,369ff.,\n1182 d.\nrelative clauses, peculiarities of,512.\nrelative compounds (possessive), 1293 c.\nrelative pronoun, 508 12.\nrepeated words, 1260.resolution, inVeda, ofsemivowels\nintovowels, andofvowels intotwo\nsyllables, 113.\nreversion ofpalatal mutes and sibil-\nant,and ofh,toguttural form,\n43,64,142, 145,147,214ff.\n,1176a.\nroots ofalanguage, 1001.106;\nforms ofroot, 107; rootsand root-\nforms according tothenative gram-\nmarians, 108.\nroot-aorist, 82945: inlaterlang-\nuage, 829;inolder, 830ff.;modes,\n835 9\n;participles, 840\n;passive\naor.3dsing.,8425.\nroot-class (second, ad-class) ofverbs,\n61141; inflection, 61223;ir-\nregularities, 625 41.\nroot-stems,their occurrence anduse,\n323, 383, 1137, 1147;asinfini-\ntives, 970a,971; independent\ncomposition, 1269;indescriptive,\n1286;inflection ofsuch stems in\na,I,u,349 ff.\n;inconsonants, 383ff.\nrwd/i-class ofverbs seenasal class.\ns-aorist, 878 97 :formation ofstem,\n878 9\n;endings andcombination\nwith stem. 880 1\n;question of\nloss ofsincertain forms, 834,\n881; inflection, 8823; irregu-\nlarities, 884 91\n;absence ofIin\n2dand3dsing,inolder language,\n88890; modes, 8926; parti-\nciples, 897\n;s-aor. stem inderi-\nvation, 1140c.\ns-futnre, 9319: formation ofstem,\n932, 936; use ofunion- voweli,\n9345; occurrence, 937; modes,\n938; participles, 939;itspreterit,\ntheconditional, 9401\n;uses, 948.\nsa-aorist, 916 20 :roots allowed later\ntomake it,916; occurrence in\nolder language, 919 20\n;formation\nofstem, 917; inflection, 918.\nsecond class ofverbs see root-\nclass.\nsecond general conjugationofverbs,\nitscharacteristics, 605, 733.\nsecondary adjective compounds, 1247,\n1292ff.\nsecondary conjugations, 540, 996\n1068: passive, 9989; intensive,\n10001025; desiderative, 1026\n40- causative, 1041 52; deno-\nminative, 105368.\nsecondary derivation, 11389, 1202\n45;relation toprimary, 1139ff.;\n31*\n"
+    },
+    {
+        "page": 500,
+        "content": "484 GENERAL INDEX.\nforms ofstem, 1203 4;accent,\n1205; meaning, 1206;sec.suffixes\nandthederivatives made withthem,\n120845.\nsecondary personal endings, 542ff.;\nnormal scheme, 553.\nsemivowels (y, r, ,v),pronunciation\netc.,518;nasal semiv., 71,206,\n213c:andseethedifferent\nletters.\nsentence,rules ofeuphonic combin-\nation in,103.\nseventh class ofverbs see nasal\nclass.\nsh-sounds (*andp),61,63.\nshort andlong quantity,76 9.\nsibilants (f, s,s),pronunciation etc.,004: andseethedifferent\nletters.\nsibilant-aorist, 824,874920: for-\nmation and classification, 8747;\n4.s-aorist, 87897; 5.is-aorist,\n898910;6. fr-aorist, 9115;\n7.sa-aorist, 91620;itsstem in\nderivation, 1140c.\nsimple aorist, 824,82855: 1.root-\naorist,82941; passive aor.3d\nsing.,8425;2.a-aorist, 846\n55.\nsis-aorist, 911 5;formation ofstem,\n'911; inflection, 911\n;forms inolder\nlanguage, 91213; modes, 914(;\nmiddle forms, 915.\nsixth class ofverbs seea-class.\nsonant andsurd sounds, 34;mutes,\n345\n;aspirates, 378\n;question\nastocharacter offt,65\n;offinal\nmute, 141end; euphonic assimi-\nlation ofthetwoclasses, 117,156\n78.\nspecial andgeneral tenses, 599.\nspirants, 59ff.: sibilants, 5964;\naspiration, 65; other breathings,679.\nstems, inflexible. 99100, 105;\ntheir derivation-see derivation,\nstrengthening andweakening process-\nes,23460.\nstrong andweak,orstrong, middle,\nand weakest, forms ofstems in\ndeclension, 311;ofrootsandstems\ningeneral, 107 -3.\nsu-class ofverbs see nu-class.\nsubjunctive mode, 533;formation and\nendings, 557 62;itsfirstpersons\nused later asimperative, 533, 574,\n578; subj. useofaugmentless pre-teritforms, 563,587\n;uses ofsubj.\nmode, 574-82.\nsuffixes forming adverbs, 10971109\n;\ndo.declinable stems see deri-\nvation.\nsurdandsonant sounds seesonant.\nsyllables, quantity of,79.\nsystem ofsounds, 1975 :vowels\nanddiphthongs, 1930; conson-\nants, 31ff.;mutes, 3250;semi-\nvowels, 518;sibilants, 5964;\naspiration, 656\n;visarga andother\nbreathings, 689;anusvara, 70\n3;unwritten sounds defined by\nHindu grammarians, 230; scheme\nofspoken alphabet,with, notice of\ncomparative frequencyofthesounds,\n75;quantity, 769;accent and\nitsdesignation, 80 97.\ntan-class ofverbs see w-class.\ntafpwrwsa-compounds see deter-\nminatives.\ntense inverbal inflection, 532;tense-\nsystems, 535; present-system, 599\n779; perfect-system, 780823;\naorist-systems, 824 930;future-\nsystems, 931950.\ntenth class ofverbs see causative\nconjugation.\nthird class ofverbs see redupli-\ncating class.\ntime, particles of,1103, 1122 f.\ntransliteration, general methodof,5;\nofsign ofelision, 16,135\n;ofcom-\nbined finaland initial vowels, 126;\nofanusvara, 73;ofaccent, 83,89.\ntad-class ofverbs seed-class.\nM-class (eighth, ton-class) ofverbs,\n697716; formation ofstem, 697;\ninflection, 698 707;roots ofthe\nclass, 713\n;irregularrootkrorfcar,\n7145;other irregularities, 716.\nuncombinable (pragrhya)final vowels,\n138.\nuninflected words see indeclin-\nables.\nunion- vowels, 254;iinpresentin-\nflection, 6301,640;inperfect,\n796_8, 803; inaorist,8767;\nins-future, 934 5\n;inperiphrastic\nfuture, 943\n;indesiderative, 1031;\ninpassive participle,956\n;inin-\nfinitive andgerund, 968, 991;in\nderivation, 1142;Iinpresent\ninflection, 6324;in2dand3d\n"
+    },
+    {
+        "page": 501,
+        "content": "GENERAL INDEX. 485\nsing., 555b;inintensive, 1004ff.;\nifori,900;aiforI,555b.\nupadhmanlya-spiT&nt, 69,170d.\nvariation ofstem-form indeclension,\n3112;inr-stems, 370; incon-\nsonantal stems, 379,3858, 421,\n4434, 458, 463;inconju-\ngation, 556\n;inpresent-stem, 604\n;\ninperfect, 792 4\n;inaorist, 831 ff.,\n879,899\n;inintensive, 1004\n;inpri-\nmary derivation, 1143\n;insecondary,\n1203 4;incomposition, 1249.\nverb see conjugation.\nverb-forms, accentuation of,inthe\nsentence, 591 8.\nverbal prefixes, 10767; kindred\nwords, 10789,1120; composition\nwith roots,1076 87,137a,b ;accent,\n1082 5;their more independent\nuse,1084, 1118; prepositional uses,\n1125; forms ofcomparison,1119\n;de-\nclinable stems from rootscompounded\nwiththem. 1141, 1282; useindes-\ncriptive composition, 1289;inpos-\nsessive, 1305;inprepositional, 1310.\nvisarga (orvisarjamya], 679\n;quan-\ntitative value, 79;occurrence, 144\n5,1702.\nvocative case, formof,307end; Vedic,\ninas,425g,454b, 462a;accent\n(along with qualifying word), 314;\nverb accented after, 594.voice inverbalinflection, 528 31.\nvowels, how written indevanagarl,\n10; sign ofabsenceof,11;\ntheir pronunciation etc., 1929:\na,i,w-vowels, 1922\n;r,Z-vowels,236; diphthongs, 279; quan-\ntity,778; accent, 80ff.;nasal\nvowels, 71\n;rules ofvowel-com-\nbination, 125 38;resulting accent,\n128, 130,135\n;exceptional cases,1368.\n7;rdd/'-strengthening, character and\noccurrence of,235 43,andpassim ;\ninprimary derivation, 1143a;in\nsecondary, 1204.\nuj-sound, belonging tov,57.\nweak, orweakest, form ofstem in\ndeclension, 311.\nweakening and strengthening pro-\ncesses, 23460.\nya-class (fourth, div-class) ofverbs,\n759_67 :formation ofstem, 759\n;\ninflection, 760;roots ofthe class\nand their classification, 761 2;\nirregularities, 763 7.\nyd-cl&ss ofverbs, orpassive present-\nsystem, 76874;formation of\nstem, 768 70;inflection,771\n;\nirregularities, 7724;t/d-formation\nfrom intensive stem, 1016 7.\n"
+    },
+    {
+        "page": 502,
+        "content": "ERRATA.\nTheremark attheend of721 belongsinstead under 688.\nAt612a,lastline ofparadigm, restore the lost iofitds;and, at1156,\nI.5.thelostaatend ofline.\nRead Iat65, last linebutone; vdvd at94e;emsandesaat176,\nII.2,3,svit at188,1.5;gordtsu at221,1.8;briivanah at268,1.5;\ntiryateat761b,1.15;dyutat785,1.5;dmoci at844,1.8;aranism at\n908,1.8;tanyuat1178e,1.1;ddhfuka at1180 a,1.3;rj(for ;)at\n1200c,1.1.\nReferences totheparagraphs 36176are(owingtoanunfortunately\nnecessary renumbering)inseveral instances wrong byadifference ofone to\nfive. Also other references asfollows; read454 at425g,1.2;547 at\n561,15from end;672 at654,1.3;1042d at1030,1.13;and1271\nat1150c,1.8.\n"
+    },
+    {
+        "page": 503,
+        "content": ""
+    },
+    {
+        "page": 504,
+        "content": ""
+    },
+    {
+        "page": 505,
+        "content": ""
+    },
+    {
+        "page": 506,
+        "content": ""
+    }
+]
\ No newline at end of file