I. ALLGEMEINES UND SPRACHE. 4. VEDIC GRAMMAR. 2. before the t of the 3. sing. of a past tense: thus vy-àvāt (AV.) ‘has shone forth', from vi-vas. This is, however, probably not a phonetic change, but is rather due to the influence of the 3. sing. of other preterites with -t;
- á-vas-t having thus, instead of *á-vas, become á-vat¹.
3. before case-terminations beginning with bh, and when final (in nom. acc. sing. neut.), in the perf. part. and in four other words: thus jāgr-vád-bhis, inst. pl., 'having awakened', tatan-vát, acc. n., 'having stretched'; uşád-bhis, from usás- 'dawn'; mad-bhis, mad-bhyás (AV.), from más- 'month'; svá-tavad- bhyas (VS.), from svá-tavas- ‘self-strong'. The change of s to t began before the bh endings (like that of s to t or d)² and was extended to the nom. acc. sing. neut. in the RV., but not till later before the -su of the loc. pl.³ 2 36 a. Allied to the change of finals of roots and stems to t, is the apparent change of the medial dental sibilant to din madgú- (VS.) 'diver', from majj- 'dive' (jj = IE. zg) 4. p. The substitution of dentals for other mutes is extremely rare. In consequence of dissimilation, a dental replaces a cerebral in anaḍutsu and anaḍúdbhyas (AV.), from anadváh- 'bull'; in dṛdhrá- firm', beside drdhá-5; in paşthavát (TS.) 'four year old bull', beside pasthavát (VS.). A dental seems to take the place of a labial in ad-bhis, ad-bhyás, beside ap- 'water': but this is probably due to the analogy of *nadbhis, nadbhyás, beside nápat- "grandson' 6. 45. The labials. These sounds as a rule represent IE. labials; e. g. pitŕ-, Gk. natýp; bhára ‘bear', Gk. pépe. But owing to the great rarity of IE. b, there are very few Vedic examples of inherited b; e. g. rambate ‘hangs down', Lat. lābī ‘glide'7. a. The number of words containing 6 has been greatly increased by new formations. I. Thus & replaces p or bh before other voiced mutes: e. g. pi-bd-aná- 'firm', beside pad-d- 'place'; rab-dhá-, beside rabhante 'they take'. 2. It is the regular substitute for bh in reduplicative syllables or when initial aspiration is lost owing to a following aspirate; e. g. ba-bhúva from bhu- 'be', bahú- 'arm', bandh- ‘bind¹8. — 3. In a few examples it takes the place of or interchanges with v9; thus pádbīša- (RV.), beside pádvīša- (VS.); band- beside vand- 'arrow'; -balsa- (AV.) beside -valsa- 'twig'; bāṇá- (AV.) 'music' beside vāṇá-; -blīna- (AV.) ‘crushed', beside vlīna- (B.)¹°⁰.- 4. It further occurs in some new onomatopoetic words; budbudá- 'bubble'; bál (AV.) interj. 'dash!'; bata interj. 'alas! and batá- 'weakling'.-5. In one instance b seems to stand for m before r, in brū- 'speak', for *mrū-¹¹, originally appearing most likely after a pause or after a final consonant ¹2 II b. In many words the origin of 7 is obscure. Most of these probably come from a foreign source: I. owing to their meaning: arbudá- and árbuda-, balbūthá-, sámbara-, sýbinda-, names of foes of Indra and of the Aryans; brbú-, a proper name; bajá- (AV.), bálbaja- (AV.), bilvá- (AV.), names of plants; bákura- and bakurá-, a musical instrument.- 2. owing to their phonetic form: kilbişa- 'sin'; bisa- 'root-fibre'; busá- 'vapour'; bát and baḍa, interjections; — 1 Cp. WACKERNAGEL I, 154. 2 See 42 d (p. 34). different scholars, see WACKERNAGEL I, 158 b, note. 3 Cp. WACKERNAGEL I, 155 a, note. 8 See above 32 d, I, 2. 4 Op. cit. 1, 155 b, note. 5 See 42 a a (p. 33). There is some confusion between forms of brh- be great', and urh- 'tear'. 6 The name in-d-ra and nánan-dr- 'hus- 10 On some doubtful or wrong explanations band's sister', are explained by some scho- of b for v (ni-byh- 'crush', bála-, bálbaja-, bát, lars as containing a Prakritic d. See WACKER-| śabála-, śámba-), seeWACKERNAGELI, 161, note. NAGEL I, 157, note. II Cp. Gk. ßporós for *uporós. 12 See WACKERNAGEL I, 159. 7 On some words (sabar-, batá-, bála-, bal- būthá) in which is regarded as IE. by
