________________
CHAPTER VII]
BOWER MANUSCRIPT
•
Ixxi
II 314 42 and haritakih. II 226 38. In the Vedas is is nowhere met with, see W. Skr. Gr. $$ 359, 363, pp. 316, 318, and M. Ved. Gr. $ 378, p. 273.--Also, accusative âpas (for a pas), II 804 63; as often in Vedic, see W. Skr. Gr. 393, p. 133.
(3) Instrumental singular neuter, once dadhina (for dadhna), II 428 47, lo suit the metre (flóka); otherwise always normal dadhna, II 149, 150 34 785, 801 62 853 65 1053 73. Also, feminine, rasnaya and balâyâ (for râsnaya and balaya), II 177 35, shown to be instrumentals by the accompanying undoubted instrumentals mûléna and madhukena ; otherwise they might be taken to be genitive substitutes. They may, but need not, be due to the metre (lóka); for we have an undoubted example in prose in vidyarajaya, VI 2 222 and (with the normal qualifying anayá), VII 6 237 ; but normal vidyârajaya, VII 3 237. In all other cases, the instrumental is normal; e.g., sarkkaraya I 81 6 107 8 II 504 51 1037 75, vachaya, II 80 30, etc. (sce Index).
(4) Genitive singular feminine, anganaya (for aiganaya), I 84 7; and súkshmailaya (for søkshmailâya), II 115 32; or also, gudikâya (for gudikaya), II 1035 72, and madhurasaya (for madhurasaya), II 67 29. In all other cases, normal, e.g., súkshmelayah, II 6! 29; gurlikayah, II 1022 72; chidayah, II 856 65111-Again, plural masculine, gridhrasinam (for gridhrasinâm), II 377 45; pittinâm (for pittinám), II 164 34 418, 423 49; pramêhinam (for pramehinâm), II 230, 243 38 971 71, rôginam (for rôginam), II 254 39. In all other cases, normal, e. K., a pasmarinam, II 378 45, udarinâm, II 971 71, käsinâm, II 164 34, kshirinam, I 89 7, II 291 41, dêhinâm, III 33 183 prâņinâm, I 46 5, mêlinam, II 606 56, Saririnâm, II 244 38, sôshinâm, II 940 69, etc., the proportion of abnormal to normal cases being 8: 14. Similarly, once, parvânâm (for parvaņâır), II 335 43.-Also plur. masc., only once, varadam ( for varadânám.) II 774 61; elsewhere normal, as naranám, I 92 7 II 37 28, etc. All the preceding abnormalities occur in verse composition, and seens to be due to the exigencies of the metre; but there is one example in prose, sarvavadinúm (for 'râdinâm), IV 3 192.
(5) Locative singular neuter : once the syncopated form nâmni, II 918 69, and optionally ahni, I 20 2 II 908 68, by the side of aliani, 1 63 5 II 723 58 784 61; but elsewhere the full form, as múrdhani, I 11 2 II 79 35, karmani, II 962 71, etc.
IV.-VERBAL INFLECTION, In the main the abnormalities in verbal inflection refer to changes with respect to « class" and " voice." Most of them have the support of Vedic and Epic usage.
(1) Change of " class": Thus I. class for II., rôdatê (for rôditi), II 1041 73. but normal II. class, rudyât, I 99 8; both classes also in Vedic and Epic 119,- Again, VI. class for IInd, parasm., lihêt (for lihyat), II 475 50 590, 594, 596 55 608 56 1081, 1088 75, or atm.. lihêta (for lihita), III 21 182; but almost equally frequent (7:8) is the normal lihyât, I 128 9 II 21 27 433. 439, 446 48 450 49 779 61, and the VI, class is also epic. Similarly VI. class for I Ind, dvishasi (for dvështi), IV 52 195; also epic.-Again, VI. class for VIIth, pishet (for pimshyat), II 850 65 896 67; in this case, as well as in the compound prapish, there is the abnormal lengthening of the root vowel, which, however, is restricted to the tenses; for the participles are pishtva, II 41 28, etc., or pishya, II 887 67 and prapishya, II 82 30. pishta, I 35 4 II 430 48, etc. (see Index). The same lengthening occurs when the root is inflected normally in the Xth class or causal, pishayêt, II 404 46 550 53 871 66 889 67. and prapishayet, II 97 31 211 37 577 54. The lengthened root vowel occurs once also in the
111 It would seem that the abnormal forms occur only when the normal visarga drops off by reason of sandhi.
112 la the Rigveda the II. class does not occur; See M. Ved. Gr., $ 450a. footnote 8, p. 395.
