________________
No. 8.]
SANJAN PLATES OF BUDDHAVARASA.
147
pusvāra occurs in runshtra., 1. 18 ; -sanndhit, 11. 25, 26; -an kan., 1. 30, and a superfluous visarga in -ūdhipatih., 1.8; -ograh, 1.8; sutahs, 1.9 ; -ānviyah-, 1. 29; -Hiranyakësih., 1. 30; bahubhihr-, 1. 34 ; bhūmis-, 1. 35, while the visarga has been wrongly omitted in samāsādita, 1.5; -pratapa, 1. 8; bähudanda, 1. 10 ; -raja, 1. 17.
The rules of Sardhi are constantly violated. Thus we find visarga instead of s in -parikarah, I. 27; visargs instead of sin mahārājnh, ll. 9, 14; and visarga before sonants in -yāgāh yats, 1.6; -bhogikaih nekakulapradhānaih Mätsi-, l. 23; nadhayirah Rēva., 1. 31. On the other hand, we find o before surds in -vijayo, 1. 15; Mahindarāmo, 1.21 ; parivano, I. 26; -vāpako, 1. 27. When two vowels follow each other, they are sometimes wrongly combined; thus, -bhishanātur- instead of bhishana atur., 1. 12. In other cases an m is inserted as a kind of Sandhi-consonant. Compare dakshina-m=iva bahudanda, 1. 10; -Oshadhi=m=ida tushţikaro m=Arjuna-m=iva, 1. 12; -chandramá=m=iva, 1. 16; -Hiranyakësi()=m=anēka-, 1. 30; sahasrēna= muafva-, l. 37. Other instances of wrong Sandhi are =vinashtam ava, 1. 6; -nyāyēna achatabhatapravēsyah a., 1. 28.
The language is Sanskrit. There are however several mistakes. The nominatives in o of a-bases where no sonant follows; forms such 49 simasy=äpi, 1. 23, perhaps -chaturvēdasya, 1. 31. and the use of the genitive instead of the dative in l. 31, and so forth, make it. probable that the person who drafted the plates was more familiar with Prakrit than with Sanskrit. A form such as Mahindāramo, l. 21, is pure Prakrit, and the form mätarā instead of mātri in saptamātarabhisiktānā, 1. 3, is in accordance with the rule in Hēmachandra's Prakrit grammar III, 46, that the Sanskrit word mātri becomes maarā if it denotes & deity, mātarā being a Sanskritised form of this maarā. Also the use of two different l's points in the same direction. That the grant was drafted by & person who was not very well versed in Sanskrit is also shown by the use of parenthetic sentences in l. 4 (bhagava-pratyaksha-Harina tushțēna varo daktah samāsādita[*] pratyayo varahalaichhanam cha), l. 3 (kesitachatva mēdha-rojasuya-paundarikayagah yat-kifiohiktu(t-ku) laduritan tadvinashtam ... dhanushmata) and so forth. Note also - Ramadēva-tat-pratimänamziva, 1.7; -Anudhyāta, 11. 10, 17; -yaso, 11. 11, 16, and so on.
Sach oocurrences cannot fail to throw some doubt on the genuineness of the plates. And other considerations lead to the same result.
I have already mentioned that the seal shows the figure of a lion. Now the Western Chalnkyas used the boar crest," and this is in accordance with 1.5 of the Sanjan grant. The invocation of the Variha-incarnation of Vishnu at the head of many Chalukya grants should, no doubt, be seen in connection with the use of this crest. It speaks strongly against the genuineness of the present grant that this stanza has been so much changed.
On the other hand, it cannot be urged against the genuineness of the plates that the grantor Buddhavarasa is not known from other sources; for a Mahasamanta Buddhavarasa of the Salakika family is known from the Torkhodo grant of Saka 735.8 This Buddhavarasa was in possession of an estate known as the Sıharakhi Twelve, which Dr. Fieet has identified with the present Serkbi, near Baroda, at 73° 8' E. and 22° 21' N. This place is not too far removed from Sanjân to make it unlikely that he was a descendant of the Buddhavarasa of the Sanjan plates, and the name of his family Salukika can hardly be anything else than Ohalukya. The later Buddhavarasa was a feudatory of the Rashtrakota king Govinda III; and nothing prevents as from supposing that members of the old imperial Chalukya family, after its overthrow by the Rashtrakata's, served under the new rulers. I am thus inclined to think that the older
1 ct. Kielhorn, Ep. Ind., Vol. IX, p. 268 and n. 8. * ct. Fleet, Gasettoor of the Bombay Presidency, VOL I, Part II, p. 299 n. 4.
Ep. Ind., Vol. III, pp. 63 f.