________________
320
EPIGRAPHIA INDICA.
[Vol. IX.
11. 11 and 22, and viharo in 11, 23 and 26, on the other hand, are vouched for by the metre Kumara instead of kumára is common in Maharashtri Prakrit, and a form Kumaravala for the usual Kumarapala occurs in H&machandra's Désinamamala, I. 104, 88. And vihara instead of vihdra is justified by Papini VI, 3. 122.
The language is Sanskrit, and, with the exception of the invocation to Vasudhára in l. 1, the whole of the inscription is in verse. There are altogether twenty-six verses. Of these thirteen (Nos. 3, 5, 6, 7, 11, 12, 13, 17, 18, 19, 21, 23 and 24) are in the Sárd ûlavikridita, five (Nos. 1, 10, 14, 15 and 20) in the Malini, foar (Nos. 4, 16, 22 and 25) in the Vasantatilaka; three (Nos. 8, 9 and 26) in the Anushțubh, and one (No. 2) in the Sragdhará metre.
The object of the inscription is to record the construction of a vihdra by Kumaradevi, one of the queens of Govindachandra of Kanauj. The wording of verses 21-23, in which her gift is mentioned, is not quite clear. We are first told, in verse 21, that a vihara was constructed, which apparently contained an image of the goddess Vasudhara. The following verses are not quite clear. Rai Bahadur V. Venkayya has suggested the following explanation, which I think is a very plausible one. Jambaki drew up a copper-plate, in which she represented to Kumaradêvî that the Dharmachakra-Jina originally set up by Dharmásoka required to be repaired or set up again. This copper-plate must have contained information about the original setting up of the Dharmachakra-Jina and further details about its maintenance and repairs. Kumaradêvî, who was apparently a stranger to the country round Benares, accepted her representation and raised her to the rank of the foremost of pattalikäs. Moreover, she restored the Jina or set up a new one and placed it in the vihára built for Vasudhårå, or in another one constructed for the purpose, and the wish is expressed that, after having been placed there, he may remain there for ever. It seems necessary to infer that the SridharmachakraJina, which is said to have existed in Dharmaśoka's time, was an image of the Buddha, and that the vihara built by the orders of Queen Kumaradevi for him, was a shrine, a gandhakuti. It is difficult to explain the wish that he, i.e. the image, may reside there for ever, under any other supposition.
The inscription can be divided into four parts. After an invocation of Vasudhårå (v. 1) and the moon (v. 2) vv. 3-6 give some information about some rulers or generals of Pithi or Pithikâ. We learn that, in the lunar race, there arose a chief called Vallabharaja, the lord of broad Pithika (v. 3). The following verse introduces the lord of Pithi Devarakshita, without xaying anything about his relationship to Vallabharaja. He is described as the full moon of the lotus of the Chhikkora-wara, and we are told that he even surpassed Gajapati in splendour, Devarakshita is again referred to as the lord of Pithi in the second part of the inscription, and it therefore svoms neceBeary to interpret ve. 5-6, which apparently refer to a son of his (tasmad-usa, etc.) as an explanation of his relationship to Vallabharaja, who would then be his father.
The second part of the inscription, vv. 7-13, contains the information that Devarakshita was defeated by Mahaņa, the maternel uncle of the Gauda king, who thus firmly established the throne of R&mapala, and subsequently bestowed bis daughter Sankaradêvi on the Pithi lord. Their daughter was Kumaradêvi, in whose praise the present inscription was written.
The third part, vv. 14-20, then contains the genealogy of the Gahadavåla family, to which Kumaradevi's husband Govindachandra belonged. It agrees with the list given in most ioscriptions of this latter king, but does not carry us farther back than to his grandfather. We are first introduced to Chandra, the Chandradeva of Govindachandra's inscriptions. His son
See Pischel, Grammatik der Prakrit-Sprachen $ 81.
1 Pattalikd is the feminine form of pattalaka, which occurs in the Buguda plates of Madhavavarman, above, Vol. III, p. 46, 1, 83, in connexion with paildsika.