________________
268
EPIGRAPHIA INDICA.
(Vol. XXV.
A.D. 554.! But, as I have said elsewhere, this identification is highly doubtful, for (1) it is by no means certain that the Maukhari Suryavarman ever came to the throne, as his name is not mentioned in the Asirgadh and Nālandă seals of the Maukharis, and as no coins belonging to him have as yet been found, and (2) whereas the Süryavarman of the Sirpur inscription is called king of Magadha, the Maukharis of the line of Harivarman were never a characteristically Magadhan dynasty, their headquarters being at Kanauj.
It is now practically certain that the Sarabhapura line, consisting of Prasannamätra, his sons Mahājayarāja and Mānamatra, and Mänamätra's sons Mahāsudēvarāja and Mahāpravararāja, preceded and not followed the Pandava kings in Southern Kösala. Professor Mirashi holds that Mahapravararāja (who issued his Thakurdiya plates from Sripura) ruled in the first half of the sixth century and was ousted by Tivara, whom, as has been said above, he proposes to place in A.D. 530. It is, however, not possible to subscribe to this view : the Kharöd inscription of Indrabala and Iśānadēva shows that even before Tivara the Pandavas were masters of much the same area as had been held by the Sarabhapura kings. The fact that Tīvara is the first king of his line who is known to have made Sripura his capital does not warrant the belief that the capital of his father and grandfather was situated elsewhere, as no copper-plate (which alone mentions the capital as the place of issue of the charter) belonging to them has as yet been found.
Having seen that there is no sure ground for placing Tivara in the second quarter of the sixth century, we may turn to an examination of the palæographical chart attached hereto. The Arang plates of Bhimasēna of unknown lineage are dated in the Gupta year 282=A.D. 601 ;* it is the only dated inscription of the locality and the period with which we are now concerned, and as such affords a convenient standard of comparison. It will be seen that the characters show typical Gupta forms throughout. Comparing this record with the Bhandak inscription of Nannadēva, which according to Professor Mirashi's chronology must belong to c. A.D. 500, we find that every letter in the latter presents a more developed appearance ; in bh and there is now & hollow wedge attached to the left corner; looped tripartite y has given place to the bipartite form : the right vertical of 6 projects a bit above the curved top-an important development which led to the ultimate separation of the left and right limbs of the letter, leaving the left limb to develop independently into the Någari and proto-Bengali forms (which tendency of separation is already noticeable in the Sirpur inscription of Mahāśivagupta, where the horizontal cross-bar of the letter has turned into a curved inward prolongation of the left member, the right member being thus a separate entity); in h the left limb is no longer a vertical straight line but a curve, and the right downward stroke shows a tendency to hang below the base line. The persistent occurrence of late forms in the Bhandak inscription marks it out as definitely much later than the Arang plates: a fortiori the possibility of its being placed a century prior to the latter cannot be considered. It is also evident from the chart that the letters of the Kõndēdda plates of the Sailodbhava Dharmaraja have practically the same forms as those of the Bhandak inscription.
Härähá inscription, above, Vol. XIV, p. 110. This theory was first started by H. C. Raychaudhuri, Political History of Ancient India, 4th ed., p. 512 n.
2Two Maukhari Seals from NĀlandā", above, Vol. XXIV, p. 288. • Cf. above, Vol. XXII, p. 16. .P.R., 4.8.W.C., 1903-04, p. 54.
The letters in the chart have been traced out of the published impressions of the respective inscriptions. They do not claim the preciseness of mechanical reproductions but may be regarded as accurate for all practical purposes.
. Above, Vol. IX, p. 342.
