A. P. R. Hoernle— Inscribed Seal of Kumdra Gupta II. [No. 2, 



had not possessed. This inscription is not dated, but its date cannot 

 have been very different from that of another Mandasor stone inscription 

 of Yasodharman,* dated in A. D. 533-534, because they were both en- 

 graved by the same person Govinda. In this inscription, it is stated 

 that Yasodharman, who was originally only a tribal chieftain (jinendra 

 or narddldpati), succeeded in conquering the countries around him and 

 thus founding an empire, after which he took the name of Vishnuvar- 

 dhana and the imperial titles of rdjddhirdja and paramesvara. Of these 

 two inscriptions, the latter would seem to be— if anything— the later 

 in date. In any case Mihirakula's overthrow would fall some time pre- 

 viously to A. D. 533 ; and it may be set down in A. D. 530, or perhaps 

 even a little earlier. f It follows, that Bahiditya, in whose reign Mihira- 

 kula's overthrow took place, must have reigned down to about A. D. 530. 

 The circumstance, that the overthrow is ascribed to both Bahiditya 

 and Yasodharman, would seem to be best explained thus, that Bahiditya 

 was, at least nominally, the paramount ruler or Emperor (Mdhdrdja- 

 d/urdja), and that Yasodharman, at that time a mere ' tribal chieftain,' 

 was one of his feudatories or lieutenants, who actually accomplished 

 the defeat of Mihirakula, but thereupon took advantage of his groat 

 success to found an empire for himself. In fact, it was probably 

 Yasodharman (rather than the Hiinas) that supplanted Kumaragupta II., 

 some time after 530 A. D., thus finally breaking up the Early Gupta 

 empire, and building up his own enrpire on its ruins. 



On the other hand Skandagupta's earliest recorded date is A D 

 455. F rom this date down to A. P>. 530 there are 75 years * and for 

 this interval we have three names Skandagupta, Puragupta and Nara- 

 siinhagupta. The interval can be more easily filled up by two genera- 

 tions including three reigns, than by two generations including only two 

 reigns ; i. e., by assuming that Skandagupta and Puragupta were bro- 

 thers, succeeding one another aud being themselves succeeded by Nara- 

 siinhagupta. Skaudagupta is known to have been still reigning in A. D. 

 466 or 468. He may have beeu succeeded by his (younger) brother 

 Puragupta c. A. D. 470, aud the latter, c. A. D. 485, by his son Narasimha- 

 gupta. This would give to Narasimhagupta the long reign of about 45 



* No. 35, ibid., Vol. Ill, p. 150. 



t In these calculations I follow, in the main, Mr. Fleet's remarks, iu Corp. 

 Insvr. Ind,, Vol. Ill, pp. 10-12 (Indrod.), 146, 152. Bnt I do not agree wit'h his view 

 about Vishnuvardhana, ibid., p. 151. (See Proceedings for August, 1889.) It seems 

 to mo hotter to accept the words, in 1. 5 of the inscription, in their obvious sense 

 that Yasodharman is identical with Vishnuvardhana, and to suppose that Yasodhar- 

 man assumed the title Vishnuvardhana after (and perhaps in commemoration) of his 

 great, victory. It is most, improbable that the imperial titles of rdjddhirdja and 

 paramesvara should be ascribed to a niero feudatory. 



