________________
196
EPIGRAPHIA INDICA.
vol. III, p. 30, and plate xiii, No. 1). On the evidence of his coins, which resemble those of Azes, Sir A. Cunningham placed Sodasa about 80-70 B.C., and conjectured that he was a son of Rajubula, another Satrap of Mathura. The latter conjecture is confirmed, as he has stated of late, by a passage on Dr. Bhagvânlal's lion-pillar, where Sodása is called the son of Rajula.' Though the precise date assigned to Sodasa may be doubted, still he must have ruled at Mathurâ in the first century B.C., before the time of Kanishka and his successors. I do not dare to offer for the present any conjecture regarding the era which Soḍâsa uses. Sir A. Cunningham (loc. cit.) is inclined to identify it with that used by the Mahârâja Moga and other foreign rulers of Northern India. The inscription No. ii is also in pure Prakrit of the Pali type. The next inscription, No. iii, mentions also a Mahakshatrapa, whose name seems to have begun with Ma. It is to this circumstance that it owes its place. Its appearance indicates that it is much later than No. ii, and as it is incised on a sculpture cut out of the back of that on which No. x is found, it must be later than the latter. Still its date will fall before the time of Kanishka, as the Mahâkshatrapas of Mathurâ must have passed away before the Kushanas reigned there. The mangala of the inscription, the only portion fully preserved, is in pure Sanskrit.
The next seven inscriptions, Nos. iv-x, which have been grouped under the name "archaic," all belong in my opinion to the period before Kanishka. But I am not able to say anything regarding their relative position towards each other or towards Nos. ii and iii. The chief differences which separate them from the documents of the Kushanaperiod are the use of the tripartite subscribed ya, the want of the loop on the left side of the ordinary ya, the da, slightly open to the left (which occurs in Nos. iv and ix), and the well developed vertical stroke and symmetrical shape of the ta. The va has a very curious shape in the word Sivayasa, No. v, 1. 2a, as it consists of two triangles with the apexes joined. The language of most of them is pure Prakrit of the Pali type, but No. iv shows one Sanskrit form and No. vii is apparently in pure Sanskrit. With respect to No. ix I must add that I am inclined to identify the Gotiputra Idrapala or Idrapálita with Gotiputra, "the black cobra for the Sakas and the Pothayas" mentioned in No. xxxiii of the collection, published ante, vol. I, p. 396. If that is correct, the document must go back to the times before the consolidation of the foreign rule at Mathura.
The next group, Nos. xi-xxiv, consists of the dated inscriptions which in my opinion all belong to the time of Kanishka, Huvishka and Vasudeva. Not one shows the name of a king. Nevertheless, I believe that nobody, who carefully compares them with the dated documents, mentioning the three kings, will come to a different conclusion. The dates range within the well-known limits from Samvat 4 to Samvat 98. The characters and the language are of the same mixed types described in the introduction to my first paper, ante, vol. I, p. 371ff. I have nothing to add to the remarks made there on the alphabet and the spelling. As regards the language, I will only call attention to a few points. The few verbal forms, bhavatu, No. xiii, astu, No. xviii, and nirvartayati, No. xx, are pure Sanskrit. No. xxiii, where we have fishyasya ganisya
Academy of April 25th, 1891, p. 397. On an impression of the inscriptions on the lion-pillar, which I owe to the kindness of Dr. Burgess, I, too, read Rajulasa putra Sudase chhatrava. But I do not feel quite certain that Rajula and Rajubula are the same person, nor that the words given above, which stand in two consecutive lines with other signs in between, are really connected.