added as a gloss, or be incorporated even in the text, by those who transmitted it either in writing or in instructing their pupils. But an argument of more weight is the fact that in the Siddhânta we find no traces of Greek astronomy. In fact the Gaina astronomy is a system of incredible absurdity, which would have been impossible, if its author had had the least knowledge of the Greek science. As the latter appears to have been introduced in India about the third or fourth century A.D., it follows that the sacred books of the Gainas were composed before that time.
Another argument which offers itself for fixing the period of the composition of the sacred books, is the language in which they are written. But, unfortunately, it is not at all clear whether the sacred books have been handed down in that language in which they were composed, or in that in which they were pronounced, and transcribed in later generations, according to the then current idiom, till Devarddhi's edition put an end to the modernising of the language of the sacred books. I am inclined to believe the latter view to be correct, and look upon the absence of a self-consistent orthography of the Gaina Prâkrit as the effect of the gradual change of the vernacular language in which the sacred books were recited. In all MSS. of Gaina texts, the same word is not always spelt in the same way. The differences of spelling refer chiefly to the retention, omission, or attenuation of single consonants between vowels, and the retention of the vowels e, o, before two consonants, or their change in i, u. It is hardly possible that the different spellings of a word should all correctly represent the pronunciation of that word at any given time, eg. bhûta, bhûya; udaga, udaya, uaya; lobha, loha, [1] &c.; but probably we must regard these methods of spelling as historical spellings, that is to say, that all different spellings presented in the MSS. which formed the materials for Devarddhi's edition of the Siddhânta, were looked upon as authentical and were preserved in all later copies of the sacred texts. If this assumption is correct, we
- ↑ I do not contend that no double forms of any word were current at any time,for there must have been a good many double forms, but I doubt that r early every word should have existed in two or three forms.
