________________
171
languages distinguish between V and b; most have either v or b. It is a well-known fact that this phenomenon has greatly affected the transmission of Sanskrit texts, in which originally v and b were distinguished (see AIG I,S 161). It is in itself not surprising that it is precisely in the South that apparently a tradition was preserved with regard to writing. v or b in particular words. This almost certainly has to do with the fact that most South-Indian languages do distinguish between v and b.
The decision to write in the text v throughout introduces yet another problem in connection with the variants. MS K, edited in the Abh., has only v. For B, P, S and T, edited in the first place in the Retr., the deviations from K in respect of this feature have been carefully noted; at least, Weber does not state explicitly that he makes an exception for the variation between b and v. The text in Ed. has only v. On page 5 Weber remarks that he does not specify the variants, unless the word in question should happen to be mentioned for other reasons. Bh, edited in ISt., seems only to have v; a few instances of b in this MS have been noted.
My procedure is the following. Where I have adopted a form with v I first mention those MSS which positively have a variant with b, and subsequently add within brackets those MSS of which the reading cannot be definitely ascertained.
As I have shown above (p. 40) the present manuscript tradition goes back to one single MS the text of which was not entirely flawless. Some errors in the MSS can be traced back to this particular Ms. Where possible an emendation has been suggested. In some cases the way to emend the text is pointed out in the MSS themselves. See 292 in which the archetype seems to have read taņuãi (cf. Bh tanuāi and Ti and Tp tanuai(d)a) which in all the other MSS was corrected into tanuāai. In other cases the emendation could be based on peculiarities of the script of the MS of the archetype, which was Jaina Nāgart. In yet a few other cases the emendation could be based on an informed guess, for instance, gammihii for gammihisi of the MSS in 609 (see the note to *10). In a few cases, though, no convincing emendation seems available.
The present manuscript tradition seems to have started in a milieu of Jains. This explains the occasional occurrence in the MS of the arche
