86 A. F. R. Hoernle— On the date of the Bower Manuscript. [No. 2, 
tions, which are not written in the North-Eastern or proper Nepalese 
alphabet, but in the North-Western characters; thus the Gaidbara 
inscription of 688 A. D. (see Mr. Bendall’s Journey in Nepal , p. 77), the 
Jaisi inscription of 751 A. D. (ibid., p. 79), the inscription of S'iva 
Deva, of 748 A. D., another of 750 A. D., and the inscription of Jaya- 
deva, of 758 A. D. (see Indian Antiquary , vol. IX, pp. 176-78). In all 
these inscriptions the modern cursive form is used exclusively. 
Another piece of evidence, in the same direction, is the Tibetan 
tradition respecting the introduction of the Northern Indian alphabet 
into Tibet (see Journal , Asiatic Society of Bengal , vol. LVII, pp. 41 ff.). 
It is said that these characters were introduced into Tibet by the sage 
Sambhota, who brought them fromMagadha, where he had resided from 
A. D. 630-650. These characters are known in Tibet as the “ Wartu” 
characters of Magadha; their forms, as traditionally preserved in Tibet, 
may be seen in Plate I of the Journal (ibidem) ; and it will be seen that 
among these the letter ya has the cursive form. This shows that at the 
time of Sambhota’s visit to Magadha, in the second quarter of the 
seventh century A. D., the cursive from of ya was in current use in 
North India.* 
I am not aware of the existence of a single dated inscription in North 
India , written in the North - Western alphabet , which indubitably proves any 
use , still less the exclusive or almost exclusive use of the old form of YA, 
after 600 A. D. It follows from this evidence that, since the old form 
of ya had entirely disappeared from inscriptions, from the end of the 
sixth century (say from about 580 A. D.), it must have disappeared 
from the cursive writing of ordinary manuscripts long before. Accord¬ 
ingly a manuscript, like the Bower MS., in which the old form is still 
used almost exclusively, must be placed long before the end of the 
sixth century, and much nearer the beginning of it. 
This conclusion is fully supported by the evidence of all the ancient 
dated (or practically dated) MSS. that are, as yet, known to exist. The 
oldest is the Horiuzi MS. The date of its writing has been shown by 
Professor Biihler to be somewhere in the middle of the sixth century, 
* The “Wartu” characters exhibit in all test points the characteristics of the 
North-Western alphabet. This shows, what I have already observed (supra, p. 82), 
that the North-Eastern alphabet, which was once current in Magadha, was there in 
very early times displaced by the North-Western alphabet. It is said, however, 
that Sambhota only “partly” adopted the “Wartu” characters for his Tibetan 
alphabet ( Journal , ibid., p. 41). This explains the fact that the “Wartu” or 
cursive form of ya does not appear in that alphabet. For the letter ya that sage 
appears to have drawn on the North-Eastern alphabet, which he must have known 
from Nepal, where (as I have shown) it maintained its ground from two to three 
centuries longer than in Magadha. 
