1895.] A. F. R. Hoernle — Ancient Manuscripts from Central Asia. 85 



The fragments number several hundreds, and arc mostly utterly 

 useless, but there are about a hundred of the size of one to two inches 

 square, which bear connected letters, and are legible. 



The material of the fragments are of three different kinds : palm- 

 leaf, birch-bark, and paper. The paper is of several varieties of manu- 

 facture. 



The number of manuscripts, represented by these fragments, it is 

 difficult to define with certainty. To judge from the varieties of writing 

 and material, however, there must have been not less than eight or nine. 

 There was certainly one manuscript of palm-leaf. There was also 

 certainly one, if not two, of birch-bark; for the writing on the fragments 

 seem to show two different styles of writing. Of paper manuscripts 

 there must have been, at least, five, but probably more ; this is shown 

 by the varieties of writing and make of paper. 



Particularly noticeable is that the palm-leaf and birch-bark frag- 

 ments show a purely Indian type of writing, of the North-Western 

 Gupta class, similar to that on the Horiuzi palm-leaf MS., and the 

 birch-bark Bower MS. This, indeed, might have been expected from 

 the fact that the material is palm-leaf or birch-bark, which is not 

 obtainable in Central Asia. Manuscripts on these materials must have 

 been prepared and introduced from India. The fact of the occurrence 

 of them in Central Asia may be of considerable chronological value. 

 They exhibit the old form of the tridentate y, and the old Gupta form of 

 m. The superscript r is formed on the top line of writing instead of 

 above it. 



The paper manuscripts uniformly exhibit the Central Asian kind 

 of Nagari, as shown in the Weber Manuscripts. But they are of con- 

 siderable variety. Some approach very closely to the pure Indian type, 

 as in Weber MS., Parts land II (Plate I, Fig. 1, 2; in Journal., As. Soc, 

 Beng., for 1893, pp. 9, 17), others show the purest Central Asian type 

 as in Weber MS., Parts IV to IX (ibid., PL II, Fig. I, 2, 3, PI. Ill, 

 Fig. 1-5). In particular, there are a few fragments, which are so 

 strikingly like several leaves of the Weber MSS., as to suggest that 

 they TociB^j have belonged to missing leaves of the same manuscripts. 

 The writing that remains on the fragments, however, is too defective to 

 allow of arriving at any more certain conclusion. 



A Plate of selected specimens is being prepared by Col. Waterhouse, 

 of the Survey of India, which, I hope to be able to publish, with addi- 

 tional information, in the Journal of the Society. 



The Philological Secretary exhibited two rare Assam coins for- 

 warded by Mr. E. A. Gait. One is a coin of Raghu-narayana, the 

 founder of the western branch of the Koch Kings (vide J. A. S. B. 



