FRAGMENTS OF THE SALISTAMBA-SÙTRA
THE following fragments originally formed part of a large and well-written paper pothi about 181 inches in width and 2.11 inches in height. The text is written on one side of the page only, and has been revised, apparently by the scribe himself. The pages are numbered in the usual way, seemingly by a second hand, on the left-hand margin.
As will be seen from the appended notes, the text preserved in these fragments substantially agrees with that of the Kanjur, Mdo xvi. It would seem that the text of the Kanjur represents a later revision of that given in our fragments, a few passages having been slightly expanded and ambiguous words changed to simpler language.
A number of roots which in modern Tibetan end in -r, -1, or -n have in our MS. a final -d appended. This -d is the so-called drag of the native grammarians, and is found in other ancient MSS.1 In isolated instances it is omitted in our MS. from roots that elsewhere have it, a fact indicating that it was already beginning to be dropped in actual speech. It can only be traced with certainty once in the other MSS. of Endere, which are apparently of a less canonical literary character ; and the popular sgraffiti on the same site also have it but once.
We also find a y between m and the high vowels i and e, for example myi, tnye, inyed. An exception is me ttog. After vowels, where the modern literary language writes an, we find here always yatt. Before a short pause a final -a sometimes appears to be lengthened to -â, the letter ( being added on the line whereas at the end of sentences the syllable o is added, as in the modern language. Other peculiarities need not be specially mentioned.
As regards form of writing, there is no difference from the modern dbu-can script. The letter sometimes has a small tick attached to it on the top, at the right side ; but this form is found in other old MSS. and books.
The fragments are numerous, and often minute ; but with the aid of a block-print of Mdo xvi of the Kanjur, kindly lent by the India Office Library, it has been possible to assign to each its place. In the text given below a thick vertical bar marks the point where the reading of a line in a fragment ends and is continued in another fragment. The original owner of the MS. tore each page into two or more pieces, and distributed these as offerings before the statues of divers deities [see above, p. 425]. Hence it is seldom that all the component parts of a page have survived. As a rule, however, the morsels are sufficiently numerous to make up the larger part of each of the pages represented by them ; and in such cases the gaps in the pages have been filled up by inserting the missing words from the Kanjur, enclosing them in square brackets and altering their spelling to suit that of our text. Before each page or fragment of a page is added a reference to the numbers of the pages in the India Office copy of the Kanjur, Mdo xvi, upon which the corresponding passage is to be found. In collation I have disregarded variations of no importance, e. g. where the Kanjur gives pa or ba as compared with ba or pa of our MS., or when the