106 
Dr. Hoernle —Antiquities from Central Asia. [Extra No. 1, 
Eighth Set. (Plates XIV, fig. 3 and XV, fig. 1.) 
This set comprises only one book. It belongs to G. 9. It was 
purchased by Sayyid Gul Muhammad, a well-known Kashghar mer¬ 
chant, for forty rupees and was sent, as a present, to Captain Godfrey. 
The book, of course, could not be accepted as a present, but it was 
purchased on behalf of the British Government. It measures 11 x6J". 
The exact number of its forms is unknown, for the beginning and end 
are missing, and a large number of leaves exist only in fragments. 
The number of still complete forms is 29; most of these even are more 
or less damaged along the edges. The book is bound with three copper 
nails, and the guards are formed of two thin copper slips, measuring 
8 x |", and covered with ornaments like those on figs. 4-9, Plate IV. 
Irrespective of its script, this book strikingly differs in several 
points from the books comprised in all the previously described seven 
sets. In the first place, it is clean; there is no trace of any burn or 
fatty stain. In the second place, the paper, to all appearance, is of an 
entirely different quality. It is thin and soft and more nearly resembles 
the paper of the Weber and Macartney Manuscripts procured from 
Kuchar. It differs, however, from their paper in colour; for while their 
paper is white or whitish, the paper of this book is of a bright yellowish- 
brown. It looks as if it were artificially tinted; but the colouring, 
if any, is fast, for it is tolerant of washing. It is a pity that its find- 
place is not known; but that it comes from some spot in the Takla 
Makan is shown by the fact of all its leaves being, like those of- all 
the other blockprints, very thickly covered with the fine yellow sand 
of the desert. Another curiosity is that a small special formula, which 
occasionally occurs in it, is printed with an apparently faded, red- 
coloured fluid, which almost resembles blood. Its ordinary formula is, 
as usual, printed with black ink. Minor peculiarities are the following: 
(1) most of the existing leaves show a clean cut on one of the narrow 
sides, (2) two of the pages have the text printed diagonally across them, 
and (3) a few leaves are only printed on one side. The last mentioned 
peculiarity is due to the extreme thinness of the paper, owing to which 
the print on one side shows through on the other. The leaves have, as 
in the case of all other block-printed books, frayed edges, but in the 
present case one of the narrower sides of most leaves has been clipped 
with a sharp knife or scissors, for it shows a clean cut, which occasion¬ 
ally passes right through a line of print, showing that the clipping 
was not done with little care. 
The text of the book consists of two formulas which I shall call 
VIII a and VIII b. The formula VIII a consists of three long lines, 
containing apparently about 16 letters each. It is the proper formula 
