xviii Dr. Hoemle— Antiquities from Central Asia. [Exfra No. 1, 
informs me, are very untrustworthy in their estimates of distances. 
Allowing a discount of 25 per cent, for this uncertainty, and also for 
windiugs of the march, the distance measured on the map may be 
taken to be about 90 miles. This places Aq Tala Tuz not far to the 
west of the Qara Qasli river, at a distance of about 30 or 40 miles 
northwards of Khotan city. In favour of this determination it may 
be noted that the itinerary does not mention the passage of either the 
Qara Qash or the Yurung Qasli rivers. As Islam Akhhun’s expedition 
took place in January, both rivers would have been, at that season, in 
a very low state: still the total omission of the mention of the passage 
of either river, if it did take place, would be very strange. The pro¬ 
bability, therefore, is that Aq Tala Tuz as well as all the other place 
mentioned in the itinerary are situated in that part of the Takla Makan 
desert, which lies to the west of the Qara Qasli river, and at a distance 
from 10 to 20 miles north of the caravan route from Guma to Khotan. 
At Aq Tala Tuz those nine block-prints are said to have been found 
which are comprised in M. 7. Of their discovery an exceptionally 
circumstantial account is given, which must be taken for what it may 
be worth. As a rule, the only information obtainable about the block- 
prints was that they were found near Khotan. The tinders or the 
Khotanese merchants from whom they were obtained either could or 
would give no further information. 
At Qara Qol Mazar ( Jjj or ‘ black lake shrine of 
my lord’), where there is said to be “an immense grave-yard in 
ruins, about 10 miles long ” was found by Islam Akhun, in August, 
1895, the manuscript M. 1, Set. IV, described and figured by me in 
my Report in the Journal, As. Soc. Beng., Vol. LXVI, pp. 238 
and 253, Plates xviii and xix. It was found simply lying on the sand, 
probably uncovered by the action of the wind which had blown away the 
superincumbent sand. When found, it is said to have been “ bound 
between two little wooden boards” in the Indian fashion. These, having 
been broken on Islam Akhuu’s journey to Kaskghar, where he sold the 
manuscript to Munshl Ahmad Din for Mr. Macartney, were apparently 
thrown away by him. This is a pity, as in the present state of our 
knowledge of these Central Asian manuscripts every means of informa¬ 
tion is valuable. 
At Kok Giumbaz ( or ‘blue dome’) were found by the 
same Islam Akhun, in October, 1895, the two manuscripts M. 1, 
Set. V and M. 1, Set. VI, also described and figured in the same 
Report, pp. 238 and 253, 254, Plate xx. The latter manuscript is said 
to have been simply “picked up from the ground” similarly to that 
(M. 1, Set. IV) found at Qara Qol Mazar; but the other manuscript 
