104 
Dr. Hoernle —Antiquities from Central Asia. [Extra No. 1, 
indistinct; otherwise well preserved. Bound with three twists of 
paper. Printing stops on the penultimate, but begins, as usual, on 
the fourth page. 
The arrangement of the text is exactly the same as in No. II, 
except that the columns only consist of three impressions of the formula, 
whence a wide margin is left at the top and bottom of every page. 
There are, however, four exceptional pages, probably misprints, on 
which the columns run in the same direction. 
No. IV. Book. 
Belongs to M. 8. Found at Kiang Tuz. Size, 13 x 8f". Number 
of forms, 20. Paper, variety IIIcZ. Surfaces greasy, and print rather 
indistinct; otherwise well preserved. Bound with two nails. 
Text printed in three columns on each page, running parallel to 
the narrower side of the book, and consisting each of two impressions 
(a b c d e f) of the formula. The columns, as a rule, run in the same 
direction (e.g., pp. 4, 5); but there are many exceptions (<?.y., pp. S, 9), 
as shown in the subjoined diagram. 
Ordinary. Exceptional. 
Page 4 
u 
Page 5. 
Page 8. 
Page 9 
• 
o 
c 
c 
c 
ft. 
a 
ft. 
a 
c 
C 
rO 
pO 
cd 
rO 
CD 
rO 
CD 
Cb 
CT¬ 
Cb 
or¬ 
rO 
<D 
rO 
CD 
c 
03 
03 
03 
os 
03 
53 
OS 
OS 
es 
os 
03 
OS 
OS 
c 
C 
c 
c 
fti 
a 
ft, 
a 
c 
c 
rO 
c 
rO 
<d 
rO 
CD 
rO 
<D 
Cb 
QJ- 
Cb 
or¬ 
fO 
<X> 
CD 
os 
03 
03 
03 
e 
03 
a 
03 
'-S 
Cb 
es 
03 
03 
03 
'OS 
Cl 
C 
c 
o 
4 
ft. 
a 
ft. 
a 
c 
c 
rO 
c 
rO 
<D 
rO 
CD 
rO 
<D 
Cb 
o- 
Cb 
CJ- 
fO 
<D 
pO 
<D 
C 
03 
e 
03 , 
S3 
03 
a 
03 
os 
Cb 
a 
03 
os 
03 
The three columns do not fill up the whole available space; there is 
a wide margin on both sides. Accordingly on 14 pages (out of a total 
of 74 printed ones) a half-column is added, consisting of two half¬ 
impressions of the formula divided vertically. As a 
rule, this half-impression stands on the margin of the 
same side, but on page 16, shown on the margin, it 
stands once on the left, and once on the right margin. 
As it is invariably the same half of the formula, this 
circumstance might be considered to indicate the 
side with which the formula commences; and 1 
have used it for that purpose in arranging the 
facsimile on Plate XIV. But, as already observed, 
the argument is not altogether conclusive. 
Page 16. 
, __ 
pO 
<D 
c 
c 
03 
rO 
<D 
03 
03 
o 
CO 
pO 
CD 
c 
03 
03 
pO 
<D 
03 
03 
Cl 
C 
rO 
S 
CD 
c 
03 
03 
pO 
c 
03 
03 
C 
pO 
c 
rO 
CD 
e 
03 
03 
03 
