14 
Dr. Hoernle —Antiquities from Central Asia. [Extra No. 1, 
form of y would be tlie 5th century A.D. Accordingly, Pothis which 
use exclusively the old three-pronged form of y may be placed before the 
5th century; and Pothis which use exclusively the modern boot-shaped 
form of y fall after the 5th century. Of course, the form of y is not the 
only test; the forms of other letters must also be taken into account; 
but in a general way, the y-test is a convenient and fairly safe one. In the 
present case, the result yielded by this test is sufficiently well confirmed 
by a comparison of the forms of the other letters shown in Table II. 
Parts I—III of the Bower MSS. show the use of the old and transi¬ 
tional forms of y in the proportion of 1: 3, while Parts IV-VII, which 
are written in a different hand, use only the old three-pronged form. 
As the last page of Part III, and the first page of Part IV, are written 
on the same leaf, occupying the obverse and reverse of it respectively, it 
follows that all the Parts are contemporary. It further follows that the 
Bower MSS. must have been written at a time when the use of the 
transitional form of y had not yet found general acceptance among liter¬ 
ary men. Accordingly they must be referred to a time not later than 
the middle of the 5th century A.D. The other Pothis, viz., the Weber 
MSS. and the Macartney MSS., having been found in the Kuchar stupa 
together with the Bower MSS., must belong to nearly the same period. 
Hence Pothi, No. 1 of Set I, which is written in the Indian type of the 
Gupta script, but never uses the transitional form of y, must be dated 
early in the 5th century, and may possibly go back to the latter part of 
the 4th century (see the evidence in the Journal, Asiatic Society Bengal, 
Vol. LXVI, pp. 245-247). The Pothis, No. 2 of Set I, and No. 1 of Set II, 
which are written in the Central Asian type of Gupta, must also belong 
to the 5th century. The two-ringleted form of y (see No. 14 in Cols. 21 
and 22 of Table II) which they use can only be viewed as having 
originated from the old three-pronged form: the boot-shaped form of y 
could not have produced it. The script of those two Pothis is a Central 
Asian modification of the contemporary Indian Gupta script. The inter¬ 
mediate stage—the Indian Gupta in the process of transition to the 
Central Asian—is exhibited in Part V of the Weber MSS. (see Column 
20 of Table II; also Journal, Asiatic Society of Bengal, Vol. LXII (1893), 
Plate II, fig. 1). The transition can clearly be seen by comparing 
the forms of m and y, in Columns 20-22 of Table II. The North- 
Indian transitional form of y of the 5th century, and the modern boot¬ 
shaped form of y, which originated in the 6th century, never obtained 
a footing in Central Asia, as little as it penetrated into Tibet or into 
Southern India. The Bower MSS. found in Eastern Turkestan are only 
an apparent exception. For as their material—birch-bark—shows, those 
manuscripts were really written in India (probably Kashmir or Udyana) 
