356 Journal of the Asiatic Society of Bengal. {September, 1912. 
/ 
The ‘one’ is of the ancient form, consisting as it does of a 
curved stroke that is approximately horizontal, but we find 
examples of this usage even to the present day. The three is 
similar to the symbols used in the eleventh and sixteenth 
centuries. 
The ‘ four ’ is of ancient type, but instances of the same 
type occur in inscriptions of the twelfth century a.p.” 
e six is the most remarkable, but closely resemble the 
Devi-ri-kothi ‘ six.? The eight is also peculiar, but resembles 
the modern Kashmir ‘eight’ as much as anything. The 
Peshawar Museum Inscription of Vanhadaka has figures whose 
resemblance with those of the Bakhshali manuscript is striking. 
The following table illustrates these points :— 
VII. 
evidence, viz. the mathematical contents of the work. These, 
it will be 
tion. Indeed, in all probability, the work is comparatively 
__ The form in which the problems are written differs con- 
siderably from the usual early Hindu practice. These earl 
nomical work and not in his arithmetic !). Brahmagupta used 
symbols which differ from those of the Bakhshali manuscript 
very considerably, but he only used them in certain classes of 
1 LA, sii 008; Adasen Veda (ioctl oe aa 
S:Awk Chatnbll. p Sites — — 
