808 Journal of the Asiatic Society of Bengal. [December, 1911. 
of their mathematics to the Hindus. The rule means that on 
division by nine of any numbers the possible remainders are 
for linear numbers .. 1,8, 2,7, 4,5 3,6,9 
Hee es eee 
js; eGUare: - .;, es 1 4 < 9 
Not only is the translation of hindasi by ‘Indian’ here philo- 
logically wrong but the term ‘Indian’ cannot be applied to the - 
Hindu work the slightest indication that would lead one to 
expect to come across such arule. The same termh indasi is 
applied to another rule obviously deduced from the former by 
Avicenna. This second rule may be expressed thus— 
n= +1 mod 9 
No such rule occurs in any Hindu work. 
aie 
‘* Ther was a kyng of Inde the quich heyth Algor and he made this 
craft... Algorisms, in quych we use teen figurys of Inde.’’ ! ; 
The cases in which the term ‘Indian’ is applied to numeri- 
cal symbols are not at first glance as easy to deal with since ib — 
appears incongruous to apply the term ‘ geometrical’ to them. 
The term hindasi is, however, quite broad enough to include 
them as will be seen later on. 
The symbols given by Maximus Planudes, el-Qalasadi, and 
other Western writers, together with some other forms for the 
sake of comparison, are exhibited in the accompanying table. 
_ _ The table on opposite page shows that the sets of symbols ~ 
under consideration (a, b, c, d, e) are not of Hindu but Arabic 
forms. Where they differ from the Arabic they generally 
bers ; (3) Commentary on the Talchis of el-Benna, etc.” 
PERE ne Bee 2 an : 
P 1, Quoted by Smith and Karpinski.—The Hindu-Arabic Numerals, 
, * A. Cherbonnean. Notice bibliographique sur Kalacadi, mathémati- 7 
aa du XV° Siécle, Journal reba 1859, Tome XIV, 5¢ Série, — 
See also Suter, p. 180, 
