50 Br. R. Mitra — 0>i a Copper "Plate Inscrijjiion from Dacca. [MarcS, 



The most important part of the record is the date. Its letters and 

 figures are perfectly clear in their outlines, and there is no reason to doubt 

 the correctness of the reading (see Plate II). The first two letters of 

 the date (in line 15) are unmistakably Samva, an abbreviation of Samvat, 

 Then follow three figures of which the last is unquestionably a Nagari 3, 

 and the first may be accepted either as a 7, or a 9. Its exact counterpart had 

 nowhere been seen. It makes some approach to a modern Bengali 7, and 

 also to the Valabhi 9, but to no other figure of any of the different ancient 

 Indian systems of numerals known to us. Dr. Mitra took it for 7. The 

 second figure has been often met with in ancient records, and has been ac- 

 cepted by all the later antiquarians and epigraphists to be a symbol for 100. 

 It is formed of two arched lines joined to each other by their nearest legs by 

 a slight curvature, and the off leg of the right hand circle drawn downwards 

 in a line and terminating in a spur projecting forward. In ancient inscrip- 

 tions these details are not constant. In Prinsep's facsimiles, as given in 

 Mr. Thomas' edition of his essays, the juncture of the two arches is indi- 

 cated by a point, but in the Indian Antiquary, Yol. XI, p. 270, the slight 

 curvature is clear. The line in front is sometimes straight and some- 

 times curved, and the spur is not constant. Indian Antiquarij, VI, p. 45. 

 On one occasion Prinsep took the curved figure without a spur for 300. 

 It is obvious, however, that the several variants are intended for the same 

 primary type, and the variations are due to copyists' peculiarities. There 

 is no valid reason to show that the figure on the copper-plate under 

 notice is other than what have been noticed by Prinsep and others. Dr. 

 Biihler takes the arched figure to be the ancient Indian s as the initial of 

 sat a, a hundred, and the spur at foot the vowel-mark u. After this 

 group of three figures, occurs the word vaiMJcha, the name of the first 

 month of the Indian year. The next is the syllable di, the initial of 

 divasa a day, and it is followed by two figures, which are counterparts 

 of the second and the third figures of the first group, and this leaves 

 no room for any centenary value being attached to the second figure 

 of the first group. It would be absurd to suppose that the month 

 VaisdJcha included 103 days, and the idea must, therefore, be at once 

 abandoned. The symbol must be taken as a numeral of a decimal 

 system, and to ascertain its value the indication of the last figure is the 

 only available guide. It is well-known that no Indian month has ever 

 been made to comprehend more" than 32 days, and, the last figure being 

 an unavoidable and unquestionable 3, the preceding one must be 1 or 2, 

 having the value, by its decimal position, of 10 or 20, i. e., the date was 

 either the 13th or the 23rd of the month named. To take it as a 3, as 

 Prinsep once imagined, would be to make the month extend to 33 days, 

 and to make two dissimilar figures do duty for the same value. In 



