﻿480 APPENDIX. 



no suspicion of the real intention of the figu r e till I 

 saw the use here made of it. It seems, however, an ar- 

 gument in favour of the identity of the Druids and 

 Bram'ms as well as another well-known diagram, 

 visually caiied the Walls of Troy, which was used 

 originally in the IT:v.doo astrology. These figures, how- 

 ever, appear to haveilowed from a much higher source, 

 and to have relation to what Leibnitz had a distant 

 idea of in his Analysis of Situation, Euclid in his Po- 

 risms, and Girard perhaps in his Restitution of them. 

 In fact., as the modern Algebraist.- hive the advantage 

 of transferring a great part of their labour from the 

 head to the hands, so there is reason to believe that 

 the Hindoos had mecha?iical methods of reasoning geome- 

 trically, much more extensive than the elementary 

 methods made use of at present; and that even their 

 games were deduced from and intended perhaps to 

 be examples of them : but this deserves to be treated 

 more at length elsewhere. 



The same apology may perhaps excuse my men- 

 tioning here, that the idea of the JSile\ deriving its 

 floods from the melted snows, as well as the Ganges, 

 appears to be rather imaginary : they seem to be caused 

 principally by the rains ; for the high hills beyond the 

 fferd-zvar apparently retain the snow all the year, and 

 therefore the quantity melted could never produce 

 the enormous swell of the Ganges ; not to mention 

 that the effect of a thaw seems different from what 

 would arise from the mere difference of heat, and 

 therefore might partly take place in winter and the dry 

 season. 1 hat the rains are sufficient for the purpose, 

 without recurring to the hypothesis of melted snows* 

 appears from the following fact : — A little before I 

 observed die aforesaid pyramid, I had been a consi- 

 derable distance up the Gunduc ; the river was low 

 for the time of the year, and the hills that skirt the 

 borders of Nefmtl were clear, and apparently not above 



