JStRONOMT of the BRAHMINS. 163 



one of thofe vibrations, which many centuries are required to 

 complete ; and that the year was then longer than it has ever 

 been fince, or than it had been for many ages before. It was 

 40"^ longer than it is at prefent ; but, at the year 5500 before 

 Christ, it was only 29" longer than at prefent, inftead of 2', 

 50", which is the refult of M. Bailly's fuppofition. During 

 all the intervening period of 2400 years, the variation of the year 

 was between thefe two quantities ; and we cannot therefore, by 

 any admiflible fuppofition, reduce the error of the tables to lefs 

 than 1', 5". The fmallnefs of this error, though extremely fa- 

 vourable to the antiquity, as well as the accuracy of the Indian 

 aftronomy, is a circumftance from which a more precife con- 

 clusion can hardly be deduced. 



33. The equation of the fun's centre is an element in the 

 Indian aftronomy, which has a more unequivocal appearance 

 of belonging to an earlier period than the Calyougham. The 

 maximum of that equation is fixed, in thefe tables, at 2°, 10', 

 32". It is at prefent, according to M. de la Caille, i°, 55'^ 

 that is, 15' lefs than with the Brahmins. Now, M. de laGrange 

 has fhewn, that the fun's equation, together with the eccentri- 

 city of the earth's orbit, on which it depends, is fubject to al- 

 ternate diminution and increafe, and accordingly has been di- 

 minifhing for many ages. In the year 3102 before our era, 

 that equation was 2°, 6', 28"f ; lefs, only by 4', than in the 

 tables of the Brahmins. But if we fuppofe the Indian aftro- 

 nomy to be founded on obfervations that preceded the Calyoug- 

 ham, the determination of this equation will be found to be 

 ftill more exac"l. Twelve hundred years before the commence- 

 ment of that period, or about 4300 years before our era, it ap- 

 pears, by computing from M. de la Grange's formula, that 

 the equation of the fun's centre was actually 2°. 8', 16"; £b 

 that if the Indian aftronomy be as old as that period, its error 

 with refpect to this equation is but of 2' *. 



x 2 34. The 



* Aft. Ind. p. 163. 



