TRIGONOMET'Rr of the BRAHMINS, 85 



. This principle may therefore have guided the aftronomers, 

 both of the Eaft and of the Weft, to the fame divilion of the^ 

 circle, without any intercourfe having taken place between them. 

 It has certainly direcfled the Chinefe in their divifion, though it 

 has led them to adopt one different from the Hindoo and Egyp- 

 tian aftronomers. They divide the circle into ^65 parts and -^^ 

 which can have no other origin than the fun's annual motion :. 

 and fome fuch divifion as this, may perhaps have been the firft 

 that was employed by other nations, who changed it however 

 to the number 360, which nearly anfwered the fame purpofe,. 

 and had befides the great advantage of being divifible into ma- 

 ny aliquot parts. The Chinefe, again, with whom the fciences 

 became ftationary almoft from their birth, have never attempt- 

 ed to improve on the method that firft occurred to them. 



3. The next thing to be mentioned, is alfo a matter of arbi- 

 trary arrangement, but one in which the Brahmins follow a 

 method peculiar to themfelves. They exprefs the radius of the 

 circle in parts of the circumference, and fuppofe it equal to. 

 3438 minutes, or 6oths of a degree. In this they are quite fin- 

 gular. Ptolemy, and the Greek mathematicians, after dividing 

 the circumference, as we have already defcribed, fuppofed the 

 radius to be divided into 60 equal parts, without feeking to af- 

 certain, in this divifion, any thing of the relation of the dia- 

 meter to the circumference : and thus, throughout the whole of 

 their tables, the chords are expreffed in fexagefimals of the ra- 

 dius, and the arches in fexagefimals of the circumference. They 

 had therefore two meafures, and two units ; one for the circum- 

 ference, and another for the diameter. The Hindoo mathema- 

 ticians, again, have but one meafure and one unit for both, viz. 

 a minute of a degree, or one of thofe parts whereof the circum- 

 ference contains 2 1 600. From this identity of meafures, they 

 derive no inconfiderable advantage in many calculations, though 

 it muft be confeffed, that the meafuring of a ftraight line, the 



radius,. 



