328 ANTIENT METAPHYSICS. Book IH. 



Latins, whofe cuftom it was to change the afperated T or e into the 

 middle letter D. 



And here the reader may obferve, that as the Latin is the moft 

 antient diale<51: of the Greek, many of the words of the Shanfcrir 

 have a greater refemblance to the Latin than to the Gre^k. Thus, 

 as I have obferved before, the word nafa^ denoting a nofc^ is plain- 

 ly Latin, but entirely different from the Greek word cxprefling that 

 feature of the face, which is giv. Apa^ which in Slianfcrit figniiies 

 ivatefy has the greateft refemblance to the Latin word aqua^ but 

 none at all to the Greek word 'y^a»o; and here is another word which 

 mult have been oiiginally In all languages. And I fliall only men- 

 tion two words more, veedbava^ in Shaufcrit, which is the vidua of 

 the Latins, but is quite different from the Greek word j^>7f a ; and 

 laka, which is the Shanfcrit word for locus.y but is quite different 

 from the Greek word ror.f. 



In thefe words the reader will obferve, that there is a good deal 

 of difference in the found betwixt the Greek and Latin words and 

 the Shanfcrit: But' there is nothing in language fo changeable as the 

 pronounciation of it, even in the fame nation. But, when the lan- 

 guage goes to a different nation, efpecially to one at fuch a diftancc 

 as India is from Egypt, when there mull be an intercourfe betwixt 

 the two nations, fuch as might preferve, in fome degree, the original 

 pronunciation of the language, the change mull be very great. 



To the words which Mr Y/ilklns has given me, I will add fome 

 that are preferved in Diodorus Siculus, and Arrian's Indica: And,. 

 as they are the names of places and perfons in India, when Alexan- 

 der was there, they are undoubtedly very antient words of the Shan- 

 fcrit, as the names of perfons and places arc the moft antient words 

 in all languages ; they have, therefore, more of the Greek found, 



in 



