﻿1 N T s - > 

 ON THE VLu OF INDIA. 347 



orders, of which he hath given a rough sketch, and 

 which may hereafter, perhaps, be completed : but the 

 distribution of vegetables into classes, according to the 

 number, length, and position of the stamens and pis- 

 tils, and of those classes into kinds and species, ac- 

 cording to certain marks of discrimination, will 

 ever be found the clearest and most convenient of me- 

 thods, and should therefore be studiously observed 

 in the work which I now suggest; but [ must be 

 forgiven, if I propose to reject the Linn&an appella- 

 tions of the twenty-four classes, because, although 

 they appear to be Greek (and, if they really were so, 

 that alone might be thought a sufficient objection) yet 

 in truth they are not Greek, nor even formed by ana- 

 logy to the language of Grecians ; for Polygamos, Mo- 

 nandros, and the rest of that form, are both masculine 

 and feminine ; Polyandra, in the abstract, never oc- 

 curs, and Polyandrion means a public cemitery j 

 dicecia and hiacus are not found in books of authority ; 

 nor, if they were, would they be derived from dis, but 

 from dia, which would include the tri&cia ; let me add 

 that the twelfth and thirteenth classes are ill distin- 

 guished by their appellations, independently of other 

 exceptions to them, since the real distinction between 

 them consists not so much in the number of their (ta- 

 mens, as in the place where they are inserted ; and 

 that the fourteenth and fifteenth are not more accu- 

 rately discriminated by two words formed in defiance 

 of grammatical analogy, since there are but two pow- 

 ers, or two diversities of 'length in each of those classes. 

 Calycopolyandros might, perhaps, not inaccurately de- 

 note a flower of the twelfth class ; but such a com- 

 pound would still savour of barbarism or pedantry ; 

 and the best way to amend such a system of words is 

 to efface it, and supply its place by a more simple 

 nomenclator, which may easily be found. Numerals 

 may be used for the eleven first classes, the former of 

 two numbers being always appropiated to the stamens, 

 and the latter to the pistils. Short phrases, as on th 



