358 



Notes on Indian Bo 'any. 



characters of his Genera Plan tar um of the ovary. This 

 character passed current among Botanical writers until the 

 publication of the 4th volume of De Candolle's Prodroums 

 in 1830, when he altered it by the addition of the word 

 "abortu" " Bacca abortu l-sperma" thereby implying that 

 there were in the ovary a plurality of ovules, all except one 

 of which aborted in progress towards maturity. This addition 

 has, since then, been admitted by all writers, so far as I am 

 aware, except Professors Endlicher and Lindley. The former 

 describing the ovary i( Ovarium inferum triloculare, Ovula 

 in loculis solitaria ex apice anguli centralis pendula," but 

 adds " Bacca abortu unilocularis monosperma/' evidently 

 implying that he had himself examined the ovary of at least 

 one species, if not more, and found it as here described, 

 plurilocular. Lindley, in his School Botany, takes no notice 

 of the ovary, but allows a 3-seeded fruit ; " Fruit succulent, 

 3-seeded," though in two of the three species he defines, 

 I find the ovary 1 -celled with a single ovule. 



While examining the four Neilgherry species with reference 

 to the articles on Caprifoliacece for my " Illustrations of 

 Indian Botany" and "Neilgherry Plants," I found in all a 

 1 -celled ovary with a single pendulous ovule, and naturally 

 inferred, on comparing them with Endlicher' s character, that 

 they must form a distinct genus. But before finally sepa- 

 rating them, and adding I knew not how many synonyms 

 to our already overgrown list, I determined to examine the 

 structure of the ovary in every species to which I had access, 

 among which fortunately was V, Lantana and V. Opulus, 

 two British species, in both of which the same structure 

 exists. Three American species were next examined with the 

 same result ; then five Nepaul ones, still the same ; and lastly, 

 three from other parts of India, in all seventeen species, in all 

 of which the ovary is 1-celled with 1 ovule. 



It results from these observations that the inference already 

 drawn in regard to this genus becomes almost inevitable : 

 namely, that if either Endlicher or Lindley's characters are the 



