Contributions towards a Flora of Ceylon. 455 



persistent tube of the perianth, about 1^ line long, containing a 

 single seed : putamen black, crustaceous. Seed inverse : albumen 

 none : Embryo orthotropous : Cotyledons plano-convex : radical short, 

 superior. 



Hab. — A very distinct species from any of those hitherto 

 described. 



GNIDIA, Linn. 



Some very interesting links connecting the Flora of Western 

 India with that of Southern Africa have lately been made 

 known by Dr. Wight. Thus in the last part of his " Icones 

 Plantarum," (Vol. iii, part 4,) he has published drawings and 

 descriptions of a species of Vogelia and of Apodytes,* both 

 African genera hitherto unknown in India. This connection 

 will be further illustrated by the present article. All the 

 hitherto published species of Gnidia are natives of South 

 Africa, but I have now to make known three Indian species, 

 one of which is peculiar to Ceylon, another to the Neilgherry 

 mountains, and a third common to both countries. One of 

 them has long been known to Botanists by the name of 

 Daphne eriocephala, that being the appellation given to it by 

 Dr. Wallich in his " Catalogue," but, so far as I am aware, 

 no description of it, or of any of the others, has yet been 

 published. According to Wallich the same plant was referred 

 by Heyne to Lachncea, but it, as well as the other two, differ 

 from both these genera in having faucial scales. With 

 Gnidia, to which I now refer them, they agree in every thing 

 except the number of the parts of the flower, the African 

 species being tetramerous, while the Indian ones are penta- 

 merous. This, however, is not of sufficient importance to 

 exclude them from the genus, of which they will form a 

 section, and to which I propose to give the name of Dingia, 



* I have lately met with the same or an allied species in Ceylon. 



