:{(M.» WILLIS : THE FLORA OF RITIGALA : 



hav. not Strobilanthea stenodon and Phyllanthus rinereus. 

 both i.' p< ies, arc represented by varieties. Cyperus and 

 Colens are represented by endemic species, but we do not know 

 the ancestry of these, whether dry zone or moist, rare or 

 common. Tbe only common forms represented by endemic 

 varieties are Anotis nummularif ormis and Bxacum zeylanicum. 



\ot only do these facte stand out, but the endemic forma 

 .iic on the whole among the families which show the greatest 

 number of endemics elsewhere in Ceylon : thus 



and 1<> add to these the endemic forms among the bird and 

 wind-carried species, we have — 



Cycles Menispermaceae endemics out of 13 



Trichomanee Pilices -'.'i do. 239 



Bulbophyllum Orchidaceae 79 do. 161 



Cirrhopetarurrj do. 79 do. 161 



and the one form among the dry zone plants 



Thuhbergia \<anthaeeae 38 endemics out of 93. 



\ddiug up the-.- figures we find that the endemics of Riti- 

 gala are in all among families which possess in Ceylon 32") 

 endemic species among a total of 1,098 species. While if we 

 take the other member- of (he. Ritigala flora we find that the 



other in species with doubtful meohanisms belong t" families 

 which have only 200 endemics among !»7"> specie, and this 

 proportion is largely brought up by Mm fact that there are 



iioii-eiideinic Kubiaeeae and Kuphorhiaceae among them. 



\\ • may put the tact- in another way. thus : 



