PITTIER TREATMENT OF THE GENUS CASTILLA. 279 



In 1001, Mr. J. Poisson, of the Paris Museum, published In an Informal way 



his Castilla clastica var. Uga a , collected somewhere In Guatemala by Mr. Rene 

 Gut'rin, Chief of the Chemical Laboratory of the government of that country. 

 TIi is variety is characterized mainly by the fact that it does not produce com- 

 mercial rubber. Besides this, the leaves are said to be smaller, less hairy, and of 

 a darker color than those of Castilla clastica (or rather of a species supposed to 

 be that, since C. elastica has not yet been found in this part of Central America) ; 

 the infructescences and seeds are also smaller, but the figured specimens are 

 evidently immature ones. 



The form referred to by Mr. Poisson may be identical with Castilla tunu 

 Hemsley as to Honduran material, but this can not be decided on the fragmen- 

 tary information at hand. At best, these data indicating the presence in Guate- 

 mala of a form that does not produce rubber might be accepted as a further 

 testimony to the real existence of Hemsley's enigmatic species. 



Mr. Poisson's suggestion that his C. elastica var. liga may be the primitive 

 type of C. clastica Cervantes, this being an " ameliorated form " ( ?), is scarcely 

 worth mentioning. Only modern man has reaped the benefit of such " natural 

 selection " and times are long since p;ist when the whole creation was considered 

 as subservient to his needs. 



"Bull. Mus. Hist. Nat. Paris. 7: :!73-377. 1901. 



