THE PALMS OF BRITISR INDIA AND CEYLON. 375 



Parish, in the meantime had an opporttinity of examining a 

 few flowers taken from the type specimen of this species, collected 

 by W. Palmer at Guyamas, Mexico. The flower has the charac- 

 ters assigned to W. filif&ra so far as the filaments are concerned, 

 the character of W. rohusta as to the divided tip of the stigma, 

 and the markedly lobate ovary of W. gracilis. (Fig. 23). 



Fig. 23. — Flower of Washingtonia sonorWf S. Watson. 

 X 5 ('After Parish.) 



It must not be forgotten, however, that these characters are 

 drawn from a study of a few individuals only. 



Habitat. — First collected by Dr. Edward Palmer in 1887 " in 

 secluded canons in the mountains about Guyamas " (Watson) ; 

 "in great quantities, and of great size, in the deep canons run- 

 ning into the Gulf of California, far away from Guyamas " (E. 

 PalmerJ ; La Paz, in Lower California, and on the opposite side 

 of the gulf (collected in these localities by E. Palmer and identi- 

 fied as W. sonorce by Watson). 



" The natives of the Cape region," says Parish, " recognize 

 three distinct forms of W. sonorce. Two of these " Palma Blanca " 

 and "Palma Colorada" — are distinguished by the white or red 

 colour of the woody fibre of their trunks. The third — " Palma 

 Nigra " — is said to be the smaller tree, growing at higher altitudes, 

 where winter snows are not infrequent."^ 



^ For more detailed information on the genus and species of Washingtonia 

 we refer to the extensive ' bibliography ' on this subject by S. B. Parish in Bot. 

 Gaz., vol. 44, 1907, p. 431—434. 

 10 



