146 BOTANICAL GAZETTE [AUGUST 
undesirable, to regard all the Washingtonias as varieties of a single poly- 
morphous species, but the one now under consideration would of all be the 
least capable of such comprehension. Without question floral characters 
Fic. 3.—W. robusta Wendl., Botanical Garden, Palermo, August, 1906.—0. 
BECCARI. X3.5. 
are of greater diagnostic value than those drawn from foliage or habit; 
but when the latter are of marked distinction, and apparently constant, 
they cannot be refused great weight. 
\ ; ese 
VAS 
‘ 
Fic. 4.—W. gracilis Parish, Riverside, Cal. X5- 
Beccart had not had an opportunity of examining flower of ul 
sonorae S. Watson, and he regards it as a doubtful species, which as 
a variety of W. robusta, suspecting that the obtusely triangular insert) 
