Evans: HeEpaTicAE OF PUERTO RICO 565 
specimen in the Lindenberg herbarium is an indeterminable frag- 
= ment.* Mitten, however, according to Spruce,f considered an- 
- other Swartzian plant as identical with B. ¢enuicaulis. Whether 
_ these specimens actually represented portions of the type material 
is not apparent. The Lindenberg herbarium also contains several 
_ specimens of /. fi/icina which were communicated by Hooker, but 
4 Stephani refers them all to B. fruticulosa Tayl. In the herbarium 
at Berlin there is a Brazilian specimen collected by Raddi ; Schiff- 
ner has recently determined this as B. tenuicaulis.} Until 1863 
~ no attention was paid to the nature of the inflorescence as a spe- 
cific character. In that year Gottsche § referred to B. fiiicina 
a series of Mexican specimens in which the inflorescence was 
_ monoicous. In 1864 || he ascribed a monoicous inflorescence to 
~ the species as a definite character. Spruce also restricted the 
_ name B. filicina to monoicous plants and imagined that he saw 
_ traces of androecia in Hooker's figure of a fruiting plant. Both 
_ Stephani and Schiffner follow Spruce in thus restricting the name 
_ and the same course is pursued in the present paper. 
As thus defined B. filicina is the only member of the genus 
q in which the inflorescence is monoicous. Its closest ally is 5. 
_ fruticulosa, which has a very similar geographical distribution. In 
_ this species, however, the inflorescence is always dioicous, and the 
plants are usually smaller and more closely pinnate then in B. fill- 
: cima. Unfortunately these last two differences are inconstant, and 
: Specimens of 3B. fruticulosa are sometimes met with which are as 
As Schiffner 
Separate the species except the differences in the inflorescence, and 
the attempt to keep them apart on this ground alone is perhaps 
questionable. The only other West Indian species with which B. 
~— Silicina is likely to be confused is ZB. tenuicaulis. (n this plant the 
_ Secondary stems are even longer and tend to be more loosely pin- 
Nate, with widely spreading branches, some of which assume a 
~ flagelliform character. So far as observed the lobules in this 
Species usually show a distinct apex, and this peculiarity may also 
Species usu and this peculiarity may a's0 
Core 
f 
* Hedwigia 29: 2. 1890, 
t Hep. Amaz. et And. 114. 1884. 
THedwigia 33 : 174. 1804. 
4 Mex. Leverm. 167. 1863. 
| Ann, Sci. Nat. Bot. V. 1: 45. 1864. 
