EvANs: THREE SoUTH AMERICAN SPECIES OF ASTERELLA 475 
to the large spores with the outer wall pellucid and “‘laxe celluloso”’ 
(in apparent allusion to the coarse surface-network). Spruce 
ascribed a dioicous inflorescence to the species, probably because 
the male branches escaped his notice, and stated further that 
ventral branches were lacking, that apical dichotomies were very 
rare, that the female inflorescence was terminal (presumably on 
the main thallus or on one of its innovations), and that the pseudo- 
perianth was definitely twelve-cleft. 
Stephani, in his description, states that the sexual branches 
are ventral in position and at least implies that forking is less 
uncommon than Spruce indicates. He definitely assigns an 
autoicous inflorescence to the species and emphasizes the minute- 
ness of the male branches. He also describes certain structural 
features omitted by Spruce, such as the green tissue with narrow 
air chambers and the elevated epidermal pores, each surrounded ~ 
by six radiating series of cells with four cells in a series. In this 
last characterization no allowance is made for variability, and 
the same criticism would apply to his account of the appendages 
of the ventral scales. According to his statements these are 
borne in pairs and are approximate, elongated, parallel, lanceolate, 
and composed of very irregular cells. In the writer’s experience 
the appendages are often borne singly—the only condition men- 
tioned by Spruce—and, although the parallel position of paired 
scales is sometimes striking, it is by no means constant., Steph- 
ani’s description of the spores as 63 uw in diameter, yellowish, and 
broadly ‘‘lobate cristate’’ might also be amplified to advantage. 
Spruce’s F. canalensis was based on material which grew on 
wet and shaded rocks. As emphasized in the original description 
the plants are extremely delicate, and the thallus shows almost no 
signs of pigmentation except on the appendages of the ventral 
scales. Stephani throws doubt upon the constancy of these 
features, suggesting that plants of less sheltered situations might 
perhaps be more robust, and the writer feels that these doubts 
are amply justified. It may be further shown that the points of 
resemblance brought out in the descriptions of F. canalensis and 
F. macropoda are many and important, while the differences are 
either insignificant or inconstant, this being true not only of those 
drawn from the texture and color but also of those drawn from 
