308 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
On rocks; known only from central Mexico. The following specimens have 
been examined: 
JaLisco: West end of the Sierra de San Esteban, near Guadalajara, alt. 1,600 
meters, September 28, 1908, Barnes & Land 192 (Y.; type). 
Moretos: Near Cuernavaca, October 18, 1908, Pringle 10667 (Y.; distributed 
in Pl. Mex. as Fimbriaria echinella). 
A wide range in color is found in many species of Asterella but is particu- 
larly striking in the present species. The upper surface of the thallus is usu- 
ally green, sometimes with a glaucous cast, but the lower surface with its ven- 
tral scales and the peduncles of the female receptacles are deeply pigmented 
with purple. The greatest variety in color is shown by the receptacles them- 
selves. The upper tuberculate portion seems to be pretty constantly green when 
young, becoming yellow or brownish when old, but the involucres and pseudo- 
perianths show all gradations from a pure white to a clear reddish purple, the 
amount of pigmentation varying in both extent and degree, although always 
less than that of the thallus and peduncles, 
The slender branches mentioned in the description are almost as thick as 
broad, and represent a striking and interesting condition, Apparently an ordi- 
nary vegetative thallus grows out directly into a process of this character, the 
keel becoming strongly rounded, the upper surface concave, and the thin mar- 
ginal portions very narrow and incurved. The branch thus takes on the ap- 
pearance of a narrow purple cylinder, the upper surface being more or less 
completely inclosed. In all probability these branches may be looked upon as 
a xerophytic adaptation, this idea being supported by the incurved margins of 
the normal flat thallus under conditions of dryness and by the compactness of 
the green tissue. 
In the structure of the thallus, in the method of branching, and in the restric- 
tion of the inflorescences to ventral branches, A. versicolor shows a close rela- 
tionship to A. elegans. Even the ventral scales have much in common, although 
the slime papillae tend to be more persistent in the Mexican species and the 
appendages are usually smaller. Further resemblances are to be noted in the 
tuberculate female receptacles, with their short, obliquely spreading lobes, 
broad and undivided involucres, and pseudoperianths with coherent, lanceolate 
segments. The most striking differential characters are those derived from 
the male inflorescences and the spores. In A. versicolor the male branch is 
very short, the antheridia form a small, indefinite group without marginal 
paleae, and the spores are covered over with a fine and very irregular reticulum. 
In A. elegans, on the other hand, the male branch varies greatly in length, 
the antheridia are in a larger and more clearly defined group with a fringe of 
marginal paleae, and the spores, which are usually paler, are covered over with 
a coarse and usually regular reticulum, The tubercles on the female receptacle 
are about as long as those of A. elegans, averaging perhaps 0.38 mm. in length, 
and are therefore appreciably shorter than those of A. echinella, with which 
the new species has been confused, 
The dark spores and purple elaters of A. versicolor have much in common 
with those of A. lindenbergiana, although the relationship between the species 
is exceedingly remote. Even the spore markings, which consist of a delicate 
network formed by low anastomosing ridges, are much the same, It is not 
diflicult, however, to detect differences in the spores. The reticulum in A. versi- 
color, for example, is more irregular than that of A. lindenbergiana, the meshes 
showing a wider range of variation in size, and the wings are more conspicu- 
ous and distinct, owing largely to the tendency ef the outer spore wall to 
become separated in A. lindenbergiana. The spores, moreover, are a trifle 
larger, and the elaters are usually longer. 
dae 
i}. 
