273 
Evans .— Vegetative Reproduction in Metzgeria. 
consequently does not connect the specialized branches studied by Ruge 
and Goebel with processes of reproduction. He attaches but slight impor¬ 
tance to these branches (from the standpoint of the taxonomist), and implies 
that they probably occur in many species where they have not yet been 
detected. He suggests that their development may perhaps be due to 
heliotropic stimuli, but makes no further attempt to account for them. 
The study of an abundant material of Metzgeria , mostly from North 
America, soon made it evident to the writer that the gemmae showed a 
much greater variety in form and structure than had been supposed. It also 
became evident that many of their peculiarities were specific in character, 
and that they often afforded a convenient means for distinguishing between 
closely related plants. The individual peculiarities of the gemmae have 
scarcely been noticed in earlier publications. This, however, is not sur¬ 
prising, because it is only within comparatively recent times that gemmae 
in other genera of the Bryophytes have been at all adequately described. 
Perhaps another reason why Metzgeria has been especially neglected in this 
respect is because the gemmae sometimes undergo reversion to a greater or 
less extent, and under these circumstances may fail to show their normal 
characteristics. This phenomenon has already been discussed by Goebel 
(’98 b ) in the case of M. furcata , and allusion will again be made to it in the 
following pages. 
Description of Gemmae. 
In the present paper the gemmae of twelve species are described, 
and fall naturally into three groups according to their position on the 
thallus. In the first group, including five species, the gemmae are 
marginal ; in the second group, including six species, they arise from 
the antical surface of the wings ; in the third group, including but a single 
species, they are indefinite in position. The gemmiparous branches in 
the first and second groups are very similar to normal branches, but in 
the third group they are much more specialized. In a few other species 
gemmae have also been noted, but they are not present in sufficient number 
for careful investigation. The remaining species examined seem to be 
entirely destitute of gemmae. Since these include both M. conjugata 
and M. hamata , where gemmae have been indicated by other observers, 
it would appear either that their statements were based on incorrect 
determinations or that the gemmae in these two species required very 
exceptional conditions for their development. Since several of the gemmi¬ 
parous species discussed are apparently new, their diagnoses will follow 
the descriptions of their gemmae. 
Metzgeria uncigera, sp. nov. 
A gemmiparous branch in this species tends to be narrower than 
a normal branch and to develop a less highly differentiated costa. When 
T 
