218 Evans: NOTES ON GENUS HERBERTA 
observed. These are apparently never produced unless the growth 
of the main male axis is brought to an end in some way, possibly 
through the failure of an androecium to proliferate, possibly 
through an accident of some sort. In either case a branch tends 
to take the place of the main axis and to assume its functions, 
precisely as a ventral branch of a sterile axis or a subfloral innova- 
tion would do. When, therefore, an androecium is borne on a 
branch of this character, its position can not be regarded as typical. 
These considerations gain in significance when H. Hutchinsiae 
is compared with H. dicrana (Tayl.) Trevis., a closely related 
species of the Himalayas. The writer’s knowledge of this species 
is largely based on specimens received from Levier and collected 
by Decoly and Schaul (782) near Kurseong in Sikkim-Himalaya, 
in May, 1899. These specimens contain both male and female 
plants and are of especial interest because they served as the basis 
for Schiffner’s statements regarding the androecia of H. dicrana.* 
He notes their occurrence on the ends of branches, their small 
size, the small number of bracts (about four pairs) which they 
bear, and the fact that they are sterile at the apices. The branches 
which bear the androecia are ventral and vary greatly in length. 
Sometimes the development of the antheridia brings the growth 
of the branch to an end, but a sterile proliferation of the androecium 
often occurs. The formation of the male branches seems to be a 
perfectly normal process and is not preceded by a cessation of 
growth of the main stem as in H. Hutchinsiae. It follows, there- 
fore, that what occurs abnormally in the northern species has 
become normal in the Himalayan species. It should be noted, 
however, that ventral androecia are not absolutely constant even 
in H. dicrana. One example was observed where the androecium 
was borne on a main stem; it had somewhat larger bracts and 
bracteoles than the ventral androecia but resembled them in other 
respects. Aside from the difference in the position of the andro- 
ecia, H. dicrana resembles H. Hutchinsiae very closely. The 
leaves and underleaves, the bracts and bracteoles, and the perianths 
are much the same in the two species, although the divisions of the 
leaves in H. dicrana usually are less strongly curved. The latter 
species, moreover, is slightly larger, the leaves are less crowded 
and the thickening of the cell-walls is less pronounced. 
* Hedwigia 50: 148. 1910, 
