1903] ODONTOSCHISMA MACOUNII AND ITS ALLIES 327 
forms in which the branching has now become constant. It is 
clear, for example, that ina species with prostrate stems the posti- 
cal position of the flagella is of distinct advantage, because it ena- 
bles these organs at once to penetrate the substratum and to act 
as fixing and absorbing organs. Ina leafy branch, however, the 
advantages of a postical origin is not so clear. Except at the 
very beginning, when the branch needs protection from drying, 
it would be placed at a disadvantage, because it would have to 
grow out beyond the leaves of the axis before it could develop 
normally and expose its leaves to the light. In such a case a 
lateral displacement would enable it to perform its functions 
earlier, while an antical position would be most advantageous of 
all. The last, however, would be precarious, unless there were 
some provision for keeping the branch-rudiment moist; this is 
secured in Anomoclada by a copious secretion of slime. In 
sexual branches the postical position is at first of distinct advan- 
tage, because it protects the antheridia and archegonia from 
drying, and at the same time tends to insure fertilization. It 
continues to be of advantage in such genera as Kantia, where a 
subterranean sac is developed. But in most cases the female 
branch curves upward and the perianth shortly assumes a posi- 
tion at right angles to the substratum. In this way the young 
capsule is so placed that a simple elongation of its stalk will push 
it above the perianth and expose its ripened spores to currents of 
air. The latter advantage is, of course, more easily secured by 
lateral or antical branches: Lateral branching, on the whole, 
seems to be the most serviceable type. This is seen especially 
well in the large and very successful group of the Jubuloideae, 
where postical branching has entirely disappeared. In this group, 
to be sure, the branching is terminal, and the lateral branches 
are therefore laid down at the growing apex. There is little 
doubt, however, that in the ancestors of these forms the method 
of branching was less definite. 
LEAF-CELLS. 
The thickening of the cell wall has already been quoted as an 
important generic character of Odontoschisma, and the variations 
in the thickening afford excellent differential characters for the 
