26 Mitchell .— Contributions towards a Knowledge of the 
microsporangium has a smooth, regular outline, and contains numerous 
small microspores, which are, however, tetrahedral in shape. The adult 
sporangium has two walls—an outer one, containing chlorophyll until after 
the spores are shed, and considerably modified in various parts as regards 
thickness, &c., and an inner one, uniform throughout, consisting of some¬ 
what flattened, sometimes elongated cells, whose boundaries are difficult to 
distinguish. The tapetum, which is so characteristic and prominent a layer 
in developing sporangia, persists until the spores are nearly mature, but then 
degenerates into a pavement-like layer whose cellular structure can scarcely 
be determined (PI. Ill, Figs. 1 and 2). Miss Lyon (6) records the occurrence 
of an incomplete septum in the sporangium of S'. rupestris , and compares this 
with the similar, more complete, trabecular structure in Lepidostrobus Brownii. 
Bower suggests that the function of this may be either (a) supporting, or 
(1 b ) nutritive; probably the latter, since the structure in question occurs 
in the microsporangium only. I have failed to discover any similar cases in 
the other species of Selaginella which I have examined. 
The mechanism for the dispersal of the spores in the genus is the most 
elaborate contrivance in the whole of the Pteridophyta. It was at first 
suggested that when once the sporangium had opened, the spores were 
scattered by the wind. This hypothesis is feasible if reference be made to 
the microspores only, since they are very light, and are no doubt frequently 
blown about from place to place. The megaspores, however, are not only 
far too heavy, but are difficult of wind access, each being sunk in a depres¬ 
sion of the sporangium wall. Moreover it is made perfectly evident, on 
watching the bursting of the sporangia, that the scattering of the spores 
is not left to chance ; on the contrary, the spores are shot out with con¬ 
siderable violence by a definite mechanism, the force being so great that the 
sporangium itself may be jerked bodily to a distance of 3 to 4 cm., 
while the spores may travel much further, the megaspores from 6 to 10 cm., 
but the microspores from 1 to 1^ cm. only. It will be seen that the mega¬ 
spores are shot out further than the microspores, a fact scarcely substantiating 
the ‘ wind agency ’ theory of dispersal. Goebel ( 9 ) suggests that this is an 
adaptation for cross-fertilization, and this explanation is supported by the 
proterogynous nature of the strobilus, and also the facts that ( a ) the micro¬ 
spores germinate sooner than the megaspores of the same cone, that (b) the 
archegonia develop six weeks after the antheridia taken from the same 
cone are empty, and that (c) no embryos are obtained if the micro- and 
megaspores of the same strobilus are sown together. Goebel’s hypothesis, 
therefore, becomes not only feasible, but highly probable. 
Schwendener, Goebel, Steinbrinck, and others have done a considerable 
amount of research on the structure of the sporangium wall, with special 
reference to the mode of distribution of the spores, and the observations 
which I have recently made on the subject are entirely in agreement with 
