204 
F. Cavers. 
which are recognisable (both in the capsule-wall and the columella) 
before the capsule is ripe. Thus each of the four valves of the 
capsule-wall has attached to its tip a quarter of the columella, with 
its tuft of fixed elaters, and as the four valves bend outwards each 
of the four masses of spores and elaters moves into a vertical 
position on the valve, and the spores, together with the loose or free 
elaters, are actively flicked outwards. Hence the columella, after 
splitting into quarters, brings about the active dispersal of the 
spores. 
Metzgeria differs from the other three genera in having 
spherical instead of ovoid or cylindrical capsules, but the structure 
is very much the same as in Aneura ; the elaterophore is not quite 
so well developed, and its cells have only here and there irregular 
thickenings on their walls, instead of the well developed half-rings 
found in the larger elaterophore of Aneura. 
In Podomitrium and Umbraculum the capsule-wall consists of 
an outer layer of large nearly cubical cells, and one or two inner 
layers of elongated and flattened cells. According to Andreas (1), 
these inner cells are continuous with the apical pad or cap, lying 
below the layer of large outer cells, at the top of the capsule, and 
Goebel (26) therefore distinguishes this cap from the columella of 
Aneura and Metzgeria as being of different origin, comparing it to 
the apical thickening of the capsule-wall found in various Marchan- 
tiales. In the latter group, however, the apical cap cannot always 
be interpreted as arising by periclinal division of the otherwise 
single-layered capsule-wall, and it is not easy to determine the exact 
origin of the cap. In Fegatella and Reboulia, according to the 
writer’s investigation of these genera (10, 11), this cap is derived 
from the archesporium, the apical portion of which exhibits sterili¬ 
sation comparable to that seen in Aneura. In any case, however, 
the capsule-wall in Marchantiales is not differentiated until a 
relatively late stage ; that is, the separation of the archesporium is 
not determined by the first periclinal divisions in the young capsule. 
In Uinbraculuin (Hy menophy turn) flab ell at um the cap, or elaterophore, 
splits into four portions, as in Aneura and Metzgeria, one remaining 
attached to the tip of each valve of the dehisced capsule and bearing 
a number of fixed elaters. In Podomitrium (Hy menophy turn ) 
phyllanthus, however, the capsule dehisces by from two to four slits, 
but these do not extend to the apex, where the valves are coherent. 
In Umbraculum the large outer cells of the capsule-wall have 
localised but indefinite band-like thickenings on the radial cell 
