344 
F. Cavers. 
most species of Anthoceros) a group, of these organs occurs in a 
closed cavity in the upper tissue of the thallus, roofed over by a 
double layer of cells, this roofing tissue being later ruptured so 
that the antheridia are left in a cup-like depression. According to 
most recent writers, the antheridium, or group of antheridia, arises 
from a single superficial cell which divides first by a transverse 
wall into an inner and outer cell, the latter dividing again by a 
tangential wall to form the beginning of the roof, while the inner 
cell divides longitudinally into four parallel cells, which either 
remain coherent and develop into a single antheridium, or else 
separate and form four antheridia. Apart from their endogenous 
origin, the antheridia agree in structure with those of the Sphaero- 
carpales. The number of antheridia in each cavity varies a good 
deal in different species of Anthoceros, and in different individuals; 
there may be as many as nine or ten. As first described by 
Leitgeb (23), exogenous antheridia sometimes occur in Anthoceros, 
evidently through an inversion of the normal course of develop¬ 
ment— i.e., the upper half of the primary cell gives rise to the body 
of the antheridium and the lower to the stalk. Frau Lampa (18) has 
described, as exogenous antheridia what appear to be in reality the 
early stages of the tubers found in Anthoceros dichotomus. Though 
the occurrence of exogenous antheridia is apparently a rare 
abnormality in Anthoceros, it probably represents a reversion to a 
primitive condition, and it would be interesting to ascertain 
whether it occurs in other species of Anthoceros besides A. Uevis, 
and in the remaining genera of the group. 
The archegonia of the Anthocerotales are not only sunk in, 
but are from the beginning histologically continuous with, the 
tissues of the gametophyte. However, the cell-divisions that occur 
in the developing archegonium are exactly the same as in the other 
Liverworts; the mother-cell does not project from the surface, but 
it undergoes the usual three primary intersecting divisions, cutting 
off the peripheral cells (which in other Bryophytes form the venter 
and neck) from the central cell which divides to form the 
usual axial series of the Liverwort archegonium (cover-cell, 
neck-canal-cells, ventral-canal-cell, oosphere). The cover-cell- 
group often projects slightly, and it would be of great interest if 
cases were discovered in which more or less of the neck is free. 
As it is, the difference between the Anthocerotean archegonium 
and that of other Bryophytes is merely one of degree, and the gap 
is to some extent bridged over by Aneura, where the venter ie 
