438 
IOWA ACADEMY OF SCIENCE Vol. XXIV, 1917 
suggests the circular cup of Marchantia. The superficial origin 
of the gemmae is thus perfectly clear.” 
At the point where the posterior part of the Blasia cupule 
begins to increase all resemblance to Marchantia ceases (7, 8, 
9, figure 86). The tissue resulting from the accelerated growth 
in Blasia covers the depression completely leaving only a small 
opening at the anterior end (10 and 11, figure 86). Because of 
this method of growth the resulting cavity is elongate and flask- 
shaped. Papillate cells are now found pushing out from all 
sides of the cavity and are soon cut off by transverse walls. 
These are differentiated either into mucilage hairs, similar to 
those' on the thallus, or into the initials of the gemmae (12, fig- 
ure 87). Further division of the gemma initials is carried on 
first by transverse walls and later by vertical walls, so that the 
mature gemma is composed of a mass of from four to twelve 
cells, resembling the anther idium in its younger stages. 
When the first gemmae are mature, accelerated growth in the 
margin cells of the walls about the opening, forces the edges 
upward forming a chimney-like tube at the anterior end of the 
cavity (13, figure 87). This tube attains a length of 1 to 2 
mm. and varies from two to four cells in thickness. At the 
apex the edges flare outward slightly, giving the tube a bell- 
shaped opening. 
Most of the cupules were found on antheridial plants, although 
in several instances they were found on archegonial plants, 
where the arehegonia had not been fertilized. The cupules ap- 
pear later in the life cycle than do the sex organs. Leitgeb 10 
held that they were antheridial pockets, for in one instance he 
found a half grown anther idium at the posterior end of the 
cupule. He tried to find other stages but was not successful. 
Cavers 2 thought the gemma receptacles were modified archegonial 
receptacles since arehegonia. were sometimes found in them. 
The writer is inclined to think that the cupules are especially 
formed for gemma production and are not modified sex organ 
receptacles. It is true that the development of thallus tissue 
is parallel to the development of the sex-organ covering, but if 
these cupules were degenerate antheridial receptacles it seems 
that the tissue development would cease when it had enclosed 
the depression, instead of elongating to form the long necklike 
extension. 
