230 
Rickett.—Fertilization in Sphaerocarpos. 
haematoxylin was also used. This stained the chromatin and nucleolus 
very heavily—almost too heavily—but left the cytoplasm practically un¬ 
touched. When light green was used as a counterstain with the haemato¬ 
xylin, the cytoplasm became visible, but could not be seen so clearly 
as with the triple stain. 
Sections were made in a longitudinal vertical direction through the 
plants and seriated so as to include all of every plant. Only a few of these 
sections contained archegonia of the right age, and the others were 
discarded. 
Free-swimming antherozoids were prepared for study by introducing 
them into a drop of water on a slide, inverting this slide over a phial contain¬ 
ing i per cent, osmic acid, and allowing it to dry, afterwards staining in 
dilute gentian violet. 
Observations and Discussion. 
Since in Sphaerocarpos fertilization proved to be a slow process, and 
since the history was worked out from a large number of instances, it 
was found convenient to divide the process into a number of phases , each 
distinguished by some well-marked change in the male nucleus, in the 
female nucleus, or in both. The nature of these changes will be taken up 
as the phases are severally discussed. It must be borne in mind, of course, 
that these changes are in reality continuous, and that the limits of the 
phases are therefore purely arbitrary, constituting merely a method of 
handling the large amount of data. The number of cases in which fertiliza¬ 
tion was observed, with the distribution of these cases over the time covered 
by the process, and their classification into phases, is shown in Table I. 
Reference will be made later to the information deducible from such 
a statistical arrangement of the data. 
Structure of the Mature A rchegonium. 
Mature archegonia appear, as has been pointed out elsewhere 
(Rickett, 1920), just behind the growing point of the thallus. Between 
them and the group ot initial cells are other archegonia that have not yet 
reached maturity, and farther back on the thallus are others which, in the 
absence of fertilization, have disintegrated. 
The structure of the archegonium is that typical for the Liverworts. 
The neck is regularly curved, often forward, owing to the more rapid 
division of the cells on one side. 1 he venter is large and rounded, and 
consists, before fertilization, of a single layer of cells. The neck consists of 
six rows of cells, as described by Campbell (1896) and Cavers (1911). 
Hy (1884) and Gayet (1897), on the contrary, state that there are but five 
rows of cells in the archegonial neck. 
