Oogenesis & Embryogeny in Ephedra dlstachya. 131 
liquid by that time was entirely dried up. This method by which 
the pollen-grains gain an entrance within the integument is also 
suggested by Dr. Lotsy in the case of Gnetum gnenion (4). 
Both the latter investigator and M. Jaccard, who observed the 
living plants, consider that the pollen is chiefly wind-borne, hut that 
occasionally insects may play a part in pollination. This view 
corresponds with the fact that usually the grains in the chamber of 
E. distachya are few in number, but in some cases it is crowded 
with them. 
II. —The Archegonium. 
The archegonial initials which at first are only distinguishable 
from the neighbouring cells by their larger size, arise usually from 
the peripheral layer of the prothallium, hut sometimes from the 
second or third layer. Their number varies from one to eight, five 
or six fully developed archegonia being frequently found in a single 
ovule. When as many as eight occur, some of them usually lie 
below the fully developed ones and possess no necks. Some such 
cases are represented in Fig. 5 ; possibly each pair of archegonia 
arises from a single initial. The first division of the nucleus to 
form the primary neck-cell and central cell (Fig. 4) appears to have 
occurred in the ovules gathered in Brittany about May 24th. 
Divisions in the neck-cell follow rapidly, first periclinally, then anti- 
clinally, until sometimes as many as eight tiers are formed. After 
that, irregular multiplication of the cells occurs, until the arche¬ 
gonium is ready for fertilization, when degeneration commences. 
The length and complexity of the full-grown neck is well shown in 
Fig. 6, where the shaded masses represent darkly staining products 
due to the breaking down of the cells ; the absence of pollen-grains 
in the chamber above, and the undisturbed condition of the 
necks, indicate that the ovule is unpollinated. 
The spindle-shaped central cell of the archegonium grows to a 
length of about '1 mm.; as growth proceeds, the cytoplasm, which 
at first is very vacuolate, becomes much denser, particularly in the 
basal part. The nucleus lies close below the neck-cells until shortly 
before fertilization, when it begins to move towards the middle of 
the cell, and on its way divides into egg nucleus and ventral canal 
nucleus. In no case was a spindle found, hut in one archegonium 
the separation of the two nuclei was incomplete, and a curious 
condition was presented which gives rise to the suspicion that no 
spindle is formed. As can he seen from Fig. 7 the two nuclei, 
although at a considerable distance from one another, are connected 
by a row of chromatin masses running through a break in the 
