12 BOTANICAL GAZETTE [JULY 
THE ARCHEGONIUM. 
About April 1, 1903, the archegonium initials were first observed. 
They are the pyramidal form common to most gymnosperms. Two 
is the usual number, one is occasional, and three are rare. The 
primary neck-cell is quickly cut off after the initial becomes apparent 
(jig. 36), and almost immediately divides periclinally (fig. 37). Other 
periclinal walls follow (jig. 4o), sometimes as many as four or five 
tiers being cut off before anticlinal walls appear; and as many as 
eight tiers of neck cells have been observed. Each tier divides anti- 
clinally into four cells; later there may be six or eight in a tier, also 
the walls which come in later are no longer truly periclinal, thus 
giving the neck a somewhat irregular appearance (fig. 41). Thirty-two 
is probably the minimum number of cells, but it may go much higher. 
Fig. 39 shows a cross section of the neck 40 above the top of the 
central cell. Of all gymnosperms Ephedra has the longest-necked 
archegonium. This may be due to the fact that the archegonial end 
of the gametophyte is freely exposed to the air. 
Simultaneously with the appearance of the archegonium initials, 
a change is observable in the nucellus. Traces of disorganization 
become visible at the tip of the nucellus and gradually proceed down- 
ward, so that by the time the ventral nucleus is cut off, the cells at 
the apex of the nucellus have completely disappeared, leaving a 
pollen-chamber shaped like the frustrum of an inverted cone. The 
pollen-grains are thus enabled to come in direct contact with the gamet- 
ophyte and the necks of the archegonia. So far as has been reported, 
Ephedra is the only gymnosperm having any part of the gametophyte 
exposed freely to the air, except in the case of Cycas circinalis, 
where, according to Warming (77), if fertilization does not occur, 
the gametophyte continues to grow, ultimately bursting out through 
the micropyle and developing chlorophyll on exposure to light. 
Strasburger’s figures show a pollen-chamber in E. altissima, but not 
in E. campylopoda; and Jaccard finds one present in E. helvetica. 
In Cycadales and Ginkgoales the pollen-chamber, formed by dis- 
integration of the cells of the nucellar beak, is a conspicuous feature. 
In Pinus Laricio it is so small as to escape notice in most instances, 
while in Thuja occidentalis the tip of the nucellus at the time of pol- 
lination is an expanded stigma-like surface. 
i 
