248 
W. T. Saxton. 
(I) A very considerable number of free nuclei (32 or more) 
are formed in a central mass of cytoplasm in the archegonium ; these 
are then organised into three tiers of cells, of which the upper forms 
suspensors, the middle the embryo, and the lowest a penetrating 
and protective cap. The pro-embryo fills or nearly fills the arche¬ 
gonium. Occurs in Agathis, and, with reasonable certainty, though 
the details are not known, in Araucaria. 
(II) Walls are laid down during and following the transition 
from the four-nucleate to the eight-nucleate stage, preceded by the 
passage of four free nuclei to the base of the archegonium, the first 
two sporophyte divisions taking place at or near the centre of the 
archegonium. Subsequent divisions result in the formation of an 
upper tier of four free nuclei, and three tiers of four cells each below, 
known as rosette, suspensor and embryo tiers respectively. The 
pro-embryo only occupies a small part of the archegonium, at its 
base. This type, which is very constant and regular, occurs in Pinus 
and several related genera. 
(III) The type met with in Sciadopitys is similar to II, but 
differs in two particulars— (a) that the walls are formed at a later 
stage, and (b) that the embryo cells of the terminal tier divide to 
form a considerable group of cells (about 16) before the elongation 
of the suspensors, some of the cells of the group forming a “ second¬ 
ary suspensor.” 
(IV) In the next type the fusion nucleus before or during, or 
occasionally just after, its first division, travels to the base of the 
archegonium and there forms a tetrad of free nuclei (the passage to 
the base of the archegonium may occasionally be delayed till the 
four-nucleate stage); wall formation occurs as in type II at the 
transition from the four- to the eight-nucleate stage, but in this case 
only three “tiers” (or regions) of cells are usually formed; nor is 
the arrangement of cells by any means so constant as in type II; 
the number of cells in different tiers may be quite different, and 
may also vary from one individual to another in the same species, 
as does also the number of different planes in which the pro-embryo 
cells are arranged, the cells of one “tier” being not always, in fact 
rather seldom, in one plane; it is customary, though not universal, 
for one embryo cell to be in advance of the others at the extreme 
base of the archegonium. Although so much variation exists, there 
is no doubt that this is really a definite type of pro-embryo, and not 
merely a series of transitional forms between other types ; to convince 
oneself of this it is only necessary to compare a considerable number 
of the published figures of different genera conforming to this type, 
