339 
Actinostrobus pyramidalis , Miq. 
a second vertical wall in each of the resulting cells, in a plane at right 
angles to the first. There is thus formed a single tier of four cells at the 
tip of the suspensor. The next division is transverse (Fig. 51 ). The sub- 
sequent development has not been followed closely, but scattered stages, 
such as those shown in Figs. 52 and 56 (microphotograph), indicate that 
the later stages are very similar to those of Pinus and other Conifers. 
Embryonal tubes are a conspicuous feature, as in many other Conifers, and 
are well shown in Fig. 56 . It is perhaps worth noting that the four cells 
formed by the first divisions in the embryo initial always remain associated, 
constituting a single embryo. 
The number of cotyledons in the mature embryo is always two, as 
noted by Hill and de Fraine (10), and in spite of the large number of 
embryos which begin to develop, no case of more than one embryo reaching 
an advanced stage has been observed. The development of the proembryo, 
and the differentiation of suspensors and embryo initials, are indicated in the 
series of diagrams in Text-fig. 3 . 
6 . Conclusions. 
The chief result of the present study has been the more detailed 
knowledge now obtained of the proembryo and embryo development, which 
has emphasized the conclusion previously reached ( 20 , 21 ), that these stages 
in the Callitrineae are radically different from the corresponding stages in 
all other Conifers. There can be no doubt that the proembryo develop- 
ment is substantially the same in Callitris and Actinostrobus , and to a less 
extent in Widdringtonia also. 
The close resemblance between Actinostrobus and Callitris in almost 
every respect is very remarkable, and tends to emphasize the differences 
between these Australasian genera on the one hand, and the African genus 
of the Callitrineae, Widdringtonia , on the other. 
In previous studies the opinion has been expressed, both by the present 
writer and by others who have investigated those Conifers with lateral 
archegonia, that this (i. e. lateral position of archegonia) is probably an 
£ ancient ’ character. The evidence was perhaps mainly the fact that they 
were associated with many other ‘ ancient 5 characters in the Araucarians ; 
in any case, the evidence was not complete, and Coulter (4) has justly 
criticized the view, and stated sound arguments for an opposite opinion. 
Land (13) and Pearson (18) had previously taken the position that this 
tendency might represent an advance towards Gnetoid conditions. Further 
study of the Callitrineae has tended to show that Coulter’s view is likely to 
be the right one, and some further evidence has been recently brought 
forward in this direction by the writer (22) from an abnormal Pine pro- 
thallus with only lateral archegonia, although opinion will doubtless differ 
