Actinostrobus pyramidalis , Miq. 329 
isolated example of an archegonium showing two nuclei in the normal 
positions of oosphere and ventral canal nuclei, respectively. Nearly all the 
other archegonia of this prothallus have the structure shown in Fig. 14. 
This shows apparently a late stage of a division which might be interpreted 
as that cutting off the ventral canal nucleus, but it has certain peculiarities 
not characteristic of that division. In the first place, the number of chromo- 
somes at each pole appears to be less in every case than the normal reduced 
number (eight) ; secondly, the upper group of chromosomes is very much 
smaller than the lower in every case, and seems to consist of only one or 
two chromosomes. The most reasonable interpretation seems to be that 
the whole ovule is an abnormal one, and that, instead of forming a normal 
ventral nucleus, the chromosomes of the central nucleus have separated out 
and discharged one or two of their number from the rest, which have then 
remained as chromosomes instead of becoming reorganized into a nucleus. 
If this explanation is more or less correct, it is evident that the occurrence 
of an apparently normally organized ventral nucleus in a neighbouring 
archegonium can scarcely be regarded as good evidence that such a body 
is formed in normal prothalli. The third case is that of an archegonium in 
which fertilization has already taken place, but which contains what can 
only be regarded as a ventral canal cell. If such a cell were normally 
organized, or if (as would be far more probable) a nucleus were formed 
which persisted for even a few hours, it is reasonably certain that it could 
not have been missed in normal preparations. A difference in the structure, 
size, and position of the archegonium nucleus is, however, apparent between 
the stages shown in Figs. 10 and 12. This is interpreted as being due to 
the cutting off, between the two stages, of an ephemeral ventral canal 
nucleus, which completely and very rapidly disorganizes. Were it not, 
however, for parallel evidence in a related genus ( W iddringionia ), one 
would be much more inclined to come to the conclusion that no ventral 
nucleus is normally cut off, as is said to be probably the case in Torreya 
taxifolia (6). 1 It is also worth noting that no ventral nucleus was 
identified in Calliiris , which is much more nearly related to Actinostrobus 
than is W iddringionia. The series obtained in Callitris was not, however, 
such a close one as that in Actinostrobus. A possible explanation, which is 
very frequently overlooked by investigators, of the persistent absence of 
a certain stage in a particular plant, is that the stage concerned only occurs 
at certain hours of the day or night, at which collections were not made. 2 
In a recent paper by Lutman ( 15 ) on Closterium , it was shown that nuclear 
and cell division in that genus always occurred at night. I am not aware 
1 The case of Torreya taxifolia , however, is emphatically £ not proven 
2 Since the above was written I see that Burlingame, working on Araucaria (Bot. Gaz., Feb., 
I 9 I 3, p. 99), suggests a similar explanation of the difficulty experienced in securing certain stages 
of development. 
