Saxton .— Contributions to the Life-History of Callitris. 559 
tiated from the surrounding cells of the nucellus, having large and lightly 
staining nuclei and dense cytoplasm, while those of the nucellus have 
smaller, deeply staining nuclei and less dense cytoplasm. 
Only a very limited number of preparations show at all clearly the 
stages following this. Neither of the reducing divisions has been seen. 
Fig. 6 indicates that the upper of the two megaspore mother-cells may 
develop, but as this represented the only preparation in which the point could 
be definitely ascertained, it is impossible to say if there is any constancy in 
the position of the functional mother-cell. This figure also shows that 
of a row of four megaspores the second from the top became functional. 
Other preparations indicate considerable variation in this respect (see Figs. 8 
and 10). 
A considerable proportion of ovules become abortive while quite young. 
The structure of the nucellus of such an ovule is shown in Fig. 7, about the 
same age as the normal ovule shown in Fig. 8. No pollen-tube is found and 
all the central tissue of the nucellus is made up of very large cells with 
scanty cytoplasm. 
4. The Female Gametophyte . The general structure of the nucellus 
after the first division of the megaspore nucleus is shown in Fig. 8. In this 
case traces of sterile megaspores can be seen above but not below the pro- 
thallus ; i. e. the lowest of the row of megaspores has become functional. 
The relative size of the pollen-tube is shown in the tip of the nucellus. 
Instances are occasionally met with where more than one megaspore 
begins to develop. Such a case is drawn in Fig. 9. The upper embryo-sac 
here contains two nuclei, the lower contains four. 
It could not be definitely ascertained whether these two megaspores 
developed from the same or different megaspore mother-cells. 
The next figure (10) shows a somewhat older sac, containing sixteen 
nuclei. In this case the three sterile megaspores are below the embryo-sac, 
and it must therefore have been the uppermost which became functional. 
At this time and throughout the development of the prothallus, the 
absence of any kind of tapetal layer is a noteworthy feature. Physio¬ 
logically its place is possibly taken by the basal part of the nucellus, the 
cells of which are always densely packed with contents ; growth of the 
nucellus takes place mainly in this basal part, as is indicated in Fig. it. This 
represents the structure of the nucellus at a time when 128 nuclei are found 
in the prothallus. The basal meristematic tissue is shaded. It consists at 
this time of a number of exceedingly regular rows of tabular cells, somewhat 
resembling the ‘ Pavement tissue ’ in Gnetum Gnemon recently described by 
Coulter ( 4 ). The upper part of this tissue is absorbed later as the chalazal 
end of the sac penetrates more deeply into the base of the nucellus, but its 
basal part persists. As far as the species available to the writer are con¬ 
cerned this has proved a constant character. Cell-formation in the pro- 
