574 L. DONCASTER. 
together. The two outer nuclei move slowly to the edge of 
the egg and often become flattened together, and are soon 
followed by the inner nucleus, so that all three come to lie 
together at the outer edge of the polar protoplasm. From 
this stage onward there appears to be some variation in the 
course of events. Most commonly they all gradually degene- 
rate, becoming crowded together and shrunken so as to be 
hardly distinguishable apart, and at a stage later they appear 
as an indefinite faintly-stained mass, which is not traceable 
further (figs. 16, 17). In other cases, before this happens, 
the chromatin of at least the two inner nuclei becomes aggre- 
gated in the centre of each, as if they were preparing to 
break up into chromosomes, and in a feweggs this actually takes 
place, and one finds one or two irregular groups of chromosomes 
in the polar protoplasm. In no case have J found any certain 
evidence of actual conjugation of nuclei such as I have de- 
scribed in N. ribesii, but some eggs suggest that all three 
polar nuclei may come together and break up to form a common 
mass of chromosomes, as has been described by Henking in 
Pieris brassice (5, p. 545). Such an irregular mass of 
chromosomes is found in some eggs, while others of the same 
batch show the nuclei degenerating as described above. In 
a few eggs two of the polar nuclei, instead of degenerating at 
once seem to remain at the edge of the egg as large rather 
indefinite bodies containing numerous chromatin granules ; 
sometimes they have the appearance of irregular mitotic 
figures, and it seems probable that such nuclei might develop 
into the structure described below, which in a few cases has 
been observed at a rather later stage (fig. 18, a and b). 
Usually, in eggs preserved more than three hours after 
being laid, all traces of the polar nuclei had disappeared, but 
one lot, all of which were laid on one day, and almost certainly 
by the same female, differed from others in this respect. 
These eggs are of ages between three and six hours, and in 
some of them, but not in all, there is a conspicuous compact 
mass of chromosomes rather deep in the polar protoplasm. 
In some it is so compact that it looks like a large, rather ab- 
