28() 
W. E. AGAK. 
oF tlie somatic or sperinatogouial mitoses. Further evidence 
will now be produced in favour of this conclusion, and to show 
that this ti-ansverse segmentation is potentially present in all 
the chromosotues in all parts of the body, thoug-h it is most 
marked whenever the chromosomes are particularly short. 
In any larva of Lepidosiren occasional mitoses are found 
with unusually short chromosomes, and these were found to 
be extraordinarily numerous in one lai'va of (Ti-aharn Kerr’s 
stage 81 +. They are present in probably all the tissues, but 
es[)ecially in the nervous s\'stein. 
Fig. 1 shows a metaphase in a nerve-cell from this larva. 
It is cut in three sections. It will be seen that the chromo- 
somes are very short, and that the majority of them are 
marked, ly transversely segmented. As they are all completel v 
longitudinally split also, the result is to form tetrads vei'V 
similar to those found in meiosis, but present of course in the 
full somatic number (thirt 3 ’-eight) . 'I'he scattering of the 
ch romosomes through the cell at this late stage is to be noted, 
and I believe that it is a frequent, though by no means 
invariable, characteristic of this type of mitosis that no 
equatorial plate is formed. 
Figs. 2 and 3 are small fragments of two metaphases, also 
in nerve-cells, showing a few tetrads produced in the same 
way as those in Fig. 1. 
In all these figures it is noticeable that certain of the 
chromosomes ai e divided by their transverse joints into very 
unequal portions. 
Such figures as these could be multiplied indefinitely from 
the same larva, and from others. 
The question now arises, at what stage does the transAmrse 
segmentation appear? As a rule, it seems not to do so till 
the metaphase, though thei'e are numerous exceptions to this, 
as can be seen in certain chromosomes in fig. 8. Fig. 4 is a 
|)i‘ophase of a somatic mitosis from the same larva as that 
from which figs. 1, 2 aiid 8 are taken. This prophase would 
undoubtedly have resulted in very short, and hence trans- 
versely constricted metaphase chromosomes, but nevertheless, 
no certain segmentation is yet visible in them. 
