688 McAllister . — Nuclear Division in Tetraspora lubrica . 
phase. While not common in my material, this prophase was still frequent 
enough to be satisfactorily studied. The chromosomes are at first elongated 
rod-like structures (Figs. 9, io, and 13). They shorten and thicken until 
their diameter is probably twice that of the chromosome at the time of the 
segmentation of the spireme, while their length becomes about twice their 
final diameter (Figs. 11, 12, 14). 
The number of the chromosomes is small. Figs. 10 and 13 of nuclei 
immediately before metaphase and Fig. 19 of a polar view of a metaphase 
stage, — all made with no reference to the chromosome number, each show 
thirteen chromosomes. The chromosome count in other nuclei, not figured, 
gave twelve and thirteen as the number, and it is quite probable that 
thirteen is the correct number. 
The details of spindle formation could not be followed on account 
of the minuteness of the nuclei. At the time of the metaphases, when the 
spindle is most conspicuous it is distinctly bipolar (Figs. 15, 16, and 18). 
Areas of protoplasm which suggest kinoplasmic caps were in several cases 
observed (Fig. 13) and it is quite probable that they give rise to the spindle 
in a manner quite similar to that described for root tip mitosis. An exami- 
nation of many metaphases failed to reveal anything that could be in- 
terpreted as a centrosome or centrosphere. Neither in the first division of 
the cell nor in the second or the third division was a centrosome to be seen. 
The appearance of centrosome-like bodies in antheridia of some Bryophytes 
at the time of the formation of the male gametes suggests that centrosomes 
are to be expected at the time of the formation of swimming cells. If such 
bodies are present in Tetraspora I have been unable to stain them. Davis 
( 12 ) has described centrosome-like bodies in Pellia which are usually 
present only in the early prophases and disappear before the metaphases. 
No such evanescent structure could be identified in Tetraspora. 
The single group of chromosomes of the metaphases splits to form the 
two groups of the anaphases (Fig. 19). It is very probable that each 
chromosome splits to form the chromosomes of the anaphases, — though on 
account of their minuteness this could not be determined with certainty. 
The reconstruction of the daughter nuclei (Figs. 20, 22, 23, and 24) as 
far as can be determined is the same as in higher plants. The chromatic 
material is at first densely massed (Fig. 20). This mass becomes looser and 
a nuclear membrane is formed (Fig. 22). The separation of the chromatic 
material into smaller and smaller parts goes on till the reticulum is formed 
(Figs. 23 and 24). Although the second and third nuclear and cell-divisions 
of the vegetative cell to form the eight gametes followed the first in quick 
succession, nevertheless the daughter nuclei in each case enter a resting 
condition similar to that of the vegetative nuclei (Figs. 24 and 28). 
The partition walls are formed during the telophases by means of a 
central spindle not greatly unlike that in dividing cells in the higher plants. 
