Archegonium ; and on Spermatogenesis in Poly trichum. 123 
in others they lie away from the spindle in other parts of the cell (Fig. 2 6). 
It seems very improbable, for the reasons already stated, that this division is 
of normal occurrence. If such were the case, the centrosomes would then 
be regarded as making their first appearance at the time of the penultimate 
division ; and between that and the final division become refractive and 
lose their centrosome function. Six is again the number of chromosomes 
which pass to each of the daughter nuclei during this division. 
The presence of centrosomes during the last cell-division in the 
antheridium in Polytrichum , and their persistence as blepharoplasts, is of 
interest in regard to Wilson’s statement ( 22 ) that in Mnium hornum and 
A trichum undulatum centrosomes do not occur, and that the blepharoplast 
in each case is derived directly from the nucleolus of the spermatid. This 
body, which is passed out from the nucleus, Wilson believes to have been 
phylogenetically derived from a centrosome. Ikeno regards Marchantia , in 
which he states that centrosomes are present during all cell-divisions in the 
antheridium, as representing in this respect a relatively primitive con- 
dition, while in other Bryophytes they are being gradually eliminated and 
occur only in the later cell-divisions. Miyake ( 14 ), on the other hand, fails 
to find centrosomes during the early divisions of Marchantia , and believes 
that the bodies which are found at the poles of the spindle during the last 
division are blepharoplasts and are not related in any way to centrosomes. 
Escoyez (6) in his researches on Marchantia and other Liverworts supports 
the views of Miyake. Woodburn, in a recent paper ( 23 ), states that he can 
find no evidence of the existence of centrosomes in the Liverworts which 
he investigated (. Porella , Aster ella, Marchantia , and Fegatella ), or of the 
persistence of the body which occupies the poles of the spindle in the last 
division as an individual organ in the resulting sperm-cell. This observer 
believes that the blepharoplast originates through a condensation or 
aggregation of cytoplasmic material in the sperm-cell. 
The Formation of the Spermatozoid. 
Fig. 27 shows a spermatid before the onset of the series of changes 
which transform it into a spermatozoid. This cell is at first somewhat 
polygonal, but as spermatozoid formation progresses it becomes more and 
more rounded. The relatively large nucleus is usually situated somewhat 
to one side of the cell, and the prominent blepharoplast is embedded in the 
cytoplasm towards the other side. At this stage the chromatin is held in 
rather large masses upon the reticulum, but most of this substance soon 
collects to form a central nucleolus-like body (Fig. 28). The greater portion 
of this deeply staining substance is next seen to be passed out into the 
surrounding cytoplasm (Figs. 29-31), where it collects, usually in the form 
of two drops, in the vicinity of the blepharoplast. This body may still 
be easily perceived, although it commonly becomes immersed in the 
