g6 Humphrey . — The Development of 
According to Ikeno 1 the spermatid mother-cells divide diagonally, thus 
forming spermatids of triangular outline without the formation of a wall 
between them. This is true of F . longiseta and Aheura as well as other 
Liverworts thus far observed. Each one of the spermatids thus formed 
develop later into spermatozoids. Up to the division of the spermatid 
mother-cell, all divisions are at right angles or nearly so, with the spindle 
lying in the long axis of the cell. In the spindles of a number of these 
cells the number of chromosomes could be estimated, and so far as could be 
ascertained they are eight. 
In the case of the diagonal division (division of spermatid mother-cell) 
the axis of the spindle is invariably diagonal, thus dividing the cell into 
two daughter-cells, whose greater faces are parallel as in Fig. 23, PL V, 
As mentioned above, no division-wall is formed between these two 
daughter-cells, which appear to be nothing more than regular masses 
of protoplasm supplied with nuclei and surrounded at first by a delicate 
membrane, which later becomes a more evident structure. Thus inside of 
the spermatid mother-cell, which Ikeno also refers to as a c tesserale Zelle,’ 
are developed these two nucleated bodies or spermatids. 
The method of division in the spermatid mother-cell is the same 
as that described above for other cells of the antheridium. In one or 
two cases bodies were seen at or near the poles of a spindle, that, so 
far as location goes, might have been interpreted as centrosomes, but were 
in all probability protoplasmic granules ; for here again none of the nuclei 
in the spirem or prophase conditions exhibited structures such as occur 
at corresponding stages in Marchantia . The diaster phase of division 
was not observed in the mitosis of the spermatid mother-cell, but appeared 
a number of times in the younger spermatogenous cells. After the 
completion of mitosis in the spermatid mother-cell, the two daughter- 
cells (spermatids) contract, drawing away from the wall of the mother- 
cell though still preserving their original form for some length of time. 
In Marchantia polymorpha Ikeno 2 reports the disappearance of 
the centrosome at the completion of mitosis in the cells of the younger 
spermatogenous tissue, reappearing in the early spirem stage of the next 
division. While in the division of the spermatid mother-cell he figures 
a centrosome just outside of each daughter-nucleus, instead of within as in 
preceding divisions. This centrosome 3 , he states, remains unchanged up 
to the time it assumes the function of a blepharoplast. 
In Fossombronia no evidence of such a structure can be seen in the 
daughter-cells : in fact, not for some time after the contraction of the cell- 
contents of the newly formed spermatids do the blepharoplasts appear. 
These bodies may be made out with considerable ease, as they take up the 
1 loc, cit. (’ 03 ), p. 73, Fig. 17 of plate. 2 loc. cit., p. 71, Fig. 166 of plate. 
3 loc. cit., p. 86. 
