Alien . — The Spermatogenesis of Poly trichum juniperinum. 273 
By the time the growing blepharoplast has reached a length equal to 
from two-thirds to three-fourths of the circumference of the cell (Figs. 14-20) 
its position seems to be more definitely fixed. It now lies close to, and for 
the most part in contact with, the plasma membrane, with one end near, 
but apparently not yet touching, the limosphere. This end of the blepharo- 
plast will be referred to hereafter as its anterior end because it is to be con- 
cerned in forming the anterior portion of the body of the antherozoid. 
Arens ( 1907 ) finds that in both Poly trichum and Mnium the anterior end 
(constituting the whole of what he calls the blepharoplast) lies usually at 
that side of the androcyte which is towards the apex of the antheridium. 
I have not found any such regularity in the position of the blepharoplast in 
my material, nor has Wilson ( 1911 ) in his study of the androcytes of Mnium . 
In Figs. 14, 15, 1 6, and 19, only that part of the blepharoplast is shown 
which lies on the side of the androcyte turned towards the observer ; in 
Figs. 17 and 18 the blepharoplast is shown throughout most or all of its 
length. The cell represented in Fig. 20 is unusual in the fact that the 
blepharoplast is in contact throughout the greater part of its length with the 
nuclear membrane and so at some distance from the plasma membrane. 
Apparently it is at about this time that the cilia first appear. At any rate, 
the youngest cilia-bearing cells that I have seen are those shown in Figs. 16 
and 19. In each case the single cilium that is visible seems to be attached 
to the blepharoplast somewhat behind the anterior end of the latter. It 
should be noted that the methods used in staining and dehydrating the 
sections containing these cells were such as to leave little or no colour in 
structures so slender as the cilia ; and also that in sectioning the material the 
cilia of any particular androcyte are likely to have been in whole or in part 
cut away. These facts doubtless explain why cilia are often not visible, 
even in the cases of cells much older than those shown in Figs. 16 and 19 ; 
and it is quite possible that the cilia begin their growth at a period earlier than 
that at which I have first seen them. 
In the stages now under discussion, a dark-staining cytoplasmic granule, 
much smaller than the limosphere, is often conspicuous. Sometimes this 
granule seems to lie in a vacuole (Figs. 19, 25) ; sometimes it is elongated 
(Fig. 16) ; sometimes there are two similar granules (Fig. 18), of which one 
may be in a vacuole (Fig. 24, lowermost cell) ; sometimes (Figs. 14, 15) no 
such definite body is to be seen. Thus appearances are too variable at this 
time to justify any general statement concerning these bodies ; but it is not 
improbable that one or more of them may correspond to a similar though 
larger body which at a somewhat later period seems to be a nearly or quite 
constant constituent of the androcyte. 
The nucleus, which as a rule has remained thus far in the central 
region of the cell, now moves towards the periphery on that side at which 
the middle portion of the blepharoplast lies (Fig. 21). The nucleus becomes 
