Allen. — The Spermatogenesis of Poly trichum juniperinum. 275 
posterior end. It is evident from my preparations that the nucleus does 
not always begin to elongate at precisely the same stage ; for the nuclei 
of Figs. 26 (PI. XV) and 27 (PI. XVI) show no evidence of a change in shape 
at a time when the apical body has been completely separated from the 
limosphere. 
The further extension of the nucleus is the central feature of the later 
history of the androcyte. In this process the nucleus stretches out along, 
and becomes closely applied to, the blepharoplast. Usually the anterior 
beak of the nucleus increases in length (Fig. 30), the posterior end remaining 
rounded, sometimes to a relatively late period (Fig. 35). Sooner or later, 
however, the posterior end also becomes pointed. This may occur quite 
early (Figs. 28, 31), and probably always takes place before the nucleus 
has attained a length anywhere nearly equal to the circumference of the 
cell (compare Figs. 36, 38, and 39). As a rule the anterior portion of the 
nucleus continues to stretch more rapidly than the posterior portion, 
consequently the latter remains much thicker down to a very late stage 
(Figs. 39-41, 43, 44). Fig. 42 represents a condition rarely found, in which 
the posterior part of the nucleus has developed a slender prolongation nearly 
as long as that derived from the anterior part ; as a result, the nucleus in 
this instance is thickest in its middle portion. Fig. 47 shows a somewhat 
similar condition at a later period, in which, however, the posterior 
prolongation is much shorter than the anterior one. The anterior pro- 
longation of the nucleus is sometimes (Figs. 42, 43) bent into a double 
curve, as though an external resistance had been encountered in the course 
of its extension. 
Apparently the anterior end of the nucleus never extends quite to the 
anterior end of the blepharoplast (Figs. 35-38). As the elongation of the 
nucleus continues and its anterior part becomes more slender until its 
diameter is little greater than that of the blepharoplast (Figs. 39-44), 
it becomes increasingly difficult to determine exactly how far forward the 
nucleus extends. However, even at these later stages, the anterior end 
of the continuous body that is now being formed by the blepharoplast and 
nucleus is stained exactly as the blepharoplast was stained before — being, 
for example, coloured somewhat more deeply by iron-alum-haematoxylin 
than the chromatic materials of the nucleus, and sometimes retaining the 
stain when the substances that are certainly nuclear are quite decolorized ; 
it seems highly probable that this, which is to be the anterior extremity 
of the body of the antherozoid, is formed wholly from the substance of the 
blepharoplast. This conclusion is supported by the visible persistence of 
the apical body to a time when the body of the antherozoid is approaching 
maturity (Figs. 48, 49). The apical body, as we have seen, lies in contact 
with, and often flattened against, the end of the blepharoplast (Figs. 40, 41, 
44, 48, 49). If the prolongation of the nucleus were to extend to the tip 
