138 Wager.— Observations on the Structure 
satisfactory. I have only seen it under the most careful 
illumination with an apochromatic objective. It is indicated 
in Fig. 9, but is a little too strongly shaded. 
The nuclei next divide into two groups of threads which 
separate from each other to form two daughter nuclei 
(Fig- 2, 5 )- 
Simultaneously with these changes in the nu,clei of the 
oogonium, similar changes have been taking place in the 
antheridium, the nuclei of which divide up in exactly 
the same manner and at the same time (Fig. 8, a). The 
changes in the oogonium and antheridium go on very regu- 
larly together, but I have not been able to observe a distinct 
separation of the protoplasm of the antheridium into gono- 
plasm and periplasm at this stage. 
The arrangement of the nuclei in the oogonium now 
becomes somewhat irregular. Further divisions of the nuclei 
take place; the nuclei at each division becoming smaller 
(Figs. 10 and 11). A small number of these then pass along 
the protoplasmic strands towards the centre of the oogonium 
(Figs. 10 and 11, n). It would be very interesting to find 
that one portion of each of the original nuclei passed to the 
centre in this manner, but, although I have spent a con- 
siderable time on this one point, I have never been able 
to observe more than three nuclei in such a position ; I am, 
moreover, somewhat doubtful about the third nucleus, and 
can only assert positively that two nuclei pass towards the 
centre. 
While these two nuclei are passing along the protoplasmic 
strands towards the centre, the cell-wall of the oosphere 
begins to form as a delicate, slightly granular layer, separating 
the small quantity of protoplasm in the centre with its two 
nuclei from the denser and more copious periplasm with its 
numerous nuclei outside (Figs. 11 and 12, 0 ). The wall of 
the oosphere is undoubtedly formed from the periplasm, 
the nuclei presumably taking a large share in the work. The 
function of the periplasmic nuclei is thus seen to be a very 
distinct one. They divide up rapidly, so fcr as can be seen, 
