BIGELOW : NUCLEAR CYCLE OF GONIONEMUS MUKBACIIII. 329 



tinguished, a fact which seems to argue for their origin from more or 

 less dissociated archoplasmic granules scattered through the cytoplasmic 

 reticulum, rather than for their direct morphological descent from either 

 centrosomal archoplasm, or spindle remnants. 



The acrosome, or perforatorium, is lirst distinguishable in the stage 

 represented in Figure 97, where it consists of a small globule lying in 

 the cytoplasm closely apposed to the nucleus at the pole opposite the 

 inner centrosome, and showing a typical archoplasmic staining reaction. 

 It has, unfortunately, been impossible to follow in detail the stages in 

 its development, but it may not be out of place to consider its probable 

 origin. As we have already seen, the remnants of the interzonal fila- 

 ments at the time of the breaking down of the interzonal bridge lie in 

 contact with the nucleus at the pole opposite the centrosome, and at 

 first sight it might seem that they are metamorphosed directly into the 

 acrosome. But, as we have also seen, they come to lie in a very different 

 position, and then dwindle and finally disappear before the appearance 

 of the acrosome, many cells (as, e. g., that shown in Figure 95) possessing 

 neither the one nor the other. The evidence, then, is wholly against 

 the direct derivation of the acrosome from the interzonal remnants. 

 Yet the evidence afforded by staining reactions — and I think that 

 in the study of the spermatid these reactions are fairly reliable, more 

 especially that to iodine — is to the effect that the acrosome is chemi- 

 cally identical with the two archoplasmic masses just mentioned as 

 lying one on either side of the axial filament. Since there is no evidence 

 in favor of the view that any other structure — nuclear substance, cen- 

 trosome, etc. — takes part in the formation of the acrosorne, I believe 

 myself justified in saying that the acrosome is of archoplasmic origin. 

 In the earliest view of it that I have seen it is a sharply outlined, 

 deeply staining, homogeneous globule lying against the nuclear mem- 

 brane (Fig. 97), and it persists in this position without any visible 

 changes in structure or outline. 



"While the foregoing modifications have been taking place in the 

 centrosome and in the archoplasmic structures, the nucleus has also 

 undergone profound changes. In early stages, previous to the division 

 of the centrosome, the chromatic substance is condensed into an irregular 

 network, while the remainder of the nuclear area is filled with dense 

 karyoplasm (Figs. 91, 92). The latter, however, shortly disappears, 

 leaving the space apparently empty except for the chromatic net (Fig. 

 90). The nucleus itself is very small, the diameter being only 2 //, but 

 during the migration of the proximal centrosome it increases considerably 



