Some pointH in the S](eniiiito(;jenesis of Maniiiialia. 149 



network or „sohaumplasnr'. The dictyosomes may then he looked 

 upon as the condensed staining material of the cytoplasm, just as tho 

 chromosomes are the condensed staining material of the nucleus, when 

 considered as a whole. And it appears, further, that the dictyosomes are 

 directly related to the formation or increase the centrosomes in Branchi- 

 pus, and lastly, that the cijtosomes if I may coin a term, out of which 

 these dictyosomes are built have not quite the same micro-chemical 

 properties as the microsomes, but rather those of the centrosomes, as 

 indeed from their relation to these bodies we might have been led to 

 expect. Reserving this distinction between cytosomes and microsomes 

 I see no reason to reject, and a good deal to support, the supposition 

 that the centrosomes are accentuated cytosomes. ^) In this light, the 

 whole subject of the centrosomes becomes freshly interesting and 

 attractive, but I prefer to leave possible conclusions open, being con- 

 vinced that, for the present, further speculation from these data would 

 be premature. 



The conversion of the spermatids into the spermatozoa. 



When freshly formed the spermatid nuclei are, as I have already 

 stated, intensely sensitive to nuclear stains. A careful survey of the 

 cell a short time after its formation shows, besides the nucleus and 

 the collected intrazonal spindle-fibres forming the spermatid archoplasm 

 (Figs. 9, 11, 14a), one or two small granules outside the nucleus, 

 which stain as sharply as the chromatin within it (Figs. 12. 13. 14 />c'). 



As the cells giow^ older, the threads of the chromatic reticulum 

 of the nucleus (into which the chromatic "N^s degenerate) begin to 

 show a monilated appearance. This becomes more marked until there 

 are many free chromatic gi-anules, (large microsomes), within the nucleai' 

 confines (Fig. 12, 13, 14, 15). 



In all such cases it is seen that similar gi'anules exist in the 

 cell outside the nucleus, hc'j and lie free in the cell body, while in the 

 most fortunate preparations the cells appear to have been killed while 

 the granules were en route from the nucleus to the cytoplasm 



^) If this be so, we have to explain the appearance of the radiation which 

 centrosomes exhibit on some other ground than that of simple attraction. 



