364 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 



which show all the staining reactions typical of chromatin (p. 301), just 

 as Eohde contends. 



Accepting this view, as I think we must, we at once reach a satisfac- 

 tory explanation for the fact that, while the somatic and spermatogonial 

 nucleoli of Gonionemus are compound, the nucleoli of the spermato- 

 cytes are pure chromatin structures. Compound nucleoli occur in 

 this animal only in cells — e. g., somatic cells and spermatogonia — which 

 divide at such long intervals that there is an opportunity for the forma- 

 tion, hy secretion or otherwise, of plasmatic substance within the chro- 

 matin nucleolus. In the spermatocytes, however, the interval between 

 divisions is too short to allow of this alteration, so that the purely chro- 

 matic condition of the nucleolus persists throughout the short vegetative 

 period. Such chromatin nucleoli are, then, merely young stages of com- 

 pound nucleoli and have no relation whatever to the structures in insect 

 germ cells which have sometimes passed under this name (Paulmier, 

 '99), but which are in reality specialized chromosomes. (See McClung, 

 :02% : 02 b , :05 ; Blackmail, :05 ; and E. B. Wilson, :05, : 06.) 



Although my observations on Gonionemus strongly support Rohde's 

 general conclusions, there is one point on which the evidence seems to 

 me to oppose him. This is on the relationship between the chromatin 

 and plasmatic substances of the nucleolus. Rohde believes that this 

 relation closely resembles that between oxy- and basiohromatin, and 

 that the alteration by which plasmatic substance is formed from chro- 

 matin closely resembles that by which oxychromatin is formed from 

 basichromatin. I believe, however, that in propounding this theory he 

 has laid too much stress on the undoubted resemblance of the staining 

 reactions, and has overlooked the much more essential evidence resting 

 on the physiologic properties of plasmosome substance and oxychro- 

 matin. The latter substance, as we have seen (p. 360), is neither more 

 nor less than chromatin in which the nucleinic acid is combined, for the 

 time being, with an excess of albumin. But this condition is not neces- 

 sarily permanent, as Heidenhain ('92, '94) demonstrated. Plasmosome 

 substance, however, though a modification of chromatin, is a permanent 

 one, for it is not capable of regaining its original characteristics. Further- 

 more, while oxychromatin takes part in the formation of the chromo- 

 somes, plasmosome substance appears never to do so, nor to have any 

 functional activity in the nucleus. Instead, then, of being, like oxychro- 

 matin, a temporary state of chromatin, it is, I believe, a by-product, 

 the nature of which is still unknown. 



