362 bulletin: museum of compaeative zoology. 



Furthermore, we have seen in Gonionemus that, not only different nu- 

 clei, but different parts of the same nucleus may differ in this respect; 

 for while the chromatin reticulum shows reversal, the chromatin " shell" 

 of the nucleolus does not. In every case so far described the difference 

 in behavior corresponds with a difference in morphologic conditions. 

 Thus, reversal seems always associated with a diffuse, or finely divided 

 condition of the chromatin, while maintenance of an affinity for basic 

 stains is correlated with an occurrence of the chromatin in larger masses. 

 Since reversal of staining depends upon an absorption of, and combina- 

 tion with, albumin (E. B. Wilson : 00, p. 340), it is evident, as Etlckert 

 ('94) and E. B. Wilson (: 00) have already noted, that it must be greatly 

 facilitated by such an increase in the surface area of the chromatin as is 

 afforded by a division of its substance into minute granules. It must, 

 on the other hand, be correspondingly retarded by the grouping of the 

 chromatin into larger masses. The effects of such difference in the 

 morphologic condition of the chromatin are amply sufficient to account 

 for the differences in its behavior towards stains. 



2. The Nucleolus. 



The literature dealing with the nucleolus in general has been reviewed 

 in great detail by Montgomery ('98 b ), Lubosch (:02), Rohde (: 03), and, 

 in so far as concerns chromatin bodies described by this name, by Black- 

 man (: 05). 



Very few students have devoted any attention to the coelenterate 

 nucleolus. These few, though disagreeing as to various details, are 

 agreed that the nucleolus is very intimately related to the chromatin 

 structures. Thus Pfitzner ('83) describes the nucleolus in Hydra as 

 contributing, in the prophase of mitosis, to the formation of the chro- 

 matin reticulum. Chatin ('90), on meagre evidence it is true, maintains 

 that in sponges it consists wholly of chromatin. More important, be- 

 cause more detailed, are the observations of Guenther (: 04). In Hydra, 

 according to this author, the nucleolus of somatic cells is not homoge- 

 neous, as Pfitzner ('83) supposed, but is compound, consisting of a central 

 pale staining mass surrounded by a shell of chromatin, exactly as I have 

 described it in Gonionemus. The resemblance between the two forms 

 is heightened by his description of a chromatin nucleolus in the sperma- 

 tocytes. Furthermore, he is able to trace a complete series of stages 

 from the chromosomes of the last spermatogonial mitosis to this chro- 

 matin nucleolus, showing clearly that a genetic relationship exists be- 

 tween the two. 



