106 bulletin: museum of comparative zoology. 



admitted. When these results are considered, it must be acknowledged 

 that at present our knowledge of this cell structure is not sufficiently 

 full to afford a basis for any general conclusions. I believe it is to the 

 experimental work of the future that we must look for the solution of 

 this and many similar problems in cytology, although of course results 

 obtained by such abnormal methods must be interpreted with extreme 

 caution when considering general cell processes. 



One point in Wilson's work which may prove to be of great interest is 

 the significant fact that in the eggs treated with magnesium chloride both 

 archoplasra and chromatin show a decided similarity in their behavior to 

 analogous structures in the cells of Protozoa. He calls attention to the 

 likeness existing between the structure of the aster in his material and 

 the corresponding structure described by P. Hertwig in Actinosphaerium, 

 and also shows that in one series of eggs the chromatin is aggregated in 

 a mass, and in other ways behaves in a manner very much as in Protozoa. 

 From his work it would seem that under abnormal circumstances the 

 cell shows a decided tendency to revert to the primitive t} T pe. 



The first investigators to assign any important function to the centro- 

 some were Boveri ('87 a , '88, p. 754, '88 a ) and van Beneden et Neyt 

 ('87, p. 279), whose accounts, while differing in several respects, agree 

 in the essential conclusion that this structure is the "dynamic centre" 

 of the cell. In this conclusion the observations of many later cytolo- 

 gists concur. Concerning Thalassema and Zirphaea, Griffin ('99) says : 

 " By a careful study of these processes, the impression is most strongly 

 conveyed that throughout all these stages the centrosome is the cause 

 rather than the mere expression or by-product of the aster formation. 

 This is especially clear in late anaphase, where the centrosome deserts 

 the old system, and moving to a different locality furnishes there the 

 stimulus to the formation of a new one." Phenomena very similar to 

 these have also been observed by MacFarland('97,Gasteropoda),Coe ('99, 

 Cerebratulus), and Lillie ('98, :01, TJnio). My results agree with these 

 in showing indubitably that the centrosomes are not a chance product 

 derived from the forming aster, but may be more justly said to be the 

 causal, or at least the directive, force in the formation of the astral system. 



It has been seen that the first change in the prophase of the sper- 

 matocyte concerns the centrosome. This is more marked in the large 

 type of the spermatocyte than in the smaller, but nevertheless may 

 be observed in both. This fact may well be taken as supporting the 

 hypothesis of Boveri and van Beneden that the impulse for cell divi- 

 sion is given by the centrosome. 



