98 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 



molluscs. In arriving at this latter view the criterion invariably 

 used is the appearance and behavior of the elements during the two 

 mitoses. Bat during the metaphase the chromosomes are often so com- 

 pact that the planes of the cleavage shown in the prophase are entirely 

 obliterated, and therefore the manner of division cannot be determined 

 with certainty. An example of the likelihood of misinterpretation of the 

 nature of these divisions is shown by Griffin ('99) in Thalassema. Here 

 the first division is evidently longitudinal, and upon superficial observa- 

 tion the second also appears to be of the same nature. But when the 

 phenomena observed in the prophase are considered, it is evident that 

 this cannot be true, since an indubitable transverse cleavage was to be 

 seen at that stage, and upon further examination the first impression is 

 shown to be false, for the second division is in reality a reduction divi- 

 sion. Thus I believe it is to the prophase of the first spermatocyte that 

 we should look for the evidence upon which to establish our conclusions 

 regarding the sequence of the divisions. In this I agree with McClung 

 (:00), who says, " Too much importance cannot be laid upon the neces- 

 sity for a thorough understanding of the early formative periods in the 

 history of the first spermatocyte chromosomes." In all of the investiga- 

 tions with which I am acquainted it has been shown that the longitudinal 

 cleavage is the one first made manifest in the prophase. I believe it is 

 but logical to conclude that the longitudinal division is completed by the 

 first spermatoc\ r te mitosis, especially since this has been shown to be the 

 case in a great number of cells. Of course it is possible that this process 

 varies in different animals, but it is not probable, for if the sequence of 

 the actual divisions varies, we should naturally expect the prophase 

 phenomena to vary in like manner. As I have already said, the lon- 

 gitudinal splitting of the chromatin segments invariably occurs first. 

 When these facts are all considered, I believe one is justified in believing 

 that in all Invertebrata the first maturation division is an equation 

 division, the second a reduction division. 



2. Cytoplasmic Structures. 



A. Centrosome and Centrospliere. 



During the last decade no structure of the cell has been the subject of 

 more controversy among cytologists than the centrosome. This is not 

 extraordinary when we consider its extremely small size, its great varia- 

 tion in form, its apparent difference in different classes of cells and even 

 in different individuals of the same general class, and the striking irreg- 



