108 THE DEVELOPMENT OF 



plasmiqiies et les elements nutritives du vitellus." Though 

 not referred to in the text, his figure 20 on plate X was 

 apparently introduced to illustrate this point. It, however, 

 does not do so, for it does not represent the central portion 

 of the egg as divided, but can readily be interpreted to 

 agree with the opinion here maintained. Van Benedeu 

 cut no sections, but depended on surface views for his re- 

 sults. Here I believe is the cause of our difFerence, for I 

 can hardly regard it as the result of our having studied 

 diflerent species, since Ci^angon vulgaris is by far the 

 most abundant Crangonid on the shores of Europe. In 

 surface views the furrows of all crustacean segmentation 

 seem deeper than they actually are and this I am confident 

 led him into error. I may remark in passing that, to my 

 mind, Van Beneden's statement (?. c.) that Gammarus lo- 

 custa has a total, while a congeneric form has a partial, 

 segmentation needs confirmation, as the point cannot be 

 settled by surface Observation. The illustrations given 

 by Van Beneden of the segmented egg of Gammarus lo- 

 custa certainly do not prove his point. 



As to the presence or absence of karyokinesis in the 

 segmentation and cell division of Crangon, my observa- 

 tious are not conclusive. I have not had the lensesneces- 

 säry for a careful study of the subject, but even in the 

 large nuclei of the earlier stages of segmentation as well 

 as later in the large, rapidly-dividing, endodermal nuclei, 

 I have not seen anything which I could Interpret as relating 

 in any way to karyokinesis, although the nuclear reticuhim 

 was clearly visible under my highest objective (Hartnack, 

 viii). Under the circumstances, I. am inclined to believe 

 that the cell division is direct. Mayer in Eupagurus ('77) 

 does not state whether a spindle metamorphosis of the nuclei 

 occurred, but like myself he savv elongate nuclei and two 

 nuclei in a cell. In fact, I do not recall a single statement 



