CRANGON VULGARIS. 107 



germiiial area, the protoplasm is far more abundant and 

 forms a layer of considerable thickness (ga). In sections 

 this has the appearance of a syncitiura, as I am unable to 

 discover any cell boundaries. In surface views, it is true, 

 the cells here, as earlier, seem clearly marked off from each 

 other ; but as the figure shows, these lines of demarcation 

 are but superficial and do not descend to any depth. This 

 obliteration of cell walls here may be due to the action of 

 Perenyi's fluid and I regret that this idea did not occur to 

 me at the proper time to test it, so I cannot positively 

 State that this region is really a syncitium. In the figure 

 is represented, at c', one of the belated cells which has not 

 yet joined its fellows ; the complete series of sections of 

 the egg show that it was the only one which remained be- 

 hind. In surface views it is not easy to assign limits to 

 the germinal area as it shades off insensibly into the sur- 

 rounding undifferentiated blastoderm and is merely a por- 

 tion of the surface where the protoplasm is more abundant 

 and the nuclei more numerous than in other parts. In its 

 general appearance it does not differ much at this stage 

 from üg. 8, except that the blastopore shown in that figure 

 is, of course, absent. 



The variations in the character of the segmentation and 

 the method of forming the gastrula and the germinal lay- 

 ers are so closely connected that a discussion of the seg- 

 mentation of Crangon, as compared with that of other 

 arthropods, is deferred untilthe end of the next section of 

 the present paper ; but here it is necessary to mention a 

 conflict betweeu my results and those of another observer. 



Edouard van Beneden ['70» P- 142] says : — "Chez les 

 Crangons, il se produit un fractionnement total du vitellus, 

 comme chez les Gammarus locusta, et les cellules du blas- 

 toderme resultant de ce que dans chacun des Segments, il 

 s'opere en Separation complete entre les Clements proto- 



ESSKX INST. BULLETIN, VOL. XVIII. 14 



