REVIEW. 85 



meshes of which surround the rapidly multiplying " globes gene- 

 rateurs/' until each of the latter is enclosed in a polygonal area. 

 The process of division still continues, until, at last, the network 

 ceases to be distinguishable, and the egg is covered with a pave- 

 ment of closely fitting polygonal blastomeres. The food yolk, in 

 the meantime, has become aggregated into pyramidal masses 

 (Fig. l,y. y. woodcut), the base of each pyramid corresponding 

 accurately with one of the division masses, and its apex pointing 

 towards, but not quite reaching, the centre of the egg. 



At this stage Reicheubach's observations begin. He confirms 

 LerebouUet's discovery of the yolk pyramids, and shows that, in 

 addition to the finely granular substance of which they are com- 

 posed, there are two other deutoplasmic constituents, globular 

 fatty bodies, and rounded vacuolated masses — the white yolk 

 elements (fig. \,w. i/.) — which disappears at an early period. He 

 states, however, that when the pyramids are first formed they are 

 completely surrounded with protoplasm, which, as development 

 proceeds, retreats to the surface and there forms the pavement- 

 like cells of the blastoderm. The formation of the pyramids is 

 evidently a process of yolk-division, the egg, in this stage, being 

 a modified morula (perimorula), and the withdrawal of the proto- 

 plasm to the surface of the egg being quite analogous to the forma- 

 tion of a cleavage cavity and the production of a periblastula. 



At first the protoplasm forms merely the bases of the yolk 

 pyramids, and is only distinguished from them by difi'erence of 

 texture, and by the greater readiness with which it takes up 

 colouring matters. Before long, however, the protaplasmic base 

 of each pyramid is completely separated ofi' ae a distinct pave- 

 ment cell. 



This account of the appearances observed by Eeichenbach 

 seems clear enough as far as it goes, but then one is led to ask, 

 what is the meaning of the curious and repeated ebb and flow of 

 formative yolk observed by Lereboullet ? It is, at any rate, 

 certain that the Crayfish's es,^ does not undergo a regular divi- 

 sion into two, four, eight, Src, blastomeres, like that of Penseus 

 or Palsemon, and it seems probable that the process of the 

 formation of the blastoderm is similar to that observed by Ed. 

 van Beneden in Gammarus, in which '' a number of isolated cells 

 (exactly comparable in origin to the Cephalpod's autoplasts) 

 rise to the surface of the yolk and then proceed to divide, and so 

 form a complete Perimorula.^' 



Another point, as to which no explanation is offered, is the 

 presence in the centre of the t^^'g, at this stage, of a sharply 

 defined globular body (fig. 1), of considerable size, and containing 

 vacuoles and fatty yolk elements. There is no evidence as to the 

 origin of this anomalous structure, nor has the time or manner of 



