300 ME. G. BBOOK Ol^f SOME POINTS ITif 



completed. The nuclei in the periblast of Motella are also con- 

 siderably larger in proportion to the size of the cell than is the 

 case in TracJiinus. 



The exact nature of tbe so-called invagination-process to form 

 the hypoblast was not observed, but an optical section taken 

 when this process was well advanced would seem to confirm my 

 views of tbe origin of the hypoblast in TracJiinus. Figure 2 re- 

 presents a surface view of tbis stage, and fig. 2 a the same as 

 seen in section. My opinion at present is that the cells of the 

 periblast are pushed under the germinal disk until they cover 

 the whole floor of tbe segmentation-cavity, and that cells ab- 

 sorbed from this layer and free cells from the yolk contribute to 

 a very great extent to build up the invaginated layer, if indeed 

 it is not at first formed entirely from these sources. I have dis- 

 cussed this question more fully in a paper on the " Origin of the 

 Hypoblast " (Quart. Journ. of Micros. Sci. Jan. 1885, p. 29), and 

 it will be useless to renew the discussion here. The cells of the 

 new layer in Motella are, bowever, so mucb larger than is usual 

 in Teleostean ova that tbeir exact position is easily made out ; and 

 to my mind it seems impossible to maintain tbat cells so large 

 and well-defined in outline could be formed by an involution and 

 budding of the tiny cells of tbe germinal disk, which, under a 

 magnifying power of 100 diameters, are scarcely distinguishable. 



The cells in the segregated hypoblast in fig. 2 are very distinct 

 and quite sharply defined in outline, while with the same magni- 

 fying power in Trachinus these cells are indistinguishable. The 

 embryonic shield at this stage is also larger in proportion to the 

 size of the egg than that of Trachinus ; and when the embryo 

 begins to make its appearance, this shield occupies the greater 

 portion of a surface view of the egg. Before the keel can be 

 made out, the cells in the centre of tbe shield become smaller and 

 smaller, until, as shown in fig. 3, they can scarcely be made out 

 with an amplification of 50 diameters. It is quite evident at a 

 very early stage that tbe young Motella embryo will be very 

 long and narrow, and before the appearance of Kupfier's vesicle 

 it occupies balf the circumference of the egg. I have nothing 

 new to report in the early development of tbe embryo, but it will 

 be interesting to compare the times at which tbe different organs 

 appear with those of Trachinus. For instance, Kupffer's vesicle, 

 whicb in Trachinus appears before any protovertebrae are formed, 

 and long before the closure of the blastopore, does not make its 



