278 ME. a. BEOOK ON THE 



rows of uuclei around the disk ; but the outline of cells cannot 

 be made out. These nuclei in the majority of ova are fairly uni- 

 formly distributed, though not with the geometric precision of 

 KupfFer's figure (" Beobacht. ii. d. Eatw. d. Knochenfische," 

 Schultze's Archiv, Bd. iv. Taf. xvi. fig. 1). Later on considerable 

 irregularity begins to show itself, which will be referred to in 

 its proper place. The blastodisk at this stage is almost flat on 

 the under surface, or perhaps very slightly convex, and has 

 perceptibly increased in thickness at the centre. There is no 

 material change in the blastodisk 12 hours later ; the " nuclear 

 zone," however, has altered considerably. The zone has in- 

 creased in width, the nuclei (which are very distinct) have 

 increased considerably in number, and are now most irregular 

 in their arrangement ; so much so that it is impossible to distin- 

 guish a series. The majority of nuclei are in pairs ; but here 

 and there are clusters of three and four, in some cases four 

 arranged in a chain. This is possibly a consequence of the sub- 

 division of the nuclei. About 30 hours after the first furrow is 

 formed the blastodisk begins to extend over the yolk ; it also 

 becomes thiuner at the centre, and is now concave on the under 

 surface. No trace of the segmentation-cavity is yet to be found, 

 nor does there appear to be that stratification in the cells which, 

 as described by Messrs. Kingsley and Conn, is the case in the 

 embryo of the Gunner. About 4^ hours later the first beginning 

 was noticed of the rim which ultimately forms the boundary of 

 the segmentation-cavity. It showed itself on a surface view of 

 the underside of the blastodisk as a short line rising a little 

 within its margin, this Hue gradually extending itself parallel 

 with the blastodisk margin in each direction until about 3 hours 

 later the outline was sharply and definitely marked out all round. 

 The blastodermic ring thus formed is a little broader at one point 

 whence the future embryo will develop (fig. 10). Diameter at 

 this stage '88 millim. No separation of the subjacent cells from 

 the blastodisk to form the segmentation-cavity has yet taken 

 place, and its origin will be considered in the next section. 



The so-called invagination of the hypoblast is very clearly 

 made out, in optical section, as at first a single layer of cells 

 sharply defined in the upper and lower limits of layer (tig. 11). 

 The orioin of this layer in teleostean fishes is not clearly under- 

 stood. According to most recent investigations, Henneguy 

 (Comptes Eendus, xcv. 1882, pp. 1297-1299) maintains that this 



