498 EEPORT OF COMMISSIONER OF FISH AND FISHERIES. [44] 



to spread out and grow over tlie surface of the yelk sphere underlying 

 it. The blastoderm in fact is molded upon the yelk sphere as a hollow 

 spherical membrane, which finally includes the latter. It grows over 

 the yelk by what is called epiboly, the yelk itself remaining passive. 

 The rim r of the blastoderm shown in Figs. 20, 21, 22, and 23, PI. lY, 

 moves progressively toward the naked pole of the yelk, and in its prog- 

 ress more and more of the yelk sphere is covered up. As long as the 

 rim is spreading over the upper pole of the yelk its circumference con- 

 tinues to increase; as soon as it has passed the greatest diameter of the 

 yelk, as shown in Fig. 22, it begins to diminish in circumference. This 

 diminution of the diameter of the rim of the blastoderm continues until 

 it finally closes at the exposed pole of the yelk. At the moment of 

 closure, the free margin of the rim presents a wrinkled appearance, the 

 wrinkles radiating from the center of the small pore which remains 

 up to this time. After the closure of the blastoderm the radial wrink- 

 ling disappears. The pore which remains up to within a short time of 

 the closure, is the yelk blastopore of authors. The rim which has closed 

 at this point forms a discoidal plate of cells continuous with and form- 

 ing a part of the tail of the embryo. The cellular disk developed by the 

 closure of the rim may be called the caudal plate. It takes a share in 

 the development of the tail of the embryo and also in the muscle plates, 

 neurula, and hind-gut, for the whole of the substance of the rim is ap- 

 propriated in building up the caudal extremity of the body of the young 

 fish. In Figs. 24 and 25, pi. Y, I have represented the appearance of 

 the tail of the embryo cod on the eve of the closure of the blastoderm 

 at the pore bl. Other representations of its appearance are given in 

 Figs. 28 a and 29 a, 29 b, and 30 a and 30 b, the last four being side 

 views. These figures also show, in three instances, the relation of the 

 problematical vesicle Jcv first described by Kupffer, to the blastopore. 

 Their identity is difficult to follow, and I now doubt whether it is any- 

 thing more than an evanescent structure which has nothing to do with 

 the development of the urinary vesicle or the anal end of the intestine. 

 It is at any rate apparently inconstant in position, and in some fishes 

 evanescent and of temporary importance. Sometimes it appears to be 

 involved in cells, at other times it is clearly surrounded below by the 

 hbmogeneous 3'elk membrane or hypoblast alone. In Fig. 26 the blas- 

 topore of the yelk is also shown with Kupffer's vesicle lying below the 

 blastoderm as a lenticular vacuole. 



As the blastoderm of the cod's egg has spread over the yelk and in- 

 cluded it, the neurula has also further developed; the most marked 

 feature of its advancing evolution being the increased ventral promi- 

 nence of its keel or carina, as shown at cv in Fig. 22. At the same time 

 it also becomes more prominent dorsally, as is shown by the same fig- 

 ure, the fore or cephalic end of the body of the embrjo is now pro- 

 nounced in outline, and the principal paired sensory appendages of the 

 neurula become more prominent. The oii\y sensory appendages of the 



