BULLETIN OF THE UNITED STATES FLSII COMMISSION. 5 



of the body of tbe future eiiibi yo. This position of the young fishes is 

 nuiintaiiuHl as long as thej^ are attached. 



The development, as it advances, enables the young embryo within 

 the egg-membrane to finally rupture the latter immediately over the 

 back, which looks down and away from the surface to which the egg is 

 attached. When the zona or egg-membrane is ruptured (Fig. 2) the 

 young fish is, however, not set free at once, as in the case of other ad- 

 hesive ova, but remains firmly glued to the inside of the zona over a 

 part of the ventral surface of the yelk-bag. This adhesion of the yelk- 

 bag to the zona takes place over about the same area on the inside of 

 the latter as that which on the outside is adherent to the stone or other 

 object, which affords support to the whole egg and embryo. It results 

 in this way that the egg-membrane is not cast off from the embryo at 

 once, but remnants of it continue to cover the sides and lower surface 

 of the yelk some time after the embryo has ruptured the zona and com- 

 menced to respire from the surrounding water by means of its gills, 

 but while still affixed to the surface to which the eggs were originally 

 caused to adhere by the parent fishes (Figs. 1 and 2). Whether the 

 substance which causes the yelk-bag to adhere to the inside of the zona 

 is- secreted at the time of oviposition, or whether it is secreted during 

 a later stage of development has not been determined ; but it is inferred 

 that this adhesion is a secondary i^henomenon, and takes place after 

 the vitellus has been covered by the blastoderm, for the reason that the 

 latter alone is adherent. In fact, if the vitellus were primarily adher- 

 ent, the blastoderm could not grow around the vitellus and over the 

 area where the former becomes adherent to the zona radiata. 



While the embryos are still adherent, the tail is not kept constantly 

 vibrating, but the pectoral fins are kept in motion so as to keep up cur- 

 rents of water and effect the constant change of the latter, needful for 

 the respiration of the embryos. 



For a considerable time the yelk-bag is almost pyriform, with its ad- 

 herent base flattened and its upper narrowed eud in relation with the 

 embryo and its vessels and heart. Vessels are developed over the sur- 

 face of the vitellus long before hatching. With the progress of devel- 

 opment the vitellus suffers constriction (Fig. 1), so that it is divided 

 into an upper portion, which is included by the down-growing myotomes 

 of the body cavity, and a lower portion which is covered by the thinner 

 epiblastic and mesoblastic covering of the inferior i^ole of the yelk. 

 When the embryos are detached from the surface to which they adhere, 

 the free, bulbous lower portion of the yelk-sack becomes wrinkled in 

 consequence of the thin epiblastic and mesoblastic investment being 

 thrown into narrow folds, which run horizontally around the yelk. 



With the extension of the abdominal walls over the yelk, more and 

 more of the yelk is finally taken into the abdominal cavity proper, and 

 a transverse constriction around its middle is finally developed, so that 

 it becomes hour-glass shaped. The upper bulb of this yelk mass is in- 



