[91] EMBRYOGRAPHY OF OSSEOUS FISHES. 545 



from tbe yelk hypoblast; the layer luj in our figures is the homologue 

 of tbe yelk hypoblast in Alosa. 



Ill the study of the yelk circulation of Goregonus albus, or Lake white- 

 fish, the vitelline vessels in optic section appear to have an inferior as 

 well as outer wall and to be connected together by a thin membrane 

 stretching between them. Can it be that this vascular membrane is 

 continuous with the heart through the thin posterior splanchnopleural 

 pericardiac membrane? It would seem as if this might be the state of 

 affairs in this species, if not in all forms. The fact that the Cuvierian 

 ducts develop iu the upper lateral portions of this membrane is greatly 

 in favor of such an interpretation. In Goregonus we may also observe 

 that there is in the living embryos a very shallow fluid space between 

 the yelk and the vascular layer between the courses of the vessels. In 

 hardened specimens of salmon embryos sections show the v^essels de- 

 pressed; this I now suspect may be due in i)art to the compression and 

 shrinkage of the outer yelk-sack under the influence of chromic acid. 

 While we cau say positively that the posterior and ventral pericardiac 

 wall does not include or cover more than a small portion of the upper 

 surface of the yelk at the time of hatching in Alosa, it is probable that 

 when the yelk is almost absorbed that it may entirely envelop it. It 

 may also be said that a marked acceleration in the development of the 

 vascular splanchnopleural yelk-layer continaous with the venous end 

 of the heart may and does probably occur if we may be guided by the 

 evidence supplied by the investigations of G^llacher on the trout. This 

 view will account for the early development of a yelk vascular sys- 

 tem iu some forms and its absence in others. It will also explain why 

 it is that in some forms an intercommunication exists between the 

 heari and segmentation cavity while it is absent in others. The re- 

 markable law of acceleration and retardation, which was first distinctly 

 formulated by the eminent palaeontologist, Professor Cope, is exemplified 

 on every hand in a study of the development of osseous fishes, and fur- 

 nishes a clew and key to much that would otherwise be obscure. 



The embryo salmon, immediately' after hatching, has an arrangement 

 of capillaries which is in the highest degree interesting. The main ves- 

 sels now consist of a great median dorsal aorta which passes backwards 

 jnst below the subnotochordal rod to the upturned hinder extremity of 

 the chorda. Here intercommunication between the aorta and caudal 

 vein is estabbshed by way of a singular caudal network of capillaries, 

 which, first of all, enipty into a sinus-like, non-contractile dilatation be- 

 fore they pour their contents into the caudal vein, which then passes 

 forwards ventrad of the aorta towards the head, dividing into the two 

 posterior cardinals above the intestine. These cardinals then give 

 off capillaries again which pass around the intestine and unite into a 

 subintestinal vein as large in diameter as one of the cardinals. The 

 suhiiitestinal vein then passes forwards over the yelk, and, bending a 

 little to the right, ends under the liver, into which it pours its blood, 

 S. Mis. 40 :]5 



