294 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



The great bulk of the body is therefore inclosed between the epiblast 

 and liypobliist; the segmentation cavity extends in reality all round the 

 embryo's body up to the point where the mesoblast ends, and from this 

 point all round the yelk between the epiblast and hypoblast after the 

 latter has been enveloped by the blastoderm. Usually the mesoblast 

 is freed from contact with the hypoblast for some distance beneath the 

 head; in the si)ace which results the heart is developed as a ventral 

 mesoblastic outgrowth of cells annular at first, tubular at last, and .soon 

 divided into three principal chambers separated by two constrictions, 

 which are not at first truly valvular. The space around the yelk is now 

 continuous with the heart space or pericardiac cavitj-; the latter is in- 

 deed a part of the segmentation cavity ; into this space the blood cor- 

 puscles of Belone are budded Irom the yelk through the intermediation 

 of the hypoblast inclosing the latter. The vessels themselves appear to 

 be intimately related to the hypoblast, and appear indeed to be placed 

 between it and the epiblast, but to make their progress mainly along 

 the former, plowing channels through it and the adjacent yelk. I'he 

 mode of forcing or breaking open channels from one vessel to another 

 over the yelk of Belone is well shown in Fig. 13, where the blind begin- 

 nings of vessels are arising at c, c, (?, and two such from the Inrger ves- 

 sels have met and joined but d short time since so as to connect the 

 larger channels together. The median vessel which traverses the yelk 

 is fed by the caudal vein behind; the lateral venous arcs v' ^'", on the 

 other hand, are fed directly from the cardinal veins. 



It is a verj- significant fact that the segmentation cavity plays a very 

 important part in the process of the formation of the blood and the in- 

 corporation of the yelk into the body of the embryo. There is no more 

 reason whj' the segmentation cavity should disappear in the germinal 

 disk of the fish-egg than in the segmenting egg of the amphibian, where 

 it actually is as intimately concerned in the formation of the heart as in 

 the fish, according to the evidence of the plates of A. Goette's classical 

 Ent'wickelungsgeschichte der Unl-e, but this, I am aware, is not that em- 

 bryologist's view of the matter. Kupflfer* has advanced another view 

 which it is important to notice in this connection, as it is very diiferent 

 from the one advanced by the writer in the foregoing pages. He sup- 

 poses that there is a mesoblastic layer surrounding the yelk besides the 

 epiblast and hypoblast, and which lies between the two latter. The 

 blood, according to him, originates by g-ermination from the hypoblast 

 between the latter and the mesoblast. The origin of the heart is de- 

 scribed essentially in the same way as it has been observed by the 

 writer. Kupfter in all his writings has, however, completely overlooked 

 the tact that the segmentation cavity of the fish-egg persists, and he 

 was not, therefore, in a position to estimate its importance in relation, 

 to the development of the blood. As to the mesoblastic layer said to 



*Beobachtuugcu iiber die Eutwickelimg der Kuochenfisclie. Arch. f. mik. Anat. 

 ■IV, 1868. 



