OCCURRING IN THE NEIGHBOURHOOD OF PLYMOUTH. 51 



slightly later stage the space is again open beneath the anterior end of 

 the embryo, between the yolk and the epiblast forming the wall of the 

 yolk-sac ; and this space becomes the venous sinus out of which the 

 heart opens. Thus it would seem that the venous sinus in the pelagic 

 ovum, which corresponds to the vitelline veins in other ova, is 

 bounded internally by the periblast, and externally by a layer of 

 epiblast, and that it is a persistent segmentation cavity. 



But if we look at the living newly-hatched larva of PI. micro- 

 cepTialiis (fig. 8) we see over the surface of the yolk black and 

 yellow dendritic chromatophores. As these are continued over the 

 anterior surface of the yolk-mass, and are not visible in the wall 

 of the yolk-sac in front of the yolk, it follows that these chromato- 

 phores exist on the surface of the periblast and not on the inner 

 surface of the epiblastic wall of the yolk-sac. Chromatophores are 

 of course mesoblastic, and these in particular are, in all probability, 

 developed in situ from the periblast, the nucleated protoplasm 

 surrounding the yolk. It has been shown in various Vertebrate ova 

 that the nucleated protoplasm of the yolk at early stages buds off 

 cells which join the mesoblast. At this later stage in the Teleostean 

 ovum the periblast forms mesoblastic pigment-cells, and also, as 

 shown by Ryder and others, hlood-corpuscles. But there is no 

 evidence that there is any mesoblast on the internal surface of the 

 wall of the yolk-sac ; the venous sinus is bounded internally by 

 periblast with a few chromatophores on its surface, externally by a 

 layer of epiblast. It is obvious, on reflection, that all the yolk, 

 with its periblast, after the formation of the definite hypoblast, is a 

 part of the splanchnic mesoblast. But the interesting morphological 

 peculiarity about the venous sinus in the Teleostean embryo is that 

 it is the persistent segmentation cavity. The segmentation cavity 

 may partially disappear by the contact of its walls, but it is not, as 

 usually represented in the frog, obliterated by the growth of the 

 mesoblast ; and thus, when the sinus venosus appears it is not as a 

 cavity or system of veins entirely surrounded by splanchnic meso- 

 blast, but is the old segmentation cavity between the epiblastic 

 ventral wall of the yolk-sac and the periblast. This periblast 

 develops chromatophores on its surface, a process which is 

 peculiarly well illustrated by the formation of pigment-cells round 

 the oil-globule in the mackerel ; and at a later stage, no doubt, the 

 sinus venosus acquires mesoblastic walls all round it, but this is not 

 till the yolk has been absorbed. 



The opening of the posterior end of the heart into the cavity round 

 the yolk looks very surprising at first sight, and as it is very 

 conspicuous has attracted the attention of all who have studied the 

 development of pelagic Teleostean ova. I have figured it myself in 



