434 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
easy to understand if we admit of tlie liomologies between the gastrulas of the 
amphibian and the fish pointed out by Ziegler. The homologies are as follows : 
Fish. ' Amphibian. 
Yolk-f-periblast =Yolk cells 
Invagination at posterior pole of) ( Invagination around tlie dorsal lip 
the fish blastoderm. ( ( of the amphibian blastopore 
=Chorda-entoblast. 
The gastrula cavity of the fish lies between the invaginated layer and the periblast. 
The growth of the anterior pole of the blastoderm round the yolk represents the growth of the 
the small cells round the yolk cells in amphibian gastrulation. 
This takes it for granted, again, that the posterior margin of the fish blastoderm 
is stationary, which it is not, and it has recently been shown by Morgan that the 
“small cells” of amphibian eggs do not spread over the yolk cells, but are split off 
from them. The “ exact” comparison can no more be instituted here than between 
Brancliiostoma and the teleost. He nowhere explains how the periblast has arisen 
from the yolk cells. The condition in Cymatogaster shows that the periblast (com- 
posed of but 12 nuclei) is a waning structure and is in direct proportion to the size 
of the yolk, a condition which very strongly presupposes a larger yolk and a more 
active periblast, as is found in other teleosts and in the elasmobranchs. 
Wilson’s discussions of concrescence do not directly concern us, but the chief 
objections to his views should be stated. His theory and observations need to be 
stated to enable us to form a fair judgment. He says, p. 260: 
Iu the growth of the blastoderm round the yolk, the head end of the embryo does not remain a 
fixed point. * * On the contrary, the tail end of the embryo * * remains a comparatively fixed point, 
as Oellacher first showed, while the anterior pole of the blastoderm travels rapidly round the yolk. 
I have already shown that this is certainly not the case in some fishes. 
The point where the blastopore closes is thus but a short distance from the original position 
occupied by the posterior pole of the blastoderm. Owing to the constant position of the single oil- 
globule, these facts can easily be made out. 
Compare figs. “35,” “36,” and “38.” In other words, these statements hinge on the 
fixity of the oil-globule. How, I have certainly seen the oil-globule in a number of eggs 
shift its position if the egg was forced to lie in a position in which the vertical line pass- 
ing through the oil-globule did not also pass through the center of gravity of the egg; 
that is, if the globule did not occupy the highest point in the egg. Bearing this in 
mind, and also the fact that the protoplasm is heavier than the yolk, as is seen in 
the position of the germ in pelagic eggs without oil-globules and the position of the 
newly-hatched larvae of such eggs, the conditions figured by Wilson, 33, 36, and 38, 
can be explained as being due to purely mechanical causes, by supposing that the egg, 
on account of the change in position of the mass of protoplasm during the formation 
of the embryo, rotates 90° and that the oil-sphere, in order to remain at the highest 
point of the egg, gradually shifts its position 90°. To make this clear I reprodude 
Wilson’s figures and add diagrams. These figures represent the egg in a vertical 
median section through the axis of the embryo; the germinal ring is also indicated. 
Since the center of gravity of the yolk lies approximately at the center of the egg, 
and the oil-sphere occupies the highest position, while the germ occupies the lowest, 
the ceuter of gravity in fig. “35” lies somewhere iu or very near the line joining the 
center of the oil-sphere, the center of the egg, and the center of the blastoderm. 
