208 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
invaginated cells liave wandered a short way only into the egg. Now, if we examine 
the peripheral cell envelope, we find abundant evidence of cell division over the side 
of the egg where the embryo proper is soon to be built up. Nests of nuclei, often sur- 
prisingly large and numerous, are now and then seen in the midst of a spherical mass 
of yolk either at the surface or just below it. Isolated cells, though few in number, 
also occur, scattered through the peripheral parts of the yolk at this stage. What 
is their origin ? They can not be referred to invaginate cells, since none of these 
have yet wandered to remote parts. Furthermore, these cells tend, not to scatter, 
but to migrate in a body. They may be the descendants of the primary yolk cells or 
migrants from the peripheral cell envelope, or may originate in both these ways. 
At the time of invagination the egg of Alpheus is very similar to that of the lob- 
ster in its histological relations. The main difference which is apparent to the eye 
is in the larger size or greater amount of food yolk in the latter. I have described 
and figured the invagination stage of Alpheus in some detail in my work on the 
embryology of this crustacean (.94, p. 400, plate xxxi). In this species the primary 
yolk cells persist and mingle with the wandering cells derived from the invagination. 
An egg of Alpheus sauleyi in the invagination stage contains about 460 cells, of which 8 
per cent — exactly 37 were found in two separate eggs — are primary yolk cells (94, p. 
432, table 1). These yolk cells do not appear to be much more numerous in the larger 
egg of the lobster (see cuts 21 and 22, showing eggs with 15 and 28 yolk cells respect- 
ively), but in this animal they degenerate faster than in Alpheus, so that at the 
invagination period very few are left. On the other hand the occurrence in the lobster at 
this time of nests of nuclei within the yolk ball, which lies just below the surface or some 
times almost in contact with it, points to migration of cells from the surface after the 
invagination stage. In any case most, if not all, such cells degenerate and disappear. 
In an older embryo, represented in fig. 254, the pit in the invagination area is 
considerably enlarged, and below this a solid wedge-shaped column of cells is seen 
advancing straight down into the yolk or bending slightly toward the hinder end of 
the embryo. This mass of cells forms what has been called the thoracic-abdominal 
plate. It here gives rise in part to a mass of cells which migrate into the yolk and 
eventually form mesodermic and endodermic structures. I shall call this cell-mass the 
mesendoderm. Begarding these cells we notice in particular the peculiar association of 
the cytoplasm with the yolk, the pseudopodia by which they worm their way among 
the yolk spherules like so many amoebte, the evidence of cell multiplication and of the 
degeneration or breaking down of cells. 
Signs of cell degeneration are also present in a striking degree over the embryonic 
area of the egg. The surface cells form a single tier of short prismatic elements 
loaded with yolk, while beneath them we see a thin cloud of fine chromatin particles. 
These are mostly the remains of cells which have migrated thither from the meseudo- 
derrnic mass, and possibly in part also of cells which have wandered from the surface. 
The embryo at a little later period has the appearance shown in plate 53. The 
ingrowing plug of cells has a rounded, somewhat pear-shaped outline when seen from 
above, the stem end of the pear pointing backward and downward into the yolk. 
The embryonic area of the egg presents a beautiful mosaic of cells, among which 
karyokiuetic figures are abundant. The dividing planes of these cells are always 
radial — that is, parallel with a radius drawn from the center to the surface of the 
egg, but make any angle with a line drawn upon the surface of the egg, such as that 
