224 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
tliickeniug is further iucreased. The development of these organs begins in the pos- 
terior region and travels anteriorly. Thus in the same embryo several stages in , 
their ditterentiation may be observed. Fig. 53, PI. xciv, and Fig. 54, PI. xcv, are 
two seetions through the trunk of an embryo somewhat younger than that shown in 
Fig. 37, PI. xcii. In the anterior section. Fig. 53, the condition is much the same as :: 
in the earlier embryo. Fig. 50, PI. xerv, but with this difference, the under, e?t., of 
the two layers, found on each side of the incipient chorda, ?ic., has detinitely separated 
from the chorda and from the upper layer, mes. It is thus made up of separate halves 1 
which project slightly under the chorda cells. This layer constitutes the definitive 
entoderm. The intimate connection between the chorda cpUs and the layer marked mes 
(which develops into the mesoblast) is soon broken. The connection has no signifi- i 
canee, for it sometimes happens that the chorda cells separate simultaneously from 
both entoderm and mesoderm. In the posterior section (Fig. 54) the differentiation 
has gone a step farther. The chorda has separated from both layers and has assumed i 
a compact shape, though the cells have practically their former arrangement. The 
entoderm is in the same condition as farther forwards, but the unicellular mesoblast ■ 
layer of Fig. 53 has begun on each side to thicken up at its inner angle. The thick- 
ening is as yet very slight, and extends but a short distance away from the notochord, i 
Let us trace the development of the notochord and secondary layers through a 
slightly older stage than that of Figs. 53 and 54. Of t his stage a surface view from i. 
above is given in Fig. 37, PI. xoii, from the side in Fig. 30, PI. xoii, a median longitudi- > 
nal section in Fig. 55, PI. xov, and a series of transverse sections (numbered from i; 
behind forwards) in Figs. 56, 57, and 58, PI. xcv. From the surface views and the , 
longitudinal section the relative extent of the head (brain) and trunk regiv^ns may be j 
gathered. In the posterior trunk region (Figs. 55 and 50) the notochord is easily r 
recognizable. The entoderm is in its former condition, or nearly so, still consisting of , : 
two lateral unicellular sheets which project under the notochord. The projection j 
under the notochord (Fig. 56) is more marked than in earlier stages. The mesoderm ^ 1 1 
plates have thickened considerabljr (compare Figs. 54 and 50). It is only near the edge i 
of the shield that they now consist of a single layer of cells. On passing gradually ' I 
into the anterior trunk region the notochord will be found to grow appreciably thinner; ; 
(Fig. 55), likewise the mesoblast plates. Going still farther forwards into what may 'i 
be called the neck region («, Figs. 55 and 57), the notochord and entoderm fuse, and i 
tlie mesoblast plates thin away into scattered cells {mes., Fig. 57). In the head regions ji 
(Figs. 55 and 58), in connection with the vertical development of the brain, the primi-3; 
tive hypoblast has thinned away into a unicellular layer, which in later stages is inj|i 
part transformed into scattered mesoblast elements, and in part persists as the extremejf 
anterior portion of the entoderm lamella. I j 
Formation of the primitive strealc and closure of the blastopore . — Before the earlyjj, 1 
history of the notochord and layers can be concluded, it will be necessary to describe j, 
the course of development at the posterior end of the embryo. The condition at this ;) 
end, after the completion of the germ ring, is shown in Fig. 44, PI. xoiii, a median Ion- jjf 
gitudinal section through a stage such as Fig. 35, PI. xcii. When the blastoderm has i 
grown round the yolk, as far as is shown in Fig. 30, PI. xcii, the condition at this i)olej I 
becomes a slightly different one. There is now at the posterior end of the embryo (Fig.l I 
55, PI. xov, median longitudinal section) an undifferentiated mass of cells of consid-|i; 
erable extent {c. m.). Let the blastotlerm grow still farther, and the mass will be found] j 
