232 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
The cell outlines in the several stages of the furrow, which I have tried to reproduce I 
exactly in the figures, indicate that the method just described is the one followed, but > 
complications undoubtedly supervene; for instance, the gut is often closed in ven- ; 
trally before all the lateral entoderm has been drawn into its wall (Fig. 93, PL xcix). 
Exactly how the cells thus left out disappear I do not know. In some cases, it would j 
seem, they force their way between the cells which already line the tube, but again i 
their position indicates that they are merely absorbed. In general, the method of form- 
ing the gut may be spoken of as a modification of the ordinary itrocess of folding, such 
as is made use of in the Amuiota, its peculiarity depending on the fact that the lateral 
entoderm is employed in helping to form the tube. 
The condition in the branchial region is shown in the transverse section. Fig. 
79, PI. xovii. If this figure be compared with the earlier stage. Fig. 71, PL xcvii, the ' 
course of development will be evident. The lateral folds (br.f.), barely begun in Fig. 
71, are now well marked out. The further development of this tract of the enterou 
may be indicated in a few words. The apex of the fold, a, grows dorsally and ulti- 
mately fusing with the ectoderm, there is thus established the embryonic gill slit ' 
(only one in the embryo). The base, b, grows towards the median line and, meeting 
its fellow of the opposite side, closes in the foregub ven trally. As to the fate of the i 
lateral entoderm of this region, I have not been able to come to a decision. I 
In passing backwards from Fig. 79 the lateral branchial folds gradually die out, 1 
Fig. 78, and in so doing approach the median line, where, in the oesophageal I'egioii I 
(Fig. 77), they give place to a single broad low arch. The low oesophageal arch passes n 
gradually into the deeper truuk furrow of Fig. 70. | | 
Stage of 45 hours (Figs. 83 and 84, PL xoviii, and Figs. 88 to 95, PL xcviii and | 
XCIX, series of sections). In this stage the tail begins to develop. The furrow, /./.f I 
which marks it off -is shown in the lougitudiual section, Fig. 84. Eupff'er’s vesicle in 
many individuals of this age is still bounded by the periblast, as in the longitudinal j 
section, Fig. 84. In others, however (transverse section. Fig. 88), the process of fold-’ I 
ing has been completed, and the vesicle has an entire cellular wall; the lower edges of : 
the fold, m-m. Fig. 82, have met in the middle line and closed in the vesicle ventrally. I 
In front of Kupft'eFs vesicle stretches the postanal gut {p. a. (/., Fig. 84), a mere I 
thickened stripe of entoderm. At its posterior limit, just in front of the vesicle, at 3 
r. Fig. 84, it is fused with the notochord. A transverse section through this region of ' 
fusion is given in Fig. 90. 
Directly in front of the postanal gut, at about a, Fig 84,. a short tract of entoderm is 
found, in the condition shown in Fig. 91. This is the region of the anus, and the fold 
here appears last of all. In front of the anal region there is still for a short. distance 
an open furrow, Fig. 92, but in the rest of the trunk and in the branchial region the 
enteron is now closed ventrally. The closure begins anteriorly aud travels back. 
Figs. 93 aud 94, PL xcix, are through the truuk region. In the latter figure the rem- 
nant of lateral entoderm has finally disappeared. In the oesophageal and branchial 
regions the cavity of the enteron is almost obliterated. Fig. 105, PL c, is through 
the oesophagus {oes.) of a later stage than the one now under examination. Fig. 95, 
through the branchial region, is slightly oblique, and therefore it is only on one 
side that the gill slit is shown. In forming the gill slit the entoderm becomes contin- 
uous with the nervous layer of the ectoderm, as shown in the figure, while the epider- 
mic stratum at first remains unbroken, dipping slightly into the branchial cavity. 
