EMBRYOLOGY OF THE SEA BASS. 
231 
with this change in shape the entoderm is being drawn away from the sides towards 
the middle. This contraction of the entoderm towards the axis of the embryo be- 
comes more apparent in later stages. 
In this stage there is no trouble in discerning the precise anterior extent of the 
entoderm. In front of the medulla {mecl., Fig. 147, surface view, of about same age as 
the sections we are now examining) there is scattered mesoderm but no entoderm. 
The transverse section, Fig. 71, through the middle of the branchial region, lies about 
in the plane x of Fig. 147. In this region the entoderm does not thicken in the middle 
line, but along two lateral lines {br. /., Fig. 71), which in going back towards the pos- 
terior limit of the branchial region approach the middle. Along these lines the 
columnar cells have already begun to rise up in the shape of lateral folds (br.f.). 
Stage of 39 hours (Figs. 73 to 79, PI. xcvii, series of sections). In this stage Kupf- 
fer’s vesicle is nearly as shallow as in the last, but its lateral limits {ni-m.. Fig. 73) are 
more sharply marked than was the case in Fig. 66. This is due, as the direction 
of the cells and subsequent stages show, to the fact that the entoderm cells at m-m 
have begun to grow towards the middle line. The formation of KuptfePs vesicle is 
a true process of folding quite like the formation of the alimentary tube in the trunk. 
There may be one point of difference which concerns the lateral entoderm {1. en., 
Fig. 73). In the trunk the whole entoderm lamella is contracted towards the middle line 
and used up in forming the tube; but in the region of Kupffer’s vesicle the lateral 
entoderm at least retains for a long time its original extent, while growing very thin. 
I am inclined to believe that the lateral entoderm is not drawn into the walls of the 
vesicle as it is into the walls of the gut, but that after becoming an excessively thin 
membrane it is absorbed in some other way. In a slightly later stage (41 hours) than 
the one just described the lateral entoderm has disappeared (Fig. 82, PI. xcviii). In 
this stage the increase in height and diminution in width of the vesicle, in consequence 
of the progress of folding, will be apparent. 
Anterior to KupffePs vesicle there is a median tract of thickened entoderm 
(transverse section, Fig. 74, en.) most of which will never be folded off, but will form 
the postanal gut. The anterior portion of the tract will, however, form a fold, and the 
axial cells in Fig. 74 have the typical arrangement and shape which precede its first 
appearance. Farther forwards there is a well-marked furrow (Fig. 75) which reaches 
its greatest height in about the middle of the trunk (Fig. 76). The distinctness of the 
cell outlines enables us to get at least a superficial insight into the process by which 
the furrow is produced. Oji comparing the three stages. Figs. 74, 75, and 76, there 
seems to be very little if any cell multiplication; no nuclear figures were observed. 
The cells on each side of the median line early exhibit a tendency to grow towards the 
axis ; and in a general way, the process may, I think, be described as follows : In Fig. 
74 the Inner and lower ends of the cells 2, 2, grow towards the axis, and in so doing 
lift up the cells 1, 1, thus establishing an extremely minute cavity between them and 
the periblast. It is possible to follow every stage in the formation of the cavity from its 
first appearance as a minute, almost round aperture to the condition shown in Fig. 75. 
The cells 3, 3, follow the example of their pred ecessors, and thus the keystone cells 1, 1, 
are lifted still higher above the periblast. Let this course of action be followed by 
the cells 4, 5, etc., and it is easy to understand how the arch is increased in height, 
and at the same time how the lateral entoderm {1. en., Fig. 74) disappears (as it has in 
Figs. 75 and 76); its cells, oue after another, slip under their axial neighbors towards 
the median line. 
