EMBRYOLOGY (JF THE SEA BASS. 
225 
to increase in length (Fig. 59, PI. xcv, inecliau longitiulinal section through a stage 
something younger than Pig. 38, PI. xcii). In transverse seetions the exact relations of 
this mass are made clear. It is then seen that in the posterior half of the apparently 
homogeneous mass (c. ni.) there is no differentiation of layers at all, but in the anterior 
half the condition is such as is shown in Fig. 60, PI. xcv. The posterior region, in which 
there is absolutely no differentiation, will be spoken of as the caudal mass (bourgeon 
caudale, schwanzknospe), while the anterior region, in which there is fusion between the 
neural chord and the entoderm, may conveniently be ealled the neurenterie streak. The 
whole tract from the termination of the notochord to the end of the embryo has been 
regarded by Henneguy (18) and Sehwarz (39) as a primitive streak analogous to the 
streak of Amniota, and (Schwarz) homologous with that of Amphibia. I fully agree 
with these views and will use the term primitive streak in this meaning. The name 
neurenterie streak implies that it is in this region that the neurenterie canal belongs, 
or would belong if it ever came into existence. As to this location of the neurenteiric 
canal most students are agreed. (See especially Henneguy, 1. c.,aud Schwarz, 1. e.) 
During the further growth of the blastoderm round the yolk the primitive streak 
may increase in length {pr. str. Fig. 65, PI. xcvi, a median longitudinal section just 
before the closure of the blastopore), but the increase is scarcely measurable. 
Before discussing the way in which the primitive streak is established, a word 
may be said on the condition in the Salmonidw. There here aii^iears to be from the 
start, i. e., before the blastoderm has begun to encircle the yolk, a very considerable 
caudal mass (bourgeon caudale, schwanzknospe), and Henneguy, 1. c., p. 585, has sat- 
isfied himself that in the trout the region of the primitive streak does not increase 
appreciably in length during the closure of the blastopore. How, in spite of the fact 
that embryos vary considerably in length, and exact measurements can not therefore 
be implicitly trusted, it is unmistakable that in the Bass the primitive streak suffers a 
considerable increase in length during the closure of the blastopore. When the rapid 
growth of the blastoderm begins there is really no bourgeon caudale (Fig. 44, PI. xciii), 
but by the time the yolk is half encircled there is a iierceptible bourgeon (caudal 
mass), from which the fusion extends forwards in the median line (Fig. 49, PI. xciv, 
transverse section through the anterior part of the primitive streak). Still later (Pigs. 
36, PI. xoii, and Fig. 55, PL xov) the primitive streak has slightly increased in length, 
and in the yet later stages (Fig. 59, PI. xcv, and Fig. 65, PL xcvi) its increase in length 
is unquestionable. 
The apparent difference between the Trout and the Bass in this matter may not 
after all be so great as it would seem. Without going into a further comparison, 
however, I feel bound to regard the Bass development as the more typical, if only for 
the reason that it is as regards this iioint so easily harmonized with the Amphibian 
development. 
The means by which the primitive streak is brought into existence is obvious. 
As the blastopore grows smaller the extra embryonic part of the germ ring is 
imssu drawn into the tail end of the embryo, and there is thus built up in this region a 
constantly increasing mass of undifferentiated cells. As this mass is built uj) the 
mesoderm plates cut their way back, and thus give rise iu the anterior part of the 
region to what I have called the neurenterie streak. In the Bass there is no actual con- 
crescence iu the middle line (see surface view of closing blastopore. Fig. 39, PL xoiii), 
but the terminal notch observed in some fish, as well as general considerations derived 
Bull. U. S. F. C. 89-^ 15 
