DEVELOPMENT OE THE SKULL IN THE SALMON. 
125 
% 
arch. There is now no appearance, nor will there be for several days to come, of any 
further subdivision of the branchial arches ; their respiratory papillae are beginning at 
this stage as dermal growths. With the “ primitive groove” filled in, with considerable 
consistency of the brain, and with the skeletal growths so far advanced, the young 
Salmon is becoming too active and strong for further imprisonment. 
Fourth Stage. — Salmon-fry in the act of hatching and newly hatched. 
The figures given are from young Salmon with the head only protruding from the 
shell (Plate II. figs. 9 & 10, and Plate III. figs. 1-3), and of others which had enjoyed 
a day or two of freedom (Plate II. fig. 11, and Plate III. figs. 4-6). The expansion of 
the cerebral vesicles is a correlate of that peculiar bend downwards of the fore part of 
the head which is called the “ mesocephalic flexure.” Now, for the first time, the 
head assumes the form so familiar to the embryologist in the air-breathing Vertebrata. 
This overbending of the cerebral vesicles evidently is not the vera ccmsa of the arrest 
of the notochord behind the smaller, dipping vesicle, the pituitary body, but is a corre- 
late of the sudden finish of the investing mass : a cause for both these arrests has yet 
to be found. This, which may possibly be a modification taking place during long 
secular periods, is also correlated with the small amount of independence seen in the 
cartilaginous cranium, which, where it is not largely aborted by the implantation of 
the ear-bulbs, yet grows over the membranous cranium from any neighbouring cartila- 
ginous tract. I said groins over , that is in the lower types ; but the most important part 
of the nervous system is largely roofed in simply by investing bones. As yet, in this 
fourth stage, the only part of the membranous cranium floored or surrounded by cartilage 
is the “ medulla oblongata.” A sectional view, both during hatching (Plate III. fig. 3) 
and after (fig. 5), shows the form of the principal parts of the brain, and the relation 
of its parts to axial and facial structures. The earlier specimen (fig. 3), even, has now 
budded out a well-defined “prosencephalon” (Cl 4 ), the more solid “ thalamencephalon” 
(CP') lying on a pair of balks, which are the bent and flattened trabecula) (tr.) : the 
pituitary body {jgy.) has no “ rest,” but lies between the trabecular hooks, amidst gela- 
tinous blastema. Even the fore part of the “ medulla oblongata” (: m.ob .) rests upon a 
fold of the primordial enclosing membrane, which retires to line the back of the pitui- 
tary body ; this is where Patiike placed his “ median, or azygous trabecula,” evidently 
a mere membranous band projecting beyond the notochord in the mid line, and having- 
no morphological relation to the trabeculae proper : even these parts were misunderstood 
by this most excellent embryologist. If these sections be compared with the earliest 
stage (Plate I. fig. 8, net), a notable difference will be seen — namely, that the pointed 
end of the notochord has retired somewhat, and that the apex of its sheath has grown 
downwards behind the pituitary body as a free tongue-like process. 
In these sectional views we see how the walls of the face are broken into ridges 
and furrows; and in fig. 5 especially the relation of both the prseorals is shown — the 
second (subocular) forming the upper cheek-wall, and the first (the trabecular) forming a 
