HEAD CAVITIES AND NERVES OF ELASMOBRANCHS, 738 
already pointed out,' the separation of the layers of the meso- 
blast so as to give rise to a ccelomic cavity, occurs much earlier 
in the head than in the body. In the head this separation first 
occurs in the earlier part of stage p,® while in the body in does 
not commence till stages @ or H, though the two layers of the 
mesoblast, somatic and splanchnic, are distinctly formed by 
Stage D. 
My observations on the earlier stages of development of the 
head cavities accord so completely with the account given by 
Mr. Balfour that I shall deal with the subject rather summarily, 
referring the reader for a fuller account to Mr. Balfour’s work 
already cited. 
Plate V, fig. 1, represents a transverse section through the 
head of a Scylliwm embryo at the end of stage c; the mesoblast 
(mes.) is distinctly divided into two layers, which are just com- 
mencing to separate from one another. Balfour figures a similar 
section® at a slightly earlier age, when the mesoblast is one solid 
mass with no distinction into layers. 
By stage D this separation of somatic and splanchnic layers, 
which is just commencing in fig. 1, has gone so far as to give 
rise to a distinct cavity in the head region on either side, while 
at a considerably later period (stages G or H) the separation ex- 
tends further back, so as to form in the trunk the peritoneal or 
body cavity. 
The whole of the cavity so formed may be termed ccelom; 
that part of it which is contained in the trunk is the body cavity, 
while the anterior part, situated in the head, is named by Balfour 
head cavity. This latter, as Balfour has pointed out,* “ can 
only be looked on in the light of a direct continuation of the 
body or peritoneal cavity into the head.” 
Plate V, fig. 2, shows the condition of the head cavities at 
stage H, as seen in a transverse section through the hind brain. 
The cavities (4.c.) are of considerable size ; their ventral ends lie 
against the sides of the alimentary canal (a/.), with which they 
are in close contact, while dorsally they extend some distance up 
the sides of the brain. The section also passes through the 
roots (vit) of the seventh or facial nerves, whose distal ends are - 
seen to be in very close contact with the dorsal walls of the head 
cavities. The walls of the cavities are formed of a single layer 
of short columnar or almost cubical cells. 
The sections from the same embryo in front of the one figured 
1 Op. cit., p. 86. 
? Throughout the present paper I have employed, in order to distinguish 
the different stages of development, the nomenclature proposed by Mr. 
Balfour in his work cited above. 
3 Op. cit., plate ix, fig. 2. 
* Op. cit., p. 86. 
