b y th e oticus lateralis and one by th e supratemporalis lateralis vagi. Near the middle of the bone,
between th e organs innervated by the oticus and supratemporalis lateralis vagi, th e main infraorbital
canal anastomoses with the dorsal end of the preopercular canal.
The primary component of th e pterotic has a thickened anterior portion b u t is elsewhere
a th in plate of bone which lines th e v entral surface of th e angular dermal component of th e bone.
I t is traversed b y th e external semicircular canal, b u t otherwise has no bounding relations to the
cranial cavity. This is due to th e presence of a large and deep depression on the lateral surface of
the brain case, this depression pinching the skull, so to speak, to such an extent th a t the primary
component of th e p terotic has been pressed upward against the under surface of its dermal component,
has ceased to form p a rt of th e apparent lateral surface of the brain case, and simply lines th e ventral
and ventro-mesial surfaces of th e overhanging roof of th e depression. This depression includes both
th e prootic and subtemporal fossae of Scorpaena, and it has, in addition, a large posterior extension
which lies posterior to th e arch of th e external semicircular canal and is not found in Scorpaena. The
entire depression may, however, be called th e subtemporal fossa, th e fossae of opposite sides giving
to th e brain case, in ventral views, an hour-glass-shaped appearance. The adductores hyomandi-
bularis and operculi h ave their origins in the fossa, arising mainly on a b and of cartilage th a t separates
the pterotic from th e exoccipital. The levators of th e first four branchial arches have the ir origins
in the anterior portion of th e fossa, anterior to the adductor hyomandibularis, while th e fifth levator
muscle has its origin ventro-mesial to th e adductor operculi, between th a t muscle and th e foramina
for th e vagus and glossopharyngeus. This fifth levator perforates, to reach its surface of origin,
a mass of tissue th a t looks like degenerate glandular tissue and th a t must be th e homologue of what
I considered as th e thymus in Amia (’97, p. 643). Whatever i t m ay be, this tissue fills th e entire fossa
ventro-mesial and posterior to th e adductores hyomandibularis and operculi, and the fossa would
seem to b e developed in some relation to i t ra th e r th a n as any consequence of the origins of th e several
muscles here.
On th e anterior edge of th e thickened anterior portion of th e p rimary component of th e p terotic
there are two fossae, th e larger, mesial one lying on th e internal surface of th e bone while th e smaller,
lateral one lies on its external surface. The mesial fossa lodges the anterior portion of th e external
semicircular canal, and is in communication w ith th e dorso-antero-lateral end of the tall recess already
described on th e internal surface of th e prootic and sphenotic; th e posterior portion of th e semicircular
canal traversing th e th in posterior portion of the auto-pterotic, and forming a prominent ridge on
its external surface. The lateral one of th e two fossae forms th e posterior half of th e dilatator fossa,
th e anterior half of th a t fossa lying in th e sphenotic. Immediately posterior to th e dilatator fossa,
and so placed th a t i t seems to form p a rt of the mouth of th a t fossa, there is, on the external surface of the
auto-pterotic, a large round articular fossa which gives articulation to th e regular posterior articular
head of th e hyomandibular. The hind edge of this articular fossa is thickened and grooved to form
an elongated, transversely placed and slightly convex articular surface, which gives a sliding articulation
to the posterior one of th e four articular heads of th e hyomandibular. Immediately dorsolateral
to this articular surface is th e perforation b y which th e preopercular latero-sensory canal
joins th e main infraorbital canal.
The auto-pterotic is bounded anteriorly by th e sphenotic and prootic, and also by a small
interspace of cartilage between th e adjoining edges of those two bones. I t is separated from th e
exoccipital by a band of cartilage, and beneath (dorsal to) this band it is in contact with the lateral
edge of the epiotic. The hind end of the bone gives support, as already stated, to the anterior end
of th e suprascapular.
The pterotic of Dactylopterus thus has no exposed surface corresponding to the one th a t
forms p a rt of th e lateral bounding wall of th e temporal fossa in th e other fishes of the group. That
portion of the temporal fossa th a t is usually bounded by this bone is thus either wholly absent in
Dactylopterus, or it has been reduced, by the compressive action th a t has given rise to the subtemporal
fossa, /to a narrow space th a t lies between the pterotic-exoccipital band of cartilage, ju s t above referred
to, and the overlying dermal bones on th e dorsal surface of the skull. The band of cartilage is evidently
the homologue of the cartilage th a t forms th e bottom of the temporal fossa in the other fishes, and
as this cartilage is certainly not in synchondrosis with the overlying dermal bones, a th in space must
exist between them, in the place where the temporal fossa is usually found. A further possibility
regarding a portion of the fossa will be referred to when describing the suprascapular.
The BASIOCCIPITAL is broad and thin, is slightly convex on its internal and slightly concave
on its external surface, and the median longitudinal line on its ventral surface presents a slight
reentrant angle. On the hind end of the bone there is a deep median p it which extends forward to
th e line of the reentrant angle and represents the vertebral depression on the hind end of the bone.
Lateral to this pit, the wide flange-like portions of the bone give support, on their dorsal surfaces,
to th e ventral edges of th e exoccipitals. Anteriorly the bone suturates with the prootics and parasphenoid.
A very slight depression on either side of th e internal surface of the bone forms the hind
end of the saccular groove. Between these two depressions there is, in th e anterior portion of the
bone, a slight median depression, the significance of which could not be determined in my limited
material. The anterior edge of the bone is grooved, and encloses th e hind end of the th in median
sheet of cartilage th a t connects th e ventral edges of the prootics of opposite sides. The median
portion of the hind edge of the dorsal surface of the bone forms th e ventral boundary of the foramen
magnum.
The EXOCCIPITAL has a concave lateral and a strongly reentrant posterior surface. The
la tte r surface has the two usual portions, one of which forms p a rt of the hind wall of th e cranial
cavity while th e other arches over th e medulla, the two portions appearing, in the disarticulated bone
as a stout, tall, V-shaped ridge arising from the dorso-mesial surface of a sub-oval bone. That p a rt
of the posterior portion of the bone th a t arches over the medulla has a thick dorso-mesial edge, which
suturates in p a rt with its fellow of th e opposite side b u t mainly with the ventral edge of th e spina
occipitalis. The dorsal edge of th a t p a rt of the bone th a t forms p a rt of th e hind wall of the cranial
cavity suturates with th e ventral edge of a strong ridge on the ventral surface of the epiotic, this
ridge forming th e posterior surface of the la tte r bone. Slightly antero-lateral to this ridge on the
epiotic, a relatively large V-shaped portion of the ventral surface of th a t bone is in sutural contact
with a corresponding surface on the dorso-mesial surface of the lateral plate of the exoccipital; this
la tte r surface of contact extending downward from the dorsal edge of the exoccipital nearly to the
central point of the bone. Between the portions of these two bones th a t have these two sutural
connections — the lateral and posterior plates of the exoccipital, below, and th e epiotic above ^
there is a tall and narrow space which must lodge th e posterior portion of the external semicircular
canal, the larger portion of the posterior canal, and possibly also the hind end of the utriculus; b u t
want of material prevented my determining the exact relations. A broad, low and rounded ridge
Zoologies. Heft 67. 22