ox THE AXATOMiT OF FISHES. 
89 
strong, iiiexteiisible, and transversely-disposed fibrous membrane which is co-extensive 
with, and closely applied to, the mesial portion of the anterior wall of the bladder 
between the two post-temporal plates, and, although intimately applied to both, 
separates the bladder from the “ head-kidney ” in front. Laterally, on each side, the 
membrane is firmly attached to the ventral and inner margins of the post-temporal 
])late, and to the crescentic distal extremity of the transverse process of the fourth 
vertebra, blending with the ligamentous fibres by which these skeletal elements are 
rigidly connected together. The dorsal edge of the membrane is firmly attached to 
nearly the whole length of the anterior margins of the transverse processes in 
question, and towards the median line blends with the floor of each saccus para- 
vertebralis. In the median line the dorsal margin appears to be continuous posteriorly 
with the fibrous floor of the aortic canal, and, on each side of the complex centrum 
becomes firmly attached to its oblique lateral ridge and to the radial nodule, while 
posteriorly a slip of fibres curves upwards and backwards to an insertion into the 
anterior or inner extremity of the dorsal lamina. The latter slip and the fibrous 
floor of the aortic canal are all that we could definitely identify as representing this 
membrane on the dorsal surface of the air-bladder. Traced ventrally the membrane 
suddenly becomes extremely thin, but, nevertheless, may be detected as extending 
for a short distance on to the lateral and ventral walls of the anterior portion of the 
air-bladder, where, however, it soon becomes lost as a definite structure, either by 
blending with the proper wall of the bladder, or becoming indistinguishable from the 
connective tissue by which the latter is connected with its peritoneal investment. 
In close relation with the anterior face of the “ head-kidney ” there is a second 
transversely disposed aponeurotic membrane which occupies the same transverse plane 
on the dorsal side of the oesophagus as the pericardial septum on the ventral side, and 
is also attached laterally to the distal ends of the modified transverse processes and to 
the adjacent margins of the post-temporal plates, blending with the ligamentous 
fibres by which the transverse processes and the post-temporals are firmly united 
together, and also wutli the corresponding attachments of the first-mentioned 
membrane, but remains separated from the latter mesially by the “ head-kidney.” 
Ventrally, the aponeurosis blends with the peritoneum as the latter is reflected back- 
wards from the under surface of the “ liead-kidney ” on to the dorsal surface of the 
oesophagus. Dorsally, the membrane extends backwards as an Investment to the 
dorsal surface of the “ head-kidney ” and the ventral faces of the basioccipital, the 
inferior limb of the post-temporal, and the body of the first vertebra, finally becoming 
inserted into the anterior margins of the modified transverse processes of the fourth 
vertebra, and there to some extent blends with the similarly attached dorsal edge of 
the preceding membrane. The aponeurotic membrane has clearly no relation to the 
air-bladder, but may be regarded simply as a dorsal and backward extension of the 
same fibrous sheet which, ventrally, separates the pericardial and abdominal cavities. 
With regard to the first-mentioned of the two membranes it is not easy to decide 
MDCCCXCHI. — 13. 
X 
