ON THE ANATOMY OP FISHES. 
135 
attachments of the anterior wall of the air-bladder, but while agreeing with 
Arius in the insertion of the dorsal edo-e of the outer stratum of the tunica 
O 
externa into the anterior margins of the modified transverse processes of the 
fourth vertebra, Platystoma differs from that genus in having the fibres of the inner 
stratum of the lateral portions of the anterior wall connected dorsally with the 
tripodes, even though, as in Arius, the fibres of the same stratum in the lateral walls 
fail to reach the ossicles in question ; and, consequently, in this respect Platystoma 
occupies an intermediate position between Arius and Macrones. 
The more extensive attachment of the anterior wall of the bladder to immoveable 
portions of the axial skeleton is evidently a further provision to counteract any 
tendency on its part to participate in the distension or contraction of the anterior 
chamber, and consequently helps to restrict any diminution or increase in the capacity 
of that compartment to movements of the lateral walls which alone can influence the 
Weberian ossicles. 
Unlike Platystoma a transverse membrane can be readily recognized in Arius as a 
thin but tough and inextensible layer of transversely or obliquely arranged fibres 
divided into two lateral halves by the subvertebral process. Each half is attached by 
its inner edge to the lateral surface of the process, and dorsally to the corresponding 
radial nodule and dorsal lamina, and in addition to the whole extent of the ventral 
margin of the decurved anterior edge of the transverse process of the fourth vertebra, 
from its root to its pointed distal extremity. At the antero-lateral angles of the air- 
bladder the fibres of this membrane pass from the extremities of the transverse 
processes and blend with the adjacent portions of the ventral wall. 
The existence in Arius of a transverse membrane distinct from that layer of the 
tunica externa of the proper anterior wall of the bladder, which is also dorsally 
attached to the anterior margins of the modified transverse processes, removes one 
difficulty as to the exact natm^e of these structures in some Siluridse. In the case of 
Platystoma fasciatum, which, in this respect, agrees with the species dissected by us 
(P. tigrinum), that part of the anterior wall which is attached to the transverse pro- 
cesses is regarded by Sorensen (37) as representing the greatly thickened inner 
stratum of the peritoneal coat (“ la pl^vre ”), and is therefore the equivalent of that 
special portion of the superficial coat of the bladder hitherto referred to by us as the 
transverse membrane.^" In P. tigrinum, and presumably in P. fasciatum also, there 
is no transverse membrane distinct from the apparent proper anterior wall, and there- 
fore, so far as these species are concerned, it is possible that the skeletally-attached 
outer stratum of the latter may be the equivalent of Sorensen’s “ la plevre ” and our 
transverse membrane. But in Arius a skeletaUy-attached outer stratum of the 
anterior wall and a transverse membrane coexist, and consequently the one can 
scarcely be an equivalent to the other in cases where either is absent. Hence to us 
it seems more reasonable to assume that the skeletally-attached stratum of the 
• See Macrones nemurus, p. 90. 
