ON THE ANATOMY OF FISHES. 
223 
transverse processes. There is no divergence of the dorso-lateral and ventro-lateral 
muscles, and consequently the lateral walls of the bladder are widely separated from 
the external skin. On removing the ventral wall a thick and almost circular pillar 
of fibres is seen to extend vertically between the dorsal and ventral walls, thereby 
reducing tbe actual cavity of the bladder to the condition of a comparatively narrow 
circular canal surrounding the central pillar {t.s.). The antero-lateral portions of the 
canal are very slightly dilated, and into the dorsal wall of these regions tbe crescentic 
process of tbe tripus is on each side inserted in the normal fashion, but otherwise 
the calibre of the canal is fairly uniform throughout. Dorsally, the central pillar is 
closely moulded to the ventral and lateral surfaces of the centra of the complex and 
fifth vertebrae, and firmly adherent thereto, while the lateral portions of the bladder 
are applied to the ventral surfaces of the transverse processes of the fourth and fifth 
vertebrae. We can only regard the formation of the vertical pillar as due to an 
excessive thickening of the primary transverse septum, and possibly also of the 
anterior portion of an incomplete longitudinal septum, with the necessary result that 
the central or mesial portions of the bladder cavities have to a great extent become 
obliterated. As in the more normal types, the cavity of the air-bladder has a thin but 
continuous lining of tunica interna. Notwithstanding the degenerate condition of 
the bladder, the disposition and arrangement of the fibres forming the comparatively 
thin antero-lateral walls of the circular canal and their convergence in the dorsal 
wall to their ultimate insertion into the crescentic processes of the tripodes (fig. 98), 
are substantially the same as in a normal anterior chamber. Radial fibres {r.f.) may 
also be seen passing from the concavity of each crescentic process {t.r.c.), and, in the 
absence of a radial nodule, becoming attached to the free extremity of the spicular 
dorsal lamina {d.l.). We could detect no trace of a ductus pneumaticus. 
The tripus has the usual three processes, but the crescentic process (fig. 98, t.r.c.) is 
very slender and has only a slight inward curvature. The scaphium is quite normal, 
having completely formed spatulate, condylar, and ascending processes. The inter- 
calarium is a very feebly developed nodule in the interossicular ligament. Claustra 
are present with their dorsal extremities projecting into the intercalated cartilage 
between the spine of the third vertebra and the supraoccipital and exoccipital. 
We were unable to make any satisfactory examination of that part of the internal 
ear which is specially related to the Weberian mechanism, but the cavum sinus 
imparis and its atrial diverticula, as well as the relations of the latter to the scaphia, 
were perfectly normal. 
