ON THE ANATOMY OF FISHES. 
251 
affect its normal attachments to rigid portions of the axial, skeleton, except in certain 
instances where encapsulation by bone is more complete than is usually the case. 
In addition to considerable variations in the thickness of its walls, the relations 
of the air-bladder to the Weberian ossicles are subject to important variations in 
different genera and species. 
In certain genera {c.g., Bagarius, Glyptosternum, Euclyj^tosterniim, Ahjsis, and 
Acrochordonichthys) the dorsal wall of each air-sac has so far degenerated that it is 
extremely improbable that the few scattered fibres that alone remain can exert 
any influence on the tripus during the distension or contraction of the bladder. In 
most of the remaining genera, however much they may be modified in other respects, 
the walls of the air-bladder are structurally complete, and the convergence of the fibres 
forming the aiitero-lateral, lateral and dorsal walls of each air-sac to their insertion 
into the crescentic process of the tripus takes place in much the same way as in the 
normal Siluridce. There would seem, however, to be some definite relation between 
the extent to which the air-bladder retains its structural integrity, and its apparently 
functional connection with the tripodes, and the extent to which it is encapsuled by 
bone. When the dorsal walls of the air-sacs have undergone more or less complete 
degeneration, and have wholly or largely lost their normal attachments to the 
tripodes, the air-bladder is but slightly enclosed within comparatively shallow 
recesses or grooves on the ventral surfaces of the transverse processes of the fourth and 
fifth vertebrae. The partial encapsulation in such cases maybe simply due to the fact 
that reduction in the size of the only portion of the air-bladder that persists, viz., the 
anterior chamber, has been accompanied by a corresponding contraction and curvature 
of the modified transverse processes, which are normally moulded to the convexity of 
its anterior and dorsal surfaces. On the contrary, in nearly all cases in which the air- 
bladder retains much of its structural integrity, and, at the same time, its anatomical 
connection with the tripodes, it is always more completely encapsuled by bone than 
when the contrary is the case {e.g., Callomystax, Cetopsis, and Clarias). In such 
instances, the reduced air-bladder is either enclosed within deep groove-like recesses 
with greatly contracted outer or distal extremities, or within tubular or funnel-like 
cavities almost completely surrounded by bone, and the encapsulation always seems to 
be more complete than can be accounted for by the mere contraction of the associated 
skeletal elements round an atrophying air-bladder. 
Notwithstanding the reduced condition of the air-bladder, the Weberian ossicles 
almost always retain their normal size relatively to that of the Fish, but the compai-a- 
tively slight modifications which they present in other respects are invariably in the 
direction of degeneration. Claustra are occasionally absent, and even when present 
are but verv feebly developed spicules of bone. As a rule, the scaphium has no 
ascending process, hut only .spatulate and condylar processes. The intercalarium may 
be absent {e.g., Clarias), in which case the Interossicular ligament is extremely short, 
or is represented by a very small Ijony nodule in the usual position ; horizontal and 
