ON THE ANATOMY OF FISHES. 
301 
various problems arising out of the physiology of the air-bladder, through anatomical 
data alone, is again forcibly illustrated, for it is impossible entirely to exclude the 
possible relation of the extrinsic muscles of the Pimelodinse to the function of sound 
production, and it may also be the case, although perhaps less likely, that the 
same reservation will also apply to the “ elastic spring ” mechanism. That a violent 
expulsion of air from the air-bladder should produce definite sounds is extremely 
probable, but how far such sounds can be considered as related to the primary 
function of these muscles, or as merely accidental concomitants to it, must for the 
present remain an open question. Sorensen (36) has adopted the former suggestion, 
and regards both the compressor muscles and the “ elastic spring ” mechanism as 
being subordinate to sound production. Nevertheless, in the absence of confirmatory 
experimental evidence, we still think it worth while to direct attention to an 
alternative interpretation of the function of these structures, which is at least as 
consistent with their morphology as any other view at present suggested. We have 
elsewhere (p. 270 and p. 298) suggested certain difficulties, which, in our opinion, 
are serious objections to Sorensen’s views on this point. 
{(I.) The SiluridcB compared with other Ostariophysece. — In our morphological 
summary, and elsewhere, we have drawn attention to the more salient features 
wherein the air-bladder and Weberian mechanism of the Siluridae differ from the 
corresponding structures in those Ostariopliyseae which in these, as in other respects, 
are the most closely allied to them, viz. : — the Cyprinidae, and we may now briefly 
consider such distinctions from a purely physiological standpoint. As the morphological 
result of a comparison of the two families was to demonstrate the increased specialisa- 
tion of these structures in the Siluridae it is highly probable that a similar comparison, 
from a functional point of view, will prove the increased sensitiveness of the latter to 
hydrostatic pressure variations, and, at the same time, suggest a corresponding 
advance in the accuracy of the readjustment processes. An important factor in this 
direction in the Siluridae is the restriction of all variations in the capacity of the 
anterior chamber under the influence of pressure variations to lateral expansion or 
contraction, from which it must follow that the auditory organ will be rendered 
susceptible to smaller variations of pressure than is the case in the Cyprinidae where 
expansion and contraction are less strictly localised, and probably take place to a 
nearly equal extent in all directions. To the same object must also be attributed the 
subdivision of the posterior section of the air-bladder in all noimial Siluridae into two 
lateral compartments by a vertical longitudinal septum, as well as the frequent 
presence of secondary transverse septa in each chamber, such structural features can 
scarcely have any other effect than that of rendering this portion of the air-bladder 
more inelastic and indistensible than it would otherwise be, and, therefore, still 
further restricting all volumetric variations to the anterior chamber alone. It is also 
o 
possible that the presence of lateral cutaneous areas may be related to the increasing 
sensitiveness of tlie air-bladder of the Siluridiu to pressure disturbances, if, as 
