THE ANATOMY OF FISHES. 
153 
teristic crescent. The concavity of the process is related to the dorsal lamina in 
such a way that it lies to the outerside of the latter, and at the same time partially 
encircles it (fig. 55). The scaphium has no ascending process, the small rounded 
condyle for articulation with the dorso-lateral edge of the centrum of the first 
vertebra being at the contracted posterior extremity of the spatulate process. The 
intercalarium is a very small nodule of bone imbedded in the interossicular ligament. 
The claustrum is either absent altogether, as appeared to be the case in one or two 
of our specimens, or, as in others, was apparently represented by a slender spicule of 
bone growing downwards from the lower border of the inclined neural arch of the 
complex vertebra towards the body of the first, and imbedded in the fibrous wall of 
the neural canal immediately behind the exoccipital. 
As in Bagarius we had no difficulty in discovering a ductus endolymphaticus, but 
of a sinus endolymphaticus in the relatively small cavum sinus imparis we were 
unable to find any indication whatever. Our specimens of Glypto sternum were 
sufliciently well preserved to render our failure in the latter respect only explicable 
on the assumption that in this genus the sinus endolymphaticus has undergone com- 
plete suppression. 
It is therefore obvious, from a comparison of the two forms, that as regards the 
condition of the air-bladder, and the nature of the associated skeletal modifications, 
G Lyptosternum and Bagarius are in close agreement on all essential points, while 
differing to some extent in minor details, and further, that both are widely different 
from any of the normal species hitherto described. The better preserved condition 
of our examples of these genera enabled us to make a more satisfactory examination 
of the structure of the air-bladder and internal ear than was possible in the case of 
the similarly modified genera Akysis and Acrochordonichthys, and it may be worth 
while to institute a brief comparison between the former and those Siluridae which 
possess air-bladders of the normal size and structure. 
The more salient features in the air-bladders of Bagarius and Glyptosternum are 
clearly the result of degeneration, and can easily be conceived as due to the retro- 
gressive modification of an originally normal organ. The two lateral air-sacs 
obviously represent the completely separated lateral halves of an ordinary anterior 
chamber longitudinally constricted from each other, the lateral compartments, always 
the most variable portion of the air-bladder in point of size, and the ductus pneu- 
maticus having undergone total atrophy. No trace of the original communication 
between the two sacs persists unless a remnant of such a connection is to be found in 
the thin solid stratum of fibres which is continued from one sac to the other across 
the ventral surface of the centrum of the complex vertebra. But notwithstanding 
such indications of degeneration, it may be noted that the reduced air-bladder still 
retains the usual and normal attachments to fixed portions of the axial skeleton. 
Anatomically, at all events, the rigid skeletal attachments of each air-sac are 
capalfie of strict comparison with the similar connections of the walls of each half of 
MDCCCXCIII. — B. X 
