218 
PROFESSORS T. W. BRIDGE AND A. C. HADDON 
anteriorly ; the whole of the posterior half of this portion is cellular ; and so is the 
small intermediate space between the two uncelled pyriform portions. The two 
posterior divisions of the air-vessel have valvular shaped folds, partially subdividing 
its interior.” As regards the subdivision of the bladder, the two accounts are neither 
reconcilable with each other, nor with the desci'iption given above by us, and 
assuming that the specimens examined by Day and Taylor were really examples 
of P. Buchanani, we can suggest no explanation of the discrepancy. Our own 
specimen was one from Dr. Day’s collection of East Indian Siluroids, and had been 
named by him P. Buchanani, and we may also add that its characters agreed in 
almost every detail with the description of this species in the British Museum 
Catalogue of Fishes (vol. 5, p, 62). It will be noticed, however, that both Taylor 
and Day agree as to the extension of the air-bladder into the caudal muscles, which 
was certainly not the case in the specimen examined by us. 
The scaphium is normal. The intercalarium is a relatively small nodule in the 
interossicular ligament without horizontal or ascending processes. The crescentic 
process of the tripus is somewhat unusually broad and flattened. From its root 
backwards the process is directed at first slightly outwards, and then curves inwards 
and downwards in contact with the lateral surface of the complex centrum, and 
finally in the form of a pointed process, extends slightly forwards along the ventro- 
lateral margin of the centrum towards the radial nodule (fig. 92, tr.c.). A ventral 
ridge traverses the root of the crescentic process, as in Macj^ones, and, anteriorly, 
receives the insertion of a slip of fibres derived from the mesial portion of the 
anterior wall of the bladder (fig. 92), and posteriorly the attachment of the fibres 
converging from the antero-lateral and lateral walls into the dorsal wall of the 
anterior chamber. Strong radial fibres {)'./.) pass from the inner margin of the 
ridge, and from the concavity of the crescent, to the radial nodule. The crescentic 
process is in close relation with the ventral surface of the flexible root of the “ elastic- 
spring” apparatus, and it seems to us that the ventral curvature of the latter must 
restrict the lateral movements of the tripus within very narrow limits, except when 
the protractor muscle pulls each oval plate forwards, and consequently widens the 
curvature of its elastic root. Hence, as in Auchenipterus, the mobility of the tripus 
may be to some extent controlled or regulated by the action of the “ elastic-spring” 
apparatus. The claustra are exceptionally large and somewhat triangular ossicles, 
with their apices extending upwards into the intercalated cartilage, which largely 
persists between the spinous process of the third vertebra and the exoccipitals and 
supraoccipital, and their broad bases resting on the lateral margins of the thick mass 
of fibrous tissue which forms the posterior wall of the two atrial cavities. 
In the structure and relations of the auditory organ, and of the cavum sinus 
imparis and atrial cavities, Pangasius closely resembles Macrones neinurus and 
other normal Siluridm. 
