244 
PKOFESSORS T. W. BRIDGE AND A. C. HADDON 
contraction of these muscles must forcibly compress the anterior chamber, we have 
termed them the “ compressor muscles” of the air-bladder. They probably occur in 
many other Pimelodinse, but, so far as our investigations are concerned, are confined 
to that group. 
The presence of compressor muscles is invariably associated with the existence of a 
pair of much smaller muscles which arise from the exoccipitals, and are inserted into 
the anterior wall of the anterior chamber of the bladder. The tendon of each muscle 
has its insertion into the anterior wall immediately external to the complex centrum, 
and the insertion coincides with the extension of a slip of fibres from the inner 
surface of the anterior wall to the ventral ridge and concave inner margin of the 
crescentic process of the tripus. As the contraction of these muscles must evidently 
have the etiect of limiting the violent excursions of the tripodes which might other- 
wise take place when the anterior chamber is forcibly compressed by the contraction 
of its compressor muscles, we have suggested for each the name of “ tensor tripodls.” 
An “ elastic-spring ” apparatus, provided with powerful protractor muscles, has 
already been described by Johannes Muller as existing in the South American 
genera Auchenipterus, Doras, and Euanemus, and in the African Siluroids Synodontis 
and Malapterurus. To this list our investigations enable us to add the South 
American form Oxydoras hrevis, and the East Indian species Pangasius Buchanani, 
P. djamhal, P. juaro, and P. macronema. The absence of this mechanism in one 
species of Pangasius, viz., P. micronema, while present in all the remaining species 
of the genus that came under our notice, is an interesting and noteworthy fact. 
Two points in connection with this apparatus may be noted, first, the reception 
of the outer margins of the tripodes into grooves between the terminal plates and 
their flexible and elastic roots, whereby the forcible compression of the air-bladder 
which follows the recoil of the plates, is prevented from imparting a too violent shock 
to the Weberian ossicles and the fluids of the internal ear ; secondly, the geographical 
distribution of these fishes in which the mechanism is present, affords an additional 
instance of the singular and exceptional resemblances that have been noted between 
the South American, African, and East Indian faunas. 
The extent to which those species with an “ elastic-spring ” mechanism differ from 
other S. normales is subject to some variation. Thus, in Malapterurus, Pangasius 
juaro, and P. micronema, the transverse process of the fourth vertebra retains both 
its characteristic divisions, although the anterior on each side forms an elastic spring, 
and the terminal oval plates of the mechanism are formed by the ossification of the 
transverse membrane alone. In Pangasius djamhal, P. Buchanani, Synodontis, and 
Doras a further modification is introduced in the formation of the terminal plates 
from both the transverse membrane and the tunica externa, but the transverse 
process still retains a well-developed posterior division. Finally, Auchenipterus and 
Oxydoras, while agreeing with the last-mentioned Siluroids in the mode of formation 
of the terminal plates, differ from them in that the modified transverse process has 
