246 
PROFESSORS T. W. BRIDGE AND A. C. HADDON 
structure of the air-bladder iu these forms are for the most part but scanty, and, as a 
rule, are only sufficient to enable us to infer that the organ is normally developed, 
although in several instances they certainly furnish additional illustrations of the 
polymorphic condition of the air-bladder in different Siluroids. Skeletal modifications 
are but seldom referred to except by Muller in the case of Synodontis, Euanemus, 
and Doras where an “ elastic-spring ” apparatus is present, and in no instance is 
there any reference to those portions of the internal ear which are specially related to 
the Weberian mechanism. 
With regard to the genera Hava, Olyra, Pseude,utropius (Day, loc. cit.) and Chaca 
(Cuv. and Val., loc. cit.) nothing is known beyond the fact that the air-bladder is well 
developed and not enclosed by bone. Platystomci fasciatum, P. cortiscans, and Callo- 
2 )hysus are remarkable for the variable development of the curious csecal appendages 
attached to the air-bladder. We have elsewhere described the antero-lateral caeca of 
Platystomci tigrinum, and according to Johannes Muller {loc. cit.) precisely similar 
structures are also present in P. fasciatum, but in P. corv.scans it would appear, on 
the same authority, that such appendages are entirely wanting. On the other hand, 
in Callopliysiis caecal appendages are exceptionally well developed and form an 
elegant wreath surrounding the lateral margins of the bladder. In certain other 
Pimelodinae these structures are entirely absent. With the exception of the 
Pimelodinae the only other Siluroid known to possess such appendages is Doras 
maculatus, which, according to Sorensen {ante, ju 165), has a fringe of branching caecal 
tubules surrounding the lateral margins of the air-bladder. Among the variations in 
the structure of the air-bladder recorded by Muller may be mentioned the trans- 
verse constriction of the organ into three intercommunicating and longitudinally 
arranged compartments in Ariiis emphysetus — a variation which was certainly not met 
with in any of the Ariinae examined by us. In the Pimelodinae Piratinga fila- 
mentosa is also said by Muller to have the air-bladder divided into two sacs, lying 
one behind the other ; both are said to be cellular, but while the anterior sac is in 
connection with the oesophagus by a ductus pneumaticus, the posterior one has no 
communication either with the oesophagus or with its fellow in front. 
The presence of muscles in connection with the air-bladder is confined, so far as our 
experience is concerned, to the Pimelodime, but, according to Cuvier and Valen- 
ciennes {loc. cit.), powerful extrinsic muscles are also present in Arius Milberti and 
A. codatus. A. Milberti we have had no opportunity of examining, but in A. ccelatus 
thei’e is no trace whatever of such muscles, and, as the same remark apjilies also to 
every one of the eleven species of Arius dissected by us, we fancy that the statement 
quoted is as erroneous in the case of A. Milberti as it certainly is with regard to 
A. ccelatus. The same writers have also described extrinsic muscles in connection 
with the air-bladder of AElurichtliys Gronomi, of which two on the ventral surface are 
said to be able to compress the bladder. The only species of this genus {A. longispinis) 
that we dissected assuredly had no such structures. 
