ON THE ANATOMY OF FISHES. 
243 
such adventitious ossifications as those above mentioned are Invariably fonned in 
connection with the attachments of the bladder to fixed or movable parts of the 
skeleton, and that such attachments were orig-inally Avholly fibrous, but subsequently 
became invaded by bony deposit. 
The post-temporal plates, which so efficiently buttress the lateral portions of the 
anterior wall of the anterior chamber {e.g., Macrones, &c.) seem to have been formed 
by the extension of ossification from the inferior limb and stem of the post-temporal 
into the aponeurotic membrane and the lateral portions of the transverse membrane ; 
and in these instances {e.g., Macrones aor) where the wall of the bladder which is 
applied to these plates is exceptionally thin it is at least possible that ossification of 
the tunica externa has contributed to their growth. 
The various sub vertebral jDrocesses which probably subserve a similar function, in 
part, at least, owe their formation to an extension of the superficial ossifications of the 
complex centrum into the mesial portion of the transverse membrane. 
The peculiar crescentic distal extremities of the transverse processes of the fourth 
vertebra, which in some Siluroids {e.g., Macrones, Liocassis, Bagrus, &c.) partially 
encircle the post-temporal plates, are almost certainly formed by a similar extension of 
bony deposit from their proper vertebral portions into the lateral regions of the 
transverse membrane, the transition from the bony to the fibrous structure being 
readily traceable. 
Sorensen {loc. cit.) has shown that the terminal plates of the “elastic-spring” 
apparatus are often formed in a precisely similar fashion. In Malapterurus for 
example, he concludes that these structures are due solely to the ossification of the 
transverse membrane (or “ la plevre ”), but in such forms as Synodontis and Doras 
ossification implicates the outer stratum of the tunica externa of the air-bladder as 
well. Our own investigations enable us to state that Auclienipterus, Oxydoras, 
Pangasius djambal, and P. Buchanani resemble in this respect the two last-mentioned 
genera, while P. juaro and P. macronema, on the contrary, agree with Malapterurus 
in that the plates are formed by the ossification of the transverse membrane alone. 
The formation of bony deposits in the proper walls of the air-bladder or in its 
superficial coat has probably been determined by physiological considerations, and 
generally with the object of securing either a firm or more extensive connection of 
the walls of the air-bladder to adjacent fixed or movable skeletal elements, or a more 
effective buttressing of certain regions, and their significance from this point of view 
will be discussed in a future section of this paper. 
Although we have never been able to detect the presence of intrinsic muscular 
fibres in the walls of the air-bladder, powerful extrinsic muscles are present in several 
Siluroids. In Platystoma tigrinum, Pwiclodus maculatus, P. ornatus, and Piramu- 
tana piramuta, a powerful muscle takes origin from the posterior face of the skull, 
on each side of the foramen magnum, and is inserted into nearly the whole extent of 
the corresponding half of the ventral surface of the anterior chamber. As the 
2 I 2 
