164 
PROFESSORS T. W. BRIDGE AND A. C. HADDON 
aduiits of the passage of the tripus from the interior of the cylinder to its connection 
with the scaphium anteriorly. The transverse processes of the fifth vertebra have no 
share in the formation of the osseous capsules for the air-bladder, nor do they in any 
way differ from the normal rib-bearing processes that succeed them. 
Except for the slenderness of its inferior limb the post-temporal is normal, and with 
the distal portion of the cylindrical transverse process forms the articular cavity for 
the clavicle. 
file air-bladder is constricted into two small, extremely thin-walled, laterally 
placed, oval air-sacs, which occupy the slightly inflated proximal portions of the two 
bony cylinders. Each sac in a Fish 8 inches long was only 6 mm. in length, but, 
nevertheless, had structurally perfect walls. The walls of each sac are in close 
relation with those of its investing bony capsule, without, however, being specially 
attached thereto, except at one point where the inner wall is adherent to the lateral 
surface of the complex centrum. The solitary specimen examined by us had been 
eviscerated before dissection, but a most careful examination failed to reveal the 
existence of any communication between the two sacs, or of any trace of a ductus 
pneumaticus. 
A scaphium without either condylar or ascending processes is present, but neither 
intercalaria or claustra could be detected. The tripus is very similar to that of 
Bagarius, except that it has only the barest rudiment of a straight “ crescentic ” 
process for insertion into the antero-internal portion of the dorsal wall of the 
corresponding air-sac. No obviously specialized radial fibres could be made out, but 
are probably represented by that part of the inner wall of the air-sac which is 
attached to the complex centrum. There are no radial nodules. 
The cavum sinus imparls and the atrial cavities aj^peared to be perfectly normal, 
but relatively of small size. Unfortunately, the auditory organ was too badly 
preserved to admit of a satisfactory examination of its structure being made. 
Doras maculatus. 
The existence of an “ elastic-s})ring ” apparatus in this South American species was 
first described by Johannes Muller (28), but a more detailed account of the relations 
of the mechanism to the walls of the air-bladder has recently been given by Sorensen 
(37, Plate 2, figs. 1 5, 16). In an earlier paper by the same author (36, Plate 3, fig. 44, 
and Plate 4, fig. 43), the bladder itself is figured. 
From Sorensen’s figures it would appear that as regards its skeletal modifications, 
and the mode of formation and relations of its “ elastic-spring ” mechanism, Doras 
exhibits a fairly close resemblance to Aucheiiipterus and Oxydoras in all essential 
features. The thick discoidal terminal plates of the mechanism are closely applied to 
the antero-lateral regions of the bladder. There seems to he no recognisable posterior 
division to the transverse process of the fourth vertebra, and the fifth vertebra has 
