ON THE ANATOMY OF FISHES. 
257 
aperture leading into the corresponding osseous capsule. It is difficult to judge from 
Reissner’s description how far coalescence has affected the anterior vertebrae, and 
whether any forward dislocation has taken place. His “ first ” vertebra, as we have 
pointed out (p. 178), is probably a “complex,” which may include the first to the 
fifth vertebrae, inclusive, bnt there is also the alternative that the vertebra has the 
same value as an ordinary “complex” in most other Siluridae, although the possibility 
that the centrum of the first vertebra may also be included in it cannot altogether 
be left out of sight. We are of opinion that Reissner’s “first” vertebra will ulti- 
mately be found to possess the same significance as in Loricaria and its allies, but a 
careful re-examination of this curious Siluroid is necessary before perfectly trustworthy 
conclusions can be arrived at. The cavum sinus imparis, the atrial cavities, and the 
sinus endolymphaticus are apparently normal so far as their mutual relations are 
concerned, and lie on the upper surfaces of the basioccipital and the anterior portion 
of the “ first ” vertebra. The relations of the scaphia to the lateral walls of the two 
atria would also seem to be normal. The dorsal wall of the cavum sinus imparis, 
however, must be fibrous, or at all events the cavum is not roofed in by horizontal 
plates from the exoccipitals ; and in the fact that the sacculi occupy mere depressions 
in the cranial surface of the basioccipital, and are not surrounded by almost complete 
osseous capsules, Acanthicus also differs from Ilypophthalmus and most other abnormal 
Siluridae, and resembles the Loricaroid types. The relative positions of the sacculi, 
the cavum sinus imparis, the sinus endolymphaticus, and the atrial cavities almost 
negative the possibility of any forward dislocation of the anterior vertebrae into the 
cranial cavity having taken place. Claustra and iutercalaria are probably absent, 
or at all events are neither figured nor mentioned by Reissner. The scaphimn, as 
in some other S, abnormales and in the Loricaroid genera, has neither condylar nor 
ascending processes, and, as is the case with the great majority of the abnormal types, 
the tripus has a straight or but very slightly curved “crescentic” process. 
From this summary of the more salient features in the structure of its air-bladder 
and Weberian mechanism, it is clear that as regards the condition and mutual 
relations of the cavum sinus imparis, the atrial cavities, and the saccular recesses, 
the existence of a “processus bijugus,” and possibly also in the relations of the 
anterior vertebrae to the skull, Acanthicus closely resembles the “Loricaroid 
genera and differs from Ihjpophthulmus. In the mode of formation of the osseous 
capsules for the air-sacs, however, Acanthicus, while peculiar among the Siluridae, 
distinctly resembles, as Reissner suggests, the Cyprinoid Colitis fossilis. On the 
other hand, the peculiar position of the posterior halves of tlie two air-sacs and 
their bony capsules, dorsad to the first pair of ribs in Acanthicus, has no parallel 
in any known Ostariophyseae. 
The last of these extremely aberrant Siluroids, IJypophthalmus, exhibits an extreme 
type of modification, of which the most noteworthy features are the complete fusion of 
2 E 
MDCCCXCIII. — 15. 
