OEGAN OF THE SILUROID HYPOPHTHALMUS. 109 



siderable size, while the second and third are much reduced and fused with the fourth, 

 which is of great size, and has very large transverse processes. Theseroof over the anterior 

 part of the air-bladder, and articulate anteriorly with the supraclaviculars. Although the 

 fifth vertebra is fused in the adult with the fourth, it does not contribute to the formation 

 of the Weberian apparatus. In determining the homologies of the osseous parts of the 

 apparatus, the spinal nerves form excellent landmarks Of these, the first pair (hypoglossal 

 aud) emerge through the exoccipitals, the second and third in the interval between the 

 stapedes and the arch of the third vertebra, while the fourth and fifth penetrate the arches 

 of the third and fourth vertebrae respectively. 



How far my interpretation of the homologies of the Weberian ossicles differs from the 

 results arrived at by other authors has been already indicated in No. 11, p. 248. I shall 

 take occasion to refer below to certain divergences between the views just expressed 

 and those embodied in Sagemehl's paper, but I may first point out how generally the 

 numbers of the anterior vertebrœ have been misreckoned by authors dealing with the 

 parts in question. Thus Eeissuer (No. 5 Fig. 9) figures as the first vertebra of Loricaria 

 calaphrada, a complex, the greater bulk of which is formed of the fourth vertebra. Groldi 

 commits the same error (No. 3, p. 414), and "Weyenbergh figures (No. 8, PI. IX. Fig. 31), as 

 the first vertebra of H//posfomus plecostomus, what is unquestionably its fifth. These errors 

 are attributable to the extreme reduction of the anterior vertebra?, which appears to be 

 carried in some respects further than in Hypophthalmiis, although the union with the 

 occipital region is more complete in the latter. This reduction is not such in Amiurus as 

 to affect materially the spinal nerves, but it does so in Hypophthahnus, audit is extremely 

 probable that investigation will show that the second and third spinal nerves in the 

 Hypostomatiua are profoundly affected by the vertebral concrescence in this region. I 

 observe in the arches of what is termed, by Keissuer and Goldi, the first vertebra, two 

 series of foramina, which I have no doubt serve for the escape of the fourth and fifth 

 nerves. 



In some respects, Sagemehl's interpretation of the homology of the Weberian ossicles 

 differs from that adopted above. The incus, for example, is interpreted as the rib of the 

 second vertebral segment, and the claustrum as a modified occipital arch. As I have else- 

 where remarked, the Siluroids (Amiurus at least) do not afford the necessary material for 

 determining the homology of the incus on account of its reduced condition, but its ascend- 

 ing process appears to me to enter into the wall of the neural canal in Catostomus, in 

 which case it must be a modified neural arch. I have not had the opportunity of settling 

 this definitely by examining the relationship of the third spinal nerve to that process. 

 With respect to the homology of the claustrum, which I have been inclined to regard as a 

 modified " tegular-stiick," I shall merely observe that the additional spinal nerve issviing 

 between the claustrum and stapes in Silurus (cf. Sagemehl, p. 5G) is certainly quite unre- 

 presented in Amiurus. 



In Sagemehl's memoir, so frequently cited, will be found (p. 0) a synopsis of those 

 Nematognathi which have been supposed to be destitute of an air-bladder, and, conse- 

 quently, of a "Weberian apparatus. Joh. Midler refers the genera Cetopsis, Arges, Brontes, 

 Loricaria, Ehinelepis, Hypostoma, Callichthys to this category ; Valenciennes adds to the 

 list, Hypophthalmus, Bagarius, Grlyptosteruum, Trichomycteres, Eremophilus ; Reissner 



