OCR Text |
Show 300 MB. G. A. BOULENGER ON SOME CBANIAL [Apr. 2, istic of Thymallus. For this reason Coregonus is retained in the Salmonidse and Thymallus is raised to family rank; the author adding that the Thymallidae are further distinguished from the Salmonidae by the presence of epipleurals and the greater development of the dorsal fin, m a n y of the anterior rays of which are unbranched. The last feature, by itself, is too trivial to be used as a family character; and the first characteristic is unfounded, Salmo and Coregonus having well-developed epipleurals. As to the condition of the parietal bones, it seems curious that there should be divergence of opinion on a point so easy to ascertain even in spirit-specimens. It is not stated to what species of Coregonus the skull belongs on which Gill's observation is based, but all the examples I have examined in the British Museum, skeletons and alcoholics, of European and American species of that genus, show the parietals forming a more or less extensive suture between the frontals and supraoccipital, as ascertained by Cope. I a m therefore at a loss to account for Gill's statement to the contrary. I have also examined Brachymystax, which, from its very small scales combined with a small mouth, holds a position intermediate between Salmo and Coregonus; the supraoccipital separates the parietals, as in Salmo. In his paper quoted above, Gill founds a subfamily, Stenodon-tinse, in the family Salmonidae, for the curious genus Stenodus, Rich. (Luciotrutta, Gthr.), without having been able, however, to examine the skeleton. A large head of the " Inconnu," Stenodus mackenzii, from the Mackenzie River, was presented by Mr. Walter Gordon Cumming to the Museum a few weeks ago, and I have had it prepared as a skull in order to supplement this gap in our knowledge of Salmonoid osteology. This skull is here figured. It will be seen that Gill's statement, " lower jaw articulating with the quadrates behind the eyes," is not quite correct, the articulation of the mandible falling below the posterior border of the orbit, in this respect intermediate between Salmo and Coregonus. The posterior process of the quadrate is very long and slender. The supraorbital is very large and in contact with the postorbital. But what is especially noteworthy is the condition of the supraoccipital and parietals, the latter bones forming a very short suture, separating the former from the frontals, a condition which is, in fact, intermediate between the two types mentioned above. O n examining a smaller specimen of Stenodus mackenzii and a specimen of the Russian S. leucichthys, I find no union whatever between the parietals, and therefore the character to which Cope and Gill have attached so great an importance falls to the ground. It must be remarked, however, that, as I have ascertained by removing the parietals in specimens of Coregonus williamsonii, the supraoccipital joins the frontal. Therefore even in Coregonus and Thymallus the relation is a widely different one from that obtained in Cyprinoids and Characinoids, the apparent resemblance being merely due to the fact that in the Salmonoids in question the |