45] THE SKULL OF AM I URUS— KINDRED 45 



the inner perichondrium forms the walls of the cavum sinus imparis. The 

 outer lamella is unpaired and is thicker in the middle line than on the sides, a 

 contrast to the paired outer lamellae of Salmo. The paired rectus externus 

 muscles are located in this region in Salmo. Further posterior in Salmo, the 

 outer lamella has a spongy appearance, comparable to the appearance of the 

 osseous lamella in the same region of Amiurus and which Schleip says arises 

 from the ossification of fibrous connective tissue in that immediate region in 

 connexion with the perichondria] ossification of the basal plate. From this 

 comparison, I have concluded that the ossifications in this region are homolo- 

 gous and that they are the centres of the basioccipital of Amiurus as they are 

 of Salmo. 



The osseous lamella around the glossophar5mgeaI and vagus foramina and 

 the ventral ends of the posterior semicircular canal and above the recessus 

 sacculorum and cavum sinus imparis, have been described by Schleip and 

 Gaupp as the centres of the exoccipitals. In Amiurus, the part above the 

 sinus impar is preformed in membrane and only secondarily connects with the 

 perichondria! ossification and hence cannot be exoccipital, though fused with 

 it. The hypoglossus, in both Salmo and Amiurus, was not included within the 

 cranium at an earlier stage, but is now enveloped in an osseous sheath between 

 parachordal plate and the occipital arch. 



The otic capsules. The cartilage enclosing each membranous labyrinth 

 has grown considerably since the 10 mm. stage, but the relations of the septa 

 semicirculares (Fig. 7) have remained the same. The detailed description of 

 these, given for the younger stage, will also fit the 32 mm. stage and the adult, 

 except for a change in size. The parts which merit description at this 32 mm. 

 stage are the centres of the otic bones, which, in the teleosts, include prootic, 

 sphenotic, pterotic, epiotic, and opisthotic. All except the last are present in 

 Amiurus, according to McMurrich (1884b), and the problem is to locate and 

 describe them in their earliest form. 



These bones were first grouped as the otica by Huxley (1864), who called 

 attention to the fact that they were developed around the otic capsule, which, 

 as cartilage, had an independent origin and only secondarily became connected 

 with the parachordalia and the occipital arch. Hence these bones formed a 

 natural group in comparison to the other bones of the cranium. Vrolik (1873) 

 objected to grouping of these bones as otics, and, because of the relation of the 

 occipital bones to the labyrinth, maintained that these also could be included 

 with the otica. As Van Wijhe (1882) later pointed out, this was due to a mis- 

 understanding of the original statement of Huxley, because it is a well-known 

 fact that the membranous labyrinth invades other bones than those which are 

 developed around the capsule. Vrolik also claimed that the otic bones took 

 no part in the formation of the cranial wall because he found the opisthotic 

 to be variable bone, sometimes developed in the cartilage of the capsule wall, 



