THE CHONDROCRANIUM OF LEPIDOSTEUS. 255' 



most of the Teleostei is composed of two components, one of 

 latero-sensoiy and the other of membrane origin, that these twO' 

 components are found separated from each other in both AIoxo- 

 stoma and Ameiurus, and that the latero-sensory component 

 pei'sists as an independent, superficial, dermal ossicle in Moxostoma, 

 but has fused with the supraclavicula in Ameiurus. The temporal 

 fossa of Ameiurus would then be strictly similar to that of 

 Moxostoma, but greatly reduced in size, the muscle, or muscles,, 

 that primarily invaded and occupied it having been entirely 

 excluded from it by the progressive utilization of the fossa as an 

 articular cavity for the dorsal end of the supraclavicula. In 

 Macrones, the temporal fossa is apparently much more strongly 

 developed than in Ameiurus, and Bridge and Haddon ('93)' 

 definitely SRy that it serves as an articular cavity for what they 

 call the ascending process of the post-temporal, but which is, in 

 fact, the dorsnl end of the supraclavicula. In Silurus glanis, I 

 find conditions strictly similar to those in Macrones, and they 

 are indicated in Jaquet's ('98) figures of this fish. 



In Ameiurus the adductor mandibulse has invaded the dorsal" 

 surface of the cranium by passing upward over the ridge of t\\&- 

 lateral semicircular canal and external to the main latero-sensory 

 canal, and it occupies wdiat corresponds to the region of the 

 primitive supraotic depression. In Silurus it is the trunk-muscles 

 that invade this region of the dorsal surface of the cranium, but 

 they here pass mesial to the summit of the epiotic bone, thus- 

 traveising what corresponds to the mesial, instead of the lateral, 

 one of the two branch depressions at the hind end of the supra- 

 otic depression of Leindosteihs . 



In Conger conger there is a very definite supraotic depression, 

 but there is no vestige, even, of a branch depression leading 

 from it, either laterally or posteriorly, between the pterotic and 

 epiotic. The adductor mandibulse has here invaded the dorsal 

 surface of the cranium, as in Ameim'us, by passing upward over 

 the ridge of the lateral semicircular canal and external to the 

 main latero-sensory canal. 



In embr3^os of all of these teleosts there undoubtedly is a 

 supraotic depression similar to that in embryos of Lepidosteus, 

 for both Parker ('73) and Gaupp ('05) show such a depression in 

 embryos of Salmo. In Parker's fifth stage of Salmo (fry of the- 

 second week after hatching) there is as yet no noticeable post- 

 orbital process, but there is said to be, on the ventral surface of 

 the ridge of the lateral semicircular canal, a slight ridge which 

 " sets bounds externally to the facet for the extended head of the 

 hyomandibular." The conditions in embryos of the Holostei and 

 Teleostei aie thus probably strictly similar, but in the Holostei 

 the pterotic portion of the spheno-pterotic ridge persists as. 

 cartilage, while in the adults of the Teleostei above considered 

 the entire ridge is of bone, and this bone has so-called primary 

 relations to the chondrocranium. The ridge of the adult teleost 

 thus overlies the ridge of embryos, and is a ridge of the dermo- 



