— 195 — 



Swinnerton, it should here be stated, has arrived, in a study of the development of Gasteros- 

 teus, at conclusions quite difEerent from those just above proposed, in so far as regards the position 

 of the mvodorae reUxtive to the cranial walls in fishes, and the homologies of the hypophysial fenestra. 

 According to him ('02, p. 527) those parts of the parachordals that, in embryos of Gasterosteus, bound 

 laterallv the interparachordal fossa, become depressed in late embryonic stages, so as to appear as 

 mere downward processes of the proötics. These processes are said to be cajjped with cartilage, to 

 eaeh be continued posteriorly by a ridge on the ventral surface of the posterior portion of the related 

 proötic, and, posterior to that bone, by a similar ridge on the ventral surface of the basioccipital. 

 The two processes are said to enclose between themselves the anterior portion of the myodome, which 

 portion is said to accordingly be an actual derivative of the cranial cavity ; while the two ridges that 

 form posterior continuations of the processes enclose a posterior portion of the myodome, which 

 is said to be extracranial in position. These conclusions lead Swinnerton to the further conclusion 

 (1. c, p. 528) that the so-called hypophysial fenestra of the skull of adult teleosts is related to the 

 parachordals, alone, and hence can not be the homologue of the pituitary (hypophysial) fossa of 

 embryos, which fossa is related to the bind ends of the trabeculae. The so-called hypophysial fenestra 

 of the adult fish can not then be the homologue of the hypophysial (pituitary) fenestra of higher 

 animals, and Swinnerton accordingly proposes for this fenestra of the adult teleost the name inter- 

 parachordal fossa. The position, in the adult, of the pituitary fossa of embryos, Swinnerton does 

 not give; the inference being that it has wholly disappeared in that suppression of the hind ends of 

 the trabeculae that is said to take place in late embryonic stages.^) 



Those teleosts in which the myodome is said to be absent can now be considered. Vrolik 

 ('73) says that it is absent in all the Gadidae, and also inSilurus, Lophius and the eel; Gill says ('91b, 

 p. 363) that it is absent in the Cyclo pteroidea, and also ('82) in Echeneis; Swinnerton ('02, p. 576) 

 says that it is absent in Fistularia and Syngnathus; and Jordan and Evermann ('98) say that the 

 basis cranii is simple in the Hemibranchii and Lophobranchii, which include Fistularia and Syn- 

 gnathus; Cope (quoted by Gill, '88, p. 576) says that it is absent in all the fishes of the groupScypho- 

 branchii, which group (Cope, '71) includes the Uranoscopidae, Gobiidae, Blenniidae, Gobiesocidae 

 and Cottidae; Gierse ('04) says that it is wholly wanting in Cyclothone; and Starks ('05 a) says that 

 it is absent in Caularchus, Callionymus and the Batrachididae, confirming also its absence in the 

 Gobiesocidae. McMurrich ('84) says that it is rudimentary in Ameiurus; and Sagemehl ('91, p. 574) 

 says that it has undergone retrogression in Cobitis, Misgurnus, Nemachilus and Acanthophthahnus. 

 Boulenger ('04) says that the basis cranii is simple in the Mormyridae, Osteoglossidae, Pantodontidae, 

 Phractolaemidae, Stomiatidae, Gonorhynchidae, Cromeriidae, Galaxiidae, Gobiiformes, Discocephali, 

 Comephoriidae, Khamphocottidae, and in all the five families of his suborder Pediculati. He further 

 says that the basis cranii is double in the symmetrical forms of his division I of the suborder Acantho- 

 pterygii; which woiüd seem to imply that it is simple in the asymmetrical forms of the same division, 

 that is the Pleuronectidae. And the expression ,, basis cranii simple", while it is, strictly speaking, 

 descriptive of a condition of the bony skull alone, is currently considered as equivalent to saying 



') Gaupp, in Bd. 3 of Hertwig's Handbuch der vergleithenden und e.xperimentellen Entwickelungslehre der Wirbel- 

 tiere, a work that I have only seen since this manuscript was sent to press, describes practically similar conditions in 

 Salmo, and arrives at practically similar conclusions regarding the homologies of the parts. This would seem to establish 

 the fact that the basi-occipital portion of the myodome is extracranial in origin. Regarding the prootic portion of the 

 myodome, Gaupp's descriptions would .seem to confirm my contention that it is an intramural space and not an intra- 

 cranial one. 



