562 H. H. SWINNERTON. 



of Stohr on iSiredon and Triton (81, p. 85), Rana (82, p. 85), 

 and the salmon (83) ; those of Pouchet on Labrus, Gobius, 

 Atheriua, Syngnathus (78) ; those of Miss Piatt on Necturus 

 (96, p. 455) ; and from my own on Gasterosteus, it becomes 

 evident that the horizontal position is exhibited by precisely 

 those groups which^ according to Balfour, have only a weak 

 cranial flexure. 



Amongst those fishes whose development has been fully 

 worked out only the Elasmobrauchs, viz. Acantliias (Sewert- 

 zoff, 99, p. 285), Prestiurus (p. 300), Kaia and torpedo (ibid., 

 97, p. 417), exhibit the vertical position for the trabeculse. 

 The presence of a post-clinoid wall may be taken as a strong 

 indication of the former existence of the same type in many 

 of the other Elasmobrauchs. 



This may apply, but with diminished force, to the Chimse- 

 roids and Dipnoi. 



There can be little doubt that the horizontal condition is 

 the more primitive one, and that the vertical is secondary, due 

 to a strong cranial flexure, which in its turn is explained by 

 Balfour, probably correctly, " as being associated with an 

 embryonic as opposed to a larval development ; " and with 

 some advantage to be " gained by a relatively early develop- 

 ment of the brain" (81, p. 267). 



In this respect, therefore, the Cyclostomata, Ganoidei, 

 Teleostei, and Amphibia have retained a more primitive con- 

 dition than the Selachii, and if their ancestral stock was in 

 any way related to these, it must have been to some form 

 whose foetal life was not yet sufficiently lengthened out to 

 permit the development of a strongly-marked flexure, and 

 the associated ventral position of the trabecular. 



The similarity in the condition of the brain and the 

 trabeculaa in Amniota to that in Selachii must be regarded as 

 due rather to convergence than to genetic affinity. 



The Primordial Cranium. 

 As already mentioned, Gegenbaur^s work on the cranium 



