32 THE SALAMANDER 



which, as already mentioned, he considers homologous with the pos- 

 terior portion of the amphibian parasphenoid.^ 



Summary. 



Reviewing then the skull as a whole, particularly in comparison with 

 that of the Frog, the following important differences may be noticed. 



(i) The skull generally is more compactly built than in the Frog. 



(ii) The nasal capsules are more completely roofed over, since the 

 nasals are pushed forwards by the pre-frontals, which are not present 

 in the Frog. 



(iii) 'X'\\Q pre-vomers have a much greater expansion and thus form 

 a complete bony floor to the nasal capsules, while their dentigerous 

 portions extend far back into the mouth. 



(iv) Thcfrontals 2ind parietals remain distinct elements and do not 

 fuse as in the Frog, neither do they extend so far ventrally round the 

 sides of the brain case. 



(v) The brain-case is relatively wider in Salamandra than in Rana, 

 when compared with the width of the skull as a whole. 



(vi) The feature just noticed is due largely to the much greater 

 development of the suspensorium in the Frog. In Salamandra the 

 squamosals are mere splints on the dorsal surfaces of the palato- 

 quadrates, while the quadrato-jugal sends no process to meet the 

 maxilla. The pterygoid likewise does not join the maxilla so that 

 the arch of the upper jaw is incomplete. 



(vii) On the ventral surface the parasphenoid is relatively broader 

 in Salamandra, so that the sides of the brain-case — orbito-sphenoids 

 — are more nearly vertical. 



(viii) The anterior rim of the orhit on the ventral side is formed by 

 the edge of the pre-vomer in Salamandra, and by the palatines in Rana, 

 the latter elements being probably absent in the adult Salamander. 



(ix) In the ear capsule the columella auris is vestigial in Sala- 

 mandra, its place being taken by the cartilaginous operculum^ which 

 has no direct connexion with the skin, while in Rana the columella 

 is attached to the tympanic membrane and there is no operculum. 



III. THE VISCERAL ARCHES. 



I. The Mandibular (First Visceral) Arch. (PI. II, fig. 7, and PL 



III, figs. 8 and 9.) 



The shape of the lower jaw as a whole is a simple bony arch, 

 almost semicircular in plan. At the median symphysis there is a 



J See also Watson's criticism of this hypothesis, Journ. Anat. London, vol. liii, 

 pp. 239-40. 



