THE ENDOSKELETON 157 



portion, the dermal palatine and pterygoid enlarge at its ex- 

 pense, and it is retained as an unimportant bit of cartilage 

 in some amphibians, but beyond these it is not seen. The two 

 dermal pieces, which originally encased it, however, retain con- 

 siderable importance and usually appear in the skulls of am- 

 niotes along a curve approximately parallel to that of the 

 maxillaries, but interior to it, thus marking the former po- 

 sition of the lost cartilage. These bones are often tooth- 

 bearing in fishes and amphibians, retaining this much of their 

 former function. 



Of the five pairs of true branchial arches, the first four 

 are retained in ganoids and teleosts as gilf-bearers, while the 

 fifth lies in the floor of the pharynx and is often covered 

 with sharp teeth arranged in several rows. 



A notable modification, which, however, is mainly an ex- 

 ternal one, is found in the development of the opercuhim or 

 gill-flap, which appears as a fold of skin and becomes rein- 

 forced by special dermal bone. This ultimately develops pos- 

 teriorly so as to cover all the gill-slits in such a way that they 

 seem to be internal, and are not visible from the exterior, as 

 in the case of most selachians. 



The subsequent history of the visceral skeleton and its fate 

 in amphibians, reptiles, birds, and mammals may be quickly 

 outlined. These Classes are fundamentally terrestrial, and 

 never possess genuine internal gills, and thus the main changes 

 are due to a loss of function, which, by throwing the branchial 

 arches out of employment, would have caused them to disap- 

 pear, were it not that they could be in part employed to sub- 

 serve some other necessary function. It will be seen that in 

 most cases this has been their fate, but this very change of 

 function, although it saves them from complete destruction, 

 of necessity causes considerable modification, oftentimes a pro- 

 found one. 



Beginning with the most anterior arches, the mandibular 

 and hyoid, the hyomandibular seems to disappear entirely, al- 

 though doubtfully identified by some morphologists with cer- 

 tain other elements in the otic region, and leaves to the quad- 



