Evolution of Animals 455 



are unknown. We need not here speculate as to the nature 

 of horny tubercles foimd in some of the oldest rocks, and viewed 

 at times as the horny teeth of primitive cyclos tomes. 



While the Amphibia as a group has been accepted by most 

 evolutionists as contributing to the direct line of vertebrate 

 ascent, this has more been because it included attractive fossil 

 forms, often of large size, from some one phylum of which 

 reptiles may have been evolved. But, in conducting toward 

 animals like reptiles that formed scales or bony scutes, that 

 developed only one occipital condyle, that had a mandible 

 of five or more distinct pieces, that formed procoelous verte- 

 brae, and that remain oviparous, we seem wholly to have left 

 the natural pathway of advance. 



We would propose then that a more likely line of advance 

 would be from the Cyclostomata to the Amphibia apoda (Gym- 

 nophiona or Cseciliada), from these again to the Urodela, and 

 — even though the present-day gap be a serious one — from 

 the last to the simpler mammals. Considerable known re- 

 mains in the carboniferous rocks of Ohio, Ireland, and Bo- 

 hemia enable us to assert that batrachians of the group Aisto- 

 podes flourished then, and showed decided afBnities with 

 living Apoda. But the Aistopodes probably had devonian 

 ancestors, that could well unite them with more primitive 

 cyclostomatous progenitors. 



If we consider then the characters of li\^ng Apoda in the 

 above light, the view which has often been cursorily accepted, 

 that they are degraded urodele batrachians, is in no way justi- 

 fied. A detailed comparison can now be made of structural 

 characters that might favor such ancestral connections as 

 above suggested. 



The olfactory organ may first be dealt with from nemerteans 

 upward. 



In the latter two closely related structures arise on the an- 

 terior or antero-dorsal part of the snout, the cephahc glands 

 and frontal organ or organs. The former are saccular paired 

 glands that surround and often open into or beside the orifice 

 of the latter. The latter or frontal organ is usually single, 

 but in several rather primitive nemertean genera (Burger, 



