Evolution of Animals 459 



posteriorly to the second ceratobranchials in the former, and 

 on the ventral side of the posterior cornua of the hyoid in the 

 latter." 



The oesophagus in cyclostomes is continued by a rudimentary 

 cardia into the straight intestine. But, though a definitely 

 enlarged stomach is absent, the researches of Alcock and Haack 

 prove that functionally it is represented by a considerable 

 part of the oesophagus and intestine, whose gland cells secrete 

 pepsin. The morphological resemblance of the above to the 

 condition in nemerteans is more than striking, while func- 

 tionally the remark of Shipley for nemerteans (H9: 366) is 

 equally so: "There is some evidence that in this group the 

 ectoderm of the oesophagus is chiefly concerned with digestion, 

 whereas the endoderm of the intestine is limited to the absorp- 

 tion of the soluble products." 



In the simpler urodeles the stomach may be little larger 

 than the oesophagus as in SireUy or may be decidedly swollen 

 as in Amblystoma, while the intestine is slightly or only mod- 

 erately coiled on itself. In the higher urodeles the stomach 

 is usually more enlarged and the intestine is mostly much 

 coiled. Liver, gall bladder, and pancreas are a common heri- 

 tage with all the craniates. From the much elongated liver 

 of the Cyclostomata and iVpoda to the at times elongate one 

 of simpler Urodela, easy transition is made to the usually 

 short compact liver of higher urodeles, and of mammals. 



The purely cartilaginous brain-box of cyclostomes, to which 

 no lower jaw is attached, represents the simplest craniate con- 

 dition, though the opinion often expressed that the lower jaw 

 is here atrophied has only a little evidence in its favor. 



The advance shown in Apoda is most marked, and would 

 indicate that between the higher cyclostomes, that almost 

 certainly must have flourished in the early devonian or even 

 in the silurian, and the aistopodous or related batrachians 

 of the carboniferous great specialization had taken place. 



In this connection the permian genus Lysorophus, as recently 

 expanded in description by Case {150, n. 14-6: 141), seems in 

 its known characters to connect cyclostome and apodous types. 

 Thus there "is no distinct atlas or axis, and the vertebrae of 



