120 BULLETIN OF THE ESSEX INSTITUTE. 



in one rod which is joined in front to the antorbital 

 process. 



The C^cilian Chondrocranium. 



There are two views according to which the Ceecilians 

 are related to Amphiuma. According to one — the theory 

 of Cope ('89*) — the Caeciliaus are the extreme of a line 

 of degeneration from the typical Urodele stock and Am- 

 phiuma is one of the intermediates of the series nearest to 

 the Gymnophiona. Indeed, Cope goes so far as to make 

 the Csecilians merely a family of the Urodeles. The other 

 view is that of the cousins Sarasin who hold that Am- 

 phiuma is a neotenic Caicilian, a larval Csecilian become 

 sexually mature while retaining their branchial respiration. 



According to the first view Amphiuma, and to a less 

 extent the rest of the Urodeles, must be <ilosely similar in 

 cranial as well as other structures to the young Csecilian. 

 Farther, if we find that Amphiuma and the Urodeles 

 have lost certain features which belonged to the ancestral 

 Craniota, the retention of these characters by the Csecil- 

 ians would be an aro-ument a2:ainst the line of descent 

 advocated. The view of the Sarasins presents even more 

 difficulties for we have both horns of the dilemma. If 

 Amphiuma be merely a Csecilian arrested in a larval con- 

 dition, then we have to say either that Amphiuma is not 

 related to the remaining Urodeles or that they have all 

 sprung from a Csecilian ancestry. The objections to the 

 second view are so many and so weighty that we think no 

 one would care to defend it. The limbs alone are enough 

 to set it aside. As to the other horn, it would seem that 

 all the evidence we have regarding adult structure and 

 development as well goes to show that Amphiuma is far 

 more closely allied to the other Urodeles than it is to the 

 Csecilians, while the same matter of limbs, weak though 



