THE CHONDROCRANIUM IN THE ICHTHYOPSIDA. 121 



they be in Amphiuma, throws the whole view out of 

 court. 



There remains then but the view of Cope, and this is 

 to be tested by seeing if there be features in the Csecil- 

 ians which must have been inherited and which could not 

 have been inherited from an Urodele ancestor. 



The parachordals of Ichthyophis are smaller than those 

 of any other form studied. The nearest approach to 

 them is found in Desmognathus where there are bands, 

 not present in Ichthyophis, connecting the anterior end 

 of the notochord with the otic capsules. No synotic tec- 

 tum is formed in Ichthyophis, though at one period in the 

 development of the skull small crests on the dorso-median 

 walls of the otic capsules represent the first steps in the 

 formation of a tectum. This appearance, however, is 

 slight and but transitory. 



There is a difference between the manner of develop- 

 ment of the trabeculae of Ichthyophis and those of the 

 other Batrachia described. Instead of being developed 

 by the successive formations of a ventral rod, a trabecu- 

 lar crest and a connective rod uniting the crest to the otic 

 capsule, we have the dorsal rod developing simultaneously 

 with the ventral rod and equally well chondrified. The 

 dorsal rod, separated as it is from the ventral rod and at- 

 tached to the anterior point of the otic capsule, somewhat 

 resembles the supraorbital band of fishes. But the fusion 

 of the ascending process of the quadrate with the dorsal 

 rod and the relations of the two trabecular rods anteriorly 

 are clearly Urodelan characters. The quadrate is pecu- 

 liar, however, in having no otic or palato-basal processes. 

 Aside from the ascending process and stapes it is entirely 

 separated from the rest of the skull. Its position is the 

 same as that of the Urodeles. The stapes is perforated 

 for the stapedial artery. 



ESSEX INST. BULLETIN, VOL. XXVIII 8* 



