12 



other teeth in the length and diameter of its basal portion. From its posi- 

 tion it is probably an incisor. 



The anterior border of the orbit falls above the third tooth of the exter- 

 nal row (counting from behind). The inner border of the maxillary bone 

 is elevated into the ridge convex inwards, as in the other species of this 

 family. The malar base of the zygomatic arch is a moderately stout ver- 

 tical oval. 



Measurements. M. 



. Length of dental series 063 



Length of molar 6-8 series .029 



Width of the two molar series 009 



Vertical diameter malar bone 013 



The associated bones of the skeleton may belong to this or to the next 

 species, or even to a small Empedocles whose teeth occur in the same lot. 

 In the uncertainty of reference I do not describe them. 

 Helodectes isaaci Cope. Sp. nov. 

 Founded on a fragment probably of a maxillary bone, lacking both ex- 

 tremities, and considerably obscured by ferruginous deposit. 



The characters are well marked, leaving no doubt that this species is 

 distinct from those previously known. The bases of the teeth of one of 

 the rows are much more extended transversely than those of the other, 

 baving the form of some of those of Empedocles. As in that genus, they 

 shorten anteriorly. In the fragment, I count on this row, bases of nine 

 teeth. In the other row, I can only definitely count three, which are 

 opposite the second, third, and fourth of the other series (counting from 

 behind). They are wide transverse ovals, about half the long diameter of 

 the posterior teeth of the other series. 



Measurements. M. 



Length of bases of eight larger molars 032 



Diameter of large molar \ anteroposterior .004 



° ( transverse . 008 



Length of three smaller molars 012 



Long diameter of a smaller molar 004 



This species is dedicated to J. C. Isaac, the discoverer of the first species 

 of this family. 



GANOCEPHALA. 



Examination of abundant material shows the correctness of my anticipa- 

 tion (American Naturalist, 1878, 638), that the vertebras of the large batra- 

 chian Eryops, would turn out to have the structure found in Bhachifomus . 

 This genus then must be referred to the same sub-order as Trimerorhachis, 

 and probably Actinodon Gaudry, which will be characterized by the seg- 

 mented vertebral centra. If European authors are correct in stating that the 

 vertebrae of the Labyrinthodontia have undivided centra, the sub-order 

 above mentioned must probably retain the name of Ganocephala, with ad- 

 dilional characters. 



The identification of the scapular arch in Eryops, and of the pelvic arch 



