NO. 2211. OENERA DESMOQNATHUS AND LEUROGNATIIUS—DUNN. 395 



Mrs. Wilder, of Smith College, Northampton, Massachusetts, for advice 

 and specimens; and finally to Dr. H. S. Pratt, of Haverford College, 

 under whom I studied, who has shown great interest in this work, 

 and allowed me to use the resources of the college laboratory with 

 the greatest freedom, especially in the matter of procuring specimens 



RELATIONSHIPS OF THE GENERA. 



Cope established a family Desmognathidae on the strength of the 

 opisthocoelous vertebrae of DesmognatJius. Thorius, TIaptoglossa, 

 Typlilotriton, and LeurognatJius, were added at various times to the 

 family chiefly on this character. Moore (1900) showed that opis- 

 thocoelous vertebrae are in many salamanders concomitant with 

 age, and that old specimens of various species of Spelerpes, Gyrino- 

 pliilus, and Autodax have vertebrae quite as opistocoelous as has 

 Desinognatfius. Consequently the family can not stand. 



Tlionus Cope, with a boletoid tongue and prefrontal bones, tends 

 to Spelerpes, 



TypMotriton according to Cope (1893a) has the same pecuHar 

 structure of the atlas found in DesmognatJius and Leurognathus . 

 However, it has prefrontal bones and a premaxillary fontanelle. 

 The tongue is attached in front, although according to Hilton (19096), 

 it is in the larval state more hke the tongue of Spelerpes than that of 

 DesTTiognaihus. 



Its position, then, would be between these two genera. 



Leurognathus Moore seems closely related to DesmognatJius in 

 structure of tongue and skuU. The prefrontal bones are lacking in 

 both. The pecuhar structure of the atlas is alike in both. Leurogna- 

 tJius so far as known never has vomerine teeth. In this it is unique 

 save that the males of some species of DesmognatJius lose the vomerine 

 teeth on arriving at sexual maturity. 



The cranial structure of Haptoglossa Cope is unknown. 



BatracJioseps Bonaparte has no prefrontal bones, but this has been 

 shown by Cope to be an independent loss and the genus tends to 

 PletJiodon. StereocJiilus Cope is not well known. Cope says that in 

 the only skull he examined the prefrontal bone was present on one 

 side and absent on the other. Besides LeurognatJius, StereocJiilus is, 

 according to Cope, the only other genus of Plethodontidae without 

 premaxillary fontanelle. This may or may not indicate relationship. 



Therefore DesmognatJius and LeurognatJius form a closely related 

 group, which is apparently isolated from other genera of Plethod- 

 ontidae. 



However, the cranial structure of Haptoglossa and StereocJiilus 

 should be examined for possible relationships. 



