12 PROCEEDINGS OF THE ACADEMY OF [Jan., 



of Ambijstoma porphyriticufn Hallowell, which Hallowell wrongly 

 confused with Salamandra porphyritica Green. Now Hallowell's 

 material is based on "One specimen in Mus. Acad. Nat. Sc, pre- 

 sented by Dr. McMurtrie" and "Prof. Green's specimens were 

 found in French Creek, near Meadville, Crawford County, Pennsyl- 

 vania. " As we have the McMurtrie example, labelled "Wabash 

 River, Indiana/' we assume "Ohio" mentioned in 1861 to be 



erroneous. 



PLETHODONTID.^. 



The present state of our knowledge of the genera of this family 

 will not permit us to advance any new rules, though Moore in 

 abolishing the family Desmognathidse renders it imperative that 

 the generic relationships be revised. The value of the characters 

 which can be used must always remain more or less doubtful. Some- 

 times the same character in one group may not be of the same value 

 in limiting the genera as in another. Nevertheless, it appears to 

 us that the boletoid tongue characterizes a very natural group, i.e., 

 Spelerpes and allied genera. All the species of this group of genera 

 display that tendency towards opisthoccelous vertebrae which led 

 Cope to place Thorius in the Desmognathidse. Most, if not all, 

 have an acjuatic larval stage, in some cases, as Spelerpes ruber and 

 Gyrinophilus porphyriticus, -of long duration. Separate premaxil- 

 laries compel us to place G. porphyriticus and Geotriton fuscus at the 

 bottom of the list. These forms are generically separated by the 

 very doubtful character of webbed toes in Geotriton, Spelerpes follows 

 Gyrinophilus, both with free toes, Manculus with only four toes, 

 (Edipus and (Edipina with webbed and united digits, respectively, 

 and Thorius with webbed toes and extremely large nostrils. It 

 may be here stated that the young of Geotriton fuscus also have 

 extremel}' large nostrils. Probably further studies may reduce 

 some of the above to subgenera and even create a new subgenus 

 for Spelerpes ruber, which seems sufficiently distinct from *S. longi- 

 caudus, the type of the genus. Manculus, Oedipus, and CEdipina 

 may be derived from Spelerpes, which in turn may be derived from 

 Gyrinophilus, probably through Spelerpes ruber. The direct rela- 

 tions of Thorius and Geotriton are less certain, but that they are 

 closely allied is unquestionable. Next to the free-tongued group 

 would come Stereochilus, a little-known form, which in color and 

 dentition approaches Spelerpes ruber. The tongue is more free than 

 in any of the other remaining genera. The cranial structure is 

 comparatively unknown. Cope's statement that it lacks the 



