DUNN: AMPHIBIA CAUDATA. 465 



of the attached-tongued species and the larva has only three epi- 

 branchials instead of the four found in Desmognathus. Furthermore 

 the great resemblance of the larva to that of Pseudotriton ruber and 

 the angle in the vomerine tooth-row ally it to that species although 

 the prefrontal bone is of the usual type and does not reach the nares. 



65. Typhlomolge rathbuni, Stejneger. 



Tex.: San Marcos, 2,428; 4,170; 4,171; San Marcos, Frank Johnson's well, 

 4,653; 4,654. 



The latter is a recently discovered place of capture and is worth 

 noting as the species is no longer taken at the old well where it was 

 first caught. 



I cannot agree to the assignment of this animal to the Proteida 

 where it is placed in the recent Check list of Stejneger and Barbour. 

 It agrees, as Miss Emerson has shown, in practically every point in 

 its anatomy with the larvae of Pseudotriton and of Typhlotriton. 

 If the classification is to show the relationships I do not see how this 

 species can possibly be placed anywhere than among the Pletho- 

 dontidae as a permanent larva. In the absence of lungs, the fusion 

 of the otic bones, the absence of any ypsiloid apparatus, even in such 

 characters as the fusion of the premaxillaries and the number of 

 epibranchials, it agrees with the larvae of the two genera of Pletho- 

 dontidae mentioned. Indeed, every reason that can be assigned for 

 placing Typhlomolge in the Proteida applies also to the axolotl. 



66. Gyrinophilus danielsi (Blatchley). 

 N. C: Haywood County, 2,890. 



67. Gyrinophilus porphyriticus (Green). 



Ontario, 1,370 (3).— N. H.: White Mountains, 996.— Mass.: Princeton, 

 2,296; Williamstown, 4,483-4,488 ("6).— N. Y.: Adirondack Mountains, 

 McBride's Pond, 3,976; Berkshire, 198 (5); 1,240; 4,577-4,582 (6); 

 Elizabethtown, 2,279; Tapper Lake, 994; 4,274.— N. J.: Plainfield, 

 4,489 {larva). 



68. Geotriton fuscus (Laurenti). 

 Italy: Genoa, 1,944. — Sardinia, 2,166. 



