NO. 2211. GENERA DESMOONATHUS AND LEUROONATHUS—DUNN. 397 



VARIATION OF THE CHARACTERS. 



Dentition. — This has been studied chiefly by taking impressions of 

 teeth on thin shavings of a preparation of beeswax and lamp})kck. 



The shape and presence of the vomerine teeth are quite constant 

 in the various forms. The females always have them, but the males 

 of all species save quadramaculata and monticola lose them at sexual 

 maturity. Some, but not all, old male monticola lack the vomerines. 



In some specimens of fusca and in most of ocTiropJtaea and caro- 

 linensis the vomerine teeth are on a bony ridge which often persists 

 after the teeth themselves are lost. 



The parasphenoid patches change somewhat in shape with age and 

 are slightly different in male and female. Aside from this they are 

 variable and must be used only in connection with other characters. 



The mandibular dentition of male ocJirophaea is merely an exag- 

 geration of that seen in maiie fusca, but it is marked and constant. 

 The shape of the tongue is subject to so much variation, which is 

 probably due to differences of preparation, that I have been unable 

 to use it as a diagnostic character. 



Tail and limbs. — The shape of the tail is constant and not subject 

 to sexual or seasonal changes. I have found no seasonal changes in 

 these salamanders, save that "in the breeding season the lips of the 

 cloaca are somewhat everted" (Mrs. Wilder). The absence of the 

 dorsal fold is characteristic of the ochropliaea group. 



The relative length of the tail is also fairly constant. The tails are 

 frequently broken and regenerated, so that it is impossible to tell 

 the normal. Results of maximum tail-lengths range the species in 

 the same order as average tail-lengths so that any error resulting 

 from mutilation is of no consequence. 



The size of the legs is quite constant in specimens of the same age. 

 The distance between the appressed toes is less in youth than in age. 

 But the change is not the same for all forms. Thus young monticola 

 have two intercostal spaces between the appressed limbs and old 

 monticola three; while young fusca have three and a half and old ones 

 four. The larvae, young, and adult of quadramaculata have two and 

 a half to three quite constantly. 



Head. — The skin of the head is sometimes very rugose. This is 

 very noticeable in specimens of quadramaculata and monticola. 

 While the majority of specimens possess it, I have been able to find 

 no other character accompanying it and am inchned to set its absence 

 down to an artifact of preservation or an individual or local variation. 

 It is not usually found, however, in old and black specimens of 

 quadramaculata. 



The size of the head is a constant but hard to use, as it has to be 

 relative and the limits of variation are small. 



