NASO-LABIAL GROOVE OF LUNGLESS SALAMANDERS. 15 



originally derived, and to decide the question as to their possible 

 relation in origin to other glandular structures such as the aci- 

 nous glands must, however, be left to further investigation. 



THE FUNCTION OF THE NASO-LABIAL GROOVE AND ITS 

 ASSOCIATED GLANDS. 



As expressed in the introductory paragraph, the exclusive 

 occurrence of so definite and well developed a set of organs as 

 the naso-labial groove and glands in those families which are 

 characterized by absence of lungs must be correlated with the 

 lungless condition. The key to the explanation of their func- 

 tions will be found, therefore, in a study of the respiratory habits, 

 particularly those which involve a use of the nasal passages. 



Although certain species of the Plcthodontidce and Desnwg- 

 natliidce have an aquatic larval life, practically all are terrestrial 

 in the adult stage. They live in moist places under stones and 

 decaying wood or, like Desmognathus^ burrow in loose, moist 

 earth along the banks of small streams. Some species, among 

 them DesDiognatJins fnsca, do not even go into the water to de- 

 posit their eggs. Antodax lugubris has carried its terrestrial 

 habits to such an extreme that it has become arboreal, and rears 

 its young in the cavities of tree trunks [RITTER and MILLER 

 ('99) and RITTER ('03)]. I have been able to find in the 

 stomachs of both Desmognathus fnsca and Plethodon cinereus 

 only spiders, mites and terrestrial insects, and I am led to infer 

 that the food of these species consists wholly of terrestrial forms. 



Whatever other methods of respiration the lungless salaman- 

 ders may have, they maintain, so long as they have access to air, 

 a constant bucco-pharyngeal aerial respiration. This is approxi- 

 mately regular and is very rapid, sometimes exceeding a hun- 

 dred respirations a minute, although the number may be as low 

 as fifty. It is accomplished by an alternate depression and ele- 

 vation of the floor of the tightly closed mouth, as a result of 

 which air is drawn in and expelled through the nasal passages. 

 Slight occasional fluctuations of the crescentic fold which guards 

 the external naris may be observed, but, under ordinary condi- 

 tions, not a complete closure. If, however, the specimen be im- 

 mersed in water, or if a drop of water be applied to the snout so 



