694 THE AMERICAN NATURALIST. [VOL XXXIII. 
on the sides of the abdomen. As we now know, the 
smaller darker ones were of last year’s hatching; and in all 
probability the larger ones were a year older. The presence 
together of so many individuals suggests that those of the same 
size all belonged to the same brood ; and also the possibility 
that the two broods represented were both the offspring of the 
same parent. Of course there was no way of answering this 
query positively, but one other piece of information that we 
have obtained confirms the suggestion in so far as it furnishes 
farther evidence that the young of a brood may remain together 
for a considerable period after hatching. Some time during 
March, 1896, a student in zodlogy, whose testimony we regard 
as reliable, reported to one of us that he had found a fully 
grown salamander, which, from his description, was undoubtedly 
Autodax, ‘with a lot of little ones.” 
The species is decidedly nocturnal in its habits. This is not 
only proven by the fact that one practically never finds it abroad 
during the daytime in nature; but also by the alacrity with 
which specimens kept in confinement in a terrarium seek to 
secrete themselves during the daytime, but come out and run 
about freely during the night. 
As shown by Wilder, Autodax is, in common with so many 
of the long-tailed amphibians, lungless. This being so, the 
exclusively terrestrial habit of the animal makes the question 
of the seat of respiration in this species particularly interesting. 
We do not propose to go at length into a discussion of this 
subject at present, but content ourselves with noting a few 
observations that support the view recently defended with 
special emphasis from the morphological side by Bethge (98), 
that, in the absence of both gills and lungs, respiration is per- 
formed by the mouth epithelium and the integument together, 
each taking an essential part. This is in opposition to the 
conclusion drawn by Camarano ('94), in particular, from phys- 
iological studies on Salamandrina perspicillata and Spelerpes 
fuscus, both lungless salamanders. This author believes that 
the skin participates very little, if at all, in respiration. 
The facts which we interpret as meaning that the integu- 
ment takes an important part in respiration are these: In the 
