242 THE AMERICAN NATURALIST. (VoL. XXXIII. 
of this larva, as before mentioned, are identical with those of the 
larva of S. bilineatus, and are thus useless as a distinguishing 
test. 
Larve of Spelerpes bilineatus. — The larvæ of S. bilineatus 
hatch early and continue for a long time in the larval state, 
probably 2-3 years. 
Core ('89) says: “It is one of those species whose metamor- 
phoses are prolonged and which remains in the larval state 
until nearly grown.” VERRILL ('62, 63) says of it, under the 
name of D. fusca: “The young become quite large before 
losing their gills.’ This description cannot apply to the 
genuine D. fusca, as has just been shown, and as the author 
has described in the same paper the eggs of S. bilineatus, 
the larvæ he found undoubtedly belonged to this latter species, 
concerning which the statement is an accurate one. The 
growth must be exceedingly slow and dependent upon the for- 
tune of the individual in securing prey. I have caught all 
stages from 16-52 mm. at all seasons of the year, and see no 
indication that those larve collected at any one time represent 
one, two, or three years of definite growth. 
For the purpose of studying this point I went to Williams- 
town, Mass., in September, 1896, collected 90 larvæ, and meas- 
ured and tabulated each! The result of this is shown graph- 
ically in Fig. 26, in which the ordinates represent the total 
lengths in millimeters, and the abscissas the number found. 
The results seem to show, in general, merely a decrease in num- 
bers as the animal gets larger, which was to be expected. 
There are gaps in one or two places, indicating sizes that I 
did not find, but these are by no means wide enough to repre- 
sent a year’s growth. 52 mm. represents about the limit of size 
reached by the larvae under the most favorable circumstances. 
I have found adults a little smaller than this. 
Differentiation of the Larve.—The fact that the larve of 
S. bilineatus are exceedingly common, while those of D. fusca 
are rare, renders it æ priori probable that a given larva belongs 
1 I was materially assisted in the collection of these larve by my good friend, 
the late Dr. James I. Peck, whose kindness I most pleasantly remember in this 
connection. 
