1937] Cannibalism among Lepidopterous Larvae 113 
marsh land, approximately half a square mile in extent, cut 
from north to south by a river, and bounded on all sides by 
highways. The larvse were found only on the east side of 
the river. Of the thousands which were traveling back and 
forth between the river and the highway many were killed 
by passing vehicles and pedestrians. Practically every fifth 
caterpillar that had been killed was being eaten by one or 
more of the same species which were consuming all parts of 
the dead ones, except the hairs. The green vegetation in the 
vicinity was restricted to a narrow belt along each side of 
the highway. The scarcity of low-lying vegetation was evi- 
dent. Practically the only green plant was an aster, Aster 
laevis, growing to a height of four feet. The larvse which 
were not feasting on the dead specimens were eating the 
tops of the asters. Feeding on tall vegetation is unusual for 
a characteristically ground-loving species like Isia isabella. 
It is noteworthy that nearly all the larvse showed signs of 
hunger. It would seem that the summer of 1937 was for 
some reason especially favorable to the development of this 
species. With the approach of autumn the supply of food 
plants in the area to which the caterpillars were confined 
gave out in the presence of such overwhelming numbers. 
The larvse were prevented from traveling west by the river, 
hence they swarmed upon the highway where they were 
killed in large numbers. The hungry survivors either 
feasted upon the dead or climbed the asters. No cannibalism 
was observed in the few individuals of D. virginica and E. 
acrxa which were also present. 
On the occasion of an exceedingly large swarm of 
Ecpantheria, Orfila (1927) noticed three animals avidly 
devouring the remains of a fourth which was still living. 
The following are also said to be cannibalistic in nature : 
Laphygma frugiperda Smith & Abbot (Moore 1912), 
Vanessa caryae Hb. (Berg 1875, 1892) , and Heliothis obsoleta 
Fab. (Berg 1875, 1892) . Berg reported that H. obsoleta ate 
from six to seven larvse in twenty-four hours. 
The question as to why larvse show cannibalistic tendencies 
is a difficult one to answer. Three alternative solutions have 
been suggested by Orfila (1927) : 1) it is due to an internal 
physiological cause, 2) an organic disarrangement produces 
an alimentary upset, 3) cannibalistic manifestations herald 
