450 
JOURNAL OF ECONOMIC ENTOMOLOGY 
[Vol. 16 
believe this thrusting motion to be necessary on account of the compara¬ 
tive straightness of the mandibles. Were they more curved a pincher- 
like action might be used. After a hold has been secured by the man¬ 
dibles violent struggles and extrusions of glue on the part of the aphid 
usually fail to secure its release. The larvae hold on tenaciously, bracing 
themselves with their feet, and frequently anchoring themselves by the 
tip of the abdomen. Every cast skin or sucked out aphid body that is 
encountered in the search for food is attacked as though it were alive. 
Skins such as these frequently become caught on the mandibles in which 
case the larva will wipe its jaws across the surface of the leaf in order to 
rid itself of them. Occasionally the aphid is held high in the air on the 
points of the mandibles for several minutes while feeding is in progress. 
Also if the larva is disturbed it will hold the aphid in this manner while 
running away. 
A first instar larva will require from two and a half to three and a half 
hours in destroying a half grown cabbage aphid. Larvae of the third 
instar will destroy them in from five to fifteen minutes depending on the 
size of the aphid. 
In addition to aphids the larvae have been found feeding on the eggs 
of Coccinellids and of the cabbage butterfly, also on those of their own 
species. Feeding on their own eggs does not usually take place if all 
eggs are of the same oviposition period as these hatch almost at the same 
time. However if two different groups of eggs are placed together the 
larva of the group first hatching invariably find and destroy the eggs of 
the second. 
The larvae have only three instars. The method used in determining 
this point has already been described under rearing. Table I is the 
record of sixteen larvae that were carried from egg to adult. It includes 
the length of the incubation period, the length of each individual instar, 
and of the inactive period spent by the larva in the cocoon before 
the final moult to the pupal stage. Averages and totals for the table are 
included. 
From Table I we find that the first instar averages two and a half days 
in length, with a maximum of four and minimum of two days. The 
average length of the second instar is one and a half days with a maxi¬ 
mum and minimum of two and one respectively. The third instar is 
much longer averaging four and a quarter days but over half of this 
period is spent in spinning the cocoon and lying inactive before the final 
moult. The total larval period averaged slightly over eight days with 
nine as the maximum and seven as the minimum length. 
