10 
plaster of Paris cages, consisting of 2 or 3 ounce salve boxes filled 
with newly mixed plaster in which a cavity was bored before the 
plaster had set. The first season that these cages were used a great 
many larvss were reared partially but died before reaching maturity. 
Finally efforts in this direction were rewarded and six larva? were car- 
ried successfully through from egg to pupa (Table III) during the 
month of June at a mean temperature of 84° F., and during this 
time the length of the larval stage averaged from 22 to 25 days. 
Table III. — Length of larval stage of the desert corn flea-beetle (Chaetocnema ectypa) at 
Tempe, Ariz., 1915. 
Aver- 
Length 
age 
Date egg 
Date of 
mean 
hatched. 
pupation 
larval 
tem- 
stage. 
pera- 
ture. 
Days. 
°F. 
June 17 
July 9 
22 
84 
Do.... 
...do 
22 
84 
Do.... 
...do i 22 
84 
Do... 
July 10 
23 
84 
Do.... 
July 12 
25 
84 
Do.... 
...do 
25 
84 
However, since little difficulty was experienced in securing the 
respective lengths of the egg and pupal stages, and then the sum 
of the lengths of the egg, larval, and pupal stages, it was possible 
from these data to obtain the length of the larval stage by sub- 
tracting the sum of the egg and pupal stages from the sum 
of the three stages. Thus we had the approximate length of the 
larval stage during practically every month in which these beetles 
were developing. It may be seen by consulting Tables I and III that 
this approximate length of the larval stage, under periods of the same 
temperatures, compares favorably with the exact length of the stage 
as noted in the case of the six larvae mentioned. By the latter method 
the larval stage is found to vary from 20 days in the warm summer 
months to 47 days during the cooler months of the season, an average 
length for the larval stage of about 32 days for all temperatures. 
FEEDING HABITS OF THE LARVA. 
Quite soon after hatching the young larva begins eating its way into 
the tender succulent roots of the host plant. It thus constructs a 
tunnel for itself as it feeds, and soon disappears from sight. Usually 
the boring of the larva occurs in the cortex. On the roots of corn 
(fig. 6) the tunnel will often appear to be formed between the cortex 
and the stele or central cylinder. The reason for this attack on the 
cortex first is doubtless because it is more tender than the rest of 
the root. If the feeding be of prolonged duration the entire root may 
be tunneled. Many of the roots are entirely hollowed out by these 
