380 Annals Entomological Society of America [Vol. V, 
Ten larve of each representative stage, varying from full 
grown to newly hatched specimens, were placed in individual 
vials and also a large number of all the possible combinations 
in two were made. For example, eight full grown larve 7 mm. 
long were placed in eight different vials and together with each 
of these was placed one individual of each of the other repre- 
sentative stages. Thus, we had a vial containing two full 
grown larve, one containing a full grown and a larva about 
6 mm. in length, and so on down the series with a gradually 
greater and greater difference in the size of the two larve within 
the same vial, until the last one contained both a full grown 
and a newly hatched larva. 
The same process was repeated with a larva of 6 mm., 5mm., 
and so on, to the larva 1 mm. in length, and thus all the possible 
combinations between the larve of practically all sizes were 
made. The additional purpose of this latter experiment was 
to determine the extent of cannibalism among the species. 
Three such large groups of vials, as that described above, 
were made and each was placed under somewhat different 
conditions. One group was exposed to day-light in the labor- 
atory, another was kept continually in the dark, and the third 
in a box under a constant thirty-five candle power electric 
light. The last mentioned group of larve had a somewhat 
higher temperature caused by the presence of the electric 
light in the box. 
Measurements of all the individuals were made and a care- 
ful record is being kept. The vials are examined regularly 
and measurements of the several individuals of each representa- 
tive stage are taken and recorded. A record of the cast skins 
is also kept; from some of the vials the exuviz are removed as 
soon as shed and in others they are allowed to remain con- 
tinually for the purpose of determining whether the larve ever 
eat them. It was found that the larve never devour their 
own nor the skins of other, specimens. There is absolutely 
no evidence of cannibalism among the larve; even the full 
grown starving specimens never attack the much smaller 
individuals. Practically all of the insects shed their skins 
shortly after they were placed without food; but between the 
other following ecdyses a period much longer than the normal 
elapsed. Careful measurements soon revealed the surprising 
fact that the larve were actually decreasing in size. In all 
