48 Psyche [February 
Cocoons of Callosamia promethea and Tropea luna gave similar 
results. In spite of errors due to imperfect conditions, these tables 
show that the temperature inside a cocoon is practically the same 
as that of the surrounding air and that there is a constant tend- 
ency for the inside temperature to approach that of its surround- 
ings. 
One fact, however, which is readily apparent, is that sudden 
changes of temperature do not occur within the cocoon. When 
the outside temperature was suddenly lowered as from 10°C. to 
0°C., the temperatures in the cocoons fell gradually and even dur- 
ing a gradual rise, the cocoon temperature lagged behind that of 
its surroundings. This is no doubt due to the poor conducting 
qualities of air and silk. 
As the cocoon of Samia cecropia is double and the pupa thereby 
protected by two air spaces, this was somewhat more resistant to 
sudden changes of temperature than the others, a longer time 
being required for the inside temperature to reach that of the at- 
mosphere, when either a rise or fall occurred. 
From figures showing comparative conductivity! one can see 
that air is extremely poor conductor, and also wool, the thermal 
conductivity of which approaches that of silk, the exact figures 
for silk being unobtainable. It is worth noting that the rather 
thin paper-like covering of the bag worm was equally as resistant 
to sudden changes, as the well made heavier cocoons of Bombya 
mori and Callosamia promethea. 
From the standpoint of temperature alone it is doubtful if the 
pupa needs the protection afforded by a thick cocoon, especially 
when we consider that pupe of the superfamily Papilionoidea are 
without such coverings. 
From the standpoint, of moisture however, it is exceedingly 
necessary for the insect to have such a covering, especially if the 
pupal stage is passed on the ground among leaves or in situations 
likely to be moist. 
Cocoons of Callosamia promethea, the vers of which tend to 
fray out toward the proximal ends, which are not finished inside 
with coatings similar to the remainder of the interiors, when partly 
1The conductivities for heat of a number of substances is as follows: Copper, 1.041, 
Iron 0.167, Ice 0.0057, Marble 0.005, Glass 0.0025, Cork 0.0007, Wool 0.00012, Paper 0.000094, 
Air 0.000056. 
