4 BULLETIN 988, U. S. DEPT. OF AGRICULTURE. 
0° C. In locations where the outer temperatures fall much below this 
the bees are still able to maintain high temperatures, more bees taking 
part in heat production. That such high temperatures can be main- 
tained in these circumstances indicates that the shell of bees is effec- 
tive as a heat insulator, but there is obviously a serious drain on the 
vital capacity of the bees employed in producing heat. This is shown 
by the rapid slowing down of the fanning of the wings as it con- 
tinues. 
OUTLINE OF THE EXPERIMENT. 
To obtain information regarding the actual amount of work done 
by a colony of bees while in the winter cluster, a small colony on 
four combs having natural honey stores was placed in the chamber of 
a small respiration calorimeter and their carbon-dioxid production 
and oxygen consumption were measured for 10 days, while the tem- 
perature of the air surrounding the bees was kept just low enough 
so that the bees at all times would remain clustered. Throughout 
the experiment the temperature of the air surrounding the bees and 
at several points within the cluster was taken in order that this 
work might be made comparable with the work on the behavior of 
bees in winter as indicated by temperature responses. The bees were 
located in a box within the calorimeter so constructed that while they 
could not escape from it there was opportunity for abundant ventila- 
tion. There were 14 thermocouples distributed in the hive in the 
calorimeter in such manner that the temperatures in different places 
inside and outside the cluster could be ascertained, the leads from 
the thermocouples being extended through the outlet in the wall of 
the chamber to a potentiometer on the outside. The temperatures 
were read every half hour, day and night, for nearly 12 days. 
The thermocouples were so placed in the hive as to make it im-' 
possible for the clustered bees ever to occupy space in which some 
of the thermocouples were not located, thus insuring that the 
temperatures of the cluster might be obtained wherever the cluster 
might move in the hive. The temperatures of all parts of the 
hive outside the cluster could also be obtained by the arrangement 
of these thermocouples. One of the thermocouples (No. 15) was 
located outside the hive and 2 inches from it, thus giving the 
temperature of the air of the chamber at this point. The readings 
obtained with this thermometer are plotted in the charts on pages 15 
to 18. A resistance thermometer was also placed in the chamber, but 
at some distance from the thermocouple. Measurements made eres 
this thermometer are shown in the table on page 8. The two 
records did not always exactly agree because the thermometers 
were not together. 
