TEMPERATURE 
837 
If, again, he were to put a dairy ther¬ 
mometer into the center of the cluster, 
allowing the upper part of the instrument 
to project thru the packing material, and 
allow it to stand, he would get a more 
correct reading, but not until the bees have 
gone back into their quiescent state pre¬ 
vious to their disturbance. They might 
never go back; but in most cases a tempo¬ 
rary disturbance does no harm, and a clus¬ 
ter of bees will resume its normal course. 
If in a day or two after the insertion of the 
thermometer the cover is lifted gently so as 
not to disturb the bees, and if the ther¬ 
mometer sticks up thru the packing, so that 
it is not necessary to uncover the cluster, a 
fairly correct reading may be secured, pro¬ 
vided the cluster in the meantime has not 
moved. The temperature may then show 
as low as 57. If it is lower, the bees will 
proceed to raise the temperature of the 
cluster in a manner that will be explained 
further on. 
In this connection it should he stated 
that the common mercurial thermometers 
are not always correct; and, what is more, 
it is not always possible to place them so 
that they will be in the exact center of the 
cluster; and even when they are so placed, 
the bees may move from one portion of the 
brood-nest to another. As fast as stores 
are consumed in one portion the cluster 
will move to a fresh supply, provided it is 
not too cold. 
It is not difficult, with these general 
facts before us, to understand how various 
observers have been deceived in forming 
conclusions in regard to the temperature 
of the honeybee cluster during winter. It 
is also very evident how one might jump to 
the conclusion that bees hibernate like ants. 
It was not until the Bureau of Entomol¬ 
ogy, Washington, D. C., attacked this prob¬ 
lem in 1912, 1913, and 1914 that the mat¬ 
ter was definitely cleared up. Dr. Burton 
N. Gates, then Apicultural Assistant in the 
Bureau of Entomology, Washington, D. C., 
made a series of experiments in determin¬ 
ing the temperature of a colony in winter. 
His investigations are described in Bulletin 
96, U. S. Department of Agriculture. 
These experiments were carried on further 
bv Dr. E. F. Phillips, Apicultural Investi¬ 
gator in the Bureau, and by Geo. S. 
Demuth, and recorded in Bulletin No. 93, 
Department of Agriculture. Dr. Gates 
worked with mercurial thermometers, but 
because of the limitations of these instru¬ 
ments he was unable to carry his work to a 
finish. 
Phillips and Demuth conducted a series 
of experiments in wintering bees in a con¬ 
stant-temperature room at the University 
of Pennsylvania, Philadelphia, during the 
winters of 1912 and 1913, and 1913 and 
1914. Several colonies variously prepared 
were placed in a constant-temperature 
room, where.the temperature was held by 
means of coils of pipes containing a brine 
solution—much the same apparatus that is 
used in cold-storage plants. On the roof 
of the building containing this room there 
were placed several colonies of bees where 
the conditions of outdoor-wintered colonies 
could be observed. A series of electric 
thermometers, or “thermo couples,” were 
placed in one of these colonies on the roof 
and likewise in the colonies in the constant- 
temperature room before mentioned. By 
an elaborate system of wiring, these elec¬ 
tric thermometers were connected to an 
observation room in the building, entirely 
separate and distinct from the constant- 
temperature room. Here Dr. Phillips, with 
his assistant, could follow with the greatest 
precision the temperatures of every part 
of the hive and clusters of the several colo¬ 
nies inside and outdoors. 
By these temperature readings it was 
possible to determine the exact state and 
size of the cluster, when it moved, and the 
various reactions that took place as the 
result of feeding, disturbance, and the 
rising and lowering of the temperature 
outside the hives. The purpose of using 
electric thermometers was to avoid the 
disturbance incident to the use of mer¬ 
curial thermometers that require the en¬ 
tering of the bee-room and the opening of 
the hive to get the readings. Moreover, it 
would be practically impossible for an ob¬ 
server to stay in a bee-room with a temper¬ 
ature of 42 F. day and night, taking read¬ 
ings every fifteen minutes; and even if he 
could do so, the constant disturbance would 
naturally cause a rise of temperature that 
would be above the actual normal of a col¬ 
ony not so molested. 
The outside-wintered colony had 19 elec¬ 
tric thermometers, with connections to the 
