The Cycle oj the Year 
87 
decrease the amount of brood when the honey-flow stops 
while Carniolan and Caucasian bees rear more brood “out 
of season,” or after the honey-flow. But all races rear 
some brood “out of season,” so that the final discontinuance 
of brood-rearing cannot be considered as due to lack of 
incoming nectar or pollen. Even among colonies of the 
same race there is considerable variation in a single apiary. 
Of course, no colony can rear brood without food for the 
young larvae. The stoppage of brood-rearing is sometimes 
attributed to low outside temperature but, as stated earlier, 
brood is sometimes reared in the coldest months, in fact 
it is usually begun then in colonies wintered out of doors. 
As will appear later, the cold outside during January is 
the cause of a higher temperature within the cluster than 
is usual in the autumn and, combined with the effects of 
the accumulation of feces, is the cause of higher cluster 
temperature than occurs with the same intensity of cold 
in December. Furthermore, when a colony begins brood¬ 
rearing in the winter, the presence of brood seems to induce 
the production of sufficient heat to care for it, the resulting 
temperature being sufficient to induce more egg-laying, so 
that brood-rearing once begun continues through the re¬ 
mainder of the winter. Since a moderately low outer tem¬ 
perature may cause more active heat production in a small 
colony than in a strong one, this may explain some of the 
variation observed in the time that brood-rearing ceases. 
A small colony may have a higher cluster temperature than 
a strong colony, the greater activity in the center of the 
cluster being necessary to produce the required temperature 
in the shell of the cluster, which is a less efficient insulator 
in a weak colony. The structure of the cluster is described 
in the following section. If an explanation for the cessation 
of brood-rearing is sought, the paradoxical conclusion is 
reached that in the fall (1) the outside temperature is not 
high enough for brood-rearing without artificial heat pro¬ 
duction and (2) it is not low enough to cause the bees to 
produce sufficient cluster heat for brood-rearing. Egg- 
