THE TEMPERATURE OP THE BEE COLONY. 7 



The rate of consumption of stores, as is shown in figure 4, exhibits 

 a relatively constant decrease from month to month. At the begin- 

 ning of the season, before the cluster was well established, when bees 

 were more active and before settled winter weather, food consump- 

 tion was greater than in midwinter. As the season progressed, during 

 February, for instance, consumption slackened. There are several 

 factors which may account for this. In the first place, as the winter 

 advanced there were fewer and fewer bees to be fed. The winter 

 was also less severe, and consequently less generation of heat was 

 necessary. 



Humidity is another factor which noticeably influenced the daily 

 weights for a considerable part of February. This also occurred 



v &000 



\ 



OCT 



ASOK 



&&?. 



^4A< S=E&. 



/&&?. 





sSS-^O 













S&9o\^ 



sS^-eo 













■*5/0\. 













3S*0^ 



^^i3270 













£<}<£&• 



^^^£0*0 













/eoo"^ 



/sso 















Fig. 4.— Graphic representation of the loss in weight of the bee colony from November 6 to March 7, 

 due to the consumption of stores. 



periodically in other months. Although condensation tended to 

 prevent a drop or even to raise the curves during a period of bad 

 weather, as will be shown below, the increased weight due to the con- 

 densed water vapor could neither be permanent nor affect the total 

 loss of weight during so long a period as a month. Whatever water 

 condensed during inclement weather would evaporate during the 

 following days of fair weather. Thus, while the scales might reg- 

 ister an increase during bad weather, consumption of stores was 

 actually going on all the time, but could not be detected in the 

 weights until fair weather had dispelled the moisture. Conse- 

 quently the records of single days are less significant than the aver- 

 ages of a month or of the season. 



