WEATHER AND CHANGE IN WEIGHT OP BEE COLONY 33 



TEMPERATURE AND MIDDAT DECLINE 



As previously pointed out, the cause for the midday decline has 

 been attributed to high temperature and low relative humidity, both 

 prevailing at that time of the day. The minimum of the average 

 midday decline occurs from 1 to 2 p. m. (figs. 7, 8, and 10). The 

 maximum of the average temperature comes from 3 to 4 p. m. (figs. 8 

 and 10), although it occurred an hour earlier during the 1922 spring 

 honey flow (fig. 7) . Although the midday decline occurs during the 

 hot part of the day, it does not coincide with the period of maximum 

 temperature. Moreover, the coefficient of correlation between net 

 gain and average temperature is high and positive for the spring 

 period, so that it is difficult to give a satisfactory explanation of the 

 occurrence of the midday decline on the basis of high temperature 

 and low relative humidity. During the fall the relation between 

 change of colony weight and the average diurnal temperature is neg- 

 ative and the coefficient is small. From the standpoint of the net 

 gain, the midday decline during the fall is large, but from the stand- 

 point of weight changes it is small and barely noticeable. 



TEMPERATURE AND NOCTURNAL LOSS 



The relation of temperature to nocturnal loss is surprising in view 

 of the widespread importance usually attributed to it by beekeepers. 

 One naturally expects to find a higher rate of evaporation on warm 

 nights following days of good gain than on cool nights. This relation, 

 however, is small, the coefficient of correlation between net gain and 

 nocturnal average temperature being but .1754 ±.1060. 



Temperature variation seems to have a more important bearing, 

 the coefficient of correlation between net gain and nocturnal vari- 

 ation of temperature being —.3439 ±.0964. One interpretation for 

 this would be that when there is a great variation in temperature 

 during the night the bees are forced constantly to modify their tem- 

 perature-regulating organization in such a manner that the temper- 

 ature of the brood-chamber may be maintained constant. In doing 

 this the task of evaporation is interrupted, resulting in a negative 

 correlation. In other words, from the standpoint of evaporation, an 

 even temperature is more desirable than a high temperature, within, 

 of course, reasonable limits. 



During the fall period the relation between temperature and noc- 

 turnal loss is the opposite of that existing in the spring period, the 

 coefficient of correlation between the average nocturnal temperature 

 and nocturnal loss being —.3391 ±.1108. It would be contrary to 

 physical laws to expect a higher rate of evaporation during cold 

 nights than during warm nights. An explanation of this negative 

 correlation seems to lie in the fact that the relation between changes 

 of weight during the day and the respective nocturnal losses is high. 

 This in turn would indicate that the incoming nectar was promp'ly 

 cared for, and the existing outside temperature, instead of primarily 

 affecting the rate of evaporation, has its effect principally upon the 

 general activity of the colony, resulting in the consumption of larger 

 quantities of stores on cold nights. 



