34 BULLETIN 1339, U. S. DEPARTMENT OF AGRICULTURE 



The coefficient of .5136 ±.0922 between nocturnal loss during the 

 fall and nocturnal variation of temperature can be explained in the 

 same way. During the fall the amount of evaporation necessary is 

 comparatively small because of the small amount of nectar gathered 

 during this investigation, so that the various weather factors have 

 little effect upon it, but have a considerable effect upon the work of 

 the colony as a whole. The greater the range in temperature, the 

 greater the consumption of stores, owing again, perhaps, to modifi- 

 cation in the temperature-regulating arrangements of the colony. It 

 is seen that variation of temperature has a greater influence upon 

 the nocturnal loss, during both the spring and fall, than does the 

 average nocturnal temperature. Variation of temperature at night 

 produces noticeable disturbance. 



RELATIVE HUMIDITY AND CHANGES IN COLONY WEIGHT 



Even less is known regarding the influence of relative humidity 

 upon changes in the weight of the colony than regarding the influ- 

 ence of temperature, probably largely because instruments for indi- 

 cating relative humidity are not as widely distributed as are ther- 

 mometers. Plant physiologists have, of course, studied the influence 

 of relative humidity upon various phases of plant life, and to a lim- 

 ited extent upon nectar secretion. As in the case of temperature, 

 there is no general agreement as to its relative value. 



One of the difficulties encountered in determining the value of the 

 influence of relative humidity upon any phase of plant or animal life 

 is that relative humidity and temperature are intimately related. It 

 is necessary, in a careful analysis of the influence of various weather 

 factors upon physiological changes, to make due allowance for the 

 influence of these variables upon each other. Wright (34), for 

 instance, in recalculating the figures of Briggs and Shantz (4, 5) 

 found that temperature and not. wet-bulb depression was the more 

 important variable, influencing the daily transpiration of plants. 

 Patterson (28) states that relative humidity as a factor influencing the 

 growth of higher plants has been greatly overestimated. He found 

 that elongation in etiolated shoots of the common bean, growing in 

 pure silica sand, either 20 per cent or 60 per cent saturated with 

 water, proceeded as rapidly in a relative humidity of 30 per cent as in 

 a relative humidity of either 60 per cent or 90 per cent. When 

 growing in silica sand, 5 per cent saturated, they grew less rapidly 

 in a relative humidity of 30 per cent than in a relative humidity of 

 60 per cent or 90 per cent, but grew as rapidly in a relative humidity 

 of 60 per cent as they did in one of 90 per cent. 



All figures pertaining to relative humidity in the present bulletin 

 were obtained from a self-recording hygrometer maintained in the 

 laboratory apiary. This instrument tended regularly to record too 

 high percentages of relative humidity, and consequently the figures 

 in some of the tables appear to be above the average. This error in 

 the instrument should not greatly affect the coefficients of correlation 

 in which relative humidity figured as one of the variables, since the 

 error appears to be fairly constant. No attempt has been made to 

 determine the optimum or minimum relative humidity necessary for 

 any activity of the apiary. 



