WEATHER AND CHANGE IN WEIGHT OP BEE COLONY 15 



close of the honey flow colony 1 was apparently increasing in numbers 

 more rapidly than colony 2. If it is assumed that both colonies 

 were working at almost maximum efficiency, the midday decline in 

 the rate of increase in weight to some extent agrees with Bonnier's 

 statements (1) that less nectar is available early in the afternoon 

 than earlier or later. Unless there exists some unknown influence on 

 bee behavior at the time of the midday decline, it must be believed 

 that the reduction in the rate of the increase in weight at this time is 

 due to a reduction in the amount of nectar in the honey plants. 

 Surrounding the Bee Culture Laboratory there is a vast acreage of 

 tuliptrees, and there are comparatively few bees in the neighborhood 

 other than those of the bureau apiary. Bonnier's assertion that bees 

 carry partial loads of nectar when nectar is scarce is thus more prob- 

 able, although the explanation may lie in the necessity for trips of 

 longer duration at this time. There is no reason to believe that the 

 bees actually gathered every drop of nectar available in the neigh- 

 borhood at the time of this decline. 



THE SPRING PERIOD 



As has been stated, the data used in a consideration of the spring 

 and fall honey-flow periods are not entirely comparable, since for the 

 spring honey flow the only records used are those for days which 

 show a net gain of at least 980 grams, whereas in the case of the fall 

 honey flow the record for every day from September 4 to October 5 

 is used. Furthermore, the spring honey flow is much more intense 

 in the vicinity of the laboratory than is that of the fall. For these 

 reasons it seems best to consider the two periods separately. In the 

 discussion immediately following, the various phases of the changes 

 of weight during the day are taken up for the spring period, covering 

 the time when the tuliptree was in bloom. 



MORNING LOSS 



Figure 3 illustrates the changes in weight during a typical day of 

 the spring honey flow. In this case the morning loss is small and 

 covers the time from 5 a. m. to 7 a. m., the majority of this loss 

 occurring during the last hour. Usually the morning loss is quite 

 small and rarely continues more than three or four hours. The 

 amount and duration of the morning loss undoubtedly depend largely 

 upon the proximity of nectar-producing plants and the abundance of 

 their secretion, and upon weather factors prevailing at this time of 

 the day. On some days the morning loss is negligible and scarcely 

 distinguishable from the nocturnal loss, as on May 27 and 29, 1923 

 (fig. 6, &). The bees began to return to the hive almost immediately 

 on these days, and thus with their increased weight more than com- 

 pensated for the loss during the early morning hours. It is quite 

 evident that on such days nectar was abundant and within easy 

 reach of the bees. 



The coefficient of correlation between morning loss and net gain is 

 — .0350 ±.0652. This indicates that the smaller the morning loss 

 the greater the resulting gain, and emphasizes the importance of 

 locating an apiary as near as possible to the principal sources of nec- 

 tar. The significance of morning loss will oecome more apparent 

 later, in the consideration of the fall period. 



