WEATHER AND CHANGE IN WEIGHT OP BEE COLONY 35 



RELATIVE HUMIDITY AND NET GAIN 



It has been shown that temperature influences nectar secretion 

 and changes in the colony weight. Similarly, plants doubtless respond 

 to changes in relative humidity according to the species and the 

 location in which they grow. Some species of plants are known to se- 

 crete nectar under conditions of the greatest possible humidity; on 

 the other hand, arid and semiarid countries contain many nectar- 

 producing plants. 



Hommell (18, p. 194) states that, other conditions being equal, the 

 quantity of nectar increases with an increase in the relative humidity 

 •of the air. Ono (26) , in discussing the influence of moisture on nec- 

 tar secretion in extrafloral nectaries, says that "moisture is one of the 

 conditions most favorable to the secretion of nectar, and its influence 

 seems to be more or less direct. Dry atmosphere is in all cases un- 

 favorable to the secretion." He cites a case in which cut twigs 

 of Prunus laurocerasus were placed in a water bottle and held under 

 a bell jar of moist air for three weeks. Despite the fact that these 

 nectaries were washed daily (an operation which will often terminate 

 secretion), they continued throughout the experiment to secrete 

 nectar without decreasing either its quantity or the quantity of 

 sugar contained in it. Wilson (83), working with the same species 

 of plant, found that after a branch of Prunus laurocerasus had been 

 made to secrete nectar by being placed under a moist bell jar, it 

 continued to do so without water and in the dry air of an ordinary 

 room, until the whole branch had lost more than one-fourth of its 

 weight by withering. Kenoyer (20) discovered that by increasing 

 the relative humidity the production of water was increased, but not 

 that of sugar from nectaries. 



Bonnier (1) found that the secretion of nectar varies directly with 

 relative humidity. Although he realized (2) the complex rdle which 

 relative humidity plays in plant physiology, he made no corrections 

 for the effect of other variables upon relative humidity in his studies 

 on nectar secretion. In calling attention to the midday decline, and 

 accounting for it by the low relative humidity and the high temper- 

 ature prevailing at that time, he made no effort to assign more im- 

 portance to one than to the other. The writer, in calculating Bonnier's 

 data, found the coefficients of correlation, as stated under the discus- 

 sion of temperature. In order to arrive at a corrected value for the 

 effect of relative humidity by reducing temperature to a constant, 

 we substitute in the following formula the numerical values of the 

 proper coefficients (see p. 27 for correlation values), and obtain — 



Tkh - (r NT . Tt h) aa „- 



tTnji = ,. = .6605 



V(l-rV)(l-rW) 



Thus the effect of temperature reduces the coefficient of .8040 between 

 relative humidity and nectar secretion to .6605. 



The substitution in the above formula of the coefficients calculated 

 from Harrault's (15) data gives — 



T r fl „ = .1 844, where r an is .2988 



