﻿EFFECTS ON HONEYBEES OF SPEAYING FEUIT TREES 17 



On April 28, 1916, the total weight of the bees in the six hives 

 weighed in the Cooper apiary at Winchester was 298.5 ounces (Table 

 3). The daily loss (or gain) in ounces from April 28 to May 12, in- 

 clusive, beginning with April 29, was 4.5, 7.25, 6.5 (gain) , 8.75, 5.5 

 (gain), 21, 10.75, 14.75, 28.5, 3.25, 13.25, 2.5, 19, and 6.5 ounces. 

 Reference to Table 3 shows slight daily losses, and on two days (May 

 1 and 3), slight gains, from April 28 to May 3, and heavier losses 

 thereafter. Since none of the samples of dead bees collected before 

 May 8 in this apiary contained arsenic when analyzed, it is reason- 

 able to suppose that these bees had not visited the sprayed trees li/^ 

 miles south of their apiary. Before the weighed bees had visited 

 the sprayed trees they lost 9.28 ounces as a daily average, being 

 3.11 per cent of their initial weight ; after having visited the sprayed 

 trees near by, they lost 8.9 ounces as a daily average, being 3 per cent 

 of their initial weight. The theoretical daily mortality, based on 

 six weeks as the normal life of bees, is 2.38 per cent. 



RESULTS OF CHEMICAL ANALYSES 



At "Winchester, in 1915, 6 samples of dead bees, 9 samples of pollen, 

 and 6 samj)les of honey were collected and analyzed. Not one was 

 found to contain arsenic. In 1916, at Winchester, samples of dead 

 bees and pollen were, as usual, collected and analyzed. Five of 

 18 samples, each of 100 dead bees, contained no arsenic; analysis of 

 the others indicated the presence of arsenic, although not enough to 

 have killed the bees analyzed (see pp. 24 to 26). Six of 10 samples 

 of pollen analyzed contained no arsenic ; the others had only a trace 

 of it. 



EXPERIMENTS AT FENNVILLE, MICH.. 1916 



The experiments just described were repeated in 1916 at Fennville, 

 Mich. Three apiaries, surrounded by commercial orchards, were se- 

 lected. The Langley apiary of 12 colonies lay 4^^ miles west of 

 Fennville, the Pratt apiary of 14 colonies 6 miles west and 100 yards 

 from the lake shore, and the Pshea apiary of 30 colonies 5 miles 

 northwest. For the weighing experiments 6 colonies in the Langley 

 apiary were used. 



APPAEATU8 AND OBSERVATIONS 



The platform scales and chemical balances were installed in a 

 temporary shelter by the side of the Langley apiary, and three dead- 

 bee trajis (Xos. 7 to 9) were also installed in front of hives Nos. 7 to 9 

 in this apiary. Four bee traps (Nos. 10 to 13) were installed in the 

 Pratt apiary and four (Nos. 14 to 17) in the Pshea apiary. 



From May 22 to June 2 (the period of the experiments) the 

 weather was almost ideal (Table 4), the blossoms secreted nectar 

 more abundantly than they did at Winchester, and the bees worked 

 well in the trees. The number of bees visiting the trees during full 

 bloom and later agi-eed closely with the number observed during the 

 same period at Winchester-. 



On May 22, 24, and 25, two orchardists sprayed their fruit trees, 

 nearly in full bloom, located 1 mile west and southwest of the Lang- 

 ley apiary, with lime-sulphur and lead arsenate. Daily observations 



