﻿EFFECTS ON" HONEYBEES OF SPEAYING FRUIT TREES 31 



was almost ideal, the blossoms secreted more abundantly than they 

 did at Winchester, and the bees worked well in the trees. When 

 the trees were sprayed the bees had almost ceased visiting them 

 and practically all the petals had fallen. The results taken together 

 indicated that the bees tested had not been poisoned, although again 

 17 of the 28 samples collected contained small quantities of arsenic, 

 too little to have been fatal to bees. (See Tables 4 and 6 for weight 

 records. ) 



In regard to the spraying of fruit trees at the customary time, 

 when 90 per cent or more of the petals had fallen, at Winchester, 

 Va., and Fennville, Mich., in 1916, under the most certain con- 

 ditions, which are usually rare, it appears for the following reasons 

 that the bees tested were not injured: (1) The old adult bees died 

 more rapidly before than after spraying was begim, as was shown 

 both by the dailj'^ weight records and by the number of dead bees 

 counted in the bee traps. (2) The gain in weight, after the spray- 

 ing was begun, of all the old and young adult bees iil two colonies 

 tested was equal to the gain before the spraying was begun. (3) 

 The results obtained from the chemical analyses of the samples of 

 dead bees, collected after the spraying was begun, did not indicate 

 that the bees had died of arsenic poisoning. And (4) no symptoms 

 of arsenic poisoning were noticed among the bees imder observation. 



The experiments at Winchester and Fennville were repeated on 

 a small scale in 1917 at Drummond, Md., where the bee culture 

 laboratorj^ of this bureau was then located. Eleven apple trees were 

 spraj^ed when about 90 per cent of the petals had fallen and when 

 bees were to some extent still visiting these blossoms. Of the 40 

 samples of dead bees collected and analyzed only 3 showed the 

 presence of arsenic, and these samples came from colonies which 

 apparently had been suffering for some time from bee paralysis. 

 Xo arsenic was found in samples of pollen taken from these hives. 

 The other results obtained agree with those obtained at Winchester 

 and Fennville in 1916. (See Tables 5 and 6 for weight records at 

 Drummond.) 



Some of the samples of dead bees collected when the trees were 

 sprayed in full bloom contained on an average as much as 0.0004 

 milligram of arsenic (As) per bee. Of course, some of the indi- 

 vidual bees contained less and some more than this quantity, yet it 

 was not definitely known whether that quantity had actually killed 

 them. A series of laboratory experiments in which definite quan- 

 tities of several arsenicals were fed to bees decided this question in 

 the affirmative, established the minimum fatal dosage of arsenic 

 (As) per bee to be about 0.0004 or 0.0005 milligram, and showed 

 that the symptoms of arsenic poisoning of bees in confinement were 

 like those of the bees poisoned in the orchards. (See Table 7 for de- 

 tails.) 



