﻿12 BULLETIN 1364, U. S. DBPAETMENT OF AGEICULTUEE 



purpose. Ten dead-bee traps were installed and observations were 

 made daily as usual. Owing to rainy weather and irregularity in 

 the blooming of the different varieties of apple trees, spray mix- 

 tures of lead arsenate and lime-sulphur were applied whenever the 

 weather permitted. As a result, a -few trees were sprayed in nearly 

 full bloom, several when 80 per cent and many when 90 per cent of 

 their petals had dropped, and many after all of the petals had fallen. 

 Observations showed that a small percentage of the mortality of 

 the bees was caused, by arsenic poisoning, but none of the colonies 

 was apparently reduced in number of bees. Thirty-three samples 

 of dead bees were analyzed, of which 30 were found to contain 

 arsenic in very small quantities. 



EXPERIMENTS AT WINCHESTER. VA., 1915 AND 191G 



The experiments conducted at Winchester in 1915, inconclusive in 

 result because of unfavorable weather, were continued there in the 

 spring of 1916 by both writers, using the dead-bee traps and analyz- 

 ing samples of the dead bees collected, and also weighing certain 

 colonies day by day. The latter method of investigation was first 

 tried at Winchester in 1915. In both "^years- the experiments were 

 conducted in three apiaries, each free from f oulbrood and surrounded 

 by large commercial orchards: The Cooper apiary, of TO colonies, 

 lying 3 miles south of Winchester ; the Miller apiary, of 40 colonies, 

 3 miles southwest ; and the Luj^ton apiary, of 5 colonies, 2 miles west. 

 For the weighing experiments in 1915 only the Cooper apiary was 

 used, but in 1916 they were carried on in both the Cooper and 

 Lupton apiaries. Six colonies, selected from the former apiary, were 

 weighed daily, and the three following factors were considered in 

 selecting them: (1) As a precaution against an abnormal death rate 

 they were not transferred from their own apiary, nor were they 

 unnecessarily disturbed; (2) they showed the best evidence of having 

 had uninterrupted brood-rearing during the spring, and so of having 

 a well-balanced ratio of old and young bees; and (3) they occupied 

 well-marked positions in the apiary, thus preventing, so far as 

 possible, errors due to the bees entering the wrong hive. 



APPARATXrS AND METHODS 



Platform scales, sensitive to a variation of one- fourth ounce, were 

 used to weigh the colonies. At Winchester these scales were located 

 in a small building at one side of the apiary, and at Fennville, Mich., 

 a temporary shelter was constructed for them. The primary purpose 

 of the shelter, in each case, was to protect the scales from the wind 

 and so prevent or reduce errors in the readings. 



Chemical balances were used to determine the weight of the sam- 

 ples of a known number of bees and to ascertain the variation in the 

 weight of the bees from day to day. In each case these balances were 

 located in a building adjacent to the apiary. 



In order to weigh the colonies when the bees were in the hive, the 

 weighing was done in the morning, as soon as there was sufficient 

 daylight for the work (about 4.80 o'clock). Each of the six hives 

 containing the colonies under observation was in turn carried to the 



