122 



experiment was undertaken, being in accordance with an agreement 

 made at the Washington meeting of the Associatioji of Economic Ento- 

 mologists, by which a series of experiments with the same object in 

 view were to be carried on by Mr. James Fletcher, Entomologist of 

 the Dominion of Canada, Dr. J. A. Lintner, State Entomologist of New 

 York, and myself. 



A mixture of Paris green, 4 ounces to oO gallons of water, was 

 sprayed on a Lombard plum tree in fnll bloom, at 2 p. m., April 29. 

 The quantity of the mixture used was sufficient to wet thoroughly with- 

 out drii)ping. The upper portion of the tree to the lower branches was 

 covered with a square of thin brown sheeting of the brand " LTtica C" 

 and held down by ropes and stakes at the corners. The lower portion, 

 ijicluding a space of about 8 feet square, was inclosed by mosquito 

 netting sewed to the sheeting above and fastened below so as to pre- 

 vent the escape of the bees. The ground thus inclosed was covered with 

 the same material as the top cover. At 7.30 p. m. the hive, which had 

 been placed near this tree some two weeks before, was moved into the 

 inlosiire and the whole secured. Dead bees began to be observed on 

 the ground cover early on the morning of the 30th, and by 10 a. m. a 

 considerable number had died and fallen on the cloth. Others were 

 evidently exhausting themselves in trying to escape. At 1.30 p. m. 

 there were a large number of dead and dying bees on the cloth, and it 

 was thought advisable to remove the cover from the tree and allow the 

 injured bees to escape. At 5 p. m. several hundred bees were either 

 dead or dying, and enough were gathered from the cloth on the ground 

 to fill a box of 21|f cubic inches capacity, while others were clinging 

 to the upper covering nearly or quite dead. 



May 2 four analyses were made by Mr. Ealkenbach, chemist of the 

 Ohio Experiment Station, using the Marsh method, which indicates 

 only the presence or absence of arsenic without revealing the exact 

 amount when present. First, a large number of the dead bees were 

 tested and arsenic found i^resent. Second, more bees were thoroughly 

 washed to remove any of the poison which miglit have become attached 

 to their bodies, but the presence of arsenic was clearly shown. Third, 

 a large number of bees were washed as in preparing for the second analy- 

 sis, and their bodies divided, the abdomens being analyzed separately, 

 but the presence of arsenic was still shown, though but a mere trace. 

 Fourth, the remainder of the bodies, less the wings, were subjected to 

 the same analyses, and arsenic shown to be present in greater amount 

 than in the third analysis. 



The balance of the dead bees were thrown out, but several days later, 

 during which time there had been a severe thunder shower, a consid- 

 erable number were picked up and thoroughly washecl,first with water 

 and then with a weak solution of ammonia, as a still further precaution 

 toward removing all poison from the outer surface of the bodies of the 

 bees. The results of tlie analysis, however, did not materially differ 

 from those previously made. 



