liiNTOMO LOGICAL SOCIETY OF ONTARIO. 105 



On May 2ad two api)le trees in full bloom — and the blossoms wen; almndant — were 

 thoroughly sprayed with a mixture of 1 ounce of Paris green to sack 12 gallons of 

 water. After the water had evaporated the poison could be clearly observed both on 

 bloom and foliage. The application was made during the forenoon, the day being warm 

 and clear, and during the afternoon qui*'e a number of bees were caught while visiting 

 the bloom and marked with carmine ink. The hives were located but a few yards distant 

 from the trees, and both being situated at a considerable distance from any other trees 

 at that time in bloom. None of these marked bees were afterwards found dead about 

 the hives. During the night following the application there was a rainfall of 0.20 

 inch. On the following day bees were caught and killed by being dropped into a cyanide 

 bottle where the cyanide was embedded in plaster of Paris, after the usual custom. As 

 soon as the bees were dead they were dissected as follows : The po.sterior legs v/ith pollen 

 attached were severed from the bodies and placed in a small glass vial and securely 

 corked. The contents of the abdomen.s, including the honey sacs, were next dissected 

 out and placed in a separate vial, and the same mode of procedure was followed with 

 the whole inside of the thorax, this giving me the entire bee except the head, anterior 

 And middle legs, wing?, and chitinous walls of the thorax and abdomen. Besides these 

 I number of the bees v/ere kept intact. The whole series was submitted to the assistant 

 professor of chemistry of the Ohio State University, L. M. Bloomfield, to be tested for 

 jrsenio by the Marsh method. j\fr. Bloomfield found the weight of material submitted 

 in each case to be as follows : Posterior legs, with pollen attached, 0.31:98 gram ; con- 

 tents of abdomens and honey sacs, 0.0990 gram ; ditto thorax, 0.0710 gram. After the 

 isual tests to prove the absence of arsenic in the reagents it was found that no arsenic 

 was associated with the posterior legs or the pollen with which they were loaded, none 

 had been left in the thoracic matter, but the material from the abdomens gave unmis- 

 takable proof of the presence of arsenic. The entire bodies of a nvmiber of the bees, 

 taken at the same time from the same tree, were then washed with diluted ammonia 

 water, three washings failing to give a trace of arsenic, but the bodies, after being thus 

 treated, and being boiled in water slightly acidulated, gave distinct traces of the poison, 

 ihus eliminating any possibility of the poison having been introduced into the abdom- 

 inal matter at the time of dissection and from the exterior. May 15th a crabapple tree 

 iCralaigus) was sprayed with a mixture of the same ratio of Paris green as before, but 

 in this case only the contents of the abdomens were retained. This matter, to the 

 weight of 0.1463 gram, treated as in the preceding, gave unmistakable, proof of the 

 presence of arsenic. 



Just at this stage of my investigations, chance, if such a thing there be, threw in 

 my way still more conclusive proof. A few days prior to my last experiment, probably 

 about May 10th, a small apple orchard on the experiment farm was sprayed with Borde'jux 

 mixture, to which had been added Paris green at the rate of 1 ounces to each 50 gallons 

 of the mixture. The bloom had at this time nearly all fallen from the trees the excep- 

 tions being an occasional belated cluster. Three colonies of bee-s, recently brought on to 

 the premises, weie located near by, to all appearances in a perfectly healthy condition. 

 A few days after the application of the poisoned Bordeaux mixture one colony suddenly 

 became extinct and a second greatly reduced in numbers, dead bees being abundant 

 about both hives. From these colonies I was able to secure dead bees, and both honey 

 from uncapped cells and dead brood from the hive that had been so mysteriously depop- 

 ulated. When tested for arsenic by Mr. Bloomheld, precisely as with the other matter, 

 contents of abdomens of the dead bees to the amount of 0.2334 gram revealed the 

 presence of arsenic ; 3.7061 grams of honey gave no trace of poison, while 1.8481 grams 

 dead brood showed it to be prtsent, acd the entire bodies of the dead bees, thrice washed 

 in ammonia water, as before explained, gave traces of arsenic. In regard to the honey I 

 can only say that it was from uncapped cells, which might and probably did contain last 

 year's honey that was still being used for a partial food supply by the bees. 



Briefly recapitulated, arsenic was found present in the contents of the abdomens of 

 bees frequenting recently sprayed blossoms, and we are at least free to assume that more 

 or less of it was contained in the honey sacs. The dead bees three times washed in 

 ammonia water, the latter not revealing the presence of arsenic externally, when tested 



