The third syrup (i gm. sol. containing .000063 gra., arsenic — See 

 Table I) was diluted with equal parts of sugar syrup and given ..to seven 

 bees under the cage. A supply was kept near them continuously. This 

 was given them on July 19, 1 130 p. m. At 8 a. m. on July 20 there were 

 four dead and two on their backs ; on July 20, at 2 p. m., there were five 

 dead and two on their backs; at 10:30 p. m. on July 20 all were dead. 



This same syrup (.000063 gms. arsenic per gm. syrup — See Table I) 

 was diluted again. This time three parts of the poisoned syrup was diluted 

 with one part clean syrup. Eleven bees were given the syrup in abundance 

 on July 19, at 2 :i5 p. m. On July 20, at 8 a. m., there were eight dead 

 and one on its back ; at 2 p. m., nine were dead ; and at 10 130 all were dead. 



A third dilution of the .000063 gm. poisoned syrup, was made by 

 adding three parts of clean syrup to one part of the arsenic syrup. This 

 was supplied in the same way as the above to eight bees on July 19, at 

 2 p. m. On July 20, at 2 p. m., one was on its back, all others appeared 

 healthy; at 10:30 p. m., on July 20, one was dead and two on their backs. 

 On July 21, at II :3o a. m., two were dead and three were on their backs; 

 on July 22, at 8 a. m., all were dead. 



Of the 27 bees used as a check on the last three groups, only two 

 died during the period of observation.^ 



In order to see what the efifect of the ordinary commercial arsenate 

 of lead would be, a 40 per cent sugar solution was made containing the 

 equivalent of one pound of powdered arsenate of lead to 50 gallons. It 

 was fed to 26 bees on July 23, at 10:15 a. m. All bees were in one cage, 

 and no individual records were kept. Twenty-five of the 26 died during 

 the night of July 23, and all were dead at 10 a. m. on July 24. A check 

 of 31 carried alongside of this cage showed one dead bee. 



It will be seen from the data given above that the bee is capable of 

 taking into its system at one feeding .063 grams of syrup. This amount 

 of syyup in our first dilution contained .0000371 grams of arsenic, which 

 was the largest dose of poison taken in, these experiments. The bee lived 

 150 minutes after feeding. Another bee with a smaller dose of poison 

 (.0000142 grams) lived no minutes after feeding. The smallest dose 

 administered was .00000055 grams of arsenic, and the bee lived 590 min- 

 utes after feeding. The experiments offer evidence that a very small 

 amount of arsenic is fatal to the bee, and that a bee takes up lead arsenate 

 particles in feeding on a sugar solution having them in suspension. 



It was noticed in the control feeding work that the poisoned bee began 

 to behave abnormally (sick) soon after feeding, usually within 20 to 30 

 minutes. It would drop from the top or side of the cage to the floor 

 buzzing and spinning on its back like a decapitated fly. It seemed unable 

 to right itself and continued to struggle in this way until it finally died. 



1 In all the above experiments arsenic in soluble form was used, because it lent itself 

 to more accurate computation in the small doses given, The commercial arsenate of lead 

 is insoluble in water and nectar, and therefore it was impracticable to use it in these tests. 

 The small particles of arsenate being in suspension or grouped in the liquid prevents equal 

 distribution. A bee feeding on a certain amount of the liquid might take a large amount 

 of the poison particles, whereas another bee taking a like amount of liquid from the same 

 receptacle might take only a small amount of the poison particles. There would thus be 

 no reliable way of computing the amount of arsenic in the doses taken in the control feed- 

 ing experiments. 



