NOTES ON A COLORADO ANT. 75 



Experiment No. 1. — One-half ounce of 98 per cent cyanic! of potash 

 dissolved in 1 (gallon of water was used. On Au.^ust 31,1 909, at 5 p. m., 

 2 gallons of this solution were applied to a nest 2\ feet in length by 

 2 feet in width. The entire outer surface of the nest was soaked and a 

 considerable quantity was poured directly into the openings. Ants 

 which were hit died almost at once and others which returned from 

 the field and ran over the wet surface died within a few seconds. 

 When the nest was examined an hour later the surface was well 

 covered with dead specimens. There was still a fairly strong odor 

 of the cyanid from the wet soil and returning ants were soon killed, 

 although they did not die quite as rapidly as when the application 

 was first made. 



Experiment No. 2. — One ounce of 98 per cent cyanid in 1 gallon of 

 water was used. On August 31, between 5.30 and 5.45 p. m., 4 

 gallons of this solution were applied to two nests, each about 3 feet 

 long and 2 feet w4de. The conditions were as in Experiment Xo. 1 

 and the immediate results appeared to be about the same. 



Experiment No. 3. — Two ounces of 98 per cent cyanid in 1 gallon 

 of water were used. On August 31, at 6 p. m., 2 gallons of this 

 solution were applied to a nest about 3 feet long by 2 feet wide. 

 The immediate results appeared to be about the same as in Experi- 

 ments Nos. 1 and 2, although there was a somewhat stronger odor 

 of the cyanid from the wet soil. 



At the time these three tests were made the sun was warm and 

 shining brightly. The ants were very active and thousantls of them 

 were away from the nests and among the aphis-infested cantalouoe 

 vines. 



Since the larger lumps of cyanid dissolved rather slowly some 

 time was gained by breaking them up with a hammer. 



At 4 p. m. on September 1 an examination was made of the nests 

 treated in these tests. At that time there were hundreds of dead ants 

 lying on the surface of the nests and a comparatively small number 

 of specimens was running about. Most of the living ants had appar- 

 ently lost interest in the aphides and had gathered on or about the 

 treatetl nests and some w^ere carrying dead specimens. There 

 appeared to be little difference between the results of Experiments 

 Nos. 1 and 2, but there were certainly fewer live ants about the nest 

 treated in Experiment No. 3 than about the others. 



As some fear was felt that a strong solution of the cyanid might 

 kill the small elm trees which occupied tlie ant-infested strip and as 

 Experiment No. 1 gave comparatively good results, it was concluded 

 to continue the w^ork with that strength. Accordingly, between 

 4.30 and B ]). m. on September 1, the remainder of the infested .>trip, 

 about 65 yards in length and containing 21 nests, was treated witli 



