5''6 DKSTRUCTION OF THE ARGENTINE ANT. 



wander around on it from side to side in a confused manner, as 

 if loath to leave it. On three occasions an incominj^ ant seemed 

 to communicate with an ant on the ])oison. and then suddenly 

 catch it b}- a leg or an antenna and pull it olif the ])oison, and 

 then release it after struggling for a few seconds. Others would 

 skirmish along the poison and suddenly " storm " across it ; but 

 the great majority refused to cross, and finally retreated along 

 the trail. On the whole I felt satisfied that unless there was some 

 exceptional reason for crossing the ])oison they would avoid it. 

 Circumstances prevented further tests being made until February. 

 1914. when I had occasion to place some eggs of the Maize Stalk 

 Borer {Scsainia fusca Hamp.) on a young maize plant at the 

 J-ioseband Experiment Station.' Argentine Ants were abundant, 

 and to prevent the eggs being carried away I decided to isolate 

 the plant by a cordon of finely-powdered corrosive sublimate 

 before covering it with a cage.' In view ni my observations in 

 August. 1913, I instructed the Assistant bjitomologist, Mr. C. 1'. 

 van der Merwc, who was responsible for making detailed o])ser- 

 vations. to keei) careful watch and note the effect of the corrosive 

 sublimate on the ants. After ;i few days had elapsed he called 

 my attention to the fact that there were a great many dead 

 Argentine Ants in the cage around the corrosive sublimate. I 

 instructed him to continue the observations with a view to 

 determining the cause of the mortality, and he later on reported 

 that the ants, as s(X)n as they came in contact with the poison. 

 es]>ecially in bright sunlight, became confused, and in a short 

 time started " leg-pulling." as he described it — I.e.. one ani would 

 seize another ant by the leg or antenna, and start pulling; a 

 third ant would seize a leg of one of the two contestants, and 

 so on till there was a " chain "' of five or six ants. This at once 

 recalled the fact that during the first trial the ants showed a 

 tendency to ])ull each other about. < )n watching the ants closely 

 b\ means of a magnit\-ing glass, I soon saw that tliev were not 

 inerelv " leg-i)ulling." but that it was a real tug-ot war in which 

 they were literally tearing each otlicr to pieces. In the mass of 

 dead ants the great majoritv had l)een injured to some extent. 

 l)Ut many died without being mutilated. It seems ])rol)able that 

 thev were ))oisoned by trying to clean their antennae and feet. 



The finelv-powdered sublimate i^ more active in sunliyiit 

 than in the shade. 



With the assistance of Mr. \'an der .Merwe. a large niunber 

 of experiments were conducted under \arying conditions, and it 

 soon became evident that by placing a very little of the finely- 

 powdered sul)limate around the entrance to the nest, the whole 

 colony would ultimately be annihilated because they could neither 

 leave nor enter the nest without getting into the poison. Forag- 

 ing queens are affected the same as the workers, the latter attack- 

 nig the (|ueen the .same as they do each other. They are evidently 

 too confused to recognise her. 



Under some conditions the ants become excited before they 

 actually touch the powder. This suggests that there may l)c 



