98 [May, 



were strewn with dead umhratus workers, several bundreds of them, 

 though not a single niger was to be s^een. I searched the box carefully, 

 but could not find any trace of niger. 



During the years 1901 — 1907 I had no opportunity of trying a ? 

 umhratus with a queenless nest of niqer, but I tried without success to 

 induce a nest of the latter to accept a ? of their own species (as 

 already mentioned). T also tried a queenless nest of niger with a 

 fertile flavus ? . In this case niger workers, even after being a whole 

 day with the 9 , never got over their fright and hostility, and the $ 

 was attacked at once on being put into the nest. 



On July 29th of last year (1908), I took a nest of niger, but did 

 not succeed in finding the queen. The nest consisted of between 200 

 and 300 workers, and about an equal number of pupse, with three ^ 

 pupae, wliich latter were among the first to hatch. I kept the nest 

 untouched till September. 



On September 18th I picked up several ieviWe umhratus $ $ , and 

 enclosed two, each in a separate box, with a worker from my nest of 

 niger. Both ? ? instantly killed the «?9'er5, just as happened in 1896. 

 I put another niger to one of the ? $ . She killed this one also, but 

 it had fastened on to one of her legs, and she could not remove it. 

 Next day I put i\.flavus worker to her. She showed no hostility, but 

 the flainis attacked her, so I separated them. Another niger put to 

 her saluted her amicably, and she caressed the niger. This was at 

 11.15 a.m. on September 19tli. At 11.25 I put in another niger, 

 who was also friendly, and at 11.50 I put in a third, followed by three 

 others. At 12.5 p.m. I removed the lid of the box, and put the latter 

 an inch and a half from the door of the nest of niger. The nigers 

 went into the nest one by one, the ? remaining in the box. In less 

 than two minutes after the first worker entered the nest, more than 

 sixty ants came out and surrounded the umhratus ? . There was great 

 excitement, but no hostility. Still surrounded by ants, the ? of her 

 own accord found her way to the narrow door of the nest. On the way 

 she was pulled by an antenna for a moment. An ant met her at the 

 entrance, and pulled her in by an antenna. At 12. IG she was in the 

 middle of the nest, with crowds round her. The ants seemed altogether 

 calmer over her appearance than in the former case in 189G, but she 

 was evidently accepted as queen. Next morning, on removing the cover 

 of the nest, I saw a dead $ ant in the nest. It proved to be one of 

 the three young winged ? $ (niger) that had hatched in the nest. 

 The ants had killed all three and removed their wings. 



By September 22nd the workers had removed the dead niger that 



