H Y MEN OP TERA . 635 



demonstrated by repeated experiments that in the case of 

 our common carpenter-ant {Cainponotus pennsylvanicus) the 

 former method is practised. But it is not improbable that 

 with certain other species the latter method occurs. 



On many occasions we have found a queen of the car- 

 penter-ant in a small cleared space beneath the bark of a 

 dead tree or log. Some of these queens were alone ; others 

 were accompanied either by eggs, larvae, or by small workers, 

 On one occasion we collected several such queens and placed 

 each with her eggs in a cell between plates of glass in an 

 artificial ant's nest, and have thus watched the beginnings 

 of colonies. A few eggs, from ten to fifteen, are laid at 

 first ; these soon hatch, and the larvae develop quite rapidly. 

 A nest which on July 15th contained, besides the queen, 

 only seven eggs, contained July 27th thirteen eggs, three 

 larvae, and one cocoon ; and on Aug. 14th there were six 

 cocoons. In another nest, which on July 15th contained, 

 besides the queen, only young larvae, the larvae began to 

 spin cocoons on July 19th, and on Aug. 8th the workers 

 began to emerge. On Aug. i6th the workers had begun to 

 work, carrying the empty cocoons out of the nest, and on 

 Aug. 20th the workers began to take into the nest dead flies 

 that had been placed at the entrance. 



The most remarkable result of this experiment was the 

 demonstration of the fact that from the time the queen forms 

 her cell and begins to lay eggs to the time when a brood of 

 workers is matured no food is taken into the nest. The cell 

 is a closed one, and contains no store of food except what 

 may be within the body of the queen. The queen does not 

 leave the nest ; and when we placed food within a nest the 

 queen built a wall of earth about it, thus walling it out. 



To test this matter a queen was placed with some of her 

 eggs in an empty vial, and Swiss muslin was tied over the 

 mouth of it. Here, where the queen could not possibly 

 obtain food, the larvae matured, spun cocoons, and adult 

 workers emerged. The queen was often seen to apparently 



