.INSTINCT OF INSECT<5. 483 



larvsB, invariably give a convex lid to the large cells 

 of drones, and one nearly flat to the smaller cells of 

 workers : but in an experiment instituted by Huber to 

 ascertain the influence of tlie size of the cells on that 

 of the included larvae, he transferred the larvae of work- 

 ers to the cells of drones. What was the result ? Did 

 the bees still continue blindly to exercise their ordi- 

 nary instinct ? On the contrary, they now placed a near- 

 ly ^a^ lid upon these large cells, as if well aware of 

 their being occupied by a different race of inhabitants'^. 



On some occasions bees, in consequence of Hul)er's 

 arrangements in the interior of their habitations, have 

 begun to build a comb nearer to the adjoining one than 

 the usual interval ; but they soon appeared to perceive 

 their error, and corrected it by giving to the comb a 

 gradual curvature, so as to resume the ordinary di- 

 stance''* 



In another instance in which various irregularities 

 had taken place in the form of the combs, the bees, in 

 prolonging one of them, had, contrary to their usual 

 custom, begun two separate and distant continuations, 

 which in approaching instead of joining would have 

 interfered with each other, had not the bees, apparently 

 foreseeing the difficulty, gradually bent their edges so 

 as to make them join w ith such exactness that they 

 could afterwards continue them conjointly^. 



In constructing their combs, bees, as you have been 

 before told, in my letter on the habitations of insects, 

 form the first range of cells — that by which the comb 

 is attached to the top of the hive — of a different shape 

 from the rest. Each cell instead of being hexagonal 



" Huber, i. 233. " Ibid. ii. 239. * Ibid. ii. 240. 



2 1 2 



