PERFECT SOCIETIES OF INSECTS. 197 



per at the other end ; this was exactly opposite to the 

 aperture, at the distance of about an inch from it. As 

 soon as the apparatus vv as fixed, the hair with its paper 

 pendulum began to oscillate more or less, the greatest 

 oscillations on both sides being an inch, by admeasure- 

 ment, from the perpendicular : if the paper was moved 

 by force to a greater distance, the vibrations did not 

 take place, and the apparatus remained at rest. He 

 then made an opening* in the top of the hive, and 

 poured in some liquid honey : soon after there arose a 

 hum, the movement in the interior increased, and 

 some bees came out. The oscillations of the pendu- 

 lum upon this became more frequent and intense, and 

 extended to fifteen lines or an inch and a quarter from 

 the perpendicular ; but when the paper was removed 

 to a greater distance from the aperture it remained at 

 rest. 



Huber, at the proposal of M. de Saussure, in order 

 to ascertain whether artificial ventilators would pro- 

 duce an analogous eiTect, got a mechanical friend to 

 construct for him a little mill with eighteen sails of tin. 

 He also prepared a large cylindrical vase, into which 

 he could, at an aperture in the box upon which it was 

 fixed, introduce a lighted taper. In one side of this 

 box was another aperture to represent that of a hive, 

 but larger. The ventilator was placed below, and 

 luted at the points of contact, and anemometers were 

 suspended before the aperture. The first experiment 

 was the introduction of the taper, without putting- the 

 ventilator in n)otion. Though the capacity of the ves- 

 sel was about 3'2'28 cubic inches, the flame soon dimi- 

 jiishedj and went out in about eight niinutesj and the 



