DEFENCE AND SANITATION OF DWELLINGS. 241 



with wax, leaving only a narrow opening through 

 which the great robber could not penetrate. Others 

 built up before the opening a series of parallel walls, 

 leaving between them a zigzag corridor through which 

 the Hymenoptera themselves were able to enter. 

 But the intruder was much too long to perform this 

 exercise successfully. Man utilises defences of this 

 kind ; it is thus at the entrance of a field, for 

 example, he places a turnstile, or parallel bars that 

 do not face each other ; the passage is not closed for 

 him, but a cow is too long to overcome the obstacle. 

 In years when the Death's-head Moth is rare the bees 

 do not set up these barricades, which, indeed, they 

 themselves find troublesome. For two or three con- 

 secutive years they leave their doors wide open. Then 

 another invasion occurs, and they immediately close the 

 openings. It cannot be denied that in these cases their 

 acts agree with circumstances that are not habitual.^ 



Precautions against inquisitiveness. — I will finally 

 quote a fact of defence which took place under circum- 

 stances that were absolutely exceptional, and which 

 therefore exhibits genuine reflection in these insects. 

 During the first exhibition of 1855 ^n artificial hive 

 was set up, one face of which was closed by a glass 

 pane. A wooden shutter concealed this pane, but 

 passers-by opened it every moment to contemplate 

 the work of the small insects. Annoyed by this 

 inquisitiveness, the bees' resolved to put an end to it, 

 and cemented the shutter with propolis. When this 

 substance dried it was no longer possible to open the 

 shutter. The bees were visible to nobody. 



Lighting up the nests. — An improvement of another 



^ These facts have recently been observed and recorded afresh by 

 Mr. Clifford in Nature Notes, January 1893. 



16 



