32 INSECT ARCHITECTURE. 



from his bees not having been able to procure moist clay. 

 On the contraiy, sand was the chief material used by the 

 mason-bees (^Megachile muraria') ; which they had the patience 

 to select from the walks of a garden, and knead into a 

 paste or mortar, adapted to their building. They had 

 consequently to expend a much greater quantity of saliva 

 than our bees (^Osmia bicurnis), which worked with moist 

 clay. Eeaumur, indeed, ascertained that everj^ individual 

 gi-ain of sand is moistened previous to its being joined to 

 the pellet, in order to make it adhere more effectually. 

 The tenacity of the mass is, besides, rendered stronger, he 

 tells us, by adding a proportion of earth or garden-mould. 

 In this manner, a ball of mortar is formed, about the size of 

 a small shot, and carried off to the nest. \Yhen the sti'uc- 

 ture of this is examined, it has all the appearance externally 

 of being composed of earth and small stones or gravel. 

 The ancients, who were by no means accurate naturalists, 

 having observed bees carrying pellets of earth and small 

 stones, supposed that they employed these to add to their 

 weight, in order to steady their flight when impeded by 

 the wind. 



The nests thus constructed appear to have been more 

 durable edifices than those which have fallen under our ob- 

 servation ; — for Eeaumur says they were harder than many 

 sorts of stone, and could scarcely be penetrated with a 

 knife. Ours, on the contrary, do not seem harder than a 

 piece of sun-baked clay, and by no means so hard as brick. 

 One circumstance appeared inexplicable to Eeaumur and 

 his friend Du Hamel, who studied the operations of these 

 insects in concert. After taking a portion of sand from one 

 part of the garden-walk, the bees usually took another 

 portion from a spot almost twent}^ and sometimes a hundred 

 paces off, though the sand, so far as could be judged by 

 close examination, was precisely the same in the two places. 

 We should be disposed to refer this more to the restless 

 character of the insect than to any difference in the sand. 

 We have observed a wasp paring the outside of a plank, for 

 materials to form its nest ; and though the j^lank was as 

 uniform in the qualities of its surface, na}^ probably more 



