AN ISLAND SETTLEMENT. 65 



ed, reaching only one-half or one-third of its nornml size. 

 When it attempts to spin its cocoon it has not strength enough 

 to do so and thus perishes miserably among the pupae of the in- 

 terlopers, which have the advantage of developing more rapidly. 

 He has proved, by experiments upon nests transported to his 

 study, that although the invaders preserve friendly relations 

 with the rightful owner of the nest so long as food is abundant, 

 they nevertheless, at the first suggestion of scarcity, fall upon 

 the wasp larva and ruthlessly devour it. This "black action" 

 lie l^as seen with his own eyes. In view of this base ingratitude 

 we are more than ever impressed with the troubles of the poor 

 Bemhex mother as she tries to feed a dozen mouths where she 

 has bargained for only one. 



We several times saw a fly follow a wasp into her nest, re- 

 maining within for half a minute, and it is probable that they 

 go in to lay their eggs. According to Fabre, it is the habit of 

 the flies that are parasitic upon the half-dozen species of Bem- 

 bex that he has studied to seize the moment at which the fly 

 projects from under the abdomen of the wasp as she enters the 

 nest, and he has even known them to lay two or three eggs on 

 one fly in the instant of time that its body was exposed. 



When we first found the colony on August tenth, it was 

 strong in numbers, and it continued to grow up to the first of 

 September. On going over one day in the middle of August 

 we found a good deal of work going on, but to our surprise the 

 flies had disappeared. A careful search showed only two in- 

 stead of the usual numbers. Their place was taken by dozens 

 of little wasps, which proved to be Pompilus higuttatus Fabr. 

 These wa^ps seemed to be merely loafing about amusing them- 

 selves, neither eating nor working. Whether their presence was 

 in any way connected ^^viih. the absence of the flies we are un- 

 able to say but it seems probable that they had congregated on 

 the Bemheoo gTound merely because it was an open spot, since 

 on looking about we found them covering every open sunny 

 space that the island afforded. 



Fabre took a partly grown Bemhex larva from the nest, where 

 5 



