328 WASP STUDIES AFIELD 



another, not only sucking out the juices, but consuming the 

 entire carcass, excepting the dry shell of the little black 

 heads (see fig. 64, the remains of a feast). The supply of 

 food provided by the mother for the growing infant varies 

 greatly, but we have never ascertained whether that supply 

 is sometimes actually inadequate or excessive, or whether 

 the young merely accepts whatever is given it and gets 

 through somehow. In our eagerness to keep alive all that 

 we had at home for observation, we gave all of them addi- 

 tional food and it was almost always accepted greedily. 

 Once we were so daring as to offer a fat, prosperous larva 

 a small brown caterpillar instead of the customary green 

 ones. The larva turned at once to this new fresh food, 

 but mumbled about it mincingly for only a minute and then 

 went back to resume chewing at its old half -eaten green car- 

 cass. We suspect that some of them are in real need of 

 more food before they pupate, especially since some have as 

 few as three little caterpillars upon which to subsist. We 

 chanced to find one good-sized larva which must have been 

 eating dirt, for the abdomen was full of earth. Yet over- 

 feeding probably tends to make them overfat and delicate. 

 We had one which was a thriving infant and promised to 

 be a fine adult. After it had sucked dry seven caterpillars 

 it was enormously large and fat. As we transferred it to 

 a clean bottle, in the hope of seeing it spin its cocoon, we 

 let the vial fall to the table, and the jar caused the body 

 walls of the fatling to burst. Of course, in nature the larva 

 runs no risk of such a catastrophe, and yet the circumstance 

 suggests that overfeeding may make them, as well as other 

 organisms, soft. 



There is an astonishing degree of variation in the dura- 

 tion of the different periods of development in this species. 

 There seems to be no fixed time for hatching, spinning or 

 any of the functions. Moreover, we have not yet ascer- 



