106 PROC. ENT. SOC. WASH., VOL. 23, NO. 5, MAY, 1921 



more generous supply of food for some of her offspring than for 

 others. Whether this is due to mere chance or whether those 

 provided with a limited amount of food develop into males and 

 those more generously supplied into females remains to be 

 investigated. As in the case of the preceding species the wasp 

 does not deposit an egg upon the first grasshopper brought into 

 the brood chamber, at least not at the time she brings it in. I 

 found several brood chambers not sealed up containing from one 

 to three grasshoppers and in every case no egg was present. 

 From this it appears that the wasp does not place an egg in a 

 brood chamber until it has been provided with what the wasp 

 considers an adequate supply of food for her offspring. 



In the case of this species (breviventris) the egg is placed upon 

 the ventral side of the thorax just behind the front leg. The 

 anterior end of the egg rests upon the mid line of the thorax just 

 behind the prothorax. In the case of one wasp the posterior 

 end of the egg extended to the left side of the body of the grass- 

 hopper whereas in the case of another wasp it extended to the 

 right. I did not succeed in finding a newly laid egg; all that were 

 found were in a more or less advanced stage of incubation, so 

 that the time required for the egg to hatch was not learned. An 

 egg found on September 27, 1920, in a brood chamber provided 

 with four grasshoppers, hatched on September 28. The larva 

 moulted on September 29 and completed its feeding on October 

 1. I could find no evidence that the larva moulted more than 

 once; but it may have done so. On October 2 it was trying to 

 form a cocoon while lying on the surface of the sand in the breed- 

 ing cell. I left it there until October 5 but it was unable to form 

 a cocoon. I then filled a large vial with moist sand and after 

 pressing the sand down firmly made a hole in it with a lead pencil. 

 I put the larva into this hole and plugged the entrance of it with 

 moist sand and laid the vial on its side. Thus enclosed the 

 larva formed its cocoon successfully. Repeated experiments 

 convince me that the larva can not form its cocoon unless sur- 

 rounded with sand. If the larva fails to form a cocoon within 

 a limited time after completing larval growth it dies. 



The cocoon is formed of sand held together by a cementing 

 substance furnished from the mouth of the larva. In beginning 

 the formation of a cocoon the larva spins a loose network of 

 silken threads about the body. This serves as a framework to 

 hold the grains of sand, which, coated with the cement substance, 

 the larva puts in place with the aid of its mandibles. When the 

 wall of the cocoon is thus formed the larva inside, using its mouth 

 parts in much the same fashion as a painter uses a brush, coats 

 the inside surface of the cocoon with a substance that further 

 binds the grains of sand together and gives the interior a smooth 

 finish. While this work is going on the wall of the cocoon is 

 flexible and the yielding of the wall to the pressure exerted in the 



