10 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.64. 



quired to complete the first stage in development varies. Of two 

 parasites taken from the ground in the act of devouring the egg of 

 ilie host, one required 11 days to reach the second instar while an- 

 other required but 5. 



It frequently happens that two or more first instars gain entrance 

 into the same cell, but in every case observed, either in the field or 

 in the laboratory, only one survived. In removing the brood cells 

 of Colletes from the ground on June 9, 1920, one cell was crushed out 

 of shape, but no rupture occurred in its walls. When examined in 

 the laboratory a first stage Meloid larva was discovered submerged 

 in the disturbed food mass. The side of the cell was opened, the 

 parasite fished out, given a bath in water to remove the honey from 

 it, and was restored to the cell, which was placed in a breeding vial. 

 The egg of the host could not be discovered in the cell and the larva 

 showed by its appearance that it had taken food before the cell had 

 been removed from the ground. On the morning of June 14 it was 

 found that this larva had molted and was lying dead on the food 

 mass and that a second larva in the first stage was present in the 

 cell. The dead larva and its cast skin were removed from the cell at 

 once and the second parasite left in the cell. On June 19 this second 

 parasite molted. On the morning of June 20 a third parasite in 

 first stage appeared in the cell. This third one had killed the newly 

 molted second instar and was feeding on its body. The cast skin 

 and the dead body of the second instar were removed from the cell 

 and the third parasite left in possession. This third parasite, a 

 first instar, molted on July 2. 



Now, all three of these parasites were in the cell when it was taken 

 from the ground, for Colletes seals her brood cell up after she de- 

 posits her egg so that nothing can get in without rupturing the cell 

 wall. All three must have been submerged in the food mass when 

 the cell was crushed out of shape in its removal from the ground, 

 but all three survived in spite of the fact that only one was dis- 

 covered and had the honey removed from it. How the egg of the 

 host was destroyed, whether by the three jointly or by the first alone 

 (which is most probable) is not known. But that the first larva to 

 molt was killed by the second and the second by the third is quite 

 evident. Here then is a case where a first instar introduced into a 

 cell of Colletes prior to June 9 and later messed up in the food sur- 

 vived and completed this first larval stage July 2, a period of more 

 than 23 days. 



In the development of the larval beetle there are in all six instars 

 and consequently six cast skins before the pupal stage is reached. 

 Aside from that relating to the first larval stage the data obtained 

 dealing with the length of time required to complete the other larval 



