240 Journal Xew York Entomological Society. t^'"'- ^i^- 



forcibly ejected over the rim. Sometimes the pellet.s are thrown to 

 a di.stance of only half an inch, but usually one or two inches or, in 

 rare instances, four inches. Occasionally a pellet is thrown into the 

 burrow of another bee and if the latter is also digging the pellet is 

 simply removed like the others by being brought to the surface with 

 the hind legs. Frequently, however, the mouth is used and the pellet 

 tossed over the top of the turret. 



Fig. I. Larva of Eiitphor bombifoniiis Cress. 



The burrows when completed are somewhat more than one fourth 

 inch wide and of varying depth, two and one half inches being about 

 the average depth though some are four inches deep and others less 

 than one inch. At the bottom the burrow is widened into a cell. 



Not infrefiuently the bee encounters difficulties in commencing a 

 burrow. In the center of the colony where the holes are most nu- 

 merous the pellets are littered over the ground to such an extent 

 that frequently only the rim of the highest turret is visible. Instead 

 of extending the colony by working at its outer borders a bee occa- 

 sionally digs among these pellets and, as fast as the dry pellets are 

 pushed aside, others roll into the forming hole, and it is only after 

 great exertion that a good beginning is effected : sometimes the 

 project is abandoned in these places and a new hole started some- 

 where else. 



At one time a bee was observed repairing a turret which was 

 slightly broken and partly covered over with pellets. On first return- 

 ing, the nest was not positively identified by the bee; it made a 



