BEHAVIOR OF POMPILID WASPS 57 



in one corner) walked over it with wonderful lightness and 

 rapidity and entered the lair. She came out at once with- 

 out the spider, however, walked without the slightest en- 

 tanglement to the edge of the web and flew off. Perhaps 

 the spider spied her and escaped, but it was surprising that 

 the Pompiloides was not ensnared in the web, as so many 

 of the larger insects are, especially since this is the whole 

 purpose for which the web is constructed. We have else- 

 where remarked that dead Chalybion caenileum and Eu- 

 menids are often found in spiders' webs. 



Others were seen in other regions and at other times, ap- 

 parently foraging. Some were out in the grassy fields, and 

 others in barren, clay-covered areas. They seem to be gen- 

 erally distributed, whereas Bembix nubilipennis and other 

 wasps are found only in specific places. 



It seems, from what we have noted, that the primary 

 sting is insufficient to paralyze the spider and that whenever 

 P. tropicus is aware of any activity on the part of her prey, 

 she re-stings it. 



One record shows that a spider may recover from the 

 sting. A small female P. tropicus in alarm left her spider, 

 a medium-sized specimen of Pellenes coecatus Htz. [N. 

 Banks] at our feet on June 12, 1916. This tribute was 

 cherished, even though it appeared to be dead. It was 

 daily stimulated, but it was not until four days later that 

 the stimulation brought forth the slightest movement of 

 the legs. Every day the spider was mildly stimulated with 

 the same sort of response, but it was not until the fifteenth 

 day that the spider showed tendencies to revive. On that 

 day, June 27, it moved the legs more actively and jumped 

 slightly. On July 4 it was so lively that, when the box was 

 opened, it jumped out. It had also just spun a web in the 

 box. The next examination, July 10, found it very active, 

 as did also the inspection on July 20, so it was released in the 



