ART. 23. BLISTER BEETLE TRICKANIA PARKER AND BOVING. 7 



ing of the burrow, attach themselves to the males of the host, which 

 emerge first and are carried about by the males till the females ap- 

 pear and then during copulation/ transfer themselves to the females 

 and in this manner finally reach the brood chamber of the host. It 

 seems highly probable that Tricrania sanguinipenuls makes use of * 

 similar methods in gaining access to the brood cells of CoUetes riiji- 

 thorax. The eggs of the beetle are laid in a cluster as stated above 

 in the vicinity of the burrows of the host and the larvae when 

 hatched scatter about activel}^ in all directions and this dispersal 

 takes place normally at the time Avhen the males of the host are 

 most active in searching out the females. In their efforts to find the 

 females, the males range hither and thither over the nesting site, spend- 

 ing much of their time crawling about over the ground and dodg- 

 ing in and out of burrows. The parasites are thus given far greater 

 opportunities to attach themselves to the male than to the female, 

 for the latter is seldom on the ground save when entering or leaving 

 a burrow. When to these facts we add the data given above, that, 

 on infested males and females of Colletes taken in the field at a time 

 when matings were in progress, the majority of the parasites on the 

 males were found on the ventral side of the body whereas on the 

 females they were found on the dorsal side, it would all seem to 

 indicate that the male of the host is an active agent in enabling the 

 parasite to attach itself to the female. 



But there is evidence to show that the parasite is not wholly de- 

 pendent upon the male to find its way to the female of its host. In 

 opening the burrows for cells of the host containing parasites, it was 

 invariably found that where nests were opened in an area on which 

 a clutch of the eggs of the parasite had hatched or on which a large 

 number of the parasites hatched in the laboratory had been turned 

 loose, the percentage of infested brood cells of the host was always 

 greater than was the case in places more remote from such centei"S of 

 dispersal of the parasite. If the parasite depended entirely upon 

 the male to gain lodgment on the female this difference in degree of 

 infestation in different parts of tlie nesting area Avould be hard to 

 explain, since the males roam freely and uniformly over the extent 

 of the nesting area in search of the females. Furthermore, cells 

 infested with parasites were found that were constructed long after 

 all males had disappeared from the field, cells that were constructed 

 by bees whose previously constructed cells did not contain the para- 

 site. In addition, some females of Ancb^ena perplexa Smith (of 

 which species the males perish before the young of Tricrcniia hatch) , 

 taken at the time the parasite was active in the field, were found with 

 first instars attached to them. These parasites must have attached 

 themselves directly to the bee and, in spite of the fact that Andrena 



