44 BULLETIN 66, UNITED STATES NATIONAL MUSEUM. 



nests up as soon as sufficient food is stored for the nourishment of 

 the larva, belong under this grouping. It is needless to say that in 

 such cases there is but the brief period of about a day during which 

 the triungulinids may reach their prospective hosts, and that they 

 must in some cases wait several days for the host larva to leave the 

 egg stage. When the hosts are predaceous the chances are that they 

 seldom visit flowers during the nest building period, and it may be 

 necessary for a triungulinid to remain for days on a flower or on a 

 secondary host while it awaits an adventitious moment for furthering 

 its search of a permanent host. 



When the hosts are pollen-gatherers it is probable that multitudes 

 of triangulinids may be swarming over the nest builders awaiting an 

 opportunity to settle in newly made nests. 



HeteropJiagous larvse. — Those young which require to be fed by 

 others have been called heterophagous (from 'hspoc = another + 

 <f)aye~cv = to feed). Under this grouping belong the Formicoidea, 

 Vespoidea and Bembex and Monedula in the Bembecida\ Concern- 

 ing these latter genera, Hartman (1905, 25) writes that "Microbem- 

 bex, contenting herself with any insect she finds, has an advantage 

 over Bembex, her nearest relative, and Monedula, both of which feed 

 their larvae from day to day." Peckham and Peckham in "The Soli- 

 tary Wasps," page 67, corroborate this statement, but Hartman 

 shows (1905) that Bembex belfragei after storing the hole closes it up. 



As these larvae are accessible to triungulinids throughout this period 

 of their existence it is reasonable to expect that parasitism among 

 them should reach a very high percentage. 



d. Life cycle of hosts. — Homoptera. — Concerning the leaf hoppers, 

 Osborn and Ball (1897, 613) state that: 



There is a wide difference in life histories, some having one brood, the majority 

 of the grassfeeding species two, and still others three in a season, and the successive 

 stages occurring at widely different times. 



Except in the case of adult hibernation the ordinary life of a brood of adults does 

 not exceed two months, and for the individuals of a brood rarely over one. The 

 males appear a week or ten days before the females and disappear as much earlier. 

 In general one brood of adults will have disappeared before the larvae of the next 

 have matured, so that individuals collected at any time may be referred with assur- 

 ance to a particular brood. 



This summary is very useful, as it indicates in a few words the 

 exigencies of parasitic life among the Elenchidse and Halictopha- 

 gidae. There must be one to several broods of parasites, according 

 to the number in the host species. It is probable that adult fer- 

 tilized females hibernate in adult hosts. The period between broods 

 must be bridged by the triungulinids in waiting unless an alternation 

 of hosts takes place. Systematic studies do not at present support 

 the theory of alternation of hosts. The epoch of parturition must be 

 rapid. 



