Entomologia Experimentalis et Applicata 89: 249–259, 1998. © 1998 Kluwer Academic Publishers. Printed in the Netherlands. 249 Hyperparasitism by an exotic autoparasitoid: secondary host selection and the window of vulnerability of conspecific and native heterospecific hosts Martha S. Hunter & Suzanne E. Kelly Department of Entomology, 410 Forbes Bldg., University of Arizona, Tucson, AZ 85721, USA Accepted: September 10, 1998 Key words: Encarsia transvena, Eretmocerus eremicus,Trialeurodes vaporariorum, Bemisia tabaci, Aphelinidae, oviposition behavior, host suitability, heteronomous hyperparasitoid Abstract Encarsia transvena is an ‘autoparasitoid’ in the hymenopteran family Aphelinidae. In this species, female eggs are laid in whitefly nymphs. Male eggs are laid externally on immature parasitoids enclosed within the whitefly integu- ment, either their own species, or other primary parasitoids. We explored parasitism by E. transvena of conspecific female immatures and those of a native primary parasitoid, Eretmocerus eremicus, in laboratory experiments. In the first experiment, female E. transvena were offered different combinations of two stages of E. transvena (late larvae – prepupae (ET2), and early pupae (ET3)), and one stage of E. eremicus (prepupae – early pupae (EE2)) in paired choice tests. The results indicated very little parasitism of ET3 relative to the host it was paired with, either EE2 or ET2. However, when EE2 was offered with ET2, there was no statistically significant difference in parasitism. In a no-choice experiment in which oviposition patterns and male progeny development were examined in four stages of both species of wasp, clear differences were observed between the host species. Only one stage of E. transvena (ET2) was parasitized and supported development of male E. transvena to any significant degree. In contrast, in E. eremicus, EE2, EE3 (red-eyed pupae), and EE4 (late pupae) were all parasitized, and male E. transvena emerged from all three stages, although fewer males emerged from EE4. In both species, wasp larvae that were still enclosed within the wet whitefly remains (ET1 and EE1) were parasitized at a very low rate. Lastly, an experiment that determined the length of the later developmental stages of E. transvena and E. eremicus suggested that the duration of the period in which E. transvena is susceptible to parasitism by conspecific females is less than half the period of susceptibility of E. eremicus. These results taken together suggest the potential for interference of E. eremicus by E. transvena, but other factors not examined here may also influence the outcome of interactions in the field. Introduction Trophic relationships have long been used to guide in- troductions of parasitoids for classical biological con- trol. Obligate secondary parasitoids (hyperparasitoids) have been excluded from introductions because of their potential to suppress populations of primary par- asitoids (Smith, 1916; Huffaker et al., 1976; Luck et al., 1981). It is common, however, to introduce sev- eral potential competitors on the same trophic level. The more contentious principle here is that interspe- cific competition between primary parasitoids does not disrupt biological control (Turnbull & Chant, 1961; DeBach, 1966; Huffaker et al., 1976; Force, 1985). The historical record of biological control has sug- gested that even when one parasitoid species displaces or excludes another, the pest population equilibrium is reduced (Luck & Podoler, 1985; Rose & DeBach, 1991–1992) but some theory suggests an increased pest equilibrium may result under some circumstances (Briggs et al., 1993; Murdoch et al., 1996). These principles give little guidance, however, when it comes to autoparasitoids, also known as heteronomous hyperparasitoids, or adelphoparasitoids (Walter, 1983; Williams & Polaszek, 1995). In these species, females develop as primary parasitoids on ho- mopteran hosts (here called primary hosts), and males develop as obligate hyperparasitoids, either on females