Ecological costs on local adaptation of an insect herbivore imposed by host plants and enemies

Herbivore populations may become adapted to the defenses of their local hosts, but the traits that maximize host exploitation may also carry ecological costs. We investigated the patterns and costs of local adaptation in the pine processionary moth, Thaumetopoea pityocampa, to its host plants, Pinus nigra and P. sylvestris. The two hosts differ in needle toughness, a major feeding impediment for leaf-eating insects. We observed a west-to-east gradient of increasing progeny size in the Italian Alps, matching the pattern in toughness of their respective local host plant. Eastern populations that feed on the native P. nigra with tough needles had larger eggs, and neonate larvae with larger head capsules, than western populations that feed on the native P. sylvestris and the introduced P. nigra with softer foliage. In a reciprocal transfer experiment that involved the eastern-most and the western-most populations of T. pityocampa from this region, and excluded natural enemies, we found evidence for local adaptation to the host plant. Specifically, larvae from the western population only performed well when raised on their local hosts with soft needles, and they suffered near-complete mortality on the tough foliage at the eastern site. In contrast, larvae from the eastern population survived equally well at both sites. Local adaptation involved a trade-off between progeny size and the number of offspring. We hypothesized that an additional cost, imposed by natural enemies, may be associated with increased egg size: we also observed a west-to-east gradient of increased egg parasitism. We tested this hypothesis in a common garden by exposing eggs of both populations to parasitism by two native egg parasitoids, Ooencyrtus pityocampae and Baryscapus servadeii. The eastern population suffered a higher level of parasitoid attack by O. pityocampae than the western population, and performance of hatched adults of both parasitoids was enhanced in large eggs. Thus, increased neonate quality (larger eggs yielding larger larvae) confers an advantage on tough foliage but incurs the ecological cost of increased parasitism, which may constrain further adaptation by this herbivore.     

Mother really knows best: host choice of adult phytophagous insect females reflects a within-host variation in suitability as larval food

Non-random distribution patterns of specialized phytophagous insects on their hosts may depend on intraspecific differences in plant tissue quality, including nutrients and secondary compounds. Secondary compounds are involved in plant resistance, but are also important for the recognition and acceptability of plants as resources by specialized insects. If individuals within a plant species vary in their content of such secondary substances, there may also be qualitative differences between them. In such cases, natural selection will favor insects with the ability to distinguish and prefer the more suitable plants. In Sweden, the leaf beetle Gonioctena linnaeana Schrank (Coleoptera, Chrysomelidae) is highly specialized on one host, the native willow Salix triandra L (Salicaceae). Field observations reveal that some host plants in a population harbor many feeding larvae, causing severe defoliation, whereas neighboring plants may have few or no feeding larvae. Our hypothesis is that the distribution pattern of G. linnaeana larvae in this population results from qualitative differences between individual host plants in combination with the ability of G. linnaeana females to distinguish between plants that are suitable and not suitable for offspring performance. We examine whether larval survival differs depending on diet and whether the content of secondary chemical compounds explains female preference. Based on the higher survival rate of larvae reared on leaves from preferred hosts, we conclude that G. linnaeana females have evolved a behavior that maximizes offspring performance and thus positively affects female fitness. A chemical survey of the plants indicates that luteolin-7-glucoside and an unidentified flavonoid are important for separating the preferred from the non-preferred plants.     

Low maternal host-encounter rate enhances offspring proliferation in a polyembryonic parasitoid

Mothers can epigenetically influence their progeny’s characteristics in response to environmental conditions they experience, thereby increasing offspring adaptation to anticipated future conditions. When resource shortage is anticipated, females are expected to produce larger offspring, as large body size often confers competitive and dispersal advantages. We tested this hypothesis using the polyembryonic parasitoid, Copidosoma koehleri. In this wasp, each egg proliferates into a clone of genetically identical individuals within its moth host, and body size correlates negatively with clone size. We expected females anticipating resource limitation to produce fewer and larger offspring per clone than females that anticipate abundant resources. Encounter rates of parasitoid females with hosts were manipulated to simulate varying levels of resource availability. High-encounter-rate females were introduced to ten hosts successively, while low-encounter-rate females encountered each of ten hosts at 8-h intervals. To control for female age at oviposition, we also introduced females at different ages to a single host. Contrary to our predictions, low-encounter-rate females produced larger offspring clones than high-encounter-rate females, and offspring body size did not differ between treatments. Low-encounter-rate females were shorter-lived than females that encountered hosts successively. Single-oviposition females resembled the high-encounter-rate females in longevity but produced as many offspring per clone as in the low-encounter-rate treatment. Female age, and number of previous host encounters, did not affect offspring clone size. These results suggest that offspring proliferation bears a cost to mothers, thus mothers that repeatedly induce high proliferation in their offspring pay an increased price.     

Host choice and offspring sex allocation in a solitary parasitic wasp

Summary Females of the parasitic wasp Antrocephalus pandens can detect differences in the quality of their hosts (pupae of Corcyra cephalonica, a stored-product moth) and allocate offspring of either sex accordingly. Larger and younger hosts are accepted more often in both dead and live hosts; more female offspring emerge from the perceived better hosts, while more males emerge from the smaller, older ones. These patterns are consistent with a sex allocation strategy by the mother, since females from a given size host tend to be larger than males and larger females produce more eggs. However, when wasps lay their eggs in groups of hosts of different size and age rather than encountering them one at a time, no difference in number or sex ratio of offspring is detected between groups. This result and evidence from the change in offspring sex ratio with female age and with numbers of females foraging on a group of hosts are interpreted and discussed in the context of sex allocation (Charnov 1979) and local mate competition (LMC, Hamilton 1967) theories.     

Offspring sex ratio response to host size in the parasitoid wasp Spalangia endius

The host size model, an adaptive model for maternal manipulation of offspring sex ratio, was examined for the parasitoid wasp Spalangia endius. In a Florida strain, as the model predicts, daughters emerged from larger hosts than sons, but only when mothers received both small and large hosts simultaneously. The pattern appeared to result from the mother's ovipositional choice and not from differential mortality of the sexes during development. If sex ratio manipulation is adaptive in the Florida strain, it appears to be through a benefit to daughters of developing on large hosts rather than through a benefit to sons of developing on small hosts. Both female and male parasitoids were larger when they developed on larger hosts. For females, developing on a larger host (1) increased offspring production, except for the largest hosts, (2) increased longevity, (3) lengthened development, and (4) had no effect on wing loading. For males, development on a larger host had no effect on any measure of male fitness – mating success, longevity, development duration, or wing loading. In contrast, a strain from India showed no difference in the size of hosts from which daughters versus sons emerged, although both female and male parasitoids were larger when they developed on larger hosts. These results together with previous studies of Spalangia reveal no consistent connection between host-size-dependent sex ratio and host-size-dependent parasitoid size among strains of S. endius or among species of Spalangia. Received: 28 October 1998 / Received in revised form: 20 May 1999 / Accepted: 30 May 1999     

Tactics of parasitic American coots: host choice and the pattern of egg dispersion among host nests

Summary I examined the tactics adopted by a conspecific brood parasite, the American coot (Fulica americana), and the degree to which these tactics reflect sources of mortality for parasitic eggs. Only 8% of parasitic eggs produced independent offspring, compared to a 35% success rate for non-parasitic eggs, and most mortality was due to egg-rejection by hosts or the consequences of laying eggs too late in the host's nesting cycle. Parasites usually laid parasitically before initiating their own nests and usually parasitized immediate neighbours. Parasites did not remove host eggs before laying their own egg, and egg disappearance in general was not more common at parasitized nests. I found no evidence for non-random host choice, either on the basis of stage of the host's nesting cycle or the host's brood size. The absence of adaptive host choice is likely a consequence of the fact that, due to host limitation, only a small proportion of parasites had meaningful variation among potential hosts to choose from. The pattern of egg dispersion among host nests by individual parasites appears to be a compromise between constraints imposed by host limitation and the increased success obtained from spreading eggs among nests. Most females laying fewer than five parasitic eggs laid them in a single host nest while females laying five or more eggs normally parasitized two or more hosts. An examination of egg rejection and survival rates showed that parasites would maximize success by laying a single egg per host nest, and the pattern of laying several eggs per host nest is likely a consequence of host limitation. However, no egg that was the fifth laid, or later, parasitic egg in a host nest was ever successful and this probably explains why most females laying five or more eggs parasitized more than one host.     

Adaptive responses to local mate competition by the parasitoid,Telenomus remus

Summary The parasitic wasp, Telenomus remus, lays her eggs in diserete patches of moth eggs, where her offspring develop and mate before dispersal, satisfying conditions for local mate competition (LMC). In the presence of other ovipositing females, wasps lay a higher sex ratio (proportion males), as predicted by LMC theory, and achieve this by a combination of two mechanisms, (1) avoidance of superparasitism and a sequence of sex allocation initially biased towards males and (2) a direct increase in sex ratio in the presence of other wasps, sex ratio increases with the proportion of previously parasitized hosts, as predicted by LMC theory. In both cases, chemical traces left by foraging wasps are indicated as the stimuli causing wasps to increase the proportion of males allocated to hosts.     

Learning and natal host influence host preference,handling time and sex allocation behaviour in a pupal parasitoid

The host choice and sex allocation decisions of a foraging female parasitoid will have an enormous influence on the life-history characteristics of her offspring. The pteromalid Pachycrepoideus vindemiae is a generalist idiobiont pupal parasitoid of many species of cyclorrhaphous Diptera. Wasps reared in Musca domestica were larger, had higher attack rates and greater male mating success than those reared in Drosophila melanogaster. In no-choice situations, na?ve female P. vindemiae took significantly less time to accept hosts conspecific with their natal host. Parasitoids that emerged from M. domestica pupae spent similar amounts of time ovipositing in both D. melanogaster and M. domestica. Those parasitoids that had emerged from D. melanogaster spent significantly longer attacking M. domestica pupae. The host choice behaviour of female P. vindemiae was influenced by an interaction between natal host and experience. Female P. vindemiae reared in M. domestica only showed a preference among hosts when allowed to gain experience attacking M. domestica, preferentially attacking that species. Similarly, female parasitoids reared on D. melanogaster only showed a preference among hosts when allowed to gain experience attacking D. melanogaster, again preferentially attacking that species. Wasp natal host also influenced sex allocation behaviour. While wasps from both hosts oviposited more females in the larger host, M. domestica, wasps that emerged from M. domestica had significantly more male-biased offspring sex ratios. These results indicate the importance of learning and natal host size in determining P. vindemiae attack rates, mating success, host preference and sex allocation behaviour, all critical components of parasitoid fitness. Electronic Publication     

Superparasitism and ovicide in parasitic Hymenoptera: theory and a case study of the ectoparasitoid Bracon hebetor

Summary Among insect parasitoids, superparasitism is said to occur when a second clutch of eggs is laid on a previously parasitized host. Ovicide occurs when a parasitoid destroys a clutch of eggs laid on a host by a previous female. Here, general models are constructed to predict the conditions which favor superparasitism and ovicide. Major predictions for the ovicidal model were that ovicide is more likely to occur if the time necessary to kill eggs is short, if travel times and the proportion of parasitized hosts increases and if the competitive advantage of a first clutch is large. The predictions of the models were tested by examining superparasitism and ovicide in Bracon hebetor (Say), a gregarious, ectoparasitoid of phytisiine moths. Using a wild and eye color mutant of B. hebetor to distinguish first and second clutches, it was found that the competitive advantage of a first clutch over a second clutch increased with the time between ovipositions. Patterns of superparasitism and ovicide in B. hebetor were in qualitative agreement with the major predictions of the model. Most notable, ovicide increased in frequency with a decrease in the overall rate of host encounter and an increase in the proportion of parasitized hosts encountered.     

The predatory mite Phytoseiulus persimilis manipulates imprinting among offspring through egg placement

A crucial question with respect to imprinting is how animals ensure that kin imprint on kin but not on non-kin. Imprinting takes place in a sensitive ontogenetic phase, usually in an early period of life or when offspring are produced, at which time the recipient imprints on the first referents met. In the predatory mite Phytoseiulus persimilis, imprinting among immature individuals happens in the larval stage immediately after hatching. I tested the hypothesis that adult P. persimilis females manipulate offspring imprinting by influencing the likelihood of encounters among recipients and referents via egg placement and egg aggregation. I conducted two experiments, one of which addressed imprinting and cannibalism, and the other addressed egg placement and egg aggregation. The imprinting experiment suggests that larvae imprint on any conspecific individual met in a sensitive ontogenetic phase and later on treat this individual as kin, irrespective of relatedness. After molting to protonymphs, imprinted individuals preferentially cannibalized unfamiliar to familiar larvae. Irrespective of familiarity, kin were cannibalized earlier than non-kin, suggesting the involvement of self-referent phenotype matching. The egg-placement experiment provides evidence that females adjust the aggregation level of their own eggs according to the degree of relatedness to present eggs from a previously ovipositing female. Both experiments in concert suggest that egg placement is a maternal strategy influencing imprinting among immature individuals. Apart from avoiding kin cannibalism, egg placement and imprinting by larvae may have relevance to other behaviors influenced by kin recognition, such as mate choice, prey-patch choice and dispersal.Communicated by L. Simmons     

Recognition of individual-specific marked parasitized hosts by the solitary parasitoid Epidinocarsis lopezi

Summary In parasitoid wasps, self-superparasitism (oviposition into a host already parasitized by the female herself) often contributes less to the reproductive success of the parasitoid than oviposition into a host previously parasitized by a conspecific (conspecific superparasitism). It could therefore often be profitable for parasitoids to avoid self-superparasitism. This requires a mechanism for either (1) the avoidance of previously searched areas and/or (2) the rejection of hosts containing eggs laid by the searching female. We investigated whether the solitary parasitoid Epidinocarsis lopezi is able to avoid self-superparasitism. We show that visits to previously searched patches were shorter than visits to unsearched patches and conclude that E. lopezi females leave a trail odour on patches they have searched. No differences were found between the time on patches previously searched by the wasp itself and on patches visited by conspecifics. However, E. lopezi superparasitizes fewer hosts previously parasitized by itself than hosts parasitized by a conspecific. Thus, they recognize an individual-specific mark in or on the host. We discuss how patch marking and host marking enable E. lopezi to avoid self-superparasitism.     

The response of an insect parasitoid,Ormia ochracea (Tachinidae), to the uncertainty of larval success during infestation

Ormia ochracea is a parasitoid fly which lays its larvae on its hosts, the field crickets Gryllus integer and Gryllus rubens, in two distinct modes: (1) directly on the host and (2) around the host. In the field, 12.7% of male crickets were parasitized and 3.2% were super-parasitized. Despite the disadvantages of parasitizing infested hosts, there was no evidence that O. ochracea avoided superparasitism. This and other experiments suggest that the host assessment ability of O. ochracea is less than that reported for many hymenopteran parasitoids. by manipulating the number of larvae in each cricket, we determined that four to five larvae per host resulted in the largest number of adult flies. However, as larval number per host increased from one to six, pupal size, and hence adult size, declined. In the field, hosts were found with a mean of 1.7 ± 1.0 (SD) larvae per cricket, suggesting that there may be some selection pressure against larger clutch sizes. Nevertheless clutch sizes larger than the host can support were sometimes found in the field. During the first mode of larviposition, gravid flies deposited no more than three larvae directly onto the host. Larvae deposited directly on the host had a high probability of infesting it. During the second mode of larviposition, gravid flies laid a larger number of larvae around the host (6.1 ± 5.2). Larvae that were laid around the host were less likely to infest a cricket than were larvae that were deposited directly onto it. O. ochracea is unique in that its two different modes of larviposition have different probabilities of larval success. Even though the success rate for larvae laid during the second mode of larviposition was low, the possibility of parasitizing more hosts appears to have selected for flies laying more larvae (e.g. increasing clutch size) than is optimal if all the larvae successfully entered a single host.     

Delayed dispersal as a potential route to cooperative breeding in ambrosia beetles

Xyleborini are a species-rich tribe of ambrosia beetles, which are haplodiploid and typically mate among siblings within their natal brood chamber. Several characteristics of this tribe would predict the evolution of higher levels of sociality: high genetic relatedness within galleries due to inbreeding, high costs of dispersal and the potential benefit of cooperation in brood care within the natal gallery (e.g. by fungus gardening, gallery extension, offspring feeding and cleaning). However, information on the social system of these beetles is very limited. We examined the potential for cooperative breeding in Xyleborinus saxeseni by monitoring dispersal in relation to brood size and composition. Results show that adult female offspring delay dispersal despite dispersal opportunities, and apparently some females never disperse. The females’ decision to stay seems to depend on the presence of eggs and dependent siblings. We found no indication that female offspring reproduce in their natal gallery, as colonies with many mature daughters do not contain more eggs than those with few or no daughters. There is a significant positive relationship between the number of females present and the number of dependent siblings (but not eggs), which suggests that cooperative brood care of female offspring raises colony productivity by improving survival rates of immatures. Our results suggest that cooperative breeding is likely to occur in X. saxeseni and possibly other xyleborine species. We argue that a closer look at sociality within this tribe may yield important information on the factors determining the evolution of cooperative breeding and advanced social organization.     

Active protection of unrelated offspring against parasitoids. A byproduct of self defense?

Natural enemies exert selection pressure on their prey. Predators and parasitoids drive their prey into the evolution of novel traits to cope with this stress. When eggs and juveniles are the target of enemies, defense strategies may rely on adults. However, it is not easy to predict whether adults should actively protect unrelated offspring. Females of the golden egg bug (Phyllomorpha laciniata) mainly oviposit on conspecifics of either sex. Females can also lay eggs on their food plant. Eggs placed on plants suffer from a higher mortality caused by natural enemies than eggs carried by bugs. Females never carry their own eggs and males are seldom related to the eggs they carry. We experimentally explored if conspecifics protect the eggs by studying the behavioral interaction between P. laciniata individuals and the specialist egg parasitoid Gryon bolivari. All bugs exhibited active responses against parasitoids regardless of the sex of the bug, the egg load, and their mating status. Most of the responses prevented parasitoids from reaching the eggs, and thus they reduced the risk of egg parasitization. Although responses of bugs were effective to overcome parasitoid attacks, we suggest that egg protection against parasitoids has evolved as a co-opted trait from a general defense of adult bugs against enemies. In this system, egg defense is not an individual's strategy to protect the offspring, but rather a consequence of the eggs being attached to one's body. It may also explain the low parasitization carried eggs suffer in the wild. The results further highlight the idea of conspecifics as an enemy-free space in P. laciniata.     

Sex ratio and host size in a parasitoid wasp

Summary Parasitoid wasps often lay male eggs in small hosts and female eggs in larger hosts. The selective advantage of this strategy can be explained by assuming wasp fitness increases with host size and that this fitness increase is greater in females than in males. I conducted experiments to test a model based on this explanation and found the results generally supported the model with one exception; unlike what the model assumed, these wasps were unable to adjust their offspring sex ratios in each generation to different host size distributions. This finding suggests an alternate view as to how selection might operate in the evolution of parasitoid sex ratios.     

Intraspecific hyperparasitism in a primary hymenopteran parasitoid

In solitary parasitoids, in which only one individual can emerge per host, the adaptive value of conspecific superparasitism is a function of the survival probability of the egg laid by the superparasitizing female. In the few cases which these probabilities are compared, the oldest immature has an advantage over the other individuals. We measured the acceptance rate of parasitized hosts and survival rate of supernumerary larvae in Anaphes victus (Hymenoptera: Mymaridae) in relation to the interval between ovipositions. When this interval was 5–7 days, the first immature was at the prepupa and pupa stage respectively, and female Anaphes victus changed their oviposition behavior markedly. They killed the developing parasitoid of their own species before ovipositing in it. The progeny of these females, which are normally primary parasitoids, developed thereafter as hyperparasitoids. Indeed, in contrast with other species, the survival of the second female's progeny increased with the time interval between ovipositions. This type of facultative intraspecific hyperparasitism is different from autoparasitism in Aphelinidae and has never been mentioned in other parasitoids; it would be adaptive if females of this short-lived species encounter low-quality patches.     

Plant volatiles affect oviposition by codling moths

Summary. Oviposition in wild codling moth females, collected as overwintering larvae from apple, pear and walnut, was stimulated by volatiles from fruit-bearing green branches of these respective hostplants. Analysis of headspace collections showed that eight compounds present in apple, pear and walnut elicited a reliable antennal response in codling moth females: (E)--ocimene, 4,8-dimethyl-1,(E)3,7-nonatriene, (Z)3-hexenyl acetate, nonanal, -caryophyllene, germacrene D, (E,E)--farnesene, and methyl salicylate. Any one of these compounds is found in many other non-host plants, and host recognition in codling moth is thus likely encoded by a blend of volatiles. A large variation in the blend proportion of these compounds released from apple, pear and walnut suggests a considerable plasticity in the female response to host plant odours. Wild females, collected as overwintering larvae in the field, laid significantly fewer eggs in the absence of host plant volatiles. The offspring of these females, however, reared on a semi-artificial diet in the laboratory, laid as many eggs with or without plant volatile stimulus. Tests with individual females showed that this rapid change in oviposition behaviour may be explained by selection for females which oviposit in the absence of odour stimuli, rather than by preimaginal conditioning of insects when rearing them on semi-artificial diet. Oviposition bioassays using laboratory-reared females are therefore not suitable to identify the volatile compounds which stimulate egglaying in wild females.     

How temperature and habitat quality affect parasitoid lifetime reproductive success—A simulation study

Adult parasitoid females lay their eggs in or on host insects. Most species are incapable of de novo lipogenesis as adults, and lipids accumulated during the larval stage are allocated either to egg production or to adult survival. Lipid consumption increases with distance covered by the parasitoids and thus with the distance between available hosts within a habitat. Temperature should affect parasitoid fitness because it changes the constraint imposed by a limited reserve of lipids and because it influences behaviour. Climate change involves both an increase in average temperature and an increased frequency of extreme weather such as heat waves. We investigated how the predicted increase of temperature will affect parasitoid fitness and how this depends on habitat parameters (spatial distribution of hosts and lipid cost of habitat exploitation). We studied optimal behaviour and calculated fitness at different temperatures and in different habitats using a stochastic dynamic programming model for pro-ovigenic parasitoids (which mature all their eggs before becoming adult). We show that an increase in temperature decreases fitness of parasitoids adapted to lower temperatures. This decrease in fitness depends on habitat quality. In field conditions (assuming small costs of intra-patch foraging), the loss of fitness should be larger in habitats with high inter-patch distance and in habitats with a more aggregated distribution of hosts. The foraging behaviour of parasitoids is also affected; at higher temperature we show that intra-patch foraging becomes less efficient, and patch residence times are longer.     

Parasitic wasps orient to green leaf volatiles

Summary Undamaged plants emit low levels of green leaf volatiles (GLVs), while caterpillar-damaged and artificially damaged plants emit relatively higher levels of certain GLVs. Female braconid parasitoids,Microplitis croceipes, oriented to both damaged plants and to individual GLVs in no-choice tests in a wind tunnel, but seldom oriented to undamaged plants. Female ichneumonid parasitoids,Netelia heroica, also oriented to individual GLVs in a wind tunnel. Males of both wasp species failed to orient to the GLVs. These data show that leaf-feeding caterpillars can cause the release of GLVs, and that parasitic wasps can respond to these odors by flying upwind (chemoanemotactic response), which brings the wasps to their caterpillar hosts. This supports the hypothesis that plants communicate with members of the third trophic level,i.e., plants under herbivore attack emit chemical signals that guide natural enemies of herbivores to sites of plant damage. In this interaction, the GLVs serve as tritrophic plant-to-parasitoid synomones. That parasitoids from two different wasp families oriented to GLVs suggests that the response may be widespread among the Hymenoptera.Mention of a commercial or proprietary product does not constitute an endorsement by the U.S. Department of Agriculture     

Fitness consequences of body-size-dependent host species selection in a generalist ectoparasitoid

In insect parasitoids, offspring fitness is strongly influenced by the adult females choice of host, particularly in ectoparasitoids that attack non-growing host stages. We quantified the fitness consequences of size-dependent host species selection in Dirhinus giffardii, a solitary ectoparasitoid of tephritid fruit fly pupae. We first showed a positive correlation between the size of emerged D. giffardii wasps and the size of their host fruit fly species (in order of decreasing size): Bactrocera latifrons, B. cucurbitae, B. dorsalis or Ceratitis capitata. We then manipulated individual wasps to show that the parasitoid preferred to attack the largest (B. latifrons) to the smallest (C. capitata) host species when provided with a choice, and laid a greater proportion of female eggs in B. latifrons than in C. capitata. There were no differences in developmental time or offspring survival between individuals reared from these two host species. Finally, we compared the foraging efficiency of large versus small wasps (reared from B. latifrons vs C. capitata) under two different laboratory conditions: high versus low host habitat quality, given that realized fecundity in parasitoids may be influenced by either egg-limited or time-limited factors. Under both conditions, large wasps parasitized more hosts than did small ones as a consequence of high searching efficiency in the host-poor habitat, and high capacity for adjusting egg maturation in response to host availability in the host-rich habitat. Considering the flexibility of body growth, the apparent lack of cost of achieving large body size in either development or survival, and the strong dependence of realized reproductive success on a females size, we argue that body size may be a key to understanding evolution of host species selection in ectoparasitoids. We also discuss constraints upon the evolution of size-dependent host species selection in parasitoids.Communicated by D. Gwynne