Interspecific discrimination and larval competition amongMicroplitis croceipes,Microplitis demolitor,Cotesia kazak (Hym.: Braconidae), andHyposoter didymator (Hym.: Ichneumonidae), parasitoids ofHeliothis virescens (Lep.: Noctuidae)

Interspecific host discrimination by adults, and larval competition among the endoparasitoidsMicroplitis croceipes (Cresson),Microplitis demolitor Wilkinson,Cotesia kazak (Telenga) andHyposoter didymator (Thunberg) were investigated usingHeliothis virescens (F.) as the host. In ovipositional choice tests, the mean number of encounters and ovipositions for unparasitized hosts was not significantly different from the mean number of encounters and ovipositions for parasitized hosts for each treatment combination (P>0.05). Thus, none of the parasitoid species discriminated between host larvae recently parasitized once by a female of another species und unparasitized hosts. However, in all but two cases, females did discriminate between unparasitized hosts and hosts in which an early first instar of the first-attacking species was developing.Cotesia kazak andH. didymator did not discriminate between unparasitized hosts and hosts parasitized by an early first instar ofM. demolitor. Larval competition among these parasitoid species was studied for three time intervals between the first and second species parasitization: 1) second species attack immediately (5–15 sec) after the first; 2) second species attack 24 h after the first; and 3) second species attack 48 h after the first. Time until egg eclosion was shortest forM. demolitor, thenC. kazak, thenM. croceipes, and longest forH. didymator. When the second parasitoid species attacked a host immediately after the first species, the species in which egg eclosion occurred first was the victor more frequently, except whenM. demolitor competed withC. kazak andH. didymator. With a 24 h delay between the first and second species to attack, the older first instar from the first parasitization usually outcompeted the younger first instar from the second attack. A first instar from the second species to attack generally outcompeted the second instar of the first species when the second parasitization had been delayed 48 h. Competiors were eliminated mainly by physical attack, butC. kazak andM. croceipes apparently also killedH. didymator eggs by physiological processes.     

Foraging behaviour of the whitefly parasitoidEncarsia formosa on tomato leaflets

IndividualEncarsia formosa parasitoids were observed continuously until the parasitoids flew away, either on clean tomato leaflets, on leaflets with honeydew, or on leaflets with unparasitized and parasitized whitefly larvae. Encounters with unparasitized and parasitized whitefly larvae, and contact with honeydew arrested the parasitoids on the leaflet. The walking speed increased linearly from 0.179 to 0.529 mm/s between 15 and 25–30°C. The walking activity showed another relationship with temperature: it was below 10% at 15 and 18°C, and increased to about 75% at 20, 25 and 30°C. It was not affected by host encounters or by 1 to 4 ovipositions. The total handling time of hosts was between 1.8–21.8% of the total time on the leaflet. Self-superparasitism was not observed. Conspecific-superparasitism did occur in 14% of the encounters with hosts containing a parasitoid egg, but was not observed anymore when the parasitoid egg had hatched. Experienced parasitoids superparasitized as often as naive females. The foraging behaviour ofE. formosa from landing on a leaf until departure has now been quantified and is discussed.     

Dynamic information and host acceptance by a tephritid fruit fly

Abstract. 1. Female apple maggot (Rhagoletis pomonella Walsh) flies held in field cages usually oviposited in an unparasitized (non-pheromone marked) fruit when it was encountered.
2. Oviposition in a previously parasitized (pheromone marked) fruit depended upon the time since the last oviposition (TSLO) and the percentage of infested fruit encountered during search for oviposition sites.
3. Previous theories of host acceptance suggest that the acceptance or rejection of a host should depend dichotomously on time since last oviposition and the fraction of marked hosts in the last five encounters. The experiments, however, show considerable variability and are thus not consistent with the theory.
4. A new theory for the experiments is introduced. This model involves physiological (egg complement) and informational state variables and leads to intuitive understanding of the experimental results. In particular, the model shows how the plasticity in oviposition site selection may arise from fitness maximizing behaviour. Alternative models are also discussed. All of the models stress the importance of physiological and informational states.     

中红侧沟茧蜂寄生对寄主粘虫血淋巴糖类、脂类和蛋白质含量的影响

为揭示寄生蜂寄生对其寄主的生理调控机制, 室内对中红侧沟茧蜂Microplitis mediator寄生与未被寄生寄主粘虫Mythimna separata幼虫血淋巴中糖类、脂类和蛋白含量变化进行了测定。结果显示: 在滞育与非滞育条件下, 被寄生的粘虫血淋巴中糖原浓度均比未被寄生的粘虫高。滞育条件下寄生后12 d差异显著（P<0.05）, 被寄生粘虫糖原含量为7.93 μg/mL, 未被寄生粘虫糖原含量为4.70 μg/mL; 非滞育条件下寄生后6 d差异显著（P<0.05）, 被寄生粘虫糖原含量为14.35 μg/mL, 未被寄生粘虫糖原含量为5.47 μg/mL。海藻糖含量测定结果显示, 在滞育条件下寄生蜂对被寄生粘虫无明显影响, 而非滞育条件下影响效果差异显著（P<0.05）, 寄生后4 d被寄生粘虫海藻糖含量为46.82 μg/mL, 未被寄生粘虫含量为26.72 μg/mL。在滞育与非滞育两种条件下, 寄生与未被寄生寄主脂类和蛋白含量没有显著性差异。结果说明: 寄生蜂的存在使寄主血淋巴中的糖原含量增高; 非滞育条件是影响被寄生粘虫海藻糖含量变化主要因素; 粘虫对中红侧沟茧蜂的寄生表现相当强的适应性和忍受力。     

The effect of mermithid parasitism on predation of nymphal Baetis bicaudatus (Ephemeroptera) by invertebrates

We investigated how infection by the mermithid nematode Gasteromermis sp. affected predation on its nymphal mayfly host, Baetisbicaudatus, by two invertebrate predators – the stonefly nymphs of Kogotusmodestus and the caddisfly larvae of Rhyacophilahyalinata. Predation trials and behavioral observations were conducted in stream-side, flow-through experimental chambers. When parasitized and unparasitized prey were offered in equal numbers, K. modestus consumed significantly more parasitized than unparasitized nymphs. R. hyalinata consumed equal numbers of both prey types. Behavioral observations of foraging K.␣modestus on parasitized and unparasitized prey suggested that the increased consumption of parasitized nymphs was due to differences in the behavior of infected mayflies in response to the predator. Specifically, parasitized nymphs drifted less often to escape an approaching predator (non-contact encounters) compared to unparasitized nymphs, which increased the number of contact encounters and attacks that occurred between K.␣modestus and parasitized prey. Because all hosts are castrated, these behavioral alterations affect only the fitness of the parasite, which is killed along with its host by invertebrate predation. We present a number of hypotheses to explain why the parasite causes increased predation on its host. These include the large size of the parasite affecting the sensory abilities of the host, the larger energetic costs of escape behavior for parasitized individuals, and natural selection from fish predation against drifting behavior by parasitized individuals. Received: 27 May 1996 / Accepted: 30 September 1996     

Oviposition behaviour ofEphedrus cerasicola [Hym.: Aphidiidae] parasitizing different instars of its aphid host

The parasitoidEphedrus cerasicola Starý oviposited in all 4 nymphal instars and in newly moulted adults ofMyzus persicae (Sulzer). The different host categories were offered with no choice. The duration of an oviposition increased with the age of nymphs, being about 13, 18, 21, 22, and 17 s from 1 st instars to adults, respectively. Observations of number of stabbing attacks prior to oviposition, percent of the encounters not resulting in oviposition, time from first encounter to oviposition, handling time and aphid defensive behaviour also indicated that 1 st instarM. persicae are most easily parasitized. The behaviour ofE. cerasicola in encounters with unparasitized and parasitized hosts, suggested that the parasitoid could discriminate. In encounters with parasitized 1 st to 4th instar aphids,E. cerasicola used only the antennae in 80% of the encounters that resulted in discrimination.     

Adaptive self- and conspecific superparasitism in the solitary parasitoid Leptopilina heterotoma (Hymenoptera: Eucoilidae)

One of the foraging decisions facing parasitoids is whetherto accept (superparasitize) or reject hosts that have alreadybeen parasitized. An important distinction is whether the hosthas been parasitized by the female parasitoid herself or bya conspecific. In solitary parasitoids, the pay-off from anegg laid in the latter host type (conspecific superparasitism)is the probability that the second egg wins the competitionfor the host and results in an offspring. The pay-off from anegg laid in the former type (self-superparasitism) increaseswith an increasing probability that another female will superparasitizethe host in the near future. When this probability equals one,self-superparasitism and conspecific superparasitism have thesame payoff. However, conspecific superparasitism will generallyhave a higher pay-off than self-superparasitism. It will thereforebe beneficial for a female parasitoid to be able to distinguishbetween a host she parasitized and one parasitized by a conspecific.The degree of benefit depends on the probability of conspecificsuperparasitism in the near future. Using an optimal diet model,I show that when a parasitoid encounters a patch containinga mixture of unparasitized and already-parasitized hosts, afemale that can distinguish between the two types of parasitizedhosts gains more offspring than a female without this ability.However, when parasitoids search a patch together with conspecifics,it is adaptive to self-superparasitize, and the pay-off fromthis ability may be negligible. It is therefore predicted thatwhen a female parasitoid searches a partially depleted patchalone, it will reject the hosts parasitized by itself more frequentlythan hosts parasitized by conspecifics. In contrast, femaleparasitoids searching together are predicted to accept hoststhat they parasitized themselves much more often. The resultsshow that the solitary parasitoid Leptopilina heterotoma (Hymenoptera:Eucoilidae) is able to distinguish between hosts that it parasitizedand hosts parasitized by conspecifics. The predictions of themodel are met in a second experiment that shows that L. heterotomaself-superparasitizes when the probability of conspecific superparasitismis high.     

Functional shifts in the use of parasitized hosts by a tephritid fly: the role of host quality

Superparasitism, a phenomenon in which parasitic insects layeggs in already-exploited hosts, provides a useful context inwhich to examine the dynamics of parental investment. This studyexplored conditions under which female Mediterranean fruit flies(Ceratitis capitata) shift from avoiding superparasitism ofhost fruit to preferring it, even placing eggs direcdy intoexisting egg-laying cavities. An a priori hypothesis of costsand benefits was used to predict how use and avoidance of parasitizedfruit would change in -response to changes in fruit size andripeness. We predicted that avoidance would decrease with increasingfruit size, while use would increase with decreasing ripeness.Using a field-cage assay, ripeness was held constant and thesize of host coffee berries manipulated. Avoidance of parasitizedberries was significandy less pronounced on large berries thanon small ones. In a second experiment, size was held constantand ripeness manipulated. On unripe berries, females depositedthe majority of clutches directly into existing egg-laying cavities.On ripe berries, by contrast, the same females deposited mostclutches in previously unparasitized fruit. Parallel patternsin the frequency of female-female contests were observed, supportingthe notion that a fruit's value is determined by an interactionbetween fruit size or ripeness, on one hand, and the prior occurrenceof eggs, on the other. Laboratory assays suggested that useof existing sites had advantages in terms of time savings; femalebehavior thus constitutes a relatively uncommon example of adaptivesuperparasitism in which parasitized hosts are actually preferredover unparasitized ones.[Behav Ecol 7: 235-242 (1996)]     

Superparasitism behavior and host discrimination of Campoletis chlorideae (Ichneumonidae: Hymenoptera) toward Mythimna separata (Noctuidae: Lepidoptera)

The oriental army worm, Mythimna separata Walker (Lepidoptera: Noctuidae), was adopted as a host to keep Campoletis chlorideae Uchida (Hymenoptera: Ichneumonidae) populations in the laboratory, because it can be mass reared. However, wasp cocoon production in this system via single parasitism was not satisfactory. To improve the rearing efficiency of C. chlorideae on M. separata, we studied the effects of superparasitism on the offspring production of C. chlorideae, as well as the host discrimination ability of the adult wasps in the laboratory. The results showed that, compared with single parasitism, both double and quintuple parasitism significantly increased C. chlorideae cocoon production in M. separata without a significant effect on emergence, weight, or sex ratio of adult wasps at the P < 0.05 level. Host selection experiments suggested that C. chlorideae preferred to lay eggs on unparasitized hosts rather than freshly parasitized hosts but could distinguish neither the 24-h postparasitization hosts from the unparasitized hosts nor the freshly self-parasitized hosts from the conspecifically parasitized hosts. No-choice behavior tests indicated that C. chlorideae took significantly longer time to accept the freshly parasitized hosts containing more than two eggs compared with the unparasitized hosts. In the end, superparasitism behavior and its application in the mass rearing of this endoparasitoid are discussed.     

The parasitoidOoencyrtus nezarae (Hymenoptera: Encyrtidae) prefers hosts parasitized by conspecifics over unparasitized hosts

Summary Two laboratory experiments were conducted to examine the ovipositional preferences of the egg parasitoidOoencyrtus nezarae Ishii (Hymenoptera: Encyrtidae) for parasitized and unparasitizedMegacopta punctatissimum Montandon (Hemiptera: Plataspidae). Females that had never oviposited or that had not oviposite for 3 days preferred recently parasitized hosts more than unparasitized hosts. The preference for recently parasitized hosts appeared to be mediated by the punctures in already parasitized hosts made by the ovipositor of the first female. Survival of the parasitoid progeny was lower in recently parasitized hosts than in unparasitized hosts. However, handling time of parasitized hosts was extremely short relative to that of unparasitized hosts, because the superparasitizing female could use the punctures made by the previous females. It is concluded that the females preferred the parasitized hosts over unparasitized hosts because the benefit of saving time and energy for drilling was more than the cost of progeny survival.     

The Ability of Coptera haywardi (Ogloblin) (Hymenoptera: Diapriidae) to Locate and Attack the Pupae of the Mediterranean Fruit Fly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae), under Seminatural Conditions

In Latin America, the diapriid Coptera haywardi (Ogloblin) attacks the pupae of tephritid fruit flies. Anastrepha spp. are among its natural hosts, but in the laboratory it also develops in the exotic Mediterranean fruit fly, Ceratitis capitata (Wiedemann). Field cage tests demonstrated that C. haywardi could locate and parasitize Mediterranean fruit fly pupae under seminatural conditions as found in a Guatemalan coffee plantation. A mean of 18.3% of the pupae buried artificially at depths of 5 mm were parasitized by C. haywardi, while those buried at 15 mm suffered 3.2% parasitism. In a laboratory experiment, larvae that buried themselves to pupate were not significantly more likely to be parasitized than artificially buried pupae, although they may have left a physical or chemical trail that betrayed their presence. Thus, the artificial burial of pupae is unlikely to grossly underestimate C. haywardi efficacy in the field. Another field cage test found that mortality levels due to unsuccessful parasitoid attacks were similar to those resulting from successful parasitism. Thus, the actual effect of a mass-release might be considerably greater than that suggested from parasitism data alone. The results are considered sufficiently positive to encourage further testing of C. haywardi as a biological control agent of the Mediterranean fruit fly.     

Parasitoid diets: Does superparasitism pay?

The ability of insect parasitoids to discriminate between parasitized and unparasitized hosts is well documented. Despite this, hosts that have been parasitized more than once are frequently found, an occurrence known as superparasitism. Since superparasitism results in interlarval competition, it was generally assumed to be caused by discriminatory 'mistakes'. Recently, theoretical studies have suggested that under certain circumstances superparasitism can be optimal. Superparasitism can thus be viewed as an active foraging strategy rather than as a passive process. Because parasitoids show a direct link between foraging success and reproductive output, they afford important opportunities to test evolutionary models of behaviour.     

The influence of host suitability on the range of grasshopper species utilized by Blaesoxipha atlanis (Diptera: Sarcophagidae) in the field

Blaesoxipha atlanis (Aldrich) is a common parasitic fly of agriculturally important grasshoppers in Canada. The suitability of Camnula pellucida (Scudder), Melanoplus bivittatus (Say), Melanoplus packardii Scudder, and Melanoplus sanguinipes (Fabricius) as hosts was studied in the laboratory. Grasshoppers were singly-parasitized or left unparasitized and reared for 9 days. Melanoplus bivittatus and M. packardii did not support parasite development, i.e. were non-permissive hosts. In both species, parasite larvae were melanized and encapsulated, but development proceeded further in M. packardii. Melanoplus sanguinipes and C. pellucida were permissive host species with, respectively, 70% and 35% of the implanted larvae emerging from their hosts of which 86% and 50% developed into adults. Parasite development time was longer in C. pellucida. Adult B. atlanis dry mass varied with host species and host mass at parasitism, but not with host sex. Parasites developing in M. sanguinipes were larger in terms of dry mass than counterparts developing in C. pellucida. In permissive species, unparasitized grasshoppers gained in body mass while parasitized insects lost mass during the 9-day observation period. In non-permissive species, all insects gained in body mass, but parasitized females gained less mass than unparasitized conspecifics. All unparasitized grasshoppers survived while 75-95% of permissive and 30-40% of non-permissive hosts died. Variation in the intensity of field parasitism among grasshopper species may be explained, at least in part, by qualitative differences in suitability between potential host species. Novel pest management strategies emphasize preservation of a small proportion of the pest population for natural enemies. Consideration of the outcome of specific host-parasite interactions should improve the understanding of grasshopper population dynamics and increase the predictive value of models that assess potential crop losses.     

Host location by the fruit fly parasitoid Diachasmimorpha krausii: role of fruit fly species,life stage and host plant

• 1 Diachasmimorpha krausii is a braconid parasitoid of larval tephritid fruit flies, which feed cryptically within host fruit. At the ovipositor probing stage, the wasp cannot discriminate between hosts that are physiologically suitable or unsuitable for offspring development and must use other cues to locate suitable hosts.
• 2 To identify the cues used by the parasitoid to find suitable hosts, we offered, to free flying wasps, different combinations of three fruit fly species (Bactrocera tryoni, Bactrocera cacuminata, Bactrocera cucumis), different life stages of those flies (adults and larvae) and different host plants (Solanum lycopersicon, Solanum mauritianum, Cucurbita pepo). In the laboratory, the wasp will readily oviposit into larvae of all three flies but successfully develops only in B. tryoni. Bactrocera tryoni commonly infests S. lycopersicon (tomato), rarely S. mauritianum (wild tobacco) but never C. pepo (zucchini). The latter two plant species are common hosts for B. cacuminata and B. cucumis, respectively.
• 3 The parasitoid showed little or no response to uninfested plants of any of the test species. The presence of adult B. tryoni, however, increased parasitoid residency time on uninfested tomato.
• 4 When the three fruit types were all infested with larvae, parasitoid response was strongest to tomato, regardless of whether the larvae were physiologically suitable or unsuitable for offspring development. By contrast, zucchini was rarely visited by the wasp, even when infested with B. tryoni larvae.
• 5 Wild tobacco was infrequently visited when infested with B. cacuminata larvae but was more frequently visited, with greater parasitoid residency time and probing, when adult flies (either B. cacuminata or B. tryoni) were also present.
• 6 We conclude that herbivore‐induced, nonspecific host fruit wound volatiles were the major cue used by foraging D. krausii. Although positive orientation to infested host plants is well known from previous studies on opiine braconids, the failure of the wasp to orientate to some plants even when infested with physiologically suitable larvae, and the secondary role played by adult fruit flies in wasp host searching, are newly‐identified mechanisms that may aid parasitoid host location in environments where both physiologically suitable and unsuitable hosts occur.

     

Choosing suitable hosts: common cuckoos Cuculus canorus parasitize great reed warblers Acrocephalus arundinaceus of high quality

We investigated the hypothesis that the common cuckoo Cuculus canorus selects host pairs of good phenotypic quality. As there is some evidence that cuckoos may select hosts within a population non-randomly based on external cues reflecting their foster abilities, we predicted that great reed warbler Acrocephalus arundinaceus pairs parasitized by the cuckoo would exhibit higher quality than unparasitized ones. To test this assumption, we evaluated two different parameters indicating host quality: body condition and characteristics of host eggs. We found that parasitized females showed significantly better body condition than unparasitized ones, and the model showed that the probability of being parasitized by the cuckoos increased with increasing body condition. Moreover, the likelihood of being parasitized by a cuckoo within the great reed warbler population increased with decreasing colour variability within clutches: parasitized females allocated costly blue pigments to eggshells more equally compared with unparasitized ones. Our study revealed that cuckoos parasitize great reed warbler females of higher quality, as reflected in host body condition and egg colour characteristics. In highly mimetic systems, cuckoos may choose to parasitize hosts with eggs displaying low intraclutch variation, both because this leads to reduced rejection and because these hosts are of high quality.     

Competitive Superiority of Early Acting Species: A Case Study of Opiine Fruit Fly Parasitoids

Koinobiont parasitoids that attack an early host stage may have an advantage in suppressing competing parasitoids that attack later stages of the same host. We examined the competitive interaction between the two most important parasitoids of tephritid fruit flies in Hawaii, Fopius arisanus (Sonan), and Diachasmimorpha longicaudata (Ashmead) (Hymenoptera: Braconidae). The former species attacks host eggs while the latter attacks host larvae, and both species emerge as adults from the host puparia. F. arisanus physiologically suppressed egg development of D. longicaudata. Over 90% of D. longicaudata eggs died in the presence of F. arisanus larvae within host larvae of either the Mediterranean fruit fly, Ceratitis capitata (Wiedemann) or the oriental fruit fly, Bactrocera dorsalis (Hendel). D. longicaudata appeared not to discriminate against hosts previously parasitized by F. arisanus. The mechanism that F. arisanus uses to eliminate D. longicaudata is similar to that which it employs to eliminate five other larval fruit fly parasitoids so far reported in Hawaii. This suggests that there is a broad competitive superiority of the early acting species in fruit fly parasitoids. We discuss the implication of this in relation to future biological control introductions against tephritid fruit flies.     

Sex-ratios of the wasp Nasonia vitripennis from self-versus conspecifically-parasitized hosts: local mate competition versus host quality models

Sex ratio patterns in the parasitoid wasp Nasonia vitripennis are frequently cited in support of a major group of evolutionary sex ratio models referred to as local mate competition (LMC) models. It has been shown repeatedly that, as predicted by LMC models, females generally oviposit a greater proportion of sons in previously parasitized hosts than in unparasitized hosts. However, this sex ratio pattern is also a prediction of another group of sex ratio models, the host quality models. Here I test a prediction of LMC models that is not also a prediction of host quality models: a female should produce a greater proportion of sons when she parasitizes a host previously parasitized by a conspecific female than when she parasitizes a host previously parasitized by herself. Females made this predicted distinction between self- and conspecifically-parasitized hosts under some conditions. There was no evidence that a female recognizes a self-parasitized host when her exposure to the host is interrupted by exposure to an unparasitized host, or that a female can distinguish between hosts parasitized by sisters versus nonsisters.     

Oxygen consumption in mosquito larvae parasitized by Romanomermis culicivorax (Nematoda)

The effect of parasitism by Romanomermis culicivorax on the oxygen consumption of three species of mosquito larvae was studied by manometric methods. The Q02 of Culex pipiens, both parasitized and unparasitized, decreased as the host larvae developed from the second to fourth instar. In the early stages there was no significant difference in Q02 in parasitized and unparasitized hosts but the Q02 increased significantly in parasitized fourth instar larvae. The Q02 of parasitized and unparasitized fourth instars of Aedes taeniorhynchus, Anopheles quadrimaculatus and C. pipiens were compared. The Q02 of all parasitized hosts was significantly increased and the Q02 of the two culicine species displayed a dramatic increase in oxygen consumption. The oxygen consumption of the parasitic mermithid larvae was estimated from that obtained earlier for the post-parasitic larvae but did not account for more than 9-36% of the increased Q02 of the parasitized hosts. On this basis it seems probable that the increased metabolic rate as reflected by oxygen consumption of the parasitized mosquito larve was principally a manifestation of the physiological and nutritional stress induced by the mermithid nematodes.     

Two different life-history strategies determine the competitive outcome between Dirhinus giffardii (Chalcididae) and Pachycrepoideus vindemmiae (Pteromalidae), ectoparasitoids of cyclorrhaphous Diptera

Dirhinus giffardii Silvestri and Pachycrepoideus vindemmiae Rondani are solitary parasitoids attacking puparia of many cyclorrhaphous flies. They are not typical ectoparasitoids, as they feed on host pupae within puparia that develop from the exoskeleton of host larvae. Dirhinus giffardii did not kill its host until the parasitoid egg developed into a larva, while P. vindemmiae permanently paralysed its host at the time of oviposition. As a result, ovipositing into a young host puparium (< 1 day old) in which the host pupa has not yet fully formed resulted in complete death of offspring in P. vindemmiae, but D. giffardii, although suffering higher mortality than in older host puparia, still showed a level of successful development. In a choice experiment, both parasitoids preferred to attack 2- to 3-day-old puparia in which the host pupae had fully formed, rather than 1-day-old host puparia. Pachycrepoideus vindemmiae always prevailed in competition because it injected venom that not only paralysed the host, but also caused the death of D. giffardii larvae in multi-parasitized hosts. Dirhinus giffardii preferred to attack unparasitized hosts rather than hosts previously parasitized by P. vindemmiae, while P. vindemmiae did not show a preference between unparasitized hosts and hosts previously parasitized by D. giffardii.     

Inter- and intra-specific host discrimination in gregarious and solitary endoparasitoid wasps

In nature, most species of Lepidoptera are attacked by parasitoids, and some species may be hosts for several parasitoid species. When hosts are parasitized by more than one female of the same species (=superparasitism) or females of different species (=multiparasitism), then intrinsic competition occurs for control of host resources. To reduce competition, some parasitoids are able to recognize the difference between parasitized and unparasitized hosts. Inter- and intra-specific host discrimination were investigated in the two sympatric species, the gregarious Cotesia kariyai (Watanabe) and solitary Meteorus pulchricornis (Wesmael), endoparasitoids of the Oriental armyworm Mythimna separata (Walker). To measure host discrimination, choice experiments were conducted in which females of both species foraged and chose between healthy host larvae and hosts initially parasitized by either C. kariyai or M. pulchricornis. An olfactory test was also performed to examine the discrimination behavior of the two parasitoids. Our results showed that, in oviposition choice tests, both braconid female wasps were able to discriminate between unparasitized hosts and from four to seven day-old hosts previously attacked by conspecific and heterospecific wasps. On the other hand, superparasitism and multiparasitism occurred even in host larvae that were parasitized two days earlier. However, once the immature parasitoids hosts are at larval stage (1st and 2nd instar), super- and multiparasitism were avoided in the two-choice test, but the latter often occurred in the multiple-choice experiment. Host discrimination abilities may have been based on plant volatile signals incurred from damaged plants and internal mechanisms from four to seven post-parasitized hosts.