The results from our behavioural observation experiments clearly indicate that M. basicornis consumes parasitized prey eggs equally often as unparasitized prey eggs when the developing parasitoid in the prey egg is still young. However, the predator shows strongly reduced predation rates when parasitized T. absoluta eggs are older and the T. pretiosum larvae start to pupate inside the host egg. Apparently, rejection of older parasitized prey eggs takes place before contact and not because they could not be penetrated by the predator’s stylets or were judged to be no longer suitable for consumption. Our results of the Y-tube olfactometer tests indicate a role of volatile cues specifically emitted by leaflets infested with older parasitized eggs that could repel the predators to avoid contacting prey eggs with pupae of the parasitoid.
Bueno et al.20 show that M. basicornis consumes T. absoluta eggs recently parasitized by T. pretiosum in equal numbers as unparasitized eggs, but hardly attacks eggs that contain pupal stages of the parasitoid. However, this study did not reveal if the predator rejected these old parasitized eggs or what prevented them from eating these eggs. The currently held opinion about prey searching and evaluation behaviour by mirid predators, though backed by very limited evidence, is that they do not search by vision or smell, but encounter prey randomly10. In this view, old parasitized eggs are rejected after contact because they can no longer be penetrated by the rostrum of the predators. Our observations of the behaviour of M. basicornis revealed a much lower number of contacts with old parasitized eggs than expected from random search behaviour. Apparently, rejection of these old parasitized eggs takes place before they are encountered. However, if encountered, they were as easily penetrated as unparasitized eggs. Thus, prey searching and penetration of old eggs by M. basicornis appears to differ from that of the mirid species referred to in Wheeler10. The olfactometer tests show that volatile cues specifically emitted by tomato leaflets infested with old parasitized eggs may repel the predators to avoid contacting less suitable prey.
Other Heteroptera, including several mirid species, also reject older parasitized eggs, but generally do not distinguish between unparasitized eggs and eggs containing parasitoids early in their development21. Macrolophus pygmaeus Rambur (Hemiptera: Miridae) preferentially preys on unparasitized or recently (< 4 days exposed to parasitoids) T. absoluta eggs parasitized by T. achaeae when the eggs are still yellow, but hardly preys on old, black parasitized eggs in laboratory experiments19. Also, in laboratory experiments with the mirid Nesidiocoris tenuis (Reuter) (Hemiptera: Miridae) and the egg parasitoid T. achaeae, N. tenuis consumed significantly more unparasitized eggs than parasitized eggs, and significantly more parasitized eggs younger than 4-day old than eggs parasitized more than 4-days ago17. Eggs of many Lepidoptera parasitized by Trichogramma spp. become dark due to the deposition of melanin to the inner surface of the host egg chorion at the end of the larval stages and the start of prepupa formation of the parasitoid22–26. In general, melanin protects the insect egg against desiccation, UV light and natural enemies25,27. The detailed description of the development of T. pretiosum in host eggs of E. kuehniella at 25oC – the same temperature we used - shows that the egg-larval stage of the parasitoid takes on average 2.9 days, the prepupal stage lasts 1.4 days and the pupal stage is about 6.1 days long28. No detailed data are available for development of T. pretiosum in T. absoluta, but the total immature development time at 25oC of 10.3 days in T. absoluta29 is very similar to that of 10.4 days in E. kuehniella. So, in our experiments, M. basicornis was exposed to T. absoluta eggs with the parasitoid in the egg-larval stage (1- and 2-day old parasitized prey eggs) and to the early and late pupal stages of the parasitoid in the prey egg (5- and 9-day old parasitized eggs).
In order to be able to draw conclusions about preference for one category of prey over another, prey searching and selection behaviour should be known, e.g. does prey selection takes place before arrival on a host plant, after landing on the plant or only after contact with the prey? Based on the present opinion that mirids search unsystematically, we initially supposed that the lower consumption rates of 5- and 9-day old parasitized eggs were the result of difficulties to penetrate the melanized chorion of the prey egg, and/or the consequence of rejection of these eggs for consumption. The results of the behavioural and olfactometer tests show that M. basicornis avoids to contact old parasitized eggs. However, in the few cases that they do contact an old parasitized egg, they will penetrate it with the same probability as unparasitized eggs. Apparently, melanin deposition and sclerotization by Trichogramma25 does not prevent M. basicornis from penetrating these eggs. Thus, M. basicornis does not search unsystematically and does not decide to reject a certain type of egg only after having made physical contact. The results of olfactometer experiments show that volatiles – in this case a synomone - play a role in prey selection, because the predators prefer tomato leaflets with unparasitized eggs over leaflets with 5-day old parasitized eggs.
Numerous studies have shown that herbivore insect oviposition induces plant volatiles (OIPVs) attract egg and larval parasitoids and repel ovipositing herbivores (reviewed by 27,30,31,32) Lepidopteran oviposition, including that by T. absoluta, does not cause obvious damage to plants. Nevertheless, egg deposition of several lepidopteran species induces quantitative changes in the plant volatile blends33. The finding that M. basicornis prefers volatiles of leaflets with unparasitized or 1-day old parasitized eggs over volatiles of 5-day old parasitized eggs indicates that the predator uses volatile information produced by the combination ‘old parasitized eggs-tomato leaflet’. Whether the information derives from volatiles emitted by the plant and/or volatiles from the T. absoluta eggs needs further investigation. If the volatiles resulting in repellence of the predators are derived purely from the developing parasitoid inside the egg is another intriguing quesion.
As far as we know, it was hitherto unknown that eggs of T. absoluta parasitized by T. pretiosum are rejected by a mirid predator before having made physical contact. A recent study indicates that OIPV blends change when eggs are parasitized: volatiles of rice plants infested with eggs of the brown plant hopper (Nilaparvata lugens) and parasitized by Anagrus nilaparvatae Pang and Wang (Hymenoptera: Mymaridae) were less attractive to the conspecific parasitoids when compared to volatiles from plants with unparasitized eggs34. Plants infested with parasitized eggs showed increased levels of some volatile compounds, such as linalool or methyl salicylate.
Whether predatory insects can make use of OIPVs for prey location has hardly been shown yet. However, numerous studies show that predators they use of herbivore-induced plant volatiles (HIPVs) for prey location35,36. For two European and three Neotropical mirid predators, including M. basicornis, we previously shown that they use volatile cues in their prey finding process12,13. Yet, the Neotropical mirid predators did not discriminate between volatiles of tomato plants infested with eggs of T. absoluta and volatiles of clean tomato plants13. Nevertheless, oviposition on tomato plants by T. absoluta triggered emission of OIPVs attracting Trichogramma wasps37. We argue that the lack of attraction of M. basicornis and several other mirid predators to T. absoluta egg-infested plants may be due to genotypic differences in volatile emission between tomato plant cultivars. A next step would be to analyse the volatiles that are emitted by tomato leaflets infested with unparasitized and parasitzed T. absoluta eggs.
Concluding remarks:
1. The predator M. basicornis can penetrate old parasitized eggs on the rare occasion when such eggs are encountered, and they are accepted for consumption at the same rate as unparasitized eggs.
2. The predator uses volatile information emitted by old parasitized eggs on tomato leaflets to prevent encounters with old parasitized eggs.
3. Due to IGP, young parasitoid eggs and larvae are killed by the predator when both natural enemies are released at the same time. In order to strongly reduce IGP, predators should be released a week after introduction of the parasitoids.
4. The current belief that mirids search unsystematically, discover and reject prey only after having physically encountered them has to be modified for M. basicornis, as they do not search randomly and reject old parasitized eggs before contacting them