Intraguild predation of Orius niger (Hemiptera: Anthocoridae) on Trichogramma evanescens (Hymenoptera: Trichogrammatidae)

Intraguild predation of a generalist predator, Orius niger Wolff (Hemiptera: Anthocoridae) on Trichogramma evanescens Westwood (Hymenoptera: Trichogrammatidae), was determined in choice and no-choice experiments using a factitious host, Ephestia kuehniella Zeller (Lepidoptera: Pyralidae), under laboratory conditions. Choice and no-choice experiments were conducted in order to assess the level of intraguild predation of O. niger on E. kuehniella eggs parasitized by T. evanescens. In no-choice experiments, approximately 50 sterile (1) non-parasitized, (2) 3-day-old parasitized, or (3) 6-day-old parasitized E. kuehniella eggs were offered to 24-h-old females of O. niger in glass tubes. In choice experiments approximately 25 eggs of two of the three groups mentioned above were offered to 24-h-old O. niger females. In both choice and no-choice experiments, O. niger consumed more non-parasitized eggs of E. kuehniella. However, intraguild predation occurred, especially of 3-day-old parasitoids, but very few 6-day-old parasitized eggs were consumed. The preference index was nearly 1 indicating O. niger preferred mainly non-parasitized E. kuehniella eggs. A lower level of intraguild predation is expected under fi eld conditions but needs to be investigated using further experiments.


INTRODUCTION
Interspecifi c interactions in insects could be classifi ed as either (1) competition, (2) predation/parasitism, (3) mutualism, (4) commensalism or (5) amensalism (Polis et al., 1989).Intraguild predation is a combination of competition and predation/parasitism and is defi ned as the killing or eating of species that use similar limiting resources (prey) (Polis et al., 1989).Intraguild predation is common in nature when species compete for a shared (extraguild) prey in the same ecosystem (Polis et al., 1989;Holt & Polis, 1997).This situation affects the distribution, abundance and evolution of many species in natural and controlled ecosystems (Polis et al., 1989).The use of more than one natural enemy in biological control programmes can lead to direct and indirect interactions between intraguild predators (IGP) or/and competition (Janssen et al., 1999).These interactions may limit the effi ciency of biological control agents against pest species (Yano, 2005).
In the order Lepidoptera there are a large number of economically important species of pests such as the Cotton bollworm (Helicoverpa armigera Hübner, Lepidoptera: Noctuidae) and Tomato pinworm (Tuta absoluta Meyrick, Lepidoptera: Gelechidae), which occur in both fi eld and protected vegetable production areas.Several of the biological control strategies against lepidopteran species of

Experimental design
No-choice and choice experiments were carried out in order to determine if O. niger fed on Ephestia kuehniella eggs that were parasitized by T. evanescens at different stages of development; (i) non-parasitized eggs, (ii) 3-day-old parasitized eggs (yellow) and (iii) 6-day-old parasitized eggs (black) by T. evanescens.
Approximately 50 one-day-old sterilized eggs of E. kuehniella were glued on blue cards (3 × 1 cm), which when they were 3 and 6 days old, respectively, were exposed for 24 h to attack by T. evanesces, which provided the material for the above experiments (ii, iii).
Females of O. niger (< 24-h-old) that were starved for 24 h were used in each experiment.

No-choice experiments
Only one type of host eggs was provided to each female of O. niger in a glass tube (8 × 1.5 cm) plugged with cotton for 24 h.The number of E. kuehniella eggs consumed by O. niger were counted under a stereo binocular microscope (× 40 magnifi cations) (Euromex Nexius Zoom).A randomized design was used in these experiments, with one factor (at three levels) and at least 12 replications per treatment.All experiments were conducted in an incubator at 25°C, 60% relative humidity under a 16L : 8D photoperiod.
Each combination of eggs were presented to a single female of O. niger for 24 h in a glass tube (8 × 1.5 cm) plugged with cotton.The level of consumption and preferences of O. niger for eggs of E. kuehniella that were parasitized by T. evanescens were determined by examining the eggs under a stereo binocular microscope.A randomized design was used in these experiments, with one factor (at three levels) and at least 12 replications per treatment.All experiments were conducted in a climatic chamber at 25°C, 60% relative humidity under a 16L : 8D photoperiod.

Statistical analyses
In the no-choice experiments, the number of eggs consumed in each group was compared using Duncan's test at a P < 0.05 signifi cance level after fi rst subjecting them to an analysis of variance (ANOVA).The paired t-test was used in binary comparisons in the choice experiments with a confi dence level P < 0.05.In the choice experiments, total numbers of eggs consumed in each combination were also compared using Duncan's test at a P < 0.05 signifi cance level after fi rst subjecting them to an ANOVA.We used Manly's preference index (Manly, 1974) for each egg combination, i.e., non-parasitized and 3-day-old parasitized eggs where β 1 is the preference for egg type 1, A 1 and A 2 are the numbers of eggs of types 1 and 2 offered.And e 1 and e 2 are the numbers of eggs remaining at the end of the experiment.The preference index (β 1 ) can be between 0 and 1. β-values higher than 0.5 indicate a preference for egg type 1.IBM SPSS 21 (SPSS, 2012) statistical package software was used for paired t-tests and The Mediterranean fl our moth, Ephestia kuehniella Zeller (Lepidoptera: Pyrallidae) is important since its eggs and larvae are commonly used for mass rearing of many natural enemies (such as Orius spp., Macrolophus spp., Trichogramma spp.), which are released in many biological control programmes against many species of pests.The members of the Genus Trichogramma Westwood (Hymenoptera: Trichogrammatidae) are well-known egg parasitoids of many insect pests, which mainly belong to the order Lepidoptera (Smith, 1996;Oztemiz, 2007).Trichogammatid egg parasitoids are reared on sterilized eggs of the factitious host, E. kuehniella in the laboratory or in mass rearing conditions.
In this study, we planned a series of experiments to determine the intraguild interactions between Orius and Trichogrammatids using Ephestia kuehniella eggs as the laboratory target insect.The aim of the present work was to determine the feeding behaviour of O. niger when presented with eggs of E. kuehniella that are either parasitized or not parasitized by T. evanescens in no-choice and choice experiments.

Ephestia kuehniella rearing
The colony of E. kuehniella came from the Biological Control Research Institute at Adana, Turkey.The colony was reared in the Entomology Laboratory at the University of Çukurova, Faculty of Agriculture, Department of Plant Protection, Adana, Turkey for two years.E. kuehniella was reared on a wheat fl our : corn fl our (1 : 1) mixture, in climatic rooms at 25°C, 60% relative humidity in the dark.Thousand eggs of E. kuehniella were added to each 100 g food material.

Orius niger rearing
The predatory bug, Orius niger was collected from cotton fl owers at Adana, Turkey.The stock culture was reared on sterilized eggs of Ephestia kuehniella Zeller (Lepidoptera: Pyralidae) in the laboratory for six months.Orius individuals were reared in incubators at 25°C, 60% relative humidity under a 16L : 8D photoperiod.Nylon plastic cups (0.5 l capacity) with two ventilation holes covered with fi ne mesh muslin were used as rearing units.Fresh bean pods were used as an oviposition substrate and E. kuehniella sterilized eggs glued onto blue cards were provided as food for O. niger.Adults were kept in other rearing cages to prevent cannibalism.Bean pods with Orius eggs were removed from the adult units three times per week and were placed in other cups for pre-imaginal rearing.

Trichogramma evanescens rearing
The colony of T. evanescens was supplied by the Biological Control Research Institute at Adana, Turkey and reared in the laboratory.T. evanescens was reared on one-day-old E. kuehniella eggs, which were previously sterilized using ultraviolet light, in climatic chambers at 25°C, 70% relative humidity under a 14L : 10L photoperiod.The sterilized eggs of E. kuehniella were glued with 10% Arabic gum to paper strips (1 × 8 cm).The paper strips were placed in glass tubes (2.5 × 16 cm) with 50% honeywater solution that was provided in fi lm form as food for the T. evanescens.Parasitized eggs of E. kuehniella were removed from the glass tubes every 8 days, before the emergence of parasitoids and placed and reared in other tubes.

Choice experiment
In choice experiments, two combinations of types of eggs of E. kuehniella were offered at the same time to females of O. niger.The predator preferred non-parasitized eggs (t = 6.021, df = 22, p < 0.0001) and 3-day-old parasitized eggs (t = 7.070, df = 22, p < 0.0001) to 6-day old parasitized eggs.When non-parasitized eggs and 3-day-old parasitized eggs were provided at the same time, O. niger signifi cantly preferred the non-parasitized eggs (t = 4.655, df = 22, p < 0.0001; Table 1).The preference indices varied between 0.95-0.98 when non-parasitized eggs were provided.The 3-day-old parasitized eggs were preferred to 6-day-old parasitized eggs, with a preference index of 0.99 ± 0.01 (Table 2).
The total numbers of eggs consumed was statistically highest in tubes with non-parasitized eggs and 3-day-old parasitized eggs (yellow) (F (3,40) = 5.097, p < 0.004; Fig. 2).In the absence of non-parasitized eggs fewer eggs were consumed.

DISCUSSION
Like other generalist predators, species of Orius may occur at the same time as species of Trichogramma in agroecosystems and often share one or more species of prey/ host in common (Kayapınar, 1991;Romeis & Shanower, 1996;Tillman et al., 2009;Zappala et al., 2013).Results of this study reveal the occurrence of intraguild predation by O. niger on eggs of E. kuehniella, which were previously parasitized by T. evanescens.On the other hand, based on the results of both the choice and no-choice experiments, it is obvious that O. niger avoids consuming black parasitized eggs of E. kuehniella.
Previous studies mostly show a similar pattern.It is reported in many studies that anthocorid, pentatomid and mirid predators consume few parasitized eggs that contain a pupa of the Trichogramma parasitoid (Brower & Press, 1988;Oliveira et al., 2004;Chailleux et al., 2013;Cabello et al., 2015).It is reported that there is little intraguild predation by the hemipteran predators Nesidiocoris tenuis (Reuter) on Trichogramma achaeae Nagaraja & Nagarkatti (Cabello et al., 2015); Macrolophus pygmaeus Rambur on the same species of parasitoid (Chailleux et al., 2013) and Podisus maculiventris (Hemiptera: Pentatomidae) on Trichogramma brassicae Bezdenko (Oliviera et al., 2004), especially of black eggs.During the pupation of trichogrammatids, melanin is deposited within the host's egg, which darkens the colour of the egg.This material increases the mechanical resistance of the egg shell, thus Table 1.Mean numbers (± SEM) of non-parasitized (0), 3-day-old parasitized (3) and 6-day-old parasitized (6) eggs of E. kuehniella consumed in 24 h by a female O. niger in choice experiments (Means followed by "*" are signifi cantly different based on paired t tests at a signifi cance level of 0.05).
Another result of this study, is the signifi cant difference in the number of yellow parasitized and non-parasitized eggs consumed, with a marked preference for non-parasitized eggs.Oliviera et al. (2004) and Cabello et al. (2015) report the same preference.Chailleux et al. (2013), contrary to our results with O. niger, reveal that M. pygmaeus does not show any preference when feeding on yellow parasitized and non-parasitized host eggs.Even if the colour does not change visibly, the content of the eggs changes during the fi rst three days of parasitization.During egg laying, Trichogramma females inject a venom into the egg of its host, which digests the contents of the egg (Knutson, 1998).Parasitoid eggs hatch in about 24 h and the larvae start feeding on the liquid and solid contents of host's egg.All the contents are consumed by the larva within 8-10 h of the parasitoid's egg hatching during which time the larva increases 40-fold in size (Wu et al., 2000).Hence, feeding by predators on a parasitized egg, even if it is yellow, would be more diffi cult than feeding on a non-parasitized egg.Small body size (2 mm) and relatively weak mouthparts could make it harder for O. niger to eat 2-3 days old parasitized eggs, which are still yellow in colour.
On the other hand, some studies report that both chrysopids and coccinellids consume equal numbers of parasitized and non-parasitized eggs (Alrouechdi & Voegele, 1981;Roger et al., 2001).Most of the predators that do not discriminate between parasitized and non-parasitized eggs have chewing mouth parts, which would make it easier to overcome hardened egg shells.
As an exceptional result, Lingren & Wolfenbarger (1976), report equal consumption of black parasitized and non-parasitized eggs of Heliothis virescens (Fabricius) (Lepidoptera: Noctuidae) by Orius insidiosus Say.However, this report is based on fi eld observations of a few eggs (n = 23).Rosenheim & Harmon (2006) suggest that omnivorous intraguild predators are more likely to consume intermediate predators than coincidental intraguild predators, thus the risk of intraguild predation is lower for coincidental intraguild predators.According to Polis et al. (1989)'s definition, intraguild predation of O. niger on T. evanescens is coincidental intraguild predation and since O. niger consumes fewer parasitized E. kuehnilla eggs than non-parasitized eggs, our results are in accordance with Rosenheim & Harmon (2006)'s suggestion.
Similar studies that have been done in microcosms (Chailleux et al., 2013) and in greenhouse conditions (Cabello et al., 2015) show a reduced percentage of intraguild predation due to reduced encounter rates.Taking this into account, it is expected that the degree of intraguild predation of O. niger on T. evanescens is negligible but this still needs to be supported by fi eld trials.

Fig. 1 .
Fig. 1.Mean number (± SEM) of non-parasitized, 3 day old parasitized and 6 day old parasitized eggs of E. kuehniella eaten in 24 h by a female O. niger in no-choice experiments.Differences between columns with different letters are statistically signifi cant according to Duncan's test (P < 0.05).

Fig. 2 .
Fig. 2. Mean numbers (± SEM) of eggs consumed by a female of O. niger when non-parasitized eggs (control) alone were provided and when non-parasitized eggs and 3 day old parasitized eggs (0-3 days), non-parasitized eggs and 6 day old parasitized eggs (0-6 days), and 3 day old parasitized eggs and 6 day old parasitized eggs (3-6 days) of E. kuehniella were provided.Differences between columns with same letters are not statistically signifi cant according to Duncan's test (p < 0.05).

Table 2 .
Mean numbers (± SEM) of eggs consumed and mean preference indices (± SEM) of O. niger female when offered different combinations of different kinds of eggs.