Parasitoids are important insects that are commonly released into the environment to reduce the population sizes of pest species. The lifetime reproductive success of parasitoids mainly depends on host availability and the availability of mature eggs. Consequently, it is predicted that female wasps must balance the risk of egg or host (time) limitation with maximized lifetime fecundity. Typically, synovigenic females, which continue to mature eggs throughout their lifetime, have been shown to adjust their egg production rate in response to environmental variations in host availability to reduce the risk of egg limitation. In this study, we found that in a synovigenic egg parasitoid, Anastatus japonicus (Hymenoptera: Eupelmidae), the oviposition experience of Ana. japonicus females significantly enhanced the egg load and increased the rate of mature egg production. However, in contrast to other studies, the experience of contact with a host did not significantly affect the egg load in females. This result suggests that the overall oviposition experience might induce an adjustment and accelerate egg maturation in Ana. japonicus and is likely more important in egg maturation than transitory host contract. In addition to affecting the egg load, oviposition experience influenced Ana. japonicus female reproductive behaviour, which shifted virgin female behavioural preferences from mating to oviposition and laying more eggs per clutch. Our study provides an optimal strategy for the post-oviposition release of Ana. japonicus, an egg parasitoid of several lepidopteran forest pests, to improve biocontrol effectiveness.

Anastatus japonicus, biological control, egg limitation, egg load

Reproduction is crucial for all animals. Insect parasitoids are insects that parasitise other organisms, and all invertebrate life stages, including the egg, larval/nymphal, pupal, and adult stages are susceptible to parasitisation. Because of their parasitic nature, an increasing number of species have been extensively released to reduce the population sizes of pest species (Hassan 1993; Zhishan et al. 2003; Asgari and Rivers 2011; Wang et al. 2019). Many studies on parasitoid wasps have focused on applied research, with the intent to improve the attack and control of pest populations by such species (Powell 1986; Wajnberg et al. 2008; Yang et al. 2014). Generally, the lifetime reproductive success of parasitoids mainly depends on host availability and the availability of mature eggs (Jervis et al. 2001; Tylianakis et al. 2004; Hougardy et al. 2005). In the field, parasitoids generally experience one of the two following situations. Either the number of mature eggs available for laying exceeds the number of oviposition opportunities or the number of oviposition opportunities exceeds the number of mature eggs available for laying; these are defined as host/time limitation and egg limitation, respectively (Godfray 1994; Richard and Casas 2012). However, a perfect match between the number of hosts and the availability of mature eggs is rare. Consequently, to maximize reproductive success, insect parasitoids are hypothesized to balance the risk of egg limitation and time limitation (Minkenberg et al. 1992; Rosenheim 1996; Heimpel et al. 1998; Rosenheim et al. 2008).

According to the type of egg production, parasitoids can be classified as pro-ovigenic or synovigenic. Pro-ovigenic species mature all or most of their lifetime complement of eggs prior to emergence from the host, whereas synovigenic species emerge with very few or no mature eggs, and egg maturation begins at eclosion and continues throughout adult life (Flanders 1950; Jervis et al. 2001). A number of internal and environmental factors have been shown to influence egg production and maturation in both pro-ovigenic and synovigenic species. Host resources greatly influence synovigenic females that adjust their egg production rate in response to host availability (Papaj 2000). For example, in the parasitoid Eupelmus vuilleti (Crawford) (Hymenoptera: Eupelmidae), contact with a host by the female antennae can accelerate egg maturation (Casas et al. 2009), and in the aphid parasitoid Aphelinus albipodus (Hayat & Fatima) (Hymenoptera: Aphelinidae), females mature eggs faster in the presence of preferred hosts (Wu and Heimpel 2007). In those studies, females were usually manipulated by a single contact with the host (i.e., antennal) to perceive the host’s presence, and then egg loads were counted. However, contact or perceived contact with a host is only one component of complete oviposition behaviour. Therefore, this study mainly focused on the effect of complete oviposition behaviour on egg loads in female wasps.

Lymantria dispar (L.) (Lepidoptera: Erebidae) is a leaf-feeding insect that causes the large-scale defoliation of forest and urban trees worldwide during intermittent population outbreaks. Anastatus japonicus Ashmead (formerly Anastatus disparis Ruschka) (Hymenoptera: Eupelmidae) is a recorded egg parasitoid of L. dispar (Crossman 1925; Kurir 1944; Avci 2009; Alalouni et al. 2013) and is widely distributed in Europe and Asia. In addition, due to its potential as a biological control agent of L. dispar, Ana. japonicus was imported to North America in the early 1900s (Crossman 1925; Yan et al. 1989) and achieved effective control efficiency. Field investigations revealed that, together with Ooencyrtus kuvanae (Howard) (Hymenoptera: Encyrtidae), which is also a primary egg parasitoid of L. dispar, parasitism rates might reach 20–40% (Hoy 1976; Reardon 1981; Brown and Cameron 1982; Brown 1984). In addition to L. dispar, several other noxious lepidopteran species that are primarily forest pests, including Malacosoma neustria testacea (Motschulsky), Odonestis pruni (Linnaeus), Antheraea pernyi (Guerin-Meneville), Dendrolimus punctatus tabulaeformis (Walker), and Actias selene ningpoana (Felder), can be parasitised (Yan et al. 1989; Li and Lou 1992; Li et al. 2001). Great efforts have been made to mass rear Ana. japonicus for the control of Caligula japonica (Moore) in China, as the eggs of Ant. pernyi are suitable factitious hosts (Yan et al. 1989). The lifetime fecundity of a female Ana. japonicus is several hundred offspring (Liu et al. 2015), and the offspring sex ratio is mainly influenced by host quality (i.e., size) (Liu et al. 2017). The egg maturation mode of Ana. japonicus is typically synovigenic, and previous studies found that females usually contained tens of mature eggs by dissecting the abdomen (Liu and Hao 2019). Hosts of Ana. japonicus, i.e., L. dispar, typically lay clutches containing hundreds or thousands of eggs; thus, the lifetime reproductive success of a female parasitoid is limited by the supply of mature eggs, i.e., egg limitation. Several synovigenic females have been shown to adjust their egg production rate in response to environmental variations in host availability (Papaj 2000; Wu and Heimpel 2007; Casas et al. 2009). During oviposition in a host, parasitoids may learn to recognize particular visual and olfactory host stimuli and use these cues to modify subsequent behaviours (Vet et al. 1995; Vinson 1998). Thus, in this study, the effect of the complete oviposition experience on egg load was studied in Ana. japonicus. In addition, the effect of contact with a host on egg load was tested, as well. Finally, the effects of oviposition experience on subsequent behaviour and capacity, e.g., behavioural choice and host exploration capacity, were also studied in Ana. japonicus.

Anastatus japonicus is a typically synovigenic parasitoid that continuously matures eggs throughout its lifetime. Synovigenic females have been shown to adjust their egg production rate in response to host availability (Papaj 2000). Our study provides further evidence that the oviposition experience of Ana. japonicus females significantly enhances female egg load and rapidly increases the number of mature eggs (Fig. 1). Hosts are important resources for reproduction, and it is predicted that female wasps must balance the risks of egg and time limitation with maximum lifetime fecundity (Minkenberg et al. 1992; Rosenheim 1996; Heimpel et al. 1998; Rosenheim et al. 2008). Thus, Ana. japonicus females with oviposition experience are likely to reduce the risk of egg limitation by rapidly increasing their egg loads. Similar to the study in the parasitoid E. vuilleti (Crawford) (Casas et al. 2009), oviposition experience might initiate a hormonal cascade leading to egg maturation in Ana. japonicus and should be studied further. In contrast to a study in the parasitoid E. vuilleti, host contact experience in Ana. japonicus did not affect the egg loads in females. This result suggests that oviposition experience is likely more important in egg load maturation than transitory host contact in Ana. japonicus. A limitation of this study was that it did not consider the effects of host density in Ana. japonicus, as a fixed number of host eggs was provided. The density of hosts in the environment has been suggested to be directly linked to egg maturation rates in many studies (Rosenheim 1996; Rosenheim et al. 2000; Segoli and Rosenheim 2013). Usually, a high host density elicits production of a larger egg load (Bodin et al. 2009; Casas et al. 2009). However, a negative effect of host density on egg load was observed in the soybean aphid parasitoid Binodoxys communis (Gahan) (Dieckhoff et al. 2014). Thus, the effect of host density on egg load should be studied and considered in further research.

Hymenopterans are haplodiploid; thus, virgin females can produce male offspring before mating (Cook 1993; Heimpel and de Boer 2008). Hypothetically, virgin females need to evaluate the following trade-offs: (1) search for hosts and produce sons immediately or (2) search for mates and perhaps produce both sons and daughters later in life (Godfray 1990; Godfray and Hardy 1993; Godfray 1994; Fauvergue et al. 2008). In Ana. japonicus, the results showed that virgin females preferred to complete mating first. A specific local mating system likely contributed to the preference for mating. It may be a result of long-term evolution that female dispersal occurs before mating. In addition, virgin Ana. japonicus may have lower fitness than mated females, that only produces male offspring. However, our study showed that the oviposition experience of females changed the behavioural preference from mating to oviposition. A potential explanation for oviposition experience influencing subsequent oviposition behaviour in Ana. japonicus females is that females with oviposition experience may fine-tune behaviours closely associated with reproduction (Pyle et al. 1991). Alternatively, compared to females without experience, those with oviposition experience had larger egg loads, indicating that these individuals had a higher risk of time limitation. Time-limited species often have large egg loads or the ability to quickly replenish their egg supply, so their reproductive success is proportional to the number of hosts they are able to attack during their lifetime (Stephens and Krebs 1986; Charnov and Stephens 1988).

Thus, oviposition experience of females changed the behavioural preference from mating to oviposition in our species, may be more expected to achieve the reproductive success in response to the risk of time limitation.

Generally, in addition to egg load, oviposition experience influenced female Ana. japonicus reproductive behaviour, shifting virgin female behavioural preference from mating to oviposition, allowing more eggs to be laid per host clutch. The change in behavioural preference from mating to oviposition may be a direct effect of oviposition experience in females. In our study, when an artificial host clutch containing nine Anth. pernyi eggs was offered for oviposition, females with oviposition experience laid a mean of 1.78 ± 0.15 eggs, which was significantly more than that laid by females without oviposition experience. In addition, during oviposition in a host, parasitoids learn to recognize particular visual and olfactory stimuli of the host and use these cues to modify subsequent behaviours (Vet et al. 1995; Vinson 1998). Thus, oviposition experience might also be associated with an increased hatch rate and increased host acceptability, resulting in more eggs laid per clutch. In addition, among all the oviposition behaviours observed in the current study, conspecific superparasitism occurred only once (1/43), as two females in a dish oviposited in the same host. Females exhibited a host discrimination ability and could identify parasitised hosts to avoid wasting eggs (Liu et al. 2018). Our experimental design including a large host number for females may also have contributed to the general absence of conspecific superparasitism.

Anastatus japonicus is an egg parasitoid of L. dispar (Crossman 1925; Kurir 1944; Avci 2009; Alalouni et al. 2013) and a potential biological control agent of this species in North America (Crossman, 1925; Yan et al. 1989). Great efforts have been made to mass rear Ana. japonicus for the control of C. japonica in addition to L. dispar in China, and the eggs of Ant. pernyi are suitable factitious hosts (Yan et al. 1989). Our study suggested that females with oviposition experience had a higher reproductive value, with larger egg loads and a preference for oviposition. Therefore, in the rearing of Ana. japonicus for biological control, hosts can be provided before the release of wasps so that the female wasps can gain oviposition experience. This will likely improve the reproductive value of the released female wasps so that the wasps can parasitize more hosts, improving the efficacy of biological control.