Introduction
The reproductive potential of parasitoids is highly variable and dependent on many factors, including nutrition, host resources, experience and female egg load (Heimpel et al., Reference Heimpel, Rosenheim and Kattari1997; Kapranas and Luck, Reference Kapranas and Luck2008). The study of the relationship between different life-history traits and fitness in parasitoids is important to biological control, since it provides insights into the efficiency of a species to regulate agricultural pests. Thus, indirect attributes are often used to estimate the fitness of a species, which is typically measured in terms of fecundity and longevity (Takano and Takasu, Reference Takano and Takasu2019). One of the most common traits that are thought to affect a female's reproductive strategy and oviposition behaviour is body size (Roitberg et al., Reference Roitberg, Boivin and Vet2001), although others features such as the age of the females, host availability and wing length have not been extensively investigated and are likely to influence reproduction (Mills and Kuhlmann, Reference Mills and Kuhlmann2000; Kölliker-Ott et al., Reference Kölliker-Ott, Blows and Hoffmann2003; Santolamazza Carbone et al., Reference Santolamazza Carbone, Nieto and Rivera2008; Pan et al., Reference Pan, Wang, Zhang, Zhang and Liu2017; Asplen, Reference Asplen2020).
The rate at which females are capable of maturing eggs and oviposit them is used as a means to define the reproductive strategy of a species, classifying parasitoids into one of two groups: pro-ovigenic, when females present all mature eggs upon adult emergence and prior to oviposition, and synovigenic, when females are lacking mature eggs at emergence and are capable of maturing the eggs throughout life (Flanders, Reference Flanders1950). In order to quantify the egg production strategy, Jervis et al. (Reference Jervis, Heimpel, Ferns, Harvey and Kidd2001) formulated the ovigeny index (OI), which describes the relationship between the female's initial egg load upon emergence (fully mature eggs) and potential fecundity throughout life. Extreme values of the OI represent either a strictly pro-ovigenic strategy (OI = 1), or a strictly synovigenic strategy (OI = 0). However, it has been suggested that parasitoid species that exhibit strict synovigenic or pro-ovigenic behaviour are rare, and the OI mainly acquires continuous values between extremes (Liu et al., Reference Liu, Wang, Cheng, Guo and Wan2014). Defining the egg maturation strategy of parasitoids and the factors that influence their strategies could be helpful in terms of maximizing parasitism rates in a pest management strategy with biological control.
Cosmocomoidea annulicornis (Ogloblin) (Hymenoptera: Mymaridae) is a solitary egg parasitoid of Proconiini sharpshooters including Tapajosa rubromarginata (Signoret) (Hemiptera: Cicadellidae) (Triapitsyn et al., Reference Triapitsyn, Huber, Logarzo, Berezovskiy and Aquino2010; Manzano et al., Reference Manzano, Benzal, Logarzo, Coll Araoz, Virla and Luft Albarracin2021), a polyphagous sharpshooter present in citrus agroecosystems and vector of the bacterium Xylella fastidiosa Wells et al. (Xanthomonadaceae), causal agent of citrus variegated chlorosis (Dellapé et al., Reference Dellapé, Paradell, Semorile and Delfederico2016). The mymarids are among the most important natural enemies of insect pests (Peña et al., Reference Peña, Jacas, Triapitsyn, Ulmer, Duncan, Consoli, Parra and Zucchi2009), and members of this family are biological control agents of sharpshooters (Grandgirard et al., Reference Grandgirard, Hoddle, Petit, Percy, Roderick and Davies2007, Reference Grandgirard, Hoddle, Petit, Roderick and Davies2008; Banks et al., Reference Banks, Banks, Cody, Hoddle and Meade2019).
Previous field reports showed that C. annulicornis is a widely distributed (Triapitsyn et al., Reference Triapitsyn, Huber, Logarzo, Berezovskiy and Aquino2010) and abundant species, and some biological features, including parasitism, functional response and developmental rate have been evaluated in the laboratory (Manzano et al., Reference Manzano, Benzal, Logarzo, Coll Araoz, Virla and Luft Albarracin2021). However, aspects of the reproductive biology of the species are still unknown.
The aims of this paper were: (i) to describe the reproductive traits of C. annulicornis (i.e., initial egg load, fecundity and OI), (ii) to assess the influence of female age on egg maturation and oöcyte length and (iii) to evaluate the effect of host availability and feeding on the fecundity and longevity of C. annulicornis females. The results of this study will widen our basic knowledge of this species and its ability to regulate sharpshooter populations.
Materials and methods
Parasitoid rearing
A laboratory colony of C. annulicornis was reared on eggs of T. rubromarginata laid in plants of citrus rootstock Swingle citrumelo 75AB variety (Citrus paradisi Macf. × Poncirus trifoliata (L.) Raf.) under controlled conditions (25 ± 3°C, 30–45% RH and L14:10D photoperiod). T. rubromarginata individuals were collected using sweep nets in an open field in San Miguel de Tucumán, Tucumán, Argentina (S 26°48′36″–W 65°14′27″, elevation 465 m). To obtain host eggs, adults of T. rubromarginata were placed inside voile fabric bags with branches of potted plants. Sharpshooter females were allowed to oviposit for 24 h and the eggs were used for the maintenance of the parasitoid colony. All assays were performed in the Biological Control Division Laboratory at PROIMI, San Miguel de Tucumán, Tucumán, Argentina.
Ovigeny studies
To obtain data on the initial egg load and other reproductive traits of C. annulicornis, newly emerged females were randomly selected from the colony and held individually in glass vials (10 cm height × 1.5 cm diameter) with a solution of unpasteurized honey and water added in equal parts (50% v/v). This solution was used throughout all experiments performed in this study. Dissections were performed to 23 naïve wasps (<24 h old) under a stereomicroscope (Zeiss Stemi 2000c) at 50× magnification in order to determine initial egg load by counting mature oöcytes (large, pedunculated and fully chorionated) (fig. 1a). The number of immature oöcytes (small, oval, with a granular appearance and a medium constriction) (fig. 1b) present in each ovary was also registered. For each dissected female a series of morphometric measures were recorded: body size (from the front of the head, excluding the antennae, to the apex of the gaster), wing length, gaster length and mature oöcyte length (five eggs in each ovary) (morphometric measures are shown in Supplementary material).
Mature (a) and immature (b) oöcytes dissected from the ovaries of C. annulicornis females.
Effect of parasitoid adult age on egg maturation dynamics
Dissections of randomly selected naïve females were performed to evaluate the dynamics of oöcyte maturation over the females’ lifetime and the influence of adult age on egg load and oöcyte length. Individual females used for these dissections were isolated in glass vials (10 cm height × 1.5 cm diameter) immediately after they emerged and until they were dissected. Females were provided with honey on a daily basis. Naïve females were 24 (n = 23), 96 (n = 14), 120 (n = 9), 192 (n = 10) and 288 h (n = 10) post-eclosion when they were dissected, with no previous exposure to host eggs. The number of mature and immature oöcytes and oöcyte length (for five oöcyte per ovary) were recorded for each female.
Fecundity assays
The amount of eggs laid by C. annulicornis females throughout life, defined as realized lifetime fecundity was measured on 24 h host eggs. Newly emerged fertilized females were individually placed inside plastic round cages (2.5 cm height × 8.7 cm diameter) (experimental arena) and provided daily with diluted bee honey. Each female was also exposed daily to a citrus leaf carrying a freshly laid host egg mass (<24 h old) from emergence until natural death occurred. The longevity of each female (in days) was recorded. Once dead, female wasps (n = 13) were dissected and the number of remaining oöcytes was recorded.
Each exposed leaf carrying host eggs was removed from the experimental arena after 24 h and kept inside Petri dishes with a plaster base moistened with distilled water and covered with plastic film to prevent parasitoids from escaping. The number of parasitized eggs was recorded by counting host eggs that had undergone a change of coloration from yellowish to dark brown (Manzano et al., Reference Manzano, Benzal, Logarzo, Coll Araoz, Virla and Luft Albarracin2021). To confirm parasitism, emergence of the adults was observed (only one adult emerged per host egg) and the remaining host eggs with no clear evidence of parasitism or parasitoid emergence were dissected in order to detect parasitoid larvae or pupae inside the eggs.
Preliminary dissections were performed to numerous host eggs to assert the development of one parasitoid larva inside a single host egg (Carolina Manzano, unpublished data).
Ovigeny index
The potential lifetime fecundity of C. annulicornis females was calculated by adding the measured realized lifetime fecundity and the average of remaining oöcytes found in the gaster of females. Data obtained from fecundity assays and from dissections (‘Ovigeny studies’ section) allowed the further determination of C. annulicornis OI, calculated as the relationship between initial egg load and potential fecundity (Jervis et al., Reference Jervis, Heimpel, Ferns, Harvey and Kidd2001).
Effect of host availability and feeding on fecundity
The effect of host availability on fecundity was analysed by comparing the average number of laid eggs plus the remaining oöcytes found in the gaster of host-exposed females (n = 13) with the number of oöcytes (mature and immature) found in the gaster of 192 h old host-deprived females (n = 10). The latter age group was selected for this analysis because it was closer to the average lifespan of host-exposed females.
The influence of food provision on fecundity was also studied. For this, fecundity assays were also carried out following the procedure described above. Females used were subject to two different experimental treatments: (i) females provided daily with host eggs but maintained only with water (n = 12) and (ii) females provided daily only with hosts and water (with no access to honey) (n = 12). The number of eggs laid (parasitized host eggs) by each female daily and their longevity was recorded and results were statistically compared.
Effect of host availability and feeding on longevity
In order to assess the influence of food provision on the survivorship of C. annulicornis females, their longevity was measured under three different treatments: (i) honey-fed females (n = 50); (ii) honey-fed and exposed to host eggs during their entire lifetime (data obtained from fecundity assays) (n = 13) and (iii) females with no access to honey or host eggs and maintained exclusively with water (n = 15). The survivorship of females belonging to treatments (i) and (iii) was measured by isolating the parasitoids inside glass vials (10 cm height × 1.5 cm diameter) and providing them with either honey or water (added to the cotton covering the vial glass) from emergence until death. The methodology performed for treatment (ii) is described in ‘Fecundity assays’ section. The survivorship of the females was then compared and survivorship curves were constructed for each treatment.
Morphometric measurements in relation to fitness-related traits
The morphometric measures registered during the dissections carried out to C. annulicornis females and the results of other bioassays performed in this study were used to investigate whether there is a relationship between traits that are known to influence ovigeny: (1) realized fecundity vs. longevity; (2) realized fecundity vs. body size; (3) longevity vs. body size; (4) morphological measurements (body size, gaster length, wing length and oöcyte length) vs. mature and immature oöcytes recorded in 24 h C. annulicornis females and (5) parasitoid female age vs. oöcyte length.
Statistical analysis
A generalized linear model (GLM) with a Poisson distribution was used to compare the number of mature and immature oöcytes of C. annulicornis females and the oöcyte length registered.
Data on the effect of host and food availability on the fecundity of C. annulicornis females was analysed using the GLM with a Poisson distribution. Treatments were compared using likelihood ratio tests (LRT) and the Bonferroni adjustment was implemented for multiple testing of these comparisons.
A survival analysis was used to compare the effect of host and food availability on the longevity of C. annulicornis females by means of the nonparametric Kaplan–Meier survival analysis (Kaplan and Meier, Reference Kaplan and Meier1958). Log-rank tests were used to compare the survival curves from the different experimental treatments (females with access to host eggs and honey, host-deprived females fed with honey and females maintained without honey or hosts).
Spearman's rank correlations were carried out to analyse the relationship between the different morphological measures recorded in C. annulicornis females and their longevity and fecundity and between female age and oöcyte length. All statistical analyses were carried out at the 0.05 significance level using R software version 1.2.1335 (RStudio team, 2018).
Results
The dissections performed in naïve females of C. annulicornis evidenced the presence of paired ovaries containing mature (fig. 1a) and immature oöcytes (fig. 1b). An average egg load of 10.38 ± 0.51 (mature oöcytes) was observed for newly emerged C. annulicornis females (<24 h), while 5.68 ± 0.37 immature oöcytes were registered. The average dimensions recorded at this moment for the morphometric approach were: body size 1.53 ± 0.01 mm, gaster size 0.60 ± 0.01 mm, wing length 1.53 ± 0.01 mm and mature oöcyte length 0.28 ± 0.00 mm.
The number of mature and immature oöcytes varied with parasitoid age. Egg load was lowest when females were 24 and 288 h old and significantly greater when females were 96, 120 and 192 h old (GLM, LRT = 19.971, P < 0.001; fig. 2). The number of immature oöcytes recorded in the ovaries of 24 h old females was significantly higher than those in 96, 120, 192 and 288 h old females (GLM, LRT = 35.42; P < 0.001; fig. 2). Mature oöcyte length did not significantly vary among individuals of different ages (GLM, LRT = 0.01; P = 1).
Influence of parasitoid age on the number of mature and immature oöcytes found in the ovaries of host-deprived females of C. annulicornis. Different letters indicate significant differences between parasitoid ages (GLM, P < 0.05).
Of the total of host eggs offered in the fecundity assay, females of C. annulicornis parasitized, on average, 29.61 ± 2.06 freshly laid T. rubromarginata eggs during their lifetime (realized fecundity calculated on 1416 offered host eggs). The number of eggs laid by females was in all cases lower than the amount of host eggs offered. The highest average number of host eggs parasitized by females was reached during the first 24 h (4.76 ± 0.88) and thereafter the average daily fecundity decreased with age (fig. 3). The longest lifespan for females exposed to hosts was 12 days. Post-mortem dissections showed that 76.9% females contained both mature (3.92 ± 1.04) and immature oöcytes (0.61 ± 0.26) remaining inside the ovaries.
Age-specific fecundity and survivorship curves of C. annulicornis (n = 13). The curve with empty dots represents the mean number of T. rubromarginata eggs parasitized by C. annulicornis females per day and their corresponding SE. Bars show the average number of host eggs offered to females each day.
Potential fecundity of C. annulicornis, calculated as the average of remaining oöcytes (4.53 ± 1.17) plus the realized fecundity (29.61 ± 2.06), was 34.15 ± 1.88 eggs. The relation between initial egg load and potential fecundity, expressed as the OI, was calculated as 0.30 for C. annulicornis females.
Host availability markedly affected female fertility, which significantly laid a greater amount of eggs (plus the remaining eggs found inside the gaster) (potential fecundity: 34.15 ± 1.88) in the presence of host eggs during their lifetime than when they had no access to hosts (19.4 ± 0.74 eggs) (GLM; LRT = 45.99; P < 0.001). On the contrary, the potential fecundity of C. annulicornis females (laid eggs + remaining eggs inside the gaster) was significantly higher when females had daily access to honey (34.15 ± 1.88) than when they were food deprived (13.83 ± 0.92) (GLM; LRT = 110.22; P < 0.001). It is interesting to note that only two of the food-deprived females (n = 12) exposed to hosts were able to lay eggs.
Female wasps used in the fecundity assays (exposed to hosts and with honey at disposal) lived, on average, 9.84 ± 0.39 days, significantly more than females maintained exclusively with honey (6.66 ± 0.57 days) and than females with no access to food or host eggs during their lifetime (1.6 ± 0.19 days) (χ2 = 67.2; df = 2; P < 0.001). Survivorship curves of the three experimental treatments are shown in fig. 4.
Survivorship curves of C. annulicornis females subject to different experimental treatments: honey-fed, honey-fed plus host eggs and honey- and host-deprived females. Confidence intervals (95%) for each curve are represented by hatched areas.
There was a positive correlation between body size and oöcyte length (Spearman's rank correlation: S = 1158.7; R = 0.42; P = 0.04) in 24 h old females. The number of immature oöcytes was only significantly correlated with wing length (Spearman's rank correlation: S = 1089.6; R = 0.46; P = 0.02). However, no significant associations were found between the number of mature oöcytes found in the gaster of females and other morphological measurements (Spearman's rank correlation; body size vs. mature oöcytes: S = 2243.9; R = −0.10; P = 0.62; wing length vs. mature oöcytes: S = 2435.8; R = −0.20; P = 0.35; body size vs. immature oöcytes: S = 1964.6; R = 0.02; P = 0.89; oöcyte length vs. wing length: S = 1628.9; R = 0.19; P = 0.37) (table 1).
Spearman's rank correlation matrix showing statistical associations between variables measured for C. annulicornis females of 24 h (n = 23).
Asterisks denote statistical significance (P < 0.05).
In addition, the relationship between the number of parasitized host eggs (realized fecundity) and female longevity was analysed. These two variables were found to be highly correlated (Spearman's rank correlation; S = 684.63; R = 0.49, P = 0.03). No significant associations were observed between the fecundity and longevity of C. annulicornis females and their body size (Spearman's rank correlation; S = 72; R = 0.14, P = 0.75; S = 76.87; R = 0.08, P = 0.84). A significant negative correlation was found between parasitoid female age and oöcyte length (Spearman's rank correlation; S = 60572; R = −0.26; P = 0.03).
Discussion
Here, we report a highly synovigenic egg maturation strategy for the egg parasitoid C. annulicornis. This result contrasts with previous reports on other mymarids, which have traditionally been classified as strictly pro-ovigenic (Jervis et al., Reference Jervis, Heimpel, Ferns, Harvey and Kidd2001; Jepsen et al., Reference Jepsen, Rosenheim and Matthews2007). In fact, several species (including species belonging to the genus Cosmocomoidea (formerly Gonatocerus)) display this type of reproductive strategy (Irvin and Hoddle, Reference Irvin and Hoddle2009), evidenced by female emergence with a mature complement of eggs and no further egg maturation recorded. However, exceptions to this observed trend have been documented for some mymarid species: Anaphes iole Girault, A. listronoti Huber (Riddick, Reference Riddick2005; Boivin and Martel, Reference Boivin and Martel2012; Boivin and Ellers, Reference Boivin and Ellers2016) and Anagrus virlai Triapitsyn (Hill et al., Reference Hill, Aguirre, Bruzzone, Virla and Luft Albarracin2020) display OIs between 0.70 and 0.90, for which they are considered highly pro-ovigenic, but capable of maturing eggs in addition to their initial egg load.
Moreover, there are other mymarid species displaying a synovigenic strategy, although there are fewer reports of these cases. For instance, an OI similar to that obtained for C. annulicornis was reported for Cosmocomoidea ashmeadi (OI = 0.22) (Irvin and Hoddle, Reference Irvin and Hoddle2009).
Although the OI calculated for C. annulicornis would indicate that this is a highly synovigenic species, the age-specific curve of fecundity observed in C. annulicornis females (fig. 3) does not follow the behaviour of the curve described by Jervis et al. (Reference Jervis, Ellers and Harvey2008). According to the authors, synovigenic species with intermediate OI values display fecundity curves in which females present a short preoviposition period and start laying eggs after the first few days of life. In other cases, females lay few eggs immediately after adult emergence and thereafter egg deposition rapidly increases during the first days before declining. Contrary to this behaviour, during the first 24 h of oviposition C. annulicornis females display the highest fecundity and after this time period, the daily oviposition rate progressively declines until death, lacking extremely marked oviposition peaks.
The egg production rate of C. annulicornis females was affected by age. There are numerous studies parasitoids that provide information about the state of egg maturation in newly emerged females, yet few studies include data about the proportion of mature and immature oöcytes in the ovary of females at specific ages throughout life. This information provides a more complete picture of the temporal dynamics of egg maturation (Jervis et al., Reference Jervis, Heimpel, Ferns, Harvey and Kidd2001).
The pattern of egg maturation used by C. annulicornis females suggests the possibility that oosorption of mature oöcytes occurs in the absence of hosts in order to return their content to the reserve of somatic nutrients. Egg oosorption has been described in other parasitoids (Collier, Reference Collier1995; Asplen and Byrne, Reference Asplen and Byrne2006), including mymarid species (Santolamazza Carbone et al., Reference Santolamazza Carbone, Nieto and Rivera2008), which redirect the energy acquired to maintenance, survival, host searching, development or production of new mature eggs (Irvin and Hoddle, Reference Irvin and Hoddle2009). If oosorption occurred in C. annulicornis, it would be highly advantageous in the field were the resources reallocated to increasing longevity, which translates to additional time for host searching. Another aspect that is in need of consideration is the cost involved in the egg oosorption process, that is, after nutrients are reallocated, females experience a transient egg limitation, during which the production on new mature eggs is a time consuming process before oviposition can be resume (Rosenheim et al., Reference Rosenheim, Heimpel and Mangel2000).
However, histological studies are necessary to confirm the occurrence of this phenomenon (i.e., through the presence of pignotic residues of the follicular cells) in C. annulicornis females.
The age of C. annulicornis females negatively affected oöcyte length, a size-related trait that has been associated in other parasitoid species with an increase in fitness (Visser, Reference Visser1994; Durocher-Granger et al., Reference Durocher-Granger, Martel and Boivin2011). The causes of variation in the egg producing strategies between species are still debated, yet it is thought that female oviposition behaviour is highly dynamic and influenced by various factors, including variations in host abundance and quality (Papaj, Reference Papaj2000) and adult nutrition (Azzouz et al., Reference Azzouz, Giordanengo, Wäckers and Kaiser2004; Benelli et al., Reference Benelli, Giunti, Tena, Desneux, Caselli and Canale2017).
In addition, C. annulicornis females that were host deprived during 192 h had significantly fewer mature and immature oöcytes inside their gaster compared to females of approximately the same age that had access to host eggs during their lifetime. These results support the hypothesis that egg oosorption takes place in the absence of hosts. Furthermore, it is possible that C. annulicornis females are able to modify its rate of egg production in response to host availability (Hougardy et al., Reference Hougardy, Bezemer and Mills2005). A similar behaviour has been reported in females of the parasitoid Eupelmus vuilleti (Craw) (Hymenoptera: Eupelmidae), which after long host-deprivation periods display a decrease in the amount of immature oöcytes and oöcyte resorption (Bodin et al., Reference Bodin, Jaloux, Delbecque, Vannier, Monge and Mondy2009).
The influence of sugary sources on reproductive traits in parasitoids is a key point to be assessed for biological control (Tena et al., Reference Tena, Pekas, Cano, Wäckers and Urbaneja2015). In our study, food provision had a positive effect on the reproductive traits of C. annulicornis. When provided daily with host eggs and honey, C. annulicornis females laid a significantly larger amount of eggs than when they were food deprived. The fact that most females failed to lay eggs in the absence of a sugar source suggests that this is an ‘anautogenous’ species that needs to feed before being able to parasitize host eggs (Jervis and Kidd, Reference Jervis and Kidd1986).
Host and food availability also influenced the longevity of C. annulicornis females. Females that were food and host deprived lived significantly less than those who were supplied daily with honey. This, added to the decrease in the fecundity of honey-deprived females, indicates that non-host food nutrients acquired during adult stage are essential for the reproductive success of C. annulicornis females.
When comparing the two experimental treatments with food availability, results showed that females that additionally had daily availability of host eggs lived significantly more days. A possible explanation for this is that continuous host availability induces further oogenesis and oviposition (Hougardy et al., Reference Hougardy, Bezemer and Mills2005), increasing longevity and therefore, fitness.
Parasitoid fecundity is not independent from other reproductive traits, and trade-offs between these are observed among different species (Jervis et al., Reference Jervis, Ellers and Harvey2008). In that regard, positive associations between some reproductive traits were found in 24 h old C. annulicornis females. On the one hand, larger females presented larger oöcytes. Body size has been commonly associated in parasitoids with egg production (Roitberg et al., Reference Roitberg, Boivin and Vet2001). The increase in number, size and quality of eggs is directly related to an increase in fitness, although to date this association has not been extensively studied in parasitoids (Martel et al., Reference Martel, Darrouzet and Boivin2011). A positive correlation between body size and gamete size was also found in the trichogrammatid wasp Trichogramma euproctidis (Girault) (Durocher-Granger et al., Reference Durocher-Granger, Martel and Boivin2011). The authors suggest that this pattern would be a result induced by phenotypic plasticity in response to the resources acquired during the larval stages of the parasitoid, that is, resources acquired from the host during development, which would have an influence on the production of gametes. The increase in gamete size could be advantageous if eggs develop into individuals with higher competitive abilities, which translates into a gain in fitness. Although in this study individuals of C. annulicornis developed within eggs of the same host and age, it is possible that there are nutritional differences in the content of the host eggs that affect certain female reproductive traits.
The same concept could be applied to explain the absence of correlations between body size and fecundity and between body size and longevity in C. annulicornis females. Even though the size of parasitoids is positively related to these life history traits in several species (Ellers et al., Reference Ellers, Van Alphen and Sevenster1998; Lauzière et al., Reference Lauzière, Pérez-Lachaud and Brodeur2000; Sagarra et al., Reference Sagarra, Vincent and Stewart2001; Jervis et al., Reference Jervis, Ferns and Heimpel2003), according to Blackburn (Reference Blackburn1991), the absence of these correlations could be due to a limitation imposed by the size of the host or by the environment.
Furthermore, there was a significantly positive correlation between wing length and the number of immature oöcytes present in C. annulicornis females. However, wing length did not affect the number of mature oöcytes. The trade-off between the gonads and the flight apparatus (wings) appears to be a common pattern in some migratory insects (Sappington and Showers, Reference Sappington and Showers1992; Dixon et al., Reference Dixon, Horth and Kindlmann1993). Johnson (Reference Johnson1969) was the first to relate flight and reproduction in insects, naming this association as ‘the oogenesis-flight syndrome’, and suggesting that it is an antagonistic process in which migration inhibits egg production and vice-versa. This syndrome has been documented in many studies conducted in insects with wing dimorphism, including aphids and other groups of insects (Oliveira et al., Reference Oliveira, Baptista, Guimarães-Motta, Almeida, Masuda and Atella2006; Lorenz, Reference Lorenz2007; Zhang et al., Reference Zhang, Wu, Wyckhuys and Heimpel2009). Although this syndrome is widely accepted in entomology, many studies carried out in insects that do not present wing dimorphism suggest that some species present no negative correlation between flight and fertility, while in others species flight seems to stimulate egg production (Tigreros and Davidowitz, Reference Tigreros and Davidowitz2019). In fact, most of the species in which a trade-off between flight and fecundity is reported were subjected to low availability of resources during development or to forced flights. The positive flight-fecundity correlations reported for some species could be explained taking into account that natural selection could be favouring maximizing both traits under certain conditions.
For parasitoid wasps facing scarce or low-quality hosts, and with mature eggs available, rapid dispersal to locate host patches could be advantageous (Asplen, Reference Asplen2020). In the same way, C. annulicornis females presenting larger wings and a higher amount of immature oöcytes would have a higher complement of eggs to gradually mature as they disperse searching for new host egg patches.
Similarly to our findings on C. annulicornis, a positive association between fecundity and longevity has been observed in many parasitoid species belonging to different families, such as Dolichogenidea tasmanica (Cameron) (Hymenoptera: Braconidae), Pholetesor ornigis (weed) (Hymenoptera: Braconidae), Trichogramma nubilale Ertle and Davis (Hymenoptera: Trichogrammatidae) and C. ashmeadi (Hymenoptera: Mymaridae) (Olson and Andow, Reference Olson and Andow1998; Berndt and Wratten, Reference Berndt and Wratten2005; Irvin and Hoddle, Reference Irvin and Hoddle2007). In accordance with our findings, these studies also demonstrated that diet positively influenced longevity in females and that, when fed with sugary substances, females lived longer and laid more eggs. Nevertheless, in the case of the highly pro-ovigenic wasp A. virlai, although feeding influenced longevity, it did not significantly affect the fecundity of the females, since they laid almost all of their mature eggs during the first few days of life (Hill et al., Reference Hill, Aguirre, Bruzzone, Virla and Luft Albarracin2020).
In synovigenic species such as C. annulicornis, a greater longevity would increase the probabilities of surviving unfavourable conditions and, therefore, increase oviposition (Boivin, Reference Boivin2010), added to the fact that the gradual maturation of the eggs throughout life would allow them to temporarily cope with host shortage (Savino, Reference Savino2014).
Knowledge of reproductive traits of parasitoids is necessary to assess their potential as biocontrol agents (Hegazi et al., Reference Hegazi, El-Aziz, El-Shazly and Khafagi2007). The highly synovigenic strategy displayed by C. annulicornis females can be considered as an advantageous attribute in a biological control scenario, compared to pro-ovigenic females, since synovigenic species would present the intrinsic capacity to reproduce and maintain their host population at low host densities, as a consequence of a higher longevity and their ability to maintain constant reserves of mature eggs throughout life (Flanders, Reference Flanders1950).
The results provided here on the key reproductive traits of the species, their synovigenic strategy and long lifespan, together with the high parasitism levels previously reported (Manzano et al., Reference Manzano, Benzal, Logarzo, Coll Araoz, Virla and Luft Albarracin2021) are key for breeding and pest management programmes and support evidence for C. annulicornis as a promising biocontrol agent for the sharpshooter T. rubromarginata.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0007485321000766
Acknowledgements
This study was supported by PICT 2015 1147 (FONCYT, Fondo para la Investigación Científica y Tecnológica, Argentina). We thank ‘Vivero Lules’ for providing the citrus plants used in assays. Carolina Manzano thanks CONICET (Consejo Nacional de Investigaciones Científicas y Técnicas, Argentina) for the scholarship granted.
