P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

C 2001) Journal of Insect Behavior, Vol. 14, No. 5, September 2001 (°

Lifetime Reproductive Success in the Solitary Endoparasitoid, Venturia canescens Jeffrey A. Harvey,1,2 Ian F. Harvey,1 and David J. Thompson1 Accepted December 12, 2000; revised April 24, 2001

Parasitoid wasps have long been considered excellent organisms in studies examining the evolution of reproductive and life-history strategies. In examining the lifetime reproductive success of parasitoids in the laboratory, most investigations have provided the insects with excess hosts and food, where they exist in a relatively constraint-free environment. Importantly, these conditions may not accurately reflect the true heterogeneity of natural systems, where suitable hosts and food sources are likely to be limiting. This study examines the influence of differences in host and food availability on reproductive and lifehistory parameters in an asexual strain of the solitary endoparasitoid, Venturia canescens (Hymenoptera: Ichneumonidae). Lifetime reproductive success in V. canescens was measured in response to temporal variations in host and food (honey solution) access. Cohorts of parasitoids were provided with 200 fifthinstar larvae of the Indian meal moth, Plodia interpunctella (Lepidoptera: Pyralidae), and food for variable periods daily after eclosion. V. canescens is synovigenic, and host-deprived wasps continued to mature eggs over the first few days after eclosion until the egg storage capacity was reached in the oviducts. When these wasps were subsequently provided with hosts, oogenesis resumed and continued until later in adult life. Constantly fed wasps lived longer and produced more progeny than wasps from cohorts which were alternately fed and starved or were starved from eclosion. Moreover, wasps with constant host and food access produced most progeny early in life and usually experienced prolonged periods of postreproductive survival. In contrast, 1Population

Biology Research Group, School of Biological Sciences, University of Liverpool, Liverpool, U.K. 2To whom correspondence should be addressed at Centre for Terrestrial Ecology, Netherlands Institute of Ecology, Postbox 40, 6666 ZG Heteren, The Netherlands. E-mail: [email protected]. 573 C 2001 Plenum Publishing Corporation 0892-7553/01/0900-0573$19.50/0 °

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

9:30

574

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

the reproductive period of wasps with limited host access was more evenly distributed throughout the adult life. Consequently, the cumulative progeny production by V. canescens with constant food access was fairly uniform irrespective of host availability. Longevity and fecundity in V. canescens were positively correlated with adult size. However, variable host access had little effect on the longevity of wasps which were constantly supplied with honey. Over the first 2 days of adult life, variation in food access also had no effect on progeny production by V. canescens. We argue that manipulating temporal host and food access to parasitoids in the laboratory more closely approximates natural conditions, where these resources are likely to be spatially separated. Moreover, our findings suggest that many highly synovigenic parasitoids like V. canescens, which produce microtype (=hydropic) eggs, have a considerably higher reproductive potential than ovary dissections have revealed. Our findings are discussed in relation to life-history evolution in the parasitic Hymenoptera. KEY WORDS: Parasitoid; reproductive success; Venturia canescens; Plodia interpunctella; synovigenic; life-history strategy.

INTRODUCTION The lifetime reproductive success of parasitoids has important implications for our understanding of the evolution of life-history strategies and the demographics of host–parasitoid interactions (Godfray, 1994; Heimpel et al., 1998). Reproductive success of female parasitoids is determined by several factors, including the accessibility and abundance of suitable hosts and the number of eggs that are available for oviposition (Rosenheim, 1996; Heimpel et al., 1997). Furthermore, reproductive success also depends on the life expectancy of the female parasitoid and on the rate and timing of egg maturation during her lifetime. Finally, among many possible factors, parasitoid foraging behavior and oviposition decisions are influenced by the rate at which suitable hosts are encountered and the physiological state of the parasitoid. Very few studies have measured the lifetime reproductive success of parasitoids in the field, because of the obvious difficulty of monitoring the foraging behavior of female wasps from eclosion until death (but see Visser, 1994; Heimpel et al., 1998). For this reason, the vast majority of studies examining lifetime offspring production in parasitoids have been performed in the laboratory (e.g., Mackauer, 1982; Orr and Boethel, 1990; Kopelman and Chabora, 1992; Bai and Smith, 1993). However, in most of these studies the insects have been provided with a surfeit of hosts and food, which may not accurately reflect the heterogeneity of natural systems, where these resources are likely to be limiting.

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

Lifetime Reproductive Success in a Parasitoid Wasp

9:30

Style file version Feb 08, 2000

575

In nature, three main food sources are used by adult parasitoids: hosts, honeydew, and certain plant exudates, such as nectar (Jervis and Kidd, 1986). These are potentially extremely important sources of adult nutrition because they may significantly influence parasitoid longevity and fecundity, thus increasing parasitism (Allen and Smith, 1958; Syme, 1975; Jervis and Kidd, 1986; Jervis et al., 1993). Since hosts may be located a considerable distance from suitable food sources, the costs and benefits of leaving host patches to search for food are likely to depend upon many factors. These include mortality risks incurred when dispersing to search for (and return from) the food source, the quality of the host patch, and the physiological condition of the parasitoid (Sirot and Bernstein, 1996; Sirot et al., 1997). After eclosion the females of many parasitoid species continue to mature eggs over a significant portion of their lifetime [=synovigeny (Flanders, 1950; Jervis and Kidd, 1986)]. Synovigenic parasitoids generally fall into two categories: (a) wasps that produce small, yolk-deficient (=hydropic) eggs which are matured for variable periods throughout adult life, without the requirement for access to an extrinsic source of protein, and (b) wasps that produce large, yolk-rich (=anhydropic) eggs, where continued oogenesis after eclosion requires an extrinsic source of protein, frequently obtained though feeding on host blood by the adult parasitoid female (Flanders, 1950; Jervis and Kidd, 1986; Heimpel et al., 1996; Ueno, 1999). Importantly, different types of feeding behavior in the two groups can have markedly different effects on their reproductive success. For parasitoids producing hydropic eggs, the presence or absence of a source of adult nutrition influences fecundity through its concomitant effects on longevity (Sirot and Bernstein, 1996; Ellers et al., 1998, 2001). Alternately, the consumption of sugar-rich foods potentially interacts with host-feeding to influence lifespan and fecundity in parasitoids producing anhydropic eggs (Jervis and Kidd, 1986). If metabolic energy reserves are insufficient for anhydropic egg-producing parasitoids to achieve maximum longevity and fecundity, the investment of resources may be partitioned between gametic (=egg) production and somatic tissues for maintenance and repair (Kirkwood, 1981). A female parasitoid that invests more in egg production may have less to invest in maintenance, and vice versa. Predictive models have frequently been employed in describing how parasitoids forage most profitably in a variable and changing environment (e.g., Charnov, 1979, 1982; Charnov and Skinner, 1985; Mangel, 1987; Roitberg et al., 1992; Heimpel et al., 1996; Sirot and Bernstein, 1996). Most of these models assume trade-offs in life-history characters and predict that individual parasitoid behavior varies in response to changes in their physiological state (e.g., egg load, hunger level) brought about by differences in environmental quality (e.g., the number and distribution of potential host types). However, the influence of these interacting states on reproductive success in

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

576

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

parasitoids has been largely ignored. Several workers also suggest that fitness returns, in terms of reproductive success, vary with differences in adult parasitoid body size (e.g., Hurlbutt, 1987; Visser, 1994; Petersen and Hardy, 1995; Kazmer and Luck, 1995; Harvey et al., 1998). Similarly, few investigations have actually determined the influence of differences in body size on total progeny production in life-history studies. In this paper we examine the influence of host and food availability on lifetime reproductive success in an asexual strain of the solitary endoparasitoid, Venturia canescens Gravenhorst (Hymenoptera: Ichneumonidae), developing in its habitual host, the Indian meal moth, Plodia interpunctella Hubner (Lepidoptera: Pyralidae). V. canescens is synovigenic and produces large numbers of hydropic eggs, so that adult feeding behavior is utilized exclusively for metabolic maintenance (somatic tissue) rather than for egg production [germinal tissue (Harvey, 1995)]. The host range of V. canescens is fairly broad (Salt, 1976), and includes a number of microlepidoptera which are common pests in granaries and flour mills, including P. interpunctella (Richards and Thomson, 1932). Sexual strains of V. canescens also occur in southern Europe and appear to attack hosts which occur singly in fruits such as figs (Driessen and Bernstein, 1996). In either of these situations, parasitoids are likely to emerge into an environment where hosts may have shifted spatially or where suitable sources of nutrition (e.g., nectar) are located a considerable distance from host patches (Sirot and Bernstein, 1996). The importance of adult nutrition on host and food searching behavior in V. canescens in flour mills was first reported by Beling (1932). She observed that newly emerged wasps frequently left the host environment soon after eclosion, whereas other wasps were captured returning to mills with droplets of nectar in their mandibles. Therefore, reproductive success in V. canescens is probably determined in large measure by trade-offs between the immediate fitness returns of searching for hosts and the indirect fitness returns of searching for sources of adult nutrition. Here we determine if lifetime reproductive success in V. canescens varies with temporal differences in host and food access and that, when food is limiting, changes in the physiological state of the parasitoid lead to a concomitant increase in the rate of oviposition by the parasitoid in accordance with the prediction of several recent models (e.g., Roitberg et al., 1992; Sirot and Bernstein 1996; Sirot et al., 1997). Furthermore, we explore the effects of parasitoid size on progeny production in response to temporal variability in host and food access. Finally, we attempt to establish if foraging costs reduce the life expectancy in V. conescens in comparison with parasitoids deprived of hosts. Our findings are discussed in relation to the hypothesis of trade-offs in life-history characters.

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Lifetime Reproductive Success in a Parasitoid Wasp

577

MATERIALS AND METHODS All experiments were conducted and cultures maintained at 25 ± 2◦ C with a 16:8-h light:dark photoperiod. Insects The original culture of P. interpunctella was supplied from an outbred colony reared at Dundee University, Scotland. They were reared in clear glass jars (500 ml) on a 10:1:1 mixture of wheatmeal, dried brewer’s yeast, and glycerol containing ≈25 g of food and 200 P. interpunctella eggs (per Rogers, 1972). P. interpunctella passes through five (or, rarely, six) instars, and egg-to-adult development is completed in about 4 weeks (Harvey, 1995). An asexual strain of V. canescens was obtained from a culture maintained at the University of Dundee, Scotland. Parasitoids were reared in clear plastic boxes (17.5 × 11.0 × 5.0 cm) with ≈200 fourth- and fifth-instar (L4 and L5 ) P. interpunctella larvae. Approximately 10 adult V. canescens were placed into each container with hosts. Upon eclosion, several drops of honey were smeared inside the box, providing a source of adult nutrition for the wasps. To segregate parasitoids for experiments, some parasitized host pupae were removed from the culture before parasitoid eclosion and placed singly in small plastic vials. Newly emerged wasps were transferred to experimental arenas (see below). Experimental Setup Wandering L5 P. interpunctella larvae were isolated from culture, chilled for approximately 5 min at 5◦ C, and then placed in groups of 100 onto a patch which consisted of a standard-sized petri dish filled to within 5 mm of the surface with plaster of Paris. Approximately 2 g of finely milled wheatmeal was evenly spread over the patch containing hosts, and this was covered by fine nylon mesh which was firmly secured by two elastic bands. The mesh was sufficiently coarse to allow for V. canescens to probe for hosts while preventing host escape. The patches were then isolated for 24 h before being presented to parasitoids. This enabled the hosts to release copious quantities of mandibiular secretions, silk and, frass, which are known to contain kairomones that elicit probing behavior by V. canescens (Corbet, 1971; Waage, 1978). The following day, parasitoids were placed individually into “arenas” consisting of two patches (200 larvae) that were placed in a large plastic boxes and contained sufficient wheatmeal to reach the top rim of each patch. At the conclusion of each experimental period hosts were transferred from patches

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

578

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

to large glass jars containing excess food medium, and emerging adult parasitoids/moths were counted. Influence of Oviposition on Egg Maturation and Ovulation in V. canescens The aim of this study was to determine if V. canescens are able to mature (ovulate) additional eggs after attaining the maximum egg load in the lateral oviducts and following a number of oviposition experiences. A previous study reported that V. canescens attains its maximum egg load 5 to 7 days after adult eclosion (Harvey et al., 1994). Thus, 40 wasps were deprived of hosts for 5 days, then placed in groups of 5 in boxes containing two patches for 24 h. At the conclusion of this period 30 wasps were transferred singly to plastic vials, with a drop of honey smeared on the inside lid of each vial. The other 10 wasps were immediately frozen, then dissected under a stereomicroscope. The number of mature eggs present in the calyx glands and lateral oviducts was counted. This procedure was repeated over the following 3 days, with 10 wasps removed daily and dissected and their egg loads determined. A separate group of 80 wasps was used as controls. Groups of 10 wasps were dissected on days 1, 3, and 5–10 after eclosion, and their egg loads counted (as above). None of these wasps had access to hosts at any time during its life. In V. canescens, egg load is a positive function of adult wasp size (Harvey et al., 1994); therefore, we ensured that wasp size (hind tibia length) was standardized for this experiment. The mean hind tibia length of wasps in each age/treatment group was consistently between 1.59 and 1.66 mm. Lifetime Reproductive Success of V. canescens in Response to Temporal Variations in Food and Host Access To test the hypothesis that V. canescens can adjust its fecundity schedule according to host availability and food access, the following experiments were undertaken. The age-specific patterns of progeny production by V. canescens were determined by exposing wasps (n = 10 for each treatment) to hosts for different periods of time and with variable access to food until the wasp died. Temporal Host Availability Constant Host Access. Individual parasitoids were exposed to 200 hosts for 24 h, after which they were immediately transferred to another box containing two fresh patches; this procedure was repeated daily until wasp death.

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Lifetime Reproductive Success in a Parasitoid Wasp

579

Intermediate Host Access. Individual parasitoids were presented with 200 hosts for 2 h daily, then were isolated without hosts in another box (see below) for the following 22 h; this procedure was repeated daily until wasp death. Limited Host Access. Individual parasitoids were presented with 200 hosts for 0.5 h daily, then were isolated without hosts in another box (see below) for the following 23.5 h; this procedure was repeated daily until wasp death. Food Availability Constant Food Access. Regardless of host availability, wasps had continuous access to 50% honey solution absorbed into a ball of cotton wool. This procedure was repeated until wasp death. Intermediate Food Access. Parasitoids had access to a 50% honey solution (as above) for 24 h, followed by 48 h of starvation. This procedure was repeated until wasp death. Starvation. Parasitoids had no access to honey solution at any time until death. Controls The aim of this experiment was to compare the effects of variable host access on longevity in V. canescens when a source of adult nutrition is constantly available. Parasitoids were individually reared in clear plastic boxes (17.5 × 11.0 × 5.0 cm) containing approximately 4 mm of wheatmeal spread evenly over the floor of the box (as per experimental conditions) and were constantly supplied with honey. At death parasitoid longevity in days was recorded, and the size of each parasitoid was determined by measuring the hind tibia length on a calibrated stereomicroscope. The longevity of host-deprived wasps was compared with hosts in all three temporal host availability treatments. RESULTS The Influence of Host Deprivation and Oviposition Experience on Parasitoid Egg Load When V. canescens is deprived of hosts after eclosion, the number of mature eggs in the oviducts increases over several days, reaching a maximum of approximately 160 at 6 days of age, declining marginally to about 150

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

580

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

Fig. 1. Mean number of mature eggs present in the lateral oviducts of Venturia canescens at different ages in parasitoids with no host access (•) or after 24 h of host access on the fifth–sixth day after eclosion (◦). Bars represent standard errors of the means. Sample size = 10 for both treatments.

after 8 days, then remaining fairly constant up to the 10th day after emergence (Fig. 1). Females provided 200 hosts for 24 h between the fifth and the sixth day after eclosion laid up to 80% of their available eggs but repelenished their egg supply over the next 4 days, acquiring the same complement of eggs as females of corresponding ages that were not provided with hosts (Fig. 1). A two-way ANOVA for egg number, with day (6–10) and treatment (oviposition experience or host deprivation) as factors revealed that the number of eggs in the oviducts of parasitoids varied significantly with day (F = 12.264,90 , P < 0.0001) and treatment (F = 102.631,90 , P < 0.0001). There was also a significant interaction between day and treatment (F = 16.354,90 , P < 0.0001). Age-Specific Fecundity of Venturia canescens with Constant Food Access Under conditions of temporal host variability and constant food access, the age-specific fecundity of V. canescens is shown in Fig. 2. Wasps provided with unlimited host access demonstrated a heavily skewed pattern

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Fig. 2. Age-specific fecundity of Venturia canescens from eclosion provided with constant food and hosts for variable periods of time per day. Bars represent standard errors of the means. (A) One-half hour of daily host access, (B) 2 h of daily host access, and (C) = 24 h of host access. Sample size = 10 for each treatment.

581

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

582

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

of early adult reproductivity, while wasps with limited host access distributed their progeny fairly evenly throughout their adult lives. Several individual parasitoids with unlimited host access displayed a prolonged period of postreproductive survival, whereas restricting host access to parasitoids usually resulted in a decrease in the duration of postreproductive survival. When wasps were constantly supplied with hosts, reproduction began to decline at a relatively early stage in their adult lives, (between 4 and 6 days after eclosion), whereas in the medium host access group this was delayed until day 7 or 8. In the limited host access group, a noticeable reproductive decline did not occur until about the 12th day after eclosion.

Relationship Among Adult Wasp Size, Host and Food Access, and Longevity The sample variances of the longevity data for the different food/host access treatments were initially analyzed using an F-max test for homogeneity of variance. The largest and smallest variances were found to be significantly different (P < 0.05), so the data were log-transformed. A further F-max test revealed that the difference in the extreme variances of the log-transformed data was not significant (P > 0.05). Prior to the analysis of covariance, the assumption of homogeneity of slopes was tested for both longevity and progeny production. This assumption was not met for either log-transformed longevity (F = 0.718,67 , P > 0.05) or log-tranformed progeny production (F = 0.898,67 , P > 0.05). A two-way ANCOVA for parasitoid longevity, with length of host and food access as factors, revealed that longevity varied significantly with food access (F = 338.322,66 , P < 0.001) (Fig. 3). Constantly fed wasps lived considerably longer than starved wasps or those provided with limited access to food. However, longevity did not vary significantly in response to duration of host access (F = 0.252,66 , P > 0.05) (Fig. 3) and the interaction between food and host access was also not significant (F = 0.914,66 , P > 0.05). Therefore, within each feeding treatment, parasitoid longevity was largely unaffected by temporal variations in host access. Figure 3 shows the mean longevity of adult V. canescens when provided with variable temporal food and host access. Length of access to a food source (honey) clearly enables the parasitoids to extend their life span greatly. Parasitoid longevity also covaried significantly with adult wasp size (F = 4.241,66 , P < 0.05), with larger wasps living considerably longer than their smaller counterparts. Parasitoid longevity did not vary significantly with duration of host access (including controls) when food was constantly provided (one-way ANOVA, F = 0.553,36 , P > 0.05). Irrespective of temporal variations in host availability, wasps lived an average of between 23 and 27 days (Fig. 3).

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Lifetime Reproductive Success in a Parasitoid Wasp

583

Fig. 3. Mean longevity of Venturia canescens from eclosion when provided with food and hosts for variable periods of time per day (24 h). Bars represent standard errors of the means. Const, constant food access; Inter, limited food access; Starv, starvation. Filled bar, control (no host access); open bars, 0.5 h of host access; hatched bars, 2 h of host access; shaded bars, 24 h of host access. Sample size = 10 for each treatment.

Relationship Among Adult Wasp Size, Host and Food Access, and Lifetime Reproductive Success As with the longevity data, the highest and lowest variances of parasitoid fecundity data were compared across all treatments by employing an F-max test. The difference was significant (P < 0.05), therefore the data were log-transformed and variances were again compared using a second F-max test. This time, the difference was not significant (P > 0.05). A two-way ANCOVA for total wasp progeny revealed that reproductive success varied significantly with length of food access (F = 27.832,66 , P < 0.001) (Fig. 4). Constantly fed wasps had much higher realized fecundities than wasps reared with limited food or under starvation (Fig. 4). Reproductive success also varied significantly with host access (F = 6.652,66 , P > 0.05). However, there was not a significant interactive effect between length of host and food access on

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

584

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

Fig. 4. Mean lifetime reproductive success of Venturia canescens from eclosion when provided with food and hosts for different periods of time per day (24 h). Bars represent standard errors of the means. Abbreviations and symbols as in the legend to Fig. 3. Sample size = 10 for each host/food treatment.

reproductive success (F = 1.124,66 , P > 0.05). Within each feeding treatment, Tukey’s pairwise comparisons revealed that the reproductive success of parasitoids did not vary significantly with the length of daily host access (P > 0.05). In fact, comparing all treatments, the highest mean lifetime fecundity occurred where wasps provided constantly with a source of nutrition had access to hosts for only 2 h daily (as opposed to 24 h) (Fig. 4). Parasitoid reproductive success covaried significantly with adult wasp size (F = 7.09 1,66 , P < 0.001). Larger wasps generally enjoyed a higher lifetime reproductive success than smaller wasps. However, individual regression analyses revealed that this relationship was significant in the 0.5- and 24-h host (and constant food) access treatments, and not when hosts were provided for 2 h daily (Fig. 5). To test if parasitoids adjust their foraging intensity in response to differences in food availability, a series of one-way ANOVAs was performed to

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Fig. 5. Relationship between adult size (hind tibia length) in Venturia canescens and lifetime reproductive success of parasitoids with variable host and constant food access per day (24 h). (A) One-half hour of host access; (B) 2 h of host access; (C) 24 h of host access. Sample size = 10 for each treatment.

585

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

586

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

compare the cumulative progeny production of parasitoids from the different treatments over the first 2 days of life. Because the experiment in-corporated reproductive data from only the first 2 days of adult life, the feeding treatments were designated 48 h of food, availability, 24 h of food, 24 h of starvation, and 48 h of starvation. The cumulative fecundity of V. canescens did not vary significantly with access to food (F = 0.162,72 , P > 0.05) but varied significantly with daily duration of host access (F = 29.372,72 , P < 0.001). However, the interactive effect between food and host availability did not significantly influence progeny production over the days tested (F = 0.124,72 , P > 0.05). Therefore, manipulating the length of food access to parasitoids did not affect their cumulative fecundity over the first 2 days of their adult lives. Figure 6 shows that within each host access treament, progeny production was very uniform, irrespective of temporal variations in food availability.

Fig. 6. Cumulative progeny production by Venturia canescens over the first 2 days after eclosion when hosts and food are provided for variable periods of time per day (24 h). Bars represent standard errors of the means. Abbreviations and symbols as in the legend to Fig. 3. Sample size = 10 for each treatment.

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

Lifetime Reproductive Success in a Parasitoid Wasp

9:30

Style file version Feb 08, 2000

587

DISCUSSION The results of this investigation clearly demonstrate that a dynamic relationship exists between host supply and oviposition rate. Parasitoids provided with constant food were affected differently by variable periods of host access, perhaps due to reproductive, behavioral, and physiological constraints. When provided with unlimited host access, parasitoids showed a heavily skewed pattern of early adult reproduction, while wasps with limited host access distributed their progeny more evenly throughout their adult lives. Because of this, cumulative reproductive success in V. canescens was fairly uniform when wasps had constant access to food. Irrespective of variations in host access, the longevity of parasitoids constantly supplied with food was approximately constant. Consequently, wasps with constant access to hosts experienced longer periods of postreproductive survival than conspecific wasps with limited host access, because they depleted their egg primordia at an earlier point of life (see also Sahragard et al., 1991). Importantly, V. canescens was able to achieve maximum fecundity only when food was constantly available; wasps in the intermediate food access and starvation treatments always died while still in the reproductive phase of their lives. Furthermore, progeny production was positively correlated with temporal host access in these treatments. Several previous studies with other parasitoid species have also reported that temporal variations in host availability or density affect oviposition rates. Trichogramma minutum provided with excess host eggs produced significantly more progeny over the first 2 days of adult life than wasps with restricted host access, although over the remainder of reproductive life daily oviposition rates were fairly uniform in both host treatments (Bai and Smith, 1993). Similarly, Kopelman and Chabora (1992) found that when the larval endoparasitoid Leptopilina boulardi was constantly provided with hosts, they exhibited an early peak and subsequent rapid decline in progeny production, followed by a prolonged period of postreproductive survival. In contrast, the authors found that wasps with limited host access extended their reproductive periods to later periods of adult life. Work with other parasitoid species has similarly shown that adjustments in reproductive rates occurs in response to temporal host limitations (e.g., Mackauer, 1982; Sahragard et al., 1991; Drost and Carde, 1992). Although V. canescens produced most progeny early in adult life when they were constantly provided with hosts, the rate of oviposition per unit of time was actually greatest when host access was limited to 30 min per day. This can be explained by closely examining the reproductive biology of the parasitoid in response to temporal variations in host availability. At eclosion, the oviducts of V. canescens contain 50–60 ovulated (=mature) eggs which

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

588

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

are ready to oviposit (Fletcher et al., 1994; Harvey et al., 1994). Trudeau and Gordon (1989) monitored the egg maturation rate in V. canescens at 25◦ C and found that wasps mature an average of almost 2 eggs per h (or about 40–50 per day). In our experiments, several parasitoids produced over 60 progeny on their first day of adult life, whereas V. canescens can lay up to 50 eggs per h if suitable hosts are available (personal observations). Parasitoids thus appeared to oviposit at their physiological limit, whether or not food was available. Consequently, after several hours of intensive foraging on the first day of adult life, the oviducts of V. canescens ostensibly became egg depleted and the parasitoid rested until it had accumulated further eggs. After this time, the reproductive output of the parasitoid was effectively constrained by the rate at which further eggs were ovulated, passing from the ovaries through the calyxes and into the oviducts. The effect of eggload on oviposition rate in V. canescens was also clearly demonstrated by comparing progeny production in host-deprived wasps and wasps presented with hosts from eclosion. Wasps presented with hosts for 24 h after 5 days of host deprivation produced an average of 120 progeny, whereas the most offspring produced by any parasitoid over the same time when provided with hosts from eclosion was 74. The number of progeny produced by V. canescens over the first 48 h of life did not vary significantly with feeding treatment. In contrast, most optimality models generally predict that female parasitoids should accept lower-quality hosts into their diet or increase their searching rate for hosts when the risk of mortality is high or environmental conditions are otherwise unfavourable (Mangel, 1987; Roitberg et al., 1992; Sirot and Bernstein, 1996; Sirot et al., 1997). Changes in the physiological state of insects alters the perception of external cues or reduces thresholds for particular responses (Fitt, 1986; Fletcher et al., 1994; Hughes et al., 1994). Therefore, starved wasps should have produced more progeny over the first 2 days of adult life than well-fed wasps because starved wasps faced a far greater risk of instantaneous mortality. However, as stipulated above, V. canescens appeared to oviposit at its physiological limit when excess hosts were available until egg supply became a limiting factor. Furthermore, starvation quickly depletes parasitoids of metabolic energy, which might compensate for any early increase in foraging intensity. The above models are probably more relevant for examining individual oviposition decisions than for comparing long-term reproductive output. For example, Fletcher et al. (1994) found that starved V. canescens more readily superparasitized hosts than well-fed wasps of the same age. Our experiments have confirmed that V. canescens has a considerably higher reproductive potential than has been suggested previously, based on actual measured progeny production (Ahmed, 1936) or on dissections of

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

Lifetime Reproductive Success in a Parasitoid Wasp

9:30

Style file version Feb 08, 2000

589

the oviducts when the egg storage capacity has been attained (Trudeau and Gordon, 1989; Harvey et al., 1994; Harvey and Thompson, 1995). The latter method of estimating fecundity is frequently seen in the literature with other parasitoid species (e.g., Waage and Ng, 1984; Hopper, 1986; Corrigan and Lashomb, 1990; Volkl and Mackauer, 1990; Harvey et al., 2000). However, for synovigenic parasitoids, including V. canescens, ovary dissections reveal only short-term egg production, and not fecundity. In this study several parasitoids produced over 400 progeny during their lifetimes, whereas the reproductive success of V. canescens in Ahmed’s (1936) study never exceeded 75. More importantly, our results may still represent a profound understimate of the reproductive potential in V. canescens, because superparasitism was not recorded and hosts suffered over 30% mortality during the course of the investigation. Therefore, it is not inconceivable that some parasitoids could produce over 600 progeny under optimal conditions. This figure represents 10–15 times the number of mature eggs in the oviducts at eclosion, and even 4 or more times the maximum egg load in V. canescens. These results should be borne in mind in future investigations examining reproductive success and oviposition behavior in synovigenic–hydropic parasitoids. Irrespective of treatment, parasitoids provided constantly with food but deprived of hosts (=controls) did not live significantly longer than parasitoids provided with food and hosts, which contradicts model predictions of tradeoffs between life-history characters (Reznick, 1985; Bell and Koufpanou, 1986; Lessells, 1991). Evidence of trade-offs between fecundity and survival in other insects has been supported in some studies (e.g., Hohmann et al., 1989; Roitberg, 1989; Orr and Boethel, 1990) but not in others (e.g., Bai and Smith, 1993). Models of resource acquisition by van Noordwijk and de Jong (1986) and de Jong and van Noordwijk (1992) suggest that positive or negative correlations between life-history traits depends largely upon variations in the fraction of resources that are allocated for different functions. When conditions are favorable, expected negative correlations between reproduction and survival may not occur because the organism is able to obtain sufficient resources from the environment, thus balancing the concomitant drain from reproduction (Bell and Koufpanou, 1986). Wasps used in this study were kept in confined arenas with ready access to hosts and food, thus reducing the metabolic costs associated with searching for these spatially separated resources which would be expected to occur under natural conditions. Studies showing positive relationships between reproduction and survival are in fact producing phenotypic correlations rather than tradeoffs (Lessells, 1991). Where phenotypic correlations are observed in measuring relationships between life-history characters in parasitoids, there may be other factors (besides food availability) which limit adult life span under conditions of host deprivation. Many synovigenic parasitoids produce

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

590

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

considerable numbers of hydropic eggs which accumulate in the oviducts pending oviposition (Flanders, 1942, 1950). However, the metabolic costs of egg maturation and storage are poorly understood. Parasitoids with constant access to hosts carry proportionately lighter egg complements than hostdeprived parasitoids, which might reduce the costs of foraging in the former group. In V. canescens we found that large wasps typically had longer life spans and produced more progeny than small wasps. Several studies with other parasitoids have also reported positive correlations between adult body size and longevity (Waage and Ng, 1984; Bellows, 1985; Heinz and Parella, 1990; Hardy et al., 1992) and fecundity (Bellows, 1985; Heinz and Parella, 1990; Nakamura, 1995). However, as Visser (1994) points out, in the field, size may affect other factors influencing fitness, such as host searching and dispersal efficiency, which are difficult to measure in the laboratory. Where hosts are spatially separated, metabolic costs associated with activity may increase, thereby benefiting large individuals which have greater storage sites than smaller individuals. In future laboratory studies, the relationship between size and fitness may be determined only where food and host access are manipulated, as we have demonstrated with V. canescens. This study has reported that lifetime reproductive success in an asexual strain of V. canescens is affected by temporal variations in host availability and that manipulating food access also markedly influences the reproductive success of the parasitoid. We suggest that the starvation or intermediate food access treatments are more indicative of conditions faced by V. canescens in the field, because sources of adult nutrition are likely to be located considerable distances from host patches. Costs imposed by searching for spatially separated hosts may be especially high for sexual strains of V. canescens in southern Europe which apparently attack hosts occuring singly in fruits, such as figs (Driessen and Bernstein, 1996). However, although wasps found in flour mills may benefit from the aggregated distribution of their hosts, the slow rate of egg maturation in V. canescens suggests that different strains of the parasitoid utilize only a fraction of their reproductive potential. If it is true that suitable hosts and food sources for parasitoids are heterogeneously distributed (Price, 1980), then future experiments should be designed that limit or interrupt host and food access to parasitoids, more closely approximating the patchy and ephemeral nature of these resources in the field. Such experiments are particularly relevant for proovigenic parasitoids, which may deplete their egg complements rapidly (Kopelman and Chabora, 1992), or parasitoids attacking highly dispersed, scarce, or concealed hosts (Jervis and Kidd, 1986). Further studies examining reproductive success in parasitoids across a wide range of parasitoid species in the laboratory and under natural conditions offer considerable promise in our

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

9:30

Lifetime Reproductive Success in a Parasitoid Wasp

Style file version Feb 08, 2000

591

understanding of trade-offs between life-history characters. Moreover, these investigations will undoubtedly help to elucidate the importance of functional constraints on parasitoid performance and will provide greater insights into life-history evolution in parasitoid wasps. ACKNOWLEDGMENTS We wish to thank Mark Jervis for reading an early draft of the manuscript, two anonymous referees for their comments on a later draft, and Mike Begon, Tom Tregenza, Steve Sait, and Louise Vet for discussion. The work was funded by NERC Grant GT4/90/TLS/72 to J.H. REFERENCES Ahmed, T. (1936). The influence of ecological factors on the Mediterranean flour moth, Ephestia kuehniella, and its parasite, Nemeritis canescens. J. Anim. Ecol. 5: 67–93. Allen, W. W., and Smith, R. F. (1958). Some factors influencing the efficiency of Apanteles medicaginis Muesbeck as a parasite of the alfalfa caterpillar, Colias philodice eurytheme Boisduval. Hilgardia 28: 1–42. Bai, B., and Smith, S. M. (1993). Effect of host availability on reproduction and survival of the parasitoid wasp, Trichogramma minutum. Ecol. Entomol. 18: 279–286. Beling, I. (1932). Zur biologie von Nemeritis canescens Grav. (Hymenoptera: Ophion.). 1. Zuchtung-serfahrungen und okologische beobachtungen. Z. Ang. Entomol. 19: 223–249. Bell, G., and Koufpanou, V. (1986). The cost of reproduction. Oxford Surv. Evol. Biol. 3: 83–131. Bellows, T. S. (1985). Effects of host age and host availability on developmental period, adult size, sex ratio, longevity and fecundity in Lariophagus distinguendis Forster (Hymenoptera: Pteromalidae). Res. Popul. Ecol. 27: 55–64. Charnov, E. L. (1979). The genetical evolution of patterns of sexuality: Darwinian fitness. Am. Nat. 113: 465–480. Charnov, E. L. (1982). The Theory of Sex Allocation, Princeton University Press, Princeton, NJ. Charnov, E. L., and Skinner, S. W. (1985). Complementary approaches to the understanding of parasitoid oviposition decisions. Environ. Entomol. 14: 383–391. Corbet, S. A. (1971). Mandibular gland secretion of larvae of the flour moth, Anagasta kuehniella, contains an epideitic pheromone and elicits oviposition movement in a hymenopteran parasite. Nature 232: 481. Corrigan, J. E., and Lashomb, J. H. (1990). Host influences on the bionomics of Edovum puttleri (Hymenoptera: Eulophidae): Effects on size and reproduction. Environ. Entomol. 19: 1496– 1502. de Jong, G., and van Noordwijk, A. J. (1992). Acquisition and allocation of resources: Genetic (co) variances, selection and life histories. Am. Nat. 139: 749–770. Driessen, G., and Bernstein, C. (1996). Patch departure mechanisms and optimal host exploitation in an insect parasitoid. J. Anim. Ecol. 68: 445–459. Drost, Y. C., and Carde, R. T. (1992). Influence of host deprivation on egg load and oviposition behaviour of Brachymeria intermedia, a parasitoid of gypsy moth. Physiol. Entomol. 17: 230–234. Ellers, J. E., van Alphen, J. J. M., and Sevenster, J. G. (1998). A field study of the size-fitness relationships in the parasitoid, Asobara tabida. J. Anim. Ecol. 67: 318–324. Ellers, J. E., Bax, M., and van Alphen, J. J. M. (2001). Seasonal changes in female size and its relation to reproduction in the parasitoid Asobara tabida. Oikos 92: 309–314. Fitt, G. P. (1986). The influence of a shortage of hosts on specificity of oviposition behavior in species of Dacus (Diptera: Tephritidae). Physiol. Entomol. 11: 133–144.

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

592

PP259-345022

September 20, 2001

9:30

Style file version Feb 08, 2000

Harvey, Harvey, and Thompson

Flanders, S. E. (1942). Oosorption and ovulation in relation to oviposition in the parasitic Hymenoptera. Ann. Entomol. Soc. Am. 35: 251–266. Flanders, S. E. (1950). Regulation of ovulation and egg disposal in the parasitic Hymenoptera. Can. Entomol. 82: 134–140. Fletcher, J. P., Hughes, J. P., and Harvey, I. F. (1994). Mortality risk and egg load affect oviposition behaviour of a solitary parasitoid. Proc. Roy. Soc. B 258: 163–167. Godfray, H. C. J. (1994). Parasitoids: Behavioral and Evolutionary Ecology, Princeton University Press, Princeton, NJ. Hardy, I. C. W., Griffiths, N. T., and Godfray, H. C. J. (1992). Clutch size in a parasitoid wasp: A manipulation experiment. J. Anim. Ecol. 61: 121–129. Harvey, J. A. (1995). The Nutritional Ecology of a Host-Parasitoid Interaction, Unpublished Ph.D. thesis, University of Liverpool, Liverpool, U.K. Harvey, J. A., and Thompson, D. J. (1995). Developmental interactions between the solitary endoparasitoid, Venturia canescens (Hymenoptera: Ichneumonidae), and two of its hosts, Plodia interpunctella and Corcyra cephalonica (Lepidoptera: Pyralidae). Eur. J. Entomol. 92: 427–435. Harvey, J. A., Harvey, I. F., and Thompson, D. J. (1994). Flexible larval growth allows use of a range of host sizes by a parasitoid wasp. Ecology 75: 1420–1428. Harvey, J. A., Vet, L. E. M., Jiang, N., and Gols, G. J. Z. (1998). Nutritional ecology of the interaction between larvae of the gregarious ectoparasitoid, Muscidifurax raptorellus (Hymenoptera: Pteromalidae), and their pupal host, Musca domestica (Diptera: Muscidae). Physiol. Entomol. 23: 113–120. Harvey, J. A., Kadash, K., and Strand, M. R. (2000). Differences in larval feeding behavior correlate with altered developmental strategies in parasitoid wasps: Implications for the size-fitness hypothesis. Oikos 88: 621–629. Heimpel, G. E., Rosenheim, J. A., and Mangel, M. (1996). Egg limitation, host quality, and dynamic behavior by a parasitoid in the field. Ecology 77: 2410–2420. Heimpel, G. E., Rosenheim, J. A., and Kattari, D. (1997). Adult feeding and lifetime reproductive success in the parasitoid, Aphytis melinus. Entomol. Exp. Appl. 83: 305–315. Heimpel, G. E., Mangel, M., and Rosenheim, J. (1998). Effects of time limitation and egg limitation on lifetime reproductive success of a parasitoid in the field. Am. Nat. 152: 273–289. Heinz, K. M., and Parella, M. P. (1990). Holarctic distribution of the leafminer parasitoid Diglyphus begini (Hymenoptera: Eulophidae) and notes on its life history attacking Liriomyza trifolii (Diptera: Agromyzidae) in chrysanthemum. Ann. Entomol. Soc. Am. 83: 916– 924. Hohmann, C. L., Luck, R. F., Oatman, E. R., and Platner, G. R. (1989). Effects of different biological factors on longevity and fecundity of Trichogramma platneri Nagarkatti (Hymenoptera: Trichogrammatidae). An. Soc. Entomol. Bras. 18: 61–70. Hopper, K. R. (1986). Preference, acceptance, and fitness components of Microplitis croceipes (Hymenoptera: Braconidae) attacking various instars of Heliothis virescens (Lepidoptera: Noctuidae). Environ. Entomol. 13: 1145–1150. Hurlbutt, B. L. (1987). Sexual size dimorphism in parasitoid wasps. Biol. J. Linn. Soc. 30: 63–89. Jervis, M. A., and Kidd, N. A. C. (1986). Host-feeding strategies in hymenopteran parasitoids. Biol. Rev. 61: 395–434. Jervis, M. A., Kidd, N. A. C., Fitton, M. G., Huddleston, T., and Dawah, H. A. (1993). Flower visiting by hymenopteran parasitoids. J. Nat. Hist. 27: 67–105. Kazmer, D. J., and Luck, R. F. (1995). Field tests of the size-fitness hypothesis in the egg parasitoid, Trichogramma pretiosum. Ecology 76: 412–425. Kirkwood, T. B. L. (1981). Repair and its evolution: Survival versus reproduction. In Townsend, C. R., and Calow, P. (eds.), Physiological Ecology: An Evolutionary Approach to Resource Use, Sinauer Press, Sunderland, MA, pp. 165–189. Kopelman, A. H., and Chabora, P. C. (1982). Resource variability and life-hsitory parameters of Leptopilina boulardi (Hymenoptera: Eucoilidae). Ann. Entomol. Soc. Am. 85: 195–199. Lessells, C. M. (1991). The evolution of life-histories. In Davies, N. B., and Krebs, J. R. (eds.), Behavioural Ecology: An Evolutionary Approach, Blackwell Scientific, Oxford, pp. 32–68.

P1: Vendor/GCP/GGT

P2: GDX/GCR

Journal of Insect Behavior [joib]

PP259-345022

September 20, 2001

Lifetime Reproductive Success in a Parasitoid Wasp

9:30

Style file version Feb 08, 2000

593

Mackauer, M. (1982). Fecundity and host utilization of the aphid parasite Aphelinus semiflavus (Hymenoptera: Aphelinidae) at two host densities. Can. Entomol. 114: 721–726. Mangel, M. (1987). Oviposition site selection and clutch size in insects. J. Math. Biol. 25: 1–22. Nakamura, S. (1995). Optimal clutch size for maximising reproductive success in a parasitoid fly, Exorista japonica. Appl. Entomol. Zool. 30: 425–431. Orr, D. B., and Boethel, D. J. (1990). Reproductive potential of Telenomus cristatus and T. podisi (Hymenoptera: Scelionidae), two egg parasitoids of pentatomids (Heteroptera). Ann. Entomol. Soc. Am. 83: 902–905. Petersen, G., and Hardy, I. C. W. (1996). The importance of being larger: Parasitoid intruderowner contests and their implications for clutch size. Anim. Behav. 51: 1363–1373. Price, P. W. (1980). Evolutionary Biology of Parasites, Princeton University Press, Princeton, NJ. Reznick, D. (1985). Costs of reproduction: an evaluation of the empirical evidence. Oikos 44: 257–267. Richards, O. W., and Thomson, W. S. (1932). A contribution to the study of the genera Ephestia (including Strymax Dyer) and Plodia Gn. (Lepidoptera: Phycitidae), with notes on parasites of the larvae. Trans. R. Entomol. Soc. Lond. 80: 169–248. Rogers, D. J. (1972). The ichneumon wasp Venturia canescens: Oviposition and avoidance of superparasitism. Entomol. Exp. Appl. 15: 190–194. Roitberg, B. D. (1989). The cost of reproduction in rosehip flies, Rhagoletis basiola: Eggs are time. Evol. Ecol. 3: 183–188. Roitberg, B. D., Mangel, M., Lalonde, R. G., van Alphen, J. J. M., and Vet, L. E. M. (1992). Seasonal dynamic shifts in patch exploitation by parasitic wasps. Behav. Ecol. 3: 156–165. Rosenheim, J. A. (1996). An evolutionary argument for egg limitation. Evolution 50: 2089–2094. Sahragard, A., Jervis, M. A., and Kidd, N. A. C. (1991). Influence of host availability on rates of oviposition and host-feeding, and on longevity in Dicondylus indianus Olmi (Hymenoptera: Dryinidae), a parasitoid of rice brown planthopper Nilaparvata lugens Stal (Hemiptera: Delphacidae). J. Appl. Entomol. 112: 153–162. Salt, G. (1976). The hosts of Nemeritis canescens, a problem in the host specificity of insect parasitoids. Ecol. Entomol. 1: 63–67. Sirot, E., and Bernstein, C. (1996). Time sharing between host searching and food searching in parasitoids: State-dependent optimal strategies. Behav. Ecol. 7: 189–194. Sirot, E., Ploye, H., and Bernstein, C. (1997). State-dependent superparasitism in a solitary parasitoid: Egg load and survival. Behav. Ecol. 8: 226–232. Syme, P. D. (1975). The effects of flowers on the longevity and fecundity of two native parasites of the European pine shoot moth in Ontario. Environ. Entomol. 4: 337–346. Trudeau, D., and Gordon, D. M. (1989). Factors determining the functional response of the parasitoid, Venturia canescens. Entomol. Exp. Appl. 50: 3–6. Ueno, T. (1999). Host-feeding and acceptance by a parasitic wasp (Hymenoptera: Ichneumonidae) as influenced by egg load and experience in a patch. Evol. Ecol. 13: 33–44. van Noordwijk, A. J., and de Jong, G. (1986). Acquisition and allocation of resources: Their influence on variation in life history tactics. Am. Nat. 128: 137–142. Visser, M. E. (1994). The importance of being large: The relationship between size and fitness in females of the parasitoid Aphaereta minuta (Hymenoptera: Braconidae). J. Anim. Ecol. 63: 963–978. Volkl, W., and Mackauer, M. (1990). Age-specific pattern of host discrimination by the aphid parasitoid, Ephedrus californicus Baker (Hymenoptera: Aphidiidae). Can. Entomol. 122: 349–361. Waage, J. K. (1978). Arrestment responses of the parasitoid, Nemeritis canescens, to a contact chemical produced by its host. Physiol. Entomol. 3: 135–146. Waage, J. K., and Ng, S. M. (1984). The reproductive strategy of a parasitic wasp. I. Optimal progeny and sex allocation in Trichogramma evanescens. J. Anim. Ecol. 53: 401–416.