Influence of Clavicipitaceous Endophyte Infection in Ryegrass on Development of the ParasitoidMicroctonus hyperodaeLoan (Hymenoptera: Braconidae) inListronotus bonariensis(Kuschel) (Coleoptera: Curculionidae)

BIOLOGICAL CONTROL ARTICLE NO.

7, 281–287 (1996)

0095

Influence of Clavicipitaceous Endophyte Infection in Ryegrass on Development of the Parasitoid Microctonus hyperodae Loan (Hymenoptera: Braconidae) in Listronotus bonariensis (Kuschel) (Coleoptera: Curculionidae) GARY M. BARKER1

AND

PAUL J. ADDISON

New Zealand Pastoral Agriculture Research Institute, Ruakura Agricultural Research Centre, Private Bag 3123, Hamilton, New Zealand Received August 28, 1995; accepted July 18, 1996

Laboratory experiments were conducted to examine the effect of ryegrass infection by the endophytic fungus Acremonium lolii Latch, Christensen and Samuels on Microctonus hyperodae Loan, a parasitoid of Listronotus bonariensis (Kuschel). Progression of parasitoids through the larval instar stages was shown to depend on adequate nutrition of the weevil host. Compared to confinement on endophyte-free ryegrass, parasitized weevils held on nonpreferred diets comprising leaf segments from endophyte-infected ryegrass and switchgrass contained parasitoid larvae with retarded development. Similarly, development of parasitoid larvae was retarded in hosts feeding on artificial diet containing diterpenes and alkaloids of A. lolii origin. Several diterpenes incorporated into the diet reduced survival of the parasitoid larvae. Attack rate of parasitoids was reduced when the quality of potential host weevils was compromised by confinement on nonpreferred A. lolii-infected ryegrass or without food for 14 days. r 1996 Academic Press, Inc. KEY WORDS: Argentine stem weevil; Listronotus bonariensis; Microctonus hyperodae; Acremonium lolii; endophyte.

INTRODUCTION

Listronotus bonariensis (Kuschel) (Coleoptera: Curculionidae) is a significant pest of forage grasses in New1 Zealand. The larvae of this weevil bore the stems and crowns, compromising grass persistence and productivity, while the adult stage can reduce seedling establishment by defoliation and cutting of the stems (Barker et al., 1984; Prestidge et al., 1991). Management of L. bonariensis in New Zealand pastures has, for the past 15 years, been largely dependent on use of resistant 1 Current address: Landcare Research, Private Bag 3127, Hamilton, New Zealand.

ryegrass (Lolium spp.) cultivars based on the clavicipitaceous endophytic fungus Acremonium lolii Latch, Christensen and Samuels (Barker et al., 1990; Prestidge and Ball, 1993; Prestidge et al., 1994). Resistance is conferred on infected ryegrass by the production of indole diterpenes and loline, ergot, and pyrollopyrazine alkaloids (Popay and Rowan, 1993; Prestidge and Ball, 1993; Rowan and Latch, 1994). Unfortunately, such resistance can produce toxicosis in livestock (Prestidge, 1993) and considerable research is currently directed at developing endophytic ryegrass cultivars that do not cause stock health problems but retain insect resistance. In recognition of the need for an integrated approach to management of L. bonariensis, the parasitoid Microctonus hyperodae Loan (Hymenoptera: Braconidae) was imported from South America and established in pastures throughout New Zealand (Goldson et al., 1990, 1993). The success of the parasitoid as a control agent may ultimately depend on its adaptation to the varied environmental and management conditions to be encountered on farms, including the widespread use of A. lolii-infected ryegrass. Chemical information from both the plant and the herbivorous insect plays a key role in the foraging and host acceptance behaviors of hymenopteran parasitoids (Doutt, 1959). There is increasing attention to the impact of host plant quality in biological control of herbivorous insects in agriculture (e.g., Dicke and Sabelis, 1992; Hare, 1992; Vet and Dicke, 1992; Walde, 1995). Through the production of various metabolites (Dahlman et al., 1991; Porter, 1994; Rowan and Latch, 1994), clavicipitaceous endophytic fungi effect significant changes in the chemistry of infected plants and hence their quality as hosts for herbivorous insects. Endophytic fungi therefore represent a further trophic influence in the interaction of plants, herbivorous insects, and parasitoids. This paper reports on experimental investigations on the influence of A. lolii infection in

281

1049-9644/96 $18.00 Copyright r 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

282

BARKER AND ADDISON

the ryegrass host of L. bonariensis on the ability of the weevil to sustain development of M. hyperodae.

L.) was included in Experiments 1 and 2 as it was shown earlier (Firth et al., 1993) to be a nonpreferred host for the weevil.

MATERIALS AND METHODS

A series of five experiments were undertaken. The adult L. bonariensis used were collected by sweep net and suction sampling of A. lolii-free perennial ryegrass (L. perenne L.) pastures near Hamilton, New Zealand and were stored at 4°C for a maximum of 3 days prior to use. The dates of these collections and the physiological states of the populations are given in Table 1. Dissection of subsamples (n 5 65–85) of the weevils collected from the field confirmed the absence of natural parasitism. The M. hyperodae adults used for the experiments were obtained from a laboratory colony of mixed ecotypic origin (Goldson et al., 1990) maintained on L. bonariensis. The parasitoids were collected from laboratory rearing cages within 2 h of emergence from pupae and given 4-h access to wicks saturated with 20% honey solution prior to confinement with the weevils. Ryegrass plants used in the experiments were grown from seed in soil-filled pots under glasshouse conditions. The endophyte status of each ryegrass plant was confirmed by staining and microscopic examination of tiller sheath tissues. Switchgrass (Panicum virgatum TABLE 1 Dates of Collection of Listronotus bonariensis Adults Used in Experimentation and Their Physiological Condition at the Time of Collection Experiment 1

2

3

4

5

Collection date 05 March 1992

Physiological status

Mixture of first and second generation weevils: 76% of population reproductively spent; remainder of population prereproductive, onset of oligopause. 09 February 1993 First generation weevils, in latter phase of reproductive activity: 65% of population sexually active, with females carrying eggs or mature oocytes; balance of population reproductively spent. 03 March 1993 Mixture of first and second generation weevils; 87% of population reproductively spent or undergoing oligopause; remainder prereproductive. 24 January 1994 First generation weevils, in latter phase of reproductive activity: 82% of population sexually active, with females carrying eggs or mature oocytes; balance of population reproductively spent. 20 September 1993 Overwintered weevils: entire population sexually active, with all females carrying eggs or mature oocytes.

Experiment 1 On 6 March 1992, under ambient laboratory conditions (18–25°C), groups of 150 L. bonariensis were each confined with five M. hyperodae in transparent polycarbonate, mesh-covered boxes 220 3 130 3 75 mm deep and provided with bouquets of A. lolii-free annual ryegrass (L. multiflorum Lam.) cv. ‘‘Concord.’’ After 72 h exposure to parasitoids, the weevils were pooled. Seven replicate groups of 30 weevils were then randomly assigned to three diet treatments comprising leaf segments from (i) A. lolii-free perennial ryegrass cv. ‘‘Ellett,’’ (ii) A. lolii-infected perennial ryegrass cv. ‘‘Ellett,’’ and (iii) switchgrass. Two replicate groups of weevils from each treatment were sacrificed to dissection to determine base levels of parasitism. The remaining five replicate groups of weevils were held on the appropriate leaf segment diet, in 150-cm3 plastic cages covered with mesh and lined with moistened paper, at 20°C and a 16:8-h light:dark photoperiod regime. At 2to 5-day intervals dead weevils and emergent parasitoid prepupae were counted and removed from the cages, and the surviving weevils were supplied with fresh grass leaf segments harvested from the appropriate plant type. Experiments 2 and 3 On 9 February (Experiment 2) and again on 4 March 1993 (Experiment 3) 180 L. bonariensis were confined to each of replicate transparent polycarbonate, meshcovered boxes 220 3 130 3 75 mm deep, with bouquets of A. lolii-free annual ryegrass. Adult M. hyperodae were introduced to one-half of the replicate weevil groups at the rate of eight/box. These weevil groups, with and without parasitoids, were then maintained in the screenhouse (16–31°C). After 96 h, the weevils were transferred to petri dishes lined with moistened filter paper and maintained under ambient laboratory conditions (Experiment 2, 12–23°C, 13.5:9.5 h light:dark; Experiment 3, 14–22°C, 12.5:11.5 h light:dark) on one of several diet treatments. In Experiment 2, 10 replicate groups of 12 weevils were provided with leaf segments harvested from A. lolii-free perennial ryegrass cv. ‘‘Yatsyn’’ plants, while the same number of weevils were provided with leaf segments from A. lolii-infected perennial ryegrass cv. ‘‘Yatsyn’’ plants. For Experiment 3, 11 replicate groups of 10 weevils were provided with leaf segments from (i) A. lolii-free perennial ryegrass cv. ‘‘Yatsyn,’’ (ii) A. lolii-infected perennial ryegrass cv. ‘‘Yatsyn,’’ or (iii) switchgrass. At 2- to 4-day intervals dead weevils were counted and removed from the petri dishes, and the survivors were supplied with fresh grass leaf segments.

283

Microctonus hyperodae DEVELOPMENT IN Listronotus

In the case of Experiment 3, the extent of weevil feeding on leaf segments within the first 24 h of introducing fresh grass was assessed on four occasions by scoring the leaf area removed on a 0–10 scale (0 5 no feeding, 10 5 ca. 250-mm2 leaf area consumed). After 21 and 12 days for Experiments 2 and 3, respectively, the weevils remaining alive were dissected to determine the incidence of parasitism and the degree of development of the parasitoid larvae. The instar stages of the parasitoid larvae were determined following Loan and Holdaway (1961), but recognizing that instars 1 to 4 occur in M. hyperodae. Experiment 4 Commencing 25 January 1994, 30 replicated groups of 80 L. bonariensis were maintained in transparent polycarbonate, mesh-covered boxes, 220 3 130 3 75 mm deep. These were held in a controlled environment chamber at 20°C, 70% relative humidity, and a 16:8-h light:dark photoperiod. Ten replicate groups were provided every 2 days with 50 leaf segments, 60 mm long, freshly harvested from A. lolii-infected perennial ryegrass, while a further 10 groups were similarly provided with leaf segments from A. lolii-free ryegrass plants. The remaining 10 groups did not receive food. At each change of ryegrass in the boxes, the intensity of weevil feeding on the ryegrass was assessed by visually scoring 10 randomly selected leaf segments on a 0–10 scale as above. After 14 days, five replicate groups of weevils from each treatment were sacrificed by dissection. For each dissected female, the degree of vitellogenesis was scored on a scale of 0–5 (Barker et al., 1988) (0 5 vitellarium undeveloped, no oocytes present; 1 5 small number of oocytes in apical vitellaria; 5 5 with numerous oocytes, including terminal oocytes in the lower vitellaria). The incidence of oocyte resorption, evidenced by b-carotene crystalline inclusions in the ovariolar pedicels (Goldson, 1983; Barker et al., 1988), was also noted. All remaining weevil groups were transferred to clean boxes and provided with leaf segments of annual ryegrass. M. hyperodae were introduced to each replicate weevil group at the rate of eight/box. The parasitoids were maintained in the boxes for 48 h, after which the weevils were recovered and dissected to determine the incidence of parasitism. The degree of vitellarium development and presence of oocyte resorption was noted for each dissected female weevil as above. Experiment 5 On 22 September 1993, 180 L. bonariensis were confined to replicate transparent polycarbonate, meshcovered boxes 220 3 130 3 75 mm deep, with bouquets of A. lolii-free annual ryegrass. Adult M. hyperodae were introduced to each replicate weevil group at the rate of eight/box. These weevil groups, with parasi-

toids, were then maintained in the screenhouse (12– 23°C). After 72 h, the weevils were pooled. A subsample of 50 weevils was dissected to determine the initial parasitism rate. The remaining weevils were randomly allocated to one of 10 diet treatments in 5 replicate groups of 10. Each replicate group of weevils was maintained on two 10-mm-diameter plugs of artificial diet in petri dishes, lined with moistened filter paper, in a controlled environment chamber at 20°C, 70% relative humidity, and 16:8-h light:dark photoperiod regime. At 2-day intervals the number of dead weevils and parasitoid prepupae/pupae were counted and removed from the petri dishes. At each assessment date, the intensity of feeding on the diet plugs by the weevils was scored on a scale of 0–10 (0 5 no feeding; 10 5 extensive feeding) before the plugs in each dish were replaced with two fresh diet plugs. The diets were prepared following Malone and Wigley (1990). Various indole diterpenes (lolitrem B, lolitriol, a-paxitriol, b-paxitriol, paxilline), ergot alkaloids (ergovaline, ergotamine, ergonovine) and the pyrollopyrazine alkaloid peramine, dissolved in dichloromethane, were added at 2 µg g21 dry weight to the molten diet at 60°C prior to pouring into petri dishes. The control diet contained the dichloromethane (1.83 ml per 100 g weight) but no diterpene or alkaloid. Once solidified the diets were stored at 4°C until required, when 10-mmdiameter plugs were cut. STATISTICAL ANALYSES

All data on percentage mortality and parasitism and scores of feeding were subjected to analysis of variance, with arcsine transformation. Experiments 1, 4, and 5 were completely randomized designs. In Experiments 2 and 3, presence or absence of parasitism was considered a main-plot effect and diet as a subplot effect in a randomized complete block design. In these two experiments, host weevil diet effects on parasitoid larval instar distributions were examined by x2 analyses. Regression analyses of percentage cumulative emergence against time, using logit link transformation, were employed to calculate median dates of parasitoid larval emergence in Experiments 1 and 5. All analyses were performed using the SAS statistical package (SAS, 1982). RESULTS

The levels of parasitism in the weevils at the commencement of Experiment 1 were low, averaging 14.5% (Table 2). High mortality occurred in the aged, reproductively spent weevils during the experiment. Mortality not attributed to parasitoid emergence did not differ

284

BARKER AND ADDISON

TABLE 2

TABLE 3

Mortality in Listronotus bonariensis and Yield of Microctonus hyperodae Prepupae after Maintenance of the Weevils on Three Grass Diets

Mortality of Listronotus bonariensis and Occurrence of Various Developmental Stages of Their Endoparasitoid Microctonus hyperodae after Maintenance of the Weevils on Three Grass Diets

Weevil diet Weevil diet A. lolii-free A. lolii-infected Switchryegrass ryegrass grass Percentage initial weevil parasitism Percentage weevil mortality a Percentage weevils yielding parasitoid prepupae Percentage initial weevils yielding parasitoids a

14.8 56

15.0 52

14.8 77

27

31

0

12

15

0

Mortality without yielding parasitoid prepupae.

A. lolii-free ryegrass

A. lolii-infected ryegrass

12 14 64

14 13 73

25

63

75

37

Percentage weevil mortality Nonparasitized a Parasitized b Percentage weevil parasitism b,c Parasitoid instar developmentb,c,d Percentage larvae in 1st instar Percentage larvae in 3rd and 4th instars a

Weevils without exposure to M. hyperodae. Weevils exposed to M. hyperodae prior to being maintained on the grass diets. c Parasitism in surviving weevils after 21 days on the grass diets. d No parasitoids present as 2nd instars. b

between weevils on A. lolii-free and A. lolii-infected ryegrass diets but was higher in those on switchgrass diet (F2,12 5 7.06, P , 0.01) (Table 2). Observations indicated that the weevils fed little on switchgrass. Parasitoid prepupae emergence from weevils on the ryegrass diets began on 26 March and was complete by 10 April, 35 days after first exposure to adult parasitoids. The A. lolii status of the ryegrass diet had no influence on the percentage of weevils yielding parasitoid prepupae (F1,8 5 1.52, P . 0.05) (Table 2). The rate of development of the parasitoid larvae was reduced, however, with the estimated median number of days to parasitoid emergence from weevils being 23.1 6 1.6 SE and 30.0 6 2.4 days for A. lolii-free and A. lolii-infected diet treatments, respectively. In contrast, no weevils on the switchgrass diet yielded parasitoids (Table 2). In Experiment 2 the mortality of weevils during the 21-day experimental period was low and was influenced neither by the endophyte status of the ryegrass diet (F1,27 5 0.68, P . 0.05) nor by parasitism (F1,27 5 1.37, P . 0.05) (Table 3). The incidence of parasitism in the weevils at the termination of the experiment was similarly not affected by the diet on which the insects were fed (F1,27 5 2.60, P . 0.05). However, the parasitoid instar distribution in the dissected weevils (Table 3) indicated that parasitoid development was considerably retarded in weevils fed leaf segments from A. lolii-infected ryegrass, with a higher proportion of larvae in the first instar (n 5 182, x2 5 51.318, df 5 2, P , 0.001). Mortality among weevils in the 12-day duration of Experiment 3 was higher in those maintained on grass leaf material from A. lolii-infected ryegrass and switchgrass than in those on a diet of A. lolii-free ryegrass (F2,50 5 3.26, P , 0.05) (Table 4). In this experiment, parasitism increased weevil mortality only in those weevils on A. lolii-free ryegrass diet (parasitism 3 diet interaction, F2,50 5 3.71, P , 0.05), lifting the death

rate closer to that which occurred on A. lolii-infected ryegrass diet. The incidence of parasitism in the weevils at the end of the experiment was not affected by the diet on which the weevils were maintained (F2,50 5 1.43, P . 0.05) (Table 4). However, as in Experiment 2, the instar distribution of parasitoid larvae in the dissected TABLE 4 Mortality of Listronotus bonariensis and Occurrence of Various Developmental Stages of Their Endoparasitoid Microctonus hyperodae after Maintenance of the Weevils on Three Grass Diets Weevil diet A. lolii-free A. lolii-infected Switchryegrass ryegrass grass Percentage weevil mortality Nonparasitized a Parasitized b Mean feeding score/24 h Nonparasitized a Parasitized b Percentage weevil parasitism b,c Parasitoid instar development b,c Percentage larvae in 1st instar Percentage larvae in 2nd and 3rd instars a

8 23 5.9 6.4

36 35 3.2 3.7

42 49 0.3 0.2

82

77

72

0

79

100

100

21

0

Weevils without exposure to M. hyperodae. Weevils exposed to M. hyperodae prior to being maintained on the grass diets. c Parasitism in surviving weevils after 12 days on the grass diets. b

285

Microctonus hyperodae DEVELOPMENT IN Listronotus

weevils was markedly affected by the diets (n 5 161, x2 5 123.05, df 5 4, P , 0.001). The majority of parasitoids in weevils fed leaf segments from A. lolii-infected ryegrass and from switchgrass were first instar larvae, while parasitoids in weevils on A. lolii-free ryegrass had developed through to second and third instar larval stages. In Experiment 3, the intensity of feeding by weevils on leaf segments was significantly higher on A. loliifree than on A. lolii-infected ryegrass (F1,50 5 7.25, P , 0.01), which in turn was fed upon more than switchgrass (F1,50 5 12.83, P , 0.001) (Table 4). Feeding by weevils was not influenced by parasitism. Confinement to A. lolii-infected ryegrass for 14 days in Experiment 4 reduced feeding in L. bonariensis compared to confinement on uninfected ryegrass (F1,18 5 4.78, P , 0.05) (Table 5). At the end of the 14-day period on the A. lolii-infected ryegrass diet, the female weevils possessed less developed vitellaria (F1,8 5 5.36, P , 0.05) and had a higher incidence of oocyte resorption (F1,8 5 13.05, P , 0.01), compared to those on the A. lolii-free diet. Female weevils held without food for 14 days possessed vitellaria of similar development to those on A. lolii-infected ryegrass (F1,8 5 2.11, P . 0.05) but had a higher incidence of oocyte resorption (F1,8 5 22.82, P , 0.001). Weevil mortality was not influenced (F2,27 5 0.48, P . 0.05) but the diet regimes resulted in differential susceptibility to parasitism among the weevils (F2,12 5 5.30, P , 0.05) (Table 5). Mortality of the weevils during Experiment 5 was low and apparently unaffected by the incorporation in the diets of alkaloids and diterpenes derived from A. lolii (F9,90 5 1.17, P . 0.05). In contrast, the feeding intensity of L. bonariensis (F9,90 5 3.01, P , 0.01) and the rate of development of their endoparasitoids (F9,90 5 2.36, P , 0.05) were reduced. Most notable in TABLE 5 Effects of Three Diet Regimens on the Feeding Intensity, Mortality, and Female Reproductive Condition of Listronotus bonariensis and the Resulting Susceptibility to Parasitism by Microctonus hyperodae Weevil diet A. lolii-free A. lolii-infected ryegrass ryegrass Mean feeding score a Percentage weevil mortality a Vitellarium score of female weevils b Percentage female weevils with oocyte resorption b Percentage weevil parasitism c a

Nil

7.8 8.0

4.1 9.6

— 8.5

4.5

3.4

3.1

5.5 78.2

28.8 66.3

71.0 50.6

During the 14-day period on the diets. At the end of 14 days on the diets. c On 24-h exposure to parasitoids following 14 days on the diets. b

TABLE 6 Feeding Intensity and Mortality of Parasitized Listronotus bonariensis on Artificial Diets Incorporating Various Diterpenes and Alkaloids of Acremonium lolii Origin and the Yield of Microctonus hyperodae Prepupae

Diet

Mean feeding score/48 h

Percentage weevil mortality

Percentage weevils yielding parasitoid prepupae

Untreated Lolitrem a-Paxitriol Lolitriol b-Paxitriol Ergonovine Paxilline Ergovaline Ergotamine Peramine

7.8 7.2 6.5 5.8 5.7 5.6 5.4 4.2 4.0 2.9

4 2 4 2 6 2 2 2 4 6

82 64 60 84 78 90 76 88 76 80

Days to 50 percent parasitoid prepupal emergence 22.2 24.5 30.0 27.8 30.9 28.8 31.2 33.5 34.2 37.0

this feeding deterrancy and parasitoid growth retardation were ergovaline, ergotamine, and peramine (Table 6). The median duration of parasitoid development in the weevils (D, in days) was strongly correlated with the feeding intensity (F) of the weevil host: D 5 45.708 2 2.849F

r2 5 0.911, P , 0.001.

Incorporation of lolitrem B and a-paxitriol in the diet reduced the percentage of weevils yielding parasitoid prepupae (F9,90 5 2.91, P , 0.05) (Table 6). DISCUSSION

The experiments reported in this paper provide clear evidence that development of M. hyperodae larvae was retarded when the host weevils were confined to a diet of A. lolii-infected ryegrass in comparison to weevils on A. lolii-free diet. The slowed progression of M. hyperodae through the various larval instars was associated with the reduced feeding intensity of the host weevils, indicating that the effect of A. lolii was to reduce the nutritional quality of the weevil as a host. This effect of inadequate nutrition was even more marked when the weevils were confined to a diet of switchgrass, a plant species largely rejected by L. bonariensis. Alkaloids and diterpenes of A. lolii origin, incorporated into artificial diet, reduced feeding by L. bonariensis, which in turn was associated with retarded development of their endoparasitoids. There was a significant linear relationship between L. bonariensis feeding intensity on the diet and the rate of development of the parasitoid larvae. Several studies have demonstrated that plant allelochemicals in the host diet may adversely affect parasi-

286

BARKER AND ADDISON

toids developing on herbivorous insects (e.g., Campbell and Duffey, 1979; Duffey et al., 1986; Barbosa et al., 1991). Bultman et al. (1993) noted that the eulophid ectoparasitoid Euplectus comstockii Howard developed more rapidly on Spodoptera fruigiperda Smith that had been fed endophyte-free Festuca arundinacea Screb. than on hosts that had been reared on Acremonium coenophialum Morgan-Jones & Gams-infected fescue. A. coenophialum is known to provide feeding deterrancy to S. fruigiperda (Clay et al., 1985). Thus, the interactions of M. hyperodae and E. comstockii with their hosts conform to the general model of the parasitoid development rate being dependent on the ‘‘quality’’ of the insect host. Clavicipitaceous endophytic infections in grass plants, which produce insect deterrent chemicals, impact on parasitoids by reducing the nutrition of their hosts. The experiments reported here suggest that the ingestion of mycotoxins by the host may have an effect additional to reduced host nutrition, namely direct toxicity to the parasitoid larval stage. At the concentration of 2 µg g21, the diterpenes lolitrem B and a-paxitriol, of A. lolii origin, did not reduce feeding of L. bonariensis but did reduce the survival of parasitoid larvae in these hosts. Reduced nutrition of L. bonariensis for 14 days in Experiment 4, either by confinement on leaf material from A. lolii-infected ryegrass plants or by the withholding of food entirely, resulted in lower rates of parasitism by M. hyperodae. Apparently the parasitoids preferred to oviposit in well-fed, reproductively active hosts. These results argue for consideration of endophytic fungi as a further trophic level affecting the interaction among plants, insect herbivores, and parasitoids. The significance of A. lolii as a factor in retarded development and mortality of parasitoid larvae in field populations of L. bonariensis is currently not known. The resistance to L. bonariensis conferred on ryegrass by endophyte infection is not absolute, with weevil individuals and populations varying in acceptance of infected plants as hosts (Dymock and Hunt, 1987; G. M. Barker and R. Sinha, unpublished). On average, there is a strong preference by L. bonariensis for endophytefree plants and in pastures dominated by A. loliiinfected ryegrass, feeding and reproductive capacity of the weevils is diminished (Barker et al., 1989, 1990). What is clear is that M. hyperodae is sensitive to the quality of its host. Model predictions on M. hyperodae phenology and impacts on L. bonariensis populations (e.g., Barlow et al., 1994) need to account for host quality as a source of variation in parasitoid searching efficacy and developmental rates. ACKNOWLEDGMENT This research was funded by the Foundation for Research, Science and Technology.

REFERENCES Barbosa, P., Gross, D., and Kemper, J. 1991. Influence of plant allelochemicals on the tobacco hornworm and its parasitoid, Cotesia congregata. Ecology 72, 1567–1575. Barker, G. M., Pottinger, R. P., and Addison, P. J. 1984. Effect of Argentine stem weevil on productivity of grasses in the Waikato. N.Z. J. Agric. Res. 27, 993–102. Barker, G. M., Prestidge, R. A., and Pottinger, R. P. 1988. Reproductive phenology of Listronotus bonariensis (Kuschel) in northern New Zealand pastures. Bull. Ent. Res. 78, 659–668. Barker, G. M., Pottinger, R. P., and Addison, P. J. 1989. Population dynamics of the Argentine stem weevil (Listronotus bonariensis) in pastures of Waikato, New Zealand. Agric. Ecosyst. Environ. 26, 79–115. Barker, G. M., Prestidge, R. A., and Pottinger, R. P. 1990. Utilization of an endophytic fungus for insect resistance in New Zealand ryegrass. Aspects Appl. Biol. 24, 279–281. Barlow, N. D., Goldson, S. L., and McNeill, M. R. 1994. A prospective model for the phenology of Microctonus hyperodae (Hymenoptera: Braconidae), a potential biological control agent of Argentine stem weevil in New Zealand. Biocontrol Sci. Technol. 4, 375–386. Bultman, T. L., Coudron, T. A., Crowder, R. J., and Davis, A. P. 1993. Effect of Acremonium coenophialum in the diet of fall armyworm on the development of the parasitoid Euplectrus comstockii. In ‘‘Proceedings of the 2nd International Symposium on Acremonium/ Grass Interactions’’ (D. E. Hume, G. C. M. Latch, and H. S. Easton, Eds.), pp. 151–154. New Zealand Pastoral Agriculture Research Institute, Palmerston North, New Zealand. Campbell, B. C., and Duffey, S. S. 1979. Tomatine and parasitic wasps: Potential incompatibility of plant antibiosis with biological control. Science 205, 700–702. Clay, K., Hardy, T. N., and Hammond, A. M. 1985. Fungal endophytes of grasses and their effects on an insect herbivore.Oecologia 66, 1–6. Dahlman, D. L., Eichenseer, H., and Siegel, M. R. 1991. Chemical perspectives on endophyte–grass interactions and their implications to insect herbivory. In ‘‘Microbial Mediation of Plant– Herbivore Interactions’’ (P. Barbosa, V. A. Krischik, and C. G. Jones, Eds), pp. 227–252. Wiley, New York. Dicke, M., and Sabelis, M. W. 1992. Costs and benefits of chemical information conveyance: Proximate and ultimate factors. In ‘‘Insect Chemical Ecology’’ (B. D. Roitberg and M. B. Isman, Eds.), pp. 122–155. Hapman & Hall, New York. Doutt, R. L. 1959. The biology of parasitic Hymenoptera. Annu. Rev. Entomol. 4, 161–182. Duffey, S. S., Bloem, K. A., and Campbell, B. C. 1986. Consequences of sequestration of plant natural products in plant–insect–parasitoid interactions. In ‘‘Interaction of Plant Resistance and Parasitoids and Predators of Insects’’ (D. J. Boethel and R. D. Eikenbury, Eds.), pp. 31–60. Horwood, Chichester, United Kingdom. Dymock, J. J., and Hunt, V. A. 1987. Feeding and oviposition behaviour of Argentine stem weevil adults on endophyte-infected perennial ryegrass. Proc. N.Z. Weed Pest Control Confr., 40, 233–235. Firth, A. C., Barker, G. M., and Addison, P. J. 1993. Susceptibility of warm-season grasses to feeding and oviposition by the Argentine stem weevil (Listronotus bonariensis). Proc. N.Z. Plant Protection Confr. 46, 215–218. Goldson, S. L. 1983. Implications of oocyte resorption bodies in the ovarioles of Argentine stem weevil, Listronotus bonariensis (Kuschel) (Coleoptera: Curculionidae). N.Z. J. Zool. 10, 111–116. Goldson, S. L., McNeill, M. R., Stufkens, M. W., Proffitt, J. R., and Farrell, J. A. 1990. Importation and quarantine of Microctonus hyperodae, a South American parasitoid of Argentine stem weevil. Proc. N.Z. Weed Pest Control Conf. 43, 334–338.

Microctonus hyperodae DEVELOPMENT IN Listronotus Goldson, S. L., McNeill, M. R., Proffitt, J. R., Barker, G. M., Addison, P. J., Barratt, B. I. P., and Ferguson, C. M. 1993. Systematic mass rearing and release of Microctonus hyperodae Loan (Hym.: Braconidae, Euphorinae), a parasitoid of the Argentine stem weevilListronotus bonariensis (Kuschel) (Col.: Curculionidae) and records of its establishment in New Zealand. Entomophaga 38, 527–536. Hare, J. D. 1992. Effects of plant variation on herbivore–natural enemy interactions. In ‘‘Plant Resistance to Herbivores and Pathogens’’ (R. S. Fritz and E. L. Simms, Eds.), pp. 278–298. Univ. of Chicago Press, Chicago. Loan, C. C., and Holdaway, F. G. 1961. Microctonus aethiops (Nees) auctt. and Perilitus rutilus (Nees) (Hymenoptera: Braconidae), European parasites of Sitona weevils (Coleoptera: Curculionidae). Can. Entomol. 93, 1057–1079. Malone, L. A., and Wigley, P. J. 1990. A practical method for rearing Argentine stem weevil, Listronotus bonariensis (Coleoptera: Curculionidae) in the laboratory. N. Z. Entomol. 13, 87–88. Popay, A. J., and Rowan, D. D. 1993. Endophytic fungi as mediators of plant–insect interactions. In ‘‘Insect–Plant Interactions’’ (E. A. Bernays, Ed.), Vol. 5, pp. 83–103. CRC Press, Ann Arbor, MI. Porter, J. K. 1994. Chemical constituents of grass endophytes. In ‘‘Biotechnology of Endophytic Fungi of Grasses’’ (C. W. Bacon and J. F. White, Eds.), pp. 103–123. CRC Press, Boca Raton.

287

Prestidge, R. A. 1993. Causes and control of perennial ryegrass staggers in New Zealand. Agric. Ecosystems Environ. 44, 283–300. Prestidge, R. A., and Ball, O. J-P. 1993. The role of endophytes in alleviating plant biotic stress in New Zealand. In ‘‘Proceedings of the 2nd International Symposium on Acremonium/Grass Interactions’’ (D. E. Hume, G. C. M. Latch, and H. S. Easton, Eds.), pp. 141–151. New Zealand Pastoral Agriculture Research Institute, Palmerston North, New Zealand. Prestidge, R. A., Barker, G. M., and Pottinger, R. P. 1991. The economic cost of Argentine stem weevil in pastures in New Zealand. Proc. N.Z. Weed Pest Control Confr. 44, 165–170. Rowan, D. D., and Latch, G. C. M. 1994. Utilization of endophyteinfected perennial ryegrasses for increased insect resistance. In ‘‘Biotechnology of Endophytic Fungi of Grasses’’ (C. W. Bacon and J. F. White, Eds.), pp. 169–183. CRC Press, Boca Raton. SAS. 1982. ‘‘SAS User’s Guide.’’ Statistical Analysis System Institute, Cary. Vet, L. E. M., and Dicke, M. 1992. Ecology of infochemical use by natural enemies in a tritropic context. Annu. Rev. Entomol. 37, 141–172. Walde, S. J. 1995. How quality of host plant affects a predator–prey interaction in biological control. Ecology 76, 1206–1219.