Host specificity assessments of Cotesia plutellae, a parasitoid of diamondback moth P.J. Cameron1, G.P. Walker1, M.A. Keller2 and J.R. Clearwater3 1New

Zealand Institute for Crop & Food Research Ltd, Private Bag 92169, Auckland, New Zealand 2Department of Crop Protection, University of Adelaide, South Australia 5064 3Horticulture and Food Research Institute of New Zealand, Private Bag 92169, Auckland, New Zealand. Current address, 63 Peter Buck Rd, Auckland.

Abstract Cotesia plutellae is being assessed as a potential biological control agent for introduction against Plutella xylostella in New Zealand. As the literature on C. plutellae provided variable assessments of its host specificity, further information was collected from the laboratory and field. Our field assessments in Fiji indicated that this parasitoid did not attack other Lepidoptera in or around vegetable brassica crops. Laboratory tests on a colony of C. plutellae in South Australia, including simple no-choice experiments and flight tunnel choice tests, showed that the parasitoid could choose to oviposit in other Lepidoptera and that successful development rarely occurred. In New Zealand, similar laboratory tests of C. plutellae collected from Fiji revealed that it was capable of ovipositing and developing in seven other species of Lepidoptera. Host suitability was assessed by comparing the ability of the parasitoid to develop in P. xylostella and other species. Host acceptability was compared by assessing the flight of adults to test larvae on their host plants, and by comparing oviposition preferences. These experiments suggested that C. plutellae may parasitise Lepidoptera other than P. xylostella in New Zealand and indicate that further assessments are required to determine its potential impact in the field. Key words: Cotesia plutellae, Plutella xylostella, host specificity, oviposition, development.

Introduction Cotesia plutellae Kurdjumov (Hymenoptera: Braconidae) is being considered as a candidate for introduction into New Zealand to augment the existing parasitoids of diamondback moth (DBM) Plutella xylostella (L.) (Lepidoptera: Yponomeutidae). These parasitoids, Diadegma semiclausum (Hellen) and Diadromus collaris (Gravenhorst), can provide high rates of parasitism in the spring and early summer, but in dry periods of the summer DBM population increases are often not controlled by existing natural enemies. Insecticide applications are then necessary (Beck and Cameron, 1992). C. plutellae was proposed for introduction because it kills DBM larvae at an earlier stage than the existing parasitoids and therefore may reduce feeding damage on host plants. This parasitoid may also supress DBM better than D. semiclausum because it is more active at higher temperatures (Talekar and Yang, 1991). The major question concerning the introduction of C. plutellae into New Zealand is its specificity to DBM and its possible effects on non-target species. Fitton and Walker (1992) point out that although C. plutellae is widely assumed to be host specific, it has been recorded from several other species of Lepidoptera. Although for some early biological control introductions native alternative hosts were considered as useful reservoirs (Cameron et al. 1993), conservation of native species, including the few attractive native butterflies in New Zealand, is now an important issue. In this study, our assessments of

the host specificity of C. plutellae have included consideration of the literature, assessments of its host range in the field, and laboratory experiments to define behavioural and physiological measures of host range. Methods Field survey The specificity of C. plutellae in the field was examined by collecting and rearing Lepidoptera larvae from crucifers and adjacent weeds in areas of Fiji where the parasitoid was known to be present (Walker et al., in press). Larvae were reared individually to allow any parasitoids to emerge, and representative Lepidoptera larvae and adults were retained for identification. Dr J.A. Berry and A.K. Walker identified the hymenopterous parasitoids and J.S. Dugdale identified the Lepidoptera. Similar observations were carried out in South Australia, although the apparent absence of C. plutellae minimised their value. Laboratory assessments Sources of insects: Laboratory experiments were performed in 1994 at the University of Adelaide with a culture of C. plutellae that had been obtained from Dr N.S. Talekar in Taiwan. Separate experiments were carried out at Crop & Food Research in Auckland with C. plutellae collected from Fiji in 1995. Both cultures were identified by Drs J.A. Berry and A.D. Austin, and voucher specimens have been deposited in the New Zealand Arthropod Collection at Landcare Research in Auckland. Biologically-based technologies 85

C. plutellae was reared on DBM larvae fed on cabbage. Adult parasitoids were held in 4 litre, ventilated, clear plastic containers at 21 °C and fed dilute honey solution on cotton wicks. From days 2 to 5 after adult emergence, approximately 60, 2nd and 3rd instar larvae on a small cabbage leaf were presented for parasitisation to 10 females (and a similar number of males) for 3 h. The larvae were then reared on cabbage, and parasitoid cocoons were collected after 11–13 days and if necessary stored at 15 °C in glass vials for up to 1 week. Lepidoptera species to be tested were generally collected as adults from light traps. Eggs were collected from gravid females and larvae reared on their usual host plant or on cabbage. The majority of test insects were collected from Auckland (Table 1); those from other sources were: Plutella antiphona Meyrick (from Chatham Island), Australian Bassaris itea F. (Adelaide, South Australia), Australian Diarsia intermixta Guenee (Devonport, Tasmania), Neumichtis saliaris Guenee (Devonport, Tasmania). Host suitability: The suitability of Lepidoptera larvae of different species for development of C. plutellae was tested by exposing individual larvae to a single mated female in a 100 x 25 mm glass vial for 5 minutes in an attempt to force oviposition. Larvae were chosen to match the approximate size of the 2nd or 3rd instar P. xylostella used as controls. Larvae were presented to one female in a sequence, alternating control and test larvae, until the parasitoid failed to oviposit in DBM. The time taken to initiate oviposition was also measured. The success of oviposition was checked by dissecting some test larvae after 48 h to determine if eggs had been deposited or larvae were

developing. The remaining test larvae were reared until parasitoid larvae emerged to form cocoons, or until the test larva died. The comparative success of parasitoid development in different test species was also assessed by exposing six replicates of 8–12 test larvae to individual females in 4 litre cages for 3 h. In similar experiments, parasitoid females were provided with a choice between larvae of DBM or another test species. The results were expressed as the number of C. plutellae cocoons per number of larvae exposed to parasitism. Flight tunnel tests: The acceptability of different test species was assessed by observing the flight of adult female C. plutellae to larvae on excised leaves in a flight tunnel. These behavioural tests were initiated at the University of Adelaide using methods developed by Keller (1990). The wind speed in the tunnel was set at 60 cm/s. Adult parasitoids were released at 70 cm from the test insects, and the experiments were run at 25 °C. At Crop & Food Research in Auckland, behavioural assays were continued using a flight tunnel based on the design of Miller and Roelofs (1978). The tunnel was operated at a wind speed of 50 cm/s and a temperature of 21 °C. Test females were fed and mated but had no experience of Lepidoptera larvae prior to release in the tunnel. Females were presented with larva-plant combinations either alternately (no-choice) or simultaneously (choice test). Five to ten test insects were placed on each plant 24 h prior to the experiments to ensure the presence of some leaf damage. Plants were presented as one or two excised leaves to provide a similar leaf area for each test. For the choice tests, the plants were placed approximately 15 cm apart

Table 1. Oviposition and development of Cotesia plutellae in test insects, measured as number per larvae exposed to parasitism Family Test insect

Ovipositions per exposure

Eggs per oviposition

Cocoons per oviposition

Cocoons per exposure

Yponomeutidae Plutella xylostella (DBM) Plutella antiphona

182/190 –

32/41 –

21/23 7/8

60/110 21/72

6/33

0/1

0/5

0/48

11/32

0/11

0/3

0/60

Tortricidae Epiphyas postvittana Pieridae Pieris rapae Nymphalidae Basaris itea Basaris itea ex Australia Arctiidae Nyctemera amica Noctuidae Agrotis ipsilon Diarsia intermixta Diarsia intermixta ex Australia Graphania mutans Graphania ustistriga Helicoverpa armigera Neumichtis saliaris ex Australia Spodoptera litura Thysanoplusia orichalcea 86

5/40 0/30

– –

–

17/24

–

–

– – –

1/12

0/3 2/6 15/30

–

– – –

16/30 – 14/24 4/10

7/22 5/12 1/8 2/31 3/17

– 2/16

– 1/8 –

Proceedings: The Management of Diamondback Moth and Other Crucifer Pests

– –

0/5 0/20 0/17 0/4

1/50 9/50 – 3/50 30/50 – – 0/47 –

across the air flow, equidistant from the centre line, and their position was alternated between each test. Results and Discussion Published host records Numerous field records suggest that C. plutellae is a narrowly oligophagous parasitoid of DBM that occasionally parasitises other Lepidoptera species. The majority of records in Shenefelt’s 1972 summary of host records report DBM as the only host. The main exception to this is the extensive list of hosts derived from Wilkinson (1939) in his description of the parasitoid. Most of Wilkinson’s records appear to be based on host identifications associated with parasitoid specimens, and six of the hosts are based on single records. After extensive rearing, Delucchi et al. (1954) considered that two of the species on Wilkinson’s list, Pieris rapae L. and Pieris brassicae L., were not attacked by C. plutellae. In his surveys of parasitism of Hyphantria cunea Drury (Arctiidae), Bogavic (1953) recorded approximately 26 C. plutellae, equivalent to less than 0.01% parasitism. Together with Wilkinson’s records of five parasitised Diacrisia urticae Esp. (Arctiidae), these records suggested that further host assessments should include representatives of this family. Wilkinson’s (1939) report of 28 host records for two species of Lymantriidae and six host records for Malacosoma castrensis L. (Lasiocampidae) also warrant consideration. The most definite information on host association is that for Aglais urticae L. (Nymphalidae). Wilkinson (1939) received specimens of C. plutellae reared from A. urticae, and maintained a culture of this parasitoid on both DBM and A. urticae. He stated that “.... Apanteles plutellae Kurdj., is able to utilise both P. maculipennis and A. urticae as a host.” The potential for rearing C. plutellae on alternative hosts in the laboratory was also demonstrated by Wang et al. (1972) who showed that although the parasitoid preferred to oviposit in DBM it survived better in the rice moth Corcyra cephalonica Stainton. Similarly, Lim (1982) (cited in Waterhouse and Norris, 1987) recorded that Crocidolomia binotalis Zeller and Hellula undalis Guenee were parasitised in the laboratory but not in the field. An additional field host record for C. plutellae was reported by Baloch et al. (1966) who recorded low rates of parasitism when assessing the potential of Oeobia (=Anania) verbascalis Schiff. (Pyralidae) for biological control of Noogoora burr, Xanthium strumarium L. These literature records, together with general criteria for selecting test species, suggested three categories of test species: 1. Close relatives, i.e. Plutellidae. 2. Species on crucifers, especially Noctuidae. 3. Species in the same family as hosts recorded in the literature, i.e. Nymphalidae, Noctuidae, Arctiidae, Pyralidae, Lymantridae, Lasiocampidae.

Field survey Collection and rearing P. xylostella in the Suva region of Fiji in 1992 and 1995 indicated that C. plutellae was common. It parasitised 74% of DBM in the 1995 survey (Walker et al. in press). Larvae of several Noctuidae and one Pyralidae were also collected from brassicas and weeds in the same area. Although other parasitoids were present, no C. plutellae were reared from 563 Spodoptera litura F., 43 Helicoverpa armigera (Hubner), 17 Chrysodeixis eriosoma Doubleday (all Noctuidae) and 130 Hymenia recurvalis (F.) (Pyralidae). Crocidolomia binotalis (Pyralidae) was similarly not parasitised, but the presence of C. plutellae in the region that this lepidopteran was collected was not confirmed. These results augment the previous observations of Lim (1982) (cited in Waterhouse and Norris, 1987) that C. binotalis and Hellula undalis were parasitised in the laboratory but not in the field. Laboratory assessments Test species: Initial assessments in South Australia in 1995 tested three native species. Two of these species, Diarsia intermixta and Neumichtis saliaris (Noctuidae), are associated with brassicas (Common, 1990), whereas the yellow admiral Bassaris itea (Nymphalidae) occurs on nettle (Urtica dioica) and is valued, particularly in New Zealand, as an attractive species. Test species in New Zealand included a near relative of DBM, the endemic species Plutella antiphona, which has a limited distribution on Cruciferae, particularly water cress, Nasturtium officinale (Dugdale, 1973). Seven test species that occur on brassicas in association with P. xylostella were Noctuidae (Table 1). Other test species from brassicas were Pieris rapae (Pieridae) and Epiphyas postvittana (Tortricidae). The remaining test species where Nyctemera amica (Arctiidae) collected from ragwort (Senecio jacobaea) and B. itea (Nymphalidae), both representing families that include hosts of C. plutellae previously reported in the literature (Wilkinson, 1939). There are no Lasiocampidae or Lymantriidae in New Zealand so assessments of species from these families were not relevant. Host suitability: In experiments with the University of Adelaide culture of C. plutellae, the parasitoid was induced to oviposit in D. intermixta, and from 31 oviposition attempts two cocoons were produced (Table 1). Oviposition responses to N. saliaris were more difficult to obtain and no cocoons were produced from 20 oviposition attempts on this species. B. itea was unacceptable to C. plutellae to the extent that no oviposition responses were obtained. C. plutellae was collected from Fiji and imported for assessment in quarantine in Auckland. This culture attempted to oviposit in all test species, but dissection of larvae revealed that eggs were not deposited in E. postvittana or P. rapae (Table 1). The oviposition rate was highest in DBM and oviposition was initiated more quickly (data not shown) in this species. There was no clear difference between oviposition rates in Biologically-based technologies 87

species other than P. xylostella, nor was the oviposition rate related to success in cocoon formation. For example, no cocoons developed from Spodoptera litura, but more than 50% of the larvae attracted oviposition attempts. This demonstrated that oviposition provided a poor estimate of the suitability of a species, possibly because this response may be elicited by plant (cabbage)-associated factors. Of those species where eggs were detected, both S. litura and Helicoverpa armigera were unsuitable for further development. The rate of cocoon formation (Table 1) indicated that six species were not hosts: B. itea, A. ipsilon and G. mutans were occasional laboratory hosts; and G. ustistriga, N. amica, P. antiphona and D. intermixta were all suitable laboratory hosts for C. plutellae. Exposure of test species in choice tests reduced the rate of cocoon formation (Table 2) indicating that mixed host populations interfered with host location and reduced the probability of attack on alternate species. Estimates of the development rate of C. plutellae provided another measure of the suitability of some test species. In P. antiphona, parasitoids developed from egg to cocoon at the same rate as in DBM. By contrast, D. intermixta and G. ustistriga required two more days to develop, and N. amica took four more days than DBM. Flight tunnel tests: Flight tunnel experiments showed that C. plutellae females could fly to all test combinations of insect and plant species (Table 3). During successful flights females frequently exhibited characteristic casting behaviour as they sampled the odour gradients produced by each source. In no-choice tests, fewer adults flew to B. itea on nettle,

but plants with larvae of H. armigera or N. amica were as attractive as DBM. The rate of successful flights declined slightly in choice tests, but apart from reduced flights to B. itea, C. plutellae showed no distinct preference. Flights to H. armigera and N. saliaris, previously demonstrated to be unsuitable for development, strongly suggested that C. plutellae is attracted by cabbage volatiles, or by the volatiles from damaged cabbage. This behaviour has also been observed in Cotesia rubecula by Agelopoulos and Keller (1994a, b) who reported that, although this parasitoid did not distinguish between damage by host or non-host Lepidoptera, the blend of volatiles emitted from frass was different for DBM and Pieris rapae. The development of behavioural measures that reflect the host range of oligophagous species such as C. plutellae will require further refinements of testing procedures. Whereas simple tests may be suitable for demonstrating high degrees of specificity such as found in C. rubecula (Cameron and Walker, in this proceedings), the specificity of C. plutellae is apparently determined more by behaviour than physiological compatibility. Testing procedures will therefore need to consider that because confinement disturbs the natural behaviour of parasitoids, laboratory tests will usually overestimate their host range in the field (Sands, 1993). To obtain more realistic measures of host range in the field, we also plan to extend our collection and rearing of Lepidoptera from regions where C. plutellae is naturally present. Acknowledgements We thank Anne Barrington, Tim Herman, Lionel Hill, John LeSueur and Chris Winks for supplying test

Table 2. Development of Cotesia plutellae cocoons from tests insects compared with Plutella xylostella (DBM) in choice and no-choice experiments, measured as number of cocoons per larvae exposed to parasitism Alternate test insect and plant

Graphania mutans on cabbage Nyctemera amica on ragwort Spodoptera litura on cabbage

Choice

No choice

Alternate

DBM on cabbage

Alternate

1/60 9/60 0/50

16/60 16/60 12/50

3/50 7/22 0/47

DBM on cabbage 41/50

Table 3. Flights of Cotesia plutellae to test insect and host plant combinations compared with Plutella xylostella (DBM) in a flight tunnel Number of flights per number of tests Alternate test insect and host plant DBM on cabbage

Alternate test combination

No choice test Bassaris itea on nettle Helicoverpa armigera on cabbage Nyctemera amica on ragwort

10/18 10/19 7/15

4/15 9/21 5/13

Choice test Bassaris itea on nettle Nyctemera amica on ragwort Diarsia intermixta on cabbage ex Australia Neumichtis saliaris on cabbage ex Australia

18/43 8/27 19/51 13/30

6/43 6/27 21/51 14/30

88

Proceedings: The Management of Diamondback Moth and Other Crucifer Pests

insects, and Darryl Jackman and Sarah Painter for rearing various Lepidoptera and assisting with experiments. Andy Austin, Jo Berry and Annette Walker provided identifications of the parasitic Hymenoptera, and John Dugdale and Lionel Hill identified the Lepidoptera. References Agelopoulos, N.G. and Keller, M.A. (1994a). Plant-natural enemy association in the tri-trophic system Cotesia rubecula-Pieris rapae-Brassicaceae (Cruciferae) II: Preference of C. rubecula for landing and searching. Journal of Chemical Ecology 20: 1 735–48. –––– (1994b). Plant-natural enemy association in the tri-trophic system Cotesia rubecula-Pieris rapae-Brassicaceae (Cruciferae) III: Collection and identification of plant and frass volatiles. Journal of Chemical Ecology 20: 1 955–67. Beck, N.G. and Cameron, P.J. (1992). Developing a reduced spray program for brassicas in New Zealand. In Diamondback moth and other crucifer pests (ed. N.S. Talekar) pp. 341–350. Baloch, G.M., Din, I.M. and Ghani, M.A. (1966). Biology and host-plant range of Oeobia verbascalis Schiff. (Pyralidae: Lepidoptera); an enemy of Xanthium strumarium L. Technical Bulletin of the Commonwealth Institute of Biological Control 7: 81–90. Bogavic, M. (1953). Some observations on fall webworm parasites. Zastita Bilja 16: 58–80. Cameron, P.J., Hill, R.L., Bain, J. and Thomas, W.P. (1993). Analysis of importations for biological control of insects pests and weeds in New Zealand. Biocontrol science and technology 3: 387–404. Cameron, P.J. and Walker, G.P. (1997). Introduction and evaluation of Cotesia plutellae, a parasitoid of Pieris rapae in New Zealand. In: A. Sivapragasam, W.H. Loke, A.K. Hussan and G.S. Lim (eds.). Proceedings of the Third International Workshop on Diamondback Moth. Kuala Lumpur, Malaysia. Common, I.F.B. (1990). Moths of Australia. E.J. Brill, The Netherlands, 535 pp. Delucchi, V., Tadic, M. and Bogavic, M. (1954). Mass rearing of Apanteles plutellae Kurdj. (Hym., Ichneumonidae), endophagous parasites of Plutella maculipennis Curt. and biological observations on these parasites. Zastita Bilja 21: 26–41

Dugdale, J.S. (1973). The genus Plutella (Hyponomeutidae) in New Zealand and the family position of Circoxena (Lepidoptera). New Zealand Journal of Science 16: 1 009–23. Fitton, M. and Walker R, A. (1992). Hymenopterous parasitoids associated with diamondback moth: the taxonomic dilemma. In Diamondback moth and other crucifer pests (ed. N.S. Talekar) pp. 225–232. Keller, M.A. (1990). Responses of the parasitoid Cotesia rubecula to its host Pieris rapae in a flight tunnel. Entomologia Experimentalis et Applicata 57: 243–249. Miller, J.R. and Roelofs, W.L. (1978). Sustained-flight tunnel for measuring insect responses to wind-borne sex pheromones. Journal of Chemical Ecology 6: 187–198. Sands, D.P.A. (1993). Effects of confinement on parasitoid/ host interactions: Interpretation and assessment for biological control of arthropod pests. In Pest Control and Sustainable Agriculture (ed S.A. Corey, D.J. Dall and W.M. Milne) pp. 196–199. Shenefelt, R.D. (1972). Hymenopterum catalogus. Part 7, Braconidae 4. Dr W. Junk N.V s-Gravenhage. Talekar, N.S. and Yang, J.C. (1991). Characteristics of parasitism of diamondback moth by two larval parasites. Entomophaga 36: 95–104. Walker, G.P., Cameron, P.J., Berry, J.A. and Lal, S.N. (in press). Parasitoids reared from lepidopteran larvae from brassicas and associated weeds in Fiji, 1992 and 1995. Proceedings of the Second Workshop on Biological Control in the Pacific. Wang, C-L., Chio, H. and Ho, K-K. (1972). The comparative study of parasitic potential of the braconid wasp (Apanteles plutellae Kurdj.) to the diamondback moth (Plutella xylostella L.) and rice moth (Corcyra cephalonica Staint.). Plant Protection Bulletin (Taipei) 14: 125–8. (In Chinese, English summary). Waterhouse, D.F. and Norris, K.R. (1987). Biological Control-Pacific Prospects. Inkata Press, Melbourne, 454 pp. Wilkinson, D.S. (1939). On two species of Apanteles (Hym. Brac.) not previously recognised from the Western Palearctic region. Bulletin of Entomological Research 30: 77–84.

Biologically-based technologies 89