Spruce stands provide natural food sources to adult hymenopteran parasitoids of bark beetles

Entomologia Experimentalis et Applicata 96: 253–263, 2000. © 2000 Kluwer Academic Publishers. Printed in the Netherlands.

253

Spruce stands provide natural food sources to adult hymenopteran parasitoids of bark beetles E. Hougardy1 & J.-C. Gr´egoire2 Laboratoire de biologie animale et cellulaire, C.P. 160/12, Universit´e Libre de Bruxelles, 50 avenue F.D. Roosevelt, 1050 Bruxelles, Belgique 1 Fonds pour la Formation a ` la Recherche dans l’Industrie et dans l’Agriculture 2 Fonds National Belge de la Recherche, Scientifique Accepted: May 11, 2000

Key words: bark beetles, Scolytidae, Ips typographus, natural enemies, parasitoids, adult feeding, natural food resources, Coeloides bostrychorum

Abstract Adults of synovigenic hymenopteran parasitoids of bark beetles need supplemental feeding to maximise their lifetime realised fecundity, and yet little is known about the natural sources that provide this food. Here we show that, in spite of the reduced diversity of flowering plants expected in even-aged spruce plantations, several plant species are present that might provide nectar and pollen to the hymenopteran parasitoid complex of Ips typographus L. (Coleoptera: Scolytidae), throughout the growing season. Conifer aphids are also abundant in these stands and produce large amounts of honeydew, which might also be consumed by the parasitoids. Laboratory experiments demonstrate that longevity is significantly increased in Coeloides bostrychorum Giraud (Hymenoptera: Braconidae) when the wasps are exposed to flowers of various species commonly found in spruce stands.

Introduction Most hymenopteran parasitoids of bark beetles are synovigenic: females emerge with a small number of large, nutrient-rich eggs and new eggs can be produced if food resources are available (Gauld & Bolton, 1988). Numerous authors have provided evidence of the major role played by adult nutrition in the biology of synovigenic parasitoids (Leius, 1960, 1961a, b, 1963, 1967; Ryan & Rudinsky, 1962; Doutt, 1964; Bushing, 1967; Jervis & Kidd, 1986; Mendel, 1988; Campos & Gonzalez, 1990; Mathews & Stephen, 1997, 1999). The availability of food and moisture can increase longevity and fecundity of parasitic wasps, increasing their potential impact as biological control agents of insect pest populations. Little is known about the range of natural food sources exploited by most adult parasitoid species (Jervis et al., 1992). Although host feeding is timeand energy-saving – because adult feeding and oviposition occur at the same place (Askew, 1971) – the

consumption of host fluids or tissues seems to be uncommon in parasitoids of bark beetles (Mendel, 1988; Stephen, 1995). Roptrocerus xylophagorum Ratzeburg (Pteromalidae) is the only parasitoid species known to host feed: females, on apparently rare occasions, build a feeding tube to get access to the haemolymph of the concealed host (Bushing, 1967). Beside this species, bark beetle parasitoids may feed from nonhost food resources including flower nectar, pollen or homopteran honeydew. The presence of food sources could be a favourable factor in the effectiveness of parasitoids. The aim of this study is to investigate the natural food resources available for forest parasitoids, especially hymenopteran parasitoids of Ips typographus L., a well-known pest of spruce in Eurasia. This study starts from an hypothesis developed by Stephen (1995) to explain the poor ability of hymenopteran parasitoids to regulate Dendroctonus frontalis Zimmermann populations in pine forests of south-eastern United States. Stephen et al. (1997) noticed in one large infestation

254 where beetle and parasitoid populations were monitored over time, that parasitism rates on D. frontalis never exceeded 10% with an average of 5–6%. They suggested that low rates could be due to intensive forest management practices, which decrease the numbers of flowering plants which may provide food to adult parasitoids, thus affecting the wasps’ reproductive capacity. This hypothesis might be applied to homogeneous spruce plantations in Southern Belgium, which are also experiencing intensive management with poorly diversified understories. Published parasitism rates on Ips typographus in Europe are scarce but it seems that parasitism success is variable. Eck (1990) reported 0–70% mortality rates of Ips typographus due to parasitoids, but for half of the trees sampled, this mortality was less than 5%. Rates of parasitism by Rhopalicus tutela Walker (Pteromalidae) vary from 4% (Bombosch, 1954, in Mills, 1983) to 20–25% (Lovasky, 1941 in Mills, 1983). Higher rates (up to 50–95%) have been reported for Coeloides bostrychorum Giraud (Braconidae) (personal observations; N. J. Mills, 1983, and references therein). Hymenopteran parasitoids can visit numerous plant species to collect nectar or pollen (Leius, 1960; Jervis et al., 1993); they also can use honeydew as a water and food resource (Györfi, 1945 in Leius, 1960). We therefore surveyed the availability of wildflowers and honeydew within or in the vicinity of spruce plantations. Furthermore, in order to establish a more explicit connection between parasitoids and the wildflowers found in the field, we carried out laboratory longevity tests with C. bostrychorum, exposing adult wasps to several species of flowering plants. Fecundity tests were not considered here because of the technical difficulties involved. In a separate experiment however (unpublished), significantly higher egg load in Coeloides bostrychorum and Rhopalicus tutela was associated with food consumption.

Materials and methods Surveys Four sites located in Wellin and one located in Bertrix (province of Luxembourg) were sampled for natural resources available to hymenopteran parasitoids. All sites are even-aged mature spruce (Picea abies Karst.) plantations (age ranging from 60 to 80 years) that have been attacked by Ips typographus in previous years.

Floristic surveys. Floristic surveys were conducted on 4 June, 1998. On each site, we visually estimated the presence and population size of the phanerogam species, using Braun-Blanquet’s abundance-dominance indices (Guinochet, 1973). Presence of honeydew. Detection devices were constructed according to a design established by H. Guerriat (pers. comm.) to collect honeydew produced by aphid populations on spruce. Each device consisted of a 10 × 10 cm piece of filter paper (Schleicher and Schuell) affixed with two rubber bands on a horizontal wooden board of the same size carried by a vertical, 70 cm wooden post. Three trees were randomly selected at each site and four devices were set at the base of each tree in the four cardinal directions, 50 cm above ground and 1 m away from the tree trunk. Detection devices were exposed for 24 h during dry weather. Spots of honeydew drops were then revealed in the laboratory by spraying a methanolic solution of aniline phthalate (0.1 M) followed by 10 min heating in an oven (130 ◦ C). Counting the honeydew spots allowed a semi-quantitative estimate of honeydew production. Detection devices were deployed on 23 June 1998 at four sites previously chosen for floristic surveys in Wellin (site 1 to 4). Time constraints permitted us to re-sample only site 3, on 8 August 1998 to assess the presence of honeydew during the summer. Longevity tests with wildflowers Experiments were run in two types of experimental cages (72 dm3 and 415 dm3 ), according to the size of the test plants. The cages had a transparent face made of clear polystyrene (respectively 40 × 60 cm and 66 × 80 cm according to the model) and were facing a window oriented to 68◦ E. Because of the position of the building, the cages were never exposed to direct sunlight. Room temperature was 23 ◦ C ± 2 ◦ C. Parasitoids were split into two groups, both receiving water on a piece of filter paper. One group (control) received no food while the second group was exposed to potted or cut wildflowers. Based on their abundance at the surveyed sites, seven plant species were selected to be tested separately: Teucrium scorodonia (Labiaceae), Digitalis purpurea (Scrophulariaceae), Senecio vulgaris (Asteraceae), Calluna vulgaris (Ericaceae), Scrophularia nodosa (Scrophulariaceae), Epilobium angustifolium (Onagraceae) and Vaccinium myrtillus (Ericaceae). Parasitoids were

255 at all the sites surveyed (Figure 1). Site 3 showed a significantly higher production of honeydew than the other sites (One-way Anova: F=19.087; P