Proceedings: Host Specificity Testing of Exotic Arthropod Biological Control Agents: The Biological Basis for Improvement in Safety Host Specificity Assessment of European Peristenus Parasitoids for Classical Biological Control of Native Lygus species in North America: Use of Field Host Surveys to Predict Natural Enemy Habitat and Host Ranges U. Kuhlmann1, P.G. Mason2 and R.G. Foottit2 1CABI Bioscience Centre, Rue des Grillons 1, 2800 Delémont, Switzerland 2Agriculture and Agri-Food Canada, Eastern Cereal and Oilseed Research Centre Ottawa, Ontario, K1A 0C6, Canada Abstract The Lygus bug complex (Het., Miridae) at many sites in North America causes economic damage to a wide variety of agricultural crops and is the focus of numerous research projects. Lygus can be best suppressed using parasitoids in unsprayed non-cropping situations as well as in crops where pollination is important for maximizing yields. The European parasitoid Peristenus digoneutis Loan (Hym., Braconidae), which attacks several species in the genera Lygus and Adelphocoris in Europe, has been established in the eastern USA to control the native pest Lygus lineolatus (Palisot de Beauvois) in alfalfa. The success of this project has stimulated interest into the potential for the establishment of additional Euro- pean species for biological control of pest Lygus bugs in several regions of North America. The research of this ongoing case study concentrates on assessing strategies and methods for host specificity testing of these parasitoids in relation to Europe and North America. Predicting the impact of European parasitoids by using existing and new knowledge of the host range and habitats in the area of origin will aid in choosing Peristenus species for further release in other regions and will have important implications for the practice of classical biological control. Introduction environmental risks of arthropod biological control agents have usually seemed negligible, but recent Need to Estimate Host Ranges of concern about the impact of alien species on Entomophagous Biological Control Agents biodiversity and natural ecosystems has caused this Concerns about reliance on chemical pesticides in crop belief to be reconsidered. protection worldwide have led to the development of integrated pest management strategies, which depend on The potential risks of arthropod biological control to natural enemy manipulations as alternatives to exclusive native non-target arthropods are being examined for use of chemicals. Classical biological control, the some cases (i.e., Goldson and Phillips, 1990; Barratt et introduction of exotic biological control agents to al., 1998, 1999). While there are no recorded examples permanently suppress exotic pests, is one such approach. of monophagous or narrowly oligophagous agents It is practiced ever more widely as successes accumulate changing their host range to cause damage to non-target and pest invasions follow trade liberalization. insects (Waterhouse, 1991), the use of polyphagous insects as biological control agents may reduce A high level of host specificity in the introduced natural populations of non-target hosts or cause their extinction enemies is desirable and should be sought during foreign (Horn, 1991; Howarth, 1991; Samways, 1994). exploration (Nechols et al., 1992). Potential However very little quantitative information is available 84 Host Specificity Assessment of European Peristenus Parasitoids Proceedings: Host Specificity Testing of Exotic Arthropod Biological Control Agents: The Biological Basis for Improvement in Safety on the influence of these agents on the population will also ensure that actual alternate hosts, occasionally densities of these non-target populations. used, are reported. Regulatory agencies that oversee biological control The laboratory methods used to assess the host ranges of introductions are responding to concerns of potential herbivorous insects being introduced as weed biological non-target impacts by requiring more rigorous control agents are often suggested as a model for efforts testing and demonstration of a high degree of host to assess host specificity of entomophagous insects. specificity by candidate arthropod biological control However, several difficulties exist in applying this agents before granting permission for release. approach. A significant problem is the major difference However, because of an earlier lack of concern for in the status of systematic knowledge of plants and non-target arthropods, methodologies for host insects. Whereas the taxonomy and phylogenetic specificity screening of arthropod natural enemies are relationships of plant species are relatively well known, much less developed than for testing of herbivorous far less is known about these relationships for insects. insects imported for weed biological control. Recent The uncertainty of the relatedness of insects within papers discuss methods to evaluate the potential families and tribes makes the choice of species for impact of parasitoids of arthropod pests on non- testing uncertain. Secondly, the large number of species target hosts before and after release of parasitoids in some insect groups often easily exceeds the number (i.e., Sands, 1997, Sands and Van Driesche, 2000; of plants in similar level taxa by an order of magnitude. Van Driesche and Hoddle, 1997; Hopper, 1998). This often precludes testing of more than a tiny Two categories of data can help in estimating the host subsample of species in related groups. If for example, ranges of insects: (1) host-natural enemy associations the target insect’s family contains 5000 species (a as seen in the published literature or in specially common possibility), most of which have biologies and conducted surveys; and (2) laboratory testing in distributions that are poorly known, choosing a list of which candidate species are presented in cages to species for host range testing is extremely difficult. natural enemies, whose oviposition and immature Consequently, systematics is a critical part of host development are then observed. Because insect faunas specificity testing programs for projects of biological are often very large, study of host-natural enemy control targeting insect pests. Reconstructions of associations by field surveys are often important in phylogenies may be needed to guide the host testing and choosing species for testing in the laboratory. agent selection process. Finally, lack of information about the biology and rearing methods for many insects makes it impractical to assemble sets of target species for Approaches to Host Ranges laboratory testing in the conventional manner used to Estimation for Parasitoids: Field test herbivory on plants (which may be collected and Surveys vs. Laboratory Testing stored as seeds until needed in most cases). for Large Faunas Once a species test list has been defined, uncertainties Traditionally, host association records from the literature exist on how to conduct and interpret laboratory host provide the first estimate of the host range of an range tests for entomophagous species. Sands (1993) entomophagous biological control agent. Such records identified several difficulties in interpreting results of are especially useful for distinguishing candidates with such laboratory tests on the physiological obviously broad host ranges from those that might have (=fundamental, see Van Klinken, 2000) host ranges of suitably narrow ones. Misidentifications, especially in entomophagous insects. It is extremely difficult to the older literature, often cause spurious host records. reproduce accurately the cues and stimuli that influence However, some of the problems posed by such errors the host searching and assessment behavior of a may be avoided by considering the quality of data, parasitoid in a natural environment. In laboratory tests, taking into account such aspects as numbers of entomophagous insects often accept a broader range of individuals examined, percent parasitism observed, hosts than in nature, over-estimating field host range spatial and temporal extent of studies, and confirmation (Loan and Holdaway, 1961). In laboratory trials, of identity by a specialist. In this way, observations of patterns of oviposition, feeding, or development in attacks on non-hosts, especially those based on single or arthropod hosts are typically assessed in small few specimens and not properly identified are less likely containers. Sands and Papacek (1993) reported that to be incorrectly entered into the host record list. This restricted space often leads to an inaccurate assessment Host Specificity Assessment of European Peristenus Parasitoids 85 Proceedings: Host Specificity Testing of Exotic Arthropod Biological Control Agents: The Biological Basis for Improvement in Safety of host specificity by disrupting the processes governing have been the focus of numerous research projects. host recognition and acceptance. For example, Parasitoids are best able to suppress Lygus species in parasitoids in small cages may oviposit in hosts that unsprayed non-cropping situations and crops in which normally do not support development of the parasitoid, pollination is important for maximizing yields. or parasitoids may oviposit in hosts that normally are Nymphal parasitoids of the subfamily Euphorinae not accepted in the field. The physiological host range (Hymenoptera: Braconidae) belonging to the genera measured in the laboratory and the realized host range Leiophron Nees and Peristenus Foerster are known to