Host specificity in arthropod biological control, methods for testing and interpretation of the data.

Abstract

Potentially, the introduction of exotic natural enemies or mass release of biological control agents may lead to unwanted non-target effects. Whether or not such effects occur will depend mainly upon the host range of the biological control agent and the presence of non-target species in the areas of release and dispersal. To predict non-target effects, risk assessments for release of exotic natural enemies have been developed and applied during the modern era of biological control. Although methods to determine host ranges of natural enemies have been proposed during the past decades, decisions about release of exotic natural enemies are often still based on short-term decisions strongly influenced by financial benefit and tend to ignore environmental ethics, especially where risks are difficult to quantify. Here, we propose a framework for host-range testing of arthropod biological control agents, and suggest methods for evaluating possible effects on those non-target species considered to be at risk. Several factors should be incorporated into a host-range assessment, including literature and museum records, field observations in the area of origin, as well as physiological, behavioural and ecological observations and experiments. Usually, laboratory-based manipulative experiments will form the core of host-range assessments. In this chapter we concentrate on the question of how to determine host ranges. Several important considerations involved in designing host-range testing are presented. Next, a framework for step-wise host-range testing is given with levels of increasing complexity that should allow over- and underestimation of the host range of a biological control agent to be avoided. Finally, the interpretation of data obtained with host-range testing is discussed and conclusions are drawn about the importance of host-range testing within the framework of future biological control projects.

Prey range of the predatory ladybird Cryptolaemus montrouzieri

“…These findings indicate that taxonomic relatedness in se may not necessarily be a sufficiently reliable criterion for determining prey ranges and even closely related prey may substantially differ in their suitability to support immature development and/or reproduction of a natural enemy. Furthermore, our study provides support for the hypothesis that in addition to non-target species that can easily be tested in a laboratory setting, prey range testing should give additional attention to economically important species, threatened or valued species and native natural enemies (Sands and van Driesche 2000;Babendreier et al 2005;van Lenteren et al 2006b).…”

“…This procedure starts with testing non-target species that are closely related to the target and then progresses to species that are more distantly related to the target organism. If none of the non-target species is attacked, one can stop testing (Wapshere 1974;Londsdale et al 2001;van Lenteren et al 2006b). In the present study, survival was high to moderate when C. montrouzieri was provided with prey species that are closely related to the mealybug target prey (M. persicae, A. pisum, B. tabaci) and overall poor to zero when the ladybird was provided with prey species that belong to a different insect order than the Hemiptera (F. occidentalis, A. bipunctata, G. mellonella) or even hemipteran prey from a different suborder (N. viridula).…”

“…Polyphagy of a non-native biological control agent can, however, also be disadvantageous when it spreads to natural ecosystems after being released and affects the abundance of non-target species, thus threatening local biodiversity (van Lenteren et al 2003;Loomans and van Lenteren 2005;De Clercq et al 2011). For this reason, the release of polyphagous biological control agents is not always considered to be environmentally safe and, therefore, an appropriate regulation concerning the import and use of natural enemies is required to prevent these undesired side-effects (van Lenteren et al 2006a;Bale 2011;Ehlers 2011). A scientific risk assessment methodology, integrating information on the potential of a biological control agent to establish (overwinter), its ability to disperse, its prey (or host) range, and its direct and indirect effects on non-target organisms, should form the basis of any regulation for biological control agents (van Lenteren et al 2003;van Lenteren and Loomans 2006;Ehlers 2011).…”

“…However, the development of a concrete experimental methodology to assess the prey range of arthropod natural enemies is still an ongoing process and several factors complicate this assessment, including the number of non-target species to be tested, the criteria for their selection, the need for reliable methods for determining prey acceptance and the discrepancy between prey ranges observed in the field and in the lab (Babendreier et al 2005;Kuhlmann et al 2006;van Lenteren et al 2006b;Hatherly et al 2009). In weed biological control, a survey across all potential host plant species in the native range of a herbivore is considered to be a good start in quantifying its degree of polyphagy (Manners et al 2011).…”

Harmonia axyridis: an environmental risk assessment for Northwest Europe

“…If the risk threshold is not crossed, the same procedure needs to be followed as for CBC natural enemies in step 4. At step 4, the host range issue (see van Lenteren et al 2006b) is addressed. If the ABC or CBC agent is either monophagous, or oligophagous/polyphagous and attacks only related AND no valued non-targets, i.e.…”