An extensive scientific literature discusses the attributes a successful insect biological control agent should possess (Harris 1973; Beddington et al. 1978; Goeden 1983; Hokkanen and Pimental 1984; Dennill 1988; Myers et al. 1989; Waage 1990). Harris (1973) was the first to propose a formal system for prioritising insects. This system was based on biological attributes including phenology of attack, number of generations per year, feeding behaviour and size. Each attribute was scored, with high fecundity, for example, being given a higher score than low fecundity. Scores for each attribute were added to give a total score for each species. However, this system and subsequent revisions (Goeden 1983; Hokkanen and Pimental 1984) are rarely used.

It seems that there are no proven scientific principles which can be used to determine in advance which biological control agents are going to establish and become abundant. Cullen (1992) comments that "this is a continual source of frustration and a waste of resources, yet attempts to do better are notoriously difficult and make little progress, to the extent that many workers feel it is not worthwhile, preferring to rely on release of the agent as the only valid test of finding whether it will be successful."

[ Next ]