Introduction

Host specificity screening is the most important step that each potential weed biological control agent (whether insect or pathogen) has to pass prior to its introduction (Schroeder, 1983). Regardless of its control potential, only the demonstrated safety of plants of economic or ecological importance in the release area, will lead to approval for release of the control agent. The use of screening-protocols for weed biological control agents (Wapshere, 1975; Schroeder, 1983; CAB, 1986) are now widely accepted. Even less rigorous tests in the first part of this century have never led to the release of a ‘new pest’ (Crawley, 1989; Howarth, 1991).

The safety record of weed biocontrol using insects is excellent, nonetheless, the growing awareness of the public towards environmental interference by man, has led to legislative regulations such as the ‘Biological Control Act of 1984’ in Australia. Today, information on the environmental impact, alternatives, the relationship between local short-term uses and long-term productivity, and a risk analysis of the proposed action are required. Conflicts of interest over the implementation or safety of a specific biological control programme or agents, may still arise (Delfosse, 1990).

During the last decades especially, the critical objections of nature conservationists who fear for already stricken and endangered plant species related to the target weed, have considerably delayed the progress of programmes (Schroeder and Goeden, 1986). Despite the excellent safety record of weed biocontrol, we have to assess the risks of introducing a foreign species into a new environment.