Harris (1973) and Goeden (1983) proposed sets of criteria against which potential control agents could be scored and compared for selection for further study. These scoring systems generated considerable debate, have been widely tested, but have not been found universally useful. Because of the uniqueness of biological interactions, such systems tend to under-state the importance of some characters in relation to a particular weed or agent. On the other hand, characters that contribute to the overall score for another weed or agent may not be relevant.

Cullen (1996) discussed the above suggestions and concluded that "If we rely on any or all of these to guide our predictions, we run an enormous risk of excluding potentially effective species and of simply getting it wrong." He proposed that during research a series of questions be asked, hypotheses be formulated and tested, and the answers should help determine the proceeding hypotheses to test. The accumulation of questions and answers should aim to understand the three major factors that influence success. They are:

Such a systematic approach to understanding ecological systems is not new. It requires researchers with the training and resources to develop the appropriate questions and pursue the answers. As the number of case studies conducted increases, researchers will at least be able to better formulate the questions that when answered will provide useful information. Eventually, accumulated knowledge may allow answers to be predicted and generalisations to be developed concerning the best strategies to achieve successful biological control.

[ Back ] [ Next ]