Thirty years of exploration for and selection of a succession of Melanterius weevil species for biological control of invasive Australian acacias in South Africa: should we have done anything differently?

The question of how we can be simpler, faster and better in exploring for and selecting successful agents for weed biological control has been on the agenda since these Symposia began. We give a brief account of the development of some of these ideas on how to “pick a winner”. For about 30 years, South African scientists have made exploratory trips to Australia to select agents for release against various alien acacias, and the related Paraserianthes lophantha. Besides two species of gall-forming wasps, a rust fungus, and far more recently, a cecidomyiid pod-galler, five seed-feeding Melanterius weevil species were chosen, and have proved to be highly successful. Melanterius ventralis was released against Acacia longifolia (in 1985); M. acaciae on A. melanoxylon (1986); M. servulus on P. lophantha (1989); M. servulus on A. cyclops (1991); M. maculatus on A. mearnsii (1994), and on A. dealbata and A. decurrens (2001); and M. compactus against A. saligna (also in 2001). With reference to this singular group of weevils, the question is, in retrospect, whether we should or could have done anything differently? The basic ingredients for success in exploration and selection still require that the agents are available, amenable and appropriate (politically, climatically, and in their niche selection and ability to inflict critical damage), and that the agents must be acceptably host-specific, and sufficiently prolific and peripatetic. We conclude, as many others have before us, that successful agent selection is a serendipitous blend of biological and ecological knowledge, and pragmatic circumstances.