Genetically engineered (GE) organisms could result in ecological harm in many ways in natural environments. Ecological harm can be assessed based on standard principles of risk assessment. Risk is the probability of harm as a result of a hazard, which in this case is a GE organism; its harm may not be known or knowable a priori, however, due to the large number of biotic interactions in nature in which it could be involved. We contend that for a GE organism to be a risk, it must be able to spread in nature. Thus, we do not have to determine all, or any, possible harms; we only need to be certain the organism will not spread if it escapes. Predicting the potential of a GE organism to spread is possible because the ultimate fate of a transgene will be determined by natural
Aquatic Sciencesselection. Thus, environmental risk assessment can be accomplished by measuring six net fitness components that are common to all organisms, transgenic or wild type: juvenile and adult viability, fecundity, fertility, age at sexual maturity, and mating success. These components can be measured in secure settings. Previously, we focused on the environmental risk posed by GE organisms created to enhance agricultural productivity. Here we review the potential of using GE biotechnology for biological control of an existing undesirable exotic species. GE biological control might be employed to induce a 'Trojan gene effect' (Muir and Howard, 1999; 2002a,b) to eliminate such species, by introducing genes which cause male-biased sex ratios, inducible fatality, or selfish gene effects.