Predicting ecotoxicological impacts of environmental contaminants on terrestrial small mammals.

This review examines whether the effects of environmental contaminants on wild small mammals can be predicted from the results of single-species, laboratory toxicity studies. Heavy metals, organochlorines, chlorinated aromatic hydrocarbons, and OP/carbamate pesticides were identified as the groups of xenobiotics for which there are toxicity data for terrestrial small mammals and that, on the basis of persistence, acute toxicity, and bio-accumulation potential, present the greatest hazard to wild mammals. Laboratory-generated toxicity data, which used lethality and reproduction as measurable endpoints, were reviewed and intake and residue LOAELs estimated for representative chemicals (lead, endrin, PCBs) from the heavy metal, organochlorine, and chlorinated aromatic hydrocarbon substance groups; the OPs and carbamates were reviewed as a whole. Intakes and residues of these compounds in wild small mammals were compared with laboratory-defined LOAELs and the likelihood of effects predicted. The accuracy of these predictions was examined and the efficacy of extrapolating toxicity data from laboratory to wild species assessed. Qualitative extrapolation from laboratory to wild species was good for all the chemicals considered, laboratory tests correctly identifying the types of effects chemicals had on a wide range of wild mammals. In contrast, the quantitative extrapolation of dose-response data was either poor or largely unvalidated. This is because interspecies variation in sensitivity to xenobiotics and the effects on toxicity of differences in exposure pattern between laboratory and wild species are largely unquantified. Based upon the limited evidence available, errors in the direct extrapolation of dose-response data from laboratory to field may be as large as three orders of magnitude. Direct extrapolation of residue-response data from laboratory to wild mammals is good both for the effects of heavy metals on specific organs and for residues and acetylcholinesterase inhibition associated with pesticide-induced mortality. The use of organ residues or biomarkers to predict the severity of sublethal effects on reproductive output may be possible, although large residues or biomarker responses are not necessarily indicative of the severity of wider physiological effect. Appropriate residues/biomarkers may differ for various xenobiotics and even between species for the same xenobiotic. Further research is required to identify suitable markers that can be correlated with the occurrence and magnitude of ecologically important effects. Xenobiotics likely to have a direct effect on population dynamics are those that are persistent and adversely affect survival and reproduction. At present, this weak correlation is the only one that can be made between single-species laboratory tests and population effects.(ABSTRACT TRUNCATED AT 400 WORDS)