ABSTRACT
A review is made of the (extensive) recent literature on field validation of predictions based on laboratory toxicity test data. The handicaps associated with the use of laboratory bioassays as surrogates for toxic threshold determination in the field are discussed. The potential and limitations of single species and multispecies laboratory tests are outlined and the ability of these two approaches to reduce the uncertainty associated with lab to field extrapolations are evaluated. Additionally, the various sources of uncertainty in extrapolations are treated and several conceptual frameworks and mathematical methods for quantifying this factor are reviewed. One of the most striking conclusions of this literature study is that, in general, “NOEC’s derived from (a selected battery of) single species laboratory tests relate relatively well to the single species and multispecies NOEC’s obtained in field studies.” This statement is exemplified by several case studies. Finally, a pragmatic approach for the use of single species toxicity data for hazard assessment purposes is proposed and suggestions for improving the predictive capacity of laboratory toxicity tests are made.
