Does a heterogeneous distribution of food or pesticide affect the outcome of toxicity tests with Collembola?

The reproduction of two closely related soil microarthropods, Folsomia candida Willem and Folsomia fimetaria L. (Insecta: Collembola), was tested under the influence of the insecticide dimethoate. Dimethoate had an adverse effect on the survival of adults and their reproduction in concentrations of about the recommended field dose, with F. fimetaria being more sensitive than F. candida. The experimental conditions were altered to evaluate the realism in the basic single species/single chemical reproductive test system. The importance of the spatial distribution of dimethoate was studied with food applied to the surface (original procedure), mixed homogeneously in the whole soil profile or only in the top layer, or mixed heterogeneously into the soil preserving the small granula of the yeast originally in the commercial formulation. Toxicity decreased significantly when exposure could be avoided in an uncontaminated bottom layer and even more if food was available in this soil horizon. But the results indicate that Collembola were not able to completely avoid dimethoate when they had the choice. For extrapolation purposes a simple test system may be sufficient as EC50 was changed less than one order of magnitude with the different test designs. In terms of EC50 the outcome of a toxicity test with a heterogeneous distribution of food and dimethoate was changed only slightly but the effects to suboptimally fed populations should be considered because they may be more vulnerable.