Collembolan reproduction in soils from a long-term fertilisation experiment opposes the Growth Rate Hypothesis

The Growth Rate Hypothesis (GRH) predicts better performance of fast-reproducing species in environments with increased P content. Thus far, most studies were performed in aquatic ecosystems; only few studies focused on soil ecosystems. In this study, soils from a long-term experiment (Static Fertilization Experiment, Bad Lauchstädt, Germany) that differ in P content by the factor 3 were used. We tested the influence of NPK fertilised, PK fertilised and unfertilised soils on the reproduction of Folsomia candida. In order to evaluate the effect of unit size, we compared the collembolan reproduction test as recommended by the OECD (large unit) with a recently introduced miniaturized version (small unit). Furthermore, the tests were combined with a predator-prey relationship using the gamasid mite Hypoaspis aculeifer. Even though significant differences between soils were found, reproduction was lowest at highest P content, which contradicts the GRH. In addition, predation and unit size had a strong influence on the reproduction. Both unit sizes proved feasible in predator-prey experiments, with stronger effects of mites on reproduction in the small unit due to higher relative predator density.