Metal pollution can be a serious threat to ecosystems at a global scale. Although the bioavailability of potentially toxic metals is determined by many biotic and abiotic factors, including pH and redox potential, total metal concentrations in the soil are used widely to assess or predict toxicity. In the present study we tested the effect of desiccation of soils differing in acidification potential and total heavy metal contamination on the growth and metal uptake of three typical, common wetland species: Caltha palustris, Juncus effusus, and Rumex hydrolapathum. We found that plant growth in wet soils mainly was determined by nutrient availability, though in dry soils the combined effects of acidification and increased metal availability prevailed. Metal uptake under anaerobic conditions was best predicted by the acidification potential (sediment S/[Ca + Mg] ratio), not by total metal concentrations. We propose that this is related to radial oxygen loss by wetland plant roots, which leads to acidification of the rhizosphere. Under aerobic conditions, plant metal uptake was best predicted by the amount of CaCl2-extractable metals. We conclude that total metal concentrations are not suitable for predicting bioavailability and that the above diagnostic parameters will provide insight into biogeochemical processes involved in toxicity assessment and soil policy.