The efficiencies of conventional extraction techniques and analytical methods (HPLC-DAD and ELISA) were investigated for analyses of microcystins (MCs) in sediments. Our results showed several limitations. First, the extraction efficiency strongly depends on the extraction solvent, and extraction with 5% acetic acid in 0.2% trifluoroacetic acid (TFA)-methanol was confirmed as being the most appropriate for three different sediments (recovery: 33.1-44.9% of total MCs according to HPLC analyses). Second, the recovery of MCs was affected by the type of sediment but did not clearly correlate with the content of organic carbon. These results suggest that the sorption of MCs onto inorganic materials such as clay minerals is probably a more important process than interactions of the MCs with organic sediment matter. Third, the structure of the MCs is another crucial factor that affects the sorption of MCs and their recovery from sediments. Hydrophilic MC-RR gave much lower recoveries (20.0-38.8%) than MC-YR (44.1-59.5%) or MC-LR (55.3-77.8%) from three different types of spiked sediments. Recovery results analysed with HPLC-DAD correlated well with ELISA analyses. Further, extraction with 5% acetic acid in 0.2% TFA-methanol was used for analyses of MCs in 34 natural sediment samples collected from Brno reservoir (Czech Republic) from April to October 2005. Concentrations of MCs in sediments ranged from 0.003 to 0.380 microg/g sediment d.m. (ELISA results) or 0.016-0.474 microg/g d.m. (HPLC results). These values are equivalent to 0.63-96.47 microg/L of sediment (ELISA) or 4.67-108.68 microg/L (HPLC), respectively. Concentrations of sediment MCs showed both temporal and spatial variability, with the highest MC contents observed in the spring (April and May) and the lowest concentrations in July and August. Our results demonstrate the suitability of the methods described here for studying the occurrence, fate and ecological role of MCs in the aquatic environment.