The present study assessed the application of tritiated leucine incorporation into protein, as a measure of bacterial biomass production, within four benthic habitats of a littoral freshwater wetland dominated by emergent vegetation. Basic assumptions underlying the method, such as linearity of leucine incorporation, saturation level of incorporation rates, and specificity of incorporation for bacterial assemblages, were tested, and two procedures for extracting radiolabeled protein were compared. TCA precipitation followed by ultrasonication, and subsequent alkaline dissolution in 0.5 M NaOH, 25 mM EDTA, and 0.1% w/v SDS, gave best results in terms of both extraction efficiency and signal-to-noise ratio. Incorporation of leucine was linear for all habitats for up to 1 h. Saturation concentrations of leucine incorporation into protein were 150 nM for littoral surface waters, >960 nM for biofilms on plant surfaces, and 50 mM for aerobic sediment and submerged plant litter. An experiment with prokaryotic and eukaryotic inhibitors designed to examine specificity of leucine incorporation into bacterial protein showed no significant leucine incorporation into eukaryotes during short-term incubations. Calculations based on kinetic parameters of fungal leucine uptake suggest, nevertheless, that significant leucine incorporation cannot be ruled out in all situations. Thus, the leucine methodology can be used for estimating bacterial production in benthic aquatic habitats, provided that substrate saturation and isotope dilution are determined and that the active biomass of eukaryotes, such as fungi, does not greatly exceed bacterial biomass.