A non-equilibrium steady state box model is reported, that predicts the fate of new chemicals in a conventional sewage treatment plant from a minimal input data set. The model, written in an electronic spreadsheet (Lotus(TM) 123), requires a minimum input: some basic properties of the chemical, its emission rate and a few parameters to account for the scale and mode of operation of the plant. Compound properties necessary as input are air-water and sludge solids-water partition coefficients. If these distribution coefficients are not available but the compound is hydrophobic, default values are calculated from input such as solubility, vapour pressure and octanol-water partition coefficient. Biodegradability data and the degree of dissociation or protonation are the required input data to account for degradation and speciation changes of the chemical in the water phase. The modelled sewage treatment installation consists of primary sedimentation and the activated sludge technique. The standard output shows the chemical's concentration in effluent and sludge. In addition the output contains the relative amounts degraded and emitted via effluent, sludge and air. The main advantage of this approach is that it provides legislators with a quick insight if a chemical will be an air, water or sludge pollutant when it is discharged into sewage works at a certain rate. Examples are evaluated and results are compared with measured concentrations. Volatilization rates derived from reported field data are in good agreement with predicted emissions to the atmosphere.