Two biological monitoring studies were carried out among workers in an explosives factory who were exposed to technical grade dinitrotoluene (DNT). In the first study urine samples from 28 workers were analysed for the metabolite 2,4-dinitrobenzoic acid (2,4-DNBA). Metabolite concentrations in urine were extremely low or non-detectable, prior to starting work at the beginning of the working week, but post-shift urine samples contained a mean 2,4-DNBA level of 17 mg/l. There were wide variations in the concentrations excreted in urine by different workers and by individual workers on consecutive days. Atmospheric levels of DNT (determined by personal monitoring) were found to be well below the recommended limit and therefore could not account for the observed excretion of 2,4-DNBA. It is suggested that skin may be the major route of absorption of DNT during this process. A second study was carried out to investigate the kinetics of absorption and excretion of DNT. Intensive urine sampling was carried out on five individuals over a 2-day period with additional samples over the subsequent 2 non-working days. Analysis for 2,4-DNBA showed that uptake of DNT is rapid and that the highest levels were normally seen in the end-of-shift specimens. Urine samples were analysed for other known metabolites of DNT which have been found in animal studies and it was shown that 2,4-DNBA is the major known metabolite which is excreted in human urine. Unchanged DNT was detected in blood samples taken during a single workshift at levels up to 250 ng/ml. It is concluded that there is a measurable uptake of DNT by the group of workers studied. The measurement of 2,4 DNBA in end-of-shift urine samples is an appropriate way of assessing the effectiveness of measures taken to reduce the absorption of DNT.