Abstract Experiments were performed to determine Cu accumulation in earthworms under laboratory conditions using soil from a Cu-contaminated site, followed by field experiments in this contaminated site. The aims of the laboratory experiments were (a) to determine Cu accumulation rate, (b) to determine the effect of soil Cu content on the steady-state concentration, and (c) to evaluate the effect of soil moisture on accumulation. The field experiments were performed to evaluate the use of accumulation data obtained from laboratory experiments for prediction of accumulation under field conditions. In the laboratory experiment, earthworms ( Lumbricus rubellus) were introduced into four homogeneously mixed Cu-contaminated soils and a reference soil. The total extractable Cu content in the soil (Cu T) varied from 10 to 130 mg kg −1, soil pH varied from 4.0 to 5.0, and soil moisture content was set to approximately 25, 35, and 45% of the dry weight for each treatment. The tissue Cu concentration (Cu W) was determined by sampling earthworms after 1, 7, 14, 28, and 56 days. In the field experiment, 500 earthworms were introduced at four different Cu-contaminated locations at a contaminated arable field. After 14, 28, and 70 days, earthworms were sampled to determine Cu W. In both experiments, soil Cu contents significantly affected Cu W. Soil moisture only significantly affected Cu accumulation for the wettest soil. Under laboratory conditions, a steady state did not seem to be achieved after 56 days; the Cu accumulation can be described by the toxicokinetic one-compartment model. The field experiment was considerably affected by variation in soil temperature resulting in significant fluctuations in tissue Cu concentrations. The tissue Cu accumulation was significantly correlated to the Cu T, which is in agreement with the results from the laboratory experiments. Variance of Cu Wat Day 14 in earthworms from the field experiments was significantly larger than in the worms from the laboratory experiment. At Day 28, the differences were not significant.