In this study, the effects of two metals, copper and cadmium, on the growth and development of juvenile Lumbricus rubellus were measured in a toxicity test in which individuals were grown in isolation. This design had a number of advantages over traditional test systems for earthworms. Importantly, the test is specifically designed to measure two juvenile traits (survival over and length of the juvenile period) that have been shown to have a high sensitivity for determining population growth rate. The test system also maximizes replication, while allowing time-series-based monitoring of individual growth. For both metals, significant exposure-dependent effects on survival, growth, development time, and (less certainly) maturation weight were observed. Comparisons of the relative toxicity of the two metals indicated different concentration-response relationships. For copper, hormesis was found at low levels, while only at the highest soil concentration tested (10.07 micromol g(-1)) were (severe) toxic effects present. For cadmium, hormesis was also evident at the lowest concentration tested; however, at soil levels above this, a graded concentration-dependent toxic effect was apparent. These differences in the exposure response patterns can be (tentatively) explained in terms of the mechanisms for handling copper (an essential metal for earthworms) and cadmium (a putative nonessential element). The applicability of the test for routine measurement of chemical effects on ecologically relevant juvenile traits is also outlined and future developments are discussed.