Early-tier studies are the initial step in the environmental risk assessment of genetically engineered plants on nontarget arthropods. They are conducted in the laboratory where surrogate species are exposed to higher concentrations of the arthropod-active compound than those expected to occur in the field. Thus, early-tier tests provide robust data and allow to make general conclusions about the susceptibility of the surrogate to the test substance. We have developed an early-tier test for assessing the toxicity of orally-active insecticidal compounds to larvae of the ladybird beetle Coccinella septempunctata L. (Coleoptera: Coccinellidae). Using potassium arsenate and the protease inhibitor E-64 as model compounds, we validated the bioassay set-up for C. septempunctata. Sucrose solution containing the test compound was offered to larvae for 24 h on the first day of each of its four larval instars. Subsequently, larvae were fed ad libitum with Ephestia kuehniella (Zeller) (Lepidoptera: Pyralidae) eggs. Both compounds negatively affected C. septempunctata larval survival and development, and adult weight, indicating that the bioassay setup was able to detect dietary effects of insecticidal substances on the ladybird. Power analyses revealed that sample sizes of 20 or 45 per treatment are sufficient to detect 50 or 20% differences between the control and treatment groups, respectively, for the various measurement endpoints.