Laboratory experiments were conducted to measure the functional response ofPodisus maculiventris(Say) in a simplified arena. Predators were isolated singly for 24 h in 9-cm Petri dishes with either 1, 2, 4, 8, 12, or 16 third-instar larvae of Mexican bean beetle (Epilachna varivestisMulsant). Results showed a typical type II response, with up to 9 prey attacked when 16 were provided. Holling's disk equation was fitted to the data to give estimates of handling time and the maximum number of prey that could be attacked in 24 h. The results differed sharply with previous field measures of predation byP. maculiventris, indicating that attack rates in a simple laboratory arena are limited by different behaviors than are important in the field. As a result we suggest that measuring functional response must be done in an appropriate test arena if the results are to be used to predict the impact of the predator on field populations of prey.