AbstractThe relationship between catch and trawling effort was explored for Penaeus aztecus, P. setiferus, and P. duorarum. Tow durations of 5, 10, 15, 20, 25, and 30 min bottom time were examined, and data were analyzed with orthogonal polynomials in a randomized complete block design. We found a significant linear relationship between catch and tow duration, which is consistent with the assumption of a constant catchability coefficient. However, tow duration accounted for only a small proportion of the total variation in catch. This indicates that catch-per-unit-effort indices of abundance may be improved when analyzed in statistical designs that consider other sources of variation in catch in addition to fishing effort. Scaling of catch data when tow durations vary is briefly discussed.