Summary
This work is a search for the fishing effort unit more suitable to the analysis of data of catch and effort provided by the fisheries in the Amazon State. Five units of fishing effort are proposed. The statistical behavior of their distributions is examined. The best two are selected: (a) days of fishing X number of fishermen, and (b) number of fishermen only. Tests of comparisons of means (and medians) are carried out and the results are different for the two fishing effort units, (a) and (b). Later (a) is selected as the best unit according to the general fishing strategy in the region.