We used activity traps as designed by Riley and Bookhout (1990. Wetlands) to sample aquatic invertebrates as part of a study to characterize wetlands on a forested and an agricultural landscape (ca. 1,000 mi'2) in northern. Maine. Eight wetlands (5 from agricultural and 3 from forested landscapes) were sampled at random from 50 wetlands surveyed for waterfowl broods. At the landscape level, insect abundance (mean no./ trap), fish abundance (mean no./trap), percent vegetation, and water chemistry variables (pH, ANC, SPCOND, Ca, Mg, K, Na, Cl) were different between landscapes. Furthermore, nearly as many fish (2,112) were caught as were insects (2,443); 47% of the 332 traps contained fish, but 84 traps accounted for 94% of the fish caught. When >4 fish were in a trap fewer insects were in the trap. Differences in water temperature among wetlands and differences in rates of escape among insect orders affected the number of different taxa caught. Until capture success of activity traps is better understood, results from activity traps should be used with care.