We studied the use that young fishes make of vegetated and nonvegetated habitats in Lawrence Lake (Pool 8, Upper Mississippi River). Habitat use was both extensive and variable. Fish were found in all portions of the macrophyte beds. Backwater macrophyte beds both near to and distant from the main channel of the Mississippi River displayed similar patterns of habitat use. The open water site was used principally by gizzard shad and unidentified Cyprinidae. Habitat use in macrophyte beds with differing vegetation densities in Lawrence Lake was probably influenced by multiple factors. Physical and chemical microhabitat data explained little of the variation in catch in light traps. Two-stage sampling analyses indicated that when fish density is highly variable, sampling at many stations (one sample per station) is needed to assess habitat use in relation to vegetation density and species complexity. We evaluated several gears for both quantitative and nonquantitative sampling of juvenile fishes in vegetated habitats. From July through September, we conducted field tests to compare catches from light traps, fyke nets, hoop nets, gill nets, and minnow traps. Three quantitative sampling gears, an electrofishing frame, a drop net, and a pop net, were used. The electrofishing frame was modified with a drop-net mechanism to assess the capture efficiency of the electrofishing device and to serve as a drop net. Light traps, which principally sampled larval fishes, were effective in vegetated habitats. Catches with all three quantitative gears were dominated by juvenile bluegills. The electrofishing frame caught significantly fewer fish in turbid vegetated areas than the drop and pop nets, due to difficulty in observing and netting stunned fish. Capture efficiencies with the electrofishing frame were higher in clear nonvegetated areas (mean 80%) than in turbid vegetated areas (mean 5%). The drop net and pop net yielded similar catches in number and species composition. With the drop net, extensive seining (8-10 hauls) was required to collect all fish within the net enclosure. However, use of a removal method to estimate absolute catch decreased seining effort to three or four hauls, more comparable to that of the pop net. We recommend either drop nets or pop nets for quantitative sampling. Replicated tests with nonquantitative sampling gears were conducted to evaluate the effectiveness of fyke nets, hoop nets, small-mesh gill nets, and minnow traps in submergent vegetation and the effectiveness of small-mesh gill nets and hoop nets in an open water habitat. Of the nonquantitative gears tested, fyke nets produced the greatest numbers and species of fish. In open water, replicated sets of small-mesh gill nets and hoop nets showed that the gill nets were effective for capturing gizzard shad and golden shiners.