Hydrologic connectivity and environmental variation can influence nekton assemblages in coastal ecosystems. We evaluated the effects of hydrologic connectivity (permanently connected pond: PCP; temporary connected pond: TCP), salinity, vegetation coverage, water depth and other environmental variables on seasonal nekton assemblages in freshwater, brackish, and saline marshes of the Chenier Plain, Louisiana, USA. We hypothesize that 1) nekton assemblages in PCPs have higher metrics (density, biomass, assemblage similarity) than TCPs within all marsh types and 2) no nekton species would be dominant across all marsh types. In throw traps, freshwater PCPs in Fall (36.0 ± 1.90) and Winter 2009 (43.2 ± 22.36) supported greater biomass than freshwater TCPs (Fall 2009: 9.1 ± 4.65; Winter 2009: 8.3 ± 3.42). In minnow traps, saline TCPs (5.9 ± 0.85) in Spring 2009 had higher catch per unit effort than saline PCPs (0.7 ± 0.67). Our data only partially support our first hypothesis as freshwater marsh PCPs had greater assemblage similarity than TCPs. As predicted by our second hypothesis, no nekton species dominated across all marsh types. Nekton assemblages were structured by individual species responses to the salinity gradient as well as pond habitat attributes (submerged aquatic vegetation coverage, dissolved oxygen, hydrologic connectivity).