Changes in nekton community structure in north Brazilian intertidal mangrove creeks were studied by setting block nets at high water in the wet season 1999. From a total of 47 species (27 families), the most abundant were: Cathorops sp. (Ariidae), Colomesus psittacus (Tetraodontidae), Hexanematichthys herzbergii (Ariidae), Pseudauchenipterus nodosus (Auchenipteridae), the shrimp Fenneropenaeus subtilis (Penaeidae), Anchovia clupeoides (Engraulidae), Mugil spp. (Mugilidae), and Anableps anableps (Anablepidae). Diversity H' was low (0.9). Mean values of fish density and biomass (± S.E.) were 0.2 ± 0.1 fish m-3 and 2.6 ± 0.7 g m−3 or 1.4 ± 0.4 g m−2, respectively. The high proportion of juvenile fish (87%) and shrimps point to the importance of intertidal mangrove creeks for fish nurseries. The number of species was highest at spring tide-night and lowest at neap tide-day. Evenness J' was lowest at spring tide-night when more rare species entered and dominant species increased in quantitative proportions. Four different species assemblages were identified as being triggered by factor combinations such as “tide” and “time of day”: spring tide-night, spring tide-day, neap tide-night and neap tide-day. High-water level influenced the species assemblages, but salinity and water temperature did not. Abundance of 14 species correlated positively with the high-water level. Strong spring tides yielded extraordinary high nekton catches, suggesting that high-water level is an important parameter in the tidal migration of fish. Most fish entered the mangrove to feed. Predation pressure was considered to be low because of the presence of only a few, mainly juvenile predators. Beach seining in the subtidal parts at low water revealed a high biomass (30 ± 14 g m−2). Larger fish withdraw to depressions, small fish and juvenile shrimp were abundant near the shallow banks, suggesting inter- and intraspecific splitting of the migration routes when returning from the intertidal zone at ebb tide. To cover the entire short-term range of tidal ichthyofauna responses, a sample design should consider the spring/neap and the day/night cycle.