In December 1982 a devastating flood occurred in Piney Creek, in the rural Ozark Mountains in Izard County, Arkansas, with vertical stage rises of 2–4 m in the headwaters, 11–12 m at downstream locations, and an estimated return time of 50–100 years. Physical effects in the watershed were catastrophic, with extreme scour and rearrangement of the stream bed, destruction of riparian forest, and deposition of huge amounts of sand from the creek in adjacent pastures or forest. In spite of the extreme nature of this winter flood, residual effects on the overall fish community of the watershed were minimal, and by eight months after the event, the community was virtually indistinguishable from that in the previous summer. In March–April 2008 flooding of equal or greater magnitude than the 1982 flood again occurred in Piney Creek. We followed effects of the spring 2008 flood on local fishes at five long-term fixed sites in the watershed, and on the fish community pooled across those sites, four months after the 2008 flood, and again in 2010 and 2012. In spite of the severity of the 2008 springtime flood, the community before and after was relatively similar qualitatively and quantitatively. But multivariate analyses of the fish community showed more change after the 2008 flood, and in a directional trajectory, than had occurred after the 1982 flood. At the five individual sites, changes in fishes after the 2008 flood were idiosyncratic, with two sites showing marked changes immediately after the flood, with only one subsequently returning toward its former structure. Fishes at all five sites showed more change in multivariate space after the 2008 than after the 1982 flood. In the summers after both floods some cyprinid and catostomid species showed sharp increases in numbers of young-of-year. Differences in the effects of the two floods on the fish community could relate to their timing, with springtime flooding having more effects on fish than the winter flood. Similarities between the two floods with respect to increased production of young-of-year could relate to the scouring of fines (silt and sand) by the floods, providing clean gravel and cobble with more interstitial spaces that could provide protection for fish eggs or larvae, and more microhabitat for food organisms used by young fishes such as micro- or macroinvertebrates. Regardless of mechanisms, much remains to be learned about the effects of extreme floods on stream fish communities, particularly in light of the potential for increased frequency of extreme events as global climate changes continue.