Inter-decadal periods of high and low flood activity have been considered to be the dominant factor driving river metamorphosis in catchments along the New South Wales coast. Recent work has questioned the data analysis techniques used in delineating the so-called flood and drought-dominated regimes (FDRs/DDRs). Concerns have also been raised about the validity of invoking a climatic control for river metamorphosis documented during the post-European period, when extensive anthropogenic alteration of catchment and riparian vegetation has also occurred. This paper reviews the evolution of the FDR/DDR concept. We examine the evidence for FDRs/DDRs, and highlight problems with the original hydrological data sets, as well as with the techniques employed in the time-series analysis. We discuss conceptual problems encountered in applying flood-frequency analysis, and the failure of the proponents of the FDR/DDR theory to consider large-scale climatic circulation patterns and the geographical boundaries of their influence. We conclude that the validity of the FDR/DDR notion has been seriously over-stated, and that managing rivers on the basis that FDRs/DDRs have occurred in the past, and will continue to occur in the future, is likely to be ineffective.