Pelagic total phosphorus (TP) concentrations in Lake Okeechobee were shown previously to be correlated with lake water levels, possibly due to littoral flooding and subsequent P release from vegetation (Canfield and Hoyer 1988). The present paper reviews the research to test this and other hypotheses to explain the water level – TP relationship. Research in the littoral region indicates that during low to intermediate water levels, there is little water or nutrient exchange between the pelagic and littoral regions, and that under high water levels, the littoral is a P sink. These findings lead to a rejection of the littoral flooding hypothesis. Alternative hypotheses, concerning internal loading from pelagic sediments, have been suggested. Maceina and Soballe (1990) proposed that wind, rather that water level, controls TP – in windy years, there is frequent resuspension of P-rich sediments into the water column. This hypothesis is supported by results of hydrodynamic model simulations, and by studies of sediment composition and sediment-water P transport. Also, there is evidence that horizontal P transport is enhanced by a combination of high wind and high water levels (Maceina 1993). There is an additional effect of high water levels that could increase TP concentrations – a greater thermal stability of the water column during summer could result in anoxic bottom waters and subsequent P release. Research results showing that Fe regulates soluble P transport between sediments and water in Lake Okeechobee, and findings that this process is a key one affecting internal loading in other shallow lakes, are supportive of this hypothesis.