Manganese (Mn) toxicity in crops is a serious problem in Hawaii, especially Kauai and Oahu, where most soils are highly weathered. To devise a management strategy to control Mn toxicity, a greenhouse experiment was conducted on an acid (pH 4.4) Oxisol (Wahiawa series) having 15g·kg-1 total Mn. Factorial combinations of lime (0, 2.0, and 4.0 g·kg-1 CaCO3) and two composts (made from chicken manure and from sewage sludge at 0, 5, and 10 g·kg-1) were applied to the soil, which was subsequently planted to watermelon (Citrullus lanatus Thunb. `Crimson Sweet'). Our preliminary results showed that: 1) liming reduced Mn extractability and phytoavailability, but the reduction in Mn per unit increase in pH was much less than predicted by theory; 2) for good watermelon growth, soluble Mn, as extracted by the saturated paste method, should be <2.0 mg·L-1 corresponding to a soil pH >5.7; 3) unlike the saturated-paste extractable Mn, the Mehlich3-extractable Mn varied less with pH in a given soil series than between soil series; 4) effects of composts on Mn toxicity varied with compost properties, especially their Ca content and pH altering capacity; and 5) the diagnostic criteria for Mn toxicity in watermelon are tentatively proposed as: leaf Mn >1000 mg·kg-1 and leaf Ca/Mn ratio (g·g-1) <25.