Phosphorus reserves in Ultisols are inherently low; but many Ultisols along the Atlantic seaboard are now high in P, both extractable and total, because P additions have exceeded P removal for many years. How long a high-P soil will maintain plant-available P above yield-limiting levels is of agroeconomic relevance. A field experiment initiated 35 yr ago on Portsmouth soil (fine sandy over sandy or sandy-skeletal, mixed, thermic Typic Umbraquult) and monitored for crop yields and soil-test P (Mehlich-1 extractant) during 8 yr of active P buildup and 26 yr of residual decline has provided quantitative data on this issue. Yields of corn (Zea mays L.) or soybean [Glycine max (L.) Merr.] were maximal with soil-test P greater-than-or-equal-to 22 g m-3, and extractable P was maintained in this range (20-24 g m-3) when P removed in harvested products (16 kg ha-1 yr-1) was replaced annually as band-applied fertilizer. High soil-test levels (greater-than-or-equal-to 50 g m-3) could not be maintained by annual replacement of crop-removed P because P reversion to unextractable forms was a larger factor than crop removal in depleting the extractable-P pool. Regardless of initial level, P disappearance into these unextractable forms was best described via equations having the form of a first-order chemical reaction; but the magnitude of the rate constant varied with size of the extractable-P pool, i.e., high-P soils have large rate constants; low-P soils have small rate constants. A Portsmouth soil testing 50 to 60 g P m-3 today will test above 22 g m-3, the approximate critical level for corn, for the next 8 to 10 yr without further P additions. Doubling the initial soil test will not double the time to reach yield-limiting P levels, however, the same soil with 100 to 120 g P m-3 initially will drop to 22 g m-3 in about 14 yr.
