Excessive fertilization with organic and/or inorganic P amendments to cropland increases the potential risk of P loss to surface waters. The objective of this study was to evaluate the effects of soil test P level, source, and application method of P amendments on P in runoff following soybean [Glycine max (L.) Merr.]. The treatments consisted of two rates of swine (Sus scrofa domestica) liquid manure surface-applied and injected, 54 kg P ha(-1) triple superphosphate (TSP) surface-applied and incorporated, and a control with and without chisel-plowing. Rainfall simulations were conducted one month (1MO) and six months (6MO) after P amendment application for 2 yr. Soil injection of swine manure compared with surface application resulted in runoff P concentration decreases of 93, 82, and 94%, and P load decreases of 99, 94, and 99% for dissolved reactive phosphorus (DRP), total phosphorus (TP), and algal-available phosphorus (AAP), respectively. Incorporation of TSP also reduced P concentration in runoff significantly. Runoff P concentration and load from incorporated amendments did not differ from the control. Factors most strongly related to P in runoff from the incorporated treatments included Bray P1 soil extraction value for DRP concentration, and Bray P1 and sediment content in runoff for AAP and TP concentration and load. Injecting manure and chisel-plowing inorganic fertilizer reduced runoff P losses, decreased runoff volumes, and increased the time to runoff, thus minimizing the potential risk of surface water contamination. After incorporating the P amendments, controlling erosion is the main target to minimize TP losses from agricultural soils.