Excessive manure phosphorous (P) application increases risk of P loss from fields. This study assessed total runoff P (TPR), bioavailable P (BAP), and dissolved reactive P (DRP) concentrations and loads in surface runoff after liquid swine (Sus scrofa domesticus) manure application with or without incorporation into soil and different timing of rainfall. Four replicated manure P treatments were applied in 2002 and in 2003 to two Iowa soils testing low in P managed with corn (Zea mays L.)-soybean [Glycine max (L.) Merr.] rotations. Total P applied each time was 0 to 80 kg P ha(-1) at one site and 0 to 108 kg P ha(-1) at the other. Simulated rainfall was applied within 24 h of P applications or after 10 to 16 d and 5 to 6 mo. Nonincorporated manure P increased DRP, BAP, and TPR concentrations and loads linearly or exponentially for 24-h and 10- to 16-d runoff events. On average for the 24-h events, DRP, BAP, and TPR concentrations were 5.4, 4.7, and 2.2 times higher, respectively, for nonincorporated manure than for incorporated manure; P loads were 3.8, 7.7, and 3.6 times higher; and DRP and BAP concentrations were 54% of TPR for nonincorporated manure and 22 to 25% for incorporated manure. A 10- to 16-d rainfall delay resulted in DRP, BAP, and TPR concentrations that were 3.1, 2.7, and 1.1 times lower, respectively, than for 24-h events across all nonincorporated P rates, sites, and years, whereas runoff P loads were 3.8, 3.6, and 1.6 times lower, respectively. A 5- to 6-mo simulated rainfall delay reduced runoff P to levels similar to control plots. Incorporating swine manure when the probability of immediate rainfall is high reduces the risk of P loss in surface runoff; however, this benefit sharply, decreases with time.