Dissolved and particulate P loss with surface runoff may occur due to excessive rainfall, although the P loss is highly dependant on the P source, soil properties, timing of rain after manure application and intensity.  This study compared short-term runoff P loss when a similar rate of 50 kg P ha^-1 was applied with diammonium-phosphate fertilizer and liquid swine, poultry (egg layers or broilers), or beef feedlot manure.  Manure from several feeding operations and the fertilizer were applied to soybean or corn residue at 21 sites. Total P concentration in swine, poultry, and beef manure were 6-97, 24-35, and 6-12 g P kg^-1 on a dry weight basis, respectively, and water extractable P was 4-15, 5-9, and 1-6 g P kg^-1, respectively.  Simulated-rainfall (71 mm hr^-1) was applied to field plots until 30 minutes of runoff was collected.  Runoff P was largest for fertilizer (on average 11 mg DRP L^-1, 8 mg bioavailable P L^-1, and 16 mg total P L^-1) and lowest for beef manure (1 mg DRP L^-1, 1.25 mg bioavailable P L^-1, and 5 mg total P L^-1).  The P loss for swine manure usually was near that for fertilizer while losses for poultry manure varied greatly depending on the source and moisture content. The ranking for P loads was similar. Water-extractable P in the manure was poorly correlated with runoff P loss across manure sources. Results indicate that risk assessment indices of short-term P loss should account for the P source but tests of manure P solubility are unreliable estimators of risk of P loss across all sources.