A better understanding of nutrient leaching in furrow irrigated agriculture is needed to optimize fertilizer use and avoid contamination of water supplies. In this field study (2003-2006), we measured deep percolation fluxes at 1.2-m depth and associated nutrient concentrations and mass losses from dairy manure nitrogen (N) or mineral N (urea, sodium nitrate [NaNO])-amended soils (372 kg available N ha in 4 yr) and nonamended controls and determined the δN-NO and δO-NO isotope ratios in the leached nitrate. Flow-weighted concentration means for individual irrigations varied widely, from near zero to as much as 250 mg L for NO-N, 480 μg L for dissolved reactive phosphorus (DRP), 43 mg L for dissolved organic C (DOC), and 390 mg L for chloride (Cl). Relative to other treatments, mineral fertilizer increased NO-N concentrations 2.6- to 3-fold and Cl concentrations 2.6- to 3.6-fold in deep leachate, particularly when NaNO was applied in 2004 and 2006, and produced maximum mean season-long NO-N and Cl losses. Manure and control treatments produced similar leachate nutrient mass losses, and for some irrigation periods, mineral fertilizer produced 85 and 97% lesser DRP losses and two times greater Cl losses compared with manure and control treatments. Four-year cumulative losses among treatments differed only for Cl. Isotopic composition of deep-leached nitrate indicates that both transformation and biologic cycling of mineral and manure N are rapid in these soils, which, with percolation volume, influence the amounts of NO-N and DOC leached. In light of the potential negative effects associated with either fertilizer type, and because even nonamended soils produced substantial amounts of leached NO-N (69.5 kg ha yr), management must minimize percolation water losses to limit nutrient losses from these fertilized, furrow-irrigated soils.