Although water from 20 to 25% of shallow farmstead wells in northern Missouri has concentrations of nitrate (NO3/-) exceeding 10 mg L-1 as nitrogen (N), many potential sources for this NO3/- are usually present. A field experiment was designed to trace and isolate the amount of a single application of N fertilizer lost to a glacial-till aquifer and runoff from a 400 m2 corn (Zea mays L.) plot with bromide (Br-) and isotopically labeled (15N) fertilizer. Soil at the plot is a Albaquic Hapludalf of the Adco Series containing a 61 cm claypan beneath 41 to 43 cm of topsoil. Groundwater levels ranged from 0.38 to 2.40 m below the land surface. Transport of water and NO3/- to the saturated zone was not substantially retarded by the claypan. Labeled-N fertilizer accounted for as much as 8.6 mg L-1 of the NO3/- (as N) in groundwater, but only in the top 1 to 2 m of the saturated zone. After two growing seasons (16 mo), <2% of the labeled-N fertilizer was lost to runoff, about 30% was in the saturated zone, 27.3% was removed with the grain, and about 5% remained in the unsaturated zone. A large part of the remaining labeled N may have been lost in gaseous N forms. The presence of labeled NO3/- only in the top 2 m of the aquifer, slow horizontal transport, and winter recharge indicate grass crops such as wheat (Triticum aestivum L.) or rye (Secale cereale L.) might be used to extract near- surface N during the winter recharge period. Also, fall fertilizations can be expected to readily leach. Because groundwater concentrations of labeled NO3/- were still increasing after two growing seasons, rotation of crops requiring small N inputs could be expected to limit the cumulative effect of large annual fertilizer applications on groundwater.