Abstract Targeted use of animal manures as a nitrogen (N) fertilizer is challenging because of their poorly predictable N fertilizer value. An enhanced understanding of their N transformation processes in soil under field conditions is necessary to better synchronize N availability and crop N demand. 15 N labeled cattle slurry, produced by feeding a heifer with 15 N labeled ryegrass hay, was used in an on-farm trial on two neighboring fields, cropped with maize or grass-clover, in order to assess crop N uptake and N dynamics in the topsoil. Recovery of applied total N in plant biomass was higher for mineral fertilizer (Min) (45–48%) than for slurry (Slu) (17–22%) when applied at the same rate of mineral N. Also, N derived from fertilizer in plant biomass was higher for Min than for Slu, due to both greater NH 3 emissions and greater initial immobilization of slurry N. Despite initial differences between the two in the relative distribution of residual fertilizer N in soil N pools, already in the following spring the majority (77–89%) of residual N from both fertilizers was found in the non-microbial organic N pool. Of the applied total N, 18–26% remained in the topsoil after the first winter for Min, compared to 32–52% for Slu. Thus, the proportion of fertilizer N not taken up by the first crop after application, enters the soil organic N pool and must be re-mineralized to become plant available.