The microbial immobilization of ammonium and nitrate was measured by 15N organic measurements after the application of labelled urea, (NH4)2SO4, KNO3 (KN) or NH4NO3 with or without glucose in four different soils. In the soils incubated without glucose, the microbial immobilization of the added ammonium varied between 1.5 and 4 mg N kg−1 soil. No immobilization occurred at the expense of NO3 when KN was applied. When glucose was added at the rate 500 mg C kg−1 soil, the immobilization was very active between the first and the third day, at 10°C. The maximal amounts of 15N immobilized were much higher for the [15N]urea, 15(NH4)2SO4, 15NH4NO3 and 15NO3K. treatments than for the NH415NO3 application. This preferential immobilization of NH4 was also observed in pure cultures of bacteria isolated from one of the soils and attributed to the inhibition of nitrate uptake by ammonium. The immobilization ratio, immobilized N: decomposed C, was calculated for glucose, accounting for pool substitution effects and immobilization due to native C. It was independent of the form of N applied and similar between soils, c 45–48 mg N g−1 C.