Slow-release 15N fertilizer formulations to measure N2-fixation by isotope dilution

Pot experiments were carried out to examine the effects of slow-release fertilizer formulations on estimates of N₂-fixation determined by the isotope dilution method. Soybeans were used as the N₂-fixing plants, with non-nodulated soybeans and maize as the non-fixing controls. The ¹⁵N-fertilizer formulations used were (¹⁵NH₄)SO₄, K¹⁵NO₃, gypsum-pelleted K¹⁵NO₃, (¹⁵NH₄)₂SO₄ + glucose, ground plant material enriched with ¹⁵N or ¹⁵N-oxamide. The estimate of the amount of N₂ fixed by the nodulated soybean plants depended upon both the control plant and the fertilizer formulation used. Maize took up N later than non-nodulated soybean and estimates of soil N-pool (soil “A” value + fertilizer N added) calculated from the enrichment of this control were about twice as large as those calculated from the enrichment of non-nodulated soybean receiving the same fertilizer treatment. As a consequence, estimates of N₂-fixation relative to this control were lower than those relative to non-nodulating soybean (mean 140 mg N per pot compared with 292 mg N per pot). With unstablilized ¹⁵N salts errors were sufficient to produce negative estimates of fixation relative to maize. Even with a “well-matched” control (non-nodulated soybean) estimates of fixation varied with fertilizer formulation.