Molybdenum (Mo) extracted from 31 New Zealand soils by an anion exchange resin method (resin Mo) was compared with that extracted by Tamm's oxalate (oxalate Mo). The amounts extracted by the two methods were correlated but, on average, oxalate Mo was 7 times resin Mo. Resin Mo was correlated with relative yield and Mo uptake of lucerne, although correlations were not high. Oxalate Mo was not correlated with plant data. Inclusion of pH in multiple regressions improved correlations with relative yield for both methods but the prediction equations accounted for only 30–35% of the variation. It is concluded that the resin method, although better than the oxalate method, is not sufficiently accurate for diagnosis of Mo deficiency’ in New Zealand pastoral soils.