In spite of the great effort that has been devoted to the search for a chemical laboratory
index to predict nitrogen (N) mineralization capability of soils, the results have not
yet been fully satisfactory. A continued effort is still needed to increase the knowledge
of the sources of variation that influence potentially available soil N. The time
of sampling has received little attention, taking into account its potential to influence
N-mineralization patterns. In this work, soil samples from three different agrosystems,
consisting of a double-crop sequence of small grains and maize, an intensively grazed
pasture, and a rainfed olive orchard, were collected at different dates. Several chemical
extractions were performed, and the results were correlated with N uptake by turnip
(Brassica campestris, L.) grown in a pot experiment. Kjeldahl N was the chemical test
that best correlated (R2 = 0.621) with N uptake by turnip. Kjeldahl N showed great
versatility relative to the origin of the soil samples. However, it was not very sensitive
to the time of sampling. It did not detect changes occurring in the soil over a short
period of time. Soil inorganic N showed the second highest coefficient of correlation
(R2 = 0.483) with N uptake by turnip. In contrast to that observed with Kjeldahl N,
soil inorganic N appeared as an index that can vary greatly over the short term. The
hot saline potassium chloride (KCl) extractions gave generally fair results. The poorest,
however, were obtained with the ultraviolet absorption of extracts of 0.01 mol L−1
sodium bicarbonate (NaHCO3) measured at 250- and 260-nm wavelengths.
