Chemical analysis is a crucial but often expensive and time consuming step in the characterisation of soils. Mid-infrared diffuse reflectance (MIR-DRIFT) spectroscopy coupled with partial least square (PLS) analysis was used to determine macro- and micronutrient concentrations of a range of alkaline soils from southern Australia. Solid phase associations of macro- and micronutrients were also investigated using the mineralogical information contained in the infrared spectra of soil samples. Results showed that MIR-PLS analysis is a powerful and rapid technique for the accurate prediction of more than 15 chemical properties from each soil sample spectrum. Correlation coefficients for MIR derived concentration versus laboratory determined values were greater than R2 = 0.80 for soil moisture, calcium carbonate concentration, total concentration of Mg, K, S, Fe, Al, Mn, Zn, Cu, and oxalate- extractable Al, Fe, Mn, and Si. In calcareous soils, sulfur was associated with carbonate and conversely Mg was more related to the clay concentration of soils. Micronutrients such as Fe, Zn, Mn, and Cu were positively associated with smectite/illite in the clay fraction and negatively with kaolinite. The potential use of these results in partitioning model to assess plant available micronutrients pools is discussed.