Two simple techniques designed to measure nitrate (NO3-N) non-destructively in small undisturbed soil samples are described. A small water-filled ‘receptor’ is placed in contact with a small soil volume (~1·7 cm3) enabling the concentration of NO3-N in the interstitial water of the receptor to approach the concentration of NO3-N in the interstitial water of the soil sample by the process of diffusion. The NO3-N concentration in the receptor water is then measured, and from this, the NO3-N concentration in the soil is estimated. The receptor used in the first technique was a disc of agar (diameter 8 mm and thickness 3 mm). This was placed on the soil volumes and allowed to equilibrate for 22 h before NO3-N in the soil and in the agar were determined. The regression of actual soil NO3-N against estimated soil NO3-N using this receptor yielded an r2 value of 0·99 and a slope coefficient of 1·01. The main disadvantage of this technique was that it was necessary to bring the soil to a water potential close to saturation before the agar discs were applied. If the soil had even a slight negative water potential (–10 kPa), water would drain out of the agar rendering it useless and would also alter the soil water potential in the process. For this reason, this technique is only recommended for use with soils which are close to saturation. The second technique used ceramic discs as the receptors to overcome the deficiencies of the agar discs. The ceramic discs had dimensions similar to the agar discs and had a gravimetric water content of 0·25 g/g at a water potential of –30 kPa. They were equilibrated with the soil for a period of 10 days. The regressions of actual soil NO3-N against estimated soil NO3-N had r2 values >0·99, and the slope coefficients ranged between 1·08 and 1·11. It was concluded that this technique is a useful tool for the non-destructive measurement of NO3-N in small undisturbed volumes of soil.