Abstract Radiance measurements of the potato crop at near-infrared (NIR) and red (R) wavelengths were related to ground cover, light interception and destructive growth measurements in 1986. The relationship between the ratio NIR/R and crop ground cover was dependent upon crop nitrogen (N) status, as a result of fertilizer applications increasing the chlorophyll concentration in leaves. Calculation of cumulative light interception, assuming a 1:1 relationship between light interception and ground cover, showed a curvilinear relationship with cumulative NIR/R (Σ-NIR/R ), with N application increasing the slope of the line. Values of Σ-NIR/R were related to total crop dry weight, giving a separate quadratic equation for crops grown with or without N fertilizers. These relationships were tested by predicting, retrospectively, the biomass of crops grown in 1985 with several different levels of fertilizer nitrogen. During the period of canopy expansion, radiometric data overestimated crop growth. After canopy closure, the mean differences between measured and estimated values of crop dry weight were not significantly ( P>0.1) different from zero, as long as the quadratic equation derived from the N-fertilized crops was used. The implications of using radiometric data for modelling crop growth are discussed.