In the 8(th) year of a long-term NPK fertilisation experiment set up in 1981, the effect of different nutrient supply levels and their combinations were examined on the mineral composition and element ratios of sugar beet. Checks were also made on the diagnostic limit values used to estimate plant nutrient status. The soil of the growing site was calcareous loamy chernozem with the following agrochemical characteristics in the ploughed layer: humus 3 %, CaCO3 5 %, clay fraction 23 %, pH(KCI) 7.3. On the basis of the available nutrient reserves the soil was judged to be poorly supplied with P and Zn, moderately supplied with N and K, and satisfactorily supplied with Fe, Mn and Cu. The factorial experiment consisted of 4N x 4P x 4K = 64 treatments in two replications, giving a total of 128 plots. The fertilisers were applied in the form of 25 % calcium ammonium nitrate, 18 % superphosphate and 50 % potassium chloride. The main results can be summarised as follows: The maximum beet yield (55.7 t/ha) and refined sugar yield(10.1 t/ha) were achieved at the following optimum concentrations in the leaf-blade at the end of June, in agreement with data in the literature: 4-5 % K, 3-4 % N, 0.3-0.4 % P. 10-15 N/P, 10-15 N/K and 1.0-1.5 P/K ratio. Over-fertilisation with N doubled the N and Na % of the beets compared with the control, thus spoiling their quality. Increasing P dominance doubled the P % of the beets and moderated the uptake of K, Mn and Zn, but was unable to counteract the negative effect of excessive N. With an improvement in the K supplies there was an increase in the K and Na concentrations and a drop in the N and Mg concentrations of the beets. An excess of K was able to counterbalance the negative effect of over-fertilisation with N to a certain extent. The K reserves of the petioles may exceed the K content of the leaf-blade by as much as 50-60 %, especially in the case of luxury supplies. The petioles stores K, so it could be used to diagnose the K supplies to the plant. Satisfactory supplies are indicated by a K concentration of 6-7 % in late June and 4-5 % in early August. NO3-N is stored in the petioles, in which 40 % of the total N may be present in nitrate form. The 1000 mg/kg nitrate limit value given in the international literature can be accepted as a satisfactory value for the achievement of maximum sugar content. In the case of repeated samplings it could help to control N nutrition during the vegetation period and to determine the harvesting order. The optimum ratios of the major nutrients could contribute to the estimation of balanced nutrition, make fertiliser recommendations more reliable and reveal whether any elements become limited during the vegetation period. The data presented in the paper could serve as guidelines for the extension service ( Tables 6, 7 and 8).