Prognosis of iron chlorosis from the mineral composition of flowers in peach

A statistical approach has been applied to study the nutrient composition of flowers in a peach orchard with 50 peach trees suffering from different degrees of Fe deficiency. Flowers from each individual tree were analysed in five crop seasons. The existence of a consistent nutrient balance over years in peach flowers, with nutrient concentrations depending to some extent on the levels of other nutrients, was revealed by a principal component-type analysis. The analysis produced a dominant first principal component, which explained 21% of the total variance. The second component explained 13% of the variance, with further components explaining less than 9%. A regression analysis revealed that the first principal component was closely related to chlorophyll (Chl) concentration determined in leaves 120 d after full bloom, as it explained 59% of average Chi concentration variance (between 40 and 52% in single year analysis), which is equivalent to a correlation coefficient of 0.77. A multiple regression model using the flower concentrations of N, P, K, Ca, Mg, Fe, Mn, Cu and Zn explained from 50 to 62% of the Chl variation among trees in the different seasons. Flower Fe was significantly correlated with Chi some years, whereas in other years the correlation was weak. Regression models including K and Zn explained 27 to 29% of leaf Chi concentration variance. This coefficient of determination is equivalent to a correlation coefficient of -0.52. This relationship was quite constant across years. We propose that the K/Zn ratio in flowers could be used, along with the flower Fe concentration, for the prognosis of iron chlorosis in peach trees.