After the confirmation of problems on cotton concerning boron nutrition in the State of São Paulo, Brazil, at the decade of 1970, experimental field tests were carried out and demonstrated the necessity of boron fertilization on deficient soils. With the purpose of evaluating the use of boron under different conditions, it was conducted a regional study through the São Paulo State cotton belt, during the period of 1979 to 1986. Rates of 0, 0.2, 0.4, 0.8, 1.6 and 3.2kg/ha of B as borax (11 % of B), were applied at planting time, in latin square design experiments. The variety IAC 17 was used during the four years, followed by the IAC 20. Boron fertilization consistently increased cotton yield, although statistically significant differences were obtained only in 20% of the events; the plants reacted better in terms of boron concentration in the leaf blade. Grouping the experiments by previous informations about use of liming and mineral manure, associated to the occurrence of boron deficiency or toxicity symptons, provided a good discrimination of B fertilization. The effect of boron was greater in tradicionally limed and fertilized (NPK) soils; it decreased on soils with acidity in phase of correction or corrected but previously fertilized with boron, and was negligible in crops after old pastures. There was a significant correlation relating cotton yield and soil boron determined with Mehlich solution (HCl 0.05M + H2SO4 0.0125M) or, specially, with hot water. Significant results were also observed in relation to yield and fifth leaf blade boron concentration. As a preliminary approach, analitycal values over ranges of 0.20-0.40 ppm and 25-40 ppm for soil boron extracted with hot water and for leaf blade boron concentration, respectively, were considered as indicative of the need for boron supplying in cotton fertilization.
boron; regional tests; cotton; yield and foliar analysis; soil analysis; correlations