EFEITO DE REGULADORES DE CRESCIMENTO SOBRE A ARQUITETURA DE PLANTAS E O RENDIMENTO DE SOJA

Abstract

Growth regulators have the potential to modulate soybean architecture and productivity, but their effects are complex and strongly dependent on interactions among the plant, the product, and the environment. This study comparatively evaluated eight regulators using a multivariate approach capable of integrating morphological, reproductive, and yield-related traits of soybean grown in the Brazilian Cerrado. The experiment was conducted in a randomized complete block design with four replications, with applications at V4, and ten agronomic variables were assessed, including plant architecture, pod classes, and yield. Although ANOVA using mixed models did not detect isolated statistical differences for most variables (p > 0.05), treatment-specific correlations revealed contrasting physiological patterns. Cluster analyses and the Smith–Hazel index proved highly informative, allowing the discrimination of three performance groups: recommended regulators (Stimulate, PR136, and Check), characterized by more balanced responses; intermediate regulators (StoppingGo and MaxCel); and non-recommended regulators (StartStop, Equaliza, and Biozyme), which showed inferior agronomic performance. Thus, even without robust univariate effects, the multivariate approach demonstrated a high capacity to distinguish the agronomic merit of the products, reinforcing that integrated analyses are essential to guide precise recommendations in soybean management. Further research should be conducted to confirm and expand these findings under different cultivation and management conditions.