G. Raykov
Department of General Agriculture and Herbology, Faculty of Agronomy, Agricultural University, 4000 Plovdiv, Bulgaria

(Manuscript received 27 February 2025; accepted for publication 14 April 2025)

Abstract. In effort to establish an approach to evaluate yield components to their importance for grain yield formation 118 winter wheat varieties ware used over three growing seasons (2013-2015) at DZI-General Toshevo, Bulgaria. Traits included in this study are grain yield (GY), Number of Grains per Square Meter (NGM), Thousand Grain Weight (TGW), Number of Productive Tillers (NPT), Number of Grains per Spike (NGS), and Plant Height (PH). The goal is to establish whether NGM can serve as a predictive tool in wheat breeding programs for enhancing yield. Significant G × E interactions were observed for all traits examined, with NGM displaying the highest interaction, indicating its potential as a marker for assessing genotypic variation across different environments. Linear mixed model was used to explore the predictive power of NGM on GY. Results showed that NGM had the highest conditional R² (0.942), suggesting that it explains a substantial portion of the variance in GY when considering both fixed effects (NGM) and random effects (Genotype, Environment, and their interaction G × E). When excluding the variation of Genotype, Environment and G x E interaction from predictive potential of the yield components NGM had the highest predictive capacity with marginal R² = 0.393. This is a sizeable portion of yield variability, supporting its use as a valuable tool in wheat breeding for yield improvement.