Cacao floral traits are shaped by the interaction of flower position with genotype

Understanding the factors influencing cacao flower morphology and abundance is essential for optimizing productivity. This study investigated the influence of vertical flower position and associated environmental and physiological gradients on these traits across two cacao genotypes, CCN51 and SCA6, under controlled greenhouse conditions. We measured flower size (lateral area, length, width, and perimeter), shape, and abundance at different developmental stages and vertical tree heights. Significant variations were observed between genotypes and across vertical positions, highlighting the roles of genetic and environmental factors in cacao reproductive biology. For example, CCN51 exhibited significantly larger flowers, with an average area of 18.63 mm² compared to 14.88 mm² for SCA6. Leveraging machine learning techniques, particularly Support Vector Machine, we successfully predicted genotypes based on flower measurements with high accuracy, achieving an area under the receiver operating characteristic curve of 0.87. These findings emphasize the phenotypic diversity of cacao flowers and demonstrate the potential of machine learning in genotype identification, offering valuable insights for breeding and cultivation strategies to enhance cacao productivity.