Fruit Morphology and Clustering Analysis Reveals Diversity among Commercial Melons (Cucumis melo L.) in Indonesia

Melon (Cucumis melo L.) is a globally important horticultural crop known for its remarkable morphological and agronomic diversity. However, commercial varieties in Indonesia remain poorly characterized. Here, we evaluated 24 C. melo varieties using 14 quantitative and 4 qualitative fruit traits related to size, shape, rind, flesh, seed morphology, and sugar content. Analysis of variance revealed significant differences among varieties (p<0.05) for all traits. Principal component analysis (PC1 = 50.8%, PC2 = 36.7%) and hierarchical clustering identified three major phenotypic groups. Random forest modeling ranked trait importance based on %IncMSE, identifying seed cavity length, seed cavity index, and fruit diameter index as top predictors (importance score >10%). Integration of PCA, heatmap clustering, and variable importance plots confirmed these three traits as robust diagnostic markers. These findings highlight key morphological features for varietal classification and breeding, and provide a phenotypic framework for future genetic studies of melon in Indonesia.