Family Evaluation for Sugarcane Yield Using Data Estimated from Stalk Number, Height, and Diameter

Abstract

Family selection in sugarcane (Saccharum spp.), as practiced in Australia, was difficult to implement in South Africa because of the non-availability of mechanical harvesting and automatic weighing machines. This study explored the potential of evaluating cane yield of sugarcane families estimated using stalk number, stalk height, and stalk diameter as compared to individual seedling selection. Data for stalk number, stalk height, and stalk diameter were collected from Bruyns Hill and irrigated populations and analyzed using the Statistical Analysis System (SAS) to estimate variance components and best linear unbiased predictors (BLUP). A sample of 500 seedlings produced a significant association (r = 89, P < 0.0001) between actual and estimated cane yield. The results showed significant differences for cane yield among families. Family estimates of broad-sense heritability (H) and percent predicted gains (%Gs) were larger than for individual seedling selection. Elite families within populations were identified using BLUP. Data from four replications and 10 seedlings per plot were adequate for family evaluation.