Indirect Selection Efficiency of Sugarcane Genotypes for Organic and Mineral Soils In Florida.

Sugarcane (Saccharum spp.) genotype selection during the past 35 yrs has been more successful at developing cultivars for the organic (muck) than for the mineral (sand) soils in Florida, more likely a consequence of conducting the early and intermediate stages of selection exclusively on organic soils. The cultivar development program at Canal Point (CP) needs to improve cultivar selection for sand soils which comprise about 20% of Florida’s sugarcane. The objective of this study was to compare the relative efficiency of indirect selection on muck and sand soils for several traits using two CP Series (CP06 and CP07), of the second clonal selection stage (Stage 2), each planted in alternating years on muck and sand. Genotype selection for several traits was evaluated in two plantings of the same Stage 2 on muck soils at the CP-Sugarcane Field Station (USDA-ARS) during 2006 and 2007, and on sand soils at Townsite Farm (US Sugar Corporation), near Clewiston, FL during 2007 and 2008. Stalk number, stalk weight, theoretical recoverable sucrose (TRS), tons of cane per acre (TCA), and tons of sucrose per acre (TSA) were measured and correlated. Relative efficiency of indirect selection for each trait was estimated as the product of its sand-muck genetic correlation with the ratio of its sand and muck repeatability. Relative efficiency under indirect selection for each trait on muck was more effective for all traits with 2-yr means (muck vs. sand) of 0.38 vs. 0.22 for stalk number, 0.52 vs. 0.44 for stalk weight, 0.43 vs. 0.24 for TRS, 0.21 vs. 0.16 for TCA, and 0.17 vs. 0.14 for TSA. The low values of indirect selection for all traits, except stalk weight, suggest that selection of high yielding genotypes for adaptation to sand soils would be improved by direct selection of Stage 2 on a sand soil.