ARS | Publication request: Repeatability and Genotype x Environment Interaction of Intermediate Stage Sugarcane Selection Conducted on Sand and Organic soils

Submitted to: American Society of Sugar Cane Technologists
Publication Type: Abstract Only
Publication Acceptance Date: May 19, 2009
Publication Date: July 1, 2009
Citation: Del Blanco, I.A., Glynn, N.C., Davidson, W.R., Comstock, J.C., Glaz, B.S., Edme, S.J., Irey, M. 2009. Repeatability and Genotype x Environment Interaction of Intermediate Stage Sugarcane Selection Conducted on Sand and Organic soils. American Society of Sugar Cane Technologists. volume 29:88

Technical Abstract: The Florida cooperative sugarcane cultivar development program conducts all of its early selection stages on muck (organic) soils at the USDA-ARS Sugarcane Field Station in Canal Point. About 25% of the locations in the final two stages (Stages 3 and 4) are conducted on sand soils, after a reduction in the number of genotypes from about 100,000 seedlings to 135 in Stage 3. This strategy might be overlooking the best sand-adapted genotypes earlier in the breeding scheme. To test this hypothesis an intermediate stage (Stage 2), with the same 1500 genotypes and a nested subset of replicated genotypes, was grown on a muck soil (Canal Point) and on a sand soil (United States - Sugar Corp. Townsite Farm, near Clewiston, FL). Genotype x environment interaction was significant for tons cane acre-1 (TCA), theoretical recoverable sucrose (TRS) (lb Ton-1), and tons sucrose acre-1 (TSA). Spearman correlations among traits measured on sand and muck were highly significant (p < 0.01), although low in value (r = 0.26 across traits), suggesting that specific selection is needed to target genotypes to the sand or muck environment. Means of TCA and TSA were higher for the muck location (75.5 ± 10.5 and 7.20 ± 1.1) than for the sand location (36.1 ± 16.1 and 4.6 ± 2.2). TRS was higher and less variable on sand (297.5 ± 9.5) than on muck (203.0 ± 20.5). Repeatability (R) was higher on muck than on sand soils for all traits, with R values 0.48 ± 0.1, 0.63 ± 0.08 and 0.46 ± 0.1 for TCA, TRS, and TSA respectively compared with respective R-values 0.43 ± 0.1, 0.47 ± 0.1 and 0.42 ± 0.11 on sand. Of 135 clones selected for advancement to Stage 3, 51 selections were common to both soil types, suggesting that some outstanding sand genotypes may be lost when intermediate selections are performed only on muck soils.