Calculation of Heritability and Genetic Correlations between Yield Traits of Sugarcane (Saccharum Spp.) in Early Selection Breeding Populations in Louisiana.

Sugarcane (Saccharum spp.) cultivars take many years to be thoroughly evaluated. During the early stages of sugarcane cultivar development, genotypes are not evaluated in multiple replications and environments because of space and labor limitations. Genetic variability is difficult to quantify in non-replicated populations. The animal model which utilizes pedigree information is one model used to evaluate this type of data. Yield data from single plot second line trials in first and second ratoon including stalk number, stalk weight, cane yield, sugar yield (SH) and tons recoverable sucrose (TRS) were analyzed in a linear mixed model using the restricted maximum likelihood (REML). Narrow sense heritability values were lower for the first ratoon ≤0.1 and higher for the second ratoon ≤0.2 crop for stalk number, stalk weight and cane yield with the highest being stalk number. Although slightly higher in the second ratoon compared with the first ratoon cane crop, the heritability values were generally low ≤0.1 for sugar traits (SH, TRS). The genetic correlations between the ratoon crops were high for stalk number (0.73) and stalk weight (0.98). The genetic covariance between stalk number and stalk weight was negative but small (-0.16) with a larger than covariate standard error (0.37) and non-significant Pearson correlation. These results suggest that in second line trials, selection would be more accurate in the second ratoon.