Effects of non-lint material on heritability estimates of cotton fiber length parameters

Fiber length is an important parameter to spin cotton fibers. Cotton breeders work to improve length, and research samples may be machine-harvested, resulting in variability for trash content between samples within the same experiment. There is evidence that trash may directly or indirectly affect the measurement of some fiber quality parameters like micronaire and strength. We hypothesize that the presence of trash particles in the samples may compromise the quality of the length measurement and screening decisions in breeding programs. In order to test this hypothesis, we developed an experiment to evaluate the heritability of length parameters for entries with the same genetic background and affected by the same environment with different trash content. The heritability estimates for samples with high trash content are different from the estimates for samples with native low trash content. Cleaning trashy samples with a laboratory-scale lint cleaner brings the heritability estimates closer to the calculated values for samples with native low trash content. Although the values are similar, the types of variation are not the same. These results indicate that breeders must avoid making decisions based on samples with high trash content. Breeders should base their decisions on samples with low original trash content because this type of sample provides research results closer to the native length distribution.