Sampling Sorghum Plots for Kernel Protein1
- W. M. Ross,
- R. Ritter and
- J. W. Maranville2
When yields of sorghum [Sorghum bicolor (L.) Moench] plots are determined from unthreshed head weights, as is common in large breeding programs, grain quality analyses must be based on head sampling. Protein contents of kernel samples composited from midpanicle branches of five to seven heads per plot were compared with the protein in samples from threshed whole plots in two grain hybrid experiments of 100 and 120 entries each and in one experiment of 20 parental lines to evaluate the accuracy of head sampling.
Protein differences between sampling methods were low (0.5, 0.1. and 0.4%), but statistically significant within the three experiments. Simple correlation coefficients between sampling methods were highly significant (0.65, 0.70, and 0.83) for the three experiments.
The method ✕ entry Interaction was significant for protein only in the first hybrid experiment, probably due to sampling problems arising from selfed heads of certain female parents. In all cases, the entries ✕ methods mean squares were relatively low, which suggested that analyses of five to seven head samples using near-infrared reflectance would be satisfactory for screening homogeneous lines and hybrids for grain protein.Please view the pdf by using the Full Text (PDF) link under 'View' to the left.
Copyright © . .