Main content area

Effectiveness of an image-based sorter to select for kernel color within early segregating hard winter wheat (Triticum aestivum L.) Populations

Brabec, D., Guttieri, M.J., Pearson, T., Carsrud, B.
Cereal research communications 2017 v.45 no.3 pp. 488-499
Triticum aestivum, agronomic traits, artificial selection, crossing, hard red winter wheat, hard white wheat, image analysis, seeds, winter wheat
This study evaluated the effectiveness of an image sorter to select for kernel color within early generations of segregating hard winter wheat populations. The wheat crosses originated from different combinations of white and red parents. Three generations (F₃, F₄ and F₅) of sorting were applied to six segregating populations. At each generation, samples of whitesorted, red-sorted, and unsorted populations, along with the parents of the populations, were planted in replicated trials at multiple locations. The sorter processed 1kg sized samples in ~30 min and samples were sorted for 108 plots per season. ~10% of the F₃ populations were sorted and planted as white-sort population. This resulted in minimal changes in the % of white kernels. ~3% of the F₄ and F₅ populations were sorted and planted as white-sorted populations and significant advancement occurred. The F₆ populations of white-sorted samples from Dakota Lake ranged from 80% to 92% white kernels. The F6 populations from Brookings ranged from 53% to 83% white kernels. Sorting for red seed decreased the frequency of white seed as compared to the unsorted reference populations; however reductions, of white seeds in the red populations, were modest and required three cycles of selection for significant effect. The effectiveness of the image-sorter varied with population and environment and sorting methods.