Abstract
While quantitative measurements of wheat (Triticum aestivum L) kernel hardness are important for market classification of cultivars, their genetic relationship to end‐use quality in breeding populations is not well established. After verifying that divergent selection for hardness score (HS) based on near‐ infrared reflectance (NIR) spectroscopy was effective, the objective was to determine correlated selection responses in milling and flour quality of two hard red winter populations differing widely in parental origin. Selection was applied in the F3 generation using replicated field plots at two locations. Selection response was evaluated in the F4 generation at the same locations the following year. Selection for high HS (harder kernels) increased kernel protein concentration in both populations, while low HS selection decreased it. Selection for HS had no consistent and detectable impact on flour yield or physical dough properties (mixograph absorption, mixing time, and mixograph rating or tolerance). Selection for high HS decreased SDS sedimentation volume adjusted for flour protein concentration in both populations, but the magnitude of the response was small (−1.7 ml in actual units; −0.3 ml after adjustment). Because correlative effects of NIR hardness were primarily expressed in protein quantity and not protein quality, milling and flour quality must be considered independently of NIR hardness if genetic improvement in those traits is desired.