The AACC Approved Method for near-infrared reflectance (NIR) spectroscopy to produce a wheat hardness score for wheat market classification can be corrected for variation in wheat moisture content. The cause of the variation in NIR spectra resulting from variation in wheat moisture was investigated. 

Ten samples each of soft red winter, soft white winter, hard red winter, and hard red spring wheats were stored at 20, 40, 60, and 80 equilibrium relative humidity. Wheats were then ground on a cyclone grinder as required by the standard method. Variation in unground wheat kernel moisture content resulted in variation in NIR data. NIR log1/reflectance values increased at all wavelengths as wheat moisture content increased. Spectral changes were related to changes in the apparent particle size of ground wheat meal as it was influenced by moisture content. 

Higher moisture contents produced slightly higher apparent particle size in meal, suggesting larger particles of pericarp that became more pliable at higher moisture (temper) levels. The apparent particle size of meal of high moisture wheats resulted in greater NIR radiation scattering and decreased reflectance. Meal moisture content itself had no effect on the two NIR wavelengths used to evaluate wheat hardness.