Cereal Chem 59:510 - 511. | VIEW
Near-Infrared Reflectance Estimates of Malt Extract.
C. F. McGuire. Copyright 1982 by the American Association of Cereal Chemists, Inc.
Traditional laboratory methods used to test malting quality of barley are poorly adapted to rapid screening of early generation breeding lines in a plant-breeding program. Selection of lines that are potentially high in malt extract would help to eliminate undesirable genotypes, preventing the need for expensive and time- consuming laboratory testing. New near-infrared reflectance (NIR) technology may provide the means to speed the development of malting barley by eliminating poor-quality lines as early as the F3 generation. Use of traditional laboratory tests could then be used on the more promising selections. We selected 54 barley genotypes having known malt extract (ME) values for use in calibration of a Technicon 400-R InfraAlyzer. The calibration techniques recommended by the equipment manufacturer were followed to obtain necessary regression coefficients. These were b = -600.2 for F10 and b = 594.9 for F14 with the intercept F00 = 58.03. With these values entered in the electronic NIR system, we measured the malt extract values of the barley samples used in the calibration procedure. The correlation coefficient between laboratory (Lab) ME values and NIR ME values was r = 0.95**, standard deviation from regression 0.833, and the regression coefficient of b = 1.00. One hundred thirty-three samples of known Lab ME not used in the calibration were analyzed for NIR ME. The correlation coefficient between Lab ME and NIR ME was r = 0.98**, the regression coefficient b = 1.00, and the standard deviation from regression 1.43. On 199 selected F3 experimental lines, the range of Lab ME was 74.1-79.2. The NIR ME values were highly correlated with Lab ME value (r = 0.36**), with a regression coefficient b = 0.38 and the standard error from regression 0.78.