Evaluation of near‐infrared reflectance spectroscopy for predicting stover quality trait in semi‐exotic populations of maize

Near‐infrared reflectance spectroscopy (NIRS) analysis was investigated as a means of predicting quality parameters in semi‐exotic maize stover. These parameters included crude protein (CP), neutral detergent fibre, acid detergent fibre and in vitro dry matter digestibility (IVDMD). Samples of semi‐exotic maize stover were collected during three growing seasons (1989, 1990, 1991) from three locations in Catalonia, Spain. Calibration equations were obtained by multiple linear regression of conventional laboratory values on NIRS data from 84 samples and verified with 20 additional samples. Separate NIRS calibration were developed also within year (1989 and 1990, respectively). A Bran + Luebbe InfraAnalyzer model 450 was used for the study. In the multi‐year calibration the coefficients of squared multiple correlation (R2) ranged from 0–81 for IVDMD to 0–92 for CP and the standard errors of calibration (SEC) ranged from 0–35 for CP to 1–46 for IVDMD. The study showed that NIRS analysis can be used to evaluate the quality of semi‐exotic maize in breeding programmes. Copyright © 1995 John Wiley & Sons, Ltd