Use of near infrared reflectance spectroscopy to screen ethiopian mustard for seed weight

Near infrared reflectance spectroscopy (NIRS) provides a method for the simultaneous measurement of multiple quality traits in oilseed Brassica species, such as oil, protein, glucosinolate, and chlorophyll content. The incorporation of new traits into NIRS analysis should increase the information obtained from analyzed samples with little additional effort. The objective of this study was to determine the potential use of NIRS to screen intact-seed samples of Ethiopian mustard (Brassica carinata Braun) for 1000-seed weight. A calibration equation for this trait developed in 1992 using modified partial squares (MPLS) regression was successively expanded with new samples from the 1993 and 1994 crops. The 3-yr calibration showed an r(2) of 0.86 in external validation, indicating the reliability of NIRS for screening Ethiopian mustard samples for seed weight. This method could be particularly useful in large breeding programs, as it is less time consuming and more cost effective than conventional methods.