Premium
Use of Near Infrared Reflectance Spectroscoppy to Screen Soybean Lines for Plant Nitrogen
Author(s) -
Mündel H. H.,
Schaalje G. B.
Publication year - 1988
Publication title -
crop science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.76
H-Index - 147
eISSN - 1435-0653
pISSN - 0011-183X
DOI - 10.2135/cropsci1988.0011183x002800010034x
Subject(s) - phytotron , autoanalyzer , kjeldahl method , cultivar , near infrared reflectance spectroscopy , biology , field experiment , horticulture , agronomy , zoology , botany , nitrogen , mathematics , chemistry , near infrared spectroscopy , chromatography , organic chemistry , neuroscience
Selecting breeding lines of soybean [ Gfycine Max (L.) Merr.] for an enhanced ability to support N fixation by Bradyrhizobium japonicum Jordan requires a fast screening method for determining N content in the plants. The objective of this study was to determine if near infrared spectroscopy (NIRS) could be used as such a method. Two cultivars grown at two locations with 10 inoculation treatments were used for field samples. One cultivar grown at two temperature regimes with 40 N fertilizer treatments was used for the phytotron samples. Samples were ground and regressions were run of log (1/relectance) of up to 19 filters (wavelengths) on a Technicon InfraAlyzer 400 (Technicon Instruments Corp., Tarrytown, NY) to laboratory Kjeldahl N determinations. Filters were selected to optimize regressions; overall stage and/or part regressions as well as individual stage and/or part (R2, R8—overall top growth, R8—seeds, and R8—top growth excluding seeds) regressions were carried out. Tests of homogeneity of regressions showed that stages and/or parts could not be combined, but that field and phytotron‐grown soybean could be combined. Predictions run on random samples were most accurate for N content in seeds and at R2 and least accurate for R8—top growth excluding seeds. Calibrations based on field and phytotron samples from 1 yr were acceptable for predicting N content in a subsequent year. The NIRS calibrations were on the average more highly repeatable for predicting N content than either one of two Kjeldahl determinations.