Premium
Variation and Evaluation of Seed Shape in Soybean
Author(s) -
Nelson Randall L.,
Wang Peiying
Publication year - 1989
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/cropsci1989.0011183x002900010033x
Subject(s) - germplasm , trait , biology , repeatability , consistency (knowledge bases) , visual inspection , mathematics , statistics , artificial intelligence , pattern recognition (psychology) , agronomy , computer science , programming language
Differences in soybean [ Glycine max (L.) Merr.] seed shape are obvious but no measure of that variability has been published. The objectives of this research were to determine the consistency of seed shape and to establish and test a visual classification scheme for soybean seed shape. Such a scheme could exploit the variability of this trait to assist in the identification and classification of germplasm. Ninety‐nine germplasm accessions in Maturity Groups I to IV were selected and grown at Urbana, IL for 3 yr. Measuring 20 seeds per seed lot provided a reasonable estimate of seed shape. A description of seed shape was established using two ratios, height to length (HLR) and height to thickness (HTR). Repeatability ues for the two ratios were 0.90 or greater based on line means. Five and three categories of HLR and HTR, respectively, were established to create a binary, visual classification scheme. All seed lots were assigned a classification based on the mean of actual seed measurements and two samples from each of the 297 seed lots were visually classified. Visual observations agreed with actual measurements for the HLR and HTR for 68 and 76% of the samples, respectively. For all of the visual misclassifications for the HTR rating and 98% of the visual misclassifications for HLR rating, the differences between the visual rating and the rating determined by actual measurement were less than half of the range of a single class. The visual classification scheme proposed here is an effective means of categorizing differences in seed shape among soybean lines and could be used to describe soybean cultivars and germplasm accessions.