High‐Throughput Precision Phenotyping of the Oil Content of Single Seeds of Various Oilseed Crops

Abstract

High‐throughput (HT) precision phenotyping of agronomic traits is important for well‐founded, rapid selection decisions in plant breeding. This applies especially to nondestructive measurement of single‐seed oil content, for which an HT platform has recently become available. The objectives of this study were (i) to evaluate the suitability of this HT platform for measuring seed mass, oil mass, and oil content in various oil crops, (ii) to determine the accuracy and repeatability of the measurements, and (iii) to discuss technical adjustments required for specific crops. Seeds of canola (Brassica napus L.), castor bean (Ricinus communis L.), cotton (Gossypium hirsutum L), jatropha (Jatropha curcas L.), maize (Zea mays L.), soybean [Glycine max (L.) Merr.], and sunflower (Helianthus annuus L.) were measured repeatedly using a randomized complete block design. Additionally, the oil content of bulks of seeds from two crops was determined by wet chemistry analysis. Repeatability of all three traits recorded via the HT platform generally exceeded 98% in all seven crops. Oil content of bulks determined by wet chemistry analysis was almost perfectly correlated (R2 > 99.9%) with the mean of nuclear magnetic resonance (NMR) measurements of single seeds from these bulks. To warrant precise results and smooth operation, yielding an average throughput of ∼600 seeds h−1, technical modifications in certain modules of the HT platform are required to accommodate the size, geometry, and oil content of seeds from different crops. In conclusion, the HT platform demonstrated high repeatability and accuracy of measurements, which opens up several fields of application in plant breeding.