A rapid, controlled-environment seedling root screen for wheat correlates well with rooting depths at vegetative, but not reproductive, stages at two field sites.

Abstract

Root length and depth determine capture of water and nutrients by plants, and are targets for crop improvement. Here we assess a controlled-environment wheat seedling screen to determine speed, repeatability and relatedness to performance of young and adult plants in the field. Recombinant inbred lines (RILs) and diverse genotypes were grown in rolled, moist germination paper in growth cabinets, and primary root number and length were measured when leaf 1 or 2 were fully expanded. For comparison, plants were grown in the field and root systems were harvested at the two-leaf stage with either a shovel or a soil core. From about the four-leaf stage, roots were extracted with a steel coring tube only, placed directly over the plant and pushed to the required depth with a hydraulic ram attached to a tractor. In growth cabinets, repeatability was greatest (r = 0.8, P < 0.01) when the paper was maintained moist and seed weight, pathogens and germination times were controlled. Scanned total root length (slow) was strongly correlated (r = 0.7, P < 0.01) with length of the two longest seminal axile roots measured with a ruler (fast), such that 100-200 genotypes were measured per day. Correlation to field-grown roots at two sites at two leaves was positive and significant within the RILs and cultivars (r = 0.6, P = 0.01), and at one of the two sites at the five-leaf stage within the RILs (r = 0.8, P = 0.05). Measurements made in the field with a shovel or extracted soil cores were fast (5 min per core) and had significant positive correlations to scanner measurements after root washing and cleaning (>2 h per core). Field measurements at two- and five-leaf stages did not correlate with root depth at flowering. The seedling screen was fast, repeatable and reliable for selecting lines with greater total root length in the young vegetative phase in the field. Lack of significant correlation with reproductive stage root system depth at the field sites used in this study reflected factors not captured in the screen such as time, soil properties, climate variation and plant phenology.