The objective of this study was to develop a phenotyping platform for the non-destructive, digital measurement of early root growth of axile and lateral roots and to evaluate its suitability for identifying maize (Zea mays L.) genotypes with contrasting root development. The system was designed to capture images of the root system within minutes and to batch process them automatically. For system establishment, roots of the inbred line Ac7729/TZSRW were grown until nine days after germination on the surface of a blotting paper in pouches. An A4 scanner was used for image acquisition followed by digital image analysis. Image processing was optimized to enhance the separation between the roots and the background and to remove image noise. Based on the root length in diameter-class distribution (RLDD), small-diameter lateral roots and large-diameter axile roots were separated. Root systems were scanned daily to model the growth dynamics of these root types. While the axile roots exhibited an almost linear growth, total lateral root length increased exponentially. Given the determined exponential growth, it was demonstrated that two plants, germinated one day apart but with the same growth rates differed in root length by 100%. From the growth rates we were able to identify contrasting genotypes from 236 recombinant inbred lines (RILs) of the CML444 x SC-Malawi cross. Differences in the growth of lateral roots of two selected RILs were due to differences in the final length and linear density of the primary lateral roots, as proven by the manual reanalysis of the digital images. The high throughput makes the phenotyping platform attractive for routine genetic studies and other screening purposes.
Plant and Soil (2009) 325 (1-2) 335-349 [doi: 10.1007/s11104-009-9984-2]