Submitted to: Sugar Bulletin
Publication Type: Trade Journal
Publication Acceptance Date: October 15, 2010
Publication Date: November 1, 2010
Citation: Tew, T.L. 2010. Multi-environment evaluation of sugarcane early in our selection and variety evaluation program. The Sugar Bulletin. 89(2):13-14. Technical Abstract: In the USDA and the LSU AgCenter breeding programs combined, fewer than 100 candidates out of about 150,000 annually-planted seedlings will receive permanent numbers and be evaluated in multiple environments for their potential as commercial varieties. This means that over 99.9% of seedlings planted to the field will be discarded before any are evaluated in more than one environment. In some sugarcane breeding and selection programs around the world, selected seedlings are advanced to multiple locations immediately after seedling stage. The hypothesis for early multi-location evaluation is that those that receive permanent numbers will have already proven themselves across multiple environments in early clonal stages (1st and 2nd line trials) of selection, and thus will be more likely to succeed as viable commercial candidates in later multi-location yield trials. To test this hypothesis in our situation, in the fall of 2011, we will advance into the 1st line stage at three locations (Schriever, St. Gabriel, and Jeanerette) 300 seedlings from 20 high-value crosses currently planted at the USDA or the LSU AgCenter. Of these 300, those 60 with the highest mean ratings across environments will be advanced to the 2nd line stage, and two years later (plant and ratoon), 12 will be assigned permanent numbers and be compared in a yield trial at these same three locations with a set of 12 candidates from the same population selected on the basis of their performance at one location (current approach). Yield data will be collected during 2015-2017 (plant cane-2nd ratoon) should tell us whether it is advantageous to conduct multi-environment evaluation of sugarcane early on in our selection and variety evaluation program.