The potential for management of reference collections in herbage variety registration trials using a cyclic planting system for reference varieties

The testing of new varieties of plants for registration and for the grantin g of Plant Breeders' Rights involves the determination of their distinctnes s, uniformity and stability (DUS) in comparison with existing reference var ieties. For most agricultural crops this is done by only comparing the new variety in field trials with a limited number of reference varieties of the same type. However, for the cross-fertilized herbage crops, where few reli able grouping characteristics are available, the scale of such trials has i ncreased significantly over the years in response to continuing breeding ef forts. A system of cyclic planting of reference varieties in two years out of ever y three has been developed with the data for each missing year compensated for by the use of historic data from two earlier years. Distinctness is ass essed within the incomplete data matrix by applying an adaptation of the Co mbined Over Years Distinctness analysis (COYD) using Modified Joint Regress ion Analysis (MJRA). Results which would have been produced from a data matrix modelled on a sys tem of cyclic planting of reference varieties have been compared to those f rom a complete data matrix and found to be in close agreement. The system t herefore has the potential to provide a robust and cost effective system fo r the determination of distinctness, uniformity and stability.