Designing improved plant types: a breeder’s viewpoint

Abstract

Plant breeders have traditionally devoted time to a consideration of the characteristics required for high performance in their target region. Some, over the last 30 years, have extended this traditional approach to encompass considerations of the basic processes of dry matter accumulation and distribution. In some cases this extension has involved a breeding approach based explicitly on the design and selection of model plants or ideotypes. This latter has not been particularly successful, but the concept that factors affecting dry matter accumulation and distribution should beltaken into account is now widely accepted. Recently, the availability of computer simulation models of crop growth has made it possible to refine the original considerations, to examine traits in a quantitative rather than qualitative manner, and to examine environmental response traits as well as morphological characteristics. They thus constitute a useful addition to the array of tools used by plant breeders when deciding on parents and crosses. Their use in this arena, however, is likely to be limited, partly because of scepticism stemming from previous experience with ‘ideotype’ concepts. They may have greater utility in some related areas. First, to analyze trial data in terms of specific genotypic traits. Second, to caIculate likely field performance over a 10 or 20 year period. And third, to calculate field scale performance from spaced plant data. Before any of these applications are possible, however, further model development to ensure that they account for all factors currently used in variety release/choice decisions may be necessary.