Evaluation of the phenotypic variation in kikuyu populations

A series of cutting experiments evaluated the phenotypic variation in populations of kikuyu in Australia to assess their capacity to provide sufficient variability to sustain a breeding and selection program. The aim of such a program would be to develop a high quality cultivar without compromising stand vigour. Single plants, generated either from seed of commercial cultivars or accessions, seed treated by mutagenic chemicals or from randomly-chosen runners from stands across Australia, were sown singularly in randomised block experiments under irrigation at two sites in subtropical eastern Australia. There were significant differences between kikuyu cultivars in leaf, stem and runner dry matter yields, plant height and quality (crude protein, neutral detergent and acid detergent fibre content and in vitro digestibility). Variation between individual plants was also substantial, with 7, 3 and 3 % improvement over the population mean for crude protein content, in vitro digestibility, and metabolisable energy respectively, and a reduction of 5 and 8 % in neutral and acid detergent fibre content. It was also associated with a forage yield twice the population mean from these elite plants. Mutagenesis of seed from a commercial cultivar produced greater within-population variation for quality, but less for agronomic, traits compared with the natural population. Gene fingerprinting suggested that the material distributed across Australia came from two main sources, one related to cv. Whittet and the other to a ‘common’ lower yielding type. It was concluded that the currentlyavailable material within Australia, although from limited genetic base, contained sufficient variation to achieve a significant improvement in quality without reducing plant vigour.