ABSTRACT
Summary Seed shattering is a problem in most pasture grasses, especially in those that have a history of recent domestication. In many of these grasses, losses from shattering are aggravated by uneven ripening of inflorescences and a low percentage of viable seed. Fragility of the rachilla appears to be the most common cause of seed shattering in grasses, and once this connection is broken, the seed is free to shed from the inflorescence. In some of the more domesticated species, the nature and configuration of the glumes and the manner in which they are packed on the inflorescence play an important part in retaining seed. Seed shattering in phalaris (Phalaris aquatica) has been overcome through the discovery of a naturally occurring seedretaining mutant possessing a strong nonbrittle rachilla that anchors the seed in the spikelet at maturity. This character has now been incorporated into a seed-retaining cultivar which can be harvested without suffering any significant loss of seed for a period extending at least three weeks after full stand maturity. It seems possible that mutations of the type found in phalaris that prevent seed shattering may be present in many grass species, but because these mutants are at a selective disadvantage in the wild, the frequency of the genes controlling them in populations will be low. Despite this low frequency it should be possible to locate such natural mutants in other important grasses, thereby overcoming one of the major problems inhibiting the domestication of these species for use in agriculture.
