Ecology of Capsella bursa-pastoris (L.) Medik. and Senecio vulgaris L. in relation to germination behaviour.

TLDR: In some cases it has been shown that a population of seeds, even from a single plant, contains distinct groups with different physiological responses which may (or may not) be correlated with morphological differences.

Abstract: In the ecology of annual weeds, it is the characteristics of the seeds which are most important in determining the distribution of the plants in both space and time. The weed plant itself lives for less than 9 months but its seeds can remain alive in the soil for many years. Furthermore, because weeds live in temporary habitats, they are always exposed to the risk of total elimination from a locality and therefore the vast reservoir of seeds in the soil is necessary to ensure survival. In the soil, germination and subsequent emergence of weed seedlings is an irregular process. Usually, at intervals throughout the year, large numbers of seeds germinate within a few days of each other and then each burst of activity is followed by a quiescent period during which only a few seeds germinate. This has been observed in natural seed populations (Chancellor 1965) and in artificially introduced populations derived from bulk seed collections (Roberts 1964). Small flushes of germination of Senecio and Capsella were observed by both these authors throughout the year but the biggest flushes occurred in May and in August and September. Salisbury (1964) confirmed this pattern of germination in Capsella using an artificially introduced population derived from a single plant. Such a pattern of seed germination would appear to be of great advantage to the species because even if an entire crop of plants is destroyed, more seedlings will soon emerge to replace those which were lost. The causes of these intermittent flushes of germination are often obscure. In some cases it has been shown that a population of seeds, even from a single plant, contains distinct groups with different physiological responses which may (or may not) be correlated with morphological differences. This phenomenon has been exposed largely by Harper and his colleagues and is referred to as seed polymorphism (Williams & Harper 1965; Cavers & Harper 1966). To be useful, the term should be restricted to those cases where it can be shown that there is a discontinuous distribution of some germination mechanism amongst the seeds, or that, when subjected to constant conditions, the distribution of the onset of germination in time contains more than one distinct mode. Even in homozygous, cultivated cereals there is great variation in the distribution of dormancy periods in a single population of seeds, even from the same plant. For example, if a population of rice seeds is stored under constant conditions after harvest and samples subjected to germination tests at regular intervals, it has been shown that the distribution of dormancy periods amongst the individual seeds is normal and that the standard deviation is relatively large (Roberts 1961, 1965): from these published data it may be calculated that the coefficient of variation