Seed Persistence in Soil and Seasonal Emergence in Plant Species from Different Habitats

Abstract

(1) The extent to which plants form a persistent seed bank and the time of year at which seedlings arise from it both have a bearing on the establishment of vegetation after disturbance. These attributes were determined for seventy dicotyledons from a range of habitats by mixing freshly-collected seeds with the top 7.5 cm of soil confined in cylinders sunk in the ground, cultivating three times yearly, and recording emergence for 5 years. (2) In about one-third of the species, few seeds persisted for longer than a year and most seedlings emerged in the autumn of sowing (Sherardia arvensis*, Galium mollugo) or early in the following spring (Chaerophyllum temulentum, Pimpinella saxifraga, Agrimonia eupatoria, Tragopogon pratensis). (3) The remaining species formed a persistent seed bank in cultivated soil from which seedlings arose in each year for which the experiments continued. Some exhibited a consistent pattern of emergence entirely (Odontites verna, Linum catharticum) or mainly (Brassica nigra, Silene dioica) in spring, while others (Hyoscyamus niger, Erodium cicutarium, Amaranthus retroflexus) emerged in late spring and summer. Emergence of Arenaria serpyllifolia, Cerastium glomeratum and Anthriscus caucalis was mainly in early autumn, while that of Legousia hybrid began in autumn and continued into spring. (4) Other species showed no pronounced pattern of emergence in successive years. Although there was often a tendency for most seedlings to emerge in spring or autumn, some appeared throughout the growing season with flushes following soil disturbance. Included in this group were Barbarea vulgaris, Coronopus didymus, Ballota nigra and Dipsacus sylvestris. (5) There were consistent differences between species in the extent of immediate emergence, seed survival in disturbed soil and in the seasonal pattern of emergence. Seeds of annual weeds of arable land were all relatively long-lived, but otherwise there appeared to be little relationship between seed persistence and habitat. The results provide indications of the potential for regeneration by different species from the seed bank after disturbance of a plant community and of the times of year at which it is likely to take place.