The weed seed population of arable soil: I. Numerical estimation of viable seeds and observations on their natural dormancy

1. Numerical estimations have been made of the number of viable weed seeds, as shown by the seedlings germinating, in soil samples of known area taken from the permanent wheat and barley fields at Rothamsted and Woburn Experimental Stations. The samples have been taken for several years in succession, both before and during specified schemes of fallowing. The actual number of living seeds per acre, being still more plentiful in certain plots. 2. Analyses of the figures obtained show that comparatively few species germinate freely throughout the year. Most species show a definite periodicity, the majority of seedlings appearing during the autumn or winter, or both, and relatively few in late spring and summer. 3. Intensive methods of cultivation indicate that many weed seeds in the soil have a period of natural dormancy, during which they will not start into growth even if they are placed under conditions favourable for germination. The length of this period varies with the species, which are considered individually. Seeds buried in the soil under conditions unsuitable for germination may retain their vitality for many years, this prolonged dormancy being termed induced in contrast to the natural dormancy. 4. It appears that a definite association exists between the weed flora and the type of long-continued manuring, this association being influenced by the nature of the soil and to a less extent by the repeated growth of an autumn or spring-sown cereal. The association is reflected in: (a) The total number of viable weed seeds of all species per unit area. (b) The preference of some species for certain manures. (c) The effect of liming on the prevalence of species occurring with certain manurial combinations. 5. Certain weeds as Alchemilla, Veronica spp., etc., show a marked preference for either the wheat or barley crop in both types of soil examined. This association is apparently caused by the method of cultivation rather than by any direct influence of the crop itself on the weeds. 6. Certain association exists between some weeds and the soil in which they grow, but the prevalence of poppy, a typical light-land weed, on the heavy soil at Rothamsted shows that corroborative evidence from a large area is necessary before such an association can be regarded as established.