Soil seed banks in Kansas prairie and resseded old fields.

Abstract

Viable seeds in the soil seed banks of a native prairie pasture and two old fields seeded with native grasses were estimated. Samples, each with a surface area of 25 cm² and a depth of 5 cm or 15 cm, were collected from these Lyon County, Kansas sites during the period of time from February 12 to March 21, 1987. They were divided into depths of 1-2.5, 2.5-5, and 5-15 cm, sieved and placed over sand in trays in a greenhouse. Samples were maintained until July 24, 1987. Seedlings were counted at least once per week, with one exception. An equivalent of 10,980 seeds•m-² (57% monocots, 41% dicots, 1% unidentified) emerged from the top 5 cm of the recently reseeded old field, 7, 260 seeds•m-² (65% monocots, 33% dicots, 2% unidentified) from the native prairie pasture, and 4,850 seeds•m-² (49% monocots, 50% dicots, 1% unidentified) from the older reseeded old field. Numbers of seedlings and monocot-dicot ratios declined with depth; the amount of decline varied with the site. In conclusion, the results of this study seem to indicate that the seed bank populations of reseeded old fields are not maintained over time. After approximately eighteen years the seed bank of an old field, in terms of seed numbers, becomes more like that of a native prairie pasture than an abandoned field. However, species composition, reflected in the monocot-dicot ratios, may remain quite different.