Gross, K. L. 1990. A comparison of methods for estimating soil seed banks. Journal of Ecology 78:1079-1093.

(1) Seed-bank species composition and seed density were determined in an annually ploughed field at the Kellogg Biological Station in south-western Michigan, U.S.A. using three different methods: direct germination, germination following cold-stratification, and washing using a modified elutriation system. A fourth method, flotation on a salt-density gradient was abandoned after preliminary trials showed it was inefficient and inaccurate. (2) Twenty soil cores (2.5 cm in diameter, 15 cm deep) were taken from each of twenty-five sampling points in the field. Cores were divided into three depth intervals (0-2 cm, 2-5 cm and 5-15 cm) and five cores from each quarter of a sampling quadrat were combined and randomly assigned to one of the four methods for estimating the seed bank. (3) A total of fifty species was detected in the seed bank of this community. Overall, more species were detected with cold-stratification (forty-two species) than with direct germination (thirty-seven) or elutriation (thirty-five). Plots of species-sampling-area curves showed that fifteen to twenty sampling locations were sufficient to determine the number of species present in the seed bank with any of these methods. (4) Seed-density estimates were significantly higher in the elutriated samples. This was due primarily to the inclusion of inviable seeds in the counts from the elutriated samples. (5) There were significant differences in seed viability among species recovered in the elutriated samples. Both Ambrosia artemisiifolia and Chenopodium album, seeds of which were common in the samples, had low viability (25% and 3%, respectively). (6) Germination methods (particularly when more than one pre-treatment method is used) provide a more complete listing of species present in the soil seed bank of a community than elutriation. However, elutriation methods may be more useful for documenting variation in seed distributions on a large spatial scale or in quantifying the distribution of species with unique, easily identifiable seeds. Elutriation seems particularly well-suited for quantifying the distribution of plants that are considered weeds in agricultural systems, because many of these species have seeds that are easily identified and retain viability when dried.

DOI: 10.2307/2260953

Associated Treatment Areas:

KBS Landscape

Download citation to endnote bibtex