Skip to main content
ARS Home » Midwest Area » Urbana, Illinois » Global Change and Photosynthesis Research » Research » Publications at this Location » Publication #229293
Title: Nondestructive Tests for Weed Seedbank Studies

Author
item Schutte, Brian
item HARAMOTO, ERIN - MT HOLYOKE COLLEGE
item Davis, Adam

Submitted to: Seed Science Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/1/2010
Publication Date: N/A
Citation: N/A

Interpretive Summary: Seed conditions and burial fates are usually unknown at the onset of experiments because viability and germinability are determined through destructive assays. We hypothesized that conductivity of seed steep can be used to nondestructively differentiate germinable, dormant, and dead seeds within seed accessions of ivyleaf morningglory, velvetleaf, common lambsquarters and giant foxtail. Practical use of steepate conductivity tests in seedbank research was then evaluated by comparing burial fate predictions with projections based on seed weight and seed imbibition status after soaking. Across species, tests based on changes in steepate conductivity over time offered a high degree of predictive power. For ivyleaf morningglory and velvetleaf, prediction accuracy rates were not improved by extending the soak interval from 24 to 168 hr, indicating short soaks are sufficient for differentiation of germinable, dead and dormant seeds of hardseeded species. According to these results, steepate conductivity tests can be used to identify viable seeds capable of remaining ungerminated. Such seeds are essential for understanding causal factors of mortality in the soil seedbank.

Technical Abstract: Seed conditions and burial fates are usually unknown at the onset of experiments because viability and germinability are determined through destructive assays. We hypothesized that conductivity of seed steep can be used to nondestructively differentiate germinable, dormant, and dead seeds within seed accessions of ivyleaf morningglory, velvetleaf, common lambsquarters and giant foxtail. Practical use of steepate conductivity tests in seedbank research was then evaluated by comparing burial fate predictions with projections based on seed weight and seed imbibition status after soaking. Across species, classification trees based on changes in steepate conductivity over time offered a high degree of predictive power (10-fold cross validation prediction accuracy rates = 95.0±2.7% to 98.3±1.7%). For ivyleaf morningglory and velvetleaf, prediction accuracy rates were not improved by extending the soak interval from 24 to 168 hr, indicating short soaks are sufficient for differentiation of germinable, dead and dormant seeds of hardseeded species. For common lambsquarters, germinable and dormant seeds were segregated with a 168-hr soak, however, the prolonged soak interval induced mortality. In a separate experiment, virtually all ivyleaf morningglory and velvetleaf seeds fated for persistence were correctly identified by changes in steepate conductivity (10-fold cross validation true positive rate = 99.1±0.9%). These true positive rates compared favorably with predictions based on seed weight and imbibition status after soaking. False discovery rates in 10-fold cross validation were lowest for predictions based on change in conductivity from 4 to 48 hr (7.2±2.5%) and imbibition status after soaking (0%). According to these results, steepate conductivity tests can be used to identify viable seeds capable of remaining ungerminated. Such seeds are essential for understanding causal factors of mortality in the soil seedbank.