Medium-term seed storage of diverse genera of forage grasses, evidence-based genebank monitoring intervals, and regeneration standards

Abstract

The frequency at which seed accessions are monitored for viability during storage in genebanks (reciprocal of the monitoring interval) must balance timely detection of loss in viability against monitoring cost (seed depletion and staff resource). Up to three decades of genebank seed germination test results of diverse grasses maintained in the International Livestock Research Institute’s medium-term store (circa 8 °C with 5% moisture content) were evaluated in an attempt to derive recommendations on seed accession monitoring intervals. Six different patterns of variation in germination test results during storage were found amongst seed lots by probit analysis within 29 genera: no trend (6 genera); contrasting trends (positive to negative) (3); common slope of loss in viability (11); common slope of increase in ability to germinate (6); common loss in viability (2); common increase in ability to germinate (1). Recommended monitoring intervals were calculated from the fitted survival curves for each of the 13 genera showing uniformity in loss in viability: the medium-term store expectation of 2–10 years’ maintenance of high viability was met in eight genera, whilst four provided greater survival periods. Furthermore, the 13 genera showing either no trend over period of storage or an increase in ability to germinate during storage also exceeded the expectations for survival periods in medium-term stores. Advice is provided on calculating monitoring intervals for different combinations of initial viability with a wide range of potential regeneration standards.