Prediction of seed longevity: a modification of the shape of the Ellis and Roberts seed survival curves

Abstract

AbstractThe prediction of the viability of stored seeds is important both for the management of germplasm collections and for the management of commercial seed production and storage. The Ellis and Roberts model for seed viability during storage is examined, and an inadequacy of the model highlighted. A modification is proposed, based on the ‘control mortality’ probit model developed for insecticide bioassays, to take proper account of variation in initial viability. This new ‘control viability’ model, relating seed viability to storage time, is fitted to data from a carrot seed storage experiment and found to fit well for a range of storage environments. A relationship, similar to that proposed by Ellis and Roberts for the effects of storage conditions on the rate of loss of viability, is fitted to the estimated rates from this new model. Data from a second carrot seed storage experiment are used to validate this relationship.