The ecophysiology of seed persistence: a mechanistic view of the journey to germination or demise.

Seed persistence is the survival of seeds in the environment once they have reached maturity. Seed persistence allows a species, population or genotype to survive long after the death of parent plants, thus distributing genetic diversity through time. The ability to predict seed persistence accurately is critical to inform long-term weed management and flora rehabilitation programs, as well as to allow a greater understanding of plant community dynamics. Indeed, each of the 420000 seed-bearing plant species has a unique set of seed characteristics that determine its propensity to develop a persistent soil seed bank. The duration of seed persistence varies among species and populations, and depends on the physical and physiological characteristics of seeds and how they are affected by the biotic and abiotic environment. An integrated understanding of the ecophysiological mechanisms of seed persistence is essential if we are to improve our ability to predict how long seeds can survive in soils, both now and under future climatic conditions. In this review we present an holistic overview of the seed, species, climate, soil, and other site factors that contribute mechanistically to seed persistence, incorporating physiological, biochemical and ecological perspectives. We focus on current knowledge of the seed and species traits that influence seed longevity under ex situ controlled storage conditions, and explore how this inherent longevity is moderated by changeable biotic and abiotic conditions in situ, both before and after seeds are dispersed. We argue that the persistence of a given seed population in any environment depends on its resistance to exiting the seed bank via germination or death, and on its exposure to environmental conditions that are conducive to those fates. By synthesising knowledge of how the environment affects seeds to determine when and how they leave the soil seed bank into a resistance-exposure model, we provide a new framework for developing experimental and modelling approaches to predict how long seeds will persist in a range of environments.