Pre-zygotic parental environment modulates seed longevity JITKA KOCHANEK, 1 * YVONNE M. BUCKLEY, 2 ROBIN J. PROBERT, 3 STEVE W. ADKINS 1 AND KATHRYN J. STEADMAN 2,4 1 School of Land, Crop and Food Sciences, University of Queensland, Gatton Campus, Gatton, Qld 4343 (Email: j.kochanek@uq.edu.au), and 2 School of Biological Sciences and 4 School of Pharmacy, University of Queensland, Brisbane, Queensland, Australia; 3 Seed Conservation Department, Royal Botanic Gardens,West Sussex, UK Abstract The potential for the pre-zygotic plant growth environment to play a role in determining seed longevity was investigated for a species that inhabits arid to semi-arid Australia. Seed longevity is particularly important for wild populations in fluctuating environments because the longer a seed-lot is able to survive in the soil seed bank the more likely it is to buffer the population from unpredictable environments. Thus Wahlenbergia tumidifructa plants received wet or dry soil moisture within a warm or cool glasshouse until flowering. Seeds subsequently produced by flowers that opened on the day that plants were moved to a common environment were collected at maturity and longevity assessed by controlled ageing at 60% relative humidity and 45°C. Mean seed longevity was similar for seeds produced by plants that grew in warm-wet, warm-dry and cool-dry conditions (P50 of about 20 days), but extended for plants in cool-wet conditions (P50 = 41.7 days). Cool temperatures resulted in seeds with a wider distribution of lifespans (s= 20 days) than warm conditions (s= 12 days); the large s caused the extended P50 for cool-wet plants, but not cool-dry as a result of a concomitant reduction in initial seed germination (Ki). After moving to the common environment, all plants generated new vegetative material, which went on to produce seeds with similar longevity (P50 approx. 20 days) irrespective of original environment.Visible phenotypic responses of the parent to environmental conditions correlated with longevity and quality parameters of the progeny seeds, suggesting that a parental effect modified seed longevity. Our study provides novel empirical data showing that environmental conditions expected under climate change scenarios may potentially cause seed longevity to decline for a species that inhabits arid to semi-arid Australia. These negative impacts on population buffering may weaken the storage effect mechanism of species coexistence in fluctuating environments. Key words: parental environment, pre-zygotic, seed longevity, storage effect, Wahlenbergia tumidifructa. INTRODUCTION Persistent seed banks increase the time to extinction of a population and can buffer populations during times of unfavourable environmental conditions.The storage effect mechanism of species coexistence requires that some buffering mechanism is present so that competi- tively inferior species can mitigate the consequences of unfavourable environmental conditions (Chesson 2000; Thompson 2000; Facelli et al. 2005). Individual seeds leave the seed bank as a result of seed-based events (germination or ageing) or external factors (predation, microbial attack). However, in arid and semi-arid environments seed ageing is believed to con- tribute significantly to soil seed bank decline (Murdoch & Ellis 2000) and there is evidence for a correlation between seed longevity and soil seed bank persistence (Long et al. 2008). Seed longevity, the time that a population of seeds will remain alive, can vary considerably between different populations of a single species. Given that seed longevity is a quantitative trait, it follows that both the environment and genetics of the plant are likely to be responsible for this variation (Clerkx et al. 2004; Sasaki et al. 2005). Certainly, there is evidence for a genetic component to seed longevity; different geno- types of a single species grown under a common envi- ronment can display different longevity as shown for soybean (Glycine max (L.) Merr.) and rice (Oryza sativa L.) (Zanakis et al. 1994; Kameswara Rao & Jackson 1996b,c).There is also evidence for an impor- tant effect of the environment, with longevity changing when the plant growth environment was manipulated in rice, wheat (Triticum aestivum L.), watermelon (Cit- rullus lanatus (Thunb.) Mtsum. and Nakai) and beans (Phaseolus vulgaris L.) (Ellis & Hong 1994; Sanhewe & Ellis 1996; Sanhewe et al. 1996; Demir et al. 2004), or when the environment during seed development was altered for rapid-cycling brassica (Brassica campestris (rapa) L.) and lettuce (Lactuca sativa L.) (Sinniah et al. 1998; Contreras et al. 2008). It is therefore not surprising that the genotype and growth environment interact to modify seed longevity, as observed when *Corresponding author. Accepted for publication December 2009. Austral Ecology (2010) ••, ••–•• © 2010 The Authors doi:10.1111/j.1442-9993.2010.02118.x Journal compilation © 2010 Ecological Society of Australia