Clinal population divergence in an adaptive parental environmental effect that adjusts seed banking. Issue 3 (2nd February 2017)

• Record Type: Journal Article
• Authors: Lampei, Christian; Metz, Johannes; Tielbörger, Katja
• Is Part Of: New phytologist. Volume 214:Issue 3(2017)
• Page Start: 1230
• Page End: 1244
• Publication Date: 2017-02-02
• Subjects: bet‐hedging -- clinal variation -- environmental autocorrelation -- maternal effect -- parental effect -- risk spreading -- seed dormancy -- transgenerational plasticity; Botany -- Periodicals; 580
• DOI: 10.1111/nph.14436
• Languages: English
• ISSNs: 0028-646X

Abstract

Summary: Bet‐hedging via between‐year seed dormancy is a costly strategy for plants in unpredictable environments. Theoretically, fitness costs can be reduced through a parental environmental effect when the environment is partly predictable. We tested whether populations from environments that differ in predictability diverged in parental effects on seed dormancy. Common garden‐produced seeds of the two annual plant species Biscutella didyma and Bromus fasciculatus collected along an aridity gradient were grown under 12 irrigation treatments. Offspring germination was evaluated and related to environmental correlations between generations and their fitness consequences at the four study sites. One species exhibited strong seed dormancy that increased with unpredictability in seasonal precipitation. The parental effect on seed dormancy also increased proportionally with the environmental correlation between precipitation in the parental season and seedling density in the following season; this correlation increased from mesic to arid environments. Because fitness was negatively related to density, this parental effect may be adaptive. However, the lack of dormancy in the second species indicates that bet‐hedging is not the only strategy for annual plants in arid environments. Our results provide the first evidence for clinal variation in the relative strength of parental effects along environmental gradients.