干旱区生态环境知识资源中心

To assess the evolutionary significance of persistent seed banks, phenotypes of naturally germinating seeds must be compared with those that remain dormant under the same environmental conditions. Dormant seeds can often be induced to germinate by application of gibberellic acid (GA). However, this method is valid only if there are no phenotypic ’’side effects’’ of GA that could confound comparisons between dormant and naturally germinating seeds. We examined this assumption in Lesquerella fendleri, a short-lived perennial mustard of the desert Southwest. We exposed 3840 seeds from 16 maternal sibships to two different GA treatments (0 or 1 g/L GA) in two different germination environments (greenhouse and growth chamber), and measured germination and postgermination traits. As expected, application of GA increased germination. GA also had strong and long-lasting effects on seedling morphology. Seeds that received GA developed into seedlings that were taller, with fewer but longer leaves, than seeds that did not receive GA. Effects of GA on both dormancy and postgermination traits varied among maternal sibships. Our results indicate that for this species and this concentration of GA, morphological effects can be substantial. Further study is required to determine whether such side effects are found for lower concentrations of GA, or under conditions that encourage faster seedling growth. Nonetheless, the present results illustrate the importance of testing potential confounding effects of GA in studies of the evolution of seed dormancy and its influence on postgermination traits.