AbstractThe ecological responses of plant populations along a successional gradient have been intensively examined; however, the evolutionary responses received much less attention. Here, I explored genetic changes of key phenotypic traits of a dominant clonal plant (Elytrigia atherica) along a saltmarsh successional gradient by collecting samples along the successional gradient in the high and low marsh and growing them in a common environment (greenhouse). Additionally, to explore whether changes in traits are driven by abiotic (e.g. clay thickness) and biotic (e.g. grazing intensity) variables along the successional gradient, I measured these two variables in the field. I found that clay thickness (a proxy of total nitrogen) increased along the successional gradient both in the high and low marsh; grazing intensity from hares (the most important herbivores) decreased along the successional gradient in the high marsh but did not change in the low marsh. Meanwhile, I found that growth in number of leaves and ramets decreased, while rhizome length increased, along the successional gradient for E. atherica collected from the high marsh. Opposite trends were found for E. atherica collected from the low marsh. Results suggest that, in the high marsh, herbivores may overrule nutrients to drive trait changes. That is, at the early successional stages, E. atherica had higher growth in number of leaves and ramets to compensate for high-intensity grazing. In the low marsh, nutrients may be the dominant driver for trait changes. That is, at the late successional stages, E. atherica had higher growth in number of leaves and ramets but shorter rhizomes to maximize its expansion under the favorable conditions (higher nutrient availability). Results suggest that ecologically important abiotic and biotic variables such as nutrients and herbivores may also have a substantial evolutionary impact on plant populations.
Genet assignment and population structure analysis in a clonal forest-floor herb, Cardamine leucantha, using RAD-seq
