Variation in seed number/mass trade-off for Pinus yunnanensis along a natural climate gradient in the mountains of Southwest China

A putative seed number/mass trade-off is fundamental to understanding the evolution of forest species and their capacity to recover from increasing losses to drought and harvest; however, such a trade-off is not always observed, and variation in this trade-off could arise if species face strong selection pressure for one or the other of these traits under different climatic factors. Yet to date few studies have considered the response of this trade-off to climatic gradients. In this study, three seed traits (seed mass, seed number, and seed abortion) at the cone level for 2,671 cones of Pinus yunnanensis collected from nine populations in the mountains of Southwest China were examined. The evidence for the potential trade-off between seed mass and number both within and across populations was assessed, In addition, the strength and direction of this trade-off along different climatic conditions was quantified. A significant negative relationship between seed mass and seed number was found across populations, which didn’t occur consistently within populations. Seed mass was strongly determined by seed abortion and the strength of the relationship between seed mass and seed abortion rate varied significantly among populations. Based on the climatic variables of populations, mean annual precipitation and temperature had a significant positive effect on seed mass and seed number, but a negative effect on seed abortion rate. However, the variation of filled seed number in the cone was much less than that of seed mass due to the stronger seed number/mass trade-off with decreasing mean annual precipitation. Thus, within populations, selection may be able to act more strongly on seed mass than on seed number in drier climate. We have found strong support for the intra-specific seed number/mass trade-off at the cone level of Pinus species. Importantly, the strength of this trade-off varied significantly among populations and seed abortion might affect seed mass through its association with this trade-off. Climatic factors generate selection both on seed mass and seed number, and the harsher climates (colder and drier climate) caused a reduction both in them. Our results highlight that climate conditions may co-vary or interact with life-history constraints, which potentially determines the natural population development and distribution.