Divergent evolution of traits and plasticity across climate gradients in a widespread tree species

Lerner, David , Gonzalez Rolfe, Rebeca , Greenbaum, Gili , Klein, Tamir

2026-06-17 JOURNAL OF ECOLOGY 2026   114(卷), 6(期), (null页)

查看原文

  • JCR分区:

    影响因子:

  • Species with large distributions that span significant environmental gradients have evolved complex mechanisms to adapt to different environments. This adaptation can be achieved both through the direct selection on traits that convey higher fitness under local environmental conditions, or through the evolution of phenotypic plasticity. Under a common-garden experimental setup, we tested for the keystone African tree species Acacia tortilis how the interaction between direct selection on traits and evolution of plasticity has led to its widespread distribution to two contrasting environments: deserts and tropical savannas. In addition, we tested the predictability of evolution by comparing populations in opposing deserts of its distribution, the Sahara and Kalahari deserts. We observed significant differences in how various phenotypes responded to the harsh desert environment compared with the more moderate tropical-savanna environment, with a stronger among-population differentiation expressed under desert conditions. We found an isolation-by-environment signal mainly for phenotypic plasticity, with a strong association to precipitation gradients, suggesting that differentiation across the tropic-desert gradient occurs in this species primarily through selection on trait plasticity rather than trait means. Interestingly, the two desert-edge populations showed similar plastic responses, despite the large geographical distances between them, suggesting convergence in plasticity. Synthesis. Taken together, our results suggest that phenotypic plasticity plays an important role in differentiation across the tropical-savanna to desert gradient in Acacia tortilis, with similar phenotypic responses observed between climatically comparable populations at opposing range edges.