Ecological differentiation and local adaptation mediate incipient speciation in the Gladiolus carneus (Iridaceae) species complex

Abstract

Ecological speciation is the process whereby ecologically-based divergent natural selection acting on populations results in the formation of new species. In the hyper-diverse Cape Floristic Region (CFR), the interactions between topographic, climatic, edaphic and biotic heterogeneity at fine-spatial scales is hypothesised to be a major driver of ecological speciation. However, investigations into the mechanisms mediating this diversity has largely focused on either identifying macroevolutionary drivers in a phylogenetic context or by documenting trait-by-environment associations between closely related taxa that are suggestive of local adaptation. To date, there has been relatively few studies that directly test whether ecologically driven divergence causes local adaptation or reproductive isolation in the CFR. In this thesis, I use Gladiolus carneus, a polymorphic geophyte that occurs along the entire elevational gradient of the CFR, to investigate whether ecological niche differentiation causes local adaptation and premating reproductive isolation. Specifically, in Chapter 2, I investigate whether ecological niche differentiation of the putative ecotypes of G. carneus causes strong premating reproductive isolation across the species complex. I found that the ecotypes of G. carneus are morphologically distinct and that they occupy distinct predicted and realised abiotic niches, resulting in, with only a few exceptions, strong ecogeographic isolation. I further found that the ecotypes occupy distinct phenological niches leading to moderate phenological isolation. For the pollinator niche, I found that all sampled populations were associated with a single highly effective functional pollinator, but at the ecotypic level, there were no consistent trends resulting in varying strengths of pollinator-mediated isolation. The strength of the ecogeographic, phenological and pollinator isolation lead to near complete premating reproductive isolation across the species complex. This chapter showcases that ecological niche differentiation can cause reproductive isolation between recently diverged taxa in the CFR. In Chapter 3, I used a common garden and a multi-year reciprocal translocation to test whether ecotypes of G. carneus that occur along an elevational gradient were locally adapted to their edaphic niche. In the common garden experiment, I found evidence that the ecotypes occupying the most extreme edaphic niches had a fitness advantage on their native soil. In the reciprocal translocations, I found that all ecotypes showed evidence of local adaptation. The two most elevationally extreme ecotypes had a fitness advantage both on their native soil, and at their native site. These results suggest that edaphic properties, along with other abiotic factors, can cause local adaptation in the CFR. Overall, my thesis provides evidence that niche differentiation can cause local adaptation and premating reproductive isolation between recently diverged taxa and that abiotic niche differentiation is likely a major driver of the early stages of speciation in the CFR.