Figure legends:

Figure 1: Host-microbiome interactions are governed by the complex interplay between host genotype, microbial genotype/community composition, and the environment (A). The environment directly affects both host and microbial genotypes/communities and microbes and host reciprocally affect each other. Reciprocal transplants of seeds into natal and non-natal environments, with and without microbes can reveal when microbes are responsible for host adaptive responses. For example, without microbes there is no pattern of adaptation (B), but when microbes are manipulated in reciprocal transplant experiments you may see either microbe-mediated local adaptation (C) when local plant genotypes have higher fitness than foreign genotypes because of a genotype-specific affiliation with locally important microbe(s) or microbe-mediated adaptive plasticity (D) when local plant phenotypes have higher fitness than foreign phenotypes as a result of interactions with locally important microbes. Squares represent plants collected from population 1 (P1), and circles represent plants collected from population 2 (P2). Population 1 plants can either be transplanted into their natal habitat (P1, represented by a square) or into a foreign habitat (P2, represented by a circle), both with (C,D) and without (B) microbes.

Figure 2: A fully factorial reciprocal transplant design can be used to transplant seeds and microbes into home and away habitats to investigate how microbes affect patterns of local adaptation. To definitively attribute effects to microbes you need to include sterilized controls (depicted here as a red line crossing out the microbial communities). Squares represent plants and microbes collected from population 1 (P1), and circles represent plants and microbes collected from population 2 (P2). Population 1 plants and microbes can either be transplanted into their natal habitat (P1, represented by a square) or into a foreign habitat (P2, represented by a circle).

Figure 3: The additional manipulation of moving microbes and plant genotypes between sites can unravel if G_pxG_mxE interactions dominate (A), in this case positive effects of microbial symbionts would only be evident in their natal habitat. Alternatively, if G_pxG_m interactions dominate (B), plants would still benefit from microbial symbionts when both were moved into novel habitats. Microbe-mediated adaptive plasticity could be the result of either microbes specialized for their particular habitat (C) or the result of either direct or indirect (plant mediated) changes to microbial function or shifts in community composition that affect plant fitness. Microbe-mediated plasticity in important plant functional traits, such as flowering time (D) might underlie microbe-mediated adaptive plasticity, where microbes push flowering time phenotype into an optimal range (depicted by the shading) for a given habitat. Squares represent plants and microbes collected from population 1 (P1), and circles represent plants and microbes collected from population 2 (P2). Population 1 plants and microbes can either be transplanted into their natal habitat (P1, represented by a square) or into a foreign habitat (P2, represented by a circle) and vice versa.

Figure 4: Experiments to identify microbe-mediated adaptation will require identifying a source of microbial inoculum (A) and determining proper controls (B) for the experimental design. Another major consideration is where the experiment will take place, evolutionary ecology experiments with plants are typically performed in common gardens (greenhouses or outdoors) or as reciprocal transplant experiments.