Fig. 1: An overview of the type of experiment and effect sizes, and species included in the dataset.

a, An outline of the experimental design we sought: a single ancestral (ANC) population is split into two or more replicate populations that are exposed to different environments or selective regimes (blue or red). Within-environment crosses are obtained between replicates evolving in the same environment (for example, A1 × A2 and B1 × B2). b, Between-environment crosses are obtained between populations evolving in different environments (for example, A1 × B1, A1 × B2, A2 × B1 and A2 × B2). If estimates of reproductive isolation are greater between environments than within environments, this is evidence for ecological speciation, whereas if they are equal, this provides evidence for mutation-order speciation. c, The dataset is predominantly comprised of data from invertebrates that are best suited for experimental evolution studies, like Drosophila melanogaster (image 1), Ceratitis capitata (image 2) and Tetranychus urticae (image 3), though some vertebrate translocation experiments, such as those done in Poecilia reticulata (image 4) also fit the criteria. d, The phylogeny of all the species included in the dataset, from yeast to nematodes to vertebrates, with the number of effect sizes (k) (and studies in parentheses) each species contributes towards the meta-analysis. Photos in c reproduced with permission from: 1, Alexis Tinker-Tsavalas; 2, Daniel Feliciano; 3, Gilles San Martin; 4, ref. ¹²², under a Creative Commons license CC BY 4.0. Silhouettes in d reproduced from PhyloPic under a Creative Commons license CC0 1.0: Sophophora melanogaster, Andy Wilson; Gallus gallus, Steven Trver; Poecilia reticulata, seung9park; Tribolium castaneum and Tetranychus urticae, Christoph Schomburg; Acanthoscelides obtectus, Birgit Lang; Caenorhabditis elegans, Jake Warner; Saccharomyces cerevisiae, Wayne Decatur.