Quantification provides a conceptual basis for convergent evolution

Abstract

While much of evolutionary biology attempts to explain the processes of diversification, there is an important place for the study of phenotypic similarity across life forms. When similar phenotypes evolve independently in different lineages this is referred to as convergent evolution. Although long recognised, evolutionary convergence is receiving a resurgence of interest. This is in part because new genomic data sets allow detailed and tractable analysis of the genetic underpinnings of convergent phenotypes, and in part because of renewed recognition that convergence may reflect limitations in the diversification of life. In this review we propose that although convergent evolution itself does not require a new evolutionary framework, none the less there is room to generate a more systematic approach which will enable evaluation of the importance of convergent phenotypes in limiting the diversity of life's forms. We therefore propose that quantification of the frequency and strength of convergence, rather than simply identifying cases of convergence, should be considered central to its systematic comprehension. We provide a non-technical review of existing methods that could be used to measure evolutionary convergence, bringing together a wide range of methods. We then argue that quantification also requires clear specification of the level at which the phenotype is being considered, and argue that the most constrained examples of convergence show similarity both in function and in several layers of underlying form. Finally, we argue that the most important and impressive examples of convergence are those that pertain, in form and function, across a wide diversity of selective contexts as these persist in the likely presence of different selection pressures within the environment.

Keywords: convergence; evolutionary ecology; homoplasy; methods; parallelism; phylogenetic comparative methods.

Figure 1

Tanglegram comparing a molecular phylogenetic tree (left) with a phenetic tree of defensive traits (right) for a set of plant species. Lines between trees link the same species and crossing lines indicate a lack of similarity in the two trees (e.g. where phenotype is more similar than implied by phylogeny, indicative of convergence). From Agrawal & Fishbein (2006), reproduced with permission of the authors and publisher.

Figure 2

Representation of the plot‐space used by the pairwise distance‐contrast method. This method plots phylogenetic distances against phenotypic distance and the results are broadly interpreted as in the differently shaded regions. Convergence (or stasis) is considered when there has been little phenotypic divergence over large phylogenetic distances (the area in green).

Figure 3

Graphical representation of output from SURFACE analyses. In this case there have been convergent shifts to both the red and blue regimes (black branches represent the ancestral regime). More specifically, the blue regime has arisen on four separate occasions (marked by #), and the red regime has arisen on only three separate occasions (marked by *), despite containing more contemporary species than the blue regime.

Figure 4

Diagrammatic example showing situations that would result in relatively low (left) and high (right) Wheatsheaf (w) index values for a given tree. In this example there are data for a single trait (inset plots show distribution) for 20 species overall, 16 non‐focals (black tips) and four focals (red tips). A lower index results from closely related focal species with trait values that overlap with non‐focals. By contrast, a higher index results from more distantly related focals with highly distinct trait values from non‐focals.

Figure 5

Examples of different kinds of animal camouflage. (A) Transparency in the drepanid moth Deroca hidda (Drepaninae). (B) Background matching by the mossy leaf‐tailed Gecko, Uroplatus sikorae. (C) Likely disruptive patterning on the Balearic toad, Bufo viridis. (D) Many caterpillars (Lepidoptera) resemble twigs, a form of camouflage known as masquerade. Here the caterpillar is somewhat out of its protective habitat, away from the twigs that it mimics. Photo credits: (A) John Hortsman/‘“itchydogimages” on Flickr’; (B–D) Michael and Richard Webster.