Plasticity and evolutionary convergence in the locomotor skeleton of Greater Antillean Anolis lizards

Abstract

Plasticity can put evolution on repeat if development causes species to generate similar morphologies in similar environments. Anolis lizards offer the opportunity to put this role of developmental plasticity to the test. Following colonization of the four Greater Antillean islands, Anolis lizards independently and repeatedly evolved six ecomorphs adapted to manoeuvring different microhabitats. By quantifying the morphology of the locomotor skeleton of 95 species, we demonstrate that ecomorphs on different islands have diverged along similar trajectories. However, microhabitat-induced morphological plasticity differed between species and did not consistently improve individual locomotor performance. Consistent with this decoupling between morphological plasticity and locomotor performance, highly plastic features did not show greater evolvability, and plastic responses to microhabitat were poorly aligned with evolutionary divergence between ecomorphs. The locomotor skeleton of Anolis may have evolved within a subset of possible morphologies that are highly accessible through genetic change, enabling adaptive convergence independently of plasticity.

Developmental plasticity reveals hidden fish phenotypes and enables morphospace diversification

“…Developmental plasticity might lead adaptive evolution especially when environmental changes are severe and the plastic responses align with selection (Feiner et al. 2020; Radersma et al. 2020).…”

Evolution of the locomotor skeleton in Anolis lizards reflects the interplay between ecological opportunity and phylogenetic inertia

“…mainland lineages may also interact differently with their environment, and more information on functional aspects of morphology, including how morphological differences impact on ecological performances, is sorely needed [32][33][34]72 . However, plastic responses to different microhabitat use appear to be far too evolutionarily labile to leave a persistent signature on morphological divergence across the phylogeny 73 .…”

“… 78 ), all linear measurements of limb bones were divided by body size. We used centroid sizes of the pelvic girdles as a proxy of body size since it has been established that these centroid sizes are tightly correlated with snout-vent-length 73 , 79 , a common measure of body size in lizards. Due to this strong correlation between centroid size and snout-vent-length, our dataset should be broadly comparable to the large body of literature on Anolis morphology that relies on snout-vent-length as an estimate of body size.…”

“…On left sides of the pectoral and the pelvic girdles, we placed 18 landmarks on informative anatomical features using the R package “geomorph” (version 3.1.3) 81 . Landmarks were developed by partially adopting published landmark sets for the pectoral 38 and pelvic girdles 39 and follow a previously described method 73 . When specimens showed damage on the left side, we landmarked the right sides of the structures and used the R package “StereoMorph” 82 (version 1.6.3) to mirror landmarks onto the left side.…”

“…All measurements of bone length and the placing of landmarks were performed blindly with respect to the identity of the specimen and by the same person. The repeatability of the landmarking procedure (including the thresholding of raw imaging data to extract mesh files), was assessed in a previous study and estimated to 0.98 for the pectoral girdle and 0.94 for the pelvic girdle 73 . We performed a “generalized Procrustes analysis” to obtain Procrustes shape variables ( X -, Y -, and Z -coordinates) in the R package geomorph.…”