Widespread convergence towards functional optimization in the lower jaws of crocodile-line archosaurs

Abstract

Extant crocodilian jaws are subject to functional demands induced by feeding and hydrodynamics. However, the morphological and ecological diversity of extinct crocodile-line archosaurs is far greater than that of living crocodilians, featuring repeated convergence towards disparate ecologies including armoured herbivores, terrestrial macropredators and fully marine forms. Crocodile-line archosaurs, therefore, present a fascinating case study for morphological and functional divergence and convergence within a clade across a wide range of ecological scenarios. Here, we build performance landscapes of two-dimensional theoretical jaw shapes to investigate the influence of strength, speed and hydrodynamics in the morphological evolution of crocodile-line archosaur jaws, and test whether ecologically convergent lineages evolved similarly optimal jaw function. Most of the 243 sampled jaw morphologies occupy optimized regions of theoretical morphospace for either rotational efficiency, resistance to Von Mises stress, hydrodynamic efficiency or a trade-off between multiple functions, though some seemingly viable shapes remain unrealized. Jaw speed is optimized only in a narrow region of morphospace whereas many shapes possess optimal jaw strength, which may act as a minimum boundary rather than a strong driver for most taxa. This study highlights the usefulness of theoretical morphology in assessing functional optimality, and for investigating form-function relationships in diverse clades.

Keywords: Crocodylomorpha; Crurotarsi; biomechanics; functional landscape; optimality.

Figure 1.

Ecological disparity in crocodile-line archosaurs. Time-calibrated tree showing the sample of crurotarsan taxa assembled in this study. Branch colour denotes habitat groupings: freshwater (black), marine (blue) and terrestrial (red). Pie charts show the distribution of dietary categories within each taxonomic group. The topology used in this tree corresponds to Tree 1 in our study (see electronic supplementary material, table S3, for alternative topologies). Taxon silhouettes are sourced from PhyloPic (see acknowledgements).

Figure 2.

Performance surfaces of functions tested: (a) Von Mises stress, (b) rotational efficiency and (c) jaw area, for (a,b) showing the mean values from 1000 iterations with pseudo-randomly sampled input parameters within a given range (see §2). Areas with darker blue denote theoretical lateral lower jaw outlines with higher values for tested functions that are interpreted as showing reduced (a,c) or increased (b) fitness. Theoretical shapes across the morphospace are superimposed with the anterior end facing right. Jaw silhouettes are scaled to the first harmonic (i.e. length) in the figure; note that this does not reflect their relative size within the functional analyses. Grey areas denote impossible morphospace.

Figure 3.

Pareto optimality. (a,c,e) Combined performance surfaces for each pair of tested functional metrics: VMS and rotational efficiency (a), VMS and jaw area (c) and rotational efficiency and jaw area (e). Areas of morphospace with higher Pareto ranks are shown by darker colours. Grey areas indicate impossible morphospace and black dots (n = 243) denote the positions of empirical taxa. (b,d,f) Performance space plotted using the functional output of each theoretical jaw shape, corresponding to the performance surface to each plot’s immediate left. Each theoretical two-dimensional lower jaw shape is represented by a black dot, showing the heterogeneity of functional occupation. The Pareto front is shown as a solid black line and the grey area represents the range of possible solutions. (g) Performance surface combining all three functional metrics overlayed with a phylomorphospace, with empirical jaws split by taxonomic group. Topology corresponds to Tree 1 (electronic supplementary material, table S3, and see figures S5 and S6 for alternative topologies).

Figure 4.

Effects of ecology on functional optimization of two-dimensional lower jaw shapes. Performance surfaces of functions tested (as in figure 2): Von Mises stress (a,b), rotational efficiency (c,d) and jaw area (e,f), overlayed with empirical taxa grouped by habitat (a,c,e) and diet (b,d,f). Areas of morphospace with higher values for tested functions are shown by darker colours as in figure 2. Grey areas denote impossible morphospace. Equivalent plots for combined performance surfaces, as in figure 3, are provided in electronic supplementary material, figure S9.