Ecomorphological diversification of squamates in the Cretaceous

Abstract

Squamates (lizards and snakes) are highly successful modern vertebrates, with over 10 000 species. Squamates have a long history, dating back to at least 240 million years ago (Ma), and showing increasing species richness in the Late Cretaceous (84 Ma) and Early Palaeogene (66-55 Ma). We confirm that the major expansion of dietary functional morphology happened before these diversifications, in the mid-Cretaceous, 110-90 Ma. Until that time, squamates had relatively uniform tooth types, which then diversified substantially and ecomorphospace expanded to modern levels. This coincides with the Cretaceous Terrestrial Revolution, when angiosperms began to take over terrestrial ecosystems, providing new roles for plant-eating and pollinating insects, which were, in turn, new sources of food for herbivorous and insectivorous squamates. There was also an early Late Cretaceous (95-90 Ma) rise in jaw size disparity, driven by the diversification of marine squamates, particularly early mosasaurs. These events established modern levels of squamate feeding ecomorphology before the major steps in species diversification, confirming decoupling of diversity and disparity. In fact, squamate feeding ecomorphospace had been partially explored in the Late Jurassic and Early Cretaceous, and jaw innovation in Late Cretaceous squamates involved expansions at the extremes of morphospace.

Keywords: Cretaceous Terrestrial Revolution; Squamata; ecomorphology; macroevolution; palaeontology.

Figure 1.

Dental disparity in Mesozoic squamate genera (n = 220). (a) The relative (proportional) diversity of eight tooth morphotypes identified among modern and Mesozoic squamates, in 14 stage-level bins from the Late Jurassic to end-Cretaceous. For each tooth morphotype, information is provided on their general phylogenetic occurrences and broad diets as observed among living and fossil forms. (b) Stacked relative (proportional) diversity of the eight tooth morphotypes through time. (c) Pairwise comparisons of Bray–Curtis dissimilarity for dental occurrences in the 14 geological stages. Large circles and darker blue shading indicate greater dissimilarity in the dental occurrences between bins. Results excluding marine taxa and based on the absolute numbers of occurrences are presented in electronic supplementary material, figures S1 and S2.

Figure 2.

Temporal trends of early squamate size evolution, based on mandible lengths (n = 116). (a) Data are plotted at the stratigraphic midpoint for each taxon with temporal ranges denoted by grey horizontal bars. Terrestrial (black circle), marine (blue circle) and mosasauroid (blue triangle) taxa are labelled separately. (b) Comparison of jaw sizes in Early and Late Cretaceous terrestrial squamates using boxplots. Note the log₁₀ scaled y-axes in both plots.

Figure 3.

Jaw morphospaces of Mesozoic squamate genera. (a) Morphospace occupation from the Late Jurassic, Early Cretaceous and Late Cretaceous taxa denoted by convex hulls (n = 89). The convex hulls of the Late Jurassic and Early Cretaceous are expanded to incorporate the morphospace centroid of snakes (dotted lines), sampled from only the Late Cretaceous. (b) Landmark and semi-landmark positions illustrated on the jaw of Polyglyphanodon sternbergi (USNM 15477). The landmarks are (1) the most posterior point of the articular, (2) the most dorsal point of coronoid process, (3) ventral point of a vertical line from landmark 2, (4) the most posterior point of the most posterior teeth, (5) ventral point of a vertical line from landmark 4, (6) the most anterior and superior point of dentary, and (7) the most anteroventral point of dentary. Twenty-six semi-landmarks were used, and all of them are marked as yellow points. (c) Sum of variances and convex hull volume, with 95% error bars, are plotted for the three temporal bins, both excluding and including the snake centroid location in the Late Jurassic and Early Cretaceous bins (including highlighted in grey with ‘sn’). (d) Morphospace of Late Cretaceous taxa divided by clades and denoted by convex hulls (n = 74). (e) Morphospace of Late Cretaceous taxa divided by dietary groups and denoted by convex hulls (n = 74). In (a), PC1 is 35.4% of total variation and PC2 is 21.8%. In (b) and (c), PC1 represents 38.1% and PC2 equals 22.1%. Morphospace occupation based on PC3 and PC4 is illustrated in electronic supplementary material, figure S6.