Date of Award

Fall 10-1-2021

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Geology and Geophysics

First Advisor

Briggs, Derek

Abstract

Echinoidea constitutes one of five major clades of living echinoderms, marine animals uniquely characterized by a pentaradial symmetry. Approximately 1,000 living and 10,000 extinct species have been described, including many commonly known as sea urchins, heart urchins and sand dollars. Today, echinoids are ubiquitous in benthic marine environments, where they strongly affect the functioning of biodiverse communities such as coral reefs and kelp forests. Given the quality of their fossil record, their remarkable morphological complexity and our thorough understanding of their development, echinoids provide unparalleled opportunities to explore evolutionary questions in deep-time, providing access to the developmental and morphological underpinnings of evolutionary innovation. These questions cannot be addressed without first resolving the phylogenetic relationships among living and extinct lineages. The goal of this dissertation is to advance our understanding of echinoid relationships and evolutionary history, as well as to explore more broadly the integration of phylogenomic, morphological and paleontological data in phylogenetic reconstruction and macroevolutionary inference.In Chapter 1, I report the results of the first phylogenomic analysis of echinoids based on the sequencing of 17 novel echinoid transcriptomes. Phylogenetic analyses of this data resolve the position of several clades—including the sand dollars—in disagreement with traditional morphological hypotheses. I demonstrate the presence of a strong phylogenetic signal for these novel resolutions, and explore scenarios to reconcile these findings with morphological evidence. In Chapter 7, I extend this approach with a more thorough taxon sampling, resulting in a robust topology with a near-complete sampling of major echinoid lineages. This effort reveals that apatopygids, a clade of three species with previously unclear affinities, represent the only living descendants of a once diverse Mesozoic clade. I also perform a thorough time calibration analysis, quantifying the relative effects of choosing among alternative models of molecular evolution, gene samples and clock priors. I introduce the concept of a chronospace and use it to reveal that only the last among the aforementioned choices affects significantly our understanding of echinoid diversification. Molecular clocks unambiguously support late Permian and late Cretaceous origins for crown group echinoids and sand dollars, respectively, implying long ghost ranges for both. Fossils have been shown to improve the accuracy of phylogenetic comparative methods, warranting their inclusion alongside extant terminals when exploring evolutionary processes across deep timescales. However, their impact on topological inference remains controversial. I explore this topic in Chapter 3 with the use of simulations, which show that morphological phylogenies are more accurate when fossil taxa are incorporated. I also show that tip-dated Bayesian inference, which takes stratigraphic information from fossils into account, outperforms uncalibrated methods. This approach is complemented in Chapter 2 with the analysis of empirical datasets, confirming that incorporating fossils reshapes phylogenies in a manner that is entirely distinct from increased sampling of extant taxa, a result largely attributable to the occurrence of distinctive character combinations among fossils. Even though phylogenomic and paleontological data are complementary resources for unraveling the relationships and divergence times of lineages, few studies have attempted to fully integrate them. Chapter 4 revisits the phylogeny of crown group Echinoidea using a total-evidence dating approach combining phylogenomic, morphological and stratigraphic information. To this end, I develop a method (genesortR) for subsampling molecular datasets that selects loci with high phylogenetic signal and low systematic biases. The results demonstrate that combining different data sources increases topological accuracy and helps resolve phylogenetic conflicts. Notably, I present a new hypothesis for the origin and early morphological evolution of the sand dollars and close allies. In Chapter 6, I compare the behavior of genesortR against alternative subsampling strategies across a sample of phylogenomic matrices. I find this method to systematically outperform random loci selection, unlike commonly-used approaches that target specific evolutionary rates or minimize sources of systematic error. I conclude that these methods should not be used indiscriminately, and that multivariate methods of phylogenomic subsampling should be favored. Finally, in Chapter 5, I explore the macroevolutionary dynamics of echinoid body size across 270 million years using data for more than 5,000 specimens in a phylogenetically explicit context. I also develop a method (extendedSurface) for parameterizing adaptive landscapes that overcomes issues with existing approaches and finds better fitting models. While echinoid body size has been largely constrained to evolve within a single adaptive peak, the disparity of the clade was generated by regime shifts driving the repeated evolution of miniaturized and gigantic forms. Most innovations occurred during the latter half of the Mesozoic, and were followed by a drastic slowdown in the aftermath of the Cretaceous-Paleogene mass extinction.

Download
COinS