Morphology and Evolution of the Ceratopsian Skull Andrew Knapp Submitted in partial fulfilment of the requirements of the Degree of Doctor of Philosophy School of Biological and Chemical Sciences Queen Mary University of London September 2019 Appendix A: Required statement of originality for inclusion in research degree theses I, Andrew Knapp, confirm that the research included within this thesis is my own work or that where it has been carried out in collaboration with, or supported by others, that this is duly acknowledged below and my contribution indicated. Previously published material is also acknowledged below. I attest that I have exercised reasonable care to ensure that the work is original, and does not to the best of my knowledge break any UK law, infringe any third party’s copyright or other Intellectual Property Right, or contain any confidential material. I accept that the College has the right to use plagiarism detection software to check the electronic version of the thesis. I confirm that this thesis has not been previously submitted for the award of a degree by this or any other university. The copyright of this thesis rests with the author and no quotation from it or information derived from it may be published without the prior written consent of the author. Signature: Andrew Knapp Date: 24th September 2019 1 Details of collaboration and publications: Publications arising from this thesis: Knapp A, Knell RJ, Farke AA, Loewen MA and Hone DWE (2018). Patterns of divergence in the morphology of ceratopsian dinosaurs: sympatry is not a driver of divergence. Proc. R. Soc. B. 285: 20180312. http://dx.doi.org/10.1098/rspb.2018.0312 Knapp A, Knell RJ and Hone DWE (2019). Modularity in the skull of the ceratopsian dinosaur Protoceratops and morphometric evidence of a socio-sexually selected origin for the frill. (prepared for submission) Collaborations: Andy Farke and Mark Loewen provided the taxonomic character matrix used in Chapter 2. Sandra Álvarez-Carretero performed the branch length estimations used to construct the time-scaled phylogeny in Chapter 4. Additional publications: O’Brien DM, Allen CE, Van Kleeck MJ, Hone DWE, Knell RJ, Knapp A, Christiansen S and Emlen DJ (2018). On the evolution of extreme structures: static scaling and the function of sexually selected signals. Animal Behaviour. 144; 95 – 108. 2 Abstract Socio-sexual selection has long been recognised as an important driver of evolution and is known to be an important influence on the morphology and behaviour of extant organisms. The long term macroevolutionary effects of socio-sexual selection are difficult to observe because they operate over long timescales. Theoretical work has suggested that socio-sexual selection can possibly influence speciation, adaptation and extinction rates, but these predictions remain largely untested. The fossil record provides a potential solution for testing macroevolutionary hypotheses regarding socio-sexual selection over millions of years. Recent studies have suggested that the growth patterns of cranial ornamentation in Ceratopsian dinosaurs resemble the growth patterns predicted for a trait under socio-sexual selection. Identifying the presence of socio-sexual selection in extinct organisms is problematic, but evidence from extant taxa provide important evidence in detecting its effects on morphology. In this project, I use a combination of traditionally defined character states and three-dimensional geometric morphometrics, on a large scale for the first time in a dinosaur clade, to test competing macroevolutionary hypotheses of socio-sexual selection. I first test the hypothesis that morphological diversity in ceratopsian dinosaurs is driven by species recognition by comparing differences in character states between ceratopsian clades. I evaluate morphological evidence for socio-sexual selection in fossil specimens, employing novel techniques to assess growth and morphological variation in the skull of the ceratopsian that is best-represented by fossil specimens, Protoceratops andrewsi. Lastly, I extend the morphological dataset to encompass other ceratopsian taxa and examine modularity, morphological disparity and evolutionary rates across the clade. My results refute the hypothesis that exaggerated ceratopsian traits evolved for species recognition and provide support for predictions that ceratopsian traits associated with socio-sexual selection, namely the frill and horns, formed distinct units that both developed and evolved at comparatively rapid rates. 3 Contents Page Abstract 3 Contents 4 List of tables 7 List of Figures 8 List of Institutional Abbreviations 10 Summary 11 1 Introduction 1.1 Background 1.1.1 Sexual selection 13 1.1.2 Social selection 15 1.1.3 Predicted effects of socio-sexual selection on macroevolution 16 1.2 Theoretical and empirical evidence for sexual selection 1.2.1 Sexual selection in extant organisms 17 1.2.2 Sexual selection in the fossil record 22 1.2.3 Ceratopsian dinosaurs 27 1.3 Methodology 1.3.1 Geometric morphometrics 30 1.3.2 Modularity and evolutionary rates 31 1.3.3 Data collection 33 1.3.4 Limitations of fossil specimens 34 1.4 Aims and objectives 35 1.5 Hypotheses 36 2 Patterns of divergence in the morphology of ceratopsian dinosaurs: sympatry is not a driver of ornament evolution. 2.1 Abstract 37 2.2 Introduction 38 2.3 Methods 41 2.4 Results 47 2.5 Discussion 53 4 3 Ontogeny and intraspecific variation of the ceratopsian dinosaur Protoceratops andrewsi 3.1 Abstract 57 3.2 Introduction 58 3.3 Methods 3.3.1 Experimental design 61 3.3.2 Statistical analysis 63 3.4 Results 3.4.1 Morphological trends in the skull of Protoceratops andrewsi 65 3.4.2 Variance and whole skull allometry 65 3.4.3 Modularity 70 3.4.4 Sexual dimorphism 74 3.4.5 Module allometry 75 3.4.6 Module disparity 77 3.5 Discussion 79 4 Modularity, morphological disparity and evolutionary rates in the ceratopsian skull 4.1 Abstract 84 4.2 Introduction 85 4.3 Methods 4.3.1 Data collection 91 4.3.2 Effects of size on shape 95 4.3.3 Modularity testing 95 4.3.4 Divergence times estimation 96 4.3.5 Phylogenetic factors affecting shape 97 4.4 Results 4.4.1 Modularity 98 4.4.2 Morphological variation 101 4.4.3 Allometry 105 4.4.4 Phylogenetic signal and evolutionary rates 111 4.4.5 Ceratopsid epiossifications 119 4.4.6 Triceratops, Torosaurus and Nedoceratops 123 4.5 Discussion 4.5.1 Overview 124 4.5.2 Modularity 124 4.5.3 Allometry and phylogenetic signal 127 4.5.4 Evolutionary rates and morphological diversity 129 4.5.5 Epiossifications 131 5 4.5.6 Status of Nedoceratops and Torosaurus 132 4.5.7 Conclusions 134 5 Conclusions 5.1 Summary of findings 135 5.2 Implications of study and future research 5.2.1 Overview 137 5.2.2 Modularity and evolution 138 5.2.3 Evolutionary rates and disparity 139 5.2.4 Allometry, Cope’s rule and heterochrony 140 5.2.5 Species recognition 141 5.2.6 A hypothesis for the role of the frill in ceratopsian evolution 142 5.2.7 Limitations of fossil data and possible solutions 143 Acknowledgements 146 References 148 Appendix 171 6 List of tables Page 2.1 Temporal calibrations and geographical locations of ceratopsian species 43 2.2 z-scores of second order polynomial regression parameters 52 3.1 Results of maximum likelihood and covariance ratio modularity analyses 72 3.2 Results of Hartigan’s dip test for Protoceratops data 74 3.3 Results of ANCOVA for Protoceratops skull size regressions 75 4.1 Results of allometry analyses for ceratopsian dataset 107 4.2 Results of PGLS analysis of evolutionary allometry 107 4.3 Results of phylogenetic signal analysis and net evolutionary rate analysis 113 Results of phylogenetic signal analysis, net evolutionary rates and morphological 121 4.4 disparity for ceratopsids 7 List of figures Page 1.1 The effect of differing species turnover on apparent species richness 23 1.2 Changes in skull shape with ontogeny in specimens of Protoceratops andrewsi, 27 illustrating growth of frill. 1.3 Examples of ceratopsian cranial diversity 28 2.1 Simplified phylogeny and examples of ceratopsian skulls 40 2.2 Time-scaled phylogeny of all known ceratopsian species 45 2.3 Pairwise comparison plots for sympatry categories and character classes 49 2.4 Distributions of simulated second-order polynomial model output parameters 50 2.5 Mean difference values of individual characters for each main sympatry category 51 3.1 Illustration of adult P. andrewsi skeleton 59 3.2 Layout of landmarks on P. andrewsi skull digital mesh model 62 3.3 Principal components analysis of 31 complete P. andrewsi skulls 66 3.4 Results of allometry analysis for whole-skull data 67 3.5 Per-landmark variance in the skull of P. andrewsi 69 3.6 Modules identified in the skull of P. andrewsi from EMMLi analysis 71 3.7 Projected shapes at minimum (Min) size and maximum (Max) size of the five 73 modules identified in P. andrewsi using EMMLi 3.8 Regression of module centroid size and CAC against skull centroid size 76 3.9 Relationship between module Procrustes variance and growth rate and 78 Procrustes variance and within-module correlation 3.10 Front view of projected shape at minimum and maximum size of the skull of 80 Protoceratops andrewsi 3.11 Hypothetical examples of allometric trajectories under different forms of sexual 82 dimorphism 4.1 Time-scaled phylogeny of ceratopsian taxa used in morphometric study 88 4.2 Landmark guide for morphometric study 93 8 4.3 Results of EMMLi modularity test 100 4.4 Phylomorphospace of whole-skull shape data 103 4.5 Phylomorphospaces of individual modules