Paleobiogeographic Analyses of Late Ordovician Faunal Migrations: Assessing Regional

Home , Asaphidae, Ascocerida, Catenipora, Cheiruridae, Colpocoryphe, Geodispersal

and Continental Pathways and Mechanisms

A thesis presented to

the faculty of

the College of Arts and Sciences of Ohio University

In partial fulfillment

of the requirements for the degree

Master of Science

Adriane R. Lam

May 2015

This thesis titled

Paleobiogeographic Analyses of Late Ordovician Faunal Migrations: Assessing Regional

and Continental Pathways and Mechanisms

ADRIANE R. LAM

has been approved for

the Department of Geological Sciences

and the College of Arts and Sciences by

Alycia L. Stigall

Associate Professor of Geological Sciences

Robert Frank

Dean, College of Arts and Sciences 3

ABSTRACT

LAM, ADRIANE R., M.S., May 2015, Geological Sciences

Paleobiogeographic Analyses of Late Ordovician Faunal Migrations: Assessing Regional and Continental Pathways and Mechanisms

Director ofThesis: Alycia L. Stigall

Late Ordovician strata of the Cincinnati Basin record an invasion during the

Richmondian Age. Competing hypotheses exist in the literature about the origin and dispersal routes used by the invasive taxa. In this thesis, a suite of traditional and novel paleobiogeographic methods are used to (1) identify geographic source regions for the invasive taxa, (2) reconstruct dispersal paths used, and (3) analyze speciation patterns that influenced biogeographic evolution of taxa. Parsimony Analysis of Endemicity (PAE) applied to over 60 genera that participated in the invasion interval for the C1-C5 depositional sequences supported multiple dispersal routes within Laurentia and between

Laurentia and Baltica. Larval type for invasive taxa was dominantly planktotrophic or planula larvae; both have high dispersal potential. Phylogenetic biogeographic analyses using parsimony and model-based approaches were applied to ten clades of Middle-Late

Ordovician taxa to compare ancestral range reconstructions characterized speciation type and identify dispersal paths. Methods recovered similar patterns, but differed when interpreting speciation mode. Based on these analyses and a review of the literature,

Richmondian invaders are determined to include taxa from multiple geographic areas.

Local and global oceanic currents and changing tectonic factors, promoted the immigration of taxa with high larval dispersal potential into the Cincinnati basin. 4

DEDICATION

To all the teachers, professors, and advisors throughout my student career who have

encouraged and shaped me into the scientist I am today, this one’s for you.

Terviseks!

ACKNOWLEDGMENTS

First and foremost, I must thank my advisor, Professor Alycia Stigall, for her insightful discussions, assistance, encouragement, and unwavering support. Not only has she helped strengthen my abilities as a scientist, but she has also instilled a sense of accomplishment and confidence in me as a young woman in the geosciences. Portions of this thesis could not have been possible through the recommendations and guidance of several researchers: Davey Wright for his encouragement working with R, David Bapst for insightful discussion into time-slicing, and Nick Matzke for assistance with

BioGeoBEARS. Correlations among basins were greatly assisted by Dr. Stig Bergström.

I would also like to thank my Paleo lab comrades who have put up with my complaints, hysterics, and sense of humor while I wrote this thesis: Jennifer Bauer, Michael Hils,

Sarah Trubovitz, Wesley Parker, Zoe Zeszut, Mackenzie Glasgow, and Timothy

Henderson. In addition, I would like to thank my committee members, Drs. R. Damian

Nance, Greg Springer, and Dave Kidder, for assistance during my time at Ohio

University. I must also thank my family for understanding that fossils are awesome, and supporting me while I follow my dreams. Last, but certainly not least, I owe a great deal to my fiancé, Jacob Uzel, for standing by me through all of this and being my biggest support system. This project was supported by funds provided by a Paleontological

Award from the Dry Dredgers of Cincinnati, Ohio, the Paleontological Society Arthur J.

Boucot Award, an Ohio University Geology Alumni Research Grant, and an Ohio

University Geology Alumni Research Fellowship. Additional support was provided by a

National Science Foundation Grant (EF-1206750, EAR-0922067) to A.L. Stigall. 6

TABLE OF CONTENTS

Page

Abstract ...... 3 Dedication ...... 4 Acknowledgments...... 5 List of Tables ...... 9 List of Figures ...... 10 Chapter 1: Introduction ...... 13 Chapter 2: Pathways and Mechanisms of Late Ordovician (Katian) Faunal Migrations of Laurentia and Baltica ...... 17 Abstract ...... 17 Introduction ...... 18 Methods ...... 19 Parsimony Analysis of Endemicity ...... 19 Analysis of Larval Type, Paleoceanography, and Paleoclimate ...... 21 Results and Interpretation ...... 22 Larval Type ...... 22 Paleobiogeography ...... 22 Paleoceanography and Paleoclimate ...... 25 Discussion ...... 26 Chapter 3: Paleobiogeographic Analyses of Middle to Late Ordovician Taxa and Comparison of Analyses Using Parsimony and Maximum-Likelihood Methods ...... 29 Introduction ...... 29 Geologic Setting ...... 32 Methods ...... 36 Phylogenetic and Species Occurrence Data ...... 37 Temporal and Geographic Framework ...... 39 Parsimony-based Phylogenetic Biogeographic Analyses ...... 41 Bayesian Analyses ...... 43 Results ...... 45 Ancestral Range Reconstructions ...... 45 7

Speciation Mode Analyses ...... 51 Biogeographic Relationships among Areas ...... 54 Area Relationships and Dispersal Patterns ...... 55 Discussion ...... 71 Comparison of Maximum Likelihood and Parsimony Analyses ...... 71 Comparison of Speciation Events to Other Time Intervals ...... 74 Biotic Influences on Speciation Patterns ...... 75 Paleoceanographic and Tectonic Influences on Species Dispersal ...... 78 Conclusions ...... 85 Chapter 4: Origin, Pathways, and Mechanisms of the Late Ordovician Richmondian Invasion: A Review...... 87 Introduction ...... 87 Overview of the Richmondian Invasion ...... 88 A True Invasion or Recurrent Fauna? ...... 95 Hypotheses of the Geographic Origin and Dispersal Pathways of Invasive Taxa ...... 97 Arctic (Equatorial) Pathway ...... 97 Northeastern Canada (Quebec, Ontario) ...... 100 Marginal Laurentian Basins ...... 102 Baltic Pathways ...... 103 Multidirectional Dispersal Paths ...... 104 Invasion dynamics ...... 107 Tectonic and Paleoclimatic Drivers of the Invasion ...... 107 Biological Facilitators of the Invasion ...... 110 Oceanographic Drivers of Long-Distance Dispersal ...... 111 Conclusions ...... 112 Chapter 5: Conclusions ...... 116 References ...... 119 Appendix 1: Species Occurrence Data ...... 134 Appendix 2: Correlation Charts ...... 142 Appendix 3: References for Species Occurrences and Stratigraphy ...... 148 Appendix 4: Stratocladograms ...... 158 8

Appendix 5: LBPA Tables ...... 160 Appendix 6: LBPA Most Parsimonious Tree Topologies ...... 172 Appendix 7: Sample BioGeoBEARS Code for Eochonetes ...... 179 Appendix 8: BioGeoBEARS Probability Models...... 192

LIST OF TABLES

Page

Table 1: Focal taxa for biogeographic analyses. Clade size indicates the number of terminal taxa after species were combined or removed from the published phylogenetic hypothesis to resolve polytomies. Refer to Figure 4 for an explanation of ages used in this study...... 37

Table 2: Distribution of cladogenetic events (raw count with percent in parenthesis) attributable to vicariance or dispersal processes per time slice as interpreted from Fitch Parsimony and best-fit models from BioGeoBEARS analyses...... 46

Table 3: Comparisons of the fit of DEC, DIVA, and BayArea ancestral range estimations with the possibility of founder-event speciation (+J) for each phylogeny from the BioGeoBEARS analyses. The log-likelihood values (lnL) from the analyses are given, as well as the Akaike Information Criterion (AIC). ΔAIC shows the difference in AIC values compared with the best-fit model. Akaike weight (ωi) values give the relative likelihood of each model. The d and e parameters are an estimate of dispersal and extinction rate, respectively, as measured along branches within each phylogeny. The j parameter is a measure of the relative weight of founder-event speciation. Best-fit models for each phylogeny are indicated in bold...... 52

Table 4: Current hypotheses as to the origins of the Richmondian invaders...... 99

LIST OF FIGURES

Page Figure 1. Paleogeographic map of Laurentia and Baltica during the Late Ordovician with inferred dispersal pathways and major paleoclimate and oceanographic influences on larval dispersal labeled. Biogeographic areas analyzed are: 1, Baltic Basin; 2, Scoto- Appalachian Basin; 3, Anticosti Island; 4, Appalachian Basin; 5, Cincinnati Basin; 6, Nashville Dome; 7, Upper Mississippi Valley; 8, Western Midcontinent (Texas and Oklahoma Basins); 9, North of the Transcontinental Arch (Hudson, Williston, and Bighorn Basins). Paleoceanography after Cocks & Torsvik (2001); current direction and storm tracks after Poussart et al. (1999), Hermann et al. (2004), and Ettensohn (2010) ...... …..20

Figure 2: Parsimony Analysis of Endemicity areagrams. Areas group together in sister- area relationships based on the number of shared endemic taxa; the nested patterns present indicate hierarchical relationships of shared endemic taxa, which is a different metric than overall similarity (similarity values in supplementary materials archive: http://dx.doi.org/10.6084/m9.figshare.1194963). Shifts in area relationships between time slices indicate dispersal between those basins...... 23

Figure 3: Paleogeographic reconstruction of the Late Ordovician Sandbian Stage (460 Ma) with oceans and approximate location of subduction zones marked. Numbers indicate geographic areas used in this study: 1, Northern Laurentia (Northwest Territories, Nunavut, British Columbia); 2, North of the Transcontinental Arch; 3, Western Midcontinent; 4, Upper Mississippi Valley; 5, Cincinnati Basin; 6, Southern Appalachian Basin; 7, Northern Appalachian Basin; 8, Southern Laurentia (Newfoundland, Quebec, Anticosti Island); 9, Scoto-Appalachia; 10, Avalonia; 11, Baltica; 12, Tarim Plate; 13, Northern Gondwana (Australia, Kazakhstan, Japan); 14, Southern Gondwana (Arabia, Turkey, Bohemia, Spain, France). Colors of areas match those used in Figure 5. Box indicates the area of focus in Figure 8. Map modified from Torsvik and Cocks (2013)...... 33

Figure 4: Chronostratigraphic chart of the Middle to Late Ordovician interval, with global stages, North American Series, Stages, and depositional sequences for the Cincinnati Basin as defined by Holland and Patzkowsky (1996) marked. Time slices used in this study (T0-T3) are indicated above the depositional sequences. Taconic orogeny tectophases indicate activity that occurred on the southern margin of Laurentia, dominant carbonate type within southern Laurentian basins, and important paleoclimatic events and carbon isotope excursions (e.g., GICE, HICE, Boda Event). Abbreviations: GICE, Guttenberg δ13C excursion; HICE, Hirnantian δ13C excursion; G, Gamachian; Ki, Kircklandian; Ro, Rocklandian; H, Hirnantian. Timescale modified from Holland and Patzkowsky (1996); Taconic tectophases after Ettensohn (2010); occurrence of GICE after Young et al. (2005); occurrence of Boda Event after Fortey and Cocks (2005); occurrence of HICE after Bergström et al. (2006)...... 34 11

Figure 5: Cladograms with area reconstructions within brachiopod and trilobite taxa using Fitch Parsimony and maximum likelihood models generated within the R package BioGeoBEARS. A) Brachiopod Fitch Parsimony, B) brachiopod maximum likelihood models, C) trilobite Fitch Parsimony, D) trilobite maximum likelihood models...... 47

Figure 6: Cladograms depicting speciation processes among brachiopod and trilobite taxa using Fitch Parsimony and maximum likelihood models generated within the R package BioGeoBEARS. A) Brachiopod Fitch Parsimony, B) brachiopod maximum likelihood models, C) trilobite Fitch Parsimony, D) trilobite maximum likelihood models...... 49

Figure 7: Strict consensus vicariance and geodispersal area cladograms from Lieberman- modified Brooks Parsimony analyses of the T0 through T3 time slice data. Vicariance trees indicate the relative order in which barriers fell, and geodispersal trees indicate the order that barriers where removed between areas. Bootstrap (plain text) and jackknife (italics) values indicate node support. T0 time slice trees are the strict consensus of two most parsimonious vicariance trees (length of 69 steps, CI 0.85, RI 0.78) and three most parsimonious geodispersal trees (length of 72 steps, CI 0.90, RI 0.81). T1 time slice trees are the strict consensus of 40 most parsimonious vicariance trees (length of 109 steps, CI 0.77, RI 0.46) and 36 most parsimonious geodispersal trees (length of 92 steps, CI 0.89, RI 0.50). T2 time slice trees are the strict consensus of 84 most parsimonious vicariance trees (length of 82 steps, CI 0.80, RI 0.68) and 24 most parsimonious geodispersal trees (length of 70 steps, CI 0.88, RI 0.76). T3 time slice trees are the strict consensus of three most parsimonious vicariance trees (length of 130 steps, CI 0.76, RI 0.55) and 10 most parsimonious geodispersal trees (length of 101 steps, CI 0.89, RI 0.65)...... 57

Figure 8: Pathways and basin associations reconstructed from Fitch Optimization and maximum-likelihood analyses. Arrows indicate paths that were identical between both methods; stars indicate basins that were reconstructed to share area relationships by both methods but in which the directionality of dispersal differed between reconstructions. Dashed lines indicate speciation by vicariance, and solid lines indicate speciation by dispersal. Colors indicate clade identity: purple, Homalonotidae; green, Tetralichinae; red, Glyptorthis; orange, Flexicalymene; black, Thaleops; blue, Bumastoides; brown, Eochonetes; white, Plaesiomys; yellow, Hebertella. See Figure 3 for area names...... 61

Figure 9: Location, extent of exposed Ordovician outcrop, lithostratigraphy and sequence stratigraphy of the Cincinnati Basin. The first phase of the Richmondian Invasion is indicated in red; second phase in orange. Outcrop belt modified from Holland (1993); stratigraphy modified from Holland and Patzkowsky (1996)...... 89

Figure 10: Stratigraphic ranges of select Richmondian Invaders in the Cincinnati Basin. List of invaders from Holland (1996). Occurrence data from the Paleobiology Database, Frey (1985), Kelly and Pope (1979), Dalvė (1948)...... 92

Figure 11: Competing hypotheses as to the geographic origin and dispersal pathways of species participating in the Richmondian Invasion. The dashed line from Avalonia indicates that this may be a source region, requiring further studies. The multidirectional pathways hypothesis proposed by Wright & Stigall (2013b) includes the Arctic invasion, northeastern Canada, and marginal basins hypotheses, and was expanded to include the Baltic basin by Lam and Stigall (2015). Figure modified after Cocks and Torsvik (2005), Cocks and Torsvik (2011)...... 101

CHAPTER 1: INTRODUCTION

This thesis is a compilation of three separate studies investigating potential source areas and pathways used by invasive taxa during the Late Ordovician Richmondian

Invasion, as well as a comparison of parsimony and maximum likelihood methods in discerning large-scale migration patterns during the Middle through Late Ordovician. The

Late Ordovician Richmondian strata in the Cincinnati Basin records a mass faunal migration event coinciding with lithologic and paleoclimatic changes within Laurentia.

The source areas and pathways used by invasive taxa have been disputed in the literature for over a century. Therefore, the goal of this thesis is to test the source regions of invasive taxa that have not previously been incorporated into a thorough analysis due to their lack of phylogenetic hypotheses.

The Middle through Late Ordovician of Laurentia was a time of increased tectonic activity as a result of the Taconic orogeny, in which multiple volcanic island arcs and Gondwanan terranes collided with the southern margin of the paleocontinent

(Shanmugam & Lash, 1982; Macdonald et al., 2014). The Blountian tectophase of the

Taconic orogeny began during the Sandbian Stage, as the locus of tectonic activity was focused on the southern margin of Laurentia from Georgia into Virginia (Bradley, 1989).

Siliciclastics were introduced into the southern Appalachian basins, but did not cause a change in tropical carbonate production on the eastern margin of Laurentia (Holland &

Patzkowsky, 1996). The Taconic tectophase began during the late Sandbian Stage in which tectonic activity shifted east towards the New York promontory (Ettensohn, 2010). 14

During this time, siliciclastics were deposited throughout the southern margin basins of

Laurentia. Flexural downwarping, combined with a transgression at the M4/M5 sequence boundary of Holland and Patzkowsky (1996), led to the introduction of cool, nutrient rich waters into the continental interior through the Sebree Trough. Coincidentally, this led to a switch from tropical to temperate carbonate deposition that took place within the southern Laurentian basins. Tropical carbonate deposition became confined to the western continental interior basins, where taxa that inhabited the southern basins retreated.

By the Richmondian Stage, tropical carbonate deposition resumed on the southern margin of Laurentia, either due to the infilling of the Sebree Trough, which led to the elimination of cool-water circulation into the craton, or to the globally recognized warming trend, the Boda Event (Fortey & Cocks, 2005). The Richmondian Invasion within the Cincinnati Basin coincides with the switch back to tropical style carbonate deposition, as well as a major transgression at the C5 sequence boundary (Holland &

Patzkowsky, 1996).

The source of the invaders has been debated in the literature for over a century.

Early authors called the invasive fauna the Black River or Arctic fauna, as several genera were found to occupy younger strata in Canada and western North America (Holland,

1996). This led to the Arctic Invasion hypothesis, even though the aforementioned areas were located equatorially during the Late Ordovician. Jin (1999) proposed many of the taxa originated from marginal basins surrounding Laurentia, moving into and occupying almost all basins within the paleocontinent by the end of the Richmondian (the marginal 15 basins hypothesis). Using phylogenetic hypotheses of orthid brachiopods, Wright and

Stigall (2013a) concluded that taxa came from both the western midcontinent as well as marginal basins, thus inferring that taxa traveled along multidirectional dispersal pathways. Bauer (2014) added to this hypothesis through her phylogenetic revision of the sowerbyellid brachiopod Eochonetes, which supported the Artic Invasion hypothesis.

Thus, recent analyses have provided support for multiple invasion routes operating simultaneously. The validity of the multiple pathway hypothesis and the single pathway hypotheses is tested in this thesis.

As the majority of the invaders lack a previously published phylogenetic hypothesis of evolutionary relationships, Parsimony Analysis of Endemicity (PAE) is used to analyze area relationships in Chapter 2. PAE is a useful approach to determine paleobiogeographic relationships using presence/absence data across time slices. The second chapter of this thesis employs this method across the C1 to C5 depositional sequences to 1) determine dispersal pathways of invasive taxa among Laurentian basins, and 2) test the hypothesis that taxa were dispersing between Baltica and Laurentia, thus indicating that Baltica was a source region for some of the invaders.

Chapter 3 focuses on analyzing biogeographic patterns within ten clades of trilobite and brachiopods for which species-level phylogenetic hypotheses do exist within a phylogenetic biogeographic analysis. Two methods of analysis are employed: a parsimony-based and a model–based approach. Parsimony-based methods have a long history of use for determining area relationships and reconstructing ancestral ranges in the fossil record. Recently, advances in maximum likelihood and Bayesian analyses 16 through the R program BioGeoBEARS (Matzke, 2013a) has allowed for determination of area relationships using probabilistic inference. Although this method is widely used by molecular phylogeneticists, it has not previously been applied to Paleozoic fossils. In chapter 3, both methods are used to 1) reconstruct ancestral ranges, 2) examine geographic influences on speciation mode, 3) analyze dispersal routes and area relationships, and 4) determine which method, if either, is best for use on deep-time phylogenetic hypotheses.

Chapter 4 combines these new analyses with previously published literature to create a synthetic review of the Richmondian Invasion. The many prior hypotheses of dispersal paths are evaluated, and the biologic, tectonic, and paleoclimatic factors that facilitated the Richmondian Invasion are reviewed. Chapter 5 synthesizes the primary results of these three distinct, but related analyses.

CHAPTER 2: PATHWAYS AND MECHANISMS OF LATE ORDOVICIAN

(KATIAN) FAUNAL MIGRATIONS OF LAURENTIA AND BALTICA

Lam, A. R., and Stigall, A. L. (2015) Pathways and mechanisms of Late Ordovician (Katian) faunal migrations of Laurentia and Baltica. Estonian Journal of Earth Sciences, 64 (1), 62-67.

Abstract

Late Ordovician strata within the Cincinnati Basin record a mass faunal migration event during the C4 and C5 depositional sequences. The geographic source region for the invaders and the paleoceanographic conditions that facilitated dispersal into the

Cincinnati Basin have previously been poorly understood. Using Parsimony Analysis of

Endemicity (PAE), biogeographic relationships among Laurentian and Baltic basins were analyzed for each of the C1-C5 depositional sequences to identify dispersal paths. The results support multiple dispersal pathways, including three separate dispersal events between Baltica and Laurentia. Within Laurentia, results support dispersal pathways between areas north of the Transcontinental Arch into the Western Midcontinent, between the Upper Mississippi Valley into the Cincinnati Basin, and between the pericratonic Scoto-Appalachian Basin and the Cincinnati Basin. These results support the hypothesis that invasive taxa entered the Cincinnati Basin via multiple dispersal pathways, and that the equatorial Iapetus current facilitated dispersal of organisms from

Baltica to Laurentia. Within Laurentia, surface currents and large storms moving from the northeast to the southwest likely influenced dispersal of organisms. Larval states were 18 characterized for the Richmondian invaders, and most invaders were found to have had planktotrophic planktic larvae. These self-feeding larvae have high dispersal potential, which—in concert with oceanographic and climatic conditions—enabled long-distance dispersal and interbasinal species migrations.

Introduction

The Richmondian Invasion, a regional invasion event that introduced over 60 genera from five phyla into the Cincinnati, Ohio, USA region, occurred during the early

Richmondian (late Katian) Age. Several competing hypotheses exist regarding the geographic source of the invaders. The ―Arctic‖ hypothesis (e.g. Foerste, 1912; Holland,

1997) postulated that invasive taxa originated in equatorial regions within Laurentia, in what today is Canada and Wyoming. Contrastingly, Jin (2001) proposed that at least some invaders originated in cool-water marginal basins around Laurentia and immigrated into the Cincinnati region as part of the broader Hiscobeccus expansion. Wright and

Stigall‘s (2013) phylogenetic analyses of brachiopods that participated in the

Richmondian Invasion identified multiple geographic source regions within Laurentia, including both a marginal basin and western epicontinental basin. Bauer and Stigall

(2014) reported support for the Artic pathway as well as the pathways of Wright and

Stigall (2014). Other authors (e.g. Anstey, 1986; Webby et al., 2004; Congreve &

Lieberman, 2010) have supported dispersal between Laurentia and Baltica or proposed

Baltica as a source area for some Richmondian taxa (e.g. Leptaena, Streptelasma). 19

Many of these proposed migration pathways have received little explicit testing.

Thus, in this contribution, we use the historical biogeographic method of Parsimony

Analysis of Endemicity (PAE) to examine patterns of endemism and dispersal within

Laurentia and between Laurentia and Baltica before and during the Richmondian

Invasion. Recovered dispersal pathways and inferred larval ecology of Richmondian invaders are compared to paleoclimate and paleoceanographic conditions within

Laurentia to determine dispersal mechanisms and test feasibility of hypothesized dispersal routes.

Methods

Parsimony Analysis of Endemicity

Biogeographic patterns and dispersal pathways were analyzed using PAE, a quantitative biogeographic method used for reconstructing hierarchical area relationships in the absence of phylogenetic information (Rosen & Smith, 1988). Presence/absence matrices for 63 genera that participated in the Richmondian Invasion were constructed for five intervals within nine geographic areas (Fig. 1, see Supplementary Materials archive: http://dx.doi.org/10.6084/m9.figshare.1194963). Occurrence data were obtained from online catalogs of natural history museums, online public databases (e.g. Paleobiology

Database: www.fossilworks.org), and published literature. Formations within each geographic area were correlated to the C1-C5 depositional sequences of Holland & 20

Patzkowsky (1996) using published chemostratigraphic and biostratigraphic data (e.g.

Young et al., 2008; Bergström et al., 2001; Bergström et al., 2015) and adjusted in

Figure 1. Paleogeographic map of Laurentia and Baltica during the Late Ordovician with inferred dispersal pathways and major paleoclimate and oceanographic influences on larval dispersal labeled. Biogeographic areas analyzed are: 1, Baltic Basin; 2, Scoto- Appalachian Basin; 3, Anticosti Island; 4, Appalachian Basin; 5, Cincinnati Basin; 6, Nashville Dome; 7, Upper Mississippi Valley; 8, Western Midcontinent (Texas and Oklahoma Basins); 9, North of the Transcontinental Arch (Hudson, Williston, and Bighorn Basins). Dark grey areas represent individual basins, light grey areas represent grouped basins. Paleoceanography after Cocks & Torsvik (2001); current directions and storm tracks after Poussart et al. (1999), Hermann et al. (2004), and Ettensohn (2010). 21 consult with Dr. Stig Bergström (pers comm. 2014). PAE areagrams were generated for each depositional sequence (C1-C5) to determine changes in area associations between depositional sequences using the parsimony tool in PAST 2.17c (Hammer et al., 2001).

Strict consensus trees were computed for analysis. Raup-Crick similarity indices were also calculated for basins in each depositional sequence (see Supplementary Materials archive: http://dx.doi.org/10.6084/m9.figshare.1194963).

Analysis of Larval Type, Paleoceanography, and Paleoclimate

The larval type and time spent in the larval phase was characterized for invasive genera using published data (e.g. Valentine & Jablonski, 1983; Chatterton & Speyer,

1989). Due to poor preservation in Late Ordovician strata, larval type was indeterminable for most molluscan taxa. Controversy exists in the literature over rhynchonelliform brachiopod larval type, and when planktotrophy evolved within the clade (e.g., Freeman & Lundelius, 2005; Peterson, 2005; Popov et al., 2011). This study follows the views of Freeman and Lundelius (2005) in which they concluded that planktotrophy evolved within the Rhynchonellata by the Early Ordovician.

Paleoceanographic data for the Late Ordovician Iapetus Ocean were obtained from published ocean modeling studies by Poussart et al. (1999) and Herrmann et al. (2004).

Paleoclimate data were interpreted from the literature (e.g. Holland & Patzkowsky, 1996;

Jin et al., 2013).

Results and Interpretation

Larval Type

Richmondian invaders with identifiable larval type are predominantly planktotrophic (self-feeding) planktic larvae (67.8%). Genera included in this group are members of Crinoidea, Strophomenata, and Rhynchonellata. Planktic planula (self- feeding) larvae, representing the class Anthozoa, are also common (28.6%).

Lecithotrophic benthic larvae, which gain energy exclusively from yolk sacs during a relatively brief larval stage, comprise only 3.4% of invasive genera. Thus, most

Richmondian invaders had long-lived larval phases with the potential for long distance transport.

Paleobiogeography

Within PAE analyses, dispersal events are identified by changes in biogeographic area relationships on subsequent areagrams. Areagrams resulting from this analysis (Fig.

2), include many such occurrences between multiple and shifting regions. Between the

C1 and C2 sequences, there is evidence for dispersal among the Western Midcontinent region, north of the Transcontinental Arch, and the Baltic Basin, as well as between the

Appalachian Basin and the Upper Mississippi Valley (Fig. 2). Similarity indices for the

C2 sequence support close area relationships between the Western Midcontinent and 23 north of the Transcontinental Arch (RC index=0.93) but lower overall similarity between these areas and the Baltic Basin (0.30 and 0.63, respectively). The overall similarity between the Appalachian Basin and Upper Mississippi Valley is 0.55. Thus, we interpret that dispersal occurred primarily between the southern Midcontinent and north of the

Transcontinental Arch with some influence from the Baltic Basin during the C1 and C2 sequences.

Figure 2. Parsimony Analysis of Endemicity areagrams. Areas group together in sister- area relationships based on the number of shared endemic taxa; the nested patterns present indicate hierarchical relationships of shared endemic taxa, which is a different metric than overall similarity (similarity values in supplementary materials archive: http://dx.doi.org/10.6084/m9.figshare.1194963). Shifts in area relationships between time slices indicate dispersal between those basins.

Between the C2 and C3 sequences there is evidence for dispersal between the

Upper Mississippi Valley and the Baltic Basin and between the Cincinnati Basin and the

Nashville Dome (Fig. 2). General similarity between the Upper Mississippi Valley and the Baltic Basin, however, is low (0.03) for the C3 sequence, indicating extremely limited faunal interchange. High general similarity (0.81) between the Cincinnati Basin and

Nashville Dome indicates substantial faunal mixing, which supports the interpretation of an active dispersal pathway during the C3 depositional sequence.

Between the C3 and C4 sequences, there is evidence for dispersal between the

Cincinnati and Scoto-Appalachian Basins, and among the Upper Mississippi Valley, north of the Transcontinental Arch, Nashville Dome, Western Midcontinent region, and the Baltic Basin (Fig. 2). Overall similarity is high for the Nashville Dome and Western

Midcontinent basins (0.99), and the Baltic Basin and Anticosti Island (0.87), which supports faunal connections between each pair of areas. The lower similarity (0.41) between the Cincinnati and Scoto-Appalachian Basins indicates selectivity along this dispersal route.

Between the C4 and C5 sequences, PAE results indicate dispersal among the

Cincinnati Basin, the Upper Mississippi Valley, and areas north of the Transcontinental

Arch, and between Scoto-Appalachia and the Baltic Basin. Overall similarity indices indicate substantial faunal exchange between the Baltic Basin and Scoto-Appalachia

(0.96) and between the Upper Mississippi Valley and north of the Transcontinental Arch

(0.90); whereas overall similarity between the Cincinnati Basin and north of the

Transcontinental Arch was lower (0.40). Combined, these results suggest that the Upper 25

Mississippi Valley region could have operated as an intermediate in the dispersal pathway between areas north of the Transcontinental Arch and the Cincinnati Basin.

Paleoceanography and Paleoclimate

Ocean circulation models for the Late Ordovician (e.g. Poussart et al., 1999;

Hermann et al., 2004) indicate the equatorial Iapetus Current flowed around Laurentia from the east, splitting around the paleocontinent to the north and south (Fig. 1). Trans-

Iapetus dispersal was likely accomplished by ―island hopping‖ among volcanic arcs surrounding Laurentia, as the majority of Richmondian invaders had planktotrophic planktic larvae with the potential to disperse over 100 km during one generation (Treml et al., 2008).

Intracontinental currents likely facilitated dispersal among Laurentian basins. The recovered dispersal pathways between the Cincinnati Basin and surrounding areas suggest that earlier (C1 through early C2 sequence) species migrations were likely influenced by the funneling of oceanic waters into Laurentia by the Sebree Trough

(Ettensohn, 2010). However, subsequent weathering and erosion of the Taconic highlands produced heavy sedimentation and the Sebree Trough was infilled by the beginning of the C2 sequence, restricting contact with the open ocean (Ettensohn, 2010).

A surface current, developed from the interaction of prevailing trade winds and the

Coriolis Effect, has been proposed to have flowed from the northeast to southwest across

Laurentia (Fig. 1; Ettensohn, 2010). Dispersal from Scoto-Appalachia and Anticosti 26

Island to the Cincinnati Basin could have been facilitated by this southwesterly oriented current.

Furthermore, cyclonic storms swept across Laurentia from the northeast during the Late Ordovician (Fig. 1). Tempestite beds in the Cincinnati region indicate the area was subjected to strong seasonal storms (Holland & Patzkowsky, 1996). Modern studies of coral larval transport have demonstrated that similar storm activity can greatly increase dispersal distance (Radford et al., 2014). Therefore, storms tracking from the northeast could have transported organisms across geographic barriers that would not normally be breached, aiding in the dispersal of taxa into surrounding basins from the Upper

Mississippi Valley and north of the Transcontinental Arch.

In addition, a substantial transgression occurred at the onset of the C5 sequence, which has been linked to global warming from the short-lived Boda Event (Fortey &

Cocks, 2005). This transgression could have flooded intracratonic arches and facilitated dispersal between geographically separated areas (Wright & Stigall, 2013), such as the basins in Cincinnati, the Upper Mississippi Valley, and north of the Transcontinental

Arch regions.

Discussion

Results of PAE support multiple dispersal pathways per timeslice and similarity analyses indicate the degree of faunal exchange along these connections. These data support three separate dispersal events between Laurentia and Baltica. Within Laurentia, 27 the basins of the Western Midcontinent and north of the Transcontinental Arch exhibit close area relationships and high similarity in all intervals, which indicates substantial dispersal between these areas throughout the study interval. For the Cincinnati Basin, the primary dispersal routes involved the Nashville Dome (C2-C3 sequences), Scoto-

Appalachian Basin (C3-C4 sequences), and the Upper Mississippi Valley and north of the

Transcontinental Arch (C4-C5 sequences).

The recovered pattern supports the multiple pathway hypothesis for the

Richmondian Invasion (cf. Wright & Stigall, 2013) and supports the idea that separate pulses of the Richmondian Invasion had distinct source areas. Specifically, PAE results indicate a clockwise shift in migration routes from the C1 to C5 depositional sequences

(Fig. 1). The primary invasion pulses occurred in the C4 and C5 sequences. The C4 sequence invasion sourced taxa from peripheral basins east of the Cincinnati region, such as the Scoto-Appalachian Basin and Anticosti Island, which is consistent with the dispersal path previously reconstructed for the brachiopod Plaesiomys (Wright & Stigall,

2013). The second phase of the Richmondian Invasion, in which over 50 new genera appeared in the Cincinnati Basin, occurred in the early C5 sequence (Holland, 1997).

This phase involved taxa immigrating from basins northwest of the Cincinnati Region

(Fig. 2), which accords with dispersal paths reconstructed for brachiopods Eochonetes and Glyptorthis and solitary corals (Elias 1983; Bauer & Stigall, 2014; Wright & Stigall,

2014). This shift in continental dispersal pathways coincides with a large transgression and a shift from temperate-style to tropical-style carbonate deposition throughout

Laurentia, thought to be the result of global warming from the short-lived Boda Event 28

(Fortey & Cocks, 2005), or termination of cool-water upwelling due to basinal infilling

(Holland & Patzkowsky, 1996).

CHAPTER 3: PALEOBIOGEOGRAPHIC ANALYSES OF MIDDLE TO LATE

ORDOVICIAN TAXA AND COMPARISON OF ANALYSES USING PARSIMONY

AND MAXIMUM-LIKELIHOOD METHODS

Introduction

Unraveling the patterns and processes that control the co-evolution of Earth and life has been an active area of fundamental research since the 1700‘s. Tremendous methodological developments over the past few decades have provided rigorous analytical techniques for examining evolutionary changes within a biogeographic context.

Specifically, phylogenetic paleobiogeography has been used to interpret dispersal pathways, discern changes in area relationships, and identify areas of origination for invasive taxa (ex., Lieberman & Eldredge, 1996; Sereno et al., 1996; Krause et al., 1997;

Lieberman, 2003a; Hembree, 2006; Folinsbee & Brooks, 2007; Maguire & Stigall, 2008;

Wright & Stigall, 2013). Phylogenetic paleobiogeography is a powerful tool and utilizes several methods, including both parsimony and model-based approaches, which have elucidated biogeographic patterns of both fossil and extant taxa.

The focus of this study is to examine biogeographic patterns of Middle to Late

Ordovician shallow marine taxa within the epicontinental seas of Laurentia, Baltica, and the intervening Iapetus Ocean. The Middle through Late Ordovician was a time of high diversification rates associated with the Great Ordovician Biodiversification Event, as well as a time of high sea level, increased volcanic activity, and widespread carbonate platforms (Algeo & Seslavinsky, 1995; Miller & Mao, 1995; Servais et al., 2010). 30

Therefore, this is an ideal interval in Earth history to examine dispersal routes among paleocontinents and effects of tectonic and paleoclimatic processes on speciation patterns. To reconstruct dispersal patterns, species-level phylogenetic hypotheses of brachiopod and trilobite clades were employed in a time-stratigraphic framework to examine speciation processes across four temporal intervals during the Middle to Late

Ordovician, each characterized by a different combination of tectonic and paleoclimatic conditions.

Traditionally, most phylogenetically informed biogeographic analyses have utilized parsimony-based approaches (e.g., Ladiges et al., 1987; Stigall Rode &

Lieberman, 2005; Wojcicki & Brooks, 2005; Escalante et al., 2007). Such approaches have produced novel and insightful results, but typically require assumptions of parsimony and congruence of biogeographic patterns among clades, which may not be valid in all circumstances. Recent advances have expanded the available methods to include maximum likelihood and Bayesian modeling approaches (e.g., Sanmartin et al.,

2001; Costa, 2010; Litsios et al., 2014; Sorenson et al., 2014; Wood et al., 2014). These methods provide a fuller exploration of probabilistic processes and idiosyncratic patterns.

The full suite of phylogenetic biogeographic methods can be utilized with extant taxa; however, the requirement of contemporaneous terminal taxa has curtailed the applicability of model-based methods (ex., LAGRANGE) with fossil data. A newly developed method, Bayesian (and likelihood) Evolutionary Analysis in R Scripts

(BioGeoBEARS) (Matzke, 2013a; R core team, 2014), removes that limitation, and is thus fully compatible with fossil data (Matzke, 2013b). Although parsimony and 31 maximum likelihood analyses differ considerably in their underlying assumptions and methodologies, both have proven to be powerful tools to discern area histories within a phylogenetic context (Lieberman, 2003b; Matzke, 2013b). Thus, both parsimony and maximum likelihood analyses are applied in this study to analyze a large data set of

Middle through Late Ordovician taxa to reconstruct biogeographic relationships and analyze the relative efficacy of these approaches.

This study is the first to use both parsimony (Fitch optimization and Lieberman- modified Brooks Parsimony analysis) and maximum likelihood methods implemented within BioGeoBEARS to examine evolutionary biogeographic patterns in fossil taxa.

Specifically, ten published species-level phylogenetic hypotheses of Middle through Late

Ordovician brachiopods and trilobites are analyzed within a comparative approach to identify vicariance patterns and dispersal pathways that facilitated evolution within these clades. The results are interpreted in a paleogeographic and paleoceanographic context through the Middle and Late Ordovician epochs in order to determine 1) similarities among dispersal pathways, and 2) the most probable reconstructed dispersal patterns when compared to paleoceanographic reconstructions for the Late Ordovician. The results of both methods will be compared across phylogenies in four time slices in order to determine the efficacy of maximum likelihood analyses and recovery of paleogeographic signals and relationships between areas.

Geologic Setting

Biogeographic patterns of brachiopods and trilobites were influenced by tectonic and paleoclimatic events that occurred during the Middle through Late Ordovician Period within Laurentia, Baltica, Avalonia, and the Intra-Iapetus region. This interval of Earth history is associated with increased tectonic activity and inferred accelerated rates of sea floor spreading (Servais et al., 2010). The subduction of the Iapetus Ocean between

Laurentia and Baltica and the Tornquist Ocean between Avalonia and Baltica produced volcanic island arc chains and led to the eventual collision of Baltica, Avalonia, and

Laurentia in the latest Late Ordovician to mid Silurian (Cocks & Torsvik, 2011).

Substantial volcanic ash produced from subduction zone volcanism produced a series of widespread Middle and Late Ordovician K-bentonite deposits in North America,

Europe, South America, and China (Huff et al., 1992; 2010). Due to inferred accelerated seafloor spreading rates and high global temperatures, the Middle through Late

Ordovician interval was characterized by high sea level, possibly the highest of the

Paleozoic (see Servais et al., 2010 and references therein). High sea level combined with widespread dispersal of the continents resulted in extensive sedimentary deposition on tropical shelves and intercratonic flooding (Fig. 3) (Algeo & Seslavinsky, 1995; Walker et al., 2002).

Figure 3. Paleogeographic reconstruction of the Late Ordovician Sandbian Stage (460 Ma) with oceans and approximate location of subduction zones marked. Numbers indicate geographic areas used in this study: 1, Northern Laurentia (Northwest Territories, Nunavut, British Columbia); 2, North of the Transcontinental Arch; 3, Western Midcontinent; 4, Upper Mississippi Valley; 5, Cincinnati Basin; 6, Southern Appalachian Basin; 7, Northern Appalachian Basin; 8, Southern Laurentia (Newfoundland, Quebec, Anticosti Island); 9, Scoto-Appalachia; 10, Avalonia; 11, Baltica; 12, Tarim Plate; 13, Northern Gondwana (Australia, Kazakhstan, Japan); 14, Southern Gondwana (Arabia, Turkey, Bohemia, Spain, France). Colors of areas match those used in Figure 5. Box indicates the area of focus in Figure 8. Map modified from Torsvik and Cocks (2013).

The Middle Ordovician Epoch (Dapingian-Darriwilian Ages) included the Great

Ordovician Biodiversity Event (GOBE), during which marine invertebrate biodiversity increased dramatically within a 25 million year timespan (Servais et al., 2010). Recent studies have suggested that homogenization of faunas among paleocontinents began to take place during this time (e.g., Harper et al., 2013) as inter-continental dispersal was facilitated by Baltica moving closer to Laurentia from the southeast and Avalonia shifting to the northeast by the closure of the Tornquist Ocean. Within Laurentia, the Blountian tectophase of the Taconic orogeny began along the southeastern margin of the 34 paleocontinent (Fig. 4) (Ettensohn, 2010). Island arcs and/or crustal fragments collided with Laurentia, forming the Sevier Basin, which stretched from the Alabama to Virginia promontories (Shanmugam & Lash, 1982; Ettensohn, 2010). Sedimentation within the epicontinental basins was mainly unaffected as tropical carbonate deposition prevailed across the paleocontinent (Holland & Patzkowsky, 1997). Additionally, faunas within

Laurentia were both relatively stable and endemic during the Middle Ordovician Epoch

(Harper et al., 2013).

Figure 4. Chronostratigraphic chart of the Middle to Late Ordovician interval, with global stages, North American Series, Stages, and depositional sequences for the Cincinnati Basin as defined by Holland and Patzkowsky (1996) marked. Time slices used in this study (T0-T3) are indicated above the depositional sequences. Taconic orogeny tectophases indicate activity that occurred on the southern margin of Laurentia, dominant carbonate type within southern Laurentian basins, and important paleoclimatic events and carbon isotope excursions (e.g., GICE, HICE, Boda Event). Abbreviations: GICE, Guttenburg δ13C excursion; HICE, Hirnantian δ13C excursion; G, Gamachian; Ki, Kircklandian; Ro, Rocklandian; H, Hirnantian. Timescale modified from Holland and Patzkowsky (1996); Taconic tectophases after Ettensohn (2010); occurrence of GICE after Young et al. (2005); occurrence of Boda Event after Fortey and Cocks (2005); occurrence of HICE after Bergström et al. (2006).

The Sebree Trough, a structural depression that stretched for hundreds of kilometers across the eastern midcontinent of Laurentia developed from the collapse of the Reelfoot Reef, funneled cool nutrient-rich waters into the craton from the Iapetus

Ocean to the south (Kolata et al., 2001). This initiated epicontinental estuarine-like circulation patterns in the midcontinent of Laurentia (Wilde, 1991; Kolata et al., 2001).

Consequently, carbonate platform deposition ceased and a mixed carbonate-clastic facies prevailed which accumulated within temperate waters across most of east-central

Laurentia (Keith, 1989; Ettensohn, 2002). Carbonate deposition resumed within

Laurentia during the Katian Age due either to the effective infilling of the Sebree Trough, which reduced upwelling, or to the globally recognized warming trend known as the

Boda Event (Fig. 4) (Fortey & Cocks, 2005, but see Cherns & Wheeley, 2007).

Significant events of faunal immigration into epicratonic basins occurred during the Katian Age. The Hiscobeccus expansion involved the migration of three brachiopod genera (Hiscobeccus, Lepidocyclus, and Hypsiptycha) into the epicontinental seas of

Laurentia from marginal intercratonic basins during the early Katian Age (Jin, 1999).

Within the Cincinnati Basin, an influx of over 60 new extra-basinal genera occurred, known as the Richmondian Invasion, which resulted in significant ecosystem restructuring (Holland, 2008). The Richmondian Invasion involved the migration of taxa from basins within Laurentia and Baltica (Wright & Stigall, 2013; Bauer & Stigall, 2014;

Lam & Stigall, 2015) and may represent a regional manifestation of the Hiscobeccus expansion (Stigall, 2010a). 36

The Late Ordovician Hirnantian Age marks the beginning of a major glaciation that led to the first mass extinction event within the Paleozoic. Mounting evidence indicates that cooling associated with the Hirnantian Age glaciation may have begun in the Middle Ordovician (Saltzman & Young, 2005; Trotter et al., 2008; Vandenbroucke et al., 2010), and the biodiversity drop may be linked to episodes of deglaciation rather than a single glacial apex (see Ghienne et al., 2014 for full discussion).

Methods

Phylogenetic biogeographic analyses were conducted using ten published species- level phylogenetic hypotheses of trilobite and brachiopod clades (Table 1) to reconstruct dispersal routes and compare dispersal mechanisms during four time intervals in the

Middle through Late Ordovician. Two methods, a parsimony-based method and a maximum likelihood method, were used to analyze speciation patterns and reconstruct biogeographic relationships using the same input data. Both methods require temporal and geographic distribution data for taxa with known phylogenetic relationships.

Table 1. Focal taxa for biogeographic analyses. Clade size indicates the number of terminal taxa after species were combined or removed from the published phylogenetic hypothesis to resolve polytomies. Refer to Figure 4 for an explanation of ages used in this study.

Clade Clade size Stratigraphic range Reference

Eochonetes Brachiopoda 14 Katian-Llandovery Bauer & Stigall, 2014 Glyptorthis Brachiopoda 23 Darriwilian-Llandovery Wright & Stigall, 2013b Hebertella Brachiopoda 10 Sandbian-Llandovery Wright & Stigall, 2014 Plaesiomys Brachiopoda 10 Katian-Llandovery Wright & Stigall, 2014 Bumastoides Trilobita 6 Sandbian-Hirnantian Carlucci et al., 2012 Deiphoninae Trilobita 13 Dapingian-Ludlow Congreve & Lieberman, 2010 Flexicalymene Trilobita 10 Sandbian-Llandovery Hunda & Hughes, 2007 Homalonotidae Trilobita 14 Darriwilian-Wenlock Congreve & Lieberman, 2008 Tetralichinae Trilobita 10 Dapingian-Hirnantian Carlucci et al., 2010 Thaleops Trilobita 17 Darriwilian-Katian Amati & Westrop, 2004

Phylogenetic and Species Occurrence Data

This study examines biogeographic patterns within brachiopod and trilobite taxa because these organisms comprise a substantial and well-sampled component of the

Ordovician benthos and phylogenetic hypotheses were most readily available for members of these groups. In addition, members of these taxa utilized a variety of larval strategies, which permits comparison of dispersal patterns across differing larval phases.

Clades selected for analysis were primarily composed of Laurentian species and species that inhabited Baltica, Avalonia, or other Iapetus-adjacent regions in order to determine dispersal routes within Laurentia and among Laurentia, Baltica, and Avalonia. Some of the focal clades (e.g., Deiphoninae, Homalonotidae, Flexicalymene, and Tetralichinae) 38 also include species from Gondwana and the Tarim Plate, which are included in the analysis as they are valuable for determining large-scale oceanographic patterns.

Phylogenetic biogeographic methods require a single fully-resolved cladogram topology for each clade. If a single most parsimonious tree (MPT) was published by the original authors, that topology was used for analysis. However, if a single MPT or strict consensus tree was not published in the original paper, the published data matrix was subjected to an exhaustive search in PAUP* 4.0b (Swofford, 2002) and the strict consensus topology was computed from the recovered set of MPT. All polytomies on calculated consensus trees were resolved because the maximum likelihood models implemented in this study cannot accommodate non-bifurcating (zero branch length) tree topologies (Matzke, 2014). Resolution was accomplished by either combining several polytomous branches into a single branch, in which case the species names and all biogeographic areas occupied were retained but treated as a single taxon, or by pruning species from the tree that lacked sufficient character support to resolve evolutionary relationships.

Temporal and geographic occurrence data were compiled for each retained species using published literature and online diversity databases such as iDigBio

(www.idigbio.org), the Paleobiology Database (www.paleobiodb.org), and the Digital

Atlas of Ordovician Life (www.ordovicianatlas.org) (Appendix 1). Stratocladograms were generated by combining phylogenetic hypotheses with temporal range data.

A detailed correlation chart was developed in order to place each species occurrence within the correct time slice for analysis (Appendix 2). Sixty stratigraphic 39 columns of the geographic area and stratigraphic formations occupied by focal species were correlated across the Middle through Late Ordovician using published literature

(Appendix 3). Correlations were based primarily on conodont biostratigraphy and chemostratigraphy when such studies were available (ex., Young et al., 2008; Bergström et al. 2010). Correlations for formations and basins within Laurentia and Baltica (e.g.,

Williston Basin, Hudson Bay Basin, and the Baltic Basin) were adjusted in consult with

Dr. Stig Bergström (pers. comm. 2014).

Within the phylogenetic hypothesis, equal split from the ancestor was assumed and ghost lineages were added to account for gaps in stratigraphic lineages following the protocol of Smith (1994) (Appendix 4). Although probabilistically-based time scaling programs for phylogenies do exist (e.g., paleotree, Bapst 2013), the use of such programs on large datasets, such as the one used here, has yet to be investigated and is outside the scope of this study.

Temporal and Geographic Framework

Four time slices were established to characterize speciation modes across major tectonic and paleoclimatic events (Fig. 4). The T0 time slice encompasses the Dapingian and the Darriwilian Ages, a time of increased global tectonic activity and major diversification associated with the Great Ordovician Biodiversification Event (Miller &

Mao, 1995). The T1 time slice spans from the start of the Sandbian Age to the M4/M5 sequence boundary (late Sandbian) and includes the Blountian tectophase of the Taconic 40 orogeny (Ettensohn, 1994). The T2 time slice includes the Taconic tectophase of the

Taconic orogeny; it begins at the M4/M5 sequence boundary in the late Sandbian Age and ends at the C3/C4 sequence boundary (mid late Katian Stage) (Ettensohn, 1994). The

T3 time slice includes the time from the C3/C4 sequence boundary to the beginning of the Hirnantian Age. This interval is associated with the globally recognized Boda Event

(Fortey & Cocks, 2005).

Because the dominant speciation processes have been previously demonstrated to vary during times of major tectonic and paleoclimatic changes (e.g., Stigall, 2010a;

Wright & Stigall, 2013a), time slice boundaries were defined to coincide with major sedimentological changes within Laurentia, indicative of changing climatic and tectonic conditions. Stratocladograms were divided by the time slices described above to characterize speciation processes occurring within clades during times of tectonic and/or climatic shifts. Speciation events optimized at the boundary between two time slices were counted as occurring within the later time interval as that interval was more likely to include documented occurrences for the species.

Fourteen biogeographic regions were defined for analysis (Fig. 3). Nine geographic areas were defined within Laurentia based upon separation of basins by physical and thermal barriers (ex., the Transcontinental Arch and the deep-water Sebree

Trough, respectively). Gondwana was separated into northern and southern regions based upon latitudinal position of areas. On a continental scale, the defined areas follow the provinces of Harper et al. (2013) for the Middle to Late Ordovician using rhynchonelliform brachiopods and the provinces defined by Meidla et al. (2013) for Late 41

Ordovician ostracods. Within Laurentia, the defined areas in this study follow those of sponge provinces identified by Carrera and Rigby (1999), as well as distinctive areas harboring faunas of conodonts, corals, brachiopods, and ostracod within the continental interior and marginal basins (Amsden, 1974; Thompson & Satterfield, 1975; Elias, 1983;

Amsden, 1986; Barrick, 1986; Elias and Young, 1992; Mohibullah et al., 2012). Basins of eastern North America follow the aquafacies of Holmden et al. (1998) based on conodont associations for the Late Ordovician of Laurentia.

Parsimony-based Phylogenetic Biogeographic Analyses

Parsimony-based analysis was used to identify biogeographic evolution of species and biogeographic relationships between areas using a modified Fitch Parsimony (FP) algorithm and Lieberman-modified Brooks Parsimony Analysis (LBPA) as described in detail in Fitch (1971), Wiley and Lieberman (2011) and Lieberman (2000). These methods have previously been used to discern dispersal pathways of invasive organisms and interpret the influence of earth system interactions upon speciation processes in the

Late Ordovician (ex., Wright & Stigall, 2013a; Bauer & Stigall, 2014) and other intervals in Earth history (ex., Lieberman, 2003c; Hendricks & Lieberman, 2007; Maguire &

Stigall, 2008).

Fitch Parsimony analysis was first performed to reconstruct the biogeographic state of the ancestral nodes within the phylogenetic hypothesis. To start, phylogenetic hypotheses were converted to taxon-area cladograms by replacing species names at the 42 terminals with numbers coding for species‘ geographic occurrences. Optimization of ancestral biogeographic areas at internal nodes within each phylogeny was performed using the modified version of Fitch Parsimony detailed in Lieberman (2000). Fitch

Parsimony is an unordered optimization algorithm under which characters (areas) are allowed to evolve and reverse along branches within a phylogeny. This method provides a framework under which to identify speciation events characterized by biogeographic changes between internal nodes. Episodes of speciation by vicariance (separation of ancestral ranges) were identified when a daughter species occupied a subset of the geographic region occupied by its immediate ancestor. Episodes of speciation by dispersal were identified when daughter species occupied areas different from or additional to the range of the ancestor.

Next, the biogeographic distribution of the FP optimized ancestral nodes and the terminal taxa were coded into separate LBPA vicariance and geodispersal matrices and analyzed using maximum parsimony to determine area relationships and identify the relative order that barriers arose between areas (vicariance) and when they were removed

(geodispersal) (Wiley & Lieberman, 2011). Vicariance and geodispersal matrices of 127 species and 117 cladogenic events were coded from the optimized taxon-area cladograms following the protocol of Lieberman (2000). Each matrix was subjected to parsimony analysis using the Branch and Bound search method in PAUP* 4.0b (Appendix 5)

(Swofford, 2002). When multiple MPT were recovered, the strict consensus tree was computed (Appendix 6). Tree support was calculated via consistency and retention indices (CI and RI, respectively). Node support was constrained using bootstrap and 43 jackknife analyses. Bootstrap values were determined from 5% resampling with replacement using a full-Heuristic search with 10,000 replicates. Jackknife values were recovered using 10,000 replicates of a full-Heuristic search with 5% character deletion.

Bayesian Analyses

Biogeographic evolution within clades and geographic changes at speciation events were also analyzed via probabilistic inferences of ancestral ranges using the R package BioGeoBEARS (Matzke, 2013a; R core team, 2014). This method has been mainly used among modern biogeographers and has been implemented to discern dispersal patterns of extant amphibians, mammals, and birds (e.g., Buckner et al., 2014;

Harris et al., 2014; Pyron, 2014). Inferences of biogeographic histories within phylogenies are determined by model testing and model choice, as speciation processes and ancestral range reconstructions are allowed to evolve along branches within six discrete models implemented in the program. Input data includes information about absolute temporal length of branches on each phylogeny, the biogeographic area(s) occupied by each species, and a fully bifurcating phylogenic topology.

Within BioGeoBEARS, biogeographic range estimation evolves along branches in a phylogeny within a maximum likelihood framework under a variety of analytical processes to determine best-fit models (Matzke, 2014). The BioGeoBEARS program calculates biogeographic patterns using the Dispersal-extinction-cladogenesis (DEC) model (Ree, 2005), dispersal-vicariance analysis (DIVA) (Ronquist, 1997), and the 44

BayArea model (Landis et al., 2013). The BioGeoBEARS program provides explicit output that facilitates model testing and model choice for each phylogeny. Furthermore, founder-event speciation (+J) is incorporated alongside classic vicariance and dispersal speciation within each method. Founder-event speciation has long been considered a crucial speciation process among island clades (Cowie & Holland, 2006; Templeton,

2008; Gillespie et al., 2012), in which small populations of individuals become genetically and geographically separated by chance from a larger population to areas conducive to reproduction, which was supported by Matzke‘s (2014) analysis of island clades within BioGeoBEARS. The implementation of BioGeoBEARS in this study will provide a means to assess whether untested founder-event speciation was important for trans-Iapetus dispersal in the Ordovician Period.

The first step in analysis was to determine the temporal duration of branches within the input phylogenies. Branch lengths used were based on the reconstructed stratocladograms described above. Because branch lengths of 0 are not allowed as input parameters, the minimum duration assigned to a branch was 50 ky, as this is considered the upper limit at which speciation processes take place (Eldredge et al., 2005).

Furthermore, when the d and e parameters (dispersal and extinction rates, respectively) are small, branch lengths have less effect on the biogeographic range estimation within the models, indicating that speciation processes occur primarily at the cladogenic splits

(Matzke, 2014; Matzke, pers. comm. 2014). As the d and e parameters calculated within

BioGeoBEARS for the ten phylogenies used in this analysis were very small (Table 3), the absolute values of branch lengths are less important within this data set and thus 45 uncertainty in branch length estimation does not have a large effect on ancestral range estimation.

The analyses within BioGeoBEARS were conducted using unconstrained analyses, thus directionality and timing of speciation events were uninhibited. The number of areas (max_range_size) was constrained based on the maximum number of areas occupied by any one species per phylogeny to avoid inference of a widespread ancestor at the tree roots. The best-fit maximum likelihood (ML) model for each phylogeny was evaluated based on the Akaike Information Criterion (AIC) as calculated from the log likelihood value for each model (Appendix 8). Sample R-script is presented in Appendix 7.

Results

Ancestral Range Reconstructions

Ancestral range optimization using modified Fitch Parsimony was able to generate ancestral reconstructions for each internal node of the input cladograms (Fig.

5A, 5C). Of the 119 internal nodes comprising the ten phylogenetic hypotheses analyzed,

117 nodes could be characterized by speciation via vicariance or dispersal processes (Fig.

6A, 6C, Table 2). Optimization of the remaining two nodes was consistent with speciation within the same biogeographic area.

Table 2. Distribution of cladogenetic events (raw count with percent in parenthesis) attributable to vicariance or dispersal processes per time slice as interpreted from Fitch Parsimony and best-fit models from BioGeoBEARS analyses.

Timeslice Fitch Parsimony BioGeoBEARS Dispersal Vicariance Dispersal Vicariance T3 5 (31%) 11 (69%) 9 (75%) 3(25%) T2 9 (33%) 18 (67%) 16 (63%) 9 (37%) T1 12 (36%) 21 (64%) 19 (66%) 10 (34%) T0 28 (67%) 14 (33%) 37 (82%) 8 (18%)

For most nodes, the BioGeoBEARS optimization produced an ancestral range reconstruction consisting of one or only few areas that were optimized with high probability of occurrence, which is indicative of a clear biogeographic signal within the data (Fig. 5B, 5D) (see Appendix 8, probability of optimized ancestral states models).

Although the optimal model of ancestral range reconstruction for individual clades varied among DEC, DIVA, and BayArea, all ten phylogenetic hypotheses were most compatible with biogeographic evolution via the ―+j‖ models within BioGeoBEARS (Table 3). The j parameter measures the weight of founder-event speciation within each model (Matzke,

2014). Within all clades, the j parameter is small relative to the maximum attainable value of 3.0, which indicates that vicariance and traditional dispersal were also important processes operating within these clades. Of the 119 internal nodes comprising the ten phylogenetic hypotheses analyzed, 111 nodes could be characterized by speciation via vicariance or dispersal processes (Fig. 6B, 6D, Table 2).

Figure 5. Cladograms with area reconstructions within brachiopod and trilobite taxa using Fitch Parsimony and maximum likelihood models generated within the R package BioGeoBEARS. A) Brachiopod Fitch Parsimony, B) brachiopod maximum likelihood models, C) trilobite Fitch Parsimony, D) trilobite maximum likelihood models.

Figure 6. Cladograms depicting speciation processes among brachiopod and trilobite taxa using Fitch Parsimony and maximum likelihood models generated within the R package BioGeoBEARS. A) Brachiopod Fitch Parsimony, B) brachiopod maximum likelihood models, C) trilobite Fitch Parsimony, D) trilobite maximum likelihood models.

Optimization of the remaining 8 nodes was consistent with speciation within the same biogeographic area. In a few clades, notably Flexicalymene, Hebertella,

Homalonotidae, and Thaleops, the ΔAIC between two or more models was very small and the Akaike weights (ωi) indicate that the relative likelihoods of the best and next-best models differ by just 0.05. Where ωi values are close, the model outputs differed only slightly. Thus, the subsequent discussion incorporates patterns observed within best-fit models with the highest likelihood values.

Speciation Mode Analyses

The number of speciation events attributable to vicariance versus dispersal processes differs considerably among time slices (Table 2, Figs. 5, 6). Under Fitch

Parsimony, the proportion of dispersal to vicariance events exhibits statistically significant differences among time slices (Χ2 test, p<0.01), whereas the proportion of dispersal to vicariance events does not vary significantly among intervals within

BioGeoBEARS optimization (Χ2 test, p=0.28). Notably, the number of vicariance events recovered per time slice by FP and ML methods are statistically different (Mann-Whitney test, p=0.03). However, the number of speciation by dispersal events per time slice recovered by the two methods does not differ statistically (Mann-Whitney test, p=0.38).

Both trilobite and brachiopod clades exhibit similar speciation patterns, and no statistical difference was recovered in either the number of dispersal or vicariance events reconstructed by time interval when 52

Table 3. Comparisons of the fit of DEC, DEC+J, DIVA, DIVA+J, BayArea, and BayArea+J ancestral range estimations each phylogeny from the BioGeoBEARS analyses. The log-likelihood values (lnL) from the analyses are given, as well as the Akaike Information Criterion (AIC). ΔAIC shows the difference in AIC values compared with the best-fit model. Akaike weight (ωi) values give the relative likelihood of each model. The d and e parameters are an estimate of dispersal and extinction rate, respectively, as measured along branches within each phylogeny. The j parameter is a measure of the relative weight of founder-event speciation. Best-fit models for each phylogeny are indicated in bold.

Model lnL AIC ΔAIC ωi d e j Bumastoides DEC -16.80 37.61 7.35 0.01 0.015 9.53E-03 0 DEC+J -12.13 30.26 0.00 0.56 0.006 1.00E-12 1.59

DIVALIKE -17.50 38.99 8.74 0.01 0.033 5.89E-02 0

DIVALIKE+J -13.40 32.80 2.54 0.16 0.013 1.00E-12 0.53

BAYAREALIKE -19.61 43.21 12.95 0.00 0.042 1.52E-01 0

BAYAREALIKE+J -12.88 31.76 1.50 0.26 0.008 1.00E-07 0.78

Deiphoninae DEC -59.88 123.77 35.96 0.00 0.022 4.91E-02 0 DEC+J -43.17 92.35 4.54 0.09 0.002 1.00E-12 0.23

DIVALIKE -60.41 124.81 37.01 0.00 0.022 4.70E-02 0

DIVALIKE+J -43.36 92.72 4.91 0.07 0.002 1.00E-12 0.29

BAYAREALIKE -62.60 129.20 41.39 0.00 0.030 8.26E-02 0

BAYAREALIKE+J -40.90 87.81 0.00 0.84 0.001 1.00E-07 0.15

Eochonetes DEC -44.86 93.71 29.10 0.00 0.077 1.36E-01 0 DEC+J -30.53 67.06 2.45 0.19 0.002 1.35E-02 0.21

DIVALIKE -44.51 93.02 28.40 0.00 0.076 1.82E-01 0

DIVALIKE+J -29.31 64.62 0.00 0.66 0.003 1.00E-12 0.19

BAYAREALIKE -45.83 95.66 31.04 0.00 0.095 2.94E-01 0

BAYAREALIKE+J -30.78 67.56 2.94 0.15 0.003 1.00E-07 0.20

Flexicalymene DEC -31.36 66.72 3.77 0.06 0.019 2.56E-02 0 DEC+J -29.28 64.57 1.62 0.18 0.012 1.00E-12 0.050

DIVALIKE -29.60 63.20 0.25 0.35 0.019 2.89E-02 0

DIVALIKE+J -28.47 62.95 0.00 0.40 0.014 4.76E-03 0.030

BAYAREALIKE -39.65 83.29 20.35 0.00 0.026 1.48E-01 0

BAYAREALIKE+J -31.87 69.74 6.79 0.01 0.013 1.00E-07 0.078

Table 3 (Continued) Glyptorthis DEC -72.23 148.46 25.25 0.00 0.017 5.19E-02 0 DEC+J -58.61 123.21 0.00 0.88 0.005 1.00E-12 0.14

DIVALIKE -75.14 154.28 31.07 0.00 0.024 6.37E-02 0

DIVALIKE+J -60.62 127.24 4.02 0.12 0.006 1.00E-12 0.14

BAYAREALIKE -70.75 145.50 22.29 0.00 0.017 1.35E-01 0

BAYAREALIKE+J -65.46 136.91 13.70 0.00 0.003 4.63E-02 0.064

Hebertella DEC -27.36 58.71 18.40 0.00 0.044 5.08E-02 0 DEC+J -17.15 40.31 0.00 0.53 0.003 1.00E-12 0.25

DIVALIKE -26.86 57.73 17.42 0.00 0.043 3.87E-02 0

DIVALIKE+J -17.62 41.23 0.93 0.33 0.004 1.00E-12 0.20

BAYAREALIKE -28.39 60.78 20.47 0.00 0.054 1.06E-01 0

BAYAREALIKE+J -18.45 42.91 2.60 0.14 0.003 1.00E-07 0.23

Homalonotidae DEC -31.35 66.70 4.06 0.06 0.014 2.47E-02 0 DEC+J -28.32 62.64 0.00 0.43 0.004 1.00E-12 0.061

DIVALIKE -33.54 71.08 8.44 0.01 0.016 2.87E-02 0

DIVALIKE+J -28.56 63.12 0.48 0.34 0.005 1.00E-12 0.059

BAYAREALIKE -34.57 73.14 10.49 0.00 0.016 3.66E-02 0

BAYAREALIKE+J -29.28 64.56 1.92 0.16 0.004 1.00E-07 0.061

Plaesiomys DEC -32.02 68.04 8.31 0.01 0.023 4.94E-02 0 DEC+J -27.64 61.28 1.55 0.27 0.013 1.00E-12 0.11

DIVALIKE -30.33 64.65 4.93 0.05 0.022 3.87E-02 0

DIVALIKE+J -26.86 59.73 0.00 0.59 0.015 1.00E-12 0.056

BAYAREALIKE -33.37 70.75 11.02 0.00 0.026 2.14E-01 0

BAYAREALIKE+J -28.86 63.72 3.99 0.08 0.010 1.00E-07 0.29

Tetralichinae DEC -32.71 69.42 24.10 0.00 0.026 6.13E-02 0 DEC+J -20.72 47.45 2.13 0.19 0.002 1.00E-12 0.21

DIVALIKE -33.37 70.75 25.42 0.00 0.037 7.59E-02 0

DIVALIKE+J -20.32 46.65 1.33 0.28 0.001 1.00E-12 0.26

BAYAREALIKE -32.71 69.42 24.09 0.00 0.042 1.08E-01 0

BAYAREALIKE+J -19.66 45.32 0.00 0.54 0.000 1.00E-07 0.31

Table 3 (Continued) Thaleops DEC -54.76 113.51 22.88 0.00 0.060 9.44E-02 0 DEC+J -42.69 91.39 0.76 0.37 0.008 1.00E-12 0.28

DIVALIKE -53.67 111.33 20.70 0.00 0.060 8.76E-02 0

DIVALIKE+J -42.31 90.63 0.00 0.54 0.009 1.00E-12 0.26

BAYAREALIKE -59.49 122.98 32.35 0.00 0.108 2.02E-01 0

brachiopod versus trilobite clades were compared (Mann-Whitney test, p=0.13 to 0.19 within maximum likelihood models; p=0.77 for Fitch optimization). Similarly, the total number of vicariance and dispersal events for all time slices did not differ between brachiopod and trilobite clades, regardless of optimization method (Mann-Whitney test, p=0.15 to 0.46 for vicariance events; p=0.37 to 0.56 for dispersal events).

Biogeographic Relationships among Areas

LBPA analyses of vicariance and geodispersal patterns produced mostly congruent patterns of area relationships throughout the T0 to T2 time slices, as is indicated by similarity in the resolved portion of the LBPA strict consensus area cladograms (Fig. 7). Within the T0 time slice, two most parsimonious vicariance area cladograms and three geodispersal cladograms were recovered. Analysis of biogeographic data for the T1 time slice recovered 40 equally parsimonious vicariance and 36 geodispersal topologies. The vicariance analysis for the T2 time slice recovered

84 most parsimonious trees, whereas the geodispersal analysis produced 24 most parsimonious trees. Finally, three most-parsimonious vicariance and ten geodispersal topologies were recovered for the T3 time slice. The strict consensus trees produced from 55 the vicariance and geodispersal analyses of the T0, T1, and T2 time slices were largely congruent, but the T3 vicariance and geodispersal consensus cladograms exhibit dissimilarities that indicate different processes operated among areas during the late

Katian Age. Recovered biogeographic patterns are well supported; the most parsimonious trees of each interval exhibit strong support in terms of CI and RI values, and the nodes resolved on the strict consensus trees have high bootstrap and jackknife values (Fig. 7).

Area Relationships and Dispersal Patterns

T0 Time Slice

Parsimony Analyses

The T0 time slice exhibited a greater percentage of speciation by dispersal compared to the T1-T3 time slices as interpreted from Fitch optimization (Table 2).

Clades that exhibit dispersal patterns within this time slice are Glyptorthis, Thaleops,

Deiphoninae, Tetralichinae, and Homalonotidae (Fig. 6A, 6C). Glyptorthis dispersed from the Western Midcontinent to the east into the Cincinnati Basin and Southern and

Northern Appalachian basins (Fig. 5A). Thaleops exhibits dispersal patterns from the

Western Midcontinent and Southern Laurentia into areas North of the Transcontinental

Arch (Fig. 5C). In addition, Thaleops dispersed from the Northern Appalachian Basin into the Southern Appalachian Basin and Southern Laurentia. Deiphoninae dispersal patterns indicate the clade originated in Baltica and dispersed into Laurentian basins

(Northern Laurentia, Northern and Southern Appalachian basins, and the Upper 56

Mississippi Valley) as well as the Tarim Plate and Northern Gondwana. Tetralichinae dispersed from Northern Gondwana, the Upper Mississippi Valley, and North of the

Transcontinental Arch into the Southern Appalachian Basin, Western Midcontinent, and

Baltica, with a subsequent dispersal event within Laurentia from the Upper Mississippi

Valley into the Southern Appalachian Basin. Homalonotid dispersal patterns indicate the clade originated in Southern Gondwana and dispersed to the northeast and southwest with species occupying Northern Gondwana, Avalonia, and basins within Laurentia (Northern and Southern Appalachian basins and the Upper Mississippi Valley). There appears to be no dominant dispersal direction or pattern among clades that underwent speciation events during this time slice. 57

Figure 7. Strict consensus vicariance and geodispersal area cladograms from LBPA analyses of the T0-T3 time slices. Vicariance trees indicate the relative order in which barriers fell, and geodispersal trees indicate the order that barriers where removed between areas. Bootstrap (plain text) and jackknife (italics) values indicate node support. T0 time slice trees are the strict consensus of two most parsimonious vicariance trees (length of 69 steps, CI 0.85, RI 0.78) and three most parsimonious geodispersal trees (length of 72 steps, CI 0.90, RI 0.81). T1 time slice trees are the strict consensus of 40 most parsimonious vicariance trees (length of 109 steps, CI 0.77, RI 0.46) and 36 most parsimonious geodispersal trees (length of 92 steps, CI 0.89, RI 0.50). T2 time slice trees are the strict consensus of 84 most parsimonious vicariance trees (length of 82 steps, CI 0.80, RI 0.68) and 24 most parsimonious geodispersal trees (length of 70 steps, CI 0.88, RI 0.76). T3 time slice trees are the strict consensus of three most parsimonious vicariance trees (length of 130 steps, CI 0.76, RI 0.55) and 10 most parsimonious geodispersal trees (length of 101 steps, CI 0.89, RI 0.65). 58

Vicariance and geodispersal area cladograms from the LBPA analysis are identical for the T0 time slice (Fig. 7), signifying that areas were separated and joined in the same order. Such patterns are produced when cyclical processes, such as eustatic sea level changes, controlled speciation and organism dispersal patterns on a global scale

(Lieberman, 2000). Both area cladograms indicate a recent separation and connection between the Western Midcontinent and Southern Laurentia and between Southern

Gondwana and Avalonia. In addition, the Upper Mississippi Valley plus North of the

Transcontinental Arch areas form a sister group relationship with Northern Gondwana; that triad is closely related to Baltica and the Southern Appalachian Basin. This group of five areas shares a closer relationship to Northern Laurentia and the Northern

Appalachian Basin than the remaining areas. Three areas: the Cincinnati Basin, Scoto-

Appalachia and the Tarim Plate are located basally on the consensus trees because they lack sufficient data to demonstrate a clear biogeographic relationship. The inability to resolve relationships of areas with fewer than three occurrence data is a well-understood limitation of the LPBA method (Lieberman, 2000).

Maximum Likelihood Models

The BioGeoBEARS optimization for the T0 time slice indicates that several dispersal events took place among Gondwana, Avalonia, Baltica, Laurentia, and the

Tarim Plate within Deiphoninae, Tetralichinae, Homalonotidae, Thaleops, and

Glyptorthis (Fig. 5B, 5D). Both Deiphoninae and Tetralichinae dispersed from Laurentia to Northern Gondwana. Additionally, deiphonine species dispersed from the Tarim Plate 59 into Laurentia and from Baltica into Laurentia twice during this time interval.

Tetralichinae exhibit several dispersal events mainly among Laurentian basins. Dispersal occurred from the Western Midcontinent and North of the Transcontinental Arch into

Northern Gondwana, as well as from the Western Midcontinent and North of the

Transcontinental Arch into the Upper Mississippi Valley. From the Upper Mississippi

Valley, additional dispersal occurred into the Southern Appalachian Basin. From the

Southern Appalachian Basin, dispersal took place into Baltica. Although the trilobite genus Thaleops is only represented within Laurentia, it displays similar patterns of dispersal to Tetralichinae, in which dispersal occurred from North of the Transcontinental

Arch and the Upper Mississippi Valley into the Southern Appalachian Basin. The ML model indicates three dispersal events took place from Avalonia into Laurentia within the

Homalonotidae and that vicariance within this clade took place between Southern

Gondwana and Avalonia. Species of Glyptorthis dispersed from the Western

Midcontinent and the Southern Appalachian Basin into the Cincinnati Basin with subsequent dispersal into the Northern Appalachian Basin.

T0 Time Slice Comparison of Parsimony and Maximum Likelihood Models

Comparison of FP and ML models of clades that exhibit speciation patterns during the T0 time slice reveals several similarities and overlaps in dispersal paths between methods (Fig. 8). Clades that underwent speciation and dispersal during this time slice include the Thaleops, Homalonotidae, Deiphoninae, Tetralichinae, and

Glyptorthis. Both FP and ML methods indicate that several dispersal events took place 60 among Laurentia, Baltica, the Tarim Plate, and Avalonia within deiphonine and homalonotid trilobites. Dispersal patterns within Glyptorthis indicate that the brachiopod genus dispersed among Laurentian basins.

The recovered patterns within Homalonotidae from FP and ML methods indicate that dispersal occurred from Southern into Northern Gondwana and from Avalonia into

Laurentia via the Cincinnati Basin, Southern Appalachian Basin, and Upper Mississippi

Valley (Fig. 8). Both FP and ML models for the Tetralichinae indicate a north to south/southeast dispersal pattern within Laurentia from the Upper Mississippi Valley into the Southern Appalachian Basin. There is a definite relationship among Northern

Gondwana, North of the Transcontinental Arch, and the Western Midcontinent, but dispersal directions between basins differ between methods. Within both methods,

Glyptorthis exhibits dispersal from the Southern into the Northern Appalachian Basin.

Dispersal from the Western Midcontinent into the Cincinnati Basin occurs within both methods for Glyptorthis. In the ML model, dispersal also occurs from the Southern

Appalachian Basin, whereas in FP, dispersal occurs exclusively from the Western

Midcontinent into the Cincinnati Basin as well as the Southern and Northern Appalachian

Basins (Fig. 5).

Figure 8. Pathways and basin associations reconstructed from Fitch Optimization and maximum-likelihood analyses. Arrows indicate paths that were identical between both methods; stars indicate basins that were reconstructed to share area relationships by both methods but in which the directionality of dispersal differed between reconstructions. Dashed lines indicate speciation by vicariance, and solid lines indicate speciation by dispersal. Colors indicate clade identity: purple, Homalonotidae; green, Tetralichinae; red, Glyptorthis; orange, Flexicalymene; black, Thaleops; blue, Bumastoides; brown, Eochonetes; white, Plaesiomys; yellow, Hebertella. See Figure 3 for area names.

T1 Time Slice

Parsimony Analyses

In the T1 time slice, which largely covers the Middle Ordovician Sandbian Age, speciation by vicariance was more common than speciation via dispersal, although both occurred (Table 2). Vicariance events during this interval indicate that southern

Laurentian basins became separated from the northern and continental interior basins within both brachiopod and trilobite phylogenies. Clades that exhibit speciation during this time interval are Glyptorthis, Hebertella, Thaleops, Flexicalymene, Bumastoides,

Deiphoninae, and Homalonotidae (Fig. 6). Flexicalymene species dispersed from

Avalonia into Baltica during four separate events and also dispersed into Baltica from

Avalonia. Similarly, homalonotid species dispersed from Avalonia into the Cincinnati

Basin. Within Bumastoides, three separate dispersal events from Northern Laurentia into southern basins within Laurentia occurred as well as a vicariance event that separated the species in the Southern Appalachian Basin from the Upper Mississippi Valley, Western

Midcontinent, and Northern Laurentia. Thaleops exhibits three dispersal events within

Laurentia and vicariance events that separated Northern and Southern Laurentia and the

Western Midcontinent. Deiphonine trilobites only experienced vicariance events during this time interval, with taxa inhabiting the Southern Appalachian Basin, Baltica,

Avalonia, and the Upper Mississippi Valley becoming separated from one another via three separate vicariance events. One vicariance and dispersal event occurs during the T1 time slice in Hebertella, both involving the Cincinnati Basin and Southern Appalachian

Basin. Within the Glyptorthis phylogeny, four vicariance events occur. 63

The LBPA vicariance area cladogram is poorly resolved, but indicates separation of the Northern Appalachian Basin from Southern Laurentia and the Upper Mississippi

Valley from Avalonia during this interval (Fig. 7). The geodispersal cladogram also indicates a close area relationship between the Upper Mississippi Valley and Avalonia through faunal exchange. The Western Midcontinent and Southern Appalachian Basin also exchanged taxa during this time. A large number of areas are resolved basally within the area cladograms, because congruent dispersal and vicariance events were unable to be determined due to conflicting patterns among the large set of MPT.

Maximum Likelihood Models

Dispersal was the primary mode of speciation during the T1 time slice, and results of biogeographic optimization indicates that dispersal occurred in all directions within

Laurentia and between Laurentia and Baltica, similar to the pattern observed in the T0 time slice (Fig. 5). The analyses of the Homalonotidae and Flexicalymene indicate that dispersal from Avalonia into Laurentia took place via the Cincinnati Basin and Northern

Appalachian Basin. Contrastingly, dispersal also occurred from the Southern

Appalachian Basin of Laurentia to Avalonia and Baltica within the deiphonine and tetralichine trilobites. Dispersal into the Cincinnati Basin from the Western Midcontinent and Upper Mississippi Valley is also evident from the analysis of Glyptorthis.

T1 Time Slice Comparison of Parsimony and Maximum Likelihood Models

Clades that underwent speciation and dispersal during this interval include all of the trilobites -- Tetralichinae, Deiphoninae, Homalonotidae, Thaleops, Flexicalymene, and Bumastoides. Hebertella and Glyptorthis dispersed between basins within Laurentia

(Fig. 8). Both FP and ML models indicate similar dispersal events within the trilobites, with the exception of Tetralichinae and Deiphoninae.

Analyses on Homalonotidae and Flexicalymene indicate dispersal occurred between Avalonia and Laurentia (Fig. 8). This result agrees with the LBPA performed solely on the homalonotid phylogeny of Congreve and Lieberman (2008), in which they concluded that a close area relationship between eastern (here considered southern)

Laurentia and Avalonia existed during the Sandbian. Indeed, this hypothesis is supported by FP and ML models of Flexicalymene, in which several dispersal events between

Avalonia and the southern margin of Laurentia are evident (Fig. 5C and D). In addition, both models also suggest that dispersal took place either between Laurentia and Baltica in the ML model or between Avalonia and Baltica as interpreted from FP. Combined, the results of the analyses indicate that trilobite dispersal happened via multiple dispersal pathways among Laurentia, Baltica, and Avalonia.

From FP and ML models of Glyptorthis, it is evident that the Laurentian Upper

Mississippi Valley, Southern Appalachian Basin, and Cincinnati Basin experienced exchange of taxa during this time (Fig. 5A and B; Fig. 8). Dispersal within the genus took place largely in a west to east direction as indicated by the results of both methods. Fitch

Parsimony and ML models recovered almost identical paleobiogeographic signals for 65

Thaleops. Both sets of results also indicate that dispersal took place from southern basins

(e.g., Northern Appalachian Basin and Southern Laurentia) to Northern Laurentia, and then into the Western Midcontinent via a counter-clockwise dispersal pattern.

T2 Time Slice

Parsimony Analyses

During the T2 time slice, which corresponds to the beginning of the Taconic tectophase of the Taconic orogeny, vicariance was the dominant speciation mode and dispersal events were comparatively reduced (Table 2). Fitch optimization of

Bumastoides indicates vicariance events separated the Southern Appalachian Basin from

Northern Laurentia, Western Midcontinent, and the Upper Mississippi Valley (Fig. 5).

Dispersal within the genus took place from Northern Laurentia into the Western

Midcontinent, Southern Appalachian Basin, and the Upper Mississippi Valley.

Eochonetes experienced several vicariance events among Laurentian interior basins, but most notable is the separation of Scoto-Appalachia from the Cincinnati Basin and North of the Transcontinental Arch. Dispersal within Eochonetes took place from North of the

Transcontinental Arch into the Western Midcontinent, Southern Laurentia, and the Upper

Mississippi Valley. Within Plaesiomys, three vicariance events occurred separating the

Laurentian western interior basins and Southern Laurentia. Flexicalymene exhibits dispersal from the Upper Mississippi Valley into the Cincinnati Basin and from the

Cincinnati Basin into the Western Midcontinent, Northern and Southern Appalachian basins, and Upper Mississippi Valley. Hebertella exhibits dispersal patterns similar to 66

Flexicalymene in that dispersal occurred from the Cincinnati Basin into the western interior basins as well as the Southern Appalachian Basin. Dispersal directionality is not apparent within this time slice, and the majority of vicariance takes place within

Laurentia among the interior basins.

The LBPA geodispersal and vicariance area cladograms are almost identical, indicating that cyclical geologic events were the major processes governing speciation globally (Fig. 7). Separation and faunal exchange occurred between the Cincinnati Basin and Scoto-Appalachian basin. Furthermore, the Upper Mississippi Valley and Western

Midcontinent exhibit a sister group relationship in both area cladograms. These areas share close biogeographic relationships with Southern Laurentia and the area North of the

Transcontinental Arch.

Maximum Likelihood Models

The T2 time slice indicates again that dispersal occurred in multiple directions.

BioGeoBEARS analyses of Hebertella and Eochonetes reveal dispersal from the

Cincinnati Basin into the Upper Mississippi Valley and Western Midcontinent. Dispersal into the Cincinnati Basin occurred from the Upper Mississippi Valley and Southern

Laurentia as interpreted from the analyses of Flexicalymene and Plaesiomys (Fig. 5B,

5D).

T2 Time Slice Comparison of Parsimony and Maximum Likelihood Models

Dispersal in this time slice occurred mainly within Laurentia, with no dispersal events recovered between paleocontinents. Clades that underwent dispersal and speciation during this time slice include Hebertella, Eochonetes, and Plaesiomys,

Bumastoides and Flexicalymene (Fig. 5).

Fitch Parsimony and ML models for Flexicalymene are nearly identical for this time slice, indicating dispersal occurred from the Upper Mississippi Valley into the

Cincinnati Basin, with subsequent dispersal into the Northern and Southern Appalachian basins, Western Midcontinent, and back into the Upper Mississippi Valley (Fig. 8).

Maximum likelihood models on Bumastoides indicate a dispersal event within Laurentia from the Upper Mississippi Valley into the Western Midcontinent. Fitch Parsimony, however, indicates that two vicariance events led to speciation within this genus during this time slice as the Upper Mississippi Valley became separated from the Western

Midcontinent and Northern Laurentia (Fig. 5 and 6, C and D). The models for

Eochonetes and Hebertella are similar in the fact that relationships between paleobiogeographic areas are evident, but the dispersal pathways and directions recovered are variable (Fig. 5A, 5B). Interestingly, FP indicates Eochonetes underwent speciation dominantly by vicariance; whereas ML models indicate this genus underwent speciation via dispersal events. Both models do, however, indicate that speciation processes within Laurentia and its marginal Scoto-Appalachian basin took place via vicariance (Fitch, Scoto-Appalachia separated from Cincinnati Basin and North of 68

Transcontinental Arch; maximum-likelihood, Scoto-Appalachia separated from

Cincinnati Basin) (Fig. 8).

T3 Time Slice

Parsimony Analyses

During the T3 time slice, which encompasses the Boda Event and increased deposition of photozoan carbonates in Laurentia, vicariance remained the dominant mode of speciation (Table 2), although dispersal events occurred commonly within brachiopod clades. Species of Hebertella dispersed in a west to east direction from the Cincinnati

Basin into Southern Laurentia, and vicariance took place within the genus as the

Cincinnati Basin became separated from the Western Midcontinent and Upper

Mississippi Valley (Fig. 5A). Glyptorthis underwent dispersal from the Western

Midcontinent in a southeastern direction into the Upper Mississippi Valley, Cincinnati

Basin, and the Southern Appalachian Basin. Dispersal events within Plaesiomys took place among all Laurentian basins, with the exception of Scoto-Appalachia and the

Northern Appalachian Basin. Within Plaesiomys, vicariance separated the Western

Midcontinent and North of the Transcontinental Arch from Southern Laurentia and the

Upper Mississippi Valley, and a second vicariance event separated Southern Laurentia from North of the Transcontinental Arch and the Western Midcontinent. Fitch optimization on deiphonine trilobites indicates that vicariance separated Northern

Gondwana from Baltica and Laurentian basins. Northern Laurentia and Baltica were also separated during this time within Plaesiomys. 69

The vicariance and geodispersal area cladograms are highly resolved among nine areas, but their topologies differ in key elements (Fig. 7). The vicariance area cladogram supports a biogeographic relationship among the Upper Mississippi Valley, Western

Midcontinent, North of the Transcontinental Arch, and Southern Laurentia. A second cluster of five regions supports recent separation between the Northern Appalachian

Basin and Northern Laurentia and between the Southern Appalachian Basin and

Baltica—which are also related to Northern Gondwana. The geodispersal cladogram indicates faunal exchange between the area North of the Transcontinental Arch and

Southern Laurentia and between the Western Midcontinent and Upper Mississippi

Valley. The last two areas also share a close relationship with the Northern and Southern

Appalachian basins. The geodispersal cladogram further indicates a relationship between

Baltica and Northern Laurentia. This incongruence between the vicariance and geodispersal trees indicates that singular, rather than cyclical, processes controlled biogeographic area relationships during this interval.

Maximum Likelihood Models

Within the T3 time slice, dispersal occurred from Northern Gondwana into

Northern Laurentia, and from there into Baltica, as reconstructed for the deiphonine analysis. Within Laurentia, Glyptorthis dispersed from the Upper Mississippi Valley and

Southern Appalachian Basin into the Cincinnati Basin. Plaesiomys dispersed from the

Upper Mississippi Valley into Northern Laurentia, the Western Midcontinent, and the

Southern Appalachian Basin (Fig. 5B and D). 70

T3 Time Slice Comparison of Parsimony and Maximum Likelihood Models

Relatively few dispersal and vicariance events occur in this time slice, and these are restricted to four clades: Plaesiomys, Hebertella, Glyptorthis, and Deiphoninae (Fig.

8). Fitch Parsimony and ML models of the brachiopod phylogenies are similar; dispersal occurred mainly between the same basins in each method, namely among western, northern, and eastern Laurentian basins. Fitch Parsimony and ML models for Plaesiomys are very similar as both methods indicate dispersal and vicariance events occur within the same lineages. Both methods for the genus indicate that dispersal occurred from western, northern, and eastern basins (e.g., North of Transcontinental Arch, Western Midcontinent into southern Laurentian basins (i.e., the Cincinnati Basin, Southern Laurentia, and

Southern Appalachian Basin). Both methods recover a dispersal pathway from the Upper

Mississippi valley into Northern Laurentia used by the species P. subcircularis, as both methods optimized the Upper Mississippi Valley as the location of the parent taxon (Fig.

5A and B). Both FP and ML models constructed for Hebertella agree that dispersal occurred from the Cincinnati Basin in an eastward direction into Southern Laurentia.

Dispersal within Glyptorthis occurred between eastern Laurentian basins. However, much like Plaesiomys, directionality and dispersal pathways differ, as FP indicates that dispersal happened from the Western Midcontinent into the Southern Appalachian Basin,

Cincinnati Basin, and Upper Mississippi Valley. The ML model, on the other hand, indicates that dispersal took place into the Cincinnati Basin from the Western

Midcontinent, Southern Appalachian Basin, and Upper Mississippi Valley, with a 71 vicariance event that separated the Western Midcontinent from the Southern Appalachian

Basin and Upper Mississippi Valley in the species G. glaseri.

Dispersal patterns reconstructed for deiphonine trilobites differ from this general trend. Ancestral range estimation for the deiphonine clade indicates that speciation occurred via dispersal from Northern Gondwana into Laurentia and from Laurentia into

Baltica. This dispersal pathway between Northern Gondwana and Laurentia was previously hypothesized by Congreve and Lieberman (2010) based on an LBPA analysis of the Deiphoninae. Fitch Parsimony analysis conducted on the dataset, however, indicates that this clade underwent speciation in the T3 time slice by vicariance between

Northern Gondwana and Laurentia.

Discussion

Comparison of Maximum Likelihood and Parsimony Analyses

Although there are several differences in dispersal patterns within clades in the ancestral range estimation from parsimony and model-based reconstructions, the larger- scale dispersal events and relationships between paleobiogeographic areas are mostly consistent between analyses. In instances of dissimilar patterns, relationships among basins are still congruent. Moreover, the results obtained from this study agree with those of previously published parsimony (LBPA) analyses conducted on single phylogenies 72 used within this study (e.g., dispersal patterns of Deiphoninae and Homalonotidae of

Congreve and Lieberman [2008; 2010]).

Minor differences in dispersal patterns within clades are evident. For example,

ML models reconstruct dispersal of Flexicalymene from Avalonia to Laurentia into

Baltica, but FP indicates that this clade dispersal from Avalonia directly into Baltica. The biggest difference between FP and ML results is the relative frequency of vicariance and dispersal events per time slice. BioGeoBEARS analyses recovered more speciation events attributable to dispersal than vicariance in all time slices, whereas Fitch parsimony indicated vicariance was the dominant speciation process in all time slices other than T0.

However, neither model favors vicariance over dispersal for any single phylogeny, nor is there a statistical difference in the number of vicariance events recovered between the methods. The relative frequencies differ primarily because BioGeoBEARS analyses recover more episodes of speciation by dispersal. The high recovery of dispersal is explained by the fact that the best-fit models for each phylogenic dataset included a ―j‖

(jump dispersal) parameter in all instances.

It should be noted that jump dispersal within BioGeoBEARS will be favored if a phylogenetic hypotheses is constructed using specimen location data instead of collective species occurrences, making it seem as though a species only inhabited a single location.

This can lead to erroneous conclusions about the origination and mode of speciation prominent within a clade (Matzke, pers. comm. 2014). However, incomplete occurrence data was not a significant factor in this study, as all phylogenetic hypotheses used were 73 constructed based upon several occurrences per species, therefore validating that jump dispersal was indeed an important speciation process that took place within the Paleozoic.

For some clades, vicariance optimizations were very dissimilar between methods.

For example, the FP analysis of the brachiopod genus Glyptorthis indicates within the T0 time slice that only one vicariance event took place, whereas the ML model indicates that several vicariance and dispersal events contributed to evolution within this Laurentian clade. Fitch Parsimony analysis for the trilobite clade Tetralichinae indicates that several vicariance events separated basins within Laurentia, but the ML model indicates that dispersal between basins was more prevalent. This is also the case for the Deiphoninae trilobites in the T3 time slice.

Vicariance and geodispersal trees produced by Lieberman-modified Brooks

Parsimony analyses were identical or nearly identical for the T0, T1, and T2 time slices but incongruent for the T3 time slice. Congruent results indicate that cyclical processes, such as eustatic sea level, controlled speciation and biogeographic patterns; whereas incongruent patterns indicate that singular events, like orogenic events, were the primary driver of biogeographic area relationships (Lieberman, 2001). Thus, cyclical processes produced biogeographic area relationships during the Dapingian to the C3/C4 sequence boundary, but the influence of localized tectonic and paleoclimate events within

Laurentia prevailed among paleobiogeographic basins in the final temporal interval.

Comparison of Speciation Events to Other Time Intervals

The relative frequency of vicariance and dispersal events recovered using FP and

ML methods for the Middle and Late Ordovician can be compared to other intervals in

Earth history. Stigall (2010a) compared the relative frequency of speciation via vicariance during the Cambrian, Ordovician, Devonian, and recent. Results of that study indicated that the relative importance of each speciation mode has varied through time, although vicariance is typically equally or more frequent than speciation by dispersal

(Stigall, 2010a). Speciation events recovered for the T0 time slice using FP (vicariance,

33%; dispersal, 67%) resemble speciation mode frequencies for Late Devonian taxa

(vicariance, 28%; dispersal, 72%) (Stigall, 2010b). Speciation percentages recovered using ML across the T0 to T3 time slices also most closely resemble the Late Devonian values due to highly elevated dispersal events. The high dispersal of the Late Devonian has been attributed the pervasive interbasinal-species invasion events at that time (Stigall,

2010a), which is comparable to the many long-distance dispersal events observed in this data set (Fig. 8). The percentage of dispersal events recovered using FP from the T1 to T3 time slices most closely resembles the modern fauna (modern: vicariance, 74%; dispersal,

26%) (Brooks & McLennan, 2002). Tectonic plates were widespread and oceans were relatively large in both the T0 interval and recent past, which may have facilitated an increased amount of vicariant speciation. Alternately, the narrowing of the Iapetus ocean and increased development of subduction-related island arcs during the T1 through T3 intervals may have promoted an increased in dispersal frequency. 75

Recently, several studies have implemented BioGeoBEARS for ancestral range estimation of modern taxa. Pyron‘s (2014) study of amphibian dispersal patterns indicated that vicariance events were elevated in both the Cretaceous (vicariance, 69%; dispersal, 31%) and Cenozoic (vicariance, 78%; dispersal, 22%) based on a DEC+J model. Litsios et al. (2014) implemented BioGeoBEARS on a phylogeny of clownfishes within the Indian Ocean, Indo-Australian Archipelago, and Central Pacific Ocean. The

BayArea+J model optimized areas within their phylogeny best, with increased dispersal events over the past 15 million years (vicariance, 15%; dispersal, 85%). The results of

Pyron (2014) match most closely with those obtained for modern fauna from Brooks and

McLennan (2002) and for the T1 to T3 speciation percentages recovered using FP in this study. This may indicate that dispersal processes that operated in the Middle through Late

Ordovician were similar to processes operational across island clades on modern timescales. As island arcs are located around subduction zones in today‘s oceans (ex,

Japan‘s Ryukyu Islands), volcanic island arcs within subducting boundaries around the majority of the Late Ordovician landmasses (Fig. 1) acted as stepping stones to dispersing taxa, leading to founder-event long distance speciation.

Biotic Influences on Speciation Patterns

Larval ecology plays an important role in dispersal distance and patterns within marine invertebrate organisms (Chatterton & Speyer, 1989; Freeman & Lundelius, 2005).

Dispersal patterns observed within this study can be partly attributed to larval type, time 76 spent in the larval phase, and larval position in the water column. Lam and Stigall (2015,

Chapter 2 of this thesis) reported that the majority of Ordovician organisms that dispersed into the Cincinnati Basin during the Richmondian (T3 time slice of this study) had planktotrophic or planula larvae. Planktotrophic larvae are self-feeding forms capable of spending weeks to months in a larval phase, whereas planula are self-feeding larvae that spend less time in a larval state compared to planktotrophic larvae (Jablonski & Lutz,

1983). Notably, both larval forms have high dispersal potential given adequate oceanographic conditions. In fact, studies of recent scleractinian corals indicated that their planula larvae can disperse up to 150 km in a single generation (Treml et al., 2008).

Studies of Paleozoic articulated brachiopod shells indicate these organisms produced large larval shells (protegulum) common among planktotrophic organisms, which suggests that Late Ordovician articulate brachiopods had planktic planktotrophic larvae

(Valentine & Jablonski, 1983; Freeman & Lundelius, 2005). As planktotrophic larvae can persist longer as plankton than planula larvae of recent corals, Ordovician brachiopods would have had the capability to travel along the dispersal pathways reconstructed using FP and ML methods in this study.

All of the trilobite clades examined in this study have previously been hypothesized to have had benthic larvae, although Flexicalymene may have had either planktonic or benthic larvae (Chatterton & Speyer, 1989). It should be noted that within the Deiphoninae and Homalonotidae trilobites, Chatterton and Speyer (1989) concluded that protaspids could have alternatively had a planktonic and/or benthic larval stage.

Compared to planktonic larvae, the dispersal potential of benthic larvae is lower. Surface 77 currents, which were likely the main influences on dispersal patterns within larval forms in the Late Ordovician (Lam & Stigall, 2015), would not have been effective at dispersing benthic larvae. Therefore, dispersal of benthic larval forms among paleocontinents, such as the dispersal of Tetralichinae trilobites among Northern

Gondwana, Baltica, and Laurentia, was more likely accomplished by shorter larval dispersals (―island-hopping‖) along volcanic island arcs over longer timescales relative to brachiopod dispersal patterns. These areas have previously been identified as ―stepping stones‖ between paleogeographic areas suitable for larval growth and development in the

Paleozoic by Neuman (1972). Island-hopping was also proposed by Congreve and

Lieberman (2008) in their explanation for Homalonotidae dispersal patterns observed within an LBPA analysis among Laurentia, Avalonia, and Arabia (Southern Gondwana).

McKerrow et al. (2000) suggested that dispersal of organisms across oceans during the

Late Ordovician were plausible, as they concluded that during this time, all oceans separating continents were less than 1000 km wide. However, a subsequent study by Lees et al. (2002) estimated distances between paleocontinents to have been from 2,000 km to

17,000 km. Even with significantly wider oceans separating shallow water habitats, larval dispersal could have been facilitated by the abundant volcanic island arc chains surrounding subduction zones of Laurentia, Baltica, Avalonia, and Gondwana (Fig. 3).

Paleoceanographic and Tectonic Influences on Species Dispersal

A number of dispersal pathways were identified from both FP and ML ancestral reconstructions. Ascertaining causal mechanisms for species dispersal requires examining the reconstructed pathways in light of published paleoclimatic and paleoceanographic reconstructions for the Middle through Late Ordovician. As no paleoclimatic reconstructions for the Middle Ordovician have been developed, the reconstructed connections between continents were compared to the ocean circulation models of

Herrmann et al. (2004) for the Caradoc (Sandbian-early Katian) and Ashgill (early

Katian-Hirnantian). Hermann et al. (2004) indicate that major gyre systems and ocean circulation patterns changed only slightly due to paleogeographic differences between the

Middle and Late Ordovician, so their analyses provide a framework for comparison in all four time slices.

During the Dapingian to Sandbian stages of the Middle Ordovician T0 time slice, four primary dispersal paths and directions were reconstructed by both FP and ML methods (Fig. 8). Dispersal of tetralichine trilobites within Laurentia is consistent with surface currents reconstructed by Ettensohn (2010), in which surface current flow was hypothesized to move from the northeast to southwest within Laurentia from the interaction of prevailing trade winds and the Coriolis Effect. These surface currents, along with strong storms originating around 10°S of the equator that swept across

Laurentia from the northeast to southwest (Poussart et al., 1999; Jin et al., 2013), would have aided dispersal of the trilobites from the Upper Mississippi Valley into the Southern 79

Appalachian Basin. Storm activity could have also influenced the dispersal of Glyptorthis from the Southern to Northern Appalachian Basin. The T0 time slice is characterized by a greater proportion of speciation events due to dispersal (Table 2), which suggests that high sea level might have eliminated barriers between basins or allowed larvae to breach barriers (Patzkowsky & Holland, 1996), thereby influencing dispersal between areas such as in Glyptorthis and tetralichine within Laurentia. Ocean circulation models of Hermann et al. (2004) include no significant current between Laurentia and Avalonia that would have produced the reconstructed dispersal between these continents directly, thus, volcanic island arcs and smaller, localized currents likely influenced immigration of taxa from Avalonia within the homalonotid trilobites. The dispersal of the homalonotid trilobites between Southern and Northern Gondwana was probably facilitated by the

South Paleotethys Convergence gyre which operated in the Paleotethys Sea and circulated water between the western coast of Gondwana and the eastern margins of

Baltica and Siberia (Hermann et al., 2004).

The T1 time slice encompassing the Sandbian Stage includes several dispersal paths within Laurentia and between Laurentia and Avalonia in the clades Flexicalymene,

Glyptorthis, Thaleops, and Homalonotidae (Fig. 8). Dispersal patterns for Flexicalymene indicate that dispersal occurred from 1) Avalonia into Laurentia, then into Baltica as interpreted from ML models, or 2) dispersal occurred directly from Avalonia into Baltica within FP (Fig. 5C and D). Surface currents for the Caradoc indicate that the Iapetus

Current flowed between Laurentia and Baltica in an east to west direction, splitting around Laurentia to the north and south (Hermann et al., 2004). The authors did not 80 identify a gyre system in this area, but their oceanographic reconstruction suggests there may have been weak development of circulating currents between Baltica and Laurentia.

A current likely flowed from the west between Avalonia and Baltica, with no direct influence into or around Laurentia (Hermann et al., 2004). Therefore, the dispersal path hypothesized by ML models for Flexicalymene is likely more correct, although it is not completely implausible that dispersal within the genus took place among the three paleocontinents. After all, as previously discussed for the Middle Ordovician, the

Homalonotidae dispersed within this time slice directly from Avalonia into Laurentia, with dispersal aided by chains of volcanic island arcs surrounding the southern margin of

Laurentia (Mac Niocaill et al., 1997). Vicariance between the Southern Appalachian

Basin and the Western Midcontinent is reconstructed for Glyptorthis within FP and ML methods. The increased tectonic activity during this time slice likely emplaced barriers between paleobiogeographic basins, especially on the southern margin of Laurentia, which would have restricted dispersal among southern margin basins (Cracraft, 1985). In addition, both methods indicate that Glyptorthis species in the Upper Mississippi Valley and Southern Appalachian Basin shared a relationship with those of the Cincinnati Basin, although dispersal directions were incongruent between methods. Dispersal among these areas was likely influenced by storm activity, especially from the Upper Mississippi

Valley into the Cincinnati Basin (Jin, 2013). Storms and hurricanes have been observed in modern clades to greatly influence larval dispersal distance (Radford et al., 2014).

Within Thaleops, dispersal occurred from the Northern Appalachian Basin and Southern

Laurentia into Northern Laurentia, with another dispersal event into the Western 81

Midcontinent. Localized currents and changing ocean circulation patterns from increased tectonic activity could have influenced such dispersal patterns as has been hypothesized for the Cenozoic (Berggren & Hollister, 1977)

Within the late Sandbian-mid late Katian T2 time slice, FP and ML methods reconstructed only two congruent dispersal paths, those of Flexicalymene and Eochonetes

(Fig. 5). After a brief quiescent period following the Blountian tectophase, the Taconic tectophase began with the locus of activity shifted towards the New York promontory

(Shanmugam & Lash, 1982; Ettensohn, 1994). The increased flexural downwarping on the southern edge of Laurentia led to the collapse of the Reelfoot Reef and development of the Sebree Trough in the midcontinent region (Kolata et al., 2001). In turn, cool, nutrient-rich oceanic waters were funneled into the continent from the Iapetus Ocean, causing counter-clockwise estuarine-like circulation patterns created by the subtropical convergence zone within Laurentia from the interaction of warm tropical waters and cool subpolar currents (Wilde, 1991; Kolata et al., 2001). Waters would have been circulated between the Upper Mississippi Valley and the Cincinnati Basin, which would have aided in dispersal of Flexicalymene between these areas. Dispersal from the Upper Mississippi

Valley into more southerly basins (Southern and Northern Appalachian basins) was likely aided by storm activity and surface currents as previously discussed. In addition, the

Maysvillian Stage Montoya Group of the Western Midcontinent basin records intense upwelling zones hypothesized to have facilitated the migration of benthic larvae across

Laurentia (Pope, 2004). Both upwelling zones from the midcontinent Sebree Trough and western midcontinent areas likely contributed to the dispersal patterns observed within 82

Flexicalymene during this time interval. FP and ML methods indicate that the strophomenid brachiopod Eochonetes underwent vicariance between Scoto-Appalachia and the Cincinnati Basin during the T2 interval (Fig. 5A and B; 8). The observed vicariance pattern accords with tectonic influences during this time, as increased barrier emplacement on the southern and southeastern margin of Laurentia due to the Taconic tectophase likely inhibited dispersal between Laurentia and its marginal Scoto-

Appalachian basin (Ettensohn, 1994). Support for increased barrier emplacement within

Laurentia is indicated by high rates of vicariant speciation for this time slice within ML models, and the second highest vicariance rates reconstructed using FP (Table 2). This is in disagreement with the results of Wright and Stigall (2013a), who performed FP on three brachiopod genera (Glyptorthis, Plaesiomys, Hebertella) used in this study and concluded that this period in time coincided with increased prevalence of speciation by dispersal. Because their analysis did not include trilobite clades, conflicting results may indicate that tectonics may have affected trilobite speciation differently, as benthic larvae responded to different paleoclimatic and oceanographic factors.

During the mid-late Katian to Hirnantian stages of the Late Ordovician T3 time slice, dispersal is observed within Plaesiomys, Hebertella, and Deiphoninae. This time interval is associated with profound paleoclimatic changes as tropical carbonate deposition returned to the southern margin of Laurentia as a consequence of the global warming interval, the Boda Event (Holland, 1997; Fortey & Cocks, 2005). Infilling of the Sebree Trough led to the eventual cessation of midcontinent counter-clockwise circulation patterns (Kolata et al., 2001). Instead, dispersal from the Upper Mississippi 83

Valley into Northern Laurentia within Plaesiomys was aided by storms in the northern hemisphere moving from the southeast to the northwest (Jin, 2013). Dispersal of the genus across the Transcontinental Arch was likely achieved by increased sea level from a major transgression at the Richmondian Age C5 sequence boundary (Holland, 1997).

Tempestite beds indicate strong storm activity in the Cincinnati Basin that could have influenced dispersal of brachiopod larvae out of the Cincinnati Basin into surrounding basins, as exhibited in Hebertella during this time (Holland & Patzkowsky, 1996). Along with elevated sea level, dispersal in an eastern direction into Southern Laurentia could have been accomplished as basins were infilled, removing physical and thermal barriers

(ex., the Transcontinental Arch and deep-water Maquoketa belt) among basins (Copper &

Grawbarger, 1978; Elias, 1983). The ocean circulation models for the Ashgill of

Hermann et al. (2004) indicate the South Paleotethys convergence zone continued to operate between the western coast of Baltica and eastern margin of Gondwana during this time. Thus, ocean circulation patterns, along with volcanic island arcs acting a stepping stones between areas of sustainability, influenced the dispersal of deiphonine trilobites from Northern Gondwana into Laurentia during this time interval.

Synthesis

The reconstructed dispersal patterns within the four time slices, but especially the

T0 and T1 time slices, can be attributed to surface currents and strong storm patterns within Laurentia. Major oceanic currents, such as the equatorial Iapetus current and the 84

South Paleotethys convergence zone, aided in dispersal of taxa among paleocontinents throughout the study interval. Cyclical processes, such as transgressive-regressive cycles, that operated on a smaller scale within Laurentia and on a global scale are evident from the congruent T0 vicariance and geodispersal area cladograms of the LBPA analysis. The

T1 areagrams are somewhat congruent, but indicate that some basins, especially on the southern margin of Laurentia, experienced different factors driving vicariance and geodispersal events. The increased tectonic activity from the Blountian tectophase on the southern margin of Laurentia emplaced barriers, thus restricting dispersal patterns during the T1 time slice. Within the T2 time slice, the development of the Sebree Trough influenced dispersal among midcontinent basins. Upwelling off the western coast of

Laurentia likely aided dispersal into basins to the east. Again, the resolved portions of

LBPA areagrams for the T2 time slice are congruent, indicating cyclical and recurrent processes influenced dispersal patterns during this time. The T3 time slice includes one of the largest transgressions of the Late Ordovician, in which dispersal of organisms was influenced by rising sea level that broke down or allowed for immigration over thermal and physical barriers that were unable to be breached in earlier time intervals. LBPA areagrams for this time slice are not congruent, and indicate that speciation processes among Laurentian basins and paleocontinents were less affected by cyclical processes and more so by local tectonic and paleoceanographic factors.

Conclusions

Ten phylogenetic hypotheses of brachiopods and trilobites spanning the Middle through Late Ordovician were subjected to Fitch Optimization and Lieberman-modified

Brooks Parsimony Analysis (LBPA) and the maximum-likelihood R program

BioGeoBEARS to compare speciation processes and dispersal routes across four time slices. Both FP and ML methods recovered similar relationships among paleogeographic areas, with differences most evident in dispersal direction and speciation type among time slices. Parsimony methods continually recovered high frequency of speciation by vicariance (with the exception of the T0 time slice), whereas BioGeoBEARS models recovered higher rates of dispersal speciation. Within BioGeoBEARS, biogeographic evolution of all clades was best fit to models that incorporated long-distance founder- event speciation, indicating this was an important speciation process for Middle to Late

Ordovician taxa. Such dispersal was facilitated by volcanic island arcs and shallow-water carbonate basins acting as stepping-stones among paleobiogeographic areas and paleocontinents. Within all time slices, dispersal occurred among major paleocontinents, which was facilitated by planktonic and planula larvae transported by paleoceanographic currents flowing between and within landmasses. In addition, paleoclimate and oceanographic factors may have impacted trilobite and brachiopod larvae differently, as the results of this analysis differ from those exclusively using brachiopod genera to reconstruct dispersal pathways within the Late Ordovician (e.g., Wright & Stigall, 2013;

Bauer & Stigall, 2014). Vicariance and geodispersal trees recovered from LBPA were 86 relatively well-resolved for all time slices. Congruent vicariance and geodispersal area cladograms for the T0 to T2 time slices indicate that cyclical processes, such as eustatic sea level changes, influenced speciation patterns on a global scale. Resolved portions of the T3 time slice areagrams indicate that vicariance and geodispersal processes operated differently among Laurentian basing during the Richmondian Age. The overall results of this analysis indicate that both parsimony and maximum likelihood Bayesian analyses are valuable tools in reconstructing and interpreting speciation processes in deep time.

Although much work remains in perfecting the methods of applying maximum likelihood analyses to parsimony-constructed phylogenies, both methods have the potential to elucidate patterns and processes of speciation and dispersal in deep time.

CHAPTER 4: ORIGIN, PATHWAYS, AND MECHANISMS OF THE LATE

ORDOVICIAN RICHMONDIAN INVASION: A REVIEW

Introduction

Late Ordovician (Richmondian) strata within the Cincinnati Arch area

(southeastern Ohio, southwestern Indiana, northern Kentucky) record a mass faunal migration event, termed the Richmondian Invasion (Fig. 9) (Holland, 1996). This invasion is characterized by the introduction of over 60 genera to the Cincinnati region that lack faunal affinities with the incumbent fauna and, thus, immigrated into the region from other basins. Speculation about the geographic origins of the invasive taxa began in the early 1900‘s with the works of Auguste F. Foerste. Foerste was one of the first stratigraphers working in the region to notice a significant faunal shift and recurrence in faunal associations within the strata that corresponded to lithologic changes (e.g., Foerste,

1905; 1912; 1924). Multiple competing hypotheses about the source of the invaders, initially termed the Red River-Stony Mountain fauna (after the Red River formation in southern Manitoba) or ―Arctic Fauna‖ by Flower in 1946 have been published over the past century (Flower, 1946; Holland, 1997; Jin, 1999; Wright & Stigall, 2013a). Many authors (ex., Flower, 1946; Sweet & Miller, 1958; Macomber, 1970; Frey, 1981; Elias,

1982; Frey, 1985; Sloan 1987; Mitchell & Sweet, 1989) hypothesized that the fauna originated from western and northern North America (northern Canada, Wyoming,

Colorado, South Dakota, Texas, Iowa, Minnesota), although some of these authors 88 considered only a few taxa (ex., corals: Elias, 1982; 1983; cephalopods: Frey, 1981;

1985). Several recent studies have applied quantitative paleobiogeographic methods such as parsimony analysis of endemicity, phylogenetic biogeography, and maximum likelihood analyses (e.g., Wright & Stigall, 2013a; Bauer & Stigall, 2014; Lam & Stigall,

2015; Chapters 2 and 3), which have indicated alternative source regions for the invasive taxa and elucidated potential dispersal pathways. This chapter synthesizes the current state of understanding of biogeographic dynamics of the Richmondian Invasion by reviewing traditional and novel biogeographic hypotheses about the biogeographic origins of the invaders, dispersal pathways utilized, and paleoclimatic and oceanographic conditions that facilitated the large-scale regional invasion event.

Overview of the Richmondian Invasion

The Richmondian Invasion involved the migration of over 60 genera from five phyla and all trophic levels into the Cincinnati Basin beginning in the early Richmondian

Stage C4 sequence (Holland, 1996). The majority of the genera that participated in the invasion were benthic invertebrates, primarily bivalves, gastropods, and brachiopods, and a few coral, trilobite, bryozoa, and crinoid genera. The invasion occurred in two main pulses. The first phase of the invasion occurred in the middle to late C4 sequence (Fig. 9).

This phase was limited to approximately ten extrabasinal genera (Fig. 10). The second phase of the Richmondian Invasion occurred at the C5 sequence boundary (Fig. 9) and is 89 marked by the introduction of over 50 extrabasinal genera into the Cincinnati Basin (Fig.

10).

Figure 9. Location, extent of exposed Ordovician outcrop, lithostratigraphy and sequence stratigraphy of the Cincinnati Basin. The first phase of the Richmondian Invasion is indicated in red; second phase in orange. Outcrop belt modified from Holland (1993); stratigraphy modified from Holland and Patzkowsky (1996).

Numerous studies have examined the paleoecology, community composition, and macroevolutionary impact of the Richmondian Invasion interval during the past twenty years (ex., Patzkowsky & Holland, 1996; Holland & Patzkowsky, 2007; Patzkowsky &

Holland, 2007; Dudei & Stigall, 2010; Stigall, 2010a, b; Malizia & Stigall, 2011; Brame

& Stigall, 2013; Stigall, 2014). Consequently, primary ecological patterns are well constrained.

During deposition of the pre-invasion C1 to C3 sequences, biogeographic and faunal communities of the Cincinnati Basin displayed a high degree of stability at the genus level. The C1 to C3 sequence strata consist of siliciclastic shale layers and skeletal packstones that reflect a heterozoan assemblage indicative of cool-water or high-nutrient conditions (Holland 2008). The frequent occurrence of phosphate grains, coupled with lack of ooids, peloids, and micrite supports the interpretation that cool, nutrient-rich waters were introduced to the Cincinnati region from upwelling along the Seebree

Trough (Holland, 2008). Taxa recorded in the C3 sequence displayed a range of geographic sizes indicative of both ecological specialist and generalist ecologies (Stigall,

2010b). Gradient analyses indicated that community structures, key environmental gradients and general taxonomic abundance patterns were broadly conserved across this interval encompassing the Edenian and Maysvillian Stages (Holland & Patzkowsky,

2007; Patzkowsky & Holland, 2007). Competitive pressure among organisms was also interpreted to have been relatively low (Patzkowsky & Holland, 2007). Similarly, ecological niche modeling (ENM) analyses examining similarity of the environmental niches of brachiopod species and genera of other common benthos (trilobites, bryozoa, 91 crinoids, corals, bivalves, gastropods) retrieved high degree of niche stability and habitat tracking of individual taxa during this interval (e.g., Dudei & Stigall, 2010; Malizia &

Stigall, 2011; Stigall, 2011, 2012, 2015; Walls & Stigall, 2011; Brame & Stigall, 2014).

During the C4 sequence, changes in sedimentology indicate an influx of water from a new oceanographic regime. Faunal assembly also undergoes a shift. The community structure and faunal gradients, which had been stable for the previous five million years, begin to break down, especially the shallow and deep subtidal facies

(Holland & Patzkowsky, 2007). Notably, certain genera that were common in the deep subtidal facies of the C2 and C3 sequences (ex., Rafinesquina, Zygospira, ramose trepostome bryozoa) come to dominate the shallow subtidal facies halfway through the

C4 sequence. Concurrently, the dominant brachiopod assemblage that previously dominated the shallow subtidal facies in earlier sequences becomes less important in the

C4 (Holland, 1996). This faunal change is quite abundant, so much so that early workers such as Ulrich (1911, 1914) considered this interval to be the base of the Silurian System

(Holland & Patzkowsky, 2007). The influx of the first wave of invasive taxa into the

Cincinnati Basin led to the extirpation of native brachiopod specialist species with narrow geographic ranges, whereas native generalist brachiopod species with larger geographic ranges were able to contract their ranges and survive the invasion interval (Stigall,

2010b). Within the C4 sequence, invasive brachiopod genera to the Cincinnati area occupied smaller geographic ranges with respect to incumbent brachiopod genera

(Stigall, 2010b). Furthermore, taxa were shown to undergo niche evolution in geographic space (Stigall, 92

Figure 10. Stratigraphic ranges of select Richmondian Invaders in the Cincinnati Basin. List of invaders from Holland (1996). Occurrence data from the Paleobiology Database, Frey (1985), Kelly and Pope (1979), Dalvė (1948).

2012) and shift the geographic centroid of their range during the C4 sequence (Dudei and

Stigall, 2010b).

The C5 sequence records the peak of the Richmondian Invasion inter-basinal immigrations as genera of tabulates, rugosans, brachiopods, bryozoa, bivalves, gastropods, cephalopods, trilobites, and echinoderms unrecorded in previous sequences became abundant and constituted the greatest portion of the fauna (Anstey, 1986;

Holland, 1997; Patzkowsky & Holland, 1997; Holland & Patzkowsky, 2007). A new, stable community structure was developed by the middle of the C5 sequence (Holland and Patzkowsky, 2007), but the primary ecological determinants and associations were substantially different from the pre-invasion communities. The influx of invaders within this interval greatly increased competition, led to decreased speciation rates, and reduced the efficacy of organisms to track their preferred habitats (Stigall, 2010b; Tyler &

Leighton, 2011, Stigall, 2014). Many of the successful taxa (both native and invasive) shared the characteristics of large geographic range size (Stigall, 2010b), broad ecological niche dimensions (Stigall, 2014), and flexibility of niche use (Brame & Stigall,

2014). Among articulated brachiopods, carryover taxa exhibited the largest geographic ranges; however invader taxa also expanded their range size compared to the C4 sequence (Stigall, 2010b). Interestingly, brachiopod species that evolved de nova during the middle to late C5 sequence were derived from incumbent ancestors and exhibited small geographic ranges. These new species have been interpreted as ecological specialists that evolved the capacity to infill the previously vacated specialist niches

(Stigall, 2010b). 94

These biotic changes also coincide with environmental changes. The carbonate layers of the C5 sequence exhibit abundant micrite and coated grains that are similar to modern photozoan carbonates. This shift in lithology has been attributed to a decline in nutrient load within the basin due to paleoceanographic conditions coinciding with the

Richmondian transgression (Holland & Patzkowsky, 1996). However, the switch from temperate- to tropical-style carbonate deposition is also associated with the onset of the globally-recognized warming trend, the Boda Event (Fig. 10) (Fortey & Cocks, 2005, but see Cherns & Wheeley 2007 for an alternate interpretation).

Notably, the introduction of invasive taxa into the Cincinnati Basin during the

Richmondian Invasion did not cause increased extinction rates for benthic taxa. Instead, native species with generalist ecology were able to persist into and through the invasion interval, which led to an increase in diversity during the C4 and C5 sequences

(Patzkowsky & Holland, 2007). Notably, the diversity increase is most evident at the genus level as it results from the addition of genera, such as Plaesiomys, that are monospecific in the basin. Speciation rates actually declined during the invasion pulses in the C4 and early C5 sequences. The community restructuring and response of incumbent taxa to the influx of invasive species during the Richmondian Invasion is of particular importance to modern studies of invasive species dynamics. Studies of modern invasive organisms and their effects on ecosystems and ecosystem restructuring are limited in temporal duration, typically to years or decades. By studying an ancient invasion interval, such as the Richmondian Invasion, long term impacts and patterns of invasive species 95 have the potential to provide insight into the future implications of modern-day invasive taxa (Stigall, 2010b, 2012).

A True Invasion or Recurrent Fauna?

The Richmondian invaders consist of taxa that were either not recorded in the

Cincinnati Basin from earlier sequences or that had reappeared after an absence of several sequences (Fig. 10). Foerste (1924) called the recurrent fauna the Black River, or

Arctic fauna, as he hypothesized that these taxa migrated northward to the Polar Sea (the shallow oceanic area around Greenland) during the latter part of the Maysvillian Age before migrating back into epicontinental Laurentian basins during the Richmondian

Age. Several other authors (ex., Elias, 1985; Frey, 1985; Mitchell and Sweet, 1989) considered the invasion a major immigration episode, whereas others (e.g., Howe, 1988) proposed instead that the new genera appeared due to an evolutionary burst within the

Cincinnati Basin. Patzkowsky and Holland (1996) demonstrated that some invaders, consist of recurrent fauna, such as the brachiopod genus Strophomena, occur in the C1 or

C2 sequence but then disappear from the Cincinnati Basin until their occurrence in the C4 or C5 sequences. Holland (1996) demonstrated through his time/environment analysis that the invasion event was in fact a series of rapid regional immigration events and not artifacts of facies controls or stratigraphic signatures. Jin et al. (2013) questioned whether the introduction of the extra-basinal taxa qualified as a coherent invasion event, and 96 instead suggested the influx of taxa resulted from a shifting latitudinal faunal gradient at the local or regional scale.

The invasion is herein examined at the generic level because most studies of the

Richmondian Invasion, including seminal work by Holland and Patzkowsky, were conducted at the genus level. To examine the degree to which Richmondian invaders could be categorized as recurrent fauna, a new and updated species occurrence database was compiled from collection-based taxonomic occurrence sites, such as the

Paleobiology Database (paleodb.org) and iDigBio (idigbio.org). The vast majorities of the Richmondian Invaders do not have ancestors in younger strata within the Cincinnati

Basin since at least the Sandbian Stage and thus do not qualify as recurrent genera (Fig.

10). Although certain components of the invasive taxa are recurrent from earlier sequences, some orders such as rhynchonellid brachiopods and rugose corals, had been absent from the basin for approximately five million years prior to the Richmondian

Invasion. The lack of close relatives within the Cincinnati basin for many invader lineages also negates the in situ evolutionary burst hypothesis of Howe (1988). Overall, the majority of the invasive genera that appeared in the C4 and C5 sequences were new to the Cincinnati Basin. Recurrent genera include organisms within the classes Bivalvia

(40% recurrent fauna), Brachiopoda (33%), Gastropoda (80%), Rugosa (66%), and

Tabulata (33%). New genera to the basin include organisms in the classes Crinoidea,

Monoplacophora, and Trilobita.

Hypotheses of the Geographic Origin and Dispersal Pathways of Invasive Taxa

Arctic (Equatorial) Pathway

The earliest and most frequently reiterated hypothesis about the origin of the

Richmondian invaders is that they came from northern and western regions of North

America via unidirectional dispersal paths (Fig. 11). This has led to the taxa being termed the ―Arctic Fauna‖ as many are thought to have come from high-latitude regions of

Canada and western North America (Nelson, 1959). These regions are now reconstructed to have been located in the Ordovician paleotropics (see Jin et al., 2013), but the name was coined prior to the plate tectonic revolution. Nelson (1959) proposed that this fauna was easily recognizable from the presence of Receptaculites, Catenipora, Manipora, and large molluscs belonging to genera such as Maclurites, Hormotoma, Fusispira,

Cyrtogomphoceras, Winnipegoceras, Lambeoceras, and Diestoceras. Holland and

Patzkowsky (2009) provided support for the Arctic Invasion hypothesis in their analysis of the Bighorn Dolomite in north-central Wyoming. The Steamboat Point Member, correlative to the C2 sequence of the Cincinnati Basin, includes genera that participated in the Richmondian Invasion (ex., Streptelasma, Grewingkia, Foerstephyllum,

Calopoecia, Rhynchotrema). The fauna involved with the Arctic Invasion hypothesis is associated with tropical carbonate platforms in warm shallow Laurentian seas (Table 4;

Fig. 11) (Flower, 1946). Jin (2001) questioned the validity of the Artic Hypothesis. He noted that although genera that include tropical species are listed among the Richmondian 98 invaders, none of the large hyper-calcified species typical of the paleoequatorial seas occur in Richmondian strata of the Cincinnati region. Thus it is more likely that extra- tropical species of invader genera were the source of the invasive taxa. 99

Table 4. Current hypotheses as to the origins of the Richmondian invaders.

Hypothesis Supporting Authors Source Regions Taxa Studied Arctic Pathway Flower, 1946 Colorado, Wyoming, Manitoba, Hudson Bay, Nunavut, Cincinnatian cephalopods Frey, 1981 Manitoba Narthecoceras Frey, 1985 Iowa, Utah, Wyoming Schuchertoceras Holland & Patzkowsky, Wyoming Streptelasm, Grewingkia, 2009 Calopoecia, Foerstephyllum, Rhynchotrema

Northeastern Canada Roy, 1941 Baffin Island, Nunavut Streptelasma Flower, 1946 Greenland, Nunavut Cincinnatian cephalopods Elias, 1982; 1983 Ontario, Quebec Grewingkia, Streptelasma Frey, 1985 Quebec, Ontario Schuchertoceras

Marginal Basins Jin, 1999 Cool, deep water marginal basins surrounding Laurentia Rhynchonelliform brachiopods

Baltic Pathway Frye, 1982 Sweden, Norway Schuchertoceras Frey, 1985 Sweden, Norway Probillingsites, Schuchertoceras Anstey, 1986 Baltica Bryozoa

Multidirectional Wright & Stigall, 2013a Anticosti Island, Western Midcontinent Glyptorthis, Hebertella, Pathways Plaesiomys Bauer & Stigall, 2014 Western Midcontinent, Scoto-Appalachia Eochonetes Lam & Stigall, 2015; Anticosti Island, Baltica, Scoto-Appalachia, Western 63 genera of Richmondian Chapters 2 & 3 Midcontinent Invaders 100

Northeastern Canada (Quebec, Ontario)

In Foerste‘s 1912 analysis of the Arnheim Formation, he alludes to the origins of

Retrorsirostra carleyi by stating that the brachiopod species must have originated from areas to the north of the Cincinnati Basin. This is because the species R. carleyi attains maximum shell size and abundance within southwestern Ohio compared to its sporadic appearance in southeastern Indiana, western Kentucky, and southwestern Tennessee. In his 1924 publication on Late Ordovician invertebrates from Ontario and Quebec, Foerste notes that the faunas of Ontario have a similar affinity with those of the Cincinnati region, and interprets many of the Waynesville Formation (=second invasion pulse) species as having a northwestern origin and corals having a northern origin.

The study and distribution of rugose corals has strengthened the hypothesis that several of the invasive taxa had a northeastern Canadian origin (Fig. 11). In a monographic revision of Ordovician fossils of Frobisher Bay, Baffin Island, Roy (1941) noted the occurrence of three species of Streptelasma that are also present in Richmondian strata of the Cincinnati region. Forty years later, Elias (1982) studied solitary rugose corals, and proposed a pathway used by solitary corals that extended from southern Ontario and

Quebec into Michigan and southward into the Cincinnati Basin and Nashville Dome region of Tennessee, thus validating Foerste‘s earlier hypothesis as to the origin of the

101

Figure 11. Competing hypotheses as to the geographic origin and dispersal pathways of species participating in the Richmondian Invasion. The multidirectional pathways hypothesis proposed by Wright & Stigall (2013b) includes the Arctic invasion, northeastern Canada, and marginal basins hypotheses, and was expanded to include the Baltic basin by Lam and Stigall (2015). The dashed line from Avalonia indicates that this may be a source region based on analyses of Chapter 3, requiring further studies. Figure modified after Cocks and Torsvik (2005), Cocks and Torsvik (2011).

102 corals. Elias named this path the Richmond Solitary Coral Province, and proposed dispersal along the route was restricted by the Queenston Delta and Taconic uplands to the south, and by the positive structure of the Canadian Shield to the north. Taking Roy‘s

(1941) observations into consideration, the coral province described by Elias could have extended further north into Nunavut. Elias found further support for his proposed

Richmond Solitary Coral Province by noting occurrences of two invasive solitary rugose corals, Grewingkia canadensis and Streptelasma divaricans, in the Cincinnati Basin

Richmondian strata. These two species attained their widest geographic extent within the

C5 sequence, and are not found outside of the Richmond Solitary Coral Province.

The occurrence of components of the Arctic Fauna within the C5 sequence was noted by Frey (1985), as he described an ascocerid cephalopod, Schuchertoceras obscurum, within the Waynesville Formation. The genus is also noted to occur within carbonate strata of Richmondian age in Quebec and Anticosti Island, indicating fauna associated with the Arctic Invasion hypothesis and fauna from northeastern Canada had some degree of area overlap.

Marginal Laurentian Basins

Another subset of the invasive taxa has been proposed to have been sourced from cooler, deeper water basins that bordered Laurentia on its southern and eastern margins

(Fig. 11). Specifically, the introduction of the rhynchonellid brachiopods, such as

Hiscobeccus and Lepidocyclus, during the Richmondian Invasion could be a regional 103 expression of the continental-scale Hiscobeccus expansion. Jin (2001) hypothesized that the Hiscobeccus fauna originated during the Trentonian Stage from smaller, thinner- shelled components of the Scoto-Appalachian fauna that inhabited deeper, cooler marginal basins surrounding Laurentia (Jin, 1999). From there, the fauna dispersed into the warm, shallow intercratonic carbonate seas of Laurentia. During the early

Cincinnatian Maysvillian Stage, the brachiopods colonized the paleotropics and subsequently attained maximum diversity and shell sizes (gigantism) compared to their pericratonic counterparts (Jin, 2001; Rasmussen, 2013). Because of his observations, Jin

(1999) proposed the majority of the brachiopod fauna originated from marginal basins surrounding Laurentia, here termed the Marginal Basin hypothesis (Fig. 11). Components of the Richmondian invasive brachiopod genera, namely Hiscobeccus and Lepidocyclus, are key components of the Hiscobeccus fauna, which evolved from forms common to the

Scoto-Appalachian fauna. From several lines of evidence (for a full discussion, see Jin,

2001), this fauna originated in deep-water marginal basins, and did not successfully invade into epicontinental basins within Laurentia until the Richmondian Age, coincident with the appearance of these taxa in the Cincinnati region (Jin, 2001).

Baltic Pathways

The occurrence of Richmondian invaders in correlative or younger strata in the

Baltic region has been hypothesized in the literature since the 1980‘s. In 1982, Frey noted the occurrence of Schuchertoceras species within the Boda Limestone of Sweden and 104

Norway. Two additional ascocerid genera, Probillingsites and Billingsites, have also been described from Late Ordovician rocks in western North America into Greenland and

Baltica (Flower, 1946; Frey, 1985). These three cephalopod genera likely indicate a

Baltica exchange of taxa during the Richmondian Stage.

Further evidence for a Baltic origin for some of the Richmondian Invaders came from Anstey‘s (1986) analysis of Late Ordovician bryozoan biomes. Many of the bryozoan genera that occur in Anstey‘s Reedsville-Lorraine Province (here termed the

Northern Appalachian Basin) or the Red River-Stony Mountain Province (Anticosti

Island, Quebec, western North America) and invade into the Cincinnati Basin during the

C4 and C5 sequences were found to have affinities to Baltic taxa. In fact, Anstey (1986) found that Baltica and Laurentia exclusively share 24 bryozoan genera, compared to just

5 genera shared between Laurentia and Canada or Siberia, indicating active dispersal pathways between Laurentia and Baltica for Late Ordovician bryozoans, with pathways that likely extended through the southern marginal basins of Laurentia.

Multidirectional Dispersal Paths

Recent quantitative paleobiogeographic analyses (e.g., Wright & Stigall, 2013a;

Wright & Stigall, 2014; Bauer & Stigall, 2014; Lam & Stigall, 2015; Chapters 2 and 3) indicate that multiple biogeographic areas of origin and several multidirectional pathways were likely used by invasive taxa. To test the competing hypotheses on the geographic origin of the Richmondian invaders, Wright & Stigall (2013b; 2014) constructed 105 phylogenetic hypotheses for two brachiopod genera that colonized the Cincinnati region during the second pulse of the Richmondian Invasion: the orthide brachiopods

Glyptorthis Foerste, 1914 and Plaesiomys Hall and Clark, 1892. Using Fitch parsimony and Lieberman-modified Brooks Parsimony analysis (LBPA), Wright & Stigall (2013a) determined that Glyptorthis insculpta dispersed in the Cincinnati region from the Western

Midcontinent whereas Plaesiomys subquadratus was sourced from the marginal Anticosti

Island basin, thereby supporting both the Arctic Invasion and marginal basin hypotheses.

Bauer (2014) conducted a similar study for the sowerbyellid brachiopod Eochonetes

Reed 1917. Her results indicated that Eochonetes clarksvillensis migrated into the

Cincinnati Basin during the Richmondian C5 sequence from areas west of the

Transcontinental Arch, consistent with the Arctic Invasion hypothesis. These analyses combined with results of Swisher (2009) provide strong support for multiple areas of origination and for multidirectional dispersal pathways from areas north of the

Transcontinental Arch and the Western Midcontinent (corresponding to the Arctic

Invasion hypothesis) as well as from the marginal Scoto-Appalachian basin.

Using Parsimony Analysis of Endemicity, Lam and Stigall (2015; Chapter 2) compiled occurrence data for over sixty Richmondian invader genera across the C1 through C5 sequences to specifically test for dispersal routes into and within Laurentia.

Because phylogenetic hypotheses are not available for the majority of the taxa that participated in the Richmondian Invasion, a presence-absence based approach weighted toward shared occurrences can provide powerful new insights. Lam and Stigall (2015;

Chapter 2) identified several dispersal events between Baltica and Laurentia during the 106

C1,C2, C3, C4, and C5 sequences via the marginal Scoto-Appalachian basin and western basins, thus providing additional support for the multidirectional pathway hypothesis. In a subsequent study, Lam (Chapter 3) conducted phylogenetic biogeographic analyses with both parsimony and maximum likelihood frameworks to examine dispersal routes for ten Middle through Late Ordovician phylogenetic brachiopod and trilobite clades across several major paleoclimatic shifts. Lam (Chapter 3) recovered support for shared trilobite taxa between Baltica and Laurentia and dispersal from the Western Midcontinent into the Cincinnati region during the Katian Age from both maximum likelihood and parsimony analyses. Dispersal was hypothesized to occur from Avalonia either directly into Laurentian basins or into Baltica with subsequent dispersal (Fig. 11). Although the trilobite clades used in their study did not participate in the Richmondian Invasion, the pathways hypothesized are likely to elucidate paths used by taxa of similar larval states.

By combining several methods (field studies, quantitative paleobiogeography, and phylogenetic biogeography), the pathways and source regions of taxa that participated in the Richmondian Invasion have been expanded upon and constrained throughout the previous century. Taxa were sourced from both northern and western North America as well as marginal basins. Baltica has been proposed as a source region to some of the taxa that participated in the invasion, but rigorous phylogenetic hypotheses have yet to be conducted on cephalopods, bivalves, gastropods, and rugose corals that invaded into the

Cincinnati Basin. As noted by Patzkowsky and Holland (1996), bryozoan genera that invaded into the basin have yet to be studied, and will likely shed light on the relationship and pathways that operated during the late Katian (Anstey, 1986). 107

Invasion dynamics

Tectonic and Paleoclimatic Drivers of the Invasion

Tectonic, oceanographic, and paleoclimatic factors have long been invoked to explain the sudden appearance of the Richmondian invaders. In an early paper, Foerste

(1924) hypothesized that the recurrence of the Black River fauna in Richmondian strata was due to tilting of the North America continent to the north during the Mohawkian and

Richmondian, allowing the fauna to invade back into epicontinental basins. Although today this idea seems rather absurd, Foerste‘s hypothesis was innovative for the time before the development of plate tectonic theory. Additionally, Foerste (1924) proposed that rising sea level flooded the craton as far south as Tennessee by way of the Hudson

Bay, facilitating the immigration of Canadian taxa into eastern North America.

More recently, Patzkowsky and Holland (1996) proposed that changes in depositional environment and oceanographic conditions related to the progression of the

Taconic Orogeny produced the regional extirpation and subsequently facilitated the

Richmondian Invasion.

The Taconic tectophase of the Taconic orogeny began during the late Sandbian to

Katian stages, when the locus of orogenic activity shifted east towards the New York promontory (Shanmugam & Lash, 1982; Ettensohn, 2010). Following a major transgression at the M4/M5 (mid Mohawkian) sequence boundary, tropical carbonate deposition ceased along the southern margin of Laurentia and to the east in Ontario 108

(Brookfield, 1988; Patzkowsky & Holland, 1996). Carbonate platforms were replaced by a mixed carbonate-siliciclastic system, in which heterozoan carbonates dominated.

Patzkowsky & Holland (1996) attributed this significant facies change to upwelling of nutrient-rich waters into the foreland basin of southern Laurentia. During the late

Mohawkian in the mid M5 sequence, the Guttenberg carbon isotope excursion (GICE) records global cooling due to a brief episode of pre-Hirnantian cooling (Saltzman &

Young, 2005). Holland (1996) suggested that the cooling due to an influx of cooler waters, along with the deposition of clastic sediment, caused the regional extirpation of the Arctic Fauna to stable carbonate platforms in western North America and Canada.

The Richmondian Stage strata in the southern Laurentian basins record a shift back to tropical carbonate deposition at the C3/C4 sequence boundary that lasted until the

Hirnantian Stage (Holland, 2008). The return of photozoan carbonates has been attributed to reduced upwelling due to infilling of the Sebree Trough, which was the previous conduit through which cool, oxygen-poor and phosphate-rich waters entered the epicontinental seas of Laurentia from the Iapetus Ocean (Mitchell & Bergström, 1991;

Kolata et al., 2001). Frey (1985) also attributed the Richmondian Invasion to warmer conditions and decreased input of clastic sediment from the Taconic highlands, but he suggested this was due to the equatorial movement of Laurentia during the Late

Ordovician. Alternatively, the return to tropical carbonate deposition may have been caused by the short-lived, globally recognized warming trend known as the Boda Event

(Fortey & Cocks, 2005). Regardless, the first phase of the Richmondian Invasion occurs during the switch back to tropical carbonate deposition within the Cincinnati Basin. 109

The second phase of the invasion at the C5 sequence boundary coincides with another major transgression. Foerste (1982) attributed the occurrence of solitary rugose corals in the Cincinnati Basin to this transgression. Holland (1996) also attributed the invasion to the transgression and exclusion of cool waters from the continental interior, as barriers that had once kept the Arctic Fauna away from the Cincinnati Basin could have become breached. In addition, he proposed that the positive structure of the Canadian

Shield and the deep-water Maquoketa belt, as well as areas of cooler waters, acted as physical and thermal barriers to dispersing taxa, respectively (Copper & Grawbarger,

1978; Elias, 1983).

Wright and Stigall‘s (2013a) analysis of the tectonic and paleoclimatic drivers of the Richmondian Invasion largely agree with those of Holland (1996) and Patzkowsky and Holland (1996). They determined that the return to tropical conditions on the southern margin of Laurentia facilitated an influx of low diversity, warm water fauna that had a high potential for dispersal due to the high degree of speciation by dispersal processes during the Richmondian Invasion interval. Similarly, Bauer and Stigall (2014) proposed that coastal upwelling along the margin of the Western Midcontinent region of

Laurentia helped to facilitate dispersal of invasive brachiopod taxa into the Cincinnati

Basin.

110

Biological Facilitators of the Invasion

As discussed above, there have been substantial contributions documenting the faunal movements between southern, eastern, and western midcontinent basins within

Laurentia throughout the Late Ordovician. Tectonic and paleoclimatic factors have been discussed as causes of the invasion interval, but few previous analyses have considered how biological attributes of the invader taxa facilitated migration between basins.

Additionally, many of the proposed drivers of the Richmondian Invasion fail to include taxa that migrated across the Iapetus Ocean from Baltica.

Larval States of the Richmondian Invaders

To determine how the Richmondian invaders were able to disperse between basins within Laurentia and among paleocontinents, Lam and Stigall (2015, Chapter 2) characterized larval states for crinoids, brachiopods, and corals that participated in the

Richmondian Invasion. They found that 96.4% of invasive taxa developed via either planktic planktotrophic or planula larvae, both of which are self-feeding larval types with the propensity for long-distance dispersal. Only 3.6% of invasive taxa were found to have lecithotrophic larval stages in which larvae feed from a yolky egg and spend a considerably shorter amount of time in a larval state. Thus the taxa that participated in the Richmondian Invasion represent only a subset of taxa present within the seas. Larval 111 type exerted a strong selection bias on which taxa were able to disperse and successfully colonize the Cincinnati region as part of the Richmondian Invasion.

Oceanographic Drivers of Long-Distance Dispersal

Several authors have hypothesized that taxa migrated between Laurentia and

Baltica before the Richmondian Stage, and some have alluded that the Baltic Basin may have been a paleogeographic source region for some of the invasive taxa (e.g., Frye,

1982; Anstey, 1986). Lam and Stigall (2015; Chapters 2 and 3) compared their hypothesized pathways among Laurentian basins and between Laurentia and Baltica to

Late Ordovician oceanographic reconstructions by Hermann et al. (2004). During the

Late Ordovician, the equatorial Iapetus current flowed around Laurentia to the west, splitting around the continent and converging off the western coast. As Laurentia was a continent that was rich in volcanic island arcs and marginal basins, Lam and Stigall

(2015) proposed taxa could easily disperse from Baltica and marginal basins via the

Iapetus current to areas of sustainability, a path they termed ―island hopping‖, given their larval state and ability for long-distance dispersal. Once into the continent, larval dispersal among basins was facilitated by strong storm activity moving from the northeast to southwest, as evidenced from tempestite beds in the Late Ordovician Cincinnati Basin strata (Lam & Stigall, 2015; Holland & Patzkowsky, 1996). This would have aided in dispersal of organisms from Midwestern areas of the North American craton into the

Cincinnati region. As epicontinental estuarine circulation ceased in the Richmondian 112

Stage from infilling of the Sebree Trough, surface currents developed from the Coriolis

Effect and trade winds would have equated to a northeast to southwest flow pattern across Laurentia, which would have also aided in dispersal from areas further east of the

Cincinnati Basin (Ettensohn, 2010).

Modern studies of larval dispersal have concluded that planktic larvae are able to traverse large distances within one generation. Treml et al. (2008) modeled coral larvae among tropical Pacific islands and found that major ocean currents largely influence dispersal direction. In addition, their models indicated that planula are able to disperse as far as 150 km in one generation. Island hopping, or dispersal along areas in a stepping stone-like fashion, has been proposed as an important mode of dispersal among modern marine invertebrate organisms between areas of habitable reefs (Treml et al., 2008; Wood et al., 2013). To partake in such lengthy dispersal patterns among suitable areas for growth and reproduction, planktic larvae are most impacted by ocean currents (Pineda et al., 2007). Therefore, the pathways proposed herein by various authors are reasonable given the planktic nature of the majority of the Richmondian invaders (Lam & Stigall,

2015).

Conclusions

The Richmondian Invasion was a significant mass faunal migration event into the

Cincinnati Basin beginning in the C4 depositional sequence. Hypotheses regarding the origin and nature of the invasion can be found in the literature dating back to the early 113

1900‘s, and several hypotheses have since been brought forth to explain the origin and pathways of the invasive organisms. The taxa were originally thought to be recurrent forms of the Black River, or Arctic Fauna, found in Middle and early Late Ordovician strata of northern Canada, Wyoming, Colorado, South Dakota, Texas, Iowa, and

Minnesota, migrating out of the Cincinnati Basin as cooler waters and shifting habitats forced them to retreat to areas of tropical carbonate deposition. However, analyses of the geographic and stratigraphic distribution of the invasive taxa from literature and modern aggregate databases demonstrate that only a minor percentage (24%) of the Richmondian

Invaders were recurrent. Thus the Richmondian Invasion represents the influx and assembly of a novel community structure. The most frequently cited hypothesis about the geographic origin of the invasive fauna is the Arctic Invasion hypothesis, which purports that taxa migrated from the western and northern North America (equatorial) regions into the Cincinnati Basin. Other authors examining cephalopods or brachiopods have suggested that components of the Richmondian Invasion came from northeastern

Canada (Quebec, Ontario) or migrated out of marginal cool-water basins surrounding

Laurentia, respectively. Other authors have advocated a biogeographic connection between the Cincinnati Basin and the Baltic region. Recent studies have evaluated each of these hypotheses via phylogenetic biogeography, parsimony methods, and maximum likelihood analyses. These findings led to the multidirectional pathways hypothesis of

Wright & Stigall (2013a), stating that several pathways were operational between Baltica and Laurentia. In addition, several invaders were sourced from marginal basins (Anticosti 114

Island and Scoto-Appalachia) and Baltica as well as from western, northern, and eastern

North America.

The influx of over sixty new genera into the Cincinnati Basin coincides with a return to tropical carbonate deposition on the southern margin of Laurentia either due to a global warming event or the retreat of cool oceanic waters from the continental interior.

A massive transgression, paired with a warming trend, would have likely facilitated dispersal over barriers that could not previously be breached. In addition, upwelling within the western midcontinent region could have influenced dispersal of invasive brachiopod species into eastern North America. Dispersal from Baltica and marginal basins was facilitated by the equatorial Iapetus current which flowed from Baltica to

Laurentia. Dispersal was further aided by volcanic island arcs acting as areas of sustainability allowing organisms to island hop between paleocontinents. This is feasible given that the majority of the invasive taxa were found to have planktotrophic planktic larval states with a propensity for long-distance dispersal within one generation.

Although several source basins have been identified, many hypothesized pathways have yet to be tested through more rigorous analyses. Additional species-level phylogenetic hypotheses of taxa that participated in the invasion will shed light on other source basins, and will aid in the interpretation of the effects of paleoclimatic and tectonic factors on organismal dispersal events. The integration of multiple data streams has already led to a holistic analysis of the invasion dynamics associated with the

Richmondian Invasion. Future analyses will likely uncover additional patterns and help comprehend complex processes associated with organism invasion events. Fundamental 115 understanding of deep-time invasion events and drivers of dispersal is crucial for understanding and predicting the behaviors and patterns of modern-day invasive species.

116

CHAPTER 5: CONCLUSIONS

Parsimony and maximum likelihood analyses are both valuable methods for determining paleobiogeographic area relationships, dispersal pathways, and speciation patterns across major tectonic and paleoclimatic shifts within the Middle through Late

Ordovician. Because no phylogenetic hypotheses are available for many taxa involved in the Richmondian Invasion, a combination of traditional and novel methods provide the most robust insights for discerning ancient areas of origination and speciation patterns across tectonic and paleoclimatic shifts.

In Chapter 2, Parsimony Analysis of Endemicity (PAE) was employed across the

C1 to C5 sequences in the Cincinnati Basin using presence/absence matrices of over 60 genera of taxa that participated in the Richmondian Invasion. Throughout all five time slices, the Western Midcontinent and North of the Transcontinental Arch shared high similarity values, indicating substantial faunal exchange between these regions. Pathways were also hypothesized to have aided in faunal dispersal into the Cincinnati Basin from the Nashville Dome (C2/C3 sequences), Scoto-Appalachia (C3/C4 sequences), and the

Upper Mississippi Valley and North of the Transcontinental Arch (C4/C5 sequences). In addition, three separate dispersal events were identified from Baltica into Laurentia through marginal and western basins. The larval states of the Richmondian invaders were characterized, and the majority of the taxa with identifiable larval stages had planktic planktotrophic or planula self-feeding states, with a high propensity for long-distance dispersal. 117

In Chapter 3, results of ancestral range and speciation mode analyses for ten brachiopod and trilobites clades conducted using Fitch parsimony and Lieberman- modified Brooks Parsimony analysis (LBPA) were compared to the results obtained from the maximum-likelihood R program BioGeoBEARS. Large-scale patterns of area relationships were found to be mostly congruent, with dispersal direction differing within some clades. Within BioGeoBEARS, biogeographic evolution of all ten clades were found to fit the ―+J‖ models best, indicating that long-distance founder event speciation was important to Paleozoic organisms. LBPA was mostly congruent between vicariance and geodispersal cladograms within each time slice, indicating that cyclical processes were significant in structuring the area relationships within Laurentian basins. When compared to paleoclimatic and oceanographic reconstructions for the Late Ordovician, pathways were found to be influenced by a variety of paleoclimatic factors as well as regional and global currents. However, it was ultimately determined that both methods have significant value for use in paleobiogeographic reconstructions, but more research should be conducted within this area.

The current state of knowledge about the Richmondian Invasion was summarized in Chapter 4, including past hypotheses from the last century about the origin and pathways used by invasive taxa. The analyses here support the multidirectional dispersal pathway hypothesis of Wright and Stigall (2013a), in which they concluded from their analyses of orthide brachiopods that Richmondian invaders were sourced from multiple geographic areas via several independent dispersal routes. PAE results support a strong relationship between the Western Midcontinent region of North America, as well as 118 dispersal pathways from Baltica and the marginal basins of Laurentia. This is congruent with the Arctic Invasion hypothesis and the marginal basins hypothesis of Jin (1999).

Analyses conducted in Chapter 4 provide additional support for multidirectional pathways that were operational before, during, and likely after the Richmondian Invasion.

The timing and nature of the Richmondian Invasion was controlled by biotic factors— such as larval type and ecological niche breadth—and earth system factors—such as cessation of upwell in southern Laurentia, shift in locus of the Taconic orogeny, and transgressive events.

119

REFERENCES

Alberstadt, L. P. (1979) The brachiopod genus Platystrophia. U. S. Geological Survey Professional Paper No. 1066-B.

Algeo, T. J., and Seslavinsky, K. B. (1995) The Paleozoic world: continental flooding, hypsometry, and sea-level. American Journal of Science, 295, 787-822.

Amati, L., and Westrop, S. R. (2004) A systematic revision of Thaleops (Trilobita: Illaenidae) with new species from the Middle and Late Ordovician of Oklahoma and New York. Journal of Systematic Palaeontology, 2 (3), 207-256.

Amsden, T. W. (1974) Late Ordovician and Early Silurian articulate brachiopods from Oklahoma, southwestern Illinois, and eastern Missouri. Oklahoma Geological Survey Bulletin, 119, 154 p.

Amsden, T. W. (1986). Paleoenvironment of the Keel-Edgewood oolitic province and the Hirnantian strata of Europe, USSR, and China. In Amsden, T. W., and Barrick, J. E. (eds.) Late Ordovician-Early Silurian Strata in the Central United States and the Hirnantian Stage, Oklahoma Geological Survey Bulletin 139, 1-55.

Anstey, R. L. (1986) Bryozoan provinces and patterns of generic evolution and extinction in the Late Ordovician of North America. Lethaia, 19, 33-51.

Ausich, W. I. (1998) Early phylogeny and subclass division of the Crinoidea (Phylum Echinodermata). Journal of Paleontology, 72 (3), 499-510.

Bapst, D. W. (2012) paleotree: an R package for paleontological and phylogenetic analyses of evolution. Methods in Ecology and Evolution, 3 (5), 803-807.

Barrick, J. E. (1986) Conodont faunas of the Keel and Cason formations. In Amsden, T. W., and Barrick, J. E. (eds.) Late Ordovician-Early Silurian Strata in the Central United States and the Hirnantian Stage, Oklahoma Geological Survey Bulletin 139, 57-68.

Bauer, J. E. (2014). A phylogenetic and paleobiogeographic analysis of the Ordovician brachiopod Eochonetes. M. S. Thesis, Ohio University, Athens, 154 pp.

Bauer, J. E., & Stigall, A. L. (2014) Phylogenetic paleobiogeography of Late Ordovician Laurentian brachiopods. Estonian Journal of Earth Sciences, 63, 189-194.

120

Berggren, W. A., and Hollister, C. D. (1977) Plate tectonics and paleocirculation- Commotion in the ocean. Tectonophysics, 38, 11-48.

Bergström, S. M., Saltzman, M. M., and Schmitz, B. (2006) First record of Hirnantian (Upper Ordovician) δ13C excursion in the North American Midcontinent and its regional implications. Geology Magazine, 143 (5), 657-678.

Bergström, S. M., Young, S., and Schmitz, B. (2010) Katian (Upper Ordovician) δ13C chemostratigraphy and sequence stratigraphy in the United States and Baltoscandia: A regional comparison. Palaeogeography, Palaeoclimatology, Palaeoecology, 296, 217-234.

Bergström, S. M., Saltzman, M. R., Leslie, S. A., Ferretti, A., and Young, S. A. (2015) Trans-Atlantic application of the Baltic Middle and Upper Ordovician carbon isotope zonation. Estonian Journal of Earth Sciences, 64 (1), 8-12.

Bradley, D. C. (1989) Taconic plate kinematics as revealed by foredeep stratigraphy, Appalachian orogen. Tectonics, 8 (5), 1037-1049.

Brame, H.-M., and Stigall, A. L. (2014) Controls on niche stability in geologic time: congruent responses to biotic and abiotic environmental changes among Cincinnatian (Late Ordovician) marine invertebrates. Paleobiology, 40 (1), 70-90.

Brookfield, M. E. (1988) A mid-Ordovician temperate carbonate shelf- The Black River and Trenton Limestone Groups of southern Ontario, Canada. Sedimentary Geology, 60, 137-154.

Brooks, D. R., and McLennan, D. A. (2002) The Nature of Diversity. Chicago Press, Chicago, Illinois, 668 p.

Buckner, J. C., Alfaro, J. L., Rylands, A. B., and Alfaro, M. E. (2014) Biogeography of the marmosets and tamarins (Callitrichidae). Molecular Phylogenetics and Evolution, 82, 413-425.

Carlucci, J. R., Westrop, S. R., and Amati, L. (2010) Tetralichine trilobites from the Upper Ordovician of Oklahoma and Virginia and phylogenetic systematics of the Tetralichini. Journal of Paleontology, 84 (6), 1099-1120.

Carlucci, J. R., Westrop, S. R., Amati, L., Adrain, J. M., and Swisher, J. E. (2012) A systematic revision of the Upper Ordovician trilobite genus Bumastoides (Illaenidea), with new species from Oklahoma, Virginia, and Missouri. Journal of Systematic Palaeontology, 10 (4), 679-723.

121

Carrera, M. G., and Rigby, J. K. (1999) Biogeography of Ordovician sponges. Journal of Paleontology, 73, 26-37.

Chatterton, B. D. E., and Speyer, S. E. (1989) Larval ecology, life history strategies, and patterns of extinction and survivorship among Ordovician trilobites. Paleobiology, 15, 118-132.

Cherns, L., and Wheeley, J. R. (2007) A pre-Hirnantian (Late Ordovician) interval of global cooling- The Boda event re-assessed. Palaeogeography, Palaeoclimatology, Palaeoecology, 251 (3), 449-460.

Cocks, L. R. M., and Torsvik, T. H. (2005) Baltica from the late Precambrian to mid- Palaeozoic times: The gain and loss of a terrane‘s identity. Earth-Science Reviews, 72, 39-66.

Cocks, L. R. M., and Torsvik, T. H. (2011) The Palaeozoic geography of Laurentia and western Laurussia: A stable craton with mobile margins. Earth-Science Reviews, 106, 1-51.

Congreve, C. R., and Lieberman, B. S. (2008) Phylogenetic and biogeographic analysis of Ordovician Homalonotid trilobites. The Open Paleontology Journal, 1, 24-32.

Congreve, C. R., and Lieberman, B. S. (2010) Phylogenetic and biogeographic analysis of deiphonine trilobites. Journal of Paleontology, 84 (1), 128-136.

Copper, P., and Grawbarger, D. J. (1978) Paleoecological succession leading to a Late Ordovician biostrome on Manitoulin Island, Ontario. Canadian Journal of Earth Sciences, 15 (12), 1987-2005.

Costa, W. J. E. M. (2010) Historical biogeography of cynolebiasine annual killifishes inferred from dispersal-vicariance analysis. Journal of Biogeography, 37, 1995- 2004.

Cowie, R. H., and Holland, B. S. (2006) Dispersal is fundamental to biogeography and the evolution of biodiversity on oceanic islands. Journal of Biogeography, 33, 193-198.

Cracraft, J. (1985) Biological diversification and its causes. Missouri Botanical Garden Annals, 72, 794-822.

Cressman, E. R., and Noger, M. C. (1976) Tidal-flat carbonate environments in the High Bridge Group (Middle Ordovician) of central Kentucky. Kentucky Geological Survey Report of Investigations, 18, 1-15.

122

Dalvė, E. (1948) The fossil fauna of the Ordovician in the Cincinnati region. University of Cincinnati, 56 p.

Dudei, N. L., and Stigall, A. L. (2010) Using ecological niche modeling to assess biogeographic and niche response of brachiopod species to the Richmondian Invasion (Late Ordovician) in the Cincinnati Arch. Palaeogeography, Palaeoclimatology, Palaeoecology, 296, 28-43.

Edgecombe, G. D. (1992) Trilobites and the Cambrian-Ordovician ―event‖: cladistics reappraisal. In Novacek, M. J., and Wheller, Q. D. (eds.) Extinction and Phylogeny, Columbian University Press, New York, 144-177.

Eldredge, N., Thompson, J. N., Brakefield, P. M., Gavrilets, S., Jablonski, D., Jackson, J. B. C., Lenski, R. E., Lieberman, B. S., McPeek, M. A., and Miller, M. (2005) The dynamics of evolutionary stasis. Paleobiology, 31 (2), 133-145.

Elias, R. J. (1982) Latest Ordovician Solitary Rugose Corals of Eastern North America. Bulletins of American Paleontology, 81 (314), 5-109.

Elias, R. J. (1983) Middle and Late Ordovician solitary rugose corals of the Cincinnati Arch region. U.S. Geological Survey Professional Paper 1066-N, N1-N13.

Elias, R. J. (1985) Solitary rugose corals of the Upper Ordovician Montoya Group, southern New Mexico and westernmost Texas. Paleontological Society Memoir, 16, 1-68.

Elias, R. J., and Young, G. A. (1992) Biostratigraphy and biogeographic affinities of latest Ordovician to earliest Silurian corals in the east-central United States. In Webby, B. D., and Laurie, J. R. (eds.) Global Perspectives on Ordovician Geology, 205-214.

Escalante, T., Rodriguez, G. Cao, N., Ebach, M. C., and Morrone, J. J. (2007) Cladistic biogeography analysis suggests an early Caribbean diversification in Mexico. Naturwissenschaften, 94, 561-565.

Ettensohn, F. R. (1994) Tectonic control on the formation and cyclicity of major Appalachian unconformities and associated stratigraphic sequences. Tectonic and Eustatic Controls on Sedimentary Cycles: SEPM, Concepts in Sedimentology and Paleontology, 4, 217-244.

Ettensohn, F. R. (2010) Origin of Late Ordovician (mid-Mohawkian) temperate-water conditions on southeastern Laurentia: Glacial or tectonic? Geological Society of America Special Paper, 466, 163-175.

123

Ettensohn, F. R., and Brett, C. E. (2002). Stratigraphic evidence for the Appalachian Basin for continuation of the Taconian Orogeny into Early Silurian time. In Mitchell, C. E., and Jacobi, R. (eds), Taconic Convergence: Orogen, Foreland Basin, and Craton, Physics and Chemistry of the Earth, 27, 279-288.

Foerste, A. F. (1905) Notes on the distribution of brachiopoda in the Arnheim and Waynesville beds. The American Geologist, 244-250.

Foerste, A. F. (1912) The Arnheim Formation within the areas traversed by the Cincinnati Geanticline. Ohio Naturalist, 12, 429-456.

Foerste, A. F. (1924) Upper Ordovician Faunas of Ontario and Quebec. Canada Department of Mines, Geological Survey, 138, 99 p.

Folinsbee, K. E., and Brooks, D. R. (2007) Miocene hominoid biogeography: pulses of dispersal and differentiation. Journal of Biogeography, 34, 383-397.

Fortey, R. A., and Cocks, L. R. M. (2005) Late Ordovician global warming-The Boda Event. Geology, 33, 405-408.

Freeman, G., and Lundelius, J. W. (2005) The transition from planktotrophy to lecithotrophy in larvae of Lower Palaeozoic rhynchonelliform brachiopods. Lethaia, 38, 219-254.

Frey, R. C. (1985) A well-preserved specimen of Schuchertoceras (Cephalopoda, Ascocerida) from the Upper Ordovician (basal Richmondian) of southwest Ohio. Paleontological Notes, 59 (6), 1506-1511.

Frye, M. W. (1982) Upper Ordovician (Harjuan) nautiloid cephalopods from the Boda Limestone of Sweden. Journal of Paleontology, 56 (5), 1274-1292.

Ghienne, J.-F., Desrochers, A., Vandenbroucke, T. R., Achab, A., Asselin, E., Dabard, M.-P., Farley, C., Loi, A., Paris, F., Wickson, S., and Veizer, J. (2014) A Cenozoic-style scenario for the end-Ordovician glaciation. Nature Communications, 5, doi:10.1038/ncomms5485.

Gillespie, R. G., Baldwin, B. G., Waters, J. M., Fraser, C. I., Nikula, R., and Roderick, G. K. (2012) Long-distance dispersal: a framework for hypothesis testing. Trends in Ecology & Evolution, 27, 47-56.

Hammer, Ø., Harper, D. A. T., and Ryan, P. D. (2001) PAST: Paleontological statistics software package for education and data analysis. Palaeontologia Electronica, 4, 1-9.

124

Harper, D. A. T., Rasmussen, C. M. Ø., Liljeroth, M., Blodgett, R. B., Candela, Y., Jin, J., Percival, I. G., Rong, J., Villas, E., and Zhan, R. B. (2013) Biodiversity, biogeography and phylogeography of Ordovician rhynchonelliform brachiopods. Geological Society, London, Memoirs, 38 (1), 127-144.

Harris, R. B., Birks, S. M., and Leache, A. D. (2014) Incubator birds: biogeographical origins and evolution of underground nesting in megapodes (Galliformes: Megapodiidae). Journal of Biogeography, 41 (11), 2045-2056.

Hembree, D. I. (2006) Amphisbaenian paleobiogeography: evidence of vicariance and geodispersal patterns. Palaeogeography, Palaeoclimatology, Palaeoecology, 235 (4), 340-354.

Hendricks, J. R., and Lieberman, B. S. (2007) Biogeography and the Cambrian radiation of arachnomorph arthropods. Memoirs of the Association of Australasian Palaeontologists, 34, 461-471.

Herrmann, A. D., Haupt, B. J., Patzkowsky, M. E., Deidov, D., and Slingerland, R. L. (2004) Response of Late Ordovician paleoceanography to changes in sea level, continental drift, and atmospheric pCO2: potential causes for long-term cooling and glaciation. Palaeogeography, Palaeoclimatology, Palaeoecology, 210, 385- 401.

Holland, S. M. (1993) Sequence stratigraphy of a carbonate-clastic ramp: The Cincinnatian Series (Upper Ordovician) in its type area. Geological Society of America Bulletin, 105 (3), 306-322.

Holland, S. M. (1997) Using time/environment analysis to recognize faunal events in the Upper Ordovician of the Cincinnati Arch. Paleontological event horizons: ecological and evolutionary implications, New York: Columbia University Press, p. 309-334.

Holland, S. M. (2008) The type Cincinnatian: An overview. Stratigraphic Renaissance in the Cincinnati Arch: Implications for Upper Ordovician Paleontology and Paleoecology, Cincinnati, Ohio: Cincinnati Museum Center Scientific Contributions 2.

Holland, S. M., and Patzkowsky, M. E. (1996) Sequence stratigraphy and long-term paleoceanographic change in the Middle and Upper Ordovician of the eastern United States. Paleozoic Sequence Stratigraphy: Views from the North American Craton, Geological Society of America Special Paper, 306, 117-129.

125

Holland, S. M., and Patzkowsky, M. E. (2009) The stratigraphic distribution of fossils in a tropical carbonate succession: Ordovician Bighorn Dolomite, Wyoming, USA. Palaios, 24, 303-317.

Holmden, C., Creaser, R. A., and Muehlenbachs, K. (1998) Isotopic evidence for geochemical decoupling between ancient epeiric seas and bordering oceans: Implications for secular curves. Geology, 26, 567-570.

Howe, H. J. (1988) Articulate brachiopods from the Richmondian of Tennessee. Journal of Paleontology, 62, 204-218.

Huff, W. D., Bergström, S. M., and Kolata, D. R. (2010) Ordovician explosive volcanism. The Ordovician Earth System: Geological Society of America Special Paper 466, p. 13-28.

Huff, W. D., Bergström, S. M., and Kolata, D. R. (1992) Gigantic Ordovician volcanic ash fall in North America and Europe: Biological, tetonomagmatic, and event- stratigraphic significance. Geology, 20, 875-878.

Hunda, B. R., and Hughes, N. C. (2007) Evaluating paedomorphic heterochrony in trilobites: the case of the diminutive trilobite Flexicalymene retrorsa minuens from the Cincinnatian Series (Upper Ordovician), Cincinnati region. Evolution & Development, 9 (5), 483-498.

Jablonski, D., and Lutz, R. A. (1983) Larval ecology of marine benthic invertebrates: paleobiological implications. Biological Reviews, 58 (1), 21-89.

Jablonski, D., Sepkoski, J. J., Bottjer, D. J., and Sheehan, P. M. (1983) Onshore-offshore patterns in the evolution of Phanerozoic shelf communities. Science, 222 (4628), 1123-1125.

Jin, J. (1999) Evolution and extinction of the Late Ordovician epicontinental brachiopod fauna of North America. Acta Universitatis Carolinae-Geologica, 43, 203-206.

Jin, J. (2001) Evolution and extinction of the North American Hiscobeccus brachiopod Fauna during the Late Ordovician. Canadian Journal of Earth Sciences, 38, 143- 151.

Jin, J., Harper, D. A. T., Cocks, L. R. M, McCausland, P. J., Rasmussen, C. M. Ø., and Sheehan, P. M. (2013) Precisely locating the Ordovician equator in Laurentia. Geology, 41, 107-110.

126

Jin, J., Sohrabi, A., and Sproat, C. (2013) Late Ordovician brachiopod endemism and faunal gradient along palaeotropical latitudes in Laurentia during a major sea level rise. GFF, 00 (1), 1-5.

Kelly, S. M., and Pope, J. K. (1979) A new camerate crinoid from the Upper Ordovician of Indiana. Journal of Paleontology, 53, 416-420.

Keith, B. D. (1989) Regional facies of the Upper Ordovician Series of eastern North America. In Keith, B. D. (ed.), The Trenton Group (Upper Ordovician series) of Eastern North America, American Association of Petroleum Geologists Studies in Geology, 29, 1-16.

Kolata, D. R., Huff, W. D., and Bergström, S. M. (2001) The Ordovician Sebree Trough: an oceanic passage to the Midcontinent United States. Geological Society of America Bulletin, 113 (8), 1067-1078.

Krause, D. W., Prasad, G. V. R., von Koenigswald, W., Sahni, A., and Grine, F. E. (1997) Cosmopolitanism among Gondwanan late Cretaceous mammals. Nature, 390, 504-507.

Ladiges, P. Y., Humphries, C. J., and Brooker, M. I. H. (1987) Cladistic and biogeographic analysis of Western Australian species of Eucalyptus L‘Herit, informal subgenus Monocalyptus Pryor & Johnson. Australian Journal of Botany, 35 (3), 251-281.

Lam, A. R., and Stigall, A. L. (2014) Pathways and mechanisms of Late Ordovician (Katian) faunal migrations of Laurentia and Baltica. Estonian Journal of Earth Sciences, 64, 62-67.

Landis, M. J., Matzke, N. J., Moore, B. R., and Huelsenbeck, J. P. (2013) Bayesian analysis of biogeography when the number of areas is large. Systematic Biology, 62, 789-804.

Lieberman, B. S. (2000) Paleobiogeography: Using Fossils to Study Global Change, Plate Tectonics, and Evolution. New York: Kluwer Academic Press/Plenum Publishing, 208 pp.

Lieberman, B. S. (2001) Phylogenetic analysis of the Olenellina Walcott, 1890 (Trilobita, Cambrian). Journal of Paleontology, 75, 96-115.

Lieberman, B. S. (2003a) Biogeography of the Trilobita during the Cambrian radiation: deducing geological processes from trilobite evolution. Special Papers in Palaeontology, 70, 59-72.

127

Lieberman, B. S. (2003b) Paleobiogeography: the relevance of fossils to biogeography. Annual Review of Ecology, Evolution, and Systematics, 34, 51-69.

Lieberman, B. S. (2003c) Taking the pulse of the Cambrian radiation. Integrative and Comparative Biology, 43 (1), 229-237.

Lieberman, B. S., Edgecombe, G. D., and Eldredge, N. (1991) Systematics and biogeography of the ‗Malvinella group‘, Calmoniidae (Trilobita, Devonian). Journal of Paleontology, 65 (5), 824-843.

Lieberman, B. S., and Eldredge, N. (1996) Trilobite biogeography in the Middle Devonian: geological processes and analytical methods. Paleobiology, 22, 66-79.

Litsios, G., Pearman, P. B., Lanterbecq, D., Tolou, N., and Salamin, N. (2014) The radiation of the clownfishes has two geographic replicates. Journal of Biogeography, 41 (11), 2140-2149.

Mac Niocaill, C., Van der Pluijm, B. A., and Van der Voo, R. (1997) Ordovician paleogeography and the evolution of the Iapetus ocean. Geology, 25 (2), 159-162.

Macdonald, F. A., Ryan-Davis, J., Coish, R. A., Crowley, J. L., and Karabinos, P. (2014) A newly identified Gondwanan terrane in the northern Appalachian Mountains: Implications for the Taconic orogeny and closure of the Iapetus Ocean. Geology, 42 (6), 539-542.

Maguire, K. C., and Stigall, A. L. (2008) Paleobiogeography of Miocene Equinae of North America: A phylogenetic biogeographic analysis of the relative roles of climate, vicariance, and dispersal. Palaeogeography, Palaeoclimatology, Palaeoecology, 267 (3), 175-184.

Malizia, R. W., and Stigall. A. L. (2011) Niche stability in Late Ordovician articulated brachiopod species before, during, and after the Richmondian Invasion. Palaeogeography, Palaeoclimatology, Palaeoecology, 311, 154-170.

Matzke, N. J. (2013a) BioGeoBEARS: BioGeography with Bayesian (and Likelihood) Evolutionary Analysis in R Scripts. University of California, Berkeley, Berkeley, CA.

Matzke, N. J. (2013b) Probabilistic historical biogeography: new models for founder- event speciation, imperfect detection, and fossils allow improved accuracy and model-testing. Frontiers of Biogeography, 5 (4), 242-248.

128

Matzke, N. J. (2014) Model selection in historical biogeography reveals that founder- event speciation is a crucial process in island clades. Systematic Biology, 63, 951- 970.

McKerrow, W. S., Niocaill, C. M., Ahlberg, P. E., Clayton, G., Cleal, C. J., and Eagar, R. M. C. (2000) The Late Palaeozoic relations between Gondwana and Laurussia. Geological Society, London, Special Publications, 179 (1), 9-20.

Meidla, T., Tinn., O., Salas, M. J., Williams, M., Siveter, D., Vandenbroucke, T. R. A., and Sabbe, K. (2013) Biogeographical patterns of Ordovician ostracods. Geological Society of London Memoirs, 38, 337-354.

Mitchell, C. E., and Bersgström, S. M. (1991) New graptolite and lithostratigraphic evidence from the Cincinnati region, U.S.A., for the definition and correlation of the base of the Cincinnatian Series (Upper Ordovician). In Barnes, C. R., and Williams, S. H. (eds). Advances in Ordovician geology: Geological Survey of Canada Paper, 90 (9), 59-77.

Mitchell, C. E., and Sweet, W. C. (1989) Upper Ordovician conodonts, brachiopods, and chronostratigraphy of the Whittaker Formation, southwestern District of Mackenzie, N.W.T., Canada. Canadian Journal of Earth Sciences, 26, 74-87.

Miller, A. I., and Mao, S. (1995) Association of orogenic activity with the Ordovician radiation of marine life. Geology, 23 (4), 305-308.

Mohibullah, M., Williams, M., Vandenbroucke, T. R. A., Sabbe, K., and Zalasiewicz, J. A. (2012) Marine ostracod provinciality in the Late Ordovician of paleocontinental Laurentia and its environmental and geographic expression. PLOS one, 7 (8), doi: 10.1371/journal.pone.0041682.

Nelson, C. S. (1988) An introductory perspective on non-tropical shelf carbonates. Sedimentary Geology, 60, 3-14.

Nelson, S. J. (1959) Arctic Ordovician fauna: An equatorial assemblage? Journal of the Alberta Society of Petroleum Geologists, 7 (8), 45-48.

Neuman, R. B. (1972) Brachiopods of Early Ordovician volcanic islands. Proceedings of the 24th International Geology Congress, 7, 297-302.

Patzkowsky, M. E., and Holland, S. M. (1996) Extinction, invasion, and sequence stratigraphy: patterns of faunal change in the Middle and Upper Ordovician of the eastern United States. In Witzke, B. J., Ludvigson, G. E., and Day, J. E. (eds.) Paleozoic sequence stratigraphy: views from the North American Craton. Geological Society of America Special Paper 306, 131–142. 129

Patzkowsky, M. E., and Holland, S. M. (2007) Diversity partitioning of a Late Ordovician marine biotic invasion: controls on diversity in regional ecosystems. Paleobiology, 33 (2), 295-309.

Peterson, K. J. (2005) Macroevolutionary interplay between planktic larvae and benthic predators. Geology, 33, 929-932.

Pineda, J., Hare, J. A., and Sponaugle, S. (2007) Larval transport and dispersal in the coastal ocean and consequences for population connectivity. Oceanography, 20 (3), 22-39.

Popov, L. E., Ghobadi Pour, M., Kebria-Ee Zadeh, M.-R., and Shahbeik, S. (2011) The first record of silicified Cambrian (Furongian) rhynchonelliform brachiopods from the Mila Formation, Alborz Range, Iran. Memoirs of the Association of Australasian Palaeontologists, 42, 193-207.

Poussart, P. F., Weaver, A. J., and Barnes, C. R. (1999) Late Ordovician glaciation under high atmospheric CO2: A coupled model analysis. Paleoceanography, 14, 542- 558.

Pyron, A. (2014) Biogeographic analysis reveals ancient continental vicariance and recent oceanic dispersal in amphibians. Systematic Biology, 63 (5), 779-797.

R Core Team. (2014) R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. http://www.R- project.org

Radford, B., Babcock, R., Van Niel, K. and Done, T. (2014) Are cyclones agents for connectivity between reefs? Journal of Biogeography, 41, 1367-1378.

Railsback, L. B. (1990) Influence of changing deep ocean circulation on the Phanerozoic oxygen isotope record. Geochimica Et Cosmochimica Acta, 54 (5), 1501-1509.

Railsback, L. B., Ackerly, S. C., Anderson, T. F., and Cisne, J. L. (1990) Paleontological and isotope evidence for warm saline deep waters in Ordovician oceans. Nature, 343, 156-159.

Rasmussen, C. M. Ø. (2013) Late Ordovician brachiopods from eastern North Greenland: Equatorial offshore migration of the Red River fauna. Palaeontology, 79, 359- 379.

Ree, R. H. (2005). Detecting the historical signature of key innovations using stochastic models of character evolution and cladogenesis. Evolution, 59, 257-265.

130

Rode, A. L., and Lieberman, B. S. (2002) Phylogenetic and biogeographic analysis of Devonian phyllocarid crustaceans. Journal of Paleontology, 76, (2), 269-284.

Ronquist, F. (1997) Dispersal-vicariance analysis: a new approach to the quantification of historical biogeography. Systematic Biology, 46, 195-203.

Rosen, B. R., and Smith, A. B. (1988) Tectonics from fossils? Analysis of reef-coral and sea-urchin distributions from late Cretaceous to Recent, using a new method. Geological Society of London, Special Papers, 37, 275-306.

Roy, S. K. (1941) The Upper Ordovician Fauna of Frobisher Bay, Baffin Island. Chicago Field Museum of Natural History Geology Memoirs, 2, 212pp.

Saltzman, M. R., and Young, S. A. (2005) Long-lived glaciations in the Late Ordovician? Isotopic and sequence-stratigraphic evidence from western Laurentia. Geology, 33 (2), 109-112.

Sanderson, M. J. (2003) r8s: inferring absolute ages of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics, 19 (2), 301- 302.

Sanmartin, I., Enghoff, H., and Ronquist, F. (2001) Patterns of animal dispersal, vicariance, and diversification in the Holarctic. Biological Journal of the Linnean Society, 73, 345-390.

Sereno, P. C., Dutheil, D. B., Iarochene, M., Larsson, H. C. E., Lyon, G. H., Magwene, M., Sidor, C. A., Varricchio, D. J., and Wilson, J. A. (1996) Predatory dinosaurs from the Sahara and Late Cretaceous faunal differentiation. Science, 272, 986- 991.

Servais, T., Harper, D. A. T., Munnecke, A., Owen, A. W., and Sheehan, P. M. (2009). Understanding the Great Ordovician Biodiversification Event (GOBE): Influences of paleogeography, paleoclimate, or paleoecology? GSA Today, 19 (4), 4-10.

Servais, T., Owen, A. W., Harper, D. A. T., Kröger, B., and Munnecke, A. (2010) The Great Ordovician Biodiversification Event (GOBE): the palaeoecological dimension. Palaeogeography, Palaeoclimatology, Palaeoecology, 294, 99-119.

Shanmugam, G., and Lash, G. G. (1982). Analogous tectonic evolution of the Ordovician foredeeps, southern and central Appalachians. Geology, 10 (11), 562-566.

Smith, A. B. (1994) Systematics and the Fossil Record: Documenting Evolutionary Patterns. Wiley Blackwell Science LTD, Oxford, 232 p.

131

Sorenson, L., Allen, G., Erdmann, M., Dai, C.-F., and Liu, S.-Y. (2014) Pleistocene diversification of the Pomacentrus coelestis species complex (Pisces: Pomacentridae): historical biogeography and species boundaries. Marine Biology, 161 (11), 2495-2507.

Stigall, A. L. (2010a) Invasive species and biodiversity crises: testing the link in the Late Devonian. Plos ONE, 5 (12), doi: 10.1371/journal.pone.0015584.

Stigall, A. L. (2010b) Using GIS to assess the biogeographic impact of species invasions on native brachiopods during the Richmondian Invasion in the type-Cincinnatian (Late Ordovician, Cincinnati region). Palaeontologia Electronica, 13 (1), 19 p.

Stigall, A.L. (2012) Using ecological niche modeling to evaluate niche stability in deep time. Journal of Biogeography, 39 (4), 772-781.

Stigall. A. L. (2013) Atlas of Ordovician Life. www.ordovicianatlas.org

Stigall, A. L. (2014) When and how do species achieve niche stability over long time scales? Ecography, 37, 1123-1132.

Stigall Rode, A. L., and Lieberman, B. S. (2005) Paleobiogeographic patterns in the Middle and Late Devonian emphasizing Laurentia. Palaeogeography, Palaeoclimatology, Palaeoecology, 222 (3-4), 272-284.

Swisher, R. E. (2009) Paleobiogeographical and evolutionary analysis of Late Ordovician, C5 sequence brachiopod species, with special reference to Rhynchonellid taxa. M.S. Thesis, Ohio University, Athens, 81-105.

Swofford, D. L. (2002) PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Sunderland, Massachusetts: Sinauer Associates.

Templeton, A. R. (2008) The reality and importance of founder speciation in evolution. BioEssays, 30, 470-479.

Thompson, T. L., and Satterfield, I. R. (1975) Stratigraphy and conodont biostratigraphy of strata contiguous to the Ordovician-Silurian boundary in eastern Missouri. Missouri Geological Survey Report of Investigations No. 57, 61-108.

Torsvik, T. H., and Cocks, L. R. M. (2013) New global palaeogeographical reconstructions for the Early Palaeozoic and their generation. Geological Society, London, Memoirs, 38 (1), 5-24.

132

Treml, E. A., Halpin, P. N., Urban, D. L., and Pratson, L. F. (2008) Modeling population connectivity by ocean currents, a graph-theoretic approach for marine conservation. Landscape Ecology, 23, 19-36.

Trotter, J. A., Williams, I. S., Barnes, C. R., Lecuyer, C., and Nicoll, R. S. (2008) Did cooling oceans trigger Ordovician biodiversification? Evidence from conodont thermometry, Science, 321, 550-554.

Valentine, J. W., and Jablonski, D. (1983) Larval adaptations and patterns of brachiopod diversity in space and time. Evolution, 5, 1052-1061.

Vandenbroucke, T. R. A., Armstrong, H. A., Williams, M., Sabbe, K., Zalasiewicz, J. A., Nolvak, J., and Verniers, J. (2010) Epipelagic chitinozoan biotopes map a steep latitudinal temperature gradient for earliest Late Ordovician seas: implications for a cooling Late Ordovician climate. Palaeogeography, Palaeoclimatology, Palaeoecology, 294, 202-219.

Walker, L. J., Wilkinson, B. H., and Ivany, L. C. (2002) Continental drift and Phanerozoic carbonate accumulation in shallow-shelf and deep-marine settings. The Journal of Geology, 110, 75-87.

Webby, B. D., Elias, R. J., Young, G. A., Neuman B. E. E., and Kaljo, D. (2004) Corals. The Great Ordovician Biodiversification Event. New York: Columbian University Press, 124-146.

Wilde, P. (1991) Oceanography in the Ordovician. In Barnes, C. R., and Williams, S. H. (eds.), Advances in Ordovician geology, Geological Survey of Canada Paper, 90 (9), 283-298.

Wiley, E. O., and Lieberman, B. S. (2011) Phylogenetics: Theory and practice of phylogenetic systematics (second edition), Hoboken: John Wiley & Sons, 406 pp.

Wood, S., Paris, C. B., Ridgwell, A., and Hendy, E. J. (2013) Modeling dispersal and connectivity of broadcast spawning corals at the global scale. Global Ecology and Biogeography, 23, 1-11.

Wright, D. F., and Stigall, A. L. (2013a) Geologic drivers of Late Ordovician faunal change in Laurentia: investigating links between tectonics, speciation, and biotic invasions. PloS one, 8 (7), doi: 10.1371/journal.pone.0068353.

Wright, D. F., and Stigall, A. L. (2013b) Phylogenetic revision of the Late Ordovician orthid brachiopod genera Plaesiomys and Hebertella from Laurentia. Journal of Paleontology, 87 (6), 1107-1128.

133

Wright, D. F., and Stigall, A. L. (2014) Species-level phylogenetic revision of the Ordovician orthide brachiopod Glyptorthis from North America. Journal of Paleontology, 12 (8), 893-906.

Young, S. A., Saltzman, M. R., and Bergstöm, S. M. (2005) Upper Ordovician (Mohawkian) carbon isotope (δ13C) stratigraphy in eastern and central North America: Regional expression of a perturbation of the global carbon cycle. Palaeogeography, Palaeoclimatology, Palaeoecology, 222, 53-76.

Young, S. A., Saltzman, M. R., Bergström, S. M., Leslie, S. A. and Xu, C. (2008) Paired 13 13 δ Ccarb and δ Corg records of Upper Ordovician (Sandbian-Katian) carbonates in North America and China: Implications for paleoceanographic change. Palaeogeography, Palaeoclimatology, Palaeoecology, 270, 166-178. 134

APPENDIX 1: SPECIES OCCURRENCE DATA

Species occurrences used to create stratocladograms (Appendix 4). Abbreviations: IGTUT, Institute of Geology at Tallinn University of Technology; KUMIP, University of Kansas Biodiversity Institute, Invertebrate Paleontology; PBDB, Paleobiology Database; SNOMNH, Sam Noble Oklahoma Museum of Natural History; YPM, Yale Peabody Museum of Natural History. Amati & Westrop, 2004 Species Occurrence Reference Thaleops marginalis lower Table Head Fm., western Newfoundland Amati & Westrop, 2004 T. utahensis Kanosh Sh. & Lehman Fm., western UT & eastern NV Amati & Westrop, 2004 T. lacertus Amadjuak Fm., Frobisher Bay, Baffin Island, Nunavut Amati & Westrop, 2004 T. viator Lourdes Ls., Port Au Port Peninsula, Newfoundland Amati & Westrop, 2004 T. mobydicki Bromide Fm., OK Amati & Westrop, 2004 T. conradi Leray Fm. & Black River Fm., Hull, Quebec Amati & Westrop, 2004 Black River Gp., Mechanicsville, Ontario Amati & Westrop, 2004 T. laurentiana Trenton Gp., Trenton Falls, NY Amati & Westrop, 2004 Galena Fm., MN & IL Amati & Westrop, 2004 Chazy Fm., Chazy, NY Amati & Westrop, 2004 T. aduncus Esbataottine Fm., District of Mackenzie Amati & Westrop, 2004 T. anusacerbissima Viola Springs Fm., OK Amati & Westrop, 2004 T. mackenziensis Esbataottine Fm., District of Mackenzie Amati & Westrop, 2004 T. fieldi Holston Ls., Catawba Valley, VA Amati & Westrop, 2004 T. latiaxiatus Black River Gp., Ottawa, Ontario Amati & Westrop, 2004 Cobourg Fm., Bowmanville, Ontario Amati & Westrop, 2004 Black River Gp., Pattersonville & Newport, NY Amati & Westrop, 2004 T. jaanussoni Bromide Fm., OK Amati & Westrop, 2004 T. raymondi Day Point & Crown Point Fm., Valcour Isl, NY Amati & Westrop, 2004 T. ovata Galena Fm., WI Amati & Westrop, 2004 Kimmswick Fm., MO Amati & Westrop, 2004 Maquoketa Gp., IA Amati & Westrop, 2004 Platteville Fm., IL & WI Amati & Westrop, 2004 Black River & Ottawa Fm., Ottawa, Ontario Amati & Westrop, 2004

135

T. longispina Crown Point & Valcour Fm., Valcour Island, NY Amati & Westrop, 2004 T. conifrons Black River Gp. & Mingan Fm., Mingan Island, Quebec Amati & Westrop, 2004

Bauer & Stigall, 2014 Species Occurrence Reference Eochonetes clarksvillensis Waynesville Fm., OH Bauer & Stigall, 2014 E. celticus Junction Beds & Killey Bridge Fm., Pomeroy Co., Ireland Bauer & Stigall, 2014 E. advena Drummuck Gp., Thraive Glen, Girvan, Scotland Bauer & Stigall, 2014 E. maearum upper Bighorn Fm., WY Bauer & Stigall, 2014 E. johnsonella Leemon Fm., MO Bauer & Stigall, 2014 E. voldemortus Saturday Mountain Fm., South Lemhi Ridge, ID Bauer & Stigall, 2014 E. dignata lower Maquoketa Fm., Clermont, IA Bauer & Stigall, 2014 E. glabra Ellis Bay Fm., Anticosti Island, Quebec Bauer & Stigall, 2014 E. mucronata Aleman Ls., TX Bauer & Stigall, 2014 E. aspera Elgin Mbr., Maquoketa Fm., IA Bauer & Stigall, 2014 E. magna Aleman Ls., TX Bauer & Stigall, 2014 E. minerva Cutter Ls. & Aleman Ls., TX Bauer & Stigall, 2014 E. vaurealensis Vaureal Fm., Anticosti Island, Quebec Bauer & Stigall, 2014 E. recedens Cincinnati Gp., MN Bauer & Stigall, 2014

Carlucci et al., 2010 Species Occurrence Reference Lyralichas bronnikovi Karatan Range, Tamdy River, Turkestan Thomas & Holloway, 1988 Probolichas robbinsi Galena Gp., MN Carlucci et al., 2010 P. kristiae Pooleville Mbr., Bromide Fm., Arbuckle Mnts, OK Carlucci et al., 2010 P. pandus Lincolnshire Ls., VA Evitt, 1951 Amphilichas subpunctatus Pooleville Mbr, Bromide Fm., Arbuckle Mountains, OK Carlucci et al., 2010 A. lineatus Boda Ls., Dalarne, Sweden Carlucci et al., 2010 Kjorrven, Bonsnes, & Husbergoya Fm., Oslo, Norway Carlucci et al., 2010 Leptaena Ls., Dalarne, Sweden Warburg, 1925 A. effnensis Effna Fm., VA Carlucci et al., 2010 A. cucullus Viola Springs Fm., OK Carlucci et al., 2010 Selkirk Mbr, Red River Fm., southern Manitoba Westrop & Ludvigsen, 1983

136

A. encyrtos Quondong Fm., Bowan Park Gp., New South Wales Webby, 1974 Bejamin Ls. & Lords Slts., Tasmania Edgecombe et al., 2004 Shale and Siltstone in Gordon Gp., Queenstown, Tasmania Edgecombe et al., 2004 Apatolichas jukesi Lower Head Fm., western Newfoundland Thomas & Holloway, 1988 Antelope Valley Ls., NV PBDB

Carlucci et al., 2012 Species Occurrence Reference Bumastoides holei Rust Fm., Trenton Gp., NY Carlucci et al., 2012 Kimmswick Ls., MO Carlucci et al., 2012 B. milleri Selby & Rockland Fm., Trenton Gp., NY & Ontario Carlucci et al., 2012 Platteville Gp., IA, WI, IL Carlucci et al., 2012 Maquoketa Fm., IA Carlucci et al., 2012 B. graffhami Pooleville Mbr, Bromide Fm., OK Carlucci et al., 2012 B. tenuirugosus Cape Calhoun Fm., northern Greenland Carlucci et al., 2012 Selkirk Mbr, Red River Fm., southern Manitoba Carlucci et al., 2012 B. bellevillensis Verulam Fm., Trenton Gp., Ontario Carlucci et al., 2012 B. billingsi Trenton Gp., Quebec Carlucci et al., 2012 Kimmswick Ls., Dunleith Fmtn, IA Carlucci et al., 2012 B. beckeri Maquoketa Fm., IA Carlucci et al., 2012 B. kimmswickensis Kimmswick Ls., Dunleith Fm., IA Carlucci et al., 2012 B. moundensis Effna Fm., VA Carlucci et al., 2012 B. lenzi Esbataottine Fm., District of Mackenzie, Northwest Territories Carlucci et al., 2012 B. solangeae Whittaker Fm., Mackenzie Mountains, Northwest Territories Carlucci et al., 2012

Congreve & Lieberman, 2008 Species Occurrence Reference Trimerus delphinocephalus Beech River Ls., Brownsport Fm., TN KUMIP Waldron Fm., Wayne Gp., IN KUMIP Platycoryphe dyaulax Qusayba Shale, Qwusayba District, Saudi Arabia Thomas, 1977 P. dentata Harknessella Beds, South Shropshire, England Dean, 1961 P. christyi Waynesville Fm., OH Whittington, 1965 P. vulcani Weston & Betton Beds, Shelve Inlier, Shropshire, England Whittard, 1960 137

Brongniartella trentonensis Salona Fm., PA Whiteley et al., 2002 B. bisulcata Alternata Ls. & Lower Cheney Longville Flags, Shropshire, England Dean, 1961 Plaesiacomia vacuvertis Hanadir Shale Mbr, Tabuk Fm., Saudi Arabia Thomas, 1977 P. oehlerti Schistes du Courijou, de Morgat, & de Raguenez, France Kerforne, 1900 Colpocoryphe arago Sarka Fmnt, Rokycany, Czech Republic iDigBio C. rouaulti Guindo Shales, Spain PBDB Calymenella boisselli Bedinan Upper Shale, Turkey PBDB C. alcantarae Iberian Peninsula, Spain Hammann & Henry, 1978 Eohomalonotus sdzuyi Iberian Peninsula, Spain Hammann & Henry, 1978

Congreve & Lieberman, 2010 Species Occurrence Reference Deiphon globifrons Slite Marl, Slite Beds, Gotland, Sweden Ramskold, 1983 Gotlandian Ls., Gotland, Sweden Whittard, 1934 D. barrandei Wenlock Shale, Malvern, Britain Lane, 1971 Woolhope Shale, Woolhope Ls., Dudley, England Whittard, 1934 Wenlock Shale, Malvern, Britain Whittard, 1934 D. longifrons St. Clair Ls., Batesville, AR Whittard, 1934 D. ellipticum Halla Beds, Gotland, Sweden Ramskold, 1983 D. bainsi Avalanche Lake Four, Mackenzie Mountains, Northwest Territories Chatterton & Perry, 1984 D. grovesi Avalanche Lake Four, Mackenzie Mountains, Northwest Territories Chatterton & Perry, 1984 Onycopyge liversidgei Quidong Fm., Quidong, New South Wales Holloway & Campbell, 1974 Sphaerocoryphe goodnovi Crown Point Fm., Chazy Grp, NY Whiteley et al., 2002 Chazy Ls., NY Raymond, 1905 Valcour Fm., NY PBDB Day Point & Crown Point Fm., Valcour Island, NY Shaw, 1968 S. gemina Edinburg Fm., VA Tripp et al., 1997 S. cranium Aseri Stage, St. Petersburg, Russia PBDB S. robustus Whittaker Fm., Northwest Territories PBDB Spillway Mbr, Rust Fm., Trenton Gp., NY PBDB S. kingi Cautley Mudstones, Murthwaite Inlier, Cautley, United Kingdom Ingham & Ingham, 1977 S. murphyi Raheen Fm., Co. Waterford, Ireland Owen & Tripp, 1986 S. longispina Edinburg Fm., VA Tripp et al., 1997 138

S. maquoketensis Clermont Mbr, Elgin Fm., IA Slocom, 1913 S. dentata Upper Jonstrop Mudstone, Vastergotland, Sweden Tripp et al., 1997 Hemisphaerocoryphe huebneri Haljala Stage, eastern Estonia Schmidt, 1881 H. elliptica Dawan Fm., Dawan, Szechuan Lu, 1975 H. pseudohemicranium Johvi Fm., Johvi Susbstage, Haljala & Keila Stages, Estonia IGTUT

Hunda & Hughes, 2007 Species Occurrence Reference Flexicalymene verecunda Winterhouse Fm., Port Au Port Peninsula, Newfoundland Dean, 1979 F. caractaci Lower Bala & Longville Flags Fm., North Wales Shirley, 1931 F. onniensis Dufton Shales, England PBDB Acton Scott Beds, Bala Fm., Shropshire, England Shirley, 1931 F. scabustula Arina & Mossen Fm., Harji Series, Estonia IGTUT F. jemtlandica Keila Stage, Estonia IGTUT F. cavei Sholeshook Ls., Llandowror, Wales Price, 1974 F. senaria Stewartville, Dubuque, & Maquoketa Fmtns, MN PBDB Beecher, Eagle Point, Fairplay, Mortimer Mbr., Dunleith Fm., IL, IA PBDB Wise Lake, Scales Shale, & Elgin Fm., IA PBDB Verulam, Bobcaygeon, & Lindsay Fm., Ontario PBDB Curdsville Mbr, Hermitage Fm., TN PBDB Chandler Falls Ls. & Groos Quarry Fm., MI PBDB Sugar River Ls. & Denley Ls., NY PBDB Sodus Fm., NY PBDB Martinsburg Fm., VA PBDB F. senaria Viola Springs Fm., OK SNOMNH Denmark Mbr, Sherman Fall Fm., VA YPM Trenton Ls., NY YPM F. quadricapita Bills Creek Shales, MI Stumm & Kauffman, 1958 F. praelongicephala Richmond Fm., MI PBDB F. granulosa Kope Fm., KY PBDB F. meeki Richmond Gp., OH, KY, IN PBDB Whitby Fm., Ontario PBDB Bills Creek Shales, MI PBDB

139

Leipers Fm., TN PBDB Lindsay Fm., Ontario PBDB Utica Shale, NY YPM Welling Fm., OK SNOMNH F. retrorsa retrorsa Arnheim Fm., OH PBDB F. retrorsa minuens Liberty Fm., OH Foerste, 1910

Wright & Stigall, 2013 Species Occurrence Reference Glyptorthis costellata Bromide Fm., OK Wright & Stigall, 2013 G. alta Kimmswick Fm., MO Wright & Stigall, 2013 G. bellarugosa Decorah Fm., WI Wright & Stigall, 2013 G. irregularis Wardell, Pierce, & Lebanon Fm., TN Wright & Stigall, 2013 G. assimilis Ridley & Advance Fm., TN Wright & Stigall, 2013 G. uniformis Benbolt, Sevier, & Wardell Fm., TN & VA Wright & Stigall, 2013 G. subcircularis Decorah Fm., IA & MN Wright & Stigall, 2013 G. multicostellata Ridley Fm., GA Wright & Stigall, 2013 G. sulcata Crown Point Fm., NY Wright & Stigall, 2013 G. equiconvexa Shippensburg Fm., MD Wright & Stigall, 2013 Boetetourt Fm., VA Wright & Stigall, 2013 G. glaseri Viola Fm., OK Wright & Stigall, 2013 Cape Ls., MO Wright & Stigall, 2013 G. insculpta Waynesville & Liberty Fm., OH & IN Wright & Stigall, 2013 G. maquoketensis Maquoketa Fm., IA Wright & Stigall, 2013 Upham & Aleman Fm., TX Wright & Stigall, 2013 Cape Ls., MO Wright & Stigall, 2013 Arnheim Fm., TN Wright & Stigall, 2013 G. senecta Poteet Fm., VA Wright & Stigall, 2013 G. obesa Bromide Fm., OK Wright & Stigall, 2013 G. bellatula Chatham Hill Fm., VA Wright & Stigall, 2013 G. concinnula Little Oak Fm., AL Wright & Stigall, 2013 Athens Fm., TN Wright & Stigall, 2013 Arline Fm., VA Wright & Stigall, 2013

140

G. unicata Bromide Fm., OK Wright & Stigall, 2013 G. glypta Effna Fm., AL Wright & Stigall, 2013 Pratt Ferry Fm., VA Wright & Stigall, 2013 G. daytonensis Brassfield Fm., OH Wright & Stigall, 2013 G. fausta Brassfield Fm., OH Wright & Stigall, 2013 G. crenulata Bromide Fm., OK Wright & Stigall, 2013 G. pulchra Maquoketa Fm., IA Wright & Stigall, 2013

Wright & Stigall, 2014 Species Occurrence Reference Hebertella alveata Liberty & Whitewater Fm., OH & IN Wright & Stigall, 2014 H. maria Vaureal & Ellis Bay Fm., Anticosti Island, Quebec Wright & Stigall, 2014 H. richmondensis Whitewater Fm., IN Wright & Stigall, 2014 H. bursa Athens Fm., TN Wright & Stigall, 2014 H. parksensis Point Pleasant & Clays Ferry Fm., KY Wright & Stigall, 2014 H. frankfortensis Bigby-Cannon Fm., KY Wright & Stigall, 2014 Catheys Fm., TN Wright & Stigall, 2014 H. subjugata Clays Ferry Fm., IN Wright & Stigall, 2014 Bellevue Fm., KY Wright & Stigall, 2014 Waynesville Fm., OH Wright & Stigall, 2014 H. occidentalis Leipers, Fernvale, & Catheys Fm., TN Wright & Stigall, 2014 Arnheim & Bellevue Fm., KY Wright & Stigall, 2014 Oregonia Fm., IN Wright & Stigall, 2014 Liberty, Whitewater, & Elkhorn Fm., OH Wright & Stigall, 2014 H. montoyensis Aleman Ls. & Cutter Fm., TX Wright & Stigall, 2014 H. prestonensis Maquoketa Fm., IA Wright & Stigall, 2014

Plaesiomys cutterensis Cutter Ls., TX Wright & Stigall, 2014 P. idahoensis Saturday Mountain Fm., ID Wright & Stigall, 2014 P. carltona Vaureal Fm., Anticosti Island, Quebec Wright & Stigall, 2014 P. subquadratus Liberty, Waynesville, & Whitewater Fm., IN & OH Wright & Stigall, 2014 Viola Fm., OK Wright & Stigall, 2014 P. subquadratus Caution Creek Fm., MN Wright & Stigall, 2014 141

Aleman Ls., TX Wright & Stigall, 2014 Avalanche Fm., British Columbia Wright & Stigall, 2014 P. anticostiensis Grindstone Mbr, Ellis Bay Fm., Anticosti Island, Quebec Wright & Stigall, 2014 P. bellistriatus Maquoketa Fm., IA & IL Wright & Stigall, 2014 Viola Fm., TN Wright & Stigall, 2014 Aleman Ls., TX Wright & Stigall, 2014 P. belilamellosus Maquoketa Fm., IA Wright & Stigall, 2014 P. subcircularis Amadjuak Fm., Baffin Island, Nunavut Wright & Stigall, 2014 P. occidentalis Maquoketa Fm., IA Wright & Stigall, 2014 P. proavitus Cape Ls., MO Wright & Stigall, 2014 Fernvale Fm., TN Wright & Stigall, 2014 Maquoketa Fm., MN Wright & Stigall, 2014 Viola Fm., IL Wright & Stigall, 2014

142

APPENDIX 2: CORRELATION CHARTS

Correlation charts of stratigraphic units listed in species occurrence data (Appendix #) and correlated using published graptolite and conodont zones for the Middle Ordovician through Silurian Ages. Chronostratigraphic chart of the Ordovician modified after Bergström et al. (2008) and Cohen et al. (2013). Cincinnatian depositional sequences, North American Series and Stages after Holland and Patzkowsky (1996). North American Ordovician midcontinent and Atlantic conodont zones and North American graptolite zones from Huff et al. (2010). North American Silurian conodont zones and graptolite zones from Pyle and Barnes (2003). References for each stratigraphic column are located at the bottom of the column.

143

144

145

146

147

148

APPENDIX 3: REFERENCES FOR SPECIES OCCURRENCES AND STRATIGRAPHY

Ainsaar, L., Kalijo, D., Martma, T., Meidla, T., Mannik, P., Nolvak, J., and Tinn, O. (2010) Middle and Upper Ordovician carbon isotope chemostratigraphy in Baltoscandia: a correlation standard and clues to environmental history. Palaeogeography, Palaeoclimatology, Palaeoecology, 294 (3), 189-201.

Al-Hajri, S. (1995) Biostratigraphy of the Ordovician chitinozoan of northwestern Saudi Arabia. Review of Palaeobotany and Palynology, 89 (1), 27-48.

Amsden, T. W. (1949) Stratigraphy and paleontology of the Brownsport Formation (Silurian) of western Tennessee. Peabody Museum of Natural History Bulletin 5, 134 p.

Amsden, T. W., and Sweet, W. C. (1983) Upper Bromide Formation and Viola Group (Middle and Upper Ordovician) in eastern Oklahoma. Oklahoma Geological Survey Bulletin, 137 (76), i0883-1351.

Andrichuk, J. M. (1959) Ordovician and Silurian stratigraphy and sedimentation in southern Manitoba, Canada. AAPG Bulletin, 43 (10), 2333-2398.

Armstrong, D. K. (2000) Lithostratigraphic logging of six bedrock boreholes in the Rouge River valley, southern Ontario. Ontario Geological Survey, Open File Report 6038.

Armstrong, D. K., and Dodge, J. P. (2007) Paleozoic geology of southern Ontario. Ontario Geological Survey, Miscellaneous Release-Data 219, 2-5.

Bassett, D. A., Whittington, H. B., and Williams, A. (1966) The stratigraphy of the Bala district, Merionethshire. Quarterly Journal of the Geological Society, 122, 219- 269.

Bassett, M. G., Cocks, L. R. M., Holland, C. H., Rickards, R. B., and Warren. P. T. (1975) The type Wenlock series. Report of the Institute of Geological Sciences, 75 (13), 1-19.

Bergström, S. M., Chen, X., Gutierrez-Marco, J. C., and Dronov, A. (2009) The new chronostratigraphic classification of the Ordovician system and its relations to major regional series and stages and to δ13C chemostratigraphy. Lethaia, 42 (1), 97-107.

149

Bergström, S. M., Riva, J., and Kay, M. (1974) Significance of conodonts, graptolites, and shelly faunas from the Ordovician of western and north-central Newfoundland. Canadian Journal of Earth Sciences, 11 (12), 1625-1660.

Bergström, S. M., Young, S. & Schmitz, B. (2010) Katian (Upper Ordovician) δ13C chemostratigraphy and sequence stratigraphy in the United States and Baltoscandia: A regional comparison. Palaeogeography, Palaeoclimatology, Palaeoecology, 296, 217-234.

Brett, C. E., McLaughlin, P. I., Cornell, S. R., and Baird, G. C. (2004) Comparative sequence stratigraphy of two classic Upper Ordovician successions, Trenton Shelf (New York-Ontario) and Lexington Platform (Kentucky-Ohio): implications for eustasy and local tectonism in eastern Laurentia. Palaeogeography, Palaeoclimatology, Palaeoecology, 210 (2), 295-329.

Bruton, D. L., and Owen, A. W. (1988) The Norwegian Upper Ordovician illaenid trilobites. Norsk Geologisk Tidsskrift, 68, 241-258.

Calner, M., Jeppsson, L., and Munnecke, A. (2004) The Silurian of Gotland-Part I: Review of the stratigraphic framework, event stratigraphy, and stable carbon and oxygen isotope development. Erlanger geologische Abhandlungen, 5, 113-131.

Catacosinos, P. A., Harrison, W. B., Reynolds, R. F., Westjohn, D. B., and Wollensak, M. S. (2000) Stratigraphic Nomenclature for Michigan. Stratigraphic Chart, Michigan Department of Environmental Quality Geological Survey Division and Michigan Basin Geological Society.

Chatterton, B. D. E., and Campbell, K. S. W. (1980) Silurian trilobites from near Canberra and some related forms from the Yass Basin. Palaeontographica Abteilung A, 1-3, 77-119.

Chatterton, B. D. E., and Perry, D. G. (1984) Silurian Cheirurid Trilobites from the Mackenzie Mountains (Northwestern Canada). Paleontographica Abteilung A, 1 (4), 1-78.

Cocks, L. R. M. (2014) The Late Ordovician brachiopods of southern Pembrokeshire and adjacent south-western Wales. Special Papers in Palaeontology, 91, 5-89.

Cocks, L. R. M., McKerrow, W. S., and van Staal, C. R. (1993) The margins of Avalonia. Geological Magazine, 134 (5), 627-636.

Cohen, K. M., Finney, S. C., and Gibbard, P. L. (2013; updated) The ICS International Chronostratigraphic Chart. Episodes, 36, 199-204.

150

Dean, W. T. (1961) The Ordovician trilobite faunas of south Shropshire, II. Bulletin of the British Museum of Natural History (Geology), 5, 311-358.

Dean, W. T. (1971) Ordovician trilobites from the Central Volcanic Mobile Belt at New World Island, northeastern Newfoundland. Geological Survey of Canada, 210, 10 p.

Dean, W. T. (1979) Trilobites from the Long Point Group (Ordovician), Port au Port Peninsula, southwestern Newfoundland. Geological Survey of Canada Bulletin, 290, 43 p.

Droste, J. B., and Shaver, R. H. (1985) Comparative stratigraphic framework for Silurian reefs; Michigan Basin to surrounding platforms. Geological Society, Special Paper, 4, 73-93.

Ebbestad, J. O. R., and Peel, J. S. (1997) Attempted predation and shell repair in Middle and Upper Ordovician gastropods from Sweden. Journal of Paleontology, 71, 1007-1019.

Edgecombe, G. D., Banks, M. R., and Banks, D. M. (2004) Late Ordovician trilobites from Tasmania: Styginidae, Asaphidae and Lichidae. Memoirs of the Association of Australasian Palaeontologists, 30, 59-77.

Evitt, W. R. (1951) Some Middle Ordovician trilobites of the families Cheiruridae, Harpidae and Lichidae. Journal of Paleontology, 25, 85-88.

Ferretti, A., Bergström, S. M., and Barnes, C. R. (2014) Katian (Upper Ordovician) conodonts from Wales. Palaeontology, 57 (4), 801-831.

Foerste, A. F. (1910) Preliminary notes on Cincinnatian and Lexington fossils of Ohio, Indiana, Kentucky, and Tennessee. Denison University Science Laboratories Bulletin, 18, 285-355.

Fortey, R. A., Harper, D. A. T., Ingham, J. K., Owen, A. W., Parkes, M. A., Rushton, A. W. A., and Woodcock, N. H. (2000) A revised correlation of Ordovician rocks in the British Isles. Geological Society of London Special Report, 24, 83 p.

Ghienne, J.-F., Monod, O., Kozlu, H., and Dean, W. T. (2010) Cambrian-Ordovician depositional sequences in the Middle East: a perspective from Turkey. Earth- Science Reviews, 101 (3), 101-146.

Gritsenko, V. P., Drygant, D. M., Ishchenko, A. A., Konstantinenko, L. I., and Tsegelnjuk, P. D. (1983) The local Silurian scheme of Podolia. In Sokolov, B. S. 151

(ed.) The Silurian of Podolia. Academy of Sciences of the Ukrainian SSR, Kiev, 61-71.

Hammann, W., and Henry, J. L. (1978) Quelques espèces de Calymenella, Eohomalonotus et Kerfornella (Trilobita, Ptychopariida) de l'Ordovicien du Massif Armoricain et de la Péninsule ibérique. Senckenbergiana lethaea, 49, 401- 429.

Harper, D. A. T. (1986) Distributional trends within the Ordovician brachiopod faunas of the Oslo Region, South Norway. In Racheboeuf, P. R., and Emig, C. C. (eds.) Les Brachiopodes Fossiles et Actuels. Biostratigraphie du Paleozoique, Universite de Bretagne Occidentale, Brest, 4, 465-475.

Hender, K. L. B., and Dix, G. R. (2008) Facies development of a Late Ordovician mixed carbonate-siliciclastic ramp proximal to the developing Taconic orogen: Lourdes Formation, Newfoundland, Canada. Facies, 54 (1), 121-149.

Holland, C. H. (2001) The Geology of Ireland. Dunedin Academic Press Ltd., Edinburgh, 568 p.

Holland, S. M, and Patzkowsky, M. E. (1996) Sequence stratigraphy and long-term paleoceanographic change in the Middle and Upper Ordovician of the eastern United States, p. 117-129. In B. J. Witzke, G. A. Ludvigson, and J. Day (eds.), Paleozoic Sequence Stratigraphy: Views from the North America Craton. Geological Society of America Special Paper 306.

Holloway, D. J. (1980) Silurian dalmanitacean trilobites from North America and the origins of the Dalmanitinae and Synphoriinae. Palaeontology, 24 (4), 695-731.

Holloway, D. J., and Campbell, K. S. W. (1974) The Silurian trilobite Onycopyge Woodward. Palaeontology, 17, 409-421.

Ingham, J. K. (1966) The Ordovician rocks in the Cautley and Dent districts of Westmoreland and Yorkshire. Proceedings of the Yorkshire Geological Society, 35, 455-405.

Ingham, J. K., and Ingham, I. K. (1977) The Upper Ordovician trilobites from the Cautley and Dent districts of Westmoreland and Yorkshire. Palaeontographical Society.

Kay, G. M. (1937) Stratigraphy of the Trenton Group. Geological Society of America Bulletin, 48, 233-302.

152

Kerforne, F. (1900) Description de trois nouveaux Trilobites de l‘Ordovicien de Bretagne. Bulletin de la Societe Geologique de France, 28 (3), 344-388.

Kiipli, T., Tsegelnyuk, P. D., and Kallaste, T. (2000) Volcanic interbeds in the Silurian of the southwestern part of the East European Platform. Proceedings of the Estonian Academy of Sciences, Geology, 49 (3), 163-176.

Kobayasht, T., and Hamada, T. (1976) A new Silurian trilobite from Ofunato, Japan. Proceedings of the Japan Academy, 52 (7), 367-370.

Landing, E., and Westrop, S. R. (2006) Lower Ordovician faunas, stratigraphy, and sea- level history of the middle Beekmantown Group, northeastern New York. Journal of Paleontology, 80 (5), 958-980.

Lane, P. D. (1971) British Cheiruridae (Trilobita). Palaeontographical Society, London, 95 p.

Lehnert, O., Fryda, J., Buggisch, W., Munnecke, A., Nutzel, A., Kriz, J., and Manda, S. (2007) δ13C records across the late Silurian Lau event: new data from middle palaeo-latitudes of northern peri-Gondwana (Prague Basin, Czech Republic). Palaeogeography, Palaeoclimatology, Palaeoecology, 245 (1), 227-244.

Lenz, A. C. (1982) Ordovician to Devonian sea-level changes in western and northern Canada. Canadian Journal of Earth Sciences, 19 (10), 1919-1932.

Leslie, S. A., and Bergström, S. M. (1997) Use of K-bentonite beds as time-planes for high-resolution lithofacies analysis and assessment of net rock accumulation rate: An example from the upper Middle Ordovician of eastern North America. Geological Society of America Special Paper, 321, 11-21.

Longman, M. W., and Haidl, F. M. (1996) Cyclic deposition and development of porous dolomites in the Upper Ordovician Red River Formation, Williston Basin. In Longman, M. W., and Sonnenfeld, M. D. (eds.) Paleozoic systems of the Rocky Mountains Region. Rocky Mountain Section, Society of Economic Paleontologists and Mineralogists, 29-46.

Lu, Y.-H. (1975) Ordovician trilobite faunas of central and southwestern China. Science Press. 463 p.

Manten, A. A. (1971) Silurian reefs of Gotland. Elsevier, Amsterdam. 539 p.

Mikulic, D. G., Sargent, M. L., Norby, R. D., and Kolata, D. R. (1985) Silurian geology of the Des Plaines river valley, northeastern Illinois. Illinois State Geological Survey Guidebook, 56 p. 153

Mitchell, C. E., and Sweet, W. C. (1989) Upper Ordovician conodonts, brachiopods, and chronostratigraphy of the Whittaker Formation, southwestern District of Mackenzie, NWT, Canada. Canadian Journal of Earth Sciences, 26 (1), 74-87.

Mossler, J. H. (2008) Paleozoic Lithostratigraphic Nomenclature for Minnesota. Minnesota Geological Survey Report of Investigations 65, 83 p.

Nolvak, J., Hints, O., and Mannik, P. (2006) Ordovician timescales in Estonia: Recent developments. Proceedings of the Estonian Academy of Sciences Geology, 55, 95- 108.

Norford, B. S., Haidl, F. M., Bezys, R. K., Cecile, M. P., McCabe, H. R., and Paterson, D. F. (1994) Middle Ordovician to Lower Devonian strata of the Western Canada Sedimentary Basin. In Geological Atlas of the Western Canada Sedimentary Basin. Canadian Society of Petroleum Geologists and Alberta Research Council, 109-127.

Owen, A. W., Tripp, R. P., and Morris, S. F. (1986) The trilobite fauna of the Raheen Formation (upper Caradoc), Co. Waterford, Ireland. Bulletin of the British Museum, Natural History, Geology, 40, 91-122.

Parkes, M. A. (1994) The brachiopods of the Duncannon group (Middle-Upper Ordovician) of southeast Ireland. Bulletin of the Natural History Museum (Geology), 50, 105-174.

Percival, I. G. (2009) Late Ordovician strophomenide and pentameride brachiopods from central New South Wales. Proceedings of the Linnean Society of New South Wales, 130, 157-178.

Pickerill, R. K., Fillion, D., and Harland, T. L. (1984) Middle Ordovician trace fossils in carbonates of the Trenton Group between Montreal and Quebec City, St. Lawrence lowland, eastern Canada. Journal of Paleontology, 58, 416-439.

Price, D. (1974) Trilobites from the Sholeshook Limestone (Ashgill) of South Wales. Palaeontology, 17, 841-868.

Ramsköld, L. (1983) Silurian cheirurid trilobites from Gotland. Palaeontology, 26 (1), 175-210.

Rasmussen, C. M. O. (2013) Late Ordovician brachiopods from eastern North Greenland: equatorial offshore migration of the Red River fauna. Palaeontology, 56 (2), 359- 379.

154

Raymond, P. E. (1905) The trilobites of the Chazy Limestone. Annuals of the Carnegie Museum, 3, 328-386.

Reyes-Abril, J., Villas, E., Gutierrez-Marco, J. C., Jimenez-Sanchez, A., and Colmenar, J. (2013) Middle Ordovician harknessellid brachiopods (Dalmanellidina) from the Mediterranean margin of Gondwana. Bulletin of Geosciences, 88 (4), 813-828.

Ross, C. A., and Ross, J. R. P. (1995) North American Ordovician depositional sequences and correlations. In Cooper, J. D., Droser, M. L., and Finney, S. C. (eds.) Ordovician Odyssey, Pacific Section, Society of Economic Paleontologists and Mineralogists, 309-313.

Ross, R. J., and James, N. P. (1987) Brachiopod biostratigraphy of the middle Ordovician Cow Head and Table Head groups, western Newfoundland. Canadian Journal of Earth Sciences, 24 (1), 70-95.

Ross, R. J., and Bergström, S. M. (1982) The Ordovician system in the United States. International Union of Geological Sciences Publication 12.

Schmidt, F. (1881) Revision der ostbaltischen silurischen Trilobiten nebst geognosticher ubersicht des ostbaltischen Silurgebiets.

Schmitz, B., and Bergström, S. M. (2007) Chemostratigraphy in Swedish Upper Ordovician: regional significance of the Hirnantian δ13C excursion (HICE) in the Boda Limestone of the Siljan region. GFF, 129 (2), 133-140.

Selleck, B. W. (2010) Stratigraphy of the Northern Appalachian Basin, Mohawk Valley, Central New York State. Unpublished, Department of Geology, Colgate University.

Shaw, F. C. (1968) Early Middle Ordovician Chazy trilobites of New York. New York State Museum and Science Service Memoir 17, University of the State of New York, Albany, 163 p.

Shaw, F. C., and Bolton, T. E. (2011) Ordovician trilobites from the Romaine and Mingan Formations (Ibexian-Late Whiterockian), Mingan Islands, Quebec. Journal of Paleontology, 85 (3), 406-441.

Shirley, J. A. (1931) A redescription of the known British Ordovician species of Calymene (s.l.). Memoirs and Proceedings of the Manchester Literary & Philosophical Society, 75, 1-35.

Slocom, A. W. (1916) Trilobites from the Maquoketa beds of Fayette County, Iowa. Iowa Geological Survey Annual Report, 25 (1), 43-83. 155

Snajdr, M. (1980) New Silurian trilobites from the Barrandian area (Czechoslovakia). Vestnik Ustredniho Ustavu Geologickeho Prague, 55, 105-110.

Soehn, K. L., Marss, T., Hanke, G. F., and Wilson, M. V. H. (2000) Preliminary vertebrate biostratigraphy of the Avalanche Lake sections (Wenlock, Silurian), southern Mackenzie Mountains, N.W.T., and review of northwestern Canadian vertebrate localities of Silurian age. In Blieck, A., and Turner, S. (eds.) Palaeozoic Vertebrate Biochronology and Global Marine/Non-marine Correlation, Final Report of IGCP 328, 129-156.

Štorch, P. (2006) Facies development, depositional settings and sequence stratigraphy across the Ordovician-Silurian boundary: a new perspective from the Barrandian area of the Czech Republic. Geological Journal, 41 (2), 163-192.

Štorch, P., and Serpagli, E. (1993) Lower Silurian graptolites from southwestern Sardinia. Bollettino della Societa Paleontologica Italiana, 32 (1), 3-57.

Stumm, E. C., and Kauffman, E. G. (1958) Calymenid trilobites from the Ordovician rocks of Michigan. Journal of Paleontology, 32 (5), 943-960.

Suchy, V., Sykorova, I., Stejskal, M., Safanda, J., Machovic, V., and Novotna, M. (2002) Dispersed organic matter from Silurian shales of the Barrandian Basin, Czech Republic: optical properties, chemical composition and thermal maturity. International Journal of Coal Geology, 53, 1-25.

Talent, J. A., Berry, W. B., Boucot, A. J., Packham, G. H., and Bischoff, G. C. (1975) Correlation of the Silurian rocks of Australia, New Zealand, and New Guinea. Geological Society of America Special Papers, 150, 1-115.

Thomas, A. T. (1977) Classification and phylogeny of homalonotid trilobites. Palaeontology, 20 (1), 159-178.

Thomas, A. T., and Holloway, D. J. (1988) Classification and phylogeny of the trilobite order Lichida. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences, 179-262.

Tremblay, J. V., and Westrop, S. R. (1991) Middle Ordovician (Whiterockian) trilobites from the Sunblood Formation, District of Mackenzie, Canada. Journal of Paleontology, 65 (5), 801-824.

Tripp, R. P. (1954) Caradocian trilobites from mudstones at Craighead Quarry, near Girvan, Ayrshire. Transactions of the Royal Society of Edinburgh, 62, 655-689.

156

Tripp, R. P. (1962) I.-Trilobites from the Confinis Flags (Ordovician) of the Girvan District, Ayrshire. Transactions of the Royal Society of Edinburgh, 60 (1), 1-40.

Tripp, R. P., Rudkin, D. M., and Evitt, W. R. (1997) Silicified trilobites of the genus Sphaerocoryphe from the Middle Ordovician of Virginia. Canadian Journal of Earth Sciences, 34 (6), 770-788.

Tuffnell, P., and Ludvigsen, R. (1984) The trilobite Triarthrus in the Whitby Formation (Upper Ordovician) of Southern Ontario and a biostratigraphic framework, in Ontario. Geological Survey, Open File Report 5516.

Vandenbroucke, T. R. A. (2005) Upper Ordovician Global Stratotype Sections and Points and the British historical type area: A chitinozoan point of view. PhD dissertation, Ghent University, Ghent, Flanders, Belgium, 295 p.

Warburg, E. (1886) The trilobites of the Leptaena Limestone in Dalarna. Bulletin of the Geological Institute, University of Uppsala, 17, 1-446.

Watkins, R. (1994) Evolution of Silurian pentamerid communities in Wisconsin. Palaios, 9 (5), 488-499.

Webby, B. D. (1974) Upper Ordovician trilobites from central New South Wales. Palaeontology, 17 (2), 203-252.

Westrop, S. R., and Ludrisgen, S. R. (1983) Systematics and paleoecology of Upper Ordovician trilobites from the Selkirk Member of the Red River Formation, southern Manitoba. Manitoba Department of Energy and Mines, Min. Res. Div. Geol. Rept., No. GR, 31 p.

Whitely, T. E., Kloc, G. J., and Brett, C. E. (2002) Trilobites of New York: an illustrated guide. Cornell University Press, Ithaca, NY.

Whittard, W. F. (1934) LII.-A revision of the trilobite genera Deiphon and Onycopyge. Journal of Natural History, 14 (83), 505-533.

Whittard, W. F. (1959) The Ordovician trilobites of the Shelve Inlier, Shropshire Part V. Palaeontographical Society Monograph, 138-142.

Whittard, W. F. (1967) The trilobite Anacheirurus frederici (Salter) from the Tremadoc series of North Wales. Geological Magazine, 104 (3), 284-288.

Whittington, H. B. (1965) Platycoryphe, and Ordovician Homalonotid trilobite. Journal of Paleontology, 39 (3), 487-491.

157

Young, S. A., Saltzman, M. R., and Bergström, S. M. (1995) Upper Ordovician (Mohawkian) carbon isotope (δ13C) stratigraphy in eastern and central North America: Regional expression of a perturbation of the global carbon cycle. Palaeogeography, Palaeoclimatology, Palaeoecology, 222 (1), 53-76.

Zhen, Y.-Y., Webby, B. D., and Barnes, C. R. (1999) Upper Ordovician conodonts from the Bowan Park succession, central New South Wales, Australia. Geobios, 32 (1), 73-104.

Zhou, Z. Y., Dean, W. T., Yuan, W. W., and Zhou, T. R. (1998) Ordovician trilobites from the Dawangou Formation, Kalpin, Xinjiang, north-west China. Palaeontology, 41, 693-735.

Zuykov, M. A., Harper, D. A. T., Terentiev, S. S., and Pelletier, E. (2011) Revision of the plectorthoid brachiopod Platystrophia dentate (Pander, 1830) from the Middle Ordovician of the East Baltic. Estonian Journal of Earth Sciences, 60 (3), 131- 136.

158

APPENDIX 4: STRATOCLADOGRAMS

Brachiopod stratocladograms. Numbered nodes and tips correspond to coding used in LBPA tables. 159

Trilobite stratocladograms. Numbered nodes and tips correspond to coding used in LBPA table. 160

APPENDIX 5: LBPA TABLES

Time Slice 0 Vicariance 1 3 4 10 12 14 16 19 21 27 28 35 37 111 113 114 115 119 121 131 133 135 137 139 141 144 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 1 1 1 1 1 1 0 S. Appalachian Basin 0 0 0 1 1 1 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 1 1 0 0 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 1 3 4 10 12 14 16 19 21 27 28 35 37 111 113 114 115 119 121 131 133 135 137 139 141 144 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 1 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 1 1 1 1 1 1 0 S. Appalachian Basin 0 0 0 2 1 1 1 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

161

Time Slice 0 (Continued) Vicariance 146 147 149 152 153 154 156 160 163 169 170 171 172 173 174 176 177 179 181 182 183 184 220 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 1 1 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 1 2 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Baltica 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Southern Gondwana 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 Northern Gondwana 1 1 0 1 1 1 1 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 2 1 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 146 147 149 152 153 154 156 160 163 169 170 171 172 173 174 176 177 179 181 182 183 184 220 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 0 0 0 0 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 1 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 0 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 1 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Baltica 0 0 0 0 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Southern Gondwana 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 Northern Gondwana 1 0 0 1 1 1 1 1 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 1 1 1 1 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

162

Time Slice 0 (Continued) Vicariance 222 223 225 233 235 236 238 Ancestor 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 Western Midcontinent 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 1 1 S. Appalachian Basin 0 0 0 0 0 1 1 Upper Mississippi Valley 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 1 1 Southern Laurentia 0 0 0 0 0 0 0 Baltica 1 1 1 1 1 1 1 Southern Gondwana 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 1 1 Avalonia 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 Geodispersal 222 223 225 233 235 236 238 Ancestor 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 Western Midcontinent 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 2 1 S. Appalachian Basin 0 0 0 0 0 2 1 Upper Mississippi Valley 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 2 1 Southern Laurentia 0 0 0 0 0 0 0 Baltica 1 1 1 1 1 1 1 Southern Gondwana 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 2 1 Avalonia 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0

163

Time Slice 1 Vicariance 2 5 7 9 13 17 18 20 29 30 31 32 33 34 36 38 39 41 43 44 65 66 70 72 112 116 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 1 2 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 1 1 1 1 0 0 Western Midcontinent 1 1 1 1 0 0 0 0 1 1 2 1 1 1 1 1 1 1 2 1 0 0 0 0 1 1 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 0 1 1 0 1 2 1 1 1 2 1 0 1 0 0 0 0 0 0 0 1 0 0 Upper Mississippi Valley 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Southern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 2 5 7 9 13 17 18 20 29 30 31 32 33 34 36 38 39 41 43 44 65 66 70 72 112 116 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 2 1 0 0 0 0 0 0 0 0 0 0 0 0 2 1 0 0 1 1 1 1 0 0 Western Midcontinent 1 1 1 0 0 0 0 1 0 1 1 1 0 1 1 1 1 1 1 1 0 0 0 0 0 1 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 0 1 1 1 2 1 1 0 1 1 1 0 2 0 0 0 0 0 0 0 2 0 0 Upper Mississippi Valley 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 Southern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

164

Time Slice 1 (Continued) Vicariance 118 120 122 123 124 126 127 128 130 132 136 138 142 143 145 151 158 161 164 165 166 168 175 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 1 2 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 2 0 0 0 0 0 N. Appalachian Basin 0 1 1 1 1 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 2 0 0 0 0 0 0 Upper Mississippi Valley 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 1 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 2 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 2 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 1 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 118 120 122 123 124 126 127 128 130 132 136 138 142 143 145 151 158 161 164 165 166 168 175 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 0 1 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 N. Appalachian Basin 0 0 1 1 1 1 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 2 1 0 0 0 0 0 0 Upper Mississippi Valley 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 1 0 0 0 0 2 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 2 0 0 0 0 0 0 0 0 2 0 0 0 1 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

165

Time Slice 1 (Continued) Vicariance 185 186 190 192 194 195 196 201 211 213 215 217 226 228 229 230 231 239 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 0 0 0 0 1 1 2 0 0 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 S. Appalachian Basin 0 0 0 0 1 2 1 0 0 0 0 0 1 1 2 1 0 1 Upper Mississippi Valley 0 0 0 0 1 1 2 1 1 1 0 0 1 1 1 1 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 1 1 1 1 2 0 0 0 0 0 0 0 0 0 0 1 Southern Laurentia 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 1 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 Avalonia 1 1 0 0 0 0 0 1 1 2 1 1 1 1 1 2 1 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 185 186 190 192 194 195 196 201 211 213 215 217 226 228 229 230 231 239 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 0 0 0 0 2 0 1 0 0 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 S. Appalachian Basin 0 0 0 0 2 1 0 0 0 0 0 0 2 1 1 0 0 0 Upper Mississippi Valley 0 0 0 0 2 0 1 1 1 0 0 0 2 1 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 1 1 1 0 1 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Avalonia 1 1 0 0 0 0 0 1 1 1 1 1 2 1 0 1 1 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

166

Time Slice 2 Vicariance 6 8 11 15 46 50 51 53 73 74 75 76 77 79 84 85 87 90 92 94 96 97 98 100 102 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 1 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 Western Midcontinent 1 1 0 0 1 1 2 1 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 2 1 North Transcontinental Arch 0 0 0 0 1 1 2 1 0 0 0 0 0 0 1 1 1 2 1 1 1 1 1 1 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 0 2 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 2 1 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 1 0 0 0 0 0 0 0 Northern Laurentia 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 1 1 2 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 6 8 11 15 46 50 51 53 73 74 75 76 77 79 84 85 87 90 92 94 96 97 98 100 102 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 Western Midcontinent 1 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 2 1 0 1 1 North Transcontinental Arch 0 0 0 0 1 1 1 1 0 0 0 0 0 0 1 1 0 1 1 1 1 1 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 1 1 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 0 1 0 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 1 1 1 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 2 1 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

167

Time Slice 2 (Continued) Vicariance 103 104 106 108 110 117 125 129 134 148 150 155 178 180 193 197 198 202 203 204 206 207 208 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 Western Midcontinent 1 1 1 1 1 1 0 1 0 0 0 1 0 0 0 1 2 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 1 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 1 2 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 2 1 1 1 2 0 1 0 0 1 1 2 0 0 0 1 1 2 1 1 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 2 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 Southern Laurentia 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 1 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 103 104 106 108 110 117 125 129 134 148 150 155 178 180 193 197 198 202 203 204 206 207 208 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 1 Western Midcontinent 0 1 1 1 0 0 0 2 0 0 0 1 0 0 0 1 1 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 1 0 1 0 0 0 0 2 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 2 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 1 1 1 1 1 0 2 0 0 1 1 1 0 0 0 1 0 1 1 1 1 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 1 0 1 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

168

Time Slice 2 (Continued) Vicariance 214 216 221 237 Ancestor 0 0 0 0 Cincinnati 0 0 0 0 Western Midcontinent 0 0 0 0 North Transcontinental Arch 0 0 0 0 N. Appalachian Basin 0 0 0 1 S. Appalachian Basin 0 0 0 2 Upper Mississippi Valley 0 0 0 0 Scoto-Appalachian 0 0 0 0 Northern Laurentia 0 0 0 1 Southern Laurentia 0 0 0 0 Baltica 0 0 1 1 Southern Gondwana 0 0 0 0 Northern Gondwana 0 0 0 1 Avalonia 1 1 0 0 Tarim 0 0 1 0 Geodispersal 214 216 221 237 Ancestor 0 0 0 0 Cincinnati 0 0 0 0 Western Midcontinent 0 0 0 0 North Transcontinental Arch 0 0 0 0 N. Appalachian Basin 0 0 0 0 S. Appalachian Basin 0 0 0 1 Upper Mississippi Valley 0 0 0 0 Scoto-Appalachian 0 0 0 0 Northern Laurentia 0 0 0 0 Southern Laurentia 0 0 0 0 Baltica 0 0 1 0 Southern Gondwana 0 0 0 0 Northern Gondwana 0 0 0 0 Avalonia 1 1 0 0 Tarim 0 0 2 0

169

Time Slice 3 Vicariance 22 23 24 25 26 40 42 45 47 48 49 52 54 55 56 57 58 59 60 61 62 63 64 67 68 69 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 1 0 2 2 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 1 0 Western Midcontinent 1 1 1 0 1 1 1 0 1 2 1 1 1 1 2 1 1 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 1 1 2 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 Upper Mississippi Valley 0 0 0 0 1 0 0 1 1 1 1 0 0 0 2 1 2 1 1 1 1 1 1 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 2 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 0 1 1 1 2 0 2 1 0 0 0 0 0 0 0 0 0 0 1 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 22 23 24 25 26 40 42 45 47 48 49 52 54 55 56 57 58 59 60 61 62 63 64 67 68 69 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 2 0 1 1 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 1 1 1 Western Midcontinent 1 1 1 1 1 0 0 1 1 1 0 0 1 0 1 1 0 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 Upper Mississippi Valley 0 0 0 0 2 0 0 2 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 0 1 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 2 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

170

Time Slice 3 (Continued) Vicariance 71 78 80 81 82 83 86 88 89 91 93 95 99 101 105 107 109 140 157 159 162 167 187 188 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 1 1 1 1 1 1 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 Western Midcontinent 0 0 0 1 2 1 0 0 0 0 1 0 1 2 2 2 1 1 1 2 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 1 0 0 1 0 1 1 0 0 0 0 0 1 2 1 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 S. Appalachian Basin 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 0 Upper Mississippi Valley 0 0 0 1 1 2 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 2 0 0 0 1 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 0 0 1 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Geodispersal 71 78 80 81 82 83 86 88 89 91 93 95 99 101 105 107 109 140 157 159 162 167 187 188 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 1 1 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 Western Midcontinent 0 0 0 2 1 0 0 0 0 0 2 0 0 1 1 1 0 2 0 1 0 0 0 0 North Transcontinental Arch 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 1 0 0 0 0 N. Appalachian Basin 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 S. Appalachian Basin 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Upper Mississippi Valley 0 0 0 2 0 1 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Southern Laurentia 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2 Avalonia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 Tarim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

171

Time Slice 3 (Continued) Vicariance 189 191 199 200 205 209 210 212 218 219 224 227 232 234 240 241 242 243 244 245 246 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 0 0 1 1 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 1 0 1 0 0 1 0 0 0 0 1 1 1 1 0 0 0 0 S. Appalachian Basin 1 0 0 0 0 1 0 0 1 0 0 1 0 0 1 1 1 0 0 0 0 Upper Mississippi Valley 1 0 2 1 1 1 0 1 1 0 0 2 0 0 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 2 2 1 2 1 Southern Laurentia 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 1 2 1 1 1 2 Southern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 1 0 0 0 0 Avalonia 0 0 0 0 0 0 0 1 0 1 0 1 1 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 Geodispersal 189 191 199 200 205 209 210 212 218 219 224 227 232 234 240 241 242 243 244 245 246 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cincinnati 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Western Midcontinent 0 0 0 1 0 2 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 North Transcontinental Arch 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N. Appalachian Basin 0 0 0 2 0 2 0 0 2 0 0 0 0 2 1 0 0 0 0 0 0 S. Appalachian Basin 2 0 0 0 0 2 0 0 2 0 0 0 0 0 1 0 0 0 0 0 0 Upper Mississippi Valley 2 0 1 1 1 2 0 0 2 0 0 1 0 0 0 0 0 0 0 0 0 Scoto-Appalachian 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Laurentia 0 1 0 1 0 0 0 0 0 0 0 0 0 2 1 0 1 1 1 1 0 Southern Laurentia 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Baltica 0 0 0 0 0 0 0 0 0 2 1 0 0 1 1 0 1 0 1 0 1 Southern Gondwana 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Northern Gondwana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 Avalonia 1 0 0 0 0 0 0 1 1 1 0 0 1 0 0 0 0 0 0 0 0 Tarim 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 172

APPENDIX 6: LBPA MOST PARSIMONIOUS TREE TOPOLOGIES

Set of most parsimonious topologies of geodispersal and vicariance trees recovered from LBPA analyses run in PAUP*4.0. All topologies are in Nexus format.

Geodispersal T0 tree PAUP_1 = [&R] (1,2,(3,10),((((4,7),13),6,11),5,9),8,(12,14),15) tree PAUP_2 = [&R] (1,2,(3,10),(((((4,7),13),(6,11)),5),9),8,(12,14),15) tree PAUP_3 = [&R] (1,2,(3,10),((((4,7),13),(6,11)),5,9),8,(12,14),15)

T1 tree PAUP_1 = [&R] (1,2,((3,6),((7,14),11)),(4,10),5,8,9,12,13,15) tree PAUP_2 = [&R] (1,2,(3,6),(4,10),5,((7,14),11),8,9,12,13,15) tree PAUP_3 = [&R] (1,2,(((3,6),(7,14)),11),(4,10),5,8,9,12,13,15) tree PAUP_4 = [&R] (1,2,((3,6),(7,14),11),(4,10),5,8,9,12,13,15) tree PAUP_5 = [&R] (1,2,((3,6),((7,14),11)),(4,10),(5,9),8,12,13,15) tree PAUP_6 = [&R] (1,(2,(3,6)),(4,10),5,((7,14),11),8,9,12,13,15) tree PAUP_7 = [&R] (1,2,((3,6),9),(4,10),5,((7,14),11),8,12,13,15) tree PAUP_8 = [&R] (1,2,((3,6),11),(4,((7,14),10)),5,8,9,12,13,15) tree PAUP_9 = [&R] (1,2,((3,6),(((7,14),10),11)),4,5,8,9,12,13,15) tree PAUP_10 = [&R] (1,2,(3,6),(4,10),(5,9),((7,14),11),8,12,13,15) tree PAUP_11 = [&R] (1,2,(3,6),(4,((7,14),10)),5,8,9,11,12,13,15) tree PAUP_12 = [&R] (1,2,(3,6),4,5,(((7,14),10),11),8,9,12,13,15) tree PAUP_13 = [&R] (1,2,(((3,6),(7,14)),11),(4,10),(5,9),8,12,13,15) tree PAUP_14 = [&R] (1,2,((3,6),((7,14),10),11),4,5,8,9,12,13,15) tree PAUP_15 = [&R] (1,(2,(3,6)),(4,10),(5,9),((7,14),11),8,12,13,15) tree PAUP_16 = [&R] (1,2,((3,6),(7,14),11),(4,10),(5,9),8,12,13,15) tree PAUP_17 = [&R] (1,2,((3,6),11),(4,((7,14),10)),(5,9),8,12,13,15) tree PAUP_18 = [&R] (1,2,((3,6),(((7,14),10),11)),4,(5,9),8,12,13,15) tree PAUP_19 = [&R] (1,(2,(3,6)),(4,((7,14),10)),(5,9),8,11,12,13,15) tree PAUP_20 = [&R] (1,(2,(3,6)),4,(5,9),(((7,14),10),11),8,12,13,15) tree PAUP_21 = [&R] (1,2,((3,6),9),4,5,(((7,14),10),11),8,12,13,15) tree PAUP_22 = [&R] (1,2,((3,6),9),(4,((7,14),10)),5,8,11,12,13,15) tree PAUP_23 = [&R] (1,2,((3,6),11),4,5,((7,14),10),8,9,12,13,15) tree PAUP_24 = [&R] (1,(2,(3,6)),(4,((7,14),10)),5,8,9,11,12,13,15) tree PAUP_25 = [&R] (1,2,(((3,6),11),((7,14),10)),4,5,8,9,12,13,15) tree PAUP_26 = [&R] (1,2,(3,6),4,5,((7,14),10),8,9,11,12,13,15) 173 tree PAUP_27 = [&R] (1,(2,(3,6)),4,5,((7,14),10),8,9,11,12,13,15) tree PAUP_28 = [&R] (1,(2,(3,6)),4,5,(((7,14),10),11),8,9,12,13,15) tree PAUP_29 = [&R] (1,2,(3,6),(4,((7,14),10)),(5,9),8,11,12,13,15) tree PAUP_30 = [&R] (1,2,(3,6),4,(5,9),((7,14),10),8,11,12,13,15) tree PAUP_31 = [&R] (1,2,((3,6),9),4,5,((7,14),10),8,11,12,13,15) tree PAUP_32 = [&R] (1,2,(3,6),4,(5,9),(((7,14),10),11),8,12,13,15) tree PAUP_33 = [&R] (1,2,((3,6),((7,14),10),11),4,(5,9),8,12,13,15) tree PAUP_34 = [&R] (1,2,((3,6),11),4,(5,9),((7,14),10),8,12,13,15) tree PAUP_35 = [&R] (1,2,(((3,6),11),((7,14),10)),4,(5,9),8,12,13,15) tree PAUP_36 = [&R] (1,(2,(3,6)),4,(5,9),((7,14),10),8,11,12,13,15)

T2 tree PAUP_1 = [&R] (1,(2,8),(((((3,7),4),10),6),11,13),5,9,(12,14),15) tree PAUP_2 = [&R] (1,(2,8),(((3,7),4,10),6,11,13),5,9,(12,14),15) tree PAUP_3 = [&R] (1,(2,8),(((((3,7),4),10),6),11,13),(5,12,14),9,15) tree PAUP_4 = [&R] (1,(2,8),(((3,7),4,10),6,11,13),(5,12,14),9,15) tree PAUP_5 = [&R] (1,(2,8),(((((3,7),4),10),6),11,13),(5,(12,14)),9,15) tree PAUP_6 = [&R] (1,(2,8),(((3,7),4,10),6,11,13),(5,(12,14)),9,15) tree PAUP_7 = [&R] (1,(2,8),((((3,7),4),10),(6,11,13)),5,9,(12,14),15) tree PAUP_8 = [&R] (1,(2,8),(((3,7),4),10),5,(6,11,13),9,(12,14),15) tree PAUP_9 = [&R] (1,(2,8),((((3,7),4),10),(6,11,13)),(5,12,14),9,15) tree PAUP_10 = [&R] (1,(2,8),(((3,7),4),10),(5,12,14),(6,11,13),9,15) tree PAUP_11 = [&R] (1,(2,8),((((3,7),4),10),(6,11,13)),(5,(12,14)),9,15) tree PAUP_12 = [&R] (1,(2,8),(((3,7),4),10),(5,(12,14)),(6,11,13),9,15) tree PAUP_13 = [&R] (1,(2,8),((((3,7),(4,10)),6),11,13),5,9,(12,14),15) tree PAUP_14 = [&R] (1,(2,8),(((3,7),(4,10)),6,11,13),5,9,(12,14),15) tree PAUP_15 = [&R] (1,(2,8),((((3,7),(4,10)),6),11,13),(5,12,14),9,15) tree PAUP_16 = [&R] (1,(2,8),(((3,7),(4,10)),6,11,13),(5,12,14),9,15) tree PAUP_17 = [&R] (1,(2,8),((((3,7),(4,10)),6),11,13),(5,(12,14)),9,15) tree PAUP_18 = [&R] (1,(2,8),(((3,7),(4,10)),6,11,13),(5,(12,14)),9,15) tree PAUP_19 = [&R] (1,(2,8),(((((3,7),10),4),6),11,13),5,9,(12,14),15) tree PAUP_20 = [&R] (1,(2,8),((((3,7),10),4),6,11,13),5,9,(12,14),15) tree PAUP_21 = [&R] (1,(2,8),(((((3,7),10),4),6),11,13),(5,12,14),9,15) tree PAUP_22 = [&R] (1,(2,8),((((3,7),10),4),6,11,13),(5,12,14),9,15) tree PAUP_23 = [&R] (1,(2,8),(((((3,7),10),4),6),11,13),(5,(12,14)),9,15) tree PAUP_24 = [&R] (1,(2,8),((((3,7),10),4),6,11,13),(5,(12,14)),9,15)

174

T3 tree PAUP_1 = [&R] (1,2,(((((3,7),6),5),(9,11)),13),(4,10),8,(12,14),15) tree PAUP_2 = [&R] (1,(2,(12,14)),(((((3,7),6),5),(9,11)),13),(4,10),8,15) tree PAUP_3 = [&R] (1,2,((((3,7),6),5),(9,11),13),(4,10),8,(12,14),15) tree PAUP_4 = [&R] (1,(2,(12,14)),((((3,7),6),5),(9,11),13),(4,10),8,15) tree PAUP_5 = [&R] (1,(2,14),(((((3,7),6),5),(9,11)),13),(4,10),8,12,15) tree PAUP_6 = [&R] (1,((2,14),12),(((((3,7),6),5),(9,11)),13),(4,10),8,15) tree PAUP_7 = [&R] (1,(2,14),((((3,7),6),5),(9,11),13),(4,10),8,12,15) tree PAUP_8 = [&R] (1,((2,14),12),((((3,7),6),5),(9,11),13),(4,10),8,15) tree PAUP_9 = [&R] (1,(2,(4,10)),(((((3,7),6),5),(9,11)),13),8,(12,14),15) tree PAUP_10 = [&R] (1,(2,(4,10)),((((3,7),6),5),(9,11),13),8,(12,14),15)

Vicariance T0 tree PAUP_1 = [&R] (1,2,(3,10),((((4,7),13),6,11),5,9),8,(12,14),15) tree PAUP_2 = [&R] (1,2,((3,10),(((((4,7),13),6,11),9),5)),8,(12,14),15)

T1 tree PAUP_1 = [&R] (1,2,(3,9),(4,(((6,(7,14)),11),13)),(5,10),8,12,15) tree PAUP_2 = [&R] (1,(2,(3,9)),(4,(((6,(7,14)),11),13)),(5,10),8,12,15) tree PAUP_3 = [&R] (1,2,(3,9),(4,((6,((7,14),11)),13)),(5,10),8,12,15) tree PAUP_4 = [&R] (1,(2,(3,9)),(4,((6,((7,14),11)),13)),(5,10),8,12,15) tree PAUP_5 = [&R] (1,2,(3,9),(4,(((6,11),(7,14)),13)),(5,10),8,12,15) tree PAUP_6 = [&R] (1,(2,(3,9)),(4,(((6,11),(7,14)),13)),(5,10),8,12,15) tree PAUP_7 = [&R] (1,2,(3,9),((4,13),((6,11),(7,14))),(5,10),8,12,15) tree PAUP_8 = [&R] (1,(2,(3,9)),((4,13),((6,11),(7,14))),(5,10),8,12,15) tree PAUP_9 = [&R] (1,2,(3,9),(4,((6,11),(7,14)),13),(5,10),8,12,15) tree PAUP_10 = [&R] (1,(2,(3,9)),(4,((6,11),(7,14)),13),(5,10),8,12,15) tree PAUP_11 = [&R] (1,2,((3,9),(5,10)),(4,(((6,(7,14)),11),13)),8,12,15) tree PAUP_12 = [&R] (1,2,((3,9),(5,10)),(4,((6,((7,14),11)),13)),8,12,15) tree PAUP_13 = [&R] (1,2,((3,9),(5,10)),(4,(((6,11),(7,14)),13)),8,12,15) tree PAUP_14 = [&R] (1,2,((3,9),(5,10)),((4,13),((6,11),(7,14))),8,12,15) tree PAUP_15 = [&R] (1,2,((3,9),(5,10)),(4,((6,11),(7,14)),13),8,12,15) tree PAUP_16 = [&R] (1,2,((3,9),4,((6,11),(7,14)),13),(5,10),8,12,15) tree PAUP_17 = [&R] (1,2,((3,9),(4,13),((6,11),(7,14))),(5,10),8,12,15) tree PAUP_18 = [&R] (1,2,(((3,9),13),4,((6,11),(7,14))),(5,10),8,12,15) tree PAUP_19 = [&R] (1,2,(((3,9),((6,(7,14)),11),13),4),(5,10),8,12,15) 175 tree PAUP_20 = [&R] (1,2,(((3,9),(6,((7,14),11)),13),4),(5,10),8,12,15) tree PAUP_21 = [&R] (1,2,((((3,9),13),((6,11),(7,14))),4),(5,10),8,12,15) tree PAUP_22 = [&R] (1,2,((3,9),4,(((6,(7,14)),11),13)),(5,10),8,12,15) tree PAUP_23 = [&R] (1,2,((3,9),4,((6,((7,14),11)),13)),(5,10),8,12,15) tree PAUP_24 = [&R] (1,2,((3,9),4,(((6,11),(7,14)),13)),(5,10),8,12,15) tree PAUP_25 = [&R] (1,2,((3,9),(4,(((6,(7,14)),11),13))),(5,10),8,12,15)

tree PAUP_26 = [&R] (1,2,((3,9),(4,((6,((7,14),11)),13))),(5,10),8,12,15) tree PAUP_27 = [&R] (1,2,((3,9),(4,(((6,11),(7,14)),13))),(5,10),8,12,15) tree PAUP_28 = [&R] (1,2,((3,9),((4,13),((6,11),(7,14)))),(5,10),8,12,15) tree PAUP_29 = [&R] (1,2,((3,9),(4,((6,11),(7,14)),13)),(5,10),8,12,15) tree PAUP_30 = [&R] (1,2,(((3,9),(((6,(7,14)),11),13)),4),(5,10),8,12,15) tree PAUP_31 = [&R] (1,2,(((3,9),((6,((7,14),11)),13)),4),(5,10),8,12,15) tree PAUP_32 = [&R] (1,2,(((3,9),(((6,11),(7,14)),13)),4),(5,10),8,12,15) tree PAUP_33 = [&R] (1,2,(((3,9),((6,11),(7,14))),4,13),(5,10),8,12,15) tree PAUP_34 = [&R] (1,2,(((3,9),((6,11),(7,14))),(4,13)),(5,10),8,12,15) tree PAUP_35 = [&R] (1,2,((((3,9),((6,(7,14)),11)),13),4),(5,10),8,12,15) tree PAUP_36 = [&R] (1,2,((((3,9),(6,((7,14),11))),13),4),(5,10),8,12,15) tree PAUP_37 = [&R] (1,2,((((3,9),((6,11),(7,14))),13),4),(5,10),8,12,15) tree PAUP_38 = [&R] (1,2,((((3,6),9),13),4,((7,14),11)),(5,10),8,12,15) tree PAUP_39 = [&R] (1,2,(((((3,6),9),13),((7,14),11)),4),(5,10),8,12,15) tree PAUP_40 = [&R] (1,2,((((3,6),9),((7,14),11),13),4),(5,10),8,12,15)

T2 tree PAUP_1 = [&R] (1,(2,8),((((3,7),10),4),6,(11,13)),(5,9),12,14,15) tree PAUP_2 = [&R] (1,(2,8),((((3,7),10),4),6,(11,13)),(5,9),(12,14),15) tree PAUP_3 = [&R] (1,(2,8),((((3,7),10),4),(6,11,13)),(5,9),12,14,15) tree PAUP_4 = [&R] (1,(2,8),((((3,7),10),4),(6,(11,13))),(5,9),12,14,15) tree PAUP_5 = [&R] (1,(2,8),((((3,7),10),4),(6,11,13)),(5,9),(12,14),15) tree PAUP_6 = [&R] (1,(2,8),((((3,7),10),4),(6,(11,13))),(5,9),(12,14),15) tree PAUP_7 = [&R] (1,(2,8),(((((3,7),10),4),6),(11,13)),(5,9),12,14,15) tree PAUP_8 = [&R] (1,(2,8),(((((3,7),10),4),6),(11,13)),(5,9),(12,14),15) tree PAUP_9 = [&R] (1,(2,8),((((3,7),10),4),6),(5,9,(11,13)),12,14,15) tree PAUP_10 = [&R] (1,(2,8),((((3,7),10),4),6),(5,9,(11,13)),(12,14),15) tree PAUP_11 = [&R] (1,(2,8),(((((3,7),10),4),6,11,13),5,9),12,14,15) tree PAUP_12 = [&R] (1,(2,8),(((((3,7),10),4),6,(11,13)),(5,9)),12,14,15) tree PAUP_13 = [&R] (1,(2,8),(((((((3,7),10),4),11,13),6),9),5),12,14,15) tree PAUP_14 = [&R] (1,(2,8),(((((3,7),10),4),6,11,13),5,9),(12,14),15) 176 tree PAUP_15 = [&R] (1,(2,8),(((((3,7),10),4),6,(11,13)),(5,9)),(12,14),15) tree PAUP_16 = [&R] (1,(2,8),(((((((3,7),10),4),11,13),6),9),5),(12,14),15) tree PAUP_17 = [&R] (1,(2,8),(((((3,7),10),4),(6,11,13)),(5,9)),12,14,15) tree PAUP_18 = [&R] (1,(2,8),(((((3,7),10),4),(6,(11,13))),(5,9)),12,14,15) tree PAUP_19 = [&R] (1,(2,8),(((((3,7),10),4),(6,11,13)),(5,9)),(12,14),15) tree PAUP_20 = [&R] (1,(2,8),(((((3,7),10),4),(6,(11,13))),(5,9)),(12,14),15) tree PAUP_21 = [&R] (1,(2,8),((((((3,7),10),4),6),11,13),5,9),12,14,15) tree PAUP_22 = [&R] (1,(2,8),((((((3,7),10),4),6),(11,13)),(5,9)),12,14,15)

tree PAUP_23 = [&R] (1,(2,8),((((((3,7),10),4),6),11,13),5,9),(12,14),15) tree PAUP_24 = [&R] (1,(2,8),((((((3,7),10),4),6),(11,13)),(5,9)),(12,14),15) tree PAUP_25 = [&R] (1,(2,8),(((((3,7),10),4),6),((5,9),11,13)),12,14,15) tree PAUP_26 = [&R] (1,(2,8),(((((3,7),10),4),6),((5,9),(11,13))),12,14,15) tree PAUP_27 = [&R] (1,(2,8),(((((3,7),10),4),6),((5,9),11,13)),(12,14),15) tree PAUP_28 = [&R] (1,(2,8),(((((3,7),10),4),6),((5,9),(11,13))),(12,14),15) tree PAUP_29 = [&R] (1,(2,8),((((3,7),10),4),(11,13)),(5,(6,9)),12,14,15) tree PAUP_30 = [&R] (1,(2,8),((((3,7),10),4),(11,13)),(5,(6,9)),(12,14),15) tree PAUP_31 = [&R] (1,(2,8),(((3,7),10),4),(5,((6,9),11,13)),12,14,15) tree PAUP_32 = [&R] (1,(2,8),(((3,7),10),4),(5,((6,9),(11,13))),12,14,15) tree PAUP_33 = [&R] (1,(2,8),(((3,7),10),4),(5,(6,9),(11,13)),12,14,15) tree PAUP_34 = [&R] (1,(2,8),(((3,7),10),4),(5,((6,9),11,13)),(12,14),15) tree PAUP_35 = [&R] (1,(2,8),(((3,7),10),4),(5,((6,9),(11,13))),(12,14),15) tree PAUP_36 = [&R] (1,(2,8),(((3,7),10),4),(5,(6,9),(11,13)),(12,14),15) tree PAUP_37 = [&R] (1,(2,8),(((3,7),10),4),(5,(6,11,13),9),12,14,15) tree PAUP_38 = [&R] (1,(2,8),(((3,7),10),4),(5,((6,(11,13)),9)),12,14,15) tree PAUP_39 = [&R] (1,(2,8),(((3,7),10),4),(5,(6,11,13),9),(12,14),15) tree PAUP_40 = [&R] (1,(2,8),(((3,7),10),4),(5,((6,(11,13)),9)),(12,14),15) tree PAUP_41 = [&R] (1,((2,8),(((3,7),10),4)),(5,((6,9),11,13)),12,14,15) tree PAUP_42 = [&R] (1,((2,8),(((3,7),10),4)),(5,((6,9),(11,13))),12,14,15) tree PAUP_43 = [&R] (1,((2,8),(((3,7),10),4)),(5,(6,9),(11,13)),12,14,15) tree PAUP_44 = [&R] (1,((2,8),(((3,7),10),4)),(5,((6,9),11,13)),(12,14),15) tree PAUP_45 = [&R] (1,((2,8),(((3,7),10),4)),(5,((6,9),(11,13))),(12,14),15) tree PAUP_46 = [&R] (1,((2,8),(((3,7),10),4)),(5,(6,9),(11,13)),(12,14),15) tree PAUP_47 = [&R] (1,((2,8),(((3,7),10),4)),(5,(6,11,13),9),12,14,15) tree PAUP_48 = [&R] (1,((2,8),(((3,7),10),4)),(5,((6,(11,13)),9)),12,14,15) tree PAUP_49 = [&R] (1,((2,8),(((3,7),10),4)),(5,(6,11,13),9),(12,14),15) tree PAUP_50 = [&R] (1,((2,8),(((3,7),10),4)),(5,((6,(11,13)),9)),(12,14),15) 177 tree PAUP_51 = [&R] (1,(2,8),(((((3,7),10),4),11,13),5,(6,9)),12,14,15) tree PAUP_52 = [&R] (1,(2,8),(((((3,7),10),4),(11,13)),(5,(6,9))),12,14,15) tree PAUP_53 = [&R] (1,(2,8),((((((3,7),10),4),11,13),(6,9)),5),12,14,15) tree PAUP_54 = [&R] (1,(2,8),(((((3,7),10),4),(6,9),11,13),5),12,14,15) tree PAUP_55 = [&R] (1,(2,8),((((3,7),10),4),(5,(6,9),11,13)),12,14,15) tree PAUP_56 = [&R] (1,(2,8),((((3,7),10),4),(5,((6,9),11,13))),12,14,15) tree PAUP_57 = [&R] (1,(2,8),((((3,7),10),4),(5,((6,9),(11,13)))),12,14,15) tree PAUP_58 = [&R] (1,(2,8),((((3,7),10),4),(5,(6,9),(11,13))),12,14,15) tree PAUP_59 = [&R] (1,(2,8),((((3,7),10),4),((5,(6,9)),11,13)),12,14,15) tree PAUP_60 = [&R] (1,(2,8),((((3,7),10),4),((5,(6,9)),(11,13))),12,14,15) tree PAUP_61 = [&R] (1,(2,8),((((3,7),10),4),(((5,9),6),11,13)),12,14,15)

tree PAUP_62 = [&R] (1,(2,8),((((3,7),10),4),((5,9),6,11,13)),12,14,15) tree PAUP_63 = [&R] (1,(2,8),((((3,7),10),4),((6,9),11,13)),5,12,14,15) tree PAUP_64 = [&R] (1,(2,8),(((((3,7),10),4),((6,9),11,13)),5),12,14,15) tree PAUP_65 = [&R] (1,(2,8),((((3,7),10),4),((5,9),(6,11,13))),12,14,15) tree PAUP_66 = [&R] (1,(2,8),((((3,7),10),4),(((5,9),11,13),6)),12,14,15) tree PAUP_67 = [&R] (1,(2,8),(((((3,7),10),4),11,13),5,(6,9)),(12,14),15) tree PAUP_68 = [&R] (1,(2,8),(((((3,7),10),4),(11,13)),(5,(6,9))),(12,14),15) tree PAUP_69 = [&R] (1,(2,8),((((((3,7),10),4),11,13),(6,9)),5),(12,14),15) tree PAUP_70 = [&R] (1,(2,8),(((((3,7),10),4),(6,9),11,13),5),(12,14),15) tree PAUP_71 = [&R] (1,(2,8),((((3,7),10),4),(5,(6,9),11,13)),(12,14),15) tree PAUP_72 = [&R] (1,(2,8),((((3,7),10),4),(5,((6,9),11,13))),(12,14),15) tree PAUP_73 = [&R] (1,(2,8),((((3,7),10),4),(5,((6,9),(11,13)))),(12,14),15) tree PAUP_74 = [&R] (1,(2,8),((((3,7),10),4),(5,(6,9),(11,13))),(12,14),15) tree PAUP_75 = [&R] (1,(2,8),((((3,7),10),4),((5,(6,9)),11,13)),(12,14),15) tree PAUP_76 = [&R] (1,(2,8),((((3,7),10),4),((5,(6,9)),(11,13))),(12,14),15) tree PAUP_77 = [&R] (1,(2,8),((((3,7),10),4),(((5,9),6),11,13)),(12,14),15) tree PAUP_78 = [&R] (1,(2,8),((((3,7),10),4),((5,9),6,11,13)),(12,14),15) tree PAUP_79 = [&R] (1,(2,8),((((3,7),10),4),((6,9),11,13)),5,(12,14),15) tree PAUP_80 = [&R] (1,(2,8),(((((3,7),10),4),((6,9),11,13)),5),(12,14),15) tree PAUP_81 = [&R] (1,(2,8),((((3,7),10),4),((5,9),(6,11,13))),(12,14),15) tree PAUP_82 = [&R] (1,(2,8),((((3,7),10),4),(((5,9),11,13),6)),(12,14),15) tree PAUP_83 = [&R] (1,(2,8),((((3,7),10),4),((5,9),11,13),6),12,14,15) tree PAUP_84 = [&R] (1,(2,8),((((3,7),10),4),((5,9),11,13),6),(12,14),15)

178

T3 tree PAUP_1 = [&R] (1,2,(((3,7),4),10),((5,9),((6,11),13)),8,12,14,15) tree PAUP_2 = [&R] (1,(2,8),(((3,7),4),10),((5,9),((6,11),13)),12,14,15) tree PAUP_3 = [&R] (1,(2,(((3,7),4),10)),((5,9),((6,11),13)),8,12,14,15)

179

APPENDIX 7: SAMPLE BIOGEOBEARS CODE FOR EOCHONETES

install.packages("BioGeoBEARS", dependencies=TRUE, repos="http://cran.rstudio.com") library(optimx) library(FD) library(parallel) library(BioGeoBEARS) source("http://phylo.wdfiles.com/local--files/biogeobears/BioGeoBEARS_basics_v1.R") source("http://phylo.wdfiles.com/local--files/biogeobears/BioGeoBEARS_classes_v1.R") source("http://phylo.wdfiles.com/local--files/biogeobears/BioGeoBEARS_models_v1.R") source("http://phylo.wdfiles.com/local--files/biogeobears/BioGeoBEARS_plots_v1.R") source("http://phylo.wdfiles.com/local--files/biogeobears/BioGeoBEARS_readwrite_v1.R") source("http://phylo.wdfiles.com/local--files/biogeobears/BioGeoBEARS_simulate_v1.R") source("http://phylo.wdfiles.com/local--files/biogeobears/BioGeoBEARS_stratified_v1.R") source("http://phylo.wdfiles.com/local--files/biogeobears/BioGeoBEARS_univ_model_v1.R") source("http://phylo.wikidot.com/local--files/biogeobears/calc_loglike_sp_v01.R") calc_loglike_sp = compiler::cmpfun(calc_loglike_sp_prebyte) # crucial to fix bug in uppass calculations calc_independent_likelihoods_on_each_branch = compiler::cmpfun(calc_independent_likelihoods_on_each_branch_prebyte) getwd() extdata_dir = np(system.file("extdata", package="BioGeoBEARS")) extdata_dir list.files(extdata_dir) trfn = np(paste(addslash(extdata_dir), "Eochonetes_newick.tre", sep="")) moref(trfn) tr = read.tree(trfn) tr plot(tr) title("Eochonetes Phylogeny") axisPhylo() geogfn = np(paste(addslash(extdata_dir), "Eochonetes_geog.txt", sep="")) moref(geogfn) tipranges = getranges_from_LagrangePHYLIP(lgdata_fn=geogfn) tipranges max_range_size = 2 numstates_from_numareas(numareas=4, maxareas=4, include_null_range=TRUE) numstates_from_numareas(numareas=4, maxareas=4, include_null_range=FALSE) numstates_from_numareas(numareas=4, maxareas=3, include_null_range=TRUE) numstates_from_numareas(numareas=4, maxareas=2, include_null_range=TRUE)

180 numstates_from_numareas(numareas=7, maxareas=2, include_null_range=TRUE) numstates_from_numareas(numareas=7, maxareas=2, include_null_range=TRUE)

####################################################### # DEC AND DEC+J ANALYSIS #######################################################

####################################################### # Run DEC #######################################################

BioGeoBEARS_run_object = define_BioGeoBEARS_run() BioGeoBEARS_run_object$force_sparse=FALSE BioGeoBEARS_run_object$speedup=TRUE BioGeoBEARS_run_object$calc_ancprobs=TRUE

BioGeoBEARS_run_object$max_range_size = max_range_size

BioGeoBEARS_run_object$num_cores_to_use=1

BioGeoBEARS_run_object$force_sparse=FALSE

BioGeoBEARS_run_object$geogfn = geogfn

BioGeoBEARS_run_object$trfn = trfn

BioGeoBEARS_run_object = readfiles_BioGeoBEARS_run(BioGeoBEARS_run_object)

BioGeoBEARS_run_object$return_condlikes_table = TRUE BioGeoBEARS_run_object$calc_TTL_loglike_from_condlikes_table = TRUE BioGeoBEARS_run_object$calc_ancprobs = TRUE

BioGeoBEARS_run_object

BioGeoBEARS_run_object$BioGeoBEARS_model_object

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table check_BioGeoBEARS_run(BioGeoBEARS_run_object) runslow = TRUE resfn = "Eochonetes_DEC_M0_unconstrained_v1.Rdata" if (runslow)

{ res = bears_optim_run(BioGeoBEARS_run_object) res

save(res, file=resfn) resDEC = res 181

} else { load(resfn) resDEC = res }

####################################################### # Run DEC+J #######################################################

BioGeoBEARS_run_object = define_BioGeoBEARS_run() BioGeoBEARS_run_object$trfn = trfn BioGeoBEARS_run_object$geogfn = geogfn BioGeoBEARS_run_object$max_range_size = max_range_size

BioGeoBEARS_run_object$use_optimx = TRUE BioGeoBEARS_run_object$num_cores_to_use=1

BioGeoBEARS_run_object$force_sparse=FALSE BioGeoBEARS_run_object$speedup=TRUE BioGeoBEARS_run_object$calc_ancprobs=TRUE

BioGeoBEARS_run_object = readfiles_BioGeoBEARS_run(BioGeoBEARS_run_object)

BioGeoBEARS_run_object$return_condlikes_table = TRUE BioGeoBEARS_run_object$calc_TTL_loglike_from_condlikes_table = TRUE BioGeoBEARS_run_object$calc_ancprobs = TRUE dstart = resDEC$outputs@params_table["d","est"] estart = resDEC$outputs@params_table["e","est"] jstart = 0.0001

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["d","init"] = dstart BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["d","est"] = dstart BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["e","init"] = estart BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["e","est"] = estart

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","type"] = "free" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","init"] = jstart BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","est"] = jstart check_BioGeoBEARS_run(BioGeoBEARS_run_object) resfn = "Eochonetes_DEC+J_M0_unconstrained_v1.Rdata" runslow = TRUE if (runslow) { #sourceall("/Dropbox/_njm/__packages/BioGeoBEARS_setup/")

res = bears_optim_run(BioGeoBEARS_run_object) res 182

save(res, file=resfn)

resDECj = res } else { load(resfn) resDECj = res } ####################################################### # PDF plots ####################################################### pdffn = "Eochonetes_DEC_vs_DEC+J_M0_unconstrained_v1.pdf" pdf(pdffn, width=8.5, height=11)

####################################################### # Plot ancestral states - DEC ####################################################### analysis_titletxt ="BioGeoBEARS DEC on Eochonetes M0_unconstrained" results_object = resDEC scriptdir = np(system.file("extdata/a_scripts", package="BioGeoBEARS")) res2 = plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="text", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="pie", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges)

####################################################### # Plot ancestral states - DECJ ####################################################### analysis_titletxt ="BioGeoBEARS DEC+J on Eochonetes M0_unconstrained" results_object = resDECj scriptdir = np(system.file("extdata/a_scripts", package="BioGeoBEARS")) res1 = plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="text", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="pie", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) dev.off() 183 cmdstr = paste("open ", pdffn, sep="") system(cmdstr)

####################################################### # DIVALIKE AND DIVALIKE+J ANALYSIS ####################################################### ####################################################### # Run DIVALIKE #######################################################

BioGeoBEARS_run_object = define_BioGeoBEARS_run() BioGeoBEARS_run_object$trfn = trfn BioGeoBEARS_run_object$geogfn = geogfn BioGeoBEARS_run_object$max_range_size = max_range_size

BioGeoBEARS_run_object$use_optimx = TRUE BioGeoBEARS_run_object$num_cores_to_use=1

BioGeoBEARS_run_object$force_sparse=FALSE BioGeoBEARS_run_object$speedup=TRUE BioGeoBEARS_run_object$calc_ancprobs=TRUE

BioGeoBEARS_run_object = readfiles_BioGeoBEARS_run(BioGeoBEARS_run_object)

BioGeoBEARS_run_object$return_condlikes_table = TRUE BioGeoBEARS_run_object$calc_TTL_loglike_from_condlikes_table = TRUE BioGeoBEARS_run_object$calc_ancprobs = TRUE

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["s","type"] = "fixed" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["s","init"] = 0.0 BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["s","est"] = 0.0

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["ysv","type"] = "2-j" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["ys","type"] = "ysv*1/2" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["y","type"] = "ysv*1/2" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["v","type"] = "ysv*1/2"

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["mx01v","type"] = "fixed" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["mx01v","init"] = 0.5 BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["mx01v","est"] = 0.5

# BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","type"] = "free" # BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","init"] = 0.01 # BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","est"] = 0.01 check_BioGeoBEARS_run(BioGeoBEARS_run_object) runslow = TRUE resfn = "Eochonetes_DIVALIKE_M0_unconstrained_v1.Rdata" if (runslow) 184

{ res = bears_optim_run(BioGeoBEARS_run_object) res

save(res, file=resfn) resDIVALIKE = res } else { load(resfn) resDIVALIKE = res }

####################################################### # Run DIVALIKE+J #######################################################

BioGeoBEARS_run_object = define_BioGeoBEARS_run() BioGeoBEARS_run_object$trfn = trfn BioGeoBEARS_run_object$geogfn = geogfn BioGeoBEARS_run_object$max_range_size = max_range_size

BioGeoBEARS_run_object$use_optimx = TRUE BioGeoBEARS_run_object$num_cores_to_use=1

BioGeoBEARS_run_object$force_sparse=FALSE BioGeoBEARS_run_object$speedup=TRUE BioGeoBEARS_run_object$calc_ancprobs=TRUE

BioGeoBEARS_run_object = readfiles_BioGeoBEARS_run(BioGeoBEARS_run_object)

BioGeoBEARS_run_object$return_condlikes_table = TRUE BioGeoBEARS_run_object$calc_TTL_loglike_from_condlikes_table = TRUE BioGeoBEARS_run_object$calc_ancprobs = TRUE dstart = resDIVALIKE$outputs@params_table["d","est"] estart = resDIVALIKE$outputs@params_table["e","est"] jstart = 0.0001

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","min"] = 0.00001 BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","max"] = 1.99999 check_BioGeoBEARS_run(BioGeoBEARS_run_object) resfn = "Eochonetes_DIVALIKE+J_M0_unconstrained_v1.Rdata" runslow = TRUE if (runslow) {

res = bears_optim_run(BioGeoBEARS_run_object) res

save(res, file=resfn)

resDIVALIKEj = res } else { load(resfn) resDIVALIKEj = res } pdffn = "Eochonetes_DIVALIKE_vs_DIVALIKE+J_M0_unconstrained_v1.pdf" pdf(pdffn, width=8.5, height=11)

####################################################### # Plot ancestral states - DIVALIKE ####################################################### analysis_titletxt ="BioGeoBEARS DIVALIKE on Eochonetes M0_unconstrained" results_object = resDIVALIKE scriptdir = np(system.file("extdata/a_scripts", package="BioGeoBEARS")) res2 = plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="text", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="pie", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges)

186

####################################################### # Plot ancestral states - DIVALIKE+J ####################################################### analysis_titletxt ="BioGeoBEARS DIVALIKE+J on Eochonetes M0_unconstrained" results_object = resDIVALIKEj scriptdir = np(system.file("extdata/a_scripts", package="BioGeoBEARS")) res1 = plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="text", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="pie", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) dev.off() cmdstr = paste("open ", pdffn, sep="") system(cmdstr)

####################################################### # BAYAREALIKE AND BAYAREALIKE+J ANALYSIS #######################################################

####################################################### # Run BAYAREALIKE #######################################################

BioGeoBEARS_run_object = define_BioGeoBEARS_run() BioGeoBEARS_run_object$trfn = trfn BioGeoBEARS_run_object$geogfn = geogfn BioGeoBEARS_run_object$max_range_size = max_range_size

BioGeoBEARS_run_object$use_optimx = TRUE BioGeoBEARS_run_object$num_cores_to_use=1

BioGeoBEARS_run_object$force_sparse=FALSE BioGeoBEARS_run_object$speedup=TRUE BioGeoBEARS_run_object$calc_ancprobs=TRUE

BioGeoBEARS_run_object = readfiles_BioGeoBEARS_run(BioGeoBEARS_run_object)

BioGeoBEARS_run_object$return_condlikes_table = TRUE BioGeoBEARS_run_object$calc_TTL_loglike_from_condlikes_table = TRUE BioGeoBEARS_run_object$calc_ancprobs = TRUE

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["v","type"] = "fixed" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["v","init"] = 0.0 BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["v","est"] = 0.0

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["ysv","type"] = "1-j" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["ys","type"] = "ysv*1/1" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["y","type"] = "1-j"

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["mx01y","type"] = "fixed" BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["mx01y","init"] = 0.9999 BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["mx01y","est"] = 0.9999 check_BioGeoBEARS_run(BioGeoBEARS_run_object) runslow = TRUE resfn = "Eochonetes_BAYAREALIKE_M0_unconstrained_v1.Rdata" if (runslow) { res = bears_optim_run(BioGeoBEARS_run_object) res

save(res, file=resfn) resBAYAREALIKE = res } else { load(resfn) resBAYAREALIKE = res } ####################################################### # Run BAYAREALIKE+J #######################################################

BioGeoBEARS_run_object = define_BioGeoBEARS_run() BioGeoBEARS_run_object$trfn = trfn BioGeoBEARS_run_object$geogfn = geogfn BioGeoBEARS_run_object$max_range_size = max_range_size

BioGeoBEARS_run_object$use_optimx = TRUE BioGeoBEARS_run_object$num_cores_to_use=1

BioGeoBEARS_run_object$force_sparse=FALSE BioGeoBEARS_run_object$speedup=TRUE BioGeoBEARS_run_object$calc_ancprobs=TRUE

BioGeoBEARS_run_object = readfiles_BioGeoBEARS_run(BioGeoBEARS_run_object)

BioGeoBEARS_run_object$return_condlikes_table = TRUE BioGeoBEARS_run_object$calc_TTL_loglike_from_condlikes_table = TRUE BioGeoBEARS_run_object$calc_ancprobs = TRUE

188 dstart = resBAYAREALIKE$outputs@params_table["d","est"] estart = resBAYAREALIKE$outputs@params_table["e","est"] jstart = 0.0001

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","max"] = 0.99999

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["d","min"] = 0.0000001 BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["d","max"] = 4.9999999

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["e","min"] = 0.0000001 BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["e","max"] = 4.9999999

BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","min"] = 0.00001 BioGeoBEARS_run_object$BioGeoBEARS_model_object@params_table["j","max"] = 0.99999 check_BioGeoBEARS_run(BioGeoBEARS_run_object) resfn = "Eochonetes_BAYAREALIKE+J_M0_unconstrained_v1.Rdata" runslow = TRUE if (runslow) { res = bears_optim_run(BioGeoBEARS_run_object) res

save(res, file=resfn) 189

resBAYAREALIKEj = res } else { load(resfn) resBAYAREALIKEj = res } pdffn = "Eochonetes_BAYAREALIKE_vs_BAYAREALIKE+J_M0_unconstrained_v1.pdf" pdf(pdffn, width=8.5, height=11)

####################################################### # Plot ancestral states - BAYAREALIKE ####################################################### analysis_titletxt ="BioGeoBEARS BAYAREALIKE on Eochonetes M0_unconstrained" results_object = resBAYAREALIKE scriptdir = np(system.file("extdata/a_scripts", package="BioGeoBEARS")) res2 = plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="text", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="pie", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges)

####################################################### # Plot ancestral states - BAYAREALIKE+J ####################################################### analysis_titletxt ="BioGeoBEARS BAYAREALIKE+J on Eochonetes M0_unconstrained" results_object = resBAYAREALIKEj scriptdir = np(system.file("extdata/a_scripts", package="BioGeoBEARS")) res1 = plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="text", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) plot_BioGeoBEARS_results(results_object, analysis_titletxt, addl_params=list("j"), plotwhat="pie", label.offset=0.45, tipcex=0.7, statecex=0.7, splitcex=0.6, titlecex=0.8, plotsplits=FALSE, cornercoords_loc=scriptdir, include_null_range=TRUE, tr=tr, tipranges=tipranges) dev.off() cmdstr = paste("open ", pdffn, sep="") system(cmdstr)

######################################################################### 190

######################################################################### # CALCULATE SUMMARY STATISTICS TO COMPARE # DEC, DEC+J, DIVALIKE, DIVALIKE+J, BAYAREALIKE, BAYAREALIKE+J ######################################################################### ######################################################################### restable = NULL teststable = NULL

####################################################### # Statistics -- DEC vs. DEC+J #######################################################

LnL_2 = get_LnL_from_BioGeoBEARS_results_object(resDEC) LnL_1 = get_LnL_from_BioGeoBEARS_results_object(resDECj) numparams1 = 3 numparams2 = 2 stats = AICstats_2models(LnL_1, LnL_2, numparams1, numparams2) stats res2 = extract_params_from_BioGeoBEARS_results_object(results_object=resDEC, returnwhat="table", addl_params=c("j"), paramsstr_digits=4) res1 = extract_params_from_BioGeoBEARS_results_object(results_object=resDECj, returnwhat="table", addl_params=c("j"), paramsstr_digits=4) rbind(res2, res1) tmp_tests = conditional_format_table(stats) restable = rbind(restable, res2, res1) teststable = rbind(teststable, tmp_tests)

####################################################### # Statistics -- DIVALIKE vs. DIVALIKE+J #######################################################

LnL_2 = get_LnL_from_BioGeoBEARS_results_object(resDIVALIKE) LnL_1 = get_LnL_from_BioGeoBEARS_results_object(resDIVALIKEj) numparams1 = 3 numparams2 = 2 stats = AICstats_2models(LnL_1, LnL_2, numparams1, numparams2) stats res2 = extract_params_from_BioGeoBEARS_results_object(results_object=resDIVALIKE, returnwhat="table", addl_params=c("j"), paramsstr_digits=4) res1 = extract_params_from_BioGeoBEARS_results_object(results_object=resDIVALIKEj, returnwhat="table", addl_params=c("j"), paramsstr_digits=4) 191

rbind(res2, res1) conditional_format_table(stats) tmp_tests = conditional_format_table(stats) restable = rbind(restable, res2, res1) teststable = rbind(teststable, tmp_tests)

####################################################### # Statistics -- BAYAREALIKE vs. BAYAREALIKE+J #######################################################

LnL_2 = get_LnL_from_BioGeoBEARS_results_object(resBAYAREALIKE) LnL_1 = get_LnL_from_BioGeoBEARS_results_object(resBAYAREALIKEj) numparams1 = 3 numparams2 = 2 stats = AICstats_2models(LnL_1, LnL_2, numparams1, numparams2) stats res2 = extract_params_from_BioGeoBEARS_results_object(results_object=resBAYAREALIKE, returnwhat="table", addl_params=c("j"), paramsstr_digits=4) res1 = extract_params_from_BioGeoBEARS_results_object(results_object=resBAYAREALIKEj, returnwhat="table", addl_params=c("j"), paramsstr_digits=4) rbind(res2, res1) conditional_format_table(stats) tmp_tests = conditional_format_table(stats) restable = rbind(restable, res2, res1) teststable = rbind(teststable, tmp_tests) ######################################################################### # RESULTS: DEC, DEC+J, DIVALIKE, DIVALIKE+J, BAYAREALIKE, BAYAREALIKE+J ######################################################################### teststable$alt = c("DEC+J", "DIVALIKE+J", "BAYAREALIKE+J") teststable$null = c("DEC", "DIVALIKE", "BAYAREALIKE") row.names(restable) = c("DEC", "DEC+J", "DIVALIKE", "DIVALIKE+J", "BAYAREALIKE", "BAYAREALIKE+J") restable teststable save(restable, file="Eochonetes_restable_v1.Rdata") load(file="Eochonetes_restable_v1.Rdata") save(teststable, file="Eochonetes_teststable_v1.Rdata") load(file="Eochonetes_teststable_v1.Rdata") 192

APPENDIX 8: BIOGEOBEARS PROBABILITY MODELS

Models generated from the R program BioGeoBEARS for brachiopod and trilobite clades used within this study. A color-coded key to areas used within each phylogeny proceeds the models produced. Models are separated by method as follows: DEC, DEC+J, DivaLike, DivaLike+J, BayAreaLike, BayAreaLike+J. Within each method, optimized ancestral states are first, followed by probability of optimized ancestral states.

Brachiopods

Eochonetes

Glyptorthis

Hebertella

Plaesiomys

Trilobites

Bumastoides

Deiphoninae

Flexicalymene

Homalonotidae

Tetralichinae

Thaleops

193

Eochonetes: Key to areas for BioGeoBEARS models

194

195

196

197

198

199

200

201

202

203

204

205

206

Glyptorthis: Key to areas for BioGeoBEARS models

207

208

209

210

211

212

213

214

215

216

217

218

219

Hebertella: Key to areas for BioGeoBEARS models

220

221

222

223

224

225

226

227

228

229

230

231

232

Plaesiomys: Key to areas for BioGeoBEARS models

233

234

235

236

237

238

239

240

241

242

243

244

245

Bumastoides: Key to Areas for BioGeoBears Models

246

247

248

249

250

251

252

253

254

255

256

257

258

Deiphoninae: Key to areas for BioGeoBEARS models

259

260

261

262

263

264

265

266

267

268

269

270

271

Flexicalymene: Key to areas for BioGeoBEARS models

272

273

274

275

276

277

278

279

280

281

282

283

284

Homalonotidae: Key to areas for BioGeoBEARS models

285

286

287

288

289

290

291

292

293

294

295

296

297

Tetralichinae: Key to areas for BioGeoBEARS models

298

299

300

301

302

303

304

305

306

307

308

309

310

Thaleops: Key to areas for BioGeoBEARS models

311

312

313

314

315

316

317

318

319

320

321

322

00 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

Thesis and Dissertation Services ! !