sign in sign up
<<
Home , Brachypterygius, Coracoid, Excalibosaurus, Ghost lineage, Ichthyopterygia, Ichthyosaur

A TOTAL EVIDENCE ANALYSIS OF THE EVOLUTIONARY HISTORY OF THE THUNNOSAUR

Jessica Danielle Lawrence

A Thesis

Submitted to the Graduate College of Bowling Green State University in partial fulfillment of the requirements for the degree of

MASTER OF SCIENCE

December 2008

Committee:

Dr. Margaret (Peg) Yacobucci, Advisor

Dr. Dan Pavuk

Dr. James E. Evans

© 2008

Jessica Danielle Lawrence

All Rights Reserved iii

ABSTRACT

Margaret (Peg) Yacobucci, Advisor

Ichthyosaurs first appear in the Early with an elongate, -shaped anatomy. The most derived ichthyosaurs, including the to Early

Eurhinosauria and the Late Triassic to Mid Thunnosauria, evolved more streamlined fish-shaped bodies. A -level cladistic analysis of this derived group of ichthyosaurs, using 60 characters for 17 taxa and one ingroup , was conducted using PAUP* (Swofford, 1998). The new analysis is compared to that of

Motani’s (1999b) -level analysis of ichthyosaurs by looking at the -defining character state changes in each. I use a total evidence approach, meaning a detailed phylogenetic analysis plus data on stratigraphic and geographic occurrences, to answer two questions. First, is it more plausible to place as a sister to the

Ophthalmosauria or to permit the longer ghost for Ophthalmosauria that would be required if is its sister taxon, a question suggested by Motani (1999b)? A incorporating known stratigraphic ranges was constructed, based on the species-level cladistic analysis, to help quantify the stratigraphic debt using the relative completeness index (RCI), stratigraphic congruency index (SCI) and the gap excess ratio

(GER). When a species-level analysis was performed, four previously defined fell apart, including , Stenopterygius, Eurhinosauria and Ophthalmosauria.

Hence, Motani’s (1999b) suggestion that Stenopterygius and Ophthalmosauria be placed as sister taxa is not supported by this analysis; nor is placing Ichthyosaurus as the supported. Rather, the ophthalmosaurian taxa were pulled down the tree in the iv

species-level analysis, creating even longer ghost lineages than were previously present.

These longer ghost lineages are the primary cause of the observed stratigraphic debt, along with Eurhinosaurus longirostris, Leptonectes solei, and costini.

Despite these increases, overall the species-level phylogeny is more stratigraphically parsimonious than Motani’s (1999b) genus-level one. The second question this study addressed was to determine whether the radiation of the Ophthalmosauria was influenced by the opening of new marine habitats from the widening of the Atlantic and/or the appearance of morphological novelties. The radiation of the Ophthalmosauria into North and South America coincides with the widening of the North and South Atlantic during the , as well as nine synapomorphies involving changes to the ear and lower jaw region, numerous changes in the forefins (including extra digits) and . These characters may be associated with to more efficient swimming in open ocean habitats, helping the groups’ biogeographic dispersion. Hence, both tectonic events and the of morphological innovations likely influenced the radiation of the

Ophthalmosauria. v

This paper is dedicated to my family with a special thanks to my Mom, Kim R.

Lawrence. Without her constant love, support, and encouragement this paper would not

have gotten done. I would also like to sincerely thank my Grandma Helen Franz

Lawrence, Grandpa Loren Eugene Richardson, and Grandma Alice Faye Richardson.

Their love and support of my education has been overwhelming and without it I know I would not be where I am today. Grandma and Grandpa Richardson, I wish you were still

here to see what all your love and support has helped me accomplish. vi

ACKNOWLEDGMENTS

I would first and foremost like to thank my advisor Peg Yacobucci, without her guidance I would have been lost at so many stages of this project. Thank you for always taking the time to answer my questions, even when you were very busy and the explanation took a bit longer than expected. Thank you to my committee members for taking the time to help make my thesis stronger. I would also like to thank the Geology department for their support as well as the Richard D. Hoare Research Scholarship and

Katzner Bookstore Award for funding my research. I would also like to thank Kevin

Seymour (ROM) and Patricia Holroyd (UCMP) for their help when looking at specimens; they were an amazing help, especially with my quick day visit to Toronto. A big thank you to all my fellow graduate students for being there to listen, help and provide a fun outlet for all the frustrations that school can bring, I will never forget these times. A special thanks to Mary Scanlan for being an amazing office mate, however short it was and for helping me get through my first , as well as listening to my rants about this paper. A special thanks to Kelsey Garner and Megan Castles who made this second year incredibly entertaining. I would also like to thank my family again, for listening to my frustrations and helping to keep me motivated when I was in need of it. Thanks to my mother for traveling with me to so that I did not have to make that drive by myself. vi

TABLE OF CONTENTS

Page

CHAPTER I. INTRODUCTION...... 1

Background...... 2

Objectives ...... 10

CHAPTER II. METHODS ...... 12

Taxa and Characters...... 12

Phylogenetic Analysis...... 13

Stratigraphic Debt Analysis ...... 15

CHAPTER III. RESULTS...... 19

Phylogenetic Analysis...... 19

PAUP ...... 19

Evolutionary Tree...... 26

Phylogenetic vs. Stratigraphic Parsimony ...... 29

Radiation of the Ophthalmosauria ...... 30

CHAPTER IV. DISCUSSION...... 31

CHAPTER V. SUMMARY AND CONCLUSIONS ...... 35

REFERENCES ...... 37

APPENDIX A. CHARACTER DESCRIPTIONS ...... 43

APPENDIX B. SPECIES DATA ...... 51

APPENDIX C. UNINFORMATIVE CHARACTERS ...... 56

APPENDIX D. CALCULATING MIG FOR GER...... 58

APPENDIX E. CALCULATING Gmin and Gmax FOR GER ...... 59 vii

APPENDIX F. GENUS-LEVEL CHARACTER STATE CHANGES...... 60

APPENDIX G. SPECIES-LEVEL CHARACTER STATE CHANGES...... 67 viii

LIST OF FIGURES

Figure Page

1 Ophthalmosaurus from the Upper Jurassic...... 3

2 Body evolution of ichthyosaurs ...... 4

3 Cladogram of ...... 5

4 Ichthyopterygian phylogeny...... 7

5 Examples of ghost lineages and ghost taxa...... 7

6 One tree from Motani’s (1999b) data set that shows where the character state changes

occur ………………………………………………………………………………… 20

7 Strict consensus tree from species-based analysis showing the one polytomy present 21

8 Bootstrap and jackknife support for subclades……………………………………… 22

9 Character state changes for species-level analysis...... 24

10 Phylogenetic tree for species ...... 27

11 Consensus tree showing the stratigraphic interval during which each of the species is

present and its common geographic locations……...... 28 ix

LIST OF TABLES

Table Page

1 Average duration of stages and ammonite biozones...... 16

1

INTRODUCTION

Ichthyosaurs, the fish shaped reptiles from the , are known for their -like

looks and fast swimming. Ichthyosaurs occur from the Triassic to the Mid-Cretaceous (Motani,

2005a). Much research has been done on the morphological possibilities of these , but not as much has been done on the , or evolutionary relationships, of the group. In the early phylogenetic studies, many paleontologists took the “Jurassicocentric” view of ichthyopterygian evolution (McGowan and Motani, 2003). This view took a top-down approach to the evolution of the group and traced characters in the reverse direction or from the Jurassic to the Triassic (McGowan and Motani, 2003). Some paleontologists took a more bottom-up approach, but this never seemed to emerge as the popular view (McGowan and Motani, 2003).

These early phylogenetic ideas came under further debate when became more

dominant as a methodology and more specimens were found at a lower stratigraphic level. The

first cladogram to use the “proto-” Grippa longirostris, almost 50

after its discovery, was conducted by Mazin (1982) (McGowan and Motani, 2003).

Calloway (1989) did the first parsimony-based analysis in 1989, focusing on shastasaurids, that

was then improved by Dal Sasso and Pinna (1996) by adding a newly described specimen

(McGowan and Motani, 2003). Motani did a general analysis of the majority of the known

ichthyosaurs at the genus level in 1999b, and some work has been done on the phylogeny of the

Family Mixosauridae (Motani, 1999b; Jiang et al., 2005). According to McGowan and Motani

(2003), Sander (2000) published another tree, which is similar to that of Mazin (1982), however

he did not include seven of the characters in the data matrix, and it is therefore not possible to

reproduce his results. Others such as Maisch and Matzke (2000) and Fernández (2007) have

done similar analyses. Maisch and Matzke’s (2000) analysis used an imaginary taxon as the 2

outgroup that was coded as all zeros for the characters, using three ancestral specimens to

polarize the character states (McGowan and Motani, 2003). The major problem with this

approach was that the authors did not explain how the characters were polarized. Fernández

(2007) re-described a specimen of Caypullisaurus from the Neuquén Basin of , and

used characters that dealt with the ; her cladistic results were similar to those of

Motani (1999b).

The more derived clades of Ichthyosauria have not been investigated on a more detailed

level. The present study takes a portion of the phylogenetic tree constructed by Motani (1999b)

and conducts a more detailed analysis of one section of this tree. The goal is to clear up some of

the unresolved clades present in Motani’s (1999b) analysis, because the more primitive groups

have been removed and therefore do not clutter up the analysis.

Background

Ichthyosaurs were originally made famous by ’s collections in the 1800s

(McGowan, 1991). Ichthyosaurs had a unique morphology that has interested many

paleontologists and amateur collectors for many years, including: large (Motani,

1999b), streamlined body (McGowan, 1991), and large numbers of phalanges or finger .

The fish-like shape of the ichthyosaur can be seen in Figure 1 and is characteristic of the more

derived ichthyosaurs. In contrast more primitive ichthyosaurs are snake- or lizard-like and do not

have the distinctive -like tail. The evolution of the ichthyosaurs includes the development

of a larger , a more streamlined body and a more shark-like tail, as can be seen in Figure 2.

Due to these features, these reptiles are considered the most specialized of the marine reptiles

(Carroll, 1988). These reptiles are classified as due to the presence of an upper temporal opening in the and solid cheek. The classification of Euryapsida is still 3 controversial, as many think that Euryapsida is just a modification of the Diapsida. Ichthyosaurs are thought to swam like and other fast swimming fish due to their streamlined shape

(Motani, 2002). The limbs of the ichthyosaur include many phalange bones and the animals may have a slight curve at the end of the tail (Carroll, 1988). In some ichthyosaurs, there is a noticeable bend in the tails in the caudal , which are referred to as the apical centra. In the apical centra there are vertebrae that are wider dorsally than ventrally, making the last part of the tail bend downward (McGowan and Motani, 2003). Before the wedged shaped vertebrae were discovered, this bend in the tail was thought to be a preservational artifact.

Figure 1. Ophthalmosaurus from the Upper Jurassic. The skeletal structure is 3.5 meters long; Streamlined body is a reconstruction. Modified from Carroll, 1988.

Ichthyosaurs are unique creatures and so it is still controversial as to where exactly they are placed on a cladogram relative to other (McGowan and Motani, 2003), as seen in

Figure 3. Figure 3 shows ichthyosaurs placed with (node 3), but Caldwell

(2002) places them within the Euryapsida, which contains most other marine reptiles (node 4).

Calloway (1997) even places ichthyosaurs outside of node 2, separating them even more from the other marine reptiles. McGowan and Motani (2003) also point out that most major 4 groups have been suggested as a possible sister group to ichthyosaurs, further highlighting the ambiguity of their placement.

Figure 2. Body evolution of ichthyosaurs. Modified from McGowan and Motani, 2003. 5

Figure 3. Cladogram of diapsid reptiles. Note where Euryapsida and ichthyosaurs are placed with respect to other diapsid reptiles. Modified from Benton, 2000.

Regardless of where the Ichthyosauria are placed within Diapsida, the group can be

subdivided into subclades. The Thunnosauria are the most derived and youngest ichthyosaurs

and are distinguished through the ratio of forefin to hindfin (forefin is at least two times the length of the hindfin, Motani, 1999b) (Figure 4). Within this clade is the Ophthalmosauria, which

first appear in the Late Jurassic after a large gap in the record. Ophthalmosauria

is defined by the absence, or extreme reduction of the basioccipital peg, the basioccipital

extracondylar area having a narrow band of concavity, the angular lateral exposure being at least

as high and anteriorly as the surangular exposure, the ridge of the being plate-like, and

the presence of the manual anterior sesamoid e and the digit distal to it. In addition, the group is

well known for its large eyes (Motani, 1999b).

One interpretation of this large gap in the Middle Jurassic ichthyosaur record is that

Ophthalmosauria have a long (Motani, 1999b). A ghost lineage is a lineage for 6 which a fossil record does not exist but whose presence is inferred from the phylogenetic relationships, and can be attributed to the imperfection of the fossil record (Figure 5). Ghost lineages are determined by a comparison of the phylogenetic tree and the occurrence of the in the stratigraphic record. Ghost lineages or ranges are often expanded to cover the interval of time from the first occurrence of a sister group to the taxon in question to that taxon’s first appearance (Hammer and Harper, 2006). An example of a ghost lineage can be seen in

Figure 5. Ghost lineages are one way to represent stratigraphic debt, or the mismatch of phylogenetic and stratigraphic patterns. Another way stratigraphic debt is represented is via ghost taxa. A ghost taxon is a hypothetical taxon for which there is no fossil record, but which must be inserted into the phylogenetic tree for the biostratigraphic data to be congruent (Figure

5) (Smith, 1994). Stratigraphic debt can be quantified using the stratigraphic congruence index

(SCI) (Huelsenbeck, 1994), relative completeness index (RCI) (Benton and Storrs, 1994) and the gap excess ratio (GER) (Willis, 1999) (Hammer and Harper, 2006).

When constructing a phylogenetic tree, the goal is to find the most parsimonious tree, that is, the one with the fewest number of required character state changes. This is done by constructing all the trees possible, and calculating the total length of each (Hammer and Harper,

2006). According to Motani (1999b), placing Stenopterygius as the sister group to the

Ophthalmosauria (as suggested by Godefroit (1993), Maisch and Matzke (2000), and Sander

(2000); see Figure 4) would only make the phylogenetic tree one step longer. This tree topology would greatly reduce the length of the ghost lineage required for the Ophthalmosauria. Hence, making the tree one step longer and therefore less phylogenetically parsimonious would make it more stratigraphically parsimonious.

7

Figure 4. Ichthyopterygian phylogeny. Modified from Motani’s (1999b) phylogenetic analysis of ichthyopterygians.

Figure 5. Examples of ghost lineages and ghost taxa. Modified from Lane et al., 2005 8

In addition to the problem of phylogenetic versus stratigraphic parsimony, there is a need

to explore the causal factors driving the polytomy of this last, most derived clade of

ichthyosaurs. A polytomy is a branching pattern on a cladogram where two or more unresolved

clades collapse into one node (Hammer and Harper, 2006). The clades present in the polytomy

have symplesiomorphies, or ancestral character states inherited from a common ancestor

(Hammer and Harper, 2006). In contrast, the clades in the polytomy do not have any

synapomorphies, or derived character states that are phylogenetically informative, in common

(Hammer and Harper, 2006). According to Motani’s (1999b) phylogenetic tree (Figure 4), there is an initial polytomy including the Eurhinosauria, Temnodontosaurus and the Thunnosauria as well as a polytomy among the genera that make up Eurhinosauria. These polytomies could be due to radiations that occurred during the Middle to Upper Jurassic, which is when the Atlantic was starting to form.

An evolutionary radiation involves events occurring in a short interval of time,

where the new species all come from a common ancestor; hence, they have symplesiomorphies

and , but no synapomorphies (shared derived characters) that could be used to

resolve branching order. A radiation is one potential explanation for the occurrence of a

polytomy on a cladogram, but it is not the only explanation. A polytomy may have formed

because of unknown or missing character states in the data matrix. An example of this would be

two species, one only being known for the skull and the other known for everything but the skull.

In this example it is impossible to resolve the relationship between the two.

If the polytomy is due to a rapid evolutionary radiation, two factors may have been

involved: 1) it was triggered by a morphological innovation that evolved in this clade, and 2) it is

due to biogeographic expansion into new marine environments being created due to the rifting of 9

the supercontinent Pangea, such as the epicontinental seas of southern South America. Recent

work has, in fact, focused on the ichthyosaurian record of the southern continents. An analysis by

Fernández et al. (2005) was based on measurements taken from the sclerotic ring of the South

American ophthalmosaurian Caypullisaurus. The objective of Fernández et al. (2005) was to compare the size of the eyes of these specimens between juvenile and adult ichthyosaurs. The main objective was not a phylogenetic analysis, but they did present a phylogenetic hypothesis.

The data for the analysis were taken from Motani’s (1999b) analysis (mentioned above) and their

own measurements from five specimens of Caypullisaurus, and four specimens of other ophthalmosaurians, and data from about Temnodontosaurus taken from McGowan (1994). This analysis was consistent with Motani’s findings, but has some species added to the phylogenetic

tree.

Fernández has written additional papers about ichthyosaurs from Argentina, which are

useful for evaluating the biogeographic expansion of the Ophthalmosauria (2001, 2003, 2007;

Fernández et al., 2005; Fernández and Aguirre-Urreta, 2005). The evidence found by Shultz et

al. (2003) supports the proposal of Fernández et al. (2005) that extension of the geographic range

of ichthyosaurs into the now closed Rocas Verdes (South Africa) backarc basin suggests that the

seaway could have been an important part of the migratory pathway to the Atlantic Ocean from

Northwest . Extension was occurring in the in South America, as evidenced

by the rift basins present on the eastern side of the Andean mountain range (Shultz et al., 2003).

This rifting represents the start of the modern South Atlantic Ocean. Based on the presence of

mafic oceanic crust in the Rocas Verdes Basin, the continent was separated in the latest Jurassic

or . Remains of ichthyosaurs and other reptiles have been found in the Neuquén

Basin, which is -Valanginian in age (Gasparini and Fernández, 1997). There are also 10 deposits found in the Rocas Verdes rift basin that are from the Early Cretaceous, specifically the

Tobifera and Zapata Formations (Wilson, 1991). An ichthyosaur has been discovered in the

Torres del Paine National Park, Ultima Esperanza region of southern Chile, and is interpreted as from the lower Zapata Formation (Lower Cretaceous) (Shultz et al., 2003). This biogeographic pathway is also supported by the presence of bivalves Megacucullaea (Reyment and Tait, 1972) in both Africa and Patagonia and by some belemnite faunas (Mutterlose, 1986). Ichthyosaur teeth have also been found on the Antarctica peninsula, meaning that ichthyosaurs were present in this area during the late Jurassic to the Early Cretaceous time (Shultz et al., 2003).

Objectives

The objective of this study is to conduct a phylogenetic analysis of the most derived

Ichthyosauria, including the clade Eurhinosauria, the genus Temnodontosaurus, and the clade

Thunnosauria, using information from multiple species of each genus (Figure 4), and including species for which I have collected my own data as well as previously published descriptions

(Bardet, 1992; Dal Sasso and Pinna, 1996; Fernández, 2001, 2007; Fernández and Aguirre-

Urreta, 2005; Fernández et al., 2005; Godefroit, 1993; Huene, 1922; Jiang et al., 2005; Lydekker,

1889; Maxwell and Caldwell, 2006; Maisch and Matzke, 1997, 2000; Mazin, 1981, 1982;

McGowan, 1976, 1994; McGowan and Motani, 2003; Motani, 1997, 1999a, 1999b, 2005a,

2005b; Sander, 2000, Shultz et al., 2003; and Wade, 1990). In the present study, a more detailed analysis of the taxa within the most derived group of ichthyosaurs, the Thunnosauria, and their sister groups, the Eurhinosauria and Temnodontosaurus, has been performed to help answer phylogenetic, morphological, and paleogeographical questions about these ichthyosaurs. I use a total evidence approach, meaning a detailed phylogenetic analysis plus data on stratigraphic and geographic occurrences, to answer two questions. First, is a long ghost lineage more plausible 11 than placing Stenopterygius as a sister to the Ophthalmosauria, a question suggested by Motani

(1999b)? This can be determined by the concept of parsimony. In a phylogenetic context, parsimony involves determining which cladogram has the fewest changes in character states. In the case of the stratigraphic record, parsimony involves determining which phylogenetic tree has the fewest range extensions and ghost lineages. What needs to be considered is whether or not it is better to have a more parsimonious cladogram or a more parsimonious stratigraphic record?

Second, if the polytomy cannot be resolved with a species-based study, then is the radiation of this derived clade of ichthyosaurs during this time due to the appearance of a morphological innovation as documented by character state changes in the phylogenetic analysis or to the rifting apart of Gondwana during this time to produce new ocean basins (Tethys, South Atlantic and

Rocas Verdes, South Africa), and hence new habitat space for open marine ichthyosaurs to invade, or both?

12

METHODS

Taxa and Characters

To help decide if the cause of the polytomy of derived ichthyosaurs could be determined,

I recollected morphological data for specimens on the species level, as opposed to the genus

level presented in Motani (1999b). The data was obtained from specimens at the Royal Ontario

Museum (ROM) in Toronto, Canada, and the University of California Museum of Paleontology

at Berkeley (UCMP). Characters and character codes used are listed in Appendix A, while the data matrices are presented in Appendix B. I was able to view the following pertinent species:

Stenopterygius quadrisscissus (UCMP 61201, 61202, 61203), a cast of Ichthyosaurus

intermedius (UCMP137398), a partial skull of Ichthyosaurus communis (ROM 28964),

Leptonectes tenuirostris (ROM 47698), Leptonectes moorei (ROM 52087), and

Excalibosauresus cabrini (ROM 26027). When collecting data on specimens from the literature,

I used the list of currently accepted taxa in McGowan and Motani (2003). I could not find

primary literature on some of the species (e.g., Arkhangelsky, 1998; Efimov, 1999; and Owen,

1881), while, for others, I could not use the primary literature due to the lack of pictures or lack

of quality pictures (e.g., McGowan, 1986; Kuhn, 1946; and Lydekker, 1888). Some articles had

the specimen redrawn, but I was unable to use those images either because the cranial sutures

were not drawn in, and these are pertinent for the characters used. There were two species,

Nannopterygius enthekiodon and Stenopterygius hauffianus for which I was able to collect a

small number of characters, but I did not feel that there was enough information to be valuable in

this analysis. These two species were therefore dropped from the analysis.

I focused on anatomical characteristics used by Motani (1999b) and Fernández (2007)

when looking at specimens. I used all 105 characters from Motani (1999b) but only used seven 13

from Fernández (2007) (characters 1, 2, 4, 7, 16, 19 and 21). These were characters that had not

been used in previous studies but I thought would be helpful in separating species; all characters

are listed in Appendix A.

My final dataset combined the results from my own observations of specimens and data

collected from the primary literature with the datasets from Motani (1999b) and Fernández

(2007) into a data matrix that was then reduced to a 18 operational taxonomic units (OTU), 112

character matrix compatible with the parsimony-based phylogenetic software package PAUP* or

Phylogenetic Analysis Using Parsimony Version 4.0b10 (Swofford, 2001) (data matrix listed in

Appendix B). PAUP* (Swofford, 2001) found that of my 112 characters, 52 of those were

parsimoniously uninformative, including 22 characters that were constant. These characters are

listed in Appendix C. So in reality, a character matrix of 18 OTU and 60 characters produced the

phylogenetic results. The reason so many of the characters were constant is due to a lot of missing data, which may reflect the way many specimens are preserved. Also, since Motani’s

(1999b) analysis included all ichthyosaurs, many of his characters varied only in the less derived

groups.

Phylogenetic Analysis

In these analyses all characters were unweighted, and polymorphisms were included,

meaning that some characters were coded with more than one state. For example, character 33

(overbite) can be absent or slight (0) or can be clearly present (1) within specimens that are plainly the same species based on other characters. PAUP* (Swofford, 2001) reads this coding as indicating that this character could be one or the other state in a single taxon, not necessarily both at the same time. The reason that the characters were unweighted is because there is no evidence that one character is more important than another. This will also keep assumptions in 14

this analysis to a minimum. Including polymorphisms can lead to more uncertainty in the final tree; however, they were included here to best reflect the true distribution of character states.

I conducted two separate phylogenetic analyses, one using Motani’s (1999b) original

dataset to determine which characters were supporting the clades that Motani’s (1999b) genus-

level analysis had produced. The second used an expanded database including seven characters

from Fernández (2007) and was done at the species level for derived ichthyosaurs only. The first

analysis was performed to help understand the radiation that is of interest in this study. I re-ran

Motani’s (1999b) original published dataset in PAUP* (Swofford, 2001) so that I could mark the

character state changes that supported each sub-clade. There are 33 OTUs and 105 characters in

the character matrix. Motani forced certain taxa to be the outgroup in PAUP* (Swofford, 2001)

and I did not recreate that for the purposes of this study. I placed as the

outgroup, because it was the most primitive according to Motani’s (1999b) interpretation. The

changes in character states help show how the clades that were created in Motani’s (1999b)

analysis are defined. This information was then used to help me to define and interpret the clades

produced by my second analysis.

The second analysis included seven new characters and the data that I collected from the

museums and from the primary literature by species. This analysis was run to determine if the

polytomies seen in the genus-level analysis could be resolved. I chose the genus placed as the

sister taxon to my group of interest in Motani’s (1999b) analysis, Suevoleviathan, as the

outgroup because it is the logical sister group to the clade of interest. My analysis was done at

the species level (other than for the outgroup), to produce a more detailed evolutionary tree for

the derived ichthyosaurs. PAUP* (Swofford, 2001) was used to find the most parsimonious tree and jackknife and bootstrap analyses (Felsenstein, 1985) were run to determine how well 15

supported each of the clades are. I determined where the character state changes occur to see

what is defining the clades that are now present.

A jackknife analysis randomly removes either taxa or characters from the original data

and then the analysis is rerun and repeated as many times as specified. The output then shows by

percentage how many times during that analysis each sub-clade occurred. If a subclade appears over 70% of the time, it is considered a strongly supported grouping. Bootstrap analysis

(Felsenstein, 1985) is similar to jackknife but it randomly samples the data, usually only characters, and then produces a brand new dataset that is the same size as the original but can have duplication of different characters while omitting others. This analysis is rerun as many times as specified and then the output is interpreted in a similar way as for jackknifing. If the bootstrap value for a subclade is 70% or above, there is a 95% probability that the subclade is real (Hillis and Bull, 1993).

Stratigraphic Debt Analysis

The Gap Excess Ratio (GER) (Willis, 1999), Relative Completeness Index (RCI) (Benton

and Storrs, 1994), and Stratigraphic Congruence Index (SCI) (Huelsenbeck, 1994) were

calculated by hand. These indices reflect how well supported the specific cladogram is when

compared to the stratigraphic record. To do the following calculations, time intervals needed to

be determined. I used the stratigraphic occurrence data for all ichthyosaur species, detailed by

ammonite biozone, presented by McGowan and Motani (2003, Table 2). Absolute date ranges

for all the ammonite biozones do not exist, but I was able to get the absolute ages for the stages

from the International Commission on Stratigraphy (2007), listed in Table 1. The durations of the stages were determined in millions of years and then divided equally among the ammonite biozones, assuming that each ammonite biozone within a stage was the same length (McGowan 16 and Motani (2003) used this same simplifying assumption). Ammonite biozone durations ranged from 0.31 to 3.25 million years (mean=1.28 m.y., standard deviation = 0.75 m.y.). These different ammonite biozone ranges were then used to calculate the stratigraphic gap and duration lengths discussed below. Because many of the stratigraphic gaps of interest are so long, deviations from the assumption of equal biozone duration would have only a small impact on the overall estimation of stratigraphic debt.

Table 1. Average duration of stages and ammonite biozones. This table shows the dates of the stages of interest along with the number of ammonite zones present and the estimated duration of those ammonite zones that were used to calculate the stratigraphic gaps. Dates from the International Commission on Stratigraphy (2007).

Stage Start Date End date Stage Number of Average Duration (Ma) (Ma) Duration Ammonite of Each (m.y.) Biozones Ammonite Biozone (m.y.) 203.6 +/- 1.5 199.6 +/- 0.6 4.0 2 2.00 Hettangian 199.6 +/- 0.6 196.5 +/- 1.0 3.1 3 1.03 196.5 +/- 1.0 189.6 +/- 1.5 6.9 6 1.15 189.6 +/- 1.5 183.0 +/- 1.5 6.6 5 1.32 183.0 +/- 1.5 175.6 +/- 2.0 7.4 6 1.23 175.6 +/- 2.0 171.6 +/- 3.0 4.0 4 1.00 171.6 +/- 3.0 167.7 +/- 3.5 3.9 7 0.56 Bathonian 167.7 +/- 3.5 164.7 +/- 4.0 3.0 8 0.38 164.7 +/- 4.0 161.2 +/- 4.0 3.5 6 0.58 161.2 +/- 4.0 155.6 +/- 4.0 5.8 8 0.70 155.6 +/- 4.0 150.8 +/- 4.0 4.8 5 0.96 Tithonian 150.8 +/- 4.0 145.5 +/- 4.0 5.3 17 0.31 145.5 +/- 4.0 140.2 +/- 3.0 5.3 2 2.65 Valanginian 140.2 +/- 3.0 133.9 +/- 2.0 6.3 6 1.05 133.9 +/- 2.0 130.0 +/- 1.5 3.9 4 0.98 Barremian 130.0 +/- 1.5 125.0 +/- 1.5 5.0 4 1.25 125.0 +/- 1.5 112.0 +/- 1.0 13.0 4 3.25 112.0 +/- 1.0 99.6 +/- 0.9 12.4 7 1.77 99.6 +/- 0.9 93.6 +/- 0.8 3.0 3 2.00 93.6 +/- 0.8 89.3 +/- 1.0 4.3 3 1.43

17

The output from the phylogenetic analysis was compared and combined with the

stratigraphic data from McGowan and Motani (2003) to form an evolutionary tree for both the

genus-level analysis and species-level analysis. When combining a cladogram and stratigraphic

data to create an evolutionary tree, there are two typical ways of doing this, coined X-trees and

A-trees. X-trees compare the cladogram to the fossil record, combining the branching order of the cladogram and the stratigraphic range of the taxa to produce the new tree (Smith, 1994). In

A-trees, taxa are taken from the cladogram and placed as direct ancestors to each other, implying

that ancestor-descendant relationships can be determined (Smith, 1994). X-trees are used most

often because they tend to retain the most detail (Smith, 1994). By making an X-tree, it is

possible to determine whether the cladogram is stratigraphically consistent. An X-tree was

therefore constructed for the derived ichthyosaurs included in this study.

The gap excess ratio (GER) (Willis, 1999) is a comparison of the duration of the ghost

lineages required by a cladogram to the best and worst-case scenarios for gaps found in the

record. The equation for the gap excess ratio is defined as:

− GMIG (Eqn. 1) GER 1−= ∑ min max − GG min

where MIG (minimum implied gap) is the length of the ghost ranges present based on a particular cladogram (Appendix D), Gmin is the minimum possible sum of ghost ranges or the

sum of distances between consecutive first appearance datums (FADs) and the Gmax is the

maximum possible sum of ghost ranges or the sum of distances from the first FAD to all of the

other FADs (Willis 1999) (Hammer and Harper, 2006) (Appendix E). The closer the GER value

is to 1, the better supported the cladogram is.

The Relative Completeness Index (RCI) (Benton and Storrs, 1994) is defined as: 18

⎛ MIG ⎞ (Eqn. 2) RCI ⎜1100 −= ∑ ⎟ ⎜ ⎟ ⎝ ∑ SRL ⎠

where SRL is the duration of the observed ranges and MIG is defined as above (Benton and

Storrs, 1994). RCI is interpreted such that the closer RCI is to 100 (that is, the smaller the MIG),

the better the completeness of the record. The RCI normally ranges between 0 and 100, but can

sometimes become negative (Hammer and Harper, 2006). This index is a more direct measure of

the fossil record and adds the aspect of time into the equation (Hammer and Harper, 2006).

The SCI (Huelsenbeck, 1994) analysis is similar to RCI, but does not use time. The SCI

is the proportion of stratigraphically consistent nodes on the cladogram. A node connecting two

taxa is stratigraphically consistent if the stratigraphically older taxon’s first occurrence is at the same age or older than its sister taxon (Hammer and Harper, 2006). The resulting value for SCI ranges from zero to one, with values closer to one indicating a more stratigraphically consistent tree (Hammer and Harper, 2006).

Institutional Abbreviations: ROM, , Toronto; UCMP

University of California Museum of Paleontology, Berkeley. 19

RESULTS

Phylogenetic Analysis

PAUP

I conducted two separate phylogenetic analyses, one using Motani’s (1998b) original genus-level dataset to determine which characters were supporting his clades, and one with an expanded dataset at the species level for the derived ichthyosaurs only. When I re-ran Motani’s

(1999b) data, I found 13 most parsimonious trees with 228 steps, which is similar to Motani’s

(1999b) findings of 12 trees and 254 steps (Consistency Index (C.I.)=0.6184, Rescaled

Consistency Index (R.C.)= 0.5202). I attribute the differences in our results to my decision not to force certain genera to be the outgroup. I then used one of the 13 trees to compile a list of where each character state change occurred on the tree (Figure 6). The major characters that define all the derived ichthyosaurs as a clade (node 37) are the presence of a reduced peripheral shaft of the , loss of an antero-medial iliac prominence, and presence medially of ischium- fusion in adults. The characters that define the Eurhinosauria clade (node 38) are loss of manual accessory digit VI and a widely separated tibia and fibula. The Eurhinosauria clade includes the genera Leptonectes, Excalibosaurus and Eurhinosaurus. The Thunnosauria clade (node 45) that was present in Motani’s (1999b) study is defined by four character state changes that include a widely open anterior process in the right and left parietals, presence of a manual pisiform, a forelimb twice as long as hindlimb, and an ischium-pubis fusion present both medially and laterally. The Thunnosauria clade includes all the genera from Stenopterygius forward on Figure

6. The characters that define Ophthalmosauria (node 49) consist of the absence or extreme reduction of the basioccipital peg, the basioccipital extracondylar area having a narrow band of 20

concavity, the angular lateral exposure being at least as high and anteriorly placed as the

surangular exposure, the ridge of the humerus being plate-like, and the presence of the manual anterior sesamoid e and the digit distal to it. The clade Ophthalmosauria includes all the genera from forward on Figure 6.

Figure 6. One tree from Motani’s (1999b) data set that shows where the character state changes occur. The bold face numbers indicate node numbers while the numbers in parentheses indicate the number of character state changes at that node. A list of the actual character state changes can be found in Appendix F. Compare to Figure 4, showing Motani’s (1999b) consensus tree. 21

When the second, species level-analysis for the derived ichthyosaurs was run, PAUP*

(Swofford, 2001) produced three most parsimonious trees with 196 steps (C.I. = 0.5918, R.C.

=0.3317). The strict consensus tree shows that the only variation among these trees is one

polytomy involving three ophthalmosaurian species (Figure 7). The results of the bootstrap and

jackknife analyses can be seen in Figure 8. In these analyses, only two clades are kept and only

one is significant. The significant clade is the one that is retained 71% of the time, meaning that there is a little more than a 95% chance that the clade is real (Hillis and Bull, 1993). This clade included the species of Ichthyosaurus, Stenopterygius, and Temnodontosaurus and two of the

Leptonectes species, Leptonectes tenuirostris and Leptonectes moorei. This clade correlates to

node 15 on the tree in Figure 9 and is supported by 13 character state changes.

Figure 7. Strict consensus tree from species-based analysis showing the one polytomy present.

22

Figure 8. Bootstrap and jackknife support for subclades. The bold numbers are from the bootstrap analysis, while the other number is from the jackknife analysis. Jackknife analysis did not support the subclade including Ophthal. discus, Brachy. extremus, and Caypull. bonaparte.

Temnodontosaurus trigonodon was pulled up the tree and the clade Ophthalmosauria was pulled down the tree relative to Motani’s (1999b) phylogeny (compare Figures 6 and 7). The polytomy also separates the two Ophthalmosaurus species, indicating that Ophthalmosaurus icenicus may not belong in the previously defined Ophthalmosauria clade (Figure 9). There are ten characteristics that separate Ophthalmosaurus icenicus from the rest of the previously defined Ophthalmosauria clade. These ten characters include the presence of a nasal and external 23

naris contact, the presence of a nasal and parietal contact lateral to the frontal, a trilunate shaped

postorbital, a short anterior process of the in lateral view, the distal radial articular facet

on the humerus is terminal and larger than the ulnar facet, the anterior flange on the humerus is

absent, the radius is longer than wide, the thyroid fenestra is absent, the tibia peripheral ‘shaft’ is

absent, and the postorbital is narrow. None of these specific characteristics were part of the

definition of the clade Ophthalmosauria (node 49) in Motani’s (1999b) analysis. The rest of the

previously defined clade is grouped together by the following nine characteristics (which include the characters defining this clade in Motani’s previous study): an absent or extremely reduced bassioccipital peg, bassioccipital extracondylar area reduced to a narrow band of concavity, the angular lateral exposure is at least as high and anterior as surangular exposure, the distal and proximal ends of the humerus exclusive of anterior flange are nearly equal, a plate-like ridge on humerus, the presence of an antero-distal facet for sesamoid on the humerus, the presence of a manual anterior sesamoid e and the digit distal to it, ischium-pubis fusion present medially and laterally, and the peripheral ‘shaft’ of the tibia is absent.

This analysis also shifted the previously defined clade of Eurhinosauria from a monophyletic clade to a paraphyletic group. Two of the Leptonectes species, Leptonectes

tenuirostris and Leptonectes moorei, are seen as more derived in this new analysis (Figure 9).

Leptonectes moorei is defined by a radius that is longer than wide, a complete or nearly complete

peripheral ‘shaft’, a mc 1 that is not ossified, and the absence of a manual digit S4-5. These

characteristics do not compare to those that defined the Eurhinosauria clade previously (loss of

manual accessory digit VI and a widely separate tibia and fibula). Leptonectes tenuirostris is

defined by the upper dental groove being present only anteriorly, presence of interdigital 24 separation, ischium-pubis fusion present medially and laterally, the atlantal pleuro centrum separate from axis, and having two distinguishable sacral ribs.

Figure 9. Character state changes for species-level analysis. The bold face numbers indicate node numbers while the numbers in parenthesis indicate the number of character state changes at that node. A list of the actual character state changes can be seen in Appendix G.

One species of Stenopterygius, S. megacephalus, is separated from the rest of the species in this genus. Stenopterygius megacephalus is defined by an anterior notch on the radiale present preaxially, and by the metacarpal 5 not being ossified. Stenopterygius megacephalus is mixed in with two Leptonectes species and the Ichthyosaurus genus. Motani’s (1999b) analysis defines the genus Stenopterygius by the presence of nasal/parietal contact lateral to frontal and the presence 25

of manual digit S4-5. In contrast, this analysis found that neither of these characteristics define

Stenopterygius megacephalus, although it does have manual digit S4-5.

Temnodontosaurus trigonodon and Stenopterygius quadrisissus are grouped together in

the species-level analysis. These species are grouped together by the following characteristics:

the width of the humerus exclusive of the anterior flange being nearly squarish, a complete or

nearly complete radius contiguous ‘shaft’ and ulna peripheral ‘shaft’, and a presence of the

metacarpal I peripheral ‘shaft’. Temnodontosaurus trigonodon is defined by the ulna peripheral

shaft being a notch or largely reduced and by the metacarpal V peripheral ‘shaft’ being complete

or nearly complete. Stenopterygius quadrisissus is defined by a complete radius peripheral

‘shaft’, the metacarpal I peripheral ‘shaft’ not being ossified, and the absence of the metacarpal

III ‘shaft”. The two character state changes that group Temnodontosaurus trigonodon and

Stenopterygius quadrisissus together also place Stenopterygius quadrisissus as its own species.

The width of humerus is a reversal while the metacarpal I peripheral ‘shaft’ is a different state,

meaning that only two characteristics really define Stenopterygius quadrisissus and three group

Temnodontosaurus trigonodon and Stenopterygius quadrisissus together. None of these species

are defined by the same characteristics set by Motani’s (1999b) analysis. Stenopterygius

cuneiceps is the only species with the presence of the nasal/parietal contact lateral to frontal and

none have the presence of the manual S4-5 digit. There are also two characters that seem unique

to Temnodontosaurus trigonodon and one other species. For the ulna contiguous ‘shaft’,

Temnodontosaurus trigonodon and Stenopterygius cuneiceps are the only ones with the character

state of notch or largely reduced, and for character metacarpal V peripheral ‘shaft’ the only other species to have a complete or nearly complete character state is sometimes Ichthyosaurus 26

conybeari. These more derived characters could be the reason that this species was pulled up the

tree.

When looking at the new cladogram associated with the species-level analysis, four

previously defined taxa fall apart: Leptonectes, Stenopterygius, and the groups Eurhinosauria and

Ophthalmosauria. With these clades falling apart, Motani’s (1999b) argument that

Stenopterygius and Ophthalmosauria could be sister taxa by adding one extra step to the cladogram falls apart as well. With the new cladogram, to make the two clades sister taxa, more than one step will need to be added to the tree, and, in fact, these groups no longer exist as true clades in the new cladogram.

Evolutionary Tree

An X-tree was created for the species-level analysis by combining the cladogram and the

stratigraphic record of the species present in the cladogram (Figure 10). The general stratigraphic range and geographic location of each species in my analysis are indicated in Figure 11. For the tree presented in Figure 9 to work, there must be a period of rapid evolution prior to the Upper

Triassic that produced the major lineages. It is possible that the genera of Stenopterygius and

Leptonectes are stratigraphically defined. This means that, for instance, if there were a specimen found in the Toarcian that did not look like Suevoleviathan or Eurhinosaurus, it would be automatically classified as Stenopterygius, and given a new species name. Hence, the taxonomic definitions of these two genera need to be evaluated and revised. The geographic locations of these ichthyosaurs did not seem to expand until the Atlantic started to form in the Late Jurassic

(Gasparini and Fernández, 1997). The species that were present in the Late Triassic and Early

Jurassic are only found in , and/or while towards the end of the

Jurassic, the geographical range of these ichthyosaurs expands into Russia and North and South 27

America, as indicated by the Upper Jurassic American species Ophthalmosaurus discus and

Caypullisaurus bonaparte, and Russian Ophthalmosaurus icenicus and Brachypterygius extremus. Note that only members of the Ophthalmosauria (sensu Motani, 1999b) participate in this biogeographic expansion.

Figure 10. Phylogenetic tree for species. The X-tree for my species-level analysis. Stage names and durations can be seen in Table 1. 28

Figure 11. Consensus tree showing the stratigraphic interval during which each of the species is present and its common geographic locations. L - Lower, M - Middle, U - Upper, Tri - Triassic, Jur - Jurassic, Cret - Cretaceous.

29

Phylogenetic vs. Stratigraphic Parsimony

When calculating the Gap Excess Ratio (GER) (Willis, 1999), Relative Completeness

Index (RCI) (Benton and Storrs, 1994) and Stratigraphic Congruency Index (SCI) (Huelsenbeck,

1994), the X-tree (Figure 10) and the durations from Table 1 were used. The Gap Excess Ratio

(GER) (Willis, 1999) was calculated using Equation (1) for Motani’s (1999b) genus-level

analysis as well as my analysis done at the species level. The GER for Motani is 0.799 and for

my species analysis is 0.955. In constructing the X-tree, ghost taxa needed to be added below

the earliest known stratigraphic occurrences of included species. When calculating the Minimum

Implied Gap (MIG) for the genus analysis, I decided to omit the length of these ghost taxa,

because there was no way to tell how long these ghost taxa would be. The assumption is that

they would be minimal when compared to the rest of the stratigraphic gaps, and likely confined to the Upper Triassic. The GER values show that my species level analysis has better stratigraphic resolution, even with the more derived ichthyosaurs being pulled down the tree,

than Motani’s genus-level analysis.

The Relative Completeness Index (RCI; see Equation (2)) (Benton and Storrs, 1994) was also calculated for Motani’s (1999b) analysis as well as my species-based analysis. The RCI for

Motani’s analysis is 1.87 and for my species analysis is 7.79. These lower numbers imply that the fossil record for ichthyosaurs is very poor (length of gaps is nearly equal to the observed durations) and is even worse on the genus level than the species level, which is surprising.

The Stratigraphic Congruency Index (SCI) (Huelsenbeck, 1994) was calculated only for

my species-based analysis, as it is not meaningful to compare genus-level and species-level

cladograms in this way. The SCI was 9/17 or 0.529. For the node to be considered

stratigraphically consistent, had to be consistent with the stratigraphic record as well as 30 consistent with the taxon next to it on the tree. This value shows us that about half of the cladogram nodes are consistent with the stratigraphic record. The species that are causing the main problems for the SCI are those of the previously defined Ophthalmosauria clade. The

Ophthalmosaurians now have an even longer ghost lineage than before. Pulling

Temnodontosaurus trigonodon up the tree actually helped with the stratigraphic debt.

Radiation of the Ophthalmosauria

The polytomy associated with the three species of Ophthalmosauria is potentially due to a radiation caused by morphological innovation, although the dataset does include numerous unknown character states for one or more of these three species. Five characters used in Motani’s

(1999b) analysis seemed to be potential morphological innovations based upon a change in character state defining the Ophthalmosauria. The characters were identified as those that unite the three taxa involved in the polytomy. These characteristics include: (1) the change from the distal end of the humerus being wider to the distal and proximal ends being nearly equal

[55(1Æ0)], (2) the ridge on the humerus changing from not plate-like to plate-like [56(0Æ2)],

(3) the anterio-distal facet on the humerus developing [57(0Æ1)], (4) the peripheral ‘shaft’ of the radius changing from a notch or largely reduced to entirely absent [59(1Æ2)], and (5) the appearance of a manual anterior sesamoid e and the digit distal to it[75(0Æ1)]. For two of these characters (2 and 4), the change is unique only to the Ophthalmosauria. It is not clear how the ridge on the radius could have contributed to the radiation of these species, but the appearance of the sesamoid e could have allowed for greater flexibility of the forefin, which could have allowed for faster, more agile swimming. Because there are only two characters that change within the Ophthalmosauria, it appears that anatomical innovations were not the sole cause of the radiation. 31

DISCUSSION

The major difference between Motani’s (1999b) study and my current study is that

Motanis’(1999) study was performed at the genus level, while this analysis is at the species

level. I also truncated the dataset by focusing only on the more derived species. When you break

up an analysis, most of the time the dataset ends up with more unknown character states in it due

to the lack of multiple specimens for each species. When conducting an analysis such as this on

the genus level, unknown characters from one species can be filled in with the known characters

from another. This pooling of data from multiple species, however, can cause problems when

different species show different character states.

In the new species-level analysis, many of the previously defined taxa fell apart and were

no longer grouped together. These taxa include Leptonectes, Stenopterygius, and the groups

Eurhinosauria and Ophthalmosauria. This is a problem that needs to be looked at to see if these

species were identified properly. The genera of Leptonectes and Stenopterygius may be

stratigraphically defined. The groups of Ophthalmosauria (sensu Motani, 1999b) were also pulled down the tree, and created larger stratigraphic debt in being placed there. One potential

explanation for the placement is that there were enough reversals in these species that when the

more primitive species were taken out, the ophthalmosaurians were interpreted as more primitive

than they actually are. To determine this, the character states of the more primative ichthyosaurs

need to be reanalyzed in a future report.

I feel that to further expand the species-level study, more specimens need to be added to

the species level analysis, especially including species from the youngest ophthalmosaurian

genus . I could not find good pictures of sufficient quality in the literature to properly code characters for this genus. This expansion along with more specimens found and 32 added to the existing stratigraphic record of each species will help to improve upon my results.

Another way to help improve these results will be to define more characters that will specifically help separate the more derived species from each other, as opposed to just the genera. Many of these characters were good for separating ichthyosaurs on the genus level; in the species level analysis, however, 52 uninformative characters were found. This large number of uninformative characters is to be expected when leaving out most of the taxa used in the original study. Some of these characters may be more useful if more specimens are found completely or almost completely articulated and not just preserved or prepared in dorsal view.

In my species level analysis, I found that there is no way to determine whether placing

Stenopterygius as a sister group to Ophthalmosauria is more parsimonious because the

Ophthalmosauria clade falls apart in the species-level analysis. Making this change would no longer be a small shuffle, as implied by Motani’s (1999b) analysis.

When comparing the new species-level cladogram to the stratigraphic record, the tree was not highly supported, but the stratigraphic problems seemed to be caused by only a few groups, specifically the ophthalmosaurian taxa and Eurhinosaurus longirostris, Leptonectes solei, and Excalibosaurus costini. These are the groups that needed to have ghost lineages added to them and in Equations (1) and (2), and the SCI calculations used. A potential explanation for this is the fact that some of the ophthalmosaurian taxa are found in places outside of Europe and may not be as well sampled. This is also coupled with the fact that these more derived ichthyosaurs could have been living in different environments from the previous taxa and may not be preserved as often. Most of the specimens I looked at were preserved in shales that came from deep water and not as many sandstones and no limestones from shallow water. There also 33 may not be many rocks of that age and location being formed at that time, meaning that even though these ichthyosaurs may have been living in this area, there is no evidence of it.

The widening of the Atlantic did not seem to relate to the polytomies from Motani’s

(1999b) study in the Late Triassic, but did affect the new radiation that appeared in the species- level analysis, in the Upper Jurassic. The North and South Atlantic started to for during the

Middle to Late Jurassic. The geographic range of the derived ichthyosaurs did not seem to expand until the Late Jurassic and into the Early Cretaceous, right after these new environments opened up. There are problems when examining geographic ranges of marine organisms, including taphonomic problems. When evaluating the geographic range of these organisms, because they were marine reptiles, you must take into account the possibility of taphonomic drift.

When a large swimming organism dies in the ocean and is decomposing, it tends to fill with gas and can sometimes float miles away from where it was really living. Many of the species are found in the Northwest Tethyan area before this radiation occurs, and this is also where you can find lots of good specimens. The Tethyan (Europe) area seems to have been thoroughly sampled and has fewer problems in this analysis because of this. To see if geographic expansion was actually happening with ichthyosaurs during this time, we should look for specimens in the potential pathways they could have taken. Some of this has been done in previous studies, but more can still be done. Looking at other marine organisms with a higher preservation rate, like ammonites or other invertebrates, could also help to show the potential biogeographic pathways of ichthyosaurs.

I feel the morphological innovation may have also contributed, but was not the sole cause of this radiation of the more derived ichthyosaurs to occur. The opening of the Atlantic, as discussed previously, seems to have contributed to the radiation of the more derived ichthyosaurs 34

by creating new habitats for them to live in, but they needed a way to get into and thrive in these new environments. There is a characteristic that is not used in this cladistic analysis that could be an important morphological innovation, eye size. Ophthalmosaurus is known for its large eyes and would have helped it to see it prey easier and probably at deeper depths. This is a characteristic that should be looked at in future analyses. 35

SUMMARY AND CONCLUSIONS

The main objectives of this study were to conduct a more detailed phylogenetic analysis

of the more derived ichthyosaurs to answer two questions. The first problem was to determine if

a long ghost lineage, or placing Stenopterygius and a sister group to Ophthalmosauria was more

parsimonious. The second was to determine if the polytomies present in Motani’s (1999b)

analysis were due to phylogenetic problems such as many unknown characters, due to a radiation

due to the biogeographic expansion of ichthyosaurs during the breaking apart of Pangea, the

evolution of morphological innovation, or a combination of the above.

My species level analysis produced three main differences from that of Motani (1999b).

The major difference is that a group of more derived ichthyosaurs (Ophthalmosauria), according

to Motani’s (1999b) tree, were pulled down the tree. Three of the taxa from the Ophthalmosauria

formed a polytomy, and include Ophthalmosaurus discus, Brachyopterygius extremus, and

Caypullisaurus Bonaparte. The polytomies that were present in Motani’s (1999b) study were no longer present. The species of the genus Leptonectes were separated from each other. The one species in the genus Temnodontosaurus was pulled up the tree. Stenopterygius megacephalus is separated from the rest of the genus in the species-level analysis as well. I feel that these species need to be further examined to see if they are placed in the right genus. Ophthalmosaurus icenicus is not grouped with the rest of the Ophthalmosaurians so either Ophthalmosaurus icenicus or Ophthalmosaurus discus should be looked at again to see if they are placed in the correct genus.

The previously defined clades of Ophthalmosauria (Motani, 1999b), Stenopterygius,

Leptonectes and Eurhinosauria fell apart in my species-level analysis. The species that would be

in the clade of Ophthalmosauria were also pulled down the cladogram and now have an even 36

longer ghost lineage to Stenopterygius, making the situation more complicated. It would not

longer be as simple as adding one extra step to the cladogram.

Through my re-analysis of Motani’s (1999b) study and my own species level analysis, I

have determined that the polytomies that were present in the genus level analysis can be cleared up when more characters are added and the analysis broken down by species. The new polytomy that is formed by the three taxa of ophthalmosaurians, may be due to a radiation enabled by the opening of the North Atlantic during the same time these taxa were expanding their geographical range. There are three characteristics that helped contribute to the geographic expansion of these species and are the ridge on the humerus being plate-like, the presence of the manual anterior sesamoid e and the digit distal to it, and the enlargement of the eyes relative to the body size.

37

REFERENCES

Arkhangelsky, M.S. 1998. [On the ichthyosaurian fossils from the Volgian Stage of the Saratov

region.] Paleontologicheskii Zhurnal. 1998(2): 87-91 [In Russian]

Bardet, N. 1992. Stratigraphic evidence for the of the ichthyosaurs. Terra Nova

4: 649-656.

Benton, M.J. 2000. Vertebrate Palaeontology. Oxford: Blackwell Publishing.

Benton, M.J. and Storrs, G.W. 1994. Testing the quality of the fossil record: Paleontological

knowledge is improving. Geology 22: 111-114.

Caldwell, M.W. 2002. From to limbs to fins: Limb evolution in fossil marine reptiles.

American Journal of Medical Genetics. 112:236-249

Calloway, J.M. 1989. , Phylogeny, and Ancestry of Triassic Ichthyosaurs.

Unpublished Ph.D. dissertation, University of Rochester, Rochester, 204 pp.

Calloway, J.M.1997. Introduction. –pp 3-16 in : Calloway, J.M. and Nicholls, E.L. (eds.) 1997.

Ancient Marine Reptiles. Academic Press, London and New York

Carroll, R.L. 1988. and Evolution. New York: W.H. Freeman

and Company.

Dal Sasso, C. and Pinna, G. 1996. Besanosaurus leptorhynchus n. gen. n. sp., a new

shastasaurid ichthyosaur from the of Besano (Lombardy, N.

Italy). Paleontologia Lombarda Nuova Serie 4: 3-23.

Efimov, V.M. 1999. [A new family of ichthyosaurus, Undorsauridae fam. nov. from the Volgian

Stage of the European part of Russia.] Paleontologicheskii Zhurnal. 1999(2): 51-58 [In

Russian]

Felsenstein, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. 38

Evolution. 39: 783-91

Fernández, M.S. 2001. Dorsal or Ventral? Homologies of the forefin of Caypullisaurus

(Ichthyosauria: Ophthalmosauria). Journal of Vertebrate Paleontology 21(3):

515-520.

Fernández, M.S. 2003. Ophthalmosauria (Ichthyosauria) forefin from the Aalenian-Bajocian

boundary of Mendoza Province, Argentina. Journal of Vertebrate Paleontology 23(3):

691-694

Fernández, M.S. 2007. Redescription and phylogenetic position of Caypullisaurus

(Ichthyosauria: Opthalmosauridae). Journal of Paleontology 81(2): 368-375.

Fernández, M.S., and Aguirre-Urreta, M.B. 2005. Revision of Platypterygius hauthali von

Huene, 1927 (Ichthyosauria: ) from the Early Cretaceous of

Patagonia, Argentina. Journal of Vertebrate Paleontology 25(3): 583-587.

Fernández, M.S., Archuby, F., Talevi, M., and Ebner, R. 2005. Ichthyosaurian eyes:

paleobiological information content in the sclerotic ring of Caypullisaurus

(Ichthyosauria, Ophthalmosauria). Journal of Vertebrate Paleontology 25(2):

330-337.

Gasparini, Z. and Fernández, M. 1997. Tithonian Marine Reptiles of the Eastern Pacific

Pp 223-253 in Calloway, J.M. and Nichols, E.L. eds. Ancient Marine Reptiles. San

Diego: Academic Press.

Godefroit, P. 1993. The skull of Stenopterygius longifrons (Owen, 1881). Revue de

Paléobiologie, Genéve. Vol. Spéciale 7:67-84.

Hammer, Ø., and Harper, D. 2006. Paleontological Data Analysis. Oxford: Blackwell

Publishing. 39

Hillis, D.M. and Bull, J. J. 1993. An empirical test of bootstrapping as a method for

assessing confidence in phylogenetic analysis. Systematic Biology 42(2): 182-192.

Huelsenbeck, J.P. 1994 . Comparing the stratigraphic record to estimates of phylogeny.

Paleobiology 20: 470-483.

Huene, F. Von. 1922. Die Ichthyosaurier des Lias und ihre Zusammenhänge. Monographien zur

Geologie und Paläontologie, 1. Verlag von Gebrüder Borntraeger, Berlin, VIII.

International Commission on Stratigraphy. 2007. Overview of Global Boundary Stratotype

Sections and Points (GSSP’s). Website: http://www.stratigraphy.org/. Accessed on

5/04/2008.

Jiang, D.-Y., Hao, W.-C., Maisch, M. W., Matzke, A. T., and Sun, Y.-L. 2005. A

mixosaurid ichthyosaur from the Middle Triassic of . Palaeontology 48: 869-882.

Kuhn, O. 1946. Ein Skelett von Ichthyosaurus (Platypterygius) hercynicus n. sp. aus dem Aptium

von Gitter. Ber. Naturf. Bericht- Naturforschende Gesellschaft, Bamberg. 29:69-82

Lane, A., Janis, C.M., and Sepkoski, J.J. Jr. 2005. Estimating paleodiversities: a test of the taxic

and phylogenetic methods. Paleobiology 31: 21-34.

Lydekker, R. 1888. Note of the classification of the (with notice of two new

species. Geological magazine, 5: 309-314.

Lydekker, R. 1889. Catalogue of the Fossil Reptilia and Amphibia in the

(Natural History). Part II. Containing the Orders Ichthyopterygia and .

London. Printed by Order of the Trustees of the British Museum, London, XXIII.

Maxwell, E.E., and Caldwell, M.W. 2006. A new genus of ichthyosaur from the Lower

Cretaceous of western Canada. Palaeontology 49: 1043-1052.

Maisch, M.W., and Matzke, A.T. 1997. Mikadocephalus gracilirostris n.gen., n. sp., a new 40

ichthyosaur from the Granzbitumenzone (-) of Monte San Giorgio

(). Paläontologische Zeitschrift 71: 267-289.

Maisch, M.W., and Matzke, A.T. 2000. The Ichthyosauria. Stuttgarter Beiträge zur Naturkunde,

Serie B 298: 1-159.

Mazin, J.M. 1981. Grippa longirostris Wiman. 1929. un Ichthyopterygia primitif du Trias

inférieur du Spitsburg. Bulletin du Muséum National d’Histoire Naturelle 4: 317-340.

Mazin, J.M. 1982. Affinités et phylogénie des Ichthyopterygia. Géobios, Mémoire

Spécial 6: 85-98.

McGowan, C. 1986. A putative ancestor for the -like ichthyosaur Eurhinosaurus.

Nature. 322(6078): 454-456.

McGowan, C. 1976. The description and phenetic relationships of a new ichthyosaur genus from

the Upper Jurassic of England. Canadian Journal of Earth Sciences 13: 668-683.

McGowan, C. 1991. , Spitfires, and Sea Dragons. Cambridge, MA: Harvard

Univ. Press.

McGowan, C. 1994. Temnodontosaurus risor is a juvenile of T. platydon (Reptilia,

Ichthyosauria). Journal of Vertebrate Paleontology 14: 472-479.

McGowan, C. and Motani, R. 2003. Ichthyopterygia. Handbook of Paleoherpetology, Part 8.

München: Verlag Dr. Friedrich Pfeil.

Motani, R. 1997. Phylogeny of the Ichthyosauria (Amniota: Reptilia) with special reference to

Triassic forms. Unpublished Ph.D. dissertation, University of Toronto, Toronto, 384 pp.

Motani, R. 1999a. On the evolution and of ichthyosaurian forefins. Journal of

Vertebrate Paleontology 19: 42-49.

Motani, R. 1999b. Phylogeny of the Ichthyopterygia. Journal of Vertebrate Paleontology 41

19: 473-496.

Motani, R. 2002. Scaling effects in caudal propulsion and the speed of ichthyosaurs.

Nature 309: 312.

Motani, R. 2005a. -shaped reptiles (Reptilia: Ichthyopterygia) in their

physical environments and constraints. Annual Review of Earth & Planetary

Sciences 33: 395-420.

Motani, R. 2005b. True skull roof configuration of Ichthyosaurus and Stenopterygius

and its implications. Journal of Vertebrate Paleontology 25(2): 338-342.

Mutterlose, J. 1986. Upper Jurassic belemnites from the Orville Coast, Western

Antarctica, and their palaeobiogeographical significance. British Antartic Survey

Bulletin 70: 1-22.

Owen, R. 1881. A monograph of the Fossil Reptilia of the Liassic Formations. Part III.

Paleontological Society, London. Ichthyopterygia pp. 83-134.

Reyment, R.A., and Tait, E.A. 1972. Biostratigraphical dating of the early history of the

South Atlantic Ocean. Philosophical Transactions, Royal Society, London B264:

55-95.

Sander, P.M. 2000. Ichthyosauria: their diversity, distribution, and phylogeny.

Paläontologische Zeitschrift 74:1-35.

Shultz, M. R., Fildani, A., and Suarez, M. 2003. Occurrence of the southernmost South

American ichthyosaur (Middle Jurassic-Lower Cretaceous), Parque Nacional Torres del

Paine, Patagonia, southernmost Chile. Palaios 18: 69-73.

Smith, A.B. 1994. Systematics and the Fossil Record: Documenting Evolutionary Patterns.

Cambridge, MA.: Blackwell Scientific Publications Inc. 42

Swofford, D.L. 2001. PAUP*. Phylogenetic Analysis Using Parsimony* (and Other Methods).

Version 4.0b10. Sunderland, Massachusetts: Sinauer Associates.

Wade, M. 1990. A review of the Australian Cretaceous longipinnate ichthyosaur

Platypterygius, (Ichthyosauria, Ichthyopterygia). Memoirs of the Queensland Museum

28(1): 115-137.

Willis, M.A. 1999. The gap excess ratio, randomization test and the goodness of fit of trees to

stratigraphy. Systematic Biology 48: 559-580.

Wilson, T.J. 1991. Transition from back-arc to foreland basin development in the

southernmost Andes: Stratigraphic record from the Ultima Esperanza District,

Chile. Geological Society of America Bulletin 103: 98-115. 43

APPENDIX A. CHARACTER DESCRIPTIONS Characters taken verbatim from Motani’s (1999b) analysis and Fernández’s (2007) analysis; the boldfaced italicized text are my interpretations and additions. Characters 1-105 are taken from Motani’s (1999b) analysis while 106-112 are taken from Fernández’s (2007) analysis.

1. Premaxilla posterior end – (0) concave and the dorsal process is longer than the ventral; (1) pointed and scarcely entering the external naris; (2) concave and the ventral process is longer than the dorsal. (Motani, 1999b)

2. Maxilla dorsal lamina – (0) absent; (1) present. [Maisch and Matzke, 1997: character 2 and 3(correlated)] (Motani, 1999b) I interpreted this character as something similar to striations or fine layering of the bone, which was hard to see in pictures.

3. Maxilla/external naris contact – (0) present; (1) absent. [Maisch and Matzke, 1997 character 1] (Motani, 1999b)

4. External naris orientation – (0) dorso-lateral; (1) lateral, scarcely visible in dorsal view. (Motani, 1999b) This character was difficult to code with many of the specimens in lateral view, but sometimes could be coded.

5. Nasal/external naris contact – (0) absent; (1) present. (Motani, 1999b)

6. Wide contact between nasal/postfrontal – (0) absent; (1) present. [Maisch and Matzke, 1997 character 9] (Motani, 1999b)

7. Nasal/parietal contact lateral to frontal – (0) absent; (1) present. (Motani, 1999b)

8. Prefrontal/postfrontal contact – (0) absent, the dorsal margin of the orbit being formed by the frontal; (1) present, eliminating the frontal from the dorsal margin of the orbit. (Motani, 1999b)

9. Postfrontal postero-lateral process – (0) absent; (1) present, overlying the post orbital. (Motani, 1999b)

10. Postfrontal participation in upper temporal fenestrate – (0) absent; (1) present. (Motani, 1999b)

11. Postorbital shape – (0) trilunate; (1) lunate, without posterior process. (Motani, 1999b)

12. Postorbital participation in upper temporal fenestra – (0) present; (1) absent. (Motani, 1999b)

13. Squamosal participation in upper temporal fenestra – (0) present; (1) absent; (2) squamosal absent. (Motani, 1999b)

44

14. Anterior terrace of upper temporal fenestra – (0) absent; (1) present, but small, reaching the posterior part of the frontal anteriorly; (2) present, and large reaching the nasal anteriorly. (Motani, 1999b) I interpreted this as a ridge on the back portion of the upper temporal fenestra that is located on the top of the skull, but it was, again, hard to see in laterally preserved specimens.

15. Frontal widest position – (0) located posteriorly; (1) at the nasal suture. (Motani, 1999b)

16. Sagittal eminence – (0) absent; (1) present, but small involving only the parietal; (2) present and large, involving the parietal, frontal, and nasal. (Motani, 1999b) I interpreted this character as the presence or absence of a ridge down the middle of the skull.

17. Parietal ridge – (0) absent; (1) present. The parietal ridge (sensu McGowan, 1973) is a feature that is only present in derived ichthyosaurs. (Motani, 1999b) I interpreted this as a ridge present between the two upper temporal fenestrate that could be seen in dorsal view.

18. Parietal supratemporal process – (0) short; (1) long. (Motani, 1999b)

19. Right and left parietals’ anterior process – (0) contacting each other anteriorly, eliminating frontal from pineal foramen; (1) narrowly separated anteriorly, forming parietal fork, and frontal dorsally visible along the pineal foramen; (2) widely open, resulting in absence of clear fork. [Mazin, 1982; Calloway, 1989: character 2] (Motani, 1999b)

20. Supratemporal posterior slope – (0) absent; (1) present. (Motani, 1999b)

21. Supratemporal posterior ridge – (0) absent; (1) present. (Motani, 1999b)

22. Supratemporal ventral process – (0) absent; (1) present. (Motani, 1999b)

23. Jugal/quadrajugal dorsal contact – (0) absent; (1) present. (Motani, 1999b)

24. Jugal shape – (0) triradiate; (1) lunate, or J-shaped. (Motani, 1999b)

25. Cheek orientation – (0) mostly lateral; (1) largely posterior. (Motani, 1999b) It was difficult to determine exactly where the check region was; but unfortunately all of the specimens I saw were in lateral view and I, therefore, was unable to code this characteristic.

26. Pterygoid, transverse flange – (0) antero-lateral; (1) posterio-lateral; (2) not well defined. (Motani, 1999b)

27. Interpterygoidal vacuity – (0) present; (1) absent, or extremely reduced. [Maisch and Matzke, 1997: character 22] (Motani, 1999b)

45

28. Ectopterygoid – (0) present; (1) absent. [Calloway, 1989 :char 9] (Motani, 1999b)

29. Basioccipital peg – (0) clearly present; (1) absent or extremely reduced. (Motani, 1999b)

30. Basioccipital extracondylar area – (0) wide; (1) reduced to a narrow band of concavity. (Motani, 1999b)

31. Basioccipital condyle – (0) flat or slightly concave; (1) hemispherical. [Calloway, 1989: character 1] (Motani, 1999b)

32. Angular lateral exposure – (0) extensive, at least as high and anteriorly as surangular exposure; (1) much smaller than surangular exposure. (Motani, 1999b) This character is referring to the angular bone present in the skull near the jaw joint and its exposure compared to the surangular, which is found right next to the angular.

33. Overbite – (0) absent or slight; (1) clearly present. (Motani, 1999b)

34. Snout extremely slender – (0) no; (1) yes. (Motani, 1999b) This character seems very subjective, and I looked at how Motani (1999b) coded different species and based my interpretations off of his previous codings.

35. Replacement teeth – (0) appear outside the pulp cavity of the predecessor; (1) inside. (Motani, 1999b) I was unable to see any replacement teeth in any specimen due to preservation of specimens.

36. Plicidentine – (0) absent; (1) present. (Motani, 1999b) To be able to see this character, a tooth must be broken or sawed across and I did not have access to any of those.

37. Tooth horizontal section – (0) circular; (1) disto-medially compressed; (2) laterally compressed. (Motani, 1999b)

38. Posterior tooth crown – (0) conical; (1) rounded; (2) flat. (Motani, 1999b)

39. Tooth size relative to skull width – (0) normal (over 0.10); (1) small (below 0.05). (Motani, 1999b)

40. Maxillary tooth row – (0) single; (1) multiple. (Motani, 1999b)

41. Upper dental groove – (0) present throughout jaw margin; (1) only present anteriorly; (2) absent. (Motani, 1999b) I coded this character based on where the teeth were seen in the jaw. Because many specimens were pictures, there was no way to tell if it was actually a dental groove present or individual sockets for the teeth.

42. Lower dental groove – (0) present throughout jaw margin; (1) only present anteriorly; (2) absent. (Motani, 1999b) See interpretation for char 41.

46

43. Bony fixation of teeth – (0) present; (1) absent. (Motani, 1999b)

44. Pterygoidal teeth – (0) present; (1) absent. (Motani, 1999b)

45. Interclavicle shape – (0) cruciform; (1) triangular; (2) T-shaped. (Motani, 1999b)

46. antero-dorsal margin – (0) fan-shaped; (1) emarginated; (2) straight. [Calloway, 1989: character 23] (Motani, 1999b)

47. Scapular blade shaft – (0) absent; (1) present at least proximally. (Motani, 1999b)

48. Scapular axis and glenoid facet orientations – (0) nearly parallel; (1) at 60 degrees or more. (Motani, 1999b) I interpreted facets as worn down areas on the bone where another bone would attach.

49. facet on scapula – (0) scapula and coracoid fused; (1) absent; (2) equal or smaller than glenoid facet of scapula; (3) twice as large as glenoid facet. (Motani, 1999b)

50. Ossified – (0) absent; (1) present. (Motani, 1999b)

51. Ossified cleithrum – (0) present; (1) absent. (Motani, 1999b)

52. Humerus distal articular facets – (0) not terminal; (1) terminal, radial facet being larger than ulnar facet; (2) terminal, two facets being nearly equal. (Motani, 1999b)

53. Humerus anterior flange – (0) absent; (1) present and complete; (2) present but reduced proximally. [Calloway, 1989: character 29] (Motani, 1999b)

54. Humerus relative width exclusive of anterior flange – (0) much longer than wide; (1) nearly squarish. [Calloway, 1989: character 28] (Motani, 1999b)

55. Humerus, distal and proximal ends, exclusive of anterior flange – (0) nearly equal; (1) distal end wider than proximal end. (Motani, 1999b)

56. Ridge on humerus plate-like – (0) no; (1) yes. (Motani, 1999b)

57. Humerus antero-distal facet for sesamoid – (0) absent; (1) present. [Godefroit, 1993: character 10] (Motani, 1999b) A sesamoid bone holds tendons away from the other bones and can be found between the metatarsals and phalanges.

58. Propodial + epipodial versus length – (0) propodial + epipodial length longer; (1) manus longer. (Motani, 1999b) I interpreted the propodial as the humerus/femur, the epipodial as the radius and ulna/ tibia and fibia, and the manus as everything distal to that.

47

59. Radius peripheral ‘shaft’ – (0) complete or nearly complete; (1) notch or largely reduced; (2) absent. (Motani, 1999b). This character was hard to interpret, but I looked at how Motani had coded the different genera and also took into account the idea that this was talking about a shaft on the peripheral or closer towards the radius than the fin.

60. Radius contiguous ‘shaft’ – (0) complete or nearly complete; (1) notch or largely reduced; (2) absent. (Motani, 1999b) Interpreted this similarly to char 59 but looked for a shaft on the edge of the radius.

61. Radius L/W ratio – (0) longer than wide; (1) wider than long. (Motani, 1999b)

62. Ulna peripheral ‘shaft’ – (0) complete or nearly complete; (1) notch or largely reduced; (2) absent. (Motani, 1999b) See interpretation for char 59.

63. Ulna contiguous ‘shaft’– (0) complete or nearly complete; (1) notch or largely reduced; (2) absent. (Motani, 1999b) See interpretation for char 60.

64. Radius/ulna relative size – (0) nearly equal; (1) radius larger than ulna. (Motani, 1999b)

65. Radiale, anterior notch – (0) absent; (1) preaxially present. (Motani, 1999b) Looked for actual notch in the radiale bone.

66. Ulnare/intermedium relative size – (0) ulnare larger than intermedium; (1) intermedium larger than ulnare; (2) intermedium lost. (Motani, 1999b)

67. Manual pisiform – (0) present; (1) absent. (Motani, 1999b) The pisiform can be found in the proximal row of the carpus, where the ulna joins the carpus and is considered a sesamoid bone.

68. Mc I peripheral ‘shaft’ – (0) complete or nearly complete; (1) notch or largely reduced; (2) absent; (3) mc 1 not ossified. (Motani, 1999b) See interpretation for char 59.

69. Mc III ‘shaft’ – (0) present; (1) absent. [Calloway, 1989: character 33] (Motani, 1999b)

70. Mc V peripheral ‘shaft’ – (0) complete or nearly complete; (1) absent; (2) mc 5 not ossified. (Motani, 1999b) See interpretation for char 59.

71. Manual digit 2 distal elements peripheral ‘shaft’ – (0) complete; (1) largely reduced or notch; (2) absent. (Motani, 1999b) See interpretation for char 59.

72. Manual accessory digit VI – (0) absent; (1) present. [Mazin, 1982: character 4] (Motani, 1999b)

73. Manual digit S4-5 – (0) absent; (1) present. (Motani, 1999b)

74. Manual centralia – (0) present; (1) absent. (Motani, 1999b) 48

75. Manual anterior sesamoid e, and the digit distal to it – (0) absent; (1) present. (Motani, 1999b) See interpretation for char 57.

76. More than one extra anterior digit – (0) absent; (1) present. (Motani, 1999b)

77. Maximum phalangeal count – (0) five or less; (1) seven or more. (Motani, 1999b)

78. Interdigital separation – (0) present; (1) absent. (Motani, 1999b)

79. Forelimb/hindlimb ratio – (0) nearly equal or hind limb longer; (1) forelimb longer but less than twice as hindlimb; (2) forelimb longer twice as much as hindlimb. [Calloway, 1989: char 26; Godefroit, 1993: character 5] (Motani, 1999b)

80. Iliac blade shape – (0) with thick shaft; (1) plate-like; (2) narrow and styloidal. [Calloway, 1989: character 24] (Motani, 1999b)

81. Iliac antero-medial prominence – (0) present; (1) absent. (Motani, 1999b)

82. Thyroid fenestra – (0) absent; (1) one median opening; (2) two openings, being medially separated. (Motani, 1999b)

83. Ischium-pubis fusion in adults – (0) complete; (1) absent; (2) present only medially; (3) present medially and laterally. [Mazin, 1982: character 13] (Motani, 1999b)

84. Pubis, obturator foramen – (0) completely enclosed in pubis; (1) mostly in pubis but open on one side; (2) part of obturator fossa. [Callaway, 1989: character 25] (Motani, 1999b)

85. Pubis, styloidal or plate-like – (0) plate-like; (1) styloidal. (Motani, 1999b)

86. Pubis/ischium relative length – (0) nearly equal or ischium slightly larger than pubis; (1) pubis twice as long as ischium. (Motani, 1999b)

87. Ischium, styloidal or plate-like – (0) plate-like; (1) styloidal. (Motani, 1999b)

88. Tibia and fibula – (0) in contact or closely placed with each other; (1) widely separated from each other. (Motani, 1999b)

89. Pes digit 1 – (0) present; (1) absent. (Motani, 1999b)

90. Tibia L/W ratio – (0) longer than wide; (1) wider than long. (Motani, 1999b)

91. Tibia contiguous ‘shaft’ – (0) complete or nearly complete; (1) absent. (Motani, 1999b)

92. Tibia peripheral ‘shaft’ – (0) complete or nearly complete; (1) notch or largely reduced; (2) absent. (Motani, 1999b) 49

93. Fibula posterior extent – (0) not fixed, fibula being mobile relative to femur; (1) much posterior to femur; (2) about the same level as femur. (Motani, 1999b)

94. Atlantal pleuro centrum – (0) separate from axis; (1) fused with axis. (Motani, 1999b) This is interpreted as the first vertebrae and its attachment to the axis, which was hard to tell when specimens were preserved in lateral view.

95. Presacral count – (0) 30 or less; (1) between 40 and 50; (2) 55 or more. [Calloway, 1989: character 13] (Motani, 1999b) (3) 31-39. This character state was added because I found some specimens with presacral vertebrae between 31 and 39.

96. Caudal peak – (0) absent; (1) present. (Motani, 1999b)

97. Posterior dorsal centra shape – (0) cylindrical; (1) discoidal. [Calloway, 1989: character 19] (Motani, 1999b)

98. Mid-caudal centra height change – (0) gradual decrease; (1) increase; (2) sudden decrease. (Motani, 1999b) This is interpreted as the change seen after the peak in the tail.

99. Cervical bicipital rib facets – (0) absent; (1) present. [Calloway, 1989 character 14] (Motani, 1999b)

100.Posterior-dorsal rib facets – (0) absent; (1) present at least near pelvic girdle. [Calloway, 1989:char 14] (Motani, 1999b)

101.Antero-dorsal rib facets – (0) confluent with anterior facet in at least some centra; (1) not confluent in any of the centra. (Motani, 1999b)

102.Anterior dorsal neural spine – (0) normal; (1) narrow, high, and straight. [Calloway,1989: character 16] (Motani, 1999b)

103.Neural spine anticlination in tail – (0) absent; (1) present. (Motani, 1999b)

104.Sacral ribs – (0) two, distinguishable; (1) not distinguishable. (Motani, 1999b)

105.Posterior gastralia – (0) present; (1) absent. (Motani, 1999b)

106.Anterior process of maxilla in lateral view – (0) long and broad; (1) long and narrow; (2) short. (Fernández, 2007: character 1)

107.Descending process of the nasal on the posterior dorsal border of the nares – (0) absent; (1) present. (Fernández, 2007: character 2)

50

108.Squamosal – (0) without posterior descending process; (1) squamosal with a posterior descending process; (2) squamosal absent. (Fernández, 2007: character 4)

109.Postorbital – (0) broad; (1) narrow. (Fernández, 2007: character 7)

110.Humerus/intermedium contact – (0) absent; (1) present. (Fernández, 2007: character 16)

111.Location of the intermedium – (0) between radius and ulna; (1) distal to the ulna. (Fernández, 2007: character 19)

112.More than one anterior accessory digit – (0) absent; (1) present. (Fernández, 2007: character 21) 51

APPENDIX B. SPECIES DATA Presented below is a compilation of all the species character state codes I collected through museums and primary literature. The “?” represents unknown character states while “&” show of character states. See Appendix A for characters and character descriptions.

Char 2 3 4 5 6 7 8 9 10 11 12 13 Species name 1 Suevoleviathan disinteger 2 0 ? 1 0 1 ? 1 1 1 1 1 ? Leptonectes tenuirostris ? ? 0 1 1 ? ? ? ? 1 1 1 ? Leptonectes solei 2 0 ? 1 0 ? ? 1 1 1 1 1 ? Leptonectes moorei 2 2 2 2 2&1 2&0 2 2 2 2 2&0 2 2 Excalibosaurus costini ? 0 ? 1 0 ? ? 1 1 ? 1 1 ? Eurhinosaurus longirostris 2 0 0 1 0 ? 0 1 1 1 1 1 ? Temnodontosaurus trigonodon ? ? ? ? ? ? ? ? ? ? ? ? ? Ichthyosaurus communis 0 1 1 0 1&0 1 ? 1 ? 1 1 1 2 Ichthyosaurus conybeari 2 ? 0 1 0 0 1 0 1 1 1 1 ? Stenopterygius quadrisissus 2 1 0 1 1 1 0 1 1 ? 1 ? 1 Stenopterygius megacephalus ? ? ? ? ? ? ? ? ? ? ? ? ? Stenopterygius megalorhinus ? ? ? ? ? ? ? ? ? ? ? ? ? Stenopterygius macrophasma 2 ? 0 ? 1 0 0 1 1 1 1 1 1 Stenopterygius cuneiceps 1 ? 0 ? 1 0 1 1 1 ? 1 ? ? Ophthalmosaurus icenicus 2 ? 1 ? 1 ? 1 ? ? ? 0 1 ? Ophthalmosaurus discus 2 0 1 1 0 1 0 1 1 1 1 1 1 Brachypterygius extremus 2 0 ? 1 1 ? ? 1 1 ? 1 1 ? Caypullisaurus bonaparte ? 0 ? 1 ? ? ? ? ? ? 1 ? ?

Species name 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Suevoleviathan disinteger 0 1 1 1 0 1 1 ? ? 1 1 0 ? ? ? Leptonectes tenuirostris ? ? ? ? ? ? ? ? ? ? 1 1 ? 0 ? Leptonectes solei 0 ? ? 1 0 ? 1 ? ? ? 1 1 ? 0 ? Leptonectes moorei 2 2 2 2 2 2 2 2 2 2&1 2&1 2 2 2 2 Excalibosaurus costini 0 ? ? ? ? ? ? ? ? ? 1 1 ? ? ? Eurhinosaurus longirostris 0 ? 0 1 0 ? 1 ? ? 1 1 1 ? ? ? Temnodontosaurus trigonodon ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Ichthyosaurus communis ? 0 ? 1 ? 2 1 1 ? ? 1 ? ? ? ? Ichthyosaurus conybeari 0 1 ? ? ? ? ? ? ? 1 1 ? ? ? ? Stenopterygius quadrisissus ? ? ? ? ? ? ? ? ? 1 1 0 ? ? ? Stenopterygius megacephalus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Stenopterygius megalorhinus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Stenopterygius macrophasma 1 0 ? 1 ? ? ? ? ? 1 0 ? ? ? ? Stenopterygius cuneiceps ? ? ? ? ? ? ? ? ? 1 0 ? ? ? ? Ophthalmosaurus icenicus ? ? ? ? ? ? ? ? ? 1 1 ? 1 0 1 Ophthalmosaurus discus 0 1 0 1 0 2 1 1 1 1 1 0 2 0 1 Brachypterygius extremus 0 ? ? ? ? ? ? ? ? 1 ? 0 ? ? ? Caypullisaurus bonaparte ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 52

Species name 29 30 31 32 33 34 35 36 37 38 39 40 41 Suevoleviathan disinteger ? ? ? 1 0 0 ? 1 0 0 0 ? ? Leptonectes tenuirostris ? ? ? ? 0 1 ? ? ? ? ? ? 1 Leptonectes solei ? 0 1 1 0 1 ? 1 0 0 1 0 0 Leptonectes moorei 2 2 2 2&0 2&0 1 ? ? ? ? ? ? ? Excalibosaurus costini ? ? ? 1 1 1 ? 1 0 0 1 0 0 Eurhinosaurus longirostris ? 0 1 1 1 1 ? 1 0 0 1 0 0 Temnodontosaurus trigonodon ? ? ? ? ? ? ? ? ? ? ? ? ? Ichthyosaurus communis 0 ? ? 0 0 0 ? ? ? ? ? ? 0 Ichthyosaurus conybeari ? ? ? 0 0 1 ? ? ? ? ? ? 0 Stenopterygius quadrisissus ? ? ? ? 0 1 ? ? ? ? ? ? 0 Stenopterygius megacephalus ? ? ? ? 0 1 ? ? ? ? ? ? 0 Stenopterygius megalorhinus ? ? ? 1 0 0 ? ? ? ? ? 0 0 Stenopterygius macrophasma ? ? ? ? 0 1 ? ? ? ? ? ? 1 Stenopterygius cuneiceps ? ? ? 0 0 0 ? ? ? ? ? ? ? Ophthalmosaurus icenicus ? ? ? ? ? ? ? ? ? ? ? 0 0 Ophthalmosaurus discus 1 1 1 0 0 0 1 1 0 0 ? 0 0 Brachypterygius extremus 1 1 1 0 0 0 ? 1 0 0 0 0 0 Caypullisaurus bonaparte ? ? ? 0 ? 0 ? ? ? ? ? ? ?

Species name 42 43 44 45 46 47 48 49 50 51 52 53 Suevoleviathan disinteger 0 ? ? ? 2 1 0 3 0 1 2 2 Leptonectes tenuirostris ? ? ? ? 0&2 1 ? ? ? ? 2&1 0&2 Leptonectes solei 0 1 1 2 2 1 0 3 0 1 2 2 Leptonectes moorei ? ? ? ? 2 1 ? ? ? ? 2 0&2 Excalibosaurus costini 0 1 1 ? 2 1 0 3 0 1 2 2 Eurhinosaurus longirostris 0 1 1 2 2 1 0 3 0 1 2 2 Temnodontosaurus trigonodon ? ? ? ? ? ? ? ? ? ? 2 2 Ichthyosaurus communis 0 ? ? ? 0 ? ? ? ? 0 2 2 Ichthyosaurus conybeari 0 ? ? ? ? ? ? ? ? ? 2 2 Stenopterygius quadrisissus 0 ? ? 2 0 1 ? ? ? ? 2 2 Stenopterygius megacephalus 0 ? ? ? ? ? ? ? ? ? 2 ? Stenopterygius megalorhinus 0 ? ? ? ? ? ? ? ? ? 2 0 Stenopterygius macrophasma 0 ? ? ? 0 1 ? ? ? ? 2 ? Stenopterygius cuneiceps ? ? ? ? ? ? ? ? ? ? 2 2 Ophthalmosaurus icenicus 0 ? ? ? 2 1 ? ? ? 1 1 0 Ophthalmosaurus discus 0 1 1 2 2 1 0 3 0 1 2 2 Brachypterygius extremus 0 1 1 ? ? ? ? ? ? ? 2 2 Caypullisaurus bonaparte ? ? ? 2 2 1 0 3 ? ? 2 2

53

Species name 54 55 56 57 58 59 60 61 62 63 64 65 Suevoleviathan disinteger 0 1 0 0 1 2 2 1 2 2 0 1 Leptonectes tenuirostris 0 1&0 ? ? 1 1 1 0&1 0&1 0 0 0 Leptonectes solei 0 0 0 0 1 1 1&2 0&1 2 1&2 0 0&1 Leptonectes moorei 1&0 0&1 0 ? 1 0&1 0 0 0 0 0 0 Excalibosaurus costini 0 1 0 0 1 1 2 1 2 2 0 0 Eurhinosaurus longirostris 0 1 0 0 1 2 2 1 2 2 0 0 Temnodontosaurus trigonodon 1 1 0 0 1 1 0 1 0 1 0 1 Ichthyosaurus communis 0&1 0 0 0 0&1 1 1 1 1 0 0 0 Ichthyosaurus conybeari 1 0 0 0 1 0 0 1 1 0 0 0 Stenopterygius quadrisissus 1 1 0 ? 1 0 0 1 0 0 0 0&1 Stenopterygius megacephalus 0 0 ? 0 1 1 0 1 1 0 0 1 Stenopterygius megalorhinus 1 1 0 0 1 0 1 1 1 0 0 0 Stenopterygius macrophasma 0 1 ? ? 1 1 1 1 1 0 0 1 Stenopterygius cuneiceps 0 0 0 ? 1 0 1 0 1 1 0 1 Ophthalmosaurus icenicus 0 1 0 ? 1 2 2 0 2 2 0 0 Ophthalmosaurus discus 0 0 1 1 1 2 2 1 2 2 0 0 Brachypterygius extremus 0 0 1 0 1 2 2 1 2 2 0 0 Caypullisaurus bonaparte 0 0 1 1 1 2 2 1 2 2 0 0

Species name 66 67 68 69 70 71 72 73 74 75 76 77 78 Suevoleviathan disinteger 1 1 3 1 1 2 1 1 1 0 0 1 1 Leptonectes tenuirostris 1 1 3 1&0 1 2&0 1&0 0&1 1 0 0 0 0 Leptonectes solei 1 1 3 1 1 2 0 0 1 0 0 1 1 Leptonectes moorei 1 1 3 1&0 1 2&0 0&1 0 1 0 0 0 1 Excalibosaurus costini 0 1 3 1 2 2 0 0 1 0 0 1&0 1 Eurhinosaurus longirostris 0 0 3 1 1 2 0 0 1 0 0 1 1 Temnodontosaurus trigonodon 1 ? 0 0 0 0 1 0 1 ? 0 0 1 Ichthyosaurus communis 0 0 1 0 1&0 1 1 1 1 ? 0&1 1 1 Ichthyosaurus conybeari 1 ? 1 0 0 0 1 1 1 ? 0 0 1 Stenopterygius quadrisissus 0&1 0&1 3 1 1 0 1 0 1 1 0 0 1 Stenopterygius megacephalus 1 ? 1 0 2 0 1 1 1 ? 0 0 1 Stenopterygius megalorhinus 0 ? 1 0 1 1 1 0 1 ? 0 0 1 Stenopterygius macrophasma 0 ? 1 0 1 0 1 0 1 ? 0 0 1 Stenopterygius cuneiceps 1 ? 1 0 1 1 1 0 1 ? 0 0 1 Ophthalmosaurus icenicus 1 ? 3 1 1 2 1 1 1 ? 0 1 1 Ophthalmosaurus discus 1 0 3 1 1 2 1 0 1 1 0 1 1 Brachypterygius extremus 1 0 3 1 1 2 1 ? 1 1 0 1 1 Caypullisaurus bonaparte 1 1 3 1 1 2 1 0 1 1 1 1 1

54

Species name 79 80 81 82 83 84 85 86 87 88 89 90 91 Suevoleviathan disinteger 1 1 0 2 1 2 1 0 1 0 1 1 1 Leptonectes tenuirostris 2 ? ? 1 3 ? 0 0 0 1 0&1 0 1&0 Leptonectes solei 2 2 1 2 2&3 2 0 1 1 1 1 1 1 Leptonectes moorei ? ? ? ? ? ? ? ? ? ? ? ? ? Excalibosaurus costini 2 2 ? ? ? 2 ? ? 0 1 0 0 ? Eurhinosaurus longirostris 1 2 1 2 1&2 2 1 0 1 1 1 1 1 Temnodontosaurus trigonodon 1 2 ? 1 2 1 0 0 0 ? ? ? ? Ichthyosaurus communis 2 ? ? ? ? ? ? ? ? 0 0 1 1&0 Ichthyosaurus conybeari 1 ? ? ? ? ? ? ? ? 0 0 1 1 Stenopterygius quadrisissus ? ? ? ? ? ? ? ? ? ? ? ? ? Stenopterygius megacephalus 2 ? ? ? ? ? ? ? ? 0 0 0 1 Stenopterygius megalorhinus 1 ? ? ? ? ? ? ? ? 0 0 1 1 Stenopterygius macrophasma 1&2 ? ? ? ? ? ? ? ? 0 0 1 1 Stenopterygius cuneiceps 1 0 ? ? ? ? ? ? ? 1 0 1 1 Ophthalmosaurus icenicus 1 1 ? 0 ? ? ? ? ? 0 1 1 1 Ophthalmosaurus discus 2 2 1 2 3 2 1 0 1 0 1 1 1 Brachypterygius extremus ? ? ? ? ? ? ? ? ? ? ? ? ? Caypullisaurus bonaparte ? ? ? ? ? ? ? ? ? ? ? ? ?

Species name 92 93 94 95 96 97 98 99 100 101 102 103 Suevoleviathan disinteger 1 2 ? 1 1 1 2 1 1 1 0 ? Leptonectes tenuirostris 1&0 2 0 1 ? 0 ? ? ? ? ? ? Leptonectes solei 1 2 1 2 1 1 2 1 1 1 0 ? Leptonectes moorei ? ? ? ? ? 0 ? ? ? ? ? ? Excalibosaurus costini ? 1 ? 1 1 1 ? ? 1 ? ? ? Eurhinosaurus longirostris 1 2 1 1 1 1 2 1 1 1 0 ? Temnodontosaurus trigonodon ? ? ? 1 ? ? ? ? ? ? ? ? Ichthyosaurus communis 2&1 2 1 1 1 1 2&0 ? ? ? 0 1 Ichthyosaurus conybeari 2 2 ? 3 1 ? 0 ? ? ? 1 ? Stenopterygius quadrisissus ? ? 1 1 1 0 2 1 1 1 0&1 1 Stenopterygius megacephalus 1 2 ? 1 1 ? 0 ? ? ? 1 1 Stenopterygius megalorhinus 0 2 ? 3 1 0 0 ? ? ? 0 0 Stenopterygius macrophasma 1 2 ? 2 1 ? 2 ? ? ? 0 0 Stenopterygius cuneiceps 0 2 ? 3 1 1 2 ? ? ? 0 1 Ophthalmosaurus icenicus 2 2 ? 1 1 ? 2 ? ? ? 0 1 Ophthalmosaurus discus 2 2 1 1 1 1 2 1 1 1 0 ? Brachypterygius extremus ? ? ? ? ? 1 ? ? ? ? ? ? Caypullisaurus bonaparte ? ? 1 3 ? 1 ? ? 1 ? 0 ?

55

Species name 104 105 106 107 108 109 110 111 112 Suevoleviathan disinteger 1 1 ? ? ? 0 0 0 0 Leptonectes tenuirostris 0 ? 1 1 ? 1 0 0&1 0 Leptonectes solei 1 ? ? ? ? ? ? ? ? Leptonectes moorei ? ? 1 1&0 1 1&0 1 1 0 Excalibosaurus costini ? ? ? ? ? ? 0 1 0 Eurhinosaurus longirostris 1 ? ? ? ? ? ? ? ? Temnodontosaurus trigonodon ? ? ? ? ? ? 0 0 0 Ichthyosaurus communis 1 1 2 0 2 1 0 0 0&1 Ichthyosaurus conybeari 0 0 1 0 ? 1 0 0 0 Stenopterygius quadrisissus 1 ? 1 ? ? 1 0 0 0 Stenopterygius megacephalus 1 1 ? ? ? ? 1 1 0 Stenopterygius megalorhinus 1 1 ? ? ? ? 0 0 0 Stenopterygius macrophasma 0 1 ? 1 0 0 0 0 0 Stenopterygius cuneiceps 1 1 0 ? ? 0 0 0 0 Ophthalmosaurus icenicus 1 1 2 1 1 1 0 0 0 Ophthalmosaurus discus 1 ? 2 1 1 1 0 0 0 Brachypterygius extremus ? ? 1 ? ? 0 1 0 0 Caypullisaurus bonaparte ? ? 1 0&1 0 0 0 1 1 56

APPENDIX C. UNINFORMATIVE CHARACTERS

A list of the parsimoniously uninformative characters from the species-level analysis.

Character Character Name Reason for Uninformative Status Number 4 External naris orientation Autapomorphies 8 Prefrontal/postfrontal contact Autapomorphies 9 Postfrontal postero-lateral process 10 Postfrontal participation in UTF Autapomorphy 12 Postorbital participation in UTF Autapomorphy 17 Parietal ridge Autapomorphy 18 Parietal supratemporal process Autapomorphy 19 Right and left parietals’ anterior process Autapomorphy 20 Supratemporal posterior slope Autapomorphy 21 Supratemporal posterior ridge Autapomorphy 22 Supratemporal ventral process Autapomorphies 23 Jugal/quadratojugal dorsal contact Autapomorphy 26 Pterygoid, transverse flange Autapomorphy 27 Interpterygoidal vacuity Autapomorphy 28 Ectopterygoid Autapomorphy 31 Basioccipital condyle Autapomorphy 33 Overbite Autapomorphies 35 Replacement teeth Constant 36 Plicidentine Constant 37 Tooth horizontal section Constant 38 Posterior tooth crown Constant 40 Maxillary tooth row Constant 42 Lower dental groove Constant 43 Bony fixation of teeth Constant 44 Pterygoidal teeth Constant 45 Interclavical shape Constant 47 Scapular blade shaft Constant 48 Scapular axis and glenoid facet Constant orientations 49 Coracoid facet on scapula Constant 50 Ossified sternum Constant 51 Ossified cleithrum Constant 52 Humerus distal articular facets Autapomorphy 53 Humerus anterior flange ? 58 Propodial + epipodial versus manus Autapomorphy length 64 Radius/ulna relative size Constant 74 Manual centralia Constant 76 More than one extra anterior digit Autapomorphy 57

78 Interdigital separation Autapomorphy 81 Iliac antero-medial prominence Autapomorphy 84 Pubis, obturator foramen Autapomorphy 86 Pubis/ischium relative length Autapomorphy 87 Ischium, styloidal or plate-like ? 91 Tibia contiguous ‘shaft’ Autapomorphy/Constant 93 Fibula posterior extent Autapomorphy 94 Atlantal pleuro centrum Autapomorphy 96 Caudal peak Constant 99 Cervical bicipital rib facet Constant 100 Posterior-dorsal bicipital rib facet Constant 101 Antero-dorsal rib facets Constant 103 Neural spine anticlination in tail ? 105 Posterior gastralia Autapomorphy 112 More than one anterior accessory digit Autapomorphy

58

APPENDIX D. Calculating MIG for GER

A diagram showing how to calculate MIG for a portion of the tree for derived ichthyosaurs (first five species). MIG is the minimum implied gap, or the length of the ghost ranges present based on a particular cladogram (Hammer and Harper, 2006).

59

APPENDIX E. Calculating Gmin and Gmax for GER

A diagram showing how to calculate Gmin and Gmax for a portion of the tree for derived ichthyosaurs (first five species). Gmin is the minimum possible sum of ghost ranges or the sum of distances between consecutive first appearance datums (FADs). Gmax is the maximum possible sum of ghost ranges or the sum of distances from the first FAD to all of the other FADs (Willis 1999, Hammer and Harper, 2006).

60

APPENDIX F. GENUS-LEVEL CHARACTER STATE CHANGES

Character state changes from the genus-level analysis using Motani’s (1999b) dataset, as seen in Figure 6.

Node/Branch# Character: Character State Change

1 82: 1Æ 0 83: 1Æ0

2 2: 0Æ1 18: 0Æ1 24: 0Æ1 27: 0Æ1 51: 0Æ1

3 10: 0Æ1 26: 0Æ1 50: 0Æ1

4 55: 0Æ1 105: 0Æ1

5 50: 0Æ1

6 11: 0Æ1 20: 0Æ1 21: 0Æ1 32: 0Æ1 36: 0Æ1 37: 0Æ1 44: 0Æ1 49: 0Æ1 52: 0Æ1 53: 0Æ1 66: 0Æ1 74: 0Æ1 79: 0Æ1 80: 0Æ1 93: 0Æ1 95: 0Æ1 104: 0Æ1

7 1: 0Æ1 2: 1Æ0 4: 0Æ1 61

8: 0Æ1 18: 1Æ0 26: 0Æ2 27: 1Æ0 67: 0Æ1

8 9: 0Æ1 10: 0Æ1 14: 0Æ1 28: 0Æ1 68: 0Æ1 70: 0Æ1 78: 0Æ1 96: 0Æ1 103: 0Æ1

9 77: 0Æ1

10 39: 0Æ1

11 44: 1Æ0 79: 1Æ0 80: 1Æ0 104: 1Æ0

12 13: 0Æ1 22: 0Æ1 55: 0Æ1 58: 0Æ1 68: 1Æ2 81: 0Æ1 99: 0Æ1 100: 0Æ1

13 34: 0Æ1 42: 0Æ2 54: 0Æ1 60: 0Æ1 66: 1Æ2 84: 0Æ1 90: 0Æ1 95: 1Æ2 100: 1Æ0

14 40: 1Æ2 78: 1Æ2 62

15 4: 0Æ1 8: 0Æ1 12: 0Æ1 16: 0Æ1 19: 0Æ1 23: 0Æ1 26: 0Æ2 37: 1Æ0 39: 1Æ0 45: 0Æ1 62: 0Æ2 77: 0Æ1 97: 0Æ1

16 5: 0Æ1 6: 0Æ1 41: 0Æ2 46: 0Æ1 95: 1Æ2 100: 1Æ0

17 42: 0Æ2 62: 2Æ1

18 16: 1Æ0 54: 0Æ1 55: 1Æ0

19 17: 0Æ1 18: 1Æ0 31: 0Æ1 94: 0Æ1

20 1: 0Æ1 14: 1Æ2 16: 1Æ2 34: 0Æ1 86: 0Æ1 98: 0Æ1 102: 0Æ1

21 38: 0Æ1 39: 0Æ1

22 37: 0Æ2 63

41: 0Æ1 42: 0Æ1 72: 0Æ1

23 38: 0Æ2 40: 0Æ1

24 2: 1Æ0 15: 0Æ1 45: 1Æ2 49: 1Æ2 55: 1Æ0 59: 0Æ1 65: 0Æ1 67: 0Æ1 68: 2Æ3 69: 0Æ1 70: 1Æ2 71: 0Æ1 89: 0Æ1 101: 0Æ1 105: 0Æ1

25 13: 1Æ0 18: 0Æ1 41: 0Æ2 42: 0Æ2 54: 0Æ1 60: 0Æ1 66: 1Æ2 84: 0Æ1 90: 0Æ1 95: 1Æ2 100: 1Æ0

26 63: 0Æ2 101: 1Æ0

27 47: 0Æ1 48: 0Æ1 59: 1Æ2 64: 0Æ1

28 5: 0Æ1 60: 1Æ0 65: 1Æ0 64

29 49: 2Æ3 53: 1Æ2 70: 2Æ1 72: 0Æ1 84: 0Æ2

30 93: 1Æ2

31 54: 0Æ1 62: 2Æ1 80: 2Æ1

32 46: 0Æ2 47: 0Æ1 60: 0Æ1 61: 0Æ1 63: 0Æ2 82: 0Æ1 85: 0Æ1 92: 0Æ1

33 55: 0Æ1 59: 1Æ2 71: 1Æ2

34 65: 1Æ2 89: 1Æ0 102: 0Æ1

35 87: 0Æ1 90: 0Æ1 91: 0Æ1 93: 1Æ2

36 73: 0Æ1 80: 2Æ1

37 59: 2Æ1 81: 0Æ1 83: 1Æ2

38 72: 1Æ0 88: 0Æ1

39 7: 0Æ1 65

65: 1Æ0 70: 1Æ2

40 3: 1Æ0 16: 1Æ0 25: 0Æ1 34: 0Æ1 39: 0Æ1

41 55: 1Æ0 79: 1Æ2

42 70: 1Æ2

43 33: 0Æ1

44 59: 1Æ2 67: 1Æ0

45 19: 1Æ2 67: 1Æ0 79: 1Æ2 83: 2Æ3

46 7: 0Æ1 73: 0Æ1

47 13: 1Æ2 55: 1Æ0 59: 1Æ2 65: 1Æ0 92: 1Æ2

48 66: 1Æ0 83: 3Æ(1/2)

49 29: 0Æ1 30: 0Æ1 32: 1Æ0 56: 0Æ1 75: 0Æ1

50 57: 0Æ1

51 13: 2Æ1 16: 1Æ0 66

52 7: 0Æ1 76: 0Æ1

53 67: 0Æ1 67

APPENDIX G. SPECIES-LEVEL CHARACTER STATE CHANGES

Character state changes from the species-level analysis, as seen in Figure 9.

Node/Branch# Character: Character State Change 1 16: 0Æ1 65: 0Æ1

2 5: 0Æ1 7: 0Æ1 11: 1Æ0 52: 2Æ1 53: 2Æ0 61: 1Æ0 82: 2Æ0 92: 1Æ2 106: 1Æ2 109: 0Æ1

3 19: 1Æ2 26: 1Æ2 73: 1Æ0 79: 1Æ2 80: 1Æ2 81: 0Æ1 83: 1Æ2

4 29: 0Æ1 30: 0Æ1 32: 1Æ0 55: 1Æ0 56: 0Æ1 57: 0Æ1 75: 0Æ1 83: 2Æ3 92: 1Æ2

5 106: 1Æ2 109: 0Æ1

6 67: 0Æ1

7 5: 0Æ1 57: 1Æ0 110: 0Æ1

68

8 76: 0Æ1 95: 1Æ3 108: 1Æ0 111: 0Æ1 112: 0Æ1

9 3: 1Æ0 15: 1Æ0 22: 1Æ2 25: 0Æ1 28: 1Æ2 34: 0Æ1 39: 0Æ1 72: 1Æ0 88: 0Æ1 109: 0Æ1

10 33: 0Æ1 66: 1Æ0 67: 1Æ0 79: 2Æ1 82: 2Æ0

11 16: 0Æ2 59: 2Æ1 85: 1Æ0

12 55: 1Æ0 86: 0Æ1 95: 1Æ2

13 18: 0Æ2 30: 0Æ2 31: 1Æ2 77: 1Æ0 82: 2Æ1 87: 1Æ0 89: 1Æ0 90: 1Æ0

14 33: 0Æ1 66: 1Æ0 70: 1Æ2 93: 2Æ1 111: 0Æ1

69

15 2: 0Æ1 5: 0Æ1 32: 1Æ0 46: 2Æ0 51: 1Æ0 60: 2Æ1 62: 2Æ1 63: 2Æ0 69: 1Æ0 71: 2Æ0 72: 0Æ1 84: 2Æ1 97: 1Æ0

16 25: 1Æ0 27: 0Æ2 68: 3Æ1 88: 1Æ0 90: 0Æ1

17 41: 0Æ1 78: 1Æ0 83: 2Æ3 94: 1Æ0 104: 1Æ0

18 7: 0Æ1 13: 1Æ2 25: 0Æ2 55: 1Æ0 73: 0Æ1 98: 2Æ0 107: 1Æ0

19 1: 2Æ0 3: 0Æ1 4: 1Æ0 34: 1Æ0 66: 1Æ0 67: 1Æ0 71: 0Æ1 77: 0Æ1 97: 0Æ1 106: 1Æ2 108: 1Æ2

70

20: 2: 1Æ2 6: 1Æ0 8: 1Æ0 15: 0Æ1 17: 1Æ2 20: 1Æ2 21: 1Æ2 29: 0Æ2 46: 0Æ2 60: 1Æ0 102: 0Æ1

21: 5: 1Æ0 54: 0Æ1 59: 1Æ0 70: 1Æ0 79: 2Æ1 92: 1Æ2 95: 1Æ3 104: 1Æ0 105: 1Æ0

22 3: 0Æ2 4: 1Æ2 7: 1Æ2 8: 0Æ2 9: 1Æ2 10: 1Æ2 11: 1Æ0 12: 1Æ2 14: 0Æ2 15: 1Æ2 90: 1Æ0 110: 0Æ1 111: 0Æ1

23 61: 1Æ0 62: 1Æ0 68: 1Æ3 73: 1Æ0

24 65: 0Æ1 70: 1Æ2

25 14: 0Æ1 71

65: 0Æ1 75: 0Æ1 79: 2Æ1 108: 1Æ0

26 54: 0Æ1 60: 1Æ0 62: 1Æ0 68: 1Æ0

27 63: 0Æ1 70: 1Æ0

28 59: 1Æ0 68: 0Æ3 69: 0Æ1

29 6: 1Æ0 24: 1Æ0 66: 1Æ0 80: 2Æ0 95: 1Æ2 103: 1Æ0 106: 1Æ0 109: 1Æ0

30 41: 0Æ1 104: 1Æ0

31 1: 2Æ1 7: 0Æ1 34: 1Æ0 59: 1Æ0 71: 0Æ1 92: 1Æ0 95: 2Æ3

32 32: 0Æ1 53: 2Æ1 54: 0Æ1 65: 1Æ0 98: 2Æ0

33 55: 1Æ0 61: 1Æ0 63: 0Æ1 72

66: 0Æ1 88: 0Æ1 97: 0Æ1 103: 0Æ1