Template: Front Matter

Home , Baiomys, Gymnodactylus darwinii, Habromys, Neotominae, Reithrodontomys, Tylomyinae

PHYLOGENETIC RELATIONSHIPS, DIVERGENCE AND RADIATION WITHIN THE SUBFAMILY NEOTOMINAE (RODENTIA: CRICETIDAE)

Megan S. Keith, B.S.

A Dissertation

Biology

Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY

Approved

Robert D. Bradley Chair of Committee

Robert J. Baker

Llewellyn D. Densmore III

David A. Ray

Darin S. Carroll

Mark Sheridan Dean of the Graduate School

December, 2015

ACKNOWLEDGMENTS There are several people that I would like to thank for their support throughout the course of my degree. First, I would like to thank my major advisor, Dr. Robert D.

Bradley, for mentoring me beginning as an undergraduate researcher and through my pursuit of a doctorate degree. I would also like to thank the remaining members of my advisory committee for their knowledge and support: Drs. Robert J. Baker, Llewellyn

D. Densmore, David A. Ray, and Darin S. Carroll.

There are numerous friends and colleagues that I would like to thank for their advice and support both personally and professionally. Thank you to Dr. R. Neal Platt

II, Dr. John D. Hanson, and Dr. Cody W. Thompson for being mentors to me in both my undergraduate and graduate research and for your continued collaboration on several projects. Several peers gave advice, support and friendship throughout the process of working toward my PhD: Dr. Matthew R. Mauldin, Dr. Faisal Ali

Anawarli Khan, Juan P. Carrera-Estupiñán, Dr. Tyla Holsombeck, Dr. Adam W.

Ferguson, Kenneth Griffith, Narayan Kandel, Dr. Molly M. McDonough, Nicté

Ordóñez-Garza, Dr. Julie A. Parlos, Dr. Caleb D. Phillips, Emma Roberts, Dr. Lizette

Siles-Mendoza, Dr. Cibele Sotero-Caio, Dr. Courtney A. Thomason, Christopher

Dunn, Dr. Miryam Venegas-Anaya, Elise Wagley, and Sheri Westerman-Ayers. In addition, I would like to thank several people that I met through field trips or through attendance at scientific meetings that also offered their support and encouragement:

Dr. Ralph Eckerlin, Dr. Walter Bulmer, Dr. John Matson and his wife Sharon, and Dr.

Texas Tech University, Megan S. Keith, December 2015

Michelle Haynie. Without all of you, my graduate school experience would not have been as exciting and rewarding.

Thank you to Heath Garner, Kathy MacDonald, and the Natural Science

Research Laboratory (Museum of Texas Tech University) staff for assistance with tissue loans. I would also like to thank the following professional and academic societies from which research funding was acquired: the Association of Biologist at

Texas Tech University and the Texas Academy of Science. Additionally, thank you to the Department of Biological Sciences and Graduate School of Texas Tech University for travel funding to attend and present my research at numerous state and national scientific meetings.

Finally I would like to thank my family for their unwavering support and encouragement throughout my academic career: my parents, Mike and Patricia Corley, my grandmother, Betty Ebarb, my brother Craig Corley as well as my extended family, and my in-laws Lonnie and Yolanda Keith, Tiffany and Noahh Dorris, and

Patrick and Mellissa Killingsworth. Most importantly, I want to thank my husband,

Eric, for his support and encouragement, and for being there through all stages in the pursuit of my graduate degree. Without the love and support of these very special people, I would not have endured the trials and pressures of research and graduate school.

iii

Texas Tech University, Megan S. Keith, December 2015

TABLE OF CONTENTS ACKNOWLEDGMENTS ...... ii

LIST OF TABLES ...... vi

LIST OF FIGURES ...... ix

ABSTRACT ...... xii

I. INTRODUCTION ...... 1

BACKGROUND INFORMATION ...... 1 OBJECTIVES ...... 6 ORGANIZATION OF CHAPTERS ...... 7 LITERATURE CITED ...... 8 II. WHAT IS PEROMYSCUS? EVIDENCE FROM NUCLEAR AND MITOCHONDRIAL DNA SEQUENCES SUGGESTS THE NEED FOR A NEW CLASSIFICATION ...... 16

ABSTRACT ...... 16 INTRODUCTION...... 17 MATERIALS AND METHODS ...... 19 RESULTS ...... 24 DISCUSSION ...... 27 ACKNOWLEDGMENTS ...... 36 LITERATURE CITED ...... 37 III. MOLECULAR DATA INDICATE THAT ISTHMOMYS IS NOT ALIGNED WITH PEROMYSCUS ...... 59

ABSTRACT ...... 59 INTRODUCTION...... 60 MATERIALS AND METHODS ...... 63 RESULTS ...... 67 DISCUSSION ...... 69 ACKNOWLEDGMENTS ...... 72 LITERATURE CITED ...... 73

Texas Tech University, Megan S. Keith, December 2015

IV. PHYLOGENETIC RELATIONSHIPS AND DIVERGENCE OF THE SUBFAMILY TYLOMYINAE (RODENTIA: CRICETIDAE) AS DETERMINED BY A MULTILOCUS DATASET ...... 94

ABSTRACT ...... 94 INTRODUCTION...... 95 MATERIALS AND METHODS ...... 98 RESULTS ...... 102 DISCUSSION ...... 103 ACKNOWLEDGMENTS ...... 106 LITERATURE CITED ...... 107 V. UTILIZATION OF MULTIPLE MOLECULAR MARKERS TO RESOLVE TRIBAL AFFILIATIONS WITHIN THE SUBFAMILY NEOTOMINAE 128

ABSTRACT ...... 128 INTRODUCTION...... 129 MATERIALS AND METHODS ...... 132 RESULTS ...... 135 DISCUSSION ...... 136 ACKNOWLEDGMENTS ...... 138 LITERATURE CITED ...... 139 VI. CONCLUSIONS ...... 157

PROJECT SUMMARY ...... 157 FUTURE DIRECTIONS ...... 162 LITERATURE CITED ...... 164

Texas Tech University, Megan S. Keith, December 2015

LIST OF TABLES 2.1 Specimens examined in this study are listed by taxon and genetic marker (Adh1- I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb-I7 - intron of the beta-fibrinogen, and Rbp3 - interphotoreceptor retinoid binding protein) and grouped by tribe, genus, and species group. GenBank accession (left of slash) and museum catalog (right of slash) numbers are given for each specimen. Museum acronyms are as follows: ASNHC (Angelo State Natural History Collection), BYU (Brigham Young University), CNMA (Colección Nacional de Mamíferos, Universidad Nacional Autónoma de México), PGSC (Peromyscus Genetic Stock Center), ROM (Royal Ontario Museum), TCWC (Texas Cooperative Wildlife Collection), and TTU (Museum of Texas Tech University). If museum catalog numbers were unavailable, specimens were referenced with the corresponding collector’s numbers or TK (special number of the Museum of Texas Tech University). ………………………………….47

2.2 Estimated genetic distances (K2P—Kimura 1980) for selected taxonomic groups based on sequences from the 4 genetic markers (Adh1-I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb-I7 -intron of the beta- fibrinogen, and Rbp3 - interphotoreceptor retinoid binding protein) examined in this study. ……………………………………………………………………...51

2.3 Three potential taxonomic solutions for Habromys, Isthmomys, Megadontomys, Neotomodon, Osgoodomys, Peromyscus and Podomys. Generic designations were identified by supported monophyletic clades within Fig. 1. Only species included in phylogenetic analyses are presented. ……………………………53

2.4 Detailed taxonomy of Peromyscus based on subsumation of Habromys, Megadontomys, Neotomodon, Osgoodomys, and Podomys to species groups; referred to as the species group taxonomy. Composition of species recognized in each species group was obtained from Musser and Carlton (2005) and other references (indicated in parentheses). …………………………………………55

3.1 Specimens examined in this study. GenBank accession numbers are provided for each gene examined. Abbreviations are as follows: Adh1-I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb-I7 - intron of the beta- fibrinogen, Dmp1- dentin matrix protein 1, Ghr- growth hormone receptor, and Rbp3 - interphotoreceptor retinoid binding protein. GenBank accession numbers (top) and museum catalogue numbers (bottom) are given for each specimen. Museum acronyms are as follows: ASNHC (Angelo State Natural History Collection), BYU (Brigham Young University), CNMA (Colección Nacional de Mamíferos, Universidad Nacional Autónoma de México), CRD (Centro Interdisciplinario de Investigación para el Desarrollo Integral Regional, Unidad Durango, Mexico), FSH (University of Texas Medical Branch at Galveston), PGSC (Peromyscus Genetic Stock Center), ROM (Royal Ontario vi

Texas Tech University, Megan S. Keith, December 2015

Museum), TCWC (Texas Cooperative Wildlife Collection), and TTU (Museum of Texas Tech University). If museum catalogue numbers were unavailable, specimens were referenced with the corresponding collector’s numbers or TK (special number of the Museum of Texas Tech University). An asterisk indicates the individuals/sequences that were used in the combined analysis. .81

3.2 Summary of results for all analyses performed in this study. The hypothesis that each dataset reflects is represented by "yes" and analyses for which relationships were statistically supported are represented by a "+". ………….87

4.1 Specimens examined by taxon, genetic marker (Adh1-I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb-I7 - intron of the beta-fibrinogen, Dmp1- dentin matrix protein 1, Ghr- growth hormone receptor, and Rbp3 - interphotoreceptor retinoid binding protein) and grouped by genus. GenBank accession (top) and museum catalogue (bottom) numbers are given for each specimen. Museum acronyms are as follows: ASNHC (Angelo State Natural History Collection), BYU (Brigham Young University), CNMA (Colección Nacional de Mamíferos, Universidad Nacional Autónoma de México), CRD (Centro Interdisciplinario de Investigación para el Desarrollo Integral Regional, Unidad Durango, Mexico), FSH (University of Texas Medical Branch at Galveston), MSB (The Museum of Southwestern Biology), PGSC (Peromyscus Genetic Stock Center), ROM (Royal Ontario Museum), TCWC (Texas Cooperative Wildlife Collection), and TTU (Museum of Texas Tech University). If museum catalogue numbers were unavailable, specimens were referenced with the corresponding collector’s numbers or TK (special number of the Museum of Texas Tech University). ………………………………….116

4.2 Analyses were conducted using the Kimura 2-parameter model (Kimura 1980) in MEGA 6.06 (Tamura et al. 2013). The number of base substitutions per site from averaging over all sequence pairs between groups are shown. Each analysis involved 48 nucleotide sequences with the exception of the Dmp1 dataset (40 nucleotide sequences) in which all sequences that were <50% of the entire gene length were removed so that common sites could be compared. Genetic distances are presented as percentages (%). ……………………….124

5.1 Specimens for which sequences were generated for this study, specifically for the individual nuclear analyses and the combined analysis are listed by taxon, genetic marker (Adh1-I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb-I7 - intron of the beta-fibrinogen, Dmp1- dentin matrix protein 1, Ghr- growth hormone receptor, and Rbp3 - interphotoreceptor retinoid binding protein) and grouped by genus and specific epithets are listed alphabetically. GenBank accession (top) and museum catalogue (bottom) numbers are given for each specimen. Museum acronyms are as follows: ASNHC (Angelo State Natural History Collection), BYU (Brigham Young

vii

Texas Tech University, Megan S. Keith, December 2015

University), CNMA (Colección Nacional de Mamíferos, Universidad Nacional Autónoma de México), CRD (Centro Interdisciplinario de Investigación para el Desarrollo Integral Regional, Unidad Durango, Mexico), FSH (University of Texas Medical Branch at Galveston), MSB (The Museum of Southwestern Biology), PGSC (Peromyscus Genetic Stock Center), ROM (Royal Ontario Museum), TCWC (Texas Cooperative Wildlife Collection), and TTU (Museum of Texas Tech University). If museum catalogue numbers were unavailable, specimens were referenced with the corresponding collector’s numbers or TK (special number of the Museum of Texas Tech University). ………………146

viii

Texas Tech University, Megan S. Keith, December 2015

LIST OF FIGURES 2.1 Phylogenetic tree obtained from maximum likelihood analysis of the a) combined mitochondrial cytochrome-b gene (Cytb) and 3 nuclear genes (alcohol dehydrogenase - Adh1-I2, beta fibrinogen - Fgb-I7, and interphotoreceptor retinoid-binding protein - Rbp3), as well as the b) mitochondrial (Cytb) phylogenetic tree, and c) combined nuclear (Adh1-I2, Fgb-I7, and Rbp3) phylogenetic tree. Taxonomic groups of interest are designated as follows: Pss (Peromyscus [sensu stricto]), Psl (Peromyscus [sensu lato]), Rei (Reithrodontomyini), and Bai (Baiomyini). Nodal support values are superimposed on the maximum likelihood tree topology. Support values are as follows: 10,000 bootstrap replicates of the maximum likelihood analysis (below branch) and clade probability values for the Bayesian inference analysis (above branch). Statistically significant clade probability values (≥ 0.95) are designated with an asterisk (*). All bootstrap support values ≥ 50 are shown. For members of Peromyscus (sensu stricto) only, species epithets are given. Peromyscus (sensu lato) affiliated genera are indicated in bold. Major nodes are indicated with roman numerals...... 57

2.2 Maximum clade credibility tree showing divergence date estimates based on a combined analysis of the mitochondrial cytochrome-b gene (Cytb) and 3 nuclear genes (alcohol dehydrogenase - Adh1-I2, beta fibrinogen - Fgb-I7, and interphotoreceptor retinoid-binding protein - Rbp3). Divergence date estimates are indicated along the x-axis in millions of years. Error bars represent the 95% highest posterior density (HPD) for node height. Peromyscus (sensu lato) affiliated genera are indicated in bold ...... 58

3.1 Phylogenetic tree obtained from the maximum likelihood analysis of the large scale sampling for the mitochondrial cytochrome-b gene (Cytb). Support values were based on 1,000 bootstrap replicates of the maximum likelihood analysis. Values ≥80 are indicated by an asterisk. The number of individuals for each species included are provided to the right of the dash; Isthmomys is represented by a dashed line. Clade I reflects a sister relationship between Reithrodontomys and Isthmomys, with Peromyscus basal to these two genera in clade II. However, there was no statistical support for this relationship. …...88

3.2 Phylogenetic trees obtained from maximum likelihood analysis of the unconstrained (A) and constrained (B) topologies. Support values were based on 1,000 bootstrap replicates of the maximum likelihood analysis. Values ≥80 are represented by an asterisk. For the unconstrained analysis (A), clade I reflects a sister relationship between Reithrodontomys and Isthmomys, with Peromyscus basal to these two genera in clade II. For the constrained analysis (B), Isthmomys was constrained to be monophyletic with Peromyscus (Clade ix

Texas Tech University, Megan S. Keith, December 2015

IV) and Reithrodontomys forms a separate clade (Clade III). The score of the best tree for the unconstrained analysis (A) was -35559.90420 and the score of the best constrained tree (B) was -35576.11781. …………………………….90

3.3 Results of the individual nuclear maximum likelihood analyses. Support values were based on 1,000 bootstrap replicates of the maximum likelihood analysis. Values ≥80 are represented by an asterisk. Analyses for Adh1-I2 (A) and Fgb-I7 (B) resulted in phylogenies in which Isthmomys was basal to both Peromyscus and Reithrodontomys, phylogenies for Ghr (C) and Rbp3 (D) indicate a clade uniting Isthmomys and Reithrodontomys with Peromyscus basal to this clade, and analysis of Dmp1 (E) resulted in a phylogeny in which Isthmomys is more closely related to Peromyscus. …………………………..92

3.4 Phylogenetic tree obtained from maximum likelihood analysis of the combined dataset (Cytb, Adh1-I2, Dmp1, Fgb-I7, Ghr, and Rbp3). Support values are based on 1,000 bootstrap replicates of the maximum likelihood analysis. Values ≥80 are shown above the supported node. Clade I reflects a strongly supported sister relationship between Reithrodontomys and Isthmomys, with a monophyletic Peromyscus basal to these two genera in clade II. ……………93

4.1 Phylogenetic tree obtained from the Bayesian inference analysis of the combined dataset. Posterior probability values ≥95% are represented by an "*" above the supported node and clade assignments are listed below each node. This analysis resulted in a clade uniting the Neotominae and Tylomyinae (Clade I) and two supported clades representing rodents of the Subfamily Neotominae (Clade A) and the Subfamily Tylomyinae (Clade B). …...……125

4.2 Maximum clade credibility tree showing divergence date estimates based on a combined analysis of the mitochondrial Cytb gene and 5 nuclear genes (Adh1- I2, Fgb-I7, Dmp1, Ghr, and Rbp3). Divergence date estimates are indicated along the x-axis in millions of years. Error bars represent the 95% highest posterior density (HPD) for node height. Labels on the right indicate taxa belonging to each subfamily. ……………………………………….………126

5.1 Phylogenetic tree obtained from Bayesian inference analysis of the combined dataset. Posterior probability values ≥95% are represented by an "*" above the supported node and clade assignments are listed below each node or as groups labeled on the right. This analysis resulted in two distinct clades corresponding to the Neotominae (I) and Tylomyinae (II). Tribes and taxa that are affiliated with them are indicated to the right of the phylogeny. ………154

5.2 Maximum clade credibility tree showing divergence date estimates based on a combined analysis of the mitochondrial Cytb gene and 5 nuclear genes (Adh1- I2, Fgb-I7, Dmp1, Ghr, and Rbp3). Divergence date estimates are indicated

Texas Tech University, Megan S. Keith, December 2015 along the x-axis in millions of years. Error bars represent the 95% highest posterior density (HPD) for node height. Labels on the right indicate taxa belonging to each tribe within the Neotominae. ……………………………155

Texas Tech University, Megan S. Keith, December 2015

ABSTRACT The primary objective of this dissertation was to assess phylogenetic relationships at multiple hierarchical levels for rodents historically classified within the subfamily Neotominae. In Chapter I, the history of the taxonomic classification of rodents in the Neotominae and incongruence among phylogenetic studies was introduced. Chapter II utilized a multilocus dataset (one mitochondrial and three nuclear markers) to assess species group relationships for Peromyscus so that higher level taxonomic questions could be addressed, including the phylogenetic relationship of the genus Isthmomys to the remainder of the Neotominae as addressed in Chapter

III. Results indicated that Peromyscus (sensu lato) requires revision, as one of its members (Isthmomys) is positioned outside of a monophyletic Peromyscus. In

Chapter IV, the taxonomic status of the tylomyine rodents which previously were classified in the Neotominae was examined utilizing one mitochondrial and five nuclear markers. The tylomyine rodents form their own subfamily separate from the

Neotominae and are more closely related to the neotomine rodents than they are to the

Sigmodontinae. Finally, Chapter V examined the number of tribes and tribal affiliations that should be recognized within the Neotominae and employed coalescent theory to date tribal origins. The Neotominae consists of five tribes: the classification of the tribes Baiomyini, Neotomini, and Ochrotomyini are maintained as recognized in previous studies, and a separate Peromyscini (Peromyscus + Habromys,

Megadontomys, Neotomodon, Onychomys, Osgoodomys, and Podomys) and

xii

Texas Tech University, Megan S. Keith, December 2015

Reithrodontomyini (Isthmomys + Reithrodontomys) should be recognized.

Concluding remarks are provided in Chapter VI.

xiii

Texas Tech University, Megan S. Keith, December 2015

CHAPTER I

INTRODUCTION

BACKGROUND INFORMATION The subfamily Neotominae is a diverse and widely distributed group of rodents occupying most of North and Central America and consisting of approximately 16 genera and over 120 species (Musser and Carleton 2005). This subfamily includes woodrats (Hodomys, Nelsonia, Neotoma, and Xenomys), deer mice and relatives

(Habromys, Isthmomys, Megadontomys, Neotomodon, Onychomys, Osgoodomys,

Peromyscus, and Podomys), harvest mice (Reithrodontomys), pygmy mice (Baiomys), singing mice (Scotinomys) and the golden mouse (Ochrotomys). Historically, the tylomyine rodents (Nyctomys, Otonyctomys, Ototylomys, and Tylomys) have been classified as a tribe within both the Neotominae (Carleton 1980; Reeder et al. 2006) and the subfamily Sigmodontinae (McKenna and Bell 1997), but currently are recognized as comprising their own subfamily (Tylomyinae- Reig 1984; Musser and

Carleton 2005). Several studies (Baker and Mascarello 1969; Greenbaum et al. 1978a,

1978b; Greenbaum and Baker 1978; Carleton 1980; Bradley et al. 2004a; Reeder et al.

2006; Bradley et al. 2007; Miller and Engstrom 2008 as well as numerous others) have examined phylogenetic relationships among members of the Neotominae utilizing morphological, allozymic, or karyotypic data, as well as mitochondrial and nuclear

DNA sequences. Various taxonomic arrangements have been proposed, including several interpretations as to the number of tribes that should be recognized within this subfamily as well as relationships of genera within each tribe (Hooper and Musser

Texas Tech University, Megan S. Keith, December 2015

1964a, 1964b; Carleton 1980; Musser and Carleton 2005; Reeder et al. 2006; Miller and Engstrom 2008).

Relationships within the Neotominae have been difficult to resolve, most likely due to a rapid radiation event from their common ancestor during the Miocene

(Bradley et al. 2004a; Reeder and Bradley 2004; Reeder et al. 2006; Fabre et al. 2012).

Additionally, the accumulation of few phylogenetically informative characters, paucity of taxa in studies, lack of congruence among data analyses, etc. have made it difficult to determine phylogenetic relationships at the onset of the origin of the

Neotominae. Carleton (1980) included the most detailed morphological dataset for rodents (79 characters for 75 taxa); however, repetitive bursts of speciation and a high level of homoplasy in morphological characters has increased difficulty in delimiting intra-familial relationships (Fabre et al. 2012) and analyses of genetic data from numerous studies produced differing results. Despite the wealth of knowledge available for this group of rodents, branching patterns and relationships within each tribal lineage remain unresolved.

Most studies focusing on this group of rodents provided support for four and in some cases five tribes; however no two studies have provided the same results on the classification of these rodents at the tribal level. Reeder et al. (2006) used molecular evidence to show support for five tribes: Baiomyini, Neotomini, Ochrotomyini,

Peromyscini, and Tylomyini. The most recent study suggested the recognition of five tribes (Baiomyini, Neotomini, Ochrotomyini, Peromyscini and Reithrodontomyini) based on DNA sequence data for one mitochondrial and two nuclear genes (Miller and

Texas Tech University, Megan S. Keith, December 2015

Engstrom 2008); however this study did not include representatives of several critical genera and species groups. Most studies suggest the placement of Ochrotomys into a separate tribe (Ochrotomyini- Musser and Carleton 2005; Reeder et al. 2006; Miller and Engstrom 2008) and others have provided opposing views on the tribal classification of Peromyscus and allies and Reithrodontomys. Further, some authors recognize the aforementioned genera as constituting the Peromyscini (Hooper and

Musser 1964a, 1964b; Reeder et al. 2006) whereas others have used

Reithrodontomyini to represent the “old” Peromyscini (Reithrodontomys +

Peromyscus and allies- Musser and Carleton 2005). Others have argued that the tribe should be split in two (Peromyscini- Peromyscus + allies and Reithrodontomyini-

Reithrodontomys + Isthmomys [Miller and Engstrom 2008]). Studies that have focused on tribal affiliations within this group of rodents may have been biased depending on the taxa that were included. Therefore, to begin to define tribal boundaries, it is necessary to first resolve relationships at the generic level and below to determine which taxa should be used to best represent the true relationships within this group.

There are several sets of taxa for which phylogenetic relationships remain unresolved within each tribe. Systematics of the genus Peromyscus have been studied extensively, and yet relationships within this group have not been fully resolved. This is due mainly to the description of new taxa (see Musser and Carleton 2005 for a review; Romo-Vázqez et al. 2005) and discovery of cryptic species (Houseal et al.

1987; Schmidly et al. 1988; Riddle et al. 2000; Bradley et al. 2004b; Bradley et al.

Texas Tech University, Megan S. Keith, December 2015

2014; Bradley et al. 2015) over the last 40 years. This rapid increase in the number of species identified within Peromyscus as well as several taxonomic rearrangements have occurred in response to the availability of more advanced methodologies

(morphology, karyology, allozymes, and genetic data). However, various datasets result in differing phylogenies, making it difficult to definitively define Peromyscus.

Complicating this issue is the conflicting evidence for the phylogenetic position of

Isthmomys which historically has been classified in the Peromyscus complex, but recent molecular data shows evidence for a sister relationship to Reithrodontomys

(Rogers et al. 2005; Miller and Engstrom 2008; Platt et al. 2015).

To begin to resolve relationships within this group, sequence data from previous studies (Reeder et al. 2006; Miller and Engstrom 2008) which have increased resolution at varying phylogenetic levels will be examined. In addition to these molecular markers, one new nuclear marker will be utilized in combined analyses to attempt to achieve higher resolution within these groups. It is well known that the evolutionary histories of individual genes may differ significantly from the underlying species tree, which can result from horizontal transfer, gene duplication, incomplete lineage sorting, and numerous other processes (Maddison 1997; Kubatko et al. 2009).

Additionally, coalescent theory suggests that incongruencies between individual gene trees and species trees will often occur when divergence times are very short relative to the effective population size of the ancestral population (Degnan and Rosenberg

2006; Kubatko and Degnan 2007; Belfiore et al. 2008). Gene trees can be utilized in species and higher hierarchical level delimitation efforts in two ways: qualitative

Texas Tech University, Megan S. Keith, December 2015 assessments of lineage boundaries can be derived from the phylogenetic estimate utilizing single-locus data (Pellegrino et al. 2005) which can be misled by coalescent stochasticity (Hudson and Coyne 2002) or individual gene trees can be combined into a single analysis. Under this scenario, the gene tree may not reflect the actual pattern of lineage divergence, but data from multiple loci can reduce the effects of lineage sorting, genetic drift, rapid divergence and several other processes (Carstens and

Dewey 2010). In order to resolve any issues that may arise due to rapid diversification over a short time period, this dissertation aims to increase the number of independent loci sampled with the expectation that the multilocus signal will: 1) increase resolution at different hierarchical levels (Bull et al. 1993; Adkins et al. 2001; Pereira et al.

2002), 2) overcome noise attributable to homoplasy (Barrett et al. 1991; de Queiroz

1993; Adkins et al. 2001), 3) overcome stochastic lineage sorting (Rokas et al. 2003) and 4) reveal hidden relationships (Gatesy et al. 2004).

There have been several publications which have argued for (Baum 1992; de

Queiroz 1993) and against (Barrett et al. 1991; Miyamoto and Fitch 1995) combining genetic datasets into one analysis, combining multiple data types (morphology, DNA sequences, karyotypes, etc.), and combining gene trees. One of the major arguments against combining datasets for phylogenetic inference is that each dataset likely reflects different phylogenetic histories (Doyle 1992; Bull et al. 1993; de Quieroz

1993). However, Wiens (1998) suggested that partitioning the available data to maximize detection of different histories (i.e. analyze each data set under its own appropriate nucleotide substitution model), performing separate analyses of the data

Texas Tech University, Megan S. Keith, December 2015 sets, and then combining the data but considering questionable or unresolved those parts of the combined tree that are strongly contested in the separate analyses until a majority of unlinked data sets support one resolution over another serves as a simple methodology for dealing with this issue. This approach allows for combining data and for accommodating the differences in phylogenetic history (de Quieroz et al. 1995;

Wiens 1998).

The individual studies within this dissertation aim to utilize combined datasets to test several phylogenetic relationships at varying taxonomic levels within the

Neotominae. Species level relationships will first be analyzed for the Peromyscus complex and several analyses will be utilized to test the phylogenetic placement of the genus Isthmomys in relation to Peromyscus and Reithrodontomys. Finally, genetic data will be employed to test the taxonomic status of the subfamily Tylomyinae and subsequently, tribal affiliations will be determined for the subfamily Neotominae.

OBJECTIVES The focus of this dissertation is to better define tribal boundaries within the subfamily Neotominae and to increase resolution of phylogenetic relationships for genera that have consistently remained unresolved. Specifically, the objectives of each study were to use a multilocus dataset or sequence data for previously unavailable taxa to:

1) Determine the relationships at the species group level for genera whose

species level relationships are controversial (i.e. Peromyscus and

Isthmomys). This also allows for the identification of crucial taxa for

Texas Tech University, Megan S. Keith, December 2015

inclusion and for the identification of genetic markers useful for resolving

relationships at these levels. (Chapters II & III)

2) Utilize a multilocus dataset to resolve the phylogenetic position of the

tylomyine rodents to determine if these rodents should be recognized in

their own subfamily. (Chapter IV)

3) Develop a robust, multi-gene phylogeny to acquire higher resolution and

determine the boundaries of tribal affiliations within the Neotominae.

(Chapter V)

4) Use coalescence theory to date species group, genus, and tribal origins to

test the hypothesis of rapid divergence over a short time frame. (Chapters

II, IV-V)

ORGANIZATION OF CHAPTERS Chapter I is an introduction to the essential background information of the research focus for this dissertation. Chapters II through V are separate manuscripts that were developed to address the objectives of this dissertation. These chapters have been or will be submitted to peer-review journals. Chapter II is titled "What is

Peromyscus? Evidence from nuclear and mitochondrial DNA sequences suggests the need for a new classification." This chapter addresses Objectives 1 and 4, is coauthored by Roy N. Platt II (first author), Brian R. Amman, Cody W. Thompson, and Robert D. Bradley, and has been published in the Journal of Mammalogy.

Chapter III also addresses Objective 1, and is titled "Molecular data indicate that

Isthmomys is not aligned with Peromyscus". This chapter is coauthored by Robert D.

Texas Tech University, Megan S. Keith, December 2015

Bradley, Jorge Salazar-Bravo and Roy N. Platt II. Chapter III will be submitted to the

Journal of Mammalogy and is cited as "Keith et al. (in prep A)" throughout this dissertation. Chapter IV addresses Objectives 2 and 4, is titled "Phylogenetic relationships and divergence of the subfamily Tylomyinae (Rodentia: Cricetidae) as determined by a multilocus dataset", is coauthored by Cody W. Thompson and Robert

D. Bradley, and will be submitted to the Journal of Mammalogy. This chapter is cited as "Keith et al. (in prep B)" throughout this dissertation. Chapter V addresses

Objectives 3 and 4, is titled "Utilization of a molecular supermatrix to resolve tribal affiliations within the subfamily Neotominae", and is coauthored by Roy N. Platt II,

Cody W. Thompson, Brian R. Amman, and Robert D. Bradley. This chapter will be submitted to Systematic Biology and will be cited as "Keith et al. (in prep C)" throughout the body of this dissertation. The final chapter, Chapter VI, summarizes the results from Chapters II through V.

LITERATURE CITED ADKINS, R. M., E. L. GELKE, D. ROWE, AND R. L. HONEYCUTT. 2001. Molecular

phylogeny and divergence time estimates for major rodent groups: evidence

from multiple genes. Molecular Biology and Evolution 18: 777-791.

BAKER, R. J., AND J. T. MASCARELLO. 1969. Karyotypic analyses of the genus

Neotoma (Cricetidae, Rodentia). Cytogenetics 8:187-198.

BARRETT, M., M. J. DONOGHUE, AND E. SOBER. 1991. Against consensus. Systematic

Zoology 40: 486-493.

Texas Tech University, Megan S. Keith, December 2015

BAUM, B. R. 1992. Combining trees as a way of combining data sets for phylogenetic

inference, and the desirability of combining gene trees. Taxon 41: 3-10.

BELFIORE N. M., L. LIU AND C. MORITZ. 2008. Multilocus phylogenetics of a rapid

radiation in the genus Thomomys (Rodentia: Geomyidae). Systematic Biology

57:294–310.

BRADLEY, R. D., C. W. EDWARDS, D. S. CARROLL, AND C. W. KILPATRICK. 2004a.

Phylogenetic relationships of neotomine-peromyscine rodents: based on DNA

sequences from the mitochondrial cytochrome-b gene. Journal of Mammalogy

85:389-395.

BRADLEY, R. D., D. J. SCHMIDLY, B. R. AMMAN, R. N. PLATT II, K. M. NEUMANN, H.

M. HUYNH, R. MUÑEZ-MARTÍNEZ, C. LÓPEZ-GONZÁLEZ, AND N. ORDÓÑEZ-

GARZA. 2015. Molecular and morphological data reveal multiple species in

Peromyscus pectoralis. Journal of Mammalogy 96: 446-459.

BRADLEY, R. D., D. S. CARROLL, M. L. HAYNIE, R. M. MARTINEZ, M. J. HAMILTON,

AND C. W. KILPATRICK. 2004b. A new species of Peromyscus from western

Mexico. Journal of Mammalogy 85: 184-1193.

BRADLEY, R. D., N. D. DURISH, D. S. ROGERS, J. R. MILLER, M. D. ENGSTROM, AND C.

W. KILPATRICK. 2007. Toward a molecular phylogeny for Peromyscus,

evidence from mitochondrial cytochrome-b sequences. Journal of

Mammalogy 88:1146-1159.

BRADLEY, R. D., N. ORDÓÑEZ-GARZA, C. G. SOTERO-CAIO, H. M. HUYNH, C. W.

KILPATRICK, L. I. IÑIGUEZ-DÁVALOS, AND D. J. SCHMIDLY. 2014.

Texas Tech University, Megan S. Keith, December 2015

Morphometric, karyotypic and molecular evidence for a new species of

Peromyscus (Cricetidae: Neotominae) from Nayarit, Mexico. Journal of

Mammalogy 95:176-186.

BULL, J. J., J. P. HUELSENBECK, C. W. CUNNINGHAM, D. L. SWOFFORD, AND P. J.

WADDELL. 1993. Partitioning and combining data in phylogenetic analysis.

Systematic Biology 42: 384-397.

CARLETON, M. D. 1980. Phylogenetic relationships in neotomine-peromyscine

rodents (Muroidea) and a reappraisal of the dichotomy within New World

Cricetinae. Miscellaneous Publications, Museum of Zoology, University of

Michigan 157:1-146.

CARSTENS, B. C. AND T. A. DEWEY. 2010. Species delimitation using a combined

coalescent and information-theoretic approach: an example from North

American Myotis bats. Systematic Biology doi: 10.1093/sysbio/syq024.

DEGNAN, J. H. AND N. A. ROSENBERG. 2006. Discordance of species trees with their

most likely gene tress. PLos Genetics 2: e68. doi:

10.1371/journal.pgen.0020068.

DE QUEIROZ, A. 1993. For consensus (sometimes). Systematic Biology 42: 368-372.

DE QUIEROZ, A., M. J. DONOGHUE, AND J. KIM. 1995. Separate versus combined

analysis of phylogenetic evidence. Annual Review of Ecology and

Systematics 26: 657-681.

DOYLE, J. J. 1992. Gene trees and species trees: molecular systematics as one-

character taxonomy. Systematic Biology 17: 144-163.

Texas Tech University, Megan S. Keith, December 2015

FABRE, P.-H., L. HAUTIER, D. DIMITROV, AND E. J. P. DOUZERY. 2012. A glimpse on

the pattern of rodent diversification: a phylogenetic approach. BMC

Evolutionary Biology 12:88.

GATESY J., R. H. BAKER AND C. HAYASHI. 2004. Inconsistencies in arguments for the

supertree approach: supermatrices versus supertrees of Crocodylia. Systematic

Biology 53:342–355.

GREENBAUM, I. F. AND R. J. BAKER. 1978. Determination of the primitive karyotype

for Peromyscus. Journal of Mammalogy 59: 820-834.

GREENBAUM, I. F., R. J. BAKER, AND J. H. BOWERS. 1978a. Chromosomal homology

and divergence between sibling species of deer mice: Peromyscus maniculatus

and P. melanotis (Rodentia, Cricetidae). Evolution 32:334-341.

GREENBAUM, I. F., R. J. BAKER, AND P. R. RAMSEY. 1978b. Chromosomal evolution

and its implication concerning the mode of speciation in three species of deer

mice of the genus Peromyscus. Evolution 32:646-654.

HOOPER, E. T. AND G. G. MUSSER. 1964a. The glans penis in Neotropical cricetines

(family Muridae), with comments on classification of muroid rodents.

Miscellaneous Publications, Museum of Zoology, University of Michigan

123:1-57.

HOOPER, E. T. AND G. G. MUSSER. 1964b. Notes on classification of the rodent genus

Peromyscus. Occasional Publications of the Museum of Zoology, University

of Michigan 635:1-13.

Texas Tech University, Megan S. Keith, December 2015

HOUSEAL, T. W., I. F. GREENBAUM, D. J. SCHMIDLY, S. A. SMITH, AND K. M. DAVIS.

1987. Karyotypic variation in Peromyscus boylii from Mexico. Journal of

Mammalogy 68: 281-296.

HUDSON, R. R., AND J. A. COYNE. 2002. Mathematical consequences of the

genealogical species concept. Evolution 56:1557-1565.

KUBATKO, L. S., B. C. CARSTENS, AND L. L. KNOWLES. 2009. STEM: species tree

estimation suing maximum likelihood for gene trees under coalescence.

Bioinformatics 25: 971-973.

KUBATKO, L. S. AND J. H. DEGNAN. 2007. Inconsistency of phylogenetic estimates

from concatenated data under coalescence. Systematic Biology 56: 17-24.

MADDISON, W. P. 1997. Gene trees in species trees. Systematic Biology 1997:523-

536.

MCKENNA, M. C., and S. K. BELL. 1997. Classification of mammals above the species

level. Columbia University Press, New York.

MILLER, J. R. AND M. D. ENGSTROM. 2008. The relationships of major lineages

within peromyscine rodents: a molecular phylogenetic hypothesis and

systematic reappraisal. Journal of Mammalogy 89:1279-1295.

MIYAMOTO, M. M. AND W. M. FITCH. 1995. Testing species phylogenies and

phylogenetic methods with congruence. Systematic Biology 44: 64-76.

MUSSER, G. G. AND M. D. CARLETON. 2005. Superfamily Muroidea. Pp. 894-1531 in

Mammal species of the world: a taxonomic and geographic reference (D. E.

Texas Tech University, Megan S. Keith, December 2015

Wilson and D. M. Reeder, Eds.). 3rd Ed. Johns Hopkins University Press,

Baltimore Maryland.

PELLEGRINO, K. C. M., M. T. RODRIGUES, A. N. WAITE, M. MORANDO, Y. Y.

YASSUDA, AND J. L. SITES. 2005. Phylogeography and species limits in the

Gymnodactylus darwinii complex (Gekkonidae: Squamata): genetic structure

coincides with river systems in the Brazilian Atlantic Forest. Biol. J. Linn.

Soc. 85: 13-26.

PEREIRA, S. L., A. J. BAKER, AND A. WAJNTAL. 2002. Combined nuclear and

mitochondrial DNA sequences resolve genetic relationships within the

Cracidae (Galliformes, Aves). Systematic Biology 51:946-958.

PLATT II, R. N., B. R. AMMAN, M. S. KEITH, C. W. THOMPSON, AND R. D. BRADLEY.

2015. What is Peromyscus? Evidence from nuclear and mitochondrial DNA

sequences for a new classification. Journal of Mammalogy 96: 708-719.

REEDER, S. A. AND R. D. BRADLEY. 2004. Molecular systematic of neotomine-

peromyscine rodents based on the dentin matrix protein 1 gene. Journal of

Mammalogy 85: 1194-1200.

REEDER, S. A., D. S. CARROLL, C. W. EDWARDS, C. W. KILPATRICK, AND R. D.

BRADLEY. 2006. Neotomine-peromyscine rodent systematics based on

combined analyses of nuclear and mitochondrial DNA sequences. Molecular

Phylogenetics and Evolution 40: 251-258.

Texas Tech University, Megan S. Keith, December 2015

REIG, O. A. 1984. Distribução geográfica e história evolutiva dos roedores muroideas

sulamericanos (Cricetidae: Sigmodontinae). Revista Brasileira de Genetica 7:

333-365.

RIDDLE, B. R., D. J. HAFNER, AND L. F. ALEXANDER. 2000. Phylogeography and

Systematics of the Peromyscus eremicus Species Group and the Historical

Biogeography of North American Warm Regional Deserts. Molecular

Phylogenetics and Evolution 17: 145-160.

ROGERS, D. S., M. D. ENGSTROM, AND E. ARELLANO. 2005. Phylogenetic relationships

among Peromyscine rodents: Allozyme evidence. Pp. 427-440 in

Contribuciones Mastozoológicas en Homenaje a Bernardo Villa (V. Sánchez-

Cordero and R. A. Medellín, Eds.). Instituto de Biología e Instituto de

Ecología, UNAM, México.

ROKAS, A., B. L. WILLIAMS, N. KING, AND S. B. CARROLL. 2003. Genome-scale

approaches to resolving incongruence in molecular phylogenies. Nature 425:

798-804.

ROMO-VÁZQUEZ, E., L. LEÓN-PANIAGUA, AND O. SÁNCHEZ. 2005. A new species of

Habromys (Rodentia: Neotominae) from México. Proceedings of the

Biological Society of Washington 118: 605-618.

SCHMIDLY, D. J., R. D. BRADLEY, AND P. S. CATO. 1988. Morphometric

differentiation and taxonomy of three chromosomally characterized groups of

Peromyscus boylii from east-central Mexico. Journal of Mammalogy 69: 462-

480.

Texas Tech University, Megan S. Keith, December 2015

WIENS, J. J. 1998. Combining data sets with differing phylogenetic histories.

Systematic Biology 47: 568-581.

Texas Tech University, Megan S. Keith, December 2015

CHAPTER II

WHAT IS PEROMYSCUS? EVIDENCE FROM NUCLEAR AND MITOCHONDRIAL DNA SEQUENCES SUGGESTS THE NEED FOR A NEW CLASSIFICATION

ABSTRACT The evolutionary relationships among Peromyscus, Habromys, Isthmomys,

Megadontomys, Neotomodon, Osgoodomys, and Podomys are poorly understood. In order to further explore the evolutionary boundaries of Peromyscus and compare potential taxonomic solutions for this diverse group and its relatives, we conducted phylogenetic analyses of DNA sequence data from alcohol dehydrogenase (Adh1-I2), beta fibrinogen (Fgb-I7), interphotoreceptor retinoid-binding protein (Rbp3), and cytochrome-b (Cytb). Phylogenetic analyses of mitochondrial and nuclear genes produced similar topologies, although levels of nodal support varied. The best- supported topology was obtained by combining nuclear and mitochondrial sequences.

No monophyletic Peromyscus clade was supported. Instead, support was found for a clade containing Habromys, Megadontomys, Neotomodon, Osgoodomys, Podomys, and Peromyscus suggesting paraphyly of Peromyscus, and confirming previous observations. Our analyses indicated an early divergence of Isthmomys from

Peromyscus (approximately 8 million years ago) whereas most other peromyscine taxa emerged within the last 6 million years. To recover a monophyletic taxonomy from

Peromyscus and affiliated lineages, we detail 3 taxonomic options in which

Habromys, Megadontomys, Neotomodon, Osgoodomys, and Podomys are retained as genera, subsumed as subgenera, or subsumed as species groups within Peromyscus.

Texas Tech University, Megan S. Keith, December 2015

Each option presents distinct taxonomic challenges, and the appropriate taxonomy must reflect the substantial levels of morphological divergence that characterize this group while maintaining the monophyletic relationships obtained from genetic data.

INTRODUCTION What is Peromyscus? More than 100 years since Osgood’s (1909) monograph the question remains unresolved. A historical perspective and overview of the taxonomic challenges affiliated with Peromyscus are provided in Bradley et al. (2007),

Carleton (1980, 1989), and Miller and Engstrom (2008). At conflict is the taxonomic status of Habromys, Isthmomys, Megadontomys, Neotomodon, Osgoodomys, and

Podomys. At various times, these taxa have been recognized at either the generic

(sensu stricto) or subgeneric (sensu lato) level, though most major classifications generally fall into 1 of 2 categories. No single historical classification fits perfectly into the sensu stricto or sensu lato categories, though most major classifications tend to reflect one interpretation over the other. Carleton (1980, 1989) and Musser and

Carleton (2005) are most closely aligned with a Peromyscus (sensu lato) taxonomy, whereas Hooper (1968) is a variation of a Peromyscus (sensu stricto) classification.

Most current classifications recognize Peromyscus (sensu stricto).

Bradley et al. (2007) completed the most comprehensive molecular study of

Peromyscus and its allies in which DNA sequences from the entire mitochondrial cytochrome-b gene (Cytb) were examined. They found 5 genera (Habromys,

Megadontomys, Neotomodon, Osgoodomys, and Podomys) to be embedded within a monophyletic clade containing Peromyscus (sensu stricto); Isthmomys was sister to

Texas Tech University, Megan S. Keith, December 2015

Reithrodontomys and basal to this group. In order to recognize Habromys,

Megadontomys, Neotomodon, Osgoodomys, and Podomys as genera, Bradley et al.

(2007) stated that at least 5 additional genera would have to be recognized to accommodate strongly supported clades under the rules of monophyly and phylogenetic principles.

Miller and Engstrom (2008) added to the molecular dataset by obtaining DNA sequences from Cytb, as well as 2 nuclear genes: interphotoreceptor retinoid-binding protein (Rbp3) and growth hormone receptor (Ghr). Their results were similar to

Bradley et al. (2007) in that Habromys, Megadontomys, Neotomodon, Osgoodomys, and Podomys were placed inside of Peromyscus (sensu stricto) and Isthmomys was sister to Reithrodontomys. Further, Miller and Engstrom (2008) agreed with Bradley et al. (2007) that additional groups would have to be elevated to avoid paraphyly in a sensu stricto interpretation of Peromyscus.

The primary objective of this study was to examine the phylogenetic relationships within the genus Peromyscus using a combination of mitochondrial and nuclear markers. Although the studies of Bradley et al. (2007) and Miller and

Engstrom (2008) recovered paraphyly within Peromyscus, each study had limitations that impact phylogenetic interpretation. Bradley et al. (2007) had a more broad taxonomic sampling scheme but was based on a single genetic marker (Cytb). Miller and Engstrom (2008) lacked representation of some species groups but included multiple genetic markers. Herein, we seek to expand upon these molecular datasets by including representatives from all species groups and to examine DNA sequences

Texas Tech University, Megan S. Keith, December 2015 from 4 markers, 2 of which were not used in the previous studies; intron 2 of the alcohol dehydrogenase (Adh1-I2) and intron 7 of the beta-fibrinogen gene (Fgb-I7).

We selected these markers based on their previous use in rodent phylogenetics

(Amman et al. 2006; Longhofer and Bradley 2006; Reeder and Bradley 2007). We combined nuclear and mitochondrial DNA sequence data from Adh1-I2, Fgb-I7, Rbp3, and Cytb to test the monophyly of Peromyscus (sensu stricto versus sensu lato) and to ascertain whether Habromys, Isthmomys, Megadontomys, Neotomodon, Osgoodomys, and Podomys are paraphyletic within Peromyscus. The genetic evidence presented herein support the need for a formal taxonomic revision of Peromyscus. We identify 3 potential taxonomic solutions that are consistent with the evidence in hand.

MATERIALS AND METHODS Samples.—Tissue samples obtained from individuals collected in naturally occurring populations or through museum loans were used to generate DNA sequences for the 4 genetic markers described below. In some cases, DNA sequences were obtained directly from GenBank. A single representative was examined from each of the following taxa: Baiomys, Habromys, Isthmomys, Megadontomys,

Neotoma, Neotomodon, Ochrotomys, Onychomys, Osgoodomys, and Podomys; 4 representatives were included for Reithrodontomys. For the subgenus Peromyscus, 2 representatives of each of the species groups were examined (except the crinitus, furvus, hooperi, megalops, and melanophrys species groups - 1 sample each).

Likewise, for the subgenus Haplomylomys, 1 sample each from the californicus and eremicus species groups were examined. An attempt was made to obtain both

Texas Tech University, Megan S. Keith, December 2015 mitochondrial and nuclear sequences from a single individual, but in a few instances this was not possible. In these cases, sequences from conspecific individuals within close geographic proximity were used to complete the dataset. Concatenation of sequence data from conspecifics to represent a composite species rather than a single individual has been successfully used in various taxa (Campbell and Lapointe 2009;

Haddrath and Baker 2012; Townsend et al. 2011). This strategy guaranteed at least 1 individual per species group was sampled. Specimens used in this study are listed in

Table 2.1.

DNA isolation and PCR.—DNA was isolated from liver samples (0.1 g) using

2 methods. Mitochondrial DNA was extracted and purified using a Wizard Miniprep kit (Promega, Madison, Wisconsin), whereas total genomic DNA was extracted from liver using DNeasy Blood and Tissue kits (Qiagen, Valencia, California) following the method of Smith and Patton (1999). The complete mitochondrial Cytb gene (1,143 bp) was amplified following methods outlined in Bradley et al. (2007) and Tiemann-

Boege et al. (2000) using primers MVZ05 (Smith and Patton 1993), H15915 (Irwin et al. 1991), and CB40 (Hanson and Bradley 2008). Intron 2 of the alcohol dehydrogenase gene (Adh1-I2, 598 bp) was amplified following the methods of

Amman et al. (2006) using primers 2340-I, 2340-II, Exon II-F, and Exon III-R. The complete intron of the beta-fibrinogen gene (Fgb-I7, 674 bp) was amplified following the methods of Carroll et al. (2005) and Wickliffe et al. (2003) using primers, Fgb-

17U-Rattus, Fgb-17L-Rattus (Wickliffe et al. 2003), B17-mammU, and B17-mammL

(Matocq et al. 2007). Exon I of interphotoreceptor retinoid binding protein gene

Texas Tech University, Megan S. Keith, December 2015

(Rbp3, 924 bp) was amplified following the methods of Chambers et al. (2009) and

Jansa and Voss (2000) using primers A and B (Stanhope et al. 1992).

Sequencing.—PCR products were purified using the QIAquick PCR Purification kit (Qiagen, Valencia, California) or ExoSAP-IT (USB Products, Cleveland, Ohio) and PCR amplicons were sequenced using ABI Prism Big Dye Terminator v3.1 ready reaction mix (Applied Biosystems, Foster City, California). Nucleotide sequences were resolved on an ABI 3100 Avant automated sequencer (Applied Biosystems,

Foster City, California) with the following primers: Cytb--PERO3' and 752R

(Tiemann-Boege et al. 2000), CWE-1 and 400F (Edwards et al. 2001), and 700L and

WDRAT400R (Peppers and Bradley 2000); Adh1-I2--Exon II-F, Exon III-R,

Adh350F, and Adh350R (Amman et al. 2006); Fgb-I7--Fgb-17U-Rattus and Fgb-17L-

Rattus (Wickliffe et al. 2003) and bFIB-I7U and bFIB-I7L (Carroll et al. 2005); and

Rbp3--A, B, and D (Stanhope et al. 1992), E2 (Weksler 2003), and 125F (DeBry and

Sagel 2001). Sequencher 5.0 software (Gene Codes Corporation, 2013) was used to align and proof individual sequencing reads into contigs representing each gene.

Conflicting base calls were verified against the associated chromatograms. For nuclear intron sequences all heterozygous sites were designated following the

International Union of Biochemistry polymorphic code. All DNA sequences were deposited in GenBank and accession numbers are provided in Table 2.1.

Phylogenetic analysis.—Nucleotide positions were treated as unordered, discrete characters with 6 possible states: A, C, G, T, gaps (-) or missing (?) for each marker. For nuclear intron sequences, polymorphic sites were designated following

Texas Tech University, Megan S. Keith, December 2015 the International Union of Biochemistry polymorphic code. However, because these polymorphisms could be the result of heterozygosity or sequencing error, to be conservative, these nucleotides were excluded from downstream analyses. Alignment of Adh1-I2 and Fgb-I7 sequences required hypothesized gaps (inserted based on homology) to represent insertion or deletion events, but gaps were not included in the phylogenetic analysis. Analyses were conducted using 3 datasets: nuclear (Adh1-I2,

Fgb-I7, Rbp3), mitochondrial (Cytb), and combined (Adh1-I2, Cytb, Fgb-I7, and

Rbp3). Neotoma mexicana was used as the outgroup taxon for all analyses (Bradley et al. 2004b).

MrModeltest and the Akaike information criterion (Nylander 2004) were used to estimate the most appropriate model of evolution for each gene region. Bayesian inference (BI) was conducted to estimate a phylogeny and generate posterior probability values for the mitochondrial, nuclear, and combined datasets using

MRBAYES v3.2.1 (Ronquist et al. 2012). Each analysis included the appropriate model identified by MrModeltest (Nylander 2004), 2 simultaneous runs of 4 Markov- chains, 10 x 109 generations, and a sample frequency of every 1,000th generation.

The number of invariable sites and gamma distribution were estimated from the data.

After a visual inspection of likelihood score distributions in Tracer v1.5 (Drummond and Rambaut 2008), the first 10,000 trees were discarded and a consensus tree (50% majority rule) was constructed from the remaining trees. Values ≥ 95% were viewed as supportive following Alfaro et al. (2003), Douady et al. (2003), and Huelsenbeck et al. (2002). For ML analyses, RaxML (Stamatakis et al. 2005) was used to generate

Texas Tech University, Megan S. Keith, December 2015 trees from each dataset. In these analyses, the GTR+G substitution model was used since the less parameter rich HKY+G model, identified by MrModeltest (Nylander

2004) as the most appropriate model, is unavailable. Nodal support was estimated with 10,000 bootstrap replicates using the “fast bootstrapping” option (Felsenstein

1985).

Topological tests.—Maximum likelihood trees from RAxML (Stamatakis et al.

2005) were used to test the difference between competing taxonomic hypotheses. Site likelihood scores generated in RAxML (Stamatakis et al. 2005) were used to score several constrained topologies. P-values were generated in Consel (Shimodaira and

Hasegawa 2001) for each topology using the approximately unbiased test (Shimodaira

2002). In particular, Peromyscus (sensu stricto) versus (sensu lato) were tested against the ML topology from the combined data analysis, as well as other alternative hypothetical taxonomic groupings.

Molecular dating.—BEAST v1.7 (Drummond et al. 2012) was used to estimate divergence dates for the sampled taxa. Sequence data from each gene was used in the analysis but were partitioned to allow modeling of each dataset. Models of substitution were the same as those used in previous Bayesian analyses (see above).

The program MEGA 5.05 (Tamura et al. 2011) was used to determine whether to accept or reject a strict molecular clock for each dataset. Given all datasets contained a single individual from each species sampled, a Yule tree prior was chosen for the

BEAST analysis. Fossil limits were used to calibrate the leucopus/maniculatus group

(~0.3 million years ago [MYA]—Dalquest 1962; Karow et al. 1996) and

Texas Tech University, Megan S. Keith, December 2015

Reithrodontomys (~1.8 MYA—Cassiliano 1999). To account for the uncertainty in the fossil record, a prior lognormal distribution was used for both calibrations with means and standard deviations adjusted to create an upper bound of 14.8 MYA to reflect the closest dated fossil outside of the taxa sampled (Behrensmeyer and Turner

2013). Test runs of 2.5 X 107 generations with a 10% burn-in were used to optimize for the final analysis. Bayes Factors (Kass and Raftery 1995; Suchard et al. 2001) were calculated to compare the results of test runs to determine final parameters. Two final runs of 1.0 x 108 generations were analyzed with log and tree files combined for final divergence date estimates. Results were examined for sufficient mixing, convergence stability, and effective sample size > 200 for all parameters using the program Tracer.

Genetic divergence.—To compare rates of genetic divergence between taxa recognized at various taxonomic ranks Kimura 2-parameter (K2P—Kimura 1980) genetic distance values were compared among currently recognized genera

(Habromys, Isthmomys, Megadontomys, Neotomodon, Osgoodomys, and Podomys).

The K2P model was selected based on its utility as a distance metric in rodent phylogenetics (Bradley and Baker 2001).

RESULTS Phylogenetic analyses.—Twenty-seven species of Peromyscus (sensu lato) and

8 additional taxa (outgroup and reference samples), representing taxonomic diversity within the Neotominae, were sampled for the nuclear introns Adh1-I2 and Fgb-I7, nuclear exon Rbp3, and the mitochondrial gene Cytb. The entire 560 bp of Adh1-I2,

Texas Tech University, Megan S. Keith, December 2015

590 bp of Fgb-I7, and 921 bp of Rbp3 were analyzed for 23 of the 27 Peromyscus

(sensu lato) species. Full gene sequences were not available for P. californicus (Rbp3:

833 of 921 bp), P. eremicus (Rbp3: 907 of 921 bp), P. furvus (Cytb: 540 of 1,143 bp), or P. ochraventer (Rbp3: 914 of 921 bp). To analyze the most complete dataset possible, other gene sequences available through GenBank, including Ghr, were not used due to lack of sequence data for many taxa. MrModeltest (Nylander 2004) identified the substitution models HKY+G for Adh1-I2 (AIC = 6,577.2827, –lnL =

3,283.6414, G = 1.1092), GTR+G for Fgb-I7 (AIC= 6931.1782, –lnL = 3,456.5891, G

= 1.3322) and GTR+I+G for Rbp3 (AIC= 6,213.5400, –lnL = 3,096.7700, I = 0.4097,

G = 0.8473) as the best-fit models.

Individual Adh1-I2, Cytb, Fgb-I7, and Rbp3 sequences were combined to generate a single concatenated sequence (3,283 bp). The ML phylogeny (-lnL =

22,631.507197) with bootstrap values and Bayesian clade probability values are shown in Figure 1A along with the mitochondrial (Cytb; Fig. 2.1B) and combined nuclear (Adh1-I2, Fgb-I7, and Rbp3; Fig. 2.1C) topologies. Specific placement of some taxa varied between the ML and BI analyses, specifically in the placement of

Megadontomys thomasi, Osgoodomys banderanus, and P. hooperi. Despite uncertain placement of these taxa, placement of terminal taxa within well-supported nodes did not conflict between markers. Significant nodal support was obtained throughout the phylogeny with most basal and terminal nodes garnering support. Middle regions of the phylogeny were less likely to receive nodal support. Both BI and ML analyses

Texas Tech University, Megan S. Keith, December 2015 were unable to recover a monophyletic Peromyscus (sensu stricto) or Peromyscus

(sensu lato) clade.

Constrained topologies reflecting various taxonomic groupings were tested using the approximately unbiased test in CONSEL (Shimodaira and Hasegawa 2001) with 10,000 replicates per test. A generalized Peromyscus (sensu lato) was unable to be rejected (P = 0.406), but Peromyscus (sensu stricto) was strongly rejected (P =

0.014) by the approximately unbiased test. An additional taxonomic scheme uniting

Peromyscus (sensu lato), but excluding Isthmomys pirrensis, was unable to be rejected

(P = 0.556).

Molecular dating.—Molecular clock tests (Tamura et al. 2011) indicated a strict molecular clock for the Cytb and Fgb-I7 datasets but a relaxed molecular clock for the Adh1-I2 and Rbp3 datasets. BEAST analyses estimated a Yule birth rate of

0.23 (95% highest posterior density [HPD]: 0.11–0.37). Mean rates of evolution (as substitutions per site per million years) were 0.007 for Adh1-I2 (95% HPD: 0.004--

0.01), 0.06 for Cytb (95% HPD: 0.03--0.09), 0.006 for Fgb-I7 (95% HPD: 0.003--

0.009), and 0.003 for Rbp3 (95% HPD: 0.001--0.004). Divergence date estimates

(Fig. 2) suggested that the split of Reithrodontomyini and Baiomyini began approximately 9.56 MYA (95% HPD: 5.65–15.27), during the Late Miocene. The split between the Isthmomys/Reithrodontomys clade and Onychomys/Peromyscus

(sensu lato) clade was estimated to occur approximately 7.93 MYA (95% HPD: 4.67–

12.59), also in the Late Miocene. In addition, the divergence of Peromyscus (sensu lato) was estimated to occur during the Late Miocene but near the Miocene/Pliocene

Texas Tech University, Megan S. Keith, December 2015 boundary (approximately 5.71 MYA, 95% HPD: 3.37–9.08). However, most species- level divergence within Peromyscus (sensu lato) occurred during the Blancan North

American land mammal age (1.8–4.9 MYA).

Genetic distances.—K2P (Table 2.2) genetic distances were used to compare taxa and provide additional information on the phylogenetic utility of each marker.

Data obtained from comparisons of Isthmomys, Onychomys, and Reithrodontomys to other genera and species groups indicated the highest levels of genetic divergence among taxa. Comparison of the 5 genera (Habromys, Megadontomys, Neotomodon,

Osgoodomys, and Podomys) to all other genera resulted in genetic distances ranging from 2.58% (Rbp3) to 15.4% (Cytb). Values obtained from comparisons of these 5 genera to currently recognized species groups within Peromyscus (sensu stricto) ranged from 1.9% (Rbp3) to 14.62% (Cytb) and were similar in magnitude to comparisons of species groups to each other. By comparing the genetic distance between all taxa examined for each respective marker, their relative rates of evolution could be compared. Overall, Adh1-I2 and Fgb-I7 exhibited rates of evolution slower than Cytb; however, Rbp3 was substantially slower than all of the other markers.

DISCUSSION Use of a combined dataset often increases resolution at different hierarchical levels with one dataset providing resolution at deep nodes and others resolving shallow nodes. More quickly evolving mitochondrial markers tend to depict more resolution at terminal nodes, whereas nuclear markers generally resolve relationships at the base of a phylogeny. Therefore, combined datasets are often advantageous in

Texas Tech University, Megan S. Keith, December 2015 studies whose phylogenetic relationships have been debated due to inconsistencies among studies or datasets. In addition, increasing the number of characters allows phylogenetic signal to assert itself over noise (homoplasy), resulting in a more accurate estimate of relationships.

Of the data analyzed, the combined dataset provided the greatest resolution and nodal support (Fig. 2.1). Additionally, the combined dataset provided resolution at several levels throughout the topology. Therefore, we use the topology from the BI analysis, as well as statistical support from ML analyses (Fig. 2.1) of the combined dataset, to discuss the phylogenetic relationships of Peromyscus. We begin by discussing Peromyscus using the current taxonomy based on the sensu stricto framework unless indicated otherwise (Carleton 1980; Musser and Carleton 2005).

Clade I contains members of Peromyscus (sensu lato), Onychomys, and

Reithrodontomys (Fig. 2.1A). This relationship agrees with a Reithrodontomyini tribal definition as proposed by Miller and Engstrom (2008) and Musser and Carleton

(2005), as well as relationships recovered in Bradley et al. (2004b, 2007), Carleton

(1980), McKenna and Bell (1997), and Reeder and Bradley (2004, 2007). The pairing of Isthmomys and Reithrodontomys in the combined analyses garnered no statistical support but generally agrees with analyses using allozymic (Rogers et al. 2005) and multiple combinations of DNA sequence data (Bradley et al. 2007; Miller and

Engstrom 2008).

Clade II most closely represents Peromyscus (sensu lato) as interpreted by

Hooper (1968), with the exclusion of Isthmomys. Analyses of the nuclear and

Texas Tech University, Megan S. Keith, December 2015 combined datasets each formed a well-supported clade similar to clade II. Although some basal branching patterns within this clade receive little or no support, it is apparent that the taxa recognized as separate genera (Habromys, Megadontomys,

Neotomodon, Osgoodomys, and Podomys) by Carleton (1980, 1989) and Musser and

Carleton (2005) are embedded within an assemblage containing Peromyscus (sensu stricto). Sub-clades within clade II support Peromyscus (sensu stricto) paraphyly.

Clade III contains the majority of peromyscine species examined and generally agrees with the findings of Bradley et al. (2007). However, the current study differs from Bradley et al. (2007) in the placement of certain taxonomic groups, although the terminal branching patterns are similar. Two strongly supported subclades recovered comprised of the aztecus (Peromyscus evides and P. spicilegus) and boylii

(Peromyscus boylii and P. levipes) species groups (Bradley et al. 2004b; Rennert and

Kilpatrick 1986, 1987; Sullivan et al. 1991, 1997; Sullivan and Kilpatrick 1991;

Tiemann-Boege et al. 2000), as well as a clade containing the megalops (Peromyscus megalops), melanophrys (Peromyscus melanophrys), and mexicanus (Peromyscus mexicanus and P. nudipes) species groups which agrees with the study by Bradley et al. (2007). Clade IV depicts the relationship among P. californicus, P. crinitus, and P. eremicus, and indicates that P. californicus and P. eremicus are members of the subgenus Haplomylomys; whereas P. crinitus is the sole member of the crinitus species group. The relationship between these 2 species groups was supported in the combined Bayesian analysis. Clade V included members of the leucopus (Peromyscus gossypinus and P. leucopus) and maniculatus (Peromyscus maniculatus and P.

Texas Tech University, Megan S. Keith, December 2015 melanotis) species groups. Strong support existed for the sister relationship between these species groups.

Several clades did not receive support in any of the analyses, including the unresolved placement of M. thomasi, N. alstoni, O. banderanus, and Podomys floridanus; although their inclusion within Clade II is supported. In addition, no support was recovered for a monophyletic group containing all members of the P. attwateri (truei species group) or for the relationships of several species groups (e.g., hooperi). A sister relationship between Reithrodontomys and Peromyscus (sensu lato, excluding Isthmomys) received strong support.

The origin of Peromyscus (sensu lato) began approximately 8 MYA (Fig. 2.2); however, the radiation of Peromyscus (sensu lato) excluding Isthmomys appears to have been focused around 5.71 MYA (95% HPD: 3.37–9.08). During this time,

Habromys, Megadontomys, Neotomodon, Osgoodomys, and Podomys originated as well as several major Peromyscus lineages including Haplomylomys (Peromyscus californicus, P. eremicus, and P. crinitus), the mexicanus (Peromyscus mexicanus and

P. nudipes) and boylii (Peromyscus boylii and P. levipes) species groups, and P. pectoralis. These lineages emerged between the minimum and maximum dates from

León-Paniagua et al. (2007) and are further evidence for an origination date of

Peromyscus (sensu lato), excluding Isthmomys, around 6 MYA followed by a rapid diversification.

Estimation of the genetic distance values among selected taxa (genera and species groups) allowed for a gross-level comparison of genetic divergence among

Texas Tech University, Megan S. Keith, December 2015 said groups (Table 2.2). For example, Isthmomys, Onychomys, and Reithrodontomys depicted substantially higher levels of genetic divergence than any other comparison.

Comparisons of genetic divergence among the 5 genera (Habromys, Megadontomys,

Neotomodon, Osgoodomys, and Podomys) to other currently recognized species groups within Peromyscus (sensu stricto) produced values similar in magnitude to comparisons of the species groups to each other.

Although several recent studies have focused on developing phylogenies for

Peromyscus (sensu lato and sensu stricto) and its affiliated genera, for a variety of reasons, none have been able to offer unambiguous taxonomic recommendations.

First, Peromyscus is a large genus with new species still being described (Bradley et al. 2004a, 2014), and complete taxonomic sampling is often difficult. Second, because several character systems have been studied and analyzed, it presents a challenge to resolve discrepancies when there are conflicting data. Third, Peromyscus may have undergone a rapid radiation that makes it difficult to reconstruct phylogenetic relationships (Fig. 2.1) with the available data. Fourth, and perhaps most important, is the occurrence of genetic conservation between taxa that exhibit substantial levels of morphological differences. For example, Carleton’s (1980) decision to elevate

Habromys, Megadontomys, Neotomodon, Osgoodomys, and Podomys to generic status was based on the occurrence of substantial morphological differentiation among taxa; yet these same taxa do not exhibit comparable levels of genetic divergence. Other morphological studies involving the glans penes and bacula (Hooper 1958; Hooper and Musser 1964) also have indicated high levels of morphological divergence among

Texas Tech University, Megan S. Keith, December 2015 these same genera. Although recent genetic studies (i.e., herein; Miller and Engstrom

2008; Rogers et al. 2005) did not examine genes coding for the external morphological characters examined in Carleton’s (1980) study, one would assume a concomitant rate of evolution. Resolving the incongruence between morphological and genetic data may be an exciting development of its own. A cursory analysis of the molecular topology produced herein, and the morphological characters analyzed in Carleton

(1980) failed to recover a fixed, derived character that unites Peromyscus.

No phylogenetic analysis, herein, recovered a clade that corresponded to

Peromyscus (sensu stricto). In our analyses, Habromys, Megadontomys, Neotomodon,

Osgoodomys, and Podomys continually were placed inside of Peromyscus (sensu stricto), producing a paraphyletic assemblage. When the molecular topology was constrained to reflect a Peromyscus (sensu stricto) framework, a significantly worse

(P < 0.05) topology was recovered. Further, we were unable to reject a Peromyscus

(sensu lato) relationship. When these results are combined with available genetic data

(Bradley et al. 2007; Miller and Engstrom 2008; Reeder and Bradley 2004, 2007;

Reeder et al. 2006; Rogers et al. 2005), it is clear that Peromyscus (sensu stricto) as currently recognized should be abandoned based on its paraphyletic nature. Providing a new, more accurate Peromyscus taxonomy is difficult because its placement can be interpreted in multiple ways due to the paraphyletic inclusion of Habromys,

Megadontomys, Neotomodon, Osgoodomys, and Podomys in most analyses.

Similarly, Peromyscus (sensu lato) requires revision, as one of its members

(Isthmomys) forms a paraphyletic assemblage with Reithrodontomys or groups outside

Texas Tech University, Megan S. Keith, December 2015 of a monophyletic Peromyscus. However, the Peromyscus (sensu lato) moniker can be recovered by simply removing Isthmomys and recognizing it as a separate genus

(regardless of its affinities outside of Peromyscus), following Bradley et al. (2007),

Miller and Engstrom (2008), and Rogers et al. (2005). BI and ML analyses both recovered topologies that supported exclusion of Isthmomys from a

Peromyscus/Onychomys clade, yet an approximately unbiased topology test did not produce a significantly worse topology when monophyly was enforced on a

Peromyscus (sensu lato) and Isthmomys clade. The inability of the approximately unbiased topology test to support exclusion of Isthmomys from Peromyscus (sensu lato) is likely due to the inclusion of only 4 of more than 20 species of

Reithrodontomys and a single representative for Isthmomys. It is expected that increased sampling will further support the exclusion of Isthmomys from Peromyscus

(sensu lato). Even with the removal of Isthmomys, paraphyly in the subgenera

Peromyscus and Haplomylomys produced by the inclusion of Habromys,

Megadontomys, Neotomodon, Osgoodomys, and Podomys would remain problematic as discussed above.

It is possible to resolve monophyly of Peromyscus with taxonomies that broadly recognize groups at the generic, subgeneric, or species group level.

Monophyletic clades from within Peromyscus (sensu stricto), as well as Habromys,

Megadontomys, Neotomodon, Osgoodomys, and Podomys, could each be recognized as genera. Similarly, Habromys, Megadontomys, Neotomodon, Osgoodomys, and

Podomys could be subsumed to subgenera within Peromyscus. Finally, Habromys,

Texas Tech University, Megan S. Keith, December 2015

Megadontomys, Neotomodon, Osgoodomys, and Podomys could be subsumed to species groups within Peromyscus. Each option presents specific taxonomic challenges that are discussed below and summarized in Table 2.3.

By retaining Habromys, Megadontomys, Neotomodon, Osgoodomys, and

Podomys at the generic level, the paraphyly within Peromyscus must be resolved by elevating monophyletic clades to the generic level (Table 2.3). Unfortunately, many of these clades originate at unsupported nodes within the phylogeny produced herein

(Fig. 2.1). Further studies may be better able to resolve these relationships. Based on the current phylogeny, the elevation of a minimum of 2 new genera would be necessary to resolve paraphyly within Peromyscus (Genus A - P. pectoralis, P. levipes, P. boylii, P. spicilegus, P. evides, P. ochraventer, P. gratus, P. attwateri, and

P. furvus; Genus B - P. mexicanus, P. nudipes, P. melanophrys, and P. megalops).

This option is unfeasible due to lack of statistical support in the phylogeny. A 2-genus option will need to be continually evaluated as new data and data types become available. If genera are designated only at supported monophyletic nodes then up to 4 new genera would require elevation from Peromyscus (Genus A - P. megalops, P. melanophyrs, P. mexicanus, and P. nudipes; Genus B - P. evides, P. boylii, P. levipes, and P. spicilegus; Genus C - P. attwateri, P. furvus, P. gratus, and P. ochraventer;

Genus Peromyscus - P. gossypinus, P. leucopus, P. maniculatus, and P. melanotis) with uncertain placement of P. hooperi and P. pectoralis. Using the subgeneric option, a genus taxonomically similar to Peromyscus can be retained by subsuming

Habromys, Megadontomys, Neotomodon, Osgoodomys, and Podomys (Table 2.3).

Texas Tech University, Megan S. Keith, December 2015

The elevation of subgenera within Peromyscus would be necessary, and newly elevated subgenera would be similar in species content to the genera created using the generic option. Finally by removing higher taxonomic ranks (genus or subgenus), paraphyletic assemblages can be resolved while continuing to recognize morphological variation and account for clades identified with genetic data (Table

2.3). Species groups have proven to be valuable units to study evolution within

Peromyscus (Bradley et al. 2004b; Durish et al. 2004; Riddle et al. 2000) and perhaps their usage would serve as a viable solution until the phylogenetic relationships of unresolved taxa are determined. Additionally, monophyly of most species groups has been somewhat resolved (except mexicanus and furvus—Bradley et al. 2007). The species group option, however, fails to recognize degrees of morphological variation that the generic and subgeneric options could offer if additional subrankings were established. Taxonomic changes would still be required in the recognition of the species groups; however, this option requires minimal changes relative to recognizing additional genera or subgenera.

In developing a revised classification for Peromyscus, standards must be agreed upon that designate distinction at a genetic level yet accommodate morphological variation. Some of these standards already are understood such as achieving monophyly and cohesion within the group. However, determining how much variation warrants generic recognition is difficult. For example, Helgen et al.

(2009) and Weksler et al. (2003) recently revised the genera formerly recognized as

Spermophilus and Oryzomys, respectively. Their revisions produced monophyly and

Texas Tech University, Megan S. Keith, December 2015 clarification of groups by the naming of additional genera to accommodate monophyletic clades produced in their analyses. Based on the data herein, it is clear that the current taxonomy of Peromyscus (sensu stricto) should be abandoned as well.

However, to resolve the paraphyly within Peromyscus, at least 3 different taxonomic options are available and should be considered. More diverse data types, including morphology, karyology, and ecology, as well as additional genetic data, will be required to develop the taxonomy that properly recognizes the diversity and distinction within Peromyscus.

ACKNOWLEDGMENTS We thank D. S. Rogers, M. D. Engstrom, and J. R. Miller for the generation of many of the sequences available in GenBank. We thank J. Light (Texas Cooperative

Wildlife Collection, Texas A&M University) and R. J. Baker (Natural Science

Research Laboratory, Museum, Texas Tech University) for kindly providing tissue loans. H. R. Huynh, M. R. Mauldin, N. O. Ordóñez-Garza, and E. K. Roberts provided helpful comments on previous versions of this manuscript. This research was supported in part by grants from the National Institutes of Health (DHHS

A141435-01 to RDB) and the National Science Foundation (MCB-0841821 and DEB-

1020865 to RNP). Additional support was provided by the Mississippi Agricultural and Forestry Experiment Station.

Texas Tech University, Megan S. Keith, December 2015

LITERATURE CITED ALFARO, M. E., S. ZOLLER, and F. LUTZONI. 2003. Bayes or Bootstrap? A simulation

study comparing the performance of Bayesian Markov Chain Monte Carlo

sampling and bootstrapping in assessing phylogenetic confidence. Molecular

Biology and Evolution 20:255-266.

AMMAN, B. R., J. D. HANSON, L. K. LONGHOFER, S. R. HOOFER, and R. D. BRADLEY.

2006. Intron 2 of the alcohol dehydrogenase gene (ADH1-I2): a nuclear DNA

marker for mammalian systematics. Occasional Papers, Museum of Texas

Tech University 256:1-16.

BEHRENSMEYER, A. K., and A. TURNER. 2013. Taxonomic occurrences of Peromyscus

recorded in the Paleobiology Database. Fossilworks. http://fossilworks.org.

Accessed 3 April 2014.

BRADLEY, R. D., and R. J. BAKER. 2001. A test of the genetic species concept:

Cytochorme-b sequences and mammals. Journal of Mammalogy 82:960-973.

BRADLEY, R. D., D. S. CARROLL, M. L. HAYNIE, R. MUÑIZ MARTÍNEZ, M. J.

HAMILTON, and C. W. KILPATRICK. 2004a. A new species of Peromyscus from

western Mexico. Journal of Mammalogy 85:1184-1193.

BRADLEY, R. D., N. D. DURISH, D. S. ROGERS, J. R. MILLER, M. D. ENGSTROM, and C.

W. KILPATRICK. 2007. Toward a molecular phylogeny for Peromyscus:

evidence from mitochondrial Cytochrome-b sequences. Journal of

Mammalogy 88:1146-1159.

Texas Tech University, Megan S. Keith, December 2015

BRADLEY, R. D., C. W. EDWARDS, D. S. CARROLL, and C. W. KILPATRICK. 2004b.

Phylogenetic relationships of Neotomine-Peromyscine rodents: based on DNA

sequences from the mitochondrial Cytochrome-b gene. Journal of Mammalogy

85:389-395.

BRADLEY, R. D., et al. 2014. Morphometric, karyotypic, and molecular evidence for a

new species of Peromyscus (Cricetidae: Neotominae) from Nayarit, Mexico.

Journal of Mammalogy 95:176-186.

CAMPBELL, V., and F.-J. LAPOINTE. 2009. The use and validity of composite taxa in

phylogenetic analysis. Systematic Biology 58:560-572.

CARLETON, M. D. 1980. Phylogenetic relationships in neotomine-peromyscine rodents

(Muroidea) and a reappraisal of the dichotomy within New World Cricetinae.

Miscellaneous Publications of the Museum of Zoology 157:146.

CARLETON, M. D. 1989. Sytematics and evolution. Pp. 7-141 in Advances in the Study

of Peromyscus (Rodentia)(G.L. Kirkland andJ.N. Layne, eds.), Texas Tech

Univeristy Press, Lubbock.

CARROLL, D. S., R. D. BRADLEY, and C. W. EDWARDS. 2005. Systematics of the genus

Sigmodon: DNA sequences from Beta-Fibrinogen and Cytochrome b. The

Southwestern Naturalist 50:342-349.

CASSILIANO, M. L. 1999. Biostratigraphy of Blancan and Irvingtonian mammals in the

Fish Creek-Vallecito section, southern California, and a review of the

Blancan–Irvingtonian boundary. Journal of Vertebrate Paleontology 19:169-

186.

Texas Tech University, Megan S. Keith, December 2015

CHAMBERS, R. R., P. D. SUDMAN, and R. D. BRADLEY. 2009. A phylogenetic

assessment of pocket gophers (Geomys): Evidence from nuclear and

mitochondrial genes. Journal of Mammalogy 90:537-547.

DALQUEST, W. W. 1962. The Good Creek Formation, Pleistocene of Texas, and its

fauna. Journal of Paleontology 36:568-582.

DEBRY, R. W., and R. M. SAGEL. 2001. Phylogeny of Rodentia (Mammalia) inferred

from the nuclear-encoded gene IRBP. Molecular Phylogenetics and Evolution

19:290-301.

DOUADY, C. J., F. DELSUC, Y. BOUCHER, W. F. DOOLITTLE, and E. J. P. DOUZERY.

2003. Comparison of Bayesian and maximum likelihood bootstrap measures of

phylogenetic reliability. Molecular Biology and Evolution 20:248-254.

DRUMMOND, A. J., and A. RAMBAUT. 2008. Tracer v1.5.

http://tree.bio.ed.ac.uk/software/tracer/. Accessed 27 October 2014

DRUMMOND, A. J., M. A. SUCHARD, D. XIE, and A. RAMBAUT. 2012. Bayesian

Phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and

Evolution 29:1969-1973.

DURISH, N. D., K. E. HALCOMB, C. W. KILPATRICK, and R. D. BRADLEY. 2004.

Molecular systematics of the Peromyscus truei species group. Journal of

Mammalogy 85:1160-1169.

EDWARDS, C. W., C. F. FULHORST, and R. D. BRADLEY. 2001. Molecular

phylogenetics of the Neotoma albigula species group: further evidence of a

paraphyletic assemblage. Journal of Mammalogy 82:267-279.

Texas Tech University, Megan S. Keith, December 2015

FELSENSTEIN, J. 1985. Confidence limits on phylogenies: an approach using the

bootstrap. Evolution 39:783-791.

HADDRATH, O., and A. J. BAKER. 2012. Multiple nuclear genes and retroposons

support vicariance and dispersal of the palaeognaths, and an Early Cretaceous

origin of modern birds. Proceedings of the Royal Society B: Biological

Sciences.

HANSON, J. D., and R. D. BRADLEY. 2008. Molecular diversity within Melanomys

caliginosus (Rodentia: Oryzomyini): evidence for multiple species. Occasional

Papers, Museum of Texas Tech University 275:1-11.

HELGEN, K. M., F. R. COLE, L. E. HELGEN, and D. E. WILSON. 2009. Generic revision

in the holarctic ground squirrel genus Spermophilus. Journal of Mammalogy

90:270-305.

HOOPER, E. T. 1958. The male phallus in mice of the genus Peromyscus.

Miscellaneous Publications of the Museum of Zoology,University of

Michigan, 105:1-24.

HOOPER, E. T. 1968. Classification. Pp. 27-74 in Biology of Peromyscus (Rodentia) (J.

A. King, ed.). Special Publication, American Society of Mammalogists, 2:1-

593.

HOOPER, E. T., and G. G. MUSSER. 1964. The glans penis in neotropical Cricetines

(Family Muridae) with comments on classification of muroid rodents.

Miscellaneous Publications of the Museum of Zoology, University of

Michigan, 123:1-57.

Texas Tech University, Megan S. Keith, December 2015

HUELSENBECK, J. P., B. LARGET, R. E. MILLER, and F. RONQUIST. 2002. Potential

applications and pitfalls of Bayesian inference of phylogeny. Systematic

Biology 51:673-688.

IRWIN, D., T. KOCHER, and A. WILSON. 1991. Evolution of the Cytochrome b gene of

mammals. Journal of Molecular Evolution 32:128-144.

JANSA, S., and R. VOSS. 2000. Phylogenetic studies on didelphid marsupials I.

Introduction and preliminary results from nuclear IRBP gene sequences.

Journal of Mammalian Evolution 7:43-77.

KAROW, P., G. MORGAN, R. PORTELL, E. SIMMONS, and K. AUFFENBERG. 1996.

Middle Pleistocene (early Rancholabrean) vertebrates and associated marine

and non-marine invertebrates from Oldsmar, Pinellas County, Florida. Pp. 97-

133 in Palaeoecology and Palaeoenuironments of Late Cenozoic mammals

Tributes to the Career of C S (Rufus) Churcher (K. Stewart and K. Seymour,

eds.). University of Tronto Press, Toronto, Canada.

KASS, R. E., and A. E. RAFTERY. 1995. Bayes Factors. Journal of the American

Statistical Association 90:773-795.

KIMURA, M. 1980. A simple method for estimating evolutionary rates of base

substitutions through comparative studies of nucleotide sequences. Journal of

Molecular Evolution 16:111-120.

LEÓN-PANIAGUA, L., A. G. NAVARRO-SIGÜENZA, B. E. HERNÁNDEZ-BAÑOS, and J. C.

MORALES. 2007. Diversification of the arboreal mice of the genus Habromys

Texas Tech University, Megan S. Keith, December 2015

(Rodentia: Cricetidae: Neotominae) in the Mesoamerican highlands. Molecular

Phylogenetics and Evolution 42:653-664.

LONGHOFER, L. K., and R. D. BRADLEY. 2006. Molecular systematics of the genus

Neotoma based on DNA seqeunces from intron 2 of the Alcohol

Dehydrogenase gene. Journal of Mammalogy 87:961-970.

MATOCQ, M. D., Q. R. SHURTLIFF, and C. R. FELDMAN. 2007. Phylogenetics of the

woodrat genus Neotoma (Rodentia: Muridae): implications for the evolution of

phenotypic variation in male external genitalia. Molecular Phylogenetics and

Evolution 42:637-652.

MCKENNA, M. C., and S. K. BELL. 1997. Classification of mammals above the species

level. Columbia University Press, New York.

MILLER, J. R., and M. D. ENGSTROM. 2008. The relationships of major lineages within

Peromyscine rodents: A molecular phylogenetic hypothesis and systematic

reappraisal. Journal of Mammalogy 89:1279-1295.

MUSSER, G. G., and M. D. CARLETON. 2005. Superfamily Muroidea. Pp. 501-775 in

Mammal species of the world: a taxonomic and geographic reference (D. E.

Wilson and D. M. Reeder, eds.). John Hopkins University Press, Baltimore,

Maryland.

NYLANDER, J. A. A. 2004. MrModeltest v2. Evolutionary Biology Centre, Uppsala

University, Uppsala, Sweden.

OSGOOD, W. H. 1909. Revisions of the mice of the American genus Peromyscus.

North American Fauna 28:1-285.

Texas Tech University, Megan S. Keith, December 2015

PEPPERS, L. L., and R. D. BRADLEY. 2000. Cryptic species in Sigmodon hispidus:

evidence from DNA seqeunces. Journal of Mammalogy 81:332-343.

REEDER, S. A., and R. D. BRADLEY. 2004. Molecular systematics of Neotomine-

Peromyscine rodents based on the Dentin Matrix Protein 1 gene. Journal of

Mammalogy 85:1194-1200.

REEDER, S. A., and R. D. BRADLEY. 2007. Phylogenetic relationships of Neotomine-

Peromyscine rodents using DNA sequences from Beta Fibrinogen and

Cytochrome b. Pp. 883-900 in The quintessential naturalist: honoring the life

and legacy of Oliver P. Pearson (D. A. Kelt, E. P. Lessa, J. A. Salazar-Bravo,

and J. L. Patton, eds.), University of California, Publications in Zoology,

Berkeley, California.

REEDER, S. A., D. S. CARROLL, C. W. EDWARDS, C. W. KILPATRICK, and R. D.

BRADLEY. 2006. Neotomine–Peromyscine rodent systematics based on

combined analyses of nuclear and mitochondrial DNA sequences. Molecular

Phylogenetics and Evolution 40:251-258.

RENNERT, P. D., and C. W. KILPATRICK. 1986. Biochemical systematics of

populations of Peromyscus boylii. I. Populations from east-central Mexico

with low fundamental numbers. Journal of Mammalogy 67:481-488.

RENNERT, P. D., and C. W. KILPATRICK. 1987. Biochemical systematics of populations

of Peromyscus boylii. II. Chromosomally variable populations from eastern

and southern Mexico. Journal of Mammalogy 68:799-811.

Texas Tech University, Megan S. Keith, December 2015

RIDDLE, B. R., D. J. HAFNER, and L. F. ALEXANDER. 2000. Phylogeography and

systematics of the Peromyscus eremicus species group and the historical

biogeography of North American warm regional deserts. Molecular

Phylogenetics and Evolution 17:145-160.

ROGERS, D. S., M. D. ENGSTROM, and E. ARELLANO. 2005. Phylogenetic relationships

among Peromyscine rodents: allozyme evidence. Pp. 427-440 in

Contribuciones mastozoológicas en Homenaje a Bernardo Villa (V. Sanchez-

Cordero and R. A. Medellin, eds.). Instituto de Biología e Instituto de

Ecología, Universidad Nacional Autónoma de México, México.

RONQUIST, F., et al. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and

model choice across a large model space. Systematic Biology 61:539-542.

GENE CODES CORPORATION, 2013, Sequencher® version 5.0 sequence analysis

software, Gene Codes Corporation, Ann Arbor Michigan.

http://www.genecodes.com

SHIMODAIRA, H. 2002. An approximately unbiased test of phylogenetic tree selection.

Systematic Biology 51:492-508.

SHIMODAIRA, H., and M. HASEGAWA. 2001. CONSEL: for assessing the confidence of

phylogenetic tree selection. Bioinformatics 17:1246-1247.

SMITH, M., and J. PATTON. 1999. Phylogenetic relationships and the radiation of

Sigmodontine rodents in South America: evidence from Cytochrome b. Journal

of Mammalian Evolution 6:89-128.

Texas Tech University, Megan S. Keith, December 2015

SMITH, M. F., and J. L. PATTON. 1993. The diversification of South American murid

rodents: evidence from mitochondrial DNA sequence data for the Akodontine

tribe. Biological Journal of the Linnean Society 50:149-177.

STAMATAKIS, A., T. LUDWIG, and H. MEIER. 2005. RAxML-III: a fast program for

maximum likelihood-based inference of large phylogenetic trees.

Bioinformatics 21:456-463.

STANHOPE, M. J., J. CZELUSNIAK, J.-S. SI, J. NICKERSON, and M. GOODMAN. 1992. A

molecular perspective on mammalian evolution from the gene encoding

interphotoreceptor retinoid binding protein, with convincing evidence for bat

monophyly. Molecular Phylogenetics and Evolution 1:148-160.

SUCHARD, M. A., R. E. WEISS, and J. S. SINSHEIMER. 2001. Bayesian selection of

continuous-time Markov chain evolutionary models. Molecular Biology and

Evolution 18:1001-1013.

SULLIVAN, J., and C. W. KILPATRICK. 1991. Biochemical systematics of the

Peromyscus aztecus assemblage. Journal of Mammalogy 72:681-691.

SULLIVAN, J., J. A. MARKERT, and C. W. KILPATRICK. 1997. Phylogeography and

molecular systematics of the Peromyscus aztecus species group (Rodentia:

Muridae) inferred using parsimony and likelihood. Systematic Biology 46:426-

440.

SULLIVAN, J. M., C. W. KILPATRICK, and P. D. RENNERT. 1991. Biochemical

systematics of the Peromyscus boylii species group. Journal of Mammalogy

72:669-680.

Texas Tech University, Megan S. Keith, December 2015

TAMURA, K., D. PETERSON, N. PETERSON, G. STECHER, M. NEI, and S. KUMAR. 2011.

MEGA5: molecular evolutionary genetics analysis using maximum likelihood,

evolutionary distance, and maximum parsimony methods. Molecular Biology

and Evolution 28:2731-2739.

TIEMANN-BOEGE, I., C. W. KILPATRICK, D. J. SCHMIDLY, and R. D. BRADLEY. 2000.

Molecular phylogenetics of the Peromyscus boylii species group (Rodentia:

Muridae) based on mitochondrial Cytochrome b sequences. Molecular

Phylogenetics and Evolution 16:366-378.

TOWNSEND, T. M., D. G. MULCAHY, B. P. NOONAN, J. W. SITES, JR., C. A. KUCZYNSKI,

J. J. WIENS, AND T. W. REEDER. 2011. Phylogeny of Iguanian lizards inferred

from 29 nuclear loci, and a comparison of concatenated and species-tree

approaches for an ancient, rapid radiation. Molecular Phylogenetics and

Evolution 61:363-380.

WEKSLER, M. 2003. Phylogeny of Neotropical Oryzomyine rodents (Muridae:

Sigmodontinae) based on the nuclear IRBP exon. Molecular Phylogenetics and

Evolution 29:331-349.

WICKLIFFE, J. K., F. G. HOFFMANN, D. S. CARROLL, Y. DUNINIA-BARKOVSKAYA, and

R. D. BRADLEY. 2003. PCR and sequencing primers for intron 7 (Fgb-I7) of the

fibrinogen, B beta polypeptide (Fgb) in mammals: a novel nuclear DNA

phylogenetic marker. Occasional Papers, Museum of Texas Tech University

219:1-6.

Texas Tech University, Megan S. Keith, December 2015

Table 2.1.—Specimens examined in this study are listed by taxon and genetic marker (Adh1-I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb-I7 - intron of the beta-fibrinogen, and Rbp3 - interphotoreceptor retinoid binding protein) and grouped by tribe, genus, and species group. GenBank accession (left of slash) and museum catalog (right of slash) numbers are given for each specimen. Museum acronyms are as follows: ASNHC (Angelo State Natural History Collection), BYU (Brigham Young University), CNMA (Colección Nacional de Mamíferos, Universidad Nacional Autónoma de México), PGSC (Peromyscus Genetic Stock Center), ROM (Royal Ontario Museum), TCWC (Texas Cooperative Wildlife Collection), and TTU (Museum of Texas Tech University). If museum catalog numbers were unavailable, specimens were referenced with the corresponding collector’s numbers or TK (special number of the Museum of Texas Tech University). ------Taxon Adh1-I2 Cytb Fgb-I7 Rbp3 ------Tribe Baiomyini Baiomys B. taylori AY994205/TTU82642 AF548469/TTU54633 AY274213/TTU54633 EF989838/ROM114886

Tribe Ochrotomyini Ochrotomys O. nuttalli JX910114/TCWC31929 AY195798/TCWC31929 AY274203/TCWC31929 EF989862/ROM113008

Tribe Neotomini Neotoma N. mexicana AY817646/TTU79129 AF294345/TTU79129 AY274200/TTU79129 JX910120/TTU79129

Tribe Reithrodontomyini Habromys H. ixtlani AY994239/TK93160 DQ000482/TK93160 FJ214701/TTU82703 EF989842/CNMA29849

Table 2.1 (Continued) 47

Texas Tech University, Megan S. Keith, December 2015

------Taxon Adh1-I2 Cytb Fgb-I7 Rbp3 ------Isthmomys I. pirrensis FJ214668/TTU39162 FJ214681/TTU39162 FJ214692/TTU39162 EF989846/ROM116308 Neotomodon N. alstoni AY994210/TK45309 AY195796/TK45302 AY274202/TK45309 EF989851/ASNHC1595 Megadontomys M. thomasi AY994208/TK93388 AY195795/TK93388 FJ214693/TK93388 EF989849/CNMA29186 Onychomys O. arenicola JX910115/TTU67559 AY195793/TTU67559 AY274204/TTU67559 EF989855/ROM114904 Osgoodomys O. banderanus AY994209/TK45952 DQ000473/TK45952 FJ214694/TK45952 EF989857/ASNHC2664 Peromyscus aztecus group P. evides FJ214670/TTU82696 FJ214685/TTU82696 FJ214700/TTU82696 JX910121/TTU82696 P. spicilegus AY994234/TK45255 FJ214669/TK45255 FJ214695/TK45255 JX910122/TK47888 boylii group P. boylii AY994227/TTU82688 AF155388/TTU81702 AY274208/TTU81702 EF989871/ASNHC3449 P. levipes AY994224/TK47819 DQ000477/TK47819 FJ214707/TTU105150 JX910123/TK47819 californicus group P. californicus AY994211/TTU83292 AF155393/TTU81275 FJ214697/TTU83291 EF989873/PGSCIS1590 crinitus group P. crinitus AY994213/DSR6171 FJ214684/TK119629 FJ214698/TK119629 EF989874/BYU16629

Table 2.1 (Continued) ------

Texas Tech University, Megan S. Keith, December 2015

Taxon Adh1-I2 Cytb Fgb-I7 Rbp3 ------eremicus group P. eremicus AY994212/TTU81850 AY322503/TTU83249 FJ214699/TTU83249 EF989876/BYU17952 furvus group P. furvus JX910116/FXG1168 AF271032/CNMA32298 JX910113/FXG1168 JX910124/FXG1168 hooperi group P. hooperi FJ214672/TTU104425 DQ973103/TTU104425 FJ214704/TTU104425 JX910125/TTU104425 leucopus group P. gossypinus FJ214671/TTU80682 DQ973102/TTU80682 FJ214702/TTU80682 JX910126/TTU80682 P. leucopus AY994240/TTU75694 DQ000483/TTU101645 FJ214706/TTU101645 EF989880/ROM101861 maniculatus group P. maniculatus AY994242/TTU97830 DQ000484/TTU38739 FJ214708/TTU97830 EF989884/ROM98941 P. melanotis FJ214673/TK70997 AF155398/TK70997 FJ214711/TK70997 EF989891/PGSC25 megalops group P. megalops AY994217/TTU82712 DQ000475/TTU82712 FJ214709/TTU82712 JX910127/TTU82712 melanophrys group P. melanophrys AY994216/TTU75509 AY322510/TTU75509 FJ214710/TTU75509 EF989890/PGSCXZ1073 mexicanus group P. mexicanus AY994236/TTU97013 JX910118/TTU105005 AY274210/TTU82759 EF989895/ROM113250 P. nudipes AY994238/TTU96972 FJ214687/ TTU96972 FJ214713/TTU96972 EF989893/ROM113216 truei group P. attwateri AY994220/TTU55688 AF155384/TTU55688 AY274207/TTU55688 JX910128/TTU55688 P. gratus AY994218/TK46354 AY376421/TK46354 FJ214703/TK46354 JX910129/TK46354 Table 2.1 (Continued) ------Taxon Adh1-I2 Cytb Fgb-I7 Rbp3 49

Texas Tech University, Megan S. Keith, December 2015

------unknown group P. ochraventer FJ214676/TTU104930 JX910119/TTU104930 FJ214715/TTU104930 JX910130/TTU104930 P. pectoralis AY994221/TK48645 AY376427/TK48642 FJ214716/TK48645 JX910131/TK48645 Podomys P. floridanus AY994214/TTU97867 DQ973109/TTU97867 FJ214723/TTU97867 EF989878/TTU97866 Reithrodontomys R. fulvescens AY994207/TTU54898 AF176257/TTU54898 AY274211/TTU54898 EF989901/ASNHC3465 R. sumichrasti JX910117/TTU54952 AF176256/TTU54952 AY274212/TTU54952 EF989924/ROM98383 R. megalotis AF176248/ TTU40942 AF176248/ TTU40942 KJ697789/ TTU40942 EF989909/ASNHC2133 R. mexicanus KJ697791/TTU85234 AY859453/ROM101508 - EF989911/ROM98468 ------

Texas Tech University, Megan S. Keith, December 2015

Table 2.2. —Estimated genetic distances (K2P—Kimura 1980) for selected taxonomic groups based on sequences from the 4 genetic markers (Adh1-I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb-I7 -intron of the beta-fibrinogen, and Rbp3 - interphotoreceptor retinoid binding protein) examined in this study.

Taxon Adh1-I2 Cytb Fgb-I7 Rbp3 Reithrodontomys & Onychomys versus all other groups 8.6/13.07 18.0/19.7 7.5/7.5 4.3/2.7 Isthmomys versus all other groups 14.0 16.5 8.9 3.6 Habromys versus other "genera" 5.2 15.0 6.0 2.0 Megadontomys versus other "genera" 5.2 15.0 6.7 2.9 Neotomodon versus other "genera" 5.3 14.7 6.1 2.6 Osgoodomys versus other "genera" 4.6 15.8 6.0 2.6 Podomys versus other "genera" 6.2 16.3 6.6 2.8 Habromys versus Peromyscus (sensu stricto) 4.0 14.0 5.2 1.3 Megadontomys versus Peromyscus (sensu stricto) 4.0 14.4 5.7 2.2 Neotomodon versus Peromyscus (sensu stricto) 3.19 13.9 5.2 2.0 Osgoodomys versus Peromyscus (sensu stricto) 3.5 15.1 5.0 2.0 Podomys versus Peromyscus (sensu stricto) 5.0 15.7 5.7 2.2 All species groups versus each other 4.4 14.2 5.1 1.6 aztecus species group versus other species groups 4.7 14.3 5.0 1.2

Texas Tech University, Megan S. Keith, December 2015

Table 2.2 (Continued) boylii species group versus other species groups 3.5 13.5 3.7 1.2 californicus species group versus other species groups 4.3 15.0 6.8 2.7 crinitus species group versus other species groups 4.4 14.2 5.8 1.8 eremicus species group versus other species groups 5.4 14.7 6.8 1.7 furvus species group versus other species groups 3.3 12.7 4.4 1.4 hooperi species group versus other species groups 5.9 12.7 4.4 1.4 leucopus species group versus other species groups 4.4 14.3 5.9 2.1 maniculatus species group versus other species groups 5.9 14.0 6.3 2.0 megalops species group versus other species groups 3.6 13.9 4.7 1.3 melanophrys species group versus other species groups 3.8 13.7 3.8 1.8 mexicanus species group versus other species groups 4.2 15.3 4.0 1.2 truei species group versus other species groups 3.9 14.2 4.5 1.3

Texas Tech University, Megan S. Keith, December 2015

Table 2.3. —Three potential taxonomic solutions for Habromys, Isthmomys, Megadontomys, Neotomodon, Osgoodomys, Peromyscus and Podomys. Generic designations were identified by supported monophyletic clades within Fig. 1. Only species included in phylogenetic analyses are presented. Generic taxonomy Subgeneric taxonomy Species group taxonomy Genus Isthmomys Genus Isthmomys Genus Isthmomys I. pirrensis I. pirrensis I. pirrensis Genus Habromys Genus Peromyscus Genus Peromyscus H. ixtlani Subgenus Habromys Species group lepturus Genus Megadontomys H. ixtlani H. ixtlani M. thomasi Subgenus Megadontomys Species group thomasi Genus Neotomodon M. thomasi M. thomasi N. alstoni Subgenus Neotomodon Species group alstoni Genus Osgoodomys N. alstoni N. alstoni O. banderanus Subgenus Osgoodomys Species group banderanus Genus Podomys O. banderanus O. banderanus P. floridanus Subgenus Podomys Species group floridanus Genus Haplomylomys P. floridanus P. floridanus P. californicus Subgenus Haplomylomys Species group californicus P. crinitus P. californicus P. californicus P. eremicus P. crinitus Species group crinitus Genus Peromyscus P. eremicus P. crinitus P. gossypinus Subgenus Peromyscus Species group eremicus P. leucopus P. gossypinus P. eremicus P. maniculatus P. leucopus Species group aztecus P. melanotis P. maniculatus P. evides New Genus A P. melanotis P. spicilegus Table 2.3 (Continued) New Subgenus A Species group boylii

Texas Tech University, Megan S. Keith, December 2015

P. megalops P. melanophrys P. megalops P. boylii P. mexicanus P. melanophrys P. levipes P. nudipes P. mexicanus Species group furvus New Genus B P. nudipes P. furvus P. boylii New Subgenus B Species group hooperi P. evides P. boylii P. hooperi P. levipes P. evides Species group leucopus P. spicilegus P. levipes P. gossypinus New Genus C P. spicilegus P. leucopus P. attwateri New Subgenus C Species group maniculatus P. furvus P. attwateri P. maniculatus P. gratus P. furvus P. melanotis P. ochraventer P. gratus Species group megalops New Genus D P. ochraventer P. megalops P. pectoralis New Subgenus D Species group melanophrys New Genus E P. pectoralis P. melanophrys P. hooperi New Subgenus E Species group mexicanus P. hooperi P. mexicanus P. nudipes Species group truei P. attwateri P. gratus P. ochraventer Species group pectoralis P. pectoralis

Texas Tech University, Megan S. Keith, December 2015

Table 2.4. —Detailed taxonomy of Peromyscus based on subsumation of Habromys, Megadontomys, Neotomodon, Osgoodomys, and Podomys to species groups; referred to as the species group taxonomy. Composition of species recognized in each species group was obtained from Musser and Carlton (2005) and other references (indicated in parentheses). ------Taxon ------Genus Isthmomys

Genus Reithrodontomys

Genus Onychomys

Genus Peromyscus

alstoni species group - alstoni

aztecus species group - aztecus, evides, hylocetes, oaxacensis (Sullivan et al. 1997), spicilegus, and winklemanni

banderanus species group - banderanus

boylii species group - beatae, boylii (Kilpatrick; Bradley et al. 2014), levipes, madrensis, schmidlyi (Bradley et al. 2004), simulus, and stephani

californicus species group - californicus, caniceps, and dickeyi

crinitus species group - crinitus

eremicus species group - eremicus, eva, fraterculus, guardia, interparietalis, merriami, pembertoni, and psuedocrinitus

floridanus species group - floridanus

furvus species group - furvus

hooperi species group - hooperi

lepturus species group - chinanteco, delicatulus, ixtlani, lepturus, lophurus, sagax = delicatulus (Rogers et al. 2007), and simulatus

leucopus species group - gossypinus and leucopus

Texas Tech University, Megan S. Keith, December 2015

Table 2.4 (Continued) ------Taxon ------

maniculatus species group - keeni, maniculatus, melanotis, polionotus, and sejugis

megalops species group - megalops and melanurus

melanophrys species group - mekisturus, melanophrys, and perfulvus

mexicanus species group - grandis, guatemalensis, gymnotis, mayensis, mexicanus, nudipes, stirtoni, yucatanicus, and zarhynchus

thomasi species group - cryophilus, nelsoni, and thomasi

truei species group - attwateri, bullatus, difficilis, gratus, nasutus, ochraventer, and truei

Incertae sedis - melanocarpus, pectoralis, polius, and slevini ------

Texas Tech University, Megan S. Keith, December 2015

Figure 2.1.—Phylogenetic tree obtained from maximum likelihood analysis of the a) combined mitochondrial cytochrome-b gene (Cytb) and 3 nuclear genes (alcohol dehydrogenase - Adh1-I2, beta fibrinogen - Fgb-I7, and interphotoreceptor retinoid- binding protein - Rbp3), as well as the b) mitochondrial (Cytb) phylogenetic tree, and c) combined nuclear (Adh1-I2, Fgb-I7, and Rbp3) phylogenetic tree. Taxonomic groups of interest are designated as follows: Pss (Peromyscus [sensu stricto]), Psl (Peromyscus [sensu lato]), Rei (Reithrodontomyini), and Bai (Baiomyini). Nodal support values are superimposed on the maximum likelihood tree topology. Support values are as follows: 10,000 bootstrap replicates of the maximum likelihood analysis (below branch) and clade probability values for the Bayesian inference analysis (above branch). Statistically significant clade probability values (≥ 0.95) are designated with an asterisk (*). All bootstrap support values ≥ 50 are shown. For members of Peromyscus (sensu stricto) only, species epithets are given. Peromyscus (sensu lato) affiliated genera are indicated in bold. Major nodes are indicated with roman numerals. 57

Texas Tech University, Megan S. Keith, December 2015

Figure 2.2—Maximum clade credibility tree showing divergence date estimates based on a combined analysis of the mitochondrial cytochrome-b gene (Cytb) and 3 nuclear genes (alcohol dehydrogenase - Adh1-I2, beta fibrinogen - Fgb-I7, and interphotoreceptor retinoid-binding protein - Rbp3). Divergence date estimates are indicated along the x-axis in millions of years. Error bars represent the 95% highest posterior density (HPD) for node height. Peromyscus (sensu lato) affiliated genera are indicated in bold.

Texas Tech University, Megan S. Keith, December 2015

CHAPTER III

MOLECULAR DATA INDICATE THAT ISTHMOMYS IS NOT ALIGNED WITH PEROMYSCUS

ABSTRACT The genus Isthmomys is composed of two species, Isthmomys flavidus and I. pirrensis. These rodents have been classified in either their own genus or as a subgenus of Peromyscus; however, recent molecular studies have alluded to a possible sister relationship with Reithrodontomys. The systematic relationships of this group have been difficult to resolve due to a vague understanding of the evolutionary history of the genus, absence of I. flavidus in phylogenetic studies, and the rarity of preserved tissue samples in museum collections. In this study, maximum likelihood analyses were used to ascertain the phylogenetic relationship of Isthmomys to Peromyscus and

Reithrodontomys using three approaches. First, a large-scale analysis of the mitochondrial cytochrome-b gene (Cytb) was conducted for Isthmomys pirrensis, and all species of Reithrodontomys, Peromyscus and selected allied taxa available on

GenBank; second, topological constraints were applied to a reduced Cytb dataset to reflect two competing hypotheses for the phylogenetic placement of Isthmomys: 1)

Isthmomys is sister to Peromyscus or 2) Isthmomys is sister to Reithrodontomys; and third, five nuclear datasets (Adh1-I2, Fgb-I7, Dmp1, Ghr, and Rbp3) were analyzed individually and then combined with Cytb data. The large-scale analysis of the mitochondrial Cytb gene indicated that Isthmomys was sister to Reithrodontomys; however, this relationship was not statistically supported. Analyses of the constrained vs. unconstrained phylogenies of the reduced Cytb dataset did not identify a 59

Texas Tech University, Megan S. Keith, December 2015 significantly better topology. Analysis of the individual nuclear datasets resulted in one topology (Dmp1) reflecting a sister relationship between Isthmomys and

Peromyscus, two topologies (Ghr and Rbp3) that supported the hypothesis that

Isthmomys is sister to Reithrodontomys, and Adh1-I2 and Fbg-I7 datasets recovered phylogenies that warranted consideration of a third, alternative hypothesis suggesting that Isthmomys was basal to Peromyscus and Reithrodontomys; although none of these relationships were statistically supported. However, the combined analysis

(mitochondrial gene plus five nuclear genes) provided statistical support for a sister- taxa relationship between Isthmomys and Reithrodontomys.

INTRODUCTION

The genus Isthmomys (family Cricetidae, subfamily Neotominae; Hooper and

Musser 1964a) consists of two species, the Yellow Isthmus rat (Isthmomys flavidus;

Bangs 1902) and the Mt. Pirri Isthmus rat (Isthmomys pirrensis; Goldman 1912). The phylogenetic relationships of these relict species have been difficult to resolve, in part, due to the rarity of voucher specimens and tissues in natural history collections.

Previous studies have been unable to consistently place Isthmomys relative to

Peromyscus and Reithrodontomys with some datasets depicting a sister relationship with Reithrodontomys and others indicating that Isthmomys is aligned with some species of peromyscine rodents and should be considered a subgenus of Peromyscus

(Hooper and Musser 1964a). Further complicating matters, a paucity of genetic data has been generated for Isthmomys pirrensis and no genetic data is available for

Isthmomys flavidus. Questions concerning these taxa involve whether Isthmomys

Texas Tech University, Megan S. Keith, December 2015 should be considered a distinct genus (Carleton 1980, 1989; Musser and Carleton

2005) or subgenus (Hooper and Musser 1964a; Hooper 1968) of Peromyscus and how a taxonomic status change could affect the phylogeny and classification of

Peromyscus (Platt et al. 2015).

Various datasets support alternative phylogenetic relationships for Isthmomys.

Carleton’s (1980) morphological study of Peromyscus is the most inclusive study of morphology for this group of rodents. Carleton (1980) suggested that Isthmomys formed a sister-group relationship with Megadontomys, whereas other morphological studies placed it within the peromyscine rodents (Osgood 1909; Hooper and Musser

1964b; Hooper 1968). Further, Isthmomys lacks the grooved incisors characteristic of

Reithrodontomys and is considerably larger than even the largest species of harvest mouse (Miller and Engstrom 2008; Reid 2009). This observation indicates a plausible morphological affiliation with Peromyscus to the exclusion of Reithrodontomys.

Alternatively, the karyotype of I. pirrensis consists of characters signifying its uniqueness relative to other species of Peromyscus. Stangl and Baker (1984) compared the karyotype of I. pirrensis to the inferred primitive karyotype for

Peromyscus and provided support for the hypothesis that Isthmomys may have been the first lineage to have diverged from Peromyscus.

The allozyme study of Rogers et al. (2005) was the first molecular study to recover a sister relationship between Isthmomys and Reithrodontomys. In their study, these genera formed a clade basal to Peromyscus (sensu lato- including Habromys,

Megadontomys, Neotomodon, Osgoodomys, and Podomys). More recently, Platt et al.

Texas Tech University, Megan S. Keith, December 2015

(2015) and other authors (Bradley et al. 2007; Miller and Engstrom 2008) examined molecular variation in DNA sequences for the peromyscine rodents and proposed additional support for a sister relationship between Isthmomys and Reithrodontomys.

In addition to recovering this phylogenetic relationship, Platt et al. (2015) reported that

Isthmomys diverged from Peromyscus approximately 8 million years ago (MYA), whereas other peromyscine rodents originated between 3.5 to 5.5 MYA.

Based on previous studies, three hypotheses have been proposed to characterize the relationship of Isthmomys to other members of the Neotominae: 1)

Isthmomys is more closely related to Peromyscus (either monophyletic with

Peromyscus-Osgood 1909; Hooper and Musser 1964a; Hooper 1968, or it is the first lineage to diverge from Peromyscus-Stangl and Baker 1984), 2) Isthmomys forms a sister-group relationship with Reithrodontomys and is basal to Peromyscus (Rogers et al. 2005; Bradley et al. 2007; Miller and Engstrom 2008; Platt et al. 2015), or 3)

Isthmomys is basal to a clade containing Peromyscus and Reithrodontomys. This study utilizes extensive phylogenetic analysis of multiple genetic datasets to test the aforementioned hypotheses. To accomplish this, the mitochondrial cytochrome-b gene (Cytb) and five individual nuclear datasets (intron 2 of the alcohol dehydrogenase gene (Adh1-I2), intron 7 of the beta-fibrinogen gene (Fgb-I7), exon 6 of the dentin matrix protein 1 gene (Dmp1), exon 10 of the growth hormone receptor gene (Ghr), and exon 1 of the interphotoreceptor retinoid-binding protein gene

(Rbp3)) were examined in a phylogenetic context for several species of peromyscine rodents available on GenBank.

Texas Tech University, Megan S. Keith, December 2015

MATERIALS AND METHODS Sequences were either generated in this study or obtained from GenBank. Six genes, which have been informative in other studies involving this group of rodents

(Reeder et al. 2006; Miller and Engstrom 2008; Platt et al. 2015), were examined to resolve the phylogenetic relationships of the genus Isthmomys: Adh1-I2, Cytb, Fgb-I7,

Dmp1, Ghr, and Rbp3. For the majority of samples, genetic data were downloaded from GenBank's Nucleotide database using the following search terms "Peromyscus

OR Reithrodontomys OR Isthmomys OR Onychomys OR Habromys OR

Megadontomys OR Neotomodon OR Osgoodomys OR Podomys OR Baiomys OR

Ochrotomys " combined with the gene symbol for each gene (ex."AND Cytb", "AND

Adh1-I2", etc.). Sequences (either obtained from GenBank or generated in this study) for datasets that were utilized in both the individual and combined analyses are listed in Table 3.1 with GenBank accession numbers and museum catalog numbers. The purpose of these searches was to recover all available data for genera/species historically affiliated with Peromyscus, Reithrodontomys, and Isthmomys from each gene of interest as well as other selected genera within the Neotominae.

Sequencing.—Genomic DNA was isolated from approximately 0.1g of frozen liver tissue using either the Puregene DNA isolation kit (Gentra, Minneapolis,

Minnesota) or the Qiagen DNeasy Blood and Tissue Kit (Qiagen, Valencia,

California). DNA fragments were amplified using polymerase chain reaction (PCR,

Saiki et al. 1988). Methods for PCR amplification and sequencing of Cytb followed protocols developed by Bradley et al. (2007) and Tiemann-Boege et al. (2000) using primers MVZ05 (Smith and Patton 1993), CB40 (Hanson and Bradley 2008) and 63

Texas Tech University, Megan S. Keith, December 2015

Pero3' (Tiemann-Boege et al. 2000) or LGL765 forward and LGL766 reverse

(Bickham et al. 1995); Adh1-I2 followed the methods of Amman et al. (2006) using primers 2340-I, 2340-II, Exon II-F, and Exon III-R (Amman et al. 2006); Fgb-I7 followed Carroll and Bradley (2005) and Wickliffe et al. (2003) using primers Fgb-

17U-Rattus, Fgb-17L-Rattus (Wickliffe et al. 2003) and B17-mammU and B17- mammL (Matocq et al. 2007); Dmp1 followed the methods of Reeder and Bradley

(2004) using primers Den-12 (F) and Den-2 (R) (Toyosawa et al. 1999); Ghr followed

Miller and Engstrom (2008) using primers GHREXON10 and GHREND (Adkins et al. 2001); and amplification of Rbp3 followed the protocols of Chambers et al. (2009) and Jansa and Voss (2000) using primers A and B (Stanhope et al. 1992).

PCR products were purified using the QIAquick PCR purification kit (Qiagen,

Valencia, California) or ExoSAP-IT (USB Products, Cleveland, Ohio) and PCR amplicons were sequenced using ABI Prism Big Dye Terminator v3.1 ready reaction mix (Applied Biosystems, Foster City, California). Nucleotide sequences were determined on an ABI 3100 or 3130-Avant automated sequencer (Applied Biosystems,

Foster City, California) with the following internal primers in addition to those used for PCR: Cytb primers 400F (Edwards et al. 2001) and 870R (Peppers et al. 2002);

Adh1-I2 primers Adh350F and Adh410R (Amman et al. 2006); Fgb-I7 primer

Bfib300F (Carroll and Bradley 2005); Dmp1 primers 900F and 900R (Reeder and

Bradley 2004); Ghr primers GHR7, GHR8, GHR9 and GHR10 (Adkins et al. 2001); and Rbp3 primers D (Stanhope et al. 1992), E2 (Weksler 2003) and 125F (DeBry and

Sagel 2001). Sequences were proofed using Sequencher 4.10 software (Gene Codes,

Texas Tech University, Megan S. Keith, December 2015

Ann Arbor, Michigan) and aligned using MEGA v6.06 (Tamura et al. 2013).

Sequences generated in this study were deposited in GenBank and are depicted in

Table 3.1.

Data Analyses.—The phylogenetic position of Isthmomys relative to

Peromyscus and Reithrodontomys was tested using multiple phylogenetic approaches.

For all datasets, sequences were aligned using MUSCLE (Edgar 2004) and jModelTest v2.1.6 (Darriba et al. 2012) was used to determine the appropriate nucleotide substitution model. Nucleotide substitution models were determined as follows: GTR+I+G for Cytb, Fgb-I7, Dmp1, and Rbp3, and HKY+I+G for Adh1-I2 and Ghr. A maximum likelihood (ML) analysis and a total of 1,000 bootstrap replicates were used to analyze the sequence data in RaxML (Stamatakis et al. 2005).

All datasets were evaluated under the GTR+I+G model (option GTRGAMMAI) as the less parameter rich HKY model was not available in the software package. Sigmodon hispidus was selected as the outgroup taxon for all analyses based on previous studies that have examined this group of rodents (Bradley et al. 2004; Reeder and Bradley

2004; Bradley et al. 2007; Miller and Engstrom 2008).

First, a large-scale analysis was conducted for the mitochondrial Cytb gene.

Sequences that were not complete were removed from the original compiled dataset; only complete sequences (≥1,143 bp) were used when possible with the exception of

Isthmomys, resulting in a dataset containing 1,063 sequences. Next, the Cytb dataset was reduced to a maximum of five sequences per species (284 total sequences). The reduced Cytb dataset was examined to test the aforementioned competing hypotheses

Texas Tech University, Megan S. Keith, December 2015 for the relationship of Isthmomys to Reithrodontomys and Peromyscus. An unconstrained analysis was performed (unconstrained = Reithrodontomys, Isthmomys, and Peromyscus) and subsequently the topology was constrained so that Isthmomys would be monophyletic with Peromyscus (Peromyscus + Isthmomys).

The approximately unbiased (AU) test (Shimodaira 2002) was conducted in

CONSEL (Shimodaira and Hasegawa 2001) to assess the confidence of the tree selection by calculating p-values for the resulting phylogenies. The AU test is the primary result of CONSEL, which uses a multiscale bootstrap technique and provides less biased results (Shimodaira and Hasegawa 2001).

To further test the placement of Isthmomys relative to Peromyscus and

Reithrodontomys; five nuclear genes were analyzed. The relationships of taxa were scored for each gene based on nodal support. Individual nuclear analyses included 33 sequences for Adh1-I2 (1- I. pirrensis, 2-Reithrodontomys, 21- Peromyscus, 8- other neotomine-peromyscine taxa, and 1- outgroup), 35 sequences for Fgb-I7 (1- I. pirrensis, 4-Reithrodontomys, 21- Peromyscus, 8- other neotomine-peromyscine taxa, and 1- outgroup), 32 sequences for Dmp1 (1- I. pirrensis, 3- Reithrodontomys, 20-

Peromyscus, 7- other neotomine-peromyscine taxa, and 1- outgroup), and 37 sequences for Ghr and Rbp3 (3- I. pirrensis, 4- Reithrodontomys, 21- Peromyscus, 8- other neotomine-peromyscine taxa, and 1- outgroup).

Individual datasets were reduced to a sampling of 32 taxa for which sequence data were available for all genes (mitochondrial and nuclear). These datasets were partitioned and analyzed in a combined analysis using the same parameters as above.

Texas Tech University, Megan S. Keith, December 2015

Individual gene trees can differ significantly from the underlying species tree due to incomplete lineage sorting and other processes (Maddison 1997; Kubatko and Degnan

2006). The use of a combined dataset can often overcome the effects of incomplete lineage sorting (Rokas et al. 2003) and help to increase resolution at multiple levels within a phylogeny with quickly evolving mitochondrial markers providing more resolution at terminal nodes and nuclear markers providing higher resolution at the base of a phylogeny (Bull et al. 1993; Adkins et al. 2001; Pereira et al. 2002; Platt et al. 2015). In this analysis, there was one representative for each species and only taxa for which there were no missing DNA sequences were included. An attempt was made to obtain mitochondrial and nuclear sequences from a single individual, but was not possible for all nuclear genes. In these cases, sequences from conspecific individuals were used to complete the dataset. Concatenation of sequence data from conspecifics to represent a composite species rather than a single individual has been successfully used in various taxa (Campbell and Lapointe 2009; Haddrath and Baker

2012; Townsend et al. 2011; Platt et al. 2015). This analysis included a total of 5,345 base pairs.

RESULTS The large-scale Cytb analysis (Fig. 3.1) generated a phylogeny in which

Isthmomys was sister to Reithrodontomys (Clade I) and Peromyscus (Clade II) was basal to both of these genera. However, nodal support was not statistically significant for this relationship. The reduced mitochondrial dataset was constrained as before to test hypotheses concerning the phylogenetic placement of Isthmomys. The

Texas Tech University, Megan S. Keith, December 2015 unconstrained analysis resulted in a phylogeny (Fig. 3.2A) in which Isthmomys was sister to Reithrodontomys and the best tree score was -35559.90420. A topological constraint was then applied to the dataset to induce monophyly of Isthmomys and

Peromyscus and the resulting phylogeny (Fig. 3.2B) received a best tree score of -

35576.117581. The AU test (CONSEL- Shimodaira and Hasegawa 2001) indicated that the unconstrained topology (Isthmomys + Reithrodontomys) was not significantly better (p-value = 0.153) than the constrained topology (Isthmomys + Peromyscus), therefore the null hypothesis could not be rejected.

Three different taxonomic relationships were recovered when the five nuclear genes were examined independently (Fig. 3.3). Analyses of Adh1-I2 and Fgb-I7 sequences (Fig. 3.3A and B) each recovered a phylogeny in which Isthmomys was basal to a clade containing Peromyscus and Reithrodontomys, analyses of Ghr and

Rbp3 sequences (Fig. 3.3C and D) both recovered a sister relationship between

Isthmomys and Reithrodontomys, and analysis of Dmp1 sequences (Fig. 3.3E) recovered a phylogeny in which Isthmomys was sister to Peromyscus. However, none of these analyses depicted statistical support for any relationship of Isthmomys to the other genera.

Finally, the combined analysis (Fig. 3.4) resulted in a phylogeny with a higher level of resolution than was recovered for the individual analyses. A strongly supported (bootstrap value = 86), monophyletic clade in which Isthmomys was sister to Reithrodontomys was recovered (Clade I). Clade I was basal to a monophyletic

Texas Tech University, Megan S. Keith, December 2015 clade containing Peromyscus and allied genera (Clade II) with a bootstrap support value of 95.

DISCUSSION The large-scale Cytb analysis (1,063 sequences) conducted herein resulted in a phylogeny that was highly unresolved and indicated a sister relationship between

Isthmomys and Reithrodontomys; however this relationship did not receive statistical support (Fig. 3.1). Therefore, inclusion of a broad sampling scheme beyond that of

Bradley et al. (2007), Miller and Engstrom (2008), and Platt et al. (2015) did not increase resolution for the phylogenetic relationships of Isthmomys. Subsequently, a topological constraint was applied to a reduced Cytb dataset. Topological tests resulted in a p-value = 0.153, therefore despite the unconstrained topology having a less negative value, the likelihood scores of the resulting topologies were not significantly different (Fig. 3.2).

Analysis of five individual nuclear genes resulted in three possible taxonomic arrangements of Isthmomys to Peromyscus and Reithrodontomys (Fig. 3.3): 1)

Isthmomys was basal to both Peromyscus and Reithrodontomys (Adh1-I2 and Fgb-I7;

Fig. 3.3A and B); 2) Isthmomys formed a sister taxa relationship with

Reithrodontomys (Ghr and Rbp3; Fig. 3.3C and D, Rogers et al. 2005; Bradley et al.

2007; Miller and Engstrom 2008; Platt et al. 2015); or 3) Isthmomys was sister to

Peromyscus (sensu lato- including Habromys, Megadontomys, Neotomodon, and

Podomys -Dmp1; Fig. 3.3E, Stangl and Baker 1984). Two of the recovered taxonomic arrangements (2 and 3) corresponded to existing hypotheses for the relationship of

Texas Tech University, Megan S. Keith, December 2015

Isthmomys to other members of the Neotominae, whereas analyses of Adh1-I2 and

Fgb-I7 recovered a relationship that required a third, alternative hypothesis (i.e.,

Isthmomys is basal to a clade uniting Peromyscus and Reithrodontomys -Table 3.2).

However, none of the individual nuclear gene phylogenies indicated statistical support for any of the recovered phylogenetic relationships of Isthmomys with relation to

Reithrodontomys and Peromyscus (Fig. 3.3).

The combined ML analysis resulted in a strongly supported clade in which

Isthmomys was sister to Reithrodontomys (Fig. 3.4, Clade I). A second clade formed a monophyletic Peromyscus (sensu lato; Fig. 3.4, Clade II). This analysis provided the greatest resolution and nodal support for a phylogenetic relationship for Isthmomys.

Both of these clades were strongly supported with bootstrap values > 85. These results were similar to those of Rogers et al. (2005), Bradley et al. (2007), Miller and

Engstrom (2008), and Platt et al. (2015) and is the only phylogenetic analysis to recover statistical support for any relationship of Isthmomys to Peromyscus or

Reithrodontomys in this study.

Systematic Conclusions.—Analysis of morphology, karyology, and genetic datasets generated different phylogenetic hypotheses concerning the relationship of

Isthmomys to other peromyscine and reithrodontomyine genera. Morphological data indicated a sister relationship for Isthmomys to Peromyscus (Hooper and Musser

1964a; Hooper 1968; Carleton 1980), especially given that Isthmomys lacks the grooved incisors that would act as a synapomorphic character for the genus

Reithrodontomys (Miller and Engstrom 2008). Karyotypic data (Stangl and Baker

Texas Tech University, Megan S. Keith, December 2015

1984) supported the hypothesis that Isthmomys may have been the first lineage to diverge from Peromyscus indicating a basal or sister relationship to Peromyscus.

However, most studies analyzing genetic data have recovered a sister relationship between Isthmomys and Reithrodontomys (Rogers et al. 2005; Bradley et al. 2007;

Miller and Engstrom 2008; Platt et al. 2015).

Increased sampling and analysis of the mitochondrial cytochrome-b gene, as well as individual analyses for five nuclear markers were unable to confidently place

Isthmomys relative to Peromyscus and Reithrodontomys. These results were a classic example of incongruencies between gene trees hindering the ability to estimate a phylogeny. Incongruencies between gene trees could be due to horizontal gene transfer, gene duplication, or incomplete lineage sorting (Avise and Ball 1990;

Maddison 1997) that may have occurred during a rapid radiation event during the late

Miocene that has increased the difficulty of placing some genera in the Neotominae

(Bradley et al. 2004; Reeder and Bradley 2004; Reeder et al. 2006). Studies examining the peromyscine rodents over the last decade have consistently predicted or recognized a sister group relationship between Isthmomys and Reithrodontomys; however, most of these studies did not provide statistical support for this relationship

(Rogers et al. 2005; Bradley et al. 2007) or statistical support was not consistent across all analyses performed in a single study (Miller and Engstrom 2008; this study).

Analysis of a combined dataset appears to be the only situation where support for a sister relationship between Isthmomys and Reithrodontomys is realized (Miller and

Engstrom 2008; Platt et al. 2015; this study). It is likely that phylogenetic signals

Texas Tech University, Megan S. Keith, December 2015 from each dataset were additive and revealed an improved phylogenetic relationship for this group. However, these studies have included a single representative of

Isthmomys and it is unclear if data for additional key taxa (i.e., I. flavidus) will have an impact on the results of the phylogenetic analysis. Sequence data for I. flavidus may be the missing piece that would allow researchers to definitively place Isthmomys within the peromyscine phylogeny.

A monophyletic clade indicating a sister taxa relationship for Isthmomys and

Reithrodontomys has been recovered (Rogers et al. 2005; Bradley et al. 2007; Miller and Engstrom 2008; Platt et al. 2015) in numerous studies examining molecular data for rodents of the Neotominae. Therefore, for consistency and because recent phylogenetic analyses have not provided support for any alternative relationship, it is best to recognize Isthmomys as forming a sister relationship with Reithrodontomys, separate from Peromyscus.

ACKNOWLEDGMENTS The authors would like to thank the Texas Academy of Science and Texas

Tech University Association of Biologists for funding necessary to complete this project. We thank R. J. Baker, H. Garner, and K. McDonald (Natural Science

Research Laboratory, Museum of Texas Tech University) and Centro

Interdisciplinario de Investigación para el Desarrollo Integral Regional, Unidad

Durango, Mexico for assistance in obtaining tissues. We would also like to thank

Nicté Ordóñez-Garza for reading and editing earlier versions of this manuscript.

Texas Tech University, Megan S. Keith, December 2015

LITERATURE CITED ADKINS, R. M., E. L. GELKE, D. ROWE, AND R. L. HONEYCUTT. 2001. Molecular

phylogeny and divergence time estimates for major rodent groups: evidence

from multiple genes. Molecular Biology and Evolution 18:777–791.

AMMAN, B. R., J. D. HANSON, L. K. LONGHOFER, S. R. HOOFER, and R. D. BRADLEY.

2006. Intron 2 of the alcohol dehydrogenase gene (ADH1-I2): a nuclear DNA

marker for mammalian systematics. Occasional Papers, Museum of Texas

Tech University 256:1-16.

AVISE, J. C., AND R. M. BALL. 1990. Principles of genealogical concordance in

species concepts and biological taxonomy. Oxford Surveys in Evolutionary

Biology 7:45- 67.

BANGS, O. 1902. Chiriqui Mammalia. Bulletin of the Museum of Comparative

Zoology at Harvard College, 39: 17-51.

BICKHAM, J. W., C. C. WOOD, AND J. C. PATTON. 1995. Biogeographic Implications

of Cytochrome b Sequences and Allozymes in Sockeye (Oncorhynchus nerka).

Journal of Heredity 86:140-144.

BRADLEY, R. D., C. W. EDWARDS, D. S. CARROLL, and C. W. KILPATRICK. 2004.

Phylogenetic relationships of Neotomine-Peromyscine rodents: based on DNA

sequences from the mitochondrial Cytochrome-b gene. Journal of Mammalogy

85:389-395.

BRADLEY, R. D., N. D. DURISH, D. S. ROGERS, J. R. MILLER, M. D. ENGSTROM, and C.

W. KILPATRICK. 2007. Toward a molecular phylogeny for Peromyscus:

Texas Tech University, Megan S. Keith, December 2015

evidence from mitochondrial Cytochrome-b sequences. Journal of

Mammalogy 88:1146-1159.

BULL, J. J., J. P. HUELSENBECK, C. W. CUNNINGHAM, D. L. SWOFFORD, AND P. J.

WADDELL. 1993. Partitioning and combining data in phylogenetic analysis.

Systematic Biology 42: 384-397.

CAMPBELL, V., and F.-J. LAPOINTE. 2009. The use and validity of composite taxa in

phylogenetic analysis. Systematic Biology 58:560-572.

CARLETON, M. D. 1980. Phylogenetic relationships in neotomine-peromyscine rodents

(Muroidea) and a reappraisal of the dichotomy within New World Cricetinae.

Miscellaneous Publications of the Museum of Zoology 157:146.

CARLETON, M. D. 1989. Systematics and evolution. Pp. 7-141 in Advances in the

Study of Peromyscus (L. KJG and Layne JN eds.), Texas Tech University

Press, Lubbock.

CARROLL, D. S., R. D. BRADLEY. 2005. Systematics of the genus Sigmodon: DNA

sequences from Beta-Fibrinogen and Cytochrome b. The Southwestern

Naturalist 50:342-349.

CHAMBERS, R. R., P. D. SUDMAN, and R. D. BRADLEY. 2009. A phylogenetic

assessment of pocket gophers (Geomys): Evidence from nuclear and

mitochondrial genes. Journal of Mammalogy 90:537-547.

DARRIBA, D., G. L. TABOADA, R. DOALLO, AND D. POSADA. 2012. jModelTest 2:

more models, new heuristics and parallel computing. Nature Methods 9:772.

doi:10.1038/nmeth.2109.

Texas Tech University, Megan S. Keith, December 2015

DEBRY, R. W., AND R. M. SAGEL. 2001. Phylogeny of Rodentia (Mammalia)

inferred from the nuclear-encoded gene Irbp. Molecular Phylogenetics and

Evolution 19:290–301.

EDGAR, R. C. 2004. MUSCLE: multiple sequence alignment with high accuracy and

high throughput. Nucleic Acids Research 32: 1792-1797.

EDWARDS, C. W., C. F. FULHORST, and R. D. BRADLEY. 2001. Molecular

phylogenetics of the Neotoma albigula species group: further evidence of a

paraphyletic assemblage. Journal of Mammalogy 82:267-279.

GOLDMAN, E. A. 1912. New small mammals from eastern Panama. Smithsonian

Miscellaneous Collections 602: 1-18.

HADDRATH, O., and A. J. BAKER. 2012. Multiple nuclear genes and retroposons

support vicariance and dispersal of the palaeognaths, and an Early Cretaceous

origin of modern birds. Proceedings of the Royal Society B: Biological

Sciences.

HANSON, J. D., and R. D. BRADLEY. 2008. Molecular diversity within Melanomys

caliginosus (Rodentia: Oryzomyini): evidence for multiple species. Occasional

Papers, Museum of Texas Tech University 275:1-11.

HOOPER, E. T. 1968. Classification. Pp. 27-74 in Biology of Peromyscus (Rodentia)

(King, J. A. ed.), The American Society of Mammalogists.

HOOPER, E. T. AND G. G. MUSSER. 1964a. Notes on classification of the rodent genus

Peromyscus. Occasional Papers of the Museum of Zoology, University of

Michigan, 635: 1-13.

Texas Tech University, Megan S. Keith, December 2015

HOOPER, E. T., AND G. G. MUSSER. 1964b. The glans penis in Neotropical cricetines

(Family Muridae) with comments on classification of muroid rodents.

Miscellaneous Publications, Museum of Zoology, University of Michigan,

123: 1-57.

HORNSBY, A. D. AND M. D. MATOCQ. 2011. Differential regional response of the

bushy-tailed woodrat (Neotoma cinerea) to late Quaternary climate change.

Journal of Biogeography 39: 289-305.

JANSA, S., and R. VOSS. 2000. Phylogenetic studies on didelphid marsupials I.

Introduction and preliminary results from nuclear IRBP gene sequences.

Journal of Mammalian Evolution 7:43-77.

KUBATKO, L. S. AND J. H. DEGNAN. 2006. Inconsistency of phylogenetic estimates

from concatenated data under coalescence. Systematic Biology 56: 17-24.

MADDISON, W. P. 1997. Gene trees in species trees. Systematic Biology 1997:523-

536.

MATOCQ, M. D., Q. R. SHURTLIFF, and C. R. FELDMAN. 2007. Phylogenetics of the

woodrat genus Neotoma (Rodentia: Muridae): implications for the evolution of

phenotypic variation in male external genitalia. Molecular Phylogenetics and

Evolution 42:637-652.

MILLER, J. R., and M. D. ENGSTROM. 2008. The relationships of major lineages within

Peromyscine rodents: A molecular phylogenetic hypothesis and systematic

reappraisal. Journal of Mammalogy 89:1279-1295.

Texas Tech University, Megan S. Keith, December 2015

MUSSER, G. G. AND M. D. CARLETON. 2005. Superfamily Muroidea. Pp. 894-1531 in

Mammal species of the world: a taxonomic and geographic reference (D. E.

Wilson and D. M. Reeder, Eds.). 3rd Ed. Johns Hopkins University Press,

Baltimore Maryland.

OSGOOD, W. H. 1909. Revisions of the mice of the American genus Peromyscus.

North American Fauna 28:1-285.

PEPPERS, L. L., D. S. CARROLL, AND R. D. BRADLEY. 2002. Molecular systematics of

the genus Sigmodon (Rodentia: Muridae): evidences from the mitochondrial

cytochrome b gene. Journal of Mammalogy 83: 396-407.

PEREIRA, S. L., A. J. BAKER, AND A. WAJNTAL. 2002. Combined nuclear and

mitochondrial DNA sequences resolve genetic relationships within the

Cracidae (Galliformes, Aves). Systematic Biology 51:946-958.

PLATT II, R. N., B. R. AMMAN, M. S. KEITH, C. W. THOMPSON, AND R. D. BRADLEY.

2015. What is Peromyscus? Evidence from nuclear and mitochondrial DNA

sequences suggests the need for a new classification. Journal of Mammalogy

96: 708-719.

REEDER, S. A., and R. D. BRADLEY. 2004. Molecular systematics of Neotomine-

Peromyscine rodents based on the Dentin Matrix Protein 1 gene. Journal of

Mammalogy 85:1194-1200.

REEDER, S. A., D. S. CARROLL, C. W. EDWARDS, C. W. KILPATRICK, AND R. D.

BRADLEY. 2006. Neotomine-peromyscine rodent systematics based on

Texas Tech University, Megan S. Keith, December 2015

combined analyses of nuclear and mitochondrial DNA sequences. Molecular

Phylogenetics and Evolution 40: 251-258.

REID, F. A. 2009. A field guide to the mammals of Central America and southeast

Mexico, second edition. Oxford University Press, New York.

ROKAS, A., B. L. WILLIAMS, N. KING, AND S. B. CARROLL. 2003. Genome-scale

approaches to resolving incongruence in molecular phylogenies. Nature 425:

798-804.

ROGERS, D. S., M. D. ENGSTROM, and E. ARELLANO. 2005. Phylogenetic relationships

among Peromyscine rodents: allozyme evidence. Pp. 427-440 in

Contribuciones mastozoológicas en Homenaje a Bernardo Villa (V. Sanchez-

Cordero and R. A. Medellin, eds.). Instituto de Biología e Instituto de

Ecología, Universidad Nacional Autónoma de México, Mé xico.

SAIKI, R. K., D. H. GELFAND, S. STOFFEL, S. J. SCHARF, R. HIGUCHI, G. T. HORN, K. B.

MULLIS, AND H. A. ERLICH. 1988. Primer-directed enzymatic amplification of

DNA with a thermostable DNA polymerase. Science 239:487–491.

SHIMODAIRA, H. 2002. An approximately unbiased test of phylogenetic tree selection.

Systematic Biology 51:492-508.

SHIMODAIRA, H. AND M. HASEGAWA. 2001. CONSEL: for assessing the confidence

of phylogenetic tree selection. Bioinformatics 17:1246-1247.

SMITH, M. F., and J. L. PATTON. 1993. The diversification of South American murid

rodents: evidence from mitochondrial DNA sequence data for the Akodontine

tribe. Biological Journal of the Linnean Society 50:149-177.

Texas Tech University, Megan S. Keith, December 2015

STAMATAKIS, A., T. LUDWIG, and H. MEIER. 2005. RAxML-III: a fast program for

maximum likelihood-based inference of large phylogenetic trees.

Bioinformatics 21:456-463.

STANGL JR., F.B. AND R. J. BAKER. 1984. Evolutionary relationships in Peromyscus:

congruence in chromosomal, genic, and classical data sets. Journal of

Mammalogy 65: 643-654.

STANHOPE, M. J., J. CZELUSNIAK, J.-S. SI, J. NICKERSON, and M. GOODMAN. 1992. A

molecular perspective on mammalian evolution from the gene encoding

interphotoreceptor retinoid binding protein, with convincing evidence for bat

monophyly. Molecular Phylogenetics and Evolution 1:148-160.

TAMURA, K., G. STECHER, D. PETERSON, A. FILIPSKI and S. KUMAR. 2013. MEGA6:

molecular evolutionary genetics analysis version 6.0. Molecular Biology and

Evolution 30: 2725-2729.

TIEMANN-BOEGE, I., C. W. KILPATRICK, D. J. SCHMIDLY, and R. D. BRADLEY. 2000.

Molecular phylogenetics of the Peromyscus boylii species group (Rodentia:

Muridae) based on mitochondrial Cytochrome b sequences. Molecular

Phylogenetics and Evolution 16:366-378.

TOWNSEND, T. M., et al. 2011. Phylogeny of Iguanian lizards inferred from 29 nuclear

loci, and a comparison of concatenated and species-tree approaches for an

ancient, rapid radiation. Molecular Phylogenetics and Evolution 61:363-380.

TOYOSAWA, S., C. OHUIGIN, H. TICHY, AND J. KLEIN. 1999. Characterization of

dentin matrix protein 1 gene in crocodilia. Gene 234: 307-314.

Texas Tech University, Megan S. Keith, December 2015

WEKSLER, M. 2003. Phylogeny of Neotropical Oryzomyine rodents (Muridae:

Sigmodontinae) based on the nuclear IRBP exon. Molecular Phylogenetics and

Evolution 29:331-349.

WICKLIFFE, J. K., F. G. HOFFMANN, D. S. CARROLL, Y. DUNINIA-BARKOVSKAYA, and

R. D. BRADLEY. 2003. PCR and sequencing primers for intron 7 (Fgb-I7) of

the fibrinogen, B beta polypeptide (Fgb) in mammals: a novel nuclear DNA

phylogenetic marker. Occasional Papers, Museum of Texas Tech University

219:1-6

Texas Tech University, Megan S. Keith, December 2015

Table 3.1.—Specimens examined in this study. GenBank accession numbers are provided for each gene examined. Abbreviations are as follows: Adh1-I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb-I7 - intron of the beta-fibrinogen, Dmp1- dentin matrix protein 1, Ghr- growth hormone receptor, and Rbp3 - interphotoreceptor retinoid binding protein. GenBank accession numbers (top) and museum catalogue numbers (bottom) are given for each specimen. Museum acronyms are as follows: ASNHC (Angelo State Natural History Collection), BYU (Brigham Young University), CNMA (Colección Nacional de Mamíferos, Universidad Nacional Autónoma de México), CRD (Centro Interdisciplinario de Investigación para el Desarrollo Integral Regional, Unidad Durango, Mexico), FSH (University of Texas Medical Branch at Galveston), PGSC (Peromyscus Genetic Stock Center), ROM (Royal Ontario Museum), TCWC (Texas Cooperative Wildlife Collection), and TTU (Museum of Texas Tech University). If museum catalogue numbers were unavailable, specimens were referenced with the corresponding collector’s numbers or TK (special number of the Museum of Texas Tech University). An asterisk indicates the individuals/sequences that were used in the combined analysis. ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Sigmodon S. hispidus* KT318181 AF155420 AY459371 AY269989 KT964999 EU635710 TTU80759 FSH33 TTU80759 TTU80759 TTU80759 TTU80759 Isthmomys I. pirrensis* FJ214668 FJ214681 FJ214692 EF989746 EF989846 TTU39162 TTU39162 TTU39162 TTU39162 ROM116308 ROM116308

EF989947 EF989848 ROM116307 ROM116307

EF989945 ROM116308

EF989946 EF989747 ROM116309 ROM116309

Texas Tech University, Megan S. Keith, December 2015

Table 3.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------I. pirrensis (cont.) DQ836298 ROM116308

DQ836299 ROM16309 Peromyscus P. attwateri* AY994220 AF155384 AY274207 AY269978 KT950905 JX910128 TTU55688 TTU55688 TTU55688 TTU55688 TTU55688 TTU55688

P. beatae* AY994223 AF131921 FJ214696 KT950901 KT950924 TK93279 GK3954 TTU105037 TK93279 TTU105037 TK93279

P. californicus* AY994211 AF155393 FJ214697 EF989772 EF989873 TTU83292 TTU81275 TTU83291 TTU83291 PGSCIS1590 PGSCIS1590

P. crinitus* AY994213 AY376413 KT375572 EF989773 EF989874 DSR6171 DSR6171 TTU108167 TTU108167 BYU16629 BYU16629

P. eremicus* AY994212 AY322503 FJ214699 EF989775 EF989876 TTU81850 TTU83249 TTU83249 TTU83249 BYU17952 BYU17952

P. evides* FJ214670 FJ214685 FJ214700 KT950904 JX910121 TTU82696 TTU82696 TTU82696 TTU82696 TTU82696 TTU82696

Texas Tech University, Megan S. Keith, December 2015

Table 3.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Peromyscus (cont.) P. furvus* JX910116 KT965004 JX910113 KT950907 JX910124 FXG1168 FXG1168 FXG1168 FXG1168 FXG1167 FXG1168

P. gossypinus* FJ214671 DQ973102 FJ214702 KT950900 JX910126 TTU80682 TTU80682 TTU80682 TTU80682 TTU80682 TTU80682

P. gratus* AY994218 AY376421 FJ214703 KT950906 JX910129 TK46354 TK46354 TK46354 TK46354 TK46354 TK46354

P. hooperi* FJ214672 DQ973103 FJ214704 KT950909 JX910125 TTU104425 TTU104425 TTU104425 TTU104425 TTU104425 TTU104425

P. leucopus* AY994241 DQ000483 FJ214706 EF989779 EF989880 TTU115505 TTU101645 TTU101645 TTU101645 ROM101861 ROM101861

P. levipes* AY994224 AY322509 FJ214707 KT950902 JX910123 TK47819 TTU82707 TTU105150 TK47819 TK47819 TK47819

P. maniculatus* AY994242 AY322508 FJ214708 EF989783 EF989884 TTU97830 TTU83249 TTU97830 TTU97830 ROM98941 ROM98941

P. megalops* AY994217 DQ000475 FJ214709 KT950908 JX910127 TTU82712 TTU82712 TTU82712 TTU82712 TTU82712 TTU82712

Texas Tech University, Megan S. Keith, December 2015

Table 3.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Peromyscus (cont.) P. melanophrys* AY994216 AY322510 FJ214710 EF989789 EF989890 TTU75509 TTU75509 TTU75509 TTU75509 PGSCXZ1073 PGSCXZ1073

P. melanotis* FJ214673 AF155398 FJ214711 EF989790 EF989891 CRD2025 CRD2025 CRD2025 CRD2025 PGSC25 PGSC25

P. mexicanus* AY994236 AY376425 AY274210 AY269981 EF989794 EF989895 TTU97013 TTU82759 TTU82759 TTU82759 ROM113250 ROM113250

P. nudipes* AY994238 FJ214687 FJ214713 EF989792 EF989893 TTU96972 TTU96972 TTU96972 TTU96972 ROM113216 ROM113216

P. ochraventer* FJ214676 FJ214689 FJ214715 KT950910 JX910130 TTU104930 TTU104930 TTU104930 TTU104930 TTU104930 TTU104930

P. pectoralis* AY994221 DQ000476 FJ214716 KT950911 JX910131 TK48645 TTU75575 TK48645 TK48645 TK48645 TK48645

P. schmidlyi* KT318182 AY322524 FJ214718 KT950903 KT950925 TTU81617 TTU81703 TTU81617 TTU81617 TK72442 TTU81703 Habromys H. lepturus* AY994239 DQ973099 FJ214701 EF989742 EF989841 TK93160 TTU82703 TTU82703 TTU82703 CMNA29970 CNMA29970

Texas Tech University, Megan S. Keith, December 2015

Table 3.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Neotomodon N. alstoni* AY994210 AY195796 AY274202 AY269973 EF989751 EF989851 TK45309 TK45302 TK45309 TK45309 TK45309 ASNHC1595 Megadontomys M. thomasi* AY994208 AY195795 FJ214693 EF989749 EF989849 TK93388 TK93388 TK93388 TK93388 TK93388 CNMA29186 Onychomys O. arenicola* JX910115 AY195793 AY274204 AY269975 EF989755 EF989855 TTU67559 TTU67559 TTU67559 TK46462 ROM114904 ROM114904

O. leucogaster* KT318183 AY195794 AY274205 AY26976 EF989758 EF989859 TTU60605 TTU60605 TTU60605 TTU60605 ASNHC4348 ASNHC4348 Osgoodomys O. banderanus* AY994209 DQ000473 AY274206 AY269977 EF989756 EF989857 TK45952 TK45952 TK45401 TK45401 ASNHC2664 ASNHC2664 Podomys P. floridanus* AY994214 KT965003 FJ214724 KT950912 EF989878 TTU97867 TTU97866 TTU97868 TTU97868 TTU97868 TTU97866 Ochrotomys O. nuttalli* JX910114 AY195798 AY274203 AY269974 EF989761 EF989862 TCWC31929 TCWC31929 TCWC31929 TCWC31929 ROM113008 ROM113008 Reithrodontomys R. fulvescens* AY994207 AF176257 AY274211 AY269982 EF989800 EF989901 TTU54898 TTU54898 TTU54898 TTU54898 ASNHC3465 ASNHC3465

Texas Tech University, Megan S. Keith, December 2015

Table 3.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Reithrodontomys (cont.) R. megalotis AF176248 KT375573 EF989808 EF989909 TTU40942 TTU40942 TTU40942 ASNHC2133 ASNHC2133

R. mexicanus EF989911 ROM98468

R. sumichrasti* JX910117 AF176256 AY274212 EF989823 EF989924 TTU54952 TTU54952 TTU54952 TK20994 ROM98383 ROM98383 ------

Texas Tech University, Megan S. Keith, December 2015

Table 3.2.—Summary of results for all analyses performed in this study. The hypothesis that each dataset reflects is represented by "yes" and analyses for which relationships were statistically supported are represented by a "+".

Large-scale Topologically Combined analysis constrained Adh1-I2 Dmp1 Fgb-I7 Ghr Rbp3 Analysis (Cytb) trees (Cytb) Hypothesis 1: Isthmomys is monophyletic with N/A Yes Peromyscus/the first lineage to diverge from Peromyscus

Hypothesis 2: Isthmomys forms a sister- Yes N/A yes yes yes (+) group relationship with Reithrodontomys

Alternative Hypothesis: Isthmomys is N/A yes yes basal to both Peromyscus and Reithrodontomys

Texas Tech University, Megan S. Keith, December 2015

Reithrodontomys (bakeri-4, chrysopsis-1, brevirostris -1, creper-8, dariensis-1, fulvescens-18, gracilis -3, hirsustus-1, humulis-2, megalotis-8, mexicanus- 31, microdon-13, montanus-2, raviventris- 3, spectabilis-3, sumichrasti -161, tenuirostris-1, unidentified species-2, and zacatecae -3)

Isthmomys pirrensis (3) Onychomys (arenicola-3 , leucogaster-3, and torridus-3)

Peromyscus and allied genera (Habromys- chinanteco-1, delicatulus-1, ixtlani- 7, lepturus-10, lophurus-6, simulatus- 3, Megadontomys- cryophilus -5, nelsoni-9, thomasi-13, Neotomodon alstoni-5 , Osgoodomys banderanus-5, Peromyscus- attwateri -91, aztecus-14, beatae-32, boylii-19, californicus-4, crinitus -4, difficilis-7, eremicus-6, furvus-5, gossypinus-4, grandis -7, gratus-17, guatemalensis-5, gymnotis-2, hylocetes -6, hooperi-1, keeni-1, leucopus-6, levipes-7, maniculatus -256, mayensis-4, megalops-3, melanocarpus-1, melanophrys -4, melanotis-6, mexicanus-7, nasutus-2, nudipes-2, ochraventer-1, pectoralis-5, perfulvus-1, polionotus -26, polius-1, sagax-1, schmidlyi-40, simulus-1, spicilegus- 5, stephani-1, truei-46, unknown-21, winkelmanni-4, zarhynchus -1, and Podomys floridanus-7)

Ochrotomys nuttalli (3) Baiomys (taylori-13 and musculus-15)

Sigmodon hispidus

Texas Tech University, Megan S. Keith, December 2015

Figure 3.1.—Phylogenetic tree obtained from the maximum likelihood analysis of the large scale sampling for the mitochondrial cytochrome-b gene (Cytb). Support values were based on 1,000 bootstrap replicates of the maximum likelihood analysis. Values ≥80 are indicated by an asterisk. The number of individuals for each species included are provided to the right of the dash; Isthmomys is represented by a dashed line. Clade I reflects a sister relationship between Reithrodontomys and Isthmomys, with Peromyscus basal to these two genera in clade II. However, there was no statistical support for this relationship.

Texas Tech University, Megan S. Keith, December 2015

Figure 3.2.—Phylogenetic trees obtained from maximum likelihood analysis of the unconstrained (A) and constrained (B) topologies. Support values were based on 1,000 bootstrap replicates of the maximum likelihood analysis. Values ≥80 are represented by an asterisk. For the unconstrained analysis (A), clade I reflects a sister relationship between Reithrodontomys and Isthmomys, with Peromyscus basal to these two genera in clade II. For the constrained analysis (B), Isthmomys was constrained to be monophyletic with Peromyscus (Clade IV) and Reithrodontomys forms a separate clade (Clade III). The score of the best tree for the unconstrained analysis (A) was -35559.90420 and the score of the best constrained tree (B) was -35576.11781.

Texas Tech University, Megan S. Keith, December 2015

Figure 3.3.—Results of the individual nuclear maximum likelihood analyses. Support values were based on 1,000 bootstrap replicates of the maximum likelihood analysis. Values ≥80 are represented by an asterisk. Analyses for Adh1-I2 (A) and Fgb-I7 (B) resulted in phylogenies in which Isthmomys was basal to both Peromyscus and Reithrodontomys, phylogenies for Ghr (C) and Rbp3 (D) indicate a clade uniting Isthmomys and Reithrodontomys with Peromyscus basal to this clade, and analysis of Dmp1 (E) resulted in a phylogeny in which Isthmomys is more closely related to Peromyscus. 92

Texas Tech University, Megan S. Keith, December 2015

Figure 3.4.—Phylogenetic tree obtained from maximum likelihood analysis of the combined dataset (Cytb, Adh1-I2, Dmp1, Fgb-I7, Ghr, Rbp3). Support values are based on 1,000 bootstrap replicates of the maximum likelihood analysis. Values ≥80 are shown above the supported node. Clade I reflects a strongly supported sister relationship between Reithrodontomys and Isthmomys, with a monophyletic Peromyscus basal to these two genera in clade II. 93

Texas Tech University, Megan S. Keith, December 2015

CHAPTER IV

PHYLOGENETIC RELATIONSHIPS AND DIVERGENCE OF THE SUBFAMILY TYLOMYINAE (RODENTIA: CRICETIDAE) AS DETERMINED BY A MULTILOCUS DATASET

ABSTRACT The tylomyine rodents consist of four genera (Nyctomys, Otonyctomys,

Ototylomys, and Tylomys) distributed from Mexico to northern South America.

Species within each of these genera can be difficult to distinguish, the diversity within this group is poorly understood, and little is known of their evolutionary history. The phylogenetic relationships of the tylomyine rodents to the neotomine and sigmodontine rodents remain enigmatic despite their inclusion in numerous studies.

Morphological data and evidence from molecular markers suggest the placement of these rodents in their own subfamily (Tylomyinae); however, genetic studies in which these rodents were examined have not specifically focused on resolving the phylogenetic relationships for these four genera. Further, previous studies have not sampled key taxa from the Neotominae or Sigmodontinae. DNA sequences were obtained from one mitochondrial (Cytb) and five nuclear markers (Adh1-I2, Fbg-I7,

Dmp1, Ghr, and Rbp3) from multiple representatives of the Neotominae,

Sigmodontinae, and Tylomyinae. Sequences were combined into a single analysis and analyzed using Bayesian inference to assess the phylogenetic position of the

Tylomyinae relative to closely affiliated subfamilies in the Cricetidae. Genetic distances were calculated and divergence dates were estimated for each subfamily to test the hypothesis that the Tylomyinae is an older clade and therefore basal to the

Texas Tech University, Megan S. Keith, December 2015 divergence of the sigmodontine and neotomine rodents. Results indicated that the tylomyine rodents should be recognized as a separate subfamily based on phylogenetic analysis with additional support from genetic distance estimates and estimations of divergence dates for each subfamily. These rodents are more closely related to the

Neotominae than the Sigmodontinae and appear to have diverged from the neotomine- peromyscine rodents approximately 12.68 million years ago.

INTRODUCTION Little is known of the evolutionary history of the rodents currently recognized within the Subfamily Tylomyinae. Tylomyines are medium to large-sized rodents that are composed of four genera: Nyctomys, Otonyctomys, Ototylomys, and Tylomys, and are distributed from Jalisco, Mexico to Ecuador (Hall 1981; Escalente et al. 2003;

Musser and Carleton 2005; Gutiérrez-García and Vázquez-Domínguez 2012; Owen and Girón 2012; Méndez-Carvajal et al. 2015). In general, these genera are not commonly collected and subspecies recognized within each genus are only distinguished in the field by geographic distribution (Reid 2009).

Few studies have examined this group of rodents due to the rarity of prepared specimens and preserved tissue samples in museum collections for some taxa, especially Otonyctomys. In addition, a consensus on the taxonomic status of this group of rodents has not been achieved. Based on morphological characters, tylomyine rodents originally were classified in the thomasomyine group of South

American cricetines (Hershkovitz 1944) and subsequently placed within a broadly defined Cricetinae as members of the tribe Oryzomyini (Vorontsov 1959). Hooper

Texas Tech University, Megan S. Keith, December 2015 and Musser (1964) viewed this group as members of an early lineage that gave rise to

Neotoma, whereas Carleton (1980) predicted that the tylomyine rodents would prove to represent an older clade that was basal to the sigmodontine and neotomine rodents

(Musser and Carleton 2005). Others have recognized this group of rodents as a tribe within the Sigmodontinae (Tribe Tylomyini- McKenna and Bell 1997), a tribe

(Tylomyini) classified in the Neotominae (Reeder et al. 2006), or as a separate subfamily (Tylomyinae- Reig 1984; Steppan et al. 2004; Musser and Carleton 2005).

Additionally, some authors have suggested that this group may represent an ancient phyletic lineage that is distantly related to the neotomine-peromyscine rodents or that they may have originated prior to both the North and South American sigmodontines

(Arata 1964; Hershkovitz 1966; Carleton 1980; Voss and Linzey 1981; Sarich 1985;

Haiduk et al. 1988; Steppan 1995; Engel et al. 1998; Musser and Carleton 2005).

More recent studies recognized a monophyletic Tylomyinae or Tylomyini consisting of two distantly related clades that correspond to two sets of sister genera

(Nyctomys - Otonyctomys and Ototylomys - Tylomys, Musser and Carleton 2005;

Reeder et al. 2006; Corley et al. 2011). Steppan et al. (2004) recovered weak support for a clade uniting the Neotominae, Sigmodontinae, and Tylomyinae based on analyses of four nuclear genes and recommended the recognition of the Tylomyinae.

Furthermore, Steppan et al. (2004) recovered a topology in which the Tylomyinae were basal to the Sigmodontinae; however, their sampling for the tylomyine rodents was restricted to a single representative each of Tylomys and Ototylomys and included few representatives of the Neotominae.

Texas Tech University, Megan S. Keith, December 2015

Recent studies have demonstrated that there may be previously undetected variation within the tylomyine rodents. For example, Corley et al. (2011) examined the cytochrome-b gene within this group and determined that the genus Nyctomys is paraphyletic with Otonyctomys and indicated a need for taxonomic revision.

Additionally, examination of genetic data recovered support for three distinct phylogeographical lineages within Ototylomys phyllotis that coincided with major geological features of Middle America, and supported monophyly of Ototylomys

(Gutiérrez-García and Vázquez-Domíguez 2012). Recent studies have raised questions about this group of rodents, including the number of genera and species that should be recognized. To gain a comprehensive understanding of this group of rodents, additional specimens and preserved tissue samples are needed for a better understanding of the evolutionary history and diversification of these rodents.

In addition to the inconsistent taxonomic status of the tylomyines, most phylogenetic studies have not examined a sufficient sampling of the Sigmodontinae and Neotominae. Most studies included multiple taxa from the Neotominae but included only one representative of sigmodontine rodents as the outgroup taxon. The objective of this study was to utilize a more robust sampling scheme and a molecular dataset comprised of six genes (the mitochondrial cytochrome-b gene (Cytb), intron 2 of the alcohol dehydrogenase gene (Adh1-I2), intron 7 of the beta-fibrinogen gene

(Fgb-I7), exon 6 of the dentin matrix protein 1 gene (Dmp1), exon 10 of the growth hormone receptor gene (Ghr), and exon 1 of the interphotoreceptor retinoid-binding protein gene (Rbp3)) to determine the phylogenetic placement of the tylomyine

Texas Tech University, Megan S. Keith, December 2015 rodents relative to the Sigmodontinae and Neotominae. Specifically, the hypothesis of

Carleton (1980) was tested: these genera constitute an older lineage basal to the divergence of the sigmodontine and neotomine rodents.

MATERIALS AND METHODS The taxonomic sampling for this study included 48 taxa: 39 species from 18 genera of North American neotomine rodents, three representatives from three genera of tylomyine rodents (preserved tissue for Otonyctomys was not available) and six species of sigmodontine rodents. At least one representative from each tribe within the Neotominae and four tribes within the Sigmodontinae (Akodontini, Oryzomyini,

Phyllotini, and Sigmodontini) were included with multiple representatives for each tribe when appropriate. Tissues were either obtained from museum collections or collected by the authors. Field collecting procedures followed the guidelines of the

American Society of Mammalogists (Sikes et al. 2011). Specimens examined are listed in Table 4.1 with GenBank accession numbers for all sequences analyzed in this study.

Sequences were either generated in this study or obtained from GenBank. Six genes were selected to resolve the taxonomic status of the tylomyine rodents: the mitochondrial Cytb gene, and five nuclear genes- Adh1-I2, Fgb-I7, Dmp1, Ghr, and

Rbp3. Genomic DNA was isolated from approximately 0.1g of frozen liver tissue using the Puregene DNA isolation kit (Gentra, Minneapolis, Minnesota) or the Qiagen

DNeasy Blood and Tissue Kit (Qiagen, Valencia, California). DNA fragments were amplified using polymerase chain reaction (PCR, Saiki et al. 1988). Methods for PCR

Texas Tech University, Megan S. Keith, December 2015 amplification and sequencing of each gene followed the protocols from Keith et al. (in prep A). Sequences were proofed using Sequencher 4.10 software (Gene Codes, Ann

Arbor, Michigan) and aligned using MEGA v6.06 software (Tamura et al. 2013).

Phylogenetic Analysis.—Nucleotide positions were treated as unordered, discrete characters with six possible states: A, C, G, T, gaps (-) or missing (?) for each marker. Alignment of nuclear intron sequences required hypothesized gaps to represent insertion or deletion events. Sequences were aligned using MUSCLE software (Edgar 2004) in MEGA v6.06 (Tamura et al. 2013). Indels were coded for intronic sequences (Adh1-I2 and Fgb-I7) using simple indel coding (SIC- Simmons and Ochoterena 2000) methods by SeqState software (Müller 2005). Gaps have been shown to provide additional phylogenetic signal and not coding indels is equivalent to discarding data (Giribet and Wheeler 1999, Graham et al. 2000, Kelchner 2000,

Simmons and Ochoterena 2000, Simmons et al. 2001). Additionally, evidence existed that indels may provide reliable phylogenetic characters of low levels of homoplasy

(Geiger 2002, Graham et al. 2000, Hamilton et al. 2003, Müller and Borsch 2005,

Simmons et al. 2001, Vogt 2002).

Individual datasets were combined and sequences from conspecific individuals were used when it was not possible to obtain mitochondrial and nuclear sequences for the same individual. The utilization of combined mitochondrial and nuclear genetic sequences generally increases resolution at multiple hierarchical levels within the phylogeny with rapidly evolving mitochondrial markers resolving terminal nodes and less quickly evolving nuclear markers resolving relationships at the base of the

Texas Tech University, Megan S. Keith, December 2015 phylogeny (Bull et al. 1993; Adkins et al. 2001; Pereira et al. 2002; Platt et al. 2015).

Additionally, concatenation of sequence data from conspecifics to represent a composite species rather than a single individual has been used successfully in various taxa (Campbell and Lapointe 2009; Haddrath and Baker 2012; Townsend et al. 2011;

Platt et al. 2015) and should not have a drastic effect on determining evolutionary relationships at higher hierarchal levels.

Molecular models of evolution were determined using jModelTest v2.1.6

(Darriba et al. 2012) and the Akaike information criterion. jModelTest v2.1.6

(Darriba et al. 2012) identified HKY+I+G for Adh1-I2 and Fgb-I7 and GTR+I+G for

Cytb, Dmp1, Ghr, and Rbp3 as the best-fit nucleotide substitution models. Datasets were partitioned to be analyzed under the designated nucleotide substitution model and indels were coded as binary data to generate a single dataset consisting of 5,599 characters. A Bayesian inference (BI) analysis was conducted to estimate a phylogeny and generate posterior probability values for the combined dataset using MrBayes v3.2.5 (Ronquist et al. 2012) with 10 million generations and a sample frequency of every thousandth generation. Posterior probability values ≥ 95% were viewed as supportive.

Sequence Divergence Estimates.—Genetic distances were calculated using the

Kimura 2-parameter model (K2P; Kimura 1980) in MEGA v6.06 (Tamura et al.

2013). Distance estimates were obtained for all genes for two sets of comparisons: the mean distance between subfamilies (Sigmodontinae, Neotominae, and Tylomyinae) and the mean distance between all tribes of the Neotominae as recognized by Reeder

100

Texas Tech University, Megan S. Keith, December 2015 et al. (2006) and Miller and Engstrom (2008): Baiomyini, Neotomini, Ochrotomyini,

Peromyscini, Reithrodontomyini, and Tylomyini. For datasets in which there were numerous partial sequences (Dmp1), any sequences that were missing 50% or more of the total length of the gene were removed so that common bases could be identified and distances could be calculated more accurately.

Molecular dating.—To estimate divergence dates between subfamilies,

BEAST v1.8.2 (Drummond et al. 2012) was used. All sequence data from each gene was used, but the data was partitioned to allow modeling for each dataset. Models of nucleotide substitution for each dataset were the same as above. The program MEGA v6.06 (Tamura et al. 2013) was used to determine whether each gene was evolving under a strict or relaxed molecular clock. A Yule tree prior was selected given only a single individual from each species was sampled.

To assess subfamily divergence, several fossil calibrations were used at a variety of taxonomic levels. Calibration points were placed at five nodes to reflect calculated first appearances of sampled taxa in the fossil record (Behrensmeyer and

Turner 2013) and to account for known complexities in the resolution of the phylogeny (e.g., see Platt et al. 2015). Calibration points were as follows: subfamily

Sigmondontinae (~5.3 million years [MYA]; Verzi and Montalvo 2008), tribe

Neotomini (~5.3 MYA; Reynolds 1991), genus Neotoma (~2.76 MYA; Hornsby and

Matocq 2012), genus Reithrodontomys (~1.8 MYA; Cassiliano 1999), and the leucopus/maniculatus group (~1.8 MYA; Dalquest 1962; Karow et al. 1996). A prior lognormal distribution was placed on each calibration point with upper limits adjusted

101

Texas Tech University, Megan S. Keith, December 2015 to reflect the boundary of the associated land mammal age (Woodburne 1987). Test runs of 1.0 X 107 generations with a 10% burn-in were used and compared with Bayes

Factors (Kass and Raftery 1995; Suchard et al. 2001) to optimize for the final analysis.

For final estimates, four final runs of 5.0 x 107 generations were analyzed with log and tree files combined. Final parameters were examined for sufficient mixing, convergence stability, and effective sample size using the program Tracer v1.5

(Drummond and Rambaut 2008).

RESULTS BI analysis of the combined dataset generated a phylogeny with statistical support (posterior probability values ≥ 95%) for all higher-level taxonomic groupings

(subfamilies and tribes- Fig. 4.1). The Neotominae and Tylomyinae were united into a single clade (Clade I) and the Sigmodontinae were basal to this clade (Clade II).

Clade I consisted of two subclades: Clade A was composed of the rodents classified within the Neotominae whereas members of Clade B represented a monophyletic

Tylomyinae. All clades were strongly supported with posterior probability values >

95%.

Genetic Distances.—K2P genetic distances (Table 4.2) were used to evaluate the taxonomic status of the tylomyine rodents by comparing genetic divergence between subfamilies and between tribes that historically have been recognized within the Neotominae. Data obtained from between subfamily comparisons indicated that the tylomyine rodents were genetically more similar to the Neotominae (13.4%) than the Sigmodontinae (15.6%). Average genetic distances from comparisons between the

102

Texas Tech University, Megan S. Keith, December 2015 tylomyine rodents and tribes within the Neotominae (Baiomyini, Neotomini,

Ochrotomyini, Peromyscini and Reithrodontomyini) ranged from 13.8% (Neotomini vs. Tylomyini) to 15.0% (Ochrotomyini vs. Tylomyini) whereas the average genetic distance across all datasets for all tribes within the Neotominae (excluding the tylomyine rodents) was 11.2%.

Molecular Dating.—A strict molecular clock was rejected for all datasets except Fgb-I7, necessitating a relaxed molecular clock for the other five datasets. The

Yule birth rate was estimated at 0.217 (95% highest posterior density [HPD]: 0.15--

0.29). Mean rates of evolution (as substitutions per site per million years) were 0.009 for Adh1-I2 (95% HPD: 0.007--0.01), 0.07 for Cytb (95% HPD: 0.056--0.085), 0.006 for Dentin (95% HPD; 0.005—0.007), 0.008 for Fgb-I7 (95% HPD: 0.006--0.009),

0.004 for Ghr (95% HPD: 0.003—0.004), and 0.004 for Rbp3 (95% HPD: 0.003--

0.004). Divergence date estimates (Fig. 4.2) suggested that the subfamily

Sigmodontinae began to split from the Neotominae/Tylomyinae complex ~14.07

MYA (95% HPD: 11.73--16.85), during the Miocene. A subsequent split between

Neotominae and Tylominae also occurred in the Miocene, ~12.68 MYA (95% HPD:

10.59--15.1).

DISCUSSION Based on the results of the BI analysis of the combined dataset (Fig. 4.1), two interpretations can explain the taxonomic status of the tylomyine rodents: 1) Clade I represents one subfamily in which both the neotomine and tylomyine rodents are united, or 2) two separate rodent subfamilies should be recognized. Under the first

103

Texas Tech University, Megan S. Keith, December 2015 interpretation, the tylomyine rodents would have to be placed in the tribe Tylomyini

(subfamily Neotominae). Alternatively, the second interpretation would require that members of Clade A (Fig. 4.1) be recognized as the Neotominae and that members of

Clade B be referred to the Tylomyinae. These results also suggest that the

Sigmodontinae (Clade II) is basal to a clade uniting the Tylomyinae and Neotominae

(Clade I) and that the sigmodontine rodents diverged from other members of the

Cricetidae prior to the divergence of the tylomyine rodents from the Neotominae.

Genetic distances between tribal comparisons indicated higher levels of divergence when comparing the tylomyine rodents to each tribe (Baiomyini,

Neotomini, Ochrotomyini, Peromyscini, and Reithrodontomyini) within the

Neotominae (averages ranged from 13.8% - 15.0%, Table 4.2) than distances observed for between tribe comparisons (11.2%). Average genetic distances calculated for comparisons between subfamilies were higher than the overall average genetic distance for tribal comparisons. Results indicated 15.0% sequence divergence for

Sigmodontinae vs. Neotominae, 15.6% for Sigmodontinae vs. Tylomyinae, and 13.4% for Neotominae vs. Tylomyinae.

Divergence date estimates (Fig. 4.2) indicated that the subfamily

Sigmodontinae began to diverge from the Neotominae/Tylomyinae complex ~14.07

MYA (95% HPD: 11.73--16.85), during the Miocene. A subsequent split between the

Neotominae and Tylominae also occurred in the Miocene, ~12.68 MYA (95% HPD:

10.59--15.1); whereas the diversification within the Tylomyinae began ~11.54 MYA

(95% HPD: 9.44--13.96).

104

Texas Tech University, Megan S. Keith, December 2015

Systematic conclusions.—Historically, based on examination of morphological characters (Reig 1984; Musser and Carleton 2005), the tylomyine rodents were hypothesized to form their own subfamily separate from the Neotominae and

Sigmodontinae. The phylogeny that resulted from BI analysis of the combined dataset supports the recognition of the Tylomyinae as was suggested by Reig (1984) based on examination of morphological characters. Genetic distance estimates further support the recognition of the Tylomyinae given that intrasubfamily observations indicated that there are approximately equal amounts of genetic differentiation between the three subfamilies included in this study. Additionally, estimates of divergence dates indicated the Sigmodontinae diverged prior to the split that occurred between the

Neotominae and Tylomyinae during the Miocene.

Two interpretations were presented to explain the taxonomic status of the tylomyine rodents based on BI analysis of the combined dataset: 1) the neotomine and tylomyine rodents are united in one clade representing the Neotominae, or 2) two separate rodent subfamilies should be recognized. A reciprocal monophyletic clade was recovered for the tylomyine rodents and genetic distance estimates for between subfamily comparisons indicated that the Tylomyinae are approximately as divergent from the Neotominae (13.4%) as the Neotominae are from the Sigmodontinae

(15.0%). Additionally, genetic distance estimates for between subfamily comparisons were higher than the average genetic distances for tribes within the Neotominae

(11.2%). Therefore, this study indicated strong phylogenetic evidence and statistical support for the recognition of the second interpretation in which two separate

105

Texas Tech University, Megan S. Keith, December 2015 subfamilies are recognized with the members of Clade A (Fig. 4.1) representing the

Neotominae and the members of Clade B representing the Tylomyinae. The

Tylomyinae does not represent an older lineage that is basal to the Sigmodontinae and

Neotominae as hypothesized by Carleton (1980). Instead, the Sigmodontinae are basal to a clade uniting the Neotominae and Tylomyinae.

ACKNOWLEDGMENTS The authors would like to thank the Texas Academy of Science and Texas

Tech University Association of Biologists for funding necessary to complete this project. We thank R. J. Baker, H. Garner, and K. McDonald (Natural Science

Research Laboratory, Museum of Texas Tech University), Centro Interdisciplinario de

Investigación para el Desarrollo Integral Regional, Unidad Durango, Mexico and

Duke S. Rogers for access and assistance in obtaining tissues. We would also like to thank Emma K. Roberts and Nicté Ordóñez-Garza for reading and editing earlier versions of this manuscript.

106

Texas Tech University, Megan S. Keith, December 2015

LITERATURE CITED ADKINS, R. M., E. L. GELKE, D. ROWE, AND R. L. HONEYCUTT. 2001. Molecular

phylogeny and divergence time estimates for major rodent groups: evidence

from multiple genes. Molecular Biology and Evolution 18: 777-791.

ARATA, A. 1964. The anatomy and taxonomic significance of the male accessory

reproductive glands of muroid rodents. Bulletin of the Florida State Museum,

Biological Sciences 9: 1-42.

BEHRENSMEYER, A. K., AND A. TURNER. 2013. Taxonomic occurrences of Cricetidae

recorded in the Paleobiology Database. Fossilworks. http://fossilworks.org.

BULL, J. J., J. P. HUELSENBECK, C. W. CUNNINGHAM, D. L. SWOFFORD, AND P. J.

WADDELL. 1993. Partitioning and combining data in phylogenetic analysis.

Systematic Biology 42: 384-397.

CAMPBELL, V., and F.-J. LAPOINTE. 2009. The use and validity of composite taxa in

phylogenetic analysis. Systematic Biology 58:560-572.

CARLETON, M.D. 1980. Phylogenetic relationships in neotomine-peromyscine

rodents (Muroidea) and a reappraisal of the dichotomy with New World

Cricetinae. Miscellaneous Publications, Museum of Zoology, University of

Michigan 157: 1-146.

CASSILIANO, M. L. 1999. Biostratigraphy of Blancan and Irvingtonian mammals in the

Fish Creek-Vallecito section, southern California, and a review of the

Blancan–Irvingtonian boundary. Journal of Vertebrate Paleontology 19:169-

186.

107

Texas Tech University, Megan S. Keith, December 2015

CORLEY, M. S., N. ORDÓÑEZ-GARZA, D. S. ROGERS, AND R. D. BRADLEY. 2011.

Molecular Evidence for Paraphyly in Nyctomys sumichrasti: Support for a

New Genus of Vesper Mice? Occasional Papers, Museum of Texas Tech

University 306: 1-10.

DALQUEST, W. W. 1962. The Good Creek Formation, Pleistocene of Texas, and its

fauna. Journal of Paleontology 36:568-582.

DARRIBA, D., G.L. TABOADA, R. DOALLO, AND D. POSADA. 2012. jModelTest 2:

more models, new heuristics and parallel computing. Nature Methods 9: 772.

DRUMMOND, A. J., and A. RAMBAUT. 2008. Tracer v1.5.

http://tree.bio.ed.ac.uk/software/tracer/.

DRUMMOND, A. J., M. A. SUCHARD, D. XIE, and A. RAMBAUT. 2012. Bayesian

Phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and

Evolution 29:1969-1973.

EDGAR, R. C. 2004. MUSCLE: multiple sequence alignment with high accuracy and

high throughput. Nucleic Acids Research 32: 1792-1797.

ENGEL, S. R., K. M. HOGAN, J. F. TAYLOR, AND S. K. DAVIS. 1998. Molecular

systematics and paleogeography of the South American sigmodontine rodents.

Molecular Biology and Evolution 15: 35-49.

ESCALANTE, T., D. ESPINSOSA, AND J. J. MORRONE. 2003. Using parsimony analysis

of endemicity to analyze the distribution of Mexian land mammals. The

Southwestern Naturalist 48: 563-578.

108

Texas Tech University, Megan S. Keith, December 2015

GEIGER, D. L. 2002. Stretch coding and block coding: two new strategies to

representent questionable aligned DNA sequences. Journal of Molecular

Evolution 54: 191-199.

GIRIBET, G., AND W. C. WHEELER. 1999. On gaps. Cladistics 13:132-143.

GRAHAM, S. W., P. A. REEVES, A. C. E. BURNS, AND R. G. OLMSTEAD. 2000.

Microstructural changes in noncoding chloroplast DNA: interpretation,

evolution, and utility of indels and inversions in basal angiosperm phylogenetic

inference. International Journal of Plant Sciences 161:S83-S96.

GUTIÉRREZ-GARCÍA, T. A. AND E. VÁZQUEZ-DOMÍNGUEZ. 2012. Biogeographically

dynamic genetic structure bridging two continents in the monotypic Central

American rodent Ototylomys phyllotis. Biological Journal of the Linnean

Society 107: 593-610.

HADDRATH, O., and A. J. BAKER. 2012. Multiple nuclear genes and retroposons

support vicariance and dispersal of the palaeognaths, and an Early Cretaceous

origin of modern birds. Proceedings of the Royal Society B: Biological

Sciences.

HAIDUK, M.W., C. SANCHEZ-HERNANDEZ, AND R. J. BAKER. 1988. Phylogenetic

relationships of Nyctomys and Xenomys to other cricetine genera based on data

from G-banded chromosomes. The Southwestern Naturalist 33: 397-403.

HALL, E. R. 1981. The Mammals of North America. Second ed. John Wiley and

Sons, New York, 2:626-632.

109

Texas Tech University, Megan S. Keith, December 2015

HAMILTON, M. B., J. M. BRAVERMAN, AND D. F. SORIA-HERNANZ. 2003. Patterns and

relative rates of nucleotide and insertion/deletion evolution at six chloroplast

intergenic regions in new world species of the lecythidaceae. Molecular

Biology and Evolution 20: 1710-1721.

HERSHKOVITZ, P. 1944. A systematic review of the Neotropical water rats of the

genus Nectomys (Cricetinae). Miscellaneous Publications, Museum of

Zoology, University of Michigan 58: 1-101.

HERSHKOVITZ, P. 1966. Mice, land bridges and Latin American faunal interchange.

Pp. 725-747, in: Ectoparasites of Panama. (R. Wenzel and V. J. Tipton, eds.)

Field Museum of Natural History, Chicago, xii + 861 pp.

HORNSBY, A. D. AND M. D. MATOCQ. 2011. Differential regional response of the

bushy-tailed woodrat (Neotoma cinerea) to late Quaternary climate change.

Journal of Biogeography 39: 289-305.

HOOPER, E. T. AND G. G. MUSSER. 1964. The glans penis in Neotropical cricetines

(family Muridae) with comments on classification of muroid rodents.

Miscellaneous Publications, Museum of Zoology, University of Michigan 120:

1-57.

KAROW, P., G. MORGAN, R. PORTELL, E. SIMMONS, and K. AUFFENBERG. 1996.

Middle Pleistocene (early Rancholabrean) vertebrates and associated marine

and non-marine invertebrates from Oldsmar, Pinellas County, Florida. Pp. 97-

133 in Palaeoecology and Palaeoenuironments of Late Cenozoic mammals

110

Texas Tech University, Megan S. Keith, December 2015

Tributes to the Career of C S (Rufus) Churcher (K. Stewart and K. Seymour,

eds.). University of Tronto Press, Toronto, Canada.

KASS, R. E., and A. E. RAFTERY. 1995. Bayes Factors. Journal of the American

Statistical Association 90:773-795.

KEITH, M. S., J. SALAZAR-BRAVO, R. D. BRADLEY, AND R. N. PLATT II. In Prep A.

Molecular data indicate that Isthmomys is not aligned with Peromyscus.

KELCHNER, S. A. 2000. The evolution of non-coding chloroplast DNA and its

application in plant systematics. Annals of the Missouri Botanical Garden 87:

482-498.

KIMURA, M. 1980. A simple method for estimating evolutionary rates of base

substitutions through comparative studies of nucleotide sequences. Journal of

Molecular Evolution 16:111-120.

MCKENNA, M. C., AND S. K. BELL. 1997. Classification of mammals above the

species level. Columbia University Press, NY.

MÉNDEZ-CARVAJAL, P., M. PEÑAFIEL, A. ZAPATA, AND G. BERGUIDO. 2015. The

Panamanian climbing rat Mammalia, Rodentia, Cricetidae, Tylomys

panamensis (Gray, 1873): New Report in Darien. Tecnociencia 17:47-56.

MÜLLER, K. 2005. SeqState-primer design and sequence statistics for phylogenetic

DNA data sets. Applied Bioinformatics 4: 65-89.

MÜLLER, K. AND T. BORSCH. 2005. Phylogenetics of Utricularia (Lentibulariaceae)

and molecular evolution of the trnK intron in a lineages with high

substitutional rates. Plant Systematics and Evolution 250:39-67.

111

Texas Tech University, Megan S. Keith, December 2015

MUSSER, G. G. and M. D. CARLETON. 2005. Superfamily Muroidea. Pp. 501-775 in

Mammal species of the world: a taxonomic and geographic reference (Wilson

DE and Reeder DM eds.), John Hopkins Univ. Press, Baltimore.

OWEN, J. G., AND L. GIRÓN. 2012. Revised checklist and distributions of land

mammals of El Salvador. Occasional Papers, Museum of Texas Tech

University: 310: 1-32.

PEREIRA, S.L., A.J. BAKER, AND A. WAJNTAL. 2002. Combined nuclear and

mitochondrial DNA sequences resolve genetic relationships within the

Cracidae (Galliformes, Aves). Systematic Biology 51:946-958.

PLATT II, R. N., B. R. AMMAN, M. S. KEITH, C. W. THOMPSON, AND R. D. BRADLEY.

2015. What is Peromyscus? Evidence from nuclear and mitochondrial DNA

sequences suggests the need for a new classification. Journal of Mammalogy

96: 708-719.

REEDER, S. A., D. S. CARROLL, C. W. EDWARDS, C. W. KILPATRICK, AND R. D.

BRADLEY. 2006. Neotomine-peromyscine rodent systematics based on

combined analyses of nuclear and mitochondrial DNA sequences. Molecular

Phylogenetics and Evolution 40: 251-258.

REID, F.A. A Field Guide to the Mammals of Central America and Southeast Mexico,

Second Edition. 2009. Oxford University Press, New York, New York.

REIG, O. A. 1984. Distribução geográfica e história evolutiva dos roedores muroideas

sulamericanos (Cricetidae: Sigmodontinae). Revista Brasileira de Genetica 7:

333-365.

112

Texas Tech University, Megan S. Keith, December 2015

REYNOLDS, R. E. 1991. Biostratigraphic relationships of Tertiary small vertebrates

from Cajon Valley, San Bernardino County, California. San Bernardino

County Museum Association Quarterly 38:4

RONQUIST, F., M. TESLENKO, P. VAN DER MARK, D. L. AYRES, A. DARLING, S. HÖHNA,

B. LARGET, M. A. SUCHARD, AND J. P. HUELSENBECK. 2012. MrBayes 3.2:

Efficient Bayesian phylogenetic inference and model choice across a large

model space. Systematic Biology 61:539-542.

SAIKI, R. K., D. H. GELFAND, S. STOFFEL, S. J. SCHARF, R. HIGUCHI, G. T. HORN, K. B.

MULLIS, AND H. A. ERLICH. 1988. Primer-directed enzymatic amplification of

DNA with a thermostable DNA polymerase. Science 239:487–491.

SARICH, V. M. 1985. Rodent macromolecular systematics. Pp. 423-452, in

Evolutionary relationships among rodents, a multidisciplinary analysis (W. P.

Luckett and J.-L. Hartengerger, eds.). Plenum Press, New York, 721 pp.

SIKES, R. S., W. L. GANNON, AND THE ANIMAL CARE AND USE COMMITTEE OF THE

AMERICAN SOCIETY OF MAMMALOGISTS. 2011. Guidelines of the American

Society of Mammalogists for the use of wild mammals in research. Journal of

Mammalogy 92:235–253.

SIMMONS, M. P. AND H. OCHOTERENA. 2000. Gaps as characters in sequence-based

phylogenetic analyses. Systematic Biology 49:369-381.

SIMMONS, M. P., H. OCHOTERENA, AND T. G. CARR. 2001. Incorporation, relative

homoplasy, and effect of gap characters in sequence-based phylogenetic

analyses. Systematic Biology 50: 454-462.

113

Texas Tech University, Megan S. Keith, December 2015

STEPPAN, S. J. 1995. Revision of the Tribe Phyllotini (Rodentia: Sigmodontinae),

with a phylogenetic hypothesis for the Sigmodontinae. Fieldiana: Zoology,

n.s., 80:112 pp.

STEPPAN, S. J., R. M. ADKINS, AND J. ANDERSON. 2004. Phylogeny and divergence-

date estimates of rapid radiations in muroid rodents based on multiple nuclear

genes. Systematic Biology 53: 533-553.

SUCHARD, M. A., R. E. WEISS, and J. S. SINSHEIMER. 2001. Bayesian selection of

continuous-time Markov chain evolutionary models. Molecular Biology and

Evolution 18:1001-1013.

TAMURA, K., G. STECHER, D. PETERSON, A. FILIPSKI, AND S. KUMAR. 2013. MEGA6:

Molecular Evolutionary Genetics Analysis version 6.0. Molecular Biology and

Evolution 30:2725-2729.

TOWNSEND, T. M., D. G. MULCAHY, B. P. NOONAN, J. W. SITES, JR., C. A. KUCZYNSKI,

J. J. WIENS, AND T. W. REEDER. 2011. Phylogeny of Iguanian lizards inferred

from 29 nuclear loci, and a comparison of concatenated and species-tree

approaches for an ancient, rapid radiation. Molecular Phylogenetics and

Evolution 61:363-380.

VERZI, D. H., AND C. I. MONTALVO. 2008. The oldest South American Cricetidae

(Rodentia) and Mustelidae (Carnivora): Late Miocene faunal turnover in Central

Argentina and the Great American Biotic Interchange. Palaeogeography,

Palaeoclimatology, Palaeoecology 267:284-291.

114

Texas Tech University, Megan S. Keith, December 2015

VOGT, L. 2002. Weighting indels as phylogenetic markers of 18S rDNA sequences in

diptera and strepsiptera. Organisms Diversity and Evolution 2: 335-349.

VORONSTOV, N. N. 1959. [The system of hamsters (Cricetinae) in the sphere of th

world fauna and their phylogenetic relations.] Byulleten' Moskovskovo

Obshchestva Ispytatelei Prirody, Otdel Biologicheskii 64: 134-147 (in

Russian).

VOSS, R. S. AND A. V. LINZEY. 1981. Comparative gross morphology of male

accessory glands among Neotropical Muridae (Mammalia: Rodentia) with

comments on systeatic implications. Miscellaneous Publications, Museum of

Zoology, University of Michigan 159: 1-41.

WOODBURNE, M. O. 1987. A prospectus of the North American Mammal Ages. Pp.

285-290, in Cenozoic Mammals of North America (M. O. Woodburne, ed.).

University of California Press, Berkeley.

115

Texas Tech University, Megan S. Keith, December 2015

Table 4.1.—Specimens examined by taxon, genetic marker (Adh1-I2 - intron 2 of alcohol dehydrogenase, Cytb - cytochrome-b, Fgb- I7 - intron of the beta-fibrinogen, Dmp1- dentin matrix protein 1, Ghr- growth hormone receptor, and Rbp3 - interphotoreceptor retinoid binding protein) and grouped by genus. GenBank accession (top) and museum catalogue (bottom) numbers are given for each specimen. Museum acronyms are as follows: ASNHC (Angelo State Natural History Collection), BYU (Brigham Young University), CNMA (Colección Nacional de Mamíferos, Universidad Nacional Autónoma de México), CRD (Centro Interdisciplinario de Investigación para el Desarrollo Integral Regional, Unidad Durango, Mexico), FSH (University of Texas Medical Branch at Galveston), MSB (The Museum of Southwestern Biology), PGSC (Peromyscus Genetic Stock Center), ROM (Royal Ontario Museum), TCWC (Texas Cooperative Wildlife Collection), and TTU (Museum of Texas Tech University). If museum catalogue numbers were unavailable, specimens were referenced with the corresponding collector’s numbers or TK (special number of the Museum of Texas Tech University). ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Sigmodontinae Tribe Akodontini Akodon A. aerosus KT361504 KT965001 KT361512 KT950891 KT950916 TTU85216 TTU85216 TTU85216 TTU85216 TTU85216 TTU85216 Tribe Phyllotini Calomys C. callosus KT361506 KT965002 KT361514 KT950893 KT950918 TTU66553 TTU66553 TTU66553 TTU66553 TTU66553 TTU66553 Phyllotis P. haggardi KT361505 KT965005 KT361513 KT950892 KT950917 TTU85469 TTU85469 TTU85469 TTU85469 TTU85469 TTU85469 Tribe Oryzomyini Holochilus H. sciureus KT361503 EU074631 KT361511 KT950890 EU273418 TTU75634 TTU75634 TTU75634 TTU75634 TTU75634 TTU75634

116

Texas Tech University, Megan S. Keith, December 2015

Table 4.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Tribe Oryzomyini (cont.) Oryzomys O. palustris DQ207949 DQ185382 AY274219 AY269988 KT950889 KT950915 TK92140 TK92140 TTU49415 TTU49415 TK91240 TTU49415 Tribe Sigmodontini Sigmodon S. hispidus KT318181 AF155420 AY459371 AY269989 KT964999 EU635710 TTU80759 FSH33 TTU80759 TTU80759 TTU80759 TTU80759 Neotominae Tribe Neotomini Hodomys H. alleni AY817627 AF186801 AY274197 AY269968 KT950894 KT950919 TK45042 TK45042 TK45042 TK45042 TK45042 TK45042 Neotoma N. albigula AY817649 AF186803 KT361515 KT950895 KT950920 TTU76474 TTU76474 TTU76474 TTU76474 TTU76474 TTU76474

N. cinerea AY817635 AF186799 AY274197 AY269970 KT950896 KT950921 MSB121427 MSB121427 MSB121427 MSB121427 MSB121427 MSB121427

N. fuscipes AY817632 AF376479 KT361516 KT950897 KT950922 TTU81391 TTU81391 TTU81391 TTU79134 TTU81391 TTU81391

117

Texas Tech University, Megan S. Keith, December 2015

Table 4.1 (cont). ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Tribe Neotomini Neotoma (cont.) N. mexicana AY817646 AF294345 AY274200 AY269971 KT950898 JX910120 TTU79129 TTU79129 TTU79129 TTU79129 TTU79129 TTU79129

N. micropus AY817655 AF186827 KT361517 EF989753 EF989953 TTU80856 TTU80856 TTU80856 TTU80856 ROM114902 ROM114902 Xenomys X. nelsoni AY817628 AF307838 AY274201 AY269972 KT950899 KT950923 TTU37790 TTU37790 TTU37790 TTU37790 TTU37790 TTU37790 Tribe Peromyscini Peromyscus P. californicus AY994211 AF155393 FJ214697 EF989772 EF989873 TTU81275 TTU81275 TTU83291 TTU83291 PGSCIS1590 PGSCIS1590

P. eremicus AY994212 AY322503 FJ214699 EF989775 EF989876 TTU83249 TTU83249 TTU83249 TTU83249 BYU17952 BYU17952

P. crinitus AY994213 AY376413 KT375572 EF989773 EF989874 DSR6171 DSR6171 TTU108167 TTU108167 BYU16629 BYU16629

P. maniculatus AY994242 AY322508 FJ214708 EF989783 EF989884 TTU81622 TTU81622 TTU97830 TTU97830 ROM98941 ROM98941

118

Texas Tech University, Megan S. Keith, December 2015

Table 4.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Tribe Peromyscini Peromyscus (cont.) P. melanotis FJ214673 AF155398 FJ214711 EF989790 EF989891 CRD2025 CRD2025 CRD2025 CRD2025 PGSC25 PGSC25

P. gossypinus FJ214671 DQ973102 FJ214702 KT950900 JX910126 TTU80682 TTU80682 TTU80682 TTU80682 TTU80682 TTU80682

P. leucopus AY994241 DQ000483 FJ214706 EF989880 TTU101645 TTU101645 TTU101645 TTU101645 TTU101645 ROM101861

P. beatae AY994223 AF131921 FJ214696 KT950901 KT950924 GK3954 GK3954 TTU105037 TK93279 TTU105037 TK93279

P. levipes KT361507 AY322509 FJ214707 KT950902 JX910123 TTU82707 TTU82707 TTU105150 TK47819 TK47819 TK47819

P. schmidlyi KT318182 AY322524 FJ214718 KT950903 KT950925 TTU81703 TTU81703 TTU81617 TTU81617 TK72442 TTU81703

P. evides FJ214670 FJ214685 FJ214700 KT950904 TTU82696 TTU82696 TTU82696 TTU82696 TTU82696 TTU82696

P. attwateri AY994220 AF155384 AY274207 AY269978 KT950905 JX910128 TTU55688 TTU55688 TTU55688 TTU55688 TTU55688 TTU55688

119

Texas Tech University, Megan S. Keith, December 2015

Table 4.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Tribe Peromyscini Peromyscus (cont.) P. ochraventer FJ214676 JX910119 FJ214715 KT950910 JX910130 TTU104930 TTU104930 TTU104930 TTU104930 TTU104930 TTU104930

P. gratus AY994218 AY376421 FJ214703 KT950906 JX910129 TK46354 TK46354 TK46354 TK46354 TK46354 TK46354

P. furvus JX910116 KT965004 JX910113 KT950907 JX910124 FXG1168 FXG1168 FXG1168 FXG1168 FXG1167 FXG1168

P. megalops AY994217 FJ214709 DQ000475 KT950908 JX910127 TTU82712 TTU82712 TTU82712 TTU82712 TTU82712 TTU82712

P. pectoralis AY994221 DQ000476 FJ214716 KT950911 JX910131 TTU75575 TTU75575 TTU75575 TTU75575 TTU75575 TTU75575

P. mexicanus AY994236 AY376425 AY274210 AY269981 EF989794 EF989895 TTU82759 TTU82759 TTU82759 TTU82759 ROM113250 ROM113250

P. nudipes AY994238 FJ214687 FJ214713 EF989792 EF989893 TTU96972 TTU96972 TTU96972 TTU96972 ROM113216 ROM113216

P. hooperi FJ214672 DQ973103 FJ214704 KT950909 JX910125 TTU104425 TTU104425 TTU104425 TTU104425 TTU104425 TTU104425

120

Texas Tech University, Megan S. Keith, December 2015

Table 4.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Tribe Peromyscini (cont.) Peromyscus (cont.) P. melanophrys AY994216 AY322510 FJ214710 EF989789 EF989890 TTU75509 TTU75509 TTU75509 TTU75509 PGSCXZ1073 PSCXZ1073 Habromys H. lepturus AY994239 DQ973099 FJ214701 EF989742 EF989841 TTU82703 TTU82703 TTU82703 TTU82703 CNMA29970 CNMA29970 Megadontomys M. thomasi AY994208 AY195795 FJ214693 EF989749 EF989849 TK93388 TK93388 TK93388 TK93388 CNMA29186 CMNA29186 Neotomodon N. alstoni AY994210 AY195796 AY274202 AY269973 EF989751 EF989851 TK45302 TK45302 TK45309 TK45309 ASNHC1595 ASNHC1595 Onychomys O. arenicola JX910115 AY195793 AY274204 AY269975 EF989755 EF989855 TTU67559 TTU67559 TTU67559 TTU67559 ROM114904 ROM114904

O. leucogaster KT318183 AY264205 AY195794 AY269976 EF989758 EF989859 TK31705 TK31705 TK31705 TK31705 ASNHC4348 ASNHC4348 Osgoodomys O. banderanus AY994209 AF155383 AY274206 AY269977 EF989756 EF989857 TTU37754 TTU37754 TK45401 TK45401 ASNHC2664 ASNHC2664 Podomys P. floridanus AY994214 KT965003 FJ214724 KT950912 EF989878 TTU97866 TTU97866 TTU97868 TTU97868 TTU97868 TTU97866

121

Texas Tech University, Megan S. Keith, December 2015

Table. 4.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Tribe Reithrodontomyini Isthmomys I. pirrensis FJ214668 FJ214681 FJ214692 EF989746 EF989846 TTU39162 TTU39162 TTU39162 TTU39162 ROM116308 ROM116308 Reithrodontomys R. fulvescens AY994207 AF176257 AY274211 AY269982 EF989800 EF989901 TTU54898 TTU54898 TTU54898 TTU54898 ASNHC3465 ASNHC3465

R. megalotis KT950928 AF176248 KT375573 EF989808 EF989909 TK22460 TK22460 TK22460 TK22460 ASNHC2133 ASNHC2133

R. sumichrasti JX910117 AF176256 AY274212 EF989823 EF989924 TTU54952 TTU54952 TTU54952 TTU54952 ROM98383 ROM98983 Tribe Baiomyini Baiomys B. taylori KT361508 AY548469 AY274213 AY269983 EF989739 EF989838 TTU54633 TTU54633 TTU54633 TTU54633 ROM114886 ROM114886 Scotinomys S. teguina KT361509 JN851815 AY274214 AY269984 EF989827 EF989928 TTU104355 TTU104355 P048 P048 ROM101507 ROM101507 Tribe Ochrotomyini Ochrotomys O. nuttalli JX910114 AY195798 AY274203 AY269974 EF989761 EF989862 TCWC31929 TCWC31929 TCWC31929 TCWC31929 ROM113008 ROM113008

122

Texas Tech University, Megan S. Keith, December 2015

Table. 4.1 (cont.) ------Taxon Adh1-I2 Cytb Fgb-I7 Dmp1 Ghr Rbp3 ------Tylomyinae Tribe Nyctomyini Nyctomys N. sumichrasti KT361510 AY195801 AY274215 KT950913 KT950926 TTU84484 TTU84484 TTU88186 TTU88186 TTU88186 TTU105094 Tribe Tylomyini Ototylomys O. phyllotis AY817624 AY009789 AY274216 AY269985 EF989763 EF989864 ASNHC7236 FN32783 FN36287 FN32557 ROM95675 ROM95675 Tylomys T. nudicaudus AY817625 AF307839 AY274217 AY269986 KT950914 KT950927 TTU77530 TTU77530 TTU67347 TTU67347 TTU67347 TTU67347

123

Texas Tech University, Megan S. Keith, December 2015

Table 4.2.—Analyses were conducted using the Kimura 2-parameter model (Kimura 1980) in MEGA 6.06 (Tamura et al. 2013). The number of base substitutions per site from averaging over all sequence pairs between groups are shown. Each analysis involved 48 nucleotide sequences with the exception of the Dmp1 dataset (40 nucleotide sequences) in which all sequences that were <50% of the entire gene length were removed so that common sites could be compared. Genetic distances are presented as percentages (%).

Adh1-I2 Fgb-I7 Cytb Dmp1 Ghr Rbp3 Average Sigmodontinae vs. 17.6 17.7 23.3 13.8 9.4 8.4 15.0 Neotominae Sigmodontinae vs. Subfamily 20.6 18.6 25.7 10.0 10.0 8.7 15.6 Tylomyinae Neotominae vs. 13.9 16.3 23.7 10.2 9.4 6.7 13.4 Tylomyinae Neotomini vs. 13.2 16.5 27.8 8.3 8.5 8.6 13.8 Tylomyini Peromyscini vs. 12.8 17.9 27.0 11.4 9.5 6.1 14.0 Tylomyini Reithrodontomyini 15.4 18.0 27.6 11.3 9.2 7.5 14.8 vs. Tylomyini Baiomyini vs. Tribe 17.5 17.0 27.6 8.9 8.7 7.9 14.6 Tylomyini Ochrotomyini vs. 18.3 18.8 26.9 10.0 8.6 7.1 15.0 Tylomyini Average between tribe distances 13.4 11.6 19.9 10.0 5.8 6.2 11.2 (excluding the Tylomyini)

124

Texas Tech University, Megan S. Keith, December 2015

Figure 4.1.—Phylogenetic tree obtained from the Bayesian inference analysis of the combined dataset. Posterior probability values ≥95% are represented by an "*" above the supported node and clade assignments are listed below each node. This analysis resulted in a clade uniting the Neotominae and Tylomyinae (Clade I) and two supported clades representing rodents of the Subfamily Neotominae (Clade A) and the Subfamily Tylomyinae (Clade B).

125

Texas Tech University, Megan S. Keith, December 2015

Figure 4.2.—Maximum clade credibility tree showing divergence date estimates based on a combined analysis of the mitochondrial Cytb gene and 5 nuclear genes (Adh1-I2, Fgb-I7, Dmp1, Ghr, and Rbp3). Divergence date estimates are indicated along 126

Texas Tech University, Megan S. Keith, December 2015

the x-axis in millions of years. Error bars represent the 95% highest posterior density (HPD) for node height. Labels on the right indicate taxa belonging to each subfamily.

127

Texas Tech University, Megan S. Keith, December 2015

CHAPTER V

UTILIZATION OF MULTIPLE MOLECULAR MARKERS TO RESOLVE TRIBAL AFFILIATIONS WITHIN THE SUBFAMILY NEOTOMINAE

ABSTRACT Combined analyses of mitochondrial and nuclear DNA sequences have been analyzed in many recent studies to resolve taxonomic relationships at multiple levels within the Neotominae; however, interpretations of phylogenetic relationships within this group have been incongruent. Previous studies examined multi-locus genetic datasets to determine tribal affiliations in this group of rodents. Whereas most studies agreed on the recognition and composition of the tribes Neotomini, Baiomyini and

Ochrotomyini, the classification of Peromyscus and allied genera, Isthmomys, and

Reithrodontomys have varied depending upon the taxa included in the study and the molecular markers that were analyzed. Additionally, some studies recognized the tylomyine rodents as a tribe within the Neotominae whereas others suggested that these rodents comprised a subfamily basal to the Neotominae. A combined dataset of one mitochondrial (Cytb) and five nuclear genes (Adh1-I2, Fgb-I7, Dmp1, Ghr, and

Rbp3) was analyzed phylogenetically to test the number of tribes and tribal affiliations of genera within the Neotominae. Results of the Bayesian analysis indicated that five tribes should be recognized and divergence date estimates range from approximately

8.58 - 5.99 million years ago for each tribe.

128

Texas Tech University, Megan S. Keith, December 2015

INTRODUCTION Phylogenetic relationships and evolutionary histories of cricetid rodents have been difficult to determine due to the accumulation of few phylogenetically informative characters, a paucity of taxa in studies, and the lack of congruence among datasets and data analyses. More specifically, rodents of the subfamily Neotominae pose their own set of difficulties in systematic studies due to morphological similarity in some groups and the presence of cryptic species (Houseal et al. 1987; Schmidly et al. 1988; Riddle et al. 2000; Bradley et al. 2004; Bradley et al. 2014; Bradley et al.

2015), limited study specimens due to rarity in nature and museum collections (i.e.,

Isthmomys), and a rapid radiation event that occurred during the late Miocene that makes it difficult to place some genera (Bradley et al. 2004; Reeder and Bradley 2004;

Reeder et al. 2006).

Several studies have attempted to resolve phylogenetic relationships among members of this subfamily using morphological (Hooper and Musser 1964a, 1964b;

Carleton 1980; Musser and Carleton 2005), allozymic (Rogers et al. 2005), or karyotypic data (Baker and Mascarello 1969; Greenbaum et al. 1978a, 1978b;

Greenbaum and Baker 1978), as well as mitochondrial and nuclear DNA sequences

(Reeder et al. 2006; Bradley et al. 2007; Miller and Engstrom 2008 and numerous others). Various taxonomic arrangements have resulted, including several interpretations as to the number of tribes that should be recognized within this subfamily as well as relationships of genera within each tribe (Hooper and Musser

1964a, 1964b; Carleton 1980; Musser and Carleton 2005; Reeder et al. 2006; Miller and Engstrom 2008). 129

Texas Tech University, Megan S. Keith, December 2015

Most studies that have focused on this group of rodents provided support for four and in some cases five tribes; however no two studies have provided the same results on the classification of these rodents at the tribal level. In general, most studies agree on the identification and classification of rodents in the tribes Neotomini

(Hodomys, Neotoma, and Xenomys), Baiomyini (Baiomys and Scotinomys) (Carleton

1980; Musser and Carleton 2005; Reeder et al. 2006; Miller and Engstrom 2008) and

Ochrotomyini (Ochrotomys- Musser and Carleton 2005; Reeder et al. 2006; Miller and

Engstrom 2008). However, proposed phylogenetic relationships of other taxa in this subfamily have been inconsistent including the classification of Peromyscus and allied genera and Reithrodontomys into one tribe (Peromyscine/i- Carleton 1980; Reeder et al. 2006, Reithrodontomyini- Musser and Carleton 2005) or separate tribes

(Peromyscini- Peromyscus + allied genera and Reithrodontomyini- Isthmomys +

Reithrodontomys- Miller and Engstrom 2008). Additionally, the tylomyine rodents

(Nyctomys, Otonyctomys, Ototylomys and Tylomys) which traditionally were classified within this group have been recognized as both a tribe within the Neotominae

(Tylomyini- Carleton 1980, Reeder et al. 2006) and as constituting their own subfamily (Tylomyinae- Wilson and Reeder 2005, Miller and Engstrom 2008; Keith et al. in prep B).

Despite the wealth of knowledge available for this group of rodents, branching patterns and relationships within some tribal lineages remain unresolved. Carleton

(1980) included the most detailed morphological dataset for this group of rodents (79 characters for 75 taxa) and suggested the recognition of four tribes. However, it is

130

Texas Tech University, Megan S. Keith, December 2015 likely that repetitive bursts of speciation and a high level of homoplasy in morphological characters hinder delimitation of intra-familial relationships (Fabre et al. 2012). Reeder et al. (2006) attempted to resolve relationships within this group by examining the mitochondrial cytochrome-b gene (Cytb) and two nuclear genes (intron

7 of the beta-fibrinogen gene (Fgb-I7) and exon 6 of the dentin matrix protein 1 gene

(Dmp1)) and determined that five tribes should be recognized within this subfamily

(Neotomini, Baiomyini, Peromyscini, Ochrotomyini and Tylomyini). Subsequently,

Miller and Engstrom (2008) utilized the mitochondrial Cytb gene and two different nuclear genes (exon 10 of the growth hormone receptor gene (Ghr) and exon 1 of the interphotoreceptor retinoid-binding protein gene (Rbp3)) to attempt to achieve increased resolution of tribal affiliations within this group. Miller and Engstrom

(2008) also determined that five tribes should be recognized within this group, however they did not recognize the tylomyine rodents as belonging in the subfamily

Neotominae and recognized a separate Peromyscini (Peromyscus and allied genera ) and Reithrodontomyini (Reithrodontomys + Isthmomys).

Combined analyses have been beneficial in studies examining taxa in which single mitochondrial markers are insufficient for resolving higher level hierarchical relationships. Mitochondrial genes often increase resolution at the tips of the a phylogeny while slower-evolving nuclear genes increase resolution toward the base of a phylogeny (Bull et al. 1993; Adkins et al. 2001; Pereira et al. 2002; Platt et al. 2015).

In addition to increasing resolution at different hierarchical levels, it is expected that a multilocus signal will overcome noise attributable to homoplasy (Barrett et al. 1991;

131

Texas Tech University, Megan S. Keith, December 2015 de Queiroz 1993; Adkins et al. 2001), overcome stochastic lineage sorting (Rokas et al. 2003) and reveal hidden relationships (Gatesy et al. 2004). This study utilizes the analysis of a combined dataset consisting of genetic markers from the two most recent

Neotominae tribal studies (one mitochondrial and four nuclear genes- Reeder et al.

2006; Miller and Engstrom 2008) and information for an additional nuclear gene

(intron 2 of the alcohol dehydrogenase gene (Adh1-I2)) to resolve tribal affiliations and branching patterns for 15 genera in the subfamily Neotominae (Nelsonia was excluded due to inability to access tissues).

MATERIALS AND METHODS Sampling.—The taxonomic sampling for this study included 42 species from

15 genera of rodents classified within the Neotominae and three taxa from the

Tylomyinae. Sigmodon hispidus and Oryzomys palustris were chosen as outgroup taxa. Six genes were examined to resolve tribal affiliations within the Neotominae: the mitochondrial Cytb gene, and five nuclear genes- Adh1-I2, Fgb-I7, Dmp1, Ghr, and Rbp3. Sequences were either generated in this study or obtained from GenBank.

Sequencing.—Genomic DNA was isolated from approximately 0.1g of frozen liver tissue using the Puregene DNA isolation kit (Gentra, Minneapolis, Minnesota) or the Qiagen DNeasy Blood and Tissue Kit (Qiagen, Valencia, California). DNA fragments were amplified using polymerase chain reaction (PCR, Saiki et al. 1988).

Methods for PCR amplification and sequencing of each gene followed the protocols from Keith et al. (in prep A). PCR products were purified using the QIAquick PCR purification kit (Qiagen, Valencia, California) or ExoSAP-IT (USB Products,

132

Texas Tech University, Megan S. Keith, December 2015

Cleveland, Ohio) and PCR amplicons were sequenced using ABI Prism Big Dye

Terminator v3.1 ready reaction mix (Applied Biosystems, Foster City, California).

Nucleotide sequences were determined on an ABI 3100 or 3130-Avant automated sequencer (Applied Biosystems, Foster City, California). Sequences were proofed using Sequencher 4.10 software (Gene Codes, Ann Arbor, Michigan) and aligned using MEGA v6.06 software (Tamura et al. 2013). All DNA sequences were deposited in GenBank. GenBank accession numbers and museum catalog numbers are listed in Table 5.1.

Data Analysis.—An attempt was made to obtain mitochondrial and nuclear sequences from a single individual, but this was not possible for all nuclear genes. In these cases, sequences from conspecific individuals were used to complete the dataset.

Nucleotide positions were treated as unordered, discrete characters with six possible states: A, C, G, T, gaps (-) or missing (?) for each marker. Alignment of nuclear intron sequences required hypothesized gaps to represent insertion or deletion events.

Sequences were aligned using MUSCLE software (Edgar 2004) in MEGA v6.06

(Tamura et al. 2013). Indels were coded as binary data for Adh1-I2 and Fbg-I7 using simple indel coding (SIC- Simmons and Ochoterena 2000) methods by SeqState software (Müller 2005), resulting in a combined dataset consisting of 5,578 characters.

Molecular models of evolution were determined using jModelTest v2.1.6

(Darriba et al. 2012) and the Akaike information criterion. jModelTest v2.1.6

(Darriba et al. 2012) identified HKY+G for Adh1-I2 and Ghr, GTR+G for Fgb-I7 and

Dmp1, and GTR+I+G for Cytb and Rbp3 as the best-fit nucleotide substitution models.

133

Texas Tech University, Megan S. Keith, December 2015

Phylogenetic analysis was conducted for a combined dataset in which sequences were partitioned to be analyzed under the appropriate nucleotide substitution model.

Bayesian inference (BI) methods were utilized to estimate a phylogeny and generate posterior probability values using MrBayes 3.2.5 (Ronquist et al. 2012). Each analysis included 10 million generations and a sample frequency of every thousandth generation. Values ≥ 95% were viewed as supportive.

Molecular Dating.—BEAST v1.8.2 (Drummond et al. 2012) was used to estimate divergence dates between Neotominae tribes. Complete datasets for each gene were used, but the data was partitioned to allow independent modeling of each dataset. Models of substitution were determined using the program jModelTest v2.1.6

(Darriba et al. 2012). MEGA 6.06 (Tamura et al. 203) was used to determine whether each gene was behaving under a strict molecular clock. Given only a single individual from each species was sampled, a Yule tree prior was selected.

To reliably assess divergence at the tribe level, fossil calibrations were used at several taxonomic levels. Calibration points were as follows: tribe Neotomini (~5.3 million years [MYA]; Reynolds 1991), genus Neotoma (~2.76 MYA; Hornsby and

Matocq 2012), genus Reithrodontomys (~1.8 MYA; Cassiliano 1999), and the leucopus/maniculatus group (~1.8 MYA; Dalquest 1962; Karow et al. 1996). These calibration points were placed to reflect calculated first appearances of sampled taxa in the fossil record (Behrensmeyer and Turner 2013) and to account for known complexities in the resolution of the phylogeny (e.g., see Platt et al. 2015). A lognormal distribution was used for each calibration with upper limits modified to

134

Texas Tech University, Megan S. Keith, December 2015 reflect the boundary of the associated land mammal age (Woodburne 1987). To examine various configurations, test runs of 1.0 X 107 generations with a 10% burn-in were used and compared using Bayes Factors (Kass and Raftery 1995; Suchard et al.

2001). Final estimates were determined by analyzing combined log and tree files from four runs of 5.0 x 107 generations. Sufficient mixing, convergence stability, and effective sample size was examined using the program Tracer v1.5 (Drummond and

Rambaut 2008).

RESULTS Bayesian analysis of the combined dataset resulted in a highly resolved and statistically supported phylogeny. The resulting topology indicated five major clades within Neotominae. Clades A-E (Fig. 5.1) represented all potential tribes that have been recognized within the Neotominae by previous studies. Clade A was basal to the rest of the Neotominae and consisted of Hodomys, Neotoma, and Xenomys which together form the Neotomini, clade B consisted of Peromyscus and allied genera representing the Peromyscini, clade C united Isthmomys and Reithrodontomys to form the Reithrodontomyini, clade D consisted of Baiomys and Scotinomys which compose the Baiomyini, and clade E represented the monotypic Ochrotomyini. Each of these clades was statistically supported with a posterior probability value of 100%.

Molecular Dating.—A strict molecular clock was rejected for all datasets.

The Yule birth rate was estimated at 0.256 (95% highest posterior density [HPD]:

0.17--0.34). Mean rates of evolution (as substitutions per site per million years) were

0.009 for Adh1-I2 (95% HPD: 0.007--0.01), 0.07 for Cytb (95% HPD: 0.06--0.09),

135

Texas Tech University, Megan S. Keith, December 2015

0.008 for Dentin (95% HPD; 0.006--0.009), 0.008 for Fgb-I7 (95% HPD: 0.007--

0.01), 0.004 for Ghr (95% HPD: 0.003--0.005), and 0.004 for Rbp3 (95% HPD:

0.003--0.005). Divergence date estimates (Fig. 5.2) suggested that tribes within the subfamily Neotominae began separating ~9.1 MYA (95% HPD: 7.54--11.0), during the Late Miocene. This initial split separated the Neotomini from the other neotomine tribes. The tribe Ochrotomyini diverged ~8.58 MYA (95% HPD: 7.13--10.36) from a clade consisting of the tribes Baiomyini, Peromyscini, and Reithrodontomyini.

Subsequently, the Baiomyini diverged from a clade consisting of the Peromyscini and

Reithrodontomyini ~8.1 MYA (95% HPD: 6.68--9.81). The final split of the neotomine tribes occurred ~5.99 MYA (95% HPD: 4.92--7.27), separating the tribes

Reithrodontomyini and Peromyscini.

DISCUSSION The use of a combined dataset to resolve taxonomic relationships within this group of rodents is not a novel approach; however, inconsistencies in selected ingroup taxa and disparities in genetic datasets have resulted in differing opinions on the tribal- level classification of the Neotominae (summarized in Reeder et al. 2006; Miller and

Engstrom 2008; Platt et al. 2015). The results of the Bayesian analysis reported herein can be interpreted in two ways: 1) five tribes (Clades A-E, Fig. 5.1) should be recognized within the Neotominae following the classification suggested by Miller and Engstrom (2008) with a separate Peromyscini and Reithrodontomyini, or 2) clades

B and C (Fig. 5.1) together represent the Peromyscini and all other clades (A, D, and

E) represent individual tribes as described by Reeder et al. (2006).

136

Texas Tech University, Megan S. Keith, December 2015

Divergence within the Neotominae began ~9.1 MYA (95% HPD: 7.54--11.0) during the Late Miocene with the separation of the taxa classified within the

Neotomini from the other neotomine tribes. The tribe Ochrotomyini diverged ~8.58

MYA (95% HPD: 7.13--10.36) from a clade consisting of the tribes Baiomyini,

Peromyscini, and Reithrodontomyini. Subsequently, the Baiomyini diverged from the

Peromyscini and Reithrodontomyini ~8.1 MYA (95% HPD: 6.68--9.81). The final divergence within the Neotominae occurred ~5.99 MYA (95% HPD: 4.92--7.27), separating Peromyscus and allied genera and the clade uniting Isthmomys and

Reithrodontomys.

Tribal Affiliations and Systematic Conclusions.—Previous studies have classified the tylomyine rodents as a tribe within the Neotominae (Carleton 1980;

Reeder et al. 2006). However, recent studies have indicated that these rodents should be classified in their own subfamily (Tylomyinae- Steppan et al. 2004 and Keith et al. in prep B) and that taxonomic assignment was consistent with this study. The results provided herein indicated that the rodents of the subfamily Tylomyinae (Clade I- Fig.

5.1) were basal to the Neotominae (Clade II).

The results supported the recognition of five tribes: Neotomini, Peromyscini,

Reithrodontomyini, Baiomyini, and Ochrotomyini. The tribe Neotomini (Fig. 5.1) consists of Hodomys, Neotoma, and Xenomys. Although Nelsonia was not examined in this study, previous evidence supports its inclusion within the Neotomini (Carleton

1980; Goldman 1910; Musser and Carleton 2005). Peromyscus and its allied genera

(Habromys, Megadontomys, Neotomodon, Onychomys, Osgoodomys, and Podomys)

137

Texas Tech University, Megan S. Keith, December 2015 are affiliated with the Peromyscini whereas Isthmomys and Reithrodontomys constitute the Reithrodontomyini following the suggested classification of Miller and Engstrom

(2008). Additionally, evidence presented in previous studies and the results herein, indicate that Baiomys and Scotinomys together form the Baiomyini and Ochrotomys forms a monotypic tribe, the Ochrotomyini.

ACKNOWLEDGMENTS The authors would like to thank Emma K. Roberts and Nicté Ordóñez-Garza for comments on earlier versions of this manuscript. We thank R. J. Baker, H. Garner, and K. McDonald (Natural Science Research Laboratory, Museum of Texas Tech

University), Centro Interdisciplinario de Investigación para el Desarrollo Integral

Regional, Unidad Durango, Mexico and Duke S. Rogers for access and assistance in obtaining tissues. The Texas Academy of Science and Texas Tech University

Association of Biologists contributed funding necessary to complete this project.

138

Texas Tech University, Megan S. Keith, December 2015

LITERATURE CITED ADKINS, R. M., E. L. GELKE, D. ROWE, AND R. L. HONEYCUTT. 2001. Molecular

phylogeny and divergence time estimates for major rodent groups: evidence

from multiple genes. Molecular Biology and Evolution 18: 777-791.

BAKER, R. J., AND J. T. MASCARELLO. 1969. Karyotypic analyses of the genus

Neotoma (Cricetidae, Rodentia). Cytogenetics 8:187-198.

BARRETT, M., M. J. DONOGHUE, AND E. SOBER. 1991. Against consensus. Systematic

Zoology 40: 486-493.

BEHRENSMEYER, A. K., AND A. TURNER. 2013. Taxonomic occurrences of Cricetidae

recorded in the Paleobiology Database. Fossilworks. http://fossilworks.org.

BRADLEY, R. D., D. S. CARROLL, M. L. HAYNIE, R. M. MARTINEZ, M. J. HAMILTON,

AND C. W. KILPATRICK. 2004. A new species of Peromyscus from western

Mexico. Journal of Mammalogy 85: 184-1193.

BRADLEY, R. D., N. D. DURISH, D. S. ROGERS, J. R. MILLER, M. D. ENGSTROM, and C.

W. KILPATRICK. 2007. Toward a molecular phylogeny for Peromyscus:

evidence from mitochondrial Cytochrome-b sequences. Journal of

Mammalogy 88:1146-1159.

BRADLEY, R. D., N. ORDÓÑEZ-GARZA, C. G. SOTERO-CAIO, H. M. HUYNH, C. W.

KILPATRICK, L. I. IÑIGUEZ-DÁVALOS, AND D. J. SCHMIDLY. 2014.

Morphometric, karyotypic and molecular evidence for a new species of

Peromyscus (Cricetidae: Neotominae) from Nayarit, Mexico. Journal of

Mammalogy 95:176-186.

139

Texas Tech University, Megan S. Keith, December 2015

BULL, J. J., J. P. HUELSENBECK, C. W. CUNNINGHAM, D. L. SWOFFORD, AND P. J.

WADDELL. 1993. Partitioning and combining data in phylogenetic analysis.

Systematic Biology 42: 384-397.

CARLETON, M.D. 1980. Phylogenetic relationships in neotomine-peromyscine

rodents (Muroidea) and a reappraisal of the dichotomy with New World

Cricetinae. Miscellaneous Publications, Museum of Zoology, University of

Michigan 157: 1-146.

CASSILIANO, M. L. 1999. Biostratigraphy of Blancan and Irvingtonian mammals in the

Fish Creek-Vallecito section, southern California, and a review of the

Blancan–Irvingtonian boundary. Journal of Vertebrate Paleontology 19:169-

186.

DALQUEST, W. W. 1962. The Good Creek Formation, Pleistocene of Texas, and its

fauna. Journal of Paleontology 36:568-582.

DARRIBA, D., G.L. TABOADA, R. DOALLO, AND D. POSADA. 2012. jModelTest 2:

more models, new heuristics and parallel computing. Nature Methods 9: 772.

DE QUEIROZ, A. 1993. For consensus (sometimes). Systematic Biology 42: 368-372.

DRUMMOND, A. J., and A. RAMBAUT. 2008. Tracer v1.5.

http://tree.bio.ed.ac.uk/software/tracer/.

DRUMMOND, A. J., M. A. SUCHARD, D. XIE, and A. RAMBAUT. 2012. Bayesian

Phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and

Evolution 29:1969-1973.

140

Texas Tech University, Megan S. Keith, December 2015

EDGAR, R. C. 2004. MUSCLE: multiple sequence alignment with high accuracy and

high throughput. Nucleic Acids Research 32: 1792-1797.

FABRE, P.-H., L. HAUTIER, D. DIMITROV, AND E. J. P. DOUZERY. 2012. A glimpse on

the pattern of rodent diversification: a phylogenetic approach. BMC

Evolutionary Biology 12:88.

GATESY J., R. H. BAKER AND C. HAYASHI. 2004. Inconsistencies in arguments for the

supertree approach: supermatrices versus supertrees of Crocodylia. Systematic

Biology 53:342–355.

Goldman, E.A., 1910. Revision of the wood rats of the genus Neotoma. North

American Fauna 31: 1–124.

.GREENBAUM, I. F. AND R. J. BAKER. 1978. Determination of the primitive karyotype

for Peromyscus. Journal of Mammalogy 59: 820-834.

GREENBAUM, I. F., R. J. BAKER, AND J. H. BOWERS. 1978a. Chromosomal homology

and divergence between sibling species of deer mice: Peromyscus maniculatus

and P. melanotis (Rodentia, Cricetidae). Evolution 32:334-341.

GREENBAUM, I. F., R. J. BAKER, AND P. R. RAMSEY. 1978b. Chromosomal evolution

and its implication concerning the mode of speciation in three species of deer

mice of the genus Peromyscus. Evolution 32:646-654.

HOOPER, E. T. AND G. G. MUSSER. 1964a. The glans penis in Neotropical cricetines

(family Muridae), with comments on classification of muroid rodents.

Miscellaneous Publications, Museum of Zoology, University of Michigan

123:1-57.

141

Texas Tech University, Megan S. Keith, December 2015

HOOPER, E. T. AND G. G. MUSSER. 1964b. Notes on classification of the rodent genus

Peromyscus. Occasional Publications of the Museum of Zoology, University

of Michigan 635:1-13.

HORNSBY, A. D. AND M. D. MATOCQ. 2011. Differential regional response of the

bushy-tailed woodrat (Neotoma cinerea) to late Quaternary climate change.

Journal of Biogeography 39: 289-305.

HOUSEAL, T. W., I. F. GREENBAUM, D. J. SCHMIDLY, S. A. SMITH, AND K. M. DAVIS.

1987. Karyotypic variation in Peromyscus boylii from Mexico. Journal of

Mammalogy 68: 281-296.

KAROW, P., G. MORGAN, R. PORTELL, E. SIMMONS, and K. AUFFENBERG. 1996.

Middle Pleistocene (early Rancholabrean) vertebrates and associated marine

and non-marine invertebrates from Oldsmar, Pinellas County, Florida. Pp. 97-

133 in Palaeoecology and Palaeoenuironments of Late Cenozoic mammals

Tributes to the Career of C S (Rufus) Churcher (K. Stewart and K. Seymour,

eds.). University of Tronto Press, Toronto, Canada.

KASS, R. E., and A. E. RAFTERY. 1995. Bayes Factors. Journal of the American

Stastistical Association 90:773-795.

Keith, M. S., J. Salazar-Bravo, R. D. Bradley, and R. N. Platt II. In prep A.

Molecular data indicate that Isthmomys is not aligned with Peromyscus. Pp.

62-95.

142

Texas Tech University, Megan S. Keith, December 2015

Keith, M. S., C. W. Thompson, and R. D. Bradley. In prep B. Phylogenetic

Relationships and Divergence of the Subfamily Tylomyinae (Rodentia:

Cricetidae) as determined by a multilocus dataset.