Louisiana State University LSU Digital Commons

LSU Doctoral Dissertations Graduate School

2011 Comparative biogeography of the arid lands of central Mexico Jesus Abraham Fernandez Louisiana State University and Agricultural and Mechanical College, [email protected]

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_dissertations

Recommended Citation Fernandez, Jesus Abraham, "Comparative biogeography of the arid lands of central Mexico" (2011). LSU Doctoral Dissertations. 3479. https://digitalcommons.lsu.edu/gradschool_dissertations/3479

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Doctoral Dissertations by an authorized graduate school editor of LSU Digital Commons. For more information, please [email protected].

COMPARATIVE BIOGEOGRAPHY OF THE ARID LANDS OF CENTRAL MEXICO

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

in

The Department of Biological Sciences

by Jesús Abraham Fernández Fernández Licenciado en Biología, Universidad Autónoma de Tlaxcala, 2000 Maestría en Ciencias, Universidad Nacional Autónoma de México, 2006 December 2011

DEDICATION

This work is dedicated to the memories of my family and friends: Julia Mendoza

(R.I.P.), Carmelo Fernández (R.I.P.), Heriberta Fernández (R.I.P.), Esperanza Fernández,

Carmen Fernández, Virginia Fernández, Adriana Fernández, Pastor Fernández (R.I.P.),

Raúl Fernández (R.I.P.), Erasmo Fernández and Miguel Fernández (R.I.P.); Federico

Bahena Romero (R.I.P.); Federico García (R.I.P.), Cinthia Campusano, and Florencia

García.

! ii ACKNOWLEDGEMENTS

I want to thank my Advisor Mark Hafner, for all his help and guidance. Mark gave me all the patience, time and training necessary to succed in my field of research.

The members of my committee, Chris Austin, Robb Brumfield, Chris Green, and Mike

Hellberg were unvaluable providing advise, and always having time to give a word of support.

The day I went to the field with the Hafner brothers and collaborators, I knew my life would be different, I want to express my gratitude to Mark, Darcey, John and Dave

Hafner, and Jessica Light and Brett Riddle for all their help and support since the days in

Mexico, to the last of the Louisiana days.

I’ve had the pleasure of making friends and interact on a daily basis with great staff, faculty, and graduate students in the Museum of Natural Science and the

Department of Biological Sciences, special thanks to Steve Cardiff, Prosanta

Chakrabarty, Elizabeth Derryberry, Donna Dittman, Tammie Jackson, Prissy Milligan,

Tom Moore, Fred Sheldon, Adrienne Steele, Richard Stevens, and Bill Wischusen.

The following persons kindly donated tissues and specimens for use in this study:

Joe Cook, University of New Mexico, Museum of Southwestern Biology; Roxana Acosta and Juan Morrone, Museo de Zoología “Alfonso L. Herrera,” Facultad de Ciencias,

Universidad Nacional Autónoma de México (U.N.A.M.); Fernando Cervantes,Yolanda

Hortelano, and Julieta Vargas of the Colección Nacional de Mamíferos, Instituto de

Biología, UNAM; Mark Hafner and Jessica Light of the Louisiana State University,

Museum of Natural Science; Brett Riddle and Stacy Mantooth, University of Nevada, Las

Vegas; Dave Hafner, New Mexico Museum of Natural History; Duke Rogers, Monty L.

! iii Bean Life Science Museum, Brigham Young University; José Ramírez-Pulido and Noé.

González, Universidad Autónoma Metropolitana-Iztapalapa; Juan Carlos López Vidal,

Colección de Mamíferos, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico

Nacional; María del Carmen Corona V., Departamento de Agrobiología, Universidad

Autónoma de Tlaxcala; Robert Baker and Heath Garner, Museum of Texas Tech

University; Fred Stangl, Midwestern State University; Veronica Rosas of Universidad de

Guadalajara; and Robert Fisher, National Museum of Natural History.

The following persons kindly helped with laboratory protocols: Marjorie Matocq and C. R. Feldman; L. Márquez V., Automatic Sequencer Facility, Instituto de Biología,

U.N.A.M., Posgrado de Ciencias Biológicas, U.N.A.M.; and Roxana Acosta, Jorge

Falcón, and Juan Carlos Windfield kindly helped during fieldwork in México.

Thanks to Fernando Cervantes, Dave Hafner, and Mark Hafner, and B. Lim for their critical evaluations of manuscripts., and the anonymous reviewers for their critical evaluation of manuscripts.

Thank you very much to all the graduate students that generously gave their friendship, this tie will last for the rest of our lives: Nati Aristizabal, Gustavo Bravo, Reid

Brennan, Clare Brown, Laura Brown, Tony Chow, Santiago Claramunt, Andres Cuervo,

Adriana Dantin, Valerie Derouen, Sandra Galeano, Amber Gates, Dency Gawin, Richard

Gibbons, Heather and Nathan Jackson, Mike Harvey, Sarah Hird, Dan Lane, James

Maley, Verity Mathis, John McCormack, Caleb McMahan, John McVay, Fabi Mendoza

Luciano Naka, Lori Patrick, Whitney Pilcher, Carlos Prada, Noha Reid, Eric Rittmeyer,

Maria Sagot, César Sánchez, Sebas Tello, Yalma Vargas and Amanda Zellmer.

! iv TABLE OF CONTENTS

DEDICATION ...... ii

ACKNOWLEDGEMENTS ...... iii

LIST OF TABLES ...... vii

LIST OF FIGURES ...... viii

ABSTRACT ...... x

CHAPTER 1. INTRODUCTION ...... 1 1.1 Literature Cited ...... 10

CHAPTER 2. MOLECULAR PHYLOGENETICS AND BIOGEOGRAPHY OF A RARE ENDEMIC: NELSON’S WOODRAT (NEOTOMA LEUCODON NELSONI) IN THE ORIENTAL BASIN OF MEXICO. 2.1 Introduction ...... 16 2.2 Materials and Methods ...... 19 2.3 Results ...... 24 2.4 Discussion ...... 29 2.5 Literature Cited ...... 36

CHAPTER 3. PHYLOGENETICS AND BIOGEOGRAPHY OF THE MICROENDEMIC RODENT XEROSPERMOPHILUS PEROTENSIS (PEROTE GROUND SQUIRREL) IN THE ORIENTAL BASIN OF MEXICO. 3.1 Introduction...... 43 3.2 Materials and Methods ...... 47 3.3 Results ...... 51 3.4 Discussion ...... 56 3.5 Literature Cited ...... 59

CHAPTER 4. MOLECULAR PHYLOGENETICS AND TEMPORAL CONTEXT FOR THE DIVERSIFICATION OF THE SOUTHERN ROCK DEERMOUSE, PEROMYSCUS DIFFICILIS (RODENTIA: CRICETIDAE). 4.1 Introduction ...... 64 4.2 Materials and Methods ...... 67 4.3 Results ...... 71 4.4 Discussion ...... 74 4.5 Literature Cited ...... 81

CHAPTER 5. MOLECULAR SYSTEMATICS AND BIOGEOGRAPHY OF THE MEXICAN ENDEMIC KANGAROO RAT, DIPODOMYS PHILLIPSII

! v (RODENTIA: HETEROMYIDAE). 5.1 Introduction ...... 88 5.2 Materials and Methods ...... 91 5.3 Results ...... 97 5.4 Discussion ...... 102 5.5 Literature Cited ...... 109

CHAPTER 6. CONCLUSIONS ...... 116

APPENDIX 3.1 ...... 122

APPENDIX 4.1 ...... 127

APPENDIX 4.2 ...... 131

APPENDIX 5.1 ...... 134

APPENDIX 5.2 ...... 140

VITA ...... 142

! vi LIST OF TABLES

Table 2.1.—Cytochrome-b sequence divergence values (Kimura 2-parameter model) between and within species of the Neotoma micropus and N. floridana species group ...... 26

Table 3.1.—Mean percent cytochrome-b sequence divergence values (Kimura 2- parameter model) among species of the genus Xerospermophilus and Cynomys ...... 53

Table 4.1.—Percent sequence divergence in the cytochrome-b gene (Kimura 2- parameter model) among 8 taxa of Peromyscus, including the 5 currently recognized subspecies of P. difficilis ...... 72

Table 5.1.—Mean percent cytochrome-b sequence divergence values (Kimura 2- parameter model) and ranges (in parentheses) between and within the 2 major clades of Dipodomys phillipsii, D. elator, and D. merriami ...... 101

Table 6.1.— Overview of taxonomic conclusion, divergence values between clades of interest, and mean estimates of divergence time for clades of interest (in million of years) ...... 121

! vii LIST OF FIGURES

Fig. 1.1.—Map of Mexico showing main geographic features discussed in this dissertation ...... 3

Fig. 1.2.—Map of Central Mexico showing the Oriental Basin and surrounding volcanoes ...... 5

Fig. 2.1.—Geographical distribution of the Neotoma micropus species group ...... 17

Fig. 2.2.—Map of central Mexico showing the known geographic distribution of Nelson’s woodrat, Neotoma leucodon nelsoni) ...... 20

Fig. 2.3.—Cladogram showing the placement of Nelson’s woodrat, N. leucodon nelsoni, within the genus Neotoma ...... 27

Fig. 2.4.—Cladogram showing placement of Nelson’s woodrat, N. leucodon nelsoni, within the genus Neotoma ...... 28

Fig. 2.5.—Phylogeny and chronogram of selected species of the genus Neotoma (and outgroups), and estimated divergence times calculated from *BEAST analysis ...... 30

Fig. 3.1.—Geographical distribution of pygmy ground squirrels of the genus Xerospermophilus ...... 44

Fig. 3.2.—Relationships among major clades of Xerospermophilis based on maximum likelihood and Bayesian analyses of Cytochrome-b sequences ...... 52

Fig. 3.3.—Relationships among major clades of Xerospermophilus based on a maximum likelihood analysis of Cytb, 12S, GHR, and IRBP sequences ...... 55

Fig. 4.1.—Geographical distribution of Peromyscus difficilis and P. nasutus in Mexico and the United States ...... 65

Fig. 4.2.—Phylogenetic relationships among the 5 subspecies of P. difficilis, P. nasutus, and allied taxa based on a maximum likelihood (ML) analysis of cytochrome- b sequences ...... 73

Fig. 4.3.—Phylogenetic relationships among selected specimens of P. difficilis and P. nasutus based on a partitioned maximum likelihood (ML) analysis of 2 mitochondrial genes and 2 nuclear genes ...... 75

Fig. 4.4.—Estimates of divergence times in P. difficilis, P. felipensis, and close relatives calculated from *BEAST analysis ...... 78

! viii Fig. 5.1.—Map of central Mexico showing the geographic distribution of Phillips’ kangaroo rat, Dipodomys phillipsii ...... 89

Fig. 5.2.—Phylogram showing placement of Phillips’ kangaroo rat, D. phillipsii, within the genus Dipodomys ...... 98

Fig. 5.3.—Phylogram showing placement of Phillips’ kangaroo rat, D. phillipsii, within allied species of Dipodomys ...... 100

! ix ABSTRACT

Most biogeographic studies on the Mexican biota have assumed that the dramatic climate cycles of the Pleistocene epoch and the prominence of the Trans-Mexico

Volcanic Belt (TMVB) have played major roles in the origin and diversification of species. Here I studied the pylogenetics and biogeography of four codistributed rodent species. In each case, I have generated a phylogenetic hypothesis for the taxon and allied species using two mitochondrial (Cytochrome-b and 12S), and two nuclear genes (GHR and IRBP), I recommended appropriate taxonomic changes, and generated a temporal framework to identify events that may have produced the phylogenetic pattern.

Nelson’s woodrat Neotoma nelsoni and the Perote ground squirrel

Xerospermophilus perotensis, were confirmed as having their closest relatives in the

Mexican Plateau. My findings also confirmed that N. nelsoni and X. perotensis are genetically well-differentiated from their sister taxa. Genetic distances in combination with low levels of morphological differentiation suggest that they should be recognized only at the subspecific level as N. leucodon nelsoni and X. spilosoma perotensis.

Molecular estimates of divergence times suggested that N. l. nelsoni and X. s. perotensis diverged from their sister taxa to the north during early Pleistocene times.

The rock mouse Peromyscus difficilis was divided into two well-supported clades, a northern clade including the subspecies P. d. difficilis and P. d. petricola, and a southern clade containing the subspecies amplus, felipensis, and saxicola. Molecular- based estimates of divergence times suggested that separation of these clades occurred in the Pleistocene.

! x My study of the Phillips’ kangaroo rat, Dipodomys phillipsii, revealed a biogeographic pattern different from that seen for other taxa. D. phillipsii was divided into two well-supported clades: one distributed on the Mexican Plateau, and a southern clade in the TMVB. Several lines of evidence supported my decision to return the

Mexican Plateau clade of D. phillipsii to full species status as D. ornatus. My study showed that D. phillipsii, D. ornatus, D. elator, and D. merriami form a well-supported clade of kangaroo rats, but I was unable to resolve relationships among these four species. My molecular-based analyses of divergence times suggests that D. phillipsii, D. ornatus, D. elator, and D. merriami diverged in mid-Pliocene times, probably in or near the Mexican Plateau. Unlike the Pleistocene divergence dates reported in previous chapters this Pliocence divergence suggests that the morphotectonic processes that gave rise to the Trans-Mexico Volcanic Belt may have influenced early diversification in

Mexican species of Dipodomys.

! xi CHAPTER 1

INTRODUCTION

Some of the oldest questions in biology focus on the origin of biological diversity.

How do new species evolve? How do geography, geology, and climate influence the rate and timing of biological diversification? Do geographic barriers and environmental fluctuations affect all organisms equally? One of the disciplines that addresses these kinds of questions is biogeography. Biogeography is the study of the causes of and limitations on the geographic distribution of organisms through space and time (Nelson and Platnick 1981).

From a classical point of view, biogeography has been split into the subfields of historical biogeography and ecological biogeography. Historical biogeographers study the effect of past geographic phenomena on the distributional patterns and diversification of species, usually at or above the species level. In contrast, ecological biogeographers focus on the effects of climate fluctuations on the early diversification (incipient speciation) of populations, emphasizing demographic parameters at and below the species level (Nelson and Platnick 1981).

Populations of a species are not distributed evenly in space and time, and their distributional ranges can expand or contract directly or indirectly by environmental fluctuations, geological events, and other changes in a dynamic landscape. Through long periods of time, those changes may result in disjunct populations, and the combined affects of natural selection and genetic drift may result in differentiation between these isolated populations and their source populations. When differentiated populations of multiple species are exclusive to a particular geographic area, that area is known as an

! 1 area of endemism. Other definitions of “areas of endemism” focus on the area’s geographic delimitation by natural barriers or the distributional congruence of several species (Harold and Mooi 1994; Hausdorf 2002; Platnick 1991). The concept of endemism is central to biogeography and biological conservation and areas with high numbers of endemic species, or “biodiversity hotspots,” commonly are included in protected area networks (Myers et al. 2000).

México is widely known as a biologically megadiverse country and a biodiversity hotspot (Lamoreaux et al. 2006). The diverse Mexican biota is the historical product of complex interactions between an ever-changing topography and myriad ecological and environmental factors (Velasco de Leon et al. 2007). In central Mexico, the Mexican

Plateau and the highlands and arid valleys of the Trans-México Volcanic Belt (TMVB) are home to one of the most diverse biotas in the world (Fig. 1.1; Cartron et al. 2005;

Luna et al. 2007).

The Mexican Plateau is bounded by the Sierra Madre Occidental to the west, the

Sierra Madre Oriental to the east, the Chihuahuan desert and the Sierra Zacatecas to the north, and the TMVB to the south (Fig. 1.1). It has the form of a parallelogram covering

85,300 km2, with elevations ranging from 1,000 to 3,300 m. The climate varies from arid and hot to semiarid and temperate, and the vegetation is similar to that of nearby deserts, with xeric shrublands in the plains and pine-oak forests in the surrounding mountains

(Ferrusquía-Villafranca et al. 2005). Recent studies of rodent populations in the Mexican

Plateau have revealed complex patterns of diversification and endemism (Fernández et al.

2012; Neiswenter and Riddle 2010).

! 2

Fig. 1.1.—Map of Mexico showing main geographic features discussed in this dissertation (modified from a map produced by CONABIO (Comisión Nacional para el Conocimiento y Uso de la Biodiversidad; http://www.conabio.gob.mx/).

! 3

Among the mountains at the southeastern edge of the TMVB lies the Oriental

Basin (Cuenca Oriental; Fig. 1.2). By any definition of “area of endemism,” this semiarid, endorheic (closed drainage) basin, which covers portions of the states of

Puebla, Tlaxcala, and Veracruz, is an important area of endemism for arid-adapted organisms in North America. This small (ca. 5,000 km2) basin characterized by alkaline grasslands, bunch grasses, and aridland scrubs in the valleys and coniferous forests in the surrounding mountains (Valdéz and Ceballos 1997), is thought to be the southernmost extension of the Chihuahuan desert (Shreve 1942). The Oriental Basin supports several endemic relict taxa of plants and animals, including at least four endemic mammals: the

Oriental Basin pocket gopher (Cratogeomys fulvescens), Nelson’s woodrat (N. nelsoni), the Perote deermouse (Peromyscus bullatus), and the Perote ground squirrel

(Xerospermophilus perotensis; Best and Ceballos 1995; González-Ruíz and Álvarez-

Castaneda 2005; González-Ruíz et al. 2006; Hafner et al. 2005).

Past studies of biogeography and areas of endemism in Mexico have used a wide variety of approaches. Some researchers have made biogeographical inferences based on examination of distributional records analyzed with clustering algorithms, whereas others have used geographic information system (GIS) algorithms, often augmented by phylogenetic and phylogeographic information. These latter approaches allow the researcher to infer evolutionary relationships, estimate current and potential distributions, and model the potential effects of climatic and geological changes on the generation of diversity (Escalante et al. 2002, 2004; Marshall and Liebherr 2000).

! 4

Fig. 1.2—Map of Central Mexico showing the Oriental Basin and surrounding volcanoes: A) Iztaccíhuatl; B) Popocatépetl; C) Malinche; D) Pico de Orizaba; E) Cofre de Perote (modified from CONABIO maps). The Oriental Basin straddles the Mexican states of Puebla, Tlaxcala, and Veracruz.

! 5 Distribution-based studies.—Over the past 20 years, many studies of the Mexican flora and fauna have used panbiogeographic (or track) methodology. Panbiogeography emphasizes the spatial or geographic dimension of biodiversity and often is used in combination with Parsimony Analyses of Endemicity (PAE) and GIS (Balleza et al 2005;

Escalante et al 2002, 2005; Luna et al 2004). Panbiogeography is a phenetic approach that uses knowledge of animal and plant distributions to draw inferences about the patterns and processes responsible for centers of diversity or areas of endemism (Craw et al. 1999). Many Mexican organisms have been studied using this approach, including aquatic organisms, insects, plants, mammals, and a combination of taxa from different groups (Andrés-Hernández et al. 2006; Corona and Morrone 2005; Corona et al. 2007;

Escalante et al. 2003, 2004, 2007; Huidrobo et al. 2006; Katinas et al. 2004; Morrone

2004; Morrone and Escalante 2002; Morrone and Gutierrez 2005; Morrone and Marquez

2001; Torres and Luna 2006).

Cladistic biogeography.—Many researchers have studied the biogeography of

Mexican organisms using cladistic biogeographic methods. These methods first reconstructs the phylogenetic relationships of multiple sets of taxa, then use this information to infer the relationships among the areas they occupy. Cladistic biogeography often is used in combination with Brooks Parsimony Analysis (BPA), an approach that has been used to study fishes, birds, or combinations of several taxa

(Domínguez et al. 2006; Marshall and Liebherr 2000; Zink et al. 2000).

Phylogenetic studies.—Most systematic studies of Mexican taxa attempt to relate the phylogenetic patterns revealed in the study to current or past geographic barriers and climate. The major findings of this approach include discovery of geographic clades

! 6 congruent with historical and cryptic geographic barriers in lineages of birds and mammals (Demastes et al. 2003; Edwards et al. 2001; Edwards and Bradley 2002;

García-Moreno et al. 2004; Hafner et al. 2004, 2005; Peppers and Bradley 2000; Peppers et al. 2002).

Phylogeographic studies.—Studies relating the genetic architecture and spatial distribution of conspecific populations or closely related species are increasingly frequent in the literature (Avise 2009). Most phylogeographic studies describe population parameters, haplotype distributions related to geographic distance, patterns of expansion, retraction, and migration among populations, but also study species limits, timing of diversification, and the impact of vicariance and dispersal on phylogeographic patterns.

Examples of the phylogeographic approach include many studies of vertebrate organisms, including fishes and amphibians (e.g., Doadrio and Domínguez 2004;

Mulcahy and Mendelson 2000).

Comparative biogeography and phylogeography.—This relatively new approach, which involves the study of codistributed taxa, has been applied to few Mexican species to date. The main objective of this approach is to explore the effects of climate and geology on codistributed taxa to reveal common patterns and processes that may explain a shared evolutionary history (Avise 2000; Arbogast and Kenagy 2001). This approach has been used in studies of birds, fishes, mammals, and a mixed group of species in the

TMVB and in the Chihuahuan, Sonoran, and Peninsular deserts of Mexico (Marshall and

Lienherr 2000; Mateos 2005; Riddle and Hafner 2004; Zink 2002; Zink et al. 2000). This approach also has been implemented in conjunction with approximate Bayesian computation in studies of birds and pitvipers and colubrid snakes in the southern Mexican

! 7 highlands, the Peninsular Desert, and the Isthmus of Tehuantepec (Barber and Klicka

2010; Castoe et al. 2009; Daza et al. 2010; Leache et al. 2007).

My dissertation research uses the comparative phylogeographic approach to determine if a set of codistributed mammalian taxa show the same phylogenetic patterns and, if so, whether these patterns are the result from the same climatic and geologic events. Same topologies not necessarily imply common history for a set of codistributed taxa because geographic barriers may produce pseudocongruence in the form of soft and hard allopatric distributions. The first one implying an organismal response to environmental variations, the second implies external factors limiting distributions.

(Pyron and Burbrink 2010). Well-supported phylogenetic inferences coupled with robust estimates of divergence times in codistributed species should be a powerful analytical tool to reveal underlying causes of lineage diversification (Bermingham and Moritz 1998;

Hickerson et al. 2006).

Several arid-adapted rodents, including Nelson’s woodrat (Neotoma nelsoni), the

Perote ground squirrel (Xerospermophilus perotensis), the rock mouse (Peromyscus difficilis), and Phillips’ kangaroo rat (Dipodomys phillipsii), show almost identical distributions in the Mexican Plateau and Oriental Basin of central Mexico. Not only are these species codistributed in their general ecological requirements, but their subspecific taxonomic boundaries are also very similar, and these boundaries are thought to result from effects of the TMVB. My study will test the hypothesis that the timing of major divergence events in the four rodent species listed above is coincident with the timing of major topographical shifts in the TMVB.

Neotoma nelsoni is a member of the rodent family Cricetidae (New World rats and

! 8 mice) and Xerospermophilus perotensis is a member of the Sciuridae (squirrels, marmots, and their relatives). Both species are endemic to the Oriental Basin and both are listed as threatened by the Mexican government. The putative sister species of both taxa are found in the Mexican Plateau to the north. Both species show slight, qualitative morphological differences when compared to their sister species, but scarcity of museum specimens and collecting restrictions by the Mexican Government have, until now, prevented a thorough study of their species status using modern systematic tools. Prior to this study, nothing was known about their biogeographic history.

Dipodomys phillipsii belongs to the rodent family Heteromyidae (kangaroo rats, pocket mice, and their allies), and Peromyscus difficilis is another cricetid rodent. Both are Mexican endemics with similar distributions. Populations of both species occur in the arid and semi-arid plains and low hills in the Oriental Basin and the Mexican Plateau, however P. difficilis has disjunct populations at higher elevations in the pine forests of the TMVB. Both species are divided into several subspecies that can be distinguished primarily by where they occur, plus a few qualitative morphological characters, such as body size and fur coloration. Until this study, neither species has been studied using modern systematic techniques and analyses.

My survey of mammals of the Oriental Basin and Mexican Plateau was carried out between 2006 and 2010 and, thanks to a collecting permit granted by the Mexican government to F. Cervantes on my behalf, I was able to obtain multiple samples, including fresh tissues, for each of the four study taxa. Using the tissues I collected plus other tissues generously donated to me by colleagues in multiple museums in Mexico and the U.S., I had the opportunity to examine, for the first time, phylogenetic and

! 9 phylogeographic relationships among the different populations and subspecies of the four study taxa using mitochondrial and nuclear markers analyzed using maximum likelihood and Bayesian approaches. Use of molecular markers under a coalescent framework also allowed me to estimate divergence dates within each taxon, which placed the evolutionary history of these Mexican endemics into a larger, historical-biogeographical context and provided the opportunity to test the role of geographic barriers and climate shifts in generating the pattern of evolutionary relationships we see today in these species.

1.1 LITERATURE CITED

ANDRÉS-HERNÁNDEZ, R., J. J. MORRONE, T. TERRAZAS, AND L. LÓPEZ M. 2006. Análisis de las especies mexicanas de Rhus subgenero Lobadium (Angiospermae: Anacardiaceae). Interciencia 31:900–904.

ARBOGAST, B. S., AND G. J. KENAGY. 2001. Comparative phylogeography as an integrative approach to historical biogeography. Journal of Biogeography 28:819– 825.

AVISE, J. C. 2009. Phylogeography: retrospect and prospect. Journal of Biogeography 36:3–15.

AVISE, J. C. 2000. Phylogeography: the history and formation of species. Harvard University Press, Cambridge, MA.

BALLEZA, J. J., J. L. VILLASENOR, AND G. I. MANRIQUEZ. 2005. Regionalización biogeográfica de Zacatecas, México, con base en los patrones de distribución de la familia Asteracea. Anales del Instituto de Biología, Serie Botánica 76:71–78.

BARBER, B. R., AND J. KLICKA. 2010. Two pulses of diversification across the Isthmus of Tehuantepec in a montane Mexican bird fauna. Proceedings of the Royal Society B doi:10.1098/rspb.2010.0343.

BERMINGHAM, E. AND C. MORITZ. 1998. Comparative phylogeography: concepts and applications. Molecular Ecology 7:367–369.

BEST, T. L., AND G. CEBALLOS. 1995. Spermophilus perotensis. Mammalian Species 507:1–3.

CARTRON, J. L., G. CEBALLOS, AND R. S. FELGER. 2005. Biodiversity, ecosystems, and

! 10 conservation in Northern Mexico. Oxford University Press, New York.

CASTOE, T. A., J. M. DAZA, E. N. SMITH, M. M. SASA, U. KUCH, J. A. CAMPBELL, P. T. CHIPPINDALE, AND C. L PARKINSON. 2009. Comparative phylogeography of pitvipers suggest a consensus of ancient Middle American highland biogeography. Journal of Biogeography 36:88–103.

CORONA, A. M. AND, J. J. MORRONE. 2005. Track Analysis of the Species of Lampetis (Spinthoptera) Casey, 1909 (Coleoptera: Buprestidae) in North America, Central America, and the West Indies. Caribbean Journal of Science 41:37–41.

CORONA, A. V. H. TOLEDO, AND J. J. MORRONE. 2007. Does the Trans-mexican Volcanic Belt represent a natural biogeographical unit? An analysis of the distributional patterns of Coleoptera. Journal of Biogeography 34:1008–1015.

CRAW, R. C., J. R. GREHAN, AND M. J. HEADS. 1999. Panbiogeography: Tracking the history of life. New York and Oxford, Oxford Biogeography series 11.

DAZA, J. M., T. A. CASTOE, AND C. L. PARKINSON. 2010. Using regional comparative phylogeographic data from snake lineages to infer historical processes in Middle America. Ecography 33:343–354.

DEMASTES, J. W., A. L. BUTT, M. S. HAFNER, AND J. E. LIGHT. 2003. Systematics of a rare species of pocket gopher, Pappogeomys alcorni. Journal of Mammalogy 84:753–761.

DOADRIO, I., AND O. DOMÍNGUEZ. 2004. Phylogenetic relationships within the fish family Goodeidae based on cytochrome b sequence data. Molecular Phylogenetics and Evolution 31:416–430.

DOMÍNGUEZ-DOMÍNGUEZ, O., I. DOADRIO, AND G. PÉREZ- PONCE DE LEÓN. 2006. Historical biogeography of some river basins in central Mexico evidenced by their goodeine freshwater fishes: a preliminary hypothesis using secondary Brooks parsimony analysis. Journal of Biogeography 33:1437–1447.

EDWARDS, C. W. AND R. D. BRADLEY. 2002. Molecular systematics and historical phylobiogeography of the Neotoma mexicana species group. Journal of Mammalogy 83:20–30.

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.

ESCALANTE, T., D. ESPINOSA, AND J. J. MORRONE. 2002. Patrones de distribución geográfica de los mamíferos de México. Acta Zoológica Mexicana (Nueva Serie) 87:47–65.

! 11 ESCALANTE, T., D. ESPINOSA, AND J. J. MORRONE. 2003. Using parsimony analysis of endemicity to analize the distribution of Mexican land mammals. The Southwestern Naturalist 48:563–578.

ESCALANTE, T., G. RODRÍGUEZ, AND J. J. MORRONE. 2004. The diversification of Nearctic mammals in the Mexican transition zone. Biological Journal of the Linnean Society 83:327–339.

ESCALANTE, T., G. RODRÍGUEZ, AND J. J. MORRONE. 2005. Las provincias biogeograficas del componente mexicano de montana desde la perspectiva de los mamíferos continentales. Anales del Instituto de Biología, Serie Botánica 76:199– 205.

ESCALANTE, T., V. SÁNCHEZ-CORDERO, J. J. MORRONE, AND M. LINAJE. 2007. Areas of endemism of Mexican terrestrial mammals: a case study using species’ ecological niche modeling, parsimony analysis of endemicity and Goloboff fit. Interciencia 32:151–159.

FERNÁNDEZ, J. A., F. A. CERVANTES, AND M. S. HAFNER. In press. Molecular systematics and biogeography of the Mexican endemic kangaroo rat, Dipodomys phillipsii (Rodentia: Heteromyidae). Journal of Mammalogy.

FERRUSQUÍA-VILLAFRANCA, I., AND L. I. GONZÁLEZ. 2005. Northern Mexico’s landscape, part II: The biotic setting across time. Pp. 39–51 in Biodiversity, ecosystems, and conservation in Northern Mexico (J. L. Cartron, G. Ceballos, and R. S. Felger., eds.), Oxford University Press, New York.

GARCÍA-MORENO, J., A. G. NAVARRO-SIGUENZA, A. TOWNSEND PETERSON, AND L. A. SÁNCHEZ-GONZÁLEZ. 2004. Genetic variation coincides with geographic structure in the common bush-tanager (Chlorospingus ophthalmicus) complex from Mexico. Molecular Phylogenetics and Evolution 33:186–196.

GONZALEZ-RUIZ, N., AND S. T. ALVAREZ-CASTANEDA. 2005. Peromyscus bullatus. Mammalian Species 770:1–3.

GONZALEZ-RUIZ, N., J. RAMÍREZ-PULIDO, AND H. H. GENOWAYS. 2006. Geographic distribution, taxonomy, and conservation status of Nelson’s woodrat (Neotoma nelsoni) in Mexico. The Southwestern Naturalist 51:112–116.

HAFNER, M. S., T. A. SPRADLING, J. E. LIGHT, D. J. HAFNER, AND J. R. DEMBOSKI. 2004. Systematic revision of pocket gophers of the Cratogeomys gymnurus species group. Journal of Mammalogy 85:1170–1183.

HAFNER, M. S., J. E. LIGHT, D. J. HAFNER, S. V. BRANT, T. A. SPRADLING, AND J. W. DEMASTES. 2005. Cryptic species in the Mexican pocket gopher Cratogeomys merriami. Journal of Mammalogy 86:1095–1108.

! 12 HAROLD, A. S., AND R. D. MOOI. 1994. Areas of endemism: definition and recognition criteria. Systematic Biology 43:261–266.

HAUSDORF, B. 1988. Weighted ancestral area analysis and a solution of the redundant distribution problem. Systematic Biology 47:445–456.

HICKERSON, M. J., G. DOLMAN, AND C. MORITZ. 2006. Comparative phylogeographic summary statistics for testing simultaneous vicariance. Molecular Ecology 15:209– 223.

HUIDOBRO, L., J. J. MORRONE, J. L. VILLALOBOS AND F. ALVAREZ. 2006. Distributional patterns of freshwater taxa (fishes, crustaceans and plants) from the Mexican Transition Zone. Journal of Biogeography 33:731–741.

KATINAS, L., J. V. CRISCI, W. L. WAGNER, AND P. C. HOCH. 2004. Geographical diversification of tribes Epilobieae, Gongylocarpeae, and Onagreae (Onagraceae) in North America, based on Parsimony Analysis of Endemicity and track compatibility analysis. Annals of the Missouri Botanical Garden 91: 159–185.

LAMOREUX, J. F., J. C. MORRISON, T. H. RICKETTS, D. M. OLSON, E. DINERSTEIN, M. W. MCKNIGHT, AND H. H. SHUGART. 2006. Global tests of biodiversity concordance and the importance of endemism. Nature 440:212–214.

LEACHE, A. D., S. C. CREWS, AND M. J. HICKERSON. 2007. Two waves of diversification in mammals and reptiles of Baja California revealed by hierarchical Bayesian analysis. Biology Letters doi:10.1098/rsbl.2007.0368.

LUNA, I., J. J. MORRONE, AND D. ESPINOSA. 2007. Biodiversidad de la Faja Volcánica Transmexicana. Comisión Nacional para la Biodiversidad, México, Instituto de Biología, Universidad Nacional Autónoma de México, México, D. F.

LUNA, V. I., O. ALCANTARA, AND R. CONTRERAS-MEDINA. 2004. Patterns of diversity, endemism and conservation: an example with Mexican species of Ternstroemiaceae Mirb. ex DC. (Tricolpates: Ericales). Biodiversity and Conservation 13: 2723– 2739.

MARSHALL, C. J. AND J. K. LIEBHERR. 2000. Cladistic biogeography of the Mexican transition zone. Journal of Biogeography 27:203–216.

MATEOS, M. 2005. Comparative phylogeography of livebearing fishes in the genera Poeciliops and Poecilia (Poeciliidae: Cyprinodontiformes) in Central Mexico. Journal of Biogeography 32:775–780.

MORRONE, J. J. 2004. Panbiogeografía, componentes bióticos y zonas de transición. Revista Brasileira de Entomologia 48: 149–162.

! 13 MORRONE. J. J. AND T. ESCALANTE. 2002. Parsimony analysis of endemicity (PAE) of Mexican terrestrial mammals at different area units: when size matters. Journal of Biogeography 29:1095–1104.

MORRONE, J. J., AND A. GUTIERREZ. 2005. Do fleas (Insecta: Siphonaptera) parallel their mammal host diversification in the Mexican transition zone? Journal of Biogeography 32:1315–1325.

MORRONE, J. J. AND J. MARQUEZ. 2001. Halffter's Mexican Transition Zone, beetle generalized tracks, and geographical homology. Journal of Biogeography 28:635– 650.

MULCAHY, D. G. AND J. R. MENDELSON. 2000. Phylogeography and speciation of the morphologically variable, widespread Species Bufo valliceps, based on molecular evidence from mtDNA. Molecular Phylogenetics and Evolution 7:173–189.

MYERS, N., R. A. MITTERMEIER, C. G. MITTERMEIER, G. A. B. FONSECA, AND J. KENTS. 2000. Biodiversity hotspots for conservation priorities. Nature 403:853–858.

NEISWENTER, S., AND B. R. RIDDLE. 2010. Diversification of the Perognathus flavus species group in emerging arid grasslands of western North America. Journal of Mammalogy 91:348–362.

NELSON, G. AND N. I. PLATNICK. 1981. Systematics and biogeography: cladistics and vicariance. Columbia University Press, NY.

PEPPERS, L. L., AND R. D. BRADLEY. 2000. Cryptic species in Sigmodon hispidus: evidence from DNA sequences. Journal of Mammalogy 81:332–343.

PEPPERS, L. L., D. S. CARROLL, AND R. D. BRADLEY. 2002. Molecular systematics of the genus Sigmodon (Rodentia: Muridae): evidence from the mitochondrial Cytochrome-b gene. Journal of Mammalogy 83:396–407.

PLATNICK, N. I. 1991. On areas of endemism. Systematic Zoology 28:537–546.

PYRON, R. A., AND F. T. BURBRINK. 2010. Hard and soft allopatry: physically and ecologically mediated modes of geographic speciation. Journal of Biogeography 37:2005–2015.

RIDDLE, B. R., AND D. J. HAFNER. 2004. The past and future roles of phylogeography in historical biogeography. Pp. 93–110 in Frontiers of Biogeography. New directions in the geography of nature (M. V. Lomolino and L. R. Heaney, eds.). Sinahuer Associates, Sunderland, MA.

SHREVE, F. 1942. The desert vegetation of North America. Botanical Review 8:195–246.

! 14 TORRES, M. A., AND I. LUNA. 2006. Análisis de trazos para establecer áreas de conservación en la Faja Volcánica Transmexicana. Interciencia 31:849–855.

VALDÉZ, M., AND G. CEBALLOS. 1997. Conservation of endemic mammals of Mexico: The Perote ground squirrel (Spermophilus perotensis). Journal of Mammalogy 78:74–82.

VELASCO DE LEÓN, P. M., J. ARELLANO, A. SILVA P., AND S. Y. GUARNEROS. 2007. Aspectos geológicos y paleontológicos. Pp. 25–38 in Biodiversidad de la Faja Volcánica Transmexicana (I. Luna, J. J. Morrone and D. Espinosa., eds.), CONABIO, México.

ZINK, R. M. 2002. Methods in comparative phylogeography, and their application to studying evolution in the North American aridlands. Integrative Comparative Biology, 42:953–959.

ZINK, R. M., R. C. BLACKWELL-RAGO, AND F. RONQUIST. 2000. The shifting roles of dispersal and vicariance in biogeography. Proc. R. Soc. London.

! 15 CHAPTER 2 MOLECULAR PHYLOGENETICS AND BIOGEOGRAPHY OF A RARE ENDEMIC: NELSON’S WOODRAT (NEOTOMA LEUCODON NELSONI) IN THE ORIENTAL BASIN OF MEXICO

2.1 INTRODUCTION

Neotoma nelsoni (Nelson’s woodrat) is a rare, monotypic, and endemic Mexican mammal, with only a handful of specimens housed in museum collections worldwide

(González-Ruíz et al. 2006, Hall 1981). This species has been collected at only four localities in the states of Puebla and Veracruz (Acosta and Fernández 2009; Falcón-Ordaz et al. 2010; Goldman 1905; González-Christen et al. 2002; González-Ruíz et al. 2006), and >200 kilometers separate N. nelsoni from the nearest population of its putative sister species, N. leucodon, in the state of Hidalgo, Mexico (Fig. 2.1; González-Ruíz et al.

2006).

Neotoma nelsoni was originally described by E. A. Goldman in 1905 based on 11 specimens (the holotype and 10 paratypes) collected in the vicinity of Perote, Veracruz.

Goldman differentiated N. nelsoni from N. leucodon by presence of a palatine bone with a short posterior spine, nasals that are more wedge-shaped and pointed posteriorly, and a tail that is indistinctly bicolored and nearly unicolored near the tip (Goldman 1905;

González-Ruíz et al. 2006). The morphological differentiation and geographic isolation of N. nelsoni led Goldman (1905) to recognize it as a distinct species, presumably related to the much more widespread N. leucodon.

Hall and Genoways (1970) used morphological evidence to place N. nelsoni in the

N. albigula species group, which included N. albigula (with 14 subspecies including the current N. leucodon), N. palatina, N. varia, and N. nelsoni. Neotoma palatina, a Mexican

! 16

Fig. 2.1.—Geographical distribution of the Neotoma micropus species group. Asterisks show the approximate collection locality of the two samples of N. leucodon. Localities are listed in the methods section (redrawn from Edwards et al. 2001; González- Ruíz et al. 2006; and Hall 1981).

! 17 endemic restricted to the canyon of the Río Bolaños, its tributaries, and immediately adjacent uplands in Jalisco (Hall and Genoways 1970), is still recognized as a valid species (Musser and Carleton 2005). Neotoma varia, also a Mexican endemic restricted to Isla Dátil off the coast of Sonora, was considered inseparable from N. albigula by

Bogan (1997), who reduced it to subspecies status within N. albigula.

Several studies have surveyed genetic relationships in Neotoma, yet none of these, to date, has included a specimen of N. nelsoni. Planz et al. (1996) examined mitochondrial DNA restriction site polymorphism in Neotoma and suggested that N. a. leucodon may represent a cryptic species within N. albigula. Edwards et al. (2001) examined variation in the mitochondrial cytochrome b gene and, in agreement with Planz et al. (1996), concluded that N. albigula consists of two species, N. albigula in the southwestern United States and northwestern Mexico, and N. leucodon in the southcentral United States and northcentral Mexico (Fig. 2.1). Edwards et al. (2001) further determined that N. albigula and N. goldmani (endemic to the Mexican altiplano) are allied with the N. floridana species group (which also includes N. floridana and N. magister), whereas N. leucodon is allied with the N. micropus species group (which includes N. micropus and, possibly, N. palatina). Building on the study by Edwards et al.

(2001), Edwards and Bradley (2002) examined cytochrome b variation in taxonomically and geographically larger samples of Neotoma and recovered the same species groupings.

Most recently, a study of variation in four mitochondrial and four nuclear genes by

Matocq et al. (2007) confirmed the Neotoma species groups defined by Edwards et al.

(2001). Because specimens of N. nelsoni and N. palatina were not included in any of

! 18 these molecular studies, the taxonomic status and phylogenetic position of these species remain unclear.

During a 2007 survey of mammals of the Oriental Basin of Mexico, I trapped a specimen of N. nelsoni in the municipality of Perote, Veracruz. This capture affords the opportunity to examine, for the first time, phylogenetic relationships between N. nelsoni and its congeners based on molecular markers. Use of molecular markers also facilitates calculation of estimated divergence dates within the clade containing N. nelsoni, which places the evolutionary history of this rare Mexican endemic into a larger, historical biogeographical context.

2.2 MATERIALS AND METHODS

Sampling.—The female specimen of Neotoma nelsoni (LSUMZ 36663) was trapped on June 22, 2007, in Veracruz (3 km S El Frijol Colorado, Municipality of

Perote, 19.5723, -97.3835, 2,437 m; Fig. 2.2) under the authority of Mexican collecting permit FAUT-0002 issued to F. A. Cervantes. Traps were set in a dry plain with almost no vegetation and in low, rocky hills dominated by yucca (Yucca sp.), agave (Agave sp.), prickly pear (Opuntia sp.), and other cacti. The specimen of N. nelsoni was captured in the latter habitat. All mammal specimens were handled in accordance with guidelines approved by the American Society of Mammalogists (Kelt et al. 2010, Sikes et al. 2011).

DNA sequences generated by Edwards et al. (2001) and Matocq et al. (2007) for

14 species of Neotoma were downloaded from GenBank. The samples include one individual per species (unless indicated otherwise), and GenBank numbers are listed in the following order: Cytb; 12S; 16S. Taxa represented were N. albigula (DQ179707;

DQ179757; DQ179857), N. cinerea (DQ179705; DQ179755; DQ179855), N. floridana

! 19

Fig. 2.2.—Map of central Mexico (inset modified from http://gaia.inegi. org.mx/mdm5/viewer.html) showing the known geographic distribution of Nelson’s woodrat, Neotoma leucodon nelsoni (shaded area in main map). The shaded area in the inset shows the extent of the Oriental Basin. The specimen examined in this study was collected at locality 1 in the inset (Falcón-Ordáz et al. (2010; this study). The three other known collection localities of N. leucodon nelsoni are based on reports by Goldman (1905; locality 2), González-Christen et al. (2002; locality 3), González-Ruíz et al. (2006; locality 4).

! 20 (DQ179669; DQ179719; DQ179819), N. fuscipes (DQ179672; DQ179722; DQ179822),

N. goldmani (DQ179677; DQ179727; DQ179827), N. isthmica (DQ179678; DQ179728;

DQ179828), N. lepida (DQ179681; DQ179731; DQ179831), N. leucodon (n = 2;

DQ179665; DQ179715; DQ179815 from Texas, and DQ179689; DQ179739; DQ179839 from Durango), N. macrotis (DQ179691; DQ179741; DQ179841), N. magister

(DQ179706; DQ179756; DQ179856), N. mexicana (DQ179695; DQ179745;

DQ179845), N. micropus (DQ179668; DQ179718; DQ179818 from Texas, and

DQ179698; DQ179748; DQ179848 from New Mexico), N. picta (DQ179701;

DQ179751; DQ179851), and N. stephensi (DQ179702; DQ179752; DQ179852).

Sequences generated by Edwards et al. (2001) and Matocq et al. (2007) for Hodomys alleni (DQ179660; DQ179710; DQ179810), Peromyscus attwateri (DQ179661;

DQ179711; DQ179811), Ototylomys phyllotis (DQ179664; DQ179714; DQ179814),

Tylomys nudicaudus (DQ179662; DQ179712; DQ179812), and Xenomys nelsoni

(DQ179663; DQ179713; DQ179813) were included as outgroups. Collection localities for all of the above specimens are available in Edwards et al. (2001) and Matocq et al.

(2007).

Laboratory protocols.—Total genomic DNA was extracted from fresh tissue using a commercial kit (DNeasy Blood and Tissue Kit; Qiagen Inc., Valencia, California), and

3 mitochondrial genes were sequenced: cytochrome-b (Cytb), 12S ribosomal RNA (12S), and 16S ribosomal RNA (16S). Sequences were amplified by PCR (Saiki et al. 1988) using the following universal primers developed for rodents: MVZ-05 and H15915 for

Cytb (Irwin et al. 1991); 12S L82 and 12S H900 for 12S (Nedbal et al. 1994); and 16Sa and 16Sb for 16S (Matocq et al. 2007). The following PCR parameters were used to

! 21 amplify genes: initial denaturation at 95°C for 2 min, followed by 27 cycles of denaturation at 95°C for 1 min, annealing at 49°C for 1 min, and extension at 72°C for 2 min, with a final extension at 72°C for 7 min (Mantooth et al. 2000). Amplifications were performed in a total volume of 25 µL and 200 ng of DNA. Agarose (2%) gels were used to visualize amplified products. PCR products were purified using ExoSAP-IT

(Affymetrix, Santa Clara, California). DNA sequencing was performed for both light and heavy strands with a Big Dye Terminator v1.1, v3.1 in an automated 3100 Genetic

Analyzer (Applied Biosystems, Foster City, California) at the Museum of Natural

Science, Louisiana State University. Sequences generated in this analysis (Cytb, 12S, and

16S.) were submitted to GenBank.

Data analysis.—Editing and alignment of sequences and matrix manipulations were performed in Sequencher 4.7 (Gene Codes Corporation, Ann Arbor, Michigan).

Sequences were verified manually, and authenticity of the gene was confirmed by amino acid translation and BLAST searches in Genbank

(http://blast.ncbi.nlm.nih.gov/Blast.cgi).

To enable comparison of our results with those of previous studies, raw sequence divergence values for the Cytb gene were corrected using the Kimura 2-parameter substitution model (Kimura 1980) in PAUP* (Phylogenetic Analysis using Parsimony, version 4.0b, Swofford 2003).

Phylogenetic analyses were carried out under a maximum likelihood (ML) framework in PAUP* and PhyMl 3.0 (Guindon and Gascuel 2003). Analyses based on

Bayesian inference (BI) were conducted using MrBayes (version 3.2-cvs; Huelsenbeck and Ronquist 2001; Ronquist and Huelsenbeck 2003). In all analyses, variable nucleotide

! 22 positions were considered unordered, discrete characters with 4 possible states (A, C, G, and T). Best-fit models for ML and BI analyses were evaluated using the Akaike

Information Criterion and the program jModeltest 0.1.1 (Akaike 1973; Guindon and

Gascuel 2003; Posada 2008). The HKY+I+G model was selected for the Cytb data, and the TIM2+I+G model was selected for the 12S and 16S genes. ML clade support was assessed with 500 bootstrap replicates (Felsenstein 1985) in PAUP*, and clade support in

BI analyses was evaluated using posterior probabilities (pp).

Tree searches in the ML analyses were performed with the starting trees obtained via 100 random, stepwise additions followed by tree-bisection-reconnection (TBR) branch swapping. In the BI analyses, best-fit models were applied to each data partition with unlinked parameters and allowing rate variation. The Metropolis Markov Chain

Monte Carlo analysis consisted of 2 independent runs of 10 x 106 generations in which trees were sampled every 103 generations, resulting in 104 samples for each run. A majority-rule consensus tree was constructed using the final 2 x 104 trees. The analysis was stopped when the average standard deviation of split frequencies approached zero, and convergence also was assessed using Tracer 1.5 (Rambaut and Drummond 2007).

The combined data set was analyzed in a partitioned way (genes and model parameters) to allow for independent convergence on optimal values for each component (Ronquist and Huelsenbeck 2003). Results from each gene were analyzed separately to evaluate potential conflict among gene trees. Nodes were considered well supported if there was

>80% bootstrap support in ML analyses or >95% pp in BI analyses.

Analysis of divergence times.—The program *BEAST version 1.6.0 (Drummond and Rambaut 2007) was used to estimate the time of divergence between N. nelsoni and

! 23 other members of the N. micropus species group. A Yule tree prior was used, implicitly considering the gene tree to represent the species tree, and the HKY+I+G model was selected as the substitution model. To improve search efficiency, monophyly of the genus Neotoma was enforced. A relaxed clock with a lognormal distribution allowing rate variation among sites was used. Chains were run for 27 generations, sampling the parameter every 103 generations. Convergence statistics were checked for effective sample sizes using Tracer version 1.5 (Rambaut and Drummond 2007). Consensus trees were generated from the resulting 2 x 104 trees using TreeAnnotator version 1.6.0

(Rambaut and Drummond 2009) after elimination of the first 25% as burn-in.

Three fossil-based dates were used to calibrate the *BEAST analysis. The 1st documents separation of the Neotoma lineage from ancestral cricetine stock in the upper

(late) Miocene, during a period that lasted from approximately 11.6 to 5.3 MYA (Hibbard

1966). The 2nd and 3rd fossil-based dates establish minimum ages for N. albigula and N. mexicana, respectively, at the beginning of the Pleistocene (ca. 2.5 MYA; Álvarez 1969;

Birney 1973, 1976; Dalquest and Stangl 1984; Harris 1984; Jakway 1958; Logan and

Black 1979; Murray 1957; Schultz and Howard 1935; Van Devender et al. 1977). To account for uncertainty in the fossil-based calibrations, the dates were modeled as lognormal distributions rather than point calibrations (Ho and Phillips 2009).

2.3 RESULTS

The final dataset consisted of 1,140 base pairs (bp) of the Cytb gene, 413 bp of the

12S gene, and 561 bp of the 16S gene for a total of 2,114 bp. Cytb divergence values

(Table 2.1) show the N. nelsoni specimen from Veracruz, Mexico to be most similar genetically to the specimen of N. leucodon from Durango, Mexico (3.3% corrected

! 24 sequence divergence). Surprisingly, the two specimens of N. leucodon, one from

Durango and the other from Texas, are only 5.5% genetically similar at the Cytb locus.

Neotoma nelsoni exhibits >12% sequence divergence when compared to all other species of Neotoma in the dataset (N. albigula, N. mexicana, and N. micropus).

ML and BI analyses of the 12S gene indicated different topologies, but both trees were poorly resolved and had universally low branch support (trees not shown). Similar analyses of the 16S dataset recovered a moderately supported clade (75% bootstrap; 0.96 pp) that includes all members of the N. micropus species group (as defined by Matocq et al. 2007) plus the N. nelsoni sample. An internal clade with low branch support (50% bootstrap; <0.60 pp) grouped N. nelsoni with the two N. leucodon samples, but relationships among the 3 taxa were unresolved (trees not shown, but are available on request).

ML and BI analyses of the Cytb gene (Fig. 2.3) recovered topologies that were almost identical to those reported by Matocq et al. (2007). The N. micropus species group of Matocq et al. (2007) was recovered with strong nodal support (100% bootstrap;

1.00 pp), and N. nelsoni was included as a member of this group. Within the N. micropus species group, N. nelsoni clustered with the two samples of N. leucodon (100% bootstrap;

1.00 pp), and within this trio of taxa, N. nelsoni grouped with the specimen of N. leucodon from Durango (92% bootstrap; 0.99 pp) to the exclusion of the sample of N. leucodon from Texas (Fig. 2.3).

Although trees based on the individual genes showed no topological conflicts, a partitioned homogeneity test indicted that the genes could not be combined. As a result,

! 25 Table 2.1.—Cytochrome-b sequence divergence values (Kimura 2-parameter model) between and within species of the Neotoma micropus and N. floridana species groups.

Taxa Cytb sequence divergence

N. nelsoni–N. leucodon (average) 4.17%

N. nelsoni–N. leucodon (Durango) 3.28%

N. nelsoni–N. leucodon (Texas) 5.07%

N. leucodon (Durango)–N. leucodon (Texas) 5.52%

N. nelsoni–N. micropus 12.83%

N. nelsoni–N. albigula 14.38%

N. nelsoni–N. mexicana 13.50%

! 26

Fig. 2.3.—Cladogram showing the placement of Nelson’s woodrat, N. leucodon nelsoni, within the genus Neotoma and the N. micropus species group (shaded box) based on a maximum likelihood (ML) analysis of cytochrome-b sequences. A Bayesian analysis of the same data yielded a tree with identical topology. Numbers above branches indicate ML bootstrap support, and numbers below branches indicate Bayesian posterior probabilities. The specimen labeled “N. leucodon (D)” is from Durango, and “N. leucodon (T)” is from Texas.

! 27

Fig. 2.4.—Cladogram showing placement of Nelson’s woodrat, N. leucodon nelsoni, within the genus Neotoma and the N. micropus species group (shaded box) based on a partitioned maximum likelihood (ML) analysis of 3 mitochondrial genes (Cytb, 12S, and 16S). A Bayesian analysis of the same data yielded a tree with identical topology. Numbers above branches indicate ML bootstrap support, and numbers below branches indicate Bayesian posterior probabilities.

! 28 the three sequence datasets were analyzed in a partitioned analysis. To save computational time, only members of the N. micropus and N. floridana species groups were included in these analyses. ML and BI analyses of the partitioned data (Fig. 2.4) showed topologies identical to the Cytb tree (Fig. 2.3) and confirmed monophyly of both species groups (100% bootstrap and 1.00 pp for both clades). The sister relationship between N. nelsoni and the N. leucodon sample from Durango also was confirmed (97% bootstrap; 1.00 pp; Fig. 2.4).

Analysis of divergence times.—Except for the position of N. stephensi (which is not relevant to the present study), the tree obtained in the *BEAST analysis (Fig. 2.5) was identical to the BI tree published by Matocq et al. (2007; their Fig. 4). The Tracer analysis of *BEAST output files confirmed a high effective sample size (>3,500) for all parameters. Although error estimates are large for most nodes, results suggest that the N. micropus and N. floridana species groups diverged near the end of the Miocene (ca. 6.7 mya) and most speciation events within these species groups took place in Pliocene or early Pleistocene times (ca. 4.6 to 2.0 mya; Fig. 2.5). The split between N. leucodon

(Texas sample) and the N. nelsoni + N. leucodon (Durango) clade was placed at approximately 3 mya (late Pliocene), and divergence of N. nelsoni from N. leucodon

(Durango) is estimated to have occurred approximately 2 mya during the early

Pleistocene.

2.4 DISCUSSION

Species status of N. nelsoni.—The discovery that N. nelsoni is phylogenetically closer to N. leucodon from Durango than the latter specimen is to N. leucodon from

Texas calls into question the monophyly of N. leucodon and the species status of N.

! 29

Fig. 2.5.—Phylogeny and chronogram of selected species of the genus Neotoma (and outgroups), and estimated divergence times calculated from *BEAST analysis. Numbers at nodes are mean divergence dates (mya) and gray bars show the 95% credibility intervals for nodes of interest. Black circles identify calibration points based on fossil evidence. The specimen labeled “N. leucodon (D)” is from Durango, and “N. leucodon (T)” is from Texas.

! 30 nelsoni. Continued recognition of N. nelsoni at the species level would require elevation of the northern clade of N. leucodon to species level (as N. warreni Merriam, 1908), but there is no compelling evidence to warrant this action. The Cytb sequence divergence measured between the Texas and Durango samples of N. leucodon in this study (5.5%;

Table 2.1) is within the range of values reported for conspecific populations by Bradley and Baker (2001). The morphological characters used to distinguish N. nelsoni from N. leucodon (modest differences in pelage quality and coloration and minor, qualitative differences in a few cranial bones; González-Ruíz et al. 2006) are on the order of characters normally used to distinguish among subspecies in other rodents. Given the current absence of diagnostic morphological characters to distinguish N. nelsoni from N. leucodon and the relatively low level of Cytb sequence divergence between the 2 forms

(3.3%; well within the range of values reported for conspecific populations by Bradley and Baker, 2001), I take a conservative taxonomic approach and recognize N. nelsoni as a subspecies of N. leucodon (as N. leucodon nelsoni Goldman 1905). Designation of subspecific epitaphs for all other populations of N. leucodon must await a large-scale assessment of geographic variation within the species.

Estimates of divergence times.—Early hypotheses about the diversification of the

North American mammalian fauna suggested that Pleistocene glacial and interglacial cycles were major generators of species-level diversity (e.g., Findley 1969; Orr 1960).

Consistent with this widely held contention, Zimmerman and Nejtek (1977) used data obtained by protein electrophoresis to estimate that basal divergence events leading to the current species N. albigula, N. floridana, and N. micropus occurred between 112,000 and

155,000 years ago. Zimmerman and Nejtek (1977) implicated vegetational changes that

! 31 occurred in response to Pleistocene glaciations as the causal force driving speciation in this clade.

Recently, new distributional, paleontological, and molecular evidence has modified the “Pleistocene paradigm,” assigning a prominent role to geologic events that took place during Miocene and Pliocene times (Hafner and Riddle 1997; Riddle 1995;

Riddle et al. 2000; Vrba, 1992). For example, estimates of divergence times calculated in this study suggest that the initial split between the N. micropus and N. floridana species groups occurred near the end of the Miocene (approximately 6.7 mya; Fig. 2.5), long before the major climatic fluctuations of the Pleistocene. Mean estimates of divergence events within the N. micropus and N. albigula species groups also are pre-Pleistocene, with diversification in both groups beginning in early Pliocene (ca. 4.5 mya). The divergence between N. leucodon nelsoni and the N. leucodon sample from Durango is the only event postulated to have taken place during the Pleistocene (Fig. 2.5).

If, as these data suggest, major divergence events in the N. micropus and N. floridana species groups took place during Miocene and Pliocene times, then geological and climatic events of those times probably played major roles, either directly or indirectly, in this diversification. The Trans-Mexico Volcanic Belt (TMVB) began to form in the Miocene (Ferrusquía-Villafranca 2007) and is still active today. Presently, it consists of a large, but somewhat fragmented, chain of mountains extending from the

Pacific coast to the Gulf of Mexico at the latitude of Mexico City. Fragmentation of the

TMVB through time created a large number of interior lakes, some of which persist today, whereas others have dried up, giving rise to small pockets of alkaline desert inside the TMVB. The Oriental Basin, home to at least one population of N. leucodon nelsoni

! 32 and several endemic animals and plants, is one such alkaline desert near the eastern end of the TMVB (Ferrusquía-Villafranca 2007; Morán-Zenteno 2004; Shreve 1942;

Viniegra 1992).

The TMVB is widely recognized as a generator of biological diversity in Mexico, and the uplift of the TMVB and associated creation of interior lakes and small patches of desert may have isolated ancestral populations of many organisms, including the common ancestor of the N. leucodon populations now in Durango and N. leucodon nelsoni in the

Oriental Basin region well before the Pleistocene. Once isolated, these populations would be subject to the many geologic and climatic perturbations of the late Miocene,

Pliocene, and early Pleistocene, resulting in a mixed pattern of diversification, in which some lineages diverge well before the Pleistocene, whereas others respond to the well- known geologic, glacial, and climate changes of the Pleistocene. It seems that the evolutionary history of N. leucodon nelsoni may be yet another example of the isolating force of the TMVB (Demastes et al. 2002; Douglas et al. 2010; Hewitt 2001; Mateos et al. 2002; McCormack et al. 2008; Mulcahy and Mendelson 2000; Zink and Blackwell

1998).

Natural history and conservation status of N. leucodon nelsoni.—Given the rarity and isolation of N. leucodon nelsoni populations, a few comments on the natural history and conservations status of this taxon are in order. Only 14 specimens of Nelson’s woodrat are known to date. Goldman (1905) described the species based on 11 specimens collected in 1893, and more than a century passed before 3 additional specimens were collected, including the specimen used in this study (González-Christen et al. 2002; González-Ruíz et al. 2006). The specimens examined by Goldman (1905)

! 33 were collected near Perote, Veracruz, but it is difficult to determine the habitat at their site of capture because Goldman used the locality “Perote, Veracruz” for all specimens collected in the general vicinity of Perote, which contains a wide variety of habitat types

(González-Ruíz et al. 2006). González-Christen et al. (2002) collected a single specimen of N. nelsoni in a tropical rain forest about 30 km SE of the type locality, and González-

Ruíz et al. (2006) reported another specimen collected in a mountainous area with cloud forests and coffee plantations approximately 40 km S of the type locality. The specimen used in this study was collected in dry, low hills of volcanic origin surrounded by xeric shrubs and cacti approximately 15 km W of the type locality. Given the wide variety of habitats in which the few known specimens of N. leucodon nelsoni have been captured, it appears that this subspecies has high environmental plasticity (as suggested by González-

Ruíz et al. 2006) and can live in a broad array of vegetational associations.

Despite its high environmental plasticity, N. leucodon nelsoni seems to have a clear association with the Oriental Basin (“Cuenca Oriental”; Fig. 2.2), and all known specimens of N. leucodon nelsoni have been collected in or around this xeric basin. The

Oriental Basin is a semiarid, endorheic (closed drainage) basin that extends over portions of the Mexican states of Puebla, Tlaxcala, and Veracruz. This relatively small (ca. 5,000 km2) basin is characterized by alkaline grasslands, bunch grasses, and aridland scrub in the valleys, and coniferous forests in the surrounding mountains (Valdéz and Ceballos

1997). The basin and surrounding areas support several endemic taxa of plants and animals, including at least 4 mammal species exclusive to the Oriental Basin region: the

Oriental Basin pocket gopher (Cratogeomys fulvescens; Hafner et al. 2005), the Perote ground squirrel (Xerospermophilus perotensis; Best and Ceballos 1995), the Perote

! 34 deermouse (Peromyscus bullatus; González-Ruíz and Álvarez-Castaneda 2005), and

Nelson’s woodrat (N. leucodon nelsoni; González-Ruíz et al. 2006; this study).

The specimen of N. leucodon nelsoni used in this study was collected in sympatry with several endemic and non-endemic rodent species, including the Perote ground squirrel, Phillip’s kangaroo rat (Dipodomys phillipsii), silky pocket mouse (Perognathus flavus), western harvest mouse (Reithrodontomys megalotis), and rock mouse

(Peromyscus difficilis). Two species of rattlesnakes (Crotalus molossus nigrescens and

C. scutulatus salvini) were collected near the capture locality, and a third species (C. ravus ravus) known from this region (Camarillo 1998) also may prey on woodrats and other rodents in the community.

A new genus of nematode, Lamotheoxyuris, was described from the specimen of

N. leucodon nelsoni used in this study (Falcón-Ordaz et al. 2010), and three species of fleas were collected from this specimen, one of them (Aniomiopsyllus perotensis) new to science (Acosta and Fernández 2009).

Despite intensive collecting efforts by several research parties working in and around the Oriental Basin (Hall and Dalquest 1963; González-Christen et al. 2002;

González-Ruíz et al. 2006; this study), only a few specimens of N. leucodon nelsoni are known to science. Neotoma leucodon nelsoni is listed (as N. nelsoni) as critically endangered by the International Union for Conservation of Nature (IUCN 2011) but, ironically, the scarcity of information about the density and distribution of N. leucodon nelsoni populations has caused this taxon to be excluded from the Mexican list of endangered species (Luiselli 2002). There is little doubt that N. leucodon nelsoni populations are threatened by continued and extensive conversion of natural habitats to

! 35 agriculture in the Oriental Basin region (González-Ruíz et al. 2006). New collecting efforts are needed in the remaining areas of natural habitat to learn more about the biology of N. leucodon nelsoni so as to protect it in the future.

2.5 LITERATURE CITED

ACOSTA, R., AND J. A. FERNANDEZ. 2009. A new species of Anomiopsyllus Baker, 1904 (Insecta: Siphonaptera), and noteworthy records of flea’s from Nelson’s woodrat, Neotoma nelsoni (Rodentia: cricetidae), in the Oriental Basin, Mexico. Journal of Parasitology 95:532–535.

AKAIKE, H. 1973. Information theory as an extension of the maximum likelihood principle. Pp. 267–281 in Second International Symposium on Information Theory (B. Petrov and F. Csaki, eds.), Akademiai Kiado, Budapest, Hungary.

ALVAREZ, T. 1969. Restos fosiles de mamíferos de Tlapacaya, Estado de México (Pleistoceno-Reciente). Miscellaneus Publications of the Museum of Natural History, University of Kansas 51:93–112.

BEST, T. L., AND G. CEBALLOS. 1995. Spermophilus perotensis. Mammalian Species 507:1–3.

BIRNEY, E. C. 1973. Systematics of three species of woodrats (genus Neotoma) in Central North America. Miscellaneus Publications of the Museum of Natural History, University of Kansas 58:1–173.

BIRNEY, E. C. 1976. An assessment of relationships and effects of interbreeding among woodrats of the Neotoma floridana species-group. Journal of Mammalogy 571:103–132.

BOGAN, M. A. 1997. On the status on Neotoma varia from Isla Dátil, Sonora. Pp. 81– 87, in Life Among the Muses: Paper in Honor of James S. Findley (T.L. Yates, W.L. Gannon, and D.E. Wilson, eds.), Special Publication, The Museum of Southwestern Biology, USA.

BRADLEY, R. D., AND R. J. BAKER. 2001. A test of the genetic species concept: cytochrome –b sequences and mammals. Journal of Mammalogy 82:969–973.

CAMARILLO, J. L. 1998. Observaciones preliminares sobre los anfibios y reptiles de los lagos crater de Puebla-Veracruz. Anales del Instituto de Biologia, Universidad Nacional Autónoma de México, Serie Zoologia 69:125–127.

DALQUEST, W. W., AND F. B, STANGL, JR. 1984. Late Pleistocene and early recent mammals from Fowlkes Cave, southern Culberson County, Texas. Pp. 432–455,

! 36 in Contributions in Quaternary vertebrate paleontology: a volume in memorial to John E. Guilday (H. H. Genoways, and M. R. Dawson, eds.), Special Publications, Carnegie Museum of Natural History 8:1–538.

DEMASTES, J. W., A. L. BUTT, M. S. HAFNER, AND J. E. LIGHT. 2003. Systematics of a rare species of pocket gopher, Pappogeomys alcorni. Journal of Mammalogy 84:753–761.

DOUGLAS, M. E., M. R. DOUGLAS, G. W. SCHUETT, D. D. BECK, AND B. K. SULLIVAN. 2010. Conservation phylogenetics of helodermatid lizards using multiple molecular markers and a supertree approach. Molecular Phylogenetics and Evolution 55:153–167.

DRUMMOND A. J., AND A. RAMBAUT. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology 7, 214.

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 82:267–269.

EDWARDS, C. W., AND R. D. BRADLEY. 2002. Molecular systematics of the genus Neotoma. Molecular Phylogenetics and Evolution 25:489–500.

FALCON-ORDAZ, J., J. A. FERNANDEZ, AND L. GARCIA-PRIETO. 2010. Lamotheoxyuris ackerti n. gen., n. comb. (Nematoda: Heteroxynematidae) parasite of Neotoma spp. (Rodentia: Muridae) Lamotheoxyuris ackerti n. gen., n. comb. (Nematoda: Heteroxynematidae) parásito de Neotoma spp. (Rodentia: Muridae). Revista Chilena de Historia Natural 83:259–266.

FELSENSTEIN, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783–791.

FERRUSQUÍA-VILLAFRANCA, I. 2007. Ensayo sobre la caracterización y significación biologica. PP. 7–24 in Biodiversidad de la Faja Volcánica Transmexicana (I. Luna, J. J. Morrone, and D. Espinosa, eds.), Universidad Nacional Autónoma de México, México, D. F.

FINDLEY, J. S. 1969. Biogeography of southwestern boreal and desert mammals. Pp. 113–128 in Contributions in mammalogy (J. K. Jones, ed.), University of Kansas, Miscellaneus Publications of the Museum of Natural History.

GOLDMAN, E. A. 1905. Twelve new rats of the genus Neotoma. Proceedings of the Biological Society of Washington 18:27–34.

! 37 GONZALEZ-CHRISTEN, A., S. GAONA, AND G. LÓPEZ-ORTEGA. 2002. Registros adicionales de mamíferos para el estado de Veracruz, México. Vertebrata Mexicana 11:9–16.

GONZALEZ-RUIZ, N., AND S. T. ALVAREZ-CASTANEDA. 2005. Peromyscus bullatus. Mammalian Species 770:1–3.

GONZALEZ-RUIZ, N., J. RAMÍREZ-PULIDO, AND H. H. GENOWAYS. 2006. Geographic distribution, taxonomy, and conservation status of Nelson’s woodrat (Neotoma nelsoni) in Mexico. The Southwestern Naturalist 51:112–116.

GUINDON, S., AND O. GASCUEL. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology: 52:696–704.

HAFNER, D. J., AND B. R. RIDDLE. 1997. Biogeography of Baja California peninsular desert mammals. Pp. 19–68 in Life among the Muses: Papers in Honor of James S. Findley (T. L. Yates, W. L. Gannon, and D. E. Wilson, eds.), The Museum of Southwestern Biology, The University of New Mexico, Albuquerque.

HAFNER, M. S., J. E. LIGHT, D. J. HAFNER, S. V. BRANT, T. A. SPRADLING, AND J. W. DEMASTES. 2005. Cryptic species in the mexican pocket gopher Cratogeomys merriami. Journal of Mammalogy, 86:1095–1108.

nd HALL, E. R. 1981. The mammals of North America. 2 ed. John Wiley & Sons, Inc., New York 1:1-600 + 90.

HALL, E. R., AND W. W. DALQUEST. 1963. The mammals of Veracruz. University of Kansas Publications, Museum of Natural History 14:165–362.

HALL, E. R., AND H. H. GENOWAYS. 1970. Taxonomy of the Neotoma albigula-group of woodrats in Central Mexico. Journal of Mammalogy 51:504–516.

HARRIS, A. H. 1984. Two new species of Late Pleistocene woodrats (Cricetidae: Neotoma) from New Mexico. Journal of Mammalogy 65:560–566.

HEWITT, G. M. 2001. Speciation, hybrid zones and phylogeography — or seeing genes in space and time. Molecular Ecology: 10:537–549.

HIBBARD, C. W. 1966. New rodents from the late Conozoic of Kansas. Papers of the Michigan Academy of Science, Arts, and Letters 52:115–131.

HO S.Y.W., AND M. J. PHILLIPS. 2009. Accounting for calibration uncertainty in phylogenetic estimation of evolutionary divergence times. Systematic Biology 58:367–380.

! 38 HONEYCUTT, R. L., M. A. NEDBAL, R. M. ADKINS, AND L. L. JANACEK. 1995. Mammalian mitochondrial DNA evolution: a comparison of the cytochrome b and cytochrome c oxidase II genes. Journal of Molecular Evolution 40:260–272.

HUELSENBECK, J. P., AND F. RONQUIST. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17:754–755.

IRWIN, D. M., T. D. KOCHER, AND C. WILSON. 1991. Evolution of the cytochrome b gene in mammals. Journal of Molecular Evolution 2:37–55.

IUCN. 2011. IUCN Red List of Threatened Species. Version 2011.1. . Accessed on 25 July 2011.

JAKWAY, G. E. 1958. Pleistocene lagomorpha and rodentia from the San Josecito Cave, Nuevo Leon, Mexico. Kansas Academy of Sciences, Transactions 61:313–327.

KELT, D. A., M. S. HAFNER, AND THE AMERICAN SOCIETY OF MAMMALOGISTS' AD OC COMMITTEE FOR GUIDELINES ON HANDLING RODENTS IN THE FIELD. 2010. Updated guidelines for protection of mammalogists and wildlife researchers from hantavirus pulmonary syndrome (HPS). Journal of Mammalogy 91:1524–1527.

KIMURA, M. 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies if nucleotide sequences. Journal of Molecular Evolution 16:111–120.

LOGAN, L. E., AND C. C. BLACK. 1979. The Quaternary vertebrate fauna of Upper Sloth Cave, Guadalupe Mountains National Park, Texas. Pp. 141–158, in Biological Investigations if the Guadalupe Mountains National Park, Texas (H. H. Genoways, and R. J. Baker eds.), U.S. Dept. Interior, National Park Services Transactions Proceedings Series 4:1–442.

LUISELLI, F. C. 2002. Norma Oficial Mexicana NOM-059-SEMARNAT-2001 Protección ambiental-Especies nativas de México de flora y fauna silvestres- Categorías de riesgo y especificaciones para su inclusión, exclusión o cambio- Lista de especies en riesgo. Diario Oficial de la Federación. Miércoles 6 de marzo de 2002.

MANTOOTH, S. J., C. JONES, AND R. D. BRADLEY. 2000. Molecular systematics of Dipodomys elator (Rodentia:Heteromyidae) and its phylogeographic implications. Journal of Mammalogy 81:885–894.

MATEOS, M., O. I. SANJUR, AND R. C. VRIJENHOEK. 2002. Historical biogeography of the livebearing fish genus Poeciliopsis (Poecilidae: Cyprinodontiformes). Evolution 56:972–984.

! 39 MATOCQ, M. D., Q. R. SHURTCLIFF, 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.

MCCORMACK, J. E., A. T. PETERSON, E. BONACCORSO, AND T. B. SMITH. 2008. Speciation in the highlands of Mexico: genetic and phenotypic divergence in the Mexican jay (Aphelocoma ultramarina). Molecular Ecology 17:2505–2521.

MORÁN-ZENTENO, D. J. 1994. The geology of the Mexican Republic. 146 pp. American Association of Petroleum Geologists, Tulsa, USA.

MULCAHY, D. G., AND J. R. MENDELSON III. 2000. Phylogeography and speciation of the morphologically variable, widespread species Bufo valliceps, based on molecular evidence from mtDNA. Molecular Phylogenetics and Evolution 17:173–189.

MURRAY, K. F. 1957. Pleistocene climate and fauna of Burnet Cave, New Mexico. Ecology 38:129–132.

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. John Hopkins University Press, Baltimore, Maryland.

NEDBAL, M. A., M. W. ALLARD, AND R. L. HONEYCUTT. 1994. Molecular systematics of hystricognath rodents: Evidence from the mitochondrial 12S rRNA gene. Molecular Phylogenetics and Evolution 3:206–220.

ORR, R. T. 1960. An analysis of the Recent land mammals. Systematic Zoology 9:171– 179.

PLANZ, J. V., E. G. ZIMMERMAN, T. A. SPRADLING, AND D. R. AKINS. 1996. Molecular phylogeny of the Neotoma floridana species group. Journal of Mammalogy 77:519–535.

POSADA, D. 2008. jModeltest: phylogenetic model averaging. Molecular Biology and Evolution 25:1253–1256.

RAMBAUT A., AND A. J. DRUMMOND. 2007. Tracer v1.4, Available from http://beast.bio.ed.ac.uk/Tracer

RAMBAUT A., AND A. J. DRUMMOND. 2009. TreeAnnotator v1.6.0: MCMC Output Analysis. Available at http://beast.bio.ed.ac.uk/TreeAnnotator\

! 40 RIDDLE, B. R. 1995. Molecular biogeography in the pocket mice (Perognathus and Chaetodipus) and grasshopper mice (Onychomys): the late Cenozoic development of a North American aridlands rodent guild. Journal of Mammalogy 76:283–301.

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.

RONQUIST, F., AND J. P. HUELSENBECK. 2003. MRBAYES 3. Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572–1574.

SAIKI, R. K., G. H. GELFAND, S. STOFFEL, S. J. SHARF, R. HIGUCHI, G. T. HORN, K. B. MULLIS, AND H. A. ERLICK. 1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487–491.

SCHULTZ, C. B., AND E. B. HOWARD. 1935. The fauna of Burnet Cave, Guadalupe Mountains, New Mexico. Proceedings of the Academy of Natural Sciences, Philadelphia 87:273–297.

SHREVE, F. 1942. 1942. The desert vegetation of North America. Botanical Review 8:195–246.

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.

SWOFFORD, D. L. 2003. PAUP: Phylogenetic analysis using parsimony (and other methods). Version 4.0b2.0a. Sinauer Associates, Inc., Publishers, Sunderland, Massachussetts.

VALDÉZ, M., AND G. CEBALLOS. 1997. Conservation of endemic mammals of Mexico: The Perote ground squirrel (Spermophilus perotensis). Journal of Mammalogy, 78:74–82.

VAN DEVENDER, T. R., A. M. PHILLIPS, III, AND J. I. MEAD. 1977. Late Pleistocene reptiles and small mammals from the Lower Grand Canyon of Arizona. The Southwestern Naturalist 22:49–66.

VINEGRA, F. O. 1992. Geología histórica de México. Facultad de Ingenieria, Universidad Nacional Autónoma de México, México, D. F.

VRBA, E. S. 1992. Mammals as a key to evolutionary theory. Journal of Mammalogy 73:1–28.

! 41 ZINK, R. M., AND R. C. BLACKWELL. 1998. Molecular systematics and biogeography of aridland gnatcatchers (genus Polioptila), and evidence supporting species status of the California gnatcatcher (Poliptila californica). Molecular Phylogenetics and Evolution 9:26–32.

ZIMMERMAN, E. G., AND M. E. NEJTEK. 1977. Genetics and speciation of three semispecies of Neotoma. Journal of Mammalogy 58:391–402.

! 42 CHAPTER 3 PHYLOGENETICS AND BIOGEOGRAPHY OF THE MICROENDEMIC RODENT XEROSPERMOPHILUS PEROTENSIS (PEROTE GROUND SQUIRREL) IN THE ORIENTAL BASIN OF MEXICO

3.1 INTRODUCTION

The concept of endemism is central to the fields of biogeography and biological conservation, and areas with high numbers of endemic species, or “biodiversity hotspots,” often are included in protected area networks (Myers et al. 2000). In biogeography, an area of endemism usually is defined as a region that contains the only known occurrences of two or more taxa (i.e., multiple taxa are restricted to that region), but other definitions of “areas of endemism” focus on the area’s geographic delimitation by natural barriers or the distributional congruence of several species in the area (Harold and Mooi 1994; Hausdorf 2002; Platnick 1991).

Mexico is widely known as a biologically megadiverse country and a biodiversity hotspot (Lamoreaux et al. 2006). The diverse Mexican biota is the product of interactions between a dynamic and complex topography and myriad ecological and historical factors

(Velasco de Leon et al. 2007). In central Mexico, the highlands and arid valleys of the

Trans-Mexico Volcanic Belt (TMVB) are home to one of the most diverse biotas in the world (Luna et al. 2007).

Among the mountains at the southeastern edge of the TMVB lies the Oriental

Basin (Cuenca Oriental; Fig. 3.1). By any definition of “area of endemism,” this semiarid, endorheic (closed drainage) basin, which covers portions of Puebla, Tlaxcala, and Veracruz, is an important area of endemism in North America. This relatively small

(ca.

! 43

Fig. 3.1.—Geographical distribution of pygmy ground squirrels of the genus Xerospermophilus (modified from Hall 1981). Numbered dots show collection localities listed in Appendix 3.1 for Xerospermophilus specimens used in this study. Locality 8 and all localities outside shaded areas are collection sites for outgroup specimens (Appendix 3.1). The shaded area near locality 11 shows the approximate extent of the Oriental Basin (Cuenca Oriental).

! 44 5,000 km2) basin characterized by alkaline grasslands, bunch grasses, and aridland scrub in the valleys and coniferous forests in the surrounding mountains (Valdéz and Ceballos

1997) supports several endemic taxa of plants and animals, including at least four taxa of endemic mammals: the Oriental Basin pocket gopher (Cratogeomys fulvescens), the

Perote deermouse (Peromyscus bullatus), Nelson’s woodrat (N. nelsoni), and the Perote ground squirrel (Xerospermophilus perotensis; Best and Ceballos 1995; González-Ruíz and Álvarez-Castaneda 2005; González-Ruíz et al. 2006; Hafner et al. 2005).

Shreve (1942) referred to the Oriental Basin as the southernmost extension of the

Chihuahuan desert, and most mammals of the Oriental Basin, including X. perotensis, are arid-adapted species. Typically, the closest relatives of Oriental Basin endemics inhabit the deserts of the Mexican Plateau to the north, and this appears to be the case for X. perotensis, whose sister species is thought to be X. spilosoma (Fig. 3.1; Howell 1938).

In his classic revision of North American ground squirrels, Howell (1938) classified all ground squirrels in the genus Citellus (later transferred to Spermophilus by

Hershkovitz 1949) and divided the genus into eight subgenera: Ammospermophilus,

Callospermophilus, Citellus, Ictidomys, Notocitellus, Otospermophilus, Poliocitellus, and

Xerospermophilus. Early morphological and chromosomal studies suggested a close relationship between Spermophilus perotensis and S. spilosoma (Howell 1938; Uribe and

Ahumada 1990), however composition of and relationships among the subgenera of

Spermophilus were not clearly understood at that time. S. perotensis and S. spilosoma were placed as sister taxa in the subgenus Ictidomys, along with I. tridecemlineatus, I. mexicanus, and the more recently described I. parvidens (Harrison et al. 2003; Herron et al. 2003).

! 45 The taxonomic status and systematic affinities of S. perotensis were not investigated again until molecular studies by Harrison et al. (2003) and Herron et al.

(2004) confirmed the close affinity of S. perotensis with S. spilosoma. In fact, both molecular studies (using the same specimens of S. spilosoma and based on the cytochrome-b [Cytb] gene) showed S. spilosoma to be paraphyletic with respect to S. perotensis, with S. s. pallescens from Mexico sister to S. perotensis, and S. s. marginatus from Kansas sister to the S. s. pallescens + S. perotensis clade. These same studies showed the S. perotensis + S. spilosoma clade (subgenus Ictidomys) to be sister to the S. mohavensis + S. tereticaudus clade (subgenus Xerospermophilus), with prairie dogs

(Cynomys) sister to this group.

Helgen et al. (2009) combined new morphological data with the molecular evidence provided by Harrison et al. (2003) and Herron et al. (2004) to elevate the subgenus Xerospermophilus to full generic status. In Xerospermophilus, Helgen et al.

(2009) included the 4 species of pygmy ground squirrels adapted to arid and semi-arid conditions, X. mohavensis, X. tereticaudus, X. spilosoma, and X. perotensis. In view of the potential paraphyly of X. spilosoma (Harrison et al. 2003; Herron et al. 2003), Helgen et al. (2009) recommended future research into species-level boundaries in the spilosoma- perotensis complex.

Despite previous morphological and molecular studies of the systematic status of

X. perotensis and allied taxa, several uncertainties remain with respect to the species status of X. perotensis, monophyly of X. spilosoma, and the timing of diversification events within the genus Xerospermophilus relative to major geological and climatic events. Each of these issues is explored in this analysis using newly acquired samples of

! 46 X. perotensis and X. spilosoma and sequence evidence from multiple mitochondrial and nuclear genes.

3.2 MATERIAL AND METHODS

Sampling.—Tissue samples of Ictidomys mexicanus (n = 1 individual), I. parvidens (n = 2), I. tridecemlineatus (n = 4), Xerospermophilus mohavensis (n = 2), X. perotensis (n = 4), X. spilosoma (n = 3), and X. tereticaudus (n = 2) were either collected in the field under the authority of Mexican collecting permit FAUT-0002 (issued to F. A.

Cervantes) or donated by museums (Appendix 3.1). In addition to the DNA sequences generated in this study, 32 sequences were downloaded from Genbank for use in the molecular analyses (Appendix 3.1). Outgroups in the analyses included specimens of

Urocitellus townsendii (in the Cytb analysis), Callospermophilus lateralis (in the 12S ribosomal RNA [12S] and interphotoreceptor retinoid-binding protein [IRBP] analyses), and Sciurus niger (in the growth hormone receptor [GHR] analysis). The collection and processing of samples was undertaken following the guidelines of the American Society of Mammalogists for use of wild animals in research (Kelt et al. 2010; Sikes et al. 2011).

Laboratory protocols.—Total genomic DNA was extracted from tissue using a commercial kit (DNeasy Blood and Tissue Kit; Qiagen Inc., Valencia, California).

Portions of two nuclear genes (GHR and IRBP), and two mitochondrial genes (Cytb and

12S) were sequenced for subsequent analysis. The genes were amplified by PCR (Saiki et al. 1988) using the following universal primers developed for rodents: GHR1f and

GHRend1f for GHR (Jansa et al. 2009); IRBP-A and IRBP-B for IRBP (Stanhope et al.

1992); MVZ-05 and H15915 for Cytb (Irwin et al. 1991); and 12S L82 and 12S H900 for

12S (Nedbal et al. 1994). PCR amplification of the GHR gene was performed under the

! 47 following parameters: initial denaturation at 94°C for 5 min followed by 34 cycles of denaturation at 94°C for 15 sec, annealing at 60°C for 1 min, extension at 72°C for 1.5 min and 1 final extention at 72°C for 10 min. Amplification of the IRBP gene began with initial denaturation at 95°C for 10 min followed by 27 cycles of denaturation at 95°C for

25 sec, annealing at 58°C for 20 sec, extension of 72°C for 1 min, and 1 final extension at

72°C for 10 min. Amplification of both mitochondrial genes began with initial denaturation at 95°C for 2 min followed by 27 cycles at 95°C for 1 min, annealing at

49°C for 1 min, extension at 72°C for 2 min, and 1 final extension at 72°C for 7 min

(Mantooth et al. 2000). Amplifications were performed in a total volume of 25 µL and

200 ng of DNA. Agarose gels (2%) were used to visualize amplified products. PCR products were purified using either Polyethylene Glycol (PEG) or ExoSAP-IT

(Affymetrix, Santa Clara, California). DNA sequencing was performed for both light and heavy strands with a Big Dye Terminator v1.1, v3.1 in an automated 3100 Genetic

Analyzer (Applied Biosystems, Foster City, California) at the Museum of Natural

Science, Louisiana State University. Editing and alignment of sequences and matrix manipulations were performed in Sequencher 4.7 (Gene Codes Corporation, Ann Arbor,

Michigan), MacClade (Maddison and Maddison 2000), and Mesquite (Maddison and

Maddison 2010). Sequences were verified manually, and authenticity of the gene was confirmed by amino acid translation and BLAST searches in Genbank

(http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Estimates of genetic divergence.—To enable comparison of my results with those of previous studies, sequence divergence values for the Cytb gene were corrected using the Kimura 2-parameter substitution model (Kimura 1980) in PAUP* 4.0b10

! 48 (Phylogenetic Analysis using Parsimony, version 4.0b10, Swofford 2003) and MEGA version 5 (Tamura et al. 2011). Saturation analyses for 3rd codon positions were performed using the methods of Griffiths (1997), and maximum likelihood (ML) analyses were run with and without 3rd codon transitions to evaluate the affects of 3rd codon substitutions on phylogenetic reconstruction.

Phylogenetic analyses.—Initial phylogenetic analyses were conducted using

Bayesian inference (BI) in the program MrBayes (version 3.2-cvs; Huelsenbeck and

Ronquist 2001; Ronquist and Huelsenbeck 2003) and maximum likelihood (ML) in

PAUP*. Analysis of the Cytb data included 41 specimens of 14 species for which complete Cytb sequences were available. Phylogenetic analysis of the 12S gene included

17 specimens of 8 species, analysis of GHR included 14 specimens of 6 species, and analysis of IRBP included 18 specimens representing 8 species (Appendix 3.1).

Following analyses of individual genes, a partitioned analysis including all 4 genes was conducted using ML and BI frameworks. ML analyses were run in both PAUP* and

PhyMl 3.0 (Guindon and Gascuel 2003). Variable nucleotide positions were considered unordered, discrete characters with 4 possible states (A, C, G, T). Best-fit models for ML and BI analyses were evaluated using the Akaike Information Criterion (AIC) and the program jModeltest 0.1.1 (Guindon and Gascuel 2003; Posada 2008). The following models were selected for Cytb, 12S, GHR, and IRBP genes, respectively: TrN+G,

TIM3+G, TPM3UF+I+G, and TPM3UF+I. ML clade support was assessed with 100 bootstrap (bs) replicates in PAUP* and PhyMl 3.0, and clade support in the BI analyses was evaluated using posterior probablilities (pp).

! 49 ML analyses were performed with the starting trees obtained from 100 random, stepwise additions followed by tree-bisection-reconnection (TBR) branch swapping. In the BI analyses, best-fit models were applied to each data partition with unlinked parameters and allowing rate variation. The Metropolis Markov Chain Monte Carlo analysis consisted of two independent runs of 10 x 106 generations in which trees were sampled every 103 generations, resulting in 104 samples for each run. After discarding the initial 10% as burn-in, a majority-rule consensus tree was constructed using the final

18 x 103 trees. The analysis was stopped when the average standard deviation of split frequencies approached zero and convergence was reached, as determined using Tracer version 1.5 (Rambaut and Drummond 2007). The combined data set was analyzed in a partitioned manner (genes and model parameters) to allow for independent convergence on optimal values for each component (Ronquist and Huelsenbeck 2003). Data partitions included mitochondrial versus nuclear genes, Cytb versus 12S, Cytb versus IRBP, and

12S versus IRBP genes. Nodes were considered well supported if there was >80% bootstrap support in ML analyses or >95% posterior probability in BI analyses.

Estimates of divergence times.—The program *BEAST version 1.6.0 (Bayesian evolutionary analysis sampling trees; Drummond and Rambaut 2007) was used to generate estimates of the timing of divergence of X. perotensis from related species.

*BEAST analyses were carried out using a Yule tree prior and implicitly considering the

Cytb gene tree to represent the species tree. The estimate was calibrated using a dated fossil (Goodwin 1995; Harrison et al. 2003; Pizzimenti 1975) to constrain the minimum date for separation of Cynomys from Xerospermophilus to 2.7 mya. To account for uncertainty in the fossil-based calibration, the fossil date was modeled on a lognormal

! 50 distribution rather than a point calibration (Ho and Phillips 2009). The analysis used a relaxed clock with a lognormal distribution allowing rate variation among sites. Chains were run for 207 generations, sampling the parameter every 103 generations.

Convergence statistics were checked for effective sample sizes using Tracer version 1.5.

Consensus trees were generated from the resulting 20 x 103 trees using TreeAnnotator version 1.6.0 (Rambaut and Drummond 2009) after elimination of 10% as burn-in.

3.3 RESULTS

Analyses of DNA sequences involved a total of 3,403 base pairs (bp), including

1,141 bp of Cytb, 736 bp of 12S, 910 of GHR, and 616 bp of IRBP. ML and BI analyses using only Cytb sequences from the 41 individuals with complete Cytb sequences recovered trees with identical branch structure (Fig. 3.2). In these trees, there is strong support (pp = 1.00; bs = 100%) for monophyly of the genus Xerospermophilus. Within

Xerospermophilus, X. perotensis is shown to be sister to the X. spilosoma sample from the state of San Luis Potosí on the Mexican Plateau (locality 12 in Fig. 3.1) to the exclusion of other samples of X. spilosoma from Durango (locality 15), Kansas (14), and

New Mexico (13). X. mohavensis and X. tereticaudus also are depicted as sister taxa.

Among the many outgroups used in the analysis (listed above and in Appendix 3.1), the genus Cynomys was found to be sister to Xerospermophilus, although this relationship was not well supported (pp = 0.84 and bs = 79%) and therefore is not shown in Fig. 3.2.

Cytb divergence values (Table 3.1) show the taxa included in Fig. 3.2 to be well differentiated genetically. X. perotensis is 3.6% genetically divergent from the X. spilosoma sample from San Luis Potosí, and these two populations together show an average Cytb divergence of 6.4% from the other samples of X. spilosoma from Durango,

! 51

Fig. 3.2.—Relationships among major clades of Xerospermophilis based on maximum likelihood and Bayesian analyses of Cytochrome-b sequences. Locality numbers (mapped in Fig. 3.1 and listed in Appendix 3.1) are indicated before taxon names. Numbers at nodes are estimated mean divergence dates (mya) calculated from *BEAST analysis. Bars show 95% credibility intervals.

! 52 Table 3.1.—Mean percent cytochrome-b sequence divergence values (Kimura 2-parameter model) among species of the genus Xerospermophilus and Cynomys. The X. spilosoma specimen from San Luis Potosí is from locality 12 in Fig. 3.1, the specimen from Durango is from locality 15, and the specimens from New Mexico and Kansas are from localities 13 and 14, respectively.

X. spilosoma X. spilosoma X. spilosoma X. perotensis San Luis Potosí Durango New Mexico, Kansas X. mohavensis X. tereticaudus Cynomys

X. perotensis – 3.6 5.2 7.9 10.0 11.5 11.9

X. spilosoma – 6.0 6.4 10.6 12.0 12.0 San Luis Potosí

X. spilosoma – 5.5 9.9 11.8 11.2 Durango

X. spilosoma – 10.6 12.9 12.2 New Mexico, Kansas

X. mohavensis – 4.6 11.0

X. tereticaudus – 14.2

Cynomys –

! 53 New Mexico, and Kansas. The X. perotensis + X. spilosoma clade (including all samples of spilosoma) shows an average of 11.2% sequence divergence from the X. mohavensis + X. tereticaudus clade and 11.8% divergence from specimens of the genus Cynomys. Independent BI and ML analyses of 12S, GHR, and IRBP sequences confirmed the Cytb topology shown in Fig.

3.2, although nodal support values varied widely depending on the gene analyzed (trees not shown but available on request). In all analyses, the genus Xerospermophilus was monophyletic and sister to Cynomys, the sister species status of X. mohavensis + X. tereticaudus was confirmed with strong support, and the sister relationship between X. perotensis and the X. spilosoma sample from San Luis Potosí was recovered, but with low nodal support in the analyses of the nuclear genes (GHR and IRBP).

BI and ML analyses of the partitioned data sets (partitioned by mitochondrial genes only, nuclear genes only, and mitochondrial + nuclear genes) focused on relationships within

Xerospermophilus (Fig. 3.3). In all partitioned analyses, the genus Xerospermophilus was monophyletic with high support values and the sister status of X. perotensis and X. spilosoma

(San Luis Potosí) was confirmed with high nodal support (pp > 0.97 and bs = 100%).

Mean estimates of divergence times (Fig. 3.2) ranged from a low of 0.7 mya between X. perotensis and X. spilosoma (San Luis Potosí) to a high of 3.5 mya between the genera Cynomys and Xerospermophilus. All estimates of divergence times within the genus Xerospermophilus place these events in the Pleistocene, with the possible exception of the split between the X. mohavensis + X. tereticaudus clade and the X. perotensis + X. spilosoma clade, which was estimated at 2.7 mya with a confidence interval extending from 4.3 to 1.3 mya (Fig. 3.2).

! 54

Fig. 3.3.—Relationships among major clades of Xerospermophilus based on a maximum likelihood analysis of Cytb, 12S, GHR, and IRBP sequences partitioned by gene. A Bayesian analysis of the same data yielded identical relationships. Xerospermophilus spilosoma from Durango (locality 15) was excluded from these analyses because only Cytb sequences were available for that specimen.

! 55 3.4 DISCUSSION

This study of mitochondrial and nuclear DNA sequences confirms that X. perotensis is a genetically well-differentiated unit within Xerospermophilus. The sister relationship between X. perotensis and the X. spilosoma sample from San Luis Potosí (representing the subspecies X. s. cabrerai) also is strongly supported in this study and is consistent with evidence provided by

Uribe and Ahumada (1990) who reported chromosomal similarities between X. perotensis and X. s. cabrerai and interpreted this as an indicator of a close phylogenetic relationship between these taxa.

Species status of X. perotensis.–Since its original description by Merriam (1893), X. perotensis has been considered a valid species. However, recent morphological and molecular studies have questioned its species status, and some authors have suggested that X. perotensis is best regarded as a subspecies of X. spilosoma (Harrison et al. 2003; Helgen et al. 2009; Herron et al. 2004). The present study confirms that continued recognition of X. perotensis at the species level renders X. spilosoma paraphyletic (Figs 3.2 and 3.3). Paraphyly of X. spilosoma could be resolved taxonomically by recognizing multiple species within X. spilosoma, but unless one recognizes species based solely on degree of genetic divergence, no evidence is available at this time suggesting that X. spilosoma is a composite of multiple cryptic species. It could also be argued that X. perotensis and X. spilosoma populations from San Luis Potosí (X. s. cabrerai) should be combined into a single species. However, again, there is no evidence for species-level divergence between X. s. cabrerai and other subspecies of X. spilosoma except for the relatively large Cytb distances measured between the subspecies examined in this study (5.2–7.9%; Table

3.1). Synonymization of X. s. cabrerai with X. perotensis still would require recognition of multiple species within X. spilosoma to maintain monophyletic taxa.

! 56 Sister species within the sciurid genera Cynomys and Marmota show Cytb divergence values ranging from 1.2% to 7.7% (Harrison et al. 2003; Steppan et al. 1999), so the divergence value calculated between X. perotensis and X. s. cabrerai in this study (3.6%) lies within this range but is less than the “high genetic divergence” value of 5% suggested by Baker and Bradley

(2006:654) to signal possible cryptic species. Morphologically, X. perotensis differs from X. spilosoma in ways that are usually used to distinguish among subspecies of rodents. For example, X. perotensis resembles X. spilosoma pallescens of the northern Mexican Plateau and

Sierra Madre Oriental, except that X. perotensis is larger overall, has a shorter tail, is more yellowish dorsally, and has smaller and less conspicuous buffy spots (Best and Ceballos 1995).

The skull of X. perotensis is similar to that of X. spilosoma spilosoma (found in southern

Durango, Zacatecas, and parts of nearby states), except that the skull of X. perotensis is larger, has a relatively narrower and higher brain case, has auditory bullae that are broader and more flattened, and molariform teeth that are heavier than those of X. s. spilosoma (Best and Ceballos

1995; Hafner and Yates 1983; Helgen et al. 2009; Uribe et al. 1978).

Considering the absence of morphological or chromosomal evidence supporting the species status of X. perotensis, I herein take the conservative route and recognize perotensis as a subspecies of X. spilosoma (as X. s. perotensis). The relatively high divergence values measured between the subspecies of X. spilosoma in this study (Table 3.1) may signal presence of multiple cryptic species, but recognition of additional species must await a thorough study of geographic variation throughout the range of X. spilosoma.

Biogeography of Xerospermophilus on the Mexican Plateau and in the Oriental Basin.–

Shreve (1942) recognized that the Mexican Plateau, the Oriental Basin, and other isolated arid and semi-arid regions of central Mexico were relicts of a once continuous southern extension of

! 57 the Chihuahuan Desert. More recent research suggests that the arid and semi-arid lands of central

Mexico were continuous until mid-late Miocene, when rise of the TMVB began to act as a barrier between populations of arid-adapted species (Ferrusquía et al. 2005; Ferrusquía and González

2005). Hoffmann and Jones (1970) examined present day distributions of several mammal species in Mexico and suggested that many prairie and desert species may have reached their southernmost distributions during Pleistocene times, with subsequent range contractions leaving isolated populations in the south. They suggested that Cynomys mexicanus was one such peripheral isolate of the once more widespread species, C. ludovicianus (Hoffmann and Jones

1970). The Perote ground squirrel, X. s. perotensis, isolated in the Oriental Basin of central

Mexico, may be another example of this phenomenon.

Uribe and Ahumada (1990) speculated that X. spilosoma stock was once widespread throughout the highlands of northern Mexico and were able to disperse southward because of continuous, dry habitats found in inter-montane valleys. Subsequent tectonic or climatic events, or a combination of both, during the Pleistocene fragmented the once continuous grassland habitat in central Mexico, leaving the present day patches of arid and semi-arid habitats, including the Oriental Basin (Ferrusquía et al. 2005; Ferrusquía and González 2005; Hoffman and

Jones 1970; Pizzimenti 1975).

The results of this study underscore the importance of three under-studied biogeographic regions of Mexico: the Oriental Basin (inhabited by X. s. perotensis), the Mexican Plateau (X. s. cabrerai), and the Bolsón de Mapimí (X. s. pallescens). The origin of the Oriental Basin is closely linked with the volcanic activity that gave raise to alkaline lakes and the rain shadow effect that caused isolated pockets of arid and semi-arid land in the TMVB (Caballero et al. 2003;

Morán-Zenteno 1994). The Mexican Plateau formed as a result of uplifting of the Sierra Madre

! 58 Oriental, Sierra Madre Occidental, and TMVB, which created a dry, table land in the rain shadow of these large mountain ranges. Recent phylogenetic and biogeographic studies are beginning to show the importance of the Mexican Plateau as a center of evolutionary divergence in many rodent taxa (Fernandez et al. 2012; Neiswenter and Riddle 2010). Finally, the Bolsón de Mapimí, a closed desert basin located north of the Sierra de Zacatecas, formed during the Wisconsinan glacial period of the Pleistocene and acted as a refugium for many desert organisms (Elias 1992), including the ancestors of X. s. pallescens.

3.5 LITERATURE CITED

BAKER, R. J., AND R. D. BRADLEY. 2006. Speciation in mammals and the genetic species concept. Journal of Mammalogy 87:643–662.

BEST, T. L., AND G. CEBALLOS. 1995. Spermophilus perotensis. Mammalian Species 507:1–3.

CABALLERO, M., G. VILLACLARA, A. MARTÍNEZ, AND D. JUÁREZ. 2003. Short-term climatic change in lake sediments from lake Alchichica, Oriental, México. Geofísica Internacional 42:529– 537.

DRUMMOND A. J., AND A. RAMBAUT. 2007. *BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology 7:214.

ELIAS, S. A. 1992. Late Quaternary zoogeography of the Chihuahuan desert insect fauna based on fossil records from packrat middens. Journal of Biogeography 19:285–297.

FERNÁNDEZ, J. A., F. A. CERVANTES, AND M. S. HAFNER. In press. Molecular systematics and biogeography of the Mexican endemic kangaroo rat, Dipodomys phillipsii (Rodentia: Heteromyidae). Journal of Mammalogy.

FERRUSQUÍA-VILLAFRANCA, I., L. I. GONZÁLEZ, AND J. L. CARTRON. 2005. Northern Mexico’s landscape, part I: The physical settings and constrains on modeling biotic evolution. Pp. 11– 38 in Biodiversity, ecosystems, and conservation in Northern Mexico (J. L. Cartron, G. Ceballos, and R. S. Felger., eds.), Oxford University Press, New York.

FERRUSQUÍA-VILLAFRANCA, I., AND L. I. GONZÁLEZ. 2005. Northern Mexico’s landscape, part II: The biotic setting across time. Pp. 39–51 in Biodiversity, ecosystems, and conservation in Northern Mexico (J. L. Cartron, G. Ceballos, and R. S. Felger., eds.), Oxford University Press, New York.

! 59 GONZALEZ-RUIZ, N., AND S. T. ALVAREZ-CASTANEDA. 2005. Peromyscus bullatus. Mammalian Species 770:1–3.

GONZALEZ-RUIZ, N., J. RAMÍREZ-PULIDO, AND H. H. GENOWAYS. 2006. Geographic distribution, taxonomy, and conservation status of Nelson’s woodrat (Neotoma nelsoni) in Mexico. The Southwestern Naturalist 51:112–116.

GOODWIN, H. T. 1995. Pliocene-Pleistocene biogeographic history of prairie dogs, genus Cynomys (Sciuridae). Journal of Mammalogy 76:100–122.

GRIFFITHS, C. 1997. Correlation of functional domains and rates of nucleotide substitution in Cytochrome b. Molecular Phylogenetics and Evolution 7:352–365.

GUINDON, S., AND O. GASCUEL. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology: 52: 696–704.

HAFNER, D. J., AND T. L. YATES. 1983. Systematic status of the Mojave ground squirrel, Spermophilus mohavensis (subgenus Xerospermophilus). Journal of Mammalogy 64:397–404.

HAFNER, M. S., J. E. LIGHT, D. J. HAFNER, S. V. BRANT, T. A. SPRADLING, AND J. W. DEMASTES. 2005. Cryptic species in the Mexican pocket gopher Cratogeomys merriami. Journal of Mammalogy, 86:1095–1108.

HALL, E. R. 1981. The mammals of North America. John Wiley & Sons, New York, 1:1–600 + 90; 2:601–1181 + 90.

HAROLD, A. S., AND R. D. MOOI. 1994. Areas of endemism: definition and recognition criteria. Systematic Biology 43:261–266.

HARRISON, R. G., S. M. BOGDANOWICZ, R. S. HOFFMANN, E. YENSEN, AND P. W. SHERMAN. 2003. Phylogeny and evolutionary history of the ground squirrels (Rodentia: Marmotinae). Journal of Mammalian Evolution 10:249–276.

HAUSDORF, B. 1988. Weighted ancestral area analysis and a solution of the redundant distribution problem. Systematic Biology 47:445–456.

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.

HERRON, M. D., T. A. CASTOE, AND C. L. PARKINSON. 2004. Sciurid phylogeny and the paraphyly of holartic ground squirrels (Spermophilus). Molecular Phylogenetics and Evolution 31:1015– 1030.

HERSHKOVITZ, P. 1949. Status of names credited to Oken, 1816. Journal of Mammalogy 30:289– 301.

! 60 HO S.Y.W., AND M. J. PHILLIPS. 2009. Accounting for calibration uncertainty in phylogenetic estimation of evolutionary divergence times. Systematic Biology 58:367–380.

HOFFMANN, R. S., AND J. K. JONES, JR. 1970. Influence of late-glacial and post-glacial events on the distribution of recent mammals on the northern Great Plains. Pp. 355–394 in Pleistocene and recent environments of the central Great Plains (W. Dort and K. K. Jones Jr., eds.), The University Press of Kansas, Lawrence.

HOWELL, A. H. 1938. Revision of the North American ground squirrels, with a classification of the North American Sciuridae. North American Fauna 56:1–256.

HUELSENBECK, J. P., AND F. RONQUIST. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17:754–755.

IRWIN, D. M., T. D. KOCHER, AND A. C. WILSON. 1991. Evolution of the cytochrome b gene in mammals. Journal of Molecular Evolution 2:37–55.

JANSA, S. A., T. C. GIARLA, AND B. K. LIM. 2009. The phylogenetic position of the rodent genus Typhlomys and the geographic origin of Muroidea. Journal of Mammalogy 90:1083–1094.

KELT, D. A., M. S. HAFNER, AND THE AMERICAN SOCIETY OF MAMMALOGISTS' AD HOC COMMITTEE FOR GUIDELINES ON HANDLING RODENTS IN THE FIELD. Updated guidelines for protection of mammalogists and wildlife researchers from hantavirus pulmonary syndrome (HPS). Journal of Mammalogy 91:1524–1527.

KIMURA, M. 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies if nucleotide sequences. Journal of Molecular Evolution 16:111–120.

LAMOREUX, J. F., J. C. MORRISON, T. H. RICKETTS, D. M. OLSON, E. DINERSTEIN, M. W. MCKNIGHT, AND H. H. SHUGART. 2006. Global tests of biodiversity concordance and the importance of endemism. Nature 440:212–214.

LUNA, I., J. J. MORRONE, AND D. ESPINOSA. 2007. Biodiversidad de la Faja Volcánica Transmexicana. Comisión Nacional para la Biodiversidad, México, Instituto de Biología, Universidad Nacional Autónoma de México, México, D. F.

MADDISON, D. R., AND W. P. MADDISON. 2000. MacClade 4: Analysis of phylogeny and character evolution. Version 4.0. Sinahuer Associates, Sunderland, Massachusetts.

MADDISON, W. P., AND D.R. MADDISON. 2010. Mesquite: a modular system for evolutionary analysis. Version 2.73 http://mesquiteproject.org (accessed 20 October 2011).

MANTOOTH, S. J., C. JONES, AND R. D. BRADLEY. 2000. Molecular systematics of Dipodomys elator (Rodentia: Heteromyidae) and its phylogeographic implications. Journal of Mammalogy 81:885–894.

! 61 MCCULLOUGH, D. A., AND R. K. CHEESER. 1997. Genetic variation among populations of the Mexican prairie dog. Journal of Mammalogy 68:555–560.

MERRIAM, C. H. 1893. Description of eight new ground squirrels of the genera Spermophilus and Tamias from California, Texas and Mexico. Proceedings of the Biological Society of Washington 8:129–138.

MORÁN-ZENTENO, D. J. 1994. The geology of the Mexican Republic. 146 pp. American Association of Petroleum Geologists, Tulsa, USA.

MYERS, N., R. A. MITTERMEIER, C. G. MITTERMEIER, G. A. B. FONSECA, AND J. KENTS. 2000. Biodiversity hotspots for conservation priorities. Nature 403:853–858.

NABHOLZ, B., S. GLÉMIN, AND N. GALTIER. 2009. The erratic mitochondrial clock: variations of mutation rate, not population size, affect mtDNA diversity across birds and mammals. BMC Evolutionary Biology 9:54.

NEDBAL, M. A., M. W. ALLARD, AND R. L. HONEYCUTT. 1994. Molecular systematics of hystricognath rodents: Evidence from the mitochondrial 12S rRNA gene. Molecular Phylogenetics and Evolution 3:206–220.

NEISWENTER, S., AND B. R. RIDDLE. 2010. Diversification of the Perognathus flavus species group in emerging arid grasslands of western North America. Journal of Mammalogy 91:348–362.

PIZZIMENTI, J. J. 1975. Evolution of the prairie dog genus Cynomys. Occasional Papers of the Museum of Natural History, University of Kansas 39:1–73.

PLATNICK, N. I. 1991. On areas of endemism. Systematic Zoology 28:537–546.

POSADA, D. 2008. jModeltest: phylogenetic model averaging. Molecular Biology and Evolution 25:1253–1256.

RAMBAUT A., AND A. J. DRUMMOND. 2007. Tracer v1.4, Available from http://beast.bio.ed.ac.uk/Tracer (accessed 20 October 2011).

RAMBAUT A., AND A. J. DRUMMOND. 2009. TreeAnnotator v1.6.0: MCMC Output Analysis. Available at http://beast.bio.ed.ac.uk/TreeAnnotator (accessed 20 October 2011).

RONQUIST, F., AND J. P. HUELSENBECK. 2003. MRBAYES 3. Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572–1574.

SAIKI, R. K., G. H. GELFAND, S. STOFFEL, S. J. SHARF, R. HIGUCHI, G. T. HORN, K. B. MULLIS, AND H. A. ERLICK. 1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487–491.

! 62 SHREVE, F. 1942. The desert vegetation of North America. Botanical Review 8:195–246.

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.

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.

STEPPAN, S. J., M. R. AKHVERDYAN, E. A. LYAPUNOVA, D. G. FRASER, N. N. VORONTSOV, R. S. HOFFMANN, AND M. J. BRAUN. 1999. Molecular phylogeny of the marmots (Rodentia: Sciuridae): tests of evolutionary and biogeographic hypotheses. Systematic Biology 48:715– 734.

SWOFFORD, D. L. 2003. PAUP*: Phylogenetic analysis using parsimony (*and other methods). Version 4.0b2.0a. Sinauer Associates, Sunderland, Massachusetts.

TAMURA, K., D. PETERSON, N. PETERSON, G. N, 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.

URIBE-ALCOCER, A., R. RUÍZ C., F. RODRÍGUEZ F., AND A. LAGUARDA F. 1978. The chromosomes of Spermophilus spilosoma cabrerai. Mammalian Chromosomal Newsletter 19:81–85.

URIBE-ALCOCER, A., AND A. AHUMADA-MEDINA. 1990. The karyotype of Spermophilus perotensis (Sciuridae: Rodentia). Cytobios 64:87–91.

VALDÉZ, M., AND G. CEBALLOS. 1997. Conservation of endemic mammals of Mexico: The Perote ground squirrel (Spermophilus perotensis). Journal of Mammalogy 78:74–82.

VELASCO DE LEÓN, P. M., J. ARELLANO, A. SILVA P., AND S. Y. GUARNEROS. 2007. Aspectos geológicos y paleontológicos. Pp. 25–38 in Biodiversidad de la Faja Volcánica Transmexicana (I. Luna, J. J. Morrone and D. Espinosa., eds.), CONABIO, México.

! 63 CHAPTER 4 MOLECULAR PHYLOGENETICS AND TEMPORAL CONTEXT FOR THE DIVERSIFICATION OF THE SOUTHERN ROCK DEERMOUSE, PEROMYSCUS DIFFICILIS (RODENTIA: CRICETIDAE)

4.1 INTRODUCTION

Although most studies of desert mammals of North America have focused on species confined to the Chihuahuan, Peninsular, or Sonoran deserts, recent studies have begun to explore temporal aspects of species-level diversification in mammals whose distribution extends southward from the Mexican Plateau into the arid lands associated with the Trans-Mexico

Volcanic Belt (TMVB; Fernández et al. in press; Neiswenter and Riddle 2010). This study focuses on one such species, the southern rock deermouse (Peromyscus difficilis), whose distribution in a wide variety of habitats (Fernández et al. 2010) affords an unusual opportunity to explore the effects of geology and climate on timing of phyletic diversification in a North

American rodent species.

Peromyscus difficilis is endemic to Mexico and is found throughout the Mexican Plateau,

Sierra Madre Oriental, and southward into north central Oaxaca (Fig. 4.1; Fernández et al. 2010;

Hall 1981). Current taxonomy divides P. difficilis into five subspecies (Fernández et al. 2010;

Hall 1981; Hoffmeister and de la Torre 1961; Fig. 4.1). Habitats occupied by P. d. amplus, P. d. difficilis, P. d. petricola, and P. d. saxicola include dry rocky areas, and oak-pine forests at higher elevations (2,000–2,400 m) and dry rocky hillsides with scattered juniper trees at lower elevations

(1,500–1,900 m; Fernández et al. 2010). In contrast, P. d. felipensis is found in wet coniferous forests at high elevations (2,500–3,500 m) in mountains surrounding the Valley of

! 64

Fig. 4.1—Geographical distribution of Peromyscus difficilis and P. nasutus in Mexico and the United States (redrawn from Fernández et al. 2010 and Hall 1981). Numbered dots show collection localities listed in Appendix 4.1 for P. nasutus and the five currently recognized subspecies of P. difficilis.

! 65 Mexico and in mountains near the city of Oaxaca (Goodwin 1954; Müdespacher-Ziehl et al.

2005; Navarro-Frías et al. 2007; Villa-Ramírez 1953).

Hoffmeister and de la Torre (1961) examined morphological variation in P. difficilis throughout its range and recognized eight subspecies, the current five plus three subspecies presently assigned to P. nasutus (P. nasutus nasutus, P. n. griseus, and P. n. penicillatus; Musser and Carleton 2005). Zimmerman et al. (1975, 1978) studied allelic differentiation at 23 protein loci in populations of P. difficilis from Zacatecas and Colorado and elevated P. nasutus to species status (supported by Avise et al. 1979; Carleton 1989). More recent studies of Peromyscus relationships based on the mitochondrial cytochrome-b gene (Durish et al. 2004; Bradley et al.

2007) have reaffirmed the species status of P. difficilis and P. nasutus.

Phylogeographic relationships among the subspecies of P. difficilis are not well understood. Whereas all populations examined to date have a chromosomal diploid number of

48, variation in fundamental number (FN) within the species (Arellano-Meneses et al. 2000;

Müdespacher-Ziehl et al. 2005; Robbins and Baker 1981; Zimmerman et al. 1975) suggests possible presence of genetic variation that remains undiscovered.

In this study, phylogeographic relationships among the subspecies of P. difficilis and between P. difficilis and its close relatives are examined using nuclear and mitochondrial DNA sequences analyzed by maximum likelihood and Bayesian approaches. Estimates of divergence times based on Bayesian analyses of genetic data affords an unusual opportunity to test the role of geographic barriers and climate shifts in generating the pattern of evolutionary relationships we see today in this species.

! 66 4.2 MATERIALS AND METHODS

Sampling, amplification, and sequencing.—Two mitochondrial genes (cytochrome-b,

Cytb, and 12S ribosomal RNA, 12S) and two nuclear genes (growth hormone receptor, GHR, and interphotoreceptor retinoid-binding protein, IRBP) were sequenced for this analysis. In the initial phase of the analysis, Cytb sequences were generated for P. d. amplus (n = 12 individuals), P. d. difficilis (n = 6), P. d. felipensis (n = 3), P. d. petricola (n = 4), P. d. saxicola (n = 3), and P. nasutus (n = 5). Following analysis of the Cytb data, representatives of each major clade were sequenced for 12S, GHR, and IRBP. Sequences for the 12S gene were generated for P. d. amplus

(n = 4), P. d. difficilis (n = 4), P. d. felipensis (n = 1), P. d. petricola (n = 3), P. d. saxicola (n =

1), and P. nasutus (n = 3). GHR sequences were obtained for P. d. amplus (n = 6), P. d. difficilis

(n = 1), P. d. felipensis (n = 1), P. d. petricola (n = 4), P. d. saxicola (n = 1), and P. nasutus (n =

2). Finally, IRBP sequences were generated for P. d. amplus (n = 6), P. d. difficilis (n = 2), P. d. felipensis (n = 1), P. d. petricola (n = 4), P. d. saxicola (n = 1), and P. nasutus (n = 2). Most of the tissue samples were collected in the field by the author following the guidelines for research on wild mammals approved by the American Society of Mammalogists (Kelt et al. 2010; Sikes et al. 2011). The remainder of the tissue samples were generously donated by museums (Appendix

4.1), and 22 additional sequences (including outgroup sequences for P. attwateri, P. nasutus, P. truei, and additional outgroups listed in Appendix 4.2) were downloaded from GenBank.

Extractions of genomic DNA were performed using a commercial kit (DNeasy Blood and

Tissue Kit; Qiagen Inc., Valencia, California). Mitochondrial and nuclear genes were amplified by polymerase chain reaction using the following universal primers developed for rodents: MVZ-

05 and H15915 for Cytb (Irwin et al. 1991); 12S L82 and 12S H900 for 12S (Nedbal et al. 1994);

GHR1f and GHRend1f for GHR (Jansa et al. 2009); and IRBP-A and IRBP-B for IRBP

! 67 (Stanhope et al. 1992). Thermal-cycling parameters for both mitochondrial genes were: initial denaturation at 95°C for 2 min, 27 cycles of denaturation at 95°C for 1 min, annealing at 49°C for

1 min, extension at 72°C for 2 min, and final extension at 72°C for 7 min. Amplification parameters for GHR were: denaturation at 94°C for 5 min, 34 cycles of denaturation at 94°C for

15 sec, annealing at 60°C for 1 min, extension at 72°C for 1.5 min, and final extension for 10 min at 72°C. Parameters for IRBP were: denaturation at 95°C for 10 min, 27 cycles of denaturation at

95°C for 25 sec, annealing at 58°C for 20 sec, extension at 72°C for 1 min, and final extension for 10 min at 72°C. Agarose gels (2%) were used to visualize amplified products. Cleaning of the amplified products was performed with Polyethylene Glycol (PEG) and DNA sequencing was performed for both light and heavy strands using Big Dye Terminator v1.1, v3.1 (Applied

Biosystems, Foster City, California) in an automated 3100 Genetic Analyzer (Applied

Biosystems) at the Museum of Natural Science, Louisiana State University.

Data Analysis.—Editing and alignment of sequences were done with Sequencher 4.7

(Gene Codes Corporation, Ann Arbor, Michigan). The authenticity of each gene was confirmed by amino acid translation and by BLAST searches in GenBank

(http://blast.ncbi.nlm.nih.gov/Blast.cgi). To enable comparison of our results with those of previous studies, average uncorrected sequence divergence values for the Cytb gene were corrected using the Kimura 2-parameter substitution model (Kimura 1980) in PAUP*

(Phylogenetic Analysis using Parsimony, version 4.0b, Swofford 2003). This model has been used widely in studies of Cytb variation in mammals (Bradley and Baker 2001; Honeycutt et al.

1995).

Phylogenies were estimated using the maximum likelihood (ML) algorithms in PAUP*

4.0b (Swofford 2003) and PhyMl 3.0 (Guindon and Gascuel 2003) and the Bayesian Inference

! 68 (BI) approach in MrBayes (version 3.2-cvs; Huelsenbeck and Ronquist 2001). Best-fit models for all analyses were selected using the Akaike Information Criterion (AIC; Akaike 1973) and the programs jModeltest 0.1.1 (Guindon and Gascuel 2003; Posada 2008), MrModeltest (version 2.2;

Nylander 2004), and Modeltest (version 3.7; Posada and Buckley 2004; Posada and Crandall

1998). The TrN+G model was selected for Cytb, TIM2+G for 12S, TPM1uf+G for GHR,

TPM3uf+I+G for IRBP, and TPM1uf+G for the partitioned analyses. Clade support was assessed with 500 bootstrap replicates (Felsenstein 1985) in PAUP* 4.0b, and clade support in BI analyses was evaluated using posterior probabilities.

Tree searches in ML analyses were performed with the starting trees obtained from 100 random stepwise additions followed by tree-bisection-reconnection (TBR) branch swapping. In

BI analyses, best-fit models were applied to each data partition with unlinked parameters and allowing rate variation. The Metropolis Markov Chain Monte Carlo analysis consisted of two independent runs of 10 x 106 generations in which trees were sampled every 103 generations, resulting in 104 samples for each run. The first 103 trees of each run were discarded as burn-in, and a majority-rule consensus tree was constructed using the final 1.8 x 103 trees. The analysis was stopped when the average standard deviation of split frequencies approached zero, and convergence also was assessed using Tracer 1.5 (Rambaut and Drummond 2007). The combined data set was analyzed in a partitioned manner to allow independent convergence on optimal values for each fragment (Ronquist and Huelsenbeck 2003).

Each gene fragment was analyzed separately then compared visually to evaluate potential conflict among resulting gene trees. Nodes were considered well supported if there was > 85% bootstrap support or significant Bayesian posterior probabilities > 0.90. Because analyses of

! 69 individual data sets showed no conflicting clades that were well supported, all genes were combined and analyzed in a single matrix partitioned by gene.

Analysis of divergence times.—The program *BEAST version 1.6.0 (Drummond and

Rambaut 2007) was used to estimate the times of divergence among P. difficilis, P. nasutus, P. truei, and P. attwateri. Several representative species from clade VI of the tribe

Reithrodontomyini (Miller and Engstrom 2008) were included in the *BEAST analysis, as follows: Habromys ixtlani, Megadontomys thomasi, Neotomodon alstoni, Onychomys arenicola,

O. leucogaster, Osgoodomys banderanus, Peromyscus crinitus, P. eremicus, P. levipes, P. maniculatus, P. melanophrys, and P. melanotis (Appendix 4.2). Isthmomys pirrensis and

Reithrodontomys sumichrasti were included as outgroups. A Yule tree prior was used, implicitly considering the gene tree to represent the species tree, and the HKY+I+G model was selected as the substitution model. A relaxed clock with a lognormal distribution allowing rate variation among sites was used. Chains were run for 107 generations, sampling the parameter every 103 generations. Convergence statistics were checked for effective sample sizes using Tracer version

1.5 (Rambaut and Drummond 2007). Consensus trees were generated from the resulting trees using TreeAnnotator version 1.6.0 (Rambaut and Drummond 2009) after elimination of 25% as burn-in.

Two fossil-based dates were used to calibrate the *BEAST analysis. The first documents the separation of Onychomys from the main stock of Peromyscus in the early Pliocene (ca. 5.3 to

3.6 mya; Korth 1994). The second sets the minimum age for P. eremicus in late Pleistocene (ca.

126,000 to 5,000 years ago; Martin 1968). To account for uncertainty in the fossil-based calibrations, the fossil-based dates were modeled as lognormal distributions rather than point calibrations (Ho and Phillips 2009).

! 70 4.3 RESULTS

Phylogenetic analyses.—A total of 1,140 base pairs (bp) of Cytb, 820 bp of 12S, 654 bp of

GHR, and 775 bp of IRBP were sequenced, yielding 3,389 bp for use in analyses. Cytb divergence values (Table 4.1) show two well differentiated units within P. difficilis, one distributed in the Sierra Madre Oriental and Mexican Plateau (the subspecies petricola and difficilis, respectively; Fig. 4.1) and the other occupying the eastern end of the TMVB southward into Oaxaca (saxicola, amplus, and felipensis, respectively). These clades show an average Cytb divergence of 6.7%, whereas within-clade divergence is only 0.7% within the northern clade and

1.2% within the southern clade. Average Cytb divergence between P. nasutus and the northern clade of P. difficilis is 6.85%, and divergence between P. nasutus and the southern clade of P. difficilis is 7.87%. Finally, P. attwateri shows an average divergence of 7.9% from the P. nasutus + P. difficilis clade, and P. truei shows an average of 13.2% Cytb divergence from all other taxa included in Table 4.1.

ML and BI analyses of the Cytb sequences yielded very similar trees that showed only minor differences near the tips of the branches. The ML tree (Fig. 4.2; BI tree available on request) showed subdivision of P. difficilis into the same northern and southern clades evident from the distance data (Table 4.1). These two clades were linked strongly with P. nasutus (100% bootstrap support and 1.0 posterior probability), but relationships among these three lineages were unresolved (Fig. 4.2). ML and BI analyses of 12S generated trees that were concordant with the Cytb tree, but with lower branch support values (trees not shown but available on

! 71 Table 4.1. Percent sequence divergence in the cytochrome-b gene (Kimura 2-parameter model) among eight taxa of Peromyscus, including the five currently recognized subspecies of P. difficilis. Geographic distributions of the two major clades within P. difficilis are indicated. TMVB = Trans-Mexico Volcanic Belt.

Mexican Plateau and Eastern end of TMVB

Sierra Madre Oriental southward into Oaxaca

P. truei P. attwateri P. nasutus P. d. petricola P. d. difficilis P. d. saxicola P. d. amplus P. d. felipensis

P. truei – 13.2 13.0 12.6 13.2 13.4 13.6 13.1

P. attwateri – 7.9 7.4 7.2 8.1 8.6 8.2

P. nasutus – 6.9 6.8 7.6 7.7 8.3

P. d. petricola – 0.7 6.8 6.7 6.3

P. d. difficilis – 6.9 6.8 6.7

P. d. saxicola – 1.5 1.6

P. d. amplus – 0.4

P. d. felipensis –

! 72

Fig. 4.2.—Phylogenetic relationships among the 5 subspecies of P. difficilis (sensu Durish et al. 2004), P. nasutus, and allied taxa based on a maximum likelihood (ML) analysis of cytochrome-b sequences. A Bayesian analysis of the same data yielded a tree with identical topology. Locality numbers are indicated for ingroup taxa and refer to mapped localities in Fig. 4.1 and localities listed in Appendix 4.1. Numbers above branches indicate ML bootstrap support, and numbers below branches indicate Bayesian posterior probabilities. The subspecies amplus, felipensis, and saxicola, formerly assigned to P. difficilis, are recognized as junior synonyms of P. felipensis (Merriam 1898) in this study.

! 73 request). Finally, ML and BI analyses of the two nuclear genes yielded trees with well- supported resolution only at the node defining the northern clade of P. difficilis and at the node defining the southern clade of P. difficilis (trees not shown but available on request).

ML and BI analyses of the combined data set (partitioned by gene) yielded identical and well-resolved trees (Fig. 4.3). Again I observed the 3 clades evident in the

Cytb analysis (northern P. difficilis, southern P. difficilis, and P. nasutus; Fig. 4.2), but the trichotomy that was unresolved in the Cytb analysis is now resolved to show P. nasutus sister to the northern clade of P. difficilis (90% bootstrap support and 0.94 posterior probability).

Estimates of divergence time.—The tree obtained in the *BEAST analysis (Fig.

4.4) showed the same northern and southern P. difficilis clades evident in the Cytb (Fig.

4.2) and 4-gene (Fig. 4.3) analyses. However, the *BEAST tree differed from the other trees in showing P. nasutus sister to P. attwateri. The Tracer analysis in *BEAST confirmed a high effective sample size (> 3,000) for all parameters. All mean estimates of divergence time for the P. attwateri, P. difficilis, and P. nasutus clades were < 1.0 mya

(Pleistocene). The split between the northern and southern clades of P. difficilis was estimated at approximately 0.7 mya.

4.4 DISCUSSION

P. difficilis is composed of two genetically well-differentiated lineages, one found in the Sierra Madre Oriental and Mexican Plateau (the subspecies petricola and difficilis, respectively; Fig. 4.1) and the other occupying the eastern end of the TMVB southward into Oaxaca (saxicola, amplus, and felipensis, respectively). This subdivision of P.

! 74 Fig. 4.3.—Phylogenetic relationships among selected specimens of P. difficilis and P. nasutus based on a partitioned maximum likelihood (ML) analysis of two mitochondrial genes (Cytb and 12S) and 2 nuclear genes (GHR and IRBP). A Bayesian analysis of the same data yielded a tree with identical topology. Locality numbers are indicated for ingroup taxa and refer to mapped localities in Fig. 4.1 and localities listed in Appendix 4.1. Numbers above branches indicate ML bootstrap support, and numbers below branches indicate Bayesian posterior probabilities. The subspecies amplus, felipensis, and saxicola, formerly assigned to P. difficilis, are recognized as junior synonyms of P. felipensis (Merriam 1898) in this study. Original subspecies assignments are indicated in parentheses: a = amplus, d = difficilis, f = felipensis, p = petricola, s = saxicola.

! 75

difficilis into two divergent clades is consistent with the findings of Durish et al. (2004), although they did not include samples of P. d. petricola or P. d. felipensis in their study.

The 4-gene partitioned analysis (Fig. 4.3) shows a sister relationship between the northern clade of P. difficilis and P. nasutus, with the southern clade of P. difficilis outside this group.

Specimens from Durango originally identified as P. difficilis by Avise et al.

(1979) and Durish et al. (2004) were represented by the specimen from locality 5 in this study (Fig. 4.1). Durish et al. (2004) reported that their specimens from Durango were more closely related to P. nasutus than to P. difficilis and my results concur in showing strong evidence (Fig. 4.2) that these specimens from Durango represent a southward range extension of P. nasutus into the Sierra Madre Occidental of Mexico. This finding also is consistent with the chromosomal findings of Robbins and Baker (1981), who reported a P. nasutus-like karyotype from Peromyscus specimens from Durango. A detailed reassessment of the distributions of P. difficilis and P. nasutus in northern

Mexico is needed.

Average Cytb genetic distances between the three major clades of Peromyscus identified in this analysis (P. nasutus and the northern and southern clades of P. difficilis) range from 6.7% to 7.9% (Table 4.1) and compare favorably with Cytb distances measured between sister species of Peromyscus reported in previous studies (5.8 –7.7%;

Baker and Bradley 2006).

Biogeographical considerations.— Recent distributional, paleontological, and molecular studies of Mexican mammals have shown that most phyletic diversification in

! 76 desert-dwelling lineages took place during Miocene and Pliocene times (Hafner and

Riddle 1997; Riddle 1995; Riddle et al. 2000; Vrba, 1992). In contrast, phyletic diversification in lineages of mammals inhabiting the highlands of the TMVB, the Sierra

Madre Oriental, and the Sierra Madre Occidental seems to have been influenced more strongly by Pleistocene climate cycles (Findley 1969; León-Paniagua et al. 2007; Orr

1960). Because populations of mammals currently recognized as P. difficilis inhabit both xeric habitats at mid-elevations and mesic habitats at higher elevations in Mexico, it was impossible to predict prior to this study whether phyletic diversification in P. difficilis was influenced more by Pleistocene or pre-Pleistocene climatic and geologic events.

The divergence times estimated in this study (Fig. 4.4) suggest that the two lineages of P. difficilis diverged during the Pleistocene. If so, the major mountain chains of northern Mexico (TMVB, Sierra Madre Oriental, and Sierra Madre Occidental), instead of acting as a barrier to range expansion in rodents, may have acted as a dispersal corridor connecting the three mountain ranges during warm interglacial cycles. Repeated expansion and contraction of the geographic range of P. difficilis may have resulted in isolation of the southern lineage in the eastern TMVB promoting the genetic differentiation we see today between the northern and southern clades of P. difficilis. The same or a similar process has been hypothesized to explain phyletic diversification in other peromyscine taxa of Mexico, including Reithrodontomys, Habromys, the P. aztecus species group, and the P. maniculatus complex (Arellano et al 2005; Avise et al. 1979;

Bradley et al. 2004a, 2004b; Dawson 2005; Hibbard 1968; Kalkvik et al. 2011, León-

Paniagua et al. 2007; Sullivan et al. 1997, 2000).

! 77

Fig. 4.4.—Estimates of divergence times in P. difficilis, P. felipensis, and close relatives calculated from *BEAST analysis. Numbers at nodes are mean divergence dates, and bars show 95% credibility intervals. Additional outgroup taxa (listed in the text) were included in the analysis, which used two dated fossils to calibrate the tree.

! 78 Taxonomic conclusions.—P. difficilis was originally described by Allen (1891) from a specimen taken in Sierra de Valparaiso, Zacatecas. Seven years later, P. felipensis was described (Merriam 1898) from specimens collected in Cerro San Felipe, near

Oaxaca City, and in 1909, Osgood placed felipensis in synonymy under P. difficilis.

Herein, I return P. felipensis to full species status based on phylogenetic analysis of mitochondrial and nuclear genes, estimates of divergence time from its sister lineage

(most likely P. nasutus), geographic distribution, and morphology. Synonymies of P. difficilis and P. felipensis follow.

Peromyscus difficilis (J. A. Allen, 1891)

Zacatecan deermouse

Vesperimus difficilis J. A. Allen, 1891:518. Type locality “Sierra de Valparaiso,

Zacatecas, Mexico.”

[Peromyscus] difficilis Trouessart, 1897:518. First use of current name combination.

P. d. petricola Hoffmeister and de la Torre, 1959:167–168. Type locality “12 mi. E. San

Antonio de las Alazanas, 9000 ft., Coahuila, Mexico.”

Geographic range.— Peromyscus difficilis is endemic to Mexico and is found throughout the Sierra Madre Oriental and adjacent mountain ranges from southwestern

Chihuahua and southeastern Coahuila, southward through the low hills of Durango and

Zacatecas. This species probably occurs in the Sierra Madre Occidental as well. Its range continues southward from Zacatecas into parts of San Luis Potosi and mountainous regions of Guanajuato (Fig. 4.1; Fernández et al. 2010; Hall 1981; Hoffmeister and de la

Torre 1961; Osgood 1909).

! 79 Description.—Size medium to large for Peromyscus; total length 230 mm in adult males (range 223–237 mm); 237 mm in adult females (range 227–250 mm); tail length

128.6–132.5 mm. Fur dull and dark dorsally with little or no ochraceous, and heavily overlaid with black. Auditory bullae greatly inflated; brain case and interorbital region broad; skull, nasals, and toothrow long (Hoffmeister and de la Torre 1961; Fernández et al. 2010).

Comments.—The currently recognized subspecies P. d. difficilis and P. d. petricola were not recognized as monophyletic clades in this analysis (Fig. 4.2), so until additional research on subspecific variation within P. difficilis is completed, no formal subspecies are recognized in this study.

Peromyscus felipensis Merriam, 1898

Southern rock deermouse

P. felipensis Merriam, 1898:122–123. Type locality “Cerro San Felipe, Oaxaca, Mexico

(alt. 10,200 ft).”

P. amplus Osgood, 1904:62–63. Type locality “[San Juan Bautista] Coixtlahuaca,

Oaxaca, Mexico.”

P. difficilis saxicola Hoffmeister and de la Torre, 1959:168–169. Type locality

“Cadereyta, 2100 meters, Querétaro, México.”

Geographic range.— P. felipensis occurs in Querétaro, northern Hidalgo, the mountains of southern Hidalgo, northern México state, Tlaxcala, Puebla, westcentral

Veracruz, and northcentral Oaxaca (Goodwin 1954; Hall 1981; Hoffmeister and de la

Torre 1961; Osgood 1904, 1909; Villa-Ramírez 1953; Fig. 4.1).

! 80 Description.—Size medium to large for Peromyscus; total length of 10 adult topotypes (including both sexes) 241.5 mm (range 225–248 mm); tail length 118–132 mm. Fur glossy and unicolored dorsally; dorsal coloration ochraceous, brownish or reddish in northern populations to blackish in the southernmost populations. Southern populations of P. felipensis (the former subspecies P. d. amplus and P. d. felipensis) have a narrower brain case and interorbital region compared to P. difficilis; northern populations (the former P. d. saxicola) are smaller in overall size than southern populations (Hoffmeister and de la Torre 1959, 1961; Fernández et al. 2010).

Comments.— The currently recognized subspecies P. d. felipensis, P. d. amplus, and P. d. saxicola were not recognized as monophyletic clades in this analysis (Fig. 4.2), so until additional research on subspecific variation within P. felipensis is completed, no formal subspecies are recognized in this study.

4.5 LITERATURE CITED

AKAIKE, H. 1973. Information theory as an extension of the maximum likelihood principle. In: B.N. Petrov and F. Csaki, Editors, Second International Symposium on Information Theory, Akademiai Kiado, Budapest.

ALLEN, J. A. 1891. Notes on new or little-known North American Mammals, based on recent additions to the Collection of Mammals in the American Museum of Natural History. Bulletin of the American Museum of Natural History 3:263– 310.

ARELLANO, E., F. X. GONZALEZ-COZATL, AND D. S. ROGERS. 2005. Molecular systematics of the Middle American harvest mice Reithrodontomys (Muridae), estimated from mitochondrial cytochrome b gene sequences. Molecular Phylogenetics and Evolution 37:529–540.

ARELLANO-MENESES, A. G., L. A. HERNÁNDEZ-CARBAJAL, I. E. LIRA-GALERA, G. RUIZ- GUZMÁN, AND C. MUDESPACHER-ZIEHL. 2000. Karyotypical studies on Peromyscus difficilis amplus (Rodentia: Muridae). Cytologia 65:25–28.

AVISE, J. C., M. H. SMITH, AND R. K. SELANDER. 1979. Biochemical polymorphism and systematics in the genus Peromyscus. VII. Geographic differentiation in

! 81 members of the truei and maniculatus species groups. Journal of Mammalogy 60:177–192.

BAKER, R. J., AND R. D. BRADLEY. 2006. Speciation in mammals and the genetic species concept. Journal of Mammalogy 87:643–662.

BRADLEY, R. D., AND R. J. BAKER. 2001. A test of the genetic species concept: cytochrome–b sequences and mammals. Journal of Mammalogy 82:969–973.

BRADLEY, R. D., D. S. CARROLL, M. L. HAYNIE, R. M. MARTINEZ, 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., F. MENDEZ-HARCLERODE, M. J. HAMILTON, AND G. CEBALLOS. 2004b. A new species of Reithrodontomys from Guerrero, Mexico. Occasional Papers, Museum of Texas Tech University 231:1–12.

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.

CARLETON, M. D. 1989. Systematics and evolution. Pp. 7–141, in Advances in the study of Peromyscus (Rodentia), (G. L. Kirkland and J. N. Layne, eds.). Texas Tech University Press, Lubbock, Texas.

DAWSON, W. D. 2005. Peromyscine biogeography. Pp. 145–156, in Contribuciones mastozoológicas en homenaje a Bernardo Villa, (V. Sánchez-Cordero and R. Medellín, eds.). Instituto de Biología, Instituto de Ecología, Uiversidad Nacional Autónoma de México, Conabio, México, D. F.

DRUMMOND, A. J., AND A. RAMBAUT. 2007. *BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology 7:214.

DUBOIS, J. Y., D. RAKOTONDRAVONY, C. HANNI, P. SOURROUILLE, AND F. M. CATZEFLIS. 1996. Molecular ecolutionary relationships of three genera of Nesomyinae, endemic rodent taxa from Madagascar. Journal of Mammalian Evolution 3:239– 260.

DURISH, N. V., 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.

FELSENSTEIN, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783–791.

! 82 FERNÁNDEZ, J. A., F. A. CERVANTES, AND M. S. HAFNER. In press. Molecular systematics and biogeography of the Mexican endemic kangaroo rat, Dipodomys phillipsii (Rodentia: Heteromyidae). Journal of Mammalogy.

FERNÁNDEZ, J. A., F. GARCÍA-CAMPUSANO, AND M. S. HAFNER. 2010. Peromyscus difficilis. Mammalian Species 42:220–229.

FINDLEY, J. S. 1969. Biogeography of southwestern boreal and desert mammals. Pp. 113–128 in Contributions in mammalogy (J. K. Jones, ed.), University of Kansas, Miscellaneus Publications of the Museum of Natural History.

GOODWIN, G. G. 1954. Mammals from Mexico collected by Marian Martin for the American Museum of Natural History. American Museum Novitates 1689:1–16.

GUINDON, S., AND O. GASCUEL. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52:696–704.

HAFNER, D. J., AND B. R. RIDDLE. 1997. Biogeography of Baja California peninsular desert mammals. Pp. 19–68 in Life among the Muses: Papers in Honor of James S. Findley (T. L. Yates, W. L. Gannon, and D. E. Wilson, eds.), The Museum of Southwestern Biology, The University of New Mexico, Albuquerque.

nd HALL, E. R. 1981. The Mammals of North America. Vol. 2, 2 ed. John Wiley & Sons, New York.

HIBBARD, C. W. 1968. Paleontology. Pp. 6–26, in Biology of Peromyscus (Rodentia), (J. A. King, ed.). American Society of Mammalogists Special Publication No. 2.

HO, S. Y. W., AND M. J. PHILLIPS. 2009. Accounting for calibration uncertainty in phylogenetic estimation of evolutionary divergence times. Systematic Biology 58:367–380.

HOFFMEISTER, D. F., AND L. DE LA TORRE. 1959. Two new subspecies of Peromyscus difficilis from Mexico. Proceedings of the Biological Society of Washington 72:167–170.

HOFFMEISTER, D. F., AND L. DE LA TORRE. 1961. Geographic variation in the mouse Peromyscus difficilis. Journal of Mammalogy 42:1–13.

HONEYCUTT, R. L., M. A NEDBAL, R. M. ADKINS, AND L. L. JANACEK. 1995. Mammalian mitochondrial DNA evolution: a comparison of the cytochrome b and cytochrome c oxidase II genes. Journal of Molecular Evolution 40:260–272.

HUELSENBECK, J. P., AND F. RONQUIST. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17:754–755.

! 83 IRWIN, D. M., T. D. KOCHER, AND A. C. WILSON. 1991. Evolution of the cytochrome b gene in mammals. Journal of Molecular Evolution 2:37–55.

JANSA, S. A., T. C. GIARLA, AND B. K. LIM. 2009. The phylogenetic position of the rodent genus Typhlomys and the geographic origin of Muroidea. Journal of Mammalogy 90:1083–1094.

KALKVIK, H. M., I. J. STOUT, T. J. DOONAN, AND C. L. PARKINSON. 2011. Investigating niche and lineage diversification in widely distributed taxa: phylogeography and ecological niche modeling of the Peromyscus maniculatus species group. Ecography 34:1-11.

KELT, D. A., M. S. HAFNER, AND THE AMERICAN SOCIETY OF MAMMALOGISTS' AD HOC COMMITTEE FOR GUIDELINES ON HANDLING RODENTS IN THE FIELD. 2010. Updated guidelines for protection of mammalogists and wildlife researchers from hantavirus pulmonary syndrome (HPS). Journal of Mammalogy 91:1524–1527.

KIMURA, M. 1980. A simple method for estimating evolutionary rate of base substitution through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16:111–120.

KORTH, W. W. 1994. The Tertiary record of rodents in North America. Topics in Geobiology, Vol. 12. Plenum Press, New York.

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 (Rodentia: Cricetidae: Neotominae) in the Mesoamerican highlands. Molecular Phylogenetics and Evolution 42:653–664.

MARTIN, R. A. 1968. Further study of the Friesenhahn cave Peromyscus. The Southwestern Naturalist 12:253–266.

MERRIAM, C. H. 1898. Descriptions of twenty new species and a new subgenus of Peromyscus from Mexico and Guatemala. Proceedings of the Biological Society of Washington 12:115–125.

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.

MÜDESPACHER-ZIEHL, C., R. ESPIRITU-MORA, M. MARTÍNEZ-CORONEL, AND S. GAONA. 2005. Chromosomal studies in two populations of Peromyscus difficilis felipensis (Rodentia: Muridae). Cytologia 70:243–248.

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.

! 84 Wilson and D. M. Reeder, eds.). 3rd ed. Johns Hopkins University Press, Baltimore, Maryland.

NAVARRO-FRÍAS J., N. GONZÁLEZ RUIZ, AND S.T. ÁLVAREZ CASTAÑEDA. 2007. Los mamíferos silvestres de Milpa Alta, Distrito Federal: lista actualizada y consideraciones para su conservación. Acta Zoológica Mexicana (nueva serie) 23:103–124.

NYLANDER, J. A. 2004. MrModeltest version 2.2. Program distributed by the author, Evolutionary Biology Centre, Uppsala University, Uppsala, Sweden.

NEDBAL, M. A., M. W. ALLARD, AND R. L. HONEYCUTT. 1994. Molecular systematics of Histricognath rodents: Evidence from the mitochondrial 12S rRNA gene. Molecular Phylogenetics and Evolution 3:206–220.

NEISWENTER, S., AND B. R. RIDDLE. 2010. Diversification of the Perognathus flavus species group in emerging arid grasslands of western North America. Journal of Mammalogy 91:348–362.

ORR, R. T. 1960. An analysis of the Recent land mammals. Systematic Zoology 9:171– 179.

OSGOOD, W. H. 1904. Thirty new mice of the genus Peromyscus from Mexico and Guatemala. Proceedings of the Biological Society of Washington 17:55–77.

OSGOOD, W. H. 1909. Revision of the mice of the American genus Peromyscus. North American Fauna No. 28. U.S. Department of Agriculture, Bureau of Biological Survey. Government Printing Office, Washington, D.C.

POSADA, D. 2008. jModeltest: phylogenetic model averaging. Molecular Biology and Evolution 25:1253–1256.

POSADA, D., AND T. R. BUCKLEY. 2004. Model Selection and Model Averaging in Phylogenetics: Advantages of Akaike Information Criterion and Bayesian approaches over likelihood ratio tests. Systematic Biology 53:793–808.

POSADA, D., AND K. A. CRANDALL. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:817–818.

RAMBAUT A., AND A. J. DRUMMOND. 2007. Tracer v1.4, Available from http://beast.bio.ed.ac.uk/Tracer. Accessed 14 October 2011.

RAMBAUT A., AND A. J. DRUMMOND. 2009. TreeAnnotator v1.6.0: MCMC Output Analysis. Available at http://beast.bio.ed.ac.uk/TreeAnnotator. Accessed 14 October 2011.

! 85 RIDDLE, B. R. 1995. Molecular biogeography in the pocket mice (Perognathus and Chaetodipus) and grasshopper mice (Onychomys): the late Cenozoic development of a North American aridlands rodent guild. Journal of Mammalogy 76:283–301.

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.

ROBBINS, L. W., AND R. J. BAKER. 1981. An assessment of the nature of chromosomal rearrangements in 18 species of Peromyscus (Rodentia: Cricetidae). Cytogenetics and Cell Genetics 31:194–202.

RONQUIST, F., AND J. P. HUELSENBECK. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572–1574.

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.

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.

SULLIVAN, J. K., 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. K., E. ARELLANO, AND D. ROGERS. 2000. Comparative phylogeography of Mesoamerican highland rodents: concerted versus independent response to past climatic fluctuations. The American Naturalist 155:755–768.

SWOFFORD, D. L. 2003. PAUP*: Phylogenetic analysis using parsimony (*and other related methods). Version 4. Sinauer Associates, Inc., Sunderland, Massachusetts.

VILLA-RAMÍREZ, B. 1953. Mamíferos silvestres del Valle de México. Anales del Instituto de Biología, Universidad Nacional Autónoma de México (Serie Zoología) 23:269–492.

VRBA, E. S. 1992. Mammals as a key to evolutionary theory. Journal of Mammalogy 73:1–28.

! 86 ZIMMERMAN, E. G., B. J. HART, AND C. W. KILPATRICK. 1975. Biochemical genetics of the boylii and truei species groups of the genus Peromyscus (Rodentia). Comparative Biochemistry and Physiology 52b:541–545.

ZIMMERMAN, E. G., C. J. KILPATRICK, AND B. J. HART. 1978. The genetics of speciation in the rodent genus Peromyscus. Evolution 32:565–579.

! 87 CHAPTER 5 MOLECULAR SYSTEMATICS AND BIOGEOGRAPHY OF THE MEXICAN ENDEMIC KANGAROO RAT, DIPODOMYS PHILLIPSII (RODENTIA: HETEROMYIDAE) JESÚS A. FERNÁNDEZ*, FERNANDO A. CERVANTES, AND MARK S. HAFNER Department of Biological Sciences and Museum of Natural Science, Louisiana State University, Baton Rouge, Louisiana 70803 (JAF, MSH) Departamento de Zoología, Instituto de Biología, Universidad Nacional Autónoma de México, Apartado Postal 70-245, México, D. F., 04510, México (FAC, JAF)

5.1 INTRODUCTION

Phillips’ kangaroo rat (Dipodomys phillipsii) is a threatened, endemic rodent of

Mexico (Luiselli 2002). Only a few taxonomic studies (all based on morphology) have focused on this species, and the systematic status of the four currently recognized subspecies is unknown. Dipodomys phillipsii is a medium sized kangaroo rat (total length 230–304 mm) with four toes on the hind limbs, a relatively short tail with black stripes united at the distal one-third of the tail, and a whitish tuft on the tip of the tail

(Hall 1981; Jones and Genoways 1975; Schmidly et al. 1993). This species is distributed along a narrow band of dry, semi-desert habitat extending from southeastern Durango to the Tehuacán–Cuicatlán Valley of southern Puebla and northern Oaxaca (McMahon

1979; Schmidly et al. 1993; Fig. 5.1). The current taxonomy of D. phillipsii, which is based on morphometric and qualitative morphological traits, is as follows: D. p. phillipsii is found mainly in the Valley of México, in the Distrito Federal, and in the states of

México and Hidalgo; D. p. ornatus is known from Durango, Zacatecas, Aguascalientes,

San Luis

! 88

Fig. 5.1.—Map of central Mexico modified from Hall (1981) and Jones and Genoways (1975) showing the geographic distribution, sampling localities, and current subspecies of Phillips’ kangaroo rat, Dipodomys phillipsii, in Mexico. Localities are listed in Appendix 5.1.

! 89 Potosí, Querétaro, Guanajuato, and Jalisco; D. p. perotensis is found in eastern Tlaxcala, central-western Veracruz, and Puebla; and D. p. oaxacae is known from southern Puebla and northern Oaxaca (Hall 1981; Jones and Genoways 1975; Wilson and Reeder 2005;

Fig. 5.1).

Dipodomys phillipsii was described by Gray (1841) based on one specimen from near Real del Monte, Hidalgo. Some 50 years later, Merriam (1894) collected specimens in the northern portion of the species range (Berriozábal, Zacatecas) and in the Oriental

Basin (Cuenca Oriental) in central Mexico (Perote, Veracruz) and described D. ornatus and D. perotensis, respectively. Merriam (1894) emphasized differences in size and fur color between these taxa, and he noted that the skulls were very similar, showing only variation in proportions. Davis (1944) compared individuals of D. phillipsii and D. perotensis and was unable to see the differences that Merriam (1894) reported. As a result, Davis (1944) suggested that D. perotensis be regarded as a subspecies of D. phillipsii. Hooper (1947) trapped specimens of D. phillipsii in Teotitlán, Oaxaca, and described D. p. oaxacae based also on size and fur color, recognizing the cranial similarities between his specimens and those of the geographically nearest subspecies, D. p. perotensis.

Genoways and Jones (1971) analyzed specimens from several populations throughout the range of D. phillipsii and D. ornatus and suggested that a suite of morphological characters, including mastoid breadth, maxillar breadth, interorbital breadth, body size, and pelage coloration could be used to distinguish among the subspecies of D. phillipsii. They reported low levels of within-population variation in cranial dimensions, but high levels of within-population variation in pelage coloration.

! 90 Because they could find no consistent morphological difference between D. ornatus and

D. phillipsii, they reduced D. ornatus to subspecific status under D. phillipsii. Genoways and Jones (1971) generated a phenogram based on a morphological distance matrix but found no congruence between the phenogram and the geographic locations of their samples.

Evolutionary relationships among the geographic subunits of D. phillipsii have remained controversial since publication of the study by Genoways and Jones (1971). In this study we use mitochondrial and nuclear DNA sequence data to test species monophyly and assess phylogenetic relationships within D. phillipsii. To place our results in a larger, historical biogeographical context, we estimate the time of major divergence events within D. phillipsii and among its close relatives.

5.2 MATERIALS AND METHODS

Sampling, amplification, and sequencing.—Samples of D. phillipsii tissue or ear clips were either collected in the field under the authority of the Mexican collecting permit FAUT-0002 issued to FAC or donated to us by museums (Appendix 5.1). We sequenced specimens from 17 localities in northern and central Mexico, including six localities of D. p. ornatus (n = 7 individuals), one locality of D. p. phillipsii (n = 1); eight localities of D. p. perotensis (n = 8), and two localities of D. p. oaxacae (n = 5; Fig. 5.1;

Appendix 5.1). Outgroup sequences for D. agilis, D. californicus, D. compactus, D. deserti, D. elator, D. heermanni, D. merriami, D. microps, D. nelsoni, D. ordii, D. panamintinus, D. spectabilis, and Heteromys irroratus (use of Heteromys follows the taxonomy of Hafner et al. 2007) were obtained from Genbank or generated for this study

(Appendix 5.1). The collection and processing of samples were undertaken following the

! 91 guidelines of the American Society of Mammalogists for use of wild animals in research

(Kelt et al. 2010; Sikes et al. 2011).

Total genomic DNA was extracted from tissue using a standard phenol- chloroform protocol (Darbre 2001; Saunders and Parkes 1999) and a commercial kit

(DNeasy Blood and Tissue Kit; Qiagen Inc., Valencia, California). Two nuclear and two mitochondrial genes were sequenced. The nuclear genes code for the growth hormone receptor (GHR) and the interphotoreceptor retinoid-binding protein (IRBP), and the mitochondrial genes code for a subunit of the respiratory chain protein Cytochrome-b

(Cytb) and the 12S ribosomal RNA gene (12S). Sequences were amplified by PCR (Saiki et al. 1988) using universal primers developed for rodents (Appendix 5.2). The following

PCR parameters were used to amplify both mitochondrial genes from fresh tissue samples: initial denaturation at 95°C for 2 min, followed by 27 cycles of denaturation at

95°C for 1 min, annealing at 49°C for 1 min, and extension at 72°C for 2 min, with a final extension at 72°C for 7 min (Mantooth et al. 2000). PCR parameters for mitochondrial genes obtained from dry skin clips were: denaturation at 94°C for 1 min, followed by 35 cycles at 94°C for 1 min, 50°C for 1 min, and 72°C for 1 min, with a final extension at 72°C for 5 min. PCR parameters for the nuclear gene GHR were: denaturation at 94°C for 5 min, followed by 34 cycles at 94°C for 15 s, 60°C for 1 min, and 72°C for 1.5 min, with a final extension at 72°C for 10 min. PCR amplification of the IRBP gene was performed as follows: denaturation at 95°C for 10 min, followed by

27 cycles at 95°C for 25 s, 58°C for 20 s, and 72°C for 1 min, with a final extension at

72°C for 10 min. Amplifications were performed with 200 ng of DNA in a total volume of 25 µL. Agarose (2%) gels were used to visualize amplified products. PCR products

! 92 were purified using a QIAquick PCR Purification Kit (Qiagen Inc.) with either

Polyethylene Glycol or ExoSAP-IT (Affymetrix, Santa Clara, California). DNA sequencing was performed for both light and heavy strands with a Big Dye Terminator v1.1, v3.1 in an automated 3100 Genetic Analyzer (Applied Biosystems, Foster City,

California) at the Instituto de Biología, Universidad Nacional Autónoma de México and at the Museum of Natural Science, Louisiana State University.

Distance analysis.—Editing and alignment of sequences and matrix manipulations were performed in Sequencher 4.7 (Gene Codes Corporation, Ann Arbor, Michigan),

MacClade (Maddison and Maddison 2000), and Mesquite (Maddison and Maddison

2010). Sequences were verified manually, and authenticity of the gene was confirmed by amino acid translation and BLAST searches in Genbank

(http://blast.ncbi.nlm.nih.gov/Blast.cgi).

To enable comparison of our results with those of previous studies, average uncorrected sequence divergence values for the Cytb gene were corrected using the

Kimura 2-parameter substitution model (Kimura 1980) in PAUP* (Phylogenetic Analysis using Parsimony, version 4.0b, Swofford 2003). This model has been used widely in studies of Cytb variation in mammals (Bradley and Baker 2001; Honeycutt et al. 1995).

We also calculated average maximum and minimum divergence values for each of the main clades in our study. To detect phylogenetic “noise,” saturation analyses for 3rd codon positions were performed as described by Griffiths (1997). To explore the effects of 3rd position substitutions in phylogenetic reconstruction, we ran maximum likelihood

(ML) analyses including all 3rd position substitutions, omitting all 3rd position substitutions, and excluding only 3rd position transitions.

! 93 Phylogenetic analyses.—Our preliminary phylogenetic analysis included all specimens (n = 21 individuals of D. phillipsii) for which we had partial Cytb sequences

(1,022 base pairs, bp). To test monophyly of D. phillipsii, we included sequences from the closely related species D. merriami (n = 2), and D. elator (n = 2) as well as one sequence from several other species of the genus Dipodomys and Heteromys irroratus

(Appendix 5.1; Alexander and Riddle 2005; Hafner et al. 2007). We selected five individuals from the Trans-Mexico Volcanic Belt (TMVB) clade and five individuals from the Mexican Plateau (MP) clade and sequenced these individuals for portions of the

12S, GHR, and IRBP genes. A partition homogeneity test in PAUP* (Swofford 2003;

1,000 replicates, P = 0.68) showed no significant heterogeneity among individual data sets, so the mitochondrial and nuclear data sets were concatenated for the remaining analyses, which were carried out with at least five representatives of each major clade.

Variable nucleotide positions were considered unordered, discrete characters with four possible states (A, C, G, T). Phylogenetic analyses were carried out under a ML framework in PAUP* or PhyMl 3.0 (Guindon and Gascuel 2003). Analyses based on

Bayesian inference (BI) were conducted using MrBayes (version 3.2-cvs; Huelsenbeck and Ronquist 2001; Ronquist and Huelsenbeck 2003). The best-fit models for ML and

BI analyses were evaluated using the Akaike Information Criterion and the programs jModeltest 0.1.1 (Akaike 1973; Guindon and Gascuel 2003; Posada 2008), MrModeltest

(version 2.2; Nylander 2004), or Modeltest (version 3.7; Posada and Buckley 2004;

Posada and Crandall 1998). The GTR+I+G model was selected for the mitochondrial sequences, and the HKY+G model was selected for the nuclear sequences. ML clade

! 94 support was assessed with 500 bootstrap (Felsenstein 1985) replicates in PAUP*, and clade support in BI analyses was evaluated using posterior probabilities.

Tree searches in the ML analyses were performed with the starting trees obtained via 100 random, stepwise additions followed by tree-bisection-reconnection (TBR) branch swapping. To test preliminary results that showed a paraphyletic D. phillipsii, we conducted an additional ML analysis forcing D. phillipsii to be monophyletic and compared this tree to the unconstrained tree using the Kishino-Hasegawa test (Kishino and Hasegawa 1989; Schmidt 2009). In the BI analyses, best-fit models were applied to each data partition with unlinked parameters and allowing rate variation. The Metropolis

Markov Chain Monte Carlo analysis consisted of two independent runs of 10 x 106 generations in which trees were sampled every 103 generations, resulting in 104 samples for each run. The first 103 trees of each run were discarded as burn-in, and a majority- rule consensus tree was constructed using the final 1.8 x 103 trees. The analysis was stopped when the average standard deviation of split frequencies approached zero, and convergence also was assessed using Tracer 1.5 (Rambaut and Drummond 2007). The combined data set was analyzed as partitioned (genes and model parameters) to allow for independent convergence on optimal values for each component (Ronquist and

Huelsenbeck 2003).

The data sets were partitioned into mitochondrial versus nuclear genes, Cytb versus 12S genes, and GHR versus IRBP genes. Whenever possible, each gene fragment was analyzed separately to evaluate potential conflict among gene trees. Nodes were considered well supported if there was > 80% bootstrap support in ML analyses or > 95% posterior probability in BI analyses.

! 95 Divergence time analysis.—We used the program *BEAST version 1.6.0

(Drummond and Rambaut 2007) to generate an estimate of the timing of divergence of the basal polytomy involving D. phillipsii and its close relatives. The *BEAST analyses included one Cytb sequence for each of the ingroup and outgroup species. A Yule tree prior was used, implicitly considering the gene tree to represent the species tree. The absence of dated fossils of D. phillipsii prevented us from using fossil-based time calibrations, so instead we used global substitution rates to calibrate our analyses

(Drummond et al. 2007). Substitution rates calculated exclusively for the genus

Dipodomys are not available in the literature, so we used three published estimates for other rodent species. The first substitution rate estimate was 4.0% per 1.0 x 106 years calculated for the split between the pocket gopher genera Pappogeomys and Cratogeomys

(DeWalt et al. 1993). Pappogeomys and Cratogeomys belong to the family Geomyidae, which is sister to the Heteromyidae. The second rate estimate we used was 4.78% per 1.0 x 106 years generated for the more distantly related house mouse Mus musculus, and the third was 5.23% per 1.0 x 106 years generated for the brown rat Rattus norvegicus

(Bininda-Emonds 2007). We excluded the frequently used rate of 2% per 1.0 x 106 years because it has been shown recently to be an underestimate of actual rates (Nabholz et al.

2009). We used a relaxed clock with a lognormal distribution allowing rate variation among sites. Chains were run for 107 generations, sampling the parameter every 103 generations. Convergence statistics were checked for effective sample sizes using Tracer version 1.5 (Rambaut and Drummond 2007). Consensus trees were generated from the resulting 104 trees using TreeAnnotator version 1.6.0 (Rambaut and Drummond 2009) after elimination of 10% as burn-in.

! 96 5.3 RESULTS

Distance and phylogenetic analyses.—We sequenced 1,022 bp of the Cytb gene,

796 bp of the 12S gene, 820 bp of GHR, and 470 bp of IRBP, yielding a total of 3,110 bp for use in the analyses. Initial ML and BI analyses using only Cytb sequences from the

21 individuals generated trees that differed only in relationships among terminal taxa

(Fig. 5.2). In these trees, D. phillipsii is divided into two well-supported clades, the MP clade, which contains all specimens currently assigned to D. p. ornatus (localities 1–6 in

Fig.5.1) and the TMVB clade, which contains all other D. phillipsii samples in this analysis (localities 7–17 in Fig. 5.1). Initial analyses grouped the MP clade of D. phillipsii with D. elator, suggesting paraphyly of D. phillipsii, but support for this group was weak, causing us to collapse the two D. phillipsii clades plus D. elator and D. merriami into a strongly supported four-way polytomy (Fig. 5.2). The Kishino-Hasegawa test showed the tree with the MP clade of D. phillipsii sister to D. elator to be a slightly better fit to the sequence data than the tree forcing D. phillipsii to be monophyletic (-lnL

= 3529.6743 and 3673.9037, respectively), but the difference was not statistically significant (P > 0.285; 1-tailed test).

The analysis of substitutional saturation showed Cytb 3rd position transitions to be saturated, so ML analyses were re-run twice, once excluding all 3rd position substitutions and again excluding only 3rd position transitions. Trees resulting from both analyses showed the same two clades of D. phillipsii (as in Fig. 5.22) but, strong bootstrap support for relationships within the D. phillipsii + D. elator + D. merriami clade was lacking.

! 97 Representatives of the D. phillipsii TMVB clade (five individuals from four localities), D. phillipsii MP clade (five individuals from three localities), and two representatives each of D.

Fig. 5.2.—Phylogram showing placement of Phillips’ kangaroo rat, D. phillipsii, within the genus Dipodomys based on a maximum likelihood analysis of Cytochrome-b sequences. Numbers before state names refer to localities mapped in Fig. 5.1. The gray dot corresponds to the divergence time of approximately 4 mya estimated using *Beast. Numbers above branches are Bayesian posterior probabilities and numbers below branches are ML bootstrap support values. Black dots represent nodes with Bayesian posterior probabilities ! 0.95 and ML bootstrap values ! 85. A Bayesian analysis of the same data yielded a tree that differed only in minor rearrangements of the terminal branches.

! 98 D. elator and D. merriami were sequenced for the other three genes (12S, GHR, and IRBP). ML and BI analyses of the 12S sequences showed basically the same tree as in Fig. 5.2, with only minor differences near the tips of the branches (tree not shown).

Separate analyses of the nuclear genes showed little or no resolution at the shallow nodes but recovered the same major clades seen in the analyses of the mitochondrial genes (Fig.

5.2). The partition homogeneity test showed no significant heterogeneity among individual data sets, so the mitochondrial and nuclear sequences were concatenated for all remaining analyses.

Bayesian analyses of the concatenated data set using the data partitions defined earlier showed the same subdivision of D. phillipsii into two well-defined clades (MP and

TMVB) and the same four-way polytomy involving the two clades of D. phillipsii, D. elator, and D. merriami (Fig. 5.3). As in the other analyses, branch suppport values were generally higher for the basal nodes and slightly lower for some of the shallower nodes in the trees.

Cytb divergence values (Table 5.1) show the two lineages of D. phillipsii (MP and

TMVB) to be approximately 9.8% genetically divergent. In contrast, average within- clade divergence was only 1.5% for the MP clade and 1.2% for the TMVB clade.

Genetic distances between the two D. phillipsii clades and D. elator ranged from 11.3% to 11.4% and distances between the two D. phillipsii clades and D. merriami ranged from

11.6% to 12.3%.

Estimates of divergence time.— Except for minor rearrangement of terminal taxa, the tree obtained in the *BEAST analysis was identical to those generated in the ML and

! 99 Fig. 5.3.—Phylogram showing placement of Phillips’ kangaroo rat, D. phillipsii, within allied species of Dipodomys based on a partitioned maximum likelihood (ML) analysis of two mitochondrial genes (Cytb and 12S) and two nuclear genes (GHR and IRBP). Numbers before state names refer to localities mapped in Fig. 5.1. Numbers above branches are Bayesian posterior probabilities and numbers below branches are ML bootstrap support values. A Bayesian analysis of the same data yielded a tree with identical topology.

! 100 Table 5.1.—Mean percent cytochrome-b sequence divergence values (Kimura 2- parameter model) and ranges (in parentheses) between and within the two major clades of Dipodomys phillipsii (Trans-Mexico Volcanic Belt, n = 14; Mexican Plateau, n = 7), D. elator (n = 2), and D. merriami (n = 2).

Dipodomys phillipsii

Trans-Mexico Mexican D. elator D. merriami Volcanic Belt Plateau

Trans-Mexico 1.2 9.8 11.4 12.3 Volcanic Belt (0.7–2.2) (9.2–10.2) (10.8–12.0) (12.0–12.8)

Mexican 1.5 11.3 11.6 Plateau (0.6–4.7) (10.5–12.0) (11.1–12.6)

D. elator - 12.7

! 101 BI analyses (Fig. 5.2). The Tracer analysis in *BEAST confirmed a high effective sample size (> 3,000) for all parameters. Estimates of divergence time (95% highest posterior density interval) for the node that marks the separation of the two D. phillipsii clades, D. elator, and D. merriami (gray dot in Fig. 5.2) were 2.86–6.46 mya when the

Pappogeomys–Cratogeomys rate calibration was used, 2.74–6.16 mya when the Mus musculus calibration was used, and 2.83–6.24 mya when the Rattus norvegicus calibration was used. The mean node age based on these three estimates ranged from

4.19 to 4.29 mya.

5.4 DISCUSSION

All phylogenetic analyses (ML, BI, and *BEAST) of the concatenated sequence data recovered a strongly supported clade composed of four genetically well- differentiated lineages: D. phillipsii (TMVB), D. phillipsii (MP), D. elator, and D. merriami (Fig. 5.3). The composition of this clade is consistent with the molecular findings of Mantooth et al. (2000) and Hafner et al. (2007), although neither of those studies included a sample of D. phillipsii from the TMVB and Hafner et al. (2007) did not include a sample of D. elator. Our trees (Figs. 5.2 and 5.3) show D. ordii and D. compactus to be sister to the above-mentioned clade, which is consistent with the findings of Mantooth et al. (2000; although that study did not include D. compactus) and does not conflict with the findings of Hafner et al. (2007) if 1 branch (with questionable support in that study: maximum parsimony bootstrap < 75%, ML bootstrap = 77%, and

BI posterior probability = 1.0) that lead to D. ordii and D. compactus as sister to a clade formed by D. panamintinus, D. heermanni, D. agilis, D. microps, and D. californicus is collapsed.

! 102 Although phyletic resolution within the phillipsii + elator + merriami clade was poor, certain of our analyses suggested (with only weak branch support) that the two D. phillipsii clades may not be sister lineages. In contrast, the smallest Cytb genetic distance between clades in Table 5.1 (9.8%) was between the two D. phillipsii clades, thus supporting their sister status, followed by D. phillipsii (MP clade) and D. elator (11.3%).

Cytb distance calculations should be viewed with caution, however, because it is well known that distance values can be misleading when substitution rates are heterogeneous among lineages (Rzhetsky and Sitnikova 1996), and our test for homogeneous substitution rates in our dataset led us to reject the assumption of clock-like behavior

(PAUP*; P = 0.0001). It also is important to note that accuracy of distance values lessen when they exceed 10% (Jin and Nei 1990), which is the case in most of our inter-clade comparisons in Table 5.1.

Although our results are ambiguous with respect to the question of D. phillipsii monophyly, this question may be moot considering that the two lineages within what is now considered D. phillipsii (MP and TMVB) approach 10% Cytb sequence divergence.

According to our estimates, divergence of these two lineages occurred approximately 4 mya, roughly contemporaneous with divergence of D. merriami and D. elator. We were unable to resolve this four-way polytomy despite using a large number of base pairs from multiple genes (both mitochondrial and nuclear) and multiple phylogenetic approaches.

We interpret our inability to resolve the branching sequence in this clade as resulting from near-simultaneous and rapid divergence of these four clades such that few, if any, molecular changes that would reveal the order in which these taxa diverged remain to be discovered today (Lanyon 1988; Spradling et al. 2004).

! 103 Even if future studies show the two clades of D. phillipsii to be sister lineages, this finding would be irrelevant to the question of whether these lineages are genetically isolated and may represent distinct species. Although we are reluctant to base species- level designations primarily on level of Cytb divergence, the level of Cytb sequence divergence between these lineages (9.8%) is not trivial—it is well above the mean level of divergence (7.3%) measured between 19 pairs of sister species of rodents reported by

Baker and Bradley (2006) in their discussion of the genetic species concept.

Biogeographical considerations.—It has been proposed that the North American mammalian biota is young by geological standards and that most species-level diversification within mammals was largely in response to glacial and interglacial changes during the Pleistocene (Findley 1969; Orr 1960; Schmidly et al. 1993).

However, new distributional, paleontological, and molecular evidence has shown that most phyletic diversification of the mammals of North American deserts took place during Miocene and Pliocene times (Hafner and Riddle 1997; Riddle 1995; Riddle et al.

2000; Vrba, 1992). Our time estimates suggest that diversification of the two lineages of

D. phillipsii also occurred prior to the major climatic fluctuations of the Pleistocene, and we contend that the Trans-Mexico Volcanic Belt, which is widely recognized as a generator of biological diversity, may have played a causal role in this divergence. The long and slow rise of this complex of mountain ranges that extends nearly from the

Pacific ocean to the Gulf of Mexico has been implicated as a causal force in phyletic diversification at multiple taxonomic levels in many organisms, frequently resulting in genetically isolated lineages north and south of the TMVB (Devitt 2006; Douglas et al.

2010; Esteva et al. 2010; Mateos et al. 2002; McCormack et al. 2008; Mulcahy and

! 104 Mendelson 2000; Sosa et al. 2009; Zink and Blackwell 1998). We believe that the two lineages of D. phillipsii are yet another example of the powerful isolating force of the

TMVB.

Researchers have proposed multiple hypotheses for diversification of D. elator,

D. merriami, and D. phillipsii, most assuming that this occurred in the Mexican Plateau and northward (Alexander and Riddle 2005; Lidicker 1960; Mantooth et al. 2000;

Morafka 1977; Savage 1960). Most of these scenarios suggest that diversification took place in response to development of extensive desert areas in North America, coincident with the uplifting of mountains in the United States and Mexico during middle and late

Tertiary (Ferrari et al. 2000; Ferrusquía 1998). One hypothesis (Lidicker 1960) suggests that D. merriami originally had a wider distribution than it does today and by peripheral isolation gave rise to D. elator in Oklahoma and Texas and to D. phillipsii in Mexico.

Another hypothesis (Mantooth et al. 2000) suggests that a D. merriami-like ancestor give rise to D. phillipsii, which originally had a wider distribution than it does today and eventually gave rise to D. elator in the southwestern United States. A third scenario (also proposed by Mantooth et al. 2000) suggests that the D. merriami-like ancestor first gave rise to D. elator, then D. elator, in turn, gave rise to D. phillipsii. Because our analysis could not resolve the sequence of branching events in this clade, our findings cannot contribute to this debate. However, our findings suggest that each of these hypotheses is missing an important phyletic event: subdivision of ancestral D. phillipsii into two distinct clades (the MP and TMVB clades), probably in response to the final uplift of the

TMVB in central Mexico.

! 105 Taxonomic conclusions.—The Mexican Plateau clade of D. phillipsii (currently recognized as D. p. ornatus) was originally described as a distinct species, D. ornatus, by

Merriam (1894). Genoways and Jones (1971) synonymized D. ornatus under D. phillipsii for lack of consistent morphological differences between the two forms. If D. ornatus and D. phillipsii are, in fact, genetically isolated species, then the evidence presented by Genoways and Jones (1971) only confirms that they are crypic, and possibly even sibling, species. Today, populations of the MP clade (D. p. ornatus) are separated from populations of the TMVB clade of D. phillipsii by > 130 km, and the likelihood that these clades come into contact, much less interbreed, is extremely low. Moreover, the

Cytb evidence suggests that these populations have been genetically isolated for approximately four million years. We believe that the sum total of the evidence suggests that D. ornatus should again be recognized as a full species distinct from D. phillipsii, and below we present a key that can be used to distinguish between these morphologically cryptic species with reasonable (77.4%) accuracy.

Dipodomys ornatus Merriam, 1894

Plateau Kangaroo Rat

Dipodomys ornatus Merriam, 1894:110. Type locality “Berriozábal, Zacatecas.”

Geographic range.—This species is known only from the Mexican Plateau, where it has been collected in the states of Aguascalientes, Durango, Guanajuato, Jalisco,

Querétaro, San Luís Potosí, and Zacatecas.

Description.—Total body length 252–302 mm; 4 toes on hind foot; dorsal coloration pale; interorbital region narrow (57.9–60.9% of mastoid breadth; Genoways

! 106 and Jones 1971); mastoid breadth of skull narrow relative to maxillary breadth (ratio between mastoid breadth and maxillary breadth ! 1.08).

Dipodomys phillipsii Gray, 1841

Phillips’ Kangaroo Rat

(Synonymy under subspecies)

Description.—Total body length 230–304 mm; four toes on hind foot; dorsal coloration variable. Populations in the Valley of Mexico (D. p. phillipsii) show dark dorsal coloration, medium body size for the species, and broad maxillary and interorbital regions relative to mastoid breadth (Genoways and Jones 1971). Populations in the south

(D. p. oaxacae) are small for the species (hind foot < 37 mm), with pale dorsal coloration and narrow maxillary and interorbital breadths relative to mastoid breadth. Populations in the Oriental Basin (D. p. perotensis) are slightly darker and larger in body size (hind foot > 37 mm) than D. p. oaxacae and have a somewhat longer cranium than D. p. phillipsii.

Geographic range.—Known from the arid, semi-desert regions of southwestern

Hidalgo, Mexico, Distrito Federal, Tlaxcala, west-central Veracruz, Puebla, and northern

Oaxaca.

D. phillipsii oaxacae Hooper, 1947

D. phillipsii oaxacae Hooper, 1947:48. Type locality “Teotitlán, 950 m, Oaxaca.”

Geographic range.—Known only from extreme southern Puebla and adjacent

northern Oaxaca.

D. phillipsii perotensis Merriam, 1894

D. perotensis Merriam, 1894:111. Type locality “Perote, Vera Cruz [Veracruz].”

! 107 Geographic range.—Known only from the Oriental Basin, west-central Veracruz

(vicinity of Perote) and adjacent parts of Puebla and Tlaxcala.

D. phillipsii phillipsii Gray, 1841

Dipodomys phillipii [sic] Gray, 1841:522. Type locality "Mexico, near Real del Monte.”

Type locality determined to be in the state of Hidalgo, “in the mountains at the

extreme north end of the Valley of Mexico, about 50 miles northeast of the City

of Mexico” by Merriam (1893:91). Species name spelled phillipii by

typographical error corrected to phillipsii by same author a few months later

(Gray 1842. American Journal of Science 42:335).

Macrocolus halticus A. Wagner, 1846:172. Type locality “Mexico.”

Distribution.—Known only from the Valley of Mexico and immediately adjacent areas of Hidalgo, Mexico, and the Distrito Federal.

KEY TO THE SPECIES AND SUBSPECIES OF D. PHILLIPSII AND D. ORNATUS

1 Hind foot length ! 37.0………………………………………………..D. p. oaxacae

Hind foot length > 37.0.. ………………………………………………………...…2

2 Middorsal pelage very dark (red reflectance < 13%) and specimen taken from vicinity of the Valley of Mexico (Hidalgo, State of Mexico, or Distrito Federal)…D. p. phillipsii

Middorsal pelage very dark to light, but if very dark (red reflectance < 13%) then specimen not taken from vicinity of the Valley of Mexico …………………………... 3

3 Ratio between mastoid breadth of skull and maxillary breadth of skull (mastoid

breadth divided by maxillary breadth) > 1.08………………………… D. p. perotensis

*Ratio between mastoid breadth and maxillary breadth ! 1.08 …….... D. ornatus

! 108

*This character is 77.4% reliable based on examination of 31 randomly

selected adult specimens of D. phillipsii perotensis (n = 17) and D. ornatus (n =

14).

5.5 LITERATURE CITED

AKAIKE, H. 1973. Information theory as an extension of the maximum likelihood principle. Pp. 267–281 in Second International Symposium on Information Theory (B. Petrov and F. Csaki, eds.) Akademiai Kiado, Budapest, Hungary.

ALEXANDER, L. F., AND B. R. RIDDLE. 2005. Phylogenetics of the New World rodent family Heteromyidae. Journal of Mammalogy 86:366–379.

BAKER, R. J., AND R. D. BRADLEY. 2006. Speciation in mammals and the genetic species concept. Journal of Mammalogy 87:643–662.

BININDA-EMONDS, O. R. P. 2007. Fast genes and slow clades: comparative rates of molecular evolution in mammals. Evolutionary Bioinformatics 3:59–85.

BRADLEY, R. D., AND R. J. BAKER. 2001. A test of the genetic species concept: cytochrome–b sequences and mammals. Journal of Mammalogy 82:969–973.

DARBRE, P. D. 2001. Basic molecular biology: essential techniques. John Wiley and Sons, New York.

DAVIS, W. B. 1944. Notes on Mexican mammals. Journal of Mammalogy 25:370–403.

DEVITT, T. J. 2006. Phylogeography of the western lyresnake (Trimorphodon biscutatus): testing aridland biogeographical hypotheses across the Neartic- Neotropical transition. Molecular Ecology 15:4387–4407.

DEWALT, T. S., P. D. SUDMAN, M. S. HAFNER, AND S. K. DAVIS. 1993. Phylogenetic relationships of pocket gophers (Cratogeomys and Pappogeomys) based on mitochondrial DNA Cytochrome b sequences. Molecular Phylogenetics and Evolution 2:193–204.

DOUGLAS, M. E., M. R. DOUGLAS, G. W. SCHUETT, D. D. BECK, AND B. K. SULLIVAN. 2010. Conservation phylogenetics of helodermatid lizards using multiple molecular markers and a supertree approach. Molecular Phylogenetics and Evolution 55:153–167.

DRUMMOND, A. J., S. Y. W. HO, N. RAWLENCE, AND A. RAMBAUT. 2007. A rough guide to Beast 1.4. Available online at

! 109 http://molecularevolution.org/molevolfiles/beast/BEAST14_MANUAL-7-6-07. Accessed 1 June 2011.

DRUMMOND, A. J., AND A. RAMBAUT. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology 7:214.

ESTEVA, M., F. A. CERVANTES, S. V. BRANT, AND J. A. COOK. 2010. Molecular phylogeny of long-tailed shrews (genus Sorex) from México and Guatemala. Zootaxa 2615: 47–65.

FELSENSTEIN, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783–791.

FERRARI, L., S. CONTICELLI, G. VAGGELLI, C. M. PETRONE, AND P. MANETTI. 2000. Late Miocene volcanism and intra-arc tectonics during the early development of the trans-Mexican volcanic belt. Tectonophysics 318:161–185.

FERRUSQUÍA, V. I. 1998. Geología de México. Pp. 3–108 in Diversidad biológica de México: orígenes y distribución (T. P. Ramammoorthy, R. Bye, A. Lot, and J. Fa., eds.), Instituto de Biología, UNAM, México, D. F.

FINDLEY, J. S. 1969. Biogeography of southwestern boreal and desert mammals. Pp. 113–128 in Contributions in mammalogy (J. K. Jones, ed.), University of Kansas, Miscellaneus Publications of the Museum of Natural History.

GENOWAYS, H. H., AND J. K. JONES, JR. 1971. Systematics of southern banner-tailed kangaroo rats of the Dipodomys phillipsi group. Journal of Mammalogy 52:265– 287.

GRAY, J. E. 1841. A new genus of Mexican Glirine Mammalia. Annals and Magazine of Natural History, ser. 1, 7:521–522.

GRIFFITHS, C. 1997. Correlation of functional domains and rates of nucleotide substitution in Cytochrome b. Molecular Phylogenetics and Evolution 7:352–365.

GUINDON, S., AND O. GASCUEL. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology: 52:696–704.

HAFNER, D. J., AND B. R. RIDDLE. 1997. Biogeography of Baja California peninsular desert mammals. Pp. 19–68 in Life among the Muses: papers in Honor of James S. Findley (T. L. Yates, W. L. Gannon, and D. E. Wilson, eds.), The Museum of Southwestern Biology, The University of New Mexico, Albuquerque.

HAFNER, J. C., J. E. LIGHT, D. J. HAFNER, M. S. HAFNER, E. REDDINGTON, D. S. ROGERS, AND B. R. RIDDLE. 2007. Basal clades and molecular systematics of heteromyid rodents. Journal of Mammalogy 88:1129–1145.

! 110 nd HALL, E. R. 1981. The mammals of North America. 2 ed. John Wiley and Sons, Inc., New York 1:1-600 + 90.

HONEYCUTT, R. L., M. A NEDBAL, R. M. ADKINS, AND L. L. JANACEK. 1995. Mammalian mitochondrial DNA evolution: a comparison of the cytochrome b and cytochrome c oxidase II genes. Journal of Molecular Evolution 40:260–272.

HOOPER, E. T. 1947. Notes on Mexican mammals. Journal of Mammalogy, 28:40–57.

HUELSENBECK, J. P., AND F. RONQUIST. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17:754–755.

IRWIN, D. M., T. D. KOCHER, AND A. C. WILSON. 1991. Evolution of the cytochrome b gene in mammals. Journal of Molecular Evolution 2:37–55.

JANSA, S. A., T. C. GIARLA, AND B. K. LIM. 2009. The phylogenetic position of the rodent genus Typhlomys and the geographic origin of Muroidea. Journal of Mammalogy 90:1083–1094.

JIN, L., AND M. NEI. 1990. Limitations of the evolutionary parsimony method of phylogenetic analysis. Molecular Biology and Evolution 7:82–102.

JONES, J. K., JR., AND H. H. GENOWAYS. 1975. Dipodomys phillipsi. Mammalian species 51:1–3.

KELT, D. A., M. S. HAFNER, AND THE AMERICAN SOCIETY OF MAMMALOGISTS' AD HOC COMMITTEE FOR GUIDELINES ON HANDLING RODENTS IN THE FIELD. 2010. Updated guidelines for protection of mammalogists and wildlife researchers from hantavirus pulmonary syndrome (HPS). Journal of Mammalogy 91:1524–1527.

KIMURA, M. 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16:111–120.

KISHINO, H., AND M. HASEGAWA. 1989. Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Hominoidea. Journal of Molecular Evolution 29:170–179.

LANYON, S. M. 1988. The stochastic mode of molecular evolution: What consequences for systematic investigation? Auk 105:565–573.

LIDICKER, W. Z., JR. 1960. The baculum of Dipodomys ornatus and its implications for superspecific groupings of kangaroo rats. Journal of Mammalogy 41:495–499.

LUISELLI, F. C. 2002. Norma Oficial Mexicana NOM-059-SEMARNAT-2001 Protección ambiental-Especies nativas de México de flora y fauna silvestres-

! 111 Categorías de riesgo y especificaciones para su inclusión, exclusión o cambio- Lista de especies en riesgo. Diario Oficial de la Federación. Miércoles 6 de marzo de 2002.

MADDISON, D. R., AND W. P. MADDISON. 2000. MacClade 4: Analysis of phylogeny and character evolution. Version 4.0. Sinauer Associates, Sunderland, Massachusetts.

MADDISON, W. P., AND D. R. MADDISON. 2010. Mesquite: a modular system for evolutionary analysis. Version 2.73. Available online at http://mesquiteproject.org. Accessed 1 June 2011.

MANTOOTH, S. J., C. JONES, AND R. D. BRADLEY. 2000. Molecular systematics of Dipodomys elator (Rodentia: Heteromyidae) and its phylogeographic implications. Journal of Mammalogy 81:885–894.

MATEOS, M., O. I. SANJUR, AND R. C. VRIJENHOEK. 2002. Historical biogeography of the livebearing fish genus Poeciliopsis (Poecilidae: Cyprinodontiformes). Evolution 56:972–984.

MCCORMACK, J. E., A. T. PETERSON, E. BONACCORSO, AND T. B. SMITH. 2008. Speciation in the highlands of Mexico: genetic and phenotypic divergence in the Mexican jay (Aphelocoma ultramarina). Molecular Ecology 17:2505–2521.

MCMAHON, J. A. 1979. North American deserts: their floral and faunal components. Pp. 21–82 in Arid-land ecosystems: structure, functioning and management (D. W. Goodall, R. A. Perry, and K. M. W. Howes, eds.), Vol. I,. Cambridge University Press, New York.

MERRIAM, C. H. 1894. Preliminary descriptions of eleven new kangaroo rats of the genera Dipodomys and Perodipus. Proceedings of the Biological Society of Washington 9:109–116.

MORAFKA, D. J. 1977. A biogeographical analysis of the Chihuahuan desert through its herpetofauna. Biogeographica IX, Dr. W. Junk B. V. Publishers, The Hague.

MULCAHY, D. G., AND J. R. MENDELSON III. 2000. Phylogeography and speciation of the morphologically variable, widespread species Bufo valliceps, based on molecular evidence from mtDNA. Molecular Phylogenetics and Evolution 17:173–189.

NABHOLZ, B., S. GLÉMIN, AND N. GALTIER. 2009. The erratic mitochondrial clock: variations of mutation rate, not population size, affect mtDNA diversity across birds and mammals. BMC Evolutionary Biology 9:54.

NEDBAL, M. A., M. W. ALLARD, AND R. L. HONEYCUTT. 1994. Molecular systematics of hystricognath rodents: Evidence from the mitochondrial 12S rRNA gene. Molecular Phylogenetics and Evolution 3:206–220.

! 112 NYLANDER, J. A. A. 2004. MrModeltest v2. Program distributed by the author. Evolutionary Biology Centre, Uppsala University.

ORR, R. T. 1960. An analysis of the Recent land mammals. Systematic Zoology 9:171– 179.

POSADA, D. 2008. jModeltest: phylogenetic model averaging. Molecular Biology and Evolution 25:1253–1256.

POSADA, D., AND T. R. BUCKLEY. 2004. Model selection and model averaging in phylogenetics: advantages of Akaike information criterion and Bayesian approaches over likelihood ratio tests. Systematic Biology 53:793–808.

POSADA, D., AND K. A. CRANDALL. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:817–818.

RAMBAUT A., AND A. J. DRUMMOND. 2007. Tracer v1.4, Available from http://beast.bio.ed.ac.uk/Tracer. Accessed 1 June 2011.

RAMBAUT A., AND A. J. DRUMMOND. 2009. TreeAnnotator v1.6.0: MCMC Output Analysis. Available at http://beast.bio.ed.ac.uk/TreeAnnotator. Accessed 1 June 2011.

RIDDLE, B. R. 1995. Molecular biogeography in the pocket mice (Perognathus and Chaetodipus) and grasshopper mice (Onychomys): the late Cenozoic development of a North American aridlands rodent guild. Journal of Mammalogy 76:283–301.

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.

RONQUIST, F., AND J. P. HUELSENBECK. 2003. MRBAYES 3. Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572–1574.

RZHETSKY, A., AND T. SITNIKOVA. 1996. When is it safe to use an oversimplified substitution model in tree-making? Molecular Biology and Evolution 13:1255– 1265.

SAIKI, R. K., G. H. GELFAND, S. STOFFEL, S. J. SHARF, R. HIGUCHI, G. T. HORN, K. B. MULLIS, AND H. A. ERLICK. 1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487–491.

SAUNDERS, C. G., AND H. C. PARKES (EDS.) 1999. Analytical molecular biology: quality and validation. Royal Society of Chemistry, Cambridge, United Kingdom.

! 113 SAVAGE, J. M. 1960. Evolution of a peninsular herpetofauna. Systematic Zoology 9:184–212.

SCHMIDLY, D. J., K. T. WILKINS, AND J. M. DERR. 1993. Biogeography. Pp. 319–356 in Biology of the Heteromyidae (H. H. Genoways, and J. H. Brown, eds.), American Society of Mammalogists.

SCHMIDT, H. A. 2009. Testing tree topologies. Pp. 381–395 in The Phylogenetic Handbook (P. Lemey, M. Salemi, and A. M. Vandamme, eds.), Cambridge University Press.

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.

SMITH, M. F., AND J. L. PATTON. 1993. The diversification of South American murid rodents: evidence from mitochodrial DNA sequence data for the akodontine tribe. Biological Journal of the Linnean Society 50:149–177.

SOSA, V., E. RUÍZ-SÁNCHEZ, AND F. C. RODRÍGUEZ-GÓMEZ. 2009. Hidden phylogeographic complexity in the Sierra Madre Oriental: the case of the Mexican tulip poppy Hunnemannia fumariifolia (Papaveraceae). Journal of Biogeography 36:18–27.

SPRADLING, T. A., S. V. BRANT, M. S. HAFNER, AND C. J. DICKERSON. 2004. DNA data support a rapid radiation of pocket gopher genera (Rodentia: Geomyidae). Journal of Mammalian Evolution 11:105–125.

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.

SWOFFORD, D. L. 2003. PAUP*: Phylogenetic analysis using parsimony (and other methods). Version 4.0b2.0a. Sinauer Associates Inc., Sunderland, Massachussetts.

VRBA, E. S. 1992. Mammals as a key to evolutionary theory. Journal of Mammalogy 73:1–28.

WILSON, D. E., AND D. A. REEDER. 2005. Mammal species of the world: a taxonomic and geographic reference. The Johns Hopkins University Press, Baltimore.

ZINK, R. M., AND R. C. BLACKWELL. 1998. Molecular systematics and biogeography of aridland gnatcatchers (genus Polioptila), and evidence supporting species status

! 114 of the California gnatcatcher (Poliptila californica). Molecular Phylogenetics and Evolution 9:26–32.

! 115 CHAPTER 6

CONCLUSIONS

A vast number of biological studies have focused on the Mexican biota, and at first glance, one’s impression might be that several of the major biological questions have been addressed. However, a careful eye will detect that many species in many parts of the country have yet to be studied, and large numbers of taxa that have been studied previously should be reexamined using modern techniques and analyses. In studies of the

Mexican biota, one cannot help but assume that the dramatic climate cycles of the

Pleistocene epoch and the prominence of the Trans-Mexico Volcanic Belt (TMVB) played major roles in the origin and diversification of Mexican species. However, only recently have studies been designed to clarify and test the role of Pleistocene climate cycles and the TMVB in generating the biological diversity we see today.

The four nearly codistributed rodent species studied in this dissertation offer excellent opportunities to investigate individual and shared patterns of evolutionary diversification. In each case, I have generated a phylogenetic hypothesis for the taxon, recommended appropriate taxonomic changes, as necessary, and placed the phylogeny in a temporal framework to identify events that may have generated the phylogenetic pattern.

Two of the study taxa, Nelson’s woodrat Neotoma nelsoni (Chapter 2) and the

Perote ground squirrel Xerospermophilus perotensis (Chapter 3), are both endemics to the

Oriental Basin in east-central Mexico. Historically, both taxa have been treated as valid species, based mainly on a few qualitative morphological characters and the fact that they are geographically isolated from congeneric populations. Previous studies have

! 116 postulated that the sister species of N. nelsoni and X. perotensis occur in the Mexican

Plateau and that current populations of these taxa are remnants of once more widespread desert-adapted forms whose distribution was fragmented by Pleistocene climatic oscillations. It is thought that these taxa evolved in isolation in the Oriental Basin following extinction of intermediate populations.

My analyses confirmed that the closest relatives of N. nelsoni and X. perotensis occur in the Mexican Plateau (N. leucodon and X. spilosoma, respectively), and my findings also confirmed that N. nelsoni and X. perotensis are genetically well- differentiated from their sister taxa to the north. However, the Cytb genetic distances

(3.3% between N. nelsoni and N. leucodon and 3.6% between X. perotensis and X. spilosoma; Table 6.1) are not large, and this in combination with low levels of morphological differentiation between the Oriental Basin and Mexican Plateau populations of these taxa suggest that they should be recognized only at the subspecific level as N. leucodon nelsoni and X. spilosoma perotensis.

Molecular estimates of divergence times suggested that N. l. nelsoni and X. s. perotensis diverged from their sister taxa to the north during early Pleistocene times ;

Table 6.1). Thus, climatic fluctuations during the Pleistocene may have played a role in the early diversification of these lineages, causing repeated fragmentation of their geographic distributions followed by extintion of geographically intermediate populations. Because divergence of N. l. nelsoni and X. s. perotensis from their relatives to the north following the Miocene/Pliocene uplift of the TMVB, it appears that the

TMVB played only a minor, if any, role in the diversification of these lineages. In the

! 117 Xerospermophilus study (Chapter 3), a well-differentiated lineage of X. spilosoma was discovered in the Bolsón de Mapimí in Durango, México.

The rock mouse Peromyscus difficilis (Chapter 4) is another Mexican endemic found in the dry foothills of the Sierra Madre Oriental, the Mexican Plateau, the low dry hills of the TMVB, and some apparently disjunct populations in pine forests of central

México as far south as Oaxaca. Historically, P. difficilis has been divided into five subspecies based primarily on distribution, but also on a few qualitative morphological characters. The results of my analyses divided this species into two well-supported clades (ca. 6.7% Cytb divergence; Table 6.1)), a northern clade including the subspecies

P. d. difficilis and P. d. petricola, and a southern clade containing the subspecies amplus, felipensis, and saxicola. Molecular-based estimates of divergence times suggested that separation of these clades occurred in the Pleistocene, again (as in Chapters 2 and 3) suggesting that glacial and interglacial cycles of the Pleistocene may have influenced genetic differentiation in the common ancestors of these lineages. The southernmost subspecies of P. difficilis (P. d. felipensis) was originally described as P. felipensis.

Several lines of evidence, including phylogenetic analyses of mitochondrial and nuclear genes, estimates of divergence times, disjunct geographic distributions, and morphological differences supported my decision to return the southern clade of P. difficilis to full species status as P. felipensis.

My study of the endangered Phillips’ kangaroo rat, Dipodomys phillipsii (Chapter

5), revealed a biogeographic pattern different for that seen in Chapters 2–4. D. phillipsii occurs on the Mexican Plateau and in the arid lowlands of the TMVB. Past studies divided this taxon into several subspecies based mainly on geographic distribution, but

! 118 also supported by several qualitative morphological characters. In my analysis, D. phillipsii was divided into two well-supported clades (ca. 9.8% Cytb divergence; Table

6.1), one distributed on the Mexican Plateau (formerly recognized as D. ornatus), and a southern clade in the TMVB. Several lines of evidence, including phylogenetic analyses of mitochondrial and nuclear genes, estimates of divergence times, disjunct geographic distributions, and morphological differences supported my decision to return the Mexican

Plateau clade of D. phillipsii to full species status as D. ornatus. My study showed that

D. phillipsii, D. ornatus, D. elator, and D. merriami form a well-supported clade of kangaroo rats, but I was unable to resolve relationships among these four species. My inability to resolve this polytomy using a large number of base pairs from a combination of mitochondrial and nuclear genes analyzed in multiple ways suggests that diversification among these four species may have occurred over a relatively short period of time.

My molecular-based analyses of divergence times suggests that D. phillipsii, D. ornatus, D. elator, and D. merriami diverged in mid-Pliocene times; Table 6.1), probably in or near the Mexican Plateau. Unlike the Pleistocene divergence dates reported in

Chapters 2–4, this Pliocence divergence suggests that the morphotectonic processes that gave rise to the Trans-Mexico Volcanic Belt may have influenced early diversification in

Mexican species of Dipodomys.

Many studies of the mammals of Mexico have focused on the rich fauna of the tropical and semitropical regions of southern Mexico. To some extent, the desert environments of Mexico have been ignored, especially by researchers using modern systematic techniques and methods of analysis. By means of this dissertation, I wish to

! 119 reaffirm and highlight the importance of Mexico’s deserts, especially the southern extension of the deserts into central Mexico, as natural laboratories and important centers of evolutionary diversification. I invite my colleagues to test the phylogenetic and biogeographic hypotheses contained in this dissertation with their own studies of the flora and fauna of central Mexico.

! 120

Table 6.1.—Overview of taxonomic conclusion, divergence values between clades of interest, and mean estimates of divergence time for clades of interest (in million years ago).

Monophyly Divergence values Mean estimated

between clades (%) divergence time

(mya)

Neotoma nelsoni Yes 3.2 2.10 Pleistocene

Xerospermophilus Yes 3.6 0.75 Pleistocene perotensis

Peromyscus Yes 6.7 0.80 Pleistocene difficilis

Dipodomys Yes 9.8 4.20 Pliocene phillipsii

! 121 APPENDIX 3.1

Specimens used in the analysis of X. s. perotensis relationships listed alphabetically by taxon, locality, geographic coordinates, elevation, catalogue number, sequenced genes, and GenBank numbers. Specimens are housed in the following museums: Colección Nacional de Mamíferos, Instituto de Biología, Universidad

Nacional Autónoma de México (CNMA), Cornell University DNA collection (CU; samples from CU are followed by the collector’s field number in parentheses), Los

Angeles County Museum of Natural History (LACM), Louisiana State University

Museum of Natural Science (LSUMZ), Museo de Zoología, Facultad de Ciencias,

Universidad Nacional Autónoma de México (MZFC); New Mexico Museum of Natural

History (NMMNH), and University of New Mexico Museum of Southwestern Biology

(MSB). Numbers in parentheses before localities indicate localities mapped in Fig. 3.1.

Callospermophilus lateralis

Collection locality not available. LACM 85487, 12S = AY227530, IRBP = AY227586.

Cynomys gunnisoni

United States: (1) Arizona, Apache Co., Petrified Forest National Park, 34.909, -109.806,

1,641 m, S 75 (73), Cytb = AF157923; CU 82 (WA1), Cytb = AF157930.

Cynomys leucurus

United States: (2) Utah, Uintah Co., 8 km E Jensen on Highway 40, CU 1 (EY 1138),

40.369, -109.240, 1,689 m, Cytb = AF157838; 12S = AY227528, IRBP = AY227584.

Utah, Uintah Co., 11 km NW Bonanza on Hwy 45, CU 2 (EY1137), 40.094, -109.269,

1,572 m, Cytb = AF157879.

! 122 Cynomys ludovicianus

United States: Nebraska, Omaha Zoo CU 38 (1A), Cytb = AF157890; CU 41 (4A), Cytb

= AF157892; Collection locality not available, CU 1 (EY 1138), IRBP = AY227584.

Cynomys mexicanus

Mexico: (3) Nuevo León, Ejido El Tokio, 18 km E highway 57 on highway 58, 24.6, -

100.2, 1,891 m, CU 101 (EY 1180), Cytb = AF157841; CU 102 (EY 1181), Cytb =

AF157842.

Cynomys parvidens

United States: (4) Utah, Kane Co., Bryce Canyon National Park, 1 km from Visitor

Center, 37.58, -112.18, 2,466 m, CU 74 (BCU 1), Cytb = AF157922; Utah, Kane Co.,

Bryce Canyon National Park, 20 m from boundary, 37.58, -112.18, 2,466 m, CU 81

(BC1), Cytb = AF157929.

Ictidomys mexicanus mexicanus

Mexico: (5) México, Parque Nacional Zoquiapan, 15 km SW Rio Frío, 19.3, -98.6, 2,980 m, CU 108 (EY 1210), Cytb = AF157848.

Ictidomys parvidens

United States: (6) New Mexico, Chaves Co., Eastern NM University, Roswell, West

Wells & University Street, 33.39, -104.52, 1,097 m, MSB 135244, Cytb = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank;

Mexico: (7) Nuevo León, 3 km NE Apodaca, ca. 23 km NE Monterrey, 25.80, -100.16,

408 m, CU 111 (EY 1197), Cytb = AF157852; CU 112 (MVA 105), Cytb = AF157853.

Ictidomys tridecemlineatus

! 123 United States: (8) Kansas, Finney Co., Garden City, 37.9, -100.8, 866 m, CU 13 (EY

1147), Cytb = AF157870; CU 14 (EY 1148), Cytb = AF157877; Collection locality not available (from personal collection of R. L. Honeycutt), H 2147, 12S = U67290, IRBP =

AF287278; (9) South Dakota, Union Co., 1 mi. N, 1.5 mi. W Junction City, 42.800, -

96.815, 377 m, LSUMZ 10692, Cytb = b = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank; South Dakota, Clay Co., 3.5 mi. N, 3.5 mi. E

Vermillion, 42.829, -96.857, 376 m, LSUMZ 10728, Cytb = b = sequences submitted to

GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank.

Xerospermophilus mohavensis

United States: (10) California, San Bernardino Co., 9 mi NNE Johannesburg, 35.473, -

117.589, 1,075 m, MSB 40496, Cytb = b = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank; MSB 40503, Cytb = b = sequences submitted to

GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank.

Xerospermophilus perotensis

Mexico: (11) Puebla, Tepayahualco, 19.490, -97.489, 2,336 m, CNMA b = sequences submitted to GenBank (EY 1176), Cytb = AF157948, 12S = b = sequences submitted to

GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank; CNMA b = sequences submitted to GenBank (EY 1179) Cytb = AF157840,

12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to

! 124 GenBank, IRBP = b = sequences submitted to GenBank; Veracruz: Municipality of

Perote, 5 km W Perote, 19.587, -97.330, 2,400 m MZFC XXXX (JAFF2248), Cytb = b = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank, GHR = b

= sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank;

Veracruz, Municipality of Perote, 3 km S El Frijol Colorado, 19.572, 97.383, 2,435 m

MZFC XXXX (JAFF2154), Cytb = b = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank.

Xerospermophilus spilosoma cabrerai

Mexico: (12) San Luis Potosi, 10 mi. S Villa de Ramos, 22.666, -101.953, 2,200 m,

NMMNH 3651, Cytb = b = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank.

Xerospermophilus spilosoma marginatus

United States: (13) New Mexico, Bernalillo Co., 8 km W Albuquerque, 35.084, -106.738,

1,631 m, LSUMZ 01, Cytb = b = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank; LSUMZ 05, Cytb = b = sequences submitted to

GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank. (14) Kansas, Finney Co., 13 km S, 2 km E Holcomb, 37.872, -100.964, 893 m, CU 3 (EY1142), Cytb = AF157885;

CU 6 (EY1146), Cytb = AF157911.

Xerospermophilus spilosoma pallescens

! 125 Mexico: (15) Durango, 4 km E Ceballos, 26.523, -104.089, 1,117 m, CU 105 (EY 1195),

Cytb = AF157845; Durango, Ejido La Flor, 20 km E Ceballos, 26.524, -103.929, 1,146 m, CU 106 (EY 1193), Cytb = AF157846.

Xerospermophilus tereticaudus neglectus

United States: (16) Arizona, Pima Co., Tucson, 32.221, -110.926, 917 m, MSB 86022,

Cytb = b = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank,

GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to

GenBank; MSB 92638, Cytb = b = sequences submitted to GenBank, 12S = b = sequences submitted to GenBank, GHR = b = sequences submitted to GenBank, IRBP = b = sequences submitted to GenBank; Arizona, Pima Co., 18 km W Tucson, Ryan Field,

32.223, -111.116, 735 m, CU 91 (EY 1169), Cytb = AF157940; CU 92 (EY 1167), Cytb,

= AF157941.

Poliocitellus franklinii

United States: Nebraska, Omaha Zoo CU 42 (1A), Cytb = AF157893; CU 43 (2A), Cytb

= AF157894.

Urocitellus townsendii idahoensis

Collection locality not available, CU 86 (EY 1062), Cytb = AF157949.

Urocitellus townsendii

Collection locality not available, CU 87 (EY 1064), Cytb = AF157938.

! 126 APPENDIX 4.1

Specimens examined in the molecular analysis of the rock mouse Peromyscus difficilis listed by taxon, state, collecting locality, geographic coordinates, elevation, catalogue number, and GenBank numbers listed in the following order: Cytb, 12S, GHR, and IRBP. A dash (–) indicates that a specimen was not sequenced for that gene.

Mammal collections where voucher specimens are housed are CWK = C. William

Kilpatrick; GK=Ira I. Greenbaum; TK=K. Nutt and Texas Tech University Tissue

Number. Carnegie Museum of Natural History (CM); Colección Regional Durango,

Centro Interdisciplinario de Investigacion para el Desarrollo Integral Regional-Durango,

Instituto Politécnico Nacional (CDR); Louisiana State University Museum of Natural

Science (LSUMZ); Museum of Southwestern Biology, University of New Mexico

(MSB); Texas A&M University, Texas Cooperative Wildlife Collection (TCWC); Texas

Tech University (TTU); Universidad Nacional Autónoma de México, Instituto de

Biología (CNMA); Universidad Nacional Autónoma de México, Museo de Zoología

“Alfonso L. Herrera”(MZFC); University of Vermont, Zadock Thompson Natural

History Collections (ZTNH);. Numbers in parentheses indicate localities mapped in Fig.

4.1. All newly generated sequences were submitted to genbank.

P. d. amplus

PUEBLA: (13) 8 mi. SE Chignahuapan, 19.753, -97.947, 8,808 ft., (ZTNHC: CWK

2770) AY376414 Cytb only; (12) 3 km S Ciudad Serdan, crossroad between Ciudad

Serdan-Esperanza, towards Santa Catarina, 18.932, -97.421, 2,536 m, CNMA 44007

Cytb, –, GHR, IRBP; (12) 1 km S Coyotepec, 19.009, -97.555, 2,435 m, CNMA 44001

Cytb, –, GHR, IRBP (12) 2 km W Guadalupe Victoria, 19.280, -97.378, 2,406 m., CNMA

! 127 43992 Cytb, 12S, GHR, IRBP; (11) 1.5 km S Oriental, 19.352, -97.635, 2,360 m, CNMA

43966 Cytb only. TLAXCALA: (13) 18 km N, 9 km E Apizaco, 19.58, -98.051, 9,142 ft., CM55804, AY387488 Cytb only; (12) 2.5 km NW El Carmen Tequexquitla, 19.35, -

97.665, 2,378 m, CNMA 43957 Cytb, 12S, GHR, IRBP; CNMA 43960 Cytb, 12S, GHR,

IRBP; (13) Barranca Huehuetitla, 2 km NE San Ambrosio Texantla, 19.316, -98.25,

2,272 m, CNMA 44269 Cytb only; (13) Mt. Malinche, 19.265, -98.022, 10,709 ft.,

TCWC: GK 3904) AY376415 Cytb only; (13) 2 km NE Tepetitla, 19.277, -98.365, 7,293 ft., TTU 82690, AY376416 Cytb only. VERACRUZ: (10) 3 km S El Frijol Colorado,

19.572, -97.383, 2,435 m, CNMA 43978 Cytb, 12S, GHR, IRBP.

P. d. difficilis

AGUASCALIENTES: (7) 6 mi. W Rincon de Romos, 22.229, -102.413, 6,852 ft.,

(TCWC GK:4129) AY376418 Cytb only; DURANGO: (5) 50 km W Las Herreras,

25.185, -106.036, 8,526 ft., CRD 1143, AY376417 Cytb only; ZACATECAS: (6) 12.4 mi. NW, 16.2 mi. NE (by road) Sombrerete, 23.9, -103.5, 2,268 m, MSB 54457 Cytb,

12S, GHR, IRBP; (6) 13.2 mi. NW, 5.4 mi. NE (by road) Sombrerete, 23.9, -103.5, 2,214 m, MSB 55604 Cytb, 12S, –, IRBP; (6) 12.4 mi. NW, 6.2 mi. NE Sombrerete, 23.9, -

103.5, 2,010 m, MSB 55616 Cytb, 12S, –, –; MSB 57701 Cytb, 12S, –, –.

P. d. felipensis

MEXICO: (15) Highway Ocuilan de Artega-Cuernavaca, km 14, 18.978, -99.416,

7,711 ft., MZFC 5689 Cytb, –, GHR, IRBP; (14) National Park Izta-Popo-Zoquiapan, 7.3 km SE Amecameca, 19.1014, -98.6955, 2,848 m, CNMA 45643, Cytb only; CNMA

45644 Cytb, 12S, GHR, IRBP.

P. d. petricola

! 128 COAHUILA: (3) 10 mi. E San Antonio de las Alazanas, 25.266, -100.766, 2,586 m, MSB 48201 Cytb only; (3) 13.9 mi. W San Antonio de las Alazanas, 25.266, -

100.766, 2,028 m, MSB 48177 Cytb, 12S, GHR, IRBP; NUEVO LEON: (4), Cerro

Potosi, Municipio Galeana, 24.85, -100.316, 1,655 m, CNMA 44276, Cyb, –, GHR,

IRBP; CNMA 44277 –, 12S, GHR, IRBP; M8647 Cytb, –, –, IRBP; M8667 –, 12S, GHR,

P. d. saxicola

HIDALGO: (8) 5.4 mi. SE, 3.2 mi. S Ixmiquilpan, 20.391, -99.316, 6,431 ft.,

(TCWC GK2642) AF155394 Cytb only; (9) 1.8 mi. E Jonacapa, 20.436, -99.506, 7,480 ft., (TCWC GK3076) AY376419 Cytb only. MEXICO: (11) Cerro Gordo, Santiago

Tolman, 19.743, 98.628, 2,500 m, CNMA 43035 Cytb, 12S, GHR, IRBP.

P. nasutus

NEW MEXICO: (1) Socorro Co., Sevilleta, Sepultura Canyon, 34.304, -106.613,

6,470 ft.; MSB 63666 Cytb, 12S, –, –; Cibola Co., 11 mi. S, 14 mi. W San Rafael, 34.963,

-108.128, 6,450 ft.; MSB 54841, Cytb, 12S, GHR, IRBP; MSB 54819, Cytb, 12S, GHR,

IRBP.

P. nasutus griseus

NEW MEXICO: (1) Lincoln Co., 4 mi. S Carrizozo, 33.583, -105.876, 5,965 ft.,

TTU 78401, AY155399 Cytb only.

P. nasutus nasutus

TEXAS: (2) Jeff Davis Co., Mt. Livermore Preserve, 30.750, -104.193 5,722 ft.,

TTU 78316, AY376426 Cytb only.

! 129 Neotoma nelsoni

VERACRUZ: (10) 3 km S El Frijol Colorado, 19.572, -97.383, 2,435 m, LSUMZ

36663 Cytb, 12S, GHR, IRBP.

Peromyscus attwateri

OKLAHOMA: McIntosh Co., 3.1 mi. E Dustin, 32.270, -95.975, 821 ft., TTU

55688, AF155384 Cytb only.

Peromyscus beatae

GUERRERO: Carrizal del Bravo, 17.266, -99.733, 3,625 ft., MZFC 9364 Cytb, –,

GHR, IRBP.

Peromyscus leucopus

CONNECTICUT: (T-175) X99463 12S.

Peromyscus levipes

OAXACA: San Martín Caballero, Distrito de Teotitlán, 17.057, -96.734, 5,069 ft.,

MZFC 8726, –, –, GHR, IRBP.

Peromyscus truei truei

ARIZONA: Apache Co., 35.580, -109.575, 6,832 ft., TTU 104427, AY376433

Cytb only.

! 130 APPENDIX 4.2

Specimens included in the *BEAST analysis of divergence times (Drummond and

Rambaut 2007). Specimens are arranged by taxon, state, collecting locality, geographic coordinates, elevation, catalogue number, and GenBank number for Cytb sequence.

Specimens were obtained from the Angelo State Natural History Collection (ASNHC),

Brigham Young University (BYU), the Royal Ontario Museum (ROM), and other museum collections listed in Appendix 4.1. Specimens from Appendix 4.1 used in the

*BEAST analysis include CNMA 44001, CNMA 45643, MSB 48177, MSB 54457, MSB

63666, TCWC GK3076, and TTU 104423.

Habromys ixtlani

OAXACA: Llano de la Flores, km 132 on highway from Tuxtepec–Oaxaca, 17.09,

-96.70, 5,470 ft., CNMA 29849, EF989941.

Isthmomys pirrensis

PANAMA: Darien Province, summit of Cerro Pirre, 7.933, -77.716, 396 ft., ROM

116308, EF989945.

Megadontomys thomasi

OAXACA: 11 km SW La Esperanza, 17.561, -96444, 8,147 ft., CNMA 29186,

EF989948.

Neotomodon alstoni

DISTRITO FEDERAL: 3 km S Parres, 19.137, -99.159, 9,260 ft., ASNHC 1595,

EF989950.

! 131 Onychomys arenicola

TEXAS: Presidio Co., Hip O Ranch, 5 mi. W Marfa, 30.308, -104.10, 4,805 ft.,

ROM 114904, EF989954.

Onychomys leucogaster

TEXAS: Cameron Co., 11 mi. N Port Isabel, 26.2, -97.2, 50 ft., ASNHC 4348,

EF989958.

Osgoodomys banderanus

MICHOACAN: 8 km N La Mira, 18.107, -102.328, 322 ft., ASNHC 2664,

EF989956.

Peromyscus attwateri

OKLAHOMA: McIntosh Co., 3.1 mi. E Dustin, 32.270, -95.975, 821 ft., TTU

55688, AF155384.

Peromyscus crinitus

UTAH: Uintah County, Bitter Creek Canyon, 41.1, -111.9, BYU 16629, EF989973.

Peromyscus eremicus

SONORA: 22 km S (by road) Hermosillo, 28.889, -110.963, 755 ft., BYU 17952,

EF989975.

Peromyscus levipes

CHIAPAS: Cerro Tzontehuiz, 16.7, -92.6, 7,513 ft., ROM 97624, EF989981.

Peromyscus maniculatus

CANADA: Ontario: Kwataboahegan, Kwataboahegan River, 51.15, -80.50, 20 m,

ROM 98941, EF989983.

! 132 Peromyscus melanophrys

Stock animal from Peromyscus Genetic Stock Center, University of South Carolina,

USC-PGSC XZ 1073, EF989989.

Peromyscus melanotis

Stock animal from Peromyscus Genetic Stock Center, University of South Carolina,

USC-PGSC 25, EF989990.

Reithrodontomys_sumichrasti

GUATEMALA: Huehuetenango: 10 km SW Santa Eulalia, 15.673, -91.528, 8,663 ft., ROM 98383, EF990023.

! 133 APPENDIX 5.1

Specimens examined.—Species name, collection locality, geographic coordinates, elevation, and catalogue number are listed for specimens used in this analysis of

Dipodomys phillipsii. Specimens used to develop the morphological key are indicated by

“(m)” following the catalogue number. GenBank numbers are listed for each specimen used in the molecular analyses. Mammal collections housing voucher or tissue specimens are ASNHC = Angelo State Natural History Collection, Angelo State University; CNMA

= Colección Nacional de Mamíferos, Instituto de Biología, Universidad Nacional

Autónoma de México; ENCB = Escuela Nacional de Ciencias Biológicas, Instituto

Politécnico Nacional; LSUMZ = Louisiana State University, Museum of Natural Science;

M = Louisiana State University, Museum of Natural Science Mammal Tissue Collection;

LVT = University of Nevada, Las Vegas; MLZ = Moore Laboratory of Zoology,

Occidental College; MVZ = Museum of Vertebrate Zoology, University of California,

Berkeley; MWSU = Midwestern State University; NMMNH = New Mexico Museum of

Natural History; TTU = Museum of Texas Tech University; UAM-I = Universidad

Autónoma Metropolitana-Iztapalapa; UATX = Universidad Autónoma de Tlaxcala; and

USNM = United States National Museum of Natural History. Numbers in parentheses refer to localities mapped in Fig. 5.1.

D. p. oaxacae

Puebla: (15) 4 km SSW San José Axusco, 18.022, -97.232, 1,031 m, CNMA 41881, Cytb

= JN183909, 12S = JN208377, GHR = JN661650; CNMA 41882, Cytb = JN183906, 12S

= JN208375, GHR = JN661651; IRBP = JN661671; CNMA 41883, Cytb = JN183907,

12S =JN661642, GHR = JN661652, IRBP = JN661672; CNMA 41884, Cytb =

! 134 JN183908, 12S = JN208376, GHR = JN661653; (14) 2.7 km SE San José Buenavista,

18.642, -97.566, UAMI 17257, Cytb = JN183910.

D. p. ornatus

Aguascalientes: (4) 8.8 km N, Las Fraguas, 22.233, -102.083, 2,150 m, IPN 36320, Cytb

= JN183912; Durango: (1) 5.8 km N, 2.1 km E Vicente Guerrero, 23.783, -103.962,

1,937 m, CNMA 39681, Cytb = JN183913, 12S = JN208380, GHR = JN661654, IRBP =

JN661676; TTU 75585, IRBP = GU955164; (2) 2.2 km S, 2.5 km E Vicente Guerrero,

23.711, -103.957, 1930 m, CNMA 39683, Cytb = JN183914, 12S = JN208382, GHR =

JN661655, IRBP = JN661677; Durango, 24.0, -104.6 , USNM 94622 (m) and 94624 (m);

Jalisco: (5) 4 km W Guadalupe Victoria, 21.699, –101.617, 2,185 m, CNMA 43385, Cytb

= JN183915, 12S = JN208384, GHR = JN661656, IRBP = JN661678; Lagos [de

Moreno], 21.4, -101.9, USNM 78952 (m); Querétaro: Tequisquiapan, 20.5, -99.9, USNM

78430 (m) and 78431 (m); San Luis Potosi: 1 km N Arenal [de Morelos], 22.42, -101.27,

LSUMZ 5178 (m); Bledos, 21.8, -101.1, LSUMZ 5177 (m); About 1 mi. W Bledos,

21.84, -101.13, LSUMZ 4286 (m); (6) Las Cabras, 4.6 mi. NW Bledos, 21.800, -100.933,

1,820 m, LVT 2056, Cytb = AY926376; Zacatecas: (3) 2 mi. E San Jeronimo, 22.640, -

97.231, 2,350 m, CNMA 42049, Cytb = JN183917, 12S = JN208385, GHR = JN661657,

IRBP = JN661679; CNMA 42050, 12S = JN208385; CNMA 42121, Cytb = JN183918,

12S = JN208386, GHR = JN661658, IRBP = JN661680; Berriozábal, 22.5, -102.3,

USNM 79506 (m); Hacienda San Juan Capistrano, 22.6, -104.1, USNM 90804 (m); [El]

Plateado [de Joaquín Amaro], 21.9, -103.1, USNM 90808 (m); Valparaíso, 22.8, -103.6,

USNM 91933 (m) and 91945 (m); Zacatecas, 22.8, -102.6, USNM 120167 (m).

D. p. perotensis

! 135 Puebla: (10) 4.5 km S, 9.5 km San José Alchichica, 19.766, -97.316, 2,350 m, IPN

42239, Cytb = JN183899; (11) 2 km W Guadalupe Victoria, 19.280, -97.378, 2,406 m,

CNMA 43986, Cytb = JN183904, 12S = JN208381, GHR = JN661659, IRBP =

JN661670; CNMA 43987–43988 (m); (12) 3.1 km SW Veladero, 20.591, -97.600, 2,340 m, LSUMZ 36253 (m), Cytb = JN183898; (16) 11 km (by road) SW Alchichica,19.766, -

97.316, 2,449 m, LSUMZ 36244 (m), Cytb = JN183900; (13) 6.7 km E Techachalco,

19.390, -97.408, 2,500 m; UAMI 17258, Cytb = JN183901; Chalchicomula (= Ciudad

Serdán), 19.0, -97.4, USNM 53323–53328 (m) and 53333 (m); Tlaxcala: Huamantla,

19.3, -97.9, USNM 53635 (m) and 53637 (m); (8) 2.5 km NW El Carmen Tequexquitla,

19.350, -97.665, 2,378 m, CNMA 44321, Cytb = JN183902, 12S = JN208378, GHR =

JN661660, IRBP = JN661674; (17) 6 km NE Cuapiaxtla, 19.300, -97.766, 2,425 m, UAT

M0366, Cytb = JN183903, 12S = JN208379, GHR = JN661661, IRBP = JN661673;

Veracruz: (9) 3 km S El Frijol Colorado, 19.572, -97.383, 2,435 m, CNMA 43972, Cytb

= JN183905, IRBP = JN661675; CNMA 43974 (m); Perote, 19.6, -97.2, USNM 54281–

54283 (m).

D. p. phillipsii

Mexico: (7) 5 km SE Nopaltepec, 19.753, -98.670, 2,400 m, UAMI 2779, Cytb =

JN183911.

D. agilis

Baja California: 6 km S, 17 km E, Valle de la Trinidad, 31.754, -116.364, 905 m, MVZ

153957, Cytb = U65303.

! 136 D. californicus

California: Tehama County, 2.5 miles S, 0.2 miles E Paynes Creek, 40.306, -121.904,

1,960 feet, MLZ 2061, Cytb = AY926368.

D. compactus

Texas: Cameron County, 4.5 miles N, 3.6 miles E Port Isabel, 26.13, _97.16, ASNHC

4327, Cytb = AY926379.

D. deserti

Nevada: Clark County, Corn Creek Desert Wildlife Refuge, 39.55, -119.46, NMMNH

5374, Cytb = AY926381.

D. elator

Texas: Cottle County, 1.6 km N, 1.8 km E junction FM 1033 and FM 104, 34.250, -

100.050, 500 m, TTU 45633, Cytb = AF172834, JN183919, 12S = JN208373, GHR =

JN661664, IRBP = JN661688; Wichita County, 8.6 miles N Iowa Park; 34.056, -98.746,

1117 feet, MWSU17542, Cytb = JN661645, 12S = JN208374, GHR = JN661665, IRBP

= JN661687.

D. heermanni

California: San Luis Obispo County, 15.0 miles S, 8.2 miles E Simmler, 33.134, -

119.839, 2,300 feet, MLZ 1852, Cytb = AY926369.

D. merriami

Arizona: Maricopa County, 11.2 km N Gila Bend, 33.666, -111.866, 224 m, TTU 41781,

Cytb = AF173502; Texas: Brewster County, Elephant Mountain Wildlife Management

Area, 29.66, -103.35, 3562 feet, TTU 97980, IRBP = GU985162; Presidio County, Las

Palomas Wildlife Management Area, 26.40, -97.80, 33 feet, TTU 75675, Cytb =

! 137 AF172837. Coahuila: 6 mi. E Parras, 24.86, -100.18, 2,118 m, M-8688, Cytb =

JN661644, 12S = JN661641, GHR = JN661662, IRBP = JN661686. Durango: 3 mi. N

Lazaro Cardenas, 25.39, -103.98, 1,500 m, M-8708, Cytb = JN661643, 12S = JN661640,

GHR = JN661663, IRBP = JN661685.

D. microps

California: Inyo County, 6.0 miles N, 0.5 miles W Bishop, 37.450, -118.404, 4,200 feet,

MLZ 1765, Cytb = AY926385.

D. nelsoni

Coahuila: 5 km S, 16 km W General Cepeda, 25.330, -101.631, 5,599 feet, NMMNH

4703, Cytb = AY926364; Durango: 7 mi. NNW La Zarca, 25.853, -104.877, 6292 feet,

NMMNH 2472, IRBP = GQ480799. Durango: 3 mi. N Lazaro Cardenas, 25.39, -103.98,

1,500 m, M-8709, Cytb = JN661648, 12S = JN661638, GHR = JN661667, IRBP =

JN661681; M-8710 Cytb = JN661649, 12S = JN661639, GHR = JN661666, IRBP =

JN661682.

D. ordii

New Mexico: Grant County, 2.6 miles N, 1.8 miles E Redrock, 32.724, -108.707, 1,241 m, NMMNH 4377, Cytb = AF173501. Luna County, 4.0 mi. S, 9.5 mi. W Deming,

33.760, -108.780, 6,198 feet, MVZ 150772, Cytb = JN661646, 12S = JN661636, GHR =

JN661669, IRBP = JN661684; MVZ 150775, Cytb = JN661647, 12S = JN661637, GHR

= JN661668, IRBP = JN661683.

D. panamintinus

California: San Bernardino County, 8.9 miles N, 1.1 miles E Red Mountain, 35.486, -

117.595, 3,150 feet, MLZ 1879, Cytb = AY926384.

! 138 D. spectabilis

New Mexico: Hidalgo County, 6 mi. SE Portal (Cochise County, Arizona), 31.859, -

109.061, NMMNH 4399, Cytb = AF173503; Lincoln County, 4 mi. N, 3 mi. W, 33.698, -

105.927, 5285 feet, TTU 38443, IRBP = GU985165.

Heteromys irroratus

Puebla: 6 km N Tilapa, 18.648, -98.553, 1,300 m, LSUMZ 36295, Cytb = GU647037.

! 139 APPENDIX 5.2

List of primers used for the amplification of nuclear (GHR and IRBP) and mitochondrial (Cytb and 12S) genes in the Phillips’ kangaroo rat (Dipodomys phillipsii).

Primer Gene Sequence Reference

MVZ-05 Cytb CGAAGCTTGATATGAAAAACCATCGTTG Irwin et al. 1991

H15915 Cytb AACTGCAGTCATCTCCGGTTTACAAGAC Irwin et al. 1991

MVZ-04 Cytb GCAGCCCCTCAGAATGATATTTGTCCTC Smith and Patton

1993

MVZ-45 Cytb ACJACHATAGCJACAGCATTCGTAGG Smith and Patton

1993

MVZ-16 Cytb AAATAGGAARTATCAYTCTGGTTTRAT Smith and Patton

1993

MVZ-17 Cytb ACCTCCTAggAgAYCCAgAHAAYT Smith and Patton

1993

MVZ-14 Cytb GGTCTTCATCTYHGGYTTACAAGAC Smith and Patton

1993

12S L82 12S CATAGACACAGAGGTTTGGTCC Nedbal et al. 1994

12S H900 12S TGACTGCAGAGGGTGACGGTGTGT Nedbal et al. 1994

GHR1f GHR GGRAARTTRGAGGAGGRGAACAC Jansa et al. 2009

GHRend1F GHR GATTTTGTTCAGTTGGTCTCTGCT Jansa et al. 2009

IRBP-A IRBP-A ATGGCCAACGTCCTCTTGGATAAC Stanhope et al.

1992

! 140 IRBP-B IRBP-B CGCAGGTCCATGATGAGGTGCTCC Stanhope et al.

1992

! 141 VITA

Jesús Abraham Fernández Fernández was born in Mexico City to Miguel Angel and Heriberta Fernández. He grew up in the city of Santa Ana Chiautempan, Tlaxcala,

México, where he graduated from Escuela Secundaria Técnica No. 4. He received his bachelor’s degree in biology from Universidad Autónoma de Tlaxcala in 2000 working on bird diversity. In 2002, he began graduate school at the National University of

Mexico, where he obtained the title of Master in Science under the tutelage of Dr.

Fernando Cervantes. His thesis topic was phylogenetics and biogeography of kangaroo rats. In August of 2006, Jesús Abraham Fernández Fernández entered the Graduate

School of Louisiana State University. Under the supervision of Dr. Mark S. Hafner,

Jesús will graduate with the degree of Doctor of Philosophy in December, 2011.

! 142