UNLV Retrospective Theses & Dissertations

1-1-2003

Evolutionary and biogeographic histories in a North American rodent family (Heteromyidae)

Lois Fay Alexander University of Nevada, Las Vegas

Follow this and additional works at: https://digitalscholarship.unlv.edu/rtds

Repository Citation Alexander, Lois Fay, "Evolutionary and biogeographic histories in a North American rodent family (Heteromyidae)" (2003). UNLV Retrospective Theses & Dissertations. 2561. http://dx.doi.org/10.25669/g45f-ub85

This Dissertation is protected by copyright and/or related rights. It has been brought to you by Digital Scholarship@UNLV with permission from the rights-holder(s). You are free to use this Dissertation in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s) directly, unless additional rights are indicated by a Creative Commons license in the record and/or on the work itself.

This Dissertation has been accepted for inclusion in UNLV Retrospective Theses & Dissertations by an authorized administrator of Digital Scholarship@UNLV. For more information, please contact [email protected]. NOTE TO USERS

This reproduction is the best copy available. UMI

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EVOLUTIONARY AND BIOGEOGRAPHIC HISTORIES IN A

NORTH AMERICAN RODENT FAMILY (HETEROMYIDAE)

by

Lois Fay Alexander

Bachelor of Science Oregon State University, Corvallis 1990

Master of Science Oregon State University, Corvallis 1994

A thesis submitted in partial fialfillment of the requirements for the

Doctor of Philosophy Degree in Biological Sciences Department of Biological Sciences College of Sciences

Graduate College University of Nevada, Las Vegas May 2004

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3143374

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

UMI

UMI Microform 3143374

Copyright 2004 by ProQuest Information and Learning Company.

All rights reserved. This microform edition is protected against

unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dissertation Approval uNiy The Graduate College University of Nevada, Las Vegas

A p ril 12 ■ 20 04

The Dissertation prepared by

Lois Fay Alexander

Entitled

Evolutionary and Blogeographic Histories in a North American

Rodent Family (Heteromyidae)______

is approved in partial fulfillment of the requirements for the degree of

Doctor of Philosophy In Biological Sciences_____

Examination Committee Chair

Dean of the Graduate College -y. Examination Committee Memb

Examination Committee Member

yCraduate College Faculty Representative

1017-52 11

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT

Evolutionary and Biogeographic Histories in a North American Rodent Family (Heteromyidae)

by

Lois F. Alexander

Dr. Brett R. Riddle, Examination Committee Chair Professor of Biological Sciences University of Nevada, Las Vegas

The family Heteromyidae includes six genera of rodents traditionally placed in

three subfamilies endemic to the Nearctic and northern Neotropical blogeographic

regions. Although several of these taxa represent intensively studied members of

North and Central American ecosystems (e.g., kangaroo rats, pocket mice),

phylogenetic relationships within and among subfamilies, genera, and species-groups

are poorly understood. Here, I used maximum likelihood (ML), Bayesian, and

maximum parsimony (MP) analyses of sequence data from two mitochondrial DNA

genes—COIII (699 bp) and cytochrome b (1140 bp)—to investigate phylogenetic

relationships among 55 species-level taxa. I found robust support for monophyly of

genera Dipodomys, Microdipodops, Chaetodipus, and Perognathus; sampling of

Liomys and Heteromys was inadequate to evaluate their reciprocal status. All analyses

converge on a phylogeny that robustly resolves several historically contentious issues,

including monophyly of the subfamily Dipodomyinae {Microdipodops + Dipodomys),

and a monophyletic Chaetodipus that includes C. formosus, C. baileyi, C.

m

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rudinoris and C. hispidus. However, Perognathinae {Perognathus + Chaetodipus) is

not supported, with no basal resolution among Perognathus, Chaetodipus,

Dipodomyinae, and Heteromyinae. Many intrageneric clades receive strong support

and are discussed herein.

I used the phylogenetic information to evaluate several hypotheses regarding the

evolution of the Heteromyidae. I separately evaluated the evolution of morphological

characters (body size, number of hind toes [Dipodomys only], and rump spines

[Chaetodipus only]) and macroecological evolution by coding each taxon into classes

(body size - very small, small, medium, large, or very large; number of toes - 4 or 5;

rump spines - presence or absence) and biome-level categories (grassland steppe,

chaparral, subtropical thomscrub, warm desert, or cold desert) and then tracing the

history of the morphological characters and ecological transitions during radiation of

extant groups using the Fitch optimization in MacClade. Because the mitochondrial

DNA genes chosen for these analyses resulted in very limited resolution at the basal

nodes of the Heteromyidae tree, recommendations for future directions of study are

discussed.

IV

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

ABSTRACT ...... üi

LIST OF FIGURES ...... vii

ACKNOWLEDGEMENTS ...... vüi

CHAPTER 1 INTRODUCTION ...... 1 Taxonomic Background ...... 1 Landscape Evolution ...... 4 Fossil H istoiy ...... 7 A Priori Hypotheses ...... 8 Morphological and Macroecological Evolution ...... 11

CHAPTER 2 METHODS...... 13 Phylogenetics of Heteromyidae ...... 13 Morphological and Macroecological Evolution ...... 15

CHAPTER 3 RESULTS...... 17 Phylogenetics of Heteromyidae ...... 17 Dipodomys...... 17 Chaetodipus ...... 23 Perognathus...... 25 Morphological and Macroecological Evolution ...... 27

CHAPTER 4 DISCUSSION...... 34 Phylogenetics of Heteromyidae ...... 34 Dipodomyinae ...... 34 Perognathinae ...... 36 Morphological and Macroecological Evolution ...... 37 Dipodomys...... 37 Chaetodipus ...... 40 Perognathus...... 42 Summary...... 42 Future W ork ...... 44

LITERATURE CITED ...... 48

APPENDIX Specimens Examined, Localities, and Characters Mapped ...... 59

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA ...... 64

VI

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES

F igure 1 T estable Hypotheses ...... 10 Figure 2 Maximum parsimony tree ...... 19 Figure 3 Bayesian tre e ...... 20 Figure 4 Maximum likelihood tree ...... 21 Figure 5 Gamma-corrected mean distances ...... 22 Figure 6 Dipodomys character mapping on bayesian tree ...... 28 Figure 7 Dipodomys character mapping on maximum likelihood tree ...... 29 Figure 8 Chaetodipus character mapping on bayesian tree ...... 30 Figure 9 Chaetodipus character mapping on maximum likelihood tree ...... 31 Figure 10 Perognathus character mapping on bayesian tree ...... 32 Figure 11 Perognathus character mapping on maximum likelihood tree ...... 33

Vll

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS

I extend many thanks to my committee (Drs. Brett R. Riddle, Daniel Thompson,

Peter Starkweather, David Hafiier, and Steven Rowland) for their patience and

understanding. My husband, Douglas Merkler, is a constant source of inspiration and

everlasting moral support as well as extremely valuable technical advice. I thank Troy

L. Best (Department of Biological Sciences, Auburn University, Alabama) and David

J. Hafiier (New Mexico Museum of Natural History, Albuquerque, New Mexico) for

many tissue samples without which this project would not have been possible. I thank

James Demastes (Department of Biology, University of Northern Iowa), Mark

Engstrom (Royal Ontario Museum, Toronto, Ontario), James L. Patton (Museum of

Vertebrate Zoology, University of California), and Phillip Sudman (Biological

Sciences, Tarleton State University, Stephenville, Texas) for additional tissue samples.

I thank F. A. Cervantes (Institute de Biologia, Universidad Nacional Autonoma de

Mexico), S. T. Alvarez-Castaneda and P. Cortés-Calva (Centro de Investigaciones

Biologicas del Noroeste), and their students for assistance in the field and for

arranging collecting permits in Mexico. Financial support for this project was

provided fiom Grants-in-aid of Research fiom the American Society of

Mammalogists, Sigma Xi, and the American Museum, Theodore Roosevelt Fund to

L.F.A.; several Graduate Student Association Research Grants fiom the University of

Nevada Las Vegas to L.F.A. and grants fiom the NSF to Brett R. Riddle (DEB-

9629787) and David J. Hafiier (D.E.B.-9629840).

viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1

INTRODUCTION

Taxonomic background

The family Heteromyidae is well established within the order Rodentia as a

member of the superfamily Geomyoidea along with Geomyidae (DeBry and Sagel

2001; Montgelard et al. 2002). Historically, the members of Heteromyidae and

Geomyidae have been recognized as either a single family (Geomyidae) in two

subfamilies (Geomyinae and Saccomyinae or Heteromyinae) or more recently as two

distinct families (Heteromyidae and Geomyidae) in the superfamily Geomyoidea

(Ryan 1989; Williams et al. 1993). This family of deciduous thorn-scrub and arid-

adapted rodents is comprised of six genera {Chaetodipus, Dipodomys, Heteromys,

Liomys, Microdipodops, and Perognathus) including approximately 57 species that are

distributed from southern Canada, throughout the aridlands of western North America,

south through Mexico, Central America, and into northern South America.

Chaetodipus, Dipodomys, Microdipodops, Perognathus, and Liomys are found

primarily in dry, arid habitats, whereas Heteromys occupies moist neotropical

rainforest habitats (Schmidly et al. 1993).

Heteromyidae is currently divided into three subfamilies: Perognathinae

containing Perognathus and Chaetodipus', Heteromyinae containing Heteromys and

Liomys', and Dipodomyinae containing Dipodomys and Microdipodops (Hafher 1993;

1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Patton 1993; Wahlert 1993; Williams et al. 1993). The most controversial aspect of

this arrangement has been the recent placement of Microdipodops within

Dipodomyinae (Hafiier 1978; Hafiier and Hafiier 1983; Reeder 1956; Ryan 1989;

Wahlert 1985) instead of the previously accepted placement within Perognathinae

(Hafiier 1978; Hall 1981; Wood 1935). It is possible that this group represents an

independent lineage with no close living relatives (Hafiier 1978; Hafiier 1993). In

addition, it has been suggested that Heteromys is paraphyletic relative to Liomys

(Rogers 1990). My sampling of Heteromys and Liomys is insufiBcient to address this

question and is being addressed elsewhere (Duke Rogers, pers. comm. 2003).

Phylogenetic relationships among the 22 species of kangaroo rats (genus

Dipodomys) also are in a state of confusion. Many attempts have been made to

resolve this problem, but all were prior to the advent of modern molecular systematics.

Several morphological comparisons have been attempted previously, but no satisfying

conclusions can be drawn. Many researchers have attempted to arrange kangaroo rats

into phylogenetic groups, but have generally failed partly because all 22 previously

recognized species are morphologically similar (Stock 1974). The first comprehensive

evaluation of relationships within the genus Dipodomys was that of Grinnell (1921)

who based his analysis on external and cranial morphology. Many attempts have been

made to revise the original groups suggested by Grinnell (1921). Kangaroo rats have

been grouped by cranial and skeletal characteristics (Best 1993; Schnell et al. 1978),

tooth characteristics (Nader 1966; Wood 1935), baculum (Best and Schnell 1974; Burt

1936), skeletal specialization and visceral measurements (Setzer 1949), protein

variation (Johnson and Selander 1971), chromosomes (Patton and Rogers 1993; Stock

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1974), satellite DNA (Mazrimas and Hatch 1972), immunological distance (Hatch and

Mazrimas 1977), and by attempting to combine several sources of information

(including “field experience”) to obtain groups (Lidicker 1960b). Analyses based

strictly on cranial and skeletal morphology results in “clades” that are similar

phenetically (“large” taxa grouped together; “small” taxa grouped together) but not

necessarily evolutionarily related (Best 1993). Phylogenies resulting Ifom these

various analyses have significantly different groupings of species, and a consensus

regarding phylogenetic relationships within the genus Dipodomys has not been

reached.

A phylogenetic analysis using molecular techniques is needed to evaluate this

problem. Once a reliable phylogeny is estimated, questions can be addressed

regarding the evolutionary and biogeographic history of the group. Patterns and

timing of genetic divergences can be used in conjunction with historical geologic

events to evaluate the historical biogeography of the genus Dipodomys.

Biogeographic structure within this group is not likely to be all “deep history”

(Miocene - Pliocene) or all “shallow history” (Pleistocene), but rather a combination

resulting from the variety of processes operating at different scales. Phylogenetic

history among species and species groups within the genus Dipodomys likely will

reflect blogeographic pattern congruent with landscape evolution in the Miocene -

Pliocene as well as more recent divergences in the Pleistocene.

Within particular species in the genus Dipodomys, many researchers have

evaluated specific questions such as foot drumming behavior (Randall 1994; Randall

1995), population structure (Good et al. 1997), dispersal (Jones et al. 1988; Waser and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Elliott 1991), genetic structure (Hamilton et al. 1987; Waser and Elliott 1991) and

survival and reproductive effort (Jones 1988; Waser and Jones 1991). With robust

phylogenetic hypotheses regarding evolutionary relationships among species in this

genus, many of the species-level questions that have been studied previously could be

expanded to include evolutionarily related species groups or other relevant units of

classification.

The phylogenetic relationships of subfamilies, genera, and species groups within

the family Heteromyidae need comprehensive evaluation. My first set of objectives of

this study was to estimate phylogenetic relationships among the six genera in the

family Heteromyidae. This included a re-evaluation of the three proposed subfamilies,

an evaluation of the phylogenetic relationships among the 22 species of kangaroo rats

in the genus Dipodomys, and a review of the phylogenetic relationships within

Chaetodipus and Perognathus. In order to accomplish these objectives, it is important

to consider the historical fi-amework in which Heteromyids evolved.

Landscape Evolution

Fossil records suggest that heteromyid rodents originated during the Oligocene in

western North America; major lineages diversified within the Neogene (Wahlert

1993), during which time there was pronounced geological activity and landscape

evolution. The development of the deserts of western North America began with the

uplift of the Rocky Mountains in the United States, and the Sierra Madre Oriental in

Mexico during the Eocene (Levin 1978). As large areas of the Rocky Mountain

region were uplifted, enormous volumes of material eroded Ifom the eastern slopes

and filled intermontane basins to the east, thus contributing to the formation of the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Great Plains (Levin 1978). These erustal movements continued throughout the

remaining epochs of the Cenozoic with additional uplift and erosion creating new

sedimentary layers throughout the Oligocene, Miocene, and Pliocene. During the late

Miocene and Pliocene, the Great Plains evolved fi-om a woodland savanna to a

grassland savanna to a grassland steppe.

Also beginning in the Eocene, a continuous subduction zone off the west coast

created a parallel zone of volcanic activity in the area that later became the Sierra

Madre Occidental in Mexico (Ortega-Gutiérrez and Guerrero-Garcia 1982). In the

Oligocene and Miocene, hundreds of calderas and their resulting ash-flow tuff

deposits, caused by the subduction of the pacific plate, combined to form the high

plateau that makes up the Sierra Madre Occidental (Mexican Plateau; Swanson and

McDowell 1984). Subsequent block faulting and erosion formed isolated basins on

the eastern and western aspects of the Mexican Plateau. During the Miocene, the

Coast and Cascade Ranges in southern Oregon began to uplift, causing a rainshadow

that began to create drier climates and expanding grasslands on the east side of these

ranges; this region makes up the northern Great Basin of today (Baldwin 1964).

From the Oligocene to the late Miocene, forests and savannas were gradually

replaced with steppe and semi-desert habitats. This continued during the latest

Miocene with expansion of regional deserts, grasslands and shrub-steppes. The

remains of plants and animals in the sediments suggest that the Pliocene was cooler

and drier than the Miocene (Levin 1978).

The Sierra Nevada Mountains were compressed and intruded during the Jurassic,

but were steadily reduced by erosion throughout the Tertiary. The continental crust

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. east of the Sierra Nevada began to stretch in an east-west direction in the Miocene.

The crust broke into a series of north-south-trending valleys and mountain ranges

(Fiero 1986). Less than five million years ago, through a combination of uplift of the

Sierran block and down-dropping of the area to the east, the Sierra Nevada rose again.

Rising far more steeply to the east than the west, the entire Sierra Nevada tilted with a

gentle slope westward to California's Central Valley and steep eastern slope (Levin

1978). During the late Pliocene, the continued continental crust expansion increased

the northward extent of the Gulf of California; uplifted the Peninsular, Tehachapi, and

Coast Ranges of Baja California and California; and drained California’s Central

Valley (Norris and Webb 1976).

The Cascade and Coast Ranges, the Sierra Nevada, the Transverse and Peninsular

Ranges, and the Sierra Madre Occidental blocked the prevailing western storm tracks,

whereas the Sierra Madre Oriental and the Rocky Mountains blocked the summer

monsoon moisture moving north and west from the Gulf of Mexico. The combined

effect of blocking the moisture from the Pacific Ocean to the west, and the summer

moisture from the Gulf of Mexico, was the drying and gradual formation of the Great

Basin, Mojave, Peninsular, Sonoran, and Chihuahuan Deserts. The Great Basin and

northern Mojave deserts changed from woodland savannas to shrub steppe

communities during the latest Miocene and early Pliocene. Similarly, the Mexican

Plateau changed from a semi-arid savanna to a desert-scrub steppe woodland, and the

Sonoran and southern Mojave deserts changed from semi-desert and subtropical

thomscrub to desert scrub (reviewed in Riddle 1995).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A long history of dynamic geologic and climatic shifts certainly had a pronounced

effect on the aridland biotas of western North America. Many pre-Pleistocene to late-

Pleistocene alternative explanations for spéciation events can be hypothesized and

tested with the use of modem molecular techniques that are now available. By

matching phylogenetic patterns of genetic diversity with biogeographic patterns of

aridlands taxa and postulated geologic events, hypotheses regarding causal

relationships between earth history and the evolutionary history of the biota can be

evaluated. Analyses of evolution and biogeography at different taxonomic levels can

help elucidate the biotic history of the deserts (generally) and the history of the

Heteromyidae (specifically).

Fossil History

The first heteromyid fossils (Proheteromys) appear in the fossil record during the

Oligocene (Wahlert 1993). Proheteromys may be an early member of the

Heteromyinae (Wood 1935), but Heteromyinae and Perognathinae are not clearly

distinguishable in the fossil record in the early Miocene (Wilson 1960). The earliest

identifiable Perognathus / Chaetodipus fossils appear during the Hemingfordian North

American Land Mammal Age (NALMA) of about 20 million years ago (Wahlert

1993), with representatives that are similar in size and dental morphology to extant

Perognathus species as well as extant Chaetodipus species appearing in the late

middle Miocene (Williams 1978). Therefore, Perognathus and Chaetodipus were

possibly differentiated as early as the late middle Miocene (Williams 1978).

Positively distinguishing between the two genera from fossils, however, has not been

possible thus far. Also appearing during the Hemingfordian NALMA of the Miocene,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cupidinimus previously was considered to be the earliest representative of the

Dipodomyinae, but recently it has been suggested that it may have closer affinities

with the Perognathines (Wahlert 1993). The earliest recognizable Dipodomys fossils

appear during the early Barstovian age of the middle Miocene, whereas the earliest

recognizable Microdipodops and Liomys fossüs do not appear until the Rancholabrean

age of the Pleistocene (Wahlert 1993). Thus, modem genera (with the possible

exception of Microdipodops) were in place for the majority of the tectonic

development of the west including the Basin and Range expansion and the uplifts of

the Rocky Mountains, Colorado Plateau, Sierra Madre Oriental, Sierra Madre

Occidental, Sierra Nevada, Peninsular Mountains, Coast Range, and the Cascade

Mountains. This makes the family Heteromyidae a very interesting taxonomic group

to use as a barometer of evolutionary development coinciding with development of a

geologically dynamic region.

A Priori Hypotheses

Given that the fossil record places the first Heteromyid fossils in the Oligocene,

but that the subfamilies and genera are not clearly distinguishable, this raises questions

as to the timing of subfamily and generic diversification. For example, discerning the

difference between Perognathus and Chaetodipus fi-om the fossil record has not been

possible; however, they have been treated as monophyletic groups, at least at the

subgenus level, based on molecular analyses of extant taxa (Patton 1993; Patton et al.

1981; WiUiams et al. 1993). The extent to which genera cannot be discriminated

from the fossil record may be extreme. For example, Microdipodops does not appear

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in the fossU record until the Pleistocene, although it could represent a very deep

lineage split if it is indeed the sister lineage of Dipodomys.

The three currently recognized subfamilies and previously proposed relationships

among them are presented as testable hypotheses in Figure 1. Assuming monophyly

of the three subfamilies, examples of other possible alternative hypotheses include:

Perognathinae sister to Heteromyinae with Dipodomyinae basal; and Heteromyinae

sister to Dipodomyinae with Perognathinae basal. Another set of testable hypotheses

of relationships involves the genera and species within the traditional subfamilies.

Notably, is Microdipodops related to Perognathinae {sensu stricto), some subset of

Perognathinae, or Dipodomyinae? Additional relevant hypotheses of relationships

include: inclusion of formosus, baileyi, and hispidus within a monophyletic

Chaetodipus {formosus and baileyi have been considered somewhat intermediate

morphologically between Chaetodipus and Perognathus [Patton et al. 1981]; with no

reference to the elevation of Chaetodipus to generic rank by Hafiier and Hafiier

[1983], Hoffineister [1986] considered all of the Perognathinae to be genus

Perognathus when he placed hispidus into its own subgenus, Burtognathus, based

primarily on bacular and chromosomal characteristics); and an investigation of

previously proposed species groups within Chaetodipus (Patton and Rogers 1993;

Patton et al. 1981), Dipodomys (Best and Schnell 1974; Johnson and Selander 1971;

Lidicker 1960b; Schnell et al. 1978; Stock 1974), and Perognathus (Williams 1978).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subgenus Perognathus Perognathinae formosus baileyi

subgenus Chaetodipus Perognathinae subgenus Burtognathus ? hispidus Microdipodops Dipodomyinae

Heteromyinae Dipodomys

Liomys + Heteromys

Fig. 1. Testable hypotheses of previously or currently proposed (dashed lines and

boxes) relationships among the three currently recognized subfamilies (solid lines and

boxes).

10

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Morphological and macroecological evolution in the Heteromyidae

Within the family Heteromyidae there exists a large range of body sizes from the

smallest pocket mouse (50 mm head and body length; 8 g) to the largest

kangaroo rat (180 mm head and body length; 138 g). These rodents occur as far north

as British Columbia and Saskatchewan and as far south as the Pacific coast of

Columbia and the northern coast of Ecuador, South America (Schmidly et al. 1993).

Heteromys inhabit more mesic environments than any of the other Heteromyids; they

are found primarily in lowland rainforests and tropical cloud forests of Central and

northern South America and tend to be large in body size (H. anomalus and H

australis are medium). Liomys is mostly confined to semi-arid habitats of Mexico and

Central America and tend to be medium in size {L. spectabilis is large). It has been

suggested that resource partitioning between Heteromys and Liomys has resulted in

Heteromys occurring in the higher and wetter regions of their distribution and Liomys

remaining in the lower, drier regions (Genoways 1973). Microdipodops are small

bodied and are restricted to arid regions of the central Great Basin. Dipodomys occur

throughout the arid or semi-arid regions of western North America including both

warm and cold deserts, grasslands, and chaparral. Most Dipodomys are either medium

(n=9) or large in size (n=9), but four are very large. The very small or small

Perognathus occur in desert and grassland regions that extend north to British

Columbia and Saskatchewan, east to the Mississippi River, and south into Mexico;

Perognathus alticola and P. parvus are the only medium sized Perognathus and are

found in two isolated populations in chaparral regions of California and the Great

Basin, respectively. With two exceptions, Chaetodipus occurs in warm desert,

11

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. chaparral, subtropical thom-scrub areas of the southern United States and northern

Mexico and are small or medium in body size; C. hispidus ranges farther north (North

Dakota) and east (Louisiana) than any other Chaetodipus (Schmidly et al. 1993) and

C. formosus inhabits intermountain sagebrush areas more typical of cold desert

microhabitats in spite of a primarily warm desert and chaparral distribution. My

second set of objectives was to analyze morphological (body size and other genus-

level traits) and ecological (macroecological habitats at a biome level) evolution

within selected subsets of heteromyids from a phylogenetic context.

12

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2

METHODS

Phylogenetics o f Heteromyidae

I applied maximum parsimony (MP), maximum likelihood (ML), and Bayesian

methods using PAUP 4.0b (Swofford 1999) and MrBayes 2.01 (Huelsenbeck 2000) to

analyze mitochondrial DNA sequences from nearly all species of Heteromyidae from

the United States and Mexico as well as exemplars of the neotropical genera. Within

Dipodomys, Microdipodops, Chaetodipus, and Perognathus, all species except for one

of questionable taxonomic status (Chaetodipus lineatus, Williams et al. 1993) were

sampled. I included 16 species-level taxa from the genus Chaetodipus; C. lineatus

was not included, and C. arenarius was split into two species (arenariusl and

arenariusl) based on preliminary evidence indicating presence of a “cryptic” species

that is provisionally assignable to C. dalquesti (Riddle et al. 2000b; Roth 1976; but see

Best 1993). I included 11 species-level taxa from the genus Perognathus. I included

22 previously described taxa of Dipodomys, including 20 species-level taxa; D.

venustus and D. elephantinus are now considered to be conspecific (Best et al. 1996)

and preliminary evidence suggests thatD. insularis, D. margaritae, and D. merriami

from the southern Baja California peninsula also might be conspecific (Alexander et

al. in litt.; Riddle et al. 2000b). I included both species of Microdipodops, one

representative of Heteromys (H. desmarestianus), and two representative species of

13

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Liomys (L. pictus and L. irroratus). Geomys breviceps, which belongs to the sister

family Geomyidae, served as the outgroup taxon (Appendix).

Total genomic DNA was extracted from frozen tissue following a lysis buffer

protocol (Longmire et al. 1991). A 705-bp fragment of mitochondrial DNA including

the cytochrome oxidase subunit 3 (CO III) gene was amplified via polymerase chain

reaction (PCR) with primers H8618 and L9323 (Riddle 1995). The cytochrome-b

(cyt-b) region of the mitochondria was amplified via PCR with primers MVZ05 and

MVZ14 (Smith and Patton 1993). PCR fragments were gel purified with a QBiogene

Gene Clean kit following manufacturer protocols. The same primers that were used

for PCR amplification of CO III and cyt-b were used to sequence both strands of every

individual. In addition to the primers used for PCR, one more primer (15162) was

used for sequencing the cyt-b gene (Taberlet et al. 1992). A total of 699-bp of the CO

III gene and 1140-bp of the cyt-b gene were aligned with BioEdit (Hall 1999) and

used in all analyses.

These gene regions are not independent estimates of phylogeny, but are evolving

at a similar rate (Xia 1998) and therefore are being used in combination to increase

numbers of informative characters available for analysis. These relatively rapidly

evolving genes are likely to lose phylogenetic signal at the base of the heteromyid tree,

but they should be informative for the intra- and intergeneric questions that are posed

herein. Maximum parsimony analyses (heuristic search, random addition sequence,

tbr branch swapping, 20 repetitions) were performed with equal weights, and because

of differential probability of saturation relative to site positions in a coding sequence, I

also weighted characters by site-specific transition-transversion ratios. Both of these

14

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. analyses were repeated with 500 bootstrap repetitions. With the assistance of

ModelTest 3.06 (Posada and Crandall 1998), I chose the Tamura-Nei model (Tamura

and Nei 1993), with a gamma distribution of rate heterogeneity among sites (0.6593)

and an assumption of some invariant sites (proportion = 0.4634), as the best

evolutionary model for the fidl, 2-gene set, and used the resulting likelihood settings in

the maximum likelihood analyses of the entire family Heteromyidae. A ML log-

likelihood ratio test was used to address whether the best ML tree is significantly more

likely than various alternative trees of interest.

For the bayesian analyses, I ran four Markov chain Monte Carlo (MCMC) chains

simultaneously using MrBayes for 1,000,000 generations sampling every 100

generations, resulting in 10,000 trees. The point at which likelihood scores stabilized

was noted and trees recorded prior to that point (1,000 trees) were discarded as the

‘bum-in’; posterior probabilities were generated from the remaining 9,000 trees. I

compared pruned bayesian trees from each of the three primary clades for which I had

good taxon sampling (Chaetodipus, Dipodomys, and Perognathus) to phylogenetic

hypotheses of previous studies with Kishino-Hasegawa (K-H) tests (which are more

appropriate if a priori hypotheses are available), as well as parallel likelihood tests in

the form of Shimodaira-Hasegawa (S-H) tests (which are more appropriate without a

priori hypotheses; Goldman et al. 2000; Shimodaira and Hasegawa 1999).

Morphological and macroecological evolution in the Heteromyidae

To examine the evolution of body size, I coded each taxon into body-size

categories (very small = 8-10 g, small = 11-16 g, medium = 20-60 g, large = 61-90 g,

or very large = 91-140 g) primarily based on published estimates of mean male body

15

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mass (Best 1988; Best and Lackey 1985; Best and Thomas 1991; Jones 1985; Wilson

and RufF 1999). Published weights were unavailable for 13 species but were

estimated by comparing head and body length of known-weight species and

developing an index to use for estimating weights directly from head and body data

(D. J. Hafrier, pers. comm.). For Dipodomys, mass = (1.786*head+body) - 140.708;

for Chaetodipus, mass = (0.679*head+body) - 38.233; for Heteromys, mass =

(0.681 *head+body) - 30.131, and for Liomys, mass = (1.286*head+body) - 97.857.

In the genus Dipodomys, some species have four toes on each of the hind feet and

others have five toes; in the genus Chaetodipus, some species have dorsal rump

spines, whereas others are smooth. These characters were coded as presence or

absence and traced onto phylogenetic hypotheses. In a separate analysis, I evaluated

macroecological evolution by coding each taxon into biome-level categories

(grassland steppe, chaparral, subtropical thom-scrub, warm desert, or cold desert;

Appendix). I then traced the history of morphological and ecological transitions

during radiation of extant groups using the Fitch optimization in MacClade (Maddison

and Maddison 2000) onto the trees resulting from both the ML and Bayesian analyses.

16

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3

RESULTS

Phylogenetics o f the family Heteromyidae

Regardless of method used for analysis (Figs. 2-4), I found robust support for the

monophyly of genera Dipodomys, Microdipodops, Chaetodipus, and Perognathus; for

a clade that includes Microdipodops and Dipodomys (i.e., Dipodomyinae); and for a

clade that includes exemplars representing Liomys and Heteromys (i.e.,

Heteromyinae). Gamma-corrected distances among the genera are provided in Figure

5. The parsimony analysis (Fig. 2) offers weak support for a clade comprised of

Chaetodipus and Perognathus, which is unsupported otherwise (Figs. 3, 4).

Generally, I found veiy little support for sister-group relationships at the base of the

tree. Although the ML analysis shows a fully resolved tree with a basal Perognathus,

grouped next with Chaetodipus, then Heteromyinae, then Dipodomyinae (Fig. 2-4; -

Ln likelihood = 36026.84165), this tree is not significantly better than one showing the

traditional subfamilial topology (-Ln likelihood = 36027.25254; Shimodaira-

Hasegawa test,/? = 0.459).

Dipodomys

Kangaroo rats have been divided into six to nine different species groups primarily

based on morphology (Best and Schnell 1974; Blair 1954; Davis 1942; Grinnell 1921;

Lidicker 1960a; Lidicker 1960b; Schnell et al. 1978; Setzer 1949) and chromosomes

17

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Johnson and Selander 1971 ; Stock 1974). Our analyses of the cyt-b and CO III

regions of the mitochondrial DNA are consistent with many of the previous groupings,

at least in terms of group membership (Figs. 2-4). We found support in all analyses

for a Dipodomys agilis species group with an agilis plus simulons clade grouping with

an elephantinus plus venustus clade. Dipodomys heermanni, panamintinus and

stephensi consistently group together with strong support in aU analyses. Dipodomys

gravipes, ingens, and microps exhibit affinities to the heermanni group, but do not

have any support in their specific relationships. All of these taxa together form a clade

that is moderately well supported and joins D. californicus.

I found robust support in aU analyses for a Dipodomys merriami species group

comprised of merriami, insularis, margaritae, and nitratoides that is sister to an elator

plus phillipsii clade (D. phillipsii species group). A spectabilis plus nelsoni clade {D.

spectabilis species group) is strongly supported in all analyses, as is the basal position

of D. deserti. In all analyses except for the ML, we found some support (weak in the

MP analysis) for a compactas plus ordii clade. We found very little resolution for the

relationships among species groups. The MP analyses resulted in an unresolved

polytomy amongst the heermanni plus agilis plus californicus clade, the merriami plus

phillipsii clade, ordii, compactas, and a spectabilis plus nelsoni clade. Kishino-

Hasegawa and Shimodaira-Hasegawa tests were performed to compare the mtDNA

MrBayes Dipodomys clade with that generated fi'om chromosomal data by Patton and

Rogers (1993); the mtDNA tree generated in this study was significantly better than

the chromosomal tree (K-H =p < 0.0001, mtDNA tree length = 2891, chromosome

tree length = 3165; S-H = p < 0.0001 ; -InL = 14113.069).

18

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C. goldmani C. artus C. nelsoni C. Intermedlus C. penicillatus C. eremlcus C. pernbc C. arenariusi C. arenarluSZ C. splnatus C. fallax C. calffomlcus C. hispldus C. balleyl C. rudlnoris C. formosus P. Inomatus P. longlmembrls P. amplus P. flavus P. meniaml P. parvus U- t a h P. alOcola P. parvus - Columbia Plateau P. flavescens P. apache P. fasclatus D. agllls D. simulans D. venustus D. elephantinus D. heermanni D. panamintinus D. stephensi D. gravipes D. microps D. Ingens D. californicus D. Insularis D. merriami-s. B a j a D. margaritae D. merriami D. nitratoides D. elator D. phillipsii D. ordii D. compactus D. nelsoni D. spectabilis D. deserti M. paWdus M. megacephalus L pictus L. Irroratus H. desmarestlanus G. bre\riceps

Fig. 2. MP tree with characters weighted by site-specific transition-transversion ratios

(Cl = 0.1795, RI = 0.4915, tree length = 10130.3, number of parsimony informative

characters = 854, total number of characters = 1839). Numbers along branches

summarize results of 500 bootstrap repetitions. All nodes with bootstrap support < 50

have been collapsed. The topology of the equally weighted MP tree was the same:

tree length = 9047, Cl = 0.1896, and RI = 0.1823.

19

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. heermanni panamintinus stephensi microps gravipes ingens agiiis simuians venu^us eiephantinus caiifomicus insularis merriami - S. Baja margaritae merriami nitratoides eiator phiiiipsi ordii compactus neisoni spectabilis deserti . paiiidus . megacephaius goidmani artus neisoni intermedius peniciiiatus eremicus pemix arenariusi arenarluSZ caiifomicus faiiax spinatus hispidus baiieyi rudinoris formosus fiavescens apache fasciatus parvus -Utah alticola parvus -Columbia Plateau inomatus iongimembris amplus flavus merriami pictus irroratus desmarestlanus breviceps Fig. 3. Bayesian tree generated from running 4 MCMC chains simultaneously for

1,000,000 generations and sampling every 100 generations. Numbers along branches

summarize results of 95% probabihty values from 9,000 trees after discarding the first

1,000 trees as the bum-in. All nodes with support < 95 have been collapsed.

20

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. D. heermanni D. panamintinus D. stephensi D. gravipes D. ingens D. microps D. agIKs D. simulans D. venustus D. elephantinus D. caiifomicus D. Insularis D. merriami- S. Baja D. margaritae D. merriami D. nitratoides D. elator D. phinipsil D. compactus D. ordil D. nelsoni D. spectabilis D. desartl M. pallldus M. megacephalus L pictus L Irroratus H. desmarestlanus C. goldmani C. artus C. nelsoni C. Intermedius C. eremlcus C. pemIx C. penicillatus C. aranariusl C. arenariu^ C. caldomicus C. fallax C. splnatus C. hispidus C. balleyl C. rudinoris C. formosus P. Inomatus P. longlmembrls P. amplus P. flavus P. merriami P. flavescens P. apache P. fasclatus P. parvus -Utah P. alticola P. parvus -C olum bia G. breviceps

Fig. 4. Maximum-likelihood tree for ail representatives of the family Heteromyidae

included in these analyses (Tamura-Nei + 1 + gamma model).

21

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 5 f I 5; / / / 1 Dipodomys 0.198 0.294 0.354 0.339 0.353 0.362

Microdipodops 0.016 0.207 0.354 0.288 0.361 0.370

Liomys 0.020 0.021 0.218 0.257 0.359 0.366

Heteromys 0.020 0.018 0.018 0.366 0.340

Chaetodipus 0.018 0.020 0.018 0.020 0.275 0.361

Perognathus 0.019 0.019 0.019 0.017 0.018 0.293

Fig. 5. Gamma-corrected mean distance within genera (bold diagonal); gamma-

corrected mean distance between genera (above diagonal) and corresponding standard

error below diagonal.

22

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chaetodipus

Interspecific relationships within the monophyletic genus Chaetodipus remain the

primary confiision. Sixteen species-level taxa were included in the analysis of

Chaetodipus. Several convincing lines of evidence now have placed C. formosus

solidly within the genus Chaetodipus rather than Perognathus (Patton et al. 1981).

Previously described species-groups for this genus were based primarily on general

morphologic similarities rather than actual relationship; the taxonomic breakdown has

included a species group with moderately well developed rump spines (including

intermedius, nelsoni, fallax, artus, and goldmani) and a species group without defined

rump spines (including penicillatus [+eremicus], pernix, and arenarius). Most of the

remaining taxa {californicus, spinatus, baiieyi [+rudinoris], hispidus, and formosus)

were placed in monotypic species groups (Patton et al. 1981).

Based on chromosome research summarized by Patton and Rogers (1993), C

penicillatus, C. baiieyi, and C. pernix each have significant karyotypic variation that

may reflect cryptic species. In the cases of C. penicillatus and C. baiieyi, the

previously embedded cryptic species are now recognized as C. eremicus and C.

rudinoris, respectively (Lee et al. 1996; Riddle et al. 2000a).

Chaetodipus formosus and C. baiieyi (+ rudinoris) traditionally have been

described as somewhat intermediate in morphology between Chaetodipus and

Perognathus. I consistently found robust support for a baiieyi plus rudinoris group

and strong (Bayesian - Fig. 3) to very weak support (MP - Fig. 2) for uniting this

group with formosus prior to uniting with all of the rest of the Chaetodipus. They are

23

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. clearly basal lineages, but my results support their inclusion within a monophyletic

Chaetodipus.

Morphologic and chromosomal characteristics traditionally have been used to

separate C. hispidus from the rest of the genus; it has been placed into its own

subgenus of Chaetodipus, Burtognathus (HoflSneister 1986). In each of my analyses,

C. hispidus consistently groups within the monophyletic Chaetodipus and is one step

inside the basal formosus-baileyi-rudinoris subclade (Figs. 2-4).

I found some agreement with the two traditional, morphology-based species

groups (Figs. 2-4). In all analyses, I found robust support for a goldmani, artus,

nelsoni group and support (weak in the MP - Fig. 2) for the subsequent addition of

intermedius to that group (but no support for the traditional inclusion of fallax). I

found robust support for an eremicus, pernix, penicillatus group (but no support for

the traditional inclusion of arenarius); the relationships at the tips differ slightly in the

various analyses. I found solid Bayesian support (Fig. 3) that first unites these two

groups (with ML agreement and concurrence with Riddle et al. 2000a), followed by an

unresolved trichotomy with an arenariusl, arenarius!, californicus, fallax group and

spinatus (all five taxa are grouped in the ML - Fig. 4). The MP analyses, however,

result in an unresolved polytomy of these initial two groups along with spinatus,

fallax, californicus, and a solidly supported arenariusl plus arenariusl group (Fig. 2).

Kishino-Hasegawa and Shimodaira-Hasegawa tests were performed to compare

the mtDNA MrBayes Chaetodipus clade with a tree generated from chromosomal data

by Patton and Rogers (1993) and a tree that was generated from allozyme data (Patton

et al. 1981). The mtDNA tree generated in this study was significantly better than

24

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. both the chromosomal tree (K-H =/> < 0.0001, mtDNA tree length = 3181,

chromosome tree length = 3508; S-H =/? < 0.0001, -InL = 14122.431) and the

allozyme tree (K-H = p< 0.0001, mtDNA tree length = 3181, chromosome tree length

= 3374; S-n=p< 0.0001, -InL = 14126.147).

Perognathus

Eleven species-level taxa were included in this analysis, including two species-

level taxa within P. parvus (based on the preliminary evidence of Riddle 1995).

Previously described species groups (Williams 1978) include a Iongimembris group

{Iongimembris, amplus, and inomatus), a parvus group {parvus and alticolus), a

fasciatus group {fasciatus, flavescens, and apache), and a flavus group {flavus and

merriami). These species groups were supported with high levels of bootstrap support

in the analyses of Riddle (1995) and in the current study. I found strong parsimony

bootstrap support (Fig. 2) and high Bayesian probabilities (Fig. 3) for these species

groups as well as identical relationships within species groups, regardless of analysis

method. In the MP, ML and Bayesian analyses (Figs. 2, 3, and 4, respectively), the

only differences in the resulting phylogenetic hypotheses are the relationships among

the species groups. All three methods support a Iongimembris plus flavus

arrangement. The differences arise in the relationships between this Iongimembris

plus flavus clade and the remaining two species groups. Both the parsimony (Fig. 2)

and Bayesian analyses (Fig. 3) resulted in an unresolved trichotomy of the parvus,

fasciatus, and Iongimembris plus flavus species groups, although there is non­

significant indication (probability of 88) of a fasciatus plus parvus grouping in the

Bayesian analysis, which agrees with that found by Williams (1978). The ML

25

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. analysis (Fig. 4) arranged Iongimembris plus flavus with fasciatus and then the entire

clade with parvus. The specific relationships among the four species groups remain

unresolved.

Kishino-Hasegawa and Shimodaira-Hasegawa tests were performed to compare

the mtDNA MrBayes Perognathus clade to previously suggested relationships based

on chromosomal data (Williams 1978) and mitochondrial DNA data Irom cyt-b and

CO III (Riddle 1995). The MrBayes tree generated in this study was identical to the

relationships described by Williams (1978) and by Riddle (1995) except for the

placement of the parvus group. Williams (1978) placed the parvus group with the

fasciatus group, whereas Riddle (1995) placed the parvus group with the

Iongimembris plus flavus groups. Because the current Bayesian analysis placed these

groups in an unresolved trichotomy, both of the previously published trees were

somewhat better than my bayesian tree, but neither of them were significant in the

Shimodaira-Hasegawa tests. The mtDNA tree generated by Riddle (1995) was better

than the bayesian tree (K-H - p < 0.0001, Riddle mtDNA tree length = 2186,

bayesian mtDNA tree length = 2210; S-H =p < 0.093, -InL = 10933.630); the tree

generated from chromosomal data by Williams (1978) also resulted in a somewhat

better tree (K-H - p < 0.0001, chromosome tree length = 2192, mtDNA tree length =

2210; S-H = p< 0.272, -InL = 10941.316). If the two previously published studies are

compared, the mtDNA tree of Riddle (1995) is shorter, but not significantly better (K-

H =p < 0.3547, Riddle mtDNA tree = 2186, chromosome tree length = 2192; S-H =p

< 0.117, -InL = 10933.631) than that provided by Williams (1978).

26

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Morphological and macroecological evolution in the Heteromyidae

Morphological characters were mapped onto both the bayesian and the maximum

likelihood trees from my phylogenetic analyses. Body size (mass) was mapped onto

bayesian and ML trees for the entire family Heteromyidae, but is shown one clade at a

time (Figs. 6-11). Number of hind toes was mapped onto the Dipodomys clade (Figs.

6, 7) and the presence or absence of rump spines was mapped onto the Chaetodipus

clade (Figs. 8, 9). Because the base of the parsimony and bayesian trees are

unresolved polytomies, tracing the history of the transitions of body size during the

radiation of the family Heteromyidae remains ambiguous; the ambiguity does not

improve when mapped onto the fiiUy resolved ML tree. Dipodomys, Chaetodipus, and

Perognathus were examined separately and are summarized below.

In a separate analysis, I evaluated macroecological evolution by coding each taxon

into biome-level categories (grassland steppe, chaparral, subtropical thom-scrub,

warm desert, or cold desert; Appendix), which I then mapped onto the bayesian and

ML phylogenetic trees (Figs. 3, 4). Because of the unresolved nature of the basal node

in the bayesian phylogeny, the ancestral condition for the family Heteromyidae was

equivocal between a warm desert and a subtropical thom-scrub origin. A warm desert

origin is indicated in Perognathus, Chaetodipus, and Dipodomyinae, whereas the

subfamily Heteromyinae is entirely subtropical thom-scmb. When the biome

categories were mapped onto the fully resolved ML tree, the family maintained its

warm desert origin. Similar to the body size analysis, each of the four groups is

examined separately (Figs. 6-11).

27

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. S. , =9*., & â.^fc:O*-î::*0'^S'®S-£EEEca> Q.? Il ûciciciciaûciciciciaciciQCicicicicicicia

■D fDC t . "Oi . I l ■a1 ï fb 2 2s. E 3 o i II

M C g

QQQQQQQQQdCiQQQQCiQQQQQQQ

. « ^ S S I m4-* "4" Q-O II

00

m (O

QQQQQQQQQQCiCiCidQCiCiCiQQCiQQ

î t l II

Fig. 6. A) Body size, B) number of toes on hind foot, and C) biome characters for

Dipodomys mapped onto the bayesian tree.

28

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a CO CL ! I o

QQQQQQQQQQQQ

O) 4 & “D " # Q. mc ■D t c g I g > s S' III

III QQQtDQQQQQQQQQQQQQQQQQQQ

s $ a p. HD

00

CO ” s « 5#è# %- *1 s - ‘ L % o> -S E c; © a < OQQÔOOOOÔOOOOOOOOOOQOOO (U O) ii rill II

Fig. 7. A) Body size, B) number of toes on hind foot, and C) biome characters for

Dipodomys mapped onto the ML tree.

29

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. g g I 1 ï I Ig si f s s I il 11f I II I f a » 3 Ü o t ) U Ü Ü 2 b! E o

L Si Si â

II2 I I .1 î i S' llgl

.§ ag I a 3 s I g 3

otjooüüoOciocitiociuO LU hU ill II CQ

H I i üOüOocjücJdoüooootj hU 3E ' T t "S E i S' IID

Fig. 8. A) Body size, B) presence of rump spines, and C) biome characters for

Chaetodipus mapped onto the bayesian tree.

30

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7J CD ■D O Q. C g Q.

■D CD

(/)

goldmani goldmani ■p C. goidmani 8 I C. artus artus artus nelsoni neisoni C. neisoni t Intermedius intermedius C. intermedius I0 eremicus eremicus C. eremicus CD 3 C. pemix 1 ■ ...... f pernix pemix penicillatus peniciiiatus C. peniciiiatus S' arenariusl arenariusl C. arenariusl arenariusl arenariusl C. arenariusl Io CD caiifomicus caiifomicus C. caiifomicus "O O i faiiax faiiax C. faiiax Û. c g. spinatus spinatus C. spinatus 3o W hispidus hispidus C. hispidus "O baiieyi baiieyi C. baiieyi o Î rudinoris rudinoris C. rudinoris formosus formosus C. formosus CD a. Q. 3

m edium sm ooth I# #1 grassland steppe I I I cold desert ■D small spiny CD 1**1 chaparral equivocal equivocal — i warm desert C/) (/) warm desert and chaparral f I I II subtropical thornscrub S’ equivocal CD ■D O Q. C 8 Q.

■D CD

C/) C/) f

CD 8 A TP P. flavescens flavesœns ^ P. apache apache “ ■ P. fasciatus fasciatus I- P. parvus -Utah parvus - Utah CD alticola CL C P. a/f/co/a 3 am p. pa/vus - Columbia Plateau parvus -Columbia Plateau 3. inornatus 3" TP P. inornatus CD P. Iongimembris Iongimembris I CD ampius ■D I I I I I I I I I I T P. amplus O Q. P. flavus flavus C a P. merriami merriami O W 3 K) ■D I S’ O M medium !• • ! grassland steppe small I I cold desert CD Q. very small I # # 1 cold desert and grassland steppe equivocal l# # l chaparral Mi warm desert ■D CD I 0 0 equivocal

C/) C/) I I ■DCD O Q. C 8 Q.

■D CD

C/) C/) 8

CD |P P. inornatus B inomatus 8 m ™ P. Iongimembris Iongimembris I P. amplus amplus i- F ° P- flavus flavus TTTr P. merriami memami P. flavescens flavesœns » 3. Irm P. apache apache 3" P. fasciatus CD fasciatus CD f P. parvus - Utah ■D parvus -Utah O C P. alticola Q. alticola C nxD p. parvus - Columbia a parvus -Columbia Plateau O Plateau 3 ! ■D O medium I # # 1 grassiandsteppe I I coid desert CD Q. Mill very small I # # 1 coid desert and grassiand steppe I I equivocal 1 **1 chaparral WHB warm desert ■D CD I equivocal

C/) C/) I i

8* CHAPTER 4

DISCUSSION

Phylogenetics o f the family Heteromyidae

In all analyses I found robust support for the monophyly of all genera as well as

for clades that support Dipodomyinae and Heteromyinae. The lack of resolution for

the basal nodes within the family Heteromyidae is most likely caused by the use of

mitochondrial genes that saturate fairly quickly (Simon et al. 1994). More

conservative nuclear genes might offer better resolution of the deeper nodes on this

tree; I am currently evaluating a suite of nuclear genes to address this problem.

Dipodomyinae

One of the most significant findings of these phylogenetic analyses is

corroboration of the position of Microdipodops within the subfamily Dipodomyinae

rather than within the Perognathinae. Based primarily on fossil dental characteristics,

Reeder (1956) suggested that Microdipodops is more closely related to Dipodomys

than any other extant taxa. He stated: “although it is not recently derived fi’om similar

stock, it is probably the remnant of the flourishing Cupidinimus-Ferognathoides

complexof the late Tertiary” (Reeder1956:416). In the extent of hypsodonty and the

pattern of cusps, the dentition o f Microdipodops is nearly identical to that of

Cupidinimus and very similar to that o f Perognathoides (Reeder 1956).

The deep divergence between Dipodomys and Microdipodops in my

34

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. data indicates that Microdipodops diverged at least 10 million years earlier than its

first appearance in the fossil record (Pleistocene). Dipodomys first appears in the

fossil record during the Barstovian age of the middle Miocene and radiated thereafter.

The phylogenetic split between the ancestors of the Dipodomys and Microdipodops

lineages must therefore have occurred no later than the early-middle Miocene. Early

Microdipodops tooth-only fossils may have been misidentified as Cupidinimus or

Perognathoides. The xerie, sandy habitat to which Microdipodops is specifically

adapted may not have assumed its current distribution until the interpluvial periods of

the Pleistocene. Microdipodops tends to occur along shorelines of pluvial lakes, but

M. megacephalus also occupies somewhat gravelly soils. If the ancestral form of

Dipodomyinae had a warm desert origin as our data suggests, it is possible that the

ancestral Microdipodops remained ecologically tied to sandy areas of the warm deserts

and tracked the sandy habitats northwards during the geological evolution of the Great

Basin. This, however, does not explain their absence fi-om sandy habitats in the

southern deserts. We know that the areas that are now warm deserts were not warm

deserts at the base of the Dipodomyinae radiation in the Miocene. The warm desert

origin for the ancestral Dipodomyinae is most likely an artifact of the way that the

Fitch parsimony algorithm processes nodes in MacClade (Maddison and Maddison

2000). Specific substrates, instead of general biomes, might be less plastic through the

duration of a lineage and therefore, might be a more informative character to map once

a well supported phylogeny is obtained.

35

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Perognathinae

These analyses corroborate the complete phylogenetic separation of Chaetodipus

and Perognathus. In fact, I have found no support for the historical inclusion of both

genera {Perognathus and Chaetodipus) in the subfamily Perognathinae. Levels of

interspecific divergence in both Chaetodipus and Perognathus within the cytochrome

b gene are considerably higher than other genera of rodents (Johns and Avise 1998).

The conservative nature of their morphology is in stark contrast to their molecular

divergence (Modi 2003). Even though paleontologists have been unable to distinguish

the two genera based on fossils, they are highly divergent lineages within the family

Heteromyidae. It would be interesting to revisit the fossil specimens now that this

generic dichotomy is robustly established and ask whether they ean be assigned to

genus with confidenee. Morphometric data presented by Hafiier (Fig. 3 in Hafiier

1978) demonstrated that Chaetodipus and Perognathus are not particularly similar in

morphometric space. Hafiier presented phenograms representing phenetic

relationships (Fig. 2 in Hafiier 1978) that place Microdipodops in a position that

implies paraphyly of the genus Perognathus-, this paraphyletie arrangement is removed

when we consider that in current taxonomy the Perognathus that Hafiier (1978)

referred to includes both Perognathus and Chaetodipus. Interestingly, if we consider

modem taxonomy, Hafiier (1978) found that Microdipodops is morphologically more

similar to Perognathus than Perognathus is to Chaetodipus. My analyses show no

evidence for a subfamily Perognathinae containing Perognathus and Chaetodipus.

36

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Morphological and macroecological evolution in the Heteromyidae

Examining morphological and macroecological evolution at the basal nodes was

problematic because of the general lack of resolution at the base of the tree. The

ancestral condition of the early heteromyids will be better evaluated after I am able to

resolve the basal relationships within the Heteromyidae. In the biome analyses it is

possible that the implied warm desert ancestry is an artifact of the way the Pitch

parsimony algorithm processes nodes in MacClade (Maddison and Maddison 2000).

If Heteromyinae were the basal subfamily, for example, subtropical thornscrub would

be the ancestral state. Some intriguing results emerged, however, fi-om examination of

each major clade.

Dipodomys

Even with an equivocal origin of Dipodomys having large or very large body size,

the body size evolution of kangaroo rats is relatively straightforward (Figs. 6, 7);

Dipodomys is, on average, medium to large-bodied (20-90 g). Very large bodies (> 90

g) have evolved independently at least twice, but remained very limited (the basal D.

deserti + D. spectabilis + D. nelsoni, and the very recent speeiation of D. ingens). The

lack of resolution at the base of the tree makes the precise number of separate events

ambiguous. Smaller kangaroo rats (20 - 60 g, “medium” sized Heteromyids as

classified in this study) appear to have evolved three or four times; the lack of

resolution at this point in the tree also makes this determination ambiguous. Two of

three clades of kangaroo rats that are joined in an unresolved trichotomy contain all

but two of the medium sized Dipodomys-, which may represent one or two separate

events. D. microps and D. simulans have very recently achieved smaller sized bodies

37

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in separate evolutionary events. D. elator is the only large-bodied member of an

otherwise medium-sized group, representing a recent attainment of increased size.

Five toes on each of the hind feet is undoubtedly the ancestral condition within

Heteromyidae. It appears from these analyses that the ancestral Dipodomys lost a toe

on each hind foot; this was followed by at least two subsequent reversals to the five

toed condition. D. compactus and D. ordii, which form a clade in the bayesian

analysis, and clade eontaining the D. heermanni + agilis speeies groups represent all

of the five toed Dipodomys.

Within Dipodomys, species groups frequently have been based on many different

morphological criteria (Best 1993; Burt 1936; Grinnell 1921; Lidicker 1960b; Setzer

1949; Wood 1935). Aeeording to Stock (1974:514) “use of indices of specialization

without attention to different habitat preferences has obscured phyletic relationships.”

We have tried to combine macroecological and body size evolution with the

phylogenetic hypotheses generated in this study. In an attempt to define the early

origins of the Dipodomys speeies groups. Stock (1974:519) stated that “The

occurrence of only 2N = 72 forms, except for D. merriami, throughout most of the

geographic range of the genus suggests that kangaroo rats may have first evolved in

the semiarid grasslands of northern Mexico and the central United States and may

have developed the evolutionary trends toward bipedal locomotion in response to

open, semiarid grassland situations rather than in response to true desert conditions as

is usually held to be the case.” He also suggests that the “ancestral condition” of

Dipodomys is one that possesses “brush-dwelling” characteristics (Stock 1974).

38

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Contrary to the idea that Dipodomys evolved in the semi-arid grasslands as

suggested by Stoek (1974), our analyses indicate that the subfamily Dipodomyinae

(along with the Chaetodipus and Perognathus) had a warm desert origin (Figs. 6-11).

This result will need further evaluation after the achievement of a more fully resolved

phylogenetic hypothesis. Within Dipodomys, D. deserti, whieh is restricted to the

warm desert, is basal to the rest of the genus. The root of the next branch, the D.

spectabilis plus D. nelsoni clade, is equivocal between a warm desert and a grassland

steppe origin because D. spectabilis tends to be found in grassland areas, whereas D.

nelsoni is found in warmer and drier desert areas. The ancestor of D. ordii and D.

compactus appears to have moved into the grassland steppes and cold desert regions of

western North America. The D. merriami speeies group is currently restricted to the

warm deserts, whereas their sister group, the phillipsii species group, is more restricted

to the eastern grasslands, but does occur in some dry aridlands. The habitat afiinity of

the ancestral form of both speeies groups {merriami plus phillipsii) remains equivocal

between a warm desert and a grassland steppe origin. With four exceptions {D.

ingens, gravipes, panamintinus, and microps), the remaining Dipodomys are restricted

to chaparral areas; it appears as though Dipodomys evolved chaparral affinities only

once with several subsequent reversals. D. ingens, gravipes, and panamintinus

reverted back to a warm desert ecological type, whereas D. microps adapted to the

cold desert areas of the Great Basin from a chaparral ancestor. It therefore appears

that different lineages of Dipodomys have adapted to warm desert habitats as many as

six times independently. The D. merriami species group and its sister clade {elator

and phiiiipsi) suggest a separation of habitat types (warm desert and grassland steppe)

39

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that may have chronologically coincided with the separation of the eastern grasslands

and the western deserts. The D. agilis species group {agilis + simulans + venustus +

elephantinus) and D. californicus are restricted to chaparral. Other than the D.

merriami+ group and the D. agilis+ group, the remaining kangaroo rats seem to

demonstrate signifieant amounts of plasticity in habitat association. The D. heermanni

group, including heermanni, panamintinus, stephensi, microps, gravipes, and ingens,

have adapted to three different biomes in as many as four different events (Fig. 6, 7).

The D. spectabilis speeies group, including only spectabilis and nelsoni, have adapted

to moderately different habitat niches. Similarly, D. ordii and D. compactus, whieh

are weakly grouped together as a species group, have adapted to two different biomes.

It is possible that the plasticity demonstrated at the tips of the Dipodomys tree suggest

that it would not be possible to identify the ancestral condition of this group even if

the phylogenetic relationships were clearly resolved. There has been repeated

reinvasion of biomes over time; biome eeology is clearly not a conservative eharacter.

Chaetodipus

Chaetodipus historieally has been split into species groups based on external

morphology, primarily the presence or absence of rump spines. Generally, this genus

was split into two species groups; those with moderately well developed rump bristles

(C. intermedins, nelsoni, fallax, artus, and goldmani), and those without defined rump

bristles (C. penicillatus [+ eremicus], pernix, and arenariusl), with the remaining

species left in monotypic species groups (C. californicus, baileyi [+ rudinoris\

hispidus,formosus, and the very bristly spinatus). C. dalquesti {arenariusl, herein)

was not separated from C. arenarius in spite of the presence of rump bristles. These

40

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. species groups have little to do with actual phylogenetic relationship. A smooth

pelage was most likely the ancestral condition (Figs. 8, 9). Rump bristles either

evolved twice (the very spiny C spinatus and C fallax, along with the moderately

bristly C. californicus and C. arenariusl at one time, and the C. goldmani group

including C. goldmani, artus, nelsoni and intermedius at a later date), or only once

with a subsequent return to the smooth pelage in the C. penicillatus group (including

C. penicillatus,pernix, eremicus, and arenariusl). Spines seem to be a relatively

good indicator of species groups and also tend to be a good indicator of specific

microhabitat; spiny Chaetodipus occur in rocky habitat, whereas smooth Chaetodipus

tend to occur in more sandy habitats.

Most of the species of Chaetodipus currently are either small or medium-bodied

(Figs. 8, 9). The ancestral Chaetodipus probably was a medium sized species; all

three members of the basal clade and the following branch currently are medium­

bodied. With the exeeption of four medium-sized species that evolved recently, but

separately (C. californicus, C. fallax, and C. goldmani + C. artus), all of the remaining

species are small-bodied.

The genus Chaetodipus appears to have had a warm desert origin (Figs. 8, 9).

Under this interpretation, C. hispidus secondarily evolved adaptations to the grassland

steppes, C. californicus and C. fallax developed chaparral tendencies, and three other

species have recently adapted to subtropical thomscrub areas in two different

evolutionary events (C goldmani plus C. artus, and C. pernix).

41

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Perognathus

The size of the ancestral Perognathus is equivocal between small (11-16 g) and

very small (8-10 g), but it seems likely that the genus had a very small-bodied

ancestral form (Figs. 10,11). If Perognathus began as a very small-bodied form, the

small forms evolved twice (P. amplus and P. fasciatus) and the medium forms evolved

once (P. alticola and P. parvus).

Perognathus historically has been split into four species groups: a longimembris

species group (longimembris, amplus, and inornatus); aflavus species group (flavus

and merriami); a fasciatus species group (fasciatus, flavescens, and apache); and a

parvus species group (parvus and alticola). Whether or not this genus had a warm

desert origin, each of the four species groups has adapted to different conditions (Figs.

10, 11). The longimembris species group has maintained its affinities for the warm

deserts of the southwest while its sister species group (flavus species group) has

adapted to the grassland steppe of the southeast. The other two species groups have

adapted to more northerly habitats. The fasciatus species group has adapted to cold

desert grassland areas of the upper Great Plains, whereas the parvus species group is

restrieted to the eold desert areas of the Great Basin (parvus) and California ehaparral

(alticola). Perognathus clearly demonstrates the ecological plasticity found within the

family Heteromyidae.

Summary

From the Oligocene to the late Miocene, subtropical thorn forests and woodland

savannas were gradually replaced with steppe and semi-desert habitats. This

continued during the latest Miocene with expansion of regional deserts, grasslands and

42

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. shrub-steppes. The early to middle Miocene experienced a tremendous radiation of

Heteromyid rodents (Korth 1994; Savage and Russell 1983). Perognathus (including

Chaetodipus) and Dipodomys were well established by the middle Miocene (Wahlert

1993). The remaining three genera (Microdipodops, Liomys, and Heteromys) have

such a poor fossil record that we are unable to identify their initial appearance from

fossils. My data suggests, however, that the four primary clades (Dipodomyinae

[including Dipodomys and Microdipodops], Perognathus, Chaetodipus, and

Heteromyinae [including Heteromys and Liomys]) all have had a similar amount of

time since lineage divergenee. The initial diversification of all of these groups

probably coincided with the middle Miocene formation and expansion of steppe and

semi-desert habitats into at least three provinces (Riddle et al. 2000a): eastern

grasslands and savannas (oecupied by Chaetodipus hispidus, and the ancestors of the

Perognathus fasciatus, Dipodomys spectabilis, and D. ordii species groups); semi-

deserts and woodlands of the Basin and Range (occupied by D. deserti,

Microdipodops, and the ancestors of the C. formosus and P. parvus species groups);

and semi-deserts and subtropical thorn-scrub in western Mexico (occupied by the

ancestors of the C. baileyi and D. merriami species groups). Subsequent

diversification events within each genus took place in response to continuing isolation

of the regional deserts throughout the late Mioeene, early Pliocene, and into the reeent

glacial-interglacial eyeles of the Pleistocene.

The subfamily Heteromyinae is the only subtropieal lineage in the family

Heteromyidae and has the most primitive morphology. Liomys is restricted to the arid

and semi-arid regions of Central and South America. This genus is apparently limited

43

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in its distribution by extreme aridity and high moisture; they have an apparent

requirement of at least 250 mm of rainfall per year, but Liomys is replaced by

Heteromys in the more mesic areas (Schmidly et al. 1993). If the subfamily

Heteromyinae is approximately as old as the Dipodomyinae, Perognathus, and

Chaetodipus as it appears in these analyses, this group of rodents could have evolved

in southern North America, moved south into Central America, and then on into South

America after the rejoining of Panama and Columbia as part of the Great American

Interchange (Webb 1985) during the Pliocene (~3 mya). They probably were

restricted to several réfugia in Central America during the glacial periods of the late

Pleistocene when climatic conditions became colder and drier. The Pleistocene

réfugia may have provided the isolation necessary for speeiation within each of these

genera.

Future Work

Two mitoehondrial gene regions were evaluated for this study. Even though they

are not independent estimates of phylogeny, they are evolving at a similar rate, and

therefore, combining the two increased the numbers of informative characters

available for analysis. These relatively rapidly evolving genes did, however, lose

phylogenetic signal at the base of the family-level and genus-level trees. Additional

loci need to be examined to resolve the relationships among genera within

Heteromyidae and among species groups within the genus Dipodomys.

The family relationships within the order Rodentia have remained largely

unresolved in spite of a significant amount of work done on this group of mammals

(including morphological, paleontological, and molecular studies); three clades above

44

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the family level, however, have been consistently supported: suborder Hystricognathi,

superfamily Geomyoidea, and superfamily Muroidea (DeBry 2003). Several nuclear

genes recently have been used as independent estimates of phylogeny in many

different groups of mammals. The nuclear gene that codes for the interphotoreceptor

retinoid-binding protein (IRBP) has heen used to study interordinal (Stanhope et al.

1992; Stanhope et al. 1998; Stanhope et al. 1996), interfamilial (DeBry and Sagel

2001; Huehon et al. 2002), and intergeneric (DeBry 2003) relationships within

mammals. Mercer and Roth (2003) have demonstrated that third base position

substitutions in IRBP behave in a good clock-like rate of evolution. These IRBP

studies have been able to resolve relationships that were previously ambiguous with

mitochondrial DNA. As more studies are completed with nuclear genes, and IRBP in

particular, they become more useful as comparative datasets for future research. The

IRBP dataset for Rodentia is extensive and now includes at least 22 genera (DeBry

and Sagel 2001).

In addition to IRBP, two other nuclear genes have been used with success at

deeper nodes in phylogenetic trees. The recombination activating gene 2 (RAG2) is

an intronless coding gene that has been used to resolve deep nodes among bats

(Teeling et al. 2002), needlefishes (Lovejoy and Collette 2001), as well as the early

placental mammal radiation (Murphy et al. 2001a). The gene that codes for cannaboid

receptor type I (CBl) is another intronless region of the nuclear DNA and has been

useful for evaluating relationships among mammal orders (Murphy et al. 2001a;

Murphy et al. 2001b). Other nuclear gene sequences have been used successfully at

45

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. deeper phylogenetic levels as well (e.g. Type I STS Markers; Koepfli and Wayne

2003).

At this time, I am proposing additional research and analyses of the relationships

within the family Heteromyidae using the IRBP, RAG2, and CBl genes of the nuclear

DNA. Because these genes have been sueeessful at resolving previously problematic

deep nodes in a wide variety of taxa, they are good candidate genes for resolving both

the currently ambiguous deep nodes within the family Heteromyidae and the

relationships of species within the genus Dipodomys. In a recent multigene analysis of

rodents, DeBry (2003) found strong support for the Geomyoidea plus Castoridae -

Pedetidae clade sister to a Muroidea plus Dipodidae clade. I will conduct the next

phase of this research in two stages. I will include representatives of 4 genera within

the family Heteromyidae (two Perognathus, two Chaetodipus, two Microdipodops,

and two Dipodomys; Liomys and Heteromys currently are being evaluated by other

researchers - Duke Rogers, pers. comm.) as well a representative Geomys

(Geomyoidea), Zapus (Dipodidae), and Peromyscus (Muroidea). After confirming

that these genes are in fact useful in resolving the deeper nodes that have been

problematic, and that Geomyoidea consists of Geomyidae plus Heteromyidae, the

second stage of this research will include all Heteromyidae species within the genera

Perognathus, Chaetodipus, Dipodomys, and Microdipodops and representatives of

Heteromys and Liomys. Geomys, Zapus, and Peromyscus will be used as outgroups.

Previous researchers have found that third position sites of IRBP can demonstrate

variation in nucleotide composition, whereas RAG2 is homogeneous (DeBry 2003;

Huchon et al. 2002). I will evaluate the phylogenetic signal of each gene and each

46

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. codon position separately as well as in combined analyses. By partitioning the data, it

will be possible to conduct MP, ML, and bayesian analyses using only the data with

useful phylogenetic signal.

47

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LITERATURE CITED

Alexander, L. F., B. R. Riddle, and D. J. Hafiier. in litt. Evolution and phylogeography

of the Dipodomys merriami (Merriam's kangaroo rat) species group: shallow

geographic structuring of an abundant and widespread mammal in the North

American warm deserts.

Baldwin, E. M. 1964. Geology of Oregon. University of Oregon, Eugene.

Best, T. L. 1988. Dipodomys nelsoni. Mammalian Species, 326:1-4.

Best, T. L. 1993. Patterns of morphologic and morphometric variation in heteromyid

rodents. Pp. 197-235 in H. H. Genoways and J. H. Brown, eds. Biology of the

Heteromyidae. American Society of Mammalogists.

Best, T. L. and J. A. Lackey. 1985. Dipodomys gravipes. Mammalian Species, 236:1-

4.

Best, T. L. and H. H. Thomas. 1991. Dipodomys insularis. Mammalian Species,

374:1-3.

Best, T. L., R. K. Chesser, D. A. McCullough, and G. D. Baumgardner. 1996. Genic

and morphometric variation in kangaroo rats, (genus Dipodomys), from coastal

California. Journal of Mammalogy 77:785-800.

Best, T. L., and G. D. Schnell. 1974. Bacular variation in kangaroo rats (genus

Dipodomys). The American Midland Naturalist 91:257-270.

48

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Blair, W. F. 1954. Mammals of the mesquite plains biotic district in Texas and

Oklahoma, and speeiation in the central grasslands. Texas Journal of Science

6:235-264.

Burt, W. H. 1936. A study of the baculum in the genera Perognathus and Dipodomys.

Journal of Mammalogy 17:145-156.

Davis, W. B. 1942. The systematic status of four kangaroo rats. Journal of

Mammalogy 23:328-333.

DeBry, R. W. 2003. Identifying signal conflict in a multigene analysis reveals a highly

resolved tree: the phylogeny of Rodentia (Mammalia). Systematic Biology

52:604-617.

DeBry, R. W., and R. M. Sagel. 2001. Phylogeny of Rodentia (Mammalia) inferred

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

19:290-301.

Fiero, B. 1986. Geology of the Great Basin. University of Nevada Press, Reno.

Genoways, H. H. 1973. Systematics and evolutionary relationships of spiny pocket

mice, genus Liomys. The Museum, Texas Tech University

Goldman, N., J. P. Anderson, and A. G. Rodrigo. 2000. Likelihood-based tests of

topologies in phylogenetics. Systematic Biology 49:652-670.

Good, S. V., D. F. Williams, K. Ralls, and R. C. Fleischer. 1997. Population structure

0 Ï Dipodomys ingens (Heteromyidae): the role of spatial heterogeneity in

maintaining genetic diversity. Evolution 51:1296-1310.

Grinnell, J. 1921. Revised list of the species in the genus Dipodomys. Journal of

Mammalogy 2:94-97.

49

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hafiier, J. C. 1978. Evolutionary relationships of kangaroo mice, genus

Microdipodops. Journal of Mammalogy 59:354-366.

Hafiier, J. C. 1993. Macroevolutionary diversification in Heteromyid rodents:

heterochrony and adaptation in phylogeny in H. H. Genoways and J. H.

Brown, eds. Biology of the Heteromyidae. American Society of

Mammalogists.

Hafiier, J. C., and M. S. Hafiier. 1983. Evolutionary relationships of Heteromyid

rodents. Great Basin Naturalist Memoirs 7:3-29.

Hall, E. R. 1981. The Mammals of North America. Wiley, New York.

Hall, T. A. 1999. BioEdit: a user-fiiendly biological sequence alignment editor and

analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series

41:95-98.

Hamilton, M. J., R. K. Chesser, and T. L. Best. 1987. Genetic variation in the Texas

kangaroo rat, Dipodomys elator Merriam. Journal of Mammalogy 68:775-781.

Hatch, F. T., and J. A. Mazrimas. 1977. Satellite DNA and cytogenetie evolution:

molecular aspects and implications for man. Human Cytogenetics, ICU-UCLA

Symposium, Molecular Cell. Biol. 7:395-414.

Hofifineister, D. F. 1986. Mammals of Arizona. The University of Arizona Press and

The Arizona Game and Fish Department, Tucson.

Huchon, D., O. Madsen, M. J. J. B. Sibbald, K. Ament, M. J. Stanhope, F. M.

Catzeflis, W. W. de Jong, and E. J. P. Douzery. 2002. Rodent phylogeny and a

timescale for the evolution of Glires: evidence from an extensive taxon

50

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sampling using three nuclear genes. Molecular Biology and Evolution

19:1053-1065.

Huelsenbeck, J. P. 2000. MrBayes: Bayesian inference of phylogeny. Distributed by

the author. Department of Biology, University of Rochester.

Johns, G. C., and J. C. Avise. 1998. A comparative summary of genetic distances in

the vertebrates from the cytochrome b gene. Molecular Biology and Evolution

15:1481-1490.

Johnson, W. E., and R. K. Selander. 1971. Protein variation and systematics in

kangaroo rats (genus Dipodomys). Systematic Zoology 20:377-405.

Jones, W. T. 1985. Body size and life-history variables in heteromyids. Journal of

Mammalogy, 66:128-132.

Jones, W. T. 1988. Density-related changes in survival of philopatric and dispersing

kangaroo rats. Ecology 69:1474-1478.

Jones, W. T., P. M. Waser, L. F. Elliott, N. E. Link, and B. B. Bush. 1988. Philopatry,

dispersal, and habitat saturation in the banner-tailed kangaroo rat, Dipodomys

spectabilis. Ecology 69:1466-1473.

Koepfli, K. and R. K. Wayne. 2003. Type I STS Markers are more informative than

cytochrome b in phylogenetic reconstruction of the Mustelidae (Mammalia:

Carnivora). Systematic Biology 52(5):571-593.

Korth, W. W. 1994. The tertiary record of rodents in North America. Plenum Press,

New York.

Lee, T. E., B. R. Riddle, and P. L. Lee. 1996. Speeiation in the desert pocket mouse

(Chaetodipus penicillatus Woodhouse). Journal of Mammalogy 77:58-68.

51

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Levin, H. L. 1978. The earth through time. W. B. Saunders Company, Philadelphia.

Lidicker, W. Z., Jr. 1960a. An analysis of intraspecific variation in the kangaroo rat

Dipodomys merriami. University of California Publications in Zoology

67:125-218.

Lidicker, W. Z., Jr. 1960b. The baculum of Dipodomys ornatus and its implication for

superspecific groupings of kangaroo rats. Journal of Mammalogy 41:495-499.

Lovejoy, N. R., and B. B. Collette. 2001. Phylogenetic relationships of New World

needlefishes (Teleostei: Belonidae) and the biogeography of transitions

between marine and fi-eshwater habitats. Copeia 2001:324-338.

Maddison, D. R., and W. P. Maddison. 2000. MacClade 4: Analysis of phylogeny and

eharacter evolution. Sinauer Assoeiates, Sunderland, Massachusetts.

Mazrimas, J. A., and F. T. Hatch. 1972. A possible relationship between satellite DNA

and the evolution of kangaroo rat species (genus Dipodomys). Nature New

Biology 240:102-105.

Mercer, J. M., and V. L. Roth. 2003. The effects of Cenozoic global change on

squirrel phylogeny. Science 299:1568-1572.

Modi, W. S. 2003. Morphological, chromosomal, and molecular evolution are

uncoupled in pocket mice. Cytogenetic and Genome Research 103:150-154.

Montgelard, C., S. Bentz, C. Tirard, O. Vemeau, and F. M. Catzeflis. 2002. Molecular

systematics of Sciurognathi (Rodentia): The mitochondrial cytochrome b and

12S rRNA genes support the Anomaluroidea (Pedetidae and Anomaluridae).

Molecular Phylogenetics and Evolution 22:220-233.

52

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Murphy, W. J., E. Eizirik, W. E. Johnson, Y. P. Zhang, O. A. Ryder, and S. J. O'Brien.

2001a. Molecular phylogenetics and the origins of placental mammals. Nature

1409:614-618.

Murphy, W. J., E. Eizirik, S. J. O'Brien, O. Madsen, M. Scally, J. Douady, E. C.

Teeling, O. A. Ryder, M. J. Stanhope, W. W. de Jong, and M. S. Springer.

2001b. Resolution of the early placental mammal radiation using Bayesian

phylogenetics. Science 294:2348-2351.

Nader, I. A. 1966. Roots of teeth as a generic character in the kangaroo rats,

Dipodomys. Bulletin of the Biological Research Centre (Baghdad) 2:62-69.

Norris, R. M., and R. W. Webb. 1976. Geology of California. John Wiley and Sons,

New York,

Ortega-Gutiérrez, F., and J. C. Guerrero-Garcia. 1982. The geologic regions of

Mexico. Pp. 99-104 in A. R. Palmer (ed.). Perspectives in regional geological

synthesis. D-NAG Special Publication 1, Geological Society of America.

Patton, J. L. 1993. Family Heteromyidae. Pp. 477-486 in D. E. Wilson and D. M.

Reeder, eds. Mammal species of the World. Smithsonian Institution Press,

Washington.

Patton, J. L., and D. S. Rogers. 1993. Biochemical genetics. Pp. 259-269 in H. H.

Genoways and J. H. Brown, eds. Biology of the Heteromyidae. American

Society of Mammalogists.

Patton, J. L., S. W. Sherwood, and S. Y. Yang. 1981. Biochemical systematics of

Chaetodipine pocket mice. Genus Perognathus. Journal of Mammalogy

62:477-492.

53

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Posada, D., and K. A. Crandall. 1998. Modeltest: testing the model of DNA

substitution. Bioinformatics 14:817-818.

Randall, J. A. 1994. Discrimination of footdrumming signatures by kangaroo rats,

Dipodomys spectabilis. Animal Behavior 47:45-54.

Randall, J. A. 1995. Modification of footdrumming signatures by kangaroo rats:

changing territories and gaining new neighbours. Animal Behavior 49:1227-

1237.

Reeder, W. G. 1956. A review of Tertiary rodents of the family Heteromyidae. Pp.

xxii + 618 pp. Unpublished Ph.D. dissertation. University of Michigan, Ann

Arbor.

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. Hafiier, and L. F. Alexander. 2000a. Comparative Phylogeography

of Baileys' pocket mouse {Chaetodipus baileyi) and the Peromyscus eremicus

Species-group: Historical Vicariance of the Peninsular and Continental

Deserts. Molecular Phylogenetics and Evolution 17:161-172.

Riddle, B. R., D. J. Hafiier, L. F. Alexander, and J. Jaeger. 2000b. Cryptic vicariance

in the historical assembly of a Baja California Peninsular Desert biota.

Proceedings of the National Academy of Sciences 97:14438-14443.

54

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rogers, D. S. 1990. Genic evolution, historical biogeography, and systematie

relationships among spiny poeket miee (subfamily Heteromyinae). Journal of

Mammalogy 71:668-685.

Roth, E. L. 1976. A new speeies of pocket mouse (Perognathus: Heteromyidae) from

the Cape Region of Baja California Sur, Mexico. Journal of Mammalogy

57:562-566.

Ryan, J. M. 1989. Comparative myology and phylogenetic systematics of the

Heteromyidae (Mammalia, Rodentia). Miscellaneous Publications, Museum of

Zoology, University of Michigan 176:1-103.

Savage, D. E., and D. E. Russell. 1983. Mammalian Paleofaunas of the World.

Addison-Wesley

Sehmidly, D. J., K. T. Wilkins, and J. N. Derr. 1993. Biogeography. Pp. 319-356 in H.

H. Genoways and J. H. Brown, eds. Biology of the Heteromyidae. American

Society of Mammalogists.

Schnell, G. D., T. L. Best, and M. L. Kennedy. 1978. Interspecific morphologic

variation in kangaroo rats {Dipodomys): degree of eoneordance with genic

variation. Systematic Zoology 27:34-48.

Setzer, H. W. 1949. Subspeciation in the kangaroo rat, Dipodomys ordii. University of

Kansas Publications, Museum of Natural History 1:473-573.

Shimodaira, H., and M. Hasegawa. 1999. Multiple comparisons of log-likelihoods

with applications to phylogenetic inference. Molecular Biology and Evolution

16:1114-1116.

55

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Simon, C., F. Frati, A. Beckenbach, B. Crespi, H. Lin, and P. Flook. 1994. Evolution,

weighting and phylogenetic utility of mitochondrial gene sequences and a

compilation of conserved polymerase chain reaction primers. Annals of the

Entomological Society of America 87:651-701.

Smith, M. F. and J. L. Patton. 1993. The diversification of South American murid

rodents: evidence fi"om mitochondrial sequence data for the akodontine tribe.

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

Stanhope, M. J., J. Czelusniak, J.-S. Si, J. Nickerson, and M. Goodman. 1992. A

molecular perspective on mammalian evolution fi-om the gene encoding

interphotoreceptor binding protein, with convincing evidence for bat

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

Stanhope, M. J., M. R. Smith, V. G. Waddell, C. A. Porter, M. S. Shivji, and M.

Goodman. 1996. Mammalian evolution and the interphotoreceptor binding

protein (IRBP) gene: Convincing evidence for several superordinal elades.

Journal of Molecular Evolution 43:83-92.

Stanhope, M. J., O. Madsen, V. G. Waddell, G. C. Cleven, W. W. de Jong, and M. S.

Springer. 1998. Highly congruent molecular support for a diverse superordinal

clade of endemic Afirican mammals. Molecular Phylogenetics and Evolution

9:501-508.

Stock, A. D. 1974. Chromosome evolution in the genus Dipodomys and its taxonomic

and phylogenetic implications. Journal o f Mammalogy 55:505-526.

56

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Swanson, E. R., and F. W. McDowell. 1984. Calderas of the Sierra Madré Occidental

volcanic field, western Mexico. Journal of Geophysical Research 89: 8787-

8799.

Swofford, D. L. 1999. PAUP* 4.0 beta version. Sinauer, Sunderland, MA.

Taberlet, P., A. Meyer, and J. Bouvet. 1992. Unusual mitochondrial DNA

polymorphism in two local populations of blue tit {Parus caerulens).

Molecular Ecology 1:27-36.

Tamura, K., and M. Nei. 1993. Estimation of the number of nucleotide substitutions in

the control region of mitochondrial DNA in humans and chimpanzees.

Molecular Biology and Evolution 10:512-526.

Teeling, E. C., O. Madsen, R. A. Van den Bussche, W. W. de Jong, M. J. Stanhope,

and M. S. Springer. 2002. Microbat paraphyly and the convergent evolution of

a key innovation in Old World rhinolophoid microhats. Proceedings of the

National Academy of Sciences 99:1431 -1436.

Wahlert, J. H. 1985. Skull morphology and relationships of geomyoid rodents.

American Museum Novitates 2812:1 -20.

Wahlert, J. H. 1993. The fossil record. Pp. 1-37 in H. H. Genoways and J. H. Brown,

eds. Biology of the Heteromyidae. American Society of Mammalogists.

Waser, P. M., and L. F. Elliott. 1991. Dispersal and genetic structure in kangaroo rats.

Evolution 45:935-943.

Waser, P. M., and W. T. Jones. 1991. Survival and reproductive effort in banner-tailed

kangaroo rats. Ecology 72:771-777.

57

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Webb, S. D. 1985. Late Cenozoic mammal dispersals between the Americas. Pp. 357-

386 in e. F. G. Stehli and S. D. Webb, ed. The great American biotic

interchange. Plenum Press, New York, 532 pp.

Williams, D. F. 1978. Karyological affinities of the species groups of silky pocket

mice (Rodentia, Heteromyidae). Journal of Mammalogy 59:599-612.

Williams, D. F., H. H. Genoways, and J. K. Braun. 1993. Taxonomy. Pp. 38-196 in H.

H. Genoways and J. H. Brown, eds. Biology of the Heteromyidae. The

American Society of Mammalogists.

Wilson, R. W. 1960. Early Miocene rodents and insectivores from northeastern

Colorado. The University of Kansas, Paleontological Contributions,

Vertebrata 7:1-92.

Wilson, D. E. and S. Ruff. 1999. The Smithsonian book of North American mammals.

Smithsonian Institution Press, Washington and London.

Wood, A. E. 1935. Evolution and relationship of the heteromyid rodents with new

forms from the Tertiary of western North America. Annals of the Carnegie

Museum 24:73-262.

Xia, X. 1998. The rate heterogeneity of nonsynonymous substitutions in mammalian

mitochondrial genes. Molecular Biology and Evolution 15(3):336-344.

58

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX

59

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■OCD O Q. C g Q.

■O CD

C/) C/)

Appendix - Catalog reference numbers; taxa; biome classification (some taxa occur in more than 1 biome category and are indicated with CD both); body size classification, abbreviated as follows: very small (VS), small (S), medium (M), large (L), and very large (VL); bristle 8 status, abbreviated as follows: smooth (SM), spiny (SP), and very spiny (VSP); localities; and GenBank accession numbers for ci' (c o m , cyt-b) specimens included in these analyses.

Specimen Taxon Biome Size Bristles C. arenarius 1 warm desert SSM CD LVT 2128 LVT 3645 C. arenariusl warm desert S SP LVT 1270 C. artus subtropical thomscrub M SP 3. C. baileyi warm desert MSM 3" LVT 1210 CD LVT 3682 C. californicus chaparral M SM ■oCD LVT 1160 C. eremicus warm desert S SM O LVT 3741 C. fallax chaparral, warm desert M SP Q. C LVT 987 C. formosus cold desert M SM a LVT 1264 C. goldmani subtropical thomscrub M SP o 3 o LVT 1099 C. hispidus grassland-steppe MSM "O LVT 1063 C. intermedius warm desert S SP o LVT 1075 C. nelsoni warm desert S SP LVT 1016 C. penicillatus warm desert SSM CD Q. LVT 1267 C. pernix subtropical thomscrub SSM LVT 2114 C. rudinoris warm desert MSM LVT 2119 C. spinatus warm desert S SP LVT 2035 / TLB 11237 D. agilis chaparral LSM ■ o CD LVT 2037 / TLB 10357 D. californicus chaparral L SM LVT 2060 / TLB 11476 D. compactus grassland-steppe M SM C /i C /i LVT 2083 D. deserti warm desert VL SM LVT 2059 / TLB 10566 D. elator grassland-steppe LSM LVT 2047 / TLB 10068 D. elephantinus chaparral LSM LVT 2048 / TLB 11048 D. gravipes warm desert LSM LVT 2041 / TLB 10045 D. heermanni chaparral LSM LVT 2057 / TLB 10825 D. ingens warm desert VL SM LVT 2043 / TLB 11018 D. insularis warm desert MSM LVT 2042/NK 2158 D. margaritae warm desert MSM CD ■ D O Q. C 8 Q.

■D CD

WC/) 3o"

Appendix - Continued

8 Specimen Taxon Biome Size Bristles LVT 1023 D. merriami warm desert M SM c i' LVT 3610 D. merriami warm desert M SM LVT 4935 D. microps cold desert MSM LVT 1107 D. nelsoni warm desert VL SM LVT 2045 / TLB 10241 D. nitratoides warm desert MSM LVT 1108 D. ordii grassland-steppe M SM 3 LVT 4672 D. panamintinus warm desert L SM 3" CD LVT 2056/NK 16072 D. phillipsii grassland-steppe M SM

CD LVT 2036 / TLB 10993 D. simulans warm desert, chaparral M SM ■ D O LVT 2470 D. spectabilis grassland-steppe VL SM Q. LVT 2061 / TLB 10775 D. stephensi chaparral LSM C a LVT 2046 / TLB 10292 D. venustus chaparral LSM O 3 LVT 5499 / FN31848 (ROM) H. desmarestianus subtropical thomscrub LVSP ■D LVT 2064/NK 7500 L. irroratus subtropical thomscmb MVSP O LVT 1253 L. pictus subtropical thomscrub MVSP LVT 5155 M. megacephalus cold desert S SM CD Q. LVT 1573 M. pallidus cold desert S SM MVZ 197329 P. alticola chaparral MSM LVT 403 P. amplus warm desert S SM LVT 3703 P. apache cold desert, grassland-steppe VS SM ■D P. fasciatus cold desert, grassland-steppe S SM CD LVT 2525 LVT 2527 P. flavescens cold desert, grassland-steppe VS SM C/) LVT 702 P. flavus grassland-steppe VS SM C/) LVT 601 / MSB-553 P. inornatus warm desert vs SM LVT 2191 P. longimembris warm desert vs SM LVT 603 P. merriami grassland-steppe vs SM LVT 1816 P. parvus cold desert MSM LVT 1920 P. parvus cold desert vs SM LVT 5500 / DLM 003 G. breviceps CD ■ D O Q. C 8 Q.

■D CD

C/) C/) Appendix - Continued

Specimen Locality GenBank # LVT 2128 Mexico: Baja California; 7 mi S, 7 mi E San Felipe AY009265, XXXXXXXX 8 LVT 3645 Mexico: Baja California; 11 km S Todos Santos AY010230, XXXXXXXX LVT 1270 Mexico: Sinaloa; Presa Hidalgo, 10 km NE El Fuerte AY009260, XXXXXXXX ci' LVT 1210 Mexico: Sonora; 2 km N Puerto de la Libertad AY009310, XXXXXXXX LVT 3682 Mexico: Baja California; 2 mi SW Laguna Hanson AY009259, XXXXXXXX LVT 1160 Mexico: Coahuila; 1 mi SE Hundido AY009256, XXXXXXXX LVT 3741 Mexico: Baja California; Misiôn San Fernando AY009262, XXXXXXXX LVT 987 California: Riverside County; 9 mi W, 1 mi S Quien Sabe Point XXXXXXXX, XXXXXXXX 3 LVT 1264 Mexico: Sonora; 4 km N Navajoa AY009261, XXXXXXXX 3" CD LVT 1099 Mexico: Durango; 7 mi NNW La Zarca AY009254, XXXXXXXX LVT 1063 Mexico: Chihuahua; 5 km NNW Chihuahua AY009264, XXXXXXXX CD ■D Mexico: Chihuahua; 3 mi NE Parral AY009255, XXXXXXXX O LVT 1075 Q. LVT 1016 California: Imperial County; 1.5 mi S, 6.5 mi W Glamis AY009257, XXXXXXXX C a LVT 1267 Mexico: Sonora; 4 km N Navajoa AY009258, XXXXXXXX O 3 S LVT 2114 Mexico: Baja California; 7 mi S, 7 mi E San Felipe AY009267, XXXXXXXX ■D O LVT 2119 Mexico: Baja California; 7 mi S, 7 mi E San Felipe AY009263, XXXXXXXX LVT 2035 / TLB 11237 California: Los Angeles County; near Wrightwood XXXXXXXX, XXXXXXXX LVT 2037 / TLB 10357 California: Tehama County; 6 mi NE Dales XXXXXXXX, XXXXXXXX CD Q. LVT 2060 / TLB 11476 Texas: Kleberg County; Padre Island National Seashore XXXXXXXX, XXXXXXXX LVT 2083 Nevada: Clark County; St. Thomas Gap XXXXXXXX, XXXXXXXX LVT 2059 / TLB 10566 Texas: Wilbarger County; 2 mi W Harrold XXXXXXXX, XXXXXXXX ■D LVT 2047 / TLB 10068 California: San Benito County; 1 mi N Pinnacles XXXXXXXX, XXXXXXXX CD LVT 2048 / TLB 11048 Mexico: Baja California; 12 km NE El Rosario XXXXXXXX, XXXXXXXX

C/) LVT 2041 / TLB 10045 California: San Benito County; 1 mi N Pinnacles XXXXXXXX, XXXXXXXX C/) LVT 2057 / TLB 10825 California: Kern County; Naval Petroleum Reserve No. 2 XXXXXXXX, XXXXXXXX LVT 2043 / TLB 11018 Mexico: Baja California Sur; Isla San José XXXXXXXX, XXXXXXXX LVT 2042/NK 2158 Mexico: Baja California Sur; Isla Santa Margarita XXXXXXXX, XXXXXXXX LVT 1023 California: San Bemadino County; Kelso Dunes XXXXXXXX, XXXXXXXX LVT 3610 Mexico: Baja California Sur; 30 km N Todos Santos XXXXXXXX, XXXXXXXX LVT 4935 Nevada: Nye County; 9 mi N Beatty, Oasis Valley XXXXXXXX, XXXXXXXX LVT 1107 Mexico: Durango; 7 mi NNW La Zarca AY009253, XXXXXXXX CD ■ D O Q. C 8 Q.

■D CD

C/) C/) Appendix - Continued

Specimen Locality GenBank # LVT 2045 / TLB 10241 California: Kern County; 6 mi W Buttonwillow XXXXXXXX, XXXXXXXX CD 8 LVT 1108 Mexico: Durango; 7 mi NNW La Zarca XXXXXXXX, XXXXXXXX LVT 4672 California: San Bemadino County; 9 mi NNE Johannesburg XXXXXXXX, XXXXXXXX LVT 2056/NK 16072 Mexico: San Luis Potosi; Las Cabras, 4.6 mi NW Bledos XXXXXXXX, XXXXXXXX LVT 2036 / TLB 10993 Mexico: Baja California; 3 km SE Colonia Vicente Guerrero XXXXXXXX, XXXXXXXX LVT 2470 New Mexico: Socorro County; San Mateo Mtns., Nogal Canyon XXXXXXXX, XXXXXXXX CD LVT 2061 / TLB 10775 California: Riverside County; 2 mi W, 2 mi S Moreno XXXXXXXX, XXXXXXXX LVT 2046 / TLB 10292 California: Monterey County; 14 mi SE Carmel Valley XXXXXXXX, XXXXXXXX 3. LVT 5499 / FN31848 (ROM) Guatemala: El Peten; Tikal XXXXXXXX, XXXXXXXX 3" CD LVT 2064/NK 7500 Mexico: Zacatecas; 5 mi N, 7.5 mi E Villa de Cos XXXXXXXX, XXXXXXXX LVT 1253 Mexico: Sonora; 10 km SSE Alamos XXXXXXXX, XXXXXXXX CD ■D LVT 5155 Nevada: Lincoln County; 6 mi N, 31 mi W Hiko XXXXXXXX, XXXXXXXX O Q. LVT 1573 Nevada: Lincoln County; 7 mi N, 6.45 mi W Tempiute XXXXXXXX, XXXXXXXX C a LVT 403 Arizona: Pima County; 0.5 mi N Organ Pipe Cactus National Monument XXXXXXXX, XXXXXXXX O 3 S LVT 3703 New Mexico: Socorro County; Rio Salada XXXXXXXX, XXXXXXXX ■D LVT 2525 Wyoming: Carbon County; 10 mi S Seminoe XXXXXXXX, XXXXXXXX O LVT 2527 Nebraska: Sheridan County; 27 mi N Lakeside XXXXXXXX, XXXXXXXX LVT 702 Arizona: Navajo County; 3 mi S Keyenta XXXXXXXX XXXXXXXX CD Q. LVT 601 / MSB-553 California: Madera County XXXXXXXX, XXXXXXXX LVT 2191 Mexico: Baja California; 27 km S Punta Prieta XXXXXXXX XXXXXXXX LVT 603 Texas: Val Verde County XXXXXXXX XXXXXXXX LVT 1816 Utah: Wayne County; 9 mi S, 2 mi W Hanksville XXXXXXXX XXXXXXXX ■ D CD LVT 1920 Washington: Adams County; 4 mi S, 6 mi E Ritzville XXXXXXXX XXXXXXXX

C/) MVZ 197329 California: Kern County; Cameron Creek, Tehachapi Mtns. XXXXXXXX, XXXXXXXX C/) LVT 5500 / DLM 003 Arkansas: Little River County; 3 mi. NW Alleene ______XXXXXXXX, XXXXXXXX VITA

Graduate College University of Nevada, Las Vegas

Lois Fay Alexander

Home Address; 659 Del Prado Drive Boulder City, NV 89005

Degrees: Bachelor of Science, Wildlife Science, 1990 Oregon State University, Corvallis

Master of Science, Wildlife Science, 1994 Oregon State University, Corvallis

Special Honors and Awards: 2003 Travel Award, BIOS, Dept. Biological Sciences, University of Nevada, Las Vegas, $200.00

2002 Travel Award, BIOS, Dept. Biological Sciences, University of Nevada, Las Vegas, $200.00

2001 Juanita Greer White Travel Award, Dept. Biological Sciences, University of Nevada, Las Vegas, $350.00

2001 Graduate Research Training Assistantship, University of Nevada, Las Vegas, $3267.00

2000 Elizabeth Homer Award, American Society of Mammalogists, $200.00

2000 Grant-in-Aid of Research, American Society of Mammalogists, $ 1000.00

2000 Graduate Student Association Research Grant, University of Nevada, Las Vegas, $500.00

1999 American Museum Natural History, Theodore Roosevelt Fund Grant-in-Aid of research, $ 1000.00

64

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1999 Graduate Student Association Research Grant, University of Nevada, Las Vegas, $500.00

1998 Grant-in-Aid of Research, American Society of Mammalogists, $ 1000.00

1998 Grant-in-Aid of Research, Sigma Xi, $700.00

1998 Graduate Summer Session Scholarship, University of Nevada, Las Vegas, $2000.00

1998 Graduate Student Association Research Grant, University of Nevada, Las Vegas, $500.

1998 Graduate Student Association Research Grant, University of Nevada, Las Vegas, $400.00

1998 Graduate Research Training Assistantship, University of Nevada, Las Vegas, $1500.00

1997 Outstanding Woman in Science, Women in Science and Engineering, University of Nevada, Las Vegas

1997 Graduate Student Association Research Grant, University of Nevada, Las Vegas, $495.00

1996 Proposal Writing Seminar Award, University of Nevada, Las Vegas, $250.00

1996 Graduate Student Association Research Grant, University of Nevada, Las Vegas, $380.00

1993 Oregon State University, Merit Award, $3500.00

1993 Henry Mastin Fund Award for Graduate Research, Dept. Fish. & Wildlife, Oregon State University, $300.00

1992 Outstanding Teaching Assistant Award, Dept. Fish. & Wildlife, Oregon State University

Publications: Alexander, L. F., and B. R. Riddle, in review. Phylogenetics of and trait evolution within the family Heteromyidae. Journal of Mammalogy

Alexander, L. F., B. R. Riddle, and D. J. Hafiier. in review. Evolution and phylogeography of the Dipodomys merriami (Merriam's kangaroo rat) species 65

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. group: shallow geographic structuring of an abundant and widespread mammal in the North American warm deserts.

Riddle, Brett R., David J. Hafiier, Lois F. Alexander, and Jef Jaeger. 2000. Cryptic vicariance in the historical assembly of a Baja California Peninsular Desert biota. Proceedings of the National Academy of Sciences, 97: 14438- 14443.

Riddle, Brett R., David J. Hafiier, and Lois 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.

Riddle, Brett R., David J. Hafiier, and Lois F. Alexander. 2000. Comparative Phylogeography of Baileys’ pocket mouse (Chaetodipus baileyi) and the Peromyscus eremicus Species-group: Historical Vicariance of the Peninsular and Continental Deserts. Molecular Phylogenetics and Evolution, 17:161-172.

Alexander, Lois F. 1999. Sorex bairdi. Pg. 18 in: The Smithsonian Book of North American Mammals. D. Wilson and S. Rufif (eds.). Smithsonian Institution Press. Washington and London in association with the American Society of Mammalogists.

Alexander, Lois F. 1999. Clethrionomys californicus. Pgs. 612-613 m: The Smithsonian Book of North American Mammals. D. Wilson and S. RuIF (eds.). Smithsonian Institution Press. Washington and London in association with the American Society of Mammalogists.

Alexander, Lois F. 1996. A morphometric analysis of geographic variation within Sorex monticolus. Occasional Papers, Museum of Natural History, University of Kansas, No. 88.

Alexander, Lois F., B.J. Verts, and T.P. Ferrell. 1994 [1995]. Diet of ringtails (Bassariscus astutus) in Oregon. Northwestern Naturalist, 75:97-101.

Carraway, Leslie N., Eric Yensen, B.J. Verts, and Lois F. Alexander. 1993 [1994]. Range extension and habitat of Peromyscus truei in eastern Oregon. Northwestern Naturalist, 74:81-84.

Carraway, Leslie N., Lois F. Alexander, and B.J. Verts. 1993. Scapanus townsendii. Mammalian Species, 434:1-7.

Alexander, Lois F., and B.J. Verts. 1992. Clethrionomys californicus. Mammalian Species, 406:1-6.

66

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Professional Presentations: Alexander, L.F., and B.R. Riddle. Phylogenetics and trends o f evolutionary history within the family Heteromyidae. 82*^*^ Annual Meeting, American Society of Mammalogists, Lubbock, Texas, June 2003 [oral].

Alexander, L.F., B.R. Riddle, and D.J. Hafiier. Phylogeography o f Dipodomys merriami, including a nested clade analysis o f population histories. Inaugural meeting - International Biogeography Society, Mesquite, Nevada, 6 January 2003 [poster].

Alexander, L.F., B.R. Riddle, and D.J. Hafiier. Phylogeography o f Dipodomys merriami, including a nested clade analysis o f population histories. 82"** Annual Meeting, American Society of Mammalogists, Lake Charles, Louisiana, 18 June 2002 [poster].

Alexander, Lois F., and Douglas J. Merkler. Representative native mammals of Nevada. Desert Living - Inside Out; Community Educational Event, North Las Vegas, Nevada, 20 October 2001 [poster].

Alexander, L.F., B.R. Riddle, and D.J. Hafiier. Phylogeography o f the Dipodomys merriami species complex. 81^ Annual Meeting, American Society of Mammalogists, Missoula, MT, June 2001 [oral].

Alexander, L.F., B.R. Riddle, and D.J. Hafiier. Phylogeography o f the Dipodomys merriami species complex. UNLV Graduate Student Research Forum, Las Vegas, Nevada, 24 March 2001 [oral].

Alexander, L.F., B.R. Riddle, T.L. Best, and D.J. Hafiier. Phylogeny and biogeography o f the Dipodomys spectabilis species-group. Annual 19^ Meeting, American Society of Mammalogists, Seattle, WA, June 1999 [oral].

Riddle, B.R., D.J. Hafiier, and L.F. Alexander. Phylogeography o f the Peromyscus eremicus species-group: cryptic lineages, paraphyletic species, and deep- history biogeography in North American deserts. 79* Annual Meeting, American Society of Mammalogists, Seattle, WA, June 1999 [oral].

Dissertation Title: Evolutionary and biogeographic histories in a North American rodent family (Heteromyidae).

Dissertation Examination Committee: Chairperson, Dr. Brett R. Riddle, Ph. D. Committee Member, Dr. Daniel Thompson, Ph. D. Committee Member, Dr. Peter Starkweather, Ph. D. Committee Member, Dr. David J. Hafiier, Ph. D. Graduate Faculty Representative, D. Steven Rowland, Ph. D.

67

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.