You are on page 1of 95

DIETS OF FREE-RANGING MEXICAN GRAY WOLVES IN ARIZONA AND NEW MEXICO by JANET E. REED. B.S.

A THESIS IN WILDLIFE SCIENCE Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE Approved

Chairperson of the Committee

Accepjpd

Dean of the Graduate School


May, 2004

ACKNOWLEDGEMENTS

I thank the United States Fish and Wildlife Service (USFWS) for fiinding this project I appreciated the direction of Mexican wolf recovery leader and committee member, Brian T Kelly, the assistance of Mexican wolf Interagency Field Team personnel (i e , USFWS, Arizona Game and Fish Department, New Mexico Department of Game and Fish, USD A APHIS Wildlife Services, and U S Forest Service [USFS]) and all the volunteers who coUected scats Thank you, David R Parsons and Wendy Brown, for mentoring and believing in me I thank Dr Warren B Ballard, my major advisor, for providing encouragement and guidance throughout this study Special thanks go to committee member Dr Robert J Baker for welcoming me into his genetics lab and to Celine C Perchellet, Adam Brown, and Drs Jeff Wickliffe and Federico Hoffman for their dedication, assistance, and humor in the DNA lab I also thank committee members Drs Philip S Gipson, Paul R. Krausman, and Mark C Wallace for their ad\ice and support A special thank you to Dr David B Wester for facilitating the statistical analyses I thank the USFS Alpine and Clifton Ranger Districts for their assistance on the forest A special thank you to Myron C Burnett, USFS Wilderness Ranger, for rising above the opposition to teach me to pack burros and believing that I could handle the job I also thank the residents of Alpine for welcoming me into their community and all the forest visitors who assisted with scat collecting Thank you Waypoint Enterprises, Show Low, Arizona, for your donation of UTM and Multi Waypointers

My parents, Roger C and Betty A Stone, my sister, Anita D. Moore, and my fnend, Alan B Alley, provided unwavering support and encouragement throughout my education. Thanks to my two burros, Pancho and Lefty, who carried my gear and the stinky carnivore scats tirelessly over many miles. Finally, my dog. Kaiser von Stockwerk, provided faithfiil companionship for MVJ years greatiy appreciated His unconditional love was

111

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ABSTRACT UST OF TABLES USTOFnOURES CHAPTER I INTRODUCTION Literature Cited n fflSTORICAL OVERVIEW OF THE MEXICAN GRAY WOLF

ii vi viii x

1 2

AND ITS RECOVERY Literature Cited in DIFFERENTIATING MEXICAN WOLF AND COYOTE SCATS USING DNA ANALYSIS Abstract Introduction Study Area Materials and Methods Results Discussion Acknowledgements Literature Cited

3 7

10 10 11 12 14 22 24 31 32

IV

IV.

DIETS OF FREE-RANGING MEXICAN GRAY WOLVES IN ARIZONA AND NEW MEXICO Abstract Introduction Study Area Methods Resuhs Discussion Acknowledgements Literamre Cited 45 45 47 48 51 55 57 64 65 82

SUMMARY

ABSTRACT

The Mexican gray wolf (Catiis lupus baileyi) was extirpated from the southwestern United States before any systematic smdies could be conducted, therefore littie is known about the subspecies' natural history From April 1998 through October 2001, we coUected carnivore scats {n = 1,682) from the Blue Range Wolf Recovery Area (BRWRA) in .\rizona and New Mexico to study the diets of free-ranging Mexican wolves We identified the scats to species using traditional field methods (i e , diameter, location, and sign) and odor, but expected that it would be difficult to separate Mexican wolf and sympatric coyote {Cams latrans) scats We tested this hypothesis whh fecal DNA analysis (molecular scatology) to verify the accuracy of identifying Mexican wolf and sympatric coyote scats Our DNA results showed scats > 28 mm diameter could be identified as deposited by Mexican wolves, a high overlap in scat diameters for Mexican wolves and sympatric coyotes, and a difference in the 2 species' scat diameter means To determine the diets of Mexican wolves, we used our DNA results to refine our scat identification criteria Scats for diet analysis were identified as deposited by Mexican wolves by DNA analysis, diameters ^ 28 mm, 2 den site locations, and ttacks. Diet analysis of 55 scats with diameters ^ 28 mm collected from areas where 6 Mexican wolf packs received supplemental food items indicated Cienega and Hawks Nest packs consumed non-supplemental food items during supplemental feeding and Campbell Blue, Lupine, Mule, and Pipestem packs did not Diet analysis of 251 scats revealed nonsupplememal fed Mexican wolves consumed large-sized food items, primarily elk

VI

{Cerws elaphus caiKhJensis [nelsoni]) adults and calves We compared this diet composition to that found in 26 scats identified with DNA analysis and found more largesized food items appeared in Mexican wolf scats identified by our refined traditional method than in scats identified with DNA analysis There was no difference between diameter means or number of food items per scat between the 2 scat identification methods There was a difference between diet composition for 26 Mexican wolf scats and that found in 21 sympatric coyote scats identified with DNA analysis, with Mexican wolves consuming more large-sized food items than sympatric coyotes. There was also a difference between diameter means and number of food items per scat for the 2 species, vsith sympatric coyote scats containing more food items per scat than Mexican wolf scats. Lastly, we found a difference in diet composition of Vtexican wolves when compared to diet composition reported in 7 previous diet studies of other North American gray wolves Mexican wolf diet analysis revealed more large-sized food items than the subspecies' larger, northern counterparts

vn

UST OF TABLES 3.1 Comparison of accuracy of predicting which species (Mexican wolf or coyote) deposited a scat using discriminant analysis based on combination of 3 measurements takoi from scats identified to species with DNA analysis Locations of scats (n = 47) identified as Mexican wolf or coyote with DNA analysis Food itemsfoundin scats (n = SS) ^ 28 mm diameter collectedfromareas where Mexican gray wolf packs (n = 6) were fed supplemental food items (i.e., carnivore logs androad-killedelk, deer and jackrabbit) ConqMurison of diet composition found in scats (n = 55) ^ 28 mm diameter from areas where Mexican gray wolf packs (n = 6) were fed supplemental food items (i.e., carnivore logs and road-killed elk, deer, and jackrabbit) to determine presence of non-supplemental food items (e.g., elk aduhs and calves, deo* adults and Eiwns, domestic bovine, and insects) Food items (n = 265) found infree-rangingMexican gray wolf scats (/i = 251) Comparison among years (n = 4,1998-2001) of diet composition found in scats {n = 251) offree-rangingMexican gray wolves collected from April 1998 to October 2001 in Arizona and New Mexico Comparison between seasons (n = 2,fiill-winterversus s|Hing-summer) of diet composition found in scats (/i = 251) offree-rangingMexican gray wolves in Arizona and New Mexico (April 1998 - October 2001) Comparison of diet conq)08ition among packs (n = 4) found in free-ranging Mexican wolf scats (n = 251) collectedfromAJptii 1998 to October 2001 in Arizona and New Mexico Food items (n = 33) found infree-rangingMexican gray wolf scats (n = 26) collected from April 1998 to October 2001 in Arizona and New Mexico Comparison of diet c<Hiq)Osition of Mexican wolf scats (n = 277) collected from April 1998 to Oct(A>er 2001 in Arizona and New Mexico and identified ung 2 methods: DNA analysis (n = 26) and refined traditional (/t = 251; i.e., ^ 28 mm diameter, 2 den sites, and tracks) Coiiq>ari8(ni of diet composition found infree-rangingMexican wolf scats (/f = 26) and sympatric coyote scats (/i = 21) collectedfromhipdX 1998 to October 2001 in Arizona and New Mexico

43 44

3.2

4.1

71

4.2

72 73

4.3 4.4

74

4.5

75

4.6

76 77

4.7 4.8

78

4.9

79

viu

4.10

Food items found infree-rangingMexican gray wolf scats (n = 26) and sympatric coyote scats (n = 21) coUected from April 1998 to October 2001 in Arizona and New Mexico ConqpariscHi of diet composition found infree-rangingMexican wolf scats (n = 277) and diet oompositi( reported in other North American gray wolf diet studies (n = 7)

80

4. U

81

IX

LIST OF FIGURES

2.1

The Bhie Range Wolf Recovery Area (BRWRA) in Arizona and New Mexico where Mexican gray wolves were released beginning April 1998 Agarose gd showing a portion of the mtDNA control region (D-loop) amplified via Pilgrim et al. (1998) canid-specific primers (Mexican wolf and dog, 164 bp; coyote 160 bp) Agarose gel showing restrictionfragmentdigestion with BstN I of mtDNA control r ^ o n (D-loop) purified polym-ase chain reaction (PPCR) isolated from Mexican wolf and coyote scats Comparison of diameters of coyote scats (n = 21, range 17.4 to 27.8 m; X = 22.8 nun) and Mexican wolf scats (n = 26; range 16.3 to 35.8 nun, x = 26.0 nun) identified with fecal DNA analysis

31

40

32

41

3.3

42

CHAPTER I INTRODUCTION

The following chapters constitute partial fulfillment of the requirements for the degree of Master of Science in Wildlife Science for the Graduate School at Texas Tech University These chapters are the resuh of research conducted on Mexican gray wolves (Cojus lupus baileyi) m Arizona and New Mexico from April 1998 through October 2001 Chapter II is an historical overview of the Mexican gray wolf and its recovery Chapters ID and I\' are 2 manuscripts intended for submission to peer-reviewed journals. Chapter III discusses differentiating Mexican wolf and coyote scats using fecal DNA analysis (molecular scatology) Chapter IV reports the diets of free-ranging Mexican wolves in .\rizona and New Mexico from April 1998 through October 2001, and compares the diets of Mexican wolves to sympatric coyotes and to other North American gray wolves Chapter V is a summary of all chapters ,\11 chapters represent my ideas,

analyses, and writing ability Each chapter has several coauthors, which were determined using guidelines provided by Dickson and Conner (1978) and the CBE Style Manual Committee (1994) .Authorships for chapters are as follows Chapter III Chapter IV; Janet E Reed, Robert J Baker, Warren B Ballard, and Brian T Kelly Janet E Reed, Warren B Ballard, Philip S Gipson, Brian T Kelly, Paul R Krausman, Mark C Wallace, and David B Wester

LiteraUire Cited CBE Style Manual Committee 1994 Scientific style and format: the CBE manual for authors, editors, and publishers Sixth Edition Council of Biology Editors, Cambridge University Press, New York, New York, USA. Dickson, J. G., and R N Conner 1978 Guidelines for authorship of scientific articles. Wildlife Society Bulletin 6: 260-261

CHAPTER II HISTORICAL OVERVIEW OF THE MEXICAN GRAY WOLF AND ITS RECOVERY

Historically, the Mexican gray wolf (Ca7>/5 lupus baileyi), or lobo, lived the farthest south of all gray wolves (C lupus) on North America and in the most arid environment at elevations from 1,200 to 3,300 m ranging throughout central and southeastern Arizona, western Texas, southern New Mexico, and most of Old Mexico (Young and Goldman 1944, Brown 1983, Parsons 1996) h is believed that the Mexican wolf comributed to the overall historical biological diversity and ecological functioning of both the southwestern ecosystems and the continued evolution of species the Mexican wolf preyed upon (Parsons 1998) It is uncertain if the Mexican wolf was sympatric with the coyote {Canis latrans) in the Southwest before humans altered its natural habitat (Brown 1983) Gier (1975) and Bekoff and Wells (1986) reported that historical coyote distnbutions were confined primarily to plains and deserts until the spread of civilization and the reduction of gray wolf ranges By the early 1880s, humans and associated livestock had moved into the Southwest in large numbers (Brown 1983, Shumway 1998) Due to overgrazing and unregulated subsistence and market hunting, native prey species were in decline and some predators turned to the more abundant and easily caught livestock (United States Fish and Wildlife Service 1987) Government and private predator comrol programs

designed to protect livestock in the United States and Mexico accelerated the extirpation of the Mexican wolf from the Southwest by the late 1960s (Gipson and Ballard 1998). In 1975, the Arizona-Sonora Desert Museum proposed to capture Mexican wolves from the remaining population in Mexico and to place them in a captive-breeding program for future re-estabhshment of the subspecies in its historic range (Siminski 1997) The United States government listed the Mexican wolf as endangered in 1976 and Mexico began protecting this subspecies, as well (United States Fish and Wildlife Service 1996) The last 4 Mexican wolves known to exist were captured in Durango and Chihuahua, Mexico from 1977 to 1980 (McBride 1978) and 3 of them founded an official captive-breeding program (Garcia-Moreno et al 1996, Hedrick et al 1997) To increase genetic diversity in the official Certified (renamed McBride) lineage, 2 other captive Mexican wolf populations, Arag6n and Ghost Ranch lineages, were certified genetically pure in July 1995 (Hedrick et al 1997) and incorporated into the official captive-breeding program The first offspring from cross-lineage pairings were produced in 1997 (Parsons 1998) .\s of October 2001, the global population of Mexican wolves consisted of 227 individuals, the majority of which were held in 40 zoos and wildlife sanctuaries throughout the United Stales and Mexico (P Suninski, Arizona-Sonora Desert Museum, personal communication) Recovery efforts for the Mexican wolf were outlined in the Mexican Wolf Recovery Plan (United States Fish and Wildlife Service 1982) and research flinded by the Lmted States Fish and Wildlife Service (USFWS) supported experimental releases of Mexican wolves imo isolated Southwest habitat where close monitoring could be

conducted (Bednarz 1988) The Final Envu-onmental Impact Statement (FEIS) for the Mexican wolf proposed reintroduction of the subspecies into a portion of its historical range in Arizona, New Mexico, and Texas (United States Fish and Wildlife Service 1996) Three areas were considered the Blue Range Wolf Recovery Area (BRWRA) in .\rizona and New Mexico, White Sands Missile Range Wolf Recovery Area in New Mexico, and Big Bend National Park in Texas This proposal was formally approved by Secretar, of the Interior Bmce Babbitt's March 1997 Record of Decision (United States Fish and Wildlife Service 1997). In April 1998, the first 11 captive-reared Mexican wol\ es-5 generations removed from the wild-were released within the primary recovery zone of the BRWRA on the Apache National Forest in east-central Arizona (Figure 2 1). In the spring 2000, Mexican wolves previously released in Arizona were translocated to the secondary recovery zone on the Gila National Forest and Wilderness in New Mexico As of October 2001, at least 37 Mexican wolves were resident within the BRWTIA and 31 of those were radiocollared (United States Fish and Wildlife Service, unpublished data) In June 2000, we began a study to determine the diets of free-ranging Mexican gray wolves in Arizona and New Mexico We collected 1,682 carnivore scats from April 1998 through October 2001 and identified them to species using traditional identification methods and odor We verified scat identification accuracy with fecal DNA analysis (molecular scatology) and refined our identification of Mexican wolf scats We then analyzed Mexican wolf scats identified with DNA analysis and our refined traditional

identification method to determine the diet composition offree-rangingMexican wolves in Arizona and New Mexico

Literature Cited Bednarz,! C 1988 The Mexican wolf biology, history, and prospects for reestabhshment in New Mexico Endangered species report No. 18 United States Fish and Wildlife Service, Albuquerque, New Mexico, USA. Bekoff, M . and M C Wells 1986 Social ecology and behavior of coyotes Advances in tiie Stiidy of Behavior 16: 251-338 Brown, D E 1983 The wolf in the Southwest: the making of an endangered species. The University of .\rizona Press, Tucson, Arizona, USA Garcia-Moreno, J , M D Matocq, M S Roy, E Geffen, and R K Wayne 1996 Relationships and genetic purity of the endangered Mexican wolf based on analysis of microsatellite loci Conservation Biology 10 376-389 Gier. H T 1975 Ecology and behavior of the coyotes (Co^iw/a/rans), pages 247-262/>/ M W Fox, editor The Wild Canids their systematics, behavioral ecology and evolution Van Nostrand Reinhold, New York, USA Gipson, P S , and W B Ballard 1998 Accountsof famous North American wolves Canadian Field-Naturalist 112 724-739 Hednck, P W , P S Miller, E Geffen, and R K Wayne 1997 Genetic evaluation of the three captive Mexican wolf lineages Zoo Biology 16 47-69 McBride, R T 1978 Statusof the gray wolf (ran/5/M^.vZ)a//ev/) in Mexico United States Fish and Wildlife Service Report, Albuquerque, New Mexico, USA. Parsons, D R 1996 Case Smdy the Mexican wolf Pages 101-123 inE A. Herrera and L F Huenneke, editors New Mexico's natural heritage biological diversity in the Land of Enchantment New Mexico Journal of Science 36 101-123 Parsons, D R. 1998 "Green fire" returns to the Southwest reintroduaion of the Mexican wolf Wildlife Society Bulletin 26: 799-807 Shumway, E W 1998 Alpine, Arizona a stroll through history, Americopy, Mesa, Arizona, USA Simmski, D P 1997 A history of cooperation in Mexican wolf conservation Proceedings American Zoo and Aquarium Association Regional Conference 1997 384-385

United States Fish and Wildlife Service. 1982. Mexicanwolf recovery plan. United States Fish and Wildlife Service, Albuquerque, New Mexico, USA United States Fish and Wildlife Service 1987. Restoring America's wildlife 1937-1987: the first 50 years of the Federal Aid in Wildlife Restoration (Pitman-Robertson .\ct) Unhed States Department of the Interior, United States Fish and Wildlife Service, Washington, D C , USA. United States Fish and Wildlife Service 1996 Reintroduction of the Mexican wolf within its historic range in the southwestern United States Final Environmental Impact Statement. United States Fish and Wildlife Service, Albuquerque, New Mexico, US.\ United States Fish and Wildlife Service 1997. Record of decision and statement of fmdings on the environmental impact statement on reintroduction of the Mexican gray wolf to rts historic range in the southwestern United States United States Fish and Wildlife Service, Albuquerque, New Mexico, USA Young, S P . andE A Goldman 1944 The wolves of North America The American Wildlife Institute, Washington, D C , USA

CHAPTER III DIFFERENT1.\TING MEXICAN GRAY WOLF AND COYOTE SCATS USING DNA ANALYSIS

Abstract We used fecal DNA analysis (molecular scatology) to test a subsample of scats (// = 203) identified as deposited by Mexican gray wolves (Ca^j/5 lupus baileyi) or coyotes {Cams latrans) and compared the resuhs to the identification of scats using traditional methods (i e , diameter, location, and sign) and odor We then used the scats identified with DN.A. analysis to evaluate discriminant analysis for classifying scats using 3 measurements (i e , diameter, mass, and length) Forty-nine (24 o) of the fieldcollected scats {n = 203) tested provided amplifiable DN.A., which identified 28 scats deposhed by Mexican wolves and 21 deposited by coyotes Scats identified with DNA analysis for the 2 species had a 79% diameter overiap (Mexican wolf, 16 3 to 35 8 mm, coyote, 17 4 to 27 8 mm) and scats ^ 28 mm diameter were Mexican wolf scats There was a difference (/ = -2 428. Z' = 0 019) between diameter means for the 2 species (Mexican wolf ir = 26 0 mm, coyotex = 22 8 mm) Of 45 scats that would have been field-identified as deposited by Mexican wolves based on location and odor criteria, DN.A analysis showed that 19 (42o) were deposited by coyote, and of 41 scats that would have been field-identified as deposhed by coyotes based on diameter < 30 mm criterion, 20 (49%) were deposited by Mexican wolves We then used Halfpenny's (1986) suggested diameter criterion for field-identification of scats, which identified 3 of

10

the scats as gray {Urocyon cmereoargenleus) or red (Vulpes vulpes) fox (0% correct), 24 as coyote (62o o correct), and 20 as Mexican wolf (75% correct). Discriminant analysis indicated that diameter and mass offered the best resuhs for accurately classifying coyote scats (86 o), but provided relatively low accuracy for classifying Mexican wolf scats (65 o) Molecular scatology appears to be a more definitive scat identification technique than methods previously utilized (i e , traditional and odor)

Introduction No scientific smdies were conducted on the Mexican gray wolf (Canw lupus baileyi) before it was extirpated from the wild, therefore little is known of the subspecies' natural history (Brown 1983) Previous studies indicate that the Mexican wolf is the smallest (Hall and Kelson 1959, Bogan and Mehlhop 1983). the most genetically distinct (Wayne et al 1992, Nowak 1995, Garcia-Moreno et al 1996), and the most endangered (McBride 1980. Brown 1983, Bednarz 1988, Ginsberg and Macdonald 1990) subspecies of gray wolf (C lupus) on North America In April 1998, the United States Fish and Wildlife Service (USFWS) began releasing captive Mexican wolves into the Blue Range Wolf Recovery .\rea (BRWRA) on the Apache and Gila National Forests and the Gila Wilderness in Arizona and New Mexico to restore the subspecies into a portion of its historical range In June 2000, we began a study to determine the diets of captive-bom and wildbom .Mexican gray wolves free-ranging in the southwestern United States From April 1998 through October 2001, we collected carnivore scats {n = 1,682) from the BRWRA

II

and identified them to species using traditional field methods (i e , diameter, location, and sign) and odor Mexican wolves are the smallest subspecies of gray wolves found on North America, therefore we hypothesized that traditional scat identification techniques might have a high error rate in separating Mexican wolf and coyote (Canis latrans) scats collected where the 2 species were sympatric There are no published reports using fecal DNA analysis (molecular scatology) to differentiate between gray wolf and coyote scats for diet analysis The purpose of this research was to accurately differentiate Mexican wolf and coyote scats in order to study the diets of Mexican wolves Accurate identification of scats would be beneficial for determining the level of dietary overlap of gray wolves and coyotes in areas where they are s>Tnpatric, managing gray wolf and prey populations, and investigating livestock depredation incidences in areas where gray wolves have been reintroduced or are recovering

Study .Area We conducted research in the BRW'R.A (Figure 2 1), a 17,820 km^ area that included the Apache National Forest (NF) in east-central Arizona and the Gila NF and Wilderness in west-central New Mexico Our focus was the 2,600 km^ primary recovery zone in .Arizona where captive Mexican wolves were released and wild-bom since April 1998 The secondary recovery zone (Gila National Forest and Wilderness, New Mexico) was included beginning spring 2000 due to translocations of Mexican wolves previously released in Arizona The BRWRA was bordered by the White Mountain (Fort Apache)

12

and San Carlos Apache Reservations to the west, while private lands were scattered within and bordered the east, north and south Almost all areas were grazed by domestic livestock Elevations within the BRWRA ranged from < 1,220 m along the San Francisco River to 3,350 m on Mount Baldy, Escudilla Mountain, and the Mogollon Mountains (United States Fish and Wildlife Service 1996) Rolling hills with moderately steepwalled canyons and sandy washes characterized the lower elevations, while mgged slopes, deep canyons, elevated mesas, and rock cliffs typified the higher elevations. Major vegetation included ponderosa pine (Pittusponderosa), aspen (Populus iremuloides), fir (Abies spp ), juniper (Juniperus spp ), piflon (Pirrus cembroides), mesquhe (Prosopis spp ), evergreen oaks {Quercus spp ), and a variety of grasses and forbs Annual temperatures reported for the BRWHA averaged 16 4-C maximum and

-3 \C minimum, annual precipitation averaged 52 1 cm, and annual snowfall averaged 139 3 cm (Desert Research Institute, Western Region Climate Center, Reno, Nevada, impublished data) From April 1998 through October 2(X)1, > 87 Mexican wolves were either released from captivity or bom in the wild within the BRWTIA As of October 2001, > 37 Mexican wolves resided in the BRWRA and 31 of those were fitted with radio collars (United States Fish and Wildlife Service, unpublished data) Other predators residem in the BRWRA included humans (Homo sapiens), coyote, gray (Urocyon cmereoargenleus) and red (Vulpes vulpes) fox, bobcat (Felis rufus), mountain lion (Puma concolor), and black bear (Ursus americanus) (United States Fish and Wildlife Service

13

1993, Arizona Game and Fish Department 1994) Density estimates for these predators were unavailable, but the United States Forest Service (USFS) and Arizona Game and Fish Department (AGFD) reported coyote densities to be relatively high (personal communication) These sources and the USFWS also reported that feral and stray dogs or wolf-dog hybrids were rare residents, if at all, within the BRWRA, therefore we assumed a low likelihood of collecting dog or wolf-dog hybrid scats.

Materials and Methods Sample collection Carnivore scats (/; = 1,682) were collected opportunistically from the BRWRA by Mexican wolf Interagency Field Team (IFT, i e , U S Fish and Wildlife Service, Arizona Cjame and Fish Department, New Mexico Department of Game and Fish, USDA APHIS Wildlife Services, and U S Forest Service) personnel from April 1998 through October 2001 We actively collected scats from June through August 2000 and March through October 2(X)1 in areas where captive-released, translocated, and wild-bom Mexican wolves were known to frequent The scat collection process employed an opportunistic sampling strategy (Frenzel 1974) in which the sampling areas followed Mexican wolves as they moved within the study area as reported by the USFWS Our search methods included driving forest roads, hiking or horseback riding forest trails, ridgelines, and npanan areas, and investigating forest campsites, opened release pens, den sites and kill or carcass sites

14

Scats were collected using disposable mbber or food preparation gloves and placed in brown paper bags, which were labeled with date, location, and scat number. We aged scats as old, recent, or fresh according to appearance, exposure of deposition site, and weather conditions (Ciucci et al 1996) Scats were identified initially to species using traditional identification techniques (i e , location, diameter, and sign) and odor as described below We did not autoclave the scats, as there were no reports of Echinoci\:cus granulosus in the Southwest The scats were allowed to ah dry in the brown paper bags before storing in large plastic containers at room temperature until analyzed The location criterion was established as areas frequented by Mexican wolves, such as opened release pens, den sites, and kill or carcass sites We measured the maximum diameter of each air-dried scat as described by Scott (1943). Weaver and Fritts (1979), Green and Flinders (1981), and Danner and Dodd (1982). however, we took 2 measurements at the maximum diameter to the nearest 0 1 mm with a 152 mm dial caliper (General Tools Manufacturing Co , LLC, New York, New York) and used the average for diameter size Thompson (1952) established the minimum diameter for identifying northern gray wolf scats in the field as > 24 mm and has been the accepted critenon for several smdies (Mech 1970, Stephenson and Johnson 1972, Peterson 1974, Van Ballenberghe et al 1975) However, Weaver and Fritts (1979) suggested > 30 mm minimum diameter be used for identifying gray wolf scats, which was the diameter critenon used by Arjo et al (2002) Halfpenny (1986) reviewed 3 stiidies and suggested > 25 mm diameter for identifying gray wolf scats Initially, we used the more

15

conservative > 30 mm diameter criterion for classifying scats as deposited by Mexican wolves Sign criterion was tracks of Mexican wolves Mexican wolf IFT personnel provided instmction for identifying Mexican wolf scats by odor We were provided with scats identified as Mexican wolf scats by the IFT personnel and were instmcted that the odor (i e , sweet, musky) of those scats was that of wolf No odor description was available for identify ing coyote scats.

Genetic identification We used fecal DN.A analysis to identify scats deposited by either Mexican wolves or coyotes Our design was to isolate and analyze Mexican wolf and coyote DNA from shed epithelial cells from the intestinal lining found on scats (Kohn et al 1995, Foran et al 1997, Kohn and Wayne 1997, Reed et al 1997, Frantzen et al 1998) We examined a portion of the mhochondrial DNA (mtDNA) control region (D-loop, Pilgrim et al. 1998) to ditTerentiate Mexican wolf scats from s\mpatric coyote scats collected from the field We subsampled 203 of the field-collected scats by taking 2 to 6 fecal surface samples from each scat for DNA analysis The surface of the scats were scraped to decrease the possibility of removing undigested prey parts from within the scats and to increase the probability of obtaining sloughed epithelial cells (Reed et al. 1997) Each of the subsamples, comprising ^ 1 g of fecal material taken from each scat, was placed in a 1 5 mL micro centrifuge tube and stored at room temperature (Taberlet et al 1997) The remainder of each scat was used for diet analysis To optimize laboratory efforts, scats were not subsampled randomly but were selected primarily from scats field-identified as

16

deposited by Mexican wolves (/= 169) We sorted the master database for scats identified as probably deposited by Mexican wolf scats (/= 1.111) based on traditional field metiiods and odor, tiien selected every 5*^ scat to be subsampled for DNA analysis.

DN.A isolation Fecal material is likely to contain relatively low concentrations of DNA much of which is degraded (Gerloff et al 1995, Kohn et al 1995), yet mtDNA is likely to be present on a scat m greater quantity than single-copy nuclear DNA (Reed et al 1997, Woods etal 1999) Polymerase chain reaction (PCR) amplification of short mtDNA fragments results in more consistem amplifications than longer fragments (Kohn et al 1995) Therefore, it was necessary to identify and establish a simple, rapid and reliable protocol that extracted as much DNA as possible from degraded samples and removed any potential PCR inhibitors (Boom et al 1990, Deuter et al 1995. Kohn et al 1995. Reed et al 1997) We tried 4 DNA isolation methods Proteinase K method (Sambrook etal 1989). salt extraction method (Miller et al 1988), Qiagen DNeasy* Tissue Kit (Qiagen, Valencia, California, USA), and guanidinium thiocyanate (GuSCN) and silica method (Boom et al 1990, Hoss et al 1992, Hbss & Piabo 1993) Our resuhs were similar to those of Hoss and Paabo (1993) and Reed et al (1997), with the GuSCN and silica method having the highest isolation efficiency and providing the most consistent resuhs We used this method as our DNA isolation protocol and optunized it for our application, as described below

17

We isolated genomic DNA from: whole blood (n = 38) previously collected from released Mexican wolves, museum specimen coyote tissue (liver, n = 4), and Mexican wolf scats (n = 13) collected from captive individuals, coyote scats (n = 5) collected where no wx)l\ es occurred along the New Mexico and Texas border, and domestic dog scats (/; = 5) The DNA isolated from tiiese sources provided the reference collection for genetically identifying target species' (Foran et al 1997) Addhionally, we attempted to isolate DN.A from museum specimen rodent tissue (liver, /; = 8), and hair (n = 4) and grass {n ^ 2) from field-collected scats to test for possible amplication of food items ingested We modified an extraction buffer (5 M GuSCN, 0 1 M Tns-HCl pH 6 4, 0 2 M EDT.A pH 8 0, 13% Triton X-100), bringing the volume to 50 mL with ddHjO (Boom et al 1990, Hoss and Paabo 1993, Hoss 1994) One millimeter of extraction buffer was added to each <. 1 g of dry fecal material in a 1 5 mL micro centrifiige tube The products were vortexed, then incubated overnight at 55C with constant agitation Samples were then centnfiiged for 10 min at 13.000 rpm in a Microfuge* 18 Centrifuge (Beckman Couher^, Inc , Palo Alto, California, USA) Approximately 7(X) [iL of the supernatant was transferred to a sterile 1 5 mL micro centrifuge tube and 300 (iL of fresh extraction buffer and 50 nL of silica matrix were added (Boom et al. 1990) The mixture was vortexed before incubated at 55C for 10 min with constant agitation After centrifligation (1 min, 5,000 rpm), the supernatant was decanted and discarded. The remaining silica pellet was washed twice with 1 mL of a modified wash buffer (5 M GuSCN, 0 1 M Tns-HCl pH 6 4, Boom et al 1990) and once with 1 mL of 70% ethanol,

18

centrifiiging each time for 1 min at 5,000 rpm The pellet was dried at 55C and then the extracted DNA was eluted at 55C with constant agitation for 30 min in 100 ^L elution buff'er (10 mM Tris-CI, pH 8 5) The product was centrifliged (5 min, 13,000 rpm) and 75 [iL of the supernatant was removed carefully to avoid transferring silica particles that could inhibit the PCR (Hoss and PMbo 1993) and placed in a sterile 15 mL micro centrifuge tube These samples were stored at -20C until PCR amplification DN A isolation procedures were performed under a fume hood in a laboratory exclusiN ely dedicated to DNA isolation and PCR amplification Care was taken throughout all procedures to ensure no cross-contamination To monhor contamination during DN.A isolations, negative controls (no fecal material added to the extraction buffer) were treated identically through both the isolation procedure and the subsequent PCR amplifications (Handt et al 1994)

PCR amplification We used the PCR to amplify < 200 base pairs (bp) of the mtDNA control region (D-loop) using canid-specific mtDNA primers (F-4479679, R-4479680, Integrated DNA Technologies, Inc , Coralville, Iowa, USA) designed by Pilgrim et al (1998) to produce species-specific bands of 164 bp for wolves and dogs and 160 bp for coyotes Four microliters of excremental DNA extract were placed in a sterile 0 5 mL micro centrifiige tube with a 46 nL volume containing 0 2 mM each dNTP, 3 25 mM .MgCIj, IX buff'er, 0 5 piM each pnmer, 0 5 U Tag polymerase (Ml865, Promega Corporation, .Madison, Wisconsin, USA), and 25 0 nL ddH20, then topped with 2 drops

19

of autoclaved mineral oil to prevent evaporation. The DNA and reagent mixtiires were amplified in a programmable Perkin Elmer CeUis 480 DNA thermal cycler (Perkin Elmer, Norwalk, Connecticut, USA) for 32 cycles, each cycle consisting of a denaturation step for 20 sec at 95C, an annealing step for 40 sec at 50C, and an extension step of 30 sec at 72C Three microliters of tiie amplification reaction were mixed with 3 |iL standard tracking dye and were separated at 60 milliamps on a 2% agarose gel containing etiiidium bromide in TAE buff'er (Ausubel et al 1989) Gels were visualized under ultraviolet (LTV) excitation on an Eagle Eye n (Stratagene*, La Jolla, California, US.A) Amplicoms were sized using a 50 bp DNA ladder (Fisher Scientific BP 2550-100) Negative extraction and PCR controls were used alongside the excremental DNA samples to monitor contamination of the PCR reagents As positive controls, we used the DN.A extracted from blood samples collected from previously released Mexican wolves.

DNA analysis Pilgrim et al (1998) identified a control region length difference between coyotes and wolves and dogs, which also results in a restriction-site polymorphism present in wolves and dogs but lacking in coyotes Using the restriction enzyme Mva I (BstN I, New England BioLabs Inc , Beverly, Massachusetts, USA), Pilgrim et al (1998) distinguished coyote amplicons from those of wolf and dog Twenty-five PCR products were digested with 0 35 ^L restriction enzyme BstN I, 3 fiL lOX buff'er, 3 ^L BSA and 15 95 ^L ddH20 at 37C for 3 hr Digested amplicons were analyzed on a 2 5% agarose

20

gel containing ethidium bromide in TAE buff'er, then visualized under UV excitation on an Eagle Eye n Ehgested products were sized using a 50 bp ladder Products that remained undigested at 160 bp were identified as coyote and fragments that digested into 2 fi^gments of 49 and 113 bp were identified as wolf To confirm the results of the mtDNA restriction fragment analysis, we sequenced tiie same fragment used for the digestions of 31 scats PCR products were purified using the QLAquick PCR Purification Kit (Qiagen) following the supplier's instmctions Punfied PCR (PPCR) products were cycle-sequenced in both directions (2 reactions) using a programmable PTC-200 Pehier Thermal Cycler (MJ Research, Inc , Watertown, Massachusetts, USA). 1 5 ^L PPCR, 1 0 nL 2 mM forward or reverse primer. 4 ^L Big D\e Terminator v3 0 (4390244, Applied Biosystems, Foster City, California, USA), and 3 5 (iL ddHjO Each cycle-sequenced product was purified with 3 0 ^L 3 M NaOAc pH 4 6, b2 5 ^L 100% EtOH, and 14 5 nL ddHjO for 15 min at room temperaUire, then centrifliged for 20 min at 13,(XX) rpm The supernatant was decanted and discarded, then the pellet was washed with 250 |iL 70% EtOH followed by centrifligation for 5 min at 13,(XX) rpm The EtOH was decanted and discarded, then the pellet was air-dried at room temperature overnight in the dark Twenty microliters of Template Suppression Reagent (Applied Biosystems) were added to the dried pellet, then vortexed The pellet was shaken down by hand and incubated at 95C for 3 min, then quick spun Raw sequence data were collected using a 310 Genetic Analyzer (Applied Biosystems) Resuhing sequences were aligned and analyzed using Sequencher v3 1 (Gene Codes Corporation, Inc , Arm Arbor, Michigan, USA) Aligned, proofed sequences were compared to

homologous wolf and coyote sequences ((jenbank No AF487754, AF020700) and domestic dog mtDNA sequences (Genbank No AF487751, Slade et al. 1994) The difference between scat diameter means for Mexican wolf and coyote was calculated using a standard /-test (Ott 1998) and deemed significant if P <. 0 05

Discriminant analysis Based on the measurement values (i e , diameter, mass, and length) of the scats, we used discriminant analysis (Williams 1982) to classify scats as either Mexican wolf or coyote Forty-seven of the scats identified with DNA analysis as deposited by Mexican wolf or coyote were weighed to the nearest 0 1 g on a OHAUS Precision Plus TP4000 scale (OHAUS Corporation, Florham Park, New Jersey, USA), and the total length of each scat was measured with a metric straight edge to the nearest 0 1 cm We used discriminant analysis to classify scats based on (1) diameter, (2) diameter and mass, (3) diameter and length, and (4) diameter, mass, and length Since coyote density was unknown within the smdy area, we accepted prior probability (q,) = 0 50

Resuhs Genetic identification Forty-nine (24o) of 203 scats provided amplifiable DNA (Figure 3 1) Twentyeight were identified as Mexican wolf scats and 21 were identified as coyote scats with restriction fragment analysis (n = IS, Figure 3 2), sequence analysis (n = 24), or both (/J = 7) Two of the scats identified as deposited by Mexican wolves were uncollectibles, which are loose, liquid feces containing little idemifiable prey material (Floyd et al

22

1978) Uncollectible scats cannot be measured for diameter or length and traditionally have not been included in diet stiidies Therefore, the 2 uncollectible scats were discarded, leaving 26 Mexican wolf scats for further analysis Mexican wolf scats (n = 26) ranged in diameter from 16 3 to 35.8 mm, while coyote scats (n = 21) ranged in diameter from 17 4 to 27 8 mm (Figure 3.3). We found a 79/o o%erlap in scat diameter for Mexican wolf (AJ = 16) and coyote (n = 2l) There was a difference (t = -2 428, P = 0.019) between scat diameter means for the 2 species (Mexican wolf x = :6 0 mm, coyote r = 22 8 mm. Figure 3 3) Scat diameters > 28 mm appeared adequate for identifying Mexican wolf scats without other identification criteria (i e , location and sign) Of the 45 scats that would have been identified as deposited by Mexican wolves based on location and odor criteria, 19 (42 o) were acmally deposited by coyotes, and of the 41 scats that would have been identified as deposited by coyotes based on diameter < 30 mm criterion, 20 (49%) were actually deposited by Mexican wolves Using Halfijenny's (1986) suggested diameter criterion for identifying the scats of 3 carnivore species, scat diameters < 18 mm would have been identified as fox (n = 3, 0% correct), diameters 18 mm to 25 mm would have been identified as coyote (n = 24, 62% correct), and diameters > 25 mm would have been identified as Mexican wolf (/? = 20, 75% correct) Foran et al (1997) reported that ingested prey tissue did not produce erroneous resuhs when targeting species-specific DNA and our resuhs supported this claim Chir attempts to isolate DNA from rodent tissue (liver) and ingested grass and prey hair taken

23

from scats provided no amplifiable PCR products using the canid-specific primers designed by Pilgrim et al. (1998)

Discriminant analysis For scats identified as Mexican wolf or coyote wrth DNA analysis, we used discriminant analysis for 4 classifications (Table 3 1) Scat diameter (Classification 1) accuratdy identified 81% of coyote scats, but only 50% of Mexican wolf scats Scat diameter and mass (Classification 2) accurately identified 86% of coyote scats and 65% of Mexican wolf scats Scat diameter and length (Classification 3) accurately identified 68% of coyote scats and 59% of Mexican wolf scats Lastly, a combination of scat diameter, mass, and length (Classification 4) accurately identified 79% of coyote scats and 55% of Mexican wolf scats

Discussion Fecal material is often the most common sign and easily collected source of information for rare, secretive carnivores (Putman 1984), and scat analysis appears to provide the best noninvasive sampling method for determining diets of free-ranging carnivore species However, which species actually deposited a scat must be accurately determined for the method to be vahd Traditional scat identification criteria have been based primarily on morphology (Halfpenny 1986, Foran et al 1997), which can be subjective and confounded by sympatric species that are comparably sized and share similar diets (Weaver and Fritts 1979, Green and Flinders 1981, Danner and Dodd 1982,

24

Foran et al 1997). Halfpenny (1986) reported that visual identification of scat to species by experienced naturaUsts had error rates approaching 50 to 66%. Scott (1941, 1943) reported tiiat fecal passage sizes varied approximately in proportion to the amount of food consumed Weaver and Fritts (1979) reported tiiat scat diameter might be influenced by diet composition and suggested that canid scats could not be identified to species based on diameter alone Previous dietary analysis based on scat identification using diameter alone may have biased results in favor of prey items that produce larger scat diameters (Danner and Dodd 1982) Halfjpenny (1986) reported that scat diameter and length values do not provide positive identification of a species, because both values are too variable to be used as adequate diagnostic criteria Green and Flinders (1981) reported that the dry mass of scats \aried considerably for coyotes and red foxes and suggested that diameter used in conjunction with mass could be used for identifying scats of the 2 species Discriminant analysis of diameter and mass values of scats (Classification 2) provided the most accurate identification for both species (coyote, 86%, Mexican wolf, 65%) However, 14'o of the scats deposited by coyotes were classified incorrectly as Mexican wolf scats and 35 0 of the Mexican wolf scats were classified incorrectly as coyote scats Although diameter and mass values provided a relatively high percentage of accuracy for identifying coyote scats, we considered the resuhs unsatisfactory for identifying Mexican wolf scats Uncollectible scats, as defined by Floyd et al (1978), have traditionally been discarded because it is difficult to identify them to species and uncollectibles usually

25

contain little identifiable prey material We were able to isolate DNA and identify the species of 2 uncollectibles subsampled When using the appropriate primers, it is also possible to amplify any prey DNA found in uncollectibles so that they can now be included in cami\ore diet studies, if desired Location of where scats have been deposited has also been used to identify wolf scats for diet analysis smdies Scats collected from wolf den and rendezvous sites undoubtedK have provided accurate diet information because coyotes rarely attend these areas (Ballard et al 2003) However, the information only provides wolf diet data for late spring through summer (Murie 1944, Mech 1966, Ballard et al 1987, Fuller 1989, Spaulding et al 1997) Identification of scats collected from kill and carcass sites (Thompson 1952, .Arjo et al 2(X)2) may be more problematic in areas where wolves and coyotes are svmpatric because coyotes often scavenge from wolf kills (Paquet 1992, Phillips and Smith 1996) The scats (n = 47, Table 3 2) from which we were able to isolate DN.A were collected from forest trails (21 scats), forest highways, roads, and 2-tracks (16 scats), an elk carcass (1 scat), a wolf den (1 scat), an opened release pen (2 scats), and other locations (i e , stream bed, cabin area, crater, canyon, mesa, and ridge, 6 scats) Odor has also been reported as a scat identification technique, but it is subjective (Stokes and Stokes 1986, Bang 2001) and currently cannot be quantified The only mention of odor in the literature as an identification technique was for fox (Scott 1943, Murie 1954, Wilcomb 1956, Korschgen 1980, Turkowski 1980), not for wolves or coyotes Halfpenny (1986) suggested that odor resuhed from the camivore's diet

26

However, some wolf biologists have reported through personal communications that they can identify wolf scats by odor, but tiiis claim has not been scientifically substantiated and our resuhs suggest that odor was not a reliable method for this stiidy Another attempt to identify carnivore scats has employed bile acids (Eneroth and Sjovall 1969, .Alfred 1980, Major et al 1980, Clinite 1981, Johnson et al 1984, Quinn and Jackman 1994) or pH (Green and Fhnders 1981) with httle or no success According to Halfpenny (1986), these methods have proven more successful for herbivores than carnivores Recent noninvasive sampling studies of free-ranging mammals (Foran et al 1997, Kohn and W^ayne 1997, Reed et al 1997, Ernest et al 2000, Lucchini et al 2002) have confirmed that fecal DNA analysis provides a more accurate assignment of the species that deposited a scat rather than morphology of scats With noninvasive sampling and fecal DN.A analysis, biologists can collect individual fecal samples at any time, from any location, to study free-ranging species without having to disturb them (Hoss et al 1992, Taberlet and Bouvet 1992. Morin et al 1993, Kohn and Wayne 1997, Taberlet et al 1999) We were able to collect scats from the field and isolate DNA for species identification without ever seeing or dismpting the individuals DNA from sloughed intestinal epithelial cells found on fecal material provides an objective marker for species identification, and potentially for individuals, that remains invariable during the life of an animal (Foran et al 1997) Mitochondrial DNA is inherited maternally and is non-recombining (Brown 1985), making it well suited for species identification of fecal samples (Moritz 1994, Kohn and Wayne 1997) and

27

mammalian population genetic stiidies (Randi et al. 1994, Slade et al. 1994, Taberlet et al 1995) Mitochondrial DNA's extia-nuclear genome provides several hundred copies per cell (Kohn et al 1995) and is more amenable to genetic manipulation than singlecopy nuclear genes (Kohn and Wayne 1997, Frantten et al 1998). The control region (D-loop) is usually more polymorphic and informative for differentiating species (Pilgrim et al 1998) because it evolves faster in mammals than either the rest of the mtDNA molecule or most single-copy nuclear DNA (Avise 1994) Beginning with small quantities of relatively short target sequences of DNA (Walsh et al. 1991. Mullis et al 1994. Reed et al 1997), the PCR is an enzymatic process that can experimentally synthesize a large number of copies of specific DNA sequences from degraded or impure samples (Saiki et al 1985, Mullis and Faloona 1987, Saiki et al 1988, Amheim et al 1990, Walsh et al 1991) When applied to mtDNA sequences useful for estimating genetic differences between closely related species can be amplified (Kocher et al 1989, .A\ise 1994) Species identification of scats using DN.A-based assays provides an accurate method that is rapid (3-4 days), repeatable, and relatively inexpensive (The cost is approximately $5 00 [USJ/sample for the disposable items, enzymes, and chemicals required for sequencing Other costs include the sequencer, non-disposable lab equipmem [e g , pipetors, cameras, and computers] and salaries of students and technicians ) Furthermore, the resuhs are definitive and not subject to confidence intervals or probabilistic estimation (Foran et al 1997) However, our low DNA isolation success (24%) may be cause for concem

28

DNA isolated from fecal material is often of low quantity and quality (Taberlet et al 1996) In addition, epitiielial cells usually are distributed unevenly (Kohn et al 1995) and our fecal subsamples may not have included tiie cells required for DNA isolation. Fecal material may also contain Taq DNA polymerase or PCR inhibitors, however, tiiis problem can be reduced by using tiie silica-based extraction method (Boom et al 1990, H6SS and Paabo 1993, Kohn et al 1995) Our modest DNA isolation success was attnbuted to low-quality and low-quamity DNA found on scats Furthermore, the scats we tested were dry-stored for up to 5 years at the time we conducted DNA analysis. Reed et al (1997) and Lucchini et al (2002) reported that fecal sample age affected DNA isolation success and suggested that fresh scats would be more suitable for DNA analysis -Although our DNA isolation success (24%) was low, we found, as did Foran et al (1997), that a scat's physical appearance was not a definitive guide to the DNA quality a\ailable Another possibility influencing our low DNA isolation success could have

been that some of the scats were deposited by non-target species whose DNA could not be amplified with the canid-specific primers designed for wolf and coyote that we used We found, however, that the targeted species could be identified if a PCR product could be obtained, since DNA too degraded to amplify produced no results as opposed to incorrect results (Foran et al 1997) Our results were consistent with the findings of Pilgrim et al (1998) that wolf and coyote mtDNAs were distinct and could be differentiated by a single restriction site and length polymorphism We recognize that there is a possibility that a low percentage of the scats identified as wolf could have been deposited by feral dogs or wolf-dog hybrids if they existed within the BRWRA.

29

Therefore, we recommend that ftiture wolf DNA research utilize primers (Foran et al. 1997) that differentiate between wolf and dog DNA, With a sample size of 47 scats identified to species using DNA analysis, we suggest that our results be interpreted with caution Our resuhs demonstrated, however, that identification of Mexican wolf and coyote scats using DNA analysis was more accurate than identification methods previously available Molecular scatology can facilrtate the identification of species, individuals, their gender, food habhs, and pathology This would require an experimental design with extended systematic transects from which fresh fecal samples are obtained, coupled with an appropriate preservation of fecal material and DN.A isolation method (Reed et al 1997, Wasser et al 1997, Frantzen et al 1998, Taberlet et al 1999, Lucchini et al 2(X)2), and using appropriate speciessp>ecific primers (Foran et al 1997). These data could then be used further to estimate home range (Adams et al 2(K)3), reproductive patterns, kinship stmcture and population size (Kohn and W ayne 1997) Molecular scatology also has potential to detect hybndization (Lehman et al 1991, Wayne et al 1992, Gondii et al 1994, Pilgrim et al 1998. Vila and Wayne 1999, Adams et al 2003) Finally, these data could be used for validating the presence of wolves in livestock depredation incidences Our resuhs suggest that previous diet studies using traditional scat identification methods may have misrepresented the diets of both the North American gray wolf and coyote where the 2 species were sympatric Fecal DNA analysis provides an accurate method for assessing the visual identification of scat samples collected from the field and improves diet analysis (Reed et al 1997) Molecular scatology appears to have

30

significant potential as a noninvasive sampling technique to momtor and manage freeranging Mexican wolves where the subspecies is sympatric with other carnivore species

Acknowledgements The United States Fish and Wildlife Service funded this research and provided scats from captive Mexican wolves The University of New Mexico and Arizona Veterinary Diagnostic Lab provided blood samples from free-ranging Mexican wolves The Museum of Texas Tech University provided coyote and rodent tissue samples. Kaiser von Stockwerk generously provided domestic dog scats A Brown, C C. Perchellet, and Drs F Hoffinan and J Wickliffe contributed invaluable assistance and dedication in the DN.A lab Dr D B Wester facilitated the statistical analysis This is a Texas Tech University. College of Agriculmral Sciences and Natural Resources technical fwjblication T-X-.\XX

31

Literatiire Cited Adams, J R , B T KeUy, and L P Waits 2003 Using faecal DNA sampling and GIS to monitor hybridization between red wolves (Canis rufus) and coyotes (Canis latrans) Molecular Ecology 12: 2175-2186 Alfred, D R 1980 Biochemical identification of carnivore scats Thesis, Mississippi State University, Mississippi, USA. .Amheim, N , T WTiite, and W E Rainey 1990 Application of PCR: organismal and population biology BioScience 40: 174-182 Arizona Game and Fish Department 1994 Predator population estimates Unpublished report, Arizona Game and Fish Department, Phoenix, Arizona, USA Arjo, W M , D R Pletscher, and R R Ream 2002 Dietary overlap between wolves and coyotes in northwestern Montana Journal of Mammalogy 83: 754-766. Ausubel, F M . R Brent, R E. Kingston, D D Moore, J G Seigdman, J A Smith, and K Struhl 1989 Current protocols in molecular biology John Wiley & Sons, New York, New York, USA .A\ise, J C 1994. Molecular markers, namral history, and evolution Chapman & Hall, New York, New York, USA Ballard, W B , J S Whitman, and C L Gardner 1987 Ecology of an exploited wolf population in south-central Alaska Wildlife Monograph 98 Ballard, W B , L N Carbyn, and D W Smitii 2003 Pages 259-271 m L D Mech and L Boitani, editors Wolves behavior, ecology and conservation University of Chicago Press, Chicago, Illinois, USA Bang, P 2001 Animal tracks and signs Oxford University Press, New York, USA Bednarz, J C 1988 The Mexican Wolf: biology, history, and prospects for reestabhshment in New Mexico Endangered Species Report No 18 United States Fish and Wildlife Service, Albuquerque, New Mexico, USA Bogan, M A , and P Mehlhop 1983 Systematic relationships of gray wolves (Coww lupus) in southwestern North America Occasional papers of the Museum of Southwestern Biology 1

32

Boom, R , C J A. Sol, M M M Salimans. C L Jansen, P M E Wertheim-van Dillen, and J van der Noordaa. 1990 Rapid and simple method for purification of nucleic acids Journal of Clinical Microbiology 28 495-503. Brown, D E 1983 The wolf in the Southwest the making of an endangered species. The Univwsity of Arizona Press, Tucson, Arizona, USA. BrowTvW M 1985 The mitochondrial genome of animals Pages 95-130//iR J Maclnryre, editor Molecular Evolutionary Genetics Plenum, New York, USA. Ciucci, P , L Boitani, E Raganella Pelliccioni, M Rocco, and I. Guy 1996 A comparison of scat-analysis methods to assess the diet of the wo\f Canis lupus Wildlife Biology 2 37-48 Chnrte, E W 1981 Biochemical analysis of mountain lion and bobcat scats: difference between species and sex Thesis, San Jose State University, San Jose, California, USA Danner, D .A , and N Dodd. 1982 Comparison of coyote and gray fox scat diameters Journal of Wildlife Management 46 240-241 Deuter, R , S Peitsch, S Hertel, and O Muller 1995 A method for preparation of faecal DN.A suitable for PCR Nuclei Acids Research 23 3800-3801 Eneroth, P , and J Sjovall 1969 Methodsof analysis in the biochemistry of bile acids Methods of Enzymology XV Academic Press, New ^'ork. New York, USA Ernest, R B , M C T Penedo, B P May, M Syvanen, and W M Boyce 2000 Molecular tracking of mountain lions in the Yosemite Valley region in California: genetic analysis using microsatellites and faecal DNA Molecular Ecology 9: 433-441 Floyd, T J , L D Mech, and P A. Jordan 1978 Relating wolf scat content to prey consumed Journal of Wildlife Management 42: 528-532 Foran, D R, K R Crooks, and S C Mima 1997 Species identification from scat: an unambiguous genetic method Wildlife Society Bulletin 25 835-839. Frantzen, M A J , J B Silk, J W H Ferguson, R K Wayne, and M H Kohn 1998 Empirical evaluation of preservation methods for faecal DNA. Molecular Ecology 7 1423-1428

33

Frenzel, L D 1974 Occurrence of moose in food of wolves as revealed by scat analyses: a review of North American stiidies Le Natiiraliste Canadien 101: 467-479 Fuller, T.K 1989 Population dynamics of wolves in north-central Minnesota. Wildlife Monograph 105 Garcia-Moreno. J . M D Matocq, M S Roy, E GefiFen, and R K Wayne 1996 Relationships and genetic purity of the endangered Mexican wolf based on analysisof microsatellite loci Conservation Biology 10: 376-389 Gerloff, U . C Schldtterer. K Rassmann, I. Rambold, G Hohmann, B Fmth, and D. Tautz 1995 Amplification of hypervariable simple sequence repeats (microsatellites) from excrememal DNA of wild living Bonobos (Pan paniscus) Molecular Ecology 4 515-518 Ginsberg, J R , and D W Macdonald 1990 Foxes, wolves, jackals and dogs An action plan for the conservation of canids World Conservation Union, Gland, Switzerland Gottelli, D , C Sillero-Zubiri, G D Applebaum, M S Roy. D J Girman, J. GarciaMoreno, E A. Ostrander, and R K Wayne 1994 Molecular genetics of the most endangered canid the Ethiopian wolf Ca?5 5/^1^^15/5 Molecular Ecology 3 301-312 Green, J S , and J T Flinders 1981 Diameter and pH comparisons of coyote and red fox scats Journal of Wildhfe Management 45 765-767 Halfpenny, J 1986 .A field guide to mammal tracking in North America Johnson Printing Company, Boulder, Colorado, USA Hall, R E , and K R Kelson 1959 The mammals of North .America The Ronald Press Company, New York, New York, USA Handt, O , M Hoss, M Krings, and S Paabo 1994 Ancient DNA - methodological challenges Experientia 50 524-529 Hoss, -M , M Kohn, and S Pdlbo 1992 Excrement analysis by PCR. Nature 359: 199 Hdss, M , and S Piabo 1993 DNA extraction from Pleistocene bones by a silica-based purification method Nucleic Acids Research 21 3913-3914 Hoss, M 1994 More about the sihca method Ancient DNA Newsletter 2: 10-12.

34

Johnson, M K , R C Belden, and D R Aked. 1984 Differentiating mountain lion and bobcat scats. Journal of Wildlife Management 48: 239-244. Kocher, T. D , W K Thomas, A Meyer, S V. Edwards, S Paabo, F. X Villablanca, and A. C Wilson 1989 Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers Proceedings of the National Academy of Sciences of tiie United States of America 86: 6196-6200. Kohn, M H . F Knauer, A StoflFella, W Schroder, and S Piiabo 1995 Conservation genetics of the European brown bear - a study using excremental PCR of nuclear and mitochondrial sequences Molecular Ecology 4: 95-103 Kohn, M H . and R K Wayne 1997 Facts from feces revisited Trends in Ecology and Evohition 12 223-227 Korschgen, L J 1980 Procedures for food-habits analyses Pages 113-127 mH S Mosby, R. H Giles, Jr, and S D Schemnitz, editors Wildlife management techniques manual, 4 * * * edition: revised The Wildlife Society, Washington, D C . USA Lehman, N , A Eisenhawer, K Hansen, L D Mech, R O Peterson, P J P Gogan, and R K Wayne 1991 Introgression of coyote mitochondrial DNA into sympatric North .American gray wolf populations Evolution 45 104-119 Lucchini, \'., E Fabbri, F Mamcco, S Ricci, L Boitani, and E Randi 2002 Noninvasive molecular tracking of colonizing wolf (Con/i lupus) packs in the western Italian Alps Molecular Ecology 11 857-868 Major, M . M K. Johnson, W S Davis, and T F Kellogg 1980 Identifying scats by recovery of bile acids Journal of Wildlife Management 44 290-293 .McBride, R T 1980 The Mexican wolf (Caw/5 lupus baileyi) a historical review and observations on its stams and distribution Endangered species report 8 Unrted States Fish and Wildlife Service, Albuquerque, New Mexico, USA .Mech, L D 1966 The wolves of Isle Royale, United States National Park Service Fauna Series Number 7 Mech, L D 1970 The wolf ecology and behavior of an endangered species University of MinnesoU Press, Minneapolis, Minnesota, USA. Miller, S A , D D Dykes, and H F Polesky 1988 A simple salting out procedure for extracting DNA from human nucleated cells Nucleic Acids Research 16: 1215.

35

Morin, P A . J Wallis, J J Moore. R Chakraborty, and D S Woodmff" 1993 Noninvasive sampling and DNA amplification for paternity exclusion, community stmcttire, and phylogeography m wild chimpanzees Primates 34 347-356. Moritz, C 1994 Applications of mitochondrial DNA in conservation: a critical review. Molecular Ecology 3 401 -411. Mullis, K B , andP A Faloona 1987 Specific synthesis of DNA/n v//ro via a polyraerase-catalysed chain reaction Methods Enzymol 155: 335-350 Mullis. K B . F Ferre. and R .A Gibbs, edhors 1994 The polymerase chain reaction Birkhauser, Boston, Massachusetts, USA Murie, A 1944 The wolves of Mount McKinley United States National Park Service Fauna Series, Number 5 Murie. O J 1954 Animal tracks, 2 * edhion Peterson Field Guides Houghton Mifflin Company. Boston, Massachusetts, USA Nowak, R. M 1995 Another look at wolf taxonomy Pages 375-397 in L N Carbyn, S H Fritts. and D R Seip, editors Ecology and conservation of wolves in a changing world Occasional Publication 35 Canadian Circumpyolar Instimte, Edmonton, Alberta, Canada Ott, L 1988 An introduction to statistical methods and data analysis, 3"^ edition PWSKENT Publishing Company. Boston, Massachusetts, US.A Paquet, P C 1992 Prey use strategies of sympatric wolves and coyotes in Riding Mountain National Park, Manitoba Journal of Mammalogy 73 337-343 Peterson, R O 1974 Wolf ecology and prey relationships on Isle Royale Dissertation Purdue L niversity, Lafayette, Louisiana, USA Phillips, M K . a n d D W Smith 1996 The wolves of Yellowstone Voyageur Press. Stillwater, .Minnesota, USA Pilgnm, K L , D K Boyd, and S H Forbes 1998 Testing for wolf-coyote hybridization in the Rocky Mountains using mitochondrial DNA Journal of W ildlife Management 62 683-689 Putman, R J 1984 Facts from faeces Mammal Review 14 79-97 Quinn, T , and W R Jackman 1994 Infiuenceof diet on detection of fecal bile acids by thin-layer chromatography Journal of Wildlife Management 58 295-299

36

Randi, E , L Gentile, G Boscagli, D Huber, and H. U Roth 1994 Mitochondrial DNA sequence divergence among some west European brown bear (Ursus arctos L.) populations Lessons for conservation Heredity 73 480-489. Reed, J Z , D J Tollit, P M Thompson, and W Amos 1997 Molecular scatology: the use of molecular genetic analysis to assign species, sex and individual identity to seal fifieces Molecular Ecology 6: 225-234 Saiki, R K . S Scharf, F Faloona, K B Mulhs, G T. Hom, H A Eriich, and N. Amheim 1985 Enzymatic amplification of B-globin genomic sequences and restriction she analysis for diagnosis of sickle cell anemia Science 230: 13501354 Saiki, R K , D R Gelfand, S Stoffel, S J Scharf, R Higuchi, G T Horn, K B Mullis, andH .A Eriich 1988 Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase Science 239 487-491 Sambrook, J . E F Fritsch, and T Maniatis 1989 Molecular cloning a laboratory manual. 2^ edition Cold Spring Harbour Laboratory Press. New York, USA Scott, T G 1941 Methods and computation in fecal analysis with reference to the red fox United States Department of the Imerior, Fish and Wildlife Service Iowa State College Journal of Science 15: 279-285 Scott, T G 1943 Some food coactions of the northern plains red fox Ecological Monographs 13 427-479 Slade, R W, C Moritz, and A Heideman 1994 Muhiple nuclear-gene phytogenies: apphcation to pinnipeds and comparison with a mitochondrial DNA gene phylogeny Molecular Biology and Evolution 11 341-356 Spaulding, R L , P R Krausman, and W B Ballard 1997 Summer diet of gray wolves, Canis lupus, in northwestern Alaska The Canadian Field-Naturalist 112: 262-266 Stephenson, R O , and L Johnson 1972 Wolf report, Alaska Federal Aid m Wildlife Research Report Project W-17-3, Juneau, Alaska, USA Stokes, D W , andL Q Stokes 1986 A guide to animal tracking and behavior Little, Brown and Company, Boston, Massachusetts, USA Taberlet, P , and J Bouvet 1992 Bear conservation genetics Nature 358 197

37

Taberlet, P , J. J E Swenson, F Sandegren, and A. Bjarvall. 1995. Localization of a contact zone between two highly divergent mitochondrial DNA lineages of the brown bear Ursus arctos in Scandinavia Conservation Biology 5: 1255-1261 Taberlet, P , S Griffin, B Goossens, S Questiau, V Manceau, N, Escaravage, L. P Waits, and J Bouvet 1996 Reliable genotyping of samples with very low DNA quantities using PCR Nucleic Acids Research 24 3189-3194. Taberlet. P , J -J Camarra, S Griffin, E Uhres, O Hanotte, L P Waits, C DuboisPaganon, T Burke, and J Bouvet 1997. Noninvasive genetic tiacking of the endangered Pyrenean brown bear population. Molecular Ecology 6: 869-876. Taberlet, P . L P Waits, and G Luikart 1999 Noninvasive genetic sampling: look before you leap Trends in Ecology and Evolution 14: 323-327. Thompson, D Q 1952 Travel, range, and food habits of timber wolves in Wisconsin Journal of Mammalogy 33: 429-442 Turkowski, F J 1980 Camivora food habits and habitat use in ponderosa pine forests United States Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station, Research Paper RM-215 United States Fish and Wildlife Service 1993 Biological evaluation for listed, proposed, and sensitive species of the forest-wide animal damage management plan on the Apache-Sitgreaves National Forests United States Fish and Wildlife Service. Phoenix, Arizona, USA United States Fish and W ildlife Service 1996 Rdntroduction of the Mexican wolf within its historic range in the southwestem United States Final environmental impact statement. United States Fish and Wildlife Service, Albuquerque, New Mexico, USA Van Ballenberghe, V , A W Erickson, and D Byman 1975 Ecology of the timber wolf in northeastern Minnesota Wildlife Monograph 43 \ ila, C , and R K Wayne 1999 Hybridization between wolves and dogs Conservation Biology 13: 195-198 Walsh, P S , D A Metzger, and R Riguchi 1991 Chelex* 100 as a medium for simple extt-action of DNA for PCR-based typing from forensic material BioTechniques 10: 506-513

38

Wasser, S K., C S Houston, G M Koehler, G G Cadd, and S R. Fain 1997 Techniques for application of faecal DNA methods to field studies in Ursids Molecular Ecology 6 1091-1097 W ayne, R K , N Uhman, M W Allard, and R. L Honeycutt 1992. Mitochondrial DNA variability of the gray wolf genetic consequences of population decline and habitat fiiigmentation Conservation Biology 6: 559-569. Weaver, J L , and S H Fritts 1979 Comparison of coyote and wolf scat diameters. Journal of Wildlife Management 43 786-788 W ilcomb, M S 1956 Fox populations and food habits in relation to game bird survival, W illiamette Valley, Oregon. Oregon State College Agricultursd Experiment Station Technical Bulletin 38 Williams, B K 1982. Some observations on the use of discriminant analysis in ecology Ecology 64 1283-1291. Woods, J G , D Paetkau, D Lewis, B N McLellan, M Proctor, and C Strobeck 1999 Genetic tagging of free-ranging black and brown bears Wildlife Society Bulletin 27 616-627

39

FIELD SCATS TISSUE Wl _Ci W2 C2

KNOWN SCATS Wl DJ 1

250 bp ^ 150 b p ^

Figure 3 1 Agarose gel showing a pxDrtion of the mtDNA control region (D-loop) amplified via Pilgrim et al (1998) canid-specific primers (Mexican wolf and dog, 164 bp; coyote 160 bp) Lanes 1 and 2 are field-collected scats Mexican wolf (WI) and coyote (CI) Lanes 3 and 4 are tissues Mexican wolf blood (W2) and coyote hver (C2) Lane 5 (S) is a 50 bp DNA size standard Lanes 6 through 8 are known scats: Mexican wolf (W I), domestic dog (Dl), and coyote (CI)

40

DIGESTED

UNDIGESTED

Wl

W2

CI

C2

C3

250 bp ^ 150bp-> 100bp->

Figure 3 2 Agarose gel showing restriction fragment digestion with BstN I of mtDNA control region (D-loop) purified polymerase chain reaction (PPCR) isolated from Mexican wolf and coyote scats The first lane is a 50 bp DN.A size standard Lanes 2 and 3 show digested Mexican wolf mtDNA (W1-W2, 113 bp), and the last 3 lanes show undigested coyote (C1-C3, 160 bp)

41

35 S mm 35 -

X 27 8 ^u^_r25 r =22 8mm


V5

x = 26.0 mm

2D -

174 mm

16.3 nun

Coyote

Species

Mexican wolf

Figure 3 3 Comparison of diameters of coyote scats ((// = 21. range 17 4 to 27 8 m, X = 22 8 mm) and Mexican wolf scats {n = 26, range 16 3 to 35 8 mm, x = 26 0 mm) idemified with fecal DN.A analysis

42

Table 3 1 Comparison of accuracy of predicting which species (Mexican wolf or coyote) deposited a scat using discriminant analysis based on combination of 3 measurements taken from scats identified to species with DNA analysis Scats were collected in southeastern Arizona and southwestem New MexicofromApril 1998 to October 2001
I 2 3 4 Classification Diameter (mm) Diameter and mass (g) Diameter and length Diameter, mass and length (cm) Coyote (%) 81 86 68 79 Mexican wolf (%) 50 65 59 55

43

Table 3.2. Locations of scats (n = 47) identified as Mexican wolf or coyote with DNA analysis Scats were collected in southeastern Arizona and southwestem New Mexico from April 1998 to October 2001
Location Forest trails Forest roads and highways Forest roads Forest 2-tracks Highway Elk carcass Mexican wolf den Opened release pen Otiier Total Mexican wolf 10 10 6 3 1 1 1 0 4 26 Coyote 11 6 4 2 0 0 0 2 2 21 1 1 2 6 47 Total 21 16

44

CHAPTER IV DIETS OF FREE-RANGING MEXICAN GRAY WOLVES IN ARIZONA AND NEW MEXICO

Abstract There were no systematic diet smdies of Mexican gray wolves (Canis lupus baileyi) before their extupation by the late 1960s from the southwestem United States. We collected carnivore scats (n = 1,682) from the Apache and Gila National Forests of Arizona and New Mexico from April 1998 through October 2001 and identified the scats to species using traditional field methods (i e , diameter, location, and sign) and odor We verified the accuracy of scat identification with fecal DNA analysis (molecular scatology, n = 26, Reed 2004) to determine the diets of free-ranging Mexican wolves in Arizona and New Mexico We analyzed scats > 28 mm diameter (n = 55) collected from areas where Mexican wolf packs (n = 6) were fed supplemental food items and determined if the packs consumed non-supplemental food items to detect a shift in diet as they acclimated to a wild existence We foimd a difference (Go,^ = 12 995, P = 0 023) m diet composition among packs, with the Cienega and Hawks Nest packs consuming more non-supplememal food hems than the Lupine, Campbell Blue, Mule and Pipestem packs. We analyzed Mexican wolf scats (w = 251) identified with our refined traditional field methods established with DNA analysis (Reed 2004) and found the diet composition of tiiose scats consisted of large-sized food items (92 8% percent frequency of occurrence [PFO]), primarily elk (Cervus elaphus canadensis [nelsonij) adults (36 6% PFO) and

45

calves (36 2% PFO). We found no difference in Mexican wolf diets among years (n = 4, God, = 2 588, P = 0 460). between seasons (n - 2. G^j, = 0 490, P = 0 484), or among packs (n = 4. G.^= 7.719, P = 0.052) We analyzed Mexican wolf scats (n = 26) identified witii DNA analysis and compared the resuhs to tiie diet composition of Mexican wolf scats (n = 251) identified witii our refined ti-aditional methods. There was a difference (G*^ = 9 761, P = 0 002) in diet composition between the 2 methods, with tiie scats idemified witii DNA analysis composed of fewer large-sized food items (72.7% PFO) than found in the scats identified with our refined tiaditional methods (92.8% PFO) However, we found no difference (/ = - I 95, f = 0 001) between diameter means (refined tt-aditional method x = 29 3 mm, DNA analysis r = 26 0 mm) and no difference (/ = 1 50,P = 0 005) between number of food items per scat for the 2 identification methods (refined traditional method = 1 06 food items per scat, DNA analysis 1 27 food items per scat) We then compared diet composition found in scats identified with DNA analysis as Mexican wolf scats (n = 26) and sympatric coyote scats (C. latrans: n = 2l) and found a difference (Gajj = 9 647, P = 0002) Mexican wolf diets consisted primarily of large-sized food items (72 7% PFO), while coyote diets were more variable and composed of more medium- and small-sized food items (63 4% PFO) There was a difference (/ = -2 428, P = 0 019) between diameter means for the 2 species (Mexican wolf X = 26 0 mm, coyote x = 22 8 mm. Reed 2004) There was also a difference (/ = 2 849, P = 0 007) between number of food items per scat for the 2 species (Mexican wolf X = 1 27 food hems per scat, coyote x = I 95 food items per scat) We compared the diet composition of Mexican wolf scats (n = 277) identified with our refined traditional field

46

methods (n = 251) and DNA analysis (n = 26) to tiie diets reported in previous North American gray wolf (C. lupus) stiidies (n = 7) and found a difference (Gadj = 462,492, P = < 0 0001) Mexican wolf diet consisted primarily of large-sized food items (90.6% PFO), which differed from late spring and summer- and carcass-based diets previously reported for North American gray wolves Our results suggest that free-ranging Mexican wolves consumed more large-sized food items than either sympatric coyotes or other North .American gray wolves

Introduction Little is known of the Mexican gray wolfs (Ccaiis lupus baileyi) natural history because systematic studies were not conducted on the subspecies before it was extirpated from the wild (Brown 1983) The Mexican wolf is the smallest and southern-most occurring gray wolf (C lupus) of North America (Young and (joldman 1944, Hall and Kelson 1959, Bogan and Mehlhop 1983, Nowak 1995) The subspecies is also the most genetically distinct (Wayne et al 1992, Garcia-Moreno et al 1996) and the most endangered (McBride 1980, Brown 1983, Bednarz 1988, Ginsberg and Macdonald 1990) North .American gray wolf The United States Fish and Wildlife Service (USFWS) began releasing captive-reared Mexican wolves in April 1998 into the Blue Range Wolf Recovery Area (BRWRA) in Arizona and New Mexico The purpose of our smdy was to determine the diets of free-ranging Mexican gray wolves in Arizona and New Mexico and to determine if there were any differences among years, between seasons, or among packs We also determined if non-

47

supplemental food items were consumed by Mexican wolves while they were fed supplemental food items From the resuhs of our fecal DNA analysis (molecular scatology. Reed 2004). we were able to compare the diet composhion of Mexican wolves with that of sympatric coyotes (Ca/iis latrans) Lastiy we compared the diet composition of Mexican wohes to that reported in 7 previous diet studies of northern gray wolves This information could prove beneficial for managing Mexican wolf and prey populations and investigating depredation incidences in areas where Mexican wolves are recovering in the southwestem United States

Study Area We conducted our research within the BRWTIA, which encompassed 17,700 km and included all of the .Apache and Gila National Forests in east-central Arizona and west-central New Mexico The primary recovery zone (2,600 km') in Arizona, in which capti\e-reared Mexican wolves were released since April 1998. was our principal focus In Spring 2000, our research efforts extended to the secondary recovery zone (Gila National Forest and Wilderness, New Mexico) due to translocations of Mexican wolves previously released in .Arizona The White Mountain (Fort Apache) and San Carlos Indian Reservations bordered the BRWRA on the west and private lands bordered the area to the east, north and south, and were scattered within public lands Domestic cattle grazed almost all areas (USFS, unpublished data) The BRWRA s elevations ranged from about 1,200 m along the San Francisco River to the 3,350 m mountain ranges of Mount Baldy. Escudilla, and Mogollon (United

48

States Fish and Wildlife Service 1996) Lower elevations were characterized by rolling hills with moderately steep-walled canyons and sandy washes, while higher elevations were typified by mgged slopes, deep canyons, elevated mesas, and rock cliffs (Umted States Fish and Wildlife Service 1996) Dominant vegetation included ponderosa pine (Pittus ponderosa), aspen (Popuhs tremuloides), fir (Pseudotsuga menziesii and Abies spp ), juniper (Juniperus spp ), piflon (Pinus cembroides), mesquhe (Prosopis spp ), evergreen oaks (Quercus spp ), and a variety of grasses and forbs (United States Fish and Wildlife Service 1996) Annual temperatures averaged 16 4C maximum and-3.1C minimum and most precipitation fell during thunderstorms (annual average 52.1 cm) from July through September and snow (annual average 139 3 cm) from December through March (Desert Research Institute, Western Region Climate Center, Reno, Nevada, unpublished data) From April 1998 through October 2001, USFWS reported > 87 Mexican wolves either released from captivity or bom in the wild within the BRWRA Approximately 37 Mexican wolves were free-ranging as of October 2(X)1 and 31 of those individuals were fitted with radio collars Other predators within the study area included humans (Homo sapiens), mountain lion (Puma concolor), black bear (Ursus americanus), coyote, bobcat (Felis rufus), and gray (Urocyon cmereoargenleus) and red (Vulpes vulpes) fox (United States Fish and Wildlife Service 1993, Arizona Game and Fish Department 1994) Density estimates were unavailable for these species Potential large-sized prey within the BRWRA included Rocky Mountain elk (Cerxvs elaphus canadensis [nelsoni]), Coues white-tailed (Odocoileus virginianus

49

couesi) and desert mule (O. hemionus eremicus) deer, pronghom (Antilocapra americana\ and Rocky Mountain bighorn sheep (Ovis canadensis. United States Fish and Wildlife Service 1996). It was estimated thattiieBRWRA supported 57.000 deer and 16,000 elk (Parsons 1998); however, the densities of these ungulates were unknown during our study. Potential medium- and small-sized prey included javelina (Dicotyles kyacu. United States Fish and Wildlife Service 1996), beaver (Castor canadensis), jackrabbit (Lepus spp ), cottontail rabbit (Syhnlagus spp ), skunk (Mephitis spp), various tree (Schmts and Tamiasciurus spp ) and ground squirrels (Spermophilus spp), chipmunk (Eutamiasspp),rat(Neotomaspp),mice(Per<miyscusspp), vole(Microtusspp), and

other small mammals (Hofifmeister 1986, Southwest Region U S Forest Service 1992), porcupines (Erethizon dorsatum), and Merriam's turkey (Meleagris gaJlopavo, Groebner etal. 1995). Elk calving occurred during May and June, mule deer fawning occurred June through August, and white-tailed deerfiiwningoccurred during August (AGFD, unpublished data. New Mexico Department of Game and Fish 19982001). Native ungulate harvesting occurred intermittently from late August through January for whitetailed and mule deer and from September through mid-December for elk (Arizona Game and Fish Departmem 19982001, New Mexico Department of Game and Fish 1998 2001), which could have resulted in carrion for Mexican wolves.

50

Methods We collected 1.682 carnivore scats from April 1998 through October 2001 from areas where captive-released, translocated, and wild-bom Mexican wolves were known to frequent within the BRWRA on the Apache and Gila National Forests of Arizona and New Mexico Personnel of the Mexican wolf Interagency Field Team (IFT, i e , U S Fish and W ildlife Service, .Arizona Game and Fish Department, New Mexico Department of Game and Fish, USDA APHIS Wildlife Services, and U S Forest Service) collected scats opportumstically from April 1998 through October 2001 and we collected scats actively from June through August 20(X) and March through October 2001 An opportunistic sampling strategy (Frenzel 1974) was employed as the sampling areas followed Mexican wolves as they moved within the study area We collected scats along forest roads, trails, ridgelines, and riparian areas, and from opened release pens, campsites, den sites, and kill and carcass sues Scats collected by 4 Drag Ranch personnel in areas where Mexican wolves were repyortedly depredating cattle were not available for this smdy Scat collectors wore disposable mbber or food preparation gloves and placed the scats in brown paper bags, which were labeled with date, location, and scat number Scats were aged as old, recent, or fresh based on appearance, exposure of deposition site, and weather conditions (Ciucci et al 1996) We identified scats to carnivore species using traditional field methods (i e , diameter, location, and sign) reported in the hterature and odor There were no reports of Echinococcus granulosus in Arizona or New Mexico, therefore we did not autoclave the scats The bagged scats were ah dried before storing

51

at room temperatiire in large plastic comainers until fiirther analysis We used DNA analysis to verify traditional scat idemification methods for a subsample of scats (Reed 2004). From our DNA analysis resuhs, we refined the tiaditional scat identification techniques as described below The location criterion was defined as den sites (n = 2) only We measured the maximum diameter of each dried scat (Scott 1943, Weaver and Fritts 1979, Green and Flinders 1981. Danner and Dodd 1982) using 152 mm dial calipers (General Tools Manufacturing Co . LLC. New ^ ork. New York) However, we took 2 measurements to the nearest 0 I mm and used the average for diameter size The diameter criterion for identifying gray wolf scats was > 24 mm diameter established by Thompson (1952) and has been accepted for several smdies (Mech 1970. Stephenson and Johnson 1972. Peterson 1974, Nan Ballenberghe et al 1975) Weaver and Fritts (1979) suggested scat diameters > 30 mm be used to identify gray wolf scats Halfpenny (1986) proposed > 25 mm diameters for identifying gray wolf scats after reviewing 3 studies For identifying Mexican wolf scats in the field, we chose the more conservative > 30 mm diameter criterion, then redefined it as > 28 mm based on results from DNA analysis (Reed 2004). We used wolf tracks and visual observations for sign criterion Some wolf biologists believe that they can identify wolf scats by odor (Halfpenny 1986) We attempted to idemify Mexican wolf scats based on odor ( e g , a sweet, musky odor) as instmcted by .Mexican wolf IFT personnel No distinguishing odor for coyote scats was reported. We established a reference collection of potemial prey items from the BRWRA We trapped and euthanized small prey and collected road-killed small prey, which were

52

prepared as museum specimens Hair samples were taken from various body areas (eg., mmp, hind legs, belly, back, and neck) of road-killed elk and deer Elk calf hair samples were taken from the carcass remains of a known Mexican wolf kill. Fawn hair samples were taken from a 2-month-old mule deer orphan residing at Sipes White Mountain Wildlife Refijge, Springerville, Arizona. We estimated the diets of Mexican wolves from 332 of the 1,682 scats. Twentyone additional scats were positively identified as coyote scats with DNA analysis (Reed 2004) The remaining 1,329 scats were < 28 mm diameter and could not be positively identified to carnivore species These scats were excluded from this diet study and were analyzed for another We analyzed 55 scats with diameters > 28 mm collected from areas where Mexican wolves were fed supplemental food items (i e , 11 1 kg carnivore logs [zoo diet formulation for wild canids. Central Nebraska Packing. Inc . North Platte, Nebraska, USA] and road-killed elk, deer, and jackrabbit) By comparing food items found in these scats with supplemental food records (United States Fish and Wildlife Service [USFWS], unpublished data), we were able to determine if any non-supplemental food (i e , elk aduhs and calves, deer adults and fawns, domestic bovine, and insects) were consumed. We determined overall diet of non-supplemental fed Mexican wolves by analyzing scats (n = 25\) identified as deposhed by the subspecies based on > 28 mm diameter (Reed 2004), den site location, and sign criteria, plus the scats (n = 26) identified as deposhed by Mexican wolves with DNA analysis (Reed 2004). The 251 scats identified by our refined traditional methods were used to determine differences

53

among years, between seasons, and among packs. We then compared the diet composition of the Mexican wolf scats identified with our refined fradhional methods to scats identified with DNA analysis We also compared the diet composition found in Mexican wolf scats (n = 26) to coyote scats (n = 21) identified with DNA analysis (Reed 2004) to determine a difference between the 2 species Mexican wolf scats identified with our refined tt^ditional methods (n = 251) and DNA analysis (n = 26) were combined for comparison with 7 North American gray wolf diet studies to determme any differences in diet composition Scats were broken apart by hand and undigested food items (i e , hak, bone, teeth, claw s, and hooves) were separated We identified the undigested food items macroscopically by comparing to our reference collection We calculated frequency of occurrence (FO) of prey items (i e . number of occurrences of a prey item divided by the total number of scats) and percent frequency of occurrence (PFO) of prey items (i e , number of occurrences of a prey item divided by the total number of occurrences of all prey items) We used PFO for all statistical analyses. To fiacilitate statistical comparisons, we pooled ungulate prey (i e , adult and neonate elk and deer, and domestic bovine) and considered them large-sized food hems We then pooled smaller food items (i e , medium- and small-sized mammals, birds, reptiles, insects, and vegetation) and considered them medium- and small-sized food hems. We did not include non-food hems (e g , rocks, sticks, pine needles, string, plastic, mbber, paper, and ahiminum foil) We used /-tests to examine differences in diameter means and number of food items per scat between Mexican wolf scats identified whh our refined

54

tt^ditional identification methods and DNA analysis, and between Mexican wolf and sympatric coyote scats identified with DNA analysis We used chi-square likelihood ratio contingency-table analysis (G-test. Ott 1988) corrected for continuity (Williams 1976) to determine the presence of non-supplemental food items found in scats identified with our refined traditional techniques collected where Mexican wolves were fed supplemental food items and to determine differences in diet composition among years (19982001), between 2 seasons (spring-summer MarchAugust, fall-winter: SeptemberFebmary), between packs (n = 4). between scat identification methods (refined traditional and DN.A analysis), between species (Mexican wolves and sympatric coyotes), and between Mexican wolves and northern gray wolves All tests were considered significant at the probability level of/* ^ 0 05

Results We found a difference {Godj - 12 995. P = 0 023. Table 4 1) in diet composhion of scats (n = 55) > 28 mm diameter collected from areas where Mexican wolf packs (n = 6) were fed supplemental food items Two packs, Cienega and Hawks Nest, consumed non-supplemental food items while being fed supplemental food items, and Lupine, Campbell, Mule and Pipestem packs did not (Table 4 2) Diet composition of Mexican wolf scats (n = 251) identified with our refined traditional field methods consisted mainly of native ungulates (88 6% PFO, Table 4 3). Smaller mammalian prey (5 3% PFO), birds (0 4% PFO), insects (0,8% PFO), and vegetation (0 8% PFO) were of lesser importance Cattle comprised 4 2% PFO We

55

found no differences in diet among years (n = A,G^ = 2 588, P = 0 460, Table 4.4), between seasons (n = 2. G* = 0 490, P = 0 484. Table 4 5), or among packs (// = 4, C/a4=7 719, P = 0 052. Table 4 6) All 3 comparisons indicated Mexican wolf diets were composed mainly of large-sized food hems Diet composhion of Mexican wolf scats (n = 26) identified with DNA analysis consisted of a variable diet composed of large-sized food items (72 7% PFO; Table 4 7) and medium- and small-sized food hems (27 3'o PFO) We compared these results with the diet composition of the scats identified with our refined traditional methods and found a difference (Ga^ = 9 761. P = 0 002, Table 4 8) in diets based on scat identification techniques Diet composhion of scats identified with our refined tradhional methods consisted of more large-sized food items (92 8o PFO) and fewer medium- and smallsized food Items (7 2o PFO) than found in the scats identified by DN.A analysis We found no difference between diameter means (I = -\ 95, P =0 052) for the 2 scat idemification methods (refined traditional method x = 29 3 mm diameter, DNA analysis X = 26 0 mm diameter) We also found no difference between number of food hems per Mexican wolf scat (t = I 50, P = 0 135) for the 2 scat identification methods (1 06 food hems per scat identified with our refined traditional methods. Table 4 3, 1 27 food hems per scat identified with DN.A analysis. Table 4 7) We found a difference (G^d, = 9 647, P = 0 002, Table 4 9) in diet composhion found in Mexican wolf (n = 26) and sympatric coyote (/= 21) scats when the 2 species were poshively identified with DNA analysis (Reed 2004). Mexican wolf diets consisted mainly of large-sized food rtems (72 7% PFO), while coyote diet had a more variable diet

56

consisting of medium- and small-sized food items (63.4% PFO). We also found a difference (t = 2.849, P = 0.007) in number of food hems per scat for the 2 species (1.27 food items per Mexican wolf scat; 1 95 food items per coyote scat. Table 4.9). We pooled scats identified with DNA analysis (n = 26) and those identified with our refined traditional field techniques (n = 251) to compare witii previous North American gray wolf diet studies (n = 7). We found a difference (Gad/ = 462.492, P = < 0.0001, Table 4.11) between the diets of Mexican wolves and other North American gray wolves, with Mexican wolves consuming a higher percentage of largesizedfixKlitems (90.6% PFO) than reported for North American gray wolves (range 58 5% to 82 9% PFO).

Discussion Most early researchers of the Mexican wolf based their assessment of the subspecies' diet on field observations and stomach analysis (Young and Goldman 1944). Bailey (1931) reported the subspecies fed primarily on cattie, as determined from undigested hair found in scats. Leopold (19S9) hypothesized that the Mexican wolfs diet consisted mainly of deer, but included bighorn sheep, pronghom, javelina, rabbits, rodents, and some plam foods. McBride (1980) reported that deer were the principal native prey of Mexican wolves, but they also consumed pronghom, rabbhs, and mice. Brown (1983) speculated that the Mexican wolf historically preyed almost entu-ely on the diminutive Coues white-tailed deer Parsons (1996) acknowledged that the Mexican wolfs historic diets were not well documented, but suggested that the subspecies preyed

57

primarily on whrte-tailed and mule deer, and ahematively on elk, pronghom, javelina, beaver, rabbits, hares, and other small mammals Introduced domestic livestock increased in Arizona and New Mexico from 1880 to 1890 and native prey populations decreased due to unregulated subsistence and market hunting (Unhed States Fish and Wildlife Service 1987) Historically, the BRWRA had a plentiftil supply of elk, deer, and otiier wild game (Shumway 1998). By 1890 the native elk (Cer\tis merriami) were extinct and current populations of Rocky Mountain elk were tt^nslocated to tiie area around 1925 (Nelson 1902, Shumway 1998) With the reduction of the large native ungulate prey base, the Mexican wolf reportedly turned to the more abundam and more-easily caught livestock (Young and Goldman 1944, Brown 1983, Parsons 1996) Leopold (1959) indicated that not all wolves were livestock killers and that such behavior was that of particular packs or individuals Historical records of the Mexican wolfs diets during the extirpation campaign focused primarily on wolves that depredated domestic cattle and the reports may have been exaggerated (Gipson and Ballard 1998) To smdy the diets of free-ranging Mexican wolves, it was necessary to select a sampling technique that would not interfere with recovery efforts Scat analysis is a noninvasive field technique widely used in determining carnivore diets, and scats are readily available and easily collected (Scott 1941, Putman 1984) However, carnivore scats must be identified accurately to species for scat analysis to be a valid method of determining the diet of the species under smdy (Reed 2004)

58

Frequency of occurrence of food hems per scat is the most commonly used and easily applied method of diet analysis (Leopold and Krausman 1986, Spaulding et al 1997). although this method may overemphasize tiie frequency of small prey (Corbett 1989) We used the PFO method for this study because our objective was to determine the diet composhion and relative amounts of food items consumed by Mexican wolves to compare with previous North American gray wolf diet studies If PFO did over emphasize smaller food rtems in the diet composhion of Mexican wolves, h was not apparent in our resuhs Diet analysis based upon scats identified with DNA analyses as deposhed by Mexican wolves and sympatric coyotes revealed a difference between the 2 species' diets, which was consistent with Thurber et al (1992) and Arjo et al (2002) who found that coyote diets contained more small-sized food items than sympatric gray wolf diets This could be contributed to morphological and social differences between the 2 species. Coyotes are medium-sized carnivores that usually prey on smaller animals (Mech 1970, Pilgrim et al 1998), and the presence of ungulate remains in their scats may resuh from feeding on carrion (Ozoga and Harger 1966, Berg and Chesness 1978, Weaver 1979) Wolves, on the other hand, are 3 to 4X the mass of coyotes and are more likely to hunt in cooperative packs than coyotes, which allows preying on larger animals (Mech 1970, Ballard et ai 1987, .Moehlman 1989, Thurt)er and Peterson 1993, Dale et al 1995) In our smdy area, Mexican wolves weighed 22 7 to 40 9 kg (Unhed States Fish and Wildlife Service 1996) and southwestem coyotes averaged 9 5 to 10 6 kg (Woolsey 1985)

59

Although our sample sizes were small, DNA analysis of scats may provide a more comprehensive representation of wolf and coyote diets than analysis of scats identified to species by ttadhional methods used by previous diet studies. Early northern gray wolf diet smdies (Murie 1944. Cowan 1947. Thompson 1952, Mech 1966, Pimlott et al. 1969) supported the conclusion that the wolfs diet consisted mainly of ungulate prey either killed or scavenged (acmal predation and carrion feedmg cannot be distinguished by scat analysis [Pimlott 1967]), as did more recent studies (Ballard et al 1987, Arjo et al. 2002). Although Mech (1970) reported that wolves prey on sick, weak, and unfit prey. Brown (1983) reported that there was no evidence in southwestem historic records that indicated such hunting behavior by Mexican wolves However, during the first year of Mexican wolf recovery efforts, all confirmed Mexican wolf prey carcasses were elk and most remains were young of the year and old or injured individuals (Parsons 1998), and the first documented native prey kill was by the Hawks Nest pack whhin 2 weeks after tiieir release (USFWS, unpublished data) From May 1998tiiroughOctober 2001, Defenders of Wildlife-Bailey Wildlife Foundation Wolf Compensation Tmst (unpublished data) paid $16,612 (US) for confirmed and unconfirmed Mexican wolfrelated injuries to 2 horses, 1 calf, and 1 guard dog, and deaths of 1 bull, 2 cows, 1 heifer, 11 calves, and 1 herding dog within Arizona and New Mexico Our resuhs combined with USFWS kill reports and Defenders of Wildlife-Bailey Wildlife Foundation Wolf Compensation Trust reports indicated that Mexican wolves consumed more large-sized native ungulates than domestic livestock

60

Mech (1970) and Messier and Crete (1985) indicated wolves concentrated on the smallest or easiest to catch large prey species in areas where > 2 large prey species inhabrt the same area Our resuhs suggest that Mexican wolves were consuming primarily adult elk (males ^ 454 5 kg, females ^ 227 3 kg, AGFD, unpublished data) and calves, the larger of tiie native ungulate species available within the BRWRA, and not the smaller white-tailed deer (males < 56 8 kg, females <. 36 4 kg; AGFD, unpublished data) or mule deer (males < 102 3 kg, females ^ 56 8 kg, AGFD, unpublished dau). When compared to previous diet studies for North American gray wolves, our resuhs suggest that Mexican gray wolves consumed a higher proportion of large-sized native ungulates than their northern counterparts This may have been the result of our conservative identification of Mexican wolf scats using diameters ^ 28 mm, which may have biased our diet analysis results towards the larger prey commonly found in large diameter scats (Daimer and Dodd 1982) Additionally, previous gray wolf diet smdies included scats collected from den and rendezvous sites and kill or carcass shes, while our Mexican wolf scats were collected from all locations frequented by the subspecies, during all seasons, and 2 den srtes Our results indicate that elk aduhs and calves were the primary food source for Nfexican wolves in Arizona and New Mexico from April 1998 through October 2001 Gray wolves in differem areas rely on different prey, and usually the wolfs diet is compnsed of 1 or 2 species (Mech 1970) Elk were the most abundant and largest prey available whhin the BRWRA (AGFD, unpublished data, USFS, unpubhshed data) Although we did not study prey selection, we were curious why Mexican wolves

61

consumed mostiy elk aduhs and calves We can only hypothesize based on previous lueramre of prey selection studies of gray wolves First, the density of elk within Arizona and New Mexico was unknown, however, elk densrty was reported to be high (Ballard et al 1998. AGFD, unpublished data) Therefore, h is unknown if the elk herd within the BRWRA had neared or reached ecological carrying capacity. Furthermore, we did not know if elk were subjected to winter severity (Mech et al 2001), malnutrition, disease, or injunes or death from human activity ( e g , hunting and vehicles) that possibly played a role in the Mexican wolfs consumption of elk One reason that a large proportion of elk remains were found in Mexican wolf scats may have been because of the low-densrty, early colonizing stage of wolves (Fritts and Mech 1981. Boyd et al 1994) These authors, as did Carbyn (1974. 1983) and Huggard (1992), found that wolves primarily preyed upon the most vulnerable ungulates juvenile, old, or post-mt males These data and ours suggest low-density, colonizing Mexican wolves may have preyed upon vulnerable elk at a higher rate than other wolves in established northern populations Another reason why Mexican wolves consumed primarily elk may be because of wolf-naive elk Although the resident elk population had been exposed to other predators (i e , mountain lions, black bears, and coyotes), the elk had not been exposed to wolves. This may have increased the vulnerability of resident elk that had never seen wolves until confronted with them in 1998 Naive prey confronted with new predators have been less wary than prey previously exposed to such danger (Byers 1998, Berger 1999, 2001)

62

To answer why Mexican wolves consumed primarily elk, ftuther study would be required Mexican wolves are 1 of the 4 major predator species in Arizona and New Mexico, and little research has been conducted on the other wild carnivores (i.e., mountain lion, black bear, and coyote) that now share the same prey base A muhicarmvore prey selection study would provide the information necessary for optimal management of predators and then prey within Arizona and New Mexico Our estimates of the diets of free-ranging Mexican gray wolves in Arizona and New Mexico were based on a small sample of scats This is a result of our conservative approach to identif>ing Mexican wolf scats based on DNA analysis and refined traditional identification methods (i e , > 28 mm diameter, den site locations, and tracks) Our results suggest that our refined traditional scat identification methods may have biased estimates of the diets of Mexican wolves toward large ungulate food items WTien researching the diets of colonizing and established wolf populations, h is important to provide wildlife managers and the public whh accurate information We recommend that future gray wolf diet studies incorporate DNA analysis of scats to ensure that only gray wolf scats are included in diet analyses for the species This approach could prove beneficial in resolving human-wolf conflicts in areas where gray wolves are reintroduced or recovering

63

Acknowledgements The Unhed States Fish and Wildlife Service funded this research The Mexican wolf Interagency Field Team (IFT, i e , US. Fish and Wildlife Service, Arizona Game and Fish Department, New Mexico Department of Game and Fish, USDA APHIS Wildlife Services, and U S Forest Service) p)ersonnel assisted with the collection of scats and provided Mexican wolf location data The United States Forest Service Alpine and Clifton Ranger District personnel assisted whh logistics This is a Texas Tech University, College of Agricultural Sciences and Natural Resources technical publication T-X-XXX

64

Literatiire Ched Arizona Game and Fish Departmem 1994 Predator population estimates Unpublished report. Arizona Game and Fish I>epartment, Phoenix, Arizona, USA. .\rizona Game and Fish Departmem 1998-2001. Arizona hunting regulations. Arizona Game and Fish Department, Phoenix, Arizona, USA. -Arjo, W M . D H Pletscher, and R R Ream 2002 Dietary overlap between wolves and coyotes in Northwestern Montana Journal of Mammalogy 83: 754-766. Bailey. V 1931. Mammals of New Mexico United States Department of Agriculttire, Bureau of Biological Survey of North American Fauna No 53 Ballard, W B . H A Whitlaw, B F Wakeling, R L Brown, J C deVos, Jr, and M C. Wallace 1998 Survival of female elk in northern Arizona Journal of Wildlife Management 64 500-504 Ballard, W B . J S Whitman, and C L Gardner 1987 Ecology of an exploited wolf population in south-central Alaska Wildlife Monograph 98 Bednarz, J C 1988 The Mexican wolf biology, history, and prospects for reestabhshment in New Mexico Endangered Species Report No 18 United Slates Fish and Wildlife Service, Albuquerque, New Mexico. USA Berg, W E , and R A Chesness 1978 Ecology of the coyote in northern Minnesota Pages 229-247 m M Bekoff, editor Coyotes biology, behavior, and management Academic Press, New York, USA Berger, J 1999 .Anthropegenic extinction of top carnivores and interspecific animal behaviour implications of the rapid decoupling of a web involving wolves, bears, moose and ravens Proceedings of Royal Society of London, Series B, 266: 22612267.

Berger, J , J E Swenson, and I Per-Illson 2001 Recolonizing carnivores and naive prey conservation lessons from Pleistocene extinctions Science 291 1036-1039. Bogan, M A . and P Mehlhop 1983 Systematic relationships of gray wolves (Caw/5 lupus) in southwestem North America Occasional Papers of the Museum of Southwestem Biology 1 Boyd, D K , R R Ream, D H Pletscher, and M W Fairchild 1994 Prey taken by colonizing wolves and hunters in the Glacier National Park area Journal of Wildlife Management 58 289-295

65

Brown, D E 1983 The wolf in tiie Southwest: tiie making of an endangered species. The Universrty of Arizona Press, Tucson, Arizona, USA. Byers, J A 1998. American pronghom: social adaptations and ghosts of predators past. University of Chicago Press, Chicago, Dlinois, USA CarbvTi, L N 1974 Wolf predation and behavioral interactions with elk and otiier ungulates in an area of high prey diversity Dissertation, Universrty of Toronto, Ontario, Canada Carbyn, L N 1983 Wolf predation on elk in Riding Mountain National Park, Manitoba Journal of Wildlife Management 47: 963-976 Ciucci, P , L Bortani, E Raganella Pelliccioni, M Rocco, and I Guy 1996. A comparison of scat analysis methods to assess the diet of the wolf Can/5 lupus Wildlife Biology 2 37-48 Corbett, L. K 1989 Assessing the diet of dingoes from feces a comparison of 3 methods Journal of Wildhfe Management 53 343-346 Cowan, I M 1947 The timber wolf in the Rocky Mountain national parks of Canada. Canadian Journal of Research 25 139-174 Dale. B W , L G Adams, and R T Bowyer 1995 Wimer wolf predation in a muhiple ungulate prey system. Gates of the Arctic National Park, Alaska Pages 223-230 m L N Carbyn, S H Fritts, and D R Seip. edhors Ecology and conservation of wolves in a changing world Canadian Circumpolar Instimte Occasional Publication 35 Danner. D A , and N Dodd 1982 Comparison of coyote and gray fox scat diameters. Journal of Wildlife Management 46 240-241 Frenzel, L D 1974 Occurrence of moose in food of wolves as revealed by scat analyses a review of North American smdies Le Naturaliste Canadien 101: 467-479 Fritts, S H , andL D Mech 1981 Dynamics, movements, and feedmg ecology of a newly protected wolf population in northwestern Minnesota Wildlife Monograph 80 Garcia-Moreno, J , M D Matocq, M S Roy, E Geflfen, and R K Wayne 1996 Relationships and genetic purity of the endangered Mexican wolf based on analysisof microsatellhe loci Conservation Biology 10 376-389,

66

Gmsberg, J R . and D W Macdonald 1990 Foxes, wolves, jackals, and dogs. An action plan for the conservation of canids Worid Conservation Union, Gland, Switzerland Gipson, P S . and W B Ballard 1998 Accounts of famous North American wolves. The Canadian Field-Nattiralist 112: 724-739 Green, J S , and J T Flinders 1981 EMameter and pH comparisons of coyote and red fox scats Journal of Wildhfe Management 45: 765-767 Groebner, D J, A L Girmendonk, and T B Johnson 1995 A proposed cooperative reintroduction plan for the Mexican wolf in Arizona Arizona Game and Fish Departmem, Phoenix, Technical Report 56 Halfipenny, J. 1986 A field guide to mammal tracking in North America. Johnson Printing Company, Boulder, Colorado, USA Hall, R E , and K R Kelson 1959 The mammals of North America The Ronald Press, New York, New York, USA Hoffmeister, D F 1986 Mammals of Arizona University of Arizona Press and Arizona Game and Fish Department, Tucson, .Arizona, USA Hoss, M , M Kohn, and S Paabo 1992 Excrement analysis by PCR Namre 359: 199. Huggard, D J 1992 Prey selectivity of wolves in Banff National Park II Age, sex, and condhion of elk Canadian Journal of Zoology 71:140-147 Leopold, A S 1959 Wildlife of Mexico the game birds and mammals University of California Press, Berkeley and Los Angeles, California, USA. Leopold, B D , and P R Krausman 1986 Diets of 3 predators in Big Bend National Park, Texas Journal of W ildlife Management 50 290-295 McBride, R T 1980 The Mexican wolf ((awi5/pM5 ^a/Vey;) Unhed States I>epartment of the Interior, Fish and Wildlife Service Endangered Species Report No 8 Mech, L. D 1966 The wolves of Isle Royale United States Department of the Interior, National Park Service Fauna Series No 7 Mech, L. D 1970 The wolf the ecology and behavior of an endangered species Universrty of Minnesota Press, Minneapolis, Minnesota, USA

67

Mech, L D . D W Smith, K M Murphy, and D R. MacNuhy 2001. Winter severity and wolf predation on a formerly wolf-free elk herd Journal of Wildlife Management 65 998-1003 Messier, F , and M Crete 1985. Moose-wolfpopulationdynamicsand the natural regulation of moose populations Oecologia(Berhn) 65: 503-512. Moehlman, P D 1989 Intt-aspecific variation in canid social units Pages 143-163 in J. L Gittleman, edrtor Carnivore behavior, ecology, and evolution. Cornell Umversrty Press, Ithaca, New York, USA Murie. A 1944 The wolves of Mount McKinley United States National Park Service Fauna Series No 5 Nelson, E W 1902 A new species of elk from Arizona Bulletin of the American Museum of Natural History 16 1-12 New Mexico Department of Game and Fish 19982(X)1 Big Game and Furbearer Rules and Information New Mexico Department of Game and Fish, Santa Fe, New Mexico, US.A Nowak, R M 1995 .Another look at wolf taxonomy Pages 375-397 m L N Carbyn, S H Fritts. and D R Seip, editors Ecology and conservation of wolves in a changing world Occasional Publication No 35 Canadian Chcumpolar Institute, Edmonton, Alberta, Canada Ozoga, J L , and E M Harger 1966 Winter activhies and feeding habrts of northern Michigan coyotes Journal of Wildlife Management 30 809-818 Ott, L 1988 An introduction to statistical methods and data analysis, 3"* edhion PWSKENT Publishing Company, Boston, Massachusetts, USA Parsons, D R 1996 Case stiidy the Mexican wolf Pages 101-123 /n E A. Herrera and L F Huermeke, editors New Mexico's nattiral heritage biological diversity in the land of enchantment New .Mexico Journal of Science, Volume 36 Parsons, D R 1998 "Green fire" returns to the Southwest reintroduction of the Mexican wolf Wildlife Society Bulletin 26 799-807 Peterson, R O 1974 Wolf ecology and prey relationships on Isle Royale Dissertation Purdue University, Lafayette, Louisiana, USA

68

Pilgrim, K L , D K Boyd, and S H Forties 1998 Testing for wolf-coyote hybndization in tiie Rocky Mountains using mhochondrial DNA. Journal of Wildlife Management 62 683-689 Phnlott, D H 1967 Wolf predation and ungulate populations American Zoolocv 7 2t)7-278 Pimlott. D H , J A Shannon, and G B Kolenosky 1969 The ecology of tiie timber wolf in Algonquin Provincial Park Ontario Department of Lands and Forest Research Report No 87 Putman, R J 1984 Facts from feces Mammal Review 14 79-97. Reed, J E , R J Baker, W B Ballard, and B T Kelly In review Differentiating Mexican wolf and coyote scats using DNA analysis Wildlife Society Bulletin Scott T G 1 ^541 Methods and computation in fecal analysis with reference to the red fox Iowa State College Journal of Science 15: 279-285 Scott, T. G. 1943 Some food coactions of the northern plains red fox Joumal Paper No J-1131 Iowa Agricultural Experiment Station, .Ames, Iowa, USA Southwest Region, United States Forest Service 1992 Fish, amphibians, reptiles and mammals a species checklist for the Gila National Forest Pamphlet, United States Department of Agriculture, Forest Service, Silver Citv, New Mexico, USA. Spaulding, R L , P R Krausman, and W B Ballard 1997 Calculation of prey biomass consumed by wolves in northwest Alaska Joumal of Wildlife Research 2 128132 Stephenson, R O , and L Johnson 1972 Wolf report, Alaska Federal Aid in Wildlife Restoration Report, Project W-17-3, Juneau, Alaska, USA Thompson, D Q 1952 Travel, range, and food habhs of timber wolves in Wisconsin Joumal of Mammalogy 33 429-442 Thurber, J M, R O Peterson, J D Woolington, and J A Vucetich 1992 Coyote coexistence with wolves on the Kenai Peninsula, Alaska Canadian Joumal of Zoology 70 2494-2498 Thurber, J M , and R O Peterson 1993 Effects of population density and pack size on the foraging ecology of gray wolves Joumal of Mammalogy 74: 879-889

69

United States Fish and Wildlife Service 1993 Biological evaluation for listed, proposed, and sensrtive species of the forest-wide animal damage management plan on the Apache-Srtgreaves National Forests Unhed States Fish and Wildlife Service, Phoenix, .Arizona, USA United States Fish and Wildlife Service. 1996 Reintroduction of the Mexican wolf within rts historic range in the southwestem United States: final environmental impact statement United States Fish and Wildlife Service, Albuquerque, New Mexico. USA \'an Ballenberghe, V , .A W Erickson, and D. Byman 1975 Ecology of the timber wolf in northeastern Minnesota Wildlife Monograph 43 Wayne, R K , N Lehman, M W Allard, and R L Honeycutt 1992 Mitochondrial DN.A variabihty of the gray wolf genetic consequences of population decline and habitat fragmentation. Conservation Biology 6: 559-569 Weaver, J. L 1979 Influence of elk carrion upon coyote populations in Jackson Hole, W yoming Pages 152-157 m M S Boyce and L D Hayden, editors. North American elk ecology, behavior, and management University of Wyoming Press, Laramie, Wyoming, USA Weaver, J L . and S H Fritts 1979 Comparison of coyote and wolf scat diameters Joumal of Wildlife Management 43 786-788 Williams, D A 1976 Improved hkelihood ratio tests for complete contingency tables. Biometrika 63 33-37 Woolsey, N G 1985 Coyote field guide Special report number 15, Arizona Game and Fish Commission, Phoenix, Arizona, USA Young, S P , andE A Goldman 1944 The wolves of North America The American Wildlife Instittite, Washington, D C , USA

70

en 2

PL,

z
(/)

fS

<s

r^

I
% Z

Tf

00

_*rf

s
4-*

ilSl

"lis-

o "B. o. c

5 (J S C S
71

S. I 3 i. S I

bu V3

ment supp
4>

S g-gH

(A

ON

rS

<S

i aE
a JU
^ II

-3 c

S II

(LI

C/}

food

siu
V

Table 4 2 Comparison of diet composhion found in scats (n = 55) ^ 28 mm diameter from areas where Mexican gray wolf packs (n = 6) were fed supplemental food hems (i e , carnivore logs and road-killed elk, deer, and jackrabbrt) to determine presence of non-supplemental food items ( e g , elk adults and calves, deer adults and fawns, domestic bovine, and insects) Scats were collected from April 1998 to October 2001 in Arizona and New Mexico and compared to Mexican wolf Interagency Field Team dated supplemental food records Comparison values are express as percent frequency of occurrence (PFO) Food hems Pack Lupine Campbell Blue Mule Pack Pipestem Cienega Hawks Nest Total No food rtems 3
-IT

8 5 5 12 55

SF' 3 16 5 2 1 2 29

NSF^ 1 6 3 3 4 10 27

NSF PFO (%)' 25 0 a 27.3 a 37.5 a 60 0ab 80 0ab 833 b 49 1

'SF = supplemental food items * NSF = non-supplemental food hems *PFO (%) = percent frequency of occurrence Percentages followed by the same lower case letter were not significantly different (P>0 05, G-test. adjusted for continuity)

72

vO

W P4

0\

<S

d
Ov

d d
-^

^^

<ri
^M

-^H

00
O

'

00

d d

t^

vO

r 0 ^ f < ^ S - H j s ( M

o\ ao

vO

r\

5!

r4 00

00 '

00

00

1 * 1 " I -

<N

't

<N

.<:! S

o
CO

w S

d
tH4

, '
I

cs

p d

I ^ 1 1 1

I
OS O

IN

gb9f>l f*> o o t*

>d vd
n I I

i r< s

1^ r

d
< M

<s

r^

n^

1
?
S "a
5

I:
i
s
t= z

111111
I I I I5 (^
a I .8 . <

il I . 1 I ^.
2 3 < a o.

"8 !>

!13 I

I 1 13
73

Table 4 4 Comparison among years (n = 4, 1998-2001) of diet composhion found in scats (n = 251) offree-rangingMexican gray wolves collected from April 1998 to October 2001 in Arizona and New Mexico Scats were identified to species by refined tiaditionalfieldmethods (i e, diameter, location, and sign; Reed 2004). Foodrtemswere combined as large-sized (i e . aduh and young ungulates) food items and medium- and small-sized food items (i e . medium- and small-sized mammals, birds, insects, and vegetation). Non-food rtems were not included. Comparison values are expressed as percentfrequencyof occurrence (PFO)
Large-sized food hems 135 19 53 39 246 Medium - and smallsized food rtems 14 1 2 2 19 Large-sized food hems PFO (%)' 90.6 95.0 964 95 1

Year 1998 1999 20OO 2001 Total

'PFO ("*/o) = percentfrequencyof occurrence, none differed significantly (P>0 05, G-test, adjusted for cominuity)

74

Table 4 5 Comparison between seasons (;i = 2, fall-winter versus spring-summer) of diet composhionfriundin scats (/i = 251) offree-rangingMexican gray wolves in Arizona and New Mexico (April 1998 - October 2001) Scats were identified to species by refined traditional field methods (i e , diameter, location, and sign. Reed 2004) Food items were combined as large-sized (aduh and young ungulates) food hems and mediumand small-sized foodrtems(i e , medium- and small-sized mammals, birds, msects, and vegetation) Non-food rtems were not included Comparison values are expressed as percent frequency of occurrence (PFO)
Season fall-winter spring-summer Total Large-sized food items 36 210 246 Medium- and smallsized food hems 4 15 19 Large-sized food rtems PFO (%)' 900 933

'PFO (%) = percentfrequencyof occurrence, none differed significantly (P > 0 05, Gtest. adjusted for continuity)

75

Table 4 6 Comparison of diet composhion among packs (n = 4) found in free-ranging Mexican wolf scats (n = 251) collected from April 1998 to October 2001 in Arizona and New Mexico Scats were identified by refined traditional field methods (i.e., diameter, location, and sign. Reed 2004) Food items were combined as large-sized (i.e., adult and young ungulates) food rtems and medium- and small-sized food hems (i e., medium- and small-sized mammals, bhds, insects, and vegetation) Non-food items were not included. Comparison values are expressed as percentfrequencyof occurrence (PFO).
Pack Hawks Nest Campbell Blue Francisco Cienega Total Large-sized food rtems 66 98 20 26 210 Medium- and smallsized food rtems 8 9 0 0 17 Large-sized food hems PFO (%)' 892 91 6 100 0 1000

'PFO ("o) = percentfrequencyof occurrence, none differed significantly (P > 0 05, Gtest, adjusted for continuity)

76

Table 4.7. Food items (n = 33)foundinfree-rangingMexican gray wolf scats (n = 26) collected from April 1998 to October 2001 in Arizona and New Mexico. Scats were identified by DNA analysis (Reed 2004). Non-food items were not included. Comparison values are expressed as percentfrequencyof occurrence (PFO). Food rtems Large-sized food items Elk (Cervus elaphus canadensis fnels<mij) Aduh Calf Deer (Odocoileus virginianus and 0. hemionus) aduh Unknown native ungulate Domestic cattle (Bos taunts) Medium- and Small-sized food rtems Javelina (Dicotyles tajacu) Red squirrel (Tamiasciurus hudsonicus) Mouse (Peromyscus spp.) Unknown rodent Birds Insects Total number of food items Total number of scats Number food items per scat 'PFO (%) = percentfrequencyof occurrence No. PFO (%)'

5 14 2 I 2 1 I 2 2 2 1 33 26 I 27

15.2 42.4 6.1 3.0 6.1 3.0 3.0 61 6.1 61 30

77

Table 4 8 Comparison of diet composhion of Mexican wolf scats (n = 277) collected from .April 1998 to October 2001 in Arizona and New Mexico and identified using 2 methods DN.A analysis (n = 2b) and refined traditional (n = 251, i e , > 28 mm diameter, 2 den srtes, and ttacks) Food items were combmed as large-sized (i e , aduh and young ungulates) food items and medium- and small-sized food hems (i e., medium- and smallsized mammals, bhds, insects, and vegetation) Non-food hems were not mcluded. Comparison \^ues are expressed as percent frequency of occurrence (PFO) Large-sized food hems 24 246 270 Medium- and small-sized food hems 9 19 28 Large-sized food hems PFO (%)' 72.7 a 92 8 b

Method DN.A analysis Refined traditional identification Total

'PFO ('o) = percent frequency of occurrence Percentages followed by the same lower case letter are not significantly different (P > 0 05, G-te^, adjusted for continuity)

78

Table 4 9, Comparison of diet composhion found in free-ranging Mexican wolf scats (n = 26) and sympatric coyote scats (/= 21) collected from April 1998 to October 2001 in Arizona and New Mexico Scats were identified by DNA analysis (Reed 2004). Food hems were combmed as large-sized (i e . aduh and young ungulates) food hems and medium- and small-sized food hems (i e , medium- and small-sized mammals, birds, insects, and vegetation) Non-food hems were not included Comparison values are expressed as percent frequency of occurrence (PFO) Large-sized food hems 24 15 39 Medium- and small-sized food hems 9 26 35 Large-sized food rtems PFO (%)' 72.7 a 36 6 b

Species Mexican gray wolf Coyote Total

'PFO l" o) = percem frequency of occurrence Percertages followed by the same lower case letter are not significantly different (P > 0 05, G-test. adjusted for continuity)

79

Table 4.10. Food items found infree-rangmgMexican gray wolf scats (n = 26) and sympatric coyote scats (n = 21) collected from April 1998 to October 2001 in Arizona and New Mexico. Scats were identified by DNA analysis (Reed 2004). Non-food hems were not mcluded. Comparison values are expressed as percentfrequencyof occiurence (PFO).
Food items Large-sized food items Elk (Cervus ekphia cxmadensis [nelsoni]) Adult Calf Deer (Odocoileus virginiamis and O. hemionus) Adult Fawn Unknown nati\<e nngiiiat^ Domesbc cattle (Bos Uwrns) Medium- and SnuO-sized food items Ja^'dina {Dicotyles tajacu) Nunall's (mountain) cottomail (Sytvilagus nuttallii) Red squind {Taniasdums hudsonicus) Golden-ffiantled ground squiirel (Spermophilus lauratis) Gny-coUared chipmunk Mouse {Peromyscus spp.) Unknown rodeot Shrew Bink Reptiles Insects VegEtatioa Total nomter of food items Total mmter of scats Noraber food items per scat Mexicanwolf No. PFO(%)'

No.
2 7 1 2 3 1 4 1 4 1 1 1
1 7

Coyote HfO(%)

5 14 2 1 2 1 1 2 2 2 1 _

15.2 42.4 6.1 3.0 61 30 30 61 61 61 3.0 _

4.9
17.1

2.4 4.9 7.3 2.4 9.8 2.4 9.8 2.4 2.4 2.4
2.4 17.1 12.2

33 26
1.27

5 41 21
1.95

'PFO (/) = percemfrequencyof occurrence

80

Table 4 11 Comparison of diet composhion found in free-ranging Mexican wolf scats (n = 277) and diet composition reported in other North American gray wolf diet studies (n = 7) Mexican wolf scats (/; = 270) were identified by refined traditional field methods (" = 251) and DNA analysis (n = 26. Reed 2004) Scats were collected from April 1998 to October 2001 in Arizona and New Mexico Food items were combined as large-sized (i e , adult and young ungulates) food items and medium- and small-sized food hems. Non-food items were not included Comparison values are expressed as percent frequency of occurrence (PFO) Large-sized food items 3,263 421 2,402 935 392 353 753 270 Medium- and smallsized food hems 2,316' 292' 1,082 406 124 94 155 28 Large-sized food hems PFO (%)' 58 5 a 59 0 a 68 9 b 697 b 76 0 c 79 0 cd 829 d 90 6 e

Source Ballard etal (1987) Thompson (1952) Spaulding et al (1997) Murie (1944) Mech (1966) Cowan (1947) .Arjo etal (2002) This study

'PFO (" o) = percent frequency of occurrence Percentages followed by the same lower case letter are not significantly different (P > 0 05. G-test, adjusted for continuity) ^Data included unidentified ungulates and undefined "unidentified" food hems 'Data mcluded unidentified non-food items

81

CHAPTER V SUMMARY

This was the first systematic stiidy to determine the diets offree-rangingMexican gray wolves (Carus hipus baileyi) in tiie soutiiwestem United States. We collected carnivore scats (n = 1,682)fromtiieBhie Range Wolf Recovery Area (BRWRA) in Arizona and New Mexico from April 1998tiiroughOctober 2001. We identified tiie scats to species usingti^dhionalfieldmethods (i.e., diameter, location, and sign) and odor. We then used fecal DNA analysis (molecular scatology) to verify the accuracy of identifying Mexican wolf scats with tradhional methods and odor. DNA analysis of Mexican wolf (n = 26) and sympatric coyote ( = 21) scats showed a 79% overlap in diameto- size (Mexican wolf; 16 3 to 35 8 mm, coyote, 17 4 to 27 8 mm) and scats ^ 28 mm diameter were deposhed by Mexican wolves We found a difference (/ = -2.428, P = 0 019) between diameter means for the 2 species (Mexican wolf F= 26 0 mm, coyote X = 22.8 mm). From our DNA analysis resuhs, we refined the traditional methods for identifying Mexican wolf scats as diameters ^ 28 mm, den site locations, and tracks, as well as scats identified whh DNA analysis Diet analysis of Mexican wolf scats (n = 55) with diameters > 28 mm collected from areas where 6 packs received supplemental carnivore logs and road-killed elk, deer, and jackrabbit showed that non-supplemental food items were consumed by Hawks Nest and Cienega packs (G^ = 12.995, P = 0.023), but not by Lupine, Campbell Blue, Mule or Pipestem packs. Diet analysis of Mexican wolf scats (n = 251) identified according to > 28 mm diameter, den she location, and

82

tracks revealed a diet composhion consisting mainly of large-sized food hems (92.8% PFO). primarily elk aduhs (36 6% PFO) and calves (36,2% PFO). There was no difference in diets among years (n = 4. Gad/ = 2 588, P = 0.460), between seasons (n = 2, Gad, = 0 490. P = 0 484), or among packs (n = 4, Gad, = 7 719, P = 0.052). We found a difference (G^^ = 9 761, P = 0 002) in diet composhion of Mexican wolf scats identified with our refined ttaditional methods (n = 251) when compared to that of DNA analysis identified scats (n = 26) There were more large-sized food items in the scats identified with our refined ttaditional method than found in scats identified with DNA analysis. There was no difference (/ = -1 95, P = 0 001) between diameter means (refined traditional method x = 29 3 mm, DNA method x = 26 0 mm) There was also no difference (t = \ 50, P = 0 005) in number of food items per scat when comparing the 2 methods (refined tradhional method x 1 06 food items per scat, DNA method x = 1 27

food hems per scat) We found a difference ((}ad, = 9 647, Z' = 0 002) in diet composhion of Mexican wolves and sympatnc coyotes, with Mexican wolf diets consisting primarily of large-sized food hems (72 7o PFO) and sympatnc coyote diets consisting primarily of medium- and small-sized food items (63 4% PFO) There found a difference between the 2 species' scat diameter means {t = -2 428. P = 0 019, Mexican wolf x = 26 0 mm, coyote X = 22 8 mm) There was also a difference (/ = 2 849, P = 0 007) between number of food items per scat for the 2 species (Mexican wolf F = 1 27 food items per scat, coyote x = 1 95 food items per scat) Finally, we found a difference (Gad, = 462 492, /> = < 0 0001) in diet composhion of Mexican wolves when compared to diets reported previously for other North American gray wolves Mexican wolf diet analysis revealed

83

more large-sized food hems (90 6% PFO) than that reported in 7 northern gray wolf diet smdies (range 58 5% to 82 90/0 PFO) Our results imply that free-ranging Mexican gray wolves in .\nzona and New Mexico consumed more large-sized food hems, primarily elk aduhs and calves, than sympatric coyotes and the larger, northern gray wolves. This information could prove useftil in managing both Mexican wolf and prey populations, as well as determining the presence of Mexican wolves in depredation incidences in Arizona and New Mexico

84

PERMISSION TO COPY

In presenting this diesis m partial fiilfilhnent of tiie requirements for a master's degree at Texas Tech University or Texas Tech University Health Sciences Center, I agree that the Library and my major department shall make it freely available for research purposes. Permission to copy this thesis for scholarly purposes may be granted by the Director of the Library or my major professor. It is understood that any copying or publication of this thesis for financial gain shall not be allowed without my fiuther written permission and that any user may be liable for copyright infringement.

A|gree (Permission is granted.)

^tudem Signature

Date

Disagree (Permission is not granted.)

Student Signature

Date