ASPECTS OF REPRODUCTIVE ENDOCRINOLOGY iN THE RED WOLF

(CANIS RLJFUS)

A Thesis

Presented to

The Faculty of Graduate Studies

of

The University of Guelph

by

SUSAN LORENE WALKER

[n partial fulfillment of requirements

for the degree of

Master of Science

August, 1999

OSusan Lorene Wal ker, 1999 National Library Bibliothèque nationale 1*1 of Canada du Canada Acquisitions and Acquisitions et Bibliographie Services services bibliographiques 395 Wellington Street 395. rue Wdlingtori OnawaON KlAON4 OnawaON KlAW canada canada

The author bas granted a non- L'auteur a accordé une Licence non exclusive Licence allowing the exclusive permettant à la National Library of Canada to Bibliothèque nationale du Canada de reproduce, loan, distribute or sell reproduire, prêter, distribuer ou copies of this thesis in microfom, vendre des copies de cette thèse sous paper or electronic formats. la fome de microfiche/fiim, de reproduction sur papier ou sur format électronique.

The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts fiom it Ni la thèse ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent être imprimes reproduced without the author's ou autrement reproduits sans son permission. autorisation. ABSTRACT

ASPECTS OF REPRODUCTIVE BIOLOGY IN THE RED WOLF (cANISluLtw9

Susm Lorene Wülker Advisor: University of Guelph. 1999 Dr. K. Goodrowe Co-Advisors: Dr. C. Gartley Dr. A. King Dr- J. Leiltherlmd

Reproductive endocrinology of the red wolf (Canis rufiis) wu studied in captive animals by fecd enzyme immunoassay. In çyçling fernales. estrogen metabolites inc~lised through proestrus. reüched maximal values during estrus and declined concurrent with rising propstin metabolite values. Progestin metabotite levels remained elevated until mid- to lare luteal phase and then zpdually declined to nadir levels. There wu no significünt difterence between pregnant and ovulatory non-pregniint hormone profiles. AcyclK- animais maintained basal civarian hormone concentrriticins throughout the breedin,0 season. During the peciovulritciry penod. serum and fecd hormone profiles exhibited compmblc trends. in male red wolves. changes in testostercme metabolite concentrations were consistent with photoperiod synchronization. Testcistemne metabolite levels began tr, increase prior to the breeding season (November-December). and reached peak levels coincident with estrus in late winter (late February).

It was concluded thrit sex determination thrciugh feçd steroid ratios of progestin/testosterone and testosterone/estrogen is possible only from le ikember to early May. ACKNOWLEDGMENTS

1 would like to thmk my supervisor Dr. Karen Goodrowe for her invaluable guidance, advice. support and encouragement. You have been ü wonderhl teacher and mentor. Thank-you. i would Like to thünk my farniiy and tnends for king so patient and understanding over the put 2 yeux. For my mother, thank-you for always being there when 1 nezded encouragement. Keener, thank-you for your patience and support.

1 am grateful to Will Waddell and Sue Behrns for allowing me to play such ün active role in the red wolf SSP. Thank-you for dways making me téel so cornfortable. Without your dedication to the survival of this endangered carnivore. there would have been no researçh. Great appreciation is extended to the st.atthe Burnet Park Zoo. Great Plains Zoo. Knoxville Zoo. Mill Mountain Zoo, Oglebay's Good Zoo. Point Detiance Zoo & Aquarium and Red Wolf Breeding Fzicility, Racine Park Zoo. Ross Park Zoo. Trevor Zoo and Western North Cmlina Nature Center. Without your cornmitment the knowledge gained in this study would not have been possible. You have dl played a crucial role un the rorid to understanding the basic reproductive physiology of the red wolf.

i would like to express my &mtitude to Dr. Cathy GartIey and Dr. Allen King. Your guidance and ridvice was invaluable. To Dr. John Leatherlmd. Dr. James Raeside and Rick Renaud. thank-you for teaching me and allowing me to trip into your vut knowledge of steroids and endoçrinology. Special thanks to Dr. John Leatherlmd md Dr. Suzanne MacDonald for statisticai assistance and to i4~tric.iBellem. Dr. Janine Brown and

Comlie Munro for assistance with the assays.

A very special thanks to mernbers both past and present in reproductive physiology Iaboratory rit the Toronto Zoo. Helen Bateman, Asmd Bellem. Maria Franke. Dr. Margery

Hay. Merryn Mclninch. Gabnela Mastromonaco. Leanne Otben. Drive Ryckman and Kristen Young. You have dl provided me with helptùl advice. support and laughter. di of which hüs allowed for the cornpletion of this study.

Financial support ww provided by the Moms Animal Foundation. the Zoologici

Society of Metmpolitan Toronto and Naturai Sciences and Engineering Resevrh Council of Canada. DECLARATION OF WORK PERFORMED

I declare that with the exception of the following. dl work covered in this thesis was performed by myself.

Fecal samples were couected by animal care staff rit the following institutions:

Burnet Park Zoo. NY: Great Plains Zoo, SD; Knoxville Zoo. TN: Mill Mountain Zoo. VA:

Oglebay's Good Zoo. WV; Racine Park Zoo, WI: Point Detiruice Zoo & Aquarium and

Red Wolf Breeding Facility. WA: Ross Park Zoo. NY: Trevor Zoo. NY and Western

North Catolina Nature Center. NC. BIood sarnpie coltection wuprekrmed by staff at the Point Defiance Zoo & Aquarium and Red Wolf' Breeding Fücility. WA. Assistance with tècd steroid extraction wlis provided by G. Mastromonaco and D. Ryckman during the Fa11 and Winter of 1998. Fecal steroid HPLC fmctionation was performed by R. Renaud

(University of Guelph. Guelph. Ontario). Enzyme imrnunoassay antibodies and HRP conjugate stocks were provided by C.J. Munro (University of Cidifornia, Davis. CA).

Luteinizing hormone antibody, label and standard was supplied by Dr. J.L. Brown

(National Zoofogicai Park Conservation and Resemh Centre. Front Royal. VA). Assay buffers were prepared by A. Bellem. G. Masuomonacv. L. Othen. D. Rycknian. K.

Young or rnyself. Validation rind andysis of semm progesterone vdues were conducted by staff at the Gynecologic Femde Treatment Clinic. Txomri. WA. February and Mmh

( 199 1- 1997). Validation of rridioimmunoassys for ,semm esuadiol and LH was pertormed by A. Bellsm (National Zoological Park Conservation and Re.search Centre, Front Royal. VA).

iii TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... i ... DECLARATION OF WORK PERFORMED ...... III .. LIST OF TABLES ...... VII ... LIST OF FIGURES ...... VIII

LIST OF ABBREVIATIONS ...... ix

INTRODUCTION ...... i REVIEW OF LITERATURE...... Reproduction in Canids...... -3 Femde Reproductive Biology ...... -3 General Ove~ew...... -3 Ovaian Cycle ...... -4 FoUicuJar and Hormonal Changes ...... -7 Ovulation ...... II Pregnancy and Pseudopregnancy ...... 14 Endocrinology ...... *...... *... 16 Luteal Maintenance and Life Span ...... 17 Artifical insemination ...... 20 Male Reproductive Biology ...... 21 Endocrinology and Spematogenesis ...... 21 Seawndity...... 23 Stress and Reproductive Function ...... 26 Mechanisms of Stress Impact ...... 2X Red Wolf ...... 30 Taxonomy ...... 30 Historicd Perspective ...... 32 Recavery Plan and Reintroduction ...... 33 RATIONA LE ...... 35 METHODS AND MATERIALS...... 38 Fecal Study Animals and Sample Collection ...... 3% Animais ...... 38 Fecd Collection ...... 39 Serum Study Animals and Sample Collection...... 39 Animais ...... 39 Blood Collection ...... 39

DISCUSSION ...... 77 Validation ...... 77

Female Reproductive Cycle ...... 80 Longitudinal Protiles ...... 80 Periovulatory Profdes ...... 82

Male Reproductive Cycle...... 86

Management and Conservation Applications ...... 87

SUMMARY AND CONCLUSIONS...... 89 LITERATURE CITED ...... 93 A PPENDIX [.A ...... 118 APPENDIX 1.B ...... 120 APPENDIX 1.C ...... 121 APPENDIX I.D...... 122 APPENDIX 1I.A ...... 123

APPENDIX I1.B ...... 124 APPENDIX 1II.A...... 125 LIST OF TABLES

Table I. Mem fecd progestin and estrogen metabolite concentrations durinp different stages of the reproductive cycle for pregnmt. ovulatory nonpregnant and ac yclic fernale wolves ...... 67

Table II. Mean concentrations and ratios of tècal progestin. estrogen and testosterone metabolites during diftèrent stages of the breeding serison in male and fernale red wolves ...... 76

vii LIST OF FIGURES Figure 1. Pÿr;illelisrn curves for progesterone gnerited by the senal dilution of feçd extract from male and female red wolves using extraction method 1 ...... 54

Figure 2. Parallelism curves for estndiol genemted by the serial dilution of kcal exuricts from male and femde red wolves using extraction method 1 ...... 55

Figure 3. Padlelism cumes for testosterone generated by the serial dilution of kcal extncts from male and fernale red wolves using extraction method 1 ...... 56

Figure 4. Pdlelism curves for cortisol genemted by the serial dilution of &;il extracts from female red woives using the cortisol extraction rnethod...... 57

Figure 5. Imm unoreactive progesterone proti les of fecd extracts from femde red wolves during the folticular and luteril phases ilt'ter fractionahon with HPLC ...... 5') Figure 6. Immunoreactive estndiol protiies of fecal exuacts frorn female red wolves duhg the follicular and luteai phrises ;ifter frictionation with HPLC ...... 60 Figure 7. Immunorerictive corésol and testosterone profiles of kçal extracts from female and male red wolves. respectively. after fi-actionation with HPLC. 62

Figure 8. Chromatogmph of retention times of 2 1 puritied steroid standards usinp reverse phrise HPLC ...... 63

Figure 9. Mem fecd esuogen and progesterune metabolite profiles during the breeding .season from female red wolves ...... 65

Figure 10. Fecal progesterone, esuogen and cortisol metabolites from 1 ) a femde red wolf that bscarne pregnant naturally and 2) a female red wolf that wm resuriined on several occasions to time for artificial insemination...... 66 Figure 1 1. Daily mem tècal estrogen and progestercme metabolite profiles during the periovuiatory period from non-pxgnant fernale red wolves ...... 70

Figure 12. Daily mean serum estradiol, progesterone and LH protiles during the periovulritory period from non-pregnmt ferni.de red wolves...... 7 1

Figure 13. Weekly mean tècal testosterone metabolite profile over ri one yeu period from male red wolves ...... 74 LIST OF ABBREVIATIONS

AC Nonovulatory or acyclic tèmde AI Artitical i nsem ination ANOVA Andysis of variance AAZPA (or AZA) American Association of Zoolopiçai Parks and Aquariums CL Corpus luteum/corpora lutea E2 Fecd esuogen metabolites E2, Semm estriidiol EDTA Ethylenedinitiilo-tebr;icetic ücid EiA Enzyme imrnunoassay FSH Follicle stimulating hormone GAM Goat anti-mou.se gamma globulin GnRH Gonadotropin-releasing hoimone hCG Human chorionic gonadotropin HPLC High performance liquid chromatognphy HRP Horseradish peroxidasr: LH Lutenizing hormone NMS Normal mouLwsemm NP Ovulritory nonpregant femde NS B Non-specitic binding P Pregnrint tèmale P4 Fecal progeste rone metabolites p4, Semm progesterone PDZA Point Detïance Zoo and Aquarium

RIA Radioimrnun~~ciy RWSSPO Red Wolf Species SuMvd Plan@

T Fecal ündrogen metabolites

USFWS United States Fish and Wildlik Service INTRODUCTION

In the I=t century extinction of animal species bas been occuning at rui unparileled rate (Flesness. 1989). Globd biotogicil diversity is eroding Iargely due to the spresid of rnankind and our 'innovative' technologicd advmcements. As a species. mankind has dtered ecosystems through: 1 ) habitat destruction and ti-agmentation. 2) introduction of non-native species. 3) pollution and 4) over-exploitation cif naturül resources (Wilson.

1r)8g: Tudge. 1992). [t hüs now become increasing clear that the world in which we [ive is

"poised on the edge of an unprecedented biologicd disaster'' (Hutchins & Wiese. 199 1). The challenge to iutificially preserve biologicd diversity is a daunting task. However. success possi bly can be ac hieved through the dedicated cooperation of numerous individuals and organizations. One such orpizrition is the Arnerican Association of

Znologicd Parks and Aquariums (AZA). The foundation of their conservation efforts is based on Species Suwival Plans (SSPO): breeding programs which act to coordinate the prexrvation of both a species and its habitat (Wie.se & Hutchins. 1994). Lch SSPG. under the direction of a species coordinator. is responsible for managing the demographics and genetics of an endangered species' population in a captive environment. The coopemtive actions of institutions that house the species ensure the genetic integrity of the population. Captive breeding programs serve a genetic reservoir for future reintroductions and maintenance of hedthy. stable captive and wild populations through the infusion of new genes. It is well recognized that the tightly managed captive population in no way serves ris ri substitute for wild populations.

Over the past 2 decrides. assisted reproductive technoiogies. such ris uîiticial insemination. embryo trrinstèr. long-terrn storage of gerrnphsm or non-invasive endocrine monitoring have begun to play an ever increasing mle in conservation programs. These technologies dlow for flexibility in ii breeding program. escaping restrictions such ris limited holding space and transportation cnsts. while still presevering the genetic diversity of the population. However, with only 20 yeürs of cxperience and an ever inoreuing number of animds in need of captive propagation progruns (Soulé et al.. 1986), there is still much tri lem More these technologies kcome pnctical and repeatable on ri Ixge scde. One important lesson that hris been acknowledged is that what is et'tective for one species often rnüy be totally unreliable for imother. even in si given mon (Goodrowe. 1997). Therefore, to apply these biotechno1ogies. b~sicresearch concerning reproductive norms must be conducted to improve our fundamental knowiedge of a given species.

In situ and ex situ conservation et'fcwts for the red woIf (Canis mfud have led to non-invasive endocrine stridies in our Iabonitoiy tit the Toronto Zoo. The goal of this study is to non-invasively ücquire basic knowledge conceming this endangered cünid's reproductive biology. The knowledge gained will help provide a resourceful stntegy tilt- more effective management of captive and wild red wolf populations. REVIEW OF LCTERATURE

Reproduction in Canids Female Re~rnductiveBiology General Oveniew .. . In the fernale domestic dog (Çani~farniliüns]. sexusil maturity ranges frorn 5 tci 24

months of age (Wildt et al.. 198 1; Feldman & Nelson. 1987: Concannon. 199 1) with the

average age of puberty onset king dependent on breed (Christiansen. 1984). Fenility

progressively declines once ri bitch is 27 yem ni age (Feidman & Nelson. 1987). The

nurnber of ova present in the avaries of a newborn bitch has been estimated at 700.000.

By puberty. this number declines to 250,000. and at 5 yeus of age there is a dramritic drup to 30.000. By age 10, only a few hundred ova remain (Anderson & Simpson. 1973).

Beyond 6-8 years of age. there are progressive reductions in iitter six. increases in the

interestrus interval. congenital birth defects. and problems at parturition (Feldmün &

Nelson. 1987). Although bitches typically continue to cycle throughout their litè-span (as

they do not experience a menopause; Feldman & Nelson. 1 '387), the recommended age tiir breeding is between 2 and 6 years of age (Feldman & Nefson. 1987).

Captive femaie gray wolves (Canis IUDU~have ken reported to conceive at 10

months of rige (Medjo & Mech, 1976; Zimen. 1976): however, most do not breed until at

lest 22 months rit' age (Seal et al.. 1979). Mech ( 1987 ) reported signs of reproductive activity. including mammary developrnent and pairhg behaviour. in wild wolves during

their tïrst breeding .season. However. most do not breed until they rire 4-5 years of age

(Mech. 199 1 ). Captive gray wolves may live to 16 years of age and free ranging wolves have been reported to live as long as 13 yeus (Sheldon. 1992). The reproductive age in

wild gray wolves hlis been estimated to extend to 1 1 yem of age (Mech. 1988). Captive fernide red woives (Canis rufuQ also are capable of reproducing durinp their tïrst breeding season: however. breeding is generaily more cornmon dunncg their second season (Nowak

1972: Parker. 1988. Waddell. 1995). Fecundity of the captive red wolf pe& at approxirnritely 7 years of age and deciines to .sene.wence rit 12 yem (Spark database.

Waddell. 1996). Coyotes (wlauans) also rire capable of reproducing in their tirst year (Kennelly & Johns. 1976): however. the percentage that does sri is variable. rmging between 10% and 708 (Gier. 1975). Longevity in the wild coyote is estirnrited to be 14- 16 years (Sheldon. 1992).

Domestic dogs are monoestrus. ovulating one to tw» times a yeu at intervals of 5- 13 months (Concannon et al.. 1989). Severd studies indicate thrit. as a group. domestic dogs cycle on üverdge every 7 months. md are nonseasonal breeders (Christie & Bell

197 1; Stribenîèldt. 1977: Felman & Nelson. 1987: Ccincannon. 1991). It hris ken demonstnted that domestic dogs housed outdoors cycle year round (Anderson. 1970). as do dogs housed under constant Lighting conditions of 12 hours light:12 hours dark

(Concannon, 19863). With the exception of some fox-Iike çanids (Sheldon. 1992). and the bush dog (S~eothosvenaticug Bekoff. 1975: Ponon et al.. 1987). wild canids generdiy are classified as rnonoestnis (Sheldon, 1992). However. unlike their domestic relatives. they have well-defined breeding seasons: from Febt-uary to March. for the red wolf (Carley 1979: Waddell. 1995). and January to April in the coyote (Hamlettt. 1938: Fowler. 1986) and gray wolf (Sheldon. 1992: Busch. 1995). A trend shared rimong these species is that animais hmsouthem latitudes tend to breed eulier than their northern cvnspecitïcs (Mech. 1970: Pmdiso & Nowak. 197 1; Ensely. 1986: Fuller. 1989~Fuller. 1989b: Waddell. 1995). Ovarian Cycle

The classical stages of the canine estrous cycle rire anestrus. proestrus. estrus. and metesws (Heape. 1900). In recent years. the term diestius hâ,c Irirgely replaced the terni metesws in the bitch (Olson et al.. 1984a. Concannon. 1991). in wild canids for which there is relevant information. the estrous cycle follciws the generil trends obsewed in the domestic bitch (Moller. 1973: Kennelly & Johns. 1976: Seal et al.. 1979: StelIt-Iug et al.. 1981; Sed et (il.. 1987: Pauimj et al.. 1992. Wasser et al.. 1995: Monfort et al.. 1997: Velloso et al.. 1998). Anestrus is the period of apparent reproductive quiexence following diestrus

(Concannon. 1991). In the domestic dog. anestrus ranges tiom 2-10 months. with ail average of 4 months (Feldman & Nelson, 1987: Concannon. 199 1).

Proestrus is a period of heightened folliculw activity (Feldmm & Nelson. 1987).

In resp0n.w to increcising levels of estrogen. proestrus is marked by ri discharge of serosanguinous tluid through the vagina. vaginal wrnificarion and bleeding. progressive

increases in vulvül edemü. and pheromond attraction of males (Feldrnim & Nelson 1987:

Concannon. 199 1 ). Dunng the majority of proestrus. the bitch refuses to permit rnounting by tuming and growling at the male. Near the end of proestrus the bitch müy permit

mounting. but usually prevents intromission by lying or sitting down (Concannon. 199 1 ). Proestrus in the domestic dog can be as bief as 1-3 days or as long as 3 weeks: the average duration is 9 days (Christie & Bell, 1972: Feldman & Nelson. 1987: Concruinon 1983).

Physical signs of proestms in wild canids. such as vaginal bleeding and swelling. appear to be longer in durrition compared to their domestic cousin. riging in duntion tkom 1-6

weeks in gray wolves (Seal et al.. 1979) to long ;LS 2-3 months in coyotes (Kenneily &

Johns. 1976: Stelltlug et al.. 1981). However. studies indicate that the duration of

proestrus for the maned wolf (Chrvsocvon brachvurus; mean 10.6 driys. VeUoso et ai..

1998) and raccoon dog (Nvctereuta procvonoides: mean 7.6 days. Valtonen et al.. 1977) hl1 within the range dzscribed for the domestic dog. Esuous behaviour within the genus Cani?;is chamcterized by retlex or spontaneous stiffening tuid deviation of the tail, firmly standing to permit mounting. and a retlex lordosis-like pre-sentrition of the vulva (Concannon et al.. 1979b: Seal et al., 1979: Beach et al.. 1982: Pauiraj et al., 1992. Monfort et al., 1997). In the domestic dog estrus may be ris short as 3 days or lut for severai weeks. but it is generilly 4-12 düys in length. with ui average of 9 days (Concannon. 1991). The duraticin of estnis in wild canids appears to be highly variable: griy wolf (3-15 days: Sed et al.. 1979) coyote (mean of 10.2 days:

Kennelly and Johns. 1976) silver fox CVuI- VCJI~.1 -5-5 days; Lindeberg et al.. 1993) raccoon dog (mean of 4 days: Vdtonen et al.. 1977) dhole (Cuon &- 14-39 days:

Pauimj et al., 1992) bush dog (1-12 driys: Ponon et al.. 1987) Atncan wild dog (LYc'aon ~ictus,6-9 days: Monfort et al., 1997). Chüracterimtion of red woIf behaviours associated with coumhi p and breeding are currently under investigation (T. Wagner. persona1 corn munication).

Estrous behüviours in domestic dogs are a respcmse to a decline or withdrriwal of esuogen. and art: facilitated by a rise in progesterone (Concannon et al.. IY79b. Beach et al.. 1982: Feldman & Nelson, 1987). This theoty is supponed by the administration of estradiol done followed by either the absence or administration of progesterone to ovuiectomized domestic bitches. Behaviourril estrus was not displayed until ski- the administration of estradiol wris halted, and the onset of estlus wu more synchronous when progesterone was ridministered at the tirne of estrogen withdrawd (Concannon & Hansel. L975). Estrous behaviours, including copulation. also have been observed during declining estrogen suid nsing progesterone concentrations in serum samples hmgmy wolves (Seal et al.. 1979) suid fecal samples from Ati-icrin wild dogs (Monfort et al.. 1997 ) and maned wdves (Velloso et al., 1998).

In the domestic dog it hris been reporteci thrit intromission and mriting may not t~cur until a few days rit'ter the onset of estrous behaviour. Dunng estrus. the vulva inust progress through a turgid phase to become soft and tlaccid. therefore nt) longer providing ri difficult brirrier for mde penetration (Feldman & Nelson. 1987). Mating is usually 5 to 45 minutes in duntion md usually results in a copulatory tie. during which the male disrnounis and stands beside or behind the bitch. with the penis locked in the vagina. The lock withstands tugging and pulling, and is terminated by detumescence of the penis

(Concrinnon. 1991 ). The copulatory tie that is common among domestic cnnids dso has been observed in the maned wolf(Rodden. 1995). gray wolf (Busch. 1995). red wolf (S. Be hrns personid communication). coyote (Sheldon, lW2 ). raccoon dog (Valtonen et ai..

1977) ünd the dhole (Pauiraj et al., 1992). The tie appears to be slightly shorter in du~ation

in wild canids. perhaps LS mechanism for survivd in the wild (Pauinj et ai.. 1992:

Sheldon. 1992; Busch. L995)

Diesas begins once the bitch stops accepting the male (Evans & Cole. 193 1: Olson et al.. 1984a) and is considered to last until the subsidence of the effects of luted phase progestemne. Diestms cm be chuicctsrized by mammÿry gland development or progesterone rnz~surementin the nonpregnÿnt or p.seudopregnant bitch. There is no xute luteolytic mechanism in the dog, and the transition hmdiestrus to anestrus is grridual

(Concannon. 1991). The luteal phase in a pregnant bitch ends with parturition (Concannon. l986b). Follicular and Hormonal Changes

It hW ken suggested that developing tollicles cm uccur in bitches of dl ages. breeds and striges of the estrous cycle (Dumt et al.. t 998). Throughout much of anestms. the ovaries of the domestic dog contain slow but continuously developing follicles. with most becoming atretic (Feldrnan & Nelson. 1987). Follicles that develop pardlel to gonadotropin stimulation mature and littain the ability for estrogen synthesis and secretion (Feldman & Nelson. 1987). Durin2 the onslt of proestrus. follicular atresia continues to ovedethe majjority of vesicular follicles. However. some remiun viable and continue to enlarge. A characteristic feature of the ovary at the on&setof proestrus is the presence of several follicles 1.0-1.5 mm in diameter. Durhg proestrus. maal follicles increase in size and by late proesaus reach 3-4 mm in diameter. At the onset of estrus- follicles appear rnuch as they did in late proestrus except that they measure 3-5 mm in diameter (Anderson & Simpson. 1973). In the domestic bitch preovulatory follicular- development has been detected as early as 11 days before ovulation. with the most rapid follicular maturation occuring 2 to 3 days before ovulation (Wildt et al.. 1977). Although anestnu hhas been referred to as the quiescent phase of the canine ovuian

cycle. studies suggest that neither the ovary nor the pituitxy gland are cnmpletely quiescent

during anestrus (Olson et al.. 1982). Fluctuations have been identitied in both pituitÿry

hormones ruid ovarian steroids (Feldman & Nelson. 1987). Serum estradio1 or esuogen levels in the domestic dog tluctuate throughout

anestrus. usually nnging between 5-20 pgml (Concannon. 199 1). Surges in esvogen (in

the ringe of 30-50 pg/ml) ;ils« have been observed (Olson et al.. 1982). and are mumzd to

be derived korn wlives of follicle development (Feldman & Nelson. 1087). During sarly proestrus. estrogen concentmtions range from 10-20 pgml (Concannon & Lein. 1989). Peak concentritions of estrogens range from 50- 100 pCJml ruid are attained in late proestrus or etirly estrus. generdly 0-3 days before the preovulatory surge in luteinizing hormone (LH: Edquist et al.. 1975. Concannon et al.. 1975: Hadely . 1975: Nett et al.. 1975: Wildt et al.. 1979a: Chakraborty. 1987. Concannon. 199l). Esuogen levels decline during estrus to concentrations of 5-20 pg/ml (Concannon. 199 1 ). and repre-sent the tind fidliculai- maturation process prior to ovulation (Feldman & Nelson. 1987). Concentrations of xrum estrogens in the gr~ywolf and coyote during late anestrus. proestrus and the cstrogen surge have been reported to fall within ranges reported for the domestic dog (Seal et d.. 1979.

Stelltlug et al.. 198 1 ). Changes in estrone concentritions reported throughout most of the domestic dog ovarian cycle tend to püraliel tho.se (if estradio1 (Wildt et al., 1979: Chakraborty. 1987). Plasma progesterone concentrations in bitches increrise during proestrus from baseline values of 0.4-0.6 ndml to vdues of 0.13-1 2 n@ml at the hme of peak estmdiol levels (Concannon. i986a. Concannon et al.. 1977ri). Progesterone increues mpidly above 1.0 ndml during the LH surge as estrogen concentrations faII (Wildt et al.. 1979:

Chakraborty. 1987: Concannon et al., 1977a). Progesterone concentntions are - 4-X ngml at the time of ovulation. which occurs - 2 days after the LH surge (Wildt et al.. 1979: Concrinnon. 1986a). Plüsma/.serum LH concentrations have been txported to be luw rind variable during

mesuus. and low during proestnis and estms except during the preuvufütory surge (Smith

& MacDonald, 1974; Concannon et ai.. 1975: Nett et al.. 1975: Concannon et al., 1977~

Chakraborty. 1987). Generrilly. LH semm concentrzitions dunng anestrus are well klow LH peak values. representing only IO-20% of pelik values (Concannon. 1991). However.

LH pulses of 2-25 ndml have been detected at intervals of 2-24 hours throughout suiestrus.

Inter-pulse levels were 0.2-1 -2 ng/ml. Just prior tii the onset of proestrus. pulse intervals increüse to 60-90 minutes md mean LH kvels are elevated to 3 ngml (Concannon et al-. L986; Concannon. 1989: Concannon et al.. 1989). During proestrus. LH levels are Iow and pulses are nondectecable. possibly due to estrogen negative kedback. LH

concentrations increase to peak levets, ringing ti-om 8-50 ndml during the preovulatory surge (Concannon, 199 1).

Unlike LH. follicle stimulating hormone (FSH) does not have wide tluctuations in peripherd concentrations. This hormone varies only slightly throughout anestrus before

declining with the onset of proestms. presumably due to negative feedback etYects uf estriidiol and inhibin. Follicle stimulating homone then ri.ses with the LH peak (Reimers et

al.. 1978: Olson et al.. 1982. Concannon. 1993). Mean levels of FSH durin,u anestrus

may surpms leveis ob.served during the preovulatory SUI-22.and concentrations have ken

reported to be 50-1008 of those of the periovulatory peak (Olson et al-. 1982: Concünncm.

199 1). Unlike thrit of LH. the pulsatile nature of FSH secretion during anestrus hu not

been clearly demonstrated. probably because of the relatively smdl incre~sesin magnitude rind the longer half-lik of this honnone (Concannon. 1993).

Compared to most domestic and labomtory animals. there is a paucity of information regarding replation of gonadotropin secretion or control of uvcuim activity by

gonlidotropins in the fernale dog (Concannon. 1991). It is likely a complex interaction

between environment. age. ovarian stritus. generil health. and other yet unidentitied factors that dictate roles for FSH and LH (Feldman & Nelson. 1987). Studies have demonstrated a pulsatile releslcse of LH secretion in the bitch (Wegstad et al.. 1993: Jet-fcoate. 1993: Hoffmann & Schneider. 1993). It ha-s ken suggested thrit anestrus may be naturiy terminated by an increax of this episodic release, resulting in an elevation of mean LH Ievels shortly More proestrus (Concannon et al.. 1986: Concaanon. 1989). However, the mechanisrns involved in altering the LH secretion pattern remin unknown (Concannon, 199 1 ). A minor. but signiticant increase in eswadiol ha been observed shortly betcm the onset of proestnis (Jeffcciate, 1993). Such elevations in estradio1 may result in an increwd rate of pulstltile LH secretion andhr may also be the resuI t of increased LH secretion. Together. both mechanisms could k involved in a brief positive feedbrick for the initiation of proestnis (Concannon. 1993). This theory cm be supponed by data that demonstrate the induction of proestrus by injections of LH done

(Concrinnon. 1993) rind by administration of estrogen alone (Bouchaccf et al.. 1993).

As demonstrrtted in other species. gonadotropin-releasing hormone (GnRH ) routinely causes the dease of LH and FSH in the domestic canid (McRae et a!.. 1985:

Vanderlip et al.. 1987; Vm Haaften et al. 1992: Concannon. 1993). The administration of a GnRH agonist for 14 days in lui anestrus bitch stirnulated LH rind FSH secretion and resulted in the successful induction of estrus and the production of twu pups (Crincannon.

1993). It is possible that elevated LH concentrations at the end of anestrus involve aii increase in sensitivity to GnRH in addition to an increase in pulse frequency (Concannon.

1993). It also hsis been repoited that the sensitivity of the LH response to GnRH was higher during Irtte anestrus than during early mestnis (Van Hariften et al.. 1992). The cause of the increase in sensitivity remains unknown (Crincannon, 1993). Concannon (1993) suggests that both FSH rind LH are folliculotwphic hormones.

The administration of FSH alone or LH alone can cause foliiculrir growth md induce proestnis (Concrinnon. 1993). However. alterations in LH. not FSH secretion. may be more significmt in the initiation of a functional fcilliculür phase. as FSH appears to be elevated throughout most of anestms and LH concentrations are usually low except near the end of estrus LH secretion. This assumes thtit the biopotency of FSH does not increase

during Iate anestrus. Therefore, either the start of a tidlicular phw does not 1-equii.e FS W.

or follicles acquire responsiveness to FSW at the end of anestrus. perhaps due to increwd concentrations of LH and estradio1 (Concannon. 1993). Pmlactin also may play a role in the initiation of the ovririan cycle in the domestic

bitch. Suppression of prolactin with dopinine agonists causes an extensive shortening of interestrous intervaIs (Okkens et al.. 198%: van Hmften et al.. 1989) and induction of esws in tèrnales with a prolonged anestrus (Arbeiter et al.. 1988: Jvchle et al.. 1989)- However. no obvious alterations in prolactin concentrations have been observed in Iate anesuus (Olson et al.. 1982). suggesting moditications in prolactin receptors may be involved (Concannon. 1993). As with other species. endogenous opioids appear to modulate LH relerisc: in the ovarian cycle of the domestic dog (Conçannon & Temple. 1988). The responsiveness to nüloxone. m opioid antagonist. laie in anestrus wax elevrited comprired to the proestrus pha.se. The authors suggest that there is an endogenous opioid tone inhibi~gLH release during late anestrus and a possible decline in opiciidergic tone or decline in non-opioid inhibitory mechanism which may causes the onset of proestrus. Ovulation

It haken suggested that changes in sexuril behaviour cm be rdated to oviilritioii

(Christiansen. 1984). Holst & Phemister ( 197 1 ) indicrited that civulation wcurred 24-48 hours after the bitch tirst dlciwed mating. Later. studies clernonstrated that ovulation may occur over several days following the beginning of standing estrus (Wildt et al.. 1978).

Penton et al. ( 199 1 ) used cinxlating progesterone and LH to determine time of ovulritioii and elucidate its relation with behaviour. Their results indicsite that the time of ovulation bore littie relation to alterations in sexual behaviour. It also hris been dernonsuated hat signs of proestrus. such as vulval swelling or serosanguinous discharge, mtiy occur -3 weeks to 3 days pnor to the LH peak. while first acceptance of a male cm occur as early as 4 days befor-e or as Irite as Ci days after the LH surge (Concmnon. 1983, Concannon. 6 In relation to behaviourd estrus. the time of ovulation in the coyote is highly variable. It mriy occur as early ris the 1st day of estrus or as late as the 9th (Kennelly & John. 1976). This suggests that the timing of ovulation is moe precisely determined by endocrine changes rrither than by observations of behaviounl or clinicd changes. DecIining esmdiol concentmtions with incixasing prrigesterone concentrations a asscwiated with the LH surge in the domestic bitcli (Smith & MacDonald. 1974: Concannon et al.. 1975, Concannon et al.. 1977a: Concannon et al.. I979b; Wildt et al.. 1979: Olscui et al., 1982). The pre-ovulatory LH surge hris been demonswited tci occur concurrrrntly with elevations in progesterone concentrritions: the two coincide on the same day or may differ by only a single day (Concannon et al.. 1977a: Penton et al.. 199 1). This phenomenon also hüs been demonsmted in gray wolves. in which the initial increase in progesterime concentrations from baseline values c~ccui-redduring the LH surge (Sed et al. 1979).

In the domestic dog the preovulatory LH surge represents ri 20-40 fold increase in

LH levels and a 2-4 fold increase in FSH levels. The LH surge may 1st for 1-4 days and generily occurs near the onset of estrus. 0-3 days rifter the estrogen peak of late proestrus

(Smith & MacDonald. 1974; Nett et al., 1975: Concrinnon et al.. 1975. Ccincünnon et al..

1977a. Wildt et al.. 1978). Similar to the domestic dog. the gray wolf exhibits ri preovulritory LH surge whiçh last for 3 days immediately fdlowing the estrogen peak (Seal et al.. 1979). Ovulation. which is spontaneous in the bitch. has been reported to wcur 1-3 days following the LH surge (Wildt et al.. 1978). rilthough otliers reportecf little variation in the 2 day interval between the LH surge and ovulation in the bitch (Phemister et d.. 1973: Concannon et al., 1977a. Concannon et al.. 1989). Durdtion of ovulation has beeii variably reccirded as tnnspiring over only a few hours (Holst & Phemister. 1971:

Phemister et ü1.. 1973: Concmnon et al.. 1977~:Anderson & Simpson. 1973). 12-72 hours (Graf. 1978) or even over several days (WiMt et al., 1977; Wildt et al.. 1978). Perhaps the di.screpancy regrading the duration of ovulation is related to the technique used (histologie or iaprascopic). Unlike most species. the ovriry of the domestic: bitch is completely surrounded with a fiitty bursa which inhibits the direct examinrition of the ovary (Wildt et d.. 1977).

Follicles which fail to ovulate following the LH surge undergo atresiri (Feldmtin &

Nelson, 1987). The number of ova relelised is pürtiülly dependent upon the breed of the bitch. Ba-sed on litter size. smaller breeds tend tu ovirlate kwei- ova (2- IO) than larger breeds (5-20 ova: Feldman & Nelson. 1987). An approximation of the number of ova relerised and corpora luteri (CL) fonned following an ovulatory surge of LH wüs determined through exminrition of ova recovered after tlushinp either the oviduct or uterus at different intemals post-ovulation. The number of CL identitied on the ovq seldom agreed with the nurnber oC ova coilected. with the average numbers of ovri and CL king

5.4 (range 2-9) and 7.6 (4-9). respectively (n= 10dogs: Penton et al.. 199 I ).

The timing of the LH surge can be supponed by studies in which ov~ectornized bitches received estmdiol-releasing implants and increasing doses of injected estrüdiul beiizoate. Ceuing injections. estradiol concentrations tell and patterns of LH secretion were investigrited with or wi thout the sim uItrtneous ~idminis~tionrif progesterone. The decline in esmdiol caused preovulation-like surges in LH which were tùrther Facilitated by the administration of progesterone (Concannon et üI.. 1977a. Concannon et al.. 1979a).

These temporal hormonal rehtionships have lertd tci the idea thrit in the bitch. elevated estradiol concentrations maintain a negative feedback eftéct on LH secretion. Further. the preovuIatory LH surge is initirited by a tàilure of an additional estrüdiol increüse or decline and is facilitated through an elevation in progesterone levels (Concannon et al.. 1979b).

Ovariectomy at any stage in the reproductive cycle results in a rapid luid listing incre~sein LH and FSH concentrations. Estradiol rippem to be the major, and perhaps rinly. ovruian hormone involved in this negative feedback (Concannon & Hansel. 1975: Concannon et a 1.. 197%). The administration of increrising estrridiol concentrations tu ovliriectomized bitches. even when above normal peak values. suppresses LI4 levels (Ccmcmnon et al.. 1979~~).In contriist. the administration of progesterone. or progestercme agonists does not reduce LH in civruiectomized bitches (McCmn et al.. 1987). As with LH. FSH secretion in either ov~iectcimizedor intact bitches is inhibited tiillowing the administration of estradio1 but not by the administration of the prcigestligen rnegestrcil xetate (Colon et al..

1993: Concannon, 1993). The ability of progesterone to synergize with estradio1 and prcduce a negative feedback suppression of LH htis ken investigsited in the mesuous bitch. Basal LH levels maintained at low vdues by endogenous estcosen were not tùrthei- reduced through the administration of megestrol licetate (Colon et ai.. 1993). Administration of LH or human chorionic gonadotropin (hCG) causes pro-estrual follicles to luteinize or ovulate (Concannon. 1993). However. FSH. in addition to LH. may serve as an ov~1atot-yhormone. In the absence of a distinct detectüble LH surge and associated with a ne=-nornial FSH surge. spontaneous fertile ovulations have been ob.served in bitches induced with an agonist of GnRH following proestms (Concannon.

19x9: Concannon. 1993). Factors regulriting FSH concentrations in the periovultitoq period rire likely to be similar to those which regulate LH (a decrease in the estradio1:progesterone ratio), except that, ;L,S in rither species. ovxitin inhibin promotes the suppression of FSH secretion during proestrus (Concannon. 1993). Although is has ken demonstrxed that GnRH stimulates gonadotropin rele~~sein dots (MacRae et al.. 1985: Vanderlip et al.. 1987; Van Haaften et al. 1992: Concannon. 1993). whether a preovulatory surge in GnRH ,wcreaon triggers the surge releüse of LH and FSH ha not ken investigrited in the domestic dog (Concannon. 1993). Pregnancy and Pseudopregnancy

Once released. oocytes (-120 pm in diameter: Holst & Phemister. 197 1) pass through the bursal cavity into the proximal portion of the oviduct within a few hours

(Anderson & Simpson. 1973). The bitch ovulates prirniiry oocytes (Evans & Cole. 1931 : Anderson & Simpson. 1973; Tsutsui, 1975). which rnay require 2-5 days for the completion of meiotic maturation (Tsutsui. 1989). The formation of polar bodies to complete meiosis occurs in the mid- and distlll portions of the oviduct (Holst & Phemister.

197 1 : Anderson & Simpson. 1973; Phemister et al.. 1973). On average. oocyte maturation is klieved to occur within 2-3 days post ovulation and. theretore 4-5 days üfter the LU surge (Cc~ncannonet al.. 1989). Signitïcant oocyte matuiation untranspire within 2 days (Mahi & Yanagirnachi. 1976): however. oocytes may ~xmainkrtile for an additional 2-3 days (Tsutsui & Shimizu. 1975) since fertile matings have cxcured 7-8 days rrfter the LH

Peak (Cuncmnon et al.. 1989). Fertile matings eulier thm 2 days before or 9-l0 days after the LH peak are rare and result in reduced litter sizes (Concannon. 1986a. Concannon et al.. 1989). Fertilization takes place in the distal portion of the oviduct (Anderson & Simpson. 1973). Ova may remain quiescent in the oviduct of different individuals for difièring periods of time priar to fertilization suggesting thrit individuai variation may c~cui. with respect to the timing of îërtilization (Tsutsui. 1075. Penton et al.. 199 1 ). The embryo of the bitch remains in the oviduct for -9-10 days following ovulation. then mers the

UENS either in the morula or early blastocyst stage (Anderson. 1927: Penton et al.. i Y9 1 ).

Free floating blatocysts are present in the utenis between days X and 20. and during this pend edematous areas appear in the endometnum which 1-epresent future implantation sites

(Holst & Phemister. 197 1 ). Implantation in the dog hlls been reported to wcur 13-20 days after ovulation (Griftïths et al., 1939: Thatcher. 1997).

Gestation length in the domestic dog from tèrtile müting to parturition ha ken reported from 55-68 driys; however, gestation length reccirded from the precivulritory LH peak to parturition is 64-66 days (Concrinnon et al.. 1983. Concannon et ai.. 1989). In wild canids it hzs been reported thrit gestation length tends to increase with ridult size. ranping hm48-58 düys in the smaller fox-like canids (Sheldon. 1992) tc~58-65 days in the larger woif-like canids (gray wolves, 60-65 days. Seal et al. 1979: Lentfer & Sanders.

1973: maned wolves, 65 driys. Bndy & Ditton. 1979: Vellem et al.. 1998: dhole. 62.7 driys. Pauirj et al.. 1992; coyote, 58-6 1 days: Gier 1975). Litter size in the domestic canid

ranges from 1-23 pups; in most breeds the average litter size is between 4 ruid 8 pups

(Christiansen. 1984). Reported Litter sizes in the gray wolf ( 1- 13, average 6: Sheldon.

1992). coyote (2- 12. average 4-7; Gier. 1975). and red wolf (2-8. average 3-4. Rüey &

McBride. 1972; Parker 1988) sire similar to that of the domestic dog. In geneml, litter six in wild crinids is reported to increase with increased adult weight (Moehlman. 1992).

Post-implantation pregnancy can be detected by palpation (Concannon. L983)

mdor ulwrlscuxi (Bondestam et id., 1983). To date there are no readily avtlilable

endocrine methods for the early diagnosis of pregnancy in the domestic bitch. However. relaxin is detectable in plrisrna by radioimmuno~ssay&ter the third week of gestation

(Steinetz et al.. 1987; Steinetz et al.. 1989) and total urinary estrogens are elevated 2 1 days tifter müting (Richkind. 1983). Both have ken suggested as possible pregnancy diagnostic tools and just recently a reiaxin assciy becme commericdly availlible (Repro Chek. Synbiotics).

The long luteal pha-se of the nonpregnant ovariim cycle has been termed ri physiologicd pseudopregnancy (Stabenfeldt & Shille. i977). Unlike the ferret or cat.

which experience ri pseudopregnancy. the long luteal phase in dogs occurs spcmtmeously

and is not dependent on mating-induced ovulation or activation of luteal function. Overt or- clinical pseudopregnancy varies with breed and individual and is characterized by extensive mammary gland development and behavioural changes typicd of pregnüncy and lactation (Concannon. 199 1). Endocrinology Little or no signiticant difierence hris ken reported in circulating plasma progesterone levels for pregnant and nonprqnant bitches (Hridely. 1975: Nett et al.. 1975: Austad et al.. 1976: Reimers et ai., 1978: Concannon et al.. 1989: Onclin & Verstegen.

1997). Ho wever. the presence of secondary increüses in progesterone following implantation have been suggested (Smith & MacDonald. 1974: Concannon et al.. 1977b). Concannon ( 109 1 ) vutlined that during pregnancy. progesterone ccincentmtions reach peak levels of 15-HO @ml rünging from days 12-30. slnwly decline to values of 2- 10 ndml by day 45. and les than 1 nghl by days 60- 1 10. Estriridi01 values reach values of 5-20 pgml in the euly luted phase. followed by a slight increase bzfore retuming to low variable concentration during anestrus (Concmnon. 199 1 ). Most studies report that mean esuridiol or total estrogen levels rire not signiticantly diftèrent dunng Che Iüst half of prepmçy when cornpxed to the nonpregnant bitch (Edqvist et al.. 1975: Hadely. 1975: Nett et al.. 1975: Austad et al,. 1976: Graf. 1978; Reimers et al.. 197x1, Estrone 1eveLs ftoliow similu' patterns to those of zstriidiol. except that a pregnüncy-specitk increase in cstrone levels htls been suggested (Chaknborty. 1987). It also has ken reported that mean levels of semm sex steroids were higher in the nonpregnant bitch than in pregnant bitches during weeks I through 6, and weeks 1 through 3 for progesterone md estrridiol. respectively

(Chakrriborty. 1987). With the exception of a fiw studies. diftèrences in progesterone and sstrogen concentrations between nonpregnant and pregnant bitches may not bz obvious as pregnancy is sswiated with: 1) an increase in blood volume diluting concentrations in semm and plasma. and 2) ii possible increase in clearance mts due to an elevated metribolism (Crincannon et al., 1977b). However. presnancy related differences in both progesterone and estrogen fecd metabolites have been found in the domestic dos (Gudermuth et al.. 19%) and the maned wolf (Was.sei-. 1905: Vellero. 1998). Luteal Maintenance md Life Span

Secretion of pmgesterone and maintenance of luteal tissues during pregnancy are dependent on pituitary secretion of luteotrophic hormones since pregnancy can be terminüted at any stage by hypophysectomy (Concannon. 1980). The two main luteotrophic factors involvecl in regulation of the canine CL are LH and prrdactin (Okkens et d. 1986: Concannon et al.. 1987; Okkens et al.. 1990). lt htis been suggested that LH and prolactin rnay exert their rictivities through different mechanisms (Hofiman & Schneidei- 1993). The rernoval of LH. using anti-LH senrm. lerids only to a ternporary decreüse in progesterone. The removd of prolactin. using bn)moci-iptine. izsults in an irreversible

decrease in piugesteronz concentrations (Concannon et al.. 1 987). Lutecitrophic control of the dog CL Lippem to shift from the thto second hdf of the luteal phase (Ol.wn et al.. 1989). Corporzl lutea we more resistant to luteolytic fiictors

during the early luteal vs. Iate luteal phase (Concannon. 1989). Observed decreltses in progesterone levels tollowing hypophysectomy wre more pronounced during late rither

than zarly stages of diestrus in the nonpregnant bitdi (Concannon. 1980: Okkens et al..

1986). Following administration of a LEiRH antagvnist, it was determined thrit the CL in

the pregnrint dog may be independent of pituiury luteotrophins up until -Day IO of

pregnancy rind dependent upon pituitary luteotrophins atier Day 22 (Vickety et al.. 1987). The luteotrciphic hormone requirements of the erirly developing CL have yet to k detemined (Concannon. 1995).

Pregnmcy-specific increases in prolactin (-35 days after the LE4 surge) have ken reported in the domestic dog (Knight et al.. 1977: Concannon et ai.. 1978: De Coster et al..

1983: McCun et al.. 1988: Concannon et al., 1989. Onclin & Vertegen. 1997). It has

ken suggested thrit prolactin is the main luteotrophic factor during the second hdf of pregnmcy. but its mode of action remains unknown (Onclin & Verstegen. 1996). The low prnlactin concentrritions ob-erved during the luteal pha-se of the nonpregnmt bitch suggests that its role in the nonpregnmt bitch is les signiticrint than in pregnancy (Onclin & Verstegen. 1997).

Prosesterone mriy be the only ovruian steiwid required hi- maintenance of pregnancy, since syn thetic progestins cm maintain pregnuncy in the r~vuiectomizedbitch

(Scikolowski. 197 I : Concannon et al., 1977b). However. progesterone alone may not be the only homione necessciry for al1 aspects of pregnüncy. as mammary enlugement did ncit rwcur in ovariectomized bitches during the mrrintenrinct: of pregnancy by administration of 1i synthetic progestagen (Steinetz et al.. 1989). The concentrations of progesterone during pregnancy appem to be in excess. as bitches treated with do.ses of prostaglandins (which produce an incornple~luteolysis) hve been able to maintain pregnüncy despite reductions in progesteinne levels to 2 2 @ml for severd days (Concannon & Hitnxl. 1977: Vickery

& McRae. 1980: Jackson et al. 1982). Corpora lutea remüin the main source of priigester~insand aix rzqiiiird for maintenance tif the eiitiii: pregnüncy (Concannon et al.. 1977b). Ovaiiectorny at any stage ïesulB in the telmination of pregiimcy (Sokolowski.

197 1: Andeixon & Simpson. 1973). Histvcheinical studies of the dog piricenu sugrest that there is little if any de-novti placenta1 steroidogenzsis (Kisc) & Ymaiichi. I9ti4).

Esmjgem do not increase &m;itically during pregnancy. This üppears tii k ciitical. sincc abnormally high estrogen levels un terminate a hlly establislied post- implantation pregnancy (Concannon. 1977b). Howevei-. the prescnce of esrrogens is essentiai during pregnmcy. üs they contribute tci the synthesis andor availability 01' progesterone receptors (Concannon et ai.. 1989). To date. no luteolytic mechmism has been dzmonstrated in the dog (Hoffman &

Schneider. 1993). Mechanisrns ailowing luteolysis rnay k: 1 ) reduced responsiveness tu luteotrophic stimuli andor 2) a deçrease in luteouophic support. Studies have dernonsinted concent~ationsof LH and prolactin mriy inci-exx in the .wiurn of pregnant and nonpregnant bitches during laie diestlus. Therefors. luted regression likely uccurs independently of the üvÿilability of LH md prolüctin (Olson et al.. IYXJa: Chakriiborty.

1987: Hoffman & Schneider. 1993). It also hxs been demonstnted that prolactin and LH receptor formation remain constant during diestrus. making endogenous controt of luteolysis difficult to explain (Fernandes et al.. 1987: Hofiman & Schneider. 1993).

As with some other species. the CL cif pregnant and nonpregnant bitches are responsive to the luteotytic effects of prostaglandin (PG) F-2a. Expeiirnents have dernonstratecl that the CL of pregnant dogs in the Iate luteal phase müy be more sensitive to PG-induced luteolysis than during early luteal phase (Concannon & Hansel. lW7: Pmdis et al.. 1983). However. Jogs in generd appew to be less sensitive to the luteolytic effects of cxogenous PG F-2u than most domestiç species. Pru longed dministmtion of relatively high doses. often resulting in severe side effect,~. is required tci richeive luteolysis

(Concannon et ai.. 1989). In many species. the uterus may be involved in regulating luted

cegression (Olson et al. 1989): however, hysterectomy does not intluence luteai function in

the dog (Oison et al.. 1984~:Okkens et al.. 1985b).

In the nonpregnrint bitch there appears to be rio acute luteolytic mechanisrn

(Concannon. 1995). There is a slow regression of the CL resulting in an undramatic decline of progesterone to near brisal levels 60-90 days after ovulation (Concannon et al..

1989). in the bitch. it has been suggested that pregnrincy is nomally tenninated by m rieute preparzum luteolysis and abrupt tàll in progesternrie over the 24 hour pericid privr to parturition. The mechanisms involved are speculated tci involve the maturation of the fiîti~l pituitruy ridi~nalaxis. efiects of fetd glucoconicoids on placentriluterus. md release of luteolytic levels of PGF-2a into the ckulation (Concannon et al.. 1988. Concannon. 1995). The mechanism by which endogenous PGF-2a depresses progesterone concentrritions is unknown (Concannon et al., 1989).

Artiticid Insemination

As canids rire generilly rnonoestms. detection ot' the time of peak krtility is critical to breeding management. As mentioned previciusly. for the domestic dog. oocyte maturation is believed to occur within 2-3 days cjf ovulation ünd 4-5 days after the LH surse (Concannon et al.. 1989). Dog sperrn have been reported tci remain femle for up to

6-7 days in the ferni.de reproductive tmct (Doak et al.. 1967: Concannon et ai.. 1983): however. the riverage is probably only 2-3 days. since tèitility decreses in breedings which occur more than 1 day prior to the LH surge (Holst & Phemister. 1974: Holst &: Phemister. 1975).

Natural mating of the bitch usually results in ri conception rate of greater than 90%

(Holst & Phernister. 1974: Farstad, 1984: Johnston. 1995). Vaginal artificial insemination

(AI) using fresh semen crin give comparable results tii nütud mlitings when the timing is right and the semen quality is high (Linde-Forsber: & Forsberg. 1989). The pregnmcy rate from frozen semen varies considerably (25%-80%:Olar. 1985; Linde-Forsberg &

Forsberg: 1989: Farstüd. 1984; Fmtad & Anderson-Berg: 1989: Fontbonne & Badinand.

1993: Linde-Forsberg & Fcmberg, 1993: Wilson. 1993) and is dependent upon a number of hctors. such as site of semen deposition. timing of in,wmination. .semen quality. spem do.% and frequency of insemination (Wilson. 1993). Reprited inseminations using frozen semen have been reported to ~~sultin higher conception rites compared tri single inseminritions (Farstüd & Anderson. 1989: Linde-Forsberg & Forsberg. 1989: Linde-

Forsberg & Forsberg. 199 3). Investigitors also have reported signitïcantly lower pregnancy rates with vaginal insemination than wi th intrau terine deposition of frozen thriwed semen (Anderson. 1972; Anderson. 1974: Smith & Graham. 1984: Olar. 1985).

Therefore. high Ai success rate with frozen-thawed .semen requires deposition of semen into the uterine body (Fontbonne & Badinand. 1993) by either cervical crinnulation or surgical insemination. Post-thaw Longevity of frozen spermritozoa is usually short-lived: therefore. inseminrition must be timed so that the ova pre-sent in the uterus or oviduct are mature. The peak time foi- Al in the dornestic bitch witii frozen-thawed sperrn is 4-5 days after the LH peak (Brittista et A.. 1988: Concannon & Battista. 1989).

There are very tèw accounts of successful tissisted reproduction in wolves. Seagei- et al. (1975) reported ri successful pregnancy in ri srüy wolf using frozen-thawed semen.

Timing for the 4 vaginal inseminations was b~sedon vaginal cytology and bleeding. A single litter of red wolf pups dso has been produced iising fresh .semen, obtained by electroejrtculation. and surgical deposition into the uterine horns (Wriddell & Platz. personal communication). Timing for the 2 surgical inseminritions was detemined through daily vaginal cytology and semm progesterone concentmtinns. Male Remoductive Cv& Endocrinology and Spermatogenesis

Pituitary gonridotropins are the major controlling mechanisms in speimatogenesis and andi-ogen production (Eik-Nes. 1962; DePalatis et al.. 1978: Kriwakami et al.. 1997). In most species. regdation of FSH and LH is controlled by secretions of gonadai steroids.

peptides (e-g. inhibin) md ultimately, GnRH (England. 1997). As with other species. GnRH stimulates the release of LH (Jones & Boyns. 1976: Jones et al.. 1976) and a

subsequent rise in serum testosterone levels within the male dornestio dog (Lacoste et al.

1988; Knol et al.. 1993: Purswell & Wilcke. 1993). Knoi et al. (1993) were able to

demonsuate a do.% dependant relationship between GnRH and LH response. However the

lack of relationship behveen GnRH and testostercine respunse. suggests that GnRH does not directly effect Leydig ce11 function in the male dog, Follicle stimulriting hormone levels

alsci have been shown tu inc'rerise following the administration cif a GnRH analogue:

however. the response WLS not as dramatic as WU ob.served for LH (Purswell & Wilcke-

1993). These author'; suggest thrit the slow elevrition of FSH over a 72 hour pnod rnay be due to a delayed reponse to the GnW analogue or ma-ely a normal diurnai variation. Tc)

date. very Little information is availsible regrirding FSH in the mde or kmde dog (Concannon. 1993: Puswell & Wilcke. 1993). In the domestic dog. castration elevrites the mean ccincentration of plasma LH. but

does not effict the rhythm of LH secretion (DePalatis et al.. 1978). Luteinizing hormone secretion is episodic in the male dog with pestks c~ccuringat intei-vds of 1.5 to 5 hciurs.

îbllowed by surges in testosterone 60 minutes later (DePalatis et al.. t978). In the

domestic dog there is little information regarding the eftècts of pnadal steroids

been conducted. using exogenous progesten and rindrcigens. have dernonstrated minimal effec~supon .semen quality and libido (Wright et al.. 1079: Taha. 1980: Püi-mo et al..

1993). A recent study suggests that the sensitivity of the pituimy-hypothalamus tc)

propestogen feedback in the male domestic dog may be different from that of other species.

Prcigestins rnay have a direct action upon sperm quality by xting on the epididymd phase of development (Engiand. 1997). In contrat. androgens have rn indimt action upon sperm quality. causing the suppression of gonadotrophs (England. 1997). It also hsbeen demonstrated that LH secretion is inhibited more by estndicil than by testosterone (Jones et al.. 1976: Winter et d.. 1982) and that aromatimtion of mdrogens to estrogens in the brain repulates LH secretion (Aman. 1986). One cycle of the seminiferous epitheliurn in the domestic dog hÿs a dumtion of 13.6 doys (Foote et al.. 1972). while spemütogenesis ranges from 52-70 days (Ghosÿl et al ..

1983: Shille & Shbenfeldt. 1986). The transit time of sperm cells through the epididymis

has been estirnated at 14 days (Austin & Short. 1973). The dumtion of spermiitogenesis

and epididymd transport time have dso been cstirnatcd in the coyote and closzly resemble

that of the domestiç dog (Kennelly. 1972). One cycle of the seminiferous epitheliurn wlis

13.6 days. arrivai in the wput epididyrnis riin@ from 37-40 days. and epididymal

transport time wüs estimated at 14 days. Selisondity

Male domestic dogs üre sexualiy active year-round (Concannon 199 1 ). producing sperm throughout the year with unchanged efficiency (Aman. 1986). However. Kurodï &

Hioe (1972) have reported that sperm production in the domesùc do$ declines during the summer. Falvo et al. ( 1980) also suggest the possiblity of iuiçestral seasonality in male monsrel dogs. Measurements of serum LH and testosterone indicated an dterdtion in the nnnual cyclic pattern of LH secretion and a signitiçant rise in testosterone durins late Augusb'early Septem ber. In contrast. most non-domestic cruiid species sxperience strict reproductive sesonality [red fox. Lloyd & Englund. 1973: blue fox ( Alcyex 1igo~uQ.Smith et al.. 1987: Mondüirl-Monval et al.. 1988: gray wolf. Mitsuzukn. 1987: I';lccon dog. Yongjun et al.. 1994: coyote. Hamlett. 1938; Green et al. 1984: red wolf. Wÿddell. 19951. There is a period. corresponding to the mating season. when spermatogeniç and ündrogenic functions ut' the testes reach a maximum. Beyond this period. regression of the testes renders the male infertile. Typically full sexual capability in the male is achieved some weeks before femüles are receptive (Lincoln. 1989). For the femak red wcilf. which is ünnually monestnis, receptivity towards the rnde is between Februq rrnd March (Crirley. N79.

Riley & McBride, 1972, and Waddell, 1W5). a period similx tc> thrit obsenred in the gray

wolf (Gensh, 1968. Seal et al.. 1979 and Seal et al.. 1987) md the coyote (Gipson et al..

1975. Hamlett. 1938 rind Stelltlug et al.. 198 1). Semonality in sperm production h~sbeen

doçumented through the crillection of viable sperm fnim Januq to Aprii in the gray wolf (Mitsuzukri. 1987; Seager et al.. 1974). and from November to Apnl in the coyote (Gipson

et al.. 1975. Green et al.. 1984 and Hamlett 1938). In a study by Gipscin et ai. ( 1975).

one red woIC x coyote hybrid actively produceci mature sperm from November 5th-Mah 22nd. For al1 species. including the red wolf (W.Waddel1, personal communicrition).

mature spem have been cotlected weeks prior to and fr~llowingthe suspected estrus pend of the femaie.

Semonai changes in the testes are diciated by the hypothdamus and the anterioi-

pituitxy gland through gonadotropin secretions. Many species have ri mechanism that responds to specific eues from the environment. inclriding changes in day length. food

supply and temperature tr) regulate hypothaliuniç activity (Lincoln. IL)8Y). Wild canids

such as the gray wolf are known to reproduce seasvnally. The timing of reproduction in

the gray wolf varies with latitude. occumng later üt higher latitudes (Sed a al.. 1983: Sed et al. 1987). Fluctuations in the secretory capacity of testosterone and LH in respcinse tri LHRH stimulation were ctosely related to temperature and photoperiod cycles. suggestins

that spermatogenesis ancl androgenesis are seasonally regulated (Seal et al.. 1987). It also

ha becn documented that the cape hunting dog (Cunningham. 1905) and maned wolf (Ewer. 1973) ai1 shift their breeding season 6 months when trrrnslocated across the equatoi-.

Al1 of these events UE consistent with photoperiod synchronization. It ha ken well documented that the photoperidic message within most animds is conveyed by the pineal hormone. melatonin (Bartness & Goldman, 1989: Morgan & Mercer. 1994). However. despite evidence of photoperiod influence wvithin the canid family. it has been suggested by

Asri et al. ( 1987) that the gray wolf mriy rely on a system other than the pined for controlling seasonal reproduction. Pineülectomy and superior cervical gringlioneçtomy in male and female prepubend wolves did nut alter the Ge at which puberty wuattained. In addition. animais were monitored hra minimum of 3 years post surgery for serum levels of tzsosterone/LH and progesterone/esuadioi. in males aiid fernales. epectivel y. Al1 wolves exhibited nomai reproductive cyles when cumptired with sham- md unoperated animals. It also has ken suggested that a prolactin rhythm may be involved in the replation of seasonal breeding (Mirarchi et d.. 1978: Schulk et al.. 1% 1 : Bubenik et al..

1985). Pmlaçtin levels crin k dtered by the administration of melatonin (Adam Br

Atkinson. 1984). Kreeger et al, (1990) Jemonstratecl thrit pindectomy of gray wolves does not alter the prolactin circuuiud rhythm. while. melatonin feeding significmtly

Jepressed prolactin levels. Together. the resulu obsereved by Kreeger et ;il. ( 1 990) and

Asa et al. ( 1987) suggest thût in the gray wolf. melatonin may be a primÿry messmger mediating the effect of photoperiod on reproductive events. but that the main source of melatonin is not the pineal gland.

For many species. changes in the endocrine üctivity of the testes in relation to cues from the environment have been observed by rne~suiingtesticular cuntent/vcilume ündlor serum levels of testosterme (Audy et al.. 1985: Ben Sud & Bayle. 1985: Lincoln. 1989:

Mdiuckin & Blackshaw. 1991ab: Atkinson & Gilmartin. 1992: Wieser et id., 1992:

GxsheHs et al.. 1994: Johnson et al.. 1994: Kaplan & Mead: 1994). Seasonid changes in testicular size also have been recorded for some non-domestic cünids including the gray wolf (Mitsuzuka. 1987) coyote (Hamlett. 1938: Gipson et al.. 1975: Green et d.. 1984) and blue fox (Smith et al.. 1984; Smith et al., 1985). Goodrowe et al. ( 1998) alsci have recordeci the testis size and combined testicular weigha of red wolves. This study found that there was a decreasing trend from February md March. with patterns beyond this time period remaining unknown. In addition to testicdu meüsurements in wild çanids. seüsonal variations in fecal testosterone metabolite concentration in the Atncün wild dog (Monfon et al.. 1997) and serum testosterone concentrations in the blue fox (Smith et al.. 1985) have ken documented. The results in both studies indicüted thüt peük testosterone concentrations were reliited to peak miiting activity. It should be noted that although peak

testis size and testosterone concentrations can be con-elated with the breeding season. circulriting testosterone ievels may not always be ri relirible index of normal spermatogenesis as testosterone concentrations tluctuate throughout the day (DePdatis et al.. 1978). Similar testosterone concentrations cm be fwnd in domestic dogs with impaired and normal spermütogenesis (Feldman & Nelson. 1987).

Stress and Reproductive Function Al1 stressors. whether physicd or emotiond. are pe~eivedby the centid nervous system (CNS). The intimate relationship of the pituitaiy with the CNS suggests that pituitary hormones play a central role in an animal's smtegy to cope with stress. dictriting which of the physiological systems must respond to maintain homeostasis. Under ricute stress. the sympathetic-adrend medullary system is xtivated. resulting in the release of critecholamines and a rripid increase in arousal and muxular iictivity. Chi-onic stress stimulates the rele~se of adrenocorticotrophic hormone (ACTH) from the pituitq. activriting the adrend conex to release various sterciids. which in tum allows for energy availability over prolonged penods (Moberg 1985).

Indicators such as behaviours and endocrine physiology frequently ate used tu detine ievels of 'stress' experienced by an animal. Hrwever. behrivioural expression varies both between ruid within species and there are incrmsistencies with regards tn what is recorded by observers as stress-related behaviours. On the other hrind. inçre~sesin plasma, fecal. urine and saliva concentrations of corticosteroids in wild and domesticated species have ken successfully used ris physiological indices for stress (Fmnzrnm et al.. 1975: Wildt et ai.. 1984: Hamilton & Weeks. 1985: Moberg. 1985: Wildt et al.. 1986:

Miller et al.. 199 1 : Culstead et al., 1992: Vincent et al.. 1992: Monon et al. 1995; Graham

& Brown. 1996: Monfon et al.. 1997). Futhemore. the non-invasive nature of monitoring fecd and urine steroid concentriitions makes them puticulacly inviting for use in wildlik species. Longitudinal monitoring cm wcur without manipulation of subject animais

there by eliminating their exposure to potentially 'stresstù l ' restrint episodes tbr biood

srimpling and allowing for a more accurite picture of endocrine status (Lrisley & Kirkpsitrick. 199 1).

Because the neurciendocrine system is an essential component of reproduction. it is

lcigical to =urne that myresponse of this system to stress rnay intluence kitility (Moberg. L985). In the reproductive process. stress piimarily at't'ects those steps where neurcxndocrine control is essentid (i.e. ovulation). Therehre. a high prioiity for captive

breeding progriuns is to interpret if stress impacts on ii femde's reprc~ductivephysiology during the peri-copulatory period and pregnancy. Although both male and femde îkrtility

rnay be comprornised by stress. the tèrnale apperiix morx vulnerable. Fier reproductive success is centered upon a series of catefully orchestratecl neuroendocrine events. If one of these events is disrupted. reprodüctive failure cm ciccur (Moberg. 1985).

The negative impact which acute and chnic stress have upon the reproduçtian of sei-verai animals hris ken extrensively reviewed (Rrimriley. 1% 1: Collu et al.. 1984: Mokrg. 198% Liptrip 1993: deCmtanzaro & MricNiven. 1992). Numerous studies have shown that environmental stressors cm influence ovulriticiii and estrous cycles in mmmds.

The incidence of estrus in ewes is reduced upon exposurc: to severe cold. wind and min

(MricKenzie et al.. 1975) and dairy cows subjected to high temperiture and humidity exhibit an abnormal estrous cycle (Stott & Williams. 1962). Socid stress from overcrowding shortens estrous cycles in domestic bovines (MacMillan and Watson. 197 1 : Wagon et al.. 1972). Stress wociated with social status in taliipoin and rnarmo.set monkeys intluences the physiological mechanisms essential for ovulation (Bowman et al.. 1978; Abbott et al.. 198 1). The stress rissociated with trinsportation cm dter the length of the estrous cycle or delay ovulation in crittle. sheep and swine (Lamond. 1962: Braden et al.. 1964: Naibandov 1964). In the rat. surgicd stress and chronic imtriint have ken shown to dter ovulation patterns (Schwartz. 1964: Nequin & Swm. t 97 1 ) and delay estnis (Euker and Riegle. l973), respectively. A new management routine cm resutt in a drop in the ovulation rite of ewes (Doney et ai. 1973. Doney et al.. 1976). Together the cornbined evidence of these studies supports the ideti thrit the estrous cycle is vulnemble to the influences of ri vririety of physical and emotional stressors (Moberg. 1985).

Although most experimentai evidence indiclites thrit folliculogenesis and ovulation are the phrises cif reproduction most vulnemble to stress. implantation and developrnent of the embryo may be equally at risk (Moberg, 1985). Heat sues reduces implantation success and embiyo development in ewes (Dutt. I963) rind dairy cnws (Putney et al..

1988: Liptrrtp 1993: Ulberg. 1967). Young sows stresseci by boar harassrnent at the time of mating exhibit higher levels of embryo mortality (Rich et al.. 1986). Physicd rest.m.int

(Euker & Riegle. 1973). human handling (Runner. 1959). exposure to predators

(deCantanzoaro. 1988) and exposure to strange males and their odours (Pikes & Bruce.

1962) have been shown to depress pregnancy rates in rats. Ove~r~iwdingmice results in ri gened friilure of reproduction either thmugh impertict (Christian & LzMunyan. L958). delriyed (Dickson. 1964) or blocked implantations (Weibold et al.. 1986) in mice. Mechanisms of Stress Imm

A physiologicaJ mechanisrn by which stress mriy intluence fertility is thriwgh ri disruption in gonadotropin replation and action by iiitluencing the hypothülmic-pituitary- adrenocorticiil axis (Moberg. 1985). Experimenu have shown suess-induced devations of cortisol or corticosterone for a viuiety of species. such ris exposure to loud noise in r.at,c;

(Zondek & Trimari. 1960). handling in rnonkeys (Elvidge et al.. 1976 ). restraint in gerbils

(Feiiske. 1986). white-tai led deer (Wesson et al.. 1 975). impda, zebrri and gù-cltot" (Morton et al.. 1995). repeated venipunctures in rabbits (Fenske. 198 1 ). and social dominance in captive gray wolves (McLeod et al., 1996). red foxes (Hartley et al.. 1994) and Africm wild dogs (Creel et al.. 1996). In stress related re-search. the administration of exogenous ACTH or corticosteroids h~sbeen used to evslluate physiologicd responses to stressors. It hüs ken demonstrrlted thrit the administration of either ACTH or cortisol crin block the p~ovulatoryrelease of LH in crinle (StoebeI & Moberg. 1982). swine. (Bürb et al.. 19x2) and monkeys (Moberg et al.. 1982). Injections of ACTH cm disnipt estrous cycles in sheep (Doney. 1976). rcnd rats (Hagion et al.. 1969) and rnenstrual cycles in briboons (Rowell. 1970). Exogenous

ACTH or corticosteroids cün also disrupt implantation and feetal development in sheep and rrits (Veludo. 1957; Howuth & Hawk. 1968). Elevritions ofcorticosteroids kquenty tire u.xd as masure of stress but there also are stresstùl conditions in which corticosteroid secretion is unattei-ed or the et'tècts of stress-induced tidrenril .secretions are minimal (Maon. 1968). Hensleigh & Johnson

( 197 1 ) demonsurited thrit ridrendectomy in pregnruit rrits Jid not ever.se the adverse eff"cts of heat stress. However. removai of the ovaries. followed by prosesterone and estrogen replacement, elirninated any effects of such stress on feu1 development. They sugpested that the stressnr mrry have 5i direct effect on pituitai-y-gonadai function. resulting in pregnancy disruption. and thüt the effects of adrend secretions were only minur.

The adrenal axis also secretes sex steroids. There is extensive evidence that these sterciids ci have a profound effect on pituitruy responsiveness tci GnRH. primruily decreasing sensitivity tcj GnRH (Moberg. 19851. Androgens. andcostenedione and dehydroepiandosterone. which are releûsed from the adrenal during stress events (Fuller et a[.. 1984) can expel fertilized eggs from the fernale's reproductive wrlct (Hqer. 1967:

Harper 1969). These androgens are believed to rict primruily through their conversion tii estrogens (Fenske. 1986: decantanzaro & MacNiven. 1992). Elevations of exopenous androgens and estrogens have demonstmted a complete disruption of eariy pregnmcy in a variety of marnmalian species (Deansley. 1963: Stone 1064: Smith & Biggers. 1968; delatmzarci et al.. 199 1 ) and have been found to have adver.= effects on the migrrrtion of fertilized ova. It also has ken demonsmted that large doses of exogenous androgens and estrogens calicexceteraticin of ova transport and smaller do.ses result in the retention tif ova in the fdlopirin tirbe and subsequent degener~tion(Parkes 8r Betlerby. 1926: Whitqey & Burdick. 1936: Burdick & Whitney, 1937; Greenwald. 1965; Chang & Yanagimlichi.

1965). Coinciding evidence suggests that administration of ACTH can increrise endogenous estrogen levels in sheep (Strott et. al.. 1975) md androgen levels in the domestic dog (W~sserrnanet al. 1978).

The viability of pregnançy is dependant on a specitic ratio of esuogen to progesterune. The.% two hormones act synei-gisticrill y ti ) conuol the rrite of em biyonic uavel and development through the faiiopian tubes md the precise time of implantation (Smith & Biggers. 1968: Roblero, 1979: Liptrrip. 1903). Any disruption of this balance crin result in pregnancy loss (Gidley-Baird. 198 1: Gidley-Barid et al.. 1986). Therefore any disturbance in the maternai hormone envirrment by stress-induced elevation of rindrogens and esuogens could contribute to the early demi.% of pregnancy.

Red Wolf

Taxonomv

Within the order Cmivora. the hrnily Crinidae consists of ripproximately 35 species that are critegorized into 15 genera. The genus Canis ccinsists tif 6 species. including varying species of jackals. the coyote. the grriy wolf and the red wolf (Sheldun,

1992).

The tlrst recorded description of the red wolf by Batrrun ( 179 1) wm in the southeuten state of Florida. However. it would not be until 185 1 that the tira valid scientitic nme for the red wolf Iç lu~usnifus). commonly known then as the "red Texan wcilF'. would be published by authors Audubon and Bachmün (185 1). The authors suggested that dl wolves found throughout the southern United States (US) were only variations or subspecies of the commonly known gray wolf (ç lu~us). During the late 1800's and early 1900's signiîicant revisions were made regarding the tzixonomy of the phenotypically diffèrent "red Texan wolf '. Bailey ( 1005) recognized the red wolf ris its own distinct species and renamed the "red Texan wolt" Ç nifii.~. Years later. Goldman

( 1944). who exxnined a luge number of canid spzcirnens in the US. detemined that the

recl wolf friund in Texas posed similrir characteristics. includinp key cranid md dental

kritures. to other cmids in the southeristeim US. Goldman therefore ~ssigneddl of the

wdves of the sciuthe~sternUS to one species (I= rufiis) and recognized three subspecies: C

-r rufus the western Tex* fcmn: ÇIgjggciryi. the centml Mississippi Valley red wolt': and 21 1 tloridstntis. the eastein subspecies. Goldman's classitications were generally well

xcepted until 1967 when riuthors Lawrence and Boxsert ( 1 067) suggested. btised on skulf

rneasurernents of several North Arneriçan Crinis. that meUurernents from red rrnd grdy

wolves overlapped md that the red wolf should be crmsidered only a subspecies of the

latter. Patadiso and Nowak ( 197 1) suggested Lawrence and Bossert's smple size was toc) srnail and by sampling rt laxger nurnber of skulls they came to the conclusion thrit the red

wolf wu indeed a distinct species. However, this would not be the end of the debate

regarding the trixanomic stritus of the red wolf.

Although some authorities consider the red wolf to be a distinct species and others consider it tu be only ri subspecies of the grliy wolf there is a third and controversial suggestion thrit the red wolf is ri hybrid. or cross-breed of the coyote and the grliy wolf. Tci date. the issue reguding the evolution and trvranomic validity of the red wolf as a distinct species is stitl widely debated and extends Far beycind the scope of this i-eview (Atkins & Dillon. 197 1: Dowiing et al.. 1992; Elder & Hriyden. 1977: Ferre11 et al.. 1978: Kurten Br

Anderson. 1980: Nowark. 1972: Nowark. 1979: Nowrirk. 1902: Phillips & Henry. I WI :

Roy et al.. 1994a: Roy et al.. 1994b: Wayne & Jenks. 1991: Wayne. 1992; Wayne & Gittleman; 1995). Despite tüxonomic squitbbling. the red wolf. as it exists todsy in captive populations is now considered by most mthorities to be a tnie species. However. the red woltvs place in the evolutionary ladder of the Cünidrie family may itlways remin uncertain. Historicd Perswtive

Very Iittle is known about the natud history of the red wolf. Based on physical appearance it is intermedirite in size between the Irirger grdy wolf (Ç lu~us).found in the northern and western ringes of North Americri. and the srnalltr coyote (j: ktri. a historicalfy western canid- The izd wolf s mtist distinguishing katures are its cinnümon coloured peIrise. long ears and long legs. It ha been suggested that the species' long slender legs were an adaptation to long distance running and pursuing prey in ccxistril prairies and river bottom swamps where it was especidly abundam

The red wolf was fonnerly distributed ncirth tc~the Ohio River Valley and Pennsylvrinia. south tci florida ruid West to central Texas (Culey. 1979: Nowak. 1972. Nowak 1996). It is believed that human pemcution played a major rde in the steady reduction of the species' range that would eventusilly confine thern to the coristd regicins of southwestern Louisiana and southeastern Texu (Crtrley. 1 979: McCürley & Culey: 1979.

Nowak. 1972: Riley & McBride: 1972). In the eürly 1900's. increzsing human population and subsequent changes in land use resulted in an encroachment un i-ed wolf habitat and ri decline of prey species. Red wolves were forced to approach man md agriculturd lands. Deeply rooted fears and a gros misunderstanding c~fwolves led to indixriminrite kiliing and predator conuol progruns that would eventudly Iead tu the removal of the red wolf ti-om its former range. It has been suggested thüt this wuthe final blow that eliminated Ç tloridanus around 1920 (USFWS. 1990: Goldmm. 1944). Deforestrition in euteni Texas and Oklahoma between 1920 and 1930. dlowed for an etistern surge by coyotes. The coyote W;LY ri much more adaptable and opportunistic species than the larser. more easily caught red wolf (Pimlott & Joslin. 1968). As red wolves were king extirpateci ti-om their former habitats. coyote populations were expanding md spretiding into these now uninhabited areüs (Paridiso & Nowak. 197 1). Human interference hrid created a situation that permitted interbreeding between coyotes and red wolves and as ri i-esuft further diluted the population of pure red wolves (Gipson et cil.. 1974). Todciy Ç L gggo-ii is believed to be the only "pure" red wolf subspecies remaining. as it is believed that Ç r l-ufus becme

extinct in the late 1960's (Crirley, 1975).

tt was not until 1962 that the scientific community was intiwmed by McCarley

( 1962) that the red wolf was in danger of extinction. The rd wolf population continued tci

be persecuted by man and the ever-expmding coyote population threatened t« ovenuheim

the species unless driatic actions were taken (USFWS. IWO). Recciverv Plan and Reintmduction

In L967. the red wolf was lis& as an endringered species under provisions of the

Endangered Species Preszrvations Act of 1966 and in 1073 was «ne of the tïrst species to

receive attention under the US Endangered Species Act The initial recovery pragm that

was established for the red wolf was btised on information which indicrited that a pure population of red wolves still existed in southest Texas and adjacent areas of Louisimü

(Pimlott & Joslin. 1968). However. field work demonstrated that ri hybrid "cuyute-wolf ' swlirrn had formed and wuspreading eastwmd (Carley. 1975). As a comequence of this finding. the recovery progriim wuredirected from an objective of lcxal preservaion to one of planned extirpation of the species in the wild. The decision to rernove the 1st ml wolves hmthe wild could only be just2ïed through the development of a longriangr: objective to eventually return the species to its native range. In the Ml of 1973. a fomd recovery plan was completed and dl remaining free- rmging red wolves were to be captured and placed in a captive breedin$cemfcütion progrm. To identify "pure" red wolves indicators included skull x-rüys. skull length. brüin to skull ratio. weight. total length, hind foot length. eüi. length and shoulder height

(USFWS. 1990). Canids determined to be po-ssible wolves were placed in the breedingkertitication program or releüsed with radio collars on public or private lands where Iandowners gave permission. Canids detennined not to be red wolves were euthanized. The captive-breedingcertification progmm was organized through a cooperative agreement between the US Fish and Wildlife Service (USFWS) md the Metropolitrin Park District of Tiiomü at the Point Defiance Zoo and Aquarium (PDZA) in Tacoma. Washington. The specitic goals of the program were to: 1) detelmine the purity of wild wolves 2) increwe the number of genetically pure red wolves in captivity and 3) maintain the gene pool for the reintroduction of the species in the wild and for distribution among zoo populations (USFWS. 1984). From the hl1 of 1973 to July I98O. over 400 canids were examined through the recovery program. From this pool anly 43 clinids met the morpholo~icalstandards used to identify red wolves (Cürley. 1975: McCarley &

Carley. 1979). Find proof of the genetic integrity «f the rinimals was deteimined by breeding sxpe~imzn~ÿnd exmining first and second generdtiuns for cvidencr of hybrid litters (USFWS. 1990). In the end 14 individurils. were certified as pure red wolves.

However. becriuse of unsuccessful breedings of certain offsp~g,2 genetic lines have been lost and the founding population of red wolves today is baxd on 12 individuds (Waddell.

1999). The 1st certitied red wolf was removed in 1970 and the species wu declared extinct in the wild by 1980 (USFWS, 1990). In 1984. the Americm Association of Zoologicd Parks and Aquariums (AAZPA) approved the red wolf for s species survival plan (SSPO) which would involve 1 participatinp fxilities and 63 animals. As captive husbandry techniques were reîïned and reproduction increased. prepantions to initiate a reintroduction program progressed

(Waddell. 1996). By 1987. the fust reintroducticm attempt wu conducted at Alligator

River Nationai Wildlife Refuge in northeatem North Carolina. As of 1998. there are a minimum of 75-80 red wolves considered to be free ranging. 175 animüls remaining in captivity and 33 institutions participating in the red wolf SSPO (Wriddell. 1999). The i-ed woIf recovciy program is considered to be one of the Endangered Species

Am' success stories (Waddell & Henry. 1996). Since the tirst reintmduction of the red wolf to its former range. wild populations have continued to grtiw and i-eproduce successtùlly on their own. However, although red wolves have prciven them,selves in the wild. the tntire red wolf population is founded on anly 12 individuals (Waddell. 1999) and

LS discussed rit a iecent population habitat vizibility assessrnent (PHVA) meeting. red wolf hybridiwtion with coyotes is evident arnong wild populritions (W. Wriddell. personal communication). Theretore. the tightly controllsd genetic management of the captive population acts ;t,s a bufier, ensuring the future of wild populations in terrns of genetic integrity and availability of individuals for future reintrocfuction areas.

In recent years. the growth of the captive population has slowed due to restncted reproduction as a result of ri limited number of holding are*. This ptx.sents a problem for the long-term maintenance of the captive population. With limited holding areas. and an ever decreasing nurnber of breeding age animals. the conservation of the genetic integrity of the population is rit risk. Upon recommendation by the red wolf recovery tem. =istzd reproductive technologies such ris &ficial insemination and sperm banking ÿre kin~ investigrited tn help overcorne this problem and to allow foi- more flexibility in the breeding program. B y using these technologies. geneticdly vrilulibIe individuals could repl-oducz even if they rire Icxated in dieerent mas of the country. six bzhriviourally incompatible or even if they are deceased. &isted reproductive technologies have the potential to becorne valurible management tools, ensuring a genetically diverse population in the future.

Repeated ritternpLs at artificid inseminations (Al) usinp ksh and frozen-thawed sperm have resulted in little success in red wolves. A factor impeding the potentid success of AI rnay be the frequent rest.int episodes used to xquire btocid samples for hormone analysis and timing AI. It hris been demonstrated in seveixl domestic species thrit stress mriy alter reproductive succes (Liptriip. 1993). Theretim. an dternative non-invasive approach fur monitoring ovu-iim status wu desired. In addition to femüles. knowledge of the basic reproductive chwicteristics of the mde red wolf is limited with the exception of three studies conducted tu dettne semen chaiacteiistics (Koehler et al.. 1994: Goodrciwe et al.. 1998: Koehler et al.. 1998). As sperm bsinking is a major focus of the breeding pi-opam. a detüiled endocrine evduation detining normal annual reproductive paxmeters may prove to be advrintri,WOUS.

In respon.se to concerns of the red wolf recovery prciprim. ciur iriboritory proposed to conduct endocrine studies thrit would not only increase the howledge regarding the basic reproductive physiology of both males and kmdes but potentidly enhance reproductive success. Previous studies have shciwn that long-term evduaticin of steroid metaboIite concentrations in feces is effective for evduatins ovarian. testicular and adrenal activity in numerous carnivores including the mink (Mustela vison: Most1 et al.. 1993). cheetah (Acinonvx ~&Nw:Brown et al.. 1994: Graham et al.. 1995: Brown et al.. 1W6a). domestic cat (Felis CU: Czekalii et al.. 1994: Grahm et al.. 1995: Brown et al.. 1996b:

Jurke et al.. 1997). clouded leopard (Neofelis nebulosri: Brown et al.. 1995) rnaned wolf (Chrysoçon brichyuru5; Wasser et al.. 1995: Velloso et al.. 1998). and African wild dog (Lycaon pictus: Monfort et al. 1997). Cornpu-ed to bliwd sampling. fecal monitoring permits longitudinal non-invasive studies of stemid cxcretion and allows for a more complete charticterizrition of reproductive events for a given species.

The hypothesis of this study was thrit ovririan and testiculu steroids could be qurintitritively and quditritively measured in the feces of the femde and male red wolf. respectively. The tint objective was to develop ri non-invasive technique hi- monitoring ovarilui and testicular steroidogenic ttctivity in çanids by validating m extiaction protcxol to measure fecal progestin. estrogen and testosterone metabolites by enzyme irnmunoassay (EIA). The second objective was to establish a databue of endocrine norms for both males and fernales. in general. a detailed exmination of the normal pattems and their range of variation would dlow tor the recognition of disturbances in these patterns in future analyses. These longitudinal analyses cciuld potentiaily diagnci.~cases of intërtili ty. test efiectiveness of contraceptive treatments or determine intluences of exteinal factors such ris husbandry. nutrition and environment on reproductive function. In fernales. by dexribing endocrine changes during pregnancy and pseudopregnancy. euly pregnancy dekction or insight into faiIed conception rnight be possible. From ri husbmdiy perspective. the abiiity to detect pregnancy woutd facilitate the requirement of special care for pregnant mimals. In males. longitudinal monitoring would dlow for investigation of the relationship between changes in fecal testosterone concentrations and photopeilod.

The third objective was to elucidate the relationship between fecal estrogen and prosesterone metabolites during the periovulatory period. If fecal steroid patterns were similar to those observed in serum. t'ecd monitoring could be u.sed as ti diable alternative to blood sampIing tor indexing ovarian function and timing AI.

The fcwrth objective was to develop a non-invasive technique to monitoi- adrend activity in red wolves through the validation of lin exwxtic~nprcitwol to me~sui-ecortisol metabolites by EIA. Monitoring cortisol as a rneüsure of 'stress' may dlow tiiture investigations to determine suessful situations for the wolves. Removing or limiting these stresstiil situations rnay improve reproductive efticiency.

The tifth objective wri'; to evaluate the potentiril use of tècd steroid anrilysis in gender determination. This technique WOUId facilitate the tracking and reproductive monitoring of animais in the wild. Each of these pals have great potentiril as vduable tools towards the management of red wolf population and wuld irnprove the reproductive success for this species. METHODS AND MATERIALS

Fecal Study Animals and Sample Collection Animais

Intact adult (n=l6. age range 4-13 years) and immature (n-1. yearling) tèmde red wolves were mllintliined in hcilities ticross the United States during the 3 yew study period

( 6 198) The fclcilities and their relative geogrtiphical tocritions were: Burnet Park Zoo. NY. 43.0"~76.1"W: Great Plains Zoo. SD. 43.5"N 96.7"W: Knoxville Zoo. TN. 36.0"N 83.9"~:Oglebüy's Good Zoo. WV. 40.l"N s2.RW: Point Defiance Zoo &

Aquarium md Red Wolf Breeding Ficility. WA. 47.2"N 122.TJW:Racine Park Zoo. WI. 42.7"~87.8'~: Ross Park Zoo. NY. 32. lUN 75.gi'W: Western North Carolinri Nature

Center. NC. 35.6"N 82.6"~. Intact adult (n=5. age range 4-8 years) suid immature (n=l. yearling) male red wolves were maintained during the the 1 year study perd at the fdlowing institutions: Bumet Park Zoo. NY: Knoxville Zoo. TN. 36.0"N 83.9"W: Mill Mountüin Zoo . VA.

37.3"N 80.OuW; Point Defiance Zoo & Aquarium and Red Wolf Breeding Ficility. WA: Trevor Zoo, NY. 4 1 .KUN37.7'W.

Depending on space availability md red wolf SSPo brrreding recommendutions. wolves were housed either individually. with potentiril pariners or in a fmily unit. Ali wolves were housed outdoors in chainlink pens (nci less thun 5000 square ket) which included a dry den structure. Since dl animais were considerd potential relesisc: candidates. a 'hands-«t'f policy wuimplemented to avoid wol f habituation tv Iiumüns. Therefore. wolves were subjected tii a minimal numkr cif disturbances other than Jaily husbandry routines. AI1 wolves were fed t -2.5 pounds of a high quality commercial dry dog food 6 times a week and were hted 1 day per week. During the whelping period. a îkmde's food ind-e wumonitored and additional food was prcwided if wamnted. Al1 wolves had ad Ii bitum access to fresh water. Fecd Collection Fecd simples from témales were collectecf during the breeding seacson, mid-

Decemkr until the end of May (1996. 1997. 19%). on average 4 times per week from each individuai during driily husbandry routines. Fecd .simples frorn males were conducted on average 4 times a week over ri 1 yeu- pei-iod beginning inid-November, 1997.

The entire tècal smple was retrieved into a zip-lc~kplastic brig. lribeled with the date and animal's studbook number. shipped frcizen tci the Toronto Zoo and then stored at

-20°C until stemid extraction. In crises where a study individual could not be isolateci tiom other wolves. fecal samples were marked with small 0.5 cm s 0.5 cm pieces of coloured surveyors tape (Hanson Co.. Fnnklin Park. IL) or by locdly purch~serlgreen vegetüble food colounng. The study animal would be fed a smdl mcütbdl made tkom ground rneat which contained the labeling substance. The Iabeling substance was subsequenti y excreted enabling for animal idenbîkation.

Serum Study Anirnals and Sample Collection Anirnals

Blood smples were obtained tiom intact fernale red wolves (n= 12: age range 3- 1 I years) hou& at the Red Wolf Breeding Facility in Graham. Washington frorn 1992- 1997. Fnim 1992-1997 waIves were monitored for rising progesterone Ievels tci time Ai. The husbandry and housing of riil mimals were the same as described above.

Blocid Collection

To obtain blood samples, wolves were rillowed/directed to ictreat into their dens cir a hoIding/shifting areri. The enuance was blocked with a net tci prevent the animal from escaping. Personnel fmiliar with the wolves then tirid controlled Iiccess into the contined area. A noose (5 foot pole. Ketch-Al1 Company. San Deigci. CA) wuslipped around the wolfs neck ruid the animal secured. The wolves wew manually restiained and blood sarnples taken from the cephalic vein from Iüte February through early Mmh. The process was repeated every 1 to 2 days until blood progesterone Irvéls were found to be nbove 5.5

&mi. On average. each wolf wu restrriined 1 1 times. Blood samples (3-6 mis) were placed into red top vacutüiners@ (Becton Dickinson Vxutriinzr Systems. Frmklin Lrikes.

NJ) and dlowed to cotigulate for - 15-20 mins. Stimples were then centrifuged for 10 minutes at 3500 rpm and the resultant semm fraction wu removed and stored in cryogenic vials (Fisher. Fair Lawn. NJ) at -2m until hoimonal analyses.

Fecal Steroid Hormonal Analyses Fecal Steroid Extraction Progesterone. Estrogen and Testosterone metabolites Method 1

Fecal progestin (P4). estrogen (E2) and testosterone (T) metabolites were extrricted from ail samples according to a method described by Graham et al. ( 1995). TO rnsure uni fcirmity of hormonal metabolites within Lècd smples. thawed samples were thoroughl y mixed within their plastic storage bags. A wet weight of 0.5 g of feces wxs combined with

4.0 ml methmol. 0.5 ml ciistilled water and 1 .O g alurninum oxide in a clean 15 ml glas tube (VWR/Crinlab. Missisauaga. ON). Smples were then futher mixed manually with a metai stir rod. A tetlon-lined cap wu then screwed onto the tube and the sample was vtii-texed for 30 seconds. Samples were mixzd at rciorn ternperriture foi- 1 hour on ri rotator and then centrifuged for 10 min rit 3500 rpm. The methmol friction wlis decanttrd into ;i Iübeled 5 ml polypropylene tube (VWR/Conlab. Missisauga. ON) and stored at -20°C until hormonal rinalysis. Method 2 As part of the tissay validation process. rxogenous P4 and €2 were exvacted from fecal samples according to methods moditied from Graham et al. (1995). The method was identical to that of method 1 with the exception that aluminum oxide wudeleted. Method 3

As part of the ;~xsayvalidation pmcess. in addition to methodi; I md 2. zxogenous

steroids (P4 and E2) were extmcted from feu1 samples using a moditied method from

Shideler et al. ( 1993). Once frozen sümples were thawed. 0.5 g wet fzces were combined

with 2.5 ml of extraction buffer (aqueous solution of O. 1 M phosphate buffer containint

0.149 M NaCl. 0.1% BSA. 10% methmol and 0.2% Tween) in a 5 ml polypropylene

tube. The mixture wuhomogenized by hand with a stir nid then capped and vortexed for

1 min. Tubes were placed un a rotatins shaker civernight ( - LX h) at rcic>m ternperxture.

Following centrifugation for 10 min at 3500 rpm. the supernatant was tixnsfened into a

liibeled 5 ml polypropylene tube. sealed with a cap and stored ût -20°C until malysis. Cortisol

The protoc01 for extraction of corti.wl difièrecl hmthat of ovarian and testicular

steroids as it has ken demonstnted thrit the removal of the duminum oxide hmthe extrriction proc-s greatly improved corticosteroid extraction et?-ciency in rinother carnivore species (black-footed krret: Young, 1998). Tri ensure sample unifcmnity. thawed fecal

samples were thoroughly mixed within the sample bags. For extraction. 0.5 g wet feces

were ridded to 2.5 ml of 90% methanol in a 5 ml polypropylene tube. The mixture was

then mixed with ri stir rod. capped and vortexed for I min. The smple tubes were then

plriced on a romtor and shaken for 2 h followzd by centfifugation foi 10 min üt 3500 rpm. The supernatant wuvünsferred to a çlem. labeled polypropylene tube. cüpped. md stored rit -20OC. Prior to hormone analysis. 20 pl of extract was aliquotted into 12 x 75 mm glas culture tubes (VWFUCanlab. Missisaugü. ON). submersed in n 34OC waterbath and dried down under a light flow of nitrogen gas. Samples. accounting hi- the 1:35 dilution required for hormonal anaIysis. were then reconstituted in 700 pl enzyme immunoway

(EIA) but'fer (0. I M phosphate buffer, 0.1% bovine serum albumin. pH 7.0). Enzvme Immuno~~w

Fecal extracts were andyzed for steroid hormone concentritiîins by EIA ris previously described (Munro and Stabenfeldt. 1985; Munro et al.. 199 I ) and as adapted for use in our laboratory with wood bison and blxk footed-ferrets (Matsudil et al.. 1996:

Young. 1998: Othen. 1999). Feces collected from fernale red wvlves (n= 17) were quruititled for P4 ruid E2 concentrations while those collected t'rom males (n=6) were analyzed t'or T metabolites. As ri pilot study. fecd cortisol metabolite concentmtions were measured in 2 fernde red wolves: one fernale that endurecl eptsd restrint episodes for blood smpling to time for AI and one tèrni.de thrit wunrit restriined and becme pregnant nriturrilly. Randomly selected male and femde kcal samples (n=960) were anülyzed for P4. E2 and T metabolites for the gender determination study. Pro,*esterone

The polyclonal anti.serum mised in rabbits to progesterone (R4861) was provided by C.J. Munro (University of California. Davis). The cross-rerictivities for the rintibody

(Appendix II A) were provided by Munro and Stabenfeldt ( 1985).

A primary utibody stock of 1:20 wüs prepared by dilution in 50 mM sodium bicarbonate coating buffer (pH 9.6) and stored at -20°C until required. One day prior to running the assay. the primary antibody stock kvas further cliluted in coating bufier to a working stock of 1: 6,000. Microtiter plaies (96 wells) were corited with SOflweli of working rintibody stock and tapped slightly ta sc=ttle the rintibody. Plates were ccnwed with an ricetate plate sealer (VWR/Canlrib. Missisauga. ON) to prevent evaporrition and incubated overnight (14 - 18 h) at 4°C in ri styrofoam container. The tirst column on these plates was not used becriuse of high variability in rintibody binding. Progesterone çontrols. standards and sample extracts were prepared in glass culture tubes by dilution in EIA buffer immediately prior to running the rissriy. The progesterone standard curve raoged 2-fold hm9.75-2500 pg/well. plus 'O' wells consistinp of EIA bufier only. Progesterone conuol stock was prepmd by extrrlcting feces. using the protocol of' Schideler et al. (1993). from a pregnant lowland goriIlri- The two intemal controls were prepxed by diluting the control stock 1: IO and 1: LOO. Bued on -50% binding from pmllelism resuits. .;ample extracts were diluted 1 :7.

At'ter overnight incubation, the microtiter plates wre rin.*cl 5 tirnes with ri wash solution (O. I5mM NaCl. 0.05% Tween 20) in a Dynatech Ultrtiwash II mircopiate washer to remove unbound antibody. Plates were then patted dry to cnsure the removal of excess wash solution. After washing. standards. controls and smples were loaded in 50 pi/well atiquots onto the pIrite. Dilutecl extrxts ünd contn)ls were pipetteci in cluplicate and standards in triplicate: a pair at the beginning of the plate and ri third rilicluot at the end to account for any tirne 1% or drift across the plate. Once al1 known and unknown samples had been ridded. 50 pl progesterone-horseradish per~ixid~ci(progesterone-HRP: working stock 1:60.000 dilution in EiA buftèr: primary stock supplied by C.J. Munro, University of Cidiforniri. Davis) wuridded to each well. Two weIis in the tint column of the plate served üs blanks for calibration purposes and were loaded with HRP only.

After loriding. plates were covered with an ricetate plate sealer aiid incubated for 2 Ii at room temperature. Aftzr incubation. unbound material wu removecl from the wells by washing 5 times with wash solution and patting the plates to dryness. One hundred pi of freshly prepared subsuate solution [0,4 mM azino-bis (3-cthyl-knzthiazoline-6-sulfonic acid). pH 6.0 and 1.6 mM H,O, in 0.05 M citrate buffer. pH 4.01 wuthen pipettecl into exh well. Plates were sealed with xetate plate .sealers and slowly shaken on a tlat rotüaw rit room temperature for ripproxirnately 15-30 minutes. The optical density ot' each well wris read using ri Dynatecli 700 plate reader (test tilter 405 nrn and reîèrence tilter 630 nm) inteficed with a MacIntosh computer. Mid-study. this plate reader wu updated with a Dynex Technologies MRX plate reader interticed with a PC computer. Standards. controls and unknown values (pdwell) obtained from reading across plates on both plate readers were comparable. Estrrid io 1

Fecal E2 wuquantitied with ün EIA. using a polyclonül rintibady (R4972) ri.sed in rribbits and HRP conjugate supplied by C.J. Munro (University tif California. Davis). The antibody cross-reactivities (rippendix II A) were supplied by C.J. Munro (University of Cal ifornia. Davis).

The working antitmdy suid HRP conjugüte were diluted 1 : 10.000 and 150.000 in coating and EIA bu fier. respectively. Standards were serially diluted ?-Md 1.95-500 pghell in EIA buffer. plus 'O' wells cimsisting of EIA buffer only. Using ELA buffet- fecril samples were diluted 1 :4. while interna1 contitjls were dilutecl I : l () and 1 :50 ft-om the control stock. The estrogen control stock was prepared ti-C-mthe supernatant of extrdcted feces (method of Schideler et al. 1993) from a pregnant lowland godla. With two exceptions. the estcadi01 EIA was identical to the pn)psterone EIA: 1 ) following the initial wrish to remove unbound antibcidy. plates wzre coated with EiA but'tèr 5OpVwell ruid then incubated for 2 h at rwm temperature before any samples were loaded onto the plates and 3) 20 pVwel1 of the standards. samples and contrcds were Ioaded onto the plates. Testosterone

The polyclond testosterone antibody ( R 1 56/7 ). HRP conjugrite and cross- reactivities of the antibody (rippendix II A) were supplied by C.J. Munrci (University of

California, Davis). The antibody and HRP conjugate were diluted 1: 10.000 and 1: 15.000 in cciating and EIA buffet- respectively. Standards were prepared 2-fold. 4.X- LX0 pdwell in E[A buffer. plus 'O' wells consisting of EiA buffer only. Fecal samples were diiuted

1 :40. while interna1 controls were diluted 1:65 and 1 :650 ft-om stocks. al1 in EIA buffet-.

Testosterone control stock wrts prepared from the supernatrint coilected der exmcting tètes from male red wolves during the breeding .season by extixction method 1. With one exception. the testosterone protocol is identical to the propesterone EIA; following the initiai wash to remove unbound antibody. 50pI of EIA buffet- were aliquoted to each well and the plates were incubüted for 30 minutes at rom temperature before the addition of standards. controls. smples and HRP. Cortisol The cortisol polyclond mtibody (R4072). HRP conjugate tuid mti,seium cross- reüctivities (appendix II A) were supplied by C.J. Munro (Univemity of California. Davis).

The mtibody and HRP conjugate were diluted 1 :X500 and I :50.000 in coating md EL4 buffer respectively. Standards 3.9- 1000 pdwell wei-s prepued 2-f&J in EIA buffer. plus

'O' well consisting of EIA buffer only. Fzcd samplzs were Jilutzd 1:35 with EiA buffer. while interna1 controls were diluted, in EIA buffer. 1:2 aiid 1 :4 from stocks. The control stock was prepared tiom a neat urine simple colIected from a single mde wood bison during the fall rut. This assay was conducted in a similu manner to that of the progesterone assay except plates were incubcited for I hour at ruom temperature îbllowing the loading of the standard. samples. controls and HRP.

Axsav Validation

Pmllelism Fernales Fecal simples taken pre, during and post breeding .season (6 wolves. 10 fecd sümples each) were exuücted (rnethod 1) to evaluiite parallelism tif P4. E2 and T metabolites in the tèces to a standard curve. Pooled îèciil estracts (20 pl t'rom each exmct) were secially diluted 2-hld in EIA butkr tci yield a rige OF dilutions ti-orn neat to 1 :204X. The dilutions were analyzed against progesterone. estrridiul and testostercine standard curves. Fecal samptes from 2 tèmale wolves that were undisturbed (20 fecal samples each) and ti-om 2 femde wolves that were subjected tc) restriint episodes for blood sampling (20 fecal samples each) were used to evduate parallelisni with ri coitisoI standard cui-ve. Fecd samples were exwrlcted using the methods of Young ( 1998) ris de,scribed eulier. The resulting extracts were pooled (20 pl hmexh extract). .wrially diluted 2-Md neat to

12048 and andyzed against a cortisol standard curve. Males

Mde fecal .simples (2 wolves. 50 kcal samples ericli) collected trcim November ti)

Much were extracted (method 1 ) to evduate pwillelism of P4. E2 ünd T metabolites in the feces to ri standard curve. Fecal exincc were pociled (20 pl frorn each extract). .serially diluted from neat to 1 :?O48 and mdyzed rigainst a progesterone. estriidid and testosterone standard curve. In dl cases. the mean percenmge bindinss obser-ved from the standard ulves were plotted against their expected log~thrnicdlytransfomçd concentrations (pg/well) on the x- mis. Mem percenÿige bindings observed from the diluted pooled extracts were plotted 2- hld against the transformed concentrations on the x-mis.

Recovery - Progesterone and Esuridiol

To compare extraction methods 1. 2. and 3. the peixent recrivenes of exogenous estmdiol and progesterone from tecal extracts were exiimined. The fecal exuac~swere produced by thoroughly mixing together 4 entire wet fecal samples obtriined tFom anestrus fernale wolves (nd). The p:wled fecal sample was then clivided. and for each method ( 1.

2 and 3) six 0.5 g ex~lictionswere made. The resultant supernatants frcim each extrrlction method were pooled together. producing 3 .sepante pciols (-27 mls exh. one kir each extraction protocol). One ml diquots from ex11 of the pools weir spiked with increüsing rimounts of exogenous progesterone and estmdiol standards to cibtain expected concentntions, O (unspiked). 19.5. 39. 78. 156. 3 1 2. 625. 1250 and O. 3.9. 7.8. 1 5.6.

3 1.2. 62.5. 125. 250 pg/well. respectively. Sample dupliciites were then rtndyzed for the appropriate steroids as described übove to obtüin the cibserved pg/well value. Extraction Efticiency - Progesterone and Estriidid

Feces collected from anestru adult îèmale wolves (n=4) were pooled as dexribed above. and osed to asses methds 1. 2 and 3 for exogencius steroid extmction efiiciency frorn îècal samples. Inçrasing levels of steroid standards were added to 0.5 g aliquots of pooled wet feces prior to extriiction with methods 1. 2 ruid 3. The expected spiked ccincentntions of progesterime and estridi01 wei-e O. 19.5. 39, 78. 156. 3 12. 625. 1250 and 0. 3.9. 7.8. 15.6. 3 1.2, 62.5. 125. 250 pg/well. respectiveiy. Srimple duplicates of the fecal exuacts were assriyed for the appropriate steroids as previously dewribed above to o btained the o bserved pglwell value.

Extrxtion Eftlciency - Testosterone

An endogenous T source wu useJ tii asszss estracticin efticiency f'rom fed srimples. The endogenous source wuobtained ti-cirn adult male (n=2) fecd srrrnples during the breeding season (late Febniruy to eürly March). The ficd smples were extriicted (method 1) and the supernatants were pooled together and analyzed for T concentmion

(pghvell). Because of the low T concenmtion in the pooled extract. three different volumes of the pooled extract were placed in glas tubes. dned down under a light tlow of niuogen gas and i-esuspended in an equal simount of EIA buftkr allowing for expected concentrations of 155.5 19 and 950 pg/well.

Feces coiiected from anestrus adult temale wolves (nd) were pciciled and used tri assess the endogenous testosterone extraction eff'iciency hmfecal samples. Aliquow of

0.5 g of pociled wet feces were added to the 3 spiked plass tubes containing the different levels of expected concenuzltions dong with 1 unspiked tube and rillowed tc) sit rit room temperature for 24 h prior tu extraction (methcicl 1 ). Srimple dupliutes of the extracts were assayed for the testosterone ris descnbed above to obtained the observed value pglwell. Extraction Efiïciency - Crirti.wl

An endogenous cortisol source was obtained hum a series of fecd smples collected fmm manudly resuained. adult fernales (n=4). Samples were extractecl using the cortisol extraction protocnl. pooled together. then serially 2-fold diluted and analyzed on the cortisol msay to deteimine expected concentrations of 40. 103. 167 pglwell, Feces collected from mestrous adult femrile wolves (na) were pcmled llnd used to a.sess the endogenous cortisiil extraction efticiency from fecal samples. Aliquots of 0.5 g of pooled wet feces were spiked with the expected concentrritions and dong with 1 unspiked smple. dlowed to sit rit room temperature for 24 h prior to extraction using the cortisol extriiction method. Sample dupliclites of the fecd extrxts were maty~dfor ccirtisol metabolites as previously described above to obtained the obseived pglwell value. In dl examples. the crikulation of percent estrxticin efticiency or recovery was equd to amount observed minus the biickground (unspiked smple) divided by amount expected. High Petic1rmmce Liquid Chromatognphy

Fecd extracts (n=7) chosen for high performance liquid chromatography (HPLC) included: 1 during the tolliculw phase and 1 during the luted phw tCom two ovulatory nonpregnant female wolves. 1 elevated cortisol srimple from a female woif and 2 elevated T concentrated srimples from two male wolves. Steroids in fecal exwricts were recovered for sepüration on HPLC by solid ph- extrxtion with Sep-Pa C,, cartridges. Cmidges were primed with 5 ml methmol followed by 5 ml distilled writer. Fecal extracts wese dried under nitmgen tbw. resuspended in 5% methanol/writer and allowed to pei-colate through the caiuidge. The cartridge wu washed with 5 ml distilled water and the steroids were eluted with 5 ml diethyl ether (unconjugated steroids) and 5 ml methanul (conjugated steroids). successively . After evaporation of the ether under ni trogen tlow. the residue (unconjugated fraction) was dissolved in 100 pl acemnitrile foi- HPLC. Sepriration of steroids was accomplished by injecting a sample (25 pl for progesterone and estrogen: 50 pl for testosterone and cortisol) on rr HPLC systern consistinp of twu Watei- Mode1 5 10 pumps

(Waters Associates Milford. MA). a WISP 7tOB autoinjector. and a dual channel mode1 4-1 1 absorbante deteetor. For optimum sepention of steroids. Waters Briseline 8 10

Controller Software wuused to create a binary solvent gradient of acetonitde-water on a Waters Nova.PI C,,(8 mm x LOO mm. 4pm) column. In total. 2 1 steroid standards were eluted at rmm temperature aver 35 min rit a rate of 7 ml per min. The multistep gradient mobile phase consisted of29.5.35. 39. and 75% acetonitrilc in water at Ume O. 10. 12.5. and 32 min. respectively. in the run. Retention times for cach stemid were determined by injecting 20 ng of each standard. with absorbmce monitored üt 154nrn ünd 280nm. Fractions (0.5 min. progesterone and estrogen: 1 min. testosterone and cortisol) were collected automaticrilly wi th an LKB RediFrx fraction col lector. Aliqunts (200 pl) from euch friction were then JrieJ under nitmgen gris. resuspended in an equal rirnm.int of EIA buffer and assayed in duplicate as described eürlier to cvaliiare immunorerictivity.

Blood Steroid Hormonal Analyses Rridioimrnuno~~;~us

Appropriate nidioimmunoassûys (RIA) were used to determine steroid concen~itionsin -serum samples. Blood samples collected from females (n=23) duriny breeding seasons ( 1 99 1 - 1 997) were quûntitïed for progesterone. csuüdiol. and luteinking hormone (LH) concentrati~ms. Semm progesterone and estradiol were memmxI using prupsterrint and estradici1

'"1 RIA Kits (Corit-a-Counta. Diagnostic Produc~sCoiporition. Los Angeles. CA). AH samples were malyzed at nrat concentration. Methods and cross-reactivities were provided by kit litemture (Coat-a-Countm. Diagnostic Prducts Ciqxmtion. Los Angeles. CA).

Serurn LH concentrations were quantified with a LH RIA developed by Dr. J.L. Brown (Brown et al.. 199 I; National Zoological Park Conservation and Research Centre. Front

Royal. VA). Al1 samples were malyzed at neat cunçentriition with the exception of sample< taken during the LH pe*. These samples were ünülyzed at a 150 dilution in RIA buffer

[In 1 L pink PBS (0.01 M phosphate buffrr. 0.9% NaCl. 0.0 1 % thimerosÿl) üdd 0.05%

BSA, 2 mM EDTA. pH 7-41. Senim LH waii meaured iising a rnoncxl~malmouse LH anti body (5 1 8-B7) and an ovine LH standard (NIH-LH-S 1 8) and LH label (LER- 1374A). The cross-rextivities for the antibody were supplied by Dr. J.L. Brown and are listed in appendix II B (Matteri et al.. 1987).

A primüry antibody stcxk of 1: 10.000 was prepxed by dilution of' the ruitibody in RIA bufier and stoi-ed at -20°C until required for u.se in the a~süy.The LH standard curve rünged 2-fold from O. 0.08-20ng/ml diluted in RIA buffer. Each standard dilution was stored at -20°C untif required.

Just pnor tc.1 conclucting the iissay. the primüiy antibody stock wu diluted in RIA but'fer tri a working concentmtion of 1: I .(H~O.OOO with i :-C(M) not-md mouse srum (NMS).

After thawing. smples (in duplicate) and standards (in triplicrite) wei-e riliquoted ( 100 pl 1 to 12 x 75 mm disposable culture glas tubes followed by the addition of LOO pi of the working mtibody to each tube. Duplicate ~1~1,s~tubes alsci were set aside for total counts and non-specitic binding (WB). The NSB tubes were incubrited with 100 pl of 1:400

NMS diluted in RIA buffer. The tubes were agitated (-3 min) and Ieft at mom tempeiaturz wernight. The tubes then received 100 pl labeied ovine LH which wu diluted in RIA bufier (-20.000 c.p.m J I O0 pl). The tubes were rnixed and re-incubated ovemight at rvom temperature. On the final day of the assay. antibody-bound complexes were precipictrted by incubation for 1 hr with 1 ml aliquots of goat anti-mtiu,se gamma globulin (GAM; preprired

1:400 in pink PBS with 5% polyethylene glycol). Tubes were centrifugeci at 3000 g for 25 min. decanted (except for totals) and each counted in a gamma counter foi- 1 min.

Gender Determination Fecd samples collected from males fl998: 5 profiles from 5 rinimals) and fernales

( 1996- 1998; 15 profiles from 13 animals) from December 2 1st (winter solstice) to June

2 1st (summer solstice) were used in the gender determination stiidy. To test the ability of kcaI steroid analysis to distinguish between .sexes duiing the selected time period. smples were divided into 4 penods (Dec. 21- Feb- 3. Feb. 4-March 20. Mmh 2 1 -May 5. May 6-

June 21). From each wolf. four kcal .simples within each periixl were randornly selected and analyxd for P4. E2 and T. Six variables (P4. T. E2. P4/E2. PU. E2K) were compared within each time period among 4 classes (males n=5. pregnant n=3. iivulatory nonpregnant n=9 and nonovulatory krndes n=3) using factorial ANOVA (Statview version 4.5. Abacus Concepts. Inc.. Berkley CA).

Data Evaluation Fecal Pro ti les

Fernales were ;~,~.ignedto 1 of 3 clil,sses xcording to the rxcumiict: of birth

(pregnant: P) and presence (ovulatory nonpregnant: NP) or absence (non-ovulatury or xyclic: AC) of a continued rise in P4. which presumribly retlected ovulation. Profiles from individual fernale red wdves (P. NP) were aligned to ri.= in P4 (day O) from baseline values using the following critelia: a baseline vriliie for each wolf wudetermined by calculatint each individual's mean P4 value from sümples aillected mid December until mid February. Day O was recognized as the value thrit was elevtited above badine by 2 standard deviritions (SD). with subsequent levels remstining èlevatzd for a minimum of 3 consecutive snmples. For AC animais. day O wurissigned baseci on the location of the individuai animal (Southeastern. Northeslstern and Nurthwestern United States) ünd the mean estimated date of estrus for dl wolves in that location ('Wriddell. 1995). Since fecd srimples were not dwriys available on a ddy basis. fecal hormone levels were pooled into

3 day means. Reproductive protïles were then divided into 9 cycle stages: pre-ovulatory (days -30 to - 13). periovulatory (days - 12-8). luteai phases (days 9- 17. 18-26. 27-35. 36-

44. 45-53 ruid 54-65)and post (66-71). Cornparison of P4 and €2 levels within each cycle stage between the 3 cl;~~,ses(P. NP. and AC) wxs determined by ANOVA with repeated measures usincg Statview (version 4.5. Abacus Concepts. Inc.. Berkley CA) on a Macintosh cornputer. Cornparison of P4 and €2 levels ricross dl cycle stages within each cl~ss(P. NP. and AC) were conducted using a factorial ANOVA andysis (Statview version 4.5, Abacus Concepts. Inc.. Berkley CA). Al1 p values reported rire derived from the Fisher's protected least significünt diftèrence test. Descriptive data rrre repoired as

means 2 SEM within ri cycle stage.

During the peiiovulritory pend, daily mem P4 aiid E2 values fwm NP wolves were cornpared to basai cr.incentrations using a Student's t-test. [nsuft?cientsmpling from

P animais did not permit the iinsilysis of daily means. Descriptive data are reporteci as daily means + SEM.

Oves the 1 year stlmpling period. male tecal T levels wwe pooled into wtxkly mesins. Compuisons among weekly means and daylength were detemined using correlation ruialysis (Statview version 4.5. Abxus Concepts. [ne.. Berkley CA).

Descriptive data are reportecl ris weekly mems + SEM. Daylength data was downloaded hmthe The United States Naval Observatory (www.usno.nüvy.mil/hc~me.html).

Blood Prt~tiles

As bmline progesterone values in red wolves have never ben chiuCiterized. dignment of data wu detemined in a similür mmner to that uf fecril samples. Individual semm prcit'iles were digned to a nse in progesterone (day O). BecauLeuf the smdl numbei- of pre-ovuiatory siunples. day O for each individual wu the value thüt WLS elevated abrive ri mean baseline (mean progesterone conçentmtion from al1 wolves Februüry 2nd-22nd) by 2

SD. with subsequent levels remaining elevated for ri minimum of 3 consecutive samples.

Luteinizing hormone and es~ddiolprotiles were normalized by riligning the scrum hormone values of individual animals to the day of progesterone rise. A Student's t test was employed to detect a signiticmt elevation in driily means abcive basal concentmticins of progesterone. estmdiol and luteinizing hormone. Values se repuned ris daily mems 4

SEM. RESULTS

Validation of Fecal Assays Pmllelism

The curves generited from the ~wrïdlydiluted pooled fecd estracts from femdes

for P4. E2. T and cortisol were pürallel to tlieir respective standard ciiives (Figures 1 a. 2rl.

3a. and 4). Serd dilutioiis of tècal extriicts froin males Save displacement curves pwdel

to the standard curves for P4, E2 ruid T (Figures 1 b. 2b. and 3b). Recoverv - Prwesterone and Estr;idiril

The mean percent recoveries of lin exogenous source of progesterone from fczll extracts for extraction methods t, 2 and 3 were 7 1.2% (range 62.8%-78.5%). 40.9%

(range 25.6%-48.8%) and 60-8% (lange 58.0%-88.9%). respectively. For estrddiol. the

mean percent recoveries tiir the 3 protocols were 78.54 (range 64.1%-92.9%). > 100% (range 90.4%- 1 18.3%) and > 100% (range l28.W- 173.19). respectively.

Extraction Efficiencv - Prozesterone and Es- For progesterone. the mean percentages of excigentus htirmcine detected from fwd samples for extnction methods 1. 2 and 3 were 77. 1% (rmge 32.0%-W.r)%). 64.6% (range 59.8%-7 1.6%) ünd 14.9% (range 7.7%-23.1%). respectively. Foi- es~adiolthe mean extriction efîiciencies for the 3 protocols were 763% (rmge 64. I %-92.9%).74.1 %

(range 64.0-89.756) md 3 1.6% (mge 12.2%-5 1.34). respectively. Basld iiii the resul~q of the recovery and extraction efticiencies for progesterone and estradiol. extmction method

1 wils used to monitor ovarian rictivity. Extraction Eftkiency - Testosterone and Cortisol

The testosterone ;t.ssriy detected a mean of 94.8% (range 92.2%-95.4%)of the endogenous T added to técal samples. The mean exaction efticiency of an endogenous source of cortisol from fecd samples wm 03.8% (i-ün~e85.6%-95.4%). LOO ,

LOO

75

50

2s

- Progestcronc (pglwcll)

-- serially dilutcd tèwl cxtract Figure 1. Panllelism curves resulting from anaylsis of serially diluted progesterone standards and pooled tècal extracts hmfemale (A) and male (B) red wolves. male

- serialIy dilutcd standard ---+- scriülly dilutcd fcmI cxtr~ci

Figure 2. Pwdielism curvcs resulting from anaylsis of senally diluted esvadiol stimdds md pooled fecal extracts from femde (A) and male (B) red wolves. 1seridly clilutcd standard

--+--- scrially diIuied cxurict

Figure 3. Piirallelisrn curves resulting hmanaylsis of senally diluted testosterone standards and pooled tecal extracts from female (A) and maie (B) red wolves. Cortiosl (pg/well)

Figure 4. Parallelism curve resulting tiom anaylsis of serially diluted cortisol struidanl and pooted fecal extracts from femde red wolves. Assriv variation

Assy variation was monitored using 1 individuülly prepared lots of controls for each EIA type. The inter-ÿssay çoeficients of variation (CV) for the progesterone ;cis;iy were 12.78% and 7.64% (n=X6). at mem percent bindings of 32.4% and 67.1%. respectively. The inm-assay CV were 5.2% (n=12) and 7.9% (n=12). at mean percent bindings of 30.18 and 62.1 %. respectiveIy.

For the estradiol usriy. the inter-;~,ssriyCV were 13.4% and 16.0% (n=85). at rnean percent bindings of 42.3% and 77.1%. respectively, The intri-~sayCV were 3.2%

(n= 12) and 3.6% (n= 12). at mean percent bindings of 4 1.1 C/c and 75.4%. respectively.

For testosterone. the inter-rissriy CV were 13.7% and 13-79 (1149). at mean percent bindings of 24.7% and 67.08. respectively. The in~li-i~~sriyCV were 9.3%

(n= 12) and 5.6% (n= 12). tit mean percent bindings of 23.1% and 60.8%.respectively.

The inter-assüy CV of the cortisol usay were I 1.1 C/c and 8.9% (n= 12). at rnem percent bindings of 52.6% and 69.8%. respectively. The intri-~uayCV were 2.1%

(n= 12) and 4.9% (n= 12). rit mean percent bindings of 50.( 1% and 73.0%. respectively. Hioh Performance Liauid Chromatm

Progesterone/Estrridiol

The resultrint P4 immunoreactive protiles of kcal esums subjected to HPLC from nonpregnant females during the folliculm and luteül phü-ses of the estrous cycle are illustnted in Fisures 5a md Sb. respectively. Usin2 this HPLC system. propesterone is known to elute within fraction 32 (Figure 8). There were severil unconjupted immunoreactive peaks mesured by this antiserum: however. the majority of immunoreactive metabolites were found to be les polar progesterone (fractions 33 to 35) in both the follicular and luteal phase. Seveml minor immunoresictivity peaks were obseived in both phases within fractions more polar than progesterone.

The resultünt E2 immunorenctive protiles of fecal samples subjected to HPLC from nonpregnnnt females dunng the follicular and luteal phases. are illustritted in Figures 6a ÿnd B (luteai phase) - Wolf 278 --+--.. Wolf 364

Fraction (minutes)

Figure 5. Immunoreactive profiles of fecd extracts from femde red wolves during the folliculÿr (A) and luteal (B) phases ÿfter fnctionation with HPLC md EIA of the frxtions obtained with proçesterone ÿntisenim (R486 1).

59 L 1 - Wolf 278 ---+--. WOI~364 Es tradiol

Fraction (minutes)

Figure 6. Immunorextive profiles of feecal extrricts from femde red wolves during the folliculûr (A) and luteal (B)phases &er fractionntion with HPLC and EIA of the frrlctions obtained with estrridiol antiserum (R4972). 6b. respectively. The resultant protiles reveriled a Irirge unconjugated immuncireative peik in 3 of the 4 protiles thüt hrid a retention time compwible to estriidid in the standard

(Fraction 15; Figure 8). In addition. minor immunoreactivity was detectld within frictions known to have ri sirnilu retention time estrone in the standard (Fraction 20: Figure 8).

The profile from wolf 278 (Figure 6b) suggested that no estridicd wuprc=.sent. This is not unexpected. as the tècd sample was obtaine J during the luteid p hue. however we have no explmation for the large immunoreüctive peak thüt wüs obszrved within hction 5.

In generii. with exception of relative rimcwii~s of iminunuresrctivity. the prngesterone and estmgen immunoreactive protiles obtained did not drrimatically diftèr between the follicular and luteril phases. Minor individual variation between protiles did exist. However. with the exception of the estmgen pi-otile obtriined during the luteal phase from one wolf. the patterns of immunoreactivity detected within protiles were similar between mimals. Cortisol The resulttint cortisol immunoreactive pro ti le of an e lelvated cortisol fecd sample subjected to HPLC from a kmde red wolf is illustrated in Figure 7ri. Frriction 6 was cornparabte to cortisol in the standard (Figure 8). Analysis of the HPLC fiacticms on the cortisol EIA illustrated ti large immunoreactive pek more polar than ciwtisol (tiiactiun 4-5) and a minor peak less polar than cortisol (fraction 9). Testosterone Two elevüted T concentration kcd extracts obtained from male wolves during the breeding season were subjected to HPLC. The resultins fractions anidyzed with the T EiA reveded severai immunoreactive peak'; (Figure 7b). Friction 17 wris comparable to testosterone in the standard (Figure 8). The lareest amuunts of immunoreüctivity were detected in peaks less polar than testosterone (between frictions 19 and 27). Two smdl unidentitied immunoreüctive peaks also were observai in frxtions more polar than testosterone- Fractions (minutes)

Figure 7. Immunoreactive profiles of fecd extracts hm(A) ri fernale red wolf after fractionation with HPLC and EIA of the frrlctions obtained with cortisol antiserum (R4972) and (B) male red wolves tifter fmctionation with HPLC and EIA of the fractions obtained with testosterone antiLwnim(R 156/7). Proctions (minutes)

Figure II. Chromatograph of 2 1 puritied steroid standards using reverse phase high performance liquid chromatogr-phy. Female Profiles Lon~itudinalFecal Protiles

Concentriitions of fecd P4 and E2 metabolites were analyzed for 19 cycles (n= 17 animais). Of the 1 9 cycles examined 3 were eliminated due to pooi- srimple collection and therefore could not be classitied. Figure 9 illustriites the loiigitudinril3 day merin composite fecd P4 and E2 excretory protiles for pregnünt (P: n=3). ovulatory nonpregnant (NP: n=9) and nonovulatory or acyclic rinimals (AC; n=3). Diu are aligned by a 15.w in P4 (day O).

The ovarian steroids and cortisol concenwitions during the repnxiuctive cycle of two fernale wolves. one which endured repeated restmint episodes (wdf 607) and one that did not (wolf 545) rippeür in Figure 10. Within Clases of fernales

Mean P4 and E2 values (Table 1) were compmd across cycle striges within each clriss (P. NP. and AC) bued on factorial ANOVA rinrilysis. Progestins remained rit low levels in ail fernales sampled until the periovulatory period. when a steep increw in P4 levels wa's obseived in the profiles of cycIing (P and NP) animals. Durin2 the luteal phase. mean P4 levels in P (range. 49.1-2961.0 ng/g kces: p10.003) and NP (1-ange. 52.9-

354 1 .O ng/g tèces: p10.006) mimals remained signitïcantly elevated above baseline levels (pre) until the mid- to late luteril phase. Mem P4 concenw~tionsreached maximum values iater in P tèmales (days 18-26) thm NP îëmales (driys 9- 17). Following day 54. mem P4 levels were not signiticantly different fi-orn b;1,seline levels in either P (p=0.347) or NP

(p=O. 190) animals. Mean P4 AC concentrations duiins the dit'ferent cycle stages did not signiticantly diffir from mean AC baseline vatues (p>O.MO ).

Mean E2 levels in P animais were signiticantly elevated above mean P baseline levels dunng the periovulritoiy period (p=0.034). days 27-35 (p=0.007) and days 36-44 (p= 0.041). Mean E2 NP levels were significantly elevated above mean NP briseline levels during the peiiovulütory period (p<0.00 1 ) and days 27-35 (p=O.O 1 4). Mem AC E2 Days from progestin n'se Figure 9. Mean longitudinal profiles of fecd progestin and estrogen membolite concentrations for the three femde reproductive cl^^: pregnant (A: n=3). ovulatory nonpregant (B: n=9) and acyclic (C;n=3) red wolves.

65 Month Figure 10. Longitudinal fecd prosestin. estrogen (AC) ruid cortisol (B,D) metabolite protiles from 2 femde red wolves. Wolf 607 (AB) experienced restaint episodes for serum sample collection to time for artifical insemination. Astensks indicrite restraint episodes and blood samples taken. Wolf 545 (CD) wris not restnined and becrime pregnant natudly. Nonovulatorv

Table I. Mean + SEM tècd progestin and estrogen metabolite çoncentmtions for dl cycle stages. pre-ovulatory (driys -30 to - l3), periovulatory (days - 12-8). luteal (days 9- 17. 18- 26.27-35. 36-44.45-53 and 54-65) and post (66-7 1) for pregnant. ovulritory nonpregnant and nonovulatory red wolves. concentrations tiom dl cycle stages did not signiticantly diffir hmmeiin AC badine values (~20.066). Between Clrisses of fernales Concentrations of P4 and E2 within each cycle stage were ccimpmd between the 3 classes (P. NP and AC) OF fernales by ANOVA with i-eperited merisures. During the pre- ovulatory period (days -30 to -13) P4 concentrations between P and NP wolves were relatively low rind did not diffa- (p=0.234. Figure 9). Prcigesteriine concen~itionsfrom

AC ;uiimÿls during this time period were elevated in coinparison tu P and NP wolves (p2O.O 17). Fecal progestin cuncentntions did not signiticantly diffei- betwsrn P and NP animals during my stage of the luteal phase (days 9-65: p10.270). Progesterone values remaineci elevated abve AC values longer in P mimals (days 364: p=0.035) than in NP rinimals (days 27-35: p=0.0 17). Insuffrcient samples precluded cornpuison of post- sampies. Fecal estrogen metabolite concentrations during the pre-ovulritciiy phase (days -30 to - 13) did not diftèr ktween NP and P animds (p=O.S24). However. during the pre- ovulatory phase. E2 concentrations in AC rinimals were signilicantly greater than E2 values in NP animals (p=0.034)but not P animals (pd1.088). During the periovulatory and luteal phase (days - 12-65) E2 concentmtions did not signiticantly diffei- among the three classes

(P. NP and AC) of femdes (~20.130). Similru' to P4. insuficient samples precluded cornparison of E2 post-samples. In addition to the rindysis of adult wolvcs. I yearling wolf. housed within a family unit (with mother. fathei' and 2 brothers) was examined foi- evidence of cyciicity.

Appendix III A(a) illustrates that the young wolf did experience a surge in E2 tbllowed by ri dramatic rise in P4. Concentrations of P4 and E2 were within ranges ~ibxrvedfor cycling adult femdes. Pericivulatow Protiles

For consistency. days - 16 tci - 12 were used as baseline values for cornpuison ktween daily means in both fecal and serurn protiles [driy - 16 corresponds to the euliest samples collected in the b;~selinetime fmme for .wiiim srimples (Febniru-y2nd-22nd)l. Fecal

In addition to the calculation ruid sutisticd analysis of 3 düy means. daily meün P4 and E2 concentrations were crilculated for 9 cycles (n=7 NP animrils) tci mure specifically evaluate endocrine trends c~'curingin the peiiuvulatury psi-iod (Figure I 1 ). Limited datri ti-urn P fernales did not aIlow for the cdculsition of daily msans duiing the periovulatory period. Progesterone metabolite values rosse stendily above mean basal values (days - 16 to

- 12: 142.2 t 15.6 ng/g tèçes: range 37.8 - 453.6 n@g tices) from day O (399.7 2 89.1 ng/g feces: range 222.2 - 1033.2 ng/g kces: p=O.O l O). reaching maximum daily mean concentrations by day 16 ( 1 157.8 + 348.0 n@g feces).

Basal E2 values (day - 16 to - 12) mnged from 25.4 - 153.0 ng/g fwes (mean 72.2 i

6.1 ng/g feces). Daily rnem E2 values demonsmted a gxdual increase from day - 12. achieving peak levels rit day -5 ( 187.20 + 5 1-89 ng/g feces). hllowed by a pdud deciine fr'rom days -4 to 3. However. rit no point were daily rnean E2 values si~niticantlyelevated above merin bual E2 values (~20.068). Of the 9 cycles esamined. 6 exhibited a single muked elevation in E2 [Dliy -7 (n=l), Driy -6 (n=3). Day -5 (n=l) md Day -4 (n=l)l with peak values riuiging hm169.9 - 1032.9 ng/g îèces. Semm

Daily mean concentrations of serum progestercme (P4s). estradio1 (EZs) and luteinizing hormone (LH) were calculated ruid anrilyzed foi- 23 cycles fmrn 12 NP animals.

Figure 12 illustrirites the endocrine relationship of P4s and E2s tci LH. Lack of data on specitic days (days -7 and - I ) wudue to infrequent simple collectioii and srnüll amounts of senim obtained. In al1 cases, priority wiiz; given tci P~srindysis. Days from Progestin rise

Figure 1 1. Daiiy mean +/-SEM (n=9 cycles) fecal progestin iind estrogen metabolite concentntions from ovulritory nonpregnruit kmale red wohes (n=7). Days from progesterone rise

Figure 12. Daily mean +/-SEM ( n=23 cycles) serurn progestzrone. ekdiol and luteinizing hckmone concentntions for ovulrrtory nonpregnant femaie red wolves (n= 12). The mean .semm profile demonstrrited that the1-e was minimal vruiation in mean

baseline PA concenti-aticins values (days - 16 to - 12: 0.7 @ml 2 O. 1 i-anse 0.2- 1.1

ng/ml). Despite the initial alignment of txch individual pi-ot'ile tc) ri i7.w in P4s [day O =

basal values (P4s vdues from al1 wolves Februriry 2nd to 22nd) + 2 SDI. statisticaily. the

tirs signiticrint increase in mem P4s concenmîtion from mean baseline concentrdtions

(driys - 16 to - 12) occurred at day -2 ( 1.3 F O. I ng/ml: p<( 1.05). which criir.esponded with the onset of the LH sui-ge. Serurn prcigestenme levels rose shrirply iintil day 5 (20.8 k 2.0 nghl). and ranged from X.9 - 33.6 ndml benveen days 6-l O. The greatest merin value measured was on day 9 (26.0 _t 4.6 ng/ml).

individual basal E2s vaiues (days - 16 to - 12) rmged ti-cim 10.0-20.1 &mi (mean

1 1.9 _+ 0.7 pg/ml). Estradio1 began an apparent n'.se on day - I 1. Daily mem E2s crincentrations continued to gradudly incrase and were signiticaitly gi-eriter by day -4

(pc0.05)when crimpmd to balvalues. Significant elevaticins of E3, above basal values were detected over an - 7 day intervül (day -4 to 2; p4).05) with ri perik mean vdue of 30.4 1. 4.8 pdml measured on driy -3. Possibly due to infrequent blood sampling. only 13 of the 33 cycles exmined exhibited ri marked elevation in E2s. Examination of thes 13 individual cycles reveded thrit E2s peaked on day -6 (n=2). day -5 (n=l). day -4 (II=?). driy -3 (n=3) and day -2 (n=5) with values rangincg from 17.9 - 52.1 pg/ml. Mem esucidinl values declined coincidentdly with the cinset of the LH surge. By driy 3. mean E2s concentrations returned tc, basal values and remained at nadir cuncentmtions through until driy 10 (p>0.05).

Baseline LHçnncentrritions (days - 16 to - 12) rangecl from 0.3-3.2 ngml (mean 0.9 i- 0.1 nghnl). The tiist significruit rise fiom bwline was observed on day -2 (p4.05) retlching rr perrk mean value of 1 1.0 * 5.5 &ml. Daily mean LH values i-emained above baseline until day 1 (P<0.05). From day 2 through day IO. driily mean LH levels remained

üt nadir levels (p>0.05). Of the 23 cycles examined. 15 exhibited a peak in LH with values ranging from 4.0 - 70.7 ng/ml. Individual peaks i~curredon day -4 (n=3). day -7

(n=6) and day O (n=6). Male Profile Concentriitiiins of fecal T were andyzed for 5 intact adult male red wcilves. Figure

13 illustmtes the weekly mean longitudinal fed T excretory protile iiver a 1 yeu period.

Between mid-spring and euly autumn. T ccincentrritioiis wrrt: at their I( iwest values (May - late October: mean 103.1 It 6.3 ng/g feues: range 3.6-507.7 ng/g feces). Levels increased above mean briselint: values (p=O.OO 1 ) as hylengths were JecIining Juring Iate auturnii

(November-December: weekly mem range 136.3 f 2 1.4 tii 259.7 + 48.5 ng/g Aces: individuai nnge 3 1.2 to 840.1 n& feces). During Autumn [September 2 1 st (Autumn

Equinox) to December 2 1st (Winter Solstice)l weekly T concenuitions and daylength were invérsely related (r= -0.749: p=0.006). From the winter .wlstice until mem peÿk T concentrations were reriched in late Februaiy (838.2 +- 743.8 ng/g fecès). mean weekly T concentntions and daylength were positively related (r= O.W 1 : pc0.00 1). Thereritier. T concentrations declined steadily to bual cimcentrations and from early Mnh to the summer solstice (June 2 1 st) T con~en~ticinsand daylength were inversely relüted (r= -

0.865. p

Fecal samples were collected (January - June. 1998) from one intact yearling

(February - Mmh was his tirs breeding .wuon). housed in a tmily unit (mother. hther. sister and brother). and were andyzed for T concentrations. Resuits for this animal dernonstrate the occurrence of .sea.sonality in T .secretion in a pattern similu tci adult animais

[Appendix [II. A( b)]. Figure 13. Weekly mean (+/- SEM) changes in fecd testosterone metabolite concentrations for male red wolves (n=5) in relation to changing daylength over a one year peroid. Gender Determination

ANOVA WLY perfrrmed tri determine which of 6 vaiirrbles (P4-E2. T. P4fT. PUE2 and TlE2) demonsmted signiticünt differences mung the classes (P. NP. AC and males) within the four different time periods (December 2 1 st - February 3rd. Febniary 4th - Mmh 20th. March 21st - May 5th. May 6th - lune 2 1 st). To test the ability of fec'd steroid andysis to distinguish between the sexes. rmly signiticant clifferences detected between the 3 classes of kmriles and males were of interest. Mean t SEM for each variable within erich period for P. NP. AC and male rrnimrrts are pre.sentc=d in Table Ii. Shaded uew repre-sent fernale variables thrit wei-e signiticantly different (p<0.005) hmthe respective rnüir variable. in dl four pericids. differences among tèrnüle uid mde vri~iableswere detected.

However only 3 of the 4 periods (from December 2 1 st - May 5th) demcinstrited ri variable that wusigniticantly different fi-om males for riIl 3 classes of fernales. The ritios P4lT and

TE2 were the two most informative vruiribles during these 3 time periods (December 21st to May 5th). Testosterone also was sui informative variable to chmcterize between the sexes: however. signiîicant differences from males for dl 3 clris.ses of femdes were only demonsuated from February 4th to May 5th (Table II). Amon2 the 6 variables. P4 and E2 concentrations dong with P4ES ratios were the lest inforinrttive to compare mde and femaie hormone levels. - Acyclic Nonpregnant mgüuit Male *SEM =SEM =SEM December 21st to February 3rd

February 4th to March 20th

March 21st to May 5th

May 6th to Junc 21st

Table II. Mean 2 SEh4 concentrations and ratios of fecal progestin (P4), estrogen (E2) and testosterone (T) metabolites for acyclic, nonpregant, pregnant and male rdwolves during 4 time periods surrounding and including the breeding season. Within rows, shaded areas represent significant differences (p-dI.005) between the variables for the female reproductive class(es) and males. DISCUSSION

Validation

One of the main objectives of this study wax to develop a iicm-invasive methrid to endocrinologicrilly monitor the reproductive status of both inale and female red woives. In

addition. the adaptation of a method tii non-invasively monitrir adrencice~~ticdactivity was investigated. Progestins and estri-tgens have been successfully me=ured foi- a number c-tf clirnivores in kcaI (felids: Shille et al.. 19') 1: Graham et al.. 1993: Bniwn et al.. 1994:

Czekrtla et al.. 1994: Brown et al.. 1995: Graham et al-, 19%: Brown et al.. 1996b, canids: Wzsser et al.. 1905: Hay. 1996: Monfort et al-. 1997: Gudeimuth et al.. 1998: Velloso et ai.. 1998) and unne samples (cünid: Batchelor et ai.. 1972. Monfort et al..

1997). Testosterone metabolites rilso have been quantitied in urine (Mid: Brown et al..

1996a) and kces (tèlid: Brown et al., 1996a. canid: Montiwt et al.. 1997: Velloso et al..

1998). as have corticosteroids (urine: tèlids: Clirlstead et al. 1992: Carlsterid et al.. 1993: canid: Jones et al, 1990. feces: felid: Graham & Brown. 19%. cmid: Creel et al. 1996 ).

In this study. kcd sampks were cho.sen as the substrate for rinzilysis based on the follciwing: 1) feces were readily avriilable and easily collected dui-ing the daily husbzindry routine. while urine was rapidly absorbed into the grciund or mSked by bushes and trees within the holding area. and 2) in previously studiecl czl~nivcire species. it has ken demonstrateci that feces is the primary route of excretion for the stei-oid metabolites cif interest in this study (progestins and estrogens: Shille et al.. 1984: Shille et al.. 1990:

Gross. 1992: Brown et al.. 1994: Monfort et al.. 1997. cortisol: Graham & Brown. 1996. testosterone: Brown et al.. 1996a: Velloso et al.. 1998).

Achievement of the tirst objective required the adaptation and validation of fwai extriction procedures for use with Our Iriboratory's enzyme imm~ino~stiys.Results obtained hmthe dose-response experiments indicated that ovluian hormone (method 1 ) and corticosteroid extractirin protcxols were etiective rit removing exogenous progestidestrogen and endogenous ccirtisul metabolites. i-espctively. from the téces of the fernde red wolf. Similady. the androgen hormone ext~~ctionprcitc~id (method 1) was efticient at rernovinp endopwus testostemne metabolites from tlie feces of the mde red wolf. Cornpaison between dittérent progestin and csuogeii extmction protcxols demonstrated that exogenous progestin and estrogen remi)val wu improved through the addition of aiuminum oxide (Method 1 vs. 2). The addition of aluminum oxide aided in the removal of background pigments. resulting in more coiisistent percent recoveries and exmiction efficiencies (Lucas et al.. I99i : Gixhrun et al.. 1995). In dl but one case. the removd of ovrriian fecd metabolires w* greater utilizing rnethods 1 (-90% methmol) than method 3 (-20% methancil). The.= results are not unexpected ris Palme et al. ( 1996) and

Young (1998) demonsmted that the reccivery of stsrciids wu irnproved by utilizing increlrsing amounts of methmol.

Tests of pmllzlism sugpsted thüt cxtrxted sten~idsfrom both males md femdes behave in an immunologically similar manner to stenrids iised iii the standard curve and are prescnt in quantities which are rneÿsurable by the assay systems. A more detded examination of unconjugated steroids present in fecal samples wu Fwilitÿted by HPLC analysis. Caution should be exercised when interpreting the data. ;I,Y only a few sllmples were subjected to HPLC analysis. However. results t'rom this study suggest that there is little variation behveen individuals or stage cif the repnductive cycle iegarding which prosestin and estrogen metabolites arz excreted into the feces. Exmination of the immunoreactive estrogen protile suggested thrit estradici1 and estinne are the major frirms tif metabolites present in the feces of the red wolf. Thesr: results are in agreement with previous HPLC brised st~idiesin which it habeen demcinsmted that estridiol (Brown et al.. 1994: Grhmet al.. 1995: Montort et al.. 1997: Velloso. 1998) and estmne (Brown et al,. 1994: Monfort et al.. 1997) constitute the major forms rjf estrogen metabolites excreted in carnivore feces. In contwst to estmgen. the majcinty of fecal progesterone metabolites in canids do not appear to be in rhe native form of progesterone (Monfort et al.. 1997: Most1

& Brunner. 1997: Velloso et al.. 1998). This apperirs to be tiue of the red wolf as well. Although progesterone appears to be pre.sent in the feces. the majority of immunorerictivity

WU detected in fractions tliat were les polar than progesterone. suggesting the presence of pregnruies. This result is mit unexpectecl. since recent piiblicatioiis in the domestic dog

(Most1 & Brunner. 1997) and African wilcl dog (Monfort et al.. 1997) identiQ 20-0x0- pregnanes as the major- form of progestin metabolites. Vie numeroiis minor peliks dso observed in red wolf feeces suggest the likelihocid of metabolkm of progesterime tti .several more polar prciductr;. possibly including I I[3-hydiu>xypi'ogestei-i)ne. 20a- dehydmxypro~esterone and 17a-hydroxyprugesterc)ne. C~illectively- this infwmrition suggests thrit an antibody with a broader rringe of cross-r-eactivities andor higher ci-oss- reactivity with pregniines wouId increase the efticiency of mcinitoiiiig ovririm activity in the red wolf.

[t also appears that the majority of coitisiil-like imrnunorextivity in ciunivore feces is not from the native steroid. but rather from other metabolite fiwms. The physiological relevance of cortisol-like metabolites in the kces of the At'ricrin wild dog (Cree1 et al..

1996) and domestic cat (Graham & Brown. l996) have ken demonstrated using ri coiticosterone RIA: however. neither cortisol nui- corticusteinne were pre.sent in their native fomi in the kces. Data from the present study provide similrir 1-esul~s.indicritin: thrit cortisol is not excreted in its native form in the feces of the red wolf. Although the cortisol-

Iike metabolites in the feces of the red wolf rict immunologicdly similar to a cuitisol standard. as evidenced hmpardlelism results. and repeated restrdint episodes appeued to increase corticosteroid metabolites in one wolf. Iùrther- investigation intci the validation cif this =ay system is required before it can be üpplied tci study 'stress' events in this species.

In one canid species. the African wild dog. HPLC analysis suggests that testosterone is pre.sent in the tèces (Montoit et al.. 1997). Sepamtion of radiolübelled rnetabolized testosterone in the rnaned wolf revealed that seveial testosterone merabolites exist. with free testostemne constituting only ri small percentage of the tutal metabcilites (Vellom et al.. 1998). ln contrast. although several ridic>lribelled fecal testosterone metaboli- were detected in the feces of the mide domestic ctit. iione were üwocirited with

ti-ee testostercine (Brown et al.. 1996a). The results in tliis prexnt study agree with the

~b~servationsmade by Brown et id. ( 1996a). as no free testosterone iippeared to be excreted in the teces of the red wotf. Although testcisterone was not observed. the presence of severai large immunoreactive periks less polar thm testosterone suggests the possiblity of

5a androstanediol. I7~-hydroxy-5a-andr~)strin-3oneand 3~-hydroxy-5a-and1-c>s~n-17one (strrnduds not shown) crcissreacting with the testosterone aiiti body u.wd in this study.

Female Reproductive Cycle

Another goal of this project wuto establish ri dafilbase of normal endocrine protiIes for the fernale red wolf. The tindings of this study indicate that the major hormonal events which occur during the reproductive cycle of the red wolf are retlected in fecd stemid concentrations. Patterns observed are compambie to those reponed tiw .severai canid species. including semm steroid protiles in the domestic dog (Smith & MacDonald. 1974: Concannon et al.. 1975: Concrinnon et al.. 19773: Wildt et al.. 1978; Wildt et al.. 1979:

Olson et al.. 1982; Concannon et al.. 1989) coyote (Stelltlu~et al .. 198 I ) and the grriy wolf (Sed et al., 1979: Sea1 et al.. 1987). as well as fical steruid piutiles in the maned wolf (Was.ser et ai.. 1995: Velloso et ai.. 1998). Afiican wild dog (Monfort et al.. 1997) and domestic dog (Gudeimuth et al.. 1998).

The meUurement of reproductive hormones in the kces tif the red wolf has zl prieticai application foi- detecting the pre-sence tir absence of ovulation. In conrit to pregnant and nonpregnant animais. progesterone md estrogen concentrations did not tluctuate throughout the breeding setison in acyclic red wtdves. In Pict. concentrations were elevated above cyclic animais prior to the imset of estrus. In domestic dogs the administration of ri progestin treatment as a means of cisntraception. maintains gonadouopin secretion in an anestrual state. This prevents the elevation of gonadotropin secretion which would normally terminate anestrus ÿnd initiate pniestrus (Concannon. 1995). Althouph several potential explanations exia (age. health. grnetic variability. stress) it is possible that the elevüted progestenine levels present in acyclic red wolvcs prior to the breeding season acted in a similar manner. thei-eby maintaininp the wcilves in an anestrual strite.

One objective for the femüle red wolf longitudinal study was to examine differences that wcurred betweeii the endoaine prcitiles of pregnrinl ruid non-pregnant anirnrils.

Gestation Içngth in the red wolf. as determined from a ri.= in progestcrcine. wu 64-65 days in duration. This is cumpamble to gestation lengths rzported in the gray wolf (60-65: Seal et al., 1979). maned wolf (65 dliys: Velloso et al.. IWX) and domestic dog (64-66 days: Concmnon et al.. 1989). The detection of pregnmcy-specific differences post- implantation (-day 20 in the domestic dog: Thatcher et al.. 1994) in .semm md plasma ovarian hormone concentrations is controversial. Piasinri tir serum concentrations of progesterone (Smith & iMacDondd. 1974) and estii>gen (Concannon et al.. 1977b) have ken reponed to be gi-eater in pregnant mimals than in the nonpregnant luted phÿsc of domestic dog. However. rnost studies have not ohseived these pregnancy-specific differences in either progesterone (Hadely. 1975: Nett et al.. 1975: Austad et al.. 1976:

Reimers et al.. 1978: Concannon et al.. 1989: Onclin Br Verstegen. 1997) or estrogen

(Edqvist et al.. 1975: Hadely. 1975: Nett et al.. 1975: A~istadet al.. 1976: Graf. 1978:

Reimers et al.. 1978). In generil. pregnmcy-specitic inc'eri.ses inüy not be cleuly obscived in plüsmü and serum steroid concentrations because of the increÿszd hemodilution. metribolism ruid clearance wocirited with pregnancy (Concannon et al..

1977b). However. it is possible that differences in wu-isin stci-ciid hormone production during pregnûncy could be evident in fecal siimples. In recent repoits for the domestic dos

(Gudermuth et al.. 199s) and mrtned wrilf (Velloso et ai.. 1998). pregnüncy-specitlc differences in progestin and estrogen metabolites were ob.wived in the feces. In the present study. dthough fecal progestin and estrogen metabolite teveis rernained elevüted slightly longer in pregnant thun non-pregnant animais. a dmmatic difierence between the concenmtions of fecal ovarian steroids in pregnant and non-pr-egnrint animrils wüs not

Jetected. Perhaps a Iuger pup of pregnant study rinimals siiid/or rin riiitibridy more efficient at monitoring changes in progestin concentriticins wc~.~lJallow foi- the detection of pregnancy-specitk endocrine changes in the red wolf Although not cibserved in this study. pregnancy-specitk ciifferences in ovarian steroid concentnitions follciwing implantation are believed tcj be of luteal iiiigin as thm is no evidence tif placental sterciici .secretion in the domestic dot (Ccincrinnon. 1 %6b). It dso hris been suggested thtlt, although the stimulus ti)r increri.sed sterciid pri~duction by the coipus luteum is iinkniwn. prolactin is likely involved (Giiderrnuth et al.. 1998). PI-olrictin is a required luteotrophin in the dot (Concrinnon et al.. 1987). and it hris ken dernonstrüted that kllowinp implantation there is ri pregnancy-specitic. elevtition in srum prolactin between days 35-30 (Gnif et al.. 1978: DeCoster et al.. 1983: Onclin &

Verstegen. 1997). Gudemuth et al. ( 1998) demonstmted that wheii piulactin lises in serum. pregnancy-specitic elevations in fed ovririan steroicis accur. Perhrips. pregnmcy diagnosis could be achieved through the validation of ri sensi tive irnmunoüssay (RIA or

EIA) to non-invsiveiy monitor urinary prolactin. similrii- to previous success with LH

(Czeküla et al.. 1988: Biannirin et al.. 1989: Robeck et id.. 1993: Jefti-iiate & Engiand.

1097) and FSH (Walker et al.. 1988). Periovulatow Pintlles

A tind objective {if the study concerning female red wolves wris to Jescribe the endocrine relationships behveen ovuian steroids during the pciiovulatory period.

Characterizaion of the periovulatory pericid through fecd and semm sterc~id hormone analysis indicated that the red wolf is very sirnilu tii the dcunes~icdog (Concannon et al..

1075: Concannon et al.. 1977a: Wildt et al.. 1979) and griy wolf (Seal et al.. 1979). The mean semm protile demcinstrrited that hormone values in the red wolf were within similx ringes reported for the gray wolf (Seal et al., 1979) luid domestic dog (Edqvist et al.. 1975: Austrid et ai., 1976: Concannon et al.- 1975: Concrinnon et al.. 1977~WiIdt et al.. 1979: Olson et al.. 1982). The proestrous surge in estixdiril begrin pnor to the surse in LH ;ind wcurred »ver a -7 day period üttininp pz& values 3 days pnor to a rise in progesterone. Mean esvidiol values declincd coincidentelly with the onset of the LH surge and then slowly retumed to basal values. Similu trends have been repuned for the gray wilf (Seül et al.. 1979) and domestic dog (Concannon et al.. 1975: Nett et al.. 1975: Austad et al.. 1976: Wildt et al,. 1979: Olson et al.. 1982). In addition. the preovulatoiy LH surge wÿs similu- in magnitude and duration (-4 days) to othei- canid species (Setil et al.. 1979: Concannon et ;IL. 1989) and t~ccurredon tliz .urne day ;LS the tirst significant increase iii mean progesterune cunczntrations. As blood sümples could not be drawn on a daily bais. no conclusive evidence could he diawn regÿrding the range in which LH and esmdiol surges mculred in relation tii ri risse in serum progesterone.

However. based on the data obtained. LH peüks generaily occurred within 0-2 days pnor tu a rise in progesteronr and pe& esuüdiol levels were observed 2-6 days before a rise in progesterone.

Examination cif the mem kcal pmtile for i-ed wolves during the penovulatory period of the nonpregnmt animal revealed similar endwiinz trends to thüt of the semm protile. Peak estrogen cimcentmtions were ob.served 5 days (range 4-71 before a lise in progesterone and althoug h values were not signi t'icantly e levated abuve bü,sal concentritions. the estrogen surge appewed to lut foi- -8 days. Comp~edto .semm samples. fecal pmgestercme and estrogen concentritions were - 140 to IO fold Iiigher during basal concentrations. respectively. md -40 to IO fold higher at peak levels of stei-oid concentrations. i-espectively. This is not unexpected. since feu1 concentrations tend to be much more concentrated than sterciid levels in the blocid (Hriy. 1996: Most1 &

Bninner. 1997). Behavicwrril observations. althnugh iiot i-eported. indicated that overt estrous behaviours. including tail deflection. presentation of am-genital ma md mounting. were generdly obsewed during falling estrogen levels and rising progester-one values. similai- to tliat of the domestic dog (Concriniion et al., 1 979b: Beach et al.. 1 982:

Feldmim & Nefson. 1987). Results from this study suggest that because .sei-um and fecal ovriririn steroid

protiles in the red wolf exhibit sirnila- trends and are similar to reproductive patterns ob-served in the domestic hg.the mexwrement of civarian Iiriimones in the feces of the red wolf hris a pricticül application for detecting the approxirnnte timing of ovulation. Although

mean tècal concentrritions inci-ex?& -2- 1 O hld between basal and peak Izvels cif sterciid cuncenti-atiuns. elevations in fecal sterckis concentritions during the estrogen surge and the ri.w in progesterone were only clewly observed in compaiistin ti) concentrations during the early stages of the breeding season in a given animal. The.= results are similar to ri recent report in the domestic Jug (Gudermuth et al.. 1998). Therefwe. the application cif kal steroid monitoring to detect ovulation must be applied on an individual buis. requiring the driily collection of fecal samples severil weeks prior tu the anticipatecl time of estrus te) xcurritely estimate ovulation.

When ushg feces ris an alternative to .semm for monitoring ski-oid concentrations. a tirne lag between circulriting steroid levels and conceiitixtirms meaurd in the feces is expected. In the domestic dog. fecd sterciid concentriticms have been demcinswrlted to be positively coi-related within a -24 hour period with seium concentrritions (Hay. 1996:

Gudermuth et al., 1998). Although semm and fecal concentrations were not collected simultaneously in this study. we would expect to .we similai- trends in the feces of the red wolf. AIthough one must be cüreful in assumint chat the time it takes for- food items tu pus through the gut clo.sely retlects the 1ag ktween steroid ,secretion in blood and its excretion rate in the feces. the transit time of small piecss of surveyors tape or food colouring within a meatbrill occurreci within an - 12- 16 Iiour period in the red wolf (S. Berhns. personal communication). Based on c~b.servatioiis in the red wolf and ttiose described for the domestic dog. it is suggested that stemid crmcentriitions within i-ed wolf kces most likely repipsentzd a pend of circulating sten~idconcentiations hmthe previous

12-24 hciurs.

Studies in the domestic dog have suggested that the peik îkrtility perioci is estimated to be 4-5 days followin~the LH surge (Ceincannon et al.. 1989) based on ovulation ciccumng within 1-3 days ti~flowingthe LH surge (Phernister et al.. 1973: Concanncin et al.. 1977a: Wildt et al.. 1978) and 2-3 days for oocyte maturation (Holst & Phemister.

197 1 : Anderson & Simpson. 1973; Phemister et al-- 1973: Concannon et al.. 1989).

Studies have demunstrited thrit declining estrogen ci~nczntrations concornitrint with incseasing progesterone concentrtitions are ~sscxiatedwitfi the LH surse in the domestic bitch (Smith & MacDonald. 1974: Concruinon et al.. 1975. Crincannon et al., 1977a:

Concanncin et al.. 1979b: Wildt et al.. 1979: Ol.wn et al.. 1982) and thüt the initial sharp eIevation in progesterone concenmtions rxcurs coincidentütly with the LH surge

(Concannon et al.. 1977a: Penton et al.. 199 1). Therefiire. timing for Al in the domestic dog is routinely based on citculating pmgesterone coiicentrritions (Linde-Foi-sberg &

Forsberg. 1993: Wilson. 1993: Fontbonne & Badinand: 1993). Previcius attempts rit AI for the red wolf required münual restriiint of femüles fi^ bliicid simple collection and vaginal cytology. Resulw t'rom the present study indicate that non-invasive fecal stsroid monitoring couid be used as an alternative to blorid sampliiig to estimate ovulation and time

AI. Blood .sampling tivm ri stress-susceptible non-domestic animal sucli as the red wolf bas the potential ti? negatively impact on Al succe.s. Stress inliy intei~uptnormal endixrine patterns. and possibly disrupt fertilization. implrintatioii or cmbryonic development

(Moberg. 1985: Liptrrip. 1993: deCmrinzrtm & MacNiven. 1995). The application of monitoring civarian stenicl concentrations thmugh fecal srimples hüs the potentid to improve AI success rite and possibly aid in the genetic ancl c1in.w-vaticln management of this endangered species. Male Reproductive Cycle

For the male red wolf. one objective of this study wris to investigrtte the relütianship behvezn changes in fecd testostercme levels md photopznod. Previous work in male non- domestic canids has demonstrated events consistent with pliotopetiod synchronization

(Green et al.. 1984: Smith et al.. i 985; Asa et al.. 1987: Mitsiizuka. I9S7: Sed et ai. 1987:

Monfort et al.. 1997: Vellosci et al., I 9%). Similai- to otliei- non-domestic canids. the irsults of diis study demonstrated changes in fecal testoster4 ]ne metaboli te concrnv'rltiims in relation tc) photopzroid and funher support .seasonal rsgulation tif nndrogznesis and sperrnritogenesis in the male red wolf.

It ha ken suggested that the gny wolf cmnot bz cleÿrly clÿssitied into either o spnng (long-day) or hll (short-day) breeder (Asa et al.. Ic)X7). This appears tu be true of the red wolf as well. As cib-served in this study. the petik br'eeding periods in red wcilves occurs dunng mid- to lae winter as days are lengthening. However. fecd testosterone levels began to rise in Iate üutumn. rexhing pe& Ievels coincident with estrus in late winter. Thus. although peak breeding occurs dur in^ lengtlieninp days. the suly phases of

.sexuai recrudescence oçcur when days arc: becoming shc~i-ter.As with the gray wolf. it is not known which of the-se photoperoidic phases is more dominant or important. or if both have equally signiticant niles in the mnual cycle (Asa et al.. 1987). In addition to the meauriement of fecd steroid ccmcenti-ations tirim adult wolves. fecal smples alsci were collected from two wolves entering their tirst breeding scÿson ( 1 male and 1 female). The wolves were housed in ri family unit. ciinsisting of the mother. hther and 2 yciung males and 1 young female. Their steroid pmtilzs indicated that the young mimals did in façt exhibit semonal changes in reproductive h«rm«nes md the concentrations observed were within ranges reported for ndult wolves. These endocrine data support the observation that young wolves crin prodiiçe pups in theii first breeding season (Medjo & Mech. 1976: Zimen. 1976; Waddell. ILN5). Management and Conservation Applications

Overall. this piliject demonstrates the ability tci non-invasively rnonitor civaiiaii md testicuiar endocrine tùnction in the tèmale ancl male red wolf. i-espectively. The knowledge gained from this study greatly improves our understanding of the basic endocrinology of the endangered red wdf and hüs Jirect application towartls the management of both captive and wild populations.

The ktthat îècal samples are readily available. easily collectecl mJ retlect changes which wcur durin2 the reproductive cycle in the kmolr lias impciiunt implications foi. the captive breeding program. Fecal samples can provide ciuci;tl information on peak fe~tility periods and could therefore be used to time AI. The establishment of endwrïne noms for male and fernale red wolves «ver a long-tenn pzriod dlows for itcognition of any disturbances in these patterns in future analyses of individual reproductive function.

Therefnre. fecd steroid anülysis mny aid in the diqnosis tif possible cases of infertility or test the effèctiveness of diftèrent contrxeptivc treütment... This kind of information is important with regard to management deçisions suçh as futiii-e docations andor pairing of animais. In addition. the non-invasive nature of this technique eiiminates any undue

'stress' that animrils wouid endure if bIood srunples wers 1-equired tci achieve the sme objectives and therehre may more accuritely retlect ovaiiaii hinction.

For ti-ee-riinging red wolves. information gainerf fI-oin this study could serve as an indicator of normal reproductive patterns. Additionally. results from the present study suggest that gender deteimination thrcx~ghfecal steroid analysis cm be achieved during the breeding season hmlate Decernber until the beginning of May and wuld possibly be used in certain aspects of wild population studies such as ses ixtio and sample identification within a temtory. Howevei: due to individual variation a Iürger databasz is needed befixe this technique could be u.ed in prxtical studies. Further validation of the cortisol asay is requked before adrend endocrine function can be non-invasively moiiitored in the tèces of the red wolf. However. the potentiril application of this technique cciuld be used to detemine if pcitential 'stressors'. such as daily husbandiy i*outines t>i- other cnvinminentlil factoi-S. iiegatively iinp'xt on the well- being of the wolves and possibly intertèi-e with reproductive success. SUMMARY AND CONCtUSlON

In light of recent concerns of hybridization of wild red wolvrs and coyotes. the importance of and relimce on a geneticülly Iiealthy and viable captive population is ciucial for the future management of this endangered carnivore. This investigation wuundertaken primarily as ri mesuis to imprcive our basic knowledge of rd wolf reproductive biology.

The fundamental biol~igicziidatri accumulatecl from this stiidy will pnwe to be rui efiective aid in retining and impiuving the ciptive bi-ezding münrigement of tliis species. Althciugh rzd wolves have beeii müintained in captive popdations foc- alrnost 20 years tliis is the tint report to examine femalr and male endocrinology. While it is gener~llyusumed that reproductive aspects of the red wolf would mimic tho.se of the domestic dog. it is necessruy to document al1 aspects of reproductive biology piior tii uiidei-ttiking assisted reprduction. Studies have demonstrated that the reprtxiuctive biology tif a non-domestic species cm differ from its domestic mode1 [sperm chwçteristics cliffer between 1-ed wolves and domestic dogs (Goodrowé et al.. 1998). çlouded ieopards (Neofelis nebulcis~.dthough 3-

8 times the body weight. require the sme dosage of gonadotropin the domestic cat

(Felis domesticus) to elicit comparable fdlicular growth (Howard et al., I992)1.

To facilitate the non-invasive monitoring of civaiiaii. testiculai- aiid ad[-enworticiiid steroids within the kces of the red wolf. validation of enzyme irnmunriusay systems used in our lubomtory. through do.= response and parallelism e~periments. was required.

Results from the.= experiments demonstrated thüt extraction prcitocols selected for use in this study were effective rit removing progestin. estropm. testostercine md ccirtisol metabolites ti-om fecal samples in this species. Fdcilitated by HPLC. a more detüiled examination of unconjupated steroids present in fecal samples i-evealed thrit it is best to use antiserum specitic: for estrüdicil to detect folliculür activity. but for monitoring products of luteal and testicular steroidogenesis. anti.senim with ri broader spectnim of cross-reactivities would be more hvourable. Although cortisol-like metabolites in the feces of the red wolf behaved immunologicülly similar to a cortisol standard and i-epeated restraint cpisodes appeared to increiisz wnisol metabolites in one wolf. HPLC results indicated cortisol is not present in its native fom but mther other cortisol-like metabcilites. Further investigation into the validation of this assriy system is required behrz it cm be applied to study 'stress' events in this species.

Longitudind monitoring of ovünün stemids wucoiiducted (in female ird wolf fecal samples collected during the breeding .season. In general. the major hormonal events which xcurred during the adul t reproductive cycle werr rrtlectsd in steroid cimcentmtions in the &es. Similar repniductive patterns dso were observed in ;i single female red wolf experiencing her tint breeding season. Maestmgen metabolite values in pregant ÿnd ovulatory non-pregnant femcile red wolves peaked prioi- to a rise in progestins and were signitïcantly elevated above meui baseline concentmtions ditring the mid-luteal phase. Mem progesterone metabolite values were signiticantly higlier than bsline concentritions until the mid- tci late luted phrise in pregnant and ovulatciry non-pregnünt wolves.

Although mean fecd ppropstin and estrogen metabolite concentmtions in pregnÿnt females remriined elevated longer during the luteül phase than thiise merisurecl in avulatory non- pregnant females no significant difference between the tivo pmfiles. foi- either progestei-one or estrosen. was demonstrated. Therefore. pregnancy detzction using the techniques outlined in this study is mit possible. However. merisurement of reprciductive hormones in the kces is useful for the deteetion of normal reproductive patterns. by evaluriting the presence or absence of ovulation through a signifierint ii.~in progesterone. In some females. fecd progestin and estmgen concentmtions remaineci rtt baseline concentrations throughout the breeding .wucm and were therefore considetrd to be xyclic. The tmditional method of timing Al in the femde i-ed wolf involved physiçd restra.int for vaginal cytology and blood sampling for progcsterone analysis. Femdes were generally subjected tci AI procedures 2-3 times (beginning 5-6 days after progesterone levels became elevated above 1 .O @ml) and were restmined on uverige 1 I times before the suspected pend of ovulation could be deteimined. Althoilgh senim and fecal values were not simultruieously monittired in this study. the .semm and ldcal profiles obtained during the periovulatory period exhibited similar trends and were similür to tho.se iibserved in the dornestic dog. This suggess that the non-invasive measuremeiit of r~variansteroids in the feces could serve as a pi-actical application for the deteciion of the appn>ximate tirne of ovutation. Unfoitunütely. due to individual variation. thei-2 wris no detined ccincent~xtion that coincided with ovulation. Theretbre, the u.se of non-iiivasive hoi-n~tmtilmonitoring to dztect a rise in pwgestin concentrations hmb;~seline values must be applied cm an individual buis. Perhüps the development of a more nppnip~-iateantiserum to progestin metabolites excreted in the tèces of the fernale red wolf would minimize individual variation ~llowingfor a broder application and possibiy pregnancy dctection.

Longitudinal monitoring of testiculÿr steroids wuconducted on male red wolf fecal samples collected over a I year period. The changes in Iioimond patterns dernonstmted events consistent with photoperiod synchronization. fuithri- supporting .seasonal regulation of andrognesis and spermatogenesis in the red wolf. Testicular steroid merisurementu, in a single male red wolf cxpzriencing its tkt breeding sewm éxhibited similu patterns to those ob.served iidult males. Adult fecal testosterone concentixticins i-emained at baseLine levels thrciughout the surnmsr months until late üutumn (Novernber-Decembsr) when levels significmtly mse above büseline values and were negütively correlüted with shortmina daylengths. From the wintzr solstice (December 21st) until peak concenti-ations were attai ned duri ng a pend coincident with estrus (late February ) testcistenme metabolites were positively correlated with increasing daylength. Thereafter. testosterone metabolite concentrations declined brick to nadir levels. Experiments from this study aiso demonstrate the ability to sex red wolves through fecal steroid analysis. Although pnctiçal application of this technology is yet tu be implernented. results from this study indicate thnt gender determination from unknown fecal samples is possible because of the differences that exist betwzzn kmale and male fecd progestin. estrosen and testosterone metabolite concentrations. Unfortunately. this technology can only be iitilized during the breeding seasciil (Irite Decemki- - early May) ris concentrations of steroids beyond this peiiod do not crmtwst suftlciently to detect diftèrences between the .sexes. In summriry. this stiidy demonstrates the capzibility tc, non-inwsively incmitcir ovarian and testiculru- function in fernale and male red wolves. i-espectiveIy. through ka1 steroid analysis. It ha gnxtly enhmced our basic knowkdge c)f the Iiormonal zvents which vccur during the repr~.uctivecycle ut' this spwies. Althougli the techniques described in this study are adequilte t'or prricticril applicatioii. f~itureinvestigations should focus on the develcipment of mays more specific tri metabolites exc~tedin the feces of this speçies. In addition. niminvasive fecal steroid analysis should be u.sed to time future AI attempts. thereby eliminslting stressful restiaint episodes and possibly improving AI success rate. Irnproving the management of the captive breeding program by utilizing techniques developed in this study bas the potentiril to greritly improve the reprc~ductionof genetically vdurible individuals for reintroduction and ensui-e the continued existence of red wolves in the wiid. LITERATURE CITED

Abbott DH, McNeilly AS, Lunn SF, Hulme MJ, Burden FJ. Inhibition of ovarian function in subordinate female mannoset monkeys (Callithri jacchus). J Reprod Fertil 1963; 63: 335-345.

Adam CL. Atkiiison T. Effect of feeding inelatonin to ixd deei. (Cei-vusela~hus) cm the cin.set of the breeding .sewon. J Reprci Feitil 1984: 72: 463466. Alesander G. Signoret JP. Hafez ESE. Sexual and rnatci+ii:rl behavioiii-. In: Hüfez (rd. Reproduction in Faim Animals. Philadelphia. Lea R; Febigsi-. 107-1: 222-254.

Aman RP. Reproductive physiology aiid cndociiniilogy ot-the dog- 111: Motrow DA id.) Current Thempy in Theringennlngy 2. Philadelphia: WB Saunders Co.. LWh: 532- 538.

Anderson AC. Simpson ME. The ovriry and reproductive cycle of the hg(Beagle). Los Altos. Californiri. Geron-X Press. 1973.

Anderson AC. Reproduction. In: Anderson AC (rd.) The Beagle üs an Expen mental Dog. Ames, IA. Iowa State University Press. 1970: 3 1-39.

Anderson D. The rrite of passage of the mmrndian ovum through vürious portions of the tàilopian tube. Amer J Physiol 1927: 82(3): 557-569.

Anderson K. Fenility of frozen dot semen. Acta Vet Scünd 1971: 13: 128- 130.

Anderson K. Intrauterine inseminlition with frozen .semeil in dogs. Proc i2th Nordic Vet Cong. keland 1974: 153- 154.

Arbeiter K. Bras W. Ballsibio R. JtichIe W. Estms induction and fertility with cabergoline. a prolric'tin inhibitor. in the bitch. J Sm Anim Prac 198%:29: 78 1-788.

Asa CS, Seal US. Letellier M. Plotka ED. Peterson EK. Pinealectcimy or superior cervical soanglionectomy do not alter repciduction in the wiilf (Crinis lu~iis).Biol Reprcid 1987; 37: 14-2 1.

Atkins DL, Dillon LS. Evolution of the cerebeilum in the senus Canis. J Mammol 1971; 52: 96-1 07.

Atkinson S. Ciilmutin WG. Seasonal testosterone pattern in Hawaiian monk secrls (Monrichus .schauinsl;rndi). J Reprod Fertil 1992: 96: 35-39.

Audubon JJ, Bachman J. The quadrupeds of North America. New York, 185 1, Vol 2 334p.

Audy MC. Bonnin M. Souloumiac J. Ribes C. Kerdelhue B. Mondain-Monval M. Scholler R. Canivenc R. Seasonal variations in plüsma luteinizing hormone and testosterone levels in the Eutoperin badger Meles meles L. Gen Comp Endrwiinol 1985; 57: 445-453. Austad R. Lunde A. Sjaastad OV. Penpherd plasma levels of oestrridiol- l7p and progesterone in the bitch during the oestrous cycle. in normal pregnancy and afwr dexamethasone treatrnent. J Reprod Fenil 1976: 46: 129- 136.

Austin CR. Short RV. Genn cells and Fenilization. In: Reproduction in Mümmals. London: Cambridse University Press. 1972: 82.

Bailey V. Biological survey of Texas. N American Fauna No 35, 1905. 333p.

Bangs O. The land n~aninialsof peninsular Florida and the coast region of Georgia. Proc Boston Soc Nat Hist 1 898; 28; 1 57-23 5.

Barb CR, Kraeling RR, Rampacek GB, Fonda ES, Kiser TE. Inhibition of ovulation and LH secretion in the gilt afier treatment witli ACTH or iiydrocortisone. J Reprod Fertil 1982; 64:S5-92.

Bartness TJ. Golrtman BD. Marnmalirui pineal melatonin. Esperienu lOS9: 45: 939-945.

Bartram W. Traveis. Philadelphia, 1791; 522p.

Batchelor A. Bell ET. Christie W. Uninary oesuogen excretion in the beagle bitch. Br Vet J 1972: 128: 560-565 Battista M. Parks J. Concünnon PW. Canine sperm post-thaw survival following freezins in su~wsor pellets using PIPES. lactose tris. or TEST extenders. Proc 1 1th [nt Congr Anim Reprod & AI. Dublin 1988: 3: 229-23 1. Beach FA. Dunbar IF. Buehlar MG. Sexual chmcteiistics of female dugs during successive phases of the ovaririn cycle. Hom Behav 1982: 16: 4 14-442.

Bekoff M. Social behaviour and ecology of the Afiican Canidae: a review. In: &ri MW (ed.) The Wild Crinids. New York: Van Nostrand Reinhold. 1975: 120- 142.

Ben Sadd M. Bayle JD. Seasonal changes in pliusma testosterme. tliyn~xine.md curtisuI levels in wild rsibbits (Oryctol~uscuniculus al~irus).Gen Comp Endcicrino1 1985: 57: 3x3-388.

Bondestam S. Alitrilo 1. Karkkainen M. Redtime ultr~souiidpi-egnrincy diagnosis in the bitch. J Sm Anim Pmt 1983: 24: 145- 15 1.

Bouchard GF. Gros S. Gmjsin VK, Youngquist RS. Concrinnon PW. Krsiuse GF. Reddy CS. Oestrus induction in the bitch with the synthetic oestrogen diethylstibestrol. J Reprod Fertil Suppl 1993: 47: 5 15-516.

Bowman LA, Dilley SR, Keverne EB. Suppression of oestroçen-induced LH surges by social subordination in talapoin monkeys. Nature, London 1978; 275: 56-58.

Braden AWH, Moule GR. Effects of stress on ovarian morphology and oestrus cycles in ewes. Aust J Agric Res 1 964;1 5 : 93 7-949. Bridy CA, Ditton MK. Management and breeding of mancd wolves at the National Zoologictil Park. Washington. Int Zoo Yerirbk 1979: 19: 17 1 - 176. Brannian JD. Griftin F. Terixnovü PT. Urinay androstenedione and luteinizing hormone conçentrritions during musth in a mature Ati-ican elephant. Zoo Biol 1985); 8: 165- 170.

Brown JL. Terio KA. Gi-aham LH. Fecal andr-tlgeii rnetabcdite analysis tiw non invasive monitoring of testicul:ir- steroidogmic activity in felids. Zoo Biol 1996~15: 425- 434.

Brown JL. Wris.wr SK. WiIdt DE. Graham LH. Coinpai-ritivc aspects of stzroid hoi-moiie metabolisin an J iivariaii activity in fel ids. rnetisrirt.rl ilciri-iriv:~l;iveIyin feces. Bit )l Reprod 1004; 5 1 : 776-7x6.

Brown JL. Citino SB. Birsli M. Lehnhadt J. Phillips LG. Cyclic patterns of luteinizing hormone. follicle stimulriting hormone. insulin. and progesterme secretion in the asian elephmt (Ele~k rn-& J Zoo Wild Med 199 1: 22: 49-57. Brown .IL. Wildt DE, Graham LH. Byers AP. Collin SL. Bai-rett S. Howard JG. Naturd versus churionic gonridotrophin-induced ovarian responses in the clouded leopürd (Neofdis nebuIosa) rissessed by fecal steroid analysis. Biol Reprod 1995: 53: 93- 102. Brown JL. Wildt DE. Wielebnowski N. Goodrowe KL. Graham LH. Wells S. Howard JG. Reproductive açtivity in captive female cheetrihs (Acinonvx jubaw assessed by Frlecd steroids. J Reprod Fertil 1996b: 106: 337-346. Bubenik GA. Schrtms D. Lerttherfand JF. Se~sonalrhythms of prolactin and iw role in the üntler cycle of the white-tded deer. In: Fennessy PF. Drew KR (eds.) Biology of deer production. R Soc NZ BuII 1985: 22: 257-262.

Burdick HO, Whitney R. Acceleration of the rate of passage of îèrtilized ova through the fallopian tubes of mice by massive injections of an estrogenic substance. Endocrinology 1937; 2 1 : 637-634.

Busch RH. The Wolf Almanic. New York: Lyons & Burt;~-d,19%.

Carley CJ. Activites and findings of the Red Wolf Field Recovery Program from late 1973 to July 1, 1975. USFWS Report, Albuquerque, NM, 1975, 2 15p.

Carley CJ. Status summary: The red wolfcanis nifus. USFWS Endangered Species Report No 7, Albuquerque, NM, 1979, 36p. Carlstead K, Brown JL. Swrlwn W. Behaviourd and physiological correlates of stress in Iaborritory cab. Appl Anim Behav Sci 1993: 38: 143- 158. Culstead K. Brown JL. Monfort SL. KiIlens RK. Wildt DE. Urinary monitoring of adrenrrl respon-ses to phychological stressors in domestic and nondomestic klids. Zoo Biol 1992: 1 1 : 165- 176. Chakrriborty P. Reproductive hormone concentrations dui-inp estcus. pi-egnruicy. and pseudopregnancy in the labrador bitch. Theriogenc)logy 1987: 27: 827-840.

Chang MC, Yanagimachi R. Effect of estrogens and other compounds as oral anti fertility agents on the devetopment of rabbit ova and Iianister embryo. Fertil Steril 1965; 16: 281-291.

Christian JJ, LeMunyan CD. Adverse et'fécts of crowdins on laction and reproduction of mice and two jenerations of tlieir progeny. Endocrinology I%S; 63: 5 17-539.

Cliristirinsen IJ. Repi-oductiim in the Dos ;id Cat. Loiidoti. Bailiiei-è Tiiiddl. 1984.

Christie DW. Bell ET. Some observations on the .wriscinaI incidence and krjuency of estius in breediiig bitches in Britüin. J Sm Aniin Pi;tct 197 1: 12: 159- 167. Christie DW, Bell ET. Studies on canine reproductive khavioi- during the normal estrus cycle. Amin Behriv 1972: 20: 62 1-63 1.

Collu R, Gibb W, Ducharme SR. Effects of stress on the gonadal function. J Endocnnol Invest 1984; 7: 529-537.

Colon J, Kim bal1 M. Hansen B, Concannon PW. Efkcts of contraceptive doses of the progestagen megestrol acetate on luteinizing hormone and follicle-stimulating hormone in female dogs. J Reprod Fertil Suppl 1993: 47: 5 19-52 1. Concannon PW. Effects of hypophysectomy md of LH administration on luted phase plasma progestenine levels in the beagle bitch. J Reprod Fertil 1980; 58: 407-4 10. Concannon PW. Reproductive physiology ruid endocrine patterns tif the bitch. In: Kirk RW (ed.) Cun-ent Veterinq Therripy. Small Animal Practice. Vol. VIII. Philridelphia. WB Saunders Co.. 1983: 886-900. Concannon PW. Canine physiology of reproduction. In: Burke T (cd.) Small Animal Reproduction and Infertility. A Clinical Approrich to Diagncisis and Treatment. Philadelphia. Lea and Febipr. 1986a: 23-77. Concannon PW. Canine pregnancy md partuition. Veteiinaiy Clinics of Noith America: SrnaIl Animal Pmctice 1986b; 16: 453-475. Concannon PW. Induction of fertile oestrous in moestrous dogs by constant infusion of GnRH agonist. J Reprod Fertil Suppl 1989: 39: 140- 160. Concannon PW. Reproduction in the dog and cat. In: Reprciduction in Domestic Animds (4th Edn). San Diego. Academic Press Inc.. 199 1 : 5 17-554. Concannon PW. Biology of gonüdotrophin secretion in adult and prepubertd female dugs. I Reprod Fertil Suppl 1993: 47: 3-27. Concannon PW. Reprodiictive endocrinology, contr'rlception. and pregnancy telmination in dogs. In Ettinger SJ. Feldman EC (eds.) Texbook of Vete~inriryInterna1 Medicine. Disemes of the dog ruid cat. Philadelphia. WB Saunders Co.. 1995: 1625- 1636.

Concannon PW, Battista M. Canine semen ti-eezing rind rutiticial in.semincition. In: Kirk RW (ed.) Current Veterinüry Theriipy. Small Animal Practice. Vo1.X. Philadelphia. WB Saunders CO.. 1989: 1247- 1259. Concannon PW, Cristricane VD. Serum ancirostenedione and testosterone concentrations duiing pregnancy and nonpregnünt cycfes in dogs. Biol Reprod 1985: 33: 1078- 1083.

Concannon PW. Hansel W. Effecrï of esuogens and pn)gsstrnmz on plasma LH. sexuül behavioui-. and pregnrincy in beagle bttches. Fed Proc 1975: 34: 323 (Abstrict). Concannon PW. Hansel W. Prosmglmdin F2a induced lutei~lysis.hypotheimia and abortions in beagle bitches. Pros~glandins1977: I -3: 533-542.

Concannon PW. Lein DH. Horrnal and clinicd correlates of ovarian cycles. ovulation. pscudopregnnncy. and pregnancy in dogs. In: Kirk RW (ed.) Current Veterinary Therapy X. Smrill Animal Practice. Philadelphia. WB Saunders CO.. 1989: 1269- 1282. Concannon PW. Temple M. Periovulritory dedine and late anestrus increase in ndoxone- inducible LH release in adult domestic bitches. Bi01 Reprod 1988: 38(suppl): IO 1.

Concannon PW. Cowan RG. Hansel W. LH release in ovariectomized dogs in response to estrogen withdrawl and its facilitrition by pmgestenme. Biol Reprod 1979a: 20: 523-53 1.

Concannon PW. Hansel W. McEntree K. Changes in LH. progestemnz and sexulil behaviour scxiüted with preovuhtory lutenizatioii in the bitch. Biol Reprod 1977~17: 604-6 1 3.

Concannon PW. Hllnsel W. Visek W. The ovlirian cycle of the bitch: plasma estrogen. LH and progesterone. Biol Reprod 1975: 13: 1 12- 12 1. Concannon PW. McCann 1P. Temple M. Biology and end~tinolrigyof ovulation. pregnancy and parturition in the dog. J Reprod Fertil Suppl 1989: 39: 3-25. C:mcannon PW, Whaley S. Anderson SP. Increrised LH pulse frequency associrited with termination of anestrus dunng the oviuian cycle of the dog. Biol Reprod 1986; 34 (Suppl.): 1 19 (Abstract).

Concannon PW. Powers ME. Holder W, Hansel W. Pregnancy and parturition in the bitch. Biol Reprod 1 977b; 16: 5 17-526. Concannon PW. Weigand N. Wilson S, Hansel W. Sexual behaviour in ovuïectomized bitches in respon-se to estrogen and progesterone administration. Biol Reprod 1979b: 20: 799-809, Concannon PW. Weintein R. Whaley S, Frmk D. Suppression of luteal function in do@ by luteinking hormone anti-semm and bromocriptine. J Reprod Fertil 1987: 8 1 : 175- 180. Concannon PW, Whdey S. Lein D, Wissler R. Canine gestation length: vwation relrited to the of mating and fertile Me of sperm. Am J Vet Res 1983: 44: 18 19- 182 1. Concannon PW. Bulter WR. Hanse1 W. Knight PJ. Hamilton JM. Parturition and lactation in the bitch: serum progestemne. cortisol. and prolrictin. Bi01 Reprod 1978: 19: 1113-1118.

Concannon PW. Isrimün L. Frank DA. Michel FJ. Currie WB. Elevated ccinceiitratioiis of 13. 14-diliydi-ci-15-ketci-prcistriglrinJin F-2a iii nilitziliül plasma during prCplirtUln luteolysis and paitiiition in dogs (Cmis PxniIiaris). J Repi-oc1 Fertil IWX: 84: 7 I - 77.

Creel S, Creel NM, Moiifoii SL. Stress and social doniinance. Nature. London 1996; 379. 313.

Cunningham DJ. Cüps Iiuiiting dvgs pictus) in ttic gu-dens uf die Ruyd Zoologicd Society of ireland. Proc. R. Soc. Edinbuigh 1905: 15: 843-848.

Czekalri NM. Durant BS. Callison L, Williams M. Millard S. Fecd stemid honnone malysis ris ÿn indicator of reproductive function in the çheetah. Zoo Biol 1994: 13: 1 19- 1%.

Czekala NM, Roser JF, Mortensen RB, Reichard T, Lasely BL. Urinary hormone analysis as a diagnostic tool to evaluate the ovarian function of female gorillas (Gorilla goda). J Reprod Fenil 1988; 82: 255-26 1 .

Deanesly R. Further observations on the effects of oestradiol on tubal eçgs and implantation in the guinea pig. J Reprod Fertil 1963; 5: 49-57. decatanzaro D. Effect of predator exposure upon early pregnancy in mice. Physiol Behav 1988: 43: 69 1-696. decatanzaro D, MacNiven E. Psychoçenic pregnancy disruptions in mammals. Neurosci Biobehav Rev 1992; 16: 43-53. decatanzaro D, MacNiven E, Ricciuti F. Cornparison of the adverse effects ofadrenal and ovarian steroids on early pregnancy. Psychoneuroendocrinology 1 99 1 : 16: 525-536.

De Coster T, Berickers JF. Beerens D, De May J. A homologous radioimmunoassay for canine pmlactin: plasma levels during the reproductive cycle. Acta Endwrinol Copenh 1983: 103: 473-478. DePalatis L, Moore J. Falvo RE. Plasma concentrations 01' testosterone and LH in the mde dog. J Reprod Fenil 1978: 52: 20 1-207.

Dickson AD. Delay of implantation in super-ovulated mice subjected to crowded conditions. Nature, London 1964; 20 1 : 829-840. Doak Ri, Hall A. Dale HE. Longevity of spermatozoü in tlie reproductive tract of the bitch. J Reprod Fertil 1967: 13: 51-58.

Doney IM,Gunn RG. Griftiths IG. The effect of premating stress on the onset of oestrus and on ovulating rates in Scottisli bfackface ewes. J Reprod Fertil 1973; 35: 381-384.

Doney JM, Gunn RG. Siiiitli WF, Cam WR. Effects of pi-e-mating environmental stress. ACTH, cortisone acetate or metyrapone on oestrus and ovulation in slieep. J Agric Sci 1 976; 87: 1 37- 133.

Dowling TE, Mincliley WL, Marsh PC, Demaris BD Response to Wayne. Nowak. and Phillips and Henry: Use of molecular cliaracters in conservation biology. Cons Bi01 1992; G(4): 600-603.

Durrant BS. Pratt NC. Russ KD, Bolambsi D. isolation and chmcterization of canine advanced preüntrii and exly anuiaal follicles. Theriogenology 1998: 19: 9 17-932.

Dun RH. Critical pend for early embryo mortality: ewes exposed to high arnbient temperature. J Anim Sci 1963; 22: 7 13-725. Edquist LE. Johansson EDB. Kasstrom H. Olsson SE. Richkind M. Blood plasma levels of progesterone and oestradiol in the dog during the cxstnis cycle and pregnruicy. Acta Endocinol 1975: 75: 554-564.

Eik-Nes KB. Secreticin of testosterone in mesthetized dots. Endwrindogy 1962: 7 1 : 101-106.

El der WH, Hayden CM. Use of discriminant function in taxonornic detemination of canids from Missouri. J Mammol 1977: 58: 1 7-24.

Elvidge H, Challis IRG, Robinson JS, Roper C.Thorburn, GD. Influence of handling and sedation on plasma cortisol in Rhesus monkeys (Macaca mulatta). J Endocrinol 1976; 70: 325-326. England GCW. Effect of progestens md imdrogens upon spermatogenesis and steroidgenesis in dogs. 1 Reprod Fertii Suppl 1997: 5 L : 123- 138.

Euker JS, Riegle GD. Effects of stress on pregnancy in the rat. J Reprod Fertil 1973; 34: 343-346.

Evans HM. Cole HH. An introduction to the study of the estrous cycle of the dog. Mem Univ Calif 193 1 : 9: 65- 103. Ewer RF. The Carnivores. Ithaca. NY:Comell University Press. 1973. Fülvo RE, DePdatis LR. Moore J, Kepic TA. Miller J. Annuül variations in plasma levels of testosterone and luteiniùng hormone in the laboritory maIe mongrel dog. J Endocrinol 1980; 86: 425430. Farstrid W, Anderson-Berg K. Frictors intluencing the success rate of artiticd insemination with frozen ,semen in the dog. J Reprod Fertil Suppl 19XO: 39: 289-292.

Ftiistad W. Bitcli krtility tifter ilriturd mating and aftei- tirti tkil iii.semiii;i~ic)iiwith lirsh or frozeii semeti. J Sm Anirn Pract 1984: 25: 56 1-565. Feidinan EC. Nelson RW. Canine Fernale Repr~ciductioriIII: Fe1 Jmaii EC. Nelson RW (eds.) Canine and Feline Endcicrinology xiid Repu idiiction. Phi ladel phia. W BStiundei-s CC).,1 W7: 399-524.

Fenske M, Grospietsch G, Koniç A. Plasma glucoconicostcriod and aldosterone levels during pliysiolo~icaland stress conditions in rabbits. II. Mature animals. Endokrinologie 198 1 ; 77: 59-64.

Fenske M. Adrenal function in the Mongolian çerbil (Meriones un.~iculanis):Influence of con fi nement stress upon gl ucocorti costeroid, proçesterone, dehydroepiandrosterone, testosterone and androstenedione plasma levels, adrenal content and in-vitro secretion. Exp Clin Endocrinoi 1986; 87: 15-35. Fernandes PA. Bowen RA. Kosm AC, Sawyer HR. Nett TM, OIson PN. Luteül tùnction in the bitch: changes during diestrus in pituitary concentrations and the number of luteal receptors for luteinizing hormone and prolactin. Biol Reprod W87: 37: 804-8 1 1.

Ferrell RE, Monzot DC, Hom J, Carley CJ. BiochemicaI markers in species endançered by introgression: the red wolf. Biochem Genet 1 978; l8:B-49.

Flesness NR. Mmrnalim extinction ratex Background to the black-hoted ferret drrima. In: Seal US. Thorne ET. Bogan MA. Anderson SH (eds.) Ccmservation Biology and the Blrick-hoted ferret. New Haven: Yale University Press. 1989: 3-9.

Fontbonne A. Badinand F. Canine ürtit'iciril inseminrition with frozen .semen: cornpuison of intrivaginal and intriutenne deposition of semen. J Rzprod Fertil ~uppi1993: 47: 325-327. Foote RH. Swierstra EE. Hunt WL. Spermatogenesis in the dog. Anat Rec 1972: 173: 34 1-352. Fowler ME. Zoo & Wild Animai Medicine. Philadelphia: WB Saunders Co.. 1986.

Franzmann AW, Flynn A, Arneson PD. Semm corticoid Ievels relative to handling stress in Alaskan moose. Can J Zoo1 1975; 53: 1424- 1426. Fuller GB, Hobson WC, Reyes Fi, Winter JSD, Faiman C. Influence of ketamine anesthesia on adrenal steroids, progesterone, and çonadotropins in rhesus monkeys. Proc Soc Exp Bi01 Med 1984; 175: 187-490.

Fuller TK. Denning behaviour of wolves in nortliceiitiill Minnesota. Am Mid1 Nat 19Wb: 121: 184-1x8.

Fuller TK. Pupulaticin Jyrnanics of wolves in iioithcenti-,ilMinnesota. Wildl Monogr 1989a: 105: 1-41.

Gensch W. Notes cm breeding Timber Wolves Canis lu~tist~cidentrrlis tit Dresden Zoo. Int. Zoo Yearb I9AEI: X: 15- 16.

Ghosal SK. LaMarche PH, Blirinjii P. Joai-dar S. Seguptri. A. ~MidyriT. Durition of meiosis and spennatogenesis in the Jog. Can J Genet Cytol 1983: 15: 678-68 I .

Gidley-Baird AA, O'Neill C. Sinosich MJ, Porter RN. Pike IL. Saunders DM. Failure of implantation in human in vitro fertilization and embryo transfer patients: The effects of altered prosesterone estrogen ratios in humans and mice. Fertil Steril 1986; 45 : 69-74.

Gidley-Baird AA. Endocrine control of implantation and delayed implantation in rats and mice. I Reprod Fertil Suppl 198 1 ; 29: 97- 109.

Gier HT. Ecology and behaviour of the coyote (Canjs latrans). In: Fox MW (ed.) The Wild Canids. New York: Van Nostnnd Reinhold. 1975: 247-262. Gipson PS. Gipson [K. Sealander JA. Reproductive biology of wild Cmis (Cmidae) in Arkansas. J Mamrnol 1975: 56(3): 605-6 12.

Gipson PS. Sealander JA. Dunn JE. The truconomic status rif wild Canis in Arkansas. Syst. Zoo1 1974: 23: 1 - 1 1.

Goldman EA. Classification of Wolves. 1n:Younç SP, Goldman EA (eds) The Wolves of North America. Washington: Amer Wild Inst, 1944: 389436. Goodrowe KL. Appliçütions of reproductive bioteçhnology tu wildlife conservation progams. Proceedings for Annuai Meeting; Septem ber 17-20: Montreril. Canada. Hastings. Nebraska: Society for Theriogenology: 1997: 30-37. Goodrowe KL. Hÿy MA. Platz CC, Behms SK. Jones MH. Waddell WT. Characteristics of fresh and frozen-thawed red wolf (Canis nifus) spermatozoa. An Reprod Sci 1998: 53: 299-308. Graf W. Serum oestrogen. progesterone and prolactin concentmtions in cyclic. pregnant and lactating beagle dogs. I Reprod Fertil 1978: 52: 9- 14. Graham LH, Brown JL. Cortisol metabolism in the dornestic cat and implications for non4 nvasive monitoring of adrenocortical function in endangered felids. Zoo Bi01 1996; 15: 71-82.

Graham LH, Goodrowe KL, Raeside JI, Liptrap RM. Non-invasive monitoring of ovarian function in several felid species by nieasurement of Fecal estradiol-17P and progestins. Zoo Biol 1995; 14: 223-237. Graham LH. Raeside JI. Gwdrowe KL. Liptmp RM. ~Me~~surementsoF friecril oestrrldiol and progesterune in non-pregnant and pregnrint doinestic esotic cats. J Repriid Feitil Suppl 1093: 47: 1 19- 120.

Green JS. Adam RA. Wi,odriit't' RA. Stelltlug JN. Seris~mrrlvariaticin in seinen pniduction by captive coyotes- -1 Mrrmrnol I9X4: 65: 506-509.

Greenwald GS. Luteolytic effect of estrogen on the corpora lutea of the hamster. Endocrinology 1965; 76: 1213-1219.

Griffith WFB. Arnoros~iEC. Proestrus. oestrus. ovulatirm and mating in the gmyhound bitch. Vet Rec 1939: 5 1: 1279- 1284.

Gross TS. Development and use of hecd steroid analyses in severi cmivore species. In: Proc Int Symp Faecal Steroid Monit Zoo Anirn. Zool rind Bot Gardens. Rotterdam. Gerrnan y. 1992: 55-6 1. muth DF. Concannon PW, Daels PF. Lrisely BL. Pregnancy-specitic elevations in fecal concentriitions of estradiol, testosterone and progesterone in the domestic dog (Canis t'rimi1iari.r). Theriogenology 1998: 50: 237-248. Hadley JC. Total unconjugrrted oesuogen rind progesterone concentrations in cyclic. pregnant and Irrctating beagle dogs. J Reprod Fertil 1975: 44: 453-460.

Hasion N, Watanabe M. Golzieher JW. Inhibition by adrenocorticotrophin-induced ovulation in immature fernale rats. Endocrinology 1969; 84: 308-3 14.

Hamil ton GD, Weeks HP. Cortisol and aldosterone corn pari sons of cottontail rabbi ts collected by shooting, trapping and falconry . I Wi Id1 Dis 1 985: 2 1 : 40-42.

Hmlett GWD. The reproductive cycle of the coyote. US Dept Agric Tech Bull 1938: 1 - 12.

Harper MJK. Eficts of androstenedione on pre-implantation stages of pregnancy in rats. Endocri noiogy 1967; 8 1 : 109 1- 1099.

Harper MK. Estrogenic effects of dehydroepiandrosterone and its sulfate in rats. Endoçrinology 1969; 84: 229-23 5. Hartley FGL. Follet BK, Harris S, Hirst D, McNeilly AS. The endocrinoloyy of gestation failure in foxes (Vdpes vulpes). J Reprod Fertil 1 994; 1OO:34 1-346. Hay MA. Canine gümetes - Evduation of uoctye maturation and penetmting potentid of spermatozori pre-freezc: and post thaw. Guelph, Ontario. Canada: University {if Guelpli: 1996. I S2p. PhD Thesis.

Heripe W. The -sesual -seasun uf mammals and the relationship o1'"pi-o-estnis'. tci menstruation. Pai~I Q J Micribiol Sci 1900: 44: 1 -70.

Hegstad RL. lohnston SD. Pxsternak DM. Concentrations and pulse aiia1y.w~of adrenoctiiticc~trqliiinand luteinizing Iic~t-inviiein pliisma ti-(lm Jogs in pnj-oestrus. oestrus. dioestrus or ilnoestrus. J Repi-cd Fertil Suppl 1993: 47: 85-9 1.

Hensleigh PA, Johson DC. Heat stress effects during pregnancy. II. Pitui tary çonadotropins in intact. adrenalectomized, and ovariectomized rats. Fenil Steril 1972; 22: 528-535.

Hoffmann B, Schneider S. Secretion and relese of luteinizing hormone duhg the luteal phrise of the oestrous cycle in the dog. I Repmd Fertil Suppl 1993: 47: 85-9 1.

Holst PA. Phemister RD. Onset of diestrus in the Beagie bitch: Detinition and Significance. Am J Vet Res 1974: 35: 40 1-406. Holst PA. Phemister RD. Temporal sequence oievenw in the estnius cycle of the Bitch. Am J Vet Res 1975: 36: 705-706.

Holst PA. Phemister RD. The prenatal development of the dog: pi-e-implantation events. Bi01 Reprod 1 97 1 : 5: 1 94-306.

Howard JG, Roth TL, Byers AP, Swanson WF, Wildt DE. Sensitivity to exogenous gonadotrophins for ovulation induction and laparoscopic artifical insemination in the cheetah and clouded leopard. Bi01 Reprod 1997; 56: 1059- 1068.

Howarth B. Hawk W.Effect of hydrocortisone on embryonic survival in sheep. J Anim Sci 1968; 27: 117-121.

Hutchins M, Wiese EU. Beyond genetic and dernographic management: The future of the Species Survival Plans and related AAZPA conservation efforts. Zoo Biol 199 1 ; 10: 285-292.

Jackson P. Fun- B, Hutchinson F. A preliminary study of pregnancy termination in the bitch with slow-release formulations of prostriglandin analogues. J Sm Anim Pmct t 982: 23: 287-294. Jeffcoate IA. Endocrinology of anoestrous bitches. J Reprud Fertil Suppl 1993: 37: 69-76. Jeft'coate LA. Englmd GCW. Unnary LH. plasma LH and progesterone and their clinical correlates in the periovulntory period of domestic bitches. J Repmd Fenil Suppl 1997: 5 1 : 267-275.

Jiichle W. Arbeitei K. Post K. Brilllibio R, D'Ver AS. Effects of pseudopregririncy. pregnmcy and illterc>estnwsintervals of pliumaci~lo~icriIsuppression of prolactin secretion in fernale dogs and c'rit.. . J Reprod Fertil Suppl 19XC): 39: ICW2O7.

Johnston LA. Armstrong DL. Brown JL. Seasonal çfficts on .seminal and endocrine triits in the captive snow leoprird (P;inrlie~xirncia). J Repnid Feitil I 994: 101: 229-236.

Johnston S D, Mather EC. Canine plasma cortisol (Iiydrocortisone) nieasured by radioimmunoassay: Clinicai absence of diumal variation and results of ACTH stin~ulationand dexamethasone suppression tests. Am J Vet Res 1978; 39: 1766- 1770. Johnston SD. Breeding management of the bitch. In: Ettinger SI. Feldman EC (eds.) Textbook of Veterinary Internai Medicine. Diseases tif the dog and cat. Philadelphiri. WB Saunders Co.. 1995: 1604- 1606.

Jones CA, Refsal KR. Lippen AC. Nachreiner RF. Schwacha MM. Changes in the adrend cortisol secretion as reflected in the urinaty cortisol/creatinine ratio in dogs. Dom Anim Endocrinol 1990; 7: 559-572.

Jones GE. Baker K. Fhmy DR. Boyns AR. Effect of luteinizing hormone releasing hormone un plasma levels of luteinizing hormone. oestmdiol and testosterone in the male dog. J Endocrinol 1976: 68: 469-474. Jones GE, Boyns AR. Cameron EHD. Bell ET. Christie DW. Parkes MF. Plasma oestr~diol.lutenizin8 hormone and progesterone during pregnancy in the Beagle bitch. J Reprod Fertil 1973: 35: 187- 189.

Jones GE, Boyns AR. Inhibition by oestmdiol of the pituitruy responsc: to luteinizing hormone releasing hormone in the dog. J Endocrinol 197k 68: 375-479. Jurke MW. Czekrila NM. Linburg DG. Millard SE. Fecd ciwticoid metabolite meuurement in the cheetah (AcinwW. Zoo Biol 1 997: 1 6: 133- 147. Kaplan JB, Mead RA. Seasonal changes in testiculrir function and semincil characteristics of the male eastem spotted skunk (Soiloeale putonus gmbarvilus). J Mammol 1994: 75(4): 10 13- 1020. Kawakarni E, Hori T, Tsutsui T. Changes in plasma luteinizing homne. testosterone and estmdiol- 17p levels and semen quality after injections of gonadotropin releasing hormone agonis

Kennelly JJ, Johns BE. The estrous cycle of the coyotes. J Wildl Manrige 1976: 40: 272-277. Kennelly JJ. Coyote reproduction-1.The durrition of the spermrrtogenic cycle and epididymal spem transport. J Reprod Ferti 1 1972: 3 1 : 163- 170. Kiso Y, Yamauchi S. Hisotchernicd study on hydroxystericd dehydrogenues in the trophoblast of the dog placenta. Jap J Vet Sci IYX3: 46: 2 19-233. Knight PI. Hamilton JM, Hiddleston WA. Serum pi-olrictiii duting pregnaiicy and Irictatirin in the beagle bitch. Vet Rec 1977: 101: 202.

Knol BW. Dieleman SJ. Bevers MM. van der Brom WE. GnRH in tlie male clog: do.*- response relrrtionsliip with LH and testcisterone. J Reprid Fertil 1993: 98: 159- 16 1.

Koehler JK. Platz CC Jr.. Wrrddell W. Jones MH. Smith R. Behrns S. Spermophrigy in .semen in the red wolf (Crrnis rubis)). Md Reprcid Dev 1994: 37: 45746 1. Koehler JK. Platz CC Jr.. Waddell W. Jones MH. Behrns S. Semen parrrrneters and electron microscope observations of spermütcizoa of the red wolf. Crini.'; rufus. J Reprod Fertil 1998: 114: 95-101. Kreeger TJ, Sed US. Cohen Y. Plotka ED. Asa CS. Chmcterizaticin of prolactin secretion in gray wolves (Canis JUDUS). Cm J Zool 199 1 ; 69: 1366- 1374.

Kuroda H, Hiroe K. Studies on the metabolism of dog sperm. 1. Seasonal variation on the semen quality and the of the aerobic metahlism of spennatozoa Jpn .i Anim Repmd 1972; 17: 89-98.

Kurten B, Anderson E. Pleistocene Mammals of North America. New York: Columbia University Press, 1 980.

Lacoste D, Arnaud R. Bleanger A. Labrie F. A pure mtiandrupen does not inteitkre with the LHRH agonist-induced blockage of t=ticuIar rrndrogen .secretion in the dus. Mol Ce11 Endocrinol 1988: 56: 141-147.

Lammond DR. Synchronization of oestrus and ovulation in beef heifers. Proc Aust Soc Anim Prod 1962; 4: 72-78.

Lasely BL, Kirkpatnck IF. Monitoring ovarian function in captive and free-ranging wildlife by means of urinary and fecal steroids. J Zoo Wildl Med 199 1 ; 22: 23-3 1.

Lawrence B, Bossert WH. Multiple character analysis of Canis lupus, latrans, and familiaris, with a discussion of the relationships of Canis niger. Amer Zoo1 1967; 71223-232. Lentier JW,Englund J. Notes on the captive wolf (crini&lu~us) colony. Barrow. Alaska. Crin J Zoo1 1973; 5 1 : 623-627. Lincoln GA. Seasonal Aspects of Testicular Functiun. In: Burger H. de Kretser D (eds.) The Testis. New York: Rriven Press Ltd. 1989: 329-385. Linde-Forsberg C. Forsberg M. Fertility in dogs in relation to semen qudity md the time and site of inseminrition with tkesh and frozen semen. J Reprod Fertil Suppl 1989: 39: 299-3 1O. Linde-Forsberg C. Forsberg M. Results of 527 controlled i.utificiii1 inLwminationsin dogs. J Repord Fertil Suppl 1993: 47: 3 13-323.

Lindeberg H. Jalkaneii L. Savolainen R. In vitro cultiii-c:01' silvei- t;)s ernbryos. Theriogenology 1993: 40: 779-78s.

Liptrap RM. Stress and reproduction in domestic ani nials. In: Corticotropin-Releasing Factor and Cytokines: Role in the Stress Response. Ann NY Acad Sci 1993; 697: 275-383.

Lli)yd GH. ErgLund J. Tlie reproductive cycle of the rdfox in Eiii-cqx. J Repi-iid Fei-til Suppl 1973: 1'): 119- 130.

Lucas 2, Raeside JI, Betteridge K. Non-invasive assessment of the incidences of pregnancy and pregnancy loss in the feral horses of Sable Island. J Reprod Fehl 199 1; 44: 479488.

MacKenzie AH, Thwaites CJ, Edey TN. Oestrous, ovarian and adrenal response of the ewe to fasting and cold stress. Aust J Agric Res 1975; 26: 545-552.

MacMillan KL, Watson JD. Short estrous cycles in New Zealand dairy cattle. J Dairy Sci 197 1 ; 54: 1 526- 1 529.

Mason JW."Over-ail" hormonal balance as a key to endocrine orginization. Psychosom Med 1968; 30: 791-808.

Matsuda DM, Bellem AC, Gartley CJ, Madison V, Kin3 WA, Liptrap RM, Goodrowe KL. Endocrine and behavioral events of estrous cyclicity and synchronization in wood bison (Bison bison athabascae). Theriogenology 1996; 45: 14%- 144 1. Matteri, R.L.. Roser. J.F.. Baldwin, D.M..Lipovetsky. V..rind H. Papkoff. Chmcterizrttion of a monoclonal rintibody whiçh detects luteinizing hormone frcim diverse marnmrilim species. Dom Anim Endocrinol 1987: 4: 157- 165. McCann JP. Altszuler N. Hampshire J, Concannon PW. Grvwth hormone. insulin. glucose, luteinizing hormone. and diabetes in beagle bitches trerrted with rnedroxyprogesterone acetate. Acta Endocrinol Copenh 1987: 1 16: 73-80. McCann IP. Temple M, Concannon PW. Pregnmcy-specik alteritions in the metabolic endocrinology of domestic dogs including insulin resistmce and moditied regulation of growth hormone. Proc 1 1th Int Congr Anim Reprod & Ai. Dublin 1988; 2: 103- 105. McCarley H, Carley CL. Recent changes in distribution of wild red wolves (Canis nifus). USFWS Endangered Species Report No 4, Albuquerque, NM, 1979,38p.

McCarley H. The taxonomie status of wild Canis (Canidae) in the South-central United States. S W Nat 1962; 7: 227-235.

McGuckin MA. Blackshriw AW. Mating-;~,ssoci;~tedperik in pl;~srnatestosterone concentratiofi in wild gray-herided t'lying tims (Pteroous -1. J Repi-od Ferti 1 I 99 I ri: 92: 347-352. McGuckin MA. Bliickshaw AW. Se~sonrilchanges in tssticuirii- six. plamri testosterone concenwdticiii and body weight in captive tlyiiig foxes (Ptei-cmus~fdioce~haiiis aiid -P. sca~iilritus).J Repriid Fertil IO9 I : 92: 339-340.

McLeod PJ, Moger WH, Ryon J, Gadbois S. Fentress JC The relation between urinary cortisol and social behaviour in captive timber wolves. Can J Zoo1 1996; 74: 209- 216.

McRrie GI. Roberts BB. Worden AC. Bajka A, Vickeiy BH. Long-term reversible suppression of oestrus in bitches with nafamiin ricetate. a potent LHRH agonist. J Reprod Fertil 1985: 74: 389-397.

Mech DL. Longevity in wild wotves. J Miunmai 1988: 69: 197- 198. Mech DL. Premature reproductive activity in wild wolves. J Marnmal 1987: 68: 87 1- 873. Mech DL. The Way of the Wolf. Stilwater. MN: Voyageur Press Inc.. LW 1. Mech DL. The Wolf. New York. NY: Doubleday. 1970.

Medjo D, Mech DL. Reproductive rictivity in nine and ten incinth cild wulves. J Mamrncd 1976: 57: 406-408.

Mialot JP, Thibier M. Toublrinc JE, Castanier M. Scholler R. Plasma ccincentrritions rif Iuteinizing hormone. testosterone. dehydrcxpiandrosterone. androstenedioiie between birth and one year in the male dog: Longitudinal study and hCG stimulation. Andrologia 1988; 20: 145- 154.

Miller MW, Hobbs NT, Sousa MC. Detecting stress responses in Rocky Mountain biçhom sheep (Ovi s canadensis canadensis): Rel i abi lity of cortisol concentrations in unne and feces. Can J Zoo1 1991; 69: 15-24. Mirarchi FE. Howland BE, Sçanlon PF, Kirkpatrick RL. Ssinkrd LM. Seasonal variation in plasma LH, FSH, prolactin, and testosterone concenwitions in adult male white- tailed deer. Can J Zoo1 1978; 56: 12 1- 127.

Mitsuzuka M. Collection. evduation cuid freezing of wolf sperrn. In: Frank H (ed.) Man and Wolk Advances. Issues, and Problems in Captive Wolf Research. Dordrecht. Netherlands: Dr W Junk Publishers. 1987: 127- 14 1. Moberg GP. Influence of stress on reproduction: Measure of well-being. In: Moberg GP (ed.) Animal Stress. Bethesda: Amencan Physiological Society, I985a: 245-267.

Moberg, GP, Watson JG, Hayashi K. Effects of adrenocorticotrophin treatrnent on estrogen, iutenizing Iiomone, and progesterone secretion in the female rhesus monkey. J Med Primat01 1982; I 1 : 235-24 1 .

Moehlman PD. Intnduction. In: Sheldon JW (Ed) Wild Diigs, The natrird history of the nondomestic crrndiie. San Diego: Acridernic Press Inc.. IW2: 1-3.

Miiller OM. Piugesti-mne coiicentmtion iii the peipliizial piUin:i of the blue tiix (Alopex diii-ing pregnancy and the oestt-ous cycle. J Endt~i-in011073: 59: 429-438.

Mrmdain-iMonvti1 M. Smith N.Simon P. Mollei- OM. Scholler- R. iMcNeilly AS. Effect of melatonin implantation on the seasonal variation of FSH secretion in the male blue fox (Alopex Iriso~us).J Reprod Fertil 1988: K3: 345-354. Monfort SL. Wasser SK. Machburn KL, Burke M. Brewer BA. Cree1 SR. Steroid metabolism and validation of noninvasive endocrine monitoring in the African wild dog (Lvcacon ~ictus).Zoo Bi01 1997: 16: 533-548. Morgan PJ, Mercer JG. Symposium on 'Seasonality'. Proc Nutr Soc 1994: 53: 483-493.

Morton DJ, Anderson E, Foggin CM, Kock MD, Tiran EP. Plasma cortisol as an indicator of stress due to capture and translocation in wildlife species. Vet Rec 1995; 136: 60-63.

Miistl E, Brunner 1. Cornpiirison of different progestagen ;lss;iys for measuring progesterone metabolites in tàeces of the bitch. J Vet Med 1997: 44: 573-578.

Miistl E. Lehmann H. Wenzel U. Gestagens in the t'rieces of mink and cmfor monitoring corpus luteum activity. J Reprod Fertil (Suppl) 1993: 47: 540-541.

Munro CJ, Stabenfeldt F. Development of a cortisol enzyme irnniunoassay in plasma. Clin Chem 1985; 3 1 : 956 (Abstract).

Munro CJ, Stabenfeldt GH, Cragun JR, Addiego LA, Overstree JW, Lasley BL. The relationship of serum estradio1 and progesterone concentrations to the excretion profiles of their major urinary metabolites as measured by enzymeirnrnunoassay and radioimmunoassay. Clin Chem 199 1; 37: 838-844.

Nalbandov AN. Reproductive Physiology. San Francisco, CA; Freeman. 1964, p 13 5.

Nequin LG, Schwartz NB. Adrenal participation in the timing of matinç and LH release in the cyclic rat. Endocrinology 1971 ; 88: 325-33 1. Nett TM, Akbar AM. Phemister RD, Holst PA, Reichert LE. Jr.. Niswender GD. Levels of lutenizing hormone. estndiol and progestei-one in the semm during the estrous cycle and pregnancy in the beagle bitch. Proc Soc Exp Biol Med 1975; 148: 134- 139.

Nowak RM. North Americam Quatemary Canis. Museum of Natural History, University of Kansas Monograpli 6. 1979, 154p.

Nowak RM. The mysterious wolf of the south. Nat Hist lW2; 8 1 : 5 1-53.

Nowak RM. The red wolf is not a hybrid. Cons Biol 1992; 6(J): F93-595

Nowak RM, Federoff NE. Systematics of Wolves in Eastern North Ameiica. Proc Defenders of Wildlife's Wolves cif North Amencri; Albany. NY 1996: 1 S7- 194.

Okkens AC, Bevers M. Dieleman S. Willemse AH. Shortening of the interestrous interval and lifespan of the corpus luteum of the cyclic dog by bromocriptine treatment. Vet Q 1985s; 7: 173- 176.

Okkens AC, Bevers MM, Dieleman SJ, Willemse AH. Evidence for prcilactin as the main luteotrophic factor in the cyclic dog. Vet Q 1990: 12: 193-21) 1. Okkens AC. Dieleman SJ. Bevers MM, LubbeRnk AAME. Willemse AH. intluence of hypophysectomy on the lifespan of the corpus luteiim in the cyclic dog. J Reprod Fertil 1986; 77: 187- 192. Okkens AC, Dieleman SJ. Bevers MM, Willemse AH. Evidence for the noninvolvement of the utei-us in the lifespan of the corpus luteum in the cyclic dog. Vet Q 1 %Sb: 7: 169- 173.

Olar Tï. Using frcizen canine semen: a guide for practitiiiners. Vet Med 1985: Mmh: 22- 30. Olson PN. Bowen RA. Behrendt M, Olson JD. Nett TM. Concentritions of reproductive hormones in canine serum throughout Irite rinesuus. proestrus and estrus. Biol Reprod 1982: 27: 1 196- 1206.

Olson PN, Bowen RA. Behrendt MD. Olson JD. Nett TM. Concentrations of testosterone in canine serum dunng lrite anestrus, proestrus. estius and early diestrus. Am J Vet Res L984b: 45: 145-148. Olson PN, Bowen RA. Behrendt MD, Olson JD. Nett TM. Concentritions of progesterone and luteinizing hormone in the serum of diestrous bitches before ruid after hysterectomy. Am J Vet Res 1984~:45: 149- 153. Olson PN. Husted PW. Allen TA, Nett TM. Reproductive endoçiinology and physiology of the bitch and queen. Veterinary Clinics of North Amencri: Small Animal Pnctice I984a: 14: 927-946.

Olson PN. Nett TM. Bowen RA, Sawyer HR. Niswender GD. Endocrine regulation of the corpus luteum of the bitch ris a potential target for altering tèrtility. J Reprod Fertil Suppl 1989: 39: 27-40. Onclin K, Verstegen JP. In vivo investigation of luteül tùnction in dogs: eftècts of cabergoline. a dopaine agonist, and prolac tin on progesterone secretion during mid-pregnrincy and -dioestms. Dom Anim Endocrino1 1996: 14: 25-38. Onclin K, Verstegen P.Secretion patterns of plasma pmlactin and progesterone in pregnant comparecl with nonpregnrint dioestroris bi tches. J Repriid Fertil Suppl 1997: 5 1 : 203-208.

Othen LS, Belleni AC. Gartley CJ, Aukland K, King WA, Liptrap Rb!, Goodrowe KL. Hormonal control of estrous cyclicity and attempted superovulation in wood bison (Bison-- bison athabascae). Tlieriogenoloyy 1 999; 54. 3 1 3-323.

Paradis M. Post K. Mripletoft RJ. Effect-s of pn~staglsindinF2a on corpori lutesi ti)rmritioti and function in mrited bitches. Cari Vet J 1983: 24: 23'1-242.

Paradiso JM, Nowak RM. A report on the taxonomie status and distribution of the red wolf. USFWS Spec Sci Rept Wildl 1971; No 145: 1-36. Paramo RM. Renton JP. Ferguson JM. Concannon PW. Eftèca of medroxyprogesterone acetate or gonadotrophin-releaing hormone on suppression of spermatogenesis in the dog (Cains familiaris). J Reprod Fertil Suppl 1993: 47: 387-397.

Parker WT. The red wolf. In: Chandler WJ, Labate L. (Eds.) Audubon Wildlife Report 198811989. San Diego: Academic Press, Inc.. 1988: 597-607.

Parkes AS, Bellerby CA. Studies on the interna1 secretions of the ovary. 2. The effects of injections of the oestrus producing hormone during pregnancy. J Physiol 1926; 62: 145-151.

Farkes AS, Bruce HM. Pregnancy-block in female mice placed in boxes soiled by males. J Reprod FertiI 1962;4: 303-308.

Pauinj S. Sundmarajan N. Manimozhi A. Walker S. Reprciduction of the Indiün wild dog (Cuon &inus) in captivity. Zoo Bi01 1992: 1 1 : 235-24 1.

Penton JP, Boyd JS. Eckersall PD, Ferguson JM, Hmey MJA, Mullaney J, Perry B. Ovulation. fertilization and early embryonic development in the bitch (Cani'; familiariQ. J Reprod Fertil 1991: 93: 221-231. Phemister RD, Holst PA, Spano JS, Hopwood ML. Time cif ovulation in the beagle bitch. Bi01 Reprod 1973: 8: 74-82.

Phillips MK, Henry VG. Comments on red wolf taxonomy. Cons Biol 1992; 6: 596- 599.

Pimlott DH, Joslin PWD. The status and distribution of the red wolf. Trans N Am Wildl Nat Res 1968; 33:373-389. Porton U, Kleimm DG. Rodden M. Aseasondity of bush dog reproduction and the intluence of scicial factors on the estrous cycle. J Mmmd 1987: 68: 867-87 1. Puswell BJ. Wiicke IR. Response to gonridotrophin-releuing hormone by the intact mde dog: serurn testosterme. luteinizing hormone and follicle-stimulriting hormone. J Reprod Fertil Suppl 1993: 47: 335-34 1.

Putney DJ, Drost M. Thatcher WW. Embryonic developnient in superovulated dairy cattle exposed to elevated ambient teniperatures between days I to 7 post insemi nation. Tlieriogenolo~y1988; 30: 195-209.

Ramaley JA. Stress and Fertility. In: Gilmore D. Cook B (eds.)Envioriiiental factors in rnarnmai reproduction. London:MacMillian Publistiers Ltd. 198 t : i 27- t 4 1. Reimers T, Phemister R. Niswender G. Rridioimmunological merisurement of follicle stimuhting hormone and prolactin in the dog. Biol Reprod 197%: 19: 673-679.

Rich TD, Turman EJ, Hillier K. A com parison of the ovulation rate, fertilization rate and embryo survival of hand-mat4 and lot-mated gilts. J Anim Sci 1968; 27: 443- 446. Richkind M. Possible use of early moring urine for dectection of pregnancy in dogs. Vet Med/Small Animal Clinic 1983: 78: 1067- lM8.

Riley GA, McBnde RT. A survey of the red wolf (Canis nifus). USFWS Spec Sci Rep Wildl 1972; No 162: 1-15.

Robeck TR, Schneyer AL, McBain JF, Dalton LM, Walsh MT, Czekala NM, Kraemer DC. Anal ysi s of urinary i mmunoreactive steroid metabol i tes and gonadotropi ns for characterization of the estrous cycle, breeding period, and seasonal estrous activity of the captive killer whales (Orcinis orca). Zoo Bi01 1993; 12: 173- 187.

Roblero LS, Garavaçno AC. Effect of oestradiol-I 7p and proyesterone on oviductal transport and eariy development of mouse embryos. J Reprod Fertil 1979; 57: 91-95.

Rctdden MD. Sorenson LG. Sherr A, Kieiman DG. Use of behriviourd mesures to üssess reproductive status in maned wolves (Chrysoc~onbiachvuruQ. Zoo Biol 1995: 15: 565-585.

Rowell TE. Baboon menstrual cycles affected by social environments. .J Reprod Fenil 1970; 21: 133-141. Roy MS, Geffen E, Smith D, Ostrander EA, Wayne RK. Patterns of differentiation and hybridization in North American wolfli ke canids, revealed by analysis of micosatellite loci. Mol Biol Evol 1994a; 1 1 : 553-570.

Roy MS, Girman DJ, Taylor AC, Wayne K. The use of museum specimens to reconstruct the genetic variability and relationsliips of extinct populations. Experientia 1994b; 50: 55 1-557-

Runner ML. Embryocidal effects of handling pregnant niice and its prevention witli progesterone. Anat Rec 1959; 1 33 : 330-33 1 .

Schulte BA. Sed US, Plotka ED, Letellier MA, Verine LH. Ozap JI. Pu-scins JA. Tlie effect of pineilectrrmy on seasonal changes in pizktctin .secretion in the white-titiled deer (Odrxiiileus vir-nimulm-ealis). Endociinri1 198 1 ; 1 OS: i 73- 178.

Schwartz NB. Acute effects of ovariectomy on pituitary LH, uterine weight and vaginal cornification. Am J Physiol 1964; 207: 125 1 - 1259.

Scott GH, Williams RJ. Causes of low breeding efikiency in dairy cattfe associated with seasonal high temperatures. J Dairy Sci 1 962; 45 : 1 3 69- 13 75. Seager SWJ. Platz CC. Hodge W. Successful pregnancy using frozen semen in the wolf. int Zoo Yearbk 1975: 15: 140- 143.

Seal US, and Mech LD. Blood indicators of seasonal rnetitbolic patterns in captive adult gray wolves. .i Wildl Mange t 983: 47: 704-7 15.

Seal US, Plotka ED. Mech DL, Packard JM. Seasonal metabolic md reproductive cycles in wolves. In: Frink H (ed.) Man and Wolf: Advances, Issues, and Problems in Captive Wolf Research. Dordrecht. Netherlands: Dr W Junk Publishers. 1987: 109- 125. Seal US. Plotka ED. Packard JM, Mech DL. Endocrine coirelates of reproduction in the wolf.1. Serum progesterone. estradiol and LH during the estrous cycle. Bi01 Reprod 1979; 2 1 : 1057- 1066. Sheldon JW. Wild Dogs. The natural history of the nondomestic canidtie. San Diego: Academic Press Inc.. 1992.

Shideler SE, Ortuno AM, Moran FM, Mootman EA, Lasley BL. Simple extraction and enzyme immunoassays for estrogen and progesterone metabolites in the feces of Macaca fascicularis during nonconceptive and conceptive ovarian cycles. Biol Reprod 1993; 48: 1290-1 298.

ShilleVM, Stabenfeldt GH. Clinical reproductive physiology in dogs. In: Morrow DA (ed.) Cument Therüpy in Theriogenology. Philadelphia: WB Saunders Co.. 1986: 57 1-574. Shille VM. Haggerty MA. Shackleton C. Ladey BL. Metabolites of estradio1 in serurn. bile, intestine and feces of the domestic cat (Fe- ci&&. Theriogenology 1990: 34: 779-794. Shille VM, Kollias GV. Thritcher MJ, Watermm L. Detennination of reproductive status in the serval and bobcat using a validated. direct radioimrnuno;~.s..yof fecal estrridiol. Biol Reprod 199 1 : (Suppl 1 ) 44: 12 1.

Shille VM. Wing AE. Lasley BL. Bmks JA. Excretioti of i-lidiolabeled estixdiol in the cat (FeIis -. L.): A preliminary report. Zoo Biol 1984: 3: 20 1 -?OC). Sinith AJ. Clausen OPF. Kirkhus B. Jahiiseii T. Mollei- OiM. Haiinson V, Se;~,sonal chüiiges in spzimatc~genesisin the blur fox (AIo,œx ~~ODUSJ.quantified by DNA tlow cytoineriy and ineasurement of soluble Mn-' -dependant iidenyllite cyclrise activity. J Repmd Fenil 1984: 72: 4534 1. Smith AJ, Mondain-iMrinval M, Berg KA. Gordeladze JO. Clausen OPF. Siinon P. Scholler R. Sexual development in the immature male blue fox (Ale-œx lago~us). investigated by testiscular histology, DNA tlow çytometp and measurements of plasma FSH. LH. testosterone and soluble testiculÿr Mn-' -dependan

Smith DM, Biggers ID. The oestrogen requirement for implantation and the effect of its dose on the i mpf antation response in the mouse. J Endocrino1 1968; 4 1 : 1-9. Smith FO, Giaham EF. Ci-yopreservation of canine semen: Technique and Perfoimance. Proc I 0th Int Congr Anim Reprod & AI. Urbüna. Champüign. 1984: 3 16.

Smith MS. McDonald LE. Semm levels of luteinizing hormone and progesterone during the esuous cycle. p.seudopregnmcy and pregnancy in the dog. Endocrinology 1974: 94: 404-4 12.

Soulé M, Gilpin M. Conway W. Foose T. The millenium rirk: How long a voyage. how rnany statemoms. how mmy passenprs? Zoo Biol 1986: 5: 10 1 - 1 13. Stabenkldt GH, Shille VM. Reproduction in the dog and cat. In: Cole HH. Cupps PT (eds.) Reproduciton in Domestic Animais. New York. Academic Press. 1977: 499- 527. Steinetz B. Goldsmith L, Lust G. Plasma relaxin leveis in pregnant and lactating dogs. Bi01 Reprod 1987: 37:7 1 9-725. Steinetz BG. Goldsmith LT. Harvey HJ, Lust G. Serum relüxin and progesterone concentrations in pregnant. pseudopregnant. and ovariectomized. progestin-treated. pregnant bitches: detection of relaxin as a marker of pregnancy. Am J Vet Res 1989: 50: 68-7 1. Stellflug IN. Musc PD. Evenon DO. Louis TM. Changes in semm progesterone and estrogen of the nonpregnant coyote during the breeding season. Proc Soc Exp Bio Med 198 1; 167: 220-223. Stoebel DP. Moberg GP. ERects of ridreoncorticotropin and cortisol on lutenizing hormone surge and estrous behaviour of cows. J Drriry Sci 1982: 65: 10 16- 1024.

Stone GM, Em mens CW. The action of oestradiol and di iiietliylsti lboestrol on early preçnancy and deciduoma formation in the moilse. J Endocrino1 1964: 39: 137- 145.

Strott CA, Sundel HlStalilman MT. Maternai and ktal plasma progesterone, cortisol. testosterone and 17p-estradiol in preparturient stieep: Response to fetal ACTH infusion. Endocrinolosy 1975; 95: 1327- 1332.

Taha MB. Studies on semen composition and testiculür function in the hg. with particular i.etèrence to the effect of season. age and exogenous hoimones. London: University of London: 1980. PhD Thesis. Thatcher MJD. Shille VM. Buhi WC. Alvarez IM. Concannon PW. Thibeault D. Canine conceptus appewce and de novo protein synthesis in relation to the time of implantation. Theriogenology 1994; 4 1 : 1679- 1692. Tsutsui T. Gamete physiology and timing of ovualtion and fertilization in dogs. J Reprod Fed SuppI 1989: 39: 269-275.

Tsutsui T. Studies on reproduction in the clog. V. On cleüvüge and trünspost of fertiiized ova in the oviduct. Jpn J Anim Reprod 1975: 21: 70-75. Tsuniu T. Shimizu T. Studies on the reproduction in the dog. IV. On the fertile pend of ovum atèr ovualtion. Jpn J Anim Reprod 1975: 2 1 : 65-69.

Tudge C. Lilst Animals at the Zoo. Washington DC: Island Press. I902.266p.

Ulberg LC. Eftèct of rnacro- and rnircoenvironrnent on the biolopy of mammdian reproduction. In: Shaw ROH (ed) Grvund Level Climütology. Washinton DC: Am Assoc Adv Sci. 1967: 265-276.

[USFWS]. Red Wolf reçovery plan. U.S. Fish and Wildlife Service. Atlanta. GA. 1984. 37p. [USFWS]. Red Wolf recovery plm. U.S. Fish and Wildlife Service. Atlanta. GA. 1990. 11op. Valtonen MH, Rajltkoski W. Makela Ji. Reproductive feritures in the fernale raccoon dog onoided. J Reprod Fertil 1977: 5 1 : 5 17-5 18. van Haatten B. Bevers MM, van der Brom WE, Dieleman SJ. Okkens AC. van Sluijs FJ. Willemse AH. Increasing sensitivity of the pituitary for GnRH from early to advanced ÿnoesuus in the beagle bitch. Proc 12th Inter Cong Anim Reprod & Ai 1992: 4: 1829-1831. van Hdten B. Dieleman SJ, Okkens AC. Bevers MM. Willemse AH. Induction of oestrus and ovulation in dogs by treatment with PMSG andor bromocriptine. J Reprod Fertil Suppl 1989: 39: 330-33 1.

Vanderlip SL. Winp AE. Fzlt P. Linke D. River S. Concannon PW. Lÿsley BL. Ovulation induction in anestrous bitches by puisaltile adrninistmtion of tonadotrophin- releasing hormone. Lrib Anim Sci 1987: 37: 459-464.

Vzleixdo JT. Action of aJrenocorticciti'r,piiion pi-egririiicy aiid littei- sizt: in rats, Am J Physiol 1957: 1') I : 3 19-322.

Velloso AI. WXSS~I-SK. Montim SL. Dietz JM. Limgitudiiial kcal stl=n)idtxcretion in maned wolves. Gen Comp Endciciincil I WS: 1 12: 06- 107.

Vickeiy B. McRae G. Et't'ects of synthetic PG analogue tm pi-egnriiicy in beagle bitches. Biol Reprocl IWO: 22: 43X42. Viçkery BH. Bergstrom K. Miller M. Goodpüsture JC. Effect of Detirdix. a long acting LHRH mtagonist. on luteal function and pregnancy in beagle bitches. Biol Reprod 1987; 36 (Supppl): 145 (Abstract) Vincent IC. Michel1 AR. Cornparison of cortisol concentrations in .diva and pluma of dogs. Res Vet Sci 1992; 53: 342-345. Waddell, WT. Red Wolf (Cimis nifus) International Studbook. 1995. WaddeU WT. AZA Species Survivd Plan Protile: Red Wolves. Endmgered Species Update 1996: 13 (IO & 11): 10-1 1.

ViliddeIl. WT. Red Wolf (Canis nifus) International Studbook. I99X (in preparation. 1099).

Wliddell. WT, Henry GV. Red wolfi recovery of a native spscies. ln: Wagner RO (ed.) AZA Regional Conference Proceedings. Wheeling. WV. 1Wh: 388-392.

Wagon KA, Rollins WC, Cupps PT, Carroll FD. Effects of stress factors on the estrous cycles of beef heifers. J Anim Sci 1972; 34: 1003-10 10.

Walker LA. Cornell L. Dahl KD. Czekaia NM, D~ganCM. Joseph B. Hsueh AJW. Lÿsley BL. Uainüry concentrations of ovünün steiuid hormone metabolites and bioactive follicle-stimulating hormone in killer whales (Orcinus orcd during ovarian cycles and pregnancy. Bi01 Reprod 1988: 39: 10 13- 1020. Waser SK. Velloso AL. Rodden MD. Using fecal steroids to evduüte reproductive function in female maned wolves. J Wildl Manage 1995; 59: 889-894. Wasserman GF. Bddi A. del Castillo E, Charreau E. Metabolism of ['Hl-testosterone by the adrend of the dog: the action of ACT)-I. J Steroid Biochem 1973; 4: 2 17-226. Wayne RK, Gittleman JL. The problematic red wolf. Scientific Amencri. July 1995: 36-39. Wayne RK. Jenks SM. Mitochrondnai DNA ÿndysis irnplying extensive hybridization of the endangerd red wolf JW~US.Nature. London 199 1 : 35 1 : 565-568. Wayne RK. On the use of morphologic wd molecular pnetiç characters to investigate species stanis. Cons Biol 1992: 6: 59 1-592.

Wiese RJ, Hutchins M (eds). Species Survival Plans: Strategies for Wi ldl ife Conservation. Bethesda, MD: American Zoo and Aquarium Association, 1994, 64p.

Wesson JA. Scanlon PF. Kirkpatrick RL. Mosby HS. Butcher RL. Influence of clieniical immobilization and pliysical restraint on steroid hormone levels in blood of wbite- tailed deer. Can J Zoo1 1979; 57: 768-776.

Whitney R, Burdick HO. Tube-locking of ova by oestrogenic substances. Endocrinology 1936; 20: 643-647.

Wiebold JL, Stanfield PH, Becker WC, Hillers JK. The effects of restraint stress in early pregnancy in mice. J Reprod Fenil 1986; 78: 185- 192. Wildt DE. Chÿkriborty PK. Pmko. Seûger WB. Relationship of reproductive behaviour. serum luteinizing hormone and time of ovulation in the bitch. Biol Reprod 1978: 18: 56 1-570.

Wildt DE, Howard JG, Chakraborty PK, Bush M. Reproductive phy siology of the clouded leopard. II. A circannual anal ysis of adrenal -pi tui tary-testicular relationshi ps during electroej aculation or after an adrenocorticotropin hormone challenge. Biol Reprod 1986; 34: 949-959. Wildt DE. kvinson CJ. Seager SWJ. Lapüroscopic zxposure and sequential observation of the ovuy of the cycling bitch. Anat Bi01 1977: 189: 443-450.

Wildt DE, Meltzer D, Chakraborty PK, Bush M. Adrenal-testicular-pituitary relationships in the cheetah subjected to anesthesia/electroejaculation. Bi01 Reprod 1984; 30: 665-672. Wildt DE. Panko WB. Chakraborty P, Seager SW. Relationship of se~umestrone. esrradiol- 17P and progesterone to LH, sexual behaviour and time of ovulation in the bitch. Biol Reprod 1979; 20: 648-658. Wildt DE. Seager SWJ, Chakraborty PK. Behaviorai. ovuian and endocrine relationships in the pubeml bitch. J Anim Sci 198 1: 53: 182- 19 1.

Wilson EO. Conservation: The next hundred yem. In: Western D. Pearl M (eds.) Conservation for the twenty-first century. New York: Oxford University Press. 1989: 3-7. Wilson MS. Non-surgical inmutenne artifical insemi nation in bitches using frozen semen. J Reprod Fertil Suppl 1993; 47: 307-3 1 1. Winter M. Pirmann J, Falvo RE, Schanbricher BD. Miller J. Steroid control of gonadotrophin secretion in the orchidectomized dos. J Reprod Fertil 1982: 64: 449- 455.

Worgul Tl. Santen RJ. Smojlik E. Evidence that brin uomritization regulsition LU secretion in the male dog. Am J Physiol 198 1 : 241 : E241-E250. Wright PJ. Stelrnüsiük T. Blxk D, Sykes D. Medroxyprogsstemne acetate and reproductive processes in male dogs. Aust Vet 1 1979: 55: 437-438.

Yonjun X. Shuxilin J. Yuren T. Jimhua S. and Guosen W. Seasonal changes in testiculÿi. size and .semm testosterone levels and the relationship between testosterone concentrations and reproductive performmce in male raccoon dogs. Scientifur 1994: 18: 39-43.

Young KM. Reproductive biology of the black-fwted ferret (Mt~~tei;am. Guelph. Ontario. Canada: University of Guelph; 1998, 1 19 p. MSc Thesis.

Zimen E. On the reguültion of pack size in wolves. Z Tierpsychol 19%: 40: 300-34 1. Zondek B. Tamari 1. Effeçü of ûudigenic stimulation on genital function iuid reproduction. Am J Obstet Gynecol 1960; 80: 1041- 1048. APPENDIX LA

CHEMICALS

Caledon Laboritories Ltd.. Georgetown. Ont.. Crinada ABTS 2.2' - Azino-bis Fisher Scientitic. Uniunville Ont.. Canada (3-ethylbenzthirizoline-6-suIfonicrtcid) Aluminum oxide powder BDH Ltd.. Burlington. Ont.. Canada

Bovine senim albumin (Fraction V) Sigma Chernical Gr.. St. Louis. MO. USA Citric rrcid Fisher Scientitic. Unionville Ont.. Crinadri

Cortisol antisenim (R4866) C.J. Munro. University of Calitornia. Davis. USA Cortisoihorsemdish peroxidase C.J. Munro. University of Californiri, Davis. USA Diethyl ether Cdedon Labontories Ltd,. Georgetown. Ont. Canada EDTA Sigma Chernical Co.. St. Louis. MO. USA

Estradiol- 17B antiserum (R4972) C.J. Munro. University of Calitornia. Davis. USA

Estradiol- 17P standard ( 1.3.5 [ 1O]- C.J. Munro. University of California. Davis. USA Estratriene-3.17gdiol)

Estradiol- I7P /horseradish peroxidase C.J. Munro. University of Californiri. Davis. USA

Hydrocortisone standard ( 1 1p, 17a, 2 1- Sigma Chernical Co.. St. Louis. MO. USA Tri hydroxypregn-4-ene-3.20-dione) Hydrogen peroxide Fisher Scientitïc. Unionville, Ont.. Canada Lutenizing hormone antiserum (5 18-B7) J-Brown. National Zoologiçal Park Conservation Research Center, Front Royal VA. USA

Lutenizing hormone ovine standard J.Brown. National Zoologid Park Conservation Research Center. Front Royal. VA. USA J-Bniwn. National Zoological Park Con.sei-vation Re.serirch Centec. Front Royal. VA. USA Methmol BDH Inc.. Toronto. Ont.. Canada Normal Mou.% Serum Sigma Chernical Co.. St. Louis. MO. USA Phenol red Sigma Chernicd Co.. St. Louis. MO. USA

Pol yethylglyctil (PEG) Fisher Scientitic. Neperin. Ont.. Canada

Progesterone antiseium (R486 1 ) C.J. Munro. University of California. Davis. USA C.J. Munro, University of Crriihrnia- Davis. USA Progesterone standrird (4-pregnen-3.20- Sigma Chemical Co.. St. Louis. MO. USA dione) Sodium carbonate Fisher Scientitic. Nepeün. Ont.. Canada Sodium chloride Fisher Scientitic. Nepean. Ont.. Canada Sodium hydrogen carbonate Fisher Scientit'ic. Nepean. Ont.. Canada Sodium phvsp hrite monobasic BDH Ltd.. Burlington. Ont., Canada Sodium phosphate dibasic BDH Ltd.. Burlington. Ont.. Canada

Testosterone rintisenim (R 15617) C.J. Munro. University of California. Davis. USA Testosterone/horseradish peroxidase C.J. Munro. University of Californiri. Davis. USA

Testosterone standard ( 17B-hydroxy-4- Steraloids. New Hampshire. USA androsten-3-one) Thimerosd Sigma Chemical Co.. St. Louis. MO. USA Tween 20 Sigma Chemicrrl Co.. St. Louis. MO. USA APPENDIX 1.B

EQUIPMENT

Centrifuge. Labofuge 400 VWiUCaiilab. Mississriuga, Ont., Clmada

Dynatech Ultr~wrisliII Micrtjpltite Wrtsher VWR/Canlab. Mississauga, Ont.. Canada Dytiatech 700 Plate Reader VWWCrinlab. Mississriuga. Ont., Canada Dynex MRX Plate Reade1- Dynes Tecnchgies. Chantilly. VA. USA Fisher Plritelet Miser Moclel3-W Fisher- Scientitk Co.. Ltd.. Guelpli. Ont.. Canada LKB RediFrx friction collector LKB Wal!ric. Turku. Finland

Nunc-Irnmuno 96F Muisvrp plates Mandel Scientitic Co.. Ltd.. Guelph. Ont.. Canada

Gamma Counter Isodata ICN Phannaceutids. Costa Mesta. CA. USA.

Gilson P200 Diamond tips Mandel Scientitic Co.. Ltd.. Guelph. Ont.. Canada Radial compression column Waters Scientitic. Mississaugsi. Ont. finada

Rotator Amencan Instruments R4 140 VWRKanlab. Mississriugri. O~L.Cmrida Sep-Pak C,,carti-idges Waters Scien titic. Mississauga. Ont-. Canada

Waterbath 180 Series Lab-Equip. Markham. Ont.. Canada APPENDIX 1.C ENZYME IMMUNOASSAY SOLUTIONS

Coating Buffer Na-CO, NaHCO, distilled H,O (dH1O) EIA Buffer Stock A: 0.2M NaH,PO, Stock B: 0.2M N+HPO,

Stock A Stock B dH,O ~a~1 BSA (Fraction V) Wash Solution Concentrate NaCl Tween 20 dH,O Substrate Buffer Citrate buffer Citiic Acid dH,O 40 mM ABTS AB-3s dH,O 0.5M H20, H,O, (30%) dH,O Fecal Extraction Solution Wash conc 10 X NaCl Tween 20

20% Methanol wash wash conc methanol

Stock A: NaH,PO, 27.8 @1000 ml 20% methmol wash Stock B: N-HPO, 28.4 g/ 1000 ml 20%methanol wuh

Extraction bufier Stock A Stock B NaCl BSA 20% Methanol wash APPENDIX 1.D

RADIOIMMUNOASSA Y SOLUTIONS Pink PBS 0.5 M monobasic NaP04 L9.54 ml 0.5 M dibasic NriP04 83.3 1 ml NaCI 40.85 3 Thimerosd 0.5 g Phenol red a few drops distilled H20 (dH.0) 5 litres pH to 7.4 RIA Buffer Pink PBS EDTA (non free =id) BSA (fraction V) APPENDIX I1.A EIA ANTIBODY CROSS-REACTIVITIES PROCESTERONE (R4861) progesterone 100 % 1 la-hydroxyprogesterone Sa-pregnane-3.20-dione 12.19 5% l7a- hydroxyprogesterone 20a-hydroxyprogesterane 0.13 % 20p-hydroxyproge~~terone pregnaned iol < 0.0 1 % pregnenolone estradiol- 17p < 0.01 $6 estrone testosterone c 0.0 1 % cortisol ESTRADIOL (R4972) estradiol- 17P 100 % estrone progesterone 0.8 % testosterone androstenedione 1.0 $2 cortisol dihydrotestosterone < 1.0 % TESTOSTERONE (R156/7) testosterone LOO % Sa-dihydrotestosterone androstenedione 0.27 % androsterone dehydroepiandrosterone 0.04 % cholesterol p-es trad io 1 0.02 % progesterone pregpenolone c 0.02 % hydrocortisone chohc acid < 0.02 % chenodeoxycholic acid cholic acid methyl ester < 0.02 96 dehydrocholic acid deoxycholic acid < 0.02 % Lithocholic acid glycholic acid < 0.02 % triurodeoxycholic acid triurochenodeoxycholic < 0.02 % glycochenodeoxycholic acid acid taurocholic acid CORTISOL (R4972) cortisol prednisolone prednisone cortisone corticosterone deoxy corticosterone 2 1-deoxycortisone 1 1-deoxycortisol progesterone 17a-hydroxypro,oesterone pregnenolone 17a-h ydroxypregnenolone androstenedione testosterone androsterone dehydroepiandrosterone aldosterone estndiol- 17P estrOne esmol dehydroisoandrosterone- spuonolactone 3-suifate c holester01 APPENDIX 1I.B RIA ANTIBODY CROSS-REACTIVITIES

LUTEINIZING HORMONE ANTIBODY (518-87)

equine LH equine CG ovine LH bovine LH porcine LH fetine LH canine LH mbbit LH nt LH equine FSH equine TSH ovine FSH ovine prolactin bovine FSH bovine TSH bovine growth hormone Fecal progestin and estrogen metabolite concentrations from a yearling îèmaie red wolf (A) and fecal testosterone metabolite concentmtions in relation to daylength from a yearling male red wolf (B).