Conservation genetics of grizzly bears by Frank Lance Craighead A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biological Sciences Montana State University © Copyright by Frank Lance Craighead (1994) Abstract: Grizzly bears have some behavioral characteristics that should tend to reduce the amount of genetic variation passed on from generation to generation: females tend to establish home ranges adjacent to their mother and not all males breed. Some males, however, travel widely and breed with several females. In order to examine the genetics of a virtually undisturbed grizzly bear population in the Alaskan Arctic, to determine basic population genetics parameters, and to answer questions of paternity, reproductive success, and genetic population subdivision, I used two DNA 'fingerprinting’ techniques. I report data from analyses of multi-locus minisatellite polymorphisms and single-locus microsatellite loci from 152 grizzly bears (including 30 grizzly bear family groups) in the primary study area. I compare these data with smaller samples from 3 other areas. These analyses were made possible by the use of single-locus primers which amplified both of an individual's alleles at 8 loci, and by detailed knowledge of matemal/offspring relationships which allowed the identification of paternal alleles. The alleles examined are shown to be selectively neutral, and distributed in Hardy-Weinberg proportions. The data demonstrate that each cub in a litter can be sired independently and that one third of all possible litters had multiple sires. Estimates of maximum reproductive success for males indicate that no single male is responsible for more than 11%-13% of total paternity. No more than half of breeding-age males successfully bred. Examination of genotype frequencies, genetic structure and effective population size showed no evidence of genetic structure within any of the populations and no significant difference in heterozygosity between any populations. The data indicate that high levels of heterozygosity (75%) and gene flow throughout grizzly bear range is maintained by the male segment of the population, and they contribute to an understanding of the genetic and demographic basis of male reproductive success which is of vital importance in the maintenance of small, isolated grizzly bear populations.  CONSERVATION GENETICS OF GRIZZLY BEARS by Frank Lance Craighead A thesis subm itted in partia l fulfillm ent of the requirem ents for the degree of Doctor of Philosophy in Biological Sciences MONTANA STATE UNIVERSITY Bozeman, M ontana March 1994 © COPYRIGHT by Frank Lance Craighead 1994 AU Rights Reserved 11 C%4+ APPROVAL of a thesis subm itted by Frank Lance Craighead This thesis has been read by each m em ber of the thesis com mittee and has been found to be satisfactory regarding content, English usage, format,citations, bibliographic style, and consistency, and is ready for submission to the College of G raduate Studies. ________________________ Date Co-Chairperson, Graduate Committee 9 ^ / f y Date Co-Chairperson, Graduate Committee Date Approved for the Major D epartm ent Head, Major Departm ent Approved for the College of Graduate Studies 2 ^ 7 Date Graduate Dean Ill STATEMENT OF PERMISSION TO USE In p resen ting this thesis in partia l fulfillm ent of the requ irem en ts fo r a doctoral degree a t M ontana State University, I agree th a t th e Library shall make it available to borrow ers u n d e r the ru les of th e Library. I fu rth e r agree th a t copying of th is thesis is allowable only for scholarly purposes, consistent w ith "fair use" as p rescribed in the U.S. Copyright Law. Requests for extensive copying o r rep roduction of this thesis should be referred to University Microfilms International, 300 North Zeeb Road, Ann Arbor, Michigan 48106, to whom I have gran ted "the exclusive righ t to rep roduce and d istribu te m y dissertation for sale in and from m icroform or electronic form at, along w ith the righ t to reproduce and d istribu te m y abstrac t in whole o r in part." Signature Date f - 2 / - f f ACKNOWLEDGMENTS I would like to begin by thanking m y fa the r Frank C. Craighead Jr. and m y uncle John J. Craighead for introducing m e to grizzly bears, and the rest of nature , and teaching me to enjoy the w ilderness. I w ould like to rem em ber m y g randfather, Frank C. Craighead Sr. for encouraging m e to seek this degree before I h ad even considered it, and Joseph Hickey for giving m ore encouragem ent and starting me down this ra th e r serendip itous academic road. I w ould like to thank the m em bers of m y g raduate com m ittee fo r all th e ir help and support and for sharing in terests of the ir own: Em ie Vyse fo r genetics, Pete Brussard fo r biogeography, Dave Cameron for population genetics, Dan Goodman for population dynam ics, and Mike Gilpin for tying these disciplines together and showing me th a t it can be enjoyable too. I especially w ant to thank Ernie Vyse for his enthusiasm and patience. I w ould like to thank everyone who helped to do th e w ork and m ake things possible; especially Harry Reynolds, David Paetkau, Curt Strobeck, Steve Fain, Layne Adams, Ron Warbelow and Jim Rood. Financial (and moral) support was provided by the Alaska Dept, of Fish and Game; U.S. National Park Service, Alaska; Patagonia Inc. (Yvon and M ahnda Chouinard); The Eppley Foundation for Research; The Wiancko Foundation (Tom and Sybil Wiancko); The Gamble Foundation (George and Launce Gamble); Fred and Barbara Hudoff; The Hudepohl-Schoenling Brewery; the Canadian National Science and Environmental Research Council (NSERC); Parks Canada; and the A lberta W ildlife Service. I am grateful for the facilities and suppo rt of th e University of Alberta, M ontana State University, and the Craighead Environmental Research Institute. I w ould like to acknowledge the kinship (bo th genetic and spiritual) of m y siblings and cousins (some of whom are half-sibs) who have grown w ith me and share a sim ilar outlook on life. Finally I w ould like to acknowledge the m aternal half of m y pedigree (genetic a n d /o r philosophical) who gave me m y mtDNA, an d /o r n u rtu red m e and encouraged me and helped me; Carolyn Craighead, Esther Craighead, M argaret Craighead, Shirley Craighead, Jean George, Ruth Stevens, M ardy Murie, and m ost im portantly , m y wife April. TABLE OF CONTENTS INTRODUCTION............................................................................................... I BACKGROUND..................................................................................................8 Bear phytogeny and systematics.......................................................8 Paleontological basis of bear phytogeny............................. 8 Genetic evidence of bear phytogeny............ .................... 10 Interspecies studies of bear genetics............................................14 Related wildlife genetic s tud ies .......................................... 15 Genetic variation among individual bears........................22 THE STUDY POPULATION.............................................................................24 The study area....................................................................................24 Demography............................. 27 HYPOTHESES.................................................................................................. 30 METHODS....................................................................................................... 32 Field techniques................................................................................. 32 DNA extraction from blood..............................................................33 DNA extraction from tissue........................................ 37 DNA fingerprinting with multi-locus probes.............................. 38 DNA restriction and separation........................................... 39 Southern transfer....................................................................40 Hybridization with radiolabelled p ro b e s ........................ 42 Signal detection...........................................................43 Hybridization with chemiluminescent p robes.............. 43 Signal detection............................................................45 Interpretation of multilocus DNA fingerprints.............. 45 DNA analysis of microsatellite loci.................................................47 Development of primer se ts ................................................48 Amplification of target DNA using PCR............................49 Electrophoresis of PCR products..........................................5 0 Data collection.............................. ......................................... 5 1 Data analysis............................................................................ 52 Page TABLE OF CONTENTS (Continued) RESULTS AND DISCUSSION....................... :................................................53 DNA extraction................................................................................... 53 DNA restriction and separation ......................................................54 DNA fingerprinting of minisatellite lo c i..................................... ..54 33.15 p ro b e .............................................................................55 33.6 probe........................ 58 CMMlOl and MSI p robes .......................... 59 Equiladder p ro b e ............................. 59 Interpretation of multilocus DNA fingerprints............. .60 DNA analysis of microsatellite loci................................................. 63 M utation......................................................... 64 Null alleles...................... 66 Individuals.............................................................................. 67 Shared alleles...............................................................67 Paternity....................................................................... 68 Multiple paternity.............................................69 .Probability..........................................................73 Hypothetical males........................................... 74 Pedigrees..................................................... 76 Male reproductive success................................. 113 Western Brooks Range popu la tion ................................... 118 Allele frequencies......................................................118 Genotype frequencies...............................................118 Population subdivision..............................................120 Estimates of N e ............................................................ 121 Variance in progeny n u m b e r ..................... 124 Unequal breeding sex r a t io ......................... 126 Neighborhood size ......................................... 129 Formula variations..........................................130 Effective num ber of neutral a lle les......................132 Comparisons between genera tions.............. 134 Interpopulation com parisons................. 137 Allele frequencies......................................................139 Measures of genetic d ifferen tia tion .....................140 Genotype frequencies and heterozygosity ...... 145 Selective neutrality .of a lleles.................................147 vi Page TABLE OF CONTENTS (Continued) LEVELS OF INQUIRY.................................................................................... 148 Molecular level........................................................................ 150 Organism level.................................................................................151 Male reproductive stra tegy ................................................151 Population level...............................................................................154 Synchrony of recru itm ent..................................................155 Species level..................................................................................... 156 SUMMARY..................................................................................................... 158 LITERATURE CITED..................................................................................... 162 APPENDICES................................................................... 177 Appendix A Individual bears and genotypes......................... 178 Sex/age relationships of WBR bears.................................179 WBR individual alleles......................................................... 183 ANWR individual alleles.............................................. 186 AKR individual alleles.... ......................................................186 NCDE individual alleles...................... 186 Appendix B Allele frequencies..................................... 187 WBR allele frequencies............................................. 188 ANWR allele frequencies..................................................... 190 AKR allele frequencies............................................. 192 NCDE allele frequencies.......................................................194 Allele frequencies for all populations..............................196 Appendix C Genotype frequencies and analyses..................198 Locus A genotype frequencies.................... 199 Locus B genotype frequencies........................................... 200 Locus C genotype frequencies........................................... 201 Locus D genotype frequencies.......................... 202 Locus L genotype frequencies............................................203 Locus M genotype frequencies ......................................... 204 Locus P genotype frequencies............................................205 Locus X genotype frequencies............................................206 vii Page V lll Page Appendix C (Continued) WBR Hardy-Weinberg equilib rium ...................................207 WBR heterozygosity (observed versus expected).......212 ANWR heterozygosity (observed versus expected) ..214 AKR heterozygosity (observed versus expected)........216 NCDE heterozygosity (observed versus e j e c t e d ) .....218 Appendix D Additional analyses................................ 220 Fct over all populations (expected) ....... 221 Fst over all populations (observed) ................................ 223 WBR two generation comparisons ................................... 225 Variance and covariance of known progeny numbers.................................................................. 227 TABLE OF CONTENTS (Continued) LIST OF TABLES Table Page 1. Shared 33.15 bands among three family groups...................... 56 2. Population genetic param eter estim ates from a single multi-locus probe (33.15)................................ .....58 3. Population genetic param eter estimates from multi-locus probe combinations...................................................... 62 4. Family genotypes for 1097's family...............................................72 5. Family genotypes for 1439's family.............................................. 73 6. Deduced genotype for hypothetical male no. I .......................75 7. Deduced genotypes of hypothetical males.............................. ,...76 8. Relative reproductive success of known fathers.................... 114 9. Relative reproductive success of the m inim um possible num ber of fathers............................................................. 115 10. Effective and actual alleles at each lo cu s ................................. 133 11. Combinations of homozygosity and ne with Ne and m utation rate. (After Kimura and Crow 1964)........................ 134 12. Mean allele frequencies between generations.... .....................136 13. Allele frequency divergence, Fst among d isparate grizzly bear subpopulations. (most common WBR alleles)......................................................... 141 14. Allele frequency divergence, Fst among d isparate grizzly bear subpopulations. (most common allele at each locus) .................. ........................142 15. M ean heterozygosity and Fst over 8 loci over all populations................................................................................. 144 16. Mean heterozygosity among disparate populations............................................. .146 17. Sex/age relationships of WBR bears .......................................... 179 18. WBR individual alleles....................................................................183 19. ANWR individual alleles .............. 186 20. AKR individual alleles.................................................................... 186 21. NCDE individual alleles .................................................................186 22. WBR allele frequencies...................................................................188 2 3. ANWR allele frequencies .............................................................. 190 24. AKR allele frequencies....................................................................192 25. NCDE allele frequencies ................................. 194 26. Allele frequencies for all populations........................................ 196 ix XLIST OF TABLES (Continued) Table Page 27. Locus A genotype frequencies.....................................................199 28. Locus B genotype frequencies......................................................200 29. Locus C genotype frequencies........................................................201 30. Locus D genotype frequencies..................................................... 202 31. Locus L genotype frequencies......................................................203 32. Locus M genotype frequencies......................................................204 33. Locus P genotype frequencies..................................................... 205. 34. Locus X genotype frequencies.................................................. 206 35. WBR Hardy-Weinberg equilibrium ......................................... ...207 36. WBR heterozygosity (observed versus expected).................. 212 37. ANWR heterozygosity (observed versus expec ted )............214 38. AKR heterozygosity (observed versus expected)................... 216 39. NCDE heterozygosity (observed versus expec ted )................218 40. Fst over all populations (expected)............................................ 221 41. Fgr over all populations (observed).......................................... 223 42. WBR two generation com parisons.............................................225 43. Variance and covariance of known progeny num bers.......227 LIST OF FIGURES 1. The study area.................................................................................... 25 2. Pedigrees of four grizzly bear fam ily groups exhibiting multiple paternity.......................................................... 71 3. Family 1087 ....................................................................................... 78 4. 1087's extended family......................................................................78 5. Family 1089........................................................................................ 79 6. Family 1095 ................... ;.................................................................. 80 7. Family 1097 ...................................................................................... .81 8. Family 1125 ................................................................... 83 9. Family 1136 ....................................................................................... 85 10. 1136’s extended family................................................................ 85 11. Family 1141 ......................................................................... 86 12. 1141's extended family......................................................................87 13. Family 1149 ....................................................................................... 88 14. Family 1166 ................................................................ 89 15. Family 1174 ........................................................................................ 90 16. 1174's extended family...................................................................... 90 17. Family 1177 ........................................................................................ 91 18. 1177's extended family.......................................... 91 19. Family 1179 ........................................................................................ 92 20. 1179's extended family...................................................................... 93 21. Family 1424 ........................................................................................ 94 22. Family 1425 ........................................................................................ 95 23. Family 1437 ........................................................................................ 96 24. Family 1438 ............................................................. 97 25. Family 1439 .................................................... 98 26. Family 1440 .......................................................................................100 27. Family 1454 .......................................................................................101 28. Family 1457 ................................................................... 102 29. Family 1458 ............................................. 103 30. Family 1460 .......................................................................................104 31. Family 1461 .......................................................................................105 32. Family 1464 .......................................................................................106 33. Family 1479 .......................................................................................107 34. Family 1716 .......................................................................................108 Figure Page LIST OF FIGURES (Continued) Figure Page 35. Family 1734 ................................................ 109 36. Family 1739 .......................................................................................HO 37. Family 17 4 5 .......................................................................................I l l 38. Family 1749 .......................................................................................112 39. Mean heterozygosity among disparate populations .............146 40. An approxim ate phenetic diagram of the U rsid s................... 157 Xlll ABSTRACT Grizzly bears have some behavioral characteristics th a t should tend to reduce the am ount of genetic variation passed on from generation to generation: females tend to establish hom e ranges ad jacen t to the ir m other and no t all males breed. Some males, how ever, travel w idely and breed w ith several females. In o rd e r to exam ine the genetics of a virtually und istu rbed grizzly bear popu la tion in the Alaskan Arctic, to determ ine basic population genetics param eters, and to answer questions of patern ity , reproductive success, and genetic population subdivision, I used two DNA 'fingerprin ting’ techniques. I repo rt data from analyses of m ulti-locus m inisatellite polymorphism s and single-locus m icrosatellite loci from 152 grizzly bears (including 30 grizzly bear fam ily groups) in the prim ary study area. I com pare these d a ta w ith sm aller samples from 3 o ther areas. These analyses were m ade possible by the use of single-locus prim ers which am plified bo th of an ind iv idual's alleles a t 8 loci, and by detailed knowledge of m atem al/o ffsp ring relationships which allowed the identification of p a te rn a l alleles. The alleles exam ined are shown to be selectively neu tra l, and d istribu ted in Hardy-Weinberg proportions. The d a ta dem onstrate th a t each cub in a litter can be sired independen tly and th a t one th ird of all possible litters h ad m ultiple sires. Estimates of maximum reproductive success fo r males ind icate th a t no single male is responsible for m ore th an 11%-13% of to ta l pa tern ity . No m ore than half of breeding-age males successfully bred. Examination of genotype frequencies, genetic s truc tu re and effective population size showed no evidence of genetic structu re w ithin any of the populations and no significant difference in heterozygosity between any populations. The da ta ind icate th a t h igh levels of heterozygosity (75%) and gene flow th roughou t grizzly bear range is m aintained by th e male segm ent of the population, and they contribute to an understanding of the genetic and dem ographic basis of male reproductive success which is of v ital im portance in the m aintenance of small, isolated grizzly bear populations. IINTRODUCTION A round the time of the m ost recent, Wisconsin, glaciation of the Pleistocene, over 12,000 years before p resen t (ybp), Homo sapiens and Ursus arctos began a long journey together. In uneasy proxim ity to each o th e r they crossed the Bering Land Bridge from Asia and found g reat expanses of hab ita t available. Both species s ta rted to popu la te the New W orld w ith the ir offspring. Homo sapiens has p roven to be m uch better adap ted to this, and in the ensuing centuries, due to h igher reproductive rates, greater d ispersal ability, d irec t com petition for space, w idespread modification of habitat, and superio r firepower; has reduced Ursus arctos to rem nan t populations in m ountainous hab ita t over m uch of its form er range. There is . m ore h ab ita t available for man, including th a t occupied by the bear. The carrying capacity of the environm ent is m uch g reater fo r m an th an fo r the bear. In fact, m an has no t yet reached the carrying capacity of his environm ent in North America; hum an populations continue to increase, and m an continues to com pete w ith the grizzly b ea r fo r space. Man converts grizzly bear hab ita t in to hum an h ab ita t and the bea r has been unable, in over 15,000 years, to adap t its behav ior enough to live peacefully in hum an habitat. If the bear is to survive as a species, it will be because m an learns to live peacefully, o r a t least tolerantly, w ith the bear. 2To begin with, the bear was a t a disadvantage. Both species com peted for some of the same p lan t and animal foods, and occupied sim ilar hab itat. Although, physically the grizzly is m uch m ore form idable, early encounters in Beringia with bands of m en arm ed w ith spears, knives, and clubs, were p robably a standoff. Man was able to p ro tec t him self from the grizzly, precariously, and the grizzly, presum ably, h ad m ore attractive prey. The grizzly (and man) also faced com petition to some degree from ano ther bear: Arctodus sim us. th e g reat short-faced bear. An ancestor of A rctodus h ad m igrated in to the New World during the Pliocene and had rad ia ted in to bo th North and South America. When the grizzly arrived it h ad to com pete d irectly w ith the great short faced bear fo r some resources, and was probably even preyed upon. The short-faced b ea r was in decline a t this time and w ent extinct along w ith m ost of its p rey species in the g reat wave of Pleistocene extinctions; perhaps nudged along this p a th by the grizzly and man. As the ice sheets receded, hum an populations expanded rap id ly in to N orth America and continued across the Isthm us of Panam a in to South America. The grizzly expanded its range a t a m uch slower rate , replacing the short-faced bear and eventually inhabiting m ost of North America from the eastern seaboard to cen tral Mexico. By this time m an had thoroughly occupied bo th continents and a second wave of m igration, across th e oceans in boats to the New World, h ad begun. The grizzly h ad reached its g reatest a rea of d ispersal by about the 1500's. From th a t po in t on grizzly num bers have been declining, particu larly since the 1600's 3w hen European m en w ith firearm s arrived in North America and began rem oving the grizzly, directly and indirectly, from the p e riphe ry of its range. By the 1700’s m an had greatly reduced the grizzly (or brown bear) in w estern Europe and populations were declining in eastern Europe. There was no longer an uneasy truce: th e grizzly was on the run. This m an-caused decline in grizzly popula tions accelerated in the 1800's w ith g reater availability of firearm s and poisons and has continued to the p resen t (Cowan 1972, Servheen 1990). Before the adven t of firearms, m en and grizzlies were to lerant, respectful, and wary of one another. Indigenous m en studied the b ea r and learned its ways in o rder to p ro tec t themselves, and inco rpo ra ted this knowledge in to rituals in sim ilar ways in all cu ltures w here hum an populations and grizzly populations were sym patric. Grizzlies were revered, and feared, and were symbolic of im po rtan t sp iritual values (Hallowell 1926, Clark 1966, Rockwell 1991). Scientific study of the grizzly bear in North America began in the 1920’s w hen Adolph Murie and his b ro ther Olaus began collecting d a ta in M ount McKinley National Park, Alaska. Early d a ta were reco rded incidentally to studies of wolves and ungulates, b u t afte r th e m id-fifties Adolf concentrated on studying grizzlies (Murie, 1981). In 1952 Frank and John Craighead, began an ecological study of the grizzly population in and around Yellowstone National Park (Craighead 1976, 1979; Craighead and Craighead 1963, 1965, 1972; Craighead et. al. 1974, 1976). This study continued un til 1969 and 4in troduced the use of rad io telem etry in wildlife studies. Grizzly bea r studies have continued in and around M ount McKinley and Yellowstone Parks and sim ilar research, prim arily ecological, dem ographic and behavioral studies, have focused on v irtually all ex tan t grizzly populations in North America. A recen t com pendium of grizzly bea r literature, edited in 1987 (Interagency Grizzly Bear Committee 1987a) lists 1,284 entries concerning various aspects of grizzly bear research. The longest continual study of grizzly bear ecology and dem ographics to date was begun by H any Reynolds w ith the Alaska D epartm ent of Fish and Game in 1977, on a study a rea in the n o rth e rn foothills of the Brooks Range in northw estern Alaska. Grizzhes have probably inhabited this area continuously since they firs t a rrived over 12,000 years ago. Since 1980 I have worked on various facets of th is study during p a rts of six field seasons. As w ith m ost research there have been m any m ore questions than answers. With the adven t of DNA fingerprin ting and its application to wildlife studies we saw the possibility of examining genetic com ponents of th is well-studied population . Among the questions we hoped to be able to resolve were th e genetic effective size of the population, estim ates of popu la tion genetics param eters, p a tern ity of cubs w ith known m others, and the relative reproductive success of all males thus identified. Since 1988, w ith the help and advice of Dr. Emie Vyse, b lood and tissue samples have been collected from this population . In i. 1990 I w orked in the field and then began the labo ra to ry phase of th e study in Dr. Vyse's laboratory. In 1991 and 1992 bo th Dr. Vyse and I p artic ip a ted in field work and continued the labo ra to ry work. I v isited the U. S. National Fish and Wildlife Forensic Laboratory in Ashland, Oregon in September of 1991 and learned p rocedures for using chem ilum inescent oligonucleotide probes. This technique resu lts in da ta consisting of band patterns from DNA fragm ents of la rge leng th (2 to 20 kilobases) from variable-num ber-of-tandem - rep ea t (VNTR) loci. This is DNA fingerprinting in the classic sense as described by Jeffreys et. al.( 1985a,b). This m ethod and its results will be re fe rred to as fingerp rin ting th roughou t th e text. A fter th ree years of labora to ry work, using the best available m olecu lar techniques suited to our goals a t the tim e we began, we were still unab le to answer some of ou r questions; b u t we had established a genetic baseline for grizzly bear populations using m ultilocus probes. During the sum m er of 1993 we learned of add itional work using m icrosatellite analysis of black bea r DNA by colleagues a t the University of Alberta. In Septem ber of 1993 I w ent to Edm onton to work w ith a technique developed by David Paetkau w ith Dr. Curt Strobeck using Polymerase Chain Reaction (PCR) am plification of single m icrosatellite loci. This technique provided accurate data on bo th alleles a t 8 loci fo r v irtua lly all of ou r genetic samples. I finally was able to resolve sufficient varia tion in m y genetic profiles to determ ine patern ity , distinguish individuals, establish pedigrees, and estim ate population genetic param eters. This m ethod and its results will be referred to as 5 6m icro sa te llite analysis th roughou t the text. A dditional w ork by ourselves and others, using these techniques and new er advances, will eventually extend the pedigrees of this population , and fill in the details of the p ic tu re th a t we have begun painting. We are adding ano ther level of inqu iry into our knowledge of th e grizzly bear. At th e tim e of this writing, la te in the tw entieth century , an ecological po in t of view is central to the life sciences and has even begun to sp read to o ther areas of hum an activity such as politics and commerce. Man has begun once again to be concerned abou t the effects of his activities upon o ther species. In N orth America in particu lar, laws have been enacted lim iting the harvest of game anim als, and w hen this proved to be insufficient, fu rth e r legislation has a ttem p ted to ensure th a t species are no t driven to extinction th rough m an 's activities. Grizzlies were listed as th rea tened in the conterm inous United States in 1975 and became subject to federal p ro tec tion and managem ent. Conservation of th rea tened populations a re supervised by the Interagency Grizzly Bear Committee (IGBC) based on research conducted by the Interagency Grizzly Bear S tudy Team (IGBST) as d irected by the Grizzly Bear Recovery Plan and the Interagency Grizzly Bear Guidelines (U.S. Fish and Wildlife Service 1982 and 1993, Interagency Grizzly Bear Committee 1987b, Peek et. al. 1987, Servheen 1990). The grizzly is p ro tected by law from becom ing extinct and there is once again an uneasy truce. The truce is uneasy now because it m ay no t be possible for m an to p reserve the grizzly; despite his best efforts. This dissertation, and the research it reports, are a p a rt of the effort to try and preserve Ursus 7arctos, and hopefully a host of o ther species as well, against the continuing m om entum of hum an demographics and its concom itant h ab ita t alteration. The te rm Conservation Genetics in the title of this d issertation is u sed in the sense th a t genetics da ta are the subject of this research, and they are analyzed in the context of populations; b u t the focus is narrow ed to those aspects of genetics which m ay be m ost usefu l in term s of th e conservation of genetic diversity. As fa r as I know, th e firs t use of this te rm was by Wayne et. al. (1991); Conservation genetics of the endangered Isle Royale gray wolf. The term s brown bear and grizzly bear can be considered to be in terchangeable, b u t I have tried to consistently use grizzly in reference to North American Arctic and in terio r populations, and brow n bea r in reference to coastal and Old World populations. The grizzly popu la tion we have studied is large, wild, v irtually unhun ted , and has persisted since the Wisconsin glaciations: it is a viable population . These da ta are offered as a genetic baseline, describing a viable population, for comparison with o ther grizzly populations th roughou t the world. 8BACKGROUND Bear Dhvlosenv and svstematics There are only eight extant species of bears worldwide. Two lines of evidence for bear phytogeny are reviewed: paleontological and m olecular. Both sets of data are roughly concordant and in general agreem ent as to the tim ing of speciation events. Paleontological basis of bear phytogeny The genus Ursus arose in the Old World, probably in Eurasia, in th e early Pliocene; probably derived from the Holarctic Miocene genus Ursavus (Kurten and Anderson, 1980). During the early M iocene in Asia, abou t 22 mybp, the ancestral U rsidae split in to two subfam ilies, th e A iluropodinae (represented by one ex tan t species th e G iant Panda, A iluropoda m elanoleuca) and the Ursinae. Later in th e Miocene, the Ursinae rad ia ted in to two groups; the T rem arctinae and the Ursinae. The Trem arctinae probably evolved by vicariance as a resu lt of crossing the Bering Land bridge in to th e New W orld and becoming isolated, probably about 15 mybp. Two genera, T rem arctos and A rctodus evolved in the New World, eventually becom ing extinct in North America. Tremarctos survives in South America as the spectacled bear, Tremarctos om atu s. 9During the Pliocene, about 3.5 mybp, the first ancestral Ursine bea r m igrated in to the New World. The fossil reco rd indicates th a t i t evolved in to the American black bear, Ursus am ericanus. and was a separate species by a t least 1.5 to 2.5 m ybp (Kurten 1964, 1968, 1976. Kurten and Anderson 1980), establishing itself sou th of the ice sheets during the Pleistocene glaciations. It shared this a rea w ith th ree species of Tremarctinae; Tremarctos floridanus. Arctodus p ris tinu s. and Arctodus simus. U. am ericanus is the m ost com monly found Pleistocene bear species among the fossils from this period. A. simus. th e g iant short-faced bear was th e m ost w ide-ranging bea r species during the Pleistocene in North America, and was p resen t in Alaska and the Yukon during the late Pleistocene. The no rth e rn form s were very large and it was the m ost powerful p red a to r of the Pleistocene fauna of North America (Kurten and Anderson, 1980). An ancestor of the black bear, which rem ained in the Old W orld rad ia ted in to southern Asia, also during the Pliocene, and evolved in to two o r m ore lineages th a t are rep resen ted by th ree ex tan t species: the Malaysian sun bear, Ursus m alavanus (or Helarctps), which probably became isolated on the Malay Peninsula; th e Asiatic o r Indian sloth bear, Ursus ursinus (or M elursusl. which p robab ly becam e isolated on the Indian subcontinent; and the Asiatic black bear, Ursus th ibetanus (or Selenarctos), which m ay have becom e isolated on the T ibetan p la teau o r elsewhere in the Himalayas. The grizzly bear, Ursus arctos, evolved in the Old W orld from the ancestra l Ursus etruscus during the m iddle Pleistocene abou t 1.6 10 m y bp. U1 arctos rad ia ted again about 300,000 ybp to form th e po lar bear, Ursus m aritim us. probably in no rth ern Asia (K urten 1964, 1968, 1976. Kurten and A nderson 1980). The grizzly crossed the Bering Land Bridge during the Wisconsin glaciation, abou t 13,000 ybp. The earliest fossil dates in North America are from 12,950 +- 550 ybp from Welsh Cave. It was found only no rth of the continen tal ice sheets during the Pleistocene. It coexisted w ith A. simus during th e Rancholabrean in the Alaska-Yukon refugium , and finally m igrated below the ice sheets only during the last phase of glaciation as th e ice sheets receded and the Mackenzie Corridor opened up. The only know n association of these 2 species sou th of Alaska is from Littie Box Elder Cave near Douglas Wyoming (Kurten and A nderson 1980). The grizzly was probably in com petition w ith the g iant short-faced bea r th roughout its range and eventually rep laced it: th e la test fossil record for A. simus is 12,650 +- 350 ybp from Lubbock Lake. As the grizzly expanded its range sou th of th e ice sheets it com peted w ith the black bear and restric ted th e black bear's range. By the time of the earliest historical records, the grizzly h ad expanded as far south as Mexico. Genetic evidence of bear phvlogenv Phylogenetic studies of bear genetics began w ith the work of Fred A llendorf and his colleagues a t the Institu te of Ecology and Genetics, University of Aarhus, Denmark (Allendorf, et. al. 1979). The original work focused on electrophoretically detectable p ro te in 11 varia tion (or the lack of it) in po lar bears. Thirteen loci were exam ined in 52 individual po lar bears w ith no allelic varia tion observed. Subsequent work has been continued in A llendorfs labo ra to ry a t the University of M ontana com paring allelic differences using p ro te in electrophoresis among the th ree species of North A m erican bears. Manlove and colleagues examined biochemical variation in the black b ea r and p resen ted results a t the 4 th In ternational Conference on Bear Research and M anagement (Manlove et. al. 1980). Further p ro te in com parisons between po lar bear populations were m ade by Larsen et. al. (1983). S tephen J. O'Brien's group at the National Institu te of Health laboratories also took this approach. The first resu lts were repo rted in 1987 w ith the publication of m olecular genetic estim ates among the Ursidae using pro tein electrophoresis (Goldman et. al. 1987, 1989). Allelic differences were exam ined among 289 fibroblast p ro te in s and 44 isozyme loci. These results were followed by Nash and O'Brien's (1987) karyotype comparisons among the Ursidae and o th e r carnivores which indicated th a t th ree m ajor chromosom al reorganization events had occurred during the evolutionary h istory of m odem ursids. The first was a m ultichrom osom al fissioning w hich increased the chromosome num ber from 2n=44 in the prim itive carnivore karyotype (Dutrillaux and Couturier, 1983) to 2n=74 in the ancestral Ursidae. The second was a com prehensive chrom osom e fusion in the lineage tha t led to the A iluropodinae subfam ily and resu lted in 2n=44 bu t with chromosomes d istinct 12 from the ancestral form . The th ird event was another, independen t centrom eric fusion in the lineage which led to the T rem arcdnae subfam ily and resu lted in 2n=52. The karyotype o f 2n=74 rem ains in all six species of the Ursinae subfamily (Nash and O'Brien, 1987). The chronology of these genetic events agrees w ith the paleontological d a ta if the rates of p ro te in divergence in bears are assum ed to be equivalent to th a t found in prim ates, a m ore in tensively stud ied group. Under this assum ption, betw een 22.4 and 32.3 m ybp an ancestor of the Procyonids (raccoon) and Ursids split in to these two lineages. Subsequently an ancestor of the A iluropodinae split from the Ursid line about 18 to 22 mybp. An ancesto r of the Trem arctinae split from the Ursid line abou t 10.5 to 15.0 m ybp. The six extant species of u rsine bears have diverged from a com m on ancestor w ithin the past 4 to 8 m illion years bu t this rad ia tion was no t resolved by p ro tein electrophoresis (Goldman et. al. 1989). The use of m itochondrial DNA (mtDNA) to examine Ursid phylogenetics was undertaken by Gerald Shields and Thomas Kocher (1991) who com pared restriction fragm ents of whole mtDNA as well sequence descriptions of cytochrome b genes and mtDNA control regions am ong American black bears, brown bears and po lar bears. Their work involved the first use of the Polymerase Chain Reaction (PCR) and nucleotide sequence analysis to com pare b ea r species. Their resu lts agreed w ith the fossil evidence, p ro te in electrophoresis results, and the fact th a t brown and po lar bears p roduce fertile F l hybrids in captivity (Kowalska 1965); i.e. th a t brow n and po lar bears 13 are sister taxa. If divergence of mtDNA is assum ed to occur a t a sim ilar ra te to prim ates, then the black bear lineage split from the po la r bear-brow n bea r line about 3.8 m illion years ago. Polar bear and brow n bear speciation occurred w ithin the last m illion years (Shields and Kocher 1991). M atthew Cronin and colleagues com pared mtDNA haplotypes am ong the th ree species of North American bears. Brown bears and po la r bears were found to share sim ilar m itochondrial DNA (0.23 base substitu tions p e r nucleotide) which is quite d ivergent (0.78 base substitu tions per nucleotide) from th a t of black bears. Brown and po la r bears were dem onstrated to be paraphyletic w ith regard to mtDNA. A lthough black bears and po lar bears have relatively low levels of p ro te in variation, there is considerable mtDNA varia tion in all th ree bears com pared with o ther mammals. C urrently designated grizzly subspecies are no t characterized by d ifferent mtDNA haplo types, and one haplotype found in Southeastern Alaska grizzly bears is m ore closely sim ilar to those of po lar bears th an to o ther grizzlies. Because of the close relationships of the th ree species and the large am ount of variation, Cronin et. al. (1991a) feel th a t assum ing a sequence divergence ra te sim ilar to p rim ates is unw arran ted . Taberlet and Bouvet (1992) developed a technique to amplify mtDNA sequence and were able to apply it to a single h a ir from a Pyrenean brow n bea r (Taberlet and Bouvet 1992) They then exam ined sequence variation in European brown bear mtDNA from 60 bears. Sequence divergence percentages ind icated th a t two 14 European lineages diverged about 850,000 ybp during the first ice age. Catherine Hanni was able to am plify a 140 base-pair fragm ent of mtDNA from cave bear bones which indicated th a t this species appeared a t abou t the same time (Dorozynksi 1994). Both these resu lts a re som ewhat ten tative because of the small sam ple sizes involved. Interspecies studies of bear genetics AUendorf and Knudsen have exam ined isozyme varia tion in grizzly bears and po lar bears. Varying levels of po lym orphism were found in d ifferent populations. Least variation was found in grizzly bears from Kodiak Island (Knudsen 1992) Zimmerman (1989) examined mtDNA differences among 55 black bears using 10 restric tion enzymes and com paring the resulting restric tion fragm ent length polym orphism s (RFLPs). M inor differences were found between th ree subspecies, Ursus am ericanus am ericanus. U. am ericanus floridanus: and the m ore restric ted Louisiana populations of U. am ericanus luteoulus in the A tchafalaya basin. Cronin et. al. (1991a) found 6 mtDNA haplotypes in black bears, 5 in brown bears, and 4 in po lar bears from sam ple locations in N orth America. Black bears (n=40) exam ined were from Alaska, New Hampshire, Oregon, and Montana. Brown bears (n=60) were from Alaska and M ontana, and po lar bears (n=40) were from Alaska and th e Northwest Territories. Two genetically d istinct groups of 15 black b ea r haplo types were found, and some types were found in only one location. The 2 morphological forms of brown bears, in te rio r grizzly and coastal brown bears, do no t cluster as distinct mtDNA lineages, and as m entioned above, one brown bear haplotype is m ore closely re la ted to po lar bears than to o ther brown bears. One haplo type was found only in M ontana, one only in Southcentral Alaska, one only in no rthw estern Alaska (Seward peninsula), and one only in islands of sou theastern Alaska. One haplotype was found in all 26 samples from Kodiak Island, b u t also from some samples from the nearby Alaska m ainland. Two po lar bea r haplotypes were found in all 3 sam ple locations while one hap lo type was found only in Alaska and one was found only on Ellesmere Island, Canada. These geographic d istributions are considered prelim inary in all th ree species because of small sam ple sizes (Cronin et. al. 1991a). Currently, Sandy Talbot and Gerry Shields are conducting sequence analysis of grizzly bear M itochondrial DNA looking a t the contro l region of the D loop and the cytochrome B gene. At the tim e of this writing Talbot is determ ining mtDNA sequence for m ost of the 30 fem ales fo r w hich we have m icrosatellite pedigree d a ta (Shields and Talbot, pers. comm.). Related wildlife genetic studies For th e purposes of our study we needed a m olecular techn ique th a t would resolve finer differences among individuals 16 th an e ither p ro te in electrophoresis or mtDNA analysis. DNA restric tion fragm ent length polym orphism s (RFLPs) have been invaluable genetic m arkers for use in linkage analysis (White et. al. 1985), medical diagnosis, and cancer research among o ther applications (Cronin et. al. 1991b), b u t the ir variability was generally too low for analysis of pedigrees un til the fortu itous discovery of variable num ber of tandem repea t (VNTR) loci or 'm inisatellite ' DNA (W yman and White, 1980). These hypervariable regions were found to consist of a DNA sequence found in m ultiple copies; allelic differences resu lt from varia tion in th e num ber of repeats, probably due to unequal recom bination events. The technique was first developed by Jeffreys lab (Jeffreys et. al. 1985a) to describe un ique genetic profiles of individuals using hum an DNA. Jeffreys and co-workers described a hum an m inisatellite com prised of repeats o f a 3 3-base p a ir sequence (Weller et. al. 1984) which was used to p repare probes th a t could detect polym orphism s in hum an DNA digested w ith HinfI an d HaeIIL A 10-15-base pair core sequence m ay ac t as a recom bination signal (Jeffreys et. ah 1985a). Most m inisatellite rep ea t un its used for fingerprinting are 9 to 60 base pairs in leng th (W eber and May 1989). Three probes (33.5,33.6, and 33.15) were cloned and characterized w hich consisted of tandem repeats of various versions of the core sequence. The fragm ents detected were found to be inhe rited in a M endelian fashion (Jeffreys et. al. 1985a). The p a tte rn of fragm ents which hybrid ized to a given probe was designated a 17 DNA 'fingerprint'. 'DNA fingerprinting' can thus be defined as the use o f detectable DNA probes which hybridize to these hypervariab le tandem repeat, o r VNTR, segments (Wyman and W hite 1980) of genomic DNA. F ingerprints derived from these multi-locus p robes were shown to be indiv idual specific, and could be used to determ ine first- o rd e r genetic relationships: paternity , m aternity , and sib-ship (Jeffreys et. al. 1985b). However, an assessm ent of p a te rn ity requ ires a detailed knowledge of the study population and DNA sam ples from bo th paren ts and offspring (W etton et. al. 1987). F ingerprinting da ta is inaccurate for m ore d istan t relationships (Lynch 1988) Subsequently, Jeffreys and co-workers were able to clone DNA fragm ents from DNA fingerprints to provide locus-specific probes (Wong et. al. 1986). Human probes have been shown to hybridize to anim al DNA and the multi-locus VNTR probes have been used to develop pedigrees of dogs and cats (Jeffreys and M orton 1987), and m ice (Jeffreys et. al. 1987). W etton et. al. (1987) and Burke and Bruford (1987) used Jeffreys probe 33.6 to dem onstrate m ultiple p a te rn ity in house sparrows. Jeffreys' probe 33.15 also hybridizes read ily w ith b ird DNA. Vassart et. al. (1987) found th a t a 280bp tandem repea t in M l3 phage DNA would hybridize to hypervariable loci in hum ans, cows and dogs, and probably salmon. W estneat et. al. (1988) then developed im proved conditions for hybridization of M l3. 18 A tandem repea t was purified from m erlin DNA and cloned in to M13mp8 and pUC8 to develop a probe (pMR-1) fo r falcon species (Longmire et. al. 1988). This probe revealed highly polym orphic fragm ent patterns in the family Falconidae bu t no t in th e closely re la ted Accipitridae. DNA fingerprints of peregrine falcons were used to differentiate between G reenland and A rgentina populations. Longmire et. al. used the M l 3 probe to develop fingerprin ts of falcons and to isolate clones in a hum an chromosome- 16-specific lib rary inserted in Charon 40. A 4.5kb fragm ent of th a t clone, PV47-2, was subcloned into pUC8 and propogated in JM lO l and was found to detect additional bands no t detected by the phage rep ea t (Longmire et. al. 1990). Quinn et. al. (1987) used sequence probes isolated from a snow goose genomic lib rary to reveal patern ity of nestlings. Four probes; M l3, Jeffreys 33.15, the hum an alpha-globin hypervariab le region (HVR), and a Drosophila Per probe, were used for anim al identification, p a tern ity testing, and linkage analysis in horses, dogs, pigs, chicken, and fish (Georges et. al. 1988). Multi-locus VNTR probes have been used to determ ine patern ity , and the developm ent of pedigrees for wild populations has been dem onstra ted w ith old world monkeys (Weiss et. al. 1988). They have also been used to m easure realized reproductive success. A m ouse m ajor histocom patibility complex cDNA p robe was used in conjunction w ith two m inisatellite probes (Jeffreys' 33.15 and the M2.5 rep ea t from the m ouse Per gene) to determ ine p a te rn ity and estim ate reproductive success in red-winged blackbirds (Gibbs et. al. 19 1990). Similar pa te rn ity and pedigree analysis techniques have been u sed to analyze kinship in prides of Serenghetti lions w here the h isto ry o f each lion p ride was known (Packer et. al. 199 la ,b). Jeffreys' probes detect a wide range of varia tion in d ifferen t taxa. The m ean num ber of bands detected p e r p robe includes 29.5 in hum ans using 33.15 o r 33.6 (Jeffreys et. al. 1985a,b), 30 in Old W orld m onkeys using 33.15 (Weiss et. al. 1988), 19 in dogs using 33.15 and 16 using 33.6, 13 in cats using 33.15 and 8 using 33.6 (Jeffreys and M orton 1987, Hill 1987), 15 in sparrows using 33.15 and 6 using 33.6 (Hill 1987, Burke and Bruford 1987), 23.7, 17.7, and 18.6 in d ifferen t species of swan using 33.6 (Meng et. al. 1989), 10 in naked m ole rats using 33.15 and 7 using 33.6 (Faulkes et. al. 1990, Reeve et. al. 1990). Bears exhibit less variation using these probes th an m ost o ther species with a m ean of 7.7 to 12.8 in black bears (Fain 1991) and sim ilar am ounts in grizzly and po lar bears (Fain, pers. comm.). The techniques of DNA fingerprinting with genomic DNA are also well su ited to the analysis of in ter-population genetic varia tion (Lynch 1990, 1991). Jeffreys' 33.6 probe was used to estim ate genetic variability and reconstruct evolutionary relationships in small, isolated populations of the California Channel Island Fox (Gilbert et. al. 1990). A com bination of techniques; p ro tein electrophoresis, mtDNA restric tion site analysis, and analysis of hypervariable m inisatellite DNA using Jeffreys' p robe were used by Robert W ayne's group a t UCLA to exam ine the genetics of wolves, particu larly the wolves of 20 Isle Royale in Michigan. The mtDNA analysis found a genotype on Isle Royale th a t is very ra re on the m ainland, suggesting th a t the island popula tion was founded by a single female. The DNA fingerprin ting analysis indicated th a t all the wolves on the island were re la ted and somewhat inbred , being as closely re la ted as captive siblings (Wayne et. al. 1991). M itochondrial DNA comparisons among wolf and coyote popula tions in states and provinces near Isle Royale found an in trogression of coyote mtDNA into wolf populations. A pparently fem ale coyotes occasionally have b red with wolves to p roduce hyb rid offspring (Lehman et. al. 1991). Further w ork w ith wolf genetics a t UCLA used fingerprin t sim ilarity (bandsharing) as an index of rela tedness among unknow n individuals. Wolves were considered un re la ted if they had different mtDNA genotypes (Lehman et. al 1992). Individuals (n=104) considered to be re la ted a t th e level of parent-offspring h ad sim ilarity values, S, o f betw een 0.700 and 0.889 (mean = 0.785, SE = 0.056). The fingerprinting studies reported above used genomic DNA fragm ents cloned in to appropriate vectors and then labeled using radioisotopes; prim arily Phosphorus-32. DNA fingerprin ting results were fu rth e r im proved with the use of DNA synthesis techniques to crea te shorter-length oligonucletides th a t could th en be labeled w ith radioisotopes (Zeff and Gellebter 1987). Subsequent advances in the developm ent of non-radioactive, chem ilum inescent labeling techniques (Edman et. al. 1988, Zischler et. al. 1989, N um berg et. al. 1989) resu lted in non-hazardous multi-locus VNTR probes labeled )w ith chem ilum inescence-producing enzymes, p rim arily alkaline phosphatase. This is the approach I firs t used to analyze m y DNA samples. The Polymerase Chain Reaction (PCR) was discovered in 1985 and becam e widely applicable to genetic studies w ith th e use of therm ally stable Taq (Therm us aquatilisl DNA polym erase (Saiki et. al. 1985, 1988). Im provem ents in this technique have m ade it possible to am plify a ta rget sequence of DNA (up to several hund red base pairs) one m illion fold (White et. al. 1989). The PCR requires pa ired p rim er sequences which anneal to com plim entary target DNA sequences after the ta rge t DNA has been dena tu red a t 94° C. Annealing occurs a t abou t 50° C., followed by elongation from the 3' end of the prim ers, catalyzed by Taq polymerase, a t 72° C. These 3 successive tem pera tu re stages are repea ted fo r 25 to 45 cycles and the am ount of PCR p roduct can theoretically be doubled during each cycle (Gyllensten 1989). Random I O m er oligonucleotide prim ers have been used w ith Taq polym erase to detect polymorphism s in genomic DNA of hum ans, p lants, and bacteria (Williams et. al. 1990). The sequences am plified differ in am oun t of varia tion and m ust be screened carefully in o rder to find useful prim ers. DNA microsatellites, o r relatively short lengths (<100 bp) of tandem ly repea ted sequences (1-6 bp long) have been characterized and found to be highly polym orphic in leng th am ong hum an individuals (W eber and May 1989). Most w ork has been done w ith families of dinucleotide repeats of the form (CA)n (GT)n (Beckman and W eber 1992). 21 Six iiiicrosatellite loci were characterized from p ilo t whales (Tautz 1989, Schlotterer et. al. 1991). The prim ers developed am plified 3, 5, 6, 8, 8, and 54 loci, and were used to estim ate m other/o ffsp ring relationships, and to estim ate rela tedness of males w ith in single pods which often contain over 100 individuals. Pod m em bers appear to form a single extended fam ily and males n e ither d isperse from o r m ate w ithin the ir na ta l pods (Amos et. al. 1993). M icrosatellites have proven to be very useful because loci w ith ten o r m ore repeats have been found to have m ultip le alleles, varying in the num ber of repeats. Primers can be developed to am plify these regions using the PCR, and the alleles p resen t can be resolved by runn ing the PCR products on polyacrylam ide gels. As I realized th a t multilocus chem ilum inescent oligo probes would no t reveal sufficient variation in grizzly bears, I h ad an oppo rtun ity to use m icrosatellite prim ers with m y DNA samples (see below). Genetic variation among individual bears Steve Fain a t the U.S. National Fish and WildUfe Forensic Laboratory (Fain 1991) has exam ined genetic sim ilarity betw een and w ithin black bear populations in d ifferent geographic locations using m ulti-locus VNTR probes. David Paetkau and Curt Strobeck exam ined m icrosatellite varia tion in Canadian black bear and polar bear populations. They found significant differences in allele frequency d istribu tion and am oun t of varia tion between d istan t populations. An isolated black 22 23 bear popula tion in Newfoundland was found to have only 35% average expected heterozygosity com pared with 80% p resen t in the continental populations. Polar bears from Hudson Bay had 49% average expected heterozygosity (Paetkau and Strobeck 1994a,b). > 24 THE STUDY POPULATION The p rim ary study area in the W estern Brooks Range (WBR) is described below. In addition, genetic samples were collected from th ree o the r areas for comparison; the Arctic National Wildlifp Refuge (ANWR) and the Alaska Range (AKR) in Alaska, and the N orthern Continental Divide Ecosystem (NCDE) in Montana. These areas are discussed u n de r RESULTS, Interpopulation comparisons. The study area The study area (Figure I) is located in northw est Alaska in the n o rth e rn foothills of the W estern Brooks Range a t 69 degrees no rth la titude , 161 degrees west longitude. It encom passes the headw aters of the Colville, Kokolik, and Utukok Rivers in an a rea of 5200 km2 (2080 mi2): over half the size of Yellowstone National Park. Elevation ranges from 400-1300 m eters. Field work is based ou t of Eagle Creek a irstrip in the Arctic Naval Petroleum Reserve. The landscape is open, treeless, Arctic tundra , and the climate is dom inated by long severe w inters, w ith short cool summ ers. The p redom inan t vegetation is tussock tund ra w ith cottongrass (E riophorum son .) and sedges (Carex spp.) co-dom inant. Wet sedge meadows are found in poorly drained areas, Drvas s p p . or fellfield com m unities are found on ridge slopes and m ountains, and willow 1 4 7 9 11411424 WBR 1097 1179 1454 1177 Meat Mountain 10871 ALASKA Poko Zv Mountain 1174 1 4 2 5 1125 1095,14 38 1716, 1 4 3 9 .1437 1734 1 7 4 9 1 0 8 9\1440 1 4 6 41 7 3 9 1461 11361 4 6 0 1 4 5 7 17451 4 5 8 1 1 4 9 ' ............. ... 10 km Figure I. The study area. Boldface num bers represen t the cen ter of female grizzly hom e ranges. I(Sajjix spp.) communities, usually stunted, are found along river channels. Topography varies from rugged m ountains in th e sou thern p a r t of the study a rea to rolling hills and a series o f east-west o rien ted ridges and bu ttes in the no rth (Reynolds 1992). Female hom e range centers in Figure I are taken from Reynolds (1978, 1980, and pers. comm.). This is n ea r the no rthern lim it of grizzly bear range. Grizzly bears h iberna te 7 to 8 m onths and have only 4 to 5 m onths to accum ulate fa t reserves. Development times are slow com pared w ith m ore tem perate bea r populations; offspring are generally w eaned a t 2-4 years of age (maximum known was 5 years) com pared w ith 1-2 years of age in M ontana (Reynolds 1991, Interagency Grizzly Bear Committee 1987a). 26 27 Demography The population size in a portion of the study a rea was estim ated a t 119 (24.4 bears p e r 1000 km^) during the firs t d irec t coun t census in 1977-1979 (Reynolds 1980, Reynolds and Hechtel 1984). An in tensive 6-day aerial census during June 1992 resu lted in a popu la tion estim ate of 153 bears in the en tire study area a t a density of 29.5 bears p e r 1000 km2 (Reynolds, pers. comm.). A m ean of 31 adu lt female bears were observed in 1977-1988. Of these, 14 females p roduced 26 cubs during 1986 ,8 females h ad 14 cubs in 1987, 5 females had 8 cubs in 1988. Using m ore in tensive survey efforts an annual m ean of 41 breeding-age females were p resen t in the study area from 1986-1991 (Reynolds 1992). A record 17 females p roduced 38 cubs in 1989, 7 females h ad 15 cubs in 1990, and 7 females had 14 cubs in 1991 (Reynolds 1989). Reproductive rates were fairly constan t over tim e a t 2.02 cu b s /litte r during 1977-1985 (n=48) and 2.02 cub s /litte r during 1986-1991 (n=58 litters). However, only 97 cubs were p roduced from 1977-1985 (9 year period) while 117 were p roduced from 1986-1991 (6 year period). Annual rates varied from 1.63-2.24 cu b s /litte r (Reynolds 1992). Reproductive in terval is a m inim um of 4 years. Mean reproductive ra te varied from 0.41 to 0.62 cu b /ad u lt fem a le /y ear w ith an average of 0.48 cu b /ad u lt fem ale /year (Reynolds and Hechtel 1980, 1984). An estim ate derived from observed families is th a t 6 females (SE = 4.03) each p roduce 2 cubs \ 28 p e r litte r p e r year, although with 41 breeding-age females and a rep roductive in terval of 4 years there should be abou t 10. The rep roductive in terval was determ ined from observations of females having successive litters; it m ay be th a t the m ean breed ing in terval is closer to 6 years. Mean age of first p a rtu rition from 1977-1983 was 8 years. The sex ra tio of animals over 2.5 years of age was 57% females and 43% males although younger age classes (< 3 yrs) were 66% females and 34% males for each year class (Reynolds and Hechtel 1984). The greatest source of variance in population grow th ra te seems to be in the num ber of females successfully breeding in any given year. This num ber may be m ore directly affected by w eather and food supply than in populations in m ore tem perate climates w here th e re is less varia tion in females breeding (Reynolds 1991, 1992) Mean m ortality during the first year of life was 44.1% from 1977-1983 (Reynolds and Hechtel 1984), 64.4% from 1986-1988 (Reynolds 1989), and 20.8% from 1989-1990 (Reynolds 1992). Average firs t year m ortality was 51.46 % (SE = 30.22) over 13 years Second year m ortality averaged 31.88 % (SE = 28.90) over 12 years. Th ird year m ortality averaged 45.75 % (SE = 39.04) over 11 years (from Reynolds 1992, Table 6). The largest cohort observed (1989) h ad 29% m ortality after 2 years bu t this was m uch lower th an usual. (Reynolds 1992). M ortality of adu lt females was 2.9% during m ost years from 1977-1991 (Reynolds 1992), b u t increased to 8.1% from 1986-1988 29 assum ing an average of 41 adu lt females p resen t each year (Reynolds 1989). Mean hom e range size was estim ated as 776 km 2 fo r adu lt males, 220 km 2 fo r adu lt females, 142 km 2 fo r subadu lt males and 113 km 2 for subadult females (Reynolds 1978, Reynolds and Hechtel 1980). Breeding females usually rem ained w ithin a 50 km 2 area (Reynolds and Hechtel 1984). 30 HYPOTHESES The specific nu ll hypotheses th a t I will address are: Ho I: Males cannot be excluded from patern ity of individual offspring using m inisatellite probes. Hq2: Males cannot be excluded from patern ity of individual offspring using m icrosatellite analysis. Hq3: There is no difference between the alleles observed in siblings and those expected to segregate from a single father. (i.e. m ultiple patern ity does no t occur) Hq4: There is no difference in num ber of offspring sired among males. HqS: There is no difference in sex ratio among successfully breeding adults. Ho6: There is no evidence of differential or stabilizing selection am ong m icrosatellite alleles, (i.e. alleles are selectively neutral). Hq7: There is no difference in allelic frequencies betw een generations. HqS: Genotype frequencies do no t differ from those expected u n de r Hardy-W einberg equilibrium . 31 Hq9: There is no difference between observed num ber of heterozygotes and the num ber expected from allelic frequencies given Hardy-W einberg equilibrium , (i.e. there is no W ahlund effect due to genetic structure) Ho 10: There is no difference between expected num ber of heterozygotes among disparate populations, (i.e. all populations sampled can be considered p arts of a single panm ictic population) y 32 METHODS M ethods used in this study include field techniques, labora to ry techniques, and analysis techniques. Laboratory techniques com prise two d istinct methods; DNA fingerprinting and m icrosatellite analysis as discussed in the INTRODUCTION and below. Field Techniques The field season consisted of approxim ately two weeks in June of each year. Beginning dates varied from year to year depending upon th e tim e of snow m elt and the weather. Typically, a field season com prised abou t 10 full days of helicopter time: the m ajor expense item . A field camp was set up a t the edge of th e Eagle Creek a irs trip w here a fuel cache was supplied by fixed-wing a ircraft flying ou t of Kotzebue, Alaska, a distance of 300 kilom eters. A Piper SuperCub aircraft and a Hughes 500 or Bell Jet Ranger helicopter were based ou t of the field camp. Bears were d arted from helicopters and imm obilized Cubs of th e yea r were occasionally ru n down on foot. When capturing bears, observers in a Supercub fixed-wing aircraft located a bear, and no tified the helicopter crew. Two of the four helicopter occupants were d ropped off while the pilot and a biologist w ith a d a r t gun pu rsu ed the bear. Bears were d arted w ith Telazol (50% tiletam ine 33 and 50% zolazepam, A.H. Robins CL, Richmond, Va.) a t a dosage of 8-9 m g/kg (Taylor et. aL 1989). Once the bear, or bears, were d arted and immobilized, the helicopter re tu rned for the o ther passengers and bears were m easured, weighed, m arked, and often fitted w ith rad io collars. Quite often family groups of th ree to four bears were im m obilized a t the same time. Blood samples were collected for DNA analysis from the fem oral artery . Blood was rem oved w ith non-heparin ized vacutainers since heparin m ay tend to inhibit digestion of DNA by endonucleases. Samples were then transferred to two volumes of 2X saline sodium citra te (SSC). Samples were kep t on ice un til they could be frozen a t -80° C. At the end of the field season samples w ere sh ipped to M ontana State University and kep t a t -80° C un til used. Tissue samples were rem oved from the ear w ith a lea ther punch while attaching ear tags. A disc of skin and connective tissue abou t 3m m in d iam eter and Im m thick was taken n ea r the edge of th e ear. The punch was cleaned after each sample was rem oved. Tissue sam ples were stored in plastic vials w ith desiccant and kep t on ice un til they could be frozen. DNA Extraction from blood Various protocols for extraction of DNA from blood were com pared: Maniatis, Jeffreys, Cronin, and NaCl-Chloroform techniques as described below. Whole blood frozen in IX SSC was thaw ed to 34 room tem peratu re . Freezing and thawing in buffer during sh ipm ent and storage effectively lysed red blood cells. Care was taken th roughou t the study to clearly and accurately label all samples. In forensic work this is term ed chain-of-custody p rocedu re and is vital so th a t genetic results can be accurately identified w ith individuals of concern. Initially we followed the protocol in Maniatis et. al.( 1982) for ex traction from blood. 500 pi of blood sample was d ilu ted in one volum e phosphate buffered saline (PBS buffer) and centrifuged a t 3 5OOx g to pellet white blood cells and nuclei. The superna tan t was d iscarded and the pellet resuspended in Trizma base (Tris)-EDTA- Sodium Dodecyl Sulfate (SDS) extraction buffer [IOmM Tris-Cl(pH 8.0), 0.1M EDTA (pH 8.0), 0.5% SDS]. Proteinase K a t a final concen tra tion of 100 ^gZml was added to lyse white cells and the reaction was incubated a t 50° C for 3 hours. After cooling to room tem pera tu re , p ro te in was rem oved by extracting twice w ith an equal volum e of buffered phenol [0.5M Tris-CL (pH 8.0)] and centrifuging a t 5000x g. The aqueous phase was saved by rem oval of the u p pe r layer w ith a m icropipette. This was extracted once w ith a phenol/chloroform -isoam yl alcohol (IAA) m ixture to rem ove residual proteins. Finally, DNA was p recip itated w ith 0.2 volum e 10M am m onium acetate (pH 5.2) and 2 volumes 95% ethanol, and frozen. DNA was pelle ted by centrifugation at 12,000x g for 10 m inutes and w ashed once w ith 75% ethanol. After final centrifugation, th e supernatant. 35 was d iscarded and the pellet a ir d ried a t room tem pera tu re before resuspension in 150 ^l distilled water. The Jeffreys' (Jeffreys and M orton 1987) pro tocol was slightly m odified and consisted of dilution of 500 jd blood sam ple in one volum e IX SSC and centrifugation to pellet white blood cells and nuclei. The pellet was washed in I volume IX SSC to rem ove residual serum and haemoglobin, centrifuged, resuspended in 0.2M sodium acetate (pH 7.0), and lysed by the addition of 1% SDS. Protein was rem oved by buffered phenol extraction followed by chloroform-IAA extraction. DNA was p recip itated by 2 volumes 95% ethanol, frozen, and centrifuged. The pellet was washed in 70% ethanol, centrifuged, a ir dried at room tem perature, and resuspended in distilled water. This procedure differs from Jeffreys' pub lished m ethod where DNA was p recip itated a t room tem peratu re in 95% ethanol, collected w ithout spooling or centrifugation, redissolved in sodium acetate, reprecip itated in ethanol, and rinsed in 80% ethanol. The Cronin technique (Cronin et. al. 1991a) consisted of incubation of 500 ^l blood sample in 1.0 ml 0.1M EDTA, 0.2M Tris, 1% SDS, and 25 ng/m l Proteinase K at 65° C for I hour, followed by incubation a t 35° C for 3 hours. Protein and SDS were p recip itated by add ition of 400 pi SM potassium acetate, cooling on ice for 30 m inutes, and centrifugation a t 12,000x g for 10 m inutes. The sup e rn a tan t was extracted w ith phenol-chloroform-IAA, and th en by chloroform-IAA. Extraction m ixtures were centrifuged a t 12,OOOx g fo r 10 m inutes. DNA in the aqueous layer was p rec ip ita ted w ith an 36 equal volume of 95% isopropanol, frozen, and washed w ith 70% . e thano l before a ir drying a t room tem perature and resuspension in d istilled water. The NaCl-Chloroform technique (Mullenbach et. al. 1989) began w ith d ilu tion of 500 ^l blood sample in I volume IX SSC, followed by centrifugation a t 10,OOOx g fo r 10 m inutes to pellet nuclei and w hite b lood cells. The pellet was resuspended in I volum e IX SSC, washed, and centrifuged a t 10,OOOx g for 5 m inutes. Washing and centrifuging was repea ted 3 to 4 times un til the pelle t was white, indicating rem oval of m ost residual serum and haemoglobin. The pelle t was th en resuspended in I volume SE buffer w ith 100 (d Proteinase K and 100 jxl 10% SDS and incubated a t 37° overnight to d ena tu re p ro te in and lyse white blood cells and nuclei. One-fourth volum e of sa tu ra ted NaCl was added to aggregate p ro te in which was rem oved by add ition of I volume chloroform-IAA, centrifugation a t 12,000x g for 10 m inutes, and pipetting the supernatan t. DNA in the aqueous phase was then precip itated w ith I volume 95% ethanol, frozen, and centrifuged to produce a pellet. The pellet was washed in 200 m! 70% ethanol, the supernatan t was discarded, and the pellet was a ir d ried a t room tem peratu re before resuspending in distilled w ater. This technique was then slightly modified by extracting the lysed m ix ture firs t w ith phenol-chloroform-IAA followed by chloroform-IAA extraction. DNA was then p recip ita ted w ith 95% isopropanol before washing with 75% ethanol, drying, and — 1 11 -------------- -------- 1 - l-l—I . L-J. , I I I 11 f I I t '-------- 1 37 resuspension in distilled water. This modified NaCl-Chloroform techn ique gave consistently h igher yields th an o ther m ethods tried. Extractions were done on 14 to 26 samples sim ultaneously, and if possible a technique was repea ted by ano ther person if results were poor. A to ta l of 64 different blood samples were extracted using the m odified NaCl-Chloroform technique. DNA extracted from blood tended to clump into viscous blobs in solution. The solution rem ained non-hom ogeneous and uniform samples for restric tion enzym e analysis were often difficult to obtain. DNA extraction from tissue Both blood and tissue samples were collected from 126 bears. In add ition 27 tissue samples were available from bears fo r w hich th e re were no blood samples. DNA was extracted from all tissue samples following a fu rther modification of the NaCl-Chloroform technique described above (Mullenbach et. al. 1989). Each ear plug was trim m ed w ith a razo r b lade to rem ove excess h a ir and then m inced in a Petri p la te w ith a razo r blade. M inced tissue was fu rth e r hom ogenized in I m l SE buffer using a glass rod in a 1.5 m l m icrocentrifuge tube. Samples were sometimes frozen overnight and fu rth e r homogenized. 200 ^l 10% SDS and 100 |fi Proteinase K were j added and incubated overnight a t 37° C or un til tissues h ad been com pletely digested, sometimes for two to th ree days. After sufficient digestion w ith Proteinase K, sam ples were cen trifuged a t 10,OOOx g for 5 m inutes and the superna tan t was 38 decan ted in to 2, 1.5 m l tubes. One-fourth volume sa tu ra ted NaCl was added and incubated for 15 m inutes. Protein was rem oved by ex traction w ith I volum e phenol-chloroform-IAA by incubating I h o u r a t 37° C, centrifuging a t 10,OOOx g for 5 m inutes, and removing the aqueous layer w ith a m icropipette. One volum e chloroform-IAA was added, the samples were incubated I hou r a t 370 c , centrifuged a t 10,000x g for 10 m inutes, and the aqueous layer rem oved. DNA was p rec ip ita ted w ith I volume isopropanol and frozen overnight. DNA was pelleted by centrifugation at 12,000x g for 30 m inutes. The supe rn a tan t was decanted and the pellet washed in 200 ^l 70% ethanol. After centrifuging again for 15 m inutes, th e superna tan t was decan ted and the pellet a ir d ried a t room tem peratu re before resuspending in distilled water. DNA extracted using th is p rocedure d id no t clum p up in w ater solution and the samples were hom ogeneous and easy to use for restric tion enzym e analysis. During m icrosatellite analysis a t the University of A lberta in Edmonton, I used an autom ated DNA extraction appara tu s (Applied BioSystems Inc. m odel 341) to extract 8 blood samples. DNA fingerprinting with multilocus probes One hund red and fifty two DNA samples from grizzlies in the W estern Brooks Range study population were used fo r fingerp rin t analysis. Because of the relatively large am ounts of DNA necessary fo r fingerprin ting , m ost of these samples (124) were extracted from tissue. 39 DNA restric tion and separation DNA samples of individual bears were restric ted using a varie ty of 4-base cutting restric tion endonucleases; H infL HaeIII. D pn H, DdeL AluL MsnL Bell. H ind i. SnaBL HhaI. T aaL and RsaL HinfI and HaeIII gave consistently good results. In o rd e r to facilitate com parison w ith o the r da ta sets, particu larly those a t the U. S. N ational Fish and Wildlife Forensic Lab, HinfI was selected as the p rim ary enzym e for generating restriction-fragm ent leng th polym orphism s (RFLPS). All DNA samples were res tric ted w ith HinfT overn igh t and sample aliquots were p repared to to ta l abou t 20 pi in o rd e r to fit in to wells in the gel. A DNA sample ( I ng-20 [Ag) was com bined w ith I [Ag endonuclease, 10% of the appropria te restric tion buffer (2 [Al), and double distilled water to total 20 |a1 and incubated a t 37° C for 6 to 11 hours. Restricted DNA samples were mixed w ith 3 [a1 loading dye. A test gel using 10% of the sample (2.3 [Al) in each well was ru n for abou t 30 m inutes at 25 milliamps. the test gel was th en sta ined w ith 10 [Al E thidium Bromide (EtBr) in 100 m l IX Tris-Borate-EDTA buffer (TBE) fo r 15 m inutes o r longer. The gel was rinsed twice in distilled w ater th en illum inated under u ltraviolet light. EtBr in tercalates in to the DNA strands and fluoresces under Ultraviolet (UV) light. The gel was exam ined to determ ine the condition of each sample. U ndigested samples were usually replaced w ith a lternate , d igested 40 samples. The rem aining, digested samples were loaded in to wells in an agarose gel. Initially 17 sample lanes were used .w ith 2 s tandard m arker lanes and one standard grizzly DNA lane on each gel. Later, m ore th an 3 s tandard lanes were used. Fragm ents were separated on electrophoretic gels consisting of 0.8 to 1.0% agarose in TBE buffer. A standardized electrophoresis techn ique was developed using 1.0% agarose gels in IX TBE a t 20-25 m illiamps. for 24 hours or longer. After electrophoresis, gels were sta ined w ith 10 pi EtBr in 100 ml IX TBE for 15 m inutes o r longer. EtBr in te rcala ted in to the DNA strands was illum inated un d e r UV light and pho tographed using polaroid X-ray film as a reco rd of the sam ple configuration and relative densities. A tran spa ren t ru le r was pho tog raphed w ith the gel to facilitate trimming of th e gel for transfer. Gels were trim m ed ju st below the I kilobase fragm ents, and 15 cm fa rth e r down the lanes. 2 m m were trim m ed off bo th sides w here the gel is thicker; final size of the gel was 12.5mm by 15mm. Southern transfer DNA was transferred to nylon m em branes of th e same size using th e Southern Blot technique of W estneat et. al. (1988). Initially, w hen using radiolabelled probes, gels were trea ted w ith 1.5 M NaCl /1 .5 M NaOH for 15 m inutes, and repeated, to dena tu re the double stranded DNA. Gels were then washed in 0.04 M Na OH/ I M NH4Ac (Ammonium Acetate) for 15 m inutes, and repeated , to 41 neu tra lize the denatu ring buffer. A Southern Blot appara tu s was set up fo r tran sfer to a charged m em brane w ith 0.04 M /1 .0 M NH4Ac as a tran sfe r buffer. Gels were transferred overnight. After the use of radioisotopes was discontinued in favor of using chem ilum inescent probes, transfers for use w ith chem ilum inescent probes required neu tra l nylon m em branes. These m em branes requ ired a d ifferent transfer protocol. A fter trimm ing, gels were depu rina ted for 15 m inutes in 0.25 M HC1. They were then w ashed in 1.5 M NaCi/0.5 M NaOH for 15 m inutes, and repeated, to dena tu re . Gels were then washed in 1.5 M NaCl/1.0 M Tris to neu tra lize for 30 m inutes, and repeated; and were tran sferred w ith 10X SSC buffer overnight. A fter transfer, m em branes were rem oved from th e tran sfe r appara tu s and trea ted to covalently bond (crosslink) DNA fragm ents to th e nylon. Most of the time m em branes were baked in a vacuum oven a t 80° C for 30 m inutes. Occasionally, m em branes were irrad ia ted w ith UV light in a S tratalinker UV cham ber a t 12,000 njoules. A S tratalinker was la ter purchased and all m em branes afte r num ber 40 were crosslinked in the chamber. A com parison was m ade using two m em branes p rep a red w ith iden tical samples and transfer protocol: one was oven-baked and the o th e r was irrad ia ted in the UV cham ber. The baked m em brane h ad a h igher signal th an the irrad iated m em brane after identical hybridization. This was also found to be the case a t the U. S. National Fish and Wildlife Forensic Lab (Fain, pers. comm.). In m ost cases the am oun t of DNA bonded is sufficient using UV irradiation, bu t in cases , 42 w here re ta in ing as m uch DNA as possible is necessary, i t is b e tte r to bake th e m em branes. H ybridization w ith radiolabelled probes M em branes were initially p robed for tandem repea t sequences of genomic DNA w ith radiolabelled Pv47 and M l 3 p robes (Weller et. al. 1984, Longmire et. al. 1990, Shields and Kocher. 1991). These p robes were labeled w ith 32P using nick transla tion as below. Charged m em branes were pre-hybrid ized in 20 m l of hybrid ization solution p e r m em brane comprised of 1.4 g 7% SDS, 40 M-I 0.5M EDTA (Im M total), 0.2 g Bovine Serum Albumin (BSA), 10.5 m l 0.5 M NaP04 (pH 7.2), and 9 m l double distilled water. M em branes were sealed in seal-a-meal bags and incubated overnight a t 60° C. Probes were radiolabelled using nick translation; probes were incubated a t 15« C for 60 mins, w ith dCTP, dGTP, dTTP, dATp32, DNAse, and DNA Polym erasel in a buffer solution of THIS, NaCl, EDTA, SDS and either blue Dextran o r Orange 10 (Weller et. al. 1984). The reaction solution was run through a column of Sephadex G-50 beads and labeled p robe was collected when cpm reached 100,000. A lternately, dCTP32 was used w ith the o ther dNTPs. Labeled p robe was added to the pre-hybrid ization solution and incubated overnight a t 60° C. M embranes were rem oved from solution and washed in 2X SSC/0.1% SDS 2 times a t room tem pera tu re for 15 min. and once a t 60° C for 60 m inutes. They 43 were th en washed briefly in IX SSC a t room tem peratu re , b lo tted , and w rapped in cling wrap. Signal detection . Labeled m em branes were used to expose x-ray film w hich was th en developed to reveal a characteristic banding p a tte rn o r DNA fingerprin t. After hybridization, m em branes were scanned w ith a Geiger coun ter to estim ate the relative strength of th e signal. M em branes w rapped in th in plastic sheets (cling w rap) were p laced on a sheet of Kodak O rtho film in a reflective film cassette in a darkroom . After placem ent, film cassettes were sto red in the lab from 24 to 72 hours before film development. J H ybridization w ith chem ilum inescent probes During October 1 991 ,1 visited the U. S. National Fish and Wildlife Forensic Lab in Ashland Oregon to work w ith Steve Fain and M att Cronin. I gained experience using alkaline phosphatase-labeled oligonucleotide probes. Two probes used were subunits of Jeffreys' 33.15 and 33.6 probes which detected m ultiple VNTR fragm ents in black and po la r bears, bu t which h ad no t been used extensively on grizzly b ea r DNA. Two additional probes, CMMlOl and MSI were also used. M em branes were hybridized following a protocol developed by Tropix Inc. and m odified according to advice from Steve Fain and 44 from m y own experience. Originally, m em branes were incubated in solution in open trays in an incubator shaker. Later a ro ta ting glass- bo ttle incubation cham ber was purchased which stream lined the process. M em branes were wet in IXSSC and incubated in 200 m l Blocking Buffer each for 60 min. a t room tem peratu re in separate, open trays u n d e r gentle agitation. Membranes were th en ro lled up lengthw ise and inserted in to glass bottles w ith the DNA side facing inw ard. Two o r m ore m em branes were separated by a layer of hybrid iza tion mesh; usually only two m em branes were hybrid ized together. Twenty m l of H ybridization buffer p e r m em brane was w arm ed to 42° C, m ixed with 2.0 nM probe and added to bottles. Bottles were sealed and ro ta ted gently a t 420 C for 30 m inutes. A series of hybrid iza tions a t varying tem peratures were ru n using 33.15 p robe and 42° C was chosen as the standard tem perature. After 30 m inutes, the hybrid ization solution was d iscarded and the m em branes were washed in 100 ml (approx.) each of IX SSC/1% SDS a t 50° C centigrade for 10 m inutes. This wash was repeated . M em branes were rem oved from bottles in to open trays and w ashed in IX SSC for 5 m inutes a t room tem perature w ith gentle agitation. This w ash was repeated . The wash buffer was d rained off and m em branes were w ashed separate ly in open trays in 125 ml of Detection Buffer for 5 m inutes a t room tem pera tu re w ith gentle agitation. This wash was repeated . 45 Signal detection Ten m l o f Detection Buffer p e r m em brane were set aside before these washes and HO ml AMPPD substra te p e r filte r was added . A fter washing, m em branes were p laced in plastic seal-a- m eal bags. No m ore than 2 m em branes were p laced in a bag w ith DNA sides facing out. Detection Buffer with AMPPD was added, bags were sealed, and the solution was slowly agitated by h an d fo r 5-10 m inutes. M embranes were removed, lightly b lotted , and placed inside clear plastic sheets. Edges of the plastic sheets were sealed w ith tran sp a ren t tape. M em branes in sheets were p laced in non-reflective film cassettes and a sheet of Kodak Ortho X-ray film was in se rted in a darkroom . Approximately 2 hours were requ ired fo r the chem ilum inescence reactions to p lateau. Film was exposed fo r I to 48 hou rs afte r this. Usually, films were developed afte r 4-6 hours. If resu lts were unsatisfactory , the exposure tim e was ad justed . The ligh t reac tion continued fo r over 48 hours so several d ifferent exposures could be m ade if necessary. In terp re ta tion of multilocus DNA fingerprints DNA fingerp rin t bands on exposed film were scored m anually by locating bands th a t were easy to detect, d isplayed variation am ong individuals in the study population, and were w ithin the size range of 2 kilobases to 20 kilobases. This size range was consistently J l clear on exposed films; smaller and larger fragm ent sizes were often undetectab le because of fain t signal (larger sizes) o r too little resolu tion between bands (smaller sizes). Only polym orphic bands were chosen fo r analysis; therefore estim ates of p e rcen t po lym orphism derived from this technique are inaccurate (see RESULTS, In terp retation of multilocus DNA fingerprints). Light lines were draw n across th e film w ith a pencil and ru le r to connect bands in m arker lanes an d /o r bands in lanes of known individuals. No a ttem pt was m ade to distinguish double copy fragm ents (homozygous alleles) from singly copy fragm ents (heterozygous alleles). A band was m erely scored as p resen t o r absen t in each individual. 15 bands were chosen for scoring w ith 33.15 probe; 10 w ith 33.6; 17 w ith CMM101; and 12 w ith MSI in the study population . A small sample of grizzly bears from the no rth e rn con tinen tal divide ecosystem was sim ilarly scored w ith 12 bands using 33.15 and 8 bands using 33.6. A BioImage Analyzer was used to exam ine feasibility of m achine scanning of fingerprin t transparencies. Experience w ith the analyzer and advice from Steve Fain indicated th a t m ore reference lanes were needed p e r gel in o rder to im prove resu lts over those ob tained manually. Fortran program s were w ritten to accept keyboard en try of b and num bers for individuals, and convert to a p resence/absence m atrix of ones and zeroes. Individuals were in rows w ith b and num bers in columns; a 0 indicated absence, and a I presence of a band . The p resence/absence m atrix was ed ited w ith w ord 46 47 processing software to include commas after each entry . A trubasic p rog ram was w ritten to transpose rows and columns fo r use in fingerp rin t analysis program s requiring band num bers in rows and indiv iduals in columns. Matrices were fu rth e r ed ited w ith w ord processing software fo r use as d a ta files. Matrices were then im ported in to a Lotus w orksheet fo rm at for fu rth e r editing and processing. D ata were processed w ith a Lotus 123 m acro program developed a t the U. S. National Fish and Wildlife Forensic Lab by Dr. Bruce Taylor. This program com pared all bands between all indiv iduals to derive a sim ilarity index num ber fo r each individual and estim ate various param eters according to form ulas derived by Lynch (1990, 1991). DNA analysis of m icrosatellite loci One hund red and fifty two DNA samples from grizzlies in the W estern Brooks Range study population were used fo r genotyping by m icrosatellite analysis. Most of these h ad been extracted for DNA fingerprin ting b u t 19 blood samples were re-ex tracted to provide additional DNA. A tooth and some scrapings of tissue pu lp from a fo ram en in the skull were obtained fo r male #1082 and traces of DNA w ere extracted from bo th samples. In add ition DNA was extracted from 15 blood samples of Arctic National Wildlife Refuge (ANWR) grizzlies, 17 blood samples of Alaska Range (AKR) grizzlies, 16 blood samples of Yukon Delta (YD) 48 grizzlies, 16 blood samples of M ontana grizzlies p rim arily from the N orthern Continental Divide Ecosystem (NCDE), and 2 tissue samples o f brow n bears from the Kamchatka Peninsula in th e form er USSR. These sam ples were analyzed as tim e perm itted in o rd e r to m ake prelim inary com parisons between disparate populations. The YD sam ples were no t genotyped. Developm ent of p rim er sets The p rim er sets used were developed by David Paetkau (Paetkau and Strobeck 1994a,b) a t the University of A lberta. Tandem repea t m icrosatellite sequences h ad been identified by screening 20,000 clones from black bear genomic DNA and 4 single­ locus p rim er sets h ad been used in an analysis of genetic varia tion in black b ea r populations (Paetkau and Strobeck 1994a). Four add itional p rim er sets were developed and used in an analysis of genetic varia tion in polar bear populations (Paetkau and Strobeck 1994b). These 8 p rim er sets were used by m yself and David Paetkau a t the University of A lberta in Septem ber 1993 w ith the DNA samples from ou r grizzly study populations. This was essentially th e firs t use of these prim ers w ith grizzly bear DNA, and th e first m icrosatellite analysis of grizzly bears. The am plification and electrophoresis protocols were developed by David Paetkau for black b ea r and po lar bear analysis, and were fu rth e r refined by him during the grizzly bear analysis. 49 David Paetkau constructed p rim er sets for clones which con tained a t least 15 un in te rrup ted tandem repeats. The first 8 p rim er sets th a t gave satisfactory results were used fo r am plification. Thus, they were chosen for efficacy of use in the labo ra to ry based on num ber of repea t segments, characteristics of th e p rim er sequence, and ease of amplification. We have no way of knowing w hat function is served by the alleles we exam ined. The p rim ers were given alphabetic designations based on the clone from w hich they were originally isolated; the 8 sets we used were designated A, B, C, D, L, M, P, and X. Amplification of target DNA using PCR Grizzly bear genomic DNA was amplified w ith Taq polym erase using a Perkin-Elmer 9600 therm al cycler. Twenty-five pi PCR cocktails were p repared using I pi of genomic DNA (approx. I p,g), 120 p,M each dNTP, 0.2 pM each fluorescent end-labeled prim er, lx Taq buffer, and 0.5 units (U) Taq polym erase m ixed w ith distilled, de-ionized water. Pre-mixes were p repared in 2 m l batches fo r 20 reactions. Three fluorescent dyes were available for use w ith the A pplied BioSystems 373A au tom ated sequencer. The ROX standard was labeled w ith red dye. Primers A, B, C, and D were labeled w ith b lue dye; and L, M, P, and X were labeled w ith green dye. Because the PCR p roducts th a t each prim er amplified fell in to different size classes, and 2 colors of dye were used, p rim er sets could be 50 m ultiplexed: prim ers A and L were com bined in one PCR reaction , M and X in another, B, C, and D in a third, and P in a fourth , for each DNA sample. Thus 4 cocktail pre-mixes were p repared . One jaI of DNA sam ple was m ixed with 23 ^l of each pre-m ix cocktail and I ^l of Taq polym erase was added ju st p rio r to therm al cycling. The therm al cycler was allowed to h ea t up to 94° C before the reaction m ixtures were p laced in the tray for a 'ho t s tart'. Two cycles consisting of 30 secs, a t 94° C, 20 secs, a t 58 ° C, and I sec. a t 72° C were ru n for longer synthesis a t the beginning of th e run . These were norm ally followed by 35 cycles of: 15 secs, a t 9 4 ° C, 20 secs, a t 5 8 ° C, and I sec. a t 72° C. The tem perature was then held a t 72° C fo r 30 secs, to anneal unreacted ingredients, and then d ropped to 4 ° C indefinitely . Reaction m ixtures were rem oved from the therm al cycler and stored a t about 4 ° C until loading. Electrophoresis of PCR products Polyacrylam ide gels were p repared using 45 m l of 6% acrylam ide/7 .5 M urea, 250 ^l of 10% Ammonium persu lphate, and 30 fil TEMED mixed a t room tem perature. The gel solution was pou red betw een clean glass plates using a syringe and allowed to polym erize fo r a t least 2 hours before setting in p lace in IX TBE electrophoresis buffer in an ABI 373A sequencer. The PCR products were com bined to allow the runn ing Of 4 loci in each lane; 2 lanes p e r individual sample. Because some products am plified b e tte r th an others and the scanner was m ore sensitive to - J - I J -L blue dye, the am plification products from prim ers A and L were d ilu ted 1:6 in to the M and X products (3.75 nl AL in to 25^1 MX). Similarly, BCD products were diluted 1:6 into P products. Loading buffer was p repared using 3 |ul formamide and 0.5 pi GENESCAN ROX 2500 s tandard p e r sample and 1.5 pi of each reaction m ixture was added . Thus each individual DNA sample was rep resen ted by 2 aliquots of PCR products, ALMX, and BCDP, mixed w ith loading buffer. Sample aliquots were kept on ice un til ju st p rio r to loading w hen they were hea ted to 90° C for 2 mins, to dena tu re products in to single strands. The 5 |il aliquots were loaded in to wells in the gel. A 36-well comb was used so 18 individual bear sam ples could be ru n a t one time. Gels were ru n fo r 6 hours a t 30 watts (24 m illiamps o r l3 0 0 volts) w ith a I hou r 45 min. scan delay. Gel tem pera tu res were abou t 40 ° C. I Data collection D ata were collected on an Apple Macintosh com puter using GENESCAN software. Scanning of the gel was delayed for I h ou r 45 m inutes afte r the s ta rt to save storage space; PCR p roducts d id no t show up un til abou t 2 hours after starting the gel. A laser scanner in th e ABI 373A detected flourescent m arker dye as it m oved p as t the scanning beam during electrophoresis. Since a s tandard of known size (ROX 2500) was run in each lane with the sample, PCR products could be accurately sized down to single base pairs and com parisons _________________________________________________________________________________________ 51 52 could be m ade between gels. Raw da ta files requ ired abou t 6 m egabytes (MB) of storage space p e r gel. D ata analysis D ata files were pre-processed using a GENESCAN software m atrix set up by David Paetkau to rem ove unnecessary background. This resu lted in a gel file of about 2.5 MB. Threshold detection levels fo r each dye were ad justed for each gel and lanes were delineated m anually. GENESCAN analysis software analyzed the gel according to th e lanes and thresholds given. This rem oved m ore background and crea ted a results file of abou t I MB. Once a satisfactory gel file was exam ined, th e raw da ta file was erased. Gel files were saved on th e M acintosh un til each lane was analyzed to the best obtainable resolu tion and allele pairs had been recorded for each individual. Allele pairs were recorded by h and and then en tered in to a Filemaker Pro database on another Macintosh com puter. Recorded d a ta were th en double-checked against the gel file by bo th m yself and David Paetkau. Once the data h ad been double-checked, the gel file was erased, and the results file was copied onto a floppy disc. Dr. Curt Strobeck developed a Filemaker Pro m acro p rogram to com pare genotypes among bears in the da ta set. A search of the d a ta w ould produce a list of o ther bears th a t shared a t least one allele a t each locus. Each search took from 6-10 m inutes. RESULTS AND DISCUSSION The W estern Brooks Range study population of grizzly bears has been stud ied continuously since 1977 by the Alaska D epartm ent of Fish and Game. During tha t time 489 captures of 256 individual bears and subsequent m onitoring of radio-collared anim als has resu lted in detailed knowledge of m aternal pedigrees fo r 53 fam ily groups. Whole blood an d /o r tissue samples were collected from all anim als hand led since 1988. Of these, 152 individuals were used fo r genetic analysis. DNA extraction Yield of DNA using the Maniads technique was abou t 10 ng DNA p e r 10 u l sample as determ ined by gel electrophoresis, staining w ith EtBr, and com parison to a standard. Yield of DNA using the Jeffreys' technique was about 50 ng DNA per 10 pi sample. Yield of DNA using the Cronin technique was about 100 ng DNA p e r 10 pi sam ple (10 pg p e r ml). Yield of DNA using the m odified NaCl- Chloroform technique w ith blood o r tissue was abou t 200-500 ng p e r 10 pi sample) 20-50 pg per ml) as m easured using this rough EtBr s tandard m ethod. Optical density m easurem ents a t a wavelength of 260 angstrom s provided m ore accurate measures of DNA concentration. 54 Since an O.D 260 of 1.0 is equivalent to 50 DNA total in the I m l cuvette, a 1:200 dilu tion of the DNA sample (5 (nl DNA sam ple in 1000 |xl distilled water) gives an O.D 260 reading in \ig/\d of original sample. The Cronin m ethod of extraction from blood yielded 14.65 (SE=9.91) (Ag DNA pe r ml of whole blood (n=52). The modified NaCl- Chloroform m ethod yielded 55 (SE=72.41) |xg DNA per m l of whole blood (n=87) and 787.70 (SE=449.52) fxg DNA per m l of m inced tissue in buffer [196.93 (SE=112.38) |ig DNA per ear plug; n=53]. Yield of DNA varied w ith the condition of the sample and increased w ith experience using the technique. DNA restriction and separation A sam ple size of 10 pg DNA was found to be sufficient for fingerprin ting . A sam ple size of I |xl regardless of DNA concentration was used for PCR amplification of microsatellite loci. Concentrations of DNA ranging from .025 to 1.88 ^g/^l gave satisfactory results. Consequently, m icrosatellite analysis is a t least 3 o rders of m agnitude m ore economical (or efficient) in use of DNA. DNA fingerprinting of m inisatellite loci T h irteen charged nylon m em branes were p rep a red from HinfI- d igested gels. Six charged nylon m em branes were p rep a red from HaeIIEdisested gels. Fifty-four n eu tra l m em branes were p rep a red 55 from Hinfl-d isested gels, and 2 neu tra l nylon m em branes were p rep a red from Haelll-digested gels. The pv47 radiolabelled probe was successfully hybrid ized to 19 charged m em branes (13 HinfI and 6 HaeIID. Inconsistent resu lts w ere ob tained . The best results were w ith gels th a t were ru n longer th an usual (30 hours o r more) and thus got be tte r separation of the longest fragm ents. Longer fragm ents in general h ad stronger signals th an sho rte r fragm ents. M other-offspring banding p a tte rn s were carefully observed in 8 family groups. Variation was lim ited and no bands were observed in offspring th a t were no t also p resen t in th e m other. The M l3 p robe was hybrid ized to fou r m em branes. It w orked no b e tte r th an the pv47 probe and detected fewer bands. However, M13 gave b e tte r resu lts w ith longer fragm ents. 33.15 probe Forty-nine of th e neu tra l m em branes were successfully hybrid ized w ith the 33.15 multi-locus oligo probe. One hund red tw enty-nine individuals were scored w ith 15 possible 33.15 bands. Analysis of p a te rn ity was no t possible using the resu lts from the 33.15 p robe alone. In fact, it was no t possible using a com bination of 4 probes, bu t initial results w ith the 33.15 probe gave some indication th a t using m ore probes would increase the reso lu tion enough to determ ine patern ity . When m other/o ffspring groups were com pared, m any offspring d id no t differ from the I 56 m othe r (shared all of the ir bands w ith the m other). Rarely, an offspring would exhibit one band no t found in the m other, m ore ra re ly 2 bands were found, and the offspring often lacked one o r m ore of th e m other's bands. The bands tha t offspring d id exhibit, w hich cam e from the father, were common bands Shared by a m ajority of the males in the sample. Some b and com parisons of m other's and offspring are given in Table I as an example. The first bear in each group is th e m other, followed by h e r offspring. Table I . Shared 33.15 bands among th ree fam ily groups. Bear_______________ 33.15 Bands 1749 2 4 6 8 9 10 11 12 15 1750 2 4 6 8 10 11 13 15 1751 2 4 6 8 9 10 11 1752 2 4 6 8 9 10 11 1166 2 3 6 8 12 13 15 1701 2 ' 4 6 8 11 12 15 1702 2 . 4 6 8 9 11 12 15 1734 2 4 5 6 8 9 10 11 15 1735 4 6 8 11 12 15 1736 2 4 6 8 10 15 hi th e upper fam ily group #1750 has one b and (13) no t found in the m other. Offspring #1751 and #1752 have no bands th a t are n o t shared w ith the m other. When the m icrosatellite resu lts were 57 la te r com pleted no fa ther was found in the sample fo r these 3 offspring. In the cen ter family group, offspring #1701 has 2 bands (4,11) no t found in the m other, and #1702 has 3 bands (4,9,11) no t found in th e m other. This was the m ost variation seen in the en tire sam ple. W hen the m icrosatellite results were la te r com pleted it was de te rm ined th a t each of these offspring were sired by a d ifferen t male. The fa ther of #1701 is #1478 who d id have bands 4 and 11. The fa th e r of #1702 is #1421 who was only fingerp rin ted on one chem ilum inescent blot; bands 4 and 11 were detected, b u t band 9 was no t found. In the lower fam ily group offspring #1735 has one band (12) n o t found in the m other while #1736 has no bands th a t are no t also shared w ith the m other. The m icrosatellite results de term ined th a t m ale #1712 is the fa ther of bo th offspring. Male #1712 was fingerp rin ted clearly on 2 chem ilum inescent blots, b u t b and 12 was no t detected. The initial results from the fingerprin t d a ta ou tlined above ind ica ted th a t the re was insufficient variation to determ ine patern ity , a t least w ith only one probe. Later com parison w ith the m icrosatellite results shows th a t the fingerprin t d a ta are also no t accurate enough to determ ine paternity , although results from the US N ational Fish and Wildlife Forensic Lab have shown th a t the use of 4 o r m ore probes, running a standard ladder in each lane, restric ting analysis to a single gel, and reading the b lo t w ith an 58 im age analyzer can determ ine patern ity in some cases (Fain pers. comm.). Band-sharing com parisons were m ade w ith th e Lotus 123 com puter p rogram obtained from the U. S. National Fish and Wildlife Forensic Lab. Initial analysis gave first estim ates of (Table 2): Table 2. Population genetic param eter estimates from a single m ulti-locus probe (33.15) Estimates from Lynch (1991) m ean sim ilarity, s 0.5424 standard deviation 0.1621 variance 0.0263 polym orphism , P2 3.605 m ean homozygosity 0.0471 m ean heterozygosity 0.9529 Ne, lower estim ate 52.73 Ne, u pper estim ate 136.72 m ean num ber of shared fragm ents 6.47 Subsequent run s refined these estimates (Table 3). The to ta l num ber of anim als used for analysis was reduced to 121 individuals because of uncerta in ty scoring some bands. Since only polym orphic bands were chosen for scoring, the estimate of percen t polym orphic loci is inaccurate. The m ean num ber of shared fragm ents (6.47) is lower th an th a t found in m ost o ther mammals using these probes (see INTRODUCTION). 33.6 probe 59 Fifteen of the neu tra l m em branes were successfully hybrid ized w ith the Jeffreys' 33.6 multi-locus oligo probe. Sixty-one individuals were scored w ith bo th 33.15 and 33.6 using 25 possible bands. This d a ta subset was analyzed w ith the Lotus p rogram and is included in Table 3. CMMlOl and MSI probes Six of the neu tra l m em branes were successfully hybrid ized w ith the CMMlOl multi-locus oligo probe. Twenty-four individuals were scored w ith 33.15, 33.6, and CMMlOl using 42 possible bands. This d a ta subset was analyzed w ith the Lotus p rogram and is included in Table 3. Four of th e n eu tra l m em branes were successfully hybrid ized w ith the MSI multi-locus oligo probe. Thirteen individuals were scored w ith 33.15,33.6, CMM101, and MSI using 53 possible bands. This d a ta subset was analyzed w ith the Lotus p rogram and is included in Table 3. Eauiladder probe Six of the n eu tra l m em branes were hybrid ized w ith the Equiladder multi-locus oligo p robe w ith lim ited success. At this po in t in th e lab work, all samples h ad been p robed w ith 33.15 and the supplies of the o ther 3 probes h ad been exhausted. Through correspondence w ith the forensic lab in Ashland I learned th a t Steve 60 Fain was obtaining very good results w ith the equ iladder and its p robe and was runn ing a sample of equiladder in each lane w ith each sam ple in o rder to make m ore accurate b and com parisons. I d id no t have sufficient DNA or oligo probes to re-run all samples in th is way, and m y results indicated th a t p a tern ity determ ination w ould still be questionable w ith this technique, so use of the equ iladder s tandard and probe was no t pu rsued farther. In te rp re ta tion of multilocus DNA fingerprints Results from the pv47 and M13 autorads were inconsisten t and of generally poor quality. No attem pt was m ade to score bands in any individuals. The estimates of population genetic param eters from the o the r 4 probes (Table 3) are of lim ited value because of the large degree of uncertain ty in the band scoring. In con trast to the unam biguity of the m icrosatellite d a ta (see following section), multi-locus m inisatellite resu lts were relatively uninform ative. The longer length of the m inisatellite bands (2 ,000 to 6500 base pairs) m ade small differences in b and leng th impossible to distinguish. Bands th a t appear sim ilar are considered as one. Com parisons are usually restric ted to samples ru n on the same gel; b u t we com pared bands across gels by restricting the bands analyzed to unconfusing areas of the gel and choosing bands found on Our grizzly bear DNA standard . As m entioned above, individual samples were ru n on several gels w ith the resu lt th a t we have m ore 61 confidence in the 33.15 data and the 33.6 data th an in the sm aller subsets. The m ean num ber of shared fragm ents (6.70 fo r p robe 33.15, and 5.55 fo r probe 33.6) is lower than tha t found in m ost o ther mammals using these probes (see INTRODUCTION). The m ean sim ilarity index, S, for m other/offspring pairs (kinship 0.5) was calculated as 0.699 (SE=0.133) which is similar to th a t found in wolves (Lehman et. al. 1991. see INTRODUCTION) although we h ad a m uch h igher variance. In fact the variance was so h igh th a t S can n o t be used w ith this da ta as a reliable ind icator of degree of rela tedness among individuals. 62 Table 3. Population genetic param eter estimates from multi-locus . p robe com binations (after Lynch 1991). 33.15 33.15 33.15 33.6 33.15 33.6 CMMlOl 33.15 33.6 CMMlOl MSI Number of samples n 129 121 61 24 13 Number of bands 15 15 25 42 54 Mean no. fragm ents 6.47 6.70 12.25 . 19.71 27.62 Sim ilarity Index S 0.542 0.559 0.600 0.558 0.626 Standard Deviation 0.162 0.155 0.115 0.121 0.120 Variance 0.026 0.024 0.013 0.015 0.014 Homo­ zygosity 0.047 0.053 0.202 0.799 2.268 Lower Ne NI 52.73 49.27 41.60 49.59 37.36 . Upper Ne 136.72 129.79 114.49 130.39 106.00 - N2 63 DNA analysis of m irmsateHite loci The 8 m icrosatellite p rim er sets am plified 61 unam biguous alleles in th e W estern Brooks Range (WBR) grizzly b ea r study population . Allele frequencies were calculated by m anually tallying each allele in the Filemaker Pro database. These d a ta were tran sfe rred in to an Excel 3.0 database and frequencies were double- checked by tallying using search and extract functions. All possible hom ozygotes and heterozygotes a t each locus were tallied using Excel functions. Despite the care taken in chain-of-custody to accurately label all samples th roughout sample collection, DNA extraction, and DNA analysis p rocedures, 5 samples were determ ined, from m icrosatellite results, to be m islabeled. Two of these were females who d id no t m atch the genotypes of their known offspring; the DNA had been ex tracted from tissue samples obtained from second parties and they a re discussed u n de r pedigrees (families 1125 and 1141). The sam ple labeled #1125 was re-num bered as Unknown 2 (UNK2). The sam ple labeled #1141 had a genotype identical to #1147 and apparen tly th e 7 h ad been m isread as a I during lab procedures. Blood samples from bo th females, #1125 and #1141, were subsequen tly extracted. The genotypes resulting from the new sam ples m atched the offspring and enabled pa te rn ity to be determ ined. A nother of the m islabeled samples, #1717, was an offspring th a t d id n o t m atch the genotype of its m other and was apparen tly 64 m islabeled during extraction in Anchorage. A nother DNA sam ple of #1717 was available which did m atch the m other and h ad the same fa th e r as th e sibling. This sample is discussed under fam ily 1716 below and was labeled Unknown I (UNKl). The 2 o the r m islabeled samples, #1480 and #1481, were siblings who d id n o t m atch th e ir m o the r and are discussed u nder fam ily 1097 below. The original sam ples were identified from fain t lip tattoos in th e field and the age and sex of the bears sam pled was consistent w ith th is identification. Subsequently, blood samples which had been ob ta ined from #1480 and #1481 as cubs were extracted. The genotypes from this DNA d id m atch the m other and enabled patern ity to be determ ined. The m islabeled samples were re-num bered Unknown 3 (UNK3) and Unknown 4 (UNK4). These sam ples were known to be m islabeled only because the genotypes d id no t m atch w ith known m others o r offspring. There m ay be o th e r samples in the da ta th a t were also m islabeled. For m ost purposes of population analysis the exact identification of an individual, which is no t a m em ber of a known fam ily group, is no t critically im portan t, b u t when behavioral da ta is in te rp re ted according to genetic results, accurate identification is crucial. M utation Hypervariable m icrosatellite loci are useful in distinguishing am ong individuals because of the relatively high m u ta tion ra te th a t occurs in these regions. However, if m utations in num ber of tandem 65 repea ts occur too often they can confuse the identification of patern ity : i.e. a male could contribute a gamete th a t differs from his genotype a t one locus. To assess the m utation ra te we com pared genotypes of offspring with the ir m others. Of the 57 offspring (912 alleles; 456 m aternal alleles) com pared we found one instance w here n e ith e r allele a t a locus m atched with one of the m other's. This indiv idual, #1732 h ad alleles M212 and M222. Allele M222 was un ique to this individual in our sample and was verified on 2 separate gels. It is probably a length m utation in th e m other's gam ete. If so, the m utation ra te in our sample m ay be abou t I p e r 500 alleles o r 2 x 10-3, and there are probably also one o r two m u ta tions in ou r sample from a m ale gamete. It m ay be even h igher (see also Effective num ber of alleles, Table 11). Thus, males excluded from patern ity a t only one locus m ay be possible fathers if no o ther fa ther has been determ ined. There a re 5 instances among 57 offspring where this is the case; one of these is m en tioned in Figure 2b. All these males are m entioned under pedigrees: #1712 in family 1141 (Figure 11), #1462 in fam ily 1174 (Figure 15), #1453 in family 1439 (Figure 25), #1459 in fam ily 1461 (Figure 31), and #1463 in family 1479 (Figure 33). Most of these have been identified as successful males. If #1712 d id fa th e r an additional offspring th a t was h idden by m utation, his estim ated rep roductive success would increase only about 1% (see Male rep roductive success). As additional grizzly fam ily groups are genotyped from o ther subpopulations, and the database increases, we will refine ou r estim ates of m utation. 6 6 Null alleles The m icrosatellite prim ers we used were chosen fo r efficacy of use in the labora to ry based on num ber of repea t segments, characteristics of the prim er sequence, and ease of amplification. The firs t 8 p rim er sets th a t worked well determ ined w hich loci were exam ined. Consequently, there is a slight possibility th a t ou r results m ay contain nu ll alleles: alleles th a t produce no discernible am plified p roduct. If a given allele has been deleted in a gamete, the genotype of the offspring would appear to be homozygous for the single, complete, allele. If the m other contribu ted a nu ll allele, the offspring m ay no t m atch h e r genotype. If the fa the r con tribu ted a nu ll allele, we w ould expect the fa ther to have the allele observed in the offspring although in fact he could have any allele a t th a t locus. If nu ll alleles were com mon enough they would cause an apparen t surplus o f homozygotes in the population. I exam ined the da ta for instances in which p a te rn ity was excluded on th e basis of an allele th a t was homozygous in the offspring, and for instances where offspring d id no t m atch the ir m others a t a homozygous locus; and found none. There was no excess of hom ozygotes over th a t expected under Hardy-W einberg Equilibrium (see below and Appendix C, Tables 34, 35, and 36). The presence of null alleles does no t appear to be a p rob lem in this analysis. 67 Indiv iduals The 153 bears in our genetic sample com prised 76 males, 75 fem ales, and 2 bears of unknow n sex : 35 males were of breeding age (5 years o r older), and 45 females were of breeding age (5 years o r older). We obtained complete genotypes a t 8 loci fo r 152 ind iv idual bears (75 male, 75 female, 2 unknown). Of these, 57 were offspring whose m others were known. I was able to iden tify fa thers fo r 36; the rem aining 21 were sired by males th a t h ad n o t been sampled. Individual genotypes are p resen ted in the appendices. Two form ats are used; indiv idual alleles (Appendix A) w hich were used in calculating allele frequencies (Appendix B), and allele pairs o r genotype frequencies (Appendix C) which were used in calculating num ber of homozygotes and heterozygotes to determ ine Hardy- W einberg equilibrium and examine inter- and in tra-popula tion differences. Shared alleles Searches of the database were m ade fo r each indiv idual fem ale to iden tify all o ther individuals which shared a t least 8 alleles w ith th e ta rge t female. This m ay have use as a m easure of rela tedness a lthough alleles m ay no t be identical by descent. In the case of offspring whose m others were known, the search iden tified po ten tia l I L fathers. In the case of adu lt females, the search iden tified po ten tia l siblings, m others and fathers in some cases. Searches of the database were sim ilarly m ade fo r each ind iv idual m ale to identify all o ther individuals w hich shared a t least 8 alleles w ith th e ta rge t male. In the case of offspring whose m others were known, the search identified po ten tia l fathers. This search helped corroborate pa tern ity determ inations and provide an estim ate of male reproductive success. P ate rn ity 6 8 Patern ity cannot be absolutely determ ined; ra th e r po ten tia l fa thers can be excluded, resulting in a high probability of p a te rn ity fo r th e possible father. In o rder to a tta in a high probab ility of correctly identifying the father, accurate knowledge is needed of the m other/o ffsp ring relationships, and the genetic sam ple m ust include bo th of th e paren ts and the offspring (W etton et. al. 1987). The sam ple m ust be large enough to be represen tative of th e allele frequencies th roughout the population (Lynch 1988). In o rd e r to determ ine patern ity , we com pared th e offspring 's genotype w ith the m other's to determ ine which allele m ust have come from th e father. These are referred to below as pa te rna l alleles. Then the database was searched to find all individuals th a t sh a red 8 alleles w ith each offspring whose m o ther was know n from observation. All males 5 years old o r older a t the tim e the fem ale b red , th a t shared 8 alleles w ith the offspring, were then com pared 69 w ith offspring to determ ine if they shared all p a te rna l alleles. Two cases were found (families 1097 and 1464) w here 2 males could no t be excluded from patern ity . In 34 o ther cases, only one m ale was iden tified as the father. In the rem aining 21 cases no possible fa thers were found in the database. M ultiple Paternity Among mammals capable of having litters of m ore th an one offspring, th e re is the potential for litte r m ates to be sired by d ifferen t fathers. Multiple pa tern ity in such cases has been dem onstra ted in dogs (Robinson 1982), deer mice (Birdsall and Nash 1973), and o ther species. It has been speculated to occur among the Ursidae (Bunnel and Tait 1981), bu t h a rd evidence has been lacking. Two family groups of 3 offspring each exhibited m ore allelic varia tion in the F l generation than could have been con tribu ted by a single m ale given the m other's genotype. One group of 3 offspring exhibited 3 pa te rna l alleles a t each of 2 loci (Figure 2, Table 4) and searches of th e database excluded all males except #1453 as the fa th e r of #1480, excluded all males except #1453 o r# 14 9 1 as the fa th e r o f #1481, and excluded all sam pled males as the fa the r of #1482 (Figure 2a). Parsimony suggests th a t #1453 was the fa the r of bo th #1480 and #1481. Hypothetical male 5 (Hyp 5) represen ts a genotype deduced to explain offspring #1482 as well as offspring #1489 (fam ily 1437). 70 The o ther fam ily group had 3 paternal alleles a t a single locus (Figure 2, Table 5) and database searches excluded all males except #1712 as th e fa th e r of #1753 and #1755. All sam pled males were sim ilarly excluded as the fa ther of #1754 (Figure 2b). Three alleles a t a single locus from one litte r is clear evidence th a t m ore th an one m ale sired the offspring. In addition, offspring in a t least 2 o ther fam ily groups were identified as having d ifferent fathers, a lthough th e to ta l num ber of paternal alleles could have come from one hypothetical fa the r th a t was no t sampled (Figures 2c and 2d). M ultiple p a te rn ity was evident in one-th ird of the know n litters having 2 o r m ore cubs. 71 Figure 2. Pedigrees of four grizzly bear family groups exhibiting m ultip le patern ity . M aternal pedigrees were verified by field observations. Hypothetical males (Hyp 2, Hyp 5 and Hyp 6) rep resen t males for which we had no genetic sample; possible full o r partia l genotypes were deduced from the data set. a: Offspring #1481 could also have been sired by male #1491 (not shown). b: Male #1453 was excluded from pa te rn ity of offspring #1754 by only one allele. a c d 72 Table 4. Family genotypes for 1097's family (see Figure la ) . Each allele is designated by a le tte r identifying the locus, followed by a num ber represen ting the num ber of nucleotides in the PCR product which includes the m icrosatelhte repeat. Differences in size are due to varia tion in the num ber of tandem repeats. #1097 mother's genotype #1480 cub's genotype #1481 cub's genotype #1482 cub's genotype paternal alleles #1453 male genotype Hyp 5 male genotype Al 84 At 84 A l 8 4 Al 84 A l 8 4 Al 84 Al 94 Al 92 Al 94 A l 9 2 Al 94 A l 9 2 Al 94 Al 94 AXXX BI 56 BI 52 BI 52 BI 60 BI 52 BI 52 BI 58* BI 60 BI 60 BI 60 BI 64 BI 64 BI 52 BI 64 Cl 05 Cl 05 Cl 05 cm Cl 05 Cl 05 cm Cl 13 cm Cl 05 C113 cm cm CXXX D172 D172 D 172 D172 D172 D172 D178 D177 Dl 72 D 1 7 7 Dl 78 D178 D177 D182* . LI 55 LI 55 LI 59 LI 55 LI 55 LI 57 LI 55 LI 59 LI 57 LI 61 LI 55 LI 57 LI 61 LI 61 LI 57* M208 M208 M208 M208 M208 M208 M208* M208 M212 M208 M212 M212 M212 M212 Pt 53 P153 P153 P153 P157 P159 P139* P153 P159 P159 P157 Pl 59 P161 P157 X135 X 1 3 5 X133 X137 X133 X133 X131* X137 X 1 3 7 X135 X141 X 1 3 5 X 1 3 7 X141 X137 X141 Boldface ind icates an allele th a t m ust have come from a male. Italics ind icates e ither one or the o ther allele m ust have come from a male. Genotype XXX indicates any allele could be postu lated a t this locus. *an asterisk indicates an allele th a t has been used to explain o ther offspring in the population, i.e. hypothetical male #5 (Hyp 5) has a deduced genotype th a t fits as the sire of #1482 and ano ther offspring in the population, #1489 (m other #1437). 73 Table 5. Family genotypes fo r 1439's fam ily (see Figure lb ) . Legend and footnotes follow Table 4. #1439 mother's genotype #1753 cub's genotype #1754 cub's . genotype #1755 cub's genotype paternal alleles #1712 male genotype Hyp6 male genotype Al 80 Al 92 Al 84 Al 80 A184 Al 84 Al 80*Al 94 A194 At 94 Al 84 Al 92 Al 92 Al 84 BI 60 BI 60 BI 52 BI 58 BI 52 B158 BI 52 BI 64 BI 60 BI 60 BI 64 BI 58 BI 60 BI 60 BI 60* Cl 03 Cl 03 Cl 05 C 103 C 10 3 Cl 03 Cl 05* Cl 05 cm cm C 1 0 5 C 1 0 5 cm cm cm D172 D182 D172 D182 D172 D181 D172 D182 Dl 86 D172 D186 D186 D186 D186* LI 55 LI 55 LI 55 LI 55 LI 55 LI 55 LI 55* LI 55 LI 55 LI 57 LI 55 LI 57 LI 55 LI 57 M212 M208 M208 M214 M208 M208 M208 M214 M212 M212 M214 M214 M214 M214* P153 P157 P153 Pl 57 P157 P157 P157* P161 P161 P159 P161 P159 P157 P159 X137 X137 X135 X137 X135 X133 X135 X137 X141 X137 X141 X141 X141 X141* H ypothetical m ale #6 (Hyp 6) has a deduced genotype th a t fits as the sire o f #1754 and two o ther offspring in the population; #1707 (m other #1440) and #1485 (m other #1149). Probability The probab ility of detecting pa te rn ity varies w ith the num ber o f p a te rna l alleles contributed; these increase w ith the num ber of offspring sired. Allele frequencies were calculated from the d a ta 74 (Appendix B), and probability was determ ined by m ultiplying the allele frequencies necessary to explain patern ity , assum ing independen t assortm ent. Thus, the probability of pa te rn ity in ou r sam ple ranged from 2.3 xlO"4 to fa ther one cub w hich requ ired 7 of th e 8 m ost com m on alleles (#1702); to 3.5 xlO"12 to fa th e r th ree cubs which together requ ired 16 assorted paternal alleles (#1708, #1709, #1710). In the case of figures 2c and 2d, the p robability th a t cubs were sired by known males (as identified) is h igher (2.3 x 10-4, to fa th e r #1701, 8.9 x 10-5 to fa ther #1702, and 3.9 x IO"7 to fa th e r #1750), th an if a single, unsam pled, male h ad fa thered all: 3.9 x 10-6 in one case (figure 2c), and 1.9 x 10-10 in th e o the r case (figure 2d). In all cases, except two, I found only a single possible fa ther in ou r sam ple fo r any given offspring. In 7 cases o the r m ales were iden tified th a t d iffered a t I locus from the necessary genotype. In th e o th e r 30 cases o the r males d iffered by a t least 2 loci from the necessary genotype. Hypothetical males Twenty one of the genotyped offspring (FI generation) were*not fa th e red by any of the known males. Examination of these 21 F l genotypes revealed th a t several additional, unknown, males are necessary in o rder to account for the various alleles presen t. For example, one hypothetical genotype can be deduced for the fa ther of 75 5 add itional offspring from 3 different females (Table 6). At least 6 additional m ale genotypes are needed to explain all th e allelic varia tion in the rem aining 16 cubs. These 7 hypothetical males rep resen t one possible, minimal, grouping of pa ternal alleles. The deduced genotypes are p resen ted in Table 7. Table 6. Deduced genotype for hypothetical male no. I (Hyp I) Legend and footnotes follow Table 4. #1425 mother's genotype #1708 cub's genotype #1709 cub's genotype #1710 cub's genotype paternal alleles (Hypi) #1466 cub's genotype #1499 cub's genotype At 94 Al 80 Al 94 AT80 Al 80 A192t A184t At 94 At 94 Al 94 At 94 Al 94 At 94 Al 94 BI 48 BI 60 B 14 8 BI 40 B140 B150t B140f BI 60 BI 60 B 16 0 BI 60 BI 60 BI 60 BI 40 Cl 09 cm Cl 05 cm Cl 05 C105t Cl 05cm cm cm cm cm C105 cm D177 D172 D177 D172 D172 D178 Dl 72 D177 D177 D178 D177 D178 D184f D186t LI 55 LI 55 LI 55 LI 55 LI 55 . LI 55 LI 55 LI 55 LI 57 LI 57 LI 55 LI 57 LI 57 L171t M206 M206 M206 M208 M206 M206 M208 M208 M206 M206 M208 M208 M208 M214t P157 P151 Pl 49 P151 Pl 49 Pt 49 Pt 49 Pt 59 Pt 59 P157 Pl 57 P151 P161t Pt 51 X137 X137 X133 X137 X133 X137f X141t X137 X141 X137 X141 X141 X141 X141 t indicates an allele derived from the motiier. Hypothetical male #1 (Hyp I) has a deduced genotype th a t fits as the sire of #1708, #1709, #1710, and also fits offspring #1466 (m other #1425) and #1499 (m other #1454). 76 Table 7. Deduced genotypes of hypothetical males. Genotype XXX indicates th a t any allele could be postu lated a t this locus. Hyp 1 Hyp 2 Hyp 3 Hyp 4 Hyp 5 Hyp 6 Hyp 7 Al 80 Al 94 Al 92 AXXX A186 Al 94 Al 94 AXXX Al 94 AXXX Al 80 Al 84 Al 80 Al 94 BI 40 BI 60 BI 60 BXXX BI 40 BI 60 BI 48 BXXX BI 58 BI 64 Bi 52 BI 60 BI 40 BI 40 Cl 05 c m Cl 03 c m Cl 05 CXXX Cl 03 Cl 05 c m CXXX Cl 05 c m Cl 03 C113 D172 D178 D172 D182 D181 D186 D177 D184 D178 D182 D172 D186 D172 D186 LI 55 LI 57 LI 55 LI 57 LI 55 LXXX LI 61 LXXX LI 55 LI 57 LI 55 LI 57 LI 57 LI 59 M206 M208 M206 M208 M208 M214 M206 M208 M208 M212 M208 M214 M206 M214 P149 P151 P153 P155 P153 P161 P159 PXXX P139 P157 P157 P159 P153 P157 X133 X141 X131 X135 X129 X137 X135 X137 X131 X141 X135 X141 X133 XXXX Pedigrees The genetic sample included 30 family groups, each w ith a different, known m other. Nineteen fathers (including 7 hypothetical m ales as discussed above and under Reproductive success) were iden tified or deduced. The 30 pedigrees developed are discussed below in num erical o rder using the m other's identification num ber. 77 As discussed under Null alleles, the da ta set was exam ined fo r instances w here a po ten tia l fa ther was excluded from pa te rn ity on the basis of a single locus tha t was homozygous in the offspring. Males are identified in the following section which d iffered a t one locus from the pa te rna l genotype; all of these were com pared w ith offspring and none of the loci were found to be homozygous in the offspring. In discussions of pedigree results below, cub refers to cubs-of- the-year, and undesignated offspring refers to offspring fo r which no fa th e r was found in our genetic sample. Squares rep resen t males, circles rep resen t females, and triangles rep resen t unsam pled indiv iduals of unknow n sex. The first pedigree d iagram draw n is one in w hich I have identified genotypes for all individuals (except fo r those rep resen ted by triangles). In some cases a second pedigree is d raw n encom passing 2 generations. In these cases I have no t ob ta ined DNA samples for the previous generation although it m ay be possible to in the fu tu re using blood serum samples. Data on relationships and life histories accompanying these pedigrees are from Reynolds (1978,1980, 1989, 1991, 1992, pers. comm. 1994), Reynolds and Hechtel (1980, 1984), additional personal comments, field notes and unpublished m aterial of Harry Reynolds, and from m y own field notes and observations. 78 Figure 3. Family 1087. Born in 1989. 1096 Female #1087 was cap tured and m arked as a yearling in 1977 w ith h e r m other, #1086 and sibling #1164 (who was m arked in 1979); she was 16 years old in 1992. She was seen consorting w ith #1411 in 1988 and gave b irth to 2 female offspring in 1989. She was cap tu red in 1991 and her offspring were m arked as #1483 and #1484. She was cap tu red again in #1992 w ithout h e r offspring. Male #1098 was identified as the fa ther of bo th offspring: males #1420 and #1478 both differed at one locus, B, from the p a te rna l genotype of offspring #1483. Male #1098 was the only adu lt m ale in the sample tha t shared 8 alleles w ith #1484, all of w hich were paternal alleles. Two generations are known, as shown below, bu t no genotype has been obtained for #1086 o r #1164. Figure 4. 1087's ex tended family 79 Figure 5. Family 1089. Born in 1989. Female #1089 was m arked as a 4-year old in 1977 and was 19 years old in 1992. She was seen consorting with #1411 in 1988 and gave b irth to 3 offspring in 1989 which were cap tu red w ith h e r and m arked as #1704, #1705, and #1706. Only 2 offspring were observed w ith h e r as 2-year olds in 1991. Male #1411 was identified as the fa ther of all th ree offspring. Hypothetical male I was the only o ther male tha t shared 8 alleles w ith any offspring (#1706) bu t differed a t 3 loci from the paternal genotype. 80 Figure 6. Family 1095. Born in 1990. Female #1095 was m arked as a 6-year old in 1977, and was 21 years old in 1992. In 1989 she had 2 unm arked 3-year olds and was seen consorting with #1477 and an unm arked male. In 1990 she gave b irth to two cubs, one was m arked as #1727 in 1991, bu t the o the r rem ained unm arked. During the analysis a t Edmonton in O ctober 1993, male #1477 was tentatively identified as the father; only 6 of his loci had been clearly typed, bu t he was the only male in the sam ple th a t was a possible match. Using a new blood sample in January 1994 Dave Paetkau determ ined a complete genotype for #1477 which m atched all 8 paternal alleles in #1727. 81 Figure 7. Family 1097. Born in 1989. Female #1097 was cap tured and m arked as an 8-year old in 1977; she was 23 years old in 1992. In 1977 and in 1978 she was observed consorting with male #1096 (serum sample only available). In 1979 she gave b irth to 2 unm arked cubs which she lost, and was seen consorting with male #1081 and an unm arked male. In 1980 she h ad 2 unm arked cubs which she lost, and was seen consorting w ith males #1081, #1096, and #1172 (serum only). In 1981 she h ad 3 cubs which were cap tured and m arked as #1402, #1403, and #1404. They were observed with her in 1982, 1983, 1984, and 1985 w hen they were p robably weaned as 4-year olds. In 1987 she was observed consorting with an unm arked male. She was no t observed in 1988. In 1989 she had 3 cubs which were cap tu red and m arked as #1480, #1481, and #1482. In 1990 she was observed w ith 2 yearlings; #1480 and #1481. In 1991 she w eaned h e r 2-year olds and was observed consorting with an unm arked male. In 1992, 2 3-year olds, male and female, were cap tu red and identified from faint lip tattoos as #1480 and #1481. W hen samples 8 2 from these anim als were genotyped, they differed from #1097 's genotype a t several loci, while the sample from #1482, collected as a cub, m atched #1097 ,s. We obtained blood samples stored in Fairbanks th a t h ad been collected from #1480 and #1481 as cubs and w hen they were genotyped they m atched the m other's. The sam ples originally though t to be #1480 and #1481 were subsequently labeled Unknown 3 (UNK 3) and Unknown 4 (UNK 4) Five adu lt males shared 8 alleles w ith #1480. #1453 was the only one th a t h ad the 8 necessary paternal alleles; #1459 differed a t 2 loci, B, and P; #1081 differed at 3 loci, A, B, and P; #1157 differed a t 3 loci, B, M, and P; and #1463 differed a t 5 loci, A, D, L, M, and P. Two adu lt males shared 8 alleles w ith offspring #1481. Both males #1453 and #1491 had all 8 necessary p a te rna l alleles; bo th are equally p robable fathers, b u t #1453 is a m ore parsim onius choice since he is already proven as having successfully b red w ith #1097. Three adu lt males shared 8 alleles w ith offspring #1482. Male #1124 lacked the necessary paternal alleles a t loci B, M, and P. Two o thers were 5-year old males; #1469 differed a t 4 loci, B, C, M, and P; and #1470 differed a t 5 loci, A, B, C, M, and X. Hypothetical male 5 (see Hypothetical males page 72, and Hyp 5 page 74) was deduced as th e fa th e r of #1482. 83 Figure 8. Family 1125. Born in 1988. Female #1125 was m arked as a 3-year old in 1977 and was 18 years old in 1992. In 1988 she gave b irth to at least 3 cubs. In 1991 she had 3, 3-year olds, one of which was cap tu red and m arked as 1726. The original DNA sample of #1125 used for m icrosatellite am plification was isolated from ear tissue th a t was first sen t to A llendorfs lab in Missoula. The label on the sample vial was partia lly obscured and was recorded as #1125. However, the genotype derived from this sample d id no t m atch the cub, #1726 a t 3 loci: C, D, and L. Mismatch at one loci could possibly be attributable to m utation , bu t this great a discrepancy could only be due to m islabeling. We labeled the original genotype Unknown #2 (UNK2) and sen t a blood sample from #1125 to David Paetkau in January 1993 to genotype. Male #1157 was the only adu lt male th a t shared 8 alleles w ith cub #1726, bu t w ithout the m other's genotype we cou ldn 't assign paternity . A com plete genotype for the m other was determ ined from the new blood sam ple in January 1994 which did m atch w ith the offspring, #1726. Male #1157 was excluded as a fa th e r a t 5 loci; A, B, L, M, and X. No o ther adu lt males were found to share 8 alleles. By adding the necessary alleles to the genotype of hypothetical m ale 7 (Hyp 7, page 74), I was able to explain offspring #1726 as well as offspring #1743 (family 1461). 84 85 Figure 9. Family 1136. Born in 1987. Female #1136 was m arked as a yearling in 1977 w ith her m o ther #1134 and 2 siblings, #1135 and #1137; she was 16 years old in 1992. She was weaned in 1979 and gave b irth to 2 cubs in 1987. Her offspring were m arked in 1988 as #1469 and #1470. Male #1124 was the only adu lt male th a t shared 8 alleles w ith the offspring, and he was identified as the father of both. Figure 10. 1136's ex tended family. 86 Figure 11. Family 1141, bom in 1989. Female #1141 was born in 1978 and was m arked then w ith h e r m other, #1139 and sibling #1140; she was 14 years old in 1992. She was w eaned in 1980 and gave b irth to an unm arked cub in 1986 w hich she lost. In 1988 she was seen consorting w ith #1456 and an unm arked male. In 1989 she gave b irth to one cub which was cap tu red and m arked as #1485. The original DNA sam ple used for m icrosatellite am plification was isolated from ear tissue th a t was first sent to A llendorfs lab in Missoula. The label on the sample vial was dam aged and was recorded as #1141. However, the genotype derived from this sample did not m atch her cub at 2 loci: D, and X, and was identical to the genotype of #1147. We a ttribu ted this d iscrepancy to mislabeling and sent a new blood sam ple of #1141 to David Paetkau for genotyping. Male #1712 was the only adu lt male th a t shared 8 alleles w ith cub #1485, bu t w ithout the m other's genotype we cou ldn 't assign pa tern ity . After a com plete genotype of female #1141 was determ ined in January 1994 by David Paetkau, male #1712 lacked only one paternal allele at locus C. Because of the slight possibility of 87 m utation in the m icrosatellite loci, potential males such as #1453 th a t differ a t a single locus can be considered as a rem ote possibility for sires. I was able to adjust the genotype of hypothetical male 6 (Hyp 6 page 74) to include alleles necessary to explain the genotype of offspring #1485 so Hyp 6 is the putative fa ther a lthough there is a possibility th a t #1712 may be the sire (see M utation above). Two generations of females are known, bu t no genotype has been ob tained for #1139. Figure 12. 1141's ex tended family 88 Figure 13. Family 1149. Bom in 1988. 1124 r I486 Female #1149 was m arked as a 4-year old in 1978 and was 18 years old in 1992. She gave b irth to 2 offspring in 1988 which were m arked as #1486 and #1487. Both cubs were w eaned as 3-year olds in 1991. Male #1124 was the only adu lt male th a t shared 8 alleles w ith e ithe r of the cubs, and he was identified as the fa ther of both. 89 Figure 14. Family 1166. Bom in 1989. Female #1166 was m arked at 10 years of age in 1979 and was 23 in 1992. She had 3 unm arked cubs in 1981, one of which survived until 1983. In 1986 she had a single cub whose fate is unknow n. In 1989 she gave b irth to at least 2 cubs who were m arked as #1701 and #1702 in 1990 when they were yearlings. Male #1478 was the only adu lt male tha t shared 8 alleles w ith offspring #1701 and he was identified as the father. Male #1421 was identified as the fa ther of #1702. Male #1712 shared 8 alleles w ith offspring #1702, bu t differed in paternal alleles at one locus; A. 90 Figure 15. Family 1174. Bom in 1989. Hyp3 1487 Female #1174 was m arked as a cub in 1980 w ith h e r m o ther #1105 and sibling #1173; she was 12 years old in 1992. She was w eaned in 1985 and was seen consorting w ith #1147 in 1987. She h ad no offspring in 1988 and was seen consorting with an unm arked male. In 1989 she gave b irth to a t least one cub, which was m arked as #1497 as a yearling in 1990. In 1991 she w eaned h e r 2-year old and b red w ith an unm arked male. Two adu lt males shared 8 alleles with offspring #1497. Male #1462 differed from the paternal allele necessary a t one locus, D, and can be considered as a rem ote possibility as a sire. Male #1378 differed at 2 loci, D and M. Hypothetical male 3 (Hyp 3, page 74) was deduced as a genotype to explain offspring #1462 and offspring #1498 and #1500 (family 1454). Figure 16. 1174's ex tended family. © _ ? Hyp3 ------^ 1 1 7 4 ^ ------ 1173 1487 91 Figure 17. Family 1177, Born in 1989. Female #1177 was bom in 1979 to #1104 and was m arked as a yearling in 1980; she was 13 years old in 1992. She was w eaned in 1981. In 1985 she was seen consorting with an unm arked male, bu t w asn 't observed in 1986. In 1987 she was seen consorting w ith #1083 , bu t w asn 't observed in 1988. In 1989 she gave b irth to a t least one cub who was m arked #1496 as a yearling in 1990. She w eaned #1496 as a 2-year old in 1991 and was seen consorting w ith #1147. Only one adu lt male (#1411) shared 8 alleles w ith offspring #1496; all of these were paternal alleles and he was iden tified as the father. Figure 18. 1177's ex tended family. 92 Figure 19. Family 1179. Bom in 1989. Female 1179 was born to #1178 in 1978, and was m arked as a 2-year old in 1980; she was 14 years old in 1992. In 1983 she was seen consorting with #1263, but was no t observed from 1984 th rough 1987. In 1988 she was located in the Noatak River drainage abou t 40 km south of the study area, over the crest of the Brooks Range. In 1989 she was observed in the study a rea w ith one unm arked cub. She was no t observed in 1990 , bu t was again located in h er Noatak River home range in 1991. She was cap tured in the study area in 1992. A 3-year old female, #1742 was cap tu red and m arked separately in 1992. In the spring of 1993 , bo th females were located in the Noatak area, in association with each o ther. Four adu lt females were found to share 8 alleles w ith #1742, bu t only one adu lt male was found; #1411. If he is assum ed to be the father, #1179 is the only adu lt female tha t m atches the necessary m aternal alleles in #1742. Two adu lt females d iffered a t one locus: #1136 at locus C and #1437 at locus B. The o ther adult fem ale differed at loci C and D from the m aternal alleles. It seems h ighly p robab le th a t #1742 is the offspring of #1179 th a t was bom 93 in the study area in 1989. After weaning h e r cub #1179 re tu rn ed to the Noatak drainage. In the sum m er of 1992 she revisited the study area and led h e r daughter south to the Noatak tha t fall. Figure 20. 1177's ex tended family. 1411 Q 94 Figure 21. Family 1424. Bom in 1987. Hypl 1466 Female # 1424 was m arked as an 8-year old in 1986; she was 14 years old in 1992. In 1987 she had one cub which was marked as #1466. She weaned #1466 as a 2-year old in 1989 and was seen consorting w ith #1147. In 1990 she gave b irth to 2 unm arked cubs th a t she lost the same year. In 1991 she gave b irth to a single unm arked cub. Her male offspring, #1466 did no t share 8 alleles w ith any males in ou r sample. A hypothetical male (Hyp I , page 74) whose genotype was deduced from the 3 offspring of #1425 (see next pedigree) was found to also explain the paternal alleles in #1466. 95 Figure 22. Family 1425. Bom in 1989. Female #1425 was m arked as a 7-year old in 1986; she was 13 years old in 1992. She had 2 offspring in 1987 which were m arked as #1426 and #1427. No genetic samples were taken of them and they d ied la te r th a t summ er. In 1987, female #1425 was seen consorting w ith #1147 and an unm arked male, bu t she was no t observed in 1988. She gave b irth to 3 cubs in 1989 th a t were cap tu red and m arked as #1708, #1709, and #1710. They rem ained w ith h e r th rough 1991. No adult males were found to share 8 alleles w ith #1708. One male, #1459 shared 8 alleles w ith #1709 bu t d iffered a t 3 loci from the necessary paternal alleles; C, D, and P. One m ale #1124 shared 8 alleles w ith #1710 bu t lacked the necessary paternal alleles at 5 loci: A, B, D, P, and X. The paternal alleles necessary to explain all 3 offspring described a com plete genotype of 16 alleles. This genotype was called Hypothetical male I (Hyp I, pages 73 and 74) as discussed earlier, and was en tered into the database. It was found to explain 2 o th e r undesignated offspring in the study area: #1466 and #1499 and is probably the genotype of an actual unsam pled male. 96 Figure 23. Family 1437. Bom in 1990. Female #1437 was m arked as a 9-year old in 1987, she was 14 in 1992. She had a single unm arked cub in 1987 which she lost la ter th a t year. In 1989 she was seen consorting w ith an unm arked male and she gave b irth to 2 cubs in 1990 which were cap tu red and m arked as #1488 and #1489. They were both seen w ith h e r in 1991 as yearlings. No adu lt males were sampled tha t m atched e ither offspring a t 8 alleles. In determ ining hypothetical genotypes fo r males, I used com binations of alleles th a t would explain all the undesignated offspring w ith as few males as possible. This requ ired a genotype (Hyp 2, page 74) th a t explained 4 offspring, one of which was #1488. A single male genotype (Hyp 8) could be deduced th a t explains both offspring #1488 and #1489 with one fa ther (which is m ore probable in this family), bu t explains only one o ther offspring in the sample. In o rd e r to use a m inim um num ber of males for calculating reproductive success and ratio of breeding males to breeding females, Hyp 2 and Hyp 5 (page 74) were used as the fathers in previous discussions. 97 Figure 24. Family 1438. Bom in 1991. Female #1438 was m arked as a 13-year old in 1987; she was 18 years old in 1992. She had 3 unm arked 2-year olds when she was first cap tured , which would be 9 years old in 1992. In 1991 she h ad a t least 2 unm arked cubs . These were cap tu red w ith her and m arked as yearlings in 1992; #1756 and #1757. Male #1420 was the only adu lt male tha t shared 8 alleles with e ither cub; all of which were paternal alleles. 98 Figure 25. Family 1439. Bom in 1989. Female #1439 was cap tured and m arked as a 14-year old in 1987; she was 19 years old in 1992. When she was first cap tured , she h ad 3 unm arked 2-year olds, which would be 7 years old in 1992. In 1989 she had a t least 3 cubs tha t were observed in 1990 as yearlings. They were cap tured w ith h e r and m arked in 1992 as 3-year olds; #1753, #1754, #1755. Male #1712 was identified as the fa ther of #1753 and #1755; he was the only adu lt male to share 8 alleles with #1755. Three o th e r males shared 8 alleles w ith #1753: #1421 d iffered a t a single locus, M. Both males #1098 and #1478 lacked four pa ternal alleles a t loci A,D,P, and X. Five adu lt males shared 8 alleles w ith #1754 bu t none m atched the pa te rna l genotype: #1453 differed from the necessary paternal alleles a t a single locus, X. Male #1463 lacked the pa te rna l alleles a t loci C,L, and P, #1081 lacked alleles at B,M,P, and X, #1157 lacked alleles at A,B,P, and X, and #1459 lacked paternal alleles at loci A,B,M, and P. Hypothetical male 6 (Hyp 6, page 74) was deduced as a genotype tha t explains the paternal alleles in #1754 and #1707 99 (see fam ily 1440 below). Because of the slight possibility of m u ta tion in the m icrosatellite loci, potential males such as #1453 th a t differ a t a single locus can be considered as a rem ote possibility fo r sires. There were no 7-year old bears th a t shared 8 alleles w ith fem ale #1439 and m ight be h e r offspring from 1987. 100 Figure 26. Family 1440. Bom in 1989. Female #1440 was cap tured and m arked as a 14-year old in 1987; she was 19 years old in 1992. She may be the m o ther of #1449 and #1450 who were cap tu red th a t year as yearlings, bu t d ied in a den the following year. Only serum samples are available from #1449 and #1450. Female #1440 was seen consorting with an unm arked m ale in 1987 and w ith male #1459 in 1988. In 1989 she gave b irth to a single cub th a t was m arked as #1707. He was w eaned in 1991 as a 2-year old and #1440 bred with #1459. Males #1147 and #1405 both shared 8 alleles w ith offspring #1707; #1147 lacked 2 paternal alleles (A and C), and #1405 lacked 3 pa te rna l alleles (A, P, and X). Hypothetical male 6 (Hyp 6, page 74) was deduced to explain offspring #1707 as well as #1485 (family 1141) and #1754 (family 1439). 101 Figure 27. Family 1454. Bomin 1989. Hypi O Hyps 1409 ^ 500^ Female #1454 was m arked as a 10-year old in 1988; she was 14 in 1992. She was seen consorting with male #1453 in 1988 and gave b irth to 3 cubs in 1989. They were cap tu red w ith h e r as 2- year olds and m arked in 1991 as #1498, #1499, and #1500. The m o the r and h e r th ree offspring are the only bears sam pled w ith the ra re allele L I73. No adu lt males were sampled tha t m atched any of the offspring a t 8 alleles. Hypothetical male I (Hyp I, page 74), whose genotype was deduced from family 1425, m atched the paternal alleles of offspring #1499. Hypothetical male 3 (Hyp 3, page 74) was therefore deduced to explain the paternal alleles from #1498 and #1500. This genotype also explains offspring #1497 from family 1174. 1 0 2 Figure 28. Family 1457. Bom in 1990. 1421 Female #1457 was cap tured and m arked as a 10-year old in 1988; she was 14 years old in 1992. In 1990 she gave b irth to at least 2 unm arked cubs. In 1991 she was observed w ith 2 yearlings which were m arked in 1992 as 2-year olds; #1731 and #1732. Offspring #1732 was found to have a rare allele, M222, th a t occurred only once in our sample in h e r genotype of M 212/M 222. Her genotype d id no t m atch her m other's genotype of M 218/M 218 a t this locus (see discussion on m utation). It seems likely th a t this allele m ay be due to a length m utation in the m other's gamete. Male #1421 was the only adu lt male tha t shared 8 alleles w ith e ithe r of the offspring, all of which were paternal alleles. His genotype a t the M locus was M212 /M 214 which would explain one of #1732 's alleles (M212) and add support to the supposition tha t h e r ra re M222 allele came from her m other. The sibling, #1731 had com m on alleles at the M locus (M214/M218). 103 Figure 29. Family 1458. Bom in 1989. 1124 _L 1484 Female #1458 was m arked as a 7-year old in 1988; she was 11 years old in 1992. In 1988 she was seen consorting w ith male #1459. In 1989 she gave b irth to 2 cubs who were m arked as #1494 and #1495. They were seen w ith her in 1990 and were w eaned, perhaps tem porarily, as 2-year olds in 1991. Female #1458 consorted with #1459 in 1991 and was la ter seen reun ited w ith h e r offspring, both of whom were cap tured th a t season. Despite h e r frequen t association with male #1459, he was no t iden tified as a possible father. Male #1124 shared 8 pa te rna l alleles w ith bo th offspring; male #1433 shared 8 alleles w ith #1495 bu t d iffered a t loci C, M, and P from the paternal genotype. 104 Figure 30. Family 1460. Bom in 1989. Female #1460 was cap tured and m arked as an 8-year old in 1988; she was 12 years old in 1992. She consorted w ith male #1459 in 1988 and in 1989 gave b irth to 3 cubs. She was cap tu red w ith 2 yearlings in 1990 which were m arked as #1492 and #1493. They were bo th recap tu red w ith h er in 1991. Male #1712 was identified as the fa ther of #1492 in October 1993. Offspring #1493 was not clearly genotyped a t th a t time a t loci M and P and differed from #1712 at those loci. Two o ther males shared 8 alleles w ith offspring #1492; #1147 who d iffered at loci B, C, and X, and #1490 who differed a t loci A and C. In January 1994 a com plete genotype of #1493 was ob tained from a new blood sample by David Paetkau and #1712 was determ ined to be the father. Despite he r association w ith male #1459 (see also previous pedigree), he was no t identified as a possible father. 105 Figure 31. Family 1461. Bom in 1990. Female #1461 was m arked as a 10-year old in 1988; she was 14 years old in 1992. She was observed consorting w ith male #1459 and an unm arked male in 1988. In 1989 she was no t observed. In 1990 she gave b irth to a t least 2 cubs tha t were cap tu red w ith h e r and m arked as 2-year olds in 1992; #1743 and #1744. No fa ther was identified for #1743 although 3 adu lt males shared 8 alleles. Male #1459 differed a t only one paternal locus, D, from offspring #1743; male #1411 differed a t 2 loci, B and M, and m ale #1712 differed at 3 loci, A, L, and M. Hypothetical male 7 (Hyp 7, page 74) was deduced as the possible father; this genotype also explains offspring #1726 (family 1125). Originally #1744 was not completely genotyped; loci B, D, and P were no t am plified. A nother blood sample was sent to Dave Paetkau in Edm onton in January 1994, and he was able to determ ine a com plete genotype. Male #1459 was identified as th e father. In this case, since #1459 was only one allele away from being identified as the fa the r of #1743, it is possible tha t a m utation m ay have occurred in one of #1459 's gametes (locus D), and tha t he is actually the fa ther of both offspring (see M utation above). 106 Figure 32. Family 1464. Bom in 1989. Female #1464 was cap tured and m arked as a 7-year old in 1988; she was 11 years old in 1992. In 1989 she gave b irth to 3 cubs. All 3 offspring were seen with her in 1990, and were cap tu red w ith h e r in 1991 and m arked as #1713, #1714, and #1715. Male #1463 shared 8 paternal alleles with all th ree offspring. He was the only adu lt male to share 8 alleles w ith #1713 and #1715. Male #1157 also shared 8 paternal alleles w ith #1714 and is ju s t as p robab le a fa ther as #1463. However, male #1463 is a m ore parsim onious choice since he has already been identified as the fa the r of the o ther two cubs in this litter. 107 Figure 33. Family 1479. Bom in 1992. Hyp2 1756 Female #1479 was cap tured and m arked as a 6-year old in 1989; she was 9 years old in 1992. In 1989 she was observed consorting w ith male #1478. In 1990 she was seen consorting w ith m ales #1405 and #1491. In 1991 she was seen consorting w ith m ale #1712. In 1992 she gave b irth to a single cub th a t was cap tu red w ith h e r and m arked as #1758. Three adu lt males shared 8 alleles with offspring #1758. None of them m atched all the paternal alleles. Male #1463 differed a t only one locus; D. Male #1491 differed at 4 loci; C, D, L, and M. Male #1103 differed at 6 loci; C, D, L, M, P, and X. Hypothetical male 2 (Hyp 2, page 74) was deduced as a genotype tha t explains 4 undesignated offspring; #1758, #1488 (family 1437), and #1751 and #1752 (fam ily 1749). It is interesting tha t although #1479 was seen consorting w ith 4 different males during previous field seasons, all of whom have been genotyped; none of them was the father. Because of the slight possibility of m utation in the m icrosatellite loci, po ten tia l males such as #1463 tha t differ a t a single locus can be considered as a possibility for sires. 108 Figure 34. Family 1716. Bom in 1991. Female #1716 was cap tured and m arked w ith h e r 2 cubs, #1717 and #1718, as an 8-year old in 1991; she was 9 years old in 1992. Her m other is unknown, although she shares 8 alleles w ith #1089 and #1413, both of whom could be her m other. Her offspring, #1717 and #1718 were both dead in 1992. Four adu lt males shared 8 alleles with #1717; only one male, #1081 shared 8 alleles w ith #1718. He was identified as the fa ther of bo th offspring. Male #1459 differed from the paternal alleles a t 2 loci; D, and X. Male #1466 differed a t 3 loci; A, M, and P. Male #1477 was only genotyped at 6 loci originally, bu t he differed a t one of those (L). W hen a complete genotype was obtained, he differed a t 3 loci (L, M, and X). The first sample of #1717 analyzed did no t m atch the m other at 3 loci. A nother DNA sample from a separate extraction was available and was found to m atch the m other at 8 loci. A pparently the first sam ple was m islabeled at some poin t (see Results: DNA extraction). It was subsequently re-labeled Unknown #1 (UNK I). 109 Figure 35. Family 1734. Bom in 1990. Female #1734 was cap tured and m arked as a 13-year old in 1992 w ith two 2-year olds; #1735 and #1736. She was cap tu red in the m ountainous southern region of the study area which is often difficult to work in because of poor weather conditions, and is therefo re no t as well sampled as the rest of the area. Male #1712 was the only adu lt male to share 8 alleles w ith offspring #1735, and was identified as the father of bo th offspring. Male #1729 also shared 8 alleles w ith offspring #1736 bu t lacked the necessary paternal alleles at loci P and X. HO Figure 36. Family 1739. Born in 1991. Female #1739 was cap tured and m arked as an 8-year old in 1992 w ith 2 yearlings, #1740 and #1741, in the sou thern p a rt of the study a rea n ea r the location of family #1734. Male #1081 was the only adu lt male to share 8 alleles with either offspring, all of which were p a te rna l alleles. A 26-year old female, #1737 was also cap tu red fo r the first tim e nearby. She was the only o lder female found to share 8 alleles w ith fem ale #1739 and male #1712 and may be the m other of e ither or both. I l l Figure 37. Family 1745. Bom in 1992. Female #1745 was cap tured and m arked as a 5-year old in 1992 w ith 2 cubs, #1746 and #1747. No males were found tha t shared 8 alleles w ith e ither of the offspring. Hypothetical male 4 (Hyp 4, page 74) was deduced as a genotype to explain both offspring. This genotype did no t explain any o ther undesignated offspring. The 2 cubs, who weighed only 3.8 kg, became separated from th e ir m other la ter during the field season and had no t been reun ited by the tim e we left the area. 1 1 2 Figure 38. Family 1749. Bom in 1991. 1096 O Hy p2 1752 Female #1749 was cap tured and m arked as a 15-year old in 1992 w ith 3 yearlings, #1750, #1751, and #1752. She was also caught in the sou thern part of the study area. Male #1098 was the only adu lt male to share 8 alleles, all paternal, w ith offspring #1750. No adu lt males in the sample shared 8 alleles with offspring #1751. Four adu lt males shared 8 alleles with offspring #1752: male #1098 d iffered from the paternal genotype at 2 loci, A and X; male #1420 differed at 2 loci, A and P; male #1459 differed a t 3 loci, A, D, and P; and male #1491 differed a t 4 loci, D, L, M, and P. Hypothetical male 2 (Hyp 2, page 74) was deduced to explain offspring 1751 and #1752. This pa ternal genotype also explains undesignated offspring #1758 (fam ily 1479) and #1488 (family 1437) 113 Male reproductive success Of the 36 cases of known parentage, relative reproductive success calculations (Table 8) indicated tha t 12 males (plus a t least one unsam pled male), or 39% of the 33 males over 5 years of age, were responsib le for all of the patern ity . However, the rem aining 21 genotyped offspring (FI generation) were no t fa the red by any of th e known males. Examination of these 21 F l genotypes revealed th a t several additional, unknown, males are necessary in o rder to account fo r the various alleles p resen t (see Hypothetical males, page 72). For example, one hypothetical genotype was deduced fo r the fa th e r of 5 additional offspring from 3 different females (Table 6). At least 6 additional males are needed to explain all the allelic varia tion in the rem aining 16 cubs. These 7 hypothetical males rep resen t one possible, m inimal, grouping of p a te rna l alleles. If we include them in ou r calculations as a m inim um num ber of sires we can m ore closely estim ate male reproductive success (Table 9). The estim ated reproductive success of each individual is lower, and 19, o r 57.6% of the 33 breeding-age males (over 5) are responsible for all p a te rn ity (bu t see below). If we include the hypothetical males in ou r pedigrees, th en a t least 7 of 21, o r one-third, of possible litters have m ultip le sires. From this data, no male has sired m ore than 11% of known offspring. If we include the possibility of m uta tion (see M utation) no male has sired m ore than 12.3% of all offspring. 114 Table 8. Relative reproductive success of known fathers. Age is given as th e earliest date of successful breeding. Reproductive success in this example is m easured by all offspring surviving to be sampled. Male number A ge when breeding Total offspring Percent of known offspring 1124 26 6 16.67 1712 14 6 16.67 1411 10 . 5 13.89 1081 18 4 11.11 1098 14 3 8.33 1463 9 3 8.33 1420 11 2 5.56 1453* 18 2 5.56 1421 15 2 5.56 1459 18 I 2.78 1478 11 1 2.78 1477 9 1 2.78 1103 23 0 0.00 1147 17 0 0 .00 1157 19 0 0.00 1165 16 0 0.00 1233 20 0 0.00 1405 15 0 0.00 1455 11 0 0.00 1456 14 0 0.00 1476 11 0 0.00 1491 19 0 0.00 1703 16 0 0.00 1720 17 0 0.00 1729 15 0 0.00 Total 36.0 100.00 m ean variance 0.97 3.03 * Male #1491 (age 19) m ay have sired one of the cubs a ttrib u ted to #1453. 115 Table 9. Relative reproductive success of the m inim um possible num ber of fathers. Reproductive success in th is example is m easured by all offspring surviving to be sam pled. Male , number Total offspring Percent o f known offspring 1124 6 10 .53 1712 6 10 .53 1411 5 8 .7 7 H y p l 5 8 .7 7 1081 4 7 .0 2 Hyp 2 4 7 .0 2 1098 3 5 .2 6 1463 3 5 .2 6 Hyp 3 3 5 .2 6 Hyp 6 3 5 .2 6 1420 2 3 .51 1453* 2 3.51 1421 2 3 .51 Hyp 4 2 3.51 Hyp 5 2 3.51 Hyp 7 2 3.51 1459 1 1 .75 1478 1 . 1 .7 5 1477 I 1 .7 5 13 other sam p led m a les 0 0 .0 0 (over 9 y ea r s old) ( s e e previous table) 14 un sam p led m a le s f 0 0 .0 0 Total 57 100 .0 0 m ean 1 .24 variance 3 .2 5 * Male #1491 (age 19) m ay have sired one of the cubs a ttrib u ted to #1453. t A dditional breeding-age males (over 9 years old) based on the num ber of hypothetical (successful, bu t unsam pled) males in the population. 1 1 6 We have observed breeding behavior in males as young as 5 years old. Only 39%, of males a t least this old, th a t we have genotyped actually p roduced offspring. The 7 successful (hypothetical) males rep resen t 18 unsam pled males o f breeding age (over 5). However, we d id no t have genetic samples from all males o f b reeding age. The youngest of the males we genotyped were 9 years o ld a t the tim e they successfully b red (Table 8). To take a conservative approach, then, we can consider 9 years and o lder to be breed ing age fo r males, there are 25 in ou r study area sample, and a m axim um of 13 /25 , or 52%, are successful. If we extrapolate one step fu rthe r, the 7 hypothetical males th a t also b red could rep resen t only 52% of th e unsam pled males, o r a to tal of 13.5 unsam pled males 9 years and older. This gives us a conservative estim ate of 25 +14 o r 39 to ta l males of breeding age (9 or older) in the population . A m inim um of 19 of these (49%) males over 9 years of age are successful. A sim ilar approach using 5 years o r o lder gives a figure of 33 + 18 o r 51 total males of breeding age. A m inim um of 19 of these (37%) are successful. If we accept th e lower figure of 39 breeding-age males, and com pare the annual m ean of 41 breeding-age females th a t have been observed in the study area (Reynolds 1992), th e re is essentially a 1:1 sex ra tio of breeding-age adults; if we assum e th a t the genetic sam ple is represen tative of the sex and age ratios in the actual popu la tion (as it appears to be). To use ano ther approach; including hypothetical males, the ra tio of successful breeding males to breeding females is 19 to 30 117 (1:1.58) in ou r genetic sample. Extrapolating from this ra tio gives us a to ta l of 26 males th a t would successfully breed w ith the 41 total breeding-age females observed annually on average. If these 26 successful males rep resen t 39% of all breeding-age m ales (5 years o r older) we have a to tal of 67 breeding-age males (41 unsam pled) in the population. If they rep resen t 52% (9 years o r older), th e re are 50 breeding-age males (24 unsam pled). In sum m ary, if we consider only males 9 years and o lder to be of b reed ing age, th en there are 39 to 50 males associated w ith the study a rea popula tion of 41 breeding-age females, b u t only abou t 26 of them are reproducdvely successful (52%-67%). Thus, the re are m ore breeding-age males associated w ith ou r sam ple of breeding-age females th an aerial census d a ta indicate. This m ay be explained by the fact th a t males have larger hom e ranges, less fidelity to hom e ranges, and often travel widely. Many, if no t all, of the unsam pled breeding males m ay have u tilized the study a rea as only a p a r t of the ir hom e range o r m ay have passed th rough as transien ts. Because m any of the males we have sam pled m ay have sired offspring outside the study area, th e lower end of th e rep roductive success estimates is no t accurate. However, the m ales iden tified as having the m ost offspring were m ales who cen tered m ost of th e ir activities w ithin the study a rea during the sampling period. The upper end of the reproductive success calculations (about 11% of the total) is therefore a reasonable estim ate. These estim ates of male reproductive success are concerned only w ith known m other/offspring relationships. 1 1 8 W estern Brooks Range population Since m ost females tend to rem ain faithful to n a ta l areas, establish hom e areas adjacent to the ir m others, and rem ain there th roughou t the ir lives, we propose to define our study population based upon a discrete num ber of breeding adu lt females. The area in w hich th e popula tion lives is defined by the hom e ranges of these adu lt females. The num ber of breeding males in the population consists of all males which successfully b reed w ith those females, regard less of w here they spend m ost of th e ir time. If we know the breeding females and breeding males, we can begin to define the genetic effective population. Allele frequencies Allele frequencies were calculated as described in METHODS, D ata analysis, from all the samples available from each population. A to ta l o f 2,432 alleles were enum erated in the WBR population . These frequencies are p resen ted in Appendix B. Genotype frequencies h i a large, idealized, random ly-m ating population, th e allele frequencies described in Appendix B should p red ic t the frequencies of genotypes, o r allele pairs, found throughout the population . If th e re is random m ating with respect to genotype, a gene w ith j 119 alleles shou ld exhibit genotype frequencies in Hardy-W einberg equilibrium : Pi^ fo r AiAi homozygotes 2pipj fo r AiAj heterozygotes (Hard 1988) This m odel also assumes th a t the alleles are selectively neu tra l and there is negligible m igration, insignificant m utation and nonoverlapping generations. However, Hardy-W einberg frequencies a re rela tively insensitive to departu res from m ost of these assum ptions (large population size, nonoverlapping generations, negligible m igration, insignificant m utation, and selective neutrality) so th a t Hardy-W einberg frequencies are often valid even if some of these assum ptions are no t m et (Harti 1988). A lthough the occurrence of Hardy-Weinberg proportions does no t act as p roof th a t these assum ptions are all valid, a departu re from these frequencies can indicate th a t some assumptions, particularly those of random m ating, are no t met. Accordingly, I exam ined the frequency of genotypes as p red ic ted by allele frequencies a t each locus (Appendix C) and de term ined using a Chi-square test th a t they d id exhibit Hardy- W einberg equilib rium a t all loci except for locus L w here th e re was an excess of heterozygotes of one genotype: L159/L161. The animals w ith th is genotype were #1125, #1166, #1167, #1456, #1468, #1481, #1491, #1726, #1733, #1745, and #1747. The only known rela tionsh ips among these individuals are between #1745 and #1747 (m o ther/o ffsp ring ). Any departu re from expected frequencies due 120 to popula tion genetic structure should cause an excess of hom ozygotes (instead, of the observed excess of heterozygotes observed a t th is locus), so this sole discrepancy is p robab ly due to chance. Population subdivision Because we have observed th a t female offspring tend to rem ain n ea r th e ir m aternal hom e range after weaning, and in some cases to establish an overlapping hom e range, we w ould expect th a t over time, genetic subdivision of the population would occur: ad jacen t females will be m ore closely re la ted th an d is tan t ones. If m ales also tend to b reed w ithin restric ted areas we w ould expect th a t over tim e homozygotes would accumulate in local areas and th e re would be a deficiency of heterozygotes (or W ahlund effect) w hen the popula tion is viewed as a panm ictic unit. W ithin the W estern Brooks Range population expected heterozygosity (H) was calculated for each genotype as pi2 a t each locus. Average expected H was calculated as 1-2 pi2. Observed H was calculated by enum erating all heterozygotes and homozygotes a t each locus from the database. Differences between expected and observed H were exam ined to determ ine if th e re was a deficiency of heterozygotes in the population, as evidence of genetic structu re (Appendix C). There is no evidence of population subdivision, or genetic structu re , a t this scale. Since we know th a t females do no t 121 tend to m igrate, the apparen t panm ixia seems to be due to the wide- ranging m ovem ents of the male segment of the population . Estimates of Na Ne, the genetic effective population size (W right 1931,1969), can be though t of as the size of a genetically ideal popu la tion tha t would undergo the same am ount of random genetic d rift as the actual popu la tion (Lande and Barrowclough 1983). h i o the r words Ne is the size of an idealised population leading to the same variance of gene frequency change per generation (Hill 1972) as the actual population . In one sense, the ideal population of size Ne has the sam e ra te of loss of heterozygosity as an actual, non-ideal, wild popu la tion of size N. The original form ula expresses Ne (here w ritten as N) as a function of the change in heterozygosity over time: Ht = Ho (W right 1969) N eutral genetic variation, o r heterozygosity, is lost over tim e in a fin ite population by random genetic d rift and by behavior th a t p roduces inbreeding. Since large mammals like grizzly bears tend to show m any characteristics th a t m ay reduce the am oun t of varia tion th a t is passed on from generation to generation, Ne m ay be substantially lower than N, the actual population size. Among these characteristics are unequal num ber of breeding males and females, 122 fluctuations in population size, non-Poisson d istribu tion in progeny num ber (unequal probability of an individual contributing to subsequen t generations), and geographic genetic s truc tu re in the population . W hen examining populations of wild animals though, it becom es difficult to estim ate param eters for these characteristics. In fact, Lewontin and K rakauer (1973) expressed the opinion th a t "m easurem ent of effective population sizes are....virtually impossible...". This was before the advent of techniques which allow a finer resolu tion in examining genetic variation, so hopefully we are getting closer to making realistic estimates. Harris and A llendorf (1989) felt th a t "for m anagem ent purposes it is p robab ly unnecessary to strive for g reat precision in Ne estim ates. Given the likely uncerta in ties in da ta necessary for its calculation by any m ethod , excessively rigid dependence on even the best of estim ations is unw arran ted". W ith these caveats in m ind then , I will a ttem p t to estim ate Ne- Male grizzlies travel over fairly large distances, sometimes hund red s of miles. One adu lt male, residen t in ou r study area for several years, was la te r shot nea r Barrow, a distance of abou t 300 kilom eters (Reynolds, pers. comm.). A nother male traveled 163 km to the Arctic Ocean coast and then re tu rned (Reynolds and Hechtel 1984). Females travel less widely, often staying w ith in estab lished hom e ranges of 50 km.2 for the ir en tire lives (Reynolds and Hechtel 1984). There was one apparen t exception to this rule; an adu lt fem ale (#1179, Figures 19 and 20) th a t was a residen t in ou r study 123 area suddenly m oved 100 km, over the Brooks Range, to a new area. She re tu rn ed in 1989 when she was observed w ith a cub. She and h e r d augh ter were bo th located in h e r new hom e range in 1993. She appears to b reed and perhaps bear young in the study area, bu t raise h e r young in the area to the south (or lead them there after weaning). Thus, as far as he r genetic contribution is concerned, she is a p a r t of the study population and h e r extensive range should have no effect on Ne. A useful m easure in conservation biology is th e ra tio of genetic effective population size to actual population size: how m any unique genotypes are effectively p resen t in a population of a know n num ber of individuals. Generalized models of large mammal populations have resu lted in an estim ate of Ne/N = about 0.25. A sim ilar approach w ith grizzly bear populations suggested an estim ate of Ne/N = 0.2 to 0.4 (Harris 1986). Harris and AUendorf used a sim ulation m odel to estim ate Ne by tracing the loss of heterozygosity th rough time, and th en com paring results with estimates p roduced by applying pub lished formulas. They found th a t the distribution of reproductive contribution among males was a critical com ponent for estim ating Ne (Harris and A llendorf 1989). I will discuss estim ates th a t focus on this com ponent (unequal breeding sex ratio) as well as variance in progeny num ber among bo th sexes, and m igration betw een subpopulations (the concept of neighborhood size). Ne is calculated from the genetic data. N is an ecological m easure and represen ts only the animals (adults, who can po ten tia lly con tribu te alleles) th a t actually inhabit the area. 124 Variance in orogeny num ber In an ideal population, the distribution of p rogeny p e r p a ren t approaches the Poisson d istribu tion when N is large. If the re is random sampling of gametes, the m ean num ber of gam etes should equal th e variance in num ber of gametes. One way of estim ating this is to look a t fam ily size per individual over the indiv idual's lifetime. The m ean, k , and variance Vk of fam ily size are re la ted to Ne as: Ne — 4 N - 2 V k + 2 or (Hartl 1988) and if the variance equals the mean, Ne = N. If fam ily size is un ifo rm variance approaches zero and Ne approaches 2N. D eterm ination of patern ity for 30 family groups gave us 12 know n fa thers (over 9 years old) for 36 offspring (see Male rep roductive success, Table 8). Mean progeny num ber was 0.97 w ith variance 3.03 (n=25) and is definitely no t Poisson. Estimates derived by including hypothetical males (Table 9) are similar. Mean progeny num ber was 1.24 w ith variance 3.25 (n=46). However, these are on ly sam ples from a 7-year w indow of data; they m ay no t reflect lifetime reproduction. The d a ta for the distribution of female progeny num ber is slightly better. If we consider all the da ta for m arked fem ales which we have observed w ith any offspring over a 16-year period , the m ean progeny num ber is 3.21 and the variance is 3.32 (n=65); this is 125 essentially a Poisson distribution. However, these d a ta are also incom plete, m any females were first cap tu red a t o lder ages and m ay have ra ised and w eaned several offspring before th e study began. O ther females were n o t observed for gaps of 7 to 10 years. Most fem ales are still living, and have no t been observed from b irth to death , so this is no t an estimate of lifetime progeny num ber. If we consider smaller subsets of the data; fo r m arked females w hich we have observed only w ith known offspring over a 16-year period , the m ean progeny num ber is 3.54 and the variance is 4.37 (n=50). For females first cap tu red a t 8 and younger w ith fairly com plete records, the m ean num ber of progeny is 3.37 and variance 3 .18 (n= 16). For females captured a t 5 and younger (earliest observed age of successful breeding) with fairly com plete records, the m ean num ber of progeny is 2.57 and variance 2.61 (n=7). Again, m ost of these females are still living, and the sam ple size is small. If we look a t female fam ily size for the 6-year w indow of d a ta fo r w hich we have genetic samples, for com parison w ith the d a ta on m ale p rogeny num ber, we find a m ean family size o f 2.03 w ith variance 0.52 (n=30). During one breeding interval, then, it appears th a t fem ale fam ily size is fairly uniform . Over en tire lifetimes, w ith stochastic dem ographic events such as death, and stochastic environm ental events such as high nu trition or fam ine, th e m ean fam ily size m ay approach a Poisson distribution. Probably all we can say is th a t variance in progeny num ber will n o t increase Ne ; we can clearly see from the d a ta th a t fam ily Isize, over the lifetimes of individuals, is no t uniform . There is some indication, however, th a t the distribution of p rogeny num ber for fem ales is essentially Poisson, and so ne ither increases o r decreases Ne. We should conclude th a t variance in fam ily size is p robably increased by the male contribution and the n e t resu lt effect will tend to reduce Ne, however, so our estim ate derived from unequal num ber of breeding males and females (see following section) should be considered a maximum value. Unequal breeding sex ratio Of the bears genotyped, 57 were offspring whose m others were known. We were able to identify fathers for 36; the rem aining 21 were sired by males th a t h ad no t been sam pled (see Male reproductive success, Tables 8 and 9). We obtained pedigree d a ta on 30 fam ily groups (each w ith a d ifferent m other) spanning a seven- y ea r period . Of these fam ily groups we were able to iden tify 12 m ales th a t sired cubs w ith 18 females. Samples w ere n o t available fo r all possible males, and because siblings can be fa thered by d ifferen t males, a t least one additional male was necessary to sire all th e offspring from these 18 females. This gives us an in itial estim ate of 13 breeding males and 18 breeding females (1:1.4) fo r this subsam ple of the population. If we deduce male genotypes to explain all offspring (from all 30 females) whose fathers were no t in our sample, we find th a t a m inim um of 7 additional males are necessary to prov ide all the 1 2 6 127 alleles found in offspring tha t d id no t come from the m o ther (see Paternity; Hypothetical Males). Thus we can estim ate a larger sam ple of 19 breeding males w ith 30 breeding fem ales (1:1.58). If we use these values in W right's original equation fo r Ne (W right 1931): Ne = f —L _ +_L _ ) X 4(Nm) 4(Nf) ) XT 4(NmNf) Ne = TtoTW we ob ta in m inim um values of Ne = 30.19 for the sm aller sam ple (18 females), and Ne = 46.5 for the larger sample (30 females). We can deduce as a rule-of-thum b from these ratios th a t Ne is rough ly equal to 1.6 tim es the num ber of breeding-age females in th e population. Our genetic sample contains a to tal of 45 breeding-age females, b u t we only h ad samples of offspring from 30 of them . Annual aerial counts ind icate an average of 41 breeding-age females in the s tudy a rea (Reynolds, 1992). If we assume th a t all breeding-age females do breed , we can estim ate (using a 1:1.58 ratio) th a t the re are a to ta l of abou t 26 males in the study area th a t successfully b reed w ith 41 females. Ne = 63.64 fo r the en tire s tudy area. The overall population which this represen ts should be viewed from an ecological point-of-view; the actual animals living in the area. However, the to tal num ber of genetic samples from which the genetic d a ta were derived are surprisingly rep resen ta tive of the dem ographic population. An aerial census in 1992 estim ated 153 128 bears in the study area, and ratios of adu lt males and adu lt females were sim ilar to those in the genetic sample (Reynolds, pers. comm.). The genetic sample com prised 152 individuals. Three of these (#1232, #1431 and #1433) were from outside the study a rea (200 km south), and two were unknown samples of inexact location (UNKl and UNK2). These 5 samples were included in the overall database since th e ir effect on allele frequencies was negligible fo r in trapopu la tion comparisons, and the ir proxim ity to the study area w arran ted inclusion for in terpopulation comparisons. These 5 sam ples were excluded in com paring allele frequencies betw een generations. Of th e rem aining 147 genotyped animals, 87 were adu lts over 5 years of age, and 69 were adults over 9 years of age. The num ber of animals estim ated for the study area from aerial census in 1992 was 153. We know the age of m arked adu lts b u t can x only guess how m any unm arked adults are over 5 , o r 9, years of age. Only adults th a t are possible contributors of gam etes are considered in estimating N in the ratio of Ne/N. This ratio m ay therefore be in the range of 0.731 - 0.884 (63.6/87 to 63.6/69). However, we also estim ated an additional 15-28 breeding age m ales (9 years o r older) th a t were no t included in th e genetic sam ple (see Male Reproductive Success). If we include a m ean estim ate of these anim als in bu r estimate of N, our best estim ate o f Ne/N th en is th en 63.6/(69+22) or 0.699. This is quite a b it h igher th an previously estim ated by sim ulation modeling. 129 Neighborhood size The effective size of the population can also be considered in term s o f th e num ber of breeding individuals p e r u n it a rea and the am oun t of d ispersion between an individual's b irthp lace and th a t of its offspring (Wright 1969): Ne = Jt 5 w here Jt =3.1416, 5 is the num ber of breeding individuals p e r un it area, and o2 is the one-way variance of d istance betw een b ir th and breed ing sites. The density of successful breeding adults in ou r s tudy a rea is abou t 81 /5200 km 2 o r 15 .6 /1000 km 2- If we use ou r estim ate of Ne ob tained above, we get a value of: O2 = 63.6/(3 .1416)(15 .6) or 1,298 km 2 and o = 3 6 km. Accordingly, 99% of individuals have the ir offspring w ithin a circle of rad ius 3o o r 108 km around the ir birthplace. The m ean hom e range size is J t r2 = 3.1416(108)2 = 36,644 km 2. This estim ate does no t come close to the actual hom e range sizes observed from m arked animals (see Demography). However, genetic evidence and incidental observations of w ide-ranging m ovem ents by adu lt males indicate th a t some males m ay have effective hom e ranges of very large size which could resu lt in a m ean rad ius of 108 km overall. 130 Formula variations In com paring Ne estimates derived from published form ulas w ith th e ir results from a sim ulated population, Harris and A llendorf (1989) found th a t m inor population fluctuations h ad little effect on Ne, b u t th a t variation in lifetime reproductive success am ong males (Vkm) greatly reduced Ne from its expectation under random m ating success. We have obtained estimates of reproductive success o f m ales over a 7-year period, equivalent to one breeding interval, b u t fa r sho rt of lifetime values. The d istribution of m ale progeny num ber during this window of time does no t appear to be Poisson, b u t dem ographic o r environm ental factors over time m ay tend to random ize this distribution. Harris and Allendorf also found tha t W right's original equation fo r unequal sex ratio ,which we used, "painted an overly optim istic picture", and th a t formulas by Hill (1972), Ryman et al. (1981), and Reed et al. (1986) gave m ore accurate estimates. These form ulas how ever requ ire estim ates of various param eters w hich include variance in lifetime num ber of progeny (Hill), heritab ility of fertility and variance of individual lifetime production of offspring who them selves survive to reproductive age (Ryman et al.), and p robab ility th a t a new born of each age survives and reproduces (Reed e t al.). I app lied m y best estimates of variance in progeny num ber to Hill's form ula (Hill 1972): 131 1/Ne=l/16M L[2 + O2Irun + 2(M/F)cov(mm,mf) + (MZF)2O2Inf] + 1/16FL[2 + o2ff + 2(F/M)cov(fm,ff) + (F/M)2o2fml w here M,F = the num ber of males (females) breeding yearly. L = average age of all parents. o2 = variance of num ber of progeny, cov = covariance in progeny. Specifically, if we use m inim um estimates of M = 3.8 and F = 6 (based on yearly female reproduction and the ratio of males to females), com bined w ith L= 10 (as a first approxim ation), and using variance and covariance estimates based on the da ta (Appendix D, Table 43): o2mm = 1.581 O2Hif = 0.686 o2fm = 0.804 o2ff =0.751 cov(mm,mf) = 0.370 cov(fm,ff) = -0.289, we get a value of Ne = 90. If we assum e th a t 10 females b reed yearly on average, F = 10, and M = 6.3, we get Ne = 149. Assuming an older average breeding age resu lts in h igher values of Ne . Assuming a Poisson d istribu tion in lifetim e progeny num bers changes the results significantly. The form ula reduces to IZNe= ( 1Z4ML) + (1Z4NL) from Hill (1972). Ifw e use a value of N = 91 as discussed earlier, in this form ula; instead of Ne = 90 we get Ne = 145. Instead of Ne = 149 we get Ne = 236. 132 To arrive a t results using Hill's form ula th a t are sim ilar to the resu lts using the basic form ula discussed above un d e r Unequal b reed ing sex ratio , we need to assume tha t the average age of all p a ren ts (L) is 8, the num ber of males breeding yearly (M) is 3.6, and the num ber of females breeding yearly (F) is 6. U nder these assum ptions we get a value of Ne = 71 using the variance found in the data, o r Ne = 116 assuming Poisson distribution of progeny num ber. However, the average age of males was 15 (from Table 8), and the average age of the 30 genotyped females was 11 w hen they b red , so th is is n o t a realistic assum ption. If we look a t th e da ta from ju s t one year (1989), the average breeding age of th e 15 fem ales th a t gave b ir th in th a t year was 11.7, and the 8 males averaged 14.6 years old. Using a m ean age of 13 fo r breeding adults gives us Ne = 116 if M = 3.6 and F = 6. If Hill's form ula is accurate in th is case, th en the basic form ula does no t overestim ate Ne bu t actually underestim ates it. Effective num ber of neu tra l alleles The effective num ber of alleles, ne, is defined as th e reciprocal of th e sum of the squares of allelic frequencies (Crow and Kimura 1970): ne = 1 /2 pj2 w here p i is the frequency of the ith allele. The actual num ber of alleles is the reciprocal of the m ean allelic frequency. Effective and actual alleles are p resen ted in Table 10: 133 Table 10. Effective and actual alleles a t each locus. N um ber o f alleles Locus 2 pi2 effective actual expected homozygosity (He) (observed) A alleles .268 3.73 8 B alleles .236 4.24 9 C alleles .261 3.83 7 D alleles .155 6.45 10 L alleles .337 2.97 6 M alleles .304 3.29 . 7 P alleles .207 4.83 8 X alleles .260 3.85 6 m ean .254 4.15 . 7.63 Total 33.19 61 Kimura and Crow (1964) have calculated average p roportion of hom ozygosity and effective num ber of alleles for random ly m ating popula tions w ith a range of values fo r Ne and m uta tion ra te as in Table 11. 134 Table 11. Combinations of homozygosity (upper figure) and ne (lower figure) w ith Ne and m uta tion rate . (After Kimura and Crow 1964). Mutation Rate (|li) Effective Population Size, Ne 10 64*(71) IO2 fOrH IO4 IO5 IO6 10-2 .71 .2 .024 .0025 1.4 5.0 41 401 (7.1 x 10-3) I .25 I 4.15 10-3 .96 .71 .2 .024 .0025 1.04 1.4 5.0 41 401 IQ"4 .966 .96 .71 .2 .024 .0025 1.004 • 1.04 1.4 5.0 41 401 10-5 .996 .966 .96 .71 .2 .024 1.0004 1.004 1.04 1.4 5.0 41 IO"6 .996 .966 .96 .71 .2 1.0004 1.004 1.04 1.4 5.0 io-7 .996 .966 .96 .71 1.0004 . 1.004 1.04 1.4 *Ne estim ated from the data.Q parentheses enclose estim ates in te rpo la ted from the table using our values of homozygosity and He. Estimates from the data (Table 10) when located in Table 11 above (outlined in box) indicate th a t the m utation ra te fo r ou r m icrosatellite alleles m ay be as h igh as 7.1 x 10"3. Comparisons between generations In a large population with random mating, selectively neu tra l alleles should m aintain the same frequency from generation to generation. Thus the m ean of gene frequency should rem ain the same, b u t th e variance should increase w ith time (Crow and Kimura 1970). The frequency of individual alleles will change between 135 generations because of the finite size of the population (Lewontin and K rakauer 1973). To examine this process I d iv ided the database in to two generations; those less than 5 years old in 1992 (except for one 5-year old female who had cubs) and those older. I calculated allele frequencies fo r b o th subsets (generations) and these are p resen ted in Appendix B (Table 42). A com parison o f m ean allele frequencies (Table 12) shows th a t the m ean allele frequencies are no t significantly d ifferent from one generation to th e next. There is an appa ren t increase in the variance (except a t locus P) as we would expect. 136 Table 12. Mean allele frequencies between generations. Locus 1st generation 2nd generation A: m ean 0.125 0.125 SE 0.133 0.142 B: m ean 0.111 0.111 SE 0.108 0.128 C: m ean 0.143 0.143 SE 0.124 0.138 D: m ean 0.100 0.100 SE 0.076 0.079 I: m ean 0.167 0.167 SE 0.156 0.175 M: m ean 0.143 0.143 SE 0.151 0.156 P: m ean 0.125 0.125 SE 0.116 0.094 X: m ean 0.167 0.167 SE 0.125 0.129 137 InteiDQDulation comparisons Small samples of 3 d isparate populations were genotyped in o rd e r to m ake prelim inary com parisons between them . These d a ta sets com prised 15 individuals from the Arctic National Wildlife Refuge (ANWR. Appendix A; Table 19), 17 individuals from the Alaska Range (AKR. Appendix A; Table 20) and 16 individuals from M ontana (NCDE. Appendix A; Table 21). Two of the NCDE bears were from the Cabinet-Yaak ecosystem (GE 2 and GE 11) and one was from the g reater Yellowstone ecosystem (GE 18). The rem aining 13 were from the Bob M arshall/G lacier Park area. All 16 were grouped as represen tative of the N orthern Continental Divide Ecosystem (NCDE) for interpopulation comparison purposes. These samples were selected in a basically random fashion; th ey are n o t entirely unrela ted . In the ANWR sample, #1515 and #1260 are m other/offspring and #1517, #1518, and #1515, #1520, are pairs of siblings. In the AKR sample #1385, #1386, and #1389, #1390, are pairs of siblings. In the NCDE sample #GB 19 and #GB 20, #GB 21, are m o ther/2 offspring. The num ber of animals in each sam ple th a t are known to be re la ted is sim ilar in each sam ple and should allow for comparisons to be m ade among them . No increase in hom ozygosity is apparen t among rela ted animals. The Cabinet-Yaak animals and the G reater Yellowstone animals have homozygosities w ithin the range of variation found in the sample. The 13 o ther NCDE samples have a m ean of 2.5 homozygous loci (range 1 - 6 ) . The Cabinet-Yaak samples have I and 4 138 homozygous loci, and the Yellowstone sample has 4 (m ean of 3 for these extra samples). Their inclusion in the NCDE sam ple does no t significantly affect the overall homozygosity of the group. A dditional large samples are available from th e ANWR and AKR populations th a t should be exam ined in the fu tu re . David Paetkau a t th e University of A lberta is cu rren tly genotyping a large sample of grizzly bears from a population in the Richardson M ountains in Canada's Yukon, ju st across the bo rder from ANWR. He is also acquiring samples from grizzly populations across Arctic Canada (Paetkau, pers. comm.). W hen these da ta are eventually available fo r analysis, estim ates of gene flow (Slatkin 1981), geographic structure, and inbreeding coefficients should improve. The N orthern Continental Divide Ecosystem (NCDE) can also be m ore accurately com pared w hen a larger sam ple size is available. Interesting com parisons can be m ade once the G reater Yellowstone Ecosystem (GYE) is analyzed using these techniques. The size of ou r WBR study popula tion (41 breeding-age females) is very close in size to estimates of d ie GYE (Dennis et. al. 1991, Foley 1994). The genetic effects of the recent isolation of the GYE population due to h ab ita t fragm entation, and reductions in effective num bers of males due to th a t isolation should have resulted in reduced levels of heterozygosity. The Alaskan samples can be considered subpopulations in the sense th a t there is contiguous, occupied, grizzly h ab ita t between them . The WBR subpopulation is located between th e foothills of the Brooks Range and the Arctic coastal plane. Similar hab itat, w ith 139 increasingly g reater relief in the m ountains to the south, extends for abou t 750 km to the ANWR population on the no rtheaste rn bo rder of Alaska. D irect m igration of individuals between these populations is unlikely, b u t possible. Gene flow over several generations th rough in te rm ed ia te subpopulations is very likely. The WBR and ANWR samples should be m ore sim ilar to each o ther than to any of the o th e r samples. The AKR subpopulation is about 1000 km to the southeast of th e WBR and 750 km southwest of ANWR and is separa ted from them by the Brooks Range, the Yukon River, extensive taiga forest, and strips of highway and hum an habitation. Direct m igration of indiv iduals is extrem ely unlikely, and gene flow over tim e should be m uch less th an th a t between the WBR and ANWR subpopulations. The NCDE sample, from the Rocky m ountains in and around Glacier N ational Park in Montana, was contiguous w ith the Alaska populations, although very d istan t (4000 km), un til th e 20 th century. Today it is a semi-isolated population, or a t m ost a pen insu lar subpopulation, near the southern end of occupied grizzly hab ita t along the Rocky Mountain chain. Allele frequencies Allele frequencies for all 4 da ta sets are p resen ted in Appendix B. Several alleles were found in the o ther populations th a t were no t p re sen t in the WBR sample; a to tal of 71 discrete alleles were found 140 in all samples as opposed to 61 in the WBR samples alone. Because the sam ples are small, these da ta should be considered prelim inary. The WBR sample contained 5 alleles no t found in o ther subpopulations: A188, L171, M222, P139, and X129. The ANWR sam ple contained 2 un ique alleles; A196 and X139. The AKR sample contained I unique allele; P141. The NCDE sample contained 5 un ique alleles; B162, B166, D183, D179, and D175. The samples from the Cabinet-Yaak and Yellowstone h ad no un ique differences from the o th e r M ontana samples. If there were one m igran t indiv idual betw een these subpopulations p e r generation, the same alleles should be p resen t in all. The WBR sample is the only sam ple large enough to be considered complete, b u t it lacks 4 alleles found in the o th e r 2 Alaskan samples, and lacks 2 alleles found in th e closest, ANWR, subpopulation. This should indicate th a t m igration between these areas is less th an one individual p e r generation, which is w hat we w ould expect. M easures of genetic d ifferentiation If we consider these samples to be rep resen tative of d ifferen t subpopulations o r demes of a single, panmictic grizzly population, we can estim ate the am ount of allele frequency divergence among them using W right's Fst statistic: , . j . ^St q( l -q) (W right 1943, A llendorf 1983) 141 w here q and o are the m ean and variance of allele frequencies am ong demes. Fst , o r the fixation index, is a m easure of the reduction in heterozygosity of a population due to genetic drift. I have com pared the 3 Alaskan subpopulations w ith each o ther, and then com pared all 4 subpopulations w ith each o ther in Tables 13 and 14 below. Table 13. Allele frequency divergence, Fst, among d ispara te grizzly subpopulations. Values in columns under each allele are frequencies of the m ost common allele in the WBR sample. Allele: Al 94 B160 C105 D172 LI 55 . M208 Pl 5 3 X137 Allele frequencies Subpopulation WBR 0.394 0.413 0.361 0.250 0.487 0.450 0.337 0.400 ANWR 0.233 0.433 0.677 0.167 0.533 0.233 0.167 0.433 AKR 0.353 0.235 0.563 0.088 0.382 0.344 0.177 0.281 mean q 0.327 0.361 0.530 0.168 0.468 0.342 0.227 0.372 variance 0.007 0.012 0.024 0.007 0.006 0.012 0.009 0.006 Fst 0.032 0.052 0.097 0.047 0.024 0.052 0.052 0.027 mean Fst 0.048 WBR 0.394 0.413 0.361 0.250 0.487 0.450 0.337 0.400 ANWR 0.233 0.433 0.677 0.167 0.533 0.233 0.167 0.433 AKR 0.353 0.235 0.563 0.088 0.382 0.344 0.177 0.281 NCDE 0.531 0.125 0.594 0.000 0.469 0.188 0.250 0.000 mean q . 0.378 0.302 0.546 0.126 0.468 0.304 0.233 0.279 variance 0.015 0.022 0.017 0.011 0.004 0.014 0.006 0.039 Fst 0.064 0.104 0.069 0.104 0.016 0.065 0.024 0.193 mean Fst_________0.081 I have com pared allele frequencies in two fashions: e ith e r by looking a t th e frequency of a single allele (the m ost com m on WBR allele, Table 13) th roughout all subpopulations, o r by considering the 142 m ost com m on allele a t each locus and grouping all o thers as an alternative allele (comparable to AUendorfs approach, Table 14). Table 14. Allele frequency divergence, Fst , among d ispara te grizzly subpopulations. Values in columns under each allele are frequencies of the m ost common allele a t th a t locus in each subpopulation. Locus: A B C D L M P X Allele frequencies Subpopulation WBR 0.394 0.413 0.361 0.250 0.487 0.450 0.337 0.400 ANWR 0.300 0.433 0.667 0.233 0.533 0.400 0.200 0.433 AKR 0.353 • 0.235 0.563 0.294 0.382 0.344 0.265 0.313 mean q 0.349 0.361 0.530 0.259 0.468 0.398 0.267 0.382 variance 0.002 0.012 0.024 0.001 0.006 0.003 0.005 0.004 Fst 0.010 0.052 0.097 0.005 0.024 0.012 0.024 0.017 mean Fst 0.030 WBR 0.394 0.413 0.361 0.250 0.487 0.450 0.337 0.400 ANWR 0.300 0.433 0.667 0.233 0.533 0.400 0.200 0.433 AKR ' . 0.353 0.235 0.563 0.294 0.382 0.344 0.265 0.313 NCDE 0.531 0.250 0.594 0.267 0.469 0.344 0.313 0.8,13 mean q 0.395 0.333 . 0.546 0.261 0.468 0.384 0.278 0.489 variance 0.010 0.011 0.017 0.001 0.004 0.003 0.004 0.049 Fst mean Fst 0.041 0.051 0.049 0.069 0.003 0.016 0.011 0.018 0.196 If these results are com pared w ith AUendorfs sim ulations for selectively n eu tra l alleles (A llendorf 1983, p 54) the m ean divergence among the 3 Alaskan populations is sim ilar to th a t expected w ith 5-10 m igrants pe r generation. This is clearly no t literally possible, especially between Alaska and M ontana 143 populations, b u t does underscore the fact th a t there is considerable gene flow between grizzly populations. If we com bine all d a ta on allele frequencies in to one large panm ictic popu la tion with 7 1 alleles We can calculate expected heterozygosity over all subpopulations. This approach is p resen ted in Appendix C. We can calculate values of Fst (or Ggr, the coefficient of gene differentiation, Nei 1973, Nei and Chakravarti 1977) from the equation: Fst = (Ht - Hs)/Hr Fst over all populations was calculated below in Table 15. The resu lts opposite Exp H (expected H) were derived using Ht as the expected m ean heterozygosity (over 8 loci) given panm ixia, and Hs as th e m ean of the expected heterozygosities (over 8 loci) in each subpopulation. Mean observed heterozygosity is also included for com parison. The results opposite Obs H (observed H) were d e riv ed . using Ht as the expected m ean heterozygosity (over 8 loci) given panm ixia, and Hs as the m ean of the observed heterozygosities (over 8 loci) in each subpopulation. 144 Table 15. Mean heterozygosity and Fst over 8 loci over all populations. ALL WBR ANWR AKR NCDE Exp H 0.763 0.747 0.713 0.765 0.675 Fst 0.049 0.021 0.066 -0.003 0.115 ObsH 0.773 0.750 0.741 0.657 Fst 0.043 -0.014 0.017 0.029 0.139 These results are sim ilar to those of the previous approach bu t allow each subpopulation to be com pared with the whole. They ind icate th a t although there is considerable gene flow among the various Alaskan subpopulations, the M ontana (NCDE) population is m ore isolated and allele frequencies have diverged from those found in Alaska. At least one locus, X, had very high homozygosity. In Appendix C, Fst is also calculated for each locus. D ifferent loci give varying estim ates of FSt probably due to sampling effects. The m ean values p resen ted above should be represen tative of each subpopulation. In a study of populations of Rhesus macaques. Gill et. al. (1992) found Fst values of 0.036, 0.007, 0.003, 0.044, 0.042, and 0.066 a t 6 polym orphic isozyme loci. Fst over all loci was 0.019. Using sim ilar techniques, 3 races of m an were found to have a m ean Fst value of 0.07, while the Ord kangaroo ra t h ad a m ean Fsr value of 0.70 (Cham bers and Bayless 1983). If Fst values using these d ifferen t techniques are com parable to our results, grizzly populations w ith a 145 m ean Fst value of 0.049 appear to be m ore isolated th an these populations of macaques, slightly less isolated than these ethnically d istinct tribes of man, and m uch less isolated than the Ord kangaroo rats. Fis, o r the inbreeding coefficient, is a m easure of th e reduction in heterozygosity in an individual due to nonrandom m ating w ithin its subpopulation . Fis = (Hs - Hi)/Hs where Hs is expected m ean heterozygosity w ithin a subpopulation and H% is th e heterozygosity of an individual. Calculations from the various subpopulations ind icate th a t the average individual in the WBR popu la tion is homozygous a t 2 loci while the average individual in the NCDE population is homozygous at 3 loci. O ther m easures of allele frequency divergence include p robab ility of identity , I (Nei 1972, Paetkau and Strobeck 1994a), and genetic d istance (Nei 1972). Probability of iden tity can be calculated using a Pascal program developed by Strobeck, and genetic d istance is -Ioge(I) o r -In(I). Genotype frequencies and heterozygosity A lthough there is no evidence of population subdivision a t th e -Ar scale of th e WBR study population, I exam ined the expected frequency versus the observed frequency of heterozygosity w ithin each of th e o the r populations to see if there was a deficiency of heterozygotes, as expected if they were subdivided. These results a re p resen ted in Appendix C, and chi-square tests indicate no 146 difference between observed and expected values. The sam ple sizes (120 loci m inimum) should give a reasonable estim ate, thus there is no evidence of population subdivision within any of the o ther sam pled subpopulations. If we consider all samples as p a rt of a single, large, panm ictic population, there may be genetic subdivision on the scale of each subpopulation (WBR, ANWR, AKR, NCDE) with respect to the whole. Average heterozygosity of 8 loci is sum m arized below: Table 16. Mean heterozygosity among disparate populations. Population Expected Observed m ean std m ean std WBR 0.747 0.052 0.773 0.059 ANWR 0.765 0.063 0.741 0.105 AKR 0.713 0.118 0.750 0.128 NCDE 0.675 0.161 0.657 0.162 Figure 39. Mean heterozygosity among disparate populations. S tandard error of NCDE sample to right of graph. Mean H £ I 5c c LUClUZ 147 So there appears to be no significant difference in m ean heterozygosity between any of these populations w ith these sam ple sizes. However, sample sizes from the ANWR (240 alleles), AKR (266 alleles), and NCDE (254 alleles) are relatively small com pared w ith the WBR (2432 alleles). There is an indication th a t th e M ontana sam ple (NCDE) has somewhat lower levels of heterozygosity, particu larly as it was collected over a m uch larger area. The X locus especially h ad very low levels of heterozygosity (Appendix C, Table 38) and appears to be drifting toward fixation of the X141 allele. Increased sample sizes should clarify this question in th e future. Selective neu tra lity of alleles If alleles are selectively neu tra l, variations in allele frequency betw een populations or between generations should be determ ined only by th e breeding structu re of the species (m igration rates and local popu la tion sizes, Lewontin and Krakauer 1973, Slatkin 1982). Breeding s tructu re should affect all alleles and loci in th e same way. Evidence of Hardy-W einberg equilibrium , allele frequency divergence among populations (Fst), and allele frequency divergence betw een generations indicates th a t the alleles s tud ied are selectively neu tra l. 148 LEVELS OF INQUIRY This pro ject spanned five years of field collection of genetic sam ples and th ree years of laboratory analysis. To the best of m y knowledge, the labora to ry techniques th a t I used were th e cu rren t sta te-o f-the-art a t the tim e I began using them . A dvancem ents in m olecular techniques occur so rapidly, though, th a t new er and be tte r m ethods of genetic analysis often are developed by labs and indiv iduals working prim arily w ith techniques before the o lder techniques have been widely applied. The process of learning a new techn ique and obtaining consistent results in a new lab can be tim e consuming. The m ulti-locus probe techniques th a t I learned while visiting th e National Fish and Wildlife Forensic Laboratory in Ashland, Oregon h ad been developed by Dr. Jerry Ruth and Dr. Steve Fain and w ere being perfected as I began using them . I encoun tered several problem s in getting the system running in our lab a t M ontana State U niversity and by the tim e tha t I h ad obtained consisten t results w ith m ost of m y samples, Drs. Fain and Ruth h ad im proved their techn ique by running a standard of known bands along w ith each sam ple in th e same lane. At this po in t in m y work it was becoming clear th a t the resolution of bands was no t quite precise enough to allow the k ind of analysis I h ad hoped. To re-run each of m y 149 sam ples w ith a s tandard would have taken ano ther year, w ith an inde term ina te increase in accuracy of results. It was extrem ely fo rtuna te for m y work th a t David Paetkau h ad developed his m icrosatellite p rim er sets by th a t tim e, and th a t I was able to work w ith h im and Dr. Curt Strobeck in th e ir lab a t the University of A lberta. My experience in this regard underscores the im portance of establishing contact w ith colleagues working in the same a rea and exchanging inform ation. Throughout m y career I have n o t encountered fellow researchers who were as open and help fu l and willing to share knowledge as I have found a t the two labs I visited. I consider myself extrem ely fo rtuna te to have w orked w ith these people and to have served an appren ticesh ip as these techniques were developed. Developing molecular techniques is b eyond m y skill, b u t I feel th a t I have learned to use these new tools in a craftsm anlike fashion. My d issertation research then has spanned an exciting window of developm ent in a rap id ly growing field of inquiry . My in te rest first began as p ro te in electrophoresis and then m itochondrial DNA techniques began providing answers to some population- and species-level questions. Since then advancem ents have m ade even the determ ination of nucleotide sequences a common (even au tom ated) p rocedure. My own lab work has given m e glimpses of how genes w ork in grizzly bears (as in all organisms) from the level of nucleotides to species to the evolution of species over time. A dm ittedly, these are ju st glimpses, b u t I will a ttem p t to discuss 150 som e of w hat I have learned a t each level. The m ost im po rtan t thing th a t I learned from this process was patience. Molecular level An ideal m icrosatellite p rim er is a unique, conserved sequence ad jacen t to a m icrosatellite repea t unit. If i t is a h ighly conserved reg ion it will be found in a range of rela ted taxa (species o r subspecies). If it is un ique it will am plify only a single locus. In these respects, the prim ers developed by David Paetkau are very good. They amplify single loci (Paetkau and Strobeck 1994a, 1994b) and all p rim er sets work w ith all ex tan t taxa of bears, except fo r one locus th a t does no t amplify in Pandas (Paetkau, pers. comm.). The rep ea t units, which provide the variability fo r analysis, are noncoding and therefore selectively neu tra l in themselves, b u t the ad jacen t coding regions, perhaps including the p rim er site, m ust be evolu tionaiily im portant, and thus selected for, since they are so h igh ly conserved. In m icrosatellite analysis, then, we are looking a t im po rtan t genes found in all individuals over a wide range of taxa, w hich are highly polymorphic in neutral, noncoding regions: the m icrosatellite repeats. We assume the coding po rtion of the gene is functionally equivalent (if no t identical) in each allele, b u t the re are m any alleles of d ifferent overall lengths. There is no selection for allele length as far as we know; how ever th e finite num ber of alleles, w ithin a lim ited range of sizes, indicates th a t some factor, perhaps the physics and geom etry of 151 nucleotides, tends to lim it the num ber of repea t units found in a given site. Ten alleles were the m ost found a t any locus in this study, b u t a m icrosatellite locus in p ilo t whales was found to have 54 un ique alleles (Tautz 1989, Schlotterer et. al. 1991). W hen homologous chromosomes pair up during meiosis, and recom bination occurs, m icrosatellite and m inisatellite regions will o ften be of d ifferen t lengths. This uneven pairing m ay stress some bonds in those regions (Gilpin, pers. comm.) and m ay be a factor in the h igh ra te of recom bination found in these tandem repea t segments. Organism level On the level of the individual organism , a grizzly bear, we have learned several im portan t things th a t have bearing (no p un in tended) on behavior. SibUngs m ay no t have full sibling kinship (0.5) b u t m ay be re la ted as half-sibs (kinship 0.25). Only half (or less) of all males are successful. Male rep roductive strategy Our study population was d istribu ted fairly uniform ly over 5200 square kilom eters a t a density of 29.5 bears p e r 1000 square kilom eters (Reynolds 1992). Like m ost Arctic and in te rio r grizzly populations, these bears utilize dispersed and patchy food sources du ring the breeding season. As a result, breeding adults are w idely 152 dispersed; m ale bears often need to travel considerable distances to locate females in estrous; and the optim al male reproductive strategy is p robab ly to try to monopolize a receptive female once one is located ra th e r th an to travel widely in o rder to m ate w ith m ore th an one fem ale (Herrero and Hamer 1977, Parker 1984). Long pairings such as this have also been reported in Denali National Park by Dean (1976) and Murie (1981). Five known breeding males sired offspring w ith m ore th an one fem ale in ou r sample. One of these (#1124) sired 2 offspring in each of th ree consecutive years w ith a d ifferent female each year. Two males (#1098 and #1081) each fa thered I offspring w ith one female in each o f two d ifferent years. These da ta suggest th a t these males concen tra ted the ir reproductive effort on a single fem ale each breed ing season, o r traveled so widely th a t o ther offsprin g were no t sampled. A fou rth male (#1712) p roduced 2 offspring w ith one female in one year, and 4 offspring w ith 2 females in ano ther year (1989). In 1989 he was the only sire of one litter, bu t shared paren tage of the o th e r litter. He m ay have been displaced from th a t m ating oppo rtun ity by a m ore dom inant male, o r m ay have located th a t fem ale a fte r b reeding w ith the first one. The fifth male (#1411) em ployed an a lternate strategy: he sired 3 cubs w ith one female, I cub w ith a d ifferent female, and I cub w ith ano ther female, all during the same year (1989). No cubs w ere sired by h im during any of the o ther years fo r w hich we have data. 153 There is obvious flexibility in reproductive strategy; bu t investing effort in a single female appears to be favored over searching fo r additional breeding opportunities when they are scarce. This strategy m ay change when opportunities are m ore frequen t. Our da ta com prise 20 litters w ith known fathers. Ten of those were b o m in the same year; 1989. This was the only year fo r w hich we have da ta in which any males successfully sired cubs w ith m ore th an one female. This is also the year in which female p roductiv ity was observed to be highest recorded during the en tire s tudy period: 17 females observed w ith 38 cubs (Reynolds, 1992). Environm ental variables in the Arctic, including the availability of m obile p rey such as caribou, are subject to a h igh degree of stochastic variance from year to year. Good years, such as 1988 which p roduced the 1989 cohort, are rare. Of the 14 cohorts m onitored for a t least 2 years, the 1989 cohort accounted for 25% of the to ta l num ber of cubs produced, and 34% of the surviving 2-year olds (Reynolds, 1992). It seems likely th a t when th e re is a good b reed ing yea r such as this, males m ay spend less tim e w ith ind iv idual females in estrous in o rder to b reed w ith o th e r females. A lternatively, females m ay re-en ter estrous in g rea te r num bers during good years and the spread of dates when females are in estrous m ay also be larger. It is unlikely th a t any single male is responsible fo r m ore th an abou t 11% of total reproduction during his lifetime (Table 5). Thus, th e re m ay be validity to the contention th a t killing of cubs can confer reproductive advantage: if a m ale encounters a cub the re is 154 an 89% chance th a t it was fathered by ano ther male. However, in any given year the re m ay be a slightly h igher p robab ility of encountering his own offspring. O ther grizzly populations, particularly coastal ones (Egbert and Stokes 1976), congregate during the breeding season a t concentrated food sources. At these feeding aggregations a m ore structu red dom inance h ierarchy is established particularly among males. D om inant m ales appear to spend less time with indiv idual females in estrous, often encountering additional receptive females and subsequently breeding with them (Homocker 1962, Egbert 1978, Craighead 1979, Craighead and Mitchell 1982, Modafferi 1984). It is likely th a t th is increased density of bears w ith increased breeding opportun ities m ay resu lt in somewhat d ifferent values for reproductive success. Dominant males have access to m ore females du ring the lim ited tim e th a t they are receptive. A lternatively, though, as dom inan t males change partners, m ore subord inate males have an opportun ity to b reed with the females th a t have been abandoned. Population level We believe these data, covering 7 years of rep roduction and sam pled over m ore than one complete breeding in terval, are rep resen tative of the grizzly bears in our study area. Our sample, however, consists of a group of bears th a t are contiguous w ith o ther grizzly bears from the Bering Strait, across Arctic Canada, and south 155 to M ontana and Wyoming. This is no t an isolated subpopulation. Behavioral observations indicate th a t female bears are strongly philopatric, and th a t their female offspring tend to establish hom e ranges ad jacen t to, o r overlapping, the ir m other's (Reynolds 1992, Reynolds and Hechtel 1984). Therefore, I choose to define the subpopulation, for purposes of discussion, as a discrete group of 41 breeding-age females, the ir offspring, and the breeding-age males associated w ith them . The genetic evidence dem onstrates th a t a lthough females m ay cluster in groups of relatively stationary, rela ted individuals, there are still h igh levels of gene flow throughout contiguous grizzly hab ita t. All of n o rth e rn Alaska, and perhaps Canada, is effectively one large panm icdc population. The probable m echanism for m aintain ing this genetic diversity, in the absence o f overdom inant selection, is m igration by the male segment of the population. Most m ales observed have discrete hom e ranges, bu t some travel widely. Synchrony of recru itm ent Synchrony of recruitm ent, or pulses of reproduction and cohort survival, are docum ented in bear populations (Jonkel and Cowan 1971, Kemp 1976, Young and Ruff 1982), and have usually been a ttrib u ted to regulation by intrinsic social factors such as aggression by adu lt males. More generally, nu tritional conditions regulate the reproductive rate , social interactions regulate access to sources of nu trition (Bunnell and Tait 1981), and male aggression regulates 156 cohort survival to some degree which is p robably also affected by nu tritional state. In ou r study area the num ber and density of adu lt males observed has rem ained fairly stable from year to year, so peaks of rep roduc tion are prim arily set by extrinsic, density -independen t factors, particu larly nu tritional status, which in tu rn is determ ined largely by environm ental conditions. When there is a good year, i t is good over a large area and social interactions m ay no t lim it access to sources of nu trition . Cohort survival, however, m ay be greatly affected by social mechanisms, particu larly male aggression. If food is scarce, aggression towards bo th juveniles and adults m ay increase. Species level A lthough there are only 8 extant species of bears on the p lanet, th ey have n o t been well studied from a phylogenetic po in t of view. Relationships between the th ree Asian species in particu la r are no t clear and there is some indication th a t some of them can in te rb reed (Nowak and Paradiso 1983, Van Gelder 1977). Using available paleontological and molecular da ta there are a num ber of possible phylogenetic trees which can be draw n to describe the evolu tionary rela tionsh ips of the Ursidae. I have constructed one possible tree (Figure 34) w ith rough approxim ations of genetic distances implied, relying m ainly on data from Goldman et. al. (1987, 1989), Nash and O’Brien (1987), and Kurten and Anderson (1984). ea rly M io ce ne 157 Figure 40. An approxim ate phenetic diagram of the Ursids. (after Kurten and Anderson 1980, Nash and O'Brien 1987, and Goldman et. al. 1987). Extant taxa in boldface. ---Grizzly bear — Polar bear Cave bear ____ American Black bear Ursus pristinus Tibetan Black bear ------ Sun bear -------Sloth bear . Great Short-faced bear Lesser Short-faced bear _ Florida Spectacled bear Extinct Pandas Spectacled bear Giant Panda Procyonidae (Raccoons) s - E CNJ Csl 00 c O - I I I t I O S Q- K |- i s I CO tt|l 158 SUMMARY In summary,' th e nu ll hypotheses th a t I have a ttem p ted to te s t in th is thesis have been accepted o r rejected as follows: Hoi: Males cannot be excluded from pa te rn ity of individual offspring using fou r m inisatellite probes. Accepted. I was unable to determ ine p a te rn ity using the probes available to me. Ho2: Males cannot be excluded from pa tern ity o f individual offspring using m icrosatellite analysis. Rejected. All b u t I male sam pled w ere excluded from p a te rn ity in 34 cases of patern ity . AU b u t 2 males were excluded in 2 cases. HqS : There is no difference betw een the alleles observed in siblings and those expected to segregate from a single father. (i.e. m ultiple pa tern ity does no t occur). Rejected. Two fanuly groups exhibited m ore pa te rna l alleles a t a single locus th an could be accounted for by a single father. Ho4: There is no difference in num ber of offspring sired among males. Rejected. Wide variance was found in progeny num ber among males. Only 46% of breeding-age males sam pled successfully p roduced offspring. HqS : There is no difference in sex ratio among successfuUy breeding adults. Rejected. Sex ra tio of breeding males to breeding females was abou t 1:1.5. 159 Hq6: There is no evidence of differential o r stabilizing selection am ong m icrosatellite alleles, (i.e. alleles are selectively neu tra l). Accepted. Evidence of Hardy-W einberg equilibrium , allele frequency divergence among populations (Fst) , and allele frequency divergence between generations indicates th a t the alleles stud ied are selectively neu tra l. Hq7: There is no difference in allelic frequencies between generations. Accepted. HqS: Genotype frequencies do no t differ from those expected u n d e r Hardy-W einberg equilibrium . Accepted. One locus, L, was found to be significantly different from expected p roportions of one genotype (L159/L161). AU o ther genotypes suppo rted Hardy- W einberg equihbrium . Hq9: There is no difference between observed num ber of heterozygotes and the num ber expected from aUehc frequencies given Hardy-W einberg equilibrium , (i.e. there is no W ahlund effect due to genetic structure). Accepted a t the level of each sam ple population . There is no evidence of genetic structu re w ithin each a rea sam pled. Ho 10: There is no difference between expected num ber of heterozygotes among disparate populations, (i.e. aU populations sam pled can be considered parts of a single panm ictic population). Tentatively accepted. At the level of one large panm ictic population of N orth American grizzly bears there is some evidence th a t M ontana bears have reduced levels of heterozygosity relative to th e Alaskan 1 6 0 bears sam pled, b u t the difference is n o t significant w ith these sam ple sizes. M icrosatellite analysis is rap id ly developing as a sensitive, practical, and highly informative m ethod of examining genetic varia tion am ong individuals, w ithin populations, and between populations. Comparison of this technique with multi-locus, m inisatellite analysis on the same genetic samples indicates th a t estim ates of Ne and H based on m inisatellite data, a t least by those techniques repo rted here, m ay be somewhat h igher th an w arran ted . Ne fo r th e 33 .15 /33 .6 probe com binations was 41.60 - 114.49 (m ean 78.04) and H was 0.953-0.967 (for 30 breeding females, Table 3). The corresponding values from microsatellite analysis were Ne = 63.6 and H(exp) = 0.747. Small populations of th rea tened o r endangered species are at risk of extinction due directly to dem ographic factors and indirectly to genetic factors such as low heterozygosity (Soule 1980, Ralls and Ballou 1983, A llendorf and Leary 1986, Gilpin and Soule 1986, Shaffer 1991). The findings reported here imply th a t th e num ber of b reed ing males (and therefore the genetic effective popula tion size) m ay be som ewhat h igher in wild populations th an if p a te rn ity was exclusive to one male p e r litter, and th a t non-residen t o r partia lly residen t males m ay p lay a significant role in reproduction . They suggest th a t m anagem ent of small o r captive populations m ay be im proved by encouraging m ultiple breeding opportunities for females. 1 6 1 Grizzly bear populations exhibit h igh levels of heterozygosity and gene flow. Since females tend to be strongly philopatric, the m echanism for m aintaining alleles in populations appears to be the w ide-ranging m ovem ents of males. Because o f the rap id ra te of hum an population grow th and se ttlem en t of undeveloped areas of N orth America, w ilderness areas th a t are n o t legally p ro tec ted will soon disappear, In th e w estern U nited States and Alaska we still have the option o f designing the wild landscape th a t we wish to re ta in in to the foreseeable future. The lessons of history, and of o the r m ore densely popu la ted nations, indicate th a t once w ilderness is lost i t cannot be regained in ecological tim e. If we wish to p reserve b iodiversity we need to lim it developm ent m ore aggressively over la rger tracts o f w ilderness th an are cu rren tly pro tected , especially in the conterm inous 48 states. If we can p reserve ecosystems fo r keystone species like th e grizzly we can p reserve biodiyersity . I feel th a t this thesis indicates th a t a grizzly b ea r ecosystem encom passes a very large area, such as those still found in no rth ern Alaska. Grizzly bears probably need this m uch space to m aintain norm al levels of genetic diversity and norm al popu la tion dynamics. If we wish to preserve the social m echanism s th a t affect popu la tion \ grow th and gene flow we need to m ain ta in large areas w here m ale grizzlies can m igrate betw een fem ale-based "denies". In sinall isolated populations it m ay be possible to approxim ate male dispersal by m anagem ent m ethods. 1 6 2 LITERATURE CITED 163 LITERATURE CITED AUendorf, F. W. 1983. Isolation, gene flow, and genetic d ifferentiation among populations. In: Genetics and Conservation. C. M Schonewald-Cox et. al., Eds. (Benjamin-Cummings, California), pp. 51-65. Allendorf, F.W., F.B. Christiansen, T. Dobson, W.F. Eanes, and 0 . Frydenberg. 1979. Electrophoretic variation in large mammals. I. The po la r bear, Thalarctos m aritim us. Hereditas 91:19-22 Allendorf, F. W., and R.F.Leary. 1986. Heterozygosity and fitness in n a tu ra l populations of animals. In: Conservation Biology. M. E. Soule, Ed. (S inauer, Sunderland, MA) pp. 57-76 Amos, B., C. Schlotterer, and D. Tautz. 1993. Social structure of pilot whales revealed by analytical DNA profiling. Science. 260:670-672. Beckman, J.S., and J.L. Weber. 1992. Survey of hum an and ra t m icrosatellites. Genomics 12:627-631. Birdsall, D.A., and D. Nash. 1973. Occurrence of successful m ultiple insem ination of females in natu ra l populations of deer mice (Peromvscus m aniculatus). Evolution 27:106-110. Bunnell, F.L. and D. Tait. 1981. Population dynamics of bears - im plications for m anagem ent. In: Dynamics of Large Mammal Populations . T.D. Smith & C. Fowler, Eds. (John Wiley and Sons, New York) pp.75-98. Burke, T., and M.W. Bruford. 1987. DNA fingerprinting in birds. N ature 327:149-152 Chambers, S.M., and J.W. Bayless. 1983. Systematics, conservation, and th e m easurem ent of genetic diversity. Chapter 21, In: Genetics and Conservation. C. M Schonewald-Cox et. al., Eds. (Benjamin- Cummings, California), pp. 349-363. 164 Clark, E.E. 1966. Indian legends from the no rth ern Rockies. Univ. Okla. Press. The Civilization of the American Indian Series. Vol. 82. 350 pp. Cowan, I.M. 1972. The status and conservation of bears (Ursidae) of th e w orld - 1970. Int. Conf. Bear Res. and Manage. 3:97-109. Craighead, F.C., Jr. 1976. Grizzly bear ranges and m ovem ent as determ ined by radiotracking. Int. Conf. Bear Res. and Manage. 2:343-367. Craighead, F.C., Jr. 1979. Track of the grizzly. Sierra Club Books, San Francisco. 262 pp. Craighead, F.C., Jr., and J.J. Craighead. 1963. Radiotracking of grizzly bears in Yellowstone National Park, Wyoming, pp . 59-67 In: National Geographic Research Report 1964, P.H. Oehser, Ed. Nad. Geog. Soc. Washington D.C. Craighead, F.C., Jr., and J.J. Craighead. 1965. Tracking grizzly bears. BioScience 15(2):88-92. Craighead, F.C., Jr., and J.J. Craighead. 1972. Grizzly bear p reh ib em ation and denning activities as determ ined by radiotracking. Wildl. Monog. No. 32. 35 pp. Craighead, J.J., and J.A. Mitchell. 1982. Grizzly bear (Ursus arctos). In: Wild Mammals of North America: Biology, Management, Economics J-A. Chapm an and GA. Feldhamer, Eds. (Hopkins. Baltimore, MD), pp. 515-556. Craighead, J.J., F.C. Craighead Jr., and J. Sumner. 1976. Reproductive cycles and rates in the grizzly bear, Ursus arctos horribilis. of the Yellowstone ecosystem. Int. Conf Bear Res. & Manage. 3:337-356. Craighead, J.J., J.R. Vamey, and F.C. Craighead Jr. 1974 . A population analysis of the Yellowstone grizzly bears. Mont. Coop. Wildl. Res. Unit. Bull. No. 40. Univ. Monti, Missoula. 20 pp. 165 Cronin, M.A., S.C. Amstrup, G.W. Gamer, and E.R. Vyse. 1991a. Interspecific and intraspecific m itochondrial DNA variation in North American bears (Ursus). Canadian Journal of Zoology 69(12):2985- 2992. Cronin, M.A., D.A. Palmisciano, E.R. Vyse, D.G. Cameron. 1991b. M itochondrial DNA in Wildlife Forensic Science: Species identification of tissues. Wildl. Soc. Bull 19:94-105. Crow, J.F., and M. Kimura. 1970. An introduction to population genetics theory. H arper and Row. New York. 591 pp. Dean, F.C. 1976. Aspects of grizzly bear population ecology in M ount McKinley National Park. Int. Conf. Bear Res. and Manage. 3:111-119. Dennis, B., P.L. Mulholland, and j.M. Scott. 1991. Estimation of growth and extinction param eters for endangered species. Ecology M onographs 61:115-143. Dorozynski, A. 1994. A family tree of European bears. Science 263:175. Dutrillaux, B., and J. Couturier. 1983. The ancestral karyotype of Carnivora: com parison w ith th a t of p la ty rrh ine monkeys. Cytogenet. Cell. Genet. 35:200-208. Edman, J.C., M.E. Evans-Holm, J.E. Maiich, and J. Ruth. 1988. Ohgonucleotide probes labelled with alkaline phosphatase for chem ilum inescent detection. Nucleic Acids Research 16(2):6235. Egbert, A.L. 1978. The social behavior of brown bears a t McNeil River Alaska. Unpub. Ph.D. diss., Utah State Univ. Egbert, A.L. and A.W. Stokes. 1976. The social behav iour of brown bears on an Alaskan salmon stream . Int. Conf Bear Res. & Manage. 3 :41-76. Fain, S.R. 1991. Genetic variation in populations of black bear. Unpub. presentation . Northwest association of Forensic Scientists. May 13-17, A nchorage Alaska. 1 6 6 Fain, S.R. Personal communication. 1992. discussions during collaborative laboratory work in Ashland, Oregon. October. Faulkes, C.G., D.H. Abbott, and A.L. Mellor. 1990. Investigation of genetic d iversity in wild colonies of naked m ole-rats by DNA fingerprinting. J. Zool. London 221:87-97. Foley, P. 1994. Predicting extinction times from environm ental stochasticity and carrying capacity. Conservation Biology 8(1): 124- 137. Georges, M., A.S. Lequarre, M. Castelli, R. Hanset, and G. Vassart. 1988. DNA fingerprinting in domestic animals using four d ifferen t m inisatellite probes. Cytogenet. Cell. Genet. 47:127-131. Gibbs, H.L. et al. 1990. Realized reproductive success of polygynous red-w inged blackbirds revealed by DNA m arkers. Science 250:1394- 1397. Gilbert, D.A., N. Lehman, S.J. O'Brien, and R.K, Wayne. 1990. Genetic fingerprinting reflects population differentiation in the California Channel Island fox. Nature 344:764-767. Gill, P.S., J. Blangero, G.S. Manis, J. Scheffler, M.E. KeeUng, and W.H. Stone. 1992. Genetic structure of th ree populations of Rhesus m acaques (Macaca m ula ta): impUcations for genetic m anagem ent. A m er. J. Primatology 27:85-92. Gilpin, M.E., and M.E. Soule. 1986. Minimum viable populations: Processes of species extinction. In: Conservation Biology: the science of scarcity and diversity. M.E. Soule, Ed. pp 19-34. Goldman, D., P. Rathngiri, and S.J. O'Brien. 1987. The molecular phytogeny of the ursids as indicated by two-dimensional p ro te in electrophoresis. Proc. Natn. Acad. Sci., USA 84:3307-3311. Goldman, D., P. Rathngiri, and S.J. O'Brien. 1989. Molecular genetic d istance estimates among the Ursidae as ind icated by one- and two- dim ensional p ro tein electrophoresis. Evolution 43(2X289-295. 167 Gyllenstein, U.B. 1989. PCR and DNA sequencing. BioTechniques 7:700-708. Hallo well, Irving A. 1926. Bear ceremonialism in the n o rth e rn hem isphere. American Anthropologist 28:1-175. Harris, R. 1986. Modeling sustainable harvest rates fo r grizzly bears. Appendix K, In: the grizzly bear in Northwestern M ontana. Final program m atic EIS. A.R. Dood, R. Brannon, and R.D. Mace, Eds. M ontana Dept. Fish, Wildlife, and Parks. Helena MT. pp 268-279. Harris, R., and F.W. Allendorf. 1989. Genetically effective population size of large mammals; an assessment of estimators. Conservation Biology 3:181-191. ' Hard, D.L. 1988. A prim er of population genetics, 2nd edition. S inauer Ass. Sunderland Mass. 305 pp. Herrero, S., and D. Hamer. 1977. Courtship and copulation of a pair of grizzly bears, w ith comments on reproductive p lasticity and strategy. J. Mammal. 58(3):441-444. Hill, W.G. 1972. Effective size of populations with overlapping generations. Theor. Pop. Bio. 3:278-289. Hill, W.G. 1987. DNA fingerprints applied to animal and b ird populations. Nature 327:98-99. Homocker, M.G. 1962. Population characteristics and social and reproductive behavior of the grizzly bear in Yellowstone National Park. Unpub. M.S. thesis, Univ. Mont. Interagency Grizzly Bear Committee. 1987a. Grizzly Bear Compendium . National Wildlife Federation. W ashington D.C. 540 pp. In teragency Grizzly Bear Committee. 1987b. Interagency Grizzly Bear guidelines. U.S. Dpt. Agric. For. Ser. Washington D.C. 100 pp. Jeffreys, A.J., and D.B. Morton. 1987. DNA fingerprints of dogs and cats. Animal Genetics 18:1-15. 168 Jeffreys, A J., V. Wilson, R. Kelly, BA. Taylor, and G. Bullfield. 1987. Mouse DNA figerpiints: analysis of chromosome localization and germ -line stability of hypervariable loci in recom binant inb red strains. Nucleic Acids Res. 15:2823-2836. Jeffreys, A.J., V. Wilson, and S i. Thein. 1985a. H ypervariable 'm inisatellite' regions in hum an DNA. Nature 314:67-73. Jeffreys, A.J., V. Wilson, and S.I. Thein. 1985b. Individual- specific 'fingerprin ts’ of hum an DNA. Nature 316:76-79 Jonkel, C. and I.M. Cowan. 1971. The black bear in the spruce-fir forest. Wildl. Monog. No. 27 .57 pp. Kemp, GA. 1976. The dymanics and regulation of black bear, Ursus am ericanus. populations in no rthern Alberta. Int. Conf Bear Res. and Manage, 3:191-198. Kimura, M., and J.F. Crow. 1964. The num ber of alleles th a t can be m ain ta ined in a finite population. Genetics 49: 725-738. Knight, R.R., and L. Eberhardt. 1985. Population dynam ics of Yellowstone grizzly bears. Ecology 66(2):323-334. Knudsen, K. 1992. Presentation a t the 9 th annual In ternational Bear Association meetings. Missoula, MT. Kowalska, Z. 1965. Cross-breeding between a female European brow n bea r CUrsus arctos) and a male polar bear CUrsus m aritim us) in the Logzkim Zoo. Prezegiad Zoologicznum 9(3):313-319. Original no t seen. Cited in Shields and Kocher 1991. Kurten, B. 1964. The evolution of the po lar bear, Ursus maritimns Phipps. Acta Zool. Penn. 108 1-30. Kurten, B. 1968. Plesitocene mammals of Europe. W eidenfeld and Nicolson, London. Kurten, B. 1976, The cave bear story. Columbia University Press. New York. 169 Lande, R., and G. Barrowclough. 1938. Effective popula tion size, genetics variation, and the ir use in population m anagem ent, hi: Genetics and Conservation. C. M Schonewald-Cox et. al., Eds. (Benjamin-Cummings, California), pp . 87-123. Larsen, T., H. Tegelstrom, R.K. Juneja, and M.K. Taylor. 1983. Low p ro te in variability and genetic sim ilarity between populations of th e po lar bea r (Ursus m aritim usL Polar Res. 1:97-105. Lehman, N., A. Eisenhawer, K. Hansen, L.D. Mech, R.O. Peterson, P.J.P. Gogan, and R.K. Wayne. 1991. Introgression of coyote m itochondrial DNA in to sympatric N orth American gray wolf populations. Evolution 45:104-119. Lehman, N., P. Clarkson, L.D. Mech, T.J. Meier, and R.K. Wayne. 1992. A study of th e genetic relationships w ithin and among wolf packs using DNA fingerprinting and m itochondrial DNA. Behav. EcoL Sociobiol. 30:83-94. Lewontin, R.C., and J. Krakauer. 1973. D istribution o f gene frequency as a te st of the theo ry of the selective neu tra lity of polym orphism s. Genetics 74:175-195. . Longmire, J.L., A.K. Lewis, N.C. Brown, J. M. Buckingham, L.M. Clark, M.D. Jones, L.J. Meincke, J. Meyne, R.L., Ratliff, RA. Ray, R. P. Wagner, arid R.K. Moyzis. 1988. Isolation and m olecular characterization of a highly polym orphic centrom eiic tandem repea t in the fam ily Falconidae. Genomics. 2:14-24. Longmire, J.L., Kraemer, P.,Brown, N., Hardekopf, L., Deaven,. L. 1990. A new m ulti-locus DNA fingerprin ting probe: pV47-2. Nucleic Acids Res.l8:1658. Lynch, M. 1988. Estimation of relatedness by DNA fingerprinting. Mol. Biol. Evol. 5:584-599. Lynch, M. 1990. The sim ilarity index and DNA fingerprinting. Mol. Biol. Evol. 7:478-484. Kurten, B., and E. Anderson. 1980. Pleistocene mammals of North America. Columbia University Press. New York. 170 Lynch, M. 1991. Analysis of population genetic struc tu re by DNA fingerprinting. First Inti. Conf. DNA Fingerprinting. London, UK. Maniatis, T., Fritsch, E.F. and Sambrook, J. 1982. Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Lab. Press, Cold Spring Harbor NY. Manlove, M.N., R. Baccus, M.R. Felton, M.H. Smith, and D. Graber. 1980. Biochemical variation in the black bear. In: Bears- th e ir biology and managem ent: Proceedings of 4 th In ternational Conference on Bear Research and Management, Kalispell, MT, Feb. 1977. Eds. C.J. M artinka and K.L. McArthur. Bear Biology Association, Tonto Basin, AZ. pp. 37-41. Meng, A., R.E. Carter, and D.T. Parkin. 1989. The variability of DNA fingerp rin ts in th ree species of swan. Heredity 64:73-80. Modafferi, R.D. 1984. Review of Alaska Peninsula brow n bear investigations. Final Rep. Alaska Dept. Fish and Game. Juneau. 43 pp. M ullenbach, R., P.J.L. Lagoda, and C. Welter. 1989. An efficient salt- chloroform extraction of DNA from blood and tissues. Trends in Genetics 5:391. Murie, Adolph. 1981. The grizzlies of Mount McKinley. Scientific M onograph Series: no. 14. U.S. Dept. hit. National Park Service. 251 PP- Nash, W.G., and S.J. O'Brien. 1987. A comparative chromosome banding analysis of the Ursidae and the ir relationship to o ther carnivores. Cytogenet. Cell. Genet. 45:206-212. Nei, M. 1972. Genetic distance between populations. American N aturalist 106(949): 283-292. Nei, M. 1973. Analysis of gene diversity in subdivided populations. Proc. Nat. Acad. ScL 70:3321-3323. Nei, M. and A. Chakravarti. 1977. Drift variances of Fsx anc ^Gsx statistics ob tained from a finite num ber of isolated populations. Theor. Pop. Bio. 11:307-325. 171 Nowak, R.M, J.L. Paradiso. 1983. Walker's Mammals of the World, 4 th edition. Johns Hopkins University Press, pp 964-977. N um berg, P. L. Roewer, H. Neitzel, K. Sperling, A. Popperl, J. Himdrieser, H. Poche, C. Epplen, H. Zischler, and J.T. Epplen. 1989. DNA fingerprinting with the oligonucleotide probe (CAC)g/(GTG)g: somatic stability and germ line m utations. Human Genetics 84:75-78. Packer, C., D.A. Gilbert, A.E. Pusey, and S.E. O'Brien. 1991a. A m olecular genetic analysis of kinship and cooperation in African lions. N ature 351: 562-565. Packer, C., A.E. Pusey, H. Rowley, D.A. Gilbert, J. M artenson, and S.E. O'Brien. 1991b. Case study of a population bottleneck: lions of the Ngorongoro crater. Conservation Biology 5:219-230. Parker, GA. 1989. Sperm com petition and the evolution of anim al m ating strategies. Chapter I In: Sperm com petition and the evolution o f an im al m ating systems. R.L. Smith, Ed. Academic Press, New York. 686 pp. Paetkau, D. Personal communications. 1993. Discussions during collaborative laboratory work in Edmonton, Alberta, Canada. Septem ber. Paetkau, D. and C. Strobeck. 1994a. Microsatellite analysis of genetic varia tion in black bear populations. Molecular Ecology, in press. Paetkau, D. and C. Strobeck. 1994b. Microsatellite analysis of genetic varia tion in po lar bear populations, in prep. Parker, GA. 1989. Sperm com petition and the evolution of anim al m ating strategies. Chapter I In: Sperm com petition and the evolution of anim al m ating systems. Academic Press. NY. 686 pp. Peek, J.M., M.R. Pelton, H.D. Picton, J.W. Schoen, and P. Zager. 1987. Grizzly bea r conservation and managem ent: a review. Wildl. Soc. Bull. 15:160-169. 172 Quinn. T.W., T.S. Quinn, F. Cooke, and B.N. White. 1987. DNA m arker analysis detects m ultiple m atern ity and p a te rn ity in single broods of the lesser snow goose. Nature 326:392-394. Ralls, K. and J. Ballou. 1983. Extinction: lessons from zoos, in Genetics and Conservation . C. M Schonewald-Cox et al., Eds. (Benjamin- Cummings, California), pp.164-184. Reed, J.M., P.D. Doerr, and J.R. Walters. 1986. Determining m inim um popu la tion sizes fo r b irds and mammals. Wildlife Society Bulletin 14: 255-261. Reeve, H.K., D.F. Westneat, W.A. Noom, P.W. Sherman, and C.F. A quadro. 1990. DNA fingerprinting reveals high levels of inbreeding in colonies of the eusocial naked mole-rat. Proc. Natl. Acad. Sci. 87:2496-2500. Reynolds H.V. 1978. Structure, status reproductive biology, m ovem ent, d istribu tion and hab itat utilization of a grizzly bear popu la tion in NPR-A. Fed. Aid Wildl. Rest. Proj. Final Report. 105-C studies, work group No. 3. Alaska Dept. Fish and Game. Juneau, AR. Reynolds, H.V. 1980. North slope grizzly bear studies. Alaska Dept. Fish and Game, Fed. Aid in Wildl. Rest. Final Rep. Proj. W-17-6, and 7. Juneau, AR. 20pp. Reynolds, H.V. 1989. Grizzly bear population ecology in the W estern Brooks Range, Alaska. Progress Report 1988. Alaska Dept. Fish and Game and U.S. Nat. Park Service. Fairbanks, AR. Reynolds, H.V. 1991. Progress Report: grizzly bear population ecology in the w estern brooks range, Alaska. Alaska Dept. Fish and Game, Feb. 1991. Reynolds, H.V. 1992. Grizzly bear population ecology in the W estern Brooks Range, Alaska. Progress Report 1990 and 1991. Alaska Dept. Fish and Game and U.S. Nat. Park Service. Fairbanks, AR. Reynolds, H.V. Personal communications 1994. Results of aerial census, hom e range estimates, and field observations being p repared fo r final report, Alaska Dept. Fish and Game. 173 Reynolds, H.V., and J. Hechtel. 1980. Structure, status, reproductive biology, movement, distribution, and hab ita t utilization of a grizzly bea r population . Alaska Dept. Fish and Game, Fed. Aid in Wildl. Rest. Prog. Rep. Proj. W-17-11. Juneau, AK. 65 pp. Reynolds, H.V., and J. Hechtel. 1984. Structure, status, reproductive biology, movement, distribution, and hab ita t utilization of a grizzly bea r population. Alaska Dept. Fish and Game, Fed. Aid in Wildl. Rest. Prog. Rep. Proj. W-21-1, W-22-1, and W-22-2. Juneau , AK. 29 pp. Robinson, R. 1982. Genetics for dog breeders. (Pergamon Press, Oxford, U.K.) p. 22. Rockwell, D. 1991. Giving Voice To Bear. Roberts Rhinehart, Niwot, Colo. 224 pp. Ryman, N., R. Baccus, C. Reuterwall, and M.H. Smith. 1981. Effective popula tion size, generation interval, and potential loss of genetic variab ility in game species under d ifferent hun ting regimes. Oikos 36: 257-266. Saiki, R.K., S.J. Scharf, F. Faloona, K.B. Mullis, G.T. Horn, H.A. Ehrlich, and N. Am heim . 1985. Enzymatic amplification of b-globin genomic sequences and restric tion site analysis for diagnostic o f sickle cell anem ia. Science 230: 1350-1354. Saiki, R.K., D.H. Gelfand, S. StofFel, S.J. Scharf, R. Higuchi, G.T. Horn, K.B. Mullis, and H.A. Ehrlich. 1988. Prim er-directed enzym atic am plification of DNA with a therm ostable DNA polym erase. Science 239: 487-491. Schldtterer, C., W. Amos, D. Tautz. 1991. Conservation of polymorphic sim ple sequence loci in cetacean species. Nature 63:354-355. Servheen, C. 1990. The status and conservation of the bears of the world. M onograph Series No. 2. 8 th Inti. Conf. Bear Res. Manage. Shaffer, M.L. 1991. M inimum population size for species conservation. BioScience 31:131-134. 174 Shields, G.F., and T.D. Kocher. 1991. Phylogenetic relationships of N orth American ursids based on analysis of m itochondrial DNA. Evolution 45(1):218-221. Shields, G.F., and S. Talbot. Personal communications w ith E.R. Vyse. 1994. Phone conversations, February. Slatkin, M. 1977. Estimating levels of gene flow in n a tu ra l populations. Genetics 99:323-335. Slatkin, M. 1982. Testing neu tra lity in a subdivided population. Genetics 100:533-545. Soule, M.E. 1980. Thresholds for survival: m aintaining fitness and evolu tionary potential. In: Conservation Biology. M. Soule and B. A Wilcox, Eds. (Sinauer. Sunderland, MA). pp .151 -169 .. Taberlet, P., and J. Bouvet. 1992. Molecular tools and the m anagem ent of endangered mam m alian species: fu tu re prospects and p resen t results concerning the Pyrenean brown bea r population, d ra ft subm itted to Conservation Biology. Tautz, D. 1989. Hypervariability of simple sequences as a general source for polym orphic DNA markers. Nucleic Acids Research 17:6463 Taylor, W.P. Jr., H.V. Reynolds, and W.B. Ballard. 1989. Immobilization of grizzly bears w ith Tiletam ine hydrochloride and Zolazepam hydrochloride. J. Wildl. Manage. 53(4): 978-981. U.S. Fish and Wildlife Service. 1982. Grizzly bear recovery plan. U.S. Dept. Int. Fish and Wildl. Sen W ashington D.C. 199 pp. U.S. Fish and Wildlife Service. 1993. Grizzly bear recovery plan. U.S. Dept. Int. Fish and Wildl. Sen Washington D.C. 300 pp. Vassart, g., Georges, M., Monsieur, R., Brocas, H., Lequarre, A.S., Christophe, D. 1987. A sequence in M13 phage detects hypervariab le m inisatellites in hum an and animal DNA. Science 235, 683-684. 175 Van Gelder, 1977. Mammalian hybrids and genetic lim its. Original n o t seen. Amer. Mus. Novit., No. 2635. 25 pp. cited in Nowak and Paradise. 1983. Wayne, R.K., D.A. Gilbert, N. Lehman, K. Hansen, A Eisenhawer, D. Girman, R.0 Peterson, L.D. Mech, P.J.P. Gogan, U.S. Seal, and R.J. K rum enaker. 1991. Conservation genetics of the endangered Isle Royale gray wolf. Conservation Biology 5(1):41-51. W eber, J., and P. May. 1989. A bundant class of hum an DNA polym orphism s which can be typed using the polym erase chain reaction. American Journal of Human Genetics. 44. pp. 388-396. Weiss, M.L., V. Wilson, C. Chan, T. Turner, and A.J. Jeffreys. 1988. Application of DNA fingerprinting probes to Old W orld monkeys. Am. J. Prim at. 16:73-79. W estneat, D.F., W.A. Noon, K.K. Reeve, and C.F. Aquadro. 1988. Im proved hybridization conditions for DNA 'fingerprints' p robed w ith M13. Nucleic Acids Res. 16:161. Wetton, J.H., R.E. Carter, D.T. Parkin, and D. Walters. 1987. Demographic study of a wild house sparrow population by DNA fingerprin ting . Nature 327:147-149. Weller, P.,Jeffreys, A.J., Wilson, V., and Blanchetot, A. 1984. DNA fingerprin ting w ith multi-locus probes. EMBO J. 3:439-446. W hite R., Leppert M., Bishop D.T., Barker D., Berkowitz J., Brown C., Callahan P., Holm T. and Jorominski L. 1985. Construction of linkage maps with DNA m arkers for hum an chromosom es. Nature 313:101-105. White, T.J., N. Amheim , and H.A. Erlich. 1989. The polym erase chain reaction. T rends Genet. 5:185-189. W ilhams, J.G.K., A.R. Kubelik, K.J. Livak, J.A. Rafalski, and S.V. Tingey. 1990. DNA polym orphism s am plified by arb irta ry p rim ers are useful as genetic m arkers. Nucleic Acids Res. 22:65316535 176 Wong, Z., V. Wilson, A J. Jeffreys, and S.L. Thein. 1986. Cloning a selected fragm ent from a hum an DNA 'fingerprint': isolation of an extrem ely polym orphic m inisatellite. Nucleic Acids Res. 14:4605- 4616. Wright, S. 1931. Evolution in mammalian populations. Genetics 16:97-159. W right, S. 1943. Isolation by distance. Genetics 28:114-138. Wright, S. 1969. Evolution and the genetics of populations. Vol. 2 The theory of gene frequencies. Univ. of Chicago Press, Chicago IL Wyman, A. and White, R. 1980. A highly polym orphic locus in hum an DNA. Proc. Natn. Acad. Sci. U.S.A. 77:6754-6758. Young, B.F., and R.L. Ruff. 1982. Population dynamics and m ovem ents of black bears in east central A lberta. J. Wildl. Manage. 46(4):845-860. Zeff, RA., anh J Gellebter. 1987. Oligonucleotide probes fo r genomic DNA blots. 1987. Bethesda Research Laboratories Focus 9(2): 1-2. Zischler, H., I. Nanda, R. Schafer, M. Schmid, and J.T. Epplen. 1989. Digoxigenated ologonucleotide probes specific for simple repeats in DNA fingerprinting and hybridization in situ. Human Genetics 82:227-233. Zimmerman, E.G. 1989. Mitochondrial DNA analysis of North American black bears, Ursus am ericanus. Unpub. Rep., Dep't. Biol. Sciences, Univ. North Texas, Denton TX. 14pp. 177 APPENDICES 178 APPENDIX A Individual bears and genotypes 179 Table 17. Sex/age/relationships of WBR bears. I Number I Se> Age* Known offspring I Mother Father CommentsJ 1081 M 20 I M r / I / I o / I 741 1087 F 16 1483/1484 1086 mate 10981089 F 19 1704/1705/1706 mate 14 1 1J 1095 F 21 1727 mate 14771097 F 23 1480/1481/1482 mates 1453/Hvo51098 M 18 1483/1484/1750 1103 M 23 -1124 M 34 I4 b y /14 /0 /1486 /1487 /1494 /149S 1125 F 18 1726 mate Hyp 7 |1136 F 16 1469/1470 1134 mate 1124 ' |1141 F- 14 1485 1139 mate Hyp 6 |1147 M 17 1149 F 18 1486/1487 mate 1124 |1157 M 19 I 1158 F 21 1471/1472 1165 M 16 1166 F 23 1701/1702 mates 1421/1478 I1167 F 20 1434/1435/1436 born 1987/no sample |1174 F 12 1497 1105 mate Hyp3 |1177 F 13 14 1496 1104 mate 1411J 1179 F 1742 1178 mate 14111232 M 13 not from study arpq |1233 M 20 1405 M 15 1407 F 18 cubs 1986/1989 I1411 M 14 1496 /1704 /1705 /1706 /1 7 4 ? 1413 F 16 cubs 1986/1988 I1416 F 15 cubs 1986/'87/'91 "I1417 F 15 no cubs observed |1420 M 13 1756/1757 1421 M 19 1702/1731/1732 1424 F 14 1466 nate Hypl/lost cub'9l|1425 F 13 1708/1709/1710 Inate Hypl |1431 F 13 Tom Timber Creek I1433 M 18 rom Timber Creek I1437 F 14 1488/1489 rnates Hyp2/Hvo5 f8)1438 F 18 1756/1757 rnate 14201439 F 19 1753/1754/1755 rnates 1712/Hyp 6 I1440 F 19 1707 nnate Hyp6 |1442 M 5 1441 killed by bear 1989 I1443 M 5 1441 dled in den 1989 |1444 M 5 1441 died in den 1989 II 1451 I F 19 Cubsl 988 I j 180 Table 17 (continued). Sex/age/relationships of WBR bears. I Number | Sex Age* I Known offspring^ Mother Father Comments |I 1452 F Adult from Omar RiverI 1453 M 22 1480/1481 J 1454 F 14 1498/1499/1500 mates Hypl/Hyp3I 1455 M 11 I 1456 M 14 killed by hunter 1989 p i 457 F 14 1731/1732 mate 1421 I I 1458 F 11 1494/1495 mate 1124 | I 1459 M 21 1744 ----------------1I 1460 F 12 1492/1493 mate 1712 I 1461 . F 14 1743/1744 mates 1459/Hyp7 \ 1462 M 8 I 1463 M 13 1713/1714/1715 not collared I 1464 F 11 1713/1714/1715 mate 1463 (1157?)I 1465 F 12 cubs 1989/lost 1990 I 1466 M 5 1424 Hypl 1467 F 10 cub 1988? 1468 F 12 no cubs observed I 1469 M 5 1136 1124 j 1470 M 5 1136 1124 1473 F 9 cub 1991 1474 F 7 no cubs observed | 1475 F 27 not collared marked '89/no cub obg 1476 M 11 not collared 1477 M 13 1727/UM cub? not collared I 1478 M 13 1701 not collared 1479 F 9 1758 mate Hyp2 1480 M 3 1097 1453 1481 F 3 1097 1453 1491 possible father 1482 M 3 1097 Hyp 5 died 1990 | 1483 F 3 1087 1098 j 1484 F 3 1087 1098 j 1485 M 3 1141 Hyp 6 dead 1991? | 1486 M 4 1149 1124 J 1487 F 4 1149 1124 J 1488 M 2 1437 Hyp2 /8 J 1489 M 2 1437 Hyp 5 /8 j 1490 M 8 ib of 1453? I 1492 M 3 1460 1712 J I 1493 M 3 1460 1712 j 1494 M 3 1458 1124 I I 1495 F 3 1458 1124 I I 1496 M 3 1177 1411 ' I 181 Table 17 (continued). Sex/age/relationships of WBR bears. Number Sex Age* Known offspring^ Mother Father Comments 1497 M 3 1174 1411 1498 F 3 1454 Hyp3 1499 F 3 1454 Hypl 1500 F 3 1454 Hyp3 1701 F 3 1166 1478 1702 F 3 1166 1421 1703 M 16 1704 F 3 1089 1411 1705 M 3 1089 1411 1706 F 3 1089 1411 1707 M 3 1440 Hyp6 1708 M 3 1425 Hypl 1709 M 3 1425 Hypl 1710 M 3 1425 Hypl 1711 F 5 1712 M 18 1492 /1493 /1735 /1736 /1753 /1754 /1755 1713 F 3 1464 1463 1714 M 3 1464 1463 1157 possible father 1715 F 3 1464 1463 1716 F ' 9 1717/1718 mate 1081 1717 F I 1716 1081 1718 F I 1716 1081 1719 M 4 obs with 1487 in'91 1720 M 17 1721 F 3 unknown sib 1722 1722 M 3 unknown sib 1721 1723 M 4 marked '91/weaned 1724 M 5 marked '91/weaned 1725 M 5 marked '91/weaned 1726 M 4 1125 Hyp 7 2 UM sibs 1989-1991 1727 M 2 1095 1477 I UMsib 1990-1991 1728 M 4 marked '92/weaned 1729 M 15 marked '92 1730 M 5 marked '92/weaned 1731 F 2 1457 1421 1732 F 2 . 1457 1421 1733 M 3 marked '92/weaned 1734 F 13 1735/1736 mate 1712 1735 M 2 1734 1712 1736 M 2 1734 1712 1737 F 26 17127/1739? marked '92/died '92 1738 M 4 marked '92/weaned 1739 F 8 1740/1741 mate 1081 182 Table 17 (continued). Sex/age/relationships of WBR bears. Number Sex Age* Known offspring^ Mother Father Comments 1740 M I 1739 1081 1741 M I 1739 1081 1742 F 3 1179 1411 1743 F 2 1461 Hyp7 1744 M 2 1461 1459 1745 F 5 1746/1747 1749? 1746 F Cub 1745 Hyp4 died '9Z? 1747 M Cub 1745 Hyp4 died '92? 1748 F 3 marked '92/weaned? 1749 F 15 1750/1751/1752 mates 1098/Hyp2 1750 F I 1749 1098 1751 F I 1749 HypZ 1752 M I 1749 HypZ 1753 M 3 1439 1712 1754 F 3 1439 Hy p 6 1755 M 3 1439 1712 1756 M I 1438 1420 1757 F I 1438 1420 1758 M Cub 1479 HypZ UNKI unknown mislabelled as 1717 UNK2 unknown mislabelled as 1125 UNK3 M 3 same age/sex as 1480 UNK4 F 3 same age/sex as 1481 * Age is given as age in 1992 for purposes of comparison ^Known offspring may run over into following columns 183 Table 18. WBR individual alleles. # Ioc Al Ioc A2 Ioc BI Ioc 82 Ioc Cl Ioc C2 Ioc Dl Ioc 02 IocLI Ioc L2 Ioc Ml Ioc M2 Ioc Pl Ioc P2 Ioc Xl Ioc X2 1081 Al 80 Al 84 8158 8160 C103 O i l D172 D178 LI 57 LI 59 M210 M212 Pl 53 Pl 57 Xl 37 Xl 37 1087 Al 84 A192 BI 56 BI 60 Cl 03 Cl 13 Dl 78 Dl 81 LI 55 LI 57 M208 M214 Pl 53 Pl 53 Xl 37 Xl 37 1089 Al 92 Al 94 BI 40 8164 Cl 03 C l l l Dl 72 0178 LI 55 LI 57 M208 M208 Pl 53 Pl 61 Xl 35 Xl 41 1095 Al 84 Al 94 8140 8148 Cl 03 0 1 3 Dl 72 Dl 81 LI 55 LI 57 M208 M208 Pl 57 P161 X131 Xl 37 1097 Al 84 Al 92 8156 BI 60 C105 0 1 3 Dl 72 Dl 77 LI 55 LI 59 M208 M208 Pl 53 Pl 53 Xl 35 Xl 37 1098 Al 84 Al 94 BI 50 BI 60 Cl 11 cm Dl 82 Dl 82 LI 55 LI 55 M206 M208 Pl 55 Pl 61 Xl 33 Xl 37 1103 Al 92 A200 8158 BI 60 C105 0 0 5 Dl 72 Dl 72 LI 57 LI 61 M208 M208 Pl 51 Pl 55 Xl 37 X141 1124 Al 84 Al 94 B148 BI 60 cm 0 1 3 Dl 77 Dl 78 LI 55 LI 55 M208 M214 Pl 57 Pl 59 Xl 37 Xl 37 1125 Al 94 Al 94 BI 60 BI 60 Cl 03 0 0 5 Dl 76 Dl 78 LI 59 LI 61 M208 M21 8 Pl 53 Pl 53 Xl 37 X141 1136 Al 84 Al 94 8160 8160 C103 0 0 5 D177 Dl 78 LI 55 LI 55 M208 M214 Pl 53 Pl 57 Xl 37 X137 1141 Al 84 Al 84 8140 8150 C103 O i l Dl 74 Dl 82 LI 55 LI 55 M208 M208 Pl 51 Pl 61 Xl 33 X141 1147 Al 84 Al 94 8148 8160 C105 0 0 5 Dl 72 Dl 81 LI 55 LI 59 M208 M214 Pl 49 Pl 57 Xl 35 Xl 37 1149 Al 94 Al 94 8158 8164 Cl 03 0 1 3 Dl 72 Dl 78 LI 55 LI 57 M208 M212 Pl 53 Pl S3 Xl 37 X141 1157 Al 94 Al 94 8160 8160 C103 C l l l Dl 72 Dl 76 LI 57 LI 59 M208 M218 Pl 53 Pl 57 X131 Xl 37 1158 Al 80 Al 84 8140 8158 C105 Cl 11 Dl 78 Dl 81 LI 55 LI 61 M208 M208 Pl 55 Pl 61 Xl 29 Xl 37 1165 Al 92 Al 94 8160 BI 60 Cl 11 O i l D177 D178 LI 55 LI 59 M208 M212 Pl 53 Pl 57 Xl 29 Xl 33 1166 Al 92 Al 92 8158 8160 0 0 3 0 0 5 Dl 78 Dl 80 LI 59 LI 61 M208 M214 Pl 51 Pl 57 Xl 37 X141 1167 Al 84 Al 94 8140 8160 0 0 3 O i l Dl 72 Dl 77 LI 59 LI 61 M208 M214 Pl 51 Pl 53 Xl 33 Xl 37 1174 Al 94 Al 94 8148 8160 0 0 3 0 0 5 Dl 72 Dl 81 LI 55 LI 55 M208 M208 Pl 49 Pl 53 X135 Xl 37 1177 Al 84 Al 94 8156 8158 Cl 13 0 1 3 Dl 77 Dl 78 LI 59 LI 59 M208 M208 Pl 53 Pl 59 X135 X H l 1179 Al 92 Al 94 8158 8160 0 0 3 0 1 3 Dl 77 Dl 78 LI 55 LI 57 M212 M214 Pl 53 Pl 57 Xl 37 X H l 1232 Al 80 Al 94 8148 8160 0 0 3 0 0 5 Dl 72 Dl 81 LI 55 LI 57 M208 M218 Pl 51 Pl 51 Xl 31 X H l 1233 Al 80 Al 94 8140 8160 0 0 5 O i l Dl 77 Dl 86 LI 57 LI 57 M208 M214 Pl 53 Pl 57 X137 X141 1405 Al 94 Al 94 8158 8160 0 0 5 O i l Dl 72 Dl 77 LI 57 LI 57 M208 M214 Pl 53 Pl 53 X141 X H l 1407 Al 84 Al 92 8140 8156 0 0 3 0 0 5 0174 Dl 77 LI 55 LI 59 M214 M214 Pl 53 Pl 53 Xl 29 X135 1411 Al 94 Al 94 8148 8158 0 0 3 O i l Dl 77 Dl 86 LI 57 LI 57 M210 M214 Pl SI Pl 57 Xl 33 Xl 37 1413 Al 84 Al 94 8158 8160 0 0 3 0 0 5 Dl 72 Dl 86 LI 55 LI 63 M208 M214 Pl 53 Pl 61 Xl 33 Xl 35 1416 Al 80 Al 92 8140 8160 0 0 3 O i l Dl 78 Dl 81 LI 55 LI 55 M208 M210 Pl 61 Pl 61 Xl 33 Xl 37 1417 Al 92 Al 94 8158 8160 O i l o n Dl 78 Dl 86 LI 55 LI 55 M208 M214 Pl 53 Pl 57 Xl 33 Xl 35 1420 Al 94 Al 94 8156 8160 0 0 3 O i l Dl 74 Dl 82 LI 55 LI 61 M206 M214 Pl S3 Pl 61 Xl 35 Xl 37 1421 Al 92 Al 94 8160 8160 0 0 5 Cl 11 Dl 76 Dl 86 LI 55 LI 57 M212 M214 Pl 53 Pl 57 X141 X H l 1424 Al 84 Al 92 8150 8160 0 0 5 0 0 5 Dl 81 0184 LI 55 LI 55 M208 M214 Pl 55 Pl 61 Xl 37 Xl 37 1425 Al 94 Al 94 8148 8160 0 0 9 Cl 11 Dl 77 Dl 77 LI 55 LI 55 M206 M208 Pl 57 Pl 59 Xl 37 Xl 37 1431 Al 80 Al 86 8152 8160 O O I 0 0 3 Dl 78 Dl 84 LI 55 LI 57 M208 M210 Pl 51 Pl 53 Xl 29 X H l 1433 Al 84 Al 94 8148 8160 0 0 5 0 0 5 Dl 78 Dl 84 LI 55 LI 55 M212 M214 P149 Pl 53 Xl 37 Xl 41 1437 Al 84 Al 94 8148 8148 0 0 5 0 1 3 0177 Dl 78 LI 55 LI 55 M208 M214 Pl 53 Pl 59 Xl 35 Xl 37 1438 Al 80 Al 84 8140 8164 0 0 5 0 0 7 Dl 78 Dl 82 LI 55 LI 57 M212 M214 Pl 53 Pl 61 X131 Xl 35 1439 Al 80 Al 94 8160 8164 0 0 3 0 0 5 Dl 72 Dl 82 LI 55 LI 55 M212 M214 Pl 53 Pl 61 Xl 37 Xl 37 1440 Al 94 Al 94 8148 8160 0 0 5 0 0 5 Dl 72 Dl 84 LI 57 LI 59 M208 M212 Pl 53 Pl 53 Xl 35 X141 1442 Al 84 A200 8140 8158 0 0 5 O i l Dl 72 Dl 78 LI 55 LI 61 M206 M208 Pl 55 Pl 55 Xl 37 Xl 37 1443 Al 84 A200 8140 8158 0 0 5 O i l Dl 72 Dl 78 LI 57 LI 57 M208 M208 Pl 51 Pl 55 Xl 37 X H I 1444 Al 84 Al 92 8140 8158 0 0 5 O i l Dl 72 Dl 78 LI 57 L161 M208 M208 Pl 55 Pl 55 Xl 37 X H I 1451 Al 92 Al 94 8140 8154 0 0 3 0 0 5 Dl 72 Dl 81 LI 57 L161 M208 M210 Pl 55 Pl 61 Xl 33 Xl 35 1452 Al 80 Al 94 8156 8160 0 0 5 0 0 5 Dl 77 Dl 86 LI 55 LI 55 M208 M214 Pl 57 Pl 57 X131 Xl 37 1453 Al 84 Al 94 8152 8152 0 0 5 cm Dl 72 Dl 77 LI 57 L161 M208 M212 Pl 59 Pl 61 Xl 33 Xl 37 1454 Al 84 Al 84 8140 8158 0 0 5 o n Dl 82 Dl 86 LI 59 L171 M208 M214 P149 Pl 57 Xl 35 X H I 1455 Al 84 Al 92 8140 8140 0 0 3 0 0 5 Dl 72 0176 LI 55 LI 59 M212 M214 Pl 51 Pl 61 X131 Xl 37 1456 Al 94 A200 8140 8156 O i l 0 1 3 Dl 77 Dl 86 LI 59 L161 M210 M212 Pl 53 Pl 53 Xl 29 Xl 33 1457 Al 94 Al 94 BI 60 BI 60 C103 cm Dl 78 Dl 78 LI 57 L161 M218 M21 8 Pl 53 Pl 53 Xl 33 X141 1458 Al 84 A192 8140 8160 0 0 5 0 0 5 Dl 72 Dl 74 LI 55 LI 57 M210 M214 Pl 53 Pl 61 Xl 35 Xl 35 1459 Al 80 Al 94 8158 8160 0 0 3 O i l Dl 72 Dl 77 LI 55 LI 57 M206 M212 Pl 53 Pl 57 Xl 33 Xl 3 5 1460 Al 90 Al 92 8148 8160 0 0 5 0 0 5 Dl 74 Dl 77 LI 55 LI 57 M208 M214 Pl 53 Pl 61 Xl 35 Xl 35 1461 Al 84 Al 90 8148 8148 0 0 3 0 1 3 Dl 77 Dl 78 LI 55 LI 55 M208 M214 Pl 53 Pl 57 Xl 35 Xl 37 1462 Al 94 Al 94 8160 8160 0 0 3 0 0 7 Dl 72 Dl 81 LI 55 LI 55 M206 M214 Pl 57 Pl 61 Xl 37 Xl 37 1463 Al 84 Al 92 8152 8160 0 0 3 0 0 5 0172 Dl 72 LI 55 LI 55 M206 M208 Pl 53 Pl 53 X131 Xl 35 1464 Al 84 Al 94 8140 8158 0 0 5 Cl 13 0172 Dl 78 LI 55 LI 57 M208 M212 Pl 53 Pl 53 X141 X H I 1465 Al 92 Al 94 8140 8152 0 0 7 Cl 13 Dl 72 Dl 82 LI 57 LI 57 M210 M214 Pl 51 Pl 61 Xl 29 Xl 37 1466 Al 92 Al 94 8150 8160 0 0 5 0 0 5 Dl 78 Dl 84 LI 55 LI 57 M206 M208 Pl 49 Pl 61 Xl 37 X H I 184 Table 18 (continued). WBR individual alleles. # Ioc Al Ioc A2 Ioc BI Ioc 82 Ioc Cl Ioc C2 Ioc Dl Ioc 02 Ioc LI Ioc L2 Ioc Ml Ioc M2 Ioc Pl Ioc P2 Ioc Xl Ioc X2 1467 Al 92 Al 92 B148 BI 50 C105 0 0 5 Dl 72 Dl 84 LI 55 LI 57 M208 M2M Pl 59 Pl 61 Xl 35 Xl 37 1468 Al 80 Al 92 BI 56 BI 60 cm 0 1 3 Dl 72 Dl 78 LI 59 L161 M208 M212 Pl 57 Pl 61 Xl 35 Xl 37 1469 Al 84 Al 94 BI 60 BI 60 C105 0 1 3 Dl 77 0178 LI 55 LI 55 M208 M208 Pl 57 Pl 57 Xl 37 Xl 37 1470 Al 84 Al 84 BI 60 BI 60 C105 0 1 3 Dl 77 Dl 78 LI 55 LI 55 M208 M214 Pl 57 Pl 57 Xl 37 Xl 37 1473 Al 90 Al 92 BI 60 BI 60 C103 0 0 5 0174 Dl 84 LI 55 LI 57 M208 M214 Pl 57 Pl 61 Xl 37 Xl 41 1474 Al 80 Al 94 BI 60 B164 Cl 03 0 0 7 Dl 76 Dl 78 LI 55 LI 55 M214 M2 M Pl 53 P161 Xl 35 Xl 35 1475 Al 92 Al 94 BI 40 BI 60 C103 O i l D172 0182 LI 55 LI 63 M208 M218 Pl 57 Pl 61 X133 Xl 37 1476 Al 92 Al 94 BI 60 BI 64 Cl 07 O i l Dl 74 Dl 81 LI 57 LI 57 M206 M208 Pl 53 Pl 55 Xl 35 Xl 37 1477 Al 80 Al 90 BI 40 BI 60 C103 0 0 7 Dl 78 Dl 78 LI 55 LI 61 M208 M2 M P149 Pl 57 Xl 33 Xl 35 1478 Al 94 Al 94 BI 60 BI 60 Cl OS O i l Dl 72 Dl 82 LI 55 LI 61 M208 M208 Pl 39 P161 Xl 3 1 Xl 37 1479 Al 84 Al 92 BI 50 BI 58 0 0 5 0 0 5 Dl 72 Dl 72 LI 57 LI 61 M208 M208 Pl 55 Pl 61 Xl 37 Xl 41 1480 Al 84 Al 94 BI 52 BI 60 0 0 5 O i l Dl 72 Dl 72 LI 55 LI 57 M208 M212 Pl 53 Pl 59 Xl 35 Xl 37 1481 Al 84 Al 92 BI 52 BI 60 0 0 5 0 0 5 Dl 72 Dl 77 LI 59 LI 61 M208 M208 Pl 53 Pl 59 Xl 33 Xl 35 1482 Al 84 Al 94 BI 60 BI 64 O i l 0 1 3 Dl 72 Dl 78 LI 55 LI 55 M208 M212 Pl 53 Pl 57 Xl 37 X141 1483 Al 92 Al 94 BI 50 BI 60 Cl 11 0 1 3 Dl 81 0182 LI 55 LI 55 M208 M214 Pl 53 Pl 61 Xl 37 X137 1484 Al 84 Al 92 BI 50 BI 60 0 0 3 cm Dl 78 Dl 82 LI 55 LI 55 M208 M208 Pl 53 Pl 55 Xl 37 X141 1485 Al 84 Al 94 BI 40 BI 60 0 0 5 O i l Dl 74 Dl 86 LI 55 LI 55 M208 M208 Pl 51 Pl 57 X141 X141 1486 Al 84 Al 94 BI 60 BI 64 0 0 3 0 1 3 Dl 72 Dl 78 LI 55 LI 55 M208 M212 Pl 53 Pl 57 Xl 37 X141 1487 Al 84 Al 94 B148 8164 O i l 0 1 3 D172 Dl 78 LI 55 LI 57 M208 M2M Pl 53 Pl 59 Xl 37 Xl 41 1488 Al 84 Al 92 B148 BI 60 O i l 0 1 3 Dl 72 Dl 78 LI 55 LI 57 M208 M2M Pl 53 Pl 59 X131 Xl 35 1489 Al 84 Al 94 BI 48 BI 58 0 0 5 O i l Dl 78 Dl 82 LI 55 LI 57 M208 M2M Pl 39 Pl 53 Xl 31 Xl 35 1490 Al 92 Al 92 BI 58 BI 58 0 0 5 0 0 5 Dl 77 Dl 81 LI 55 LI 55 M206 M208 P149 Pl 57 X137 Xl 41 1491 Al 84 Al 92 BI 52 BI 60 0 0 5 0 0 5 Dl 72 Dl 77 LI 59 L161 M208 M208 Pl 53 Pl 59 Xl 33 Xl 35 1492 Al 84 Al 92 BI 58 BI 60 0 0 3 0 0 5 Dl 74 Dl 81 LI 55 LI 55 M208 M2 M Pl 57 Pl 61 Xl 35 X141 1493 Al 84 Al 90 BI 58 BI 60 0 0 5 O i l Dl 77 Dl 86 LI 55 LI 57 M2M M214 Pl 53 Pl 57 Xl 3 5 Xl 3 5 1494 Al 84 Al 92 BI 60 BI 60 0 0 5 0 1 3 Dl 74 Dl 77 LI 55 LI 57 M210 M2 M Pl 59 Pl 61 Xl 35 Xl 37 1495 Al 84 Al 84 B140 B148 0 0 5 cm Dl 72 Dl 78 LI 55 LI 57 M208 M214 Pl 53 Pl 59 X135 Xl 37 1496 Al 94 Al 94 BI 58 BI 58 Cl 11 0 1 3 Dl 77 Dl 78 LI 57 LI 59 M208 M2M Pl 51 Pl 59 Xl 35 Xl 37 1497 Al 94 Al 94 BI 60 BI 60 0 0 3 0 0 5 Dl 72 Dl 86 LI 55 LI 55 M208 M214 Pl 53 Pl 61 Xl 35 Xl 37 1498 Al 84 Al 86 BI 58 BI 60 0 0 5 cm Dl 81 Dl 86 LI 55 LI 71 M208 M214 Pl 49 Pl 53 Xl 29 X141 1499 Al 84 Al 94 BI 40 BMO 0 0 5 O i l Dl 72 0186 LI 55 LI 71 M208 M2 M Pl 49 Pl 51 X141 X141 1500 Al 84 Al 94 BI 40 BMO 0 0 5 0 0 5 Dl 81 Dl 82 LI 55 LI 71 M208 M2 M Pl 53 Pl 57 Xl 29 Xl 41 1701 Al 92 Al 94 BI 58 BI 60 0 0 5 0 0 5 Dl 72 Dl 80 LI 55 LI 59 M208 M208 Pl SI Pl 61 Xl 37 Xl 37 1702 Al 92 Al 94 BI 60 BI 60 0 0 3 0 0 5 Dl 78 Dl 86 LI 55 LI 61 M208 M2M Pl 57 Pl 57 Xl 37 Xl 41 1703 Al 84 Al 84 BI 40 BM 8 O O I 0 0 3 Dl 78 0181 LI 57 LI 57 M212 M2M Pl 51 Pl 53 X131 Xl 37 1704 Al 92 Al 94 BI 58 BI 64 0 0 3 O i l 0177 Dl 78 LI 55 LI 57 M208 M2M Pl 51 Pl 61 Xl 33 Xl 35 1705 Al 94 Al 94 BI 40 BI 58 0 0 3 0 0 3 D178 Dl 86 LI 55 LI 57 M208 M2M Pl 53 Pl 57 Xl 33 X141 1706 Al 94 Al 94 BI 40 BI 58 0 0 3 O i l Dl 77 0178 LI 55 LI 57 M208 M2M Pl 51 Pl 61 Xl 33 Xl 41 1707 Al 80 A194 B148 8160 0 0 5 O i l 0172 Dl 84 LI 57 LI 59 M208 M2M Pl 53 Pl 57 Xl 35 Xl 4 1 1708 Al 80 Al 94 BI 60 BI 60 O i l O i l Dl 72 0177 LI 55 LI 57 M206 M206 P lS l Pl 59 Xl 37 Xl 41 1709 Al 94 Al 94 BI 48 BI 60 O O S cm Dl 77 0178 LI 55 LI 57 M206 M206 Pl 49 Pl 57 Xl 33 Xl 37 1710 Al 80 Al 94 BMO BI 60 cm cm Dl 72 Dl 77 LI 55 LI 55 M208 M208 Pl 51 Pl 57 Xl 37 X141 1711 Al 80 Al 94 BI 60 BI 64 0 0 5 0 0 7 Dl 76 0178 LI 55 LI 55 M212 M2 M Pl 53 Pl 61 Xl 35 Xl 37 1712 Al 84 A192 BI 58 BI 60 0 0 3 O i l Dl 81 Dl 86 LI 55 LI 55 M208 M2M Pl 57 Pl 57 Xl 33 X141 1713 Al 84 Al 94 BI 58 BI 60 0 0 3 0 1 3 0172 Dl 72 LI 55 LI 55 M208 M208 Pl 53 Pl 53 Xl 31 Xl 41 1714 Al 84 A194 BI 58 BI 60 0 0 3 0 1 3 Dl 72 Dl 78 LI 55 LI 57 M208 M212 Pl 53 Pl 53 X131 X141 1715 Al 84 Al 84 BMO BI 60 0 0 5 0 0 5 Dl 72 Dl 78 LI 55 LI 57 M208 M208 Pl 53 Pl 53 Xl 35 Xl 41 1716 Al 92 Al 94 BI 58 BI 64 0 0 3 0 0 5 Dl 72 Dl 72 LI 55 LI 55 M208 M208 Pl 55 Pl 61 X135 X135 1717 Al 80 Al 92 BI 58 BI 60 0 0 3 0 0 5 Dl 72 Dl 78 LI 55 LI 57 M208 M212 Pl 57 Pl 61 Xl 3 5 X137 1718 Al 84 Al 92 BI 58 BI 58 0 0 3 0 0 3 Dl 72 Dl 72 LI 55 LI 59 M208 M212 Pl 53 Pl 55 X135 Xl 37 1719 A192 Al 94 BI 58 BI 60 0 0 3 0 0 5 Dl 72 Dl 77 LI 57 LI 57 M212 M214 Pl 51 Pl 57 Xl 37 Xl 41 1720 Al 92 Al 94 BI 52 BI 60 0 0 3 0 0 5 Dl 81 Dl 81 LI 55 LI 55 M206 M212 Pl 53 Pl 53 Xl 35 Xl 37 1721 Al 84 Al 94 BMO BI 64 C l l l cm Dl 81 Dl 81 L161 L161 M206 M210 Pl 53 Pl 55 Xl 37 Xl 37 1722 Al 94 Al 94 BI 58 BI 60 0 0 3 0 0 5 Dl 86 Dl 86 LI 57 LI 59 M208 M210 Pl 53 Pl 57 Xl 37 X141 1723 Al 92 A200 BI 60 BI 60 0 0 5 0 0 5 Dl 72 Dl 78 LI 55 LI 55 M208 M210 Pl 51 Pl 53 Xl 29 X141 1724 Al 92 Al 94 BMO BI 60 0 0 5 O i l Dl 77 Dl 78 LI 55 LI 57 M208 M208 Pl 51 P lS l Xl 31 Xl 37 1725 Al 94 Al 94 BI 54 BI 58 0 0 5 0 0 7 Dl 77 Dl 78 LI 57 LI 57 M206 M208 Pl 53 Pl 55 X131 Xl 37 1726 Al 80 Al 94 BMO BI 60 0 0 3 0 0 3 Dl 72 Dl 86 LI 59 L161 M208 M214 Pl 53 Pl 57 Xl 33 Xl 37 1727 Al 80 Al 94 BM 8 BI 60 0 0 3 0 0 7 Dl 72 Dl 78 LI 55 LI 61 M208 M214 P149 Pl 61 X131 Xl 35 185 Table 18 (continued). WBR individual alleles. # Ioc Al Ioc A2 Ioc BI Ioc 82 Ioc Cl Ioc C2 Ioc Dl Ioc 02 Ioc LI Ioc L2 Ioc Ml Ioc M2 Ioc Pl Ioc P2 Ioc Xl Ioc X2 1728 Al 94 Al 94 8150 8160 Cl 03 0 0 5 Dl 72 Dl 78 LI 57 L161 M208 M208 Pl 57 Pl 61 Xl 37 X141 1729 Al 92 Al 94 8140 8158 C105 O i l Dl 77 0186 LI 55 LI 61 M206 M214 P149 Pl 53 Xl 37 Xl 37 1730 Al 80 Al 94 8140 8160 C105 O i l Dl 74 Dl 76 LI 57 LI 61 M208 M212 Pl 59 Pl 61 X135 X137 1731 Al 94 Al 94 8160 8160 Cl 03 O i l Dl 76 0178 LI 55 LI 57 M214 M218 Pl 53 Pl 53 Xl 33 X141 1732 Al 92 Al 94 8160 8160 C103 O i l Dl 76 0178 LI 57 LI 61 M212 M222 Pl 53 Pl 53 X141 X141 1733 Al 84 Al 92 8140 8140 C103 0 0 5 Dl 78 Dl 84 LI 59 LI 61 M212 M21 8 Pl 53 Pl 55 Xl 33 Xl 37 1734 Al 90 Al 92 8154 8160 C105 0 0 7 Dl 72 Dl 82 LI 55 LI 57 M208 M214 Pl 53 Pl 61 Xl 37 Xl 37 1735 A190 Al 92 8160 8160 C103 0 0 5 Dl 81 Dl 82 LI 55 LI 57 M208 M214 Pl 53 Pl 57 Xl 37 X141 1736 Al 92 Al 92 8158 8160 C107 O i l 0172 Dl 86 LI 55 LI 57 M208 M214 Pl 53 Pl 57 Xl 33 Xl 37 1737 Al 84 Al 90 8154 BI 60 ClOI 0 0 3 Dl 72 Dl 81 LI 55 LI 57 M214 M214 Pl 57 Pl 57 Xl 37 X141 1738 Al 84 Al 92 8140 8158 C103 0 0 3 0172 Dl 81 LI 55 LI 61 M210 M214 Pl 57 Pl 61 Xl 35 X141 1739 Al 90 Al 92 8140 8160 C103 O i l 0181 Dl 81 LI 57 LI 61 M208 M214 Pl 57 Pl 61 Xl 37 X141 1740 Al 80 Al 92 8140 8160 Cl 11 O i l Dl 78 Dl 81 LI 57 LI 57 M210 M214 Pl 53 Pl 61 Xl 37 X141 1741 Al 80 Al 90 8140 8158 C103 O i l Dl 78 0181 LI 57 LI 57 M208 M210 Pl 57 Pl 61 Xl 37 Xl 37 1742 Al 94 Al 94 8148 8160 C103 0 1 3 Dl 78 Dl 86 LI 55 LI 57 M214 M214 Pl 51 Pl 53 Xl 33 Xl 37 1743 Al 84 Al 94 8148 8160 Cl 03 0 1 3 Dl 77 Dl 86 LI 55 LI 57 M206 M214 Pl 53 Pl 57 Xl 33 Xl 35 1744 Al 90 Al 94 8148 8160 Cl 03 0 1 3 Dl 77 Dl 77 LI 55 LI 57 M206 M208 Pl 53 Pl 57 Xl 35 Xl 35 1745 Al 84 Al 94 8156 8160 CIOS 0 0 5 0172 Dl 84 LI 59 LI 61 M208 M212 Pl 53 Pl 59 Xl 37 Xl 37 1746 Al 94 Al 94 8148 8160 C103 0 0 5 Dl 84 Dl 84 LI 61 LI 61 M206 M212 Pl 59 Pl 59 Xl 35 Xl 37 1747 Al 84 Al 94 8148 8160 C105 0 0 5 Dl 77 Dl 84 LI 59 LI 61 M208 M208 Pl 53 Pl 59 X137 Xl 37 1748 Al 94 Al 94 8158 8160 C105 0 0 5 0172 Dl 72 LI 55 LI 57 M208 M210 Pl 53 Pl 61 Xl 37 X141 1749 Al 88 Al 94 8140 8156 C105 O i l Dl 77 Dl 84 LI 57 LI 61 M208 M208 Pl 53 Pl 59 Xl 37 X137 1750 Al 94 Al 94 8150 8156 cm O i l Dl 77 Dl 82 LI 55 L161 M208 M208 Pl 55 Pl 59 Xl 33 Xl 37 1751 Al 88 Al 92 8156 8160 C105 O i l Dl 82 Dl 84 LI 55 L161 M208 M208 Pl S3 Pl 53 Xl 35 X137 1752 Al 92 Al 94 8156 8160 C105 O i l Dl 77 Dl 82 LI 55 LI 61 M206 M208 Pl 53 Pl 55 X135 Xl 37 1753 Al 92 Al 94 8160 8160 Cl 03 cm Dl 82 Dl 86 LI 55 LI 55 M208 M212 Pl 57 Pl 61 Xl 37 X141 1754 Al 84 Al 94 8152 8160 C105 O i l Dl 72 Dl 72 LI 55 LI 57 M208 M212 Pl 53 Pl 59 Xl 35 Xl 37 1755 Al 80 Al 84 8158 8164 Cl 03 0 0 5 Dl 82 Dl 86 LI 55 LI 55 M214 M214 Pl 57 Pl 61 Xl 37 X141 1756 Al 84 Al 94 8140 8156 Cl 03 0 0 5 Dl 78 Dl 82 LI 55 LI 57 M214 M214 Pl 53 P161 X135 X135 1757 Al 84 Al 94 8140 8160 Cl 03 0 0 7 Dl 78 0182 LI 55 LI 57 M214 M214 Pl 61 Pl 61 Xl 35 Xl 37 1758 Al 84 Al 92 8150 8160 Cl 03 0 0 5 0172 Dl 82 LI 55 LI 61 M206 M208 Pl 53 Pl 55 Xl 35 X141 UNKl Al 80 Al 92 8150 8160 Cl 03 0 0 3 Dl 78 0180 LI 55 LI 57 M208 M212 Pl 51 Pl 59 Xl 35 Xl 37 UNK2 Al 80 Al 94 8160 8164 C105 0 0 7 Dl 76 Dl 78 LI 55 LI 55 M212 M214 Pl 53 Pl 61 Xl 35 Xl 37 UNK3 Al 84 Al 94 8158 8164 Cl 03 0 0 5 Dl 81 0182 LI 55 LI 55 M212 M214 Pl 57 Pl 61 Xl 33 Xl 37 UNK4 Al 92 Al 94 8160 8160 C103 0 0 5 Dl 76 Dl 86 LI 55 LI 55 M214 M214 Pl 51 Pl 61 Xl 35 Xl 37 HYPl Al 80 Al 94 8140 8160 C105 O i l Dl 72 Dl 78 LI 55 LI 57 M206 M208 Pl 49 Pl 51 Xl 33 Xl 41 HYP2 Al 92 AXXX 8160 BXXX 0 0 3 O i l Dl 72 Dl 82 LI 55 LI 57 M206 M208 Pl 53 Pl 55 Xl 31 Xl 35 HYP3 Al 86 Al 94 8140 8160 0 0 5 CXXX 0181 Dl 86 LI 55 LXXX M208 M214 Pl 53 P161 Xl 29 Xl 37 HYP4 Al 94 AXXX 8148 BXXX 0 0 3 0 0 5 Dl 77 Dl 84 LI 61 LXXX M206 M208 Pl 59 PXXX X135 X137 HYP 5 Al 94 AXXX 8158 8164 O i l CXXX Dl 78 Dl 82 LI 55 LI 57 M208 M212 Pl 39 Pl 57 Xl 31 X141 HYP6 Al 80 Al 84 8152 8160 0 0 5 O i l Dl 72 Dl 86 LI 55 LI 57 M208 M214 Pl 57 Pl 59 Xl 35 X141 HYP7 Al 80 Al 94 8140 8140 0 0 3 0 1 3 Dl 72 Dl 86 LI 57 LI 59 M206 M214 Pl 53 Pl 57 Xl 33 XXXX 186 Table 19. ANWR individual alleles. # Ioc A l Ioc A2 Ioc BI Ioc B2 Ioc C l Ioc C2 Ioc D l Ioc D2 Ioc LI Ioc L2 Ioc Ml Ioc M2 Ioc P l Ioc P2 Ioc Xl Ioc X2 1 1 9 7 A l 8 4 A l 9 6 BI 4 8 BI 6 0 C l 0 3 C 105 D l 7 6 D l 7 6 LI 5 7 LI 5 7 M 208 M 214 P l 5 5 P l 61 Xl 41 Xl 41 1 2 1 6 A l 9 4 A l 9 4 BI 5 4 BI 6 0 0 1 0 5 C l 0 5 D l 7 6 D l 81 LI 5 5 LI 5 7 M 206 M 210 P l 51 P l 5 5 Xl 3 7 X l 3 9 1 2 2 3 A l 9 4 A l 9 6 BI 4 0 8 1 6 0 C 1 0 5 C 1 0 5 D l 7 2 D l 7 4 LI 5 7 LI 5 7 M 214 M 214 P l 5 3 P l 5 9 X l 3 7 X141 1 2 2 6 A l 8 4 A 2 0 0 BI 6 0 BI 6 4 0 1 0 5 C 1 0 5 D l 81 D l 8 2 LI 5 5 LI 5 5 M 208 M 212 P l 5 5 P l 5 7 X l 3 7 X l 3 7 1 2 6 0 A l 8 4 A l 8 6 6 1 4 8 8 1 5 4 C l 0 3 C 1 0 5 D l 7 4 D l 81 L 1 5 5 LI 5 7 M 210 M 214 P l 5 9 P l 61 Xl 31 X l 3 7 1 2 6 2 A l 8 4 A l 8 4 BI 4 0 BI 4 0 C l 0 3 C l 0 9 D l 7 2 D l 8 2 LI 5 5 L 1 5 5 M 208 M 212 P l 51 P l 53 X l 3 5 X l 3 7 1 2 8 2 A l 8 4 A l 9 2 BI 4 8 8 1 6 0 C 1 0 5 C 1 0 5 D l 8 6 D l 8 6 LI 55 LI 5 5 M 210 M 210 P l 51 P l 5 9 X l 31 X l 31 1 5 0 5 A l 8 4 A l 9 4 8 1 5 8 8 1 6 0 0 1 0 5 0 1 0 9 D l 7 4 D l 7 6 LI 5 5 LI 5 7 M 208 M 214 P l 5 3 P l 61 X l 3 5 Xl 3 5 1 5 1 3 A l 8 4 A 2 0 0 BI 5 0 BI 5 2 C 1 0 5 C 1 0 5 D l 7 2 D l 81 LI 55 LI 5 7 M 208 M 214 P l 5 5 P l 61 Xl 3 3 Xl 41 1 5 1 5 A l 8 6 A 2 0 0 8 1 4 8 BI 6 0 C 1 0 3 0 1 0 5 D l 7 7 D l 81 LI 5 5 LI 5 5 M 214 M 214 P l 51 P l 5 9 X l 3 7 X l 41 1 5 1 6 A l 8 4 A l 9 4 BI 4 8 BI 5 0 0 1 0 1 C 1 0 5 D l 8 2 D l 8 6 LI 55 LI 5 7 M 210 M 214 P l 5 7 P l 5 9 X131 Xl 3 7 1 5 1 7 A l 9 2 A l 9 2 BI 5 4 BI 6 0 C 1 0 3 0 1 0 5 D l 81 D l 8 2 LI 5 5 LI 6 3 M 212 M 214 P l 51 P l 5 7 X l 3 7 X l 41 1 5 1 8 A l 9 6 A 2 0 0 BI 6 0 8 1 6 0 C 1 0 5 0 1 0 5 D l 81 D l 8 6 LI 5 5 LI 5 7 M 208 M 214 P l 61 P l 61 X l 3 7 X l 41 1 5 1 9 A l 9 2 A l 9 4 BI 6 0 BI 6 0 0 1 0 5 C l 1 3 D l 7 2 D l 7 4 LI 5 7 LI 5 7 M 208 M 214 P l 5 3 P l 5 5 X l 3 7 Xl 3 7 1 5 2 0 A l 8 6 A l 9 4 BI 5 2 BI 6 0 C l 0 5 0 1 0 9 D l 7 2 D l 7 4 LI 5 5 LI 5 7 M 210 M 212 P l 5 3 P l 5 5 Xl 31 X l 3 7 Table 20. AKR individual alleles. # Ioc A l Ioc A2 Ioc BI Ioc B2 Ioc C l Ioc C2 Ioc D l Ioc D2 Ioc LI Ioc L2 Ioc Ml Ioc M2 Ioc P l Ioc P2 Io cX l Ioc X2 GB6 A l 8 0 A l 8 4 BI 5 2 BI 6 0 C 1 0 5 C l 1 3 D l 7 7 D l 8 6 L 1 5 5 L 1 5 5 M 208 M 214 P l 51 P l 61 X l 31 Xl 41 GB7 A l 8 4 A l 8 4 BI 5 2 8 1 5 8 C l 0 5 C l 0 5 D l 7 8 D l 7 8 LI 5 7 L 1 5 7 M 206 M 206 P l 5 3 P l 5 5 X l 31 X l 3 7 GB8 A l 9 4 A l 9 4 BI 4 0 BI 4 0 C l OS C l 0 9 D l 7 2 D l 7 8 LI 5 5 L 155 M 208 M 210 P l 5 3 P l 5 7 X l 41 X141 GB9 A l 8 4 A l 9 4 BI 4 0 BI 6 0 C l 0 7 C 1 0 7 D l 81 D l 81 LI 5 5 LI 5 5 M 208 M 212 P l 5 3 P l 5 7 Xl 3 7 Xl 3 7 GBl 2 A l 8 0 A l 8 0 BI 4 0 BI 4 0 C IO S C l OS D l 7 8 D l 8 2 LI 55 L lS S M 208 M 208 P l 51 P l 5 7 Xl 3 3 X l 3 7 G B I3 A l 8 0 A l 9 4 BI 4 0 BI 5 8 C 1 0 5 C l 0 9 D l 7 4 D l 7 8 LI 55 LI 5 7 M 208 M 210 P l 51 P l 5 5 Xl 3 3 X141 1 3 3 2 A l 8 6 A 2 0 0 BI 5 2 BI 6 0 C IO l C l 1 3 D l 7 2 D l 81 LI 5 7 LI 5 7 M 210 M 218 P l 41 P l 5 3 Xl 3 7 X l 41 1 3 4 4 A l 9 4 A l 9 4 BI 6 0 8 1 6 0 C 1 0 5 cm D l 7 7 D l 8 2 LI 5 5 LI 5 7 M 210 M 212 P l 5 9 P l 61 Xl 41 X l 4 3 1 3 7 2 A l 8 0 A l 9 4 8 1 5 2 BI 6 0 C 1 0 5 C l 0 5 D l 7 7 D l 8 0 LI 5 9 LI 63 M 208 M 210 P l 51 P l 5 5 Xl 3 5 Xl 41 1 3 8 5 A l 9 0 A l 9 8 BI 4 8 BI 5 8 C 1 0 5 C l 1 3 D l 7 8 D l 7 8 LI 5 5 LI 5 9 M 206 M 210 P l 55 P l 5 9 X l 3 5 Xl 3 7 1 3 8 6 A l 9 0 A l 9 4 BI 4 8 BI 5 8 C l 0 5 cm D l 7 8 D l 81 L15S LI 5 9 M 206 M 208 P l 51 P l 5 9 X l 31 Xl 3 5 1 3 8 9 A l 8 0 A l 9 2 BI 4 8 8 1 6 0 C 1 0 3 C 1 0 5 D l 8 6 D l 8 6 LI 5 7 LI 5 9 M 210 M 214 P l 51 P l 5 9 X l 31 Xl 3 7 1 3 9 0 A l 8 4 A l 9 4 BI 5 0 BI 6 0 C IO I C l 0 3 D l 8 6 D l 8 6 LI 5 7 LI 5 9 M 208 M 210 P l 5 9 P l 5 9 Xl 3 7 Xl 41 1391 A l 8 4 A l 9 8 BI 4 8 8 1 5 8 C l 0 3 C l 13 D l 7 4 D l 7 8 LI 5 9 LI 61 M 206 M 206 P l 5 3 P l 5 7 X131 X 135 1 3 9 3 A l 8 0 A l 9 4 BI 4 0 8 1 5 4 C l OS cm D l 7 6 D l 8 0 L 1 5 9 LI 5 9 P l 51 P l 51 1 3 9 8 A l 8 4 A l 9 4 BI 5 4 BI 5 8 C 1 0 5 C lO S D l 7 6 D l 81 LI 5 9 L 1 5 9 M 208 M 214 P l 5 1 P l 5 3 X141 Xl 41 1 4 0 0 A l 8 4 A l 9 4 BI 4 0 8 1 5 0 C 1 0 5 CIO S D l 7 2 D l 7 8 LI 5 5 LI 5 7 M 206 M 208 P l 5 5 P l 5 7 Xl 35 Xl 3 7 Table 21. NCDE individual alleles. # Ioc A l loc A2 Ioc BI Ioc B2 Io cC I Ioc C2 Ioc D l Ioc D2 loc LI loc L2 loc Ml loc M2 loc P l loc P2 loc Xl loc X2 GBI A l 9 2 A l 9 4 BI 4 8 BI 6 0 C l 0 5 C l 0 7 D l 8 0 D l 8 4 LI 5 7 LI 5 9 M 210 M 214 P l 5 9 P l 61 Xl 41 Xl 41 GB2 A l 9 0 A l 9 8 BI 6 0 BI 6 2 C l 0 5 C l 11 D l 7 5 D l 7 9 LI 5 5 LI 5 7 M 208 M 208 P l 51 P l 5 7 X l 3 3 Xl 3 5 GB3 A l 9 2 A l 9 4 BI 5 2 8 1 5 8 C l 0 5 C 1 0 7 D l 8 0 D l 8 4 LI 5 5 LI 5 7 M 210 M 214 P l 51 P l 5 7 X l 41 Xl 41 GB5 A l 9 4 A l 9 4 BI 5 6 BI 6 2 C IO l C 1 0 5 D l 7 7 D l 7 7 LI 5 5 LI 5 7 M 210 M 214 P l 5 5 P l 5 9 Xl 41 Xl 41 GBIO A l 8 6 A l 9 0 BI 5 0 8 1 6 2 C 1 0 5 C 1 0 7 D l 7 5 D l 81 LI 5 5 LI 55 M 210 M 210 P l 4 9 P l 53 Xl 41 X l 41 G B II A l 8 4 A l 8 4 BI 4 8 BI 4 8 C 1 0 5 C l OS D l 81 D l 8 4 LI 5 7 LI 5 7 M 206 M 212 P l S I P l 5 5 X141 Xl 4 3 GBl 5 A l 9 4 A l 9 8 8 1 5 8 8 1 5 8 C l 0 5 C l 0 7 D l 8 4 D l 8 4 LI 5 7 LI 5 7 M 208 M 208 P l 5 9 P 1 5 9 X141 X141 G B I6 A l 9 4 A l 9 4 8 1 4 8 BI 5 8 C 1 0 5 C 1 0 7 D l 8 4 D l 8 4 LI 5 7 LI 5 9 M 210 M 214 P l 51 P l 5 9 Xl 41 X141 G B I7 A l 9 0 A l 9 2 8 1 5 8 BI 5 8 C l OS C 1 0 5 D l 7 5 D l 8 2 LI 5 7 LI 5 9 M 206 M 210 P l 5 3 P l 5 3 Xl 41 Xl 41 GBl 8 A l 9 4 A l 9 8 BI 5 6 BI 6 0 C 1 0 5 C 1 0 5 D l 7 7 D l 8 0 LI 5 5 LI 5 5 M 208 M 208 P l S l P l 5 9 X141 Xl 41 G B I9 A l 9 4 A l 9 4 BI 5 4 8 1 5 6 C 1 0 5 C IO S D l 7 9 D l 8 0 LI 5 5 LI 5 5 M 206 M 210 P l 51 P l 53 Xl 41 Xl 41 G B20 A l 8 6 A l 9 4 BI 5 4 8 1 6 2 C 1 0 3 C l 11 D l 83 D l 8 4 L 155 LI 55 M 206 M 206 P l 51 P l 5 3 X l 3 3 X141 GB21 A l 8 6 A l 9 4 BI 5 4 BI 5 8 C 1 0 5 C l 11 D l 8 0 D l 8 3 LI 5 5 L 1 5 7 M 206 M 210 P l 51 P l 5 3 Xl 41 Xl 41 G B22 A l 9 4 A l 9 4 8 1 5 6 BI 5 8 C 1 0 3 C 105 D l 8 0 D l 8 3 LI 5 5 LI 5 7 M 206 M 210 P l 51 P l 53 Xl 41 Xl 4 3 GB23 A l 9 2 A l 9 4 BI 5 6 BI 6 0 C l 0 5 C l 11 LI 5 5 LI 5 5 M 21 2 M 212 P l 5 9 P l 5 9 Xl 3 5 X l 41 G B24 A l 9 4 A l 9 4 BI 6 2 BI 6 6 C l 0 3 C l 0 5 D l 7 9 D l 7 9 LI 5 7 LI 5 7 M 210 M 212 P l 51 P l 5 3 Xl 41 X141 APPENDIX B Allele frequencies 188 Table 22. WBR allele frequencies. WBR bears A alleles frequencynumber Al 94 0.395 Al 84 0.240 Al 92 0.211 Al 80 0.089 Al 90 0.036 A200 0.016 Al 86 0.007 Al 88 0.007 TOTAL 1.000 B alleles frequencynumber 0.414 0.155 0.155 BI 48 0.092 0.053 BI 56 0.046 0.039 0.033 0.013 1.000TOTAL C alleles number frequency Cl 05 0.355 0.257 0.240 Cl 13 0.092 Cl 07 0.043 0.010 0.003C109 1.000TOTAL D alleles number frequency 0.243Dl 72 0.214Dl 78 0.145Dl 77 0.099 0.089Dl 86 0.082 0.049 0.036Dl 74 0.033Dl 76 0.010Dl 80 1.000TOTAL 0.400 0.300 0.200 0.100 0.000 0.400 0.300 0.200 0.100 0.000 0.500 0.400 0.300 0.200 0.100 0.000 0.250 0.200 0.150 0.100 0.050 0.000 189 Table 22 (continued). WBR allele frequencies. WBR bears L alleles number frequency 0.487 0.276 0.128 0.089 0.013 0.007 TOTAL 1.000 M alleles number frequency M208 0.454 M214 0.273 M212 0.115 M206 0.076 M210 0.053 M218 0.026 M222 0.003 TOTAL 1.000 P alleles number frequency 0.342 0.197 0.178 0.089 0.079 0.069 0.039 0.007 1.000TOTAL X alleles number frequency 0.395 0.400 0.300 0.200 0.100 0.000 0.214 0.211 0.099 0.053 0.030Xl 29 1.000TOTAL 0.000 0.200 0.400 0.600 0.500 0.400 0.300 0.200 0.100 0.000 0.400 0.300 0.200 0.100 0.000 190 Table 23. ANWR allele frequencies. ANWR bears A alleles frequencynumber Al 84 0.300 Al 94 0.233 A200 0.133 Al 92 0.133 Al 86 0.100 Al 96 0.100 TOTAL 1.000 B alleles number frequency 0.433 0.167 0.100 BI 54 0.100 0.067 0.067 BI 58 0.033 BI 64 0.033 TOTAL 1.000 frequencyC alleles number 0.800 0.600 0.400 0.200 0.000 0.667 0.167 0.100 0.033 0.033 1.000TOTAL frequencyD alleles number 0.233 0.167Dl 72 0.167Dl 74 0.133Dl 76 0.133 0.133Dl 86 0.033Dl 77 1.000TOTAL 0.300 0.200 0.100 0.000 0.100 0.000 0.200 0.300 0.200 0.000 0.600 0.400 191 Table 23 (continued). ANWR allele frequencies. ANWR bears L alleles number frequency 0.600 0.400 0.200 0.000 LI 55 16 0.533 LI 57 13 0.433 LI 63 I 0.033 TOTAL 30 1.000 I 2 3 M alleles number frequency M214 12 0.400 0.400,—020IlI0.000 I 2 3 M208 7 0.233 M210 6 0.200 M212 4 0.133 M206 I 0.033 E - TOTAL 30 1.000 4 5 P alleles number frequency Pl 55 6 0.200 Pl 61 6 0.200 0.200 i— — =IIlUi 1 2 3 4 5 6 Pl 53 5 0.167 Pl 51 5 0.167 Pl 59 5 0.167 Pl 57 3 0.100 TOTAL 30 1.000 X alleles number frequency n ennXl 37 13 0.433 Xl 41 7 0.233 0.400 0.200 n nnn ill -_ _X131 5 0.167Xl 35 3 0.100Xl 39 I 0.033 Xl 33 I 0.033 1 2 3 4 5 6TOTAL 30 1.000 192 Table 24. AKR allele frequencies. AKR bears A alleles number frequency Al 94 12 0.353 Al 84 8 0.235 0.400 | 0.300 ■ H l lo.ooo ! . ! . H 1 2 3 4 5 6 7 8 Al 80 7 0.206 Al 90 2 0.059 Al 98 2 0.059 Al 86 I 0.029 Al 92 I 0.029 A200 I 0.029 TOTAL 34 1 . 0 0 0 B alleles number frequency 0.300 I 0.200 I 1 1 1 » . . 1 2 3 4 5 6 7 BI 40 8 0.235 BI 60 8 0.235 BI 58 6 0.176 BI 48 4 0.118 BI 52 4 0.118 BI 54 2 0.059 BI 50 2 0.059 TOTAL 34 1 . 0 0 0 C alleles number frequency Cl 05 18 0.563 0.600 I 0.400 I 0.200 I o.ooo 1 2 3 4 5 6 Cl 13 4 0.125 C103 3 0.094 c m 3 0.094 CIO? 2 0.063 ClOl 2 0.063 TOTAL 32 1 . 0 0 0 D alleles number frequency 0178 1 0 0.294 0181 5 0.1471 U.30C 0.20C 0.1 OC O.OOC lllllllH 0186 5 0.147 0177 3 0.088 0172 3 0.088 0180 2 0.059 0174 2 0.059 I 9 3 vl C C 7 Q O 0182 2 0.059 0176 2 0.059 TOTAL 34 1.000 193 Table 24 (continued). AKR allele frequencies. AKR bears L alleles number frequency LI 55 13 0.382 0.400 = I I l I 1 2 3 4 5 LI 59 10 0.294 LI 57 9 0.265 1161 I 0.029 LI 63 I 0.029 TOTAL 34 1.000 M alleles number frequency M208 11 0.344 M210 8 0.250 0.400 0.300 0.200 0.100 0.000 1 1 1* . - M206 7 0.219 M214 3 0.094 M212 2 0.063 M21 8 I 0.031 TOTAL 32 1.000 1 2 3 4 5 6 P alleles number frequency Pl 51 9 0.265 Pl 59 6 0.176 0.300 I - I l I l j 0.000 1 2 3 4 . Pl 53 6 0.176 Pl 55 5 0.147 I Pl 57 5 0.147 L Pl 61 2 0.059 IMm Pl 41 I 0.029 6 7 TOTAL 34 1.000 X alleles number frequency 0.400 I 0.300 \tm 0.200 I 0.100 I I ■ ■ 0.000 I". 1 2 3 4 5 6 Xl 41 10 0.313 Xl 37 9 0.281 Xl 31 5 0.156 Xl 35 5 0.156 Xl 33 2 0.063 X143 I 0.031 TOTAL 32 1.000 194 Table 25. NCDE allele frequencies. NCDE bears A alleles Al 94 Al 92 Al 86 Al 90 Al 98 Al 84 TOTAL B alleles BI 58 BI 62 BI 56 BI 48 BI 60 BI 54 BI 66 BI 50 BI 52 number frequency 0.531 0.125 0.094 0.094 0.094 0.063 1.000 number frequency 0.250 0.156 0.156 0.125 0.125 0.094 0.031 0.031 0.031 0.250 0.200 0.150 0.100 0.050 0.000 1 2 3 4 5 6 7 8 9 TOTAL 1.000 C alleles C105 Cl 07 Clll C103 ClOl TOTAL number frequency 0.594 0.156 0.125 0.094 0.031 1.000 1 2 3 4 5 D alleles number frequency Dl 84 Dl 80 Dl 79 Dl 77 Dl 75 Dl 83 Dl 81 Dl 82 TOTAL 0.267 0.200 0.133 0.100 0.100 0.100 0.067 0.033 1.000 0.300 0.200 0.100 0.000 l l l l l l l - 12 3 4 5 6 7 8 195 Table 25 (continued). NCDE allele frequencies. NCDE bears L alleles number frequency 0.6C0 0 0 r\ B I . LI 55 15 0.469 0.4C LI 57 14 0.438 0.2CLI 59 3 0.094 0 OCTOTAL 32 1.000 I 2 3 M alleles number frequency Or ! ■ ■ ■ ■ 1 2 3 4 5 M210 11 0.344 0.40 M206 7 0.219 0.30 M208 6 0.188 0.20 M212 4 0.125 0.10 M214 4 0.125 0.00 TOTAL 32 1.000 P alleles number frequency PT 51 10 0.313 Pl 59 8 0.250 0.300 0.200 0.100 i n . Pl 53 8 0.250 Pl 55 2 0.063 Pl 57 2 0.063 P149 I 0.031 U.UUU ---- --------------- —1 2 3 4 5 6 7Pl 61 I 0.031 TOTAL 32 1.000 X alleles number frequency Xl 41 26 0.813 I .UUU 0.500 0.000 Xl 33 2 0.063 Xl 35 2 0.063 X143 2 0.063 TOTAL 32 1.000 1 2 3 4 196 Table 26. Allele frequencies for all populations. WBR f ANWR AKR NCDE ALL alleles number freq number freq number freq number freq number freq Al 94 120 0.395 7 0.233 12 0.353 17 0.531 156 0.39 Al 84 73 0.24 9 0.3 8 0.235 2 0.063 92 0.23 Al 92 64 0.211 4 0.133 I 0.029 4 0.125 73 0.183 Al 80 27 0.089 0 7 0.206 0 34 0.085 Al 90 11 0.036 0 2 0.059 3 0.094 16 0.04 A200 5 0.016 4 0.133 I 0.029 0 10 0.025 Al 86 2 0.007 3 0.1 I 0.029 3 0.094 9 0.023 Al 88 2 0.007 0 0 0 2 0.005 Al 96 0 3 0.1 0 0 3 0.008 Al 98 0 0 2 0.059 3 0.094 5 0.013 8160 126 0.414 13 0.433 8 0.235 4 0.125 151 0.378 8158 47 0.155 I 0.033 6 0.176 8 0.25 62 0.155 8140 47 0.155 3 0.1 8 0.235 0 58 0.145 8148 28 0.092 5 0.167 4 0.118 4 0.125 41 0.103 8164 16 0.053 I 0.033 0 0 17 0.043 8156 14 0.046 0 0 5 0.156 19 0.048 8150 12 0.039 2 0.067 2 0.059 I 0.031 17 0.043 8152 10 0.033 2 0.067 4 0.118 I 0.031 17 0.043 8154 4 0.013 3 0.1 2 0.059 3 0.094 ■ 12 0.03 8162 0 0 0 5 0.156 5 0.013 8166 0 0 0 I 0.031 I 0.003 Cl 05 108 0.355 20 0.667 18 0.563 19 0.594 165 0.415 Cl 03 78 0.257 5 0.167 3 0.094 3 0.094 89 0.224 Cl 11 73 0.24 0 3 0.094 4 0.125 80 0.201 CU 3 28 0.092 I 0.033 4 0.125 0 33 0.083 Cl 07 13 0.043 0 2 0.063 5 0.156 20 0.05 CIOl 3 0.01 I 0.033 2 0.063 I 0.031 7 0.018 Cl 09 I 0.003 3 0.1 0 0 4 0.01 Dl 72 74 0.243 5 0.167 3 0.088 0 82 0.206 Dl 78 65 0.214 0 10 0.294 0 75 0.188 Dl 77 44 0.145 I 0.033 3 0.088 3 0.1 51 0.128 Dl 81 30 0.099 7 0.233 5 0.147 2 0.067 44 0.111 Dl 86 27 0.089 4 0.133 5 0.147 0 36 0.09 Dl 82 25 0.082 4 0.133 2 0.059 I 0.033 32 0.08 Dl 84 15 0.049 0 0 8 0.267 23 0.058 D176 11 0.036 4 0.133 2 0.059 0 17 0.043 Dl 74 10 0.033 5 0.167 2 0.059 0 17 0.043 Dl 80 3 0.01 0 2 0.059 6 0.2 11 0.028 Dl 83 0 . 0 0 3 0.1 3 0.008 Dl 79 0 0 0 4 0.133 4 0.01 Dl 75 0 0 0 3 0.1 3 0.008 197 Table 26 (continued). Allele frequencies for all populations. WBR ANWR AKR NCDE ALL alleles number freq number freq number freq number freq number freq LI 55 148 0.487 16 0.533 13 0.382 15 0.469 192 0.48 LI 57 84 0.276 13 0.433 9 0.265 14 0.438 120 0.3 LI 61 39 0.128 0 I 0.029 0 40 0.1 L159 27 0.089 0 10 0.294 3 0.094 40 0.1 L171 4 0.013 0 0 0 4 0.01 LI 63 2 0.007 I 0.033 I 0.029 0 4 0.01 M208 138 0.454 7 0.233 11 0.344 6 0.188 162 0.407 M214 83 0.273 12 0.4 3 0.094 4 0.125 102 0.256 M212 35 0.115 4 0.133 2 0.063 4 0.125 45 0.113 M206 23 0.076 I 0.033 7 0.219 7 0.219 38 0.095 M210 16 0.053 6 0.2 8 0.25 11 0.344 41 0.103 M218 8 0.026 0 I 0.031 0 9 0.023 M222 I 0.003 0 0 0 I 0.003 Pl 53 104 0.342 5 0.167 6 0.176 8 0.25 123 0.308 P157 60 0.197 3 0.1 5 0.147 2 0.063 70 0.175 Pl 61 54 0.178 6 0.2 2 0.059 I 0.031 63 0.158 Pl 51 27 0.089 5 0.167 9 0.265 10 0.313 51 0.128 Pl 59 24 0.079 5 0.167 6 0.176 8 0.25 43 0.108 Pl 55 21 0.069 6 0.2 5 0.147 2 0.063 34 0.085 P149 12 0.039 0 0 I 0.031 13 0.033 P139 2 0.007 0 0 0 2 0.005 P141 0 0 I 0.029 0 I 0.003 Xl 37 120 0.395 13 0.433 9 0.281 0 142 0.357 Xl 35 65 0.214 3 0.1 5 0.156 2 0.063 75 0.188 X141 64 0.211 7 0.233 10 0.313 26 0.813 107 0.269 X133 30 0.099 I 0.033 2 0.063 2 0.063 35 0.088 X131 16 0.053 5 0.167 5 0.156 0 26 0.065 X129 9 0.03 0 0 0 9 0.023 Xl 39 0 I 0.033 0 0 , I 0.003 Xl 43 0 0 I 0.031 2 0.063 3 0.008 71 alleles TOTAL 2432 240 266 254 3192 198 APPENDIX C Genotype frequencies and analyses 199 Table 27. Locus A genotype frequencies. Alleles are shown on the axes of the table; the number of individuals observed with each 2-allele combination is shown in the body of the table. A l94 A l84 Al 92 Al 80 Al 90 A200 Al 86 Al 88 TOTAL Al 94 25 32 23 14 I I 0 I 97 Al 84 5 20 4 3 2 I 0 35 Al 92 4 5 5 2 0 I 17 A l 80 0 2 0 I 0 3 A l 90 0 0 0 0 0 A200 0 0 0 0 A l 86 0 0 0 A l 88 0 0 Heterozygotes 118 Homozygotes 34 152 200 Table 28. Locus B genotype frequencies. Alleles are shown on the axes of the table; the number of individuals observed with each 2-allele combination is shown in the body of the table. Heterozygotes 121 Homozygotes 31 105 28 13 4 0 I 0 I 0 152 -I 201 Table 29. Locus C genotype frequencies. Alleles are shown on the axes of the table; the number of individuals observed with each 2-allele combination is shown in the body of the table. C105 C103 C lll C113 C107 ClOl C109 TOTAL C105 21 29 26 6 5 0 0 87 C103 5 19 12 5 3 0 44 cm 8 8 2 0 I 19 C113 I I 0 0 2 C107 0 0 0 0 ClOl 0 0 0 C109 0 0 Heterozygotes 117 Homozygotes 35 152 202 Table 30. Locus D genotype frequencies. Alleles are shown on the axes of the table; the number of individuals observed with each 2-allele combination is shown in the body of the table. D172 D178 D177 Dl 81 Dl 86 Dl 82 Dl 84 D174 D176 Dl 80 TOTAL D172 9 19 10 8 5 6 4 I 2 I 65 D178 2 15 6 5 5 4 0 5 2 44 D177 2 I 7 2 2 3 0 0 17 Dl 81 3 2 4 I 2 0 0 12 Dl 86 I 3 0 I 2 0 . 7 Dl 82 I I 2 0 0 4 Dl 84 I I 0 0 2 D174 0 I 0 I D176 0 0 0 Dl 80 0 0 Heterozygotes 133 Homozygotes 19 152 203 Table 31. Locus L genotype frequencies. Alleles are shown on the axes of the table; the number of individuals observed with each 2-allele combination is shown in the body of the table. Note the excess of L159/L161 heterozygotes. L155 L157 L161 L159 L171 L163 TOTAL L155 39 45 13 7 3 2 109 L157 11 11 6 0 0 28 L161 2 11 0 0 13 L159 I I 0 2 L171 0 0 0 LI 63 0 0 Heterozygotes 9C.) Homozygotes 53 152 204 Table 32. Locus M genotype frequencies. Alleles are shown on the axes of the table; the number of individuals observed with each 2-allele combination is shown in the body of the table. M208 M214 M212 M206 M210 M218 M222 TOTAL M208 28 43 17 11 7 4 0 109 M214 9 11 4 6 I 0 31 M212 0 3 2 I I . 7 M206 2 I 0 0 3 M210 0 0 0 0 M218 I 0 I M222 0 0 Heterozygotes 112 Homozygotes 40 152 205 Table 33. Locus P genotype frequencies. Alleles are shown on the axes of the table; the number of individuals observed with each 2-allele combination is shown in the body of the table. 86 32 Heterozygotes 121 Homozygotes 31 8 2 2 0 0 152 206 Table 34. Locus X genotype frequencies. Alleles are shown on the axes of the table; the number of individuals observed with each 2-allele combination is shown in the body of the table. 100 30 19 2 0 0 152 207 Table 35. WBR Hardy-Weinberg equilibrium. LOCUS A 8 alleles homozygotes Allele Allele freq. smpl variance p2*153 exp horn obs (obs-exp)2/exp Al 94 0.395 0.0008 23.872 25 0.053 Al 84 0.240 0.0006 8.813 5 1.650 Al 92 0.211 0.0005 6.812 4 1.161 Al 80 0.089 0.0003 1.212 0 1.212 others 0.065 0.0002 0.646 0 0.646 heterozygotes het exp het obs Observed: 2(p184p194) 29.009 32 0.308 Heterozygosity 2(p1 92p194) 25.804 23 0.246 0.7763 2(p192p184) 15.496 20 1.309 Homozygosity 2(p180p194) 10.757 14 0.977 0.2237 2(p180p184) 6.536 4 0.984 Homozygotes 2(p1 80p192) 5.746 5 0.097 34 other 2pq 18.597 20 0.106 chi-square 8.750 prob (IOdf) 0.540 LOCUS B 9 alleles homozygotes Allele Allele freq. smpl variance p2*153 exp horn obs (obs-exp)2/exp BI 60 0.414 0.0008 26.224 20 1.477 BI 40 0.155 0.0004 3.676 4 0.029 BI 58 0.155 0.0004 3.676 3 0.124 BI 48 0.092 0.0003 1.295 2 0.384 others 0.184 0.0005 5.180 I 3.373 heterozygotes het exp het obs Observer : Heterozygosity 2(p140p160) 19.636 15 1.095 0.7961 2(p158p160) 19.636 20 - 0.007 Homozygosity 2(p1 58p140) 7.352 11 1.811 0.2039 2(p148p160) 11.655 17 2.451 Homozygotes 2(p148p140) 4.364 3 0.426 31 2(p148p158) 4.364 2 1.280 other 2pq 45.944 54 1.413 chi-square 8.299 prob (10 df) 0.600 208 Table 35 (continued). WBR Hardy-Weinberg equilibrium. LOCUS C 7 alleles . homozygotes Allele Allele freq. smpl variance p2*153 exp horn obs (obs-exp)2/exp Cl OS 0.355 0.0007 19.282 21 0.T53 Cl 03 0.257 0.0006 10.105 5 2.579 cm 0.240 0.0006 8.813 8 0.075 Cl 13 0.092 0.0003 1.295 I 0.067 others 0.056 0.0002 0.480 0 0.480 heterozygotes het exp het obs Observer Heterozygosity 2(pl 05p103) 27.918 29 0.042 0.7697 2(p111p105) 26.071 26 0.000 Homozygosity 2(p111 pi 03) I 8.874 19 0.001 0.2303 2(pl 13p105) 9.994 6 1.596 Homozygotes 2(p113p103) 7.235 12 3.138 35 2(p113p111) 6.756 8 0.229 other 2pq 16.176 17 0.042 chi-square 8.403 prob (IOdQ 0.600 209 Table 35 (continued). WBR Hardy-Weinberg equilibrium. LOCUS D 10 alleles homozygotes Allele Allele freq. smpl variance p2*153 exp horn obs (obs-exp)2/exp Dl 72 0.243 0.0006 9.034 9 0.000 Dl 78 0.214 0.0005 7.007 2 3.578 Dl 77 0.145 0.0004 3.217 2 0.460 Dl 81 0.099 0.0003 1.500 3 1.501 Dl 86 0.089 0.0003 1.212 I 0.037 Dl 82 0.082 0.0002 1.029 I 0.001 Dl 84 0.049 0.0002 0.367 I 1.090 others 0.079 0.0002 0.955 0 0.955 heterozygotes het exp het obs ObserverI Heterozygosity 2(p172p178) 15.913 19 0.599 0.875 2(p177p172) 10.782 10 0.057 Homozygosity 2(p177p178) 9.495 15 3.191 0.125 2(p1 81 pi 72) 7.361 8 0.055 Homozygotes 2(p181 pi 78) 6.483 6 0.036 19 2(p181 pi 77) 4.393 I 2.620 2(p186p172) 6.618 5 0.396 2(p186p178) 5.828 4 0.573 2(p186p177) 3.949 7 2.357 2(p186p181) 2.696 2 0.180 2(p1 82p172) 6.097 6 0.002 2(p182p178) 5.370 5 0.025 2(p182p177) 3.638 2 0.738 2(p182p181) 2.484 4 0.925 2(p182p186) 2.233 3 0.263 2(p184p172) 3.644 4 0.035 2(p1 84p178) 3.209 4 0.195 2(p1 84p177) 2.174 2 0.014 2(p1 84p181) 1.484 I 0.158 2(p184p186) 1.334 0 1.334 2(p184p182) 1.230 I 0.043 other 2pq 22.264 23 0.024 chi-square 21.443 prob (28 df) 0.800 210 Table 35 (continued). WBR Hardy-Weinberg equilibrium. LOCUS L 6 alleles homozygotes Allele Allele freq. smpl variance p2*153 exp horn obs (obs-exp)2/exp LI 55 0.487 0.0008 36.287 39 0.203 LI 57 0.276 0.0007 11.655 11 0.037 LI 61 0.128 0.0004 2.507 2 0.102 LI 59 0.089 0.0003 1.212 I 0.037 others 0.020 0.0001 0.061 0 0.061 Observed: heterozygotes het exp het obs Heterozygosity 2(p1 55p157) 41.130 45 0.364 0.6513 2(p161 pi 55) 19.075 13 1.935 Homozygosity 2(p161 pi 57) 10.810 11 0.003 0.3487 2(p1 59p155) 13.263 7 2.957 Homozygotes 2(p1 59p157) 7.517 6 0.306 53 2(p159p161) 3.486 11 16.197 other 2pq 5.998 6 0.000 chi-square 22.203 prob (IO df) 0.030 LOCUS M7 alleles homozygotes Allele Allele freq. smpl variance p2*153 exp horn obs (obs-exp)2/exp M208 0.454 0.0008 31.536 28 0.396 M214 0.273 0.0006 11.403 9 0.506 M212 0.115 0.0003 2.023 0 2.023 M206 0.076 0.0002 0.884 2 1.410 others 0.082 0.0002 1.029 I 0.001 heterozygotes het exp het obs Observer Heterozygosity 2(p208p214) 37.926 43 0.679 0.7368 2(p212p208) 15.976 17 0.066 Homozygosity 2(p212p214) 9.607 11 0.202 0.2632 2(p206p208) 10.558 11 0.018 Homozygotes 2(p206p214) 6.349 4 0.869 40 2(p206p212) 2.674 3 0.040 other 2pq 23.034 23 0.000 chi-square 6.211 prob (IOdQ 0.730 211 Table 35 (continued). WBR Hardy-Weinberg equilibrium. LOCUS P 8 alleles homozygotes Allele Allele freq. smpl variance p2*153 exp horn obs (obs-exp)2/exp Pl 53 0.342 0.0007 17.895 18 0.001 Pl 57 0.197 0.0005 5.938 6 0.001 Pl 61 0.178 0.0005 4.848 2 1.673 Pl 51 0.089 0.0003 1.212 2 0.512 others 0.194 0.0005 5.758 3 1.321 Observer heterozygotes het exp het obs Heterozygosity 2(p153p157) 20.616 22 0.093 0.7961 2(p1 61 pi 53) 18.628 16 0.371 Homozygosity 2(p161 pi 57) 10.730 14 0.996 0.2039 2(p151 pi 53) 9.314 5 1.998 Homozygotes 2(p1 51 pi 57) 5.365 5 0.025 31 2(p151 pi 61) 4.848 7 0.956 other 2pq 47.847 55 1.069 chi-square 9.016 prob (10 df) 0.550 LOCUS X 6 alleles homozygotes Allele Allele freq. smpl variance p2*153 exp horn obs (obs-exp)2/exp Xl 37 0.395 0.0008 23.872 20 0.628 Xl 35 0.214 0.0005 7.007 7 0.000 Xl 41 0.211 0.0005 6.812 6 0.097 Xl 33 0.099 0.0003 1.500 0 1.500 others 0.081 0.0002 1.004 0 1.004 heterozygotes het exp het obs Observer Heterozygosity 2(p137p135) 25.866 27 0.050 0.7829 2(p14-1 pi 37) 25.504 30 0.793 Homozygosity 2(p141 pi 35) 13.817 9 1.679 0.2171 2(p133p137) 11.966 13 0.089 Homozygotes 2(p133p135) 6.483 8 0.355 33 2(p133p141) 6.392 6 0.024 other 2pq 22.778 25 0.217 chi-square 5.750 prob (10 df) I 0.880 212 Table 36. WBR heterozygosity (observed versus expected). Chi-square results are shown at the end of the table. WBR bears A alleles number frequency P2 exp H obs H Al 94 120 0.395 0.156 0.733 0.776 Al 84 73 0.240 0.058 (oe)2/exp Al 92 64 0.211 0.044 0.00256573 Al 80 27 0.089 0.008 Al 90 11 0.036 0.001 ' AZO O 5 0.016 0:000 Al 86 2 0.007 0.000 Al 88 2 0.007 0.000 TOTAL 304 1.000 0.267 B alleles number frequency P2 exp H obs H BI 60 126 0.414 0.172 • 0.764 0.796 BI 40 47 0.155 0.024 (o-e)2/exp BI 58 47 0.155 0.024 0.00132171 B148 28 0.092 0.008 BI 64 16 0.053 0.003 BI 56 14 0.046 0.002 B150 12 0.039 0.002 BI 52 10 0.033 0.001 BI 54 4 0.013 0.000 TOTAL 304 1.000 0.236 C alleles number frequency p2 exp H obs H Cl 05 108 0.355 0.126 0.740 0.770 Cl 03 78 0.257 0.066 (o-e)2/exp c m 73 0.240 0.058 0.00122684 Cl 13 28 0.092 0.008 Cl 07 13 0.043 0.002 ClOl 3 0.010 0.000 Cl 09 I 0.003 0.000 TOTAL 304 1.000 0.260 D alleles number frequency • p2 exp H obs H Dl 72 74 0.243 0.059 0.845 0.875 D178 65 0.214 0.046 (o-e)2/exp Dl 77 44 0.145 0.021 0.00108199 Dl 81 30 0.099 0.010 Dl 86 27 0.089 0.008 Dl 82 25 0.082 0.007 Dl 84 15 0.049 0.002 Dl 74 11 0.036 0.001 Dl 76 10 0.033 0.001 Dl 80 3 0.010 0.000 TOTAL 304 1.000 0.155 213 Table 36 (continued). WBR heterozygosity (observed versus expected). Chi-square results are shown at the end of the table. WBR bears L alleles number frequency , p2 exp H obs H LI 55 148 0.487 0.237 0.662 0.651 LI 57 84 0.276 0.076 (o-e)2/ex'p LI 61 39 0.128 0.016 0.00018514 LI 59 27 0.089 0.008 L171 4 0.013 0.000 LI 63 2 0.007 0.000 TOTAL 304 1.000 0.338 M alleles number frequency P2 exp H obs H • M208 138 0.454 0.206 0.697 0.737 M214 83 0.273 0.075 (o-e)2/exp M212 35 0.115 0.013 0.00230316 M206 23 0.076 0.006 M210 16 0.053 0.003 M218 8 0.026 0.001 M222 I. 0.003 0.000 TOTAL 304 1.000 0.303 P alleles number frequency p2 exp H obs H PT 53 104 0.342 0.117 0.792 0.796 P157 60 0.197 0.039 (o-e)2/exp PT 61 54 0.178 0.032 2.0584E-05 PT 51 27 0.089 0.008 Pl 59 24 0.079 ■ 0.006 Pl 55 21 0.069 0.005 Pl 49 12 0.039 0.002 Pl 39 2 0.007 0.000 TOTAL 304 1.000 0.208 X alleles number frequency p2 exp H obs H X137 120 0.395 0.156 0.741 0.783 X135 65 0.214 0.046 (o-e)2/exp X141 64 0.211 0.044 0.00240873 Xl 33 30 0.099 0.010 Xl 31 16 0.053 0.003 CHI-SQUARE= 0.011 X129 9 0.030 0.001 PROB (6 df)= 1.000 TOTAL 304 1.000 0.259 214 Table 37. ANWR heterozygosity (observed versus expected). Chi-square results are shown at the end of the table. ANWR bears A alleles number frequency p2 exp H obs H Al 84 9 0 .300 0 .090 0 .800 0 .800 Al 94 7 0.233 0 .054 (o-e)2/exp A200 4 0.133 0.018 0 .000 Al 92 4 0.133 0.018 Al 86 3 0 .100 0.010 Al 96 3 0.100 0.010 TOTAL 30 1.000 0.200 B alleles number frequency p2 exp H obs H BI 60 13 0.433 0.188 0 .753 0 .800 BI 48 5 0.167 0.028 (o-e)2/exp BI 40 3 0 .100 0.010 0.003 BI 54 3 0.100 0.010 BI 50 2 0.067 0 .004 BI 52 2 0.067 0 .004 BI 58 I 0 .033 0.001 BI 64 I - 0 .033 0.001 TOTAL 30 1.000 0.247 C alleles number frequency p2 exp H obs H Cl 05 20 0.667 0 .444 0.516 0 .600 Cl 03 5 0.167 0.028 (o-e)2/exp Cl 09 3 0 .100 0.010 0 .014 CIOI I 0 .033 0.001 Cl 13 I 0 .033 0.001 TOTAL 30 1.000 0 .484 D alleles number frequency p2 exp H obs H Dl 81 7 0.233 0 .054 0.836 0.867 D172 5 0.167 0.028 (o-e)2/exp Dl 74 5 0.167 0.028 0.001 Dl 76 4 0.133 0.018 Dl 82 4 0.133 0.018 Dl 86 4 0.133 0.018 Dl 77 I 0.033 0.001 TOTAL 30 1.000 0 .164 215 Table 37 (continued). ANWR heterozygosity (observed versus expected). Chi-square results are shown at the end of the table. ANWR bears L alleles number frequency p2 exp H obs H LI 55 16 0.533 0 .284 0 .527 0 .533 LI 57 13 0.433 0.188 (o-e)2/exp LI 63 I 0 .033 0.001 0 .000 TOTAL 30 1.000 0.473 M alleles number frequency p2 exp H obs H M214 12 0.400 0.160 0 .727 0 .800 M208 7 0.233 0 .054 (o-e)2/exp M210 6 0 .200 0 .040 0 .007 M212 4 0.133 0 .018 M206 I 0 .033 o.ooi TOTAL 30 1.000 0.273 P alleles number frequency p2 exp H obs H PT 55 6 0.200 0 .040 0.827 0.933 Pl 61 6 0 .200 0.040 (o-e)2/exp Pl 53 5 0.167 0.028 ■0.014 Pl 51 5 0.167 0.028 Pl 59 5 0.167 0.028 Pl 57 3 0 .100 0.010 TOTAL 30 1.000 0.173 X alleles number frequency p2 exp H obs H Xl 37 13 0.433 0 .188 0 .718 0.667 Xl 41 7 0.233 0 .054 (o-e)2/exp Xl 31 5 0 .167 0.028 0 .004 Xl 35 3 0.100 0.010 Xl 39 I 0 .033 0.001 Xl 33 I 0.033 0.001 CHI-SQUARE= 0.043 TOTAL 30 1.000 0.282 PROB (6 df)= 1.000 216 Table 38. AKR heterozygosity (observed versus expected). Chi-square results are shown at the end of the table. AKR bears A alleles number frequency P2 exp H obs H Al 94 12 0.353 0.125 0.768 0.765 Al 84 8 0.235 0.055 (o-e)2/exp Al 80 7 0.206 0.042 0.000 Al 90 2 0.059 0.003 Al 98 2 0.059 0.003 Al 86 I 0.029 0.001 Al 92 I 0.029 0.001 A200 I 0.029 0.001 TOTAL 34 1.000 0.232 B alleles number frequency P2 exp H obs H BI 40 8 0.235 0.055 0.824 0.824 BI 60 8 0.235 0.055 (o-e)2/exp BI 58 6 0.176 0.031 0.000 BI 48 4 0.118 0.014 BI 52 4 0.118 0.014 BI 54 2 0.059 0.003 BI 50 2 0.059 0.003 TOTAL 34 1.000 0.176 C alleles number frequency p2 exp H obs H Cl 05 18 0.563 0.316 0.643 0.647 Cl 13 4 0.125 0.016 (o-e)2/exp Cl 03 3 0.094 0.009 0.000 cm 3 0.094 0.009 Cl 07 2 0.063 0.004 CIOI 2 0.063 0.004 TOTAL 32 1.000 0.357 D alleles number frequency P2 exp H obs H Dl 78 10 0.294 0.087 0.841 0.706 Dl 81 5 0.147 0.022 (o-e)2/exp Dl 86 5 0.147 0.022 0.022 Dl 77 3 0.088 0.008 Dl 72 3 0.088 0.008 Dl 80 2 0.059 0.003 Dl 74 2 0.059 0.003 Dl 82 2 0.059 0.003 Dl 76 2 0.059 0.003 TOTAL 34 1.000 0.159 217 Table 38 (continued). AKR heterozygosity (observed versus expected). Chi-square results are shown at the end of the table. AKR bears L alleles number frequency p2 exp H obs H LI 55 13 0.382 0.146 0.696 0.529 LI 59 10 0.294 0.087 (o-e)2/exp LI 57 9 0.265 0.070 0.040 LI 61 I 0.029 0.001 LI 63 I 0.029 0.001 TOTAL 34 1.000 0.304 M alleles number frequency P2 exp H obs H M208 11 0.344 0.118 0.758 0.765 M210 8 0.250 0.063 (o-e)2/exp M206 7 0.219 0.048 0.000 M214 3 0.094 0.009 M212 2 0.063 0.004 M218 I 0.031 0.001 TOTAL 32 1.000 0.242 P alleles number frequency P2 exp H obs H Pl 51 9 0.265 0.070 0.820 0.882 Pl 59 6 0.176 0.031 (o-e)2/exp Pl 53 6 0.176 0.031 0.005 Pl 55 5 0.147 0.022 Pl 57 5 0.147 0.022 Pl 61 2 0.059 0.003 Pl 41 I 0.029 0.001 TOTAL 34 1.000 0.180 X alleles number frequency p2 exp H obs H Xl 41 10 0.313 0.098 0.770 0.813 Xl 37 9 0.281 0.079 (o-e)2/exp Xl 31 5 0.156 0.024 0.002 Xl 35 5 0.156 0.024 Xl 33 2 0.063 0.004 Xl 43 I 0.031 0.001 CHI-SQUARE= 0.069 TOTAL 32 1.000 0.230 PROB (6 df)= 0.990 218 Table 39. NCDE heterozygosity (observed versus expected). Chi-square results are shown at the end of the table. NCDE bears A alleles number frequency P2 exp H obs H Al 94 17 0.531 0.282 0.672 0.625 Al 92 4 0.125 0.016 (o-e)2/exp Al 86 3 0.094 0.009 0.00327035 Al 90 3 0.094 0.009 Al 98 3 0.094 0.009 Al 84 2 0.063 0.004 TOTAL 32 1.000 0.328 B alleles number frequency P2 exp H obs H BI 58 8 0.250 0.063 0.846 0.813 BI 62 5 0.156 0.024 (o-e)2/exp BI 56 5 0.156 0.024 0.00130359 BI 48 4 0.125 0.016 BI 60 4 0.125 0.016 BI 54 3 0.094 0.009 BI 66 I 0.031 0.001 BI 50 I 0.031 0.001 BI 52 I 0.031 0.001 TOTAL 32 1.000 0.154 C alleles number frequency P2 exp H obs H Cl 05 19 0.594 0.353 0.598 0.750 Cl 07 5 0.156 0.024 (o-e)2/exp cm 4 0.125 0.016 0.03883272 Cl 03 3 0.094 6.009 CIOI I 0.031 0.001 TOTAL 32 1.000 0.402 D alleles number frequency P2 exp H obs H Dl 84 8 0.267 0.071 0.836 0.750 Dl 80 6 0.200 0.040 (o-e)2/exp Dl 79 4 0.133 0.018 0.00876034 Dl 77 3 0.100 0.010 Dl 75 3 0.100 0.010 Dl 83 3 0.100 0.010 Dl 81 2 0.067 0.004 Dl 82 I 0.033 0.001 TOTAL 30 1.000 0.164 219 Table 39 (continued). NCDE heterozygosity (observed versus expected). Chi-square results are shown at the end of the table. NCDE bears L alleles number frequency P2 exp H obs H LI 55 15 0.469 0.220 0.580 0.500 LI 57 14 0.438 0.191 (o-e)2/exp LI 59 3 0.094 0.009 0.01105456 TOTAL 32 1.000 0.420 M alleles number frequency P2 exp H obs H M210 11 0.344 0.118 0.768 0.688 M206 7 0.219 0.048 (o-e)2/exp M208 6 0.188 0.035 0.00835421 M212 4 0.125 0.016 M214 4 0.125 0.016 TOTAL 32 1.000 0.232 P alleles number frequency p2 exp H obs H Pl 51 10 0.313 0.098 0.768 0.813 Pl 59 8 0.250 0.063 (o-e)2/exp Pl 53 8 0.250 0.063 0.00262902 Pl 55 2 0.063 0.004 Pl 57 2 0.063 0.004 Pl 49 I 0.031 0.001 Pl 61 I 0.031 0.001 TOTAL 32 1.000 0.232 X alleles number frequency p2 exp H obs H X141 26 0.813 0.660 0.328 0.313 Xl 33 2 0.063 0.004 (o-e)2/exp Xl 35 2 0.063 0.004 0.00074405 Xl 43 2 0.063 0.004 CHI-SQUARE= 0.075 TOTAL 32 1.000 0.672 PROB (6 df)= 0.950 2 2 0 APPENDIX D Additional analyses 221 Table 40. Fst over all populations (expected). Mean Fst is at end of table. All populations allele number freq (p) P2 Al 94 156 0.390 0.152 exp H ALL Al 84 92 0.230 0.053 0.751 Al 92 73 0.183 0.033 Al 80 34 0.085 0.007 WBR ANWR AKR NCDE Al 90 16 0.040 0.002 exp H 0.733 0.8 0.7681 0.672 A200 10 0.025 0.001 Fst 0.024 -0.065 -0.022 0.106 Al 86 9 0.023 0.001 Al 88 2 0.005 0.000 Al 96 3 0.008 0.000 Al 98 5 0.013 0.000 BI 60 151 0.378 0.143 exp H ALL BI 58 62 0.155 0.024 0.792 B140 58 0.145 0.021 BI 48 41 0.103 0.011 WBR ANWR AKR NCDE BI 64 17 0.043 0.002 exp H 0.764 0.753 0.824 0.846 BI 56 19 0.048 0.002 Fst 0.036 0.05 -0.04 -0.068 BI 50 17 0.043 0.002 BI 52 17 0.043 0.002 BI 54 12 0.030 0.001 BI 62 5 0.013 0.000 BI 66 I 0.003 0.000 Cl 05 165 0.415 0.172 exp H ALL Cl 03 89 0.224 0.050 0.727 c m 80 0.201 0.040 Cl 13 33 0.083 0.007 WBR ANWR AKR NCDE Cl 07 20 0.050 0.003 exp H 0.740 0.516 0.643 0.598 ClOI 7 0.018 0.000 Fst -0.017 0.291 0.116 0.178 C109 4 0.010 0.000 Dl 72 82 0.206 0.042 exp H ALL Dl 78 75 0.188 0.035 0.871 Dl 77 51 0.128 0.016 Dl 81 44 0.111 0.012 WBR ANWR AKR NCDE Dl 86 36 0.090 0.008 exp H 0.845 0.836 0.841 0.836 Dl 82 32 0.080 0.006 Fst 0.03 0.04 0.035 0.04 Dl 84 23 0.058 0.003 D176 17 0.043 0.002 Dl 74 17 0.043 0.002 Dl 80 11 0.028 0.001 Dl 83 3 0.008 0.000 D179 4 0.010 0.000 Dl 75 3 0.008 0.000 2 2 2 Table 40 (continued). Fst over all populations (expected). Mean Fst is at end of table. All populations allele number freq P2 LI 55 192 0.480 0.230 exp H ALL L157 120 0.300 0.090 0.659 LI 61 . 40 0.100 0.010 1159 40 0.100 0.010 WBR ANWR AKR NCDE LI 71 4 0.010 0.000 exp H 0.662 0.527 0.696 0.58 LI 63 4 0.010 0.000 Fst -0.004 0.201 -0.055 0.12 M208 162 0.407 0.166 exp H ALL M214 102 0.256 . 0.066 0.736 M212 45 0.113 0.013 M206 38 0.095 0.009 WBR ANWR AKR NCDE M210 41 0.103 0.011 exp H 0.697 0.727 0.758 0.768 M218 9 0.023 0.001 Fst 0.053 0.012 -0.03 -0.043 M222 I 0.003 0.000 Pl 53 123 0.308 0.095 exp H ALL Pl 57 70 0.175 0.031 0.813 Pl 61 63 0.158 0.025 Pl 51 51 0.128 0.016 WBR ANWR AKR NCDE Pl 59 43 0.108 0.012 exp H 0.792 0.827 0.82 0.768 Pl 55 34 0.085 0.007 Fst 0.026 -0.017 -0.008 0.056 P149 13 0.033 0.001 Pl 39 2 0.005 0.000 Pl 41 I 0.003 0.000 Xl 37 142 0.357 0.127 exp H ALL Xl 35 75 0.188 0.035 0.752 X141 107 0.269 0.072 Xl 33 35 0.088 ' 0.008 WBR ANWR AKR ' NCDE Xl 31 26 0.065 0.004 exp H 0.741 0.718 0.77 0.329 Xl 29 9 0.023 0.001 Fst 0.015 0.046 -0.023 0.563 Xl 39 I 0.003 0.000 Xl 43 3 0.008 . 0.000 average expected H over all popns given panmixia 0.763 mean expected H averaged over all popns 0.725 Fst over 8 loci (exp) ALL WBR ANWR AKR NODE 0.0498 0.021 0.0655 -0.0026 0.115 223 Table 41. Fst over all populations (observed). Mean Fst is at end of table. All populations allele number freq (p) p2 Al 94 156 0.390 0.152 exp H ALL Al 84 92 0.230 0.053 0.751 Al 92 73 0.183 0.033 Al 80 34 0.085 0.007 WBR ANWR AKR NCDE Al 90 16 0.040 0.002 obs H 0.776 0.8 0.765 0.625 A200 10 0.025 0.001 Fst -0.033 -0.065 -0.018 0.168 Al 86 9 0.023 0.001 Al 88 2 0.005 0.000 Al 96 ■ 3 0.008 0.000 Al 98 5 0.013 0.000 8160 151 0.378 0.143 exp H ALL 8158 62 0.155 0.024 0.792 8140 58 0.145 0.021 8148 41 0.103 0.011 WBR ANWR AKR NCDE 8164 17 0.043 0.002 obs H 0.796 0.8 0.824 0.813 8156 19 0.048 0.002 Fst -0.004 -0.01 -0.04 -0.026 8150 17 0.043 0.002 8152 17 0.043 0.002 8154 12 0.030 0.001 8162 . 5 0.013 0.000 8166 I 0.003 0.000 Cl 05 165 0.415 0.172 exp H ALL Cl 03 89 0.224 0.050 0.727 c m 80 0.201 0.040 Cl 13 33 0.083 0.007 WBR ANWR AKR NCDE Cl 07 20 0.050 0.003 obs H 0.770 0.6 0.647 0.75 ClOl 7 0.018 0.000 Fst -0.058 0.175 0.111 -0.031 C109 4 0.010 0.000 DT72 82 0.206 0.042 exp H ALL D178 75 0.188 0.035 0.871 Dl 77 51 0.128 0.016 Dl 81 44 0J11 0.012 WBR ANWR AKR NCDE Dl 86 36 0.090 0.008 obs H 0.875 0.867 0.706 0.75 Dl 82 32 0.080 0.006 Fst -0.004 0.005 0.19 0.139 Dl 84 23 0.058 0.003 D176 17 0.043 0.002 Dl 74 17 - 0.043 0.002 Dl 80 11 0.028 0.001 Dl 83 3 0.008 0.000 Dl 79 4 0.010 0.000 Dl 75 3 0.008 0.000 224 Table 41 (continued). Fst over all populations (observed). Mean Fst is at end of table. All populations allele number freq P2 L155 192 0.480 0.230 exp H ALL LI 57 120 0.300 0.090 0.659 LI 61 40 0.100 0.010 LI 59 40 0.100 0.010 WBR ANWR AKR NCDE LI 71 4 0.010 0.000 obs H 0.651 0.533 0.529 0.5 LI 63 4 0.010 0.000 Fst 0.013 0.192 0.198 0.242 M208- 162 0.407 0.166 exp H ALL M214 102 0.256 0.066. 0.736 M212 45 0.113 0.013 M206 38 0.095 0.009 WBR ANWR AKR NCDE M210 41 0.103 0.011 obs H 0.737 0.8 0.765 0.688 M218 9 0.023 0.001 Fst -0.001 -0.087 -0.039 0.065 M222 I 0.003 0.000 P153 123 0.308 0.095 exp H ALL P157 70 0.175 0.031 0.813 PT 61 63 0.158 0.025 PT 51 51 0.128 0.016 WBR ANWR AKR NCDE Pl 59 43 0.108 0.012 obs H 0.796 0.933 0.882 0.813 Pl 55 34 0.085 0.007 Fst 0.021 -0.147 -0.085 3E-04 Pl 49 13 0.033 0.001 Pl 39 2 0.005 0.000 Pl 41 I 0.003 0.000 Xl 37 142 0.357 0.127 exp H ALL Xl 35 75 0.188 0.035 0.752 X141 107 0.269 0.072 X133 35 0.088 0.008 WBR ANWR AKR NCDE X131 26 0.065 0.004 obs H 0.783 0.667 0.813 . 0.313 Xl 29 9 0.023 0.001 Fst -0.041 0.113 -0.081 0.584 X139 I 0.003 0.000 X143 3 0.008 0.000 average expected H over all popns given panmixia 0.763 mean observed H averaged over all popns 0.73 Fst over 8 loci (obs) ALL WBR ANWR AKR NODE 0.0433 -0.014 0:017 0.0288 0.139 225 Table 42. WBR two generation comparisons. WBR I st generation bears WBR 2nd generation bears A alleles number frequency variance A alleles number frequency variance Al 94 54 0.380 0.002 Al 94 63 0.420 0.002 Al 84 33 0.232 0.001 Al 84 39 0.260 0.001 Al 92 35 0.246 0.001 Al 92 28 0.187 0.001 Al 80 10 0.070 0.000 Al 80 11 0.073 0.000 Al 90 7 0.049 0.000 Al 90 4 0.027 0.000 A200 2 0.014 0.000 A200 3 0.020 0.000 Al 86 0 0.000 0.000 Al 86 I 0.007 0.000 Al 88 I 0.007 0.000 Al 88 I 0.007 0.000 TOTAL 142 I TOTAL 150 I B alleles number frequency varianpe B alleles number frequency variance BI 60 55 0.387 0.002 BI 60 65 0.433 0.002 BI 40 23 0.162 0.001 B140 24 0.160 0.001 BI 58 20 0.141 0.001 BI 58 27 0.180 0.001 B148 14 0.099 0.001 BI 48 12 0.080 0.000 BI 64 7 0.049 0.000 BI 64 8 0.053 0.000 BI 56 9 0.063 0.000 BI 56 4 0.027 0.000 BI 50 5 0.035 0.000 BI 50 6 0.040 0:000 BI 52 6 0.042 0.000 BI 52 3 0.020 0.000 BI 54 .3 0.021 0.000 BI 54 I 0.007 0.000 TOTAL 142 I TOTAL 150 I C alleles number frequency variance C alleles number frequency variance 0105 49 0.345 0.002 Cl 05 53 0.353 0.002 0103 36 0.254 0.001 Cl 03 38 0.253 0.001 0111 33 0.232 0.001 cm 40 0.267 0.001 0113 14 0.099 0.001 Cl 13 14 0.093 0.001 0107 7 0.049 0.000 Cl 07 5 0.033 0.000 0101 2 0.014 0.000 ClOl 0 0.000 0.000 0109 I 0.007 0.000 Cl 09 0 0.000 0.000 TOTAL 142 I TOTAL 150 I D alleles number frequency variance D alleles number frequency variance Dl 72 37 0.261 0.001 Dl 72 36 0.240 0.001 D178 25 0.176 0.001 D178 36 0.240 0.001 Dl 77 23 0.162 0.001 Dl 77 20 0.133 0.001 Dl 81 18 0.127 0.001 Dl 81 11 0.073 0.000 Dl 86 10 0.070 0.000 Dl 86 16 0.107 0.001 Dl 82 11 0.077 0.001 Dl 82 14 0.093 0.001 Dl 84 6 0.042 0.000 Dl 84 7 0.047 0.000 Dl 74 7 . 0.049 0.000 Dl 74" 4 0.027 0.000 Dl 76 4 0.028 0.000 Dl 76 5 0.033 0.000 Dl 80 I 0.007 0.000 Dl 80 I 0.007 0.000 TOTAL 142 I TOTAL ' 150 I J. 226 Table 42 (continued). WBR two generation comparisons. WBR I st generation bears WBR 2nd generation bears L alleles number frequency variance L alleles number frequency variance L155 65 0.458 0.002 LI 55 74 0.493 0.002 LI 57 37 0.261 0.001 LI 57 44 0.293 0.001 LI 61 19 0.134 0.001 LI 61 20 0.133 0.001 LI 59 18 0.127 0.001 LI 59 9 0.060 0.000 L171 I 0.007 0.000 LI 71 3 0.020 0.000 LI 63 2 0.014 0.000 LI 63 0 0.000 0.000 TOTAL 142 I TOTAL 150 I M alleles number frequency variance M alleles number frequency variance M208 64 0.451 0.002 M208 70 0.467 0.002 M214 39 0.275 0.001 M214 41 0.273 0.001 M212 17 0.120 0.001 M212 15 0.100 0.001 M206 10 0.070 0.000 M206 13 0.087 0.001 M210 7 0.049 0.000 M210 8 0.053 0.000 M218 5 0.035 0.000 M218 2 0.013 0.000 M222 0 0.000 0.000 M222 I 0.007 0.000 TOTAL 142 I TOTAL 150 I P alleles number frequency variance P alleles number frequency variance PT 53 53 0.373 0.002 P153 48 0.320 0.001 PT 57 28 0.197 0.001 P157 30 0.200 0.001 PT 61 ' 29 0.204 0.001 Pl 61 24 0.160 0.001 P151 8 0.056 0.000 Pl 51 15 0.100 0.001 P159 9 0.063 0.000 Pl 59 14 0.093 0.001 P155 8 0.056 0.000 Pl 55 13 0.087 0.001 P149 6 0.042 0.000 Pl 49 5 0.033 0.000 Pl 39 I 0.007 0.000 Pl 39 I 0.007 0.000 TOTAL 142 I TOTAL 150 I X alleles number frequency variance X alleles number frequency variance Xl 37 58 0.408 0.002 X137 • 58 0.387 0.002 Xl 35 31 0.218 0.001 X135 32 0.213 0.001 XMI 24 0.169 0.001 X141 37 0.247 0.001 Xl 33 17 0.120 0.001 X133 13 0.087 0.001 Xl 31 7 0.049 0.000 Xl 31 7 0.047 0.000 X129 5 0.035 0.000 Xl 29 . 3 0.020 0.000 TOTAL 142 I TOTAL 150 I 227 Table 43. Variance and covariance of known progeny num ber. A d fe m a le s PROGENY fm know n ff know n A d m a le s m m know n m f know n PROGENY 1 0 8 6 4 i I 1 0 8 7 5 0 2 4 2 6 1 0 8 9 8 I 2 6 0 6 1 0 9 2 I 0 I 2 3 5 1 0 9 5 4 2 0 2 2 4 1 0 9 7 10 3 3 2 I 3 1 1 0 2 3 0 2 0 3 3 1 1 0 5 4 I I I I 2 1 1 0 6 3 0 31 T1 0 I 1 1 1 0 2 2 0 i 0 I 11 1 1 2 I I i I 2 11 2 1 2 I I i 0 I 1 1 2 5 3 I 0 i 0 I 1 1 3 4 6 I 2 0 I I 1 1 3 6 2 2 0 0 0 0 1 1 3 8 3 I 2 0 0 0 1 1 3 9 5 I I 0 0 0 1141 3 I 0 0 0 0 1 1 4 3 2 0 I 0 0 0 1 1 4 6 I 0 I 0 0 0 1 1 4 9 2 I I 0 0 0 1 1 5 4 3 I 0 0 0 0 1 1 5 8 2 2 0 0 0 0 1 1 6 6 6 I I 0 0 0 1 1 6 7 4 I 2 0 0 0 1 1 6 9 7 0 2 0 0 0 1 1 7 4 I I 0 0 0 0 1 1 7 6 5 2 0 0 0 0 1 1 7 7 I I 0 0 0 0 1 1 7 8 3 0 I 0 0 0 1 1 7 9 2 0 I 0 0 0 1 4 2 4 I I 0 0 0 0 1 4 2 5 5 4 I 0 0 0 1 4 3 7 3 2 0 0 0 0 1 4 3 8 5 2 0 0 0 0 1 4 3 9 6 2 I 0 0 0 1 4 4 0 3 2 I 0 0 0 1441 3 3 0 0 0 0 1 4 5 4 3 I 2 1 4 5 7 2 0 2 1 4 5 8 2 I I 1 4 6 0 3 2 0 1461 2 I I 1 4 6 4 6 2 I 1 4 7 9 i I 0 1 7 1 6 2 0 2 1 7 3 4 2 2 0 1 7 3 9 2 2 0 1 7 4 5 2 I I 1 7 4 9 3 I 2 COUNT 5 0 3 7 SUM 1 6 5 3 0 VARIANCE BY SEX: 0 .8 0 4 0 .7 5 1 1 .581 0 .6 8 6 COVARIANCE FEMALES COVARIANCE MALES C o lum n I C o lum n 2 C o lum n I C o lum n 2 C o lum n I 0 .7 8 8 C olum n 1 1 .5 3 8 C o lum n 2 - 0 .2 8 9 0 .7 3 6 C olum n 2 0 .3 7 0 .6 6 8 VARIANCE FEMALES m e a n VARIANCE MALES m e an 4 .3 6 6 3 .5 4 1 3 .0 2 7 0 .9 7 3 MONTANA STATE UNIVERSITY LIBRARIES 3 4 762 10220960 6