published 7 January 2003 doi 101098rspb20022184270 2003 Proc R Soc Lond B Ilpo Kojola Adriano Casulli Haringkan Sand Petter Wabakken and Hans Ellegren

Karin Sundqvist Oslashystein Flagstad Jennifer Seddon Susanne Bjouml rnerfeldtminusCarles Vilagrave Anna population by a single immigrant

)Canis lupusRescue of a severely bottlenecked wolf (

Referenceshttprspbroyalsocietypublishingorgcontent270151091related-urls

Article cited in

Email alerting service hereright-hand corner of the article or click Receive free email alerts when new articles cite this article - sign up in the box at the top

httprspbroyalsocietypublishingorgsubscriptions go to Proc R Soc Lond BTo subscribe to

on May 17 2014rspbroyalsocietypublishingorgDownloaded from on May 17 2014rspbroyalsocietypublishingorgDownloaded from

Received 20 June 2002Accepted 3 September 2002

Published online 21 November 2002

Rescue of a severely bottlenecked wolf(Canis lupus) population by a single immigrantCarles Vila1 Anna-Karin Sundqvist1 Oslashystein Flagstad1Jennifer Seddon1 Susanne Bjornerfeldt1 Ilpo Kojola2 Adriano Casulli3HaEcirc kan Sand4 Petter Wabakken5 and Hans Ellegren1

1Department of Evolutionary Biology Evolutionary Biology Centre Uppsala University Norbyvagen 18DSE-752 36 Uppsala Sweden2Finnish Game and Fisheries Research Institute Oulu Game and Fisheries Research Tutkijantic 2AFIN-90570 Oulu Finland3Laboratory of Parasitology Istituto Superiore di Sanita Viale Regina Elena 299 00161 Rome Italy4Grimso Research Station Swedish University of Agricultural Sciences SE-730 91 Riddarhyttan Sweden5Hedmark College Department of Forestry and Wilderness Management N-2480 Koppang Norway

The fragmentation of populations is an increasingly important problem in the conservation of endangeredspecies Under these conditions rare migration events may have important effects for the rescue of smalland inbred populations However the relevance of such migration events to genetically depauperate natu-ral populations is not supported by empirical data We show here that the genetic diversity of the severelybottlenecked and geographically isolated Scandinavian population of grey wolves (Canis lupus) foundedby only two individuals was recovered by the arrival of a single immigrant Before the arrival of thisimmigrant for several generations the population comprised only a single breeding pack necessarilyinvolving matings between close relatives and resulting in a subsequent decline in individual heterozygos-ity With the arrival of just a single immigrant there is evidence of increased heterozygosity signi cantoutbreeding (inbreeding avoidance) a rapid spread of new alleles and exponential population growthOur results imply that even rare interpopulation migration can lead to the rescue and recovery of isolatedand endangered natural populations

Keywords wolf genetic rescue bottleneck founding effect migration Scandinavia

1 INTRODUCTION

Human pressure is leading increasingly to the isolation ofnatural populations (Ceballos amp Ehrlich 2002) with aconsequent loss of genetic variability (Frankham 1995)and an elevated risk of extinction (Saccheri et al 1998Lande 1999 Bijlsma et al 2000 Higgins amp Lynch 2001)However laboratory and translocation experiments haveindicated that small and inbred populations may be res-cued by the contribution of only a small number of immi-grants (Westemeier et al 1998 Madsen et al 1999)preventing inbreeding depression (Spielman amp Frankham1992 Ebert et al 2002) and inducing profound changes ingenetic structures (Ball et al 2000 Saccheri amp Brake eld2002) Nevertheless the impact of such rare migration onthe survival of natural populations is unclear To addresssuch issues in a natural population would require the closemonitoring of a population at an individual level at andafter the time of immigration

Wolves have been competing with humans for millen-nia Such con ict has led to intense persecution withactive efforts towards their extermination especiallyef cient through poisoning and rearm hunting from thenineteenth century (Mech 1970) As a result of extermi-nation campaigns wolf (Canis lupus) populations havebeen decimated and fragmented worldwide On the Scan-

Author for correspondence (hansellegrenebcuuse)

Proc R Soc Lond B (2003) 270 91ndash97 91 Oacute 2002 The Royal SocietyDOI 101098rspb20022184

dinavian peninsula hundreds of wolves were killed everyyear as late as the mid nineteenth century After a dra-matic decline the population was considered extinct bythe 1960s (Wabakken et al 2001a) However in 1983 abreeding wolf pack was unexpectedly discovered in south-ern Scandinavia more than 900 km from the nearestknown occurrence in Finland and Russia sparking contro-versy as to its origin The newly established populationcomprised only one breeding pack per year and consistedof fewer than 10 individuals during the 1980s Preliminarygenetic analysis showed a loss of genetic variability due toinbreeding in this newly founded population (Ellegren etal 1996) However without changes in habitat preyavailability legal protection or the level of illegal huntingthe population suddenly started to grow exponentiallyafter 1991 (mean annual growth rate of 29 (Wabakkenet al 2001a)) Despite both legal and illegal hunting thereare now 10ndash11 breeding packs and an estimated totalpopulation size of 90ndash100 individuals ( gure 1a)

The rigorous monitoring and continuous sampling of thiswolf population since its foundation provides an idealopportunity to follow the changes in genetic diversity in asmall and isolated population Using a variety of moleculartools including maternally (mitochondrial DNA (mtDNA))paternally (Y chromosome) and bi-parentally (autosomaland X-chromosome microsatellites) inherited markers wehave been able to track the populationrsquos origin to follow thechanges in genetic diversity over time and to assess theimpact of rare immigrants on this natural population

on May 17 2014rspbroyalsocietypublishingorgDownloaded from

92 C Vila and others Rescue of a wolf population

100

90

80

70

60

50

40

30

20

10

0

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

10

09

08

07

06

05

04

03

02

01

0

12

8

6

4

2

0

no o

f br

eedi

ng p

acks

esti

mat

ed p

opul

atio

n si

zein

divi

dual

het

eroz

ygos

ity

year

a)

b)

10

Figure 1 (a) Estimated total number of individuals (line left-hand scale) and packs where breeding was con rmed or likely tohave occurred (columns right-hand scale) in the Scandinavian wolf population since the rst recorded breeding in 1983(based on Wabakken et al 2001a) (b) Individual microsatellite heterozygosity (at 19 autosomal microsatellite loci) inScandinavian wolves by year of birth Each of the circles represents one individual The leftmost grey circle with an estimateddate of birth of 1978 represents the founding female Samples with birth dates prior to 1991 when the population includedonly one pack are shown in grey Black circles indicate wolves born between 1991 and 1995 that carry a number of newalleles and express the increase in heterozygosity resulting from the reproduction of one immigrant

2 MATERIAL AND METHODS

(a) SamplesWe analysed tissue or blood samples from 94 Scandinavian

wolves collected between 1984 and 2001 93 wolves from Fin-land and northwest Russia (Tvier and Smoliensk regions) thatare referred to as the eastern population and 66 pure and mixed-breed dogs One Scandinavian wolf sampled in 1984 in south-ern Norway had a genetic composition indicating that it was amigrant from the eastern wolf population and that it did notcontribute to the extant Scandinavian wolf population (eg ithas mtDNA and Y-chromosome haplotypes that differ from anyother animal in the Scandinavian population) This individualwas excluded from all subsequent analyses We also analysedthe historical Scandinavian wolf population represented byteeth samples of 30 museum specimens obtained between 1829and 1965 ie pre-dating the extinction of the 1960s (Flagstad

Proc R Soc Lond B (2003)

et al 2003) Information on the demographic changes in theScandinavian population has been published (Wabakken et al2001a) and is available in the yearly status reports of the SwedishNature Conservation Agency (2002 see also Aronson et al1999 2000 2001) and Norwegian Nature Conservation Agency(2002 see also Aronson et al 2002 Wabakken et al 19992001b 2002)

(b) Genetic methodsGenomic DNA was extracted using conventional protocols

(Sambrook et al 1989) DNA from historical specimens wasextracted using the Isoquick DNA extraction kit (Orca ResearchInc Bothell WA USA) and analysed in a separate laboratorydedicated to low-copy DNA research to minimize the risks ofcontamination The authenticity of the results was veri ed byreplicates carried out by different researchers (Hofreiter et al

on May 17 2014rspbroyalsocietypublishingorgDownloaded from

Rescue of a wolf population C Vila and others 93

2001) Partial mitochondrial control region sequences wereobtained for the wolf samples as described in Vila et al (1999)Four Y-chromosome-linked microsatellites (Sundqvist et al2001) were typed and combined into haplotypes Five X-chromosome-linked microsatellites (REN296K08 FH2548FH2584 FH2985 and FH3027 obtained from the Fred Hutch-inson Cancer Research Center (FHCRC) Dog Genome Project(httpwwwfhcrcorgsciencedog_genomedoghtml)) and 19autosomal microsatellite markers (c2001 c2006 c2010 c2017c2054 c2079 c2088 and c2096 (Francisco et al 1996) vWF(Shibuya et al 1994) u109 u173 u225 u250 and u253(Ostrander et al 1993) and PEZ01 PEZ03 PEZ05 PEZ06 andPEZ12 and PEZ20 (PerkinndashElmer Zoogen see the FHCRCDog Genome Project)) were used to genotype the modern wolfsamples A subset of 12 autosomal microsatellite markers(c2001 c2054 c2088 c2096 vWF u109 u173 u225 u253PEZ01 PEZ03 and PEZ05) chosen because of their reliableampli cation even in samples of low quality was successfullytyped in the historical samples In dogs (Canis familiaris) 15autosomal microsatellites were typed (c2001 c2010 c2017c2054 c2079 c2088 c2096 vWF u250 u253 PEZ01 PEZ03PEZ05 PEZ06 and PEZ12) The X-linked and autosomalmicrosatellites were ampli ed by polymerase chain reaction(PCR) in 10 m l reactions containing 20 ng of DNA template20 mM MgCl2 02 mM dNTPs 32 pmol of each primer and05 units of DNA polymerase (Amplitaq Gold (Applied Biosys-tems Foster City CA USA) for recent samples and HotStarDNA Polymerase (QIAGEN GmbH Hilden Germany) forhistorical samples) The PCR pro le was identical across allmarkers and included an initial denaturation step of 95 degC for10 min (15 min for HotStar ampli cations) 11 touch-downcycles with 94 degC for 30 s 58 degC for 30 s decreasing by 05 degCin each cycle and 72 degC for 1 min 28 cycles of 94 degC for 30 s52 degC for 30 s and 72 degC for 1 min and a nal extension of 72 degCfor 10 min All microsatellite ampli cations and sequencingreactions were run on an ABI377 (Applied Biosystems) sequen-cing instrument

(c) Data analysisFactorial correspondence analysis (Benzecri 1973) which dis-

plays the genetic similarity among samples in a two-dimensionalgraphical space and an assessment of the genetic variability ineach population were undertaken using the software Genetix

v 402 (Belkhir et al 2001) The degree of differentiationbetween populations was measured by the statistic u an esti-mator of FST (Weir amp Cockerham 1984) and by Neirsquos distance(1978) and its signi cance was assessed against 1000 permu-tations in Genetix Individual pairwise relationships wereassessed in Kinship v 10 (Goodnight amp Queller 1999) bydetermining the likelihood ratio of a pair of autosomal microsat-ellite genotypes for a speci ed primary and null relationship andattributing signi cance by simulation

Although the rst breeding female was sampled the genotypeof the rst breeding male was inferred indirectly His recon-structed genotype was based on the genotypes of three wolvesavailable from the rst litter (born in 1983) which were con- rmed to be full siblings by both Kinship analysis and an exam-ination of the compatibility of all loci and that of their presumedmother The genotype of the unsampled father was recon-structed manually for each locus by deducing the inheritance ofalleles from each parent to the offspring Similarly the genotypeof a presumed immigrant breeding for the rst time in 1991was inferred by attributing to it the alleles present in its offspring

Proc R Soc Lond B (2003)

but not previously seen in the Scandinavian population togetherwith alleles at homozygous loci in the offspring At some loci inboth reconstructed genotypes we could infer only one of thetwo alleles In such cases we assumed that the individual washomozygous at these loci to facilitate the estimation of its prob-able origin The origin of the founders (including the recon-structed genotypes) was determined with a Bayesian approach asimplemented in the program Structure (Pritchard et al 2000)Individual polymorphism was estimated as the proportion ofheterozygous autosomal microsatellite loci (Ellegren et al 1996)The age of most sampled wolves from the extant Scandinavianpopulation has been estimated (Wabakken et al 2001a) but ifunknown we assumed the collection year minus one for theanalysis of the changes in heterozygosity through time

3 RESULTS AND DISCUSSION

(a) Founding of the Scandinavian wolf populationSamples for DNA analysis are available for a signi cant

part of the Scandinavian wolf population from 1984 to2001 (n = 93) ie from one year following the re-establishment of the population until the present Elevenanimals with an estimated year of birth between 1983 and1990 show no more than four alleles at any of 19 autoso-mal microsatellite loci and no more than three alleles atany of ve X-linked microsatellite loci Among these ani-mals there is a maximum of two alleles per locus that arenot carried by the rst breeding female which died in1985 and is sampled The markers used for genotypingare highly variable in outbred wolf populations (table 2)and so these results indicate that there were only two foun-ders The reconstructed paternal genotype can togetherwith the genotype of the rst breeding female explain thepresence of all alleles seen in the population among ani-mals born prior to 1991 Finally all animals born before1991 carry the same mtDNA haplotype and the six malessampled from this time-period are xed for the same Y-chromosome haplotype as revealed by four highly variableY-chromosome-linked microsatellites (Sundqvist et al2001) From this we conclude that only two individualsone male and one female founded the population

The origin of the founders was traced using a Bayesianapproach (Pritchard et al 2000) to estimate the prob-ability of their genotypes corresponding to individualsoriginating from (i) the historical Scandinavian wolf popu-lation (ii) the neighbouring eastern wolf population or(iii) dogs

For this analysis no information about the origin of thesamples was speci ed in the program Structure and theindividual genotypes (using the nine autosomal microsat-ellite loci typed in the three datasets) were divided intothree groups each one of them as close as possible toHardyndashWeinberg and linkage equilibrium The resultinggroups corresponded very closely to the three populationsThe founding female has a probability of 095 of originat-ing from the eastern wolf population but less than 001of originating from the historical population This femalealso possesses six alleles found in the eastern wolf popu-lation but not seen in the historical Scandinavian wolfpopulation Similarly the inferred founding male has ahigh probability (098) of being an immigrant from theneighbouring eastern population The probabilities of adog ancestry are lower than 005 for both founders and

on May 17 2014rspbroyalsocietypublishingorgDownloaded from

94 C Vila and others Rescue of a wolf population

Table 1 Differentiation between populations at autosomal microsatellite loci(FST estimated by u (Weir amp Cockerham 1984) above the diagonal and the distance of Nei (1978) below the diagonal Thenumber of microsatellites considered varies between the comparisons (see sect 2))

wolves

Scandinavia eastern historical dogs

Scandinavia mdash 0136 curren curren 0265 curren curren 0213 curren curren

eastern 0310 curren curren mdash 0089 curren curren 0103 curren curren

historical 0791 curren curren 0333 curren curren mdash 0184 curren curren

dogs 0603 curren curren 0399 curren curren 0704 curren curren mdash

curren curren Signi cantly different from 0 at p 001

2

1

0

ndash1

0 1axis 1

axis

2

Figure 2 Factorial correspondence analysis of wolves basedon the 12 microsatellite loci described in table 2 Thecontemporary Scandinavian wolf population (black squares)shows the lowest genetic variability and appears to bedifferentiated from both the historical Scandinavian (1829ndash1965 white squares) and neighbouring eastern (greysquares) wolf populations The axes explain 11 of the totalvariance

the probability of being F1 hybrids is similarly reduced(data not shown) As genetic variability is partitionedwithin populations and even more dramatically amongdog breeds it is dif cult to fully characterize the diversityof the possible source populations for the migrants andthis must be considered in the interpretation of the analy-sis However these results answer a long-term debate onthe origin of the population countering earlier suggestionsof a small group of wolves surviving undetected in Scandi-navian forests unauthorized releases from captive coloniesin Scandinavian zoos (Ellegren et al 1996) and hybridiz-ation between dogs and wolves

(b) Loss of genetic diversityEven though eastern immigrants founded the popu-

lation a factorial correspondence analysis ( gure 2) indi-cates that contemporary Scandinavian wolves are highlydifferentiated both from the neighbouring eastern popu-lation and from historical Scandinavian wolves All wolfpopulations are signi cantly differentiated as well aswolves and dogs (table 1 permutation test of u using 1000re-samplings p 001 in all comparisons (Belkhir et al2001)) Moreover the genetic variability is signi cantlylower in the contemporary Scandinavian wolf populationthan in the eastern or historical wolves (table 2Wilcoxonrsquos signed rank test for the expected heterozygos-ity and number of alleles p 005 in both cases) Low

Proc R Soc Lond B (2003)

genetic diversity and strong differentiation are consistentwith a very small number of individuals founding theScandinavian population and stress the in uence of gen-etic drift (Barton amp Charlesworth 1984 Merila et al 1996Tarr et al 1998) The founding effect may have formedin just 15 years a population that is highly differentiatedfrom the source population

The founding female was killed in 1985 As the geno-types of some offspring born in 1985ndash1990 are incompat-ible with them being fathered by the inferred foundingmale and with no evidence of new alleles arriving duringthis time period this necessarily implies mating betweensiblings Consistent with close inbreeding a sharp declinein individual heterozygosity ( gure 1b) is observed duringthe 1980s The expected loss of heterozygosity per gener-ation corresponds to (1 2 1(2Ne)) where Ne is the effec-tive population size (Hartl amp Clark 1997) Thus in apopulation with just two reproducing individuals weshould expect a 25 loss in heterozygosity per generationalthough the strict application of such a formula may notbe appropriate in extremely small populations Our obser-vations roughly correspond to this expectation as theaverage heterozygosity decreases from 070 in the individ-uals born in 1983 to 049 in the individuals born between1985 (after the replacement of both parents) and 1990 (adecrease in heterozygosity of 30 gure 1b) Moreoverrelatedness values between pairs of individuals from thistime (inferred from microsatellite genotypes) are similarto those observed after sibling mating in the captive Scan-dinavian zoo population (Ellegren 1999) for which mul-tiple deleterious effects associated with inbreedingdepression have been described (Laikre amp Ryman 1991Laikre et al 1993)

(c) Migration from neighbouring populationsBy contrast to the homogenous allele composition of

wolves from the 1980s ten new alleles over 19 loci sud-denly appeared in a group of six wolves that were born in1991 and 1992 and identi ed as siblings by Kinship analy-sis ( gure 1b) These siblings all males also carry a newY-chromosome haplotype showing that a male new to thepopulation fathered them Reconstruction of the genotypeof this new male and its comparison with eastern wolvesand dogs at 15 autosomal microsatellite loci assigns it witha probability of 099 to the eastern wolf population con- rming a new immigrant

The arrival of the male immigrant led to an increase inindividual heterozygosity ( gure 1b) from a mean of 049

on May 17 2014rspbroyalsocietypublishingorgDownloaded from

Rescue of a wolf population C Vila and others 95

Table 2 Variability at autosomal microsatellite loci for the Scandinavian the eastern and the historical Scandinavian wolf popu-lations and for dogs(Variability at the 12 loci shared among the wolf populations a limitation imposed by the analysis of historical specimens isshown in italics HE unbiased expected heterozygosity HO observed heterozygosity)

population N loci HE (sd) HO (sd) average no alleles (sd)

Scandinavia 93 19 055 (003) 059 (001) 374 (124)12 054 (004) 059 (001) 383 (134)

eastern 93 19 077 (002) 069 (001) 800 (356)12 077 (002) 071 (001) 800 (336)

historical 30 12 071 (002) 058 (003) 575 (171)dogs 66 15 070 (004) 058 (002) 813 (320)

(plusmn 003) in eight animals born in 1985ndash1990 to 062(plusmn 003) in 16 animals born in 1991ndash1995 (MannndashWhitney U-test p = 0014) The increase in heterozygositycan be directly linked to the immigrant as all wolves butfour that were born after 1993 (n = 72) carry at least oneof his alleles

The arrival of this immigrant coincides with the estab-lishment of a second breeding pack in 1991 and the startof rapid and exponential population growth ( gure 1)Moreover a permutation test on all animals born after1996 (n = 56) shows an excess of heterozygotes (theinbreeding coef cient FIS = 2008 p 001) implyingan avoidance of mating with close relatives (outbreeding)An initial excess of heterozygotes could be explained bythe admixture of two populations However by restrictingthe analysis to individuals born after 1996 ie after severalgenerations from the arrival of the migrant male it isimprobable that any effect of admixture would remain andtherefore we conclude that outbreeding occurred follow-ing the arrival of this immigrant

(d) Implications for conservationThe selection of dissimilar mates can be advantageous

since it may increase reproductive success (Amos et al2001) and thereby tness (Ball et al 2000 Hedrick ampKalinowski 2000 Keller amp Waller 2002) As members ofthe same population are more likely to share deleteriousmutations the arrival of a migrant can induce hybrid vig-our (Westemeier et al 1998 Madsen et al 1999 Ebert etal 2002) which may lead to immigrant alleles beingpresent in higher frequencies than predicted from neutralexpectations (Whitlock et al 2000 Ingvarsson amp Whitlock2000 Ebert et al 2002 Saccheri amp Brake eld 2002) Thearrival of the immigrant male wolf is thus likely to haveprovided the possibility for inbreeding avoidancedecreased the risk of inbreeding depression and triggeredpopulation growth Only immigrants that successfullyreproduce within the population can contribute to the res-cue of the population In this sense we have evidence ofanother immigrant wolf in Scandinavia in the early 1980sthat failed to reproduce and did not contribute to the gen-etic makeup of the current population (see sect 2a)

Many species have behavioural mechanisms to avoidinbreeding and the deleterious effects commonly associa-ted with it (Keller et al 1994 Pusey amp Wolf 1996 Smithet al 1997) Such mechanisms are likely to have contrib-uted to the low rate of population growth in Scandinavianwolves prior to 1991 It is possible that the observedexcess of heterozygotes results from a higher survival rate

Proc R Soc Lond B (2003)

of animals with ancestry from the new migrant as hasbeen observed in laboratory experiments (Spielman ampFrankham 1992 Ball et al 2000 Ebert et al 2002Saccheri amp Brake eld 2002) However to our knowledgeno eld data are available that could indicate these differ-ences in survival Instead it has been suggested thatinbreeding avoidance may have limited the expansion ofthe wolf population at Isle Royale in Lake Superior follow-ing a population decline (Wayne et al 1991) Henceavoidance of the deleterious effects of incestuous matingin a small population may enhance the extinction risk fromenvironmental or demographic stochasticity However asevidenced here migration even at a low rate can behighly bene cial for a natural population through geneticrescue and by allowing non-incestuous mating permittingincreased tness (Westemeier et al 1998 Ingvarsson ampWhitlock 2000) and rapid population growth

This work was supported by grants from the EnvironmentalProtection Agencies in Norway and Sweden Olle Engkvist andCarl Trygger foundations Swedish Hunting Association andNordic Council of Ministers HE is a Royal Swedish Acad-emy of Sciences Research Fellow supported by a grant fromKnut and the Alice Wallenberg foundationThe authors thankP Ingvarsson for his useful comments on the paper

REFERENCES

Amos W Worthington Wilmer J Fullard K Burg T MCroxall J P Bloch D amp Coulson T 2001 The in uenceof parental relatedness on reproductive success Proc R SocLond B 268 2021ndash2027 (DOI 101098rspb20011751)

Aronson AEcirc Wabakken P Sand H Steinset O K ampKojola I 1999 Varg i Skandinavien statusrapport for vintern1998ndash1999 Rapport nr 18 NaturvaEcirc rdverket Sverige Fyl-kesmannen i Hedmark OsloAkershus Ostfold fylker ampDirektoratet for naturforvaltning Norge (ISBN 82-7671-079-4) (In Swedish)

Aronson AEcirc Wabakken P Sand H Steinset O K ampKojola I 2000 Varg I Skandinavien statusrapport for vintern1999ndash2000 Rapport nr 2 NaturvaEcirc rdverket amp Samordnings-gruppen for bestaEcirc ndsovervakning av stora rovdjur SverigeFylkesmennene i Hedmark OsloAkershus Ostfold fylker ampDirektoratet for naturforvaltning Norge (ISBN 82-7671-124-3) (In Swedish)

Aronson AEcirc Wabakken P Sand H Steinset O K ampKojola I 2001 Varg I Skandinavien statusrapport for vintern2000ndash2001 Rapport nr 2-2001 NaturvaEcirc rdverket amp Samord-ningsgruppen for bestaEcirc ndsovervakning av stora rovdjurSverige Norsk institutt for naturforskning (NINA) ampDirektoratet for naturforvaltning Norge (ISBN 82-7671-190-1) (In Swedish)

on May 17 2014rspbroyalsocietypublishingorgDownloaded from

96 C Vila and others Rescue of a wolf population

Aronson AEcirc Wabakken P Sand H Steinset O-K ampKojola I 2002 The wolf of Scandinavia status report of the200102 winter Sweden Swedish Environmental Protec-tion Agency

Ball S J Adams M Possingham H P amp Keller M A2000 The genetic contribution of single male immigrants tosmall inbred populations a laboratory study usingDrosophila melanogaster Heredity 84 677ndash684

Barton N H amp Charlesworth B 1984 Genetic revolutionsfounder effects and speciation A Rev Ecol Syst 15133ndash164

Belkhir K Borsa P Chikhi L Raufaste N amp BonhommeF 2001 Genetix 402 logiciel sous Windows TM pour la gene-tique des populations Montpellier Laboratoire GenomePopulations Interactions CNRS UMR 5000 Universite deMontpellier II

Benzecri J P 1973 LrsquoAnalyse des Donnees T 2 Irsquo Analyse descorrespondances Paris Dunod

Bijlsma R Bundgaard J amp Boerema A C 2000 Doesinbreeding affect the extinction risk of small populationspredictions from Drosophila J Evol Biol 13 502ndash514

Ceballos G amp Ehrlich P R 2002 Mammal population lossesand the extinction crisis Science 296 904ndash907

Ebert D Haag C Kirkpatrick M Riek M HottingerJ W amp Pajunen V I 2002 A selective advantage to immi-grant genes in a Daphnia metapopulation Science 295485ndash488

Ellegren H 1999 Inbreeding and relatedness in Scandinaviangrey wolves Canis lupus Hereditas 130 239ndash244

Ellegren H Savolainen P amp Rosen B 1996 The geneticalhistory of an isolated population of the endangered grey wolfCanis lupus a study of nuclear and mitochondrial polymor-phisms Phil Trans R Soc Lond B 351 1661ndash1669

Flagstad Oslash Walker C W Vila C Sundqvist A-KFernholm B Hufthammar A K Wiig Oslash Kojola I ampEllegren H 2003 Two centuries of the Scandinavian wolfpopulation patterns of genetic variability and migration dur-ing an era of dramatic decline Mol Ecol (In the press)

Francisco L V Langston A A Mellersh C S NealC L amp Ostrander E A 1996 A class of highly polymorphictetranucleotide repeats for canine genetic mapping Mam-malian Genome 7 359ndash362

Frankham R 1995 Conservation genetics A Rev Genetics 29305ndash327

Goodnight K F amp Queller D C 1999 Computer softwarefor performing likelihood tests of pedigree relationship usinggenetic markers Mol Ecol 8 1231ndash1234

Hartl D L amp Clark A G 1997 Principles of population gen-etics 3rd edn Sunderland MA Sinauer

Hedrick P W amp Kalinowski S T 2000 Inbreedingdepression in conservation biology A Rev Ecol Syst 31139ndash162

Higgins K amp Lynch M 2001 Metapopulation extinctioncaused by mutation accumulation Proc Natl Acad Sci USA98 2928ndash2933

Hofreiter M Serre D Poinar H N Kuch M amp PaaboS 2001 Ancient DNA Nature Rev Genet 2 353ndash359

Ingvarsson P K amp Whitlock M C 2000 Heterosis increasesthe effective migration rate Proc R Soc Lond B 267 1321ndash1326 (DOI 101098rspb20001145)

Keller L F amp Waller D M 2002 Inbreeding effects in wildpopulations Trends Ecol Evol 17 230ndash241

Keller L F Arcese P Smith J N M HochachkaW M amp Stearns S C 1994 Selection against inbred songsparrows during a natural population bottleneck Nature 372356ndash357

Laikre L amp Ryman N 1991 Inbreeding depression in a cap-tive wolf (Canis lupus) population Conserv Biol 5 33ndash40

Proc R Soc Lond B (2003)

Laikre L Ryman N amp Thompson E A 1993 Hereditaryblindness in a captive wolf (Canis lupus) population fre-quency reduction of a deleterious allele in relation to geneconservation Conserv Biol 7 592ndash601

Lande R 1999 Extinction risks from anthropogenic ecologi-cal and genetic factors In Genetics and the extinction of species(ed L F Landweber amp A P Dobson) pp 1ndash11 PrincetonUniversity Press

Madsen T Shine R Olsson M amp Wittzell H 1999 Con-servation biology restoration of an inbred adder populationNature 402 34ndash35

Mech L D 1970 The wolf the ecology and behavior of anendangered species Garden City NY The Natural HistoryPress

Merila J Bjorklund M amp Baker A J 1996 The successfulfounder genetics of introduced Carduelis chloris (green nch)populations in New Zealand Heredity 77 410ndash422

Nei M 1978 Estimation of average heterozygosity and geneticdistance from a small number of individuals Genetics 89583ndash590

Norwegian Nature Conservation Agency 2002 Seewwwninanikunoskandulvhtm

Ostrander E A Sprague G F amp Rine J 1993 Identi cationand characterization of dinucleotide repeat (ca)n markers forgenetic-mapping in dog Genomics 16 207ndash213

Pritchard J K Stephens M amp Donnelly P 2000 Inferenceof population structure using multilocus genotype dataGenetics 155 945ndash959

Pusey A amp Wolf M 1996 Inbreeding avoidance in animalsTrends Ecol Evol 11 201ndash209

Saccheri I J amp Brake eld P M 2002 Rapid spread of immi-grant genomes into inbred populations Proc R SocLond B 269 1073ndash1078 (DOI 101098rspb20021963)

Saccheri I Kuussaari M Kankare M Vikman PFortelius W amp Hanski I 1998 Inbreeding and extinctionin a butter y metapopulation Nature 392 491ndash494

Sambrook E Fritsch F amp Maniatis T 1989 Molecular clon-ing Cold Spring Harbor NY Cold Spring Harbor Press

Shibuya H Collins B K Huang T H amp Johnson G S1994 A polymorphic (AGGAAT)n tandem repeat in anintron of the canine von Willebrand factor gene AnimGenet 25 122

Smith D Meier T Geffen E Mech L D Burch J WAdams L G amp Wayne R K 1997 Is incest common ingray wolf packs Behav Ecol 8 384ndash391

Spielman D amp Frankham R 1992 Modelling problems inconservation genetics using captive Drosophila populationsimprovement of reproductive tness due to immigration ofone individual into small partially inbred populations ZooBiol 11 343ndash351

Sundqvist A-K Ellegren H Olivier M amp Vila C 2001Y chromosome haplotyping in Scandinavian wolves (Canislupus) based on microsatellite markers Mol Ecol 101959ndash1966

Swedish Nature Conservation Agency 2002 See wwwninanikunoskandulvhtm

Tarr C L Conant S amp Fleischer R C 1998 Foundereffects and variation at microsatellite loci in an insular pass-erine bird the Laysan nch (Telespiza cantans) Mol Ecol7 719ndash731

Vila C Amorim I R Leonard J A Posada DCastroviejo J Petrucci-Fonseca F Crandall K AEllegren H amp Wayne R K 1999 Mitochondrial DNA phy-logeography and population history of the grey wolf Canislupus Mol Ecol 8 2089ndash2103

Wabakken P Aronson AEcirc Sand H Steinset O K ampKojola I 1999 Ulv i Skandinavia statusrapport for vinteren1998ndash1999 Rapport nr 19 NaturvaEcirc rdsverket Sverige Fyl-kesmennene i Hedmark Oslo amp Akershus Oslashstfold fylker amp

on May 17 2014rspbroyalsocietypublishingorgDownloaded from

Rescue of a wolf population C Vila and others 97

Direktoratet for naturforvaltning (ISBN 82-7671-080-8)(In Norwegian)

Wabakken P Sand H Liberg O amp Bjarvall A 2001a Therecovery distribution and population dynamics of wolveson the Scandinavian peninsula 1978ndash1998 Can J Zool 79710ndash725

Wabakken P Aronson AEcirc Sand H Steinset O K ampKojola I 2001b Ulv i Skandinavia statusrapport for vinteren2000ndash2001 Rapport nr 1-2001 Norsk institutt for natur-forskning (NINA) Direktoratet for naturforvaltning amp Nat-urvaEcirc rdsverket Sverige (ISBN 82-7671-165-0) (InNorwegian)

Wabakken P Aronson AEcirc Steinset O K amp Sand H 2002Foreloslashpig statusrapport om ulv i Skandinavia vinteren20012002 wwwninanikunoskandulv (In Norwegian)

Proc R Soc Lond B (2003)

Wayne R K (and 10 others) 1991 Conservation genetics of theendangered Isle Royale gray wolf Conserv Biol 5 41ndash51

Weir B S amp Cockerham C C 1984 Estimating F-statisticsfor the analysis of population structure Evolution 381358ndash1370

Westemeier R L Brawn J D Simpson S A Esker T LJansen R W Walk J W Kershner E L Bouzat J L ampPaige K N 1998 Tracking the long-term decline and recov-ery of an isolated population Science 282 1695ndash1698

Whitlock M C Ingvarsson P K amp Hat eld T 2000 Localdrift load and the heterosis of interconnected populationsHeredity 84 452ndash457

As this paper exceeds the maximum length normally permitted theauthors have agreed to contribute to production costs

on May 17 2014rspbroyalsocietypublishingorgDownloaded from