MAINTENANCE OF GENETIC DIVERSITY IN FOUR TAIGA SPECIALISTS

LEENAUIMANIEMI

Department of Biology,University of Oulu

OULU 2004

LEENA UIMANIEMI

MAINTENANCE OF GENETIC DIVERSITY IN FOUR TAIGA SPECIALISTS

Academic Dissertation to be presented with the assent ofthe Faculty of Science, University of Oulu, for publicdiscussion in Kuusamonsal i (Auditorium YB210),Linnanmaa, on August 20th, 2004, at 12 noon.

OULUN YLIOPISTO, OULU 2004

Copyright © 2004University of Oulu, 2004

Supervised byProfessor Markku OrellDocent Jaakko Lumme

Reviewed byProfessor Mats BjörklundProfessor Juha Merilä

ISBN 951-42-7409-1 (nid.)ISBN 951-42-7410-5 (PDF) http://herkules.oulu.fi/isbn9514274105/

ISSN 0355-3191 http://herkules.oulu.fi/issn03553191/

OULU UNIVERSITY PRESSOULU 2004

Uimaniemi, Leena, Maintenance of genetic diversity in four taiga specialists Department of Biology, University of Oulu, P.O.Box 3000, FIN-90014 University of Oulu, Finland 2004Oulu, Finland

AbstractGenetic diversity in three taiga specialists – the Siberian tit (Parus cinctus), the Siberian jay(Perisoreus infaustus) and the Siberian flying squirrel (Pteromys volans) – was assessed bycomparing DNA sequence variation across the mitochondrial control region and allele frequenciesof microsatellites from samples collected from Fennoscandia and Siberia. Population sizes of thesespecies have declined in association with fragmentation and loss of suitable forest habitat due tomodern forestry practices in Fennoscandia. The red squirrel (Sciurus vulgaris) served as a referencefor the flying squirrel.

Genetic differentiation among species studied ranged from a panmictic population in the Siberiantit to that of the strong differentiation of populations (θST = 53%) in the flying squirrel in Finland.MtDNA and microsatellite data, together with assignment studies, showed the Siberian jaypopulation to be significantly genetically structured and supported the existence of a metapopulationlike structuring in Fennoscandia. Division of genetic variation among flying squirrel populationsalong the ancient shoreline of the Littorina Lymnea Sea stage of the Baltic Sea (7000 BP) and twogeographically associated branches in the minimum spanning network supported a two-waycolonisation history for the species. The Finnish inland appears to have been colonised from the eastin association with the arrival of Norway spruce. At the same time, Coastal Finland was colonisedfrom the south-east through the Karelian Isthmus. Gene flow of the species appeared female biasedand restricted. Species exhibiting more restrictive dispersal characteristics and habitat requirementspossessed stronger population genetic structure than those with opposite characteristics.

Growth or contractions in population size leave characteristic signatures in mtDNA that can bestudied by comparing different sequence diversity estimates among populations. I applied thismethod to the species studied. Significant differences in nucleotide diversities indicated restrictionsin gene flow among populations in all species studied. Half of the Siberian jay populations gave asignal of population size bottleneck.

All the species studied showed differences in their population genetic structures across their entiredistribution ranges consistent with the multirefugia model, most likely to be attributable todifferences in their ecological characteristics and Pleistocene histories.

Keywords: conservation, microsatellites, mitochondrial DNA, phylogeography, taiga species

Acknowledgements

This research was carried out at the Department of Biology in the University of Oulu. I am grateful to my supervisors Professor Markku Orell and Dr. Jaakko Lumme for an interesting and important research topic and providing guidance during the work. In particular, Professor Orell has corrected numerous of versions of manuscripts. Many researchers and those ones who do nature conservation work, in addition to their normal every day work, have helped me by offering samples. All authors in the manuscripts have provided their insight and an important part to this research. I thank Jouni Aspi, Susanna Huttunen, Laura Kvist and Helmi Kuittinen for interesting scientific discussions.

The following colleagues were in the laboratory with me: Arttu Ahola, Eduardo Belda, Riikka Holopainen, Johanna Korvala, Jussi Kuusela, Laura Kvist, Minna Leppäjärvi, Maria Meinilä, Keijo Viiri and Marek Zietara, which I am grateful for. Phone calls with Maria made my days pleasant and somehow connected me to both normal and scientific life and, in addition, reminded me about the realities in this life. I am especially grateful to my three master’s thesis students: Päivi Kuuva, Heidi Soini and Pia Vaaraniemi. I am grateful to Shiang-Fan Chen, Chris Faulkes, Andrew Liech, Richard Nichols, Eva Sykorova, Patricia Mirol and the rest of the staff for taking greater care of my minor existence during my short visit to the Queen Mary Westfield College in London.

Thank you for: Soile Finne and Bohdan Sklepkovych (who also kindly checked the English of my Ph.D. thesis and improved it remarkably) for being better company at the Department of Biology and the rest of the staff for sharing. Liisa Räisänen and Marie-Héléne Muller for their friendship on my spare time at very last stages of my Ph.D. thesis work. Päivi and Stefan for their long-term friendship and joyful moments. Angela, Chris, Eeva, Hannat, Jatmo, Oskari and those friends and collagues with whom I, unfortunately, too rare was able spend my time, for better days. My family and particularly for my godmother.

Oulu, July, 2004 Leena Uimaniemi For daily planet

List of original papers

This thesis is based on the following publications, which are referred to in the text by their Roman numerals. I Uimaniemi L, Orell M, Kvist L, Jokimäki J & Lumme J (2003) Genetic variation of

the Siberian Tit Parus cinctus populations at the continental and regional levels: A mitochondrial sequence analysis. Ecography 26: 98-106.

II Uimaniemi L, Orell M, Mönkkönen M, Huhta E, Jokimäki J & Lumme J (2000) Genetic diversity in the Siberian jay Perisoreus infaustus in fragmented old-growth forests of Fennoscandia. Ecography 23: 669-677.

III Uimaniemi L, Lillandt B-G, Edenius L, Sklepkovych B & Orell M (2004) Present and Pleistocene effects in the genetic variation of the Siberian jay (Perisoreus infaustus). Manuscript.

IV Uimaniemi L, Orell M, Reunanen P & Lumme J (2004) Genetics of the Siberian flying squirrel (Pteromys volans) and the red squirrel (Sciurus vulgaris) in fragmented and continuous coniferous forests in Finland. Manuscript.

Contents

Abstract Acknowledgements List of original papers Contents 1 Introduction ................................................................................................................... 11

1.1 Environmental phenomena behind species' genetic variation.................................12 1.1.1 Ice Ages shaping phylogeography ...................................................................12 1.1.2 Fragmentation of boreal forests .......................................................................13

1.2 Evolutionary forces in population and conservation genetics.................................14 1.3 Markers...................................................................................................................16

1.3.1 Mitochondrial DNA and control region...........................................................16 1.3.2 Microsatellites .................................................................................................18

1.4 Species and populations..........................................................................................19 1.4.1 The Siberian tit (Parus cinctus)........................................................................19 1.4.2 The Siberian jay (Perisoreus infaustus) ...........................................................20 1.4.3 The Siberian flying squirrel (Pteromys volans) ...............................................21 1.4.4 The red squirrel (Sciurus vulgaris) ..................................................................22

2 Questions in focus .........................................................................................................23 3 Materials and methods...................................................................................................24

3.1 Molecular methods .................................................................................................24 3.2 Mitochondrial sequence diversity indices...............................................................25 3.3 Microsatellite allele diversity parameters ...............................................................26

4 Results and discussion...................................................................................................27 4.1 Phylogeography in Siberia......................................................................................27

4.1.1 Coalescence time estimates .............................................................................34 4.2 Colonisation of Fennoscandia.................................................................................35 4.3 Gene flow and dispersal studies .............................................................................36 4.4 Effective population size ........................................................................................38 4.5 Spatio-temporal population dynamics ....................................................................40

5 Concluding remarks and future prospects .....................................................................43 References Original papers

1 Introduction

Biodiversity conservation directs its attention to the levels of ecosystems, species and genes (UNCED 1992). Genetic diversity is affected by both past and present extrinsic environmental and intrinsic species specific factors. These factors are reflected in species distribution, population genetic structure and the amount of genetic variation within species. Ice ages have been major past factors shaping species distributions and phylogeography, the geographical distribution of genealogical lineages (Avise 1998). Ice started to recede from Fennoscandia only 11 000 years ago. Therefore, the evolutionary history of Northern Hemisphere fauna is rather young. Past range expansions are still detectable in the genomes of many species (Kvist et al. 1998, 1999a, b). However, also present-day environment is shaping species range, population sizes and amount of genetic diversity. Environmental destruction, namely the loss and fragmentation of habitats, is a serious present-day factor in impacting above mentioned species’ characteristics.

Neutral markers such as mitochondrial DNA and microsatellites are the most commonly used markers in phylogeographical and conservation genetic studies. Uniparental mode of inheritance makes mtDNA a good marker for studying species’ Pleistocene and colonisation history. MtDNA shows variation which has accumulated over thousands of generations through the random processes of drift and mutation as well as selection and adaptation, whereas microsatellites characterise more recent variation (Sherwin & Moritz 2000). The historical part of genetic variation is irreplaceable because the circumstances which generated the variation can only be surmised, and the timescale cannot be repeated. Genetic variation, which has arisen recently, is important for short-term conservation because of its potential consequences for current and future adaptation.

Because of their small numbers and often secretive manner, endangered species are generally difficult to observe in nature. Their distribution, mating system, dispersal and population size all affect the amount and allocation of genetic diversity. Molecular and population genetic studies offer important insights for example regarding the true effective dispersal rate on wide geographical scales, which is impossible to gain from pure population ecological studies. In the present thesis, I focus my attention on species important and characteristic of boreal fauna as well as on the interplay between ecological and evolutionary forces on the maintenance and loss of genetic variation.

12

1.1 Environmental phenomena behind species' genetic variation

1.1.1 Ice Ages shaping phylogeography

Pleistocene biogeographical events, which began 2.4 Myr ago by establishing an Arctic ice cap, have been ascribed a major role in promoting speciations and microevolutionary genetic diversification of avian and vertebrate taxa (Avise & Walker 1998, Hewitt 2000). Large fluctuations in climate made ice sheets to advance and recede every 41 000 years until 0.9 Myr ago (Hays et al. 1976, Gribbin 1989, Hewitt 1996 a, b). Thereafter, ice sheets increased dramatically and started to shift: increase and decrease in size with roughly a 100 000 years cycle. Our best scientific understanding comes from the Late Pleistocene, the last ice age, which started 135 000 years ago and reached its coldest period about 18 000 ago (Hewitt 1996 a,b). The ice reached 52°N at its maximum in Europe and exceeded past the Ural Mountains far to the east. However, there has been a large ice-free tundra area from Taimyrian Peninsula to Eastern Siberia (Alekseev 1997, Svendsen et al. 1999). Small parts in Eastern Siberia from present Yakutsk area over Verhoyansk Mountains to the uplands of Andur were covered by ice. The vegetation and faunal zones reached lower latitudes during Ice Ages than nowadays (Andersen & Borns 1997, Richardson 1998). Oxygen and carbon isotopes, magnetic and CO2 measures, fossil remains together with genetic information give more information about past changes in population sizes, different refugia, colonisation routes, colonisation modes and the timing of colonisations, local adaptations, extinctions and speciations during the Pleistocene (Rogers & Harpending 1992, Halkka et al. 1994, Rogers 1995, Hewitt 1996 a, b, Ibrahim et al. 1996, Bilton et al. 1998, Bermingham & Moritz 1998, Fedorov 1998, Taberlet 1998, Hewitt 2000, Lister & Bahn 2000, Ruokonen et al. 2000, Storz & Beaumont 2002).

Comparisons of intraspecific phylogeographies of codistributed species have revealed common vicariant factors and identified geographical areas in which populations have undergone independent evolution (Zink 1996, Bermingham & Moritz 1998, Smith et al. 2000). Phylogeographical studies of the European fauna have shown that populations generally have less genetic variation in previously glaciated areas (Hewitt 1996 a, b, Merilä 1997, Hewitt 2000). Colonisation of northern areas has mainly occurred from two southern Peninsulas, Iberia and Balkans, and north of Alps, whereas the Italian Peninsula has been a rather closed refugium (Bilton et al. 1998, Taberlet et al. 1998). There may have been cryptic refugia for species like the bank vole (Clethrionomys glareolus) and red deer (Cervus elaphus) in more northern Europe (Stewart & Lister 2001). Common phylogeographical features in Siberia and Asia remain to be studied. Most wide phylogeographical studies of the Siberian fauna have concentrated on avian species like the dunlin (Calidris alpina), the black-backed gull (Larus fuscus), the great tit (Parus major) and the willow tit (Parus montanus) (Wenink et al. 1993, 1994, 1996, Liebers & Helbig 2002, Kvist et al. 2001, 2003 a, b), with only a few studies being done on mammals e.g., the moose (Alces alces), the wood lemming (Myopus schisticolor) and the Arctic lemming (Lemmus lemmus) (Fedorov et al. 1996, 1998, 1999, Hundertmark et al. 2002, Liebers & Helbig 2002).

13

1.1.2 Fragmentation of boreal forests

The boreal coniferous forest, or taiga, extends across the Northern Hemisphere through Alaska, Canada, Scotland, Fennoscandia and Russia. The largest areas of the boreal forest are in Russia. Large parts of this biome have been destroyed or seriously fragmented because of modern forestry especially in Fennoscandia. Originally the forest landscape was fragmented by lakes, mires, local fires or windfalls as a part of the natural succession and renewal process (Esseen et al. 1997, Syrjänen et al. 1998). Nonetheless, boreal forests experience a slow renewal process making them and the species associated and specialized on them, seriously vulnerable to habitat alterations. In Finland, more than one third of endangered species are dependent on forest habitat (Rassi et al. 2000).

Fig. 1. Different forest use stages on time scale in Fennoscandia and Northern European Russia (Lloyd 1999).

Man has used forest since settling Finland (Fig. 1: phase 1). Pre-industrial forestry (Fig 1: phase 2) started by charcoal and tar production during the 17th and 18th century (Lloyd 1999). Populated areas, areas near harbours and navigable rivers were affected first. Large areas of old-growth forests became under intensive slash-and-burn cultivation especially in Eastern Finland during the 18th century (Heikinheimo 1915). At the same time, large-scale selective logging has been performed in Northern European Russia (Lloyd 1999). Industrial forest use (Fig 1: phase 3:1) started with selective logging for saw industry on 1890 – 1918. There were large deciduous forests in the previously burnt areas in Finland, where spruce intensively followed the succession in the beginning of the 20th century (Järvinen et al. 1977). Different political and economic histories in Fennoscandia and Northern Russia reflect their land use histories. In Russia, large-scale clear cutting began in the 1930s, although modern, also known as rotation, forestry never developed on a full scale (Lloyd 1999). In Finland, independence was obtained in 1917, after which a period of depression in the 1920’s, and the Second World War with following war taxes to Russia increased the demand of economical welfare causing increasing pressure on forestry, from which most of our national income was obtained. Most serious changes in forest landscape have been in Northern Finland since the 1950's (Järvinen et al. 1977, Virkkala 1990). Nowadays, forest use is changing towards

Stage 1 Stage 2 Stage 3:1 Stage 4

Stage 1 Stage 2 Stage 3:1

Stage 3:2Fennoscandia

NorthernEuropeanRussia

Stage 4

Stage 3:2

16008000 BC 19301900 20001950

14

sustainable forestry, where an attempt is being made to mimic natural processes and system complexity.

1.2 Evolutionary forces in population and conservation genetics

Genetic variation is introduced to a population by mutation, recombination and migration, for which antagonistic forces are selection and drift (Wright 1931). These evolutionary forces together with geographical discontinuity usually result in at least some degree of substructuring of populations over wide geographic scales (Avise 1994).

Once a mutation is introduced to a population its fate is mainly determined by selection and genetic drift. The molecular clock hypothesis predicts that the molecular evolutionary rate is constant over time (Zuckerkandl & Pauling 1965). However, molecular rate constancy has been tested against well-dated fossils or vicariant events in only a few cases. For example, mutation rates of mitochondrial cytochrome b sequence in Hawaiian drepanides, Yp1 gene in Hawaiian Drosophila and parts of the mitochondrial 12S and 16S rRNA and tRNAval genes in Laupala crickets have been calibrated to the timing of island formation (Fleischer et al. 1998). Typically molecular rates are extrapolated from point estimates relating an expected date of separation with molecular divergence between species (Bermingham & Moritz 2000). Usually molecular rates differ between taxa and genes (e.g., Vawter & Brown 1986, Ruokonen & Kvist 2002).

Extremely effective migration may prevent population differentiation for neutral markers over wide geographical scales. The willow tit (Parus montanus) is a good example of a species having a panmictic population genetic structure over its whole distribution, from the Atlantic to the Pacific Ocean (Kvist et al. 2003a). A totally panmictic population genetic structure over the whole distribution area is not usual. Dispersal and colonisation modes, on which population genetic structure is partially depended, differ between species (Ibrahim et al. 1996). Species have less genetic variation in newly colonised areas compared to refugia because of leading edge dispersal (Hewitt 1996 a b). According to this model, long distance dispersants set up colonies ahead of the main distribution and expand rapidly dominating in leading populations. Leptocurtic dispersal, where long-distance dispersants set up pocket populations far ahead of the main dispersing population, forms rather pronounced spatial structure especially when compared to normal and stepping stone dispersal models (Ibrahim et al. 1996). The normal dispersal model creates stronger spatial genetic structure than the stepping stone model, where gene flow occurs only between close neighbour populations (Ibrahim et al. 1996). Also the number of colonisation sources affects population genetic structure. Slatkin (1977), Pannell and Charlesworth (1999) have studied migrant and propagule pool migration on the degree of population genetic differentiation (FST). The propagule pool colonisation, where colonisation occurs from one source population, creates a very strong FST, whereas the migrant pool, where colonists are derived from several subpopulations, does not create a very pronounced FST.

Drift work most efficiently in small, isolated environments (Gaines et al. 1997, Clark et al. 1999, Dayanandan et al. 1999, Morden & Loeffler 1999, Rosquist & Prentice 2000, Fowler et al. 2000). Charles Darwin (1859) was the first one to study evolution on

15

isolated environments. He found that isolation, depending on the size of an area, leads to either speciation or preserves of vulnerable fauna. The very first model that described isolated populations was an island model under which evolution was fast. Traditionally substructuring of populations has been described with continent-island and stepping stone models. The metapopulation model (Fig. 2), usually characterised by population turnover rule, is a modern way to describe substructuring of populations (Hanski & Simberloff 1997). Almost all fragmented populations may be considered as metapopulations (Harrison & Taylor 1997). The Capercaillie (Tetrao urogallus) in the Alps is an excellent

example of a metapopulation (Segelbacher & Storch 2002). Capercaillie sink populations appear to have lost genetic variation as compared to source populations and they were more differentiated from others than source populations. In metapopulations, loss of heterozygosity may increase the risk of extinction (Saccheri et al. 1998).

Fig. 2. Different types of metapopulations (Harrison & Taylor 1997). 1) Classic ‘Levins’ - metapopulation, 2) ‘continent island’ also known as ‘source-sink’ metapopulation, 3) fragmented metapopulation, 4) unstable metapopulation, 5) a mixed metapopulation containing characteristics of almost all other metapopulation types.

Increasing isolation e.g., because of environmental destruction, leads to small populations having decreased genetic variation and a risk of increased inbreeding as for instance in the case of the massasauga rattlesnake (Sistrurus catenatus) populations in USA (Gibbs et al. 1997). However, some exceptions are known, e.g., the sugar maple (Acer saccharum), where gene flow has increased because of environmental fragmentation (Young et al. 1996, Porter 1998). Sometimes populations may recover from genetic bottlenecks without loosing their vitality because deleterious alleles become purged rapidly. A microsatellite study of the Mauritius kestrel (Falco tinnunculus) showed that despite extensive loss of genetic variation, the species was able to recover without addition of genetic variation (Groombridge et al. 2001). The first bottleneck signal, an excess in heterozygosity, has been detected in edge populations of the wolverine (Gulo gulo) and the Siberian flying squirrel (Pteromys volans) (Walker et al. 2001, Selonen 2002).

Endangered species often need to be monitored and proper geographical areas for conservation purposes need to be designed. Two types of conservation units have been defined based on genetic variation: management units (MUs) and evolutionary significant units (ESUs). MUs are defined based on significant haplotype or allele differences and they are units for short-term management and monitoring (Moritz 1994). ESUs should be reciprocally monophyletic for mtDNA alleles and differ significantly for allele frequencies at nuclear loci (Moritz 1994). Management units have been defined based on haplotype frequencies for Southern Queensland koala (Phascolarctos cinereus) and northern bettong (Bettongia tropica) populations in Australia (Houlden et al. 1999, Pope

1 2 3 4 5

16

et al. 2000). Although populations may fulfil genetic criteria surrounding MUs, they are not always recommended to as protective management units. For example, the red squirrel (Sciurus vulgaris) populations in Britain were not proposed to be distinct management units despite having distinct haplotype frequencies, a result most likely because of the recent fragmentation of populations (Barratt et al. 1999). Red squirrels may compensate the effects of habitat fragmentation well as increased gene flow associated with increasing connectivity of habitat has been detected by comparing microsatellite allele frequencies in museum and present-day samples of the red squirrel from previously badly fragmented forest area, which was planted during the past few decades in Scotland (Hale et al. 2001).

1.3 Markers

1.3.1 Mitochondrial DNA and control region

Mitochondrial DNA (mtDNA) is a circular molecule (16 – 20 kb) on the inner mitochondrial membrane. It constitutes a non-coding control region, 22 tRNA, 13 mRNA, 12S and 16S rRNA coding genes encoding proteins for electron transport and oxidative phosphorylation. MtDNA is translated by a unique genetic code. Instead of introns there are small intergenic regions between genes, where reading frames may overlap and stop codons are created by polyadenylation (Borst & Grivell 1981, Desjardins & Morais 1990). MtDNA gene order differs among animal taxa (Fig. 3) having multiple independent origins (Desjardins & Morais 1990, 1991, Kumazawa et al. 1996, Mindell et al. 1998, Eberhard et al. 2001, Haring et al. 2001, Väli 2002). The mutation model for explaining the changed gene order is replication slippage through which duplication of a gene region evolves (Desjardins & Morais 1990, Quinn & Wilson 1993). Replication starts from the nascent H strand of the control region. The leading 3' strand mispairs with a same kind of sequence from which replication continues. Mutations and deletions may evolve on duplicated regions. In later generations the amount of genes are reduced to the original number either by neutral or selective deletions (Quinn & Wilson 1993).

Maternal inheritance, a rather high mutation rate (10-8 base pairs/generation), a mostly uniparental mode of inheritance (Berlin & Ellegren 2001) and the relative ease of amplification (Kocher et al. 1989) have made mitochondrial DNA a popular marker for use in evolutionary studies. Exceptions to generally maternal inheritance are paternal leakage and biparental inheritance, which are found in the house mouse (Mus musculus), the great tit and the marine mussel (Musselus edulis) (Gyllensten et al. 1991, Kvist et al. 2003b, Zouros et al. 1992). A high mutation rate can be explained by the lack of proof reading activity in the replication enzyme (Brown 1983). In addition to the incapability of proof reading activity, the enzyme incorporates the wrong nucleotides during replication more often when compared to nuclear replication enzymes. Free radicals on the inner mitochondrial membrane may cause mutations. Mitochondrial DNA evolves 5 – 10 times faster than nuclear DNA (Brown et al. 1979). The main evolutionary force affecting

17

mitochondrial DNA variation is genetic drift, but also selection and genetic hitchhiking may be involved (Ballard & Kreitman 1995, Galtier & Boursot 2000). mtDNA may be exposed to mildly deleterious mutations (Fry 1999). mtDNA is inherited together with the

W chromosome in avian species and the X chromosome in mammals.

Fig. 3. MtDNA gene order in different animal phyla (Desjardins & Morais 1990, 1991, Kumazava et al. 1996, Mindell et al. 1998, Eberhard et al. 2001, Haring et al. 2001, Väli 2002).

Mutation rates vary between mitochondrial genes making different genes suitable for different purposes. Cytochrome b and 12SrRNA genes, for example, are excellent markers for phylogenetic studies (Oshida et al. 1996). The control region is a popular marker for within species studies because of its high variability. The length of the avian control region, for example, varies between 1028 and 1581 nucleotides (Boon et al. 2000, Haddrath & Baker 2001). The control region is divided into three domains based on its variability. The central part of the control region is often rather conservative because of its functional importance in regulating transcription and replication (Doda et al. 1981, Brown et al. 1986, Southern et al. 1988, Desjardins & Morais 1990, Clayton et al. 1991,

Cyt b

Cyt b

Cyt b

CR

CR

12s

12s

Nd6

ND6 Nc

Thr

Pro

Glu Phe

Phe

Gl uPro

CR 12sRNAPhe

Thr

Thr

a) typical bird mt gene order

b) Novel bird gene order

d) mammal and mt gene orderXenopus

c) mt gene orderAmazona

Cyt b CR1 CR2 12sNd6 Phe

GluPro

Thr

pGlu

PND

6

Pro

pGlu

Kl

18

L'Abbe et al. 1991). Other conserved sequence regions, specific only for mammals or birds, are also known to occur on control region (Doda et al. 1981, Brown 1986, Southern et al. 1988, Clayton 1991, Ruokonen & Kvist 2002). Mutation accumulation is fastest in the control region next to genes for two tRNAs (Desjardins & Morais 1991, Wenink 1994).

1.3.2 Microsatellites

Microsatellites are mostly neutral, highly polymorphic tandem repeats (repeat unit usually 1 - 6 nucleotides) with different loci and codominant alleles. Sometimes microsatellites are called simple sequence length polymorphins, short tandem repeats (STRs), simple sequence repeats (SSRs) or variable number tandem repeats (VNTRs) (Tautz 1989). The total number of microsatellite loci differs between taxa (Gibbs et al. 1997, Primmer et al. 1998). Microsatellites may be functionally important e.g. in gene transcription, translation, chromatin organization, recombination, DNA replication and DNA mismatch repair system (Li et al. 2002). Therefore, selection may affect SSR variation in some groups of loci (Li et al. 2002). Numerous studies show that their distribution is non-random across coding and non-coding regions (Li et al. 2002). In mammals the most common repeat is GT/CA. Microsatellite mutations are changes in number of repeat units. The most common microsatellite mutation occurs because of slippage of one repeat. Also point mutations can occur either in flanking regions or between microsatellite repeats causing interruption in a microsatellite repeat. Interrupted repeats or repeats, which consist of two or more repeats, have a lower mutation rate because they have difficulties in forming slippage intermediates (Kunst et al. 1997, Petes et al. 1997). Mutation rate estimates from pedigree analyses in humans gave mutation rate estimates of 10-3 event per locus per generation (Weber & Wong 1993). The mutation rate of dinucleotide repeats is estimated to be 5.6 × 10-4; 1.5-2.0 times higher as compared to tetranucleotide repeats (Weber & Wong 1993). The mutation rate of non-disease-causing trinucleotides has been estimated to be intermediate to di- and tetranucleotides. Longer microsatellite repeats tend to be more polymorphic than shorter ones (van Treuren et al. 1997, Primmer et al. 1998, Ellegren 2000). However, Samadi (et al. 1998) have suggested that loci with longer repeat units seem to experience stronger selection against size differences. Loci with lower mutation rates decrease heterozygosity (Amos & Harwood 1998). Mutation rates vary between species, loci and even alleles within the same locus (Jin et al. 1996, Primmer et al. 1998, Nichols et al. 2001).

At a chromosomal level, unequal crossing-over, gene conversion or slipped strand mispairing during DNA replication induce microsatellite variation (Smith 1976, Levinson & Gutman 1987, Jeffreys et al. 1994). There are different kinds of theoretical mutation models to interpret microsatellite allele frequency variation. IAM (infinite allele model) was originally developed for allozymes, but together with SMM (stepwise mutation model), it is the most often used mutation model in microsatellite allele data analysis. Under infinite alleles model every new mutation creates a new allele (Kimura & Crow 1964). In the K alleles model (KAM) there are K allelic states and all alleles have a constant probability of mutating towards any of the previous K-1 allelic states (Crow &

19

Kimura 1970). The K alleles model is more probable mutation model for microsatellite data than the infinite allele model, but only recently has it been used for interpreting microsatellite data (Groombridge et al. 2000, Nichols et al. 2002). In a stepwise mutation model small changes in repeat numbers create new alleles but all alleles have a constant probability of back mutation (Kimura & Ohta 1978). In the two-phase model (TPM), which is a simplified version of the generalised stepwise model (GSM), a limited number of mutations involve several repeats (Di Rienzo et al. 1994, Estoup et al. 2002). The TPM is intermediate to the SMM and IAM. Most microsatellite data-sets better fit TPM than SMM or IAM models (Di Rienzo et al. 1994). Length variability in SSR loci does not follow a simple mutation model (Li et al. 2002).

1.4 Species and populations

The boreal forest landscape is structurally heterogenous and there exist unsuitable breeding or dispersing habitats (Spies & Franklin 1996, Tiebout & Anderson 1997, Eeva et al. 1989, Virkkala et al. 1990). However, all focal species considered herein are able to use the landscape matrix for colonising novel forest areas and they are adapted to natural environmental fragmentation to some degree (Sklepkovych 1997, Reunanen 2001, Selonen 2002). Despite this, their population sizes have rapidly declined in association with environmental fragmentation during the last 50 years. The effect of habitat loss is obvious for the Siberian flying squirrel (Reunanen 2001). In Europe the flying squirrel is listed as a threatened species (IUCN) and as a species of special concern in Finland (EU). Both bird species studied are classified as near threatened (IUCN). The red squirrel, not suffering a decline in population size in Finland, but declining tremendously due to fragmentation in Britain, is herein used as a reference species for the flying squirrel. The three focal species are considered the most sedentary among northern taiga species with restricted dispersal (Virkkala 1991).

1.4.1 The Siberian tit (Parus cinctus)

The Siberian tits range is distributed through mixed coniferous forests dominated by the Scots pine (Pinus sylvestris) up to the timberline in the north from Norway to the Pacific Ocean in Eurasia and Alaska in North-Western of America (Cramp & Perrins 1993, Eeva et al. 1989, Virkkala & Liehu 1990). The morphological variation of the species is rather slight and largely clinal throughout its distribution range. Only four subspecies have been described: lapponicus in Fennoscandia and European Russia, the nominate cinctus from the Urals east through Siberia grading to sayanus and lathami in Alaska (Cramp & Perrins 1993, Harrap & Quinn 1996).

The Finnish Siberian tit population was estimated at 200 000 pairs in 1958, whereas the present population estimate is 40 000 pairs (Merikallio 1958, Väisänen et al. 1998). The probable reason for the population decline has been attributed to modern forestry, although amelioration in climate has also been proposed as a cause (Virkkala 1991, Järvinen & Väisänen 1979). The species has large area requirements making it sensitive

20

to habitat fragmentation: the breeding territories vary from 15-20 ha and 50-100 ha in good and marginal habitats, respectively (Haftorn 1973, Järvinen 1982). The species already suffers from thinning of forests and even on good breeding territories breeding success appears to be dependent on food resources at the end of summer (Virkkala 1990, Veistola & Lehikoinen 1994). Even though competition has been claimed to exist between the Siberian tit and the willow tit in some areas, they form mixed flocks during winters (Cramp & Perrins 1993). In addition, the Siberian tit can breed with the willow tit at the extreme distribution areas in Northern Finland and hybrid offspring of the Siberian tit female and the willow tit male are fertile (Järvinen 1987).

Even though the species is considered mainly as sedentary especially because of its food hiding during the summer, the Siberian tit may occasionally move southwards outside of the breeding season (Cramp & Perrins 1993). Mean natal dispersal of the Siberian tit has been estimated to be 4 km in the Kuusamo population (Orell et al. 1999). However, only 15 long-distance (> 10 km) recoveries have been detected according to the Finnish Ringing Centre. In NE Finland median breeding dispersal accounted only for 100 m in both sexes, while the natal dispersal distance was female biased (range 800-4900 m and 2900-10 300 m for males and females, respectively, Orell et al. 1999).

1.4.2 The Siberian jay (Perisoreus infaustus)

The Siberian jay is distributed from Norway in the west to the Pacific Ocean in the east (Cramp & Perrins 1994). From five to sixteen allopatric morphological subspecies are defined by various authors (Cramp & Perrins 1994). No more than three allopatric subspecies of the Siberian jay have been defined from west of Ural Mountains: nominate P. i. infaustus in Fennoscandia and the Kola Peninsula grading into ostjakorum and ruthenus eastwards. About thirteen allopatric subspecies have been described in Siberia. Altogether sixteen defined subspecies of the Siberian jay group into four main morphological colour lines (Cramp & Perrins 1994). Northern races: infaustus, ostjakorum, monjerensis, bungei and yakutensis show relatively pale colour, a small rufous primary-patch, and a distinct grey patch on tail-tip. Mid-latitudinal races: ruthenus, rogosovi, sibericus and tkachenkoi are typified by relatively dark colour. Southern races: opicus, caudatus and varnak show very saturated body colour with the latter species being almost uniform dark grey. The three races located near the Pacific ocean i.e., sokolnikowi, sakhalinensis and maritimus colour more rufous again.

The decline in population size of the Siberian jay is especially marked in Finland, where it has declined to one third of its original estimated population size since the 1940’s (e.g., Järvinen et al. 1978, Järvinen & Väisänen 1979, Helle 1985, Virkkala 1987, Väisänen et al. 1998). At present ~ 110 000 - 400 000 pairs have been estimated to breed in Fennoscandia, of which about 40 000 pairs are present in Finland (Tucker & Heath 1994, Väisänen et al. 1998). The decline is linked to the loss and fragmentation of suitable habitats because of modern forestry (Järvinen & Väisänen 1978). The species is long-lived and may even reach the age of 20 years (Lillandt 2001) having an average generation length of 4 years (II). Families form flocks with 3-5 juveniles that may accept juvenile migrants in winter (Ekman et al. 1994).

21

The Siberian jay is a proto-cooperative (i.e., prolonged parental investment without helping behaviour) species with delayed dispersal (Ekman et al. 1994, Sklepkovych 1997). In Fennoscandia the species is sedentary after natal dispersal, whereas in Siberia occasional winter movements have been observed (Cramp & Perrins 1994). Mean natal dispersal distance has been estimated to be 2.4 km, other dispersal distances being 1 km the record being 290 km (Haartman von et al. 1963, Lindgren 1975, Savolainen 1992, Lillandt et al. 2002, Finnish Bird Ringing Centre).

1.4.3 The Siberian flying squirrel (Pteromys volans)

The Siberian flying squirrel (hereafter flying squirrel) is distributed throughout mixed, spruce deciduous forests of Eurasia (Reunanen 2001, Selonen 2002). Its distribution extends from Finland and the Baltic countries in the west to the Pacific Ocean in the east and the Islands of Sakhalin reaching its southernmost distribution on the Island of Hokkaido (Ognev 1940). Three allopatric morphological subspecies of the flying squirrel have been defined: P. volans volans and ognevii in Eurasia and orii on the Island of Hokkaido (Ognev 1940). In Finland, the species range extends to the southern parts of Lapland, where its distribution is rather discontinuous, although large gaps also determine its occurrence elsewhere (Reunanen et al. 1998, Ympäristöministeriö 2001). Its distribution has become even more patchy as a result of environmental fragmentation (Reunanen 2001, Selonen 2002). In Russia, the flying squirrel may reach more northern latitudes than in Finland along riversides (Ognev 1940, Reunanen & Nikula 1998).

The population size of the flying squirrel has strongly declined in Finland since the 1950's (Hokkanen et al. 1982). For the last 5-17 years alone, the population has been estimated to have declined 20-38 % (Ympäristöministeriö 2001). The lowest present day population size estimate for the species is 23 200 (14 500-32 000) individuals in Southern Finland. However, exact population size estimates do not exist. Changes in the tree species composition and a decline in the amount of dead trees in forests are considered to endanger the species' survival (Rassi et al. 2000). Increasing amount of open habitat patches at local scale decreases the species probability of occurrence (Mönkkönen et al. 1997). However, the species may prefer certain types of edges (Desrochers et al. 2003).

The average home range size is 59.9 ha for males and 8.3 ha for females (Hanski et al. 2000). The Siberian flying squirrel is promiscuous with both males and females often copulating with several mates (Selonen et al. 2002). The average natal dispersal distance is 2.5 km (Selonen & Hanski 2003), with some individuals, most often females, dispersing distances that may reach over eight kilometres (Selonen & Hanski 2001). Short-distance dispersers select random directions, but long distance dispersers select spruce oriented directions (Selonen & Hanski 2004). The species is flexible in its dispersal and dispersing individuals can move also across small areas of open habitat (Selonen 2002). Nocturnal movements of females can cover areas from ten hectares to two square kilometres (Hanski et al. 2000, Reunanen et al. 2000).

22

1.4.4 The red squirrel (Sciurus vulgaris)

The red squirrel is distributed throughout in the Palaearctic from Portugal in the west to China and Siberia in the east, and from the Mediterranean Sea in the south to Fennoscandia in the north (Gurnell 1987). More than 40 allopatric subspecies have been ascribed. The species has wide habitat preferences in coniferous forests and its distribution is continuous in Finland. In Britain the species is distributed in fragmented manner and classified as endangered (Gurnell 1987, IUCN). In Finland, the species is common and has not suffered a noticeable change in population size.

2 Questions in focus

This Ph.D. thesis has arisen from a special interest towards ecology and evolution of boreal taiga species at the Department of Biology at the University of Oulu. A great deal of the previous work has focused on Parus species. The Siberian tit, the Siberian jay and the Siberian flying squirrel were selected as study objects because of their strong population size declines suggested to be associated with the loss and fragmentation of suitable forest habitat. It is therefore natural that I compare results obtained from my focal species to the published data of other Parus species and the results from the Siberian flying squirrel study to that of a reference species, the red squirrel.

I focused on the following questions:

1. Do the neutral markers show population genetic structure in the species studied? 2. Do signs of isolation or decline in population size exist? 3. Most of the populations sampled were from Fennoscandia, but I was lucky enough to

get samples from some populations in Siberia. What is the Pleistocene history of the species studied in Fennoscandia and how does it differ from the Pleistocene history of the species in Siberia?

4. What kind of conservation genetic conclusions and recommendations can be drawn from the results?

3 Materials and methods

3.1 Molecular methods

DNA was extracted from blood, muscle, skin, feather, hair and even bone samples using phenol chloroform, chelex and similar methods with some modifications (Sambrook et al. 1989, Walsh et al. 1991, Lillandt et al. 2001). MtDNA extraction (Tamura & Aotsuka 1988) was done in the beginning of laboratory work of each project to assure the mitochondrial origin of the control region sequence. Sequencing was performed and allele sizes were scored with ABI377. Populations sampled are shown in table 1. Detailed information about the PCR conditions, the primer sequences and the samples are described in the original papers (I-IV).

Table 1. Sample sizes and populations sampled in the species studied. ms = microsatellite, mtDNA = mitochondrial DNA

Sample size Populations or areas sampled The Siberian tit 56 (mtDNA) Fennoscandia: Kuusamo, Kemijärvi, Sodankylä,

Rovaniemi, Hetta 3 (mtDNA) Yakutsk The Siberian jay 101 (mtDNA) Fennoscandia: Kainuu, Kuusamo, Pomokaira, Blåkölen,

Granlandet, Arvidsjaur, Åmsele, Kristiinankaupunki Siberia: Yenisej, Taimyr, Yakutsk, Magadan

124 (ms) Fennoscandia: Kainuu, Pomokaira, Blå-kölen, Granlandet, Arvidsjaur, Åmsele, Kristiinankaupunki

The Siberian flying squirrel 125 (mtDNA) Finland: Kokkola, Luoto, Pietarsaari, Vaasa, Alavus, Turku, Nuuksio, Kausala, Anjalankoski, Karjala, Savo

The red squirrel 36 (mtDNA) Finland: Oulu, Kuopio, Helsinki

25

3.2 Mitochondrial sequence diversity indices

Most sequences were aligned using the interfaces of the GCG – program package provided by the Centre for Scientific Computing in Finland (Wisconsin Package Version 9.1). The most simple method to estimate the level of genetic variation is to calculate the number of segregating sites (S) and the average number of pairwise nucleotide differences (k) (Watterson 1975). The number of polymorphic nucleotide sites per nucleotide is calculated from: θS = S/a1, where a = 1 + ½ + 1/3 + … (n-1) - 1 (Nei 1987). Haplotype diversity and nucleotide diversity were calculated from equations h = (n/(n-1))(1-Σfi) and π = (n/(n-1))Σxixjπij, where fi is frequency of ith haplotype, xi is the population frequency of the ith type of DNA sequence and πij is the proportion of different nucleotides between the ith and jth types of DNA sequences (Nei 1987). Tajima’s D, from equation: D = (π -θS)/sqrt(varπ – varθS), tests whether the marker used is under natural selection or whether there have been drastic changes in the demographic history of a population (Tajima 1989 a, b, Aris-Brosou & Excoffier 1996).

Tukey-Kramer test (Sokal & Rohlf 1985) was used to test among population differences in sequence diversity estimates to find out possible population dynamic changes. The average nucleotide diversity (π) for sequence data is the most stable sequence diversity estimate (Nei 1987). Theta per site (θS) is sensitive for growth in population size and current population size (Nei 1975, Tajima 1989b). Only very strong and long-lasting bottlenecks affect theta (Tajima 1989b). Population bottlenecks and the original population size affect mostly to the number of the pairwise genetic difference (k) (Tajima 1989b). Significant difference between populations only in: 1) nucleotide diversities indicates long-lasting difference in population demographic conditions and limited migration between them, 2) numbers of thetas indicates increase in size in another of the populations, 3) k’s shows differences in founder number among populations or bottlenecks in population sizes. If significant differences exist in nucleotide diversities among certain populations, no further conclusions on population demographic changes can be derived despite of significant differences in thetas and k’s among those populations. When arranging significant differences between populations in a pairwise manner into diagonal matrices and comparing raw values from sequence diversity tables on significant differences, possible relative population dynamic changes or differences in founder numbers can be readily checked.

The pairwise genetic distance describes the number of nucleotide substitutions per nucleotide site between two sequences. Transitions occur more easily than transversions and mutational bias can be corrected in several ways. Some phylogenetic trees and estimates of common ancestry rely on the pairwise genetic distance (Kumar et al. 1993). The distribution of pairwise genetic distances includes information about the past demographic history and the coalescence time in a population. The observed distribution was compared to the expected under the expectation of sudden population growth (Rogers & Harpending 1992, Rogers 1995).

The divergence of populations (ΦST) was estimated by the analysis of molecular variance (AMOVA), where both haplotype diversity and pairwise genetic distance between haplotypes was taken into consideration (Excoffier et al. 1992). A hierarchical approach in analysis of molecular variance was adopted for the Siberian flying squirrel,

26

where inland and coastal populations were pooled as their own groups. The number of effective migrants (Nmf) was estimated from ΦST = 1/(2Nmf +1).

Long-term effective population size was calculated after the following expression: Ne = 106tx/g, where g is the mean generation time, tx = 0.5dx and dx is the mean pairwise genetic distance among individuals (Wilson et al. 1985). The short term effective population size assuming coalescence is estimated from the equation: θ = 2Neµ for haploid sequence data, where µ is mutation rate used. I have corrected this equation with generation time in a similar fashion to that of Wilson and co-authors (1985).

The mutation rate of 2 % for control regions of the two bird species studied is widely used conservative estimate (Klicka & Zink 1997) and provides a means for comparing results to other studies (II). The mutation rate estimate for the Siberian flying squirrel studies was 2.33%, which was obtained by comparing Tamura-Nei(gamma) distances between the sequences for 12srRNA (Oshida et al. 1996) and the control region of the Siberian flying squirrel and the red squirrel.

3.3 Microsatellite allele diversity parameters

The most common parameters calculated from microsatellite data are the mean number of alleles, allelic richness and heterozygosity. Heterozygosity (h) is calculated from: h = 1 - m∑i=1xi

2, where m is the number of alleles xi is the population frequency of the ith allele at a locus. Expected heterozygosity is the probability that two randomly chosen alleles are identical and it is calculated from: H = (n/(n – 1))(1 - kΣi = 1 pi

2), where n is the number of gene copies in the sample, k is the number of haplotypes and pi is the sample frequency of the i-th haplotype.

F-statistics was calculated after Weir and Cockerham (1984). Significance tests were calculated for allele frequencies (assuming Hardy-Weinberg) of each pair of populations using 1000 permutations.

4 Results and discussion

4.1 Phylogeography in Siberia

The genetic structure of colonizing populations of organisms is generally shaped as the result of their Pleistocene histories (Hewitt 1996, Merilä et al. 1997, Lucchini & Randi 1998). Associations between genealogies and geography among postglacial populations arise from the extent of e.g., population expansions, vicariant events, differing dispersal abilities and refugial isolation (Hewitt 1996, Ibrahim et al. 1996). Comparatively, no association between genealogy and geography would be expected if a species’ whole range were colonized from a randomly mating source (Barton & Wilson 1998). Different degrees of spatial patterns and concordance among loci arise from restrictions in gene flow (Barton & Wilson 1998).

Previously glaciated regions and past tundra in Northern Siberia are occupied by fauna and flora which have extended their distribution range since the Pleistocene. In evolutionary terms, the fauna and flora in northernmost areas of Siberia is even younger than in Fennoscandia since pine has extended its range there as late as 4000-5000 years ago (Kremenetski et al. 1998). The northernmost populations would be expected to have reduced levels of genetic variation when compared with their southern counterparts in the same way as shown for many species in Europe and North America (Hewitt 1996, Merilä et al. 1997, Klicka & Zink 1997, Palo 2003). However, levels of genetic variation in Siberian populations may be affected by admixture among populations expanding from multiple refugia south of the ice sheets or from Beringia (e.g., Fedorov 1998, Kvist et al. 1999b, 2003 a, b). Pine, for example, has had several refugia during the Late Pleistocene (Fig. 4, Kremenetski et al. 1998). However, the Scots pine (Pinus sylvestris) has had only one refugium at the River Yenisei.

For each of the taiga species studied, within species genetic differentiation reflect to some degree the morphological differentiation throughout their Palearctic distributions (I, II, III, IV). Siberian tits from Yakutsk did not differ from their Fennoscandian conspecifics in terms of mtDNA diversity (I). There were three mtDNA lineages across the distribution range of the Siberian jay (II, III). In the Siberian flying squirrel, there were at least seven deep mtDNA lineages across its distribution range (Fig. 5-6). The

28

Siberian tit has only four morphological subspecies, the Siberian jay sixteen, the Siberian flying squirrel three (but nine different fur morphs of the flying squirrel have been described) and the red squirrel has altogether 40 subspecies (Ognev 1940, Gurnell 1987, Cramp & Perrins 1993, 1994). Natural selection caused by ecological factors can cause rapid morphological divergence (Orr & Smith 1998). Morphological subspecies may sometimes be distinguished based on neutral genetic variation, but this can not be considered a general rule since genes affecting morphological evolution may evolve faster than neutral markers (Seutin et al. 1995, Merilä et al. 1997, Lucchini & Randi 1998).

Fig. 4. The continental ice (paler grey) at its maximum 20 000 yr ago (Andersen & Borns 1997). Some pine and spruce refugia in the old continent (Lang 1992, Kremenetski et al. 1998, Willis et al. 1998). The Norway spruce originates from the Lake Baikal.

Panmictic population structure would be expected if the species was nomadic (Seutin et al. 1995). The lack of genetic divergence between P. c. lapponicus and P. c. cinctus subspecies of the Siberian tit in Fennoscandia and Yakutsk was a surprise since the

Magadan

Caspian Sea

?

Taimyr

BaikalYenisei

Yakutsk

Kamchatka

Sakhalin

Amur

Nagano

Black Sea

29

species is considered sedentary. The lack of genetic divergence is most probably the result of serial bottlenecking followed by subsequent growth in population size (I). The geographical distance between the two Siberian tit subspecies is more than 5000 km. Cytochrome b sequence from Yakutsk is similar to the Fennoscandian sample supporting recent divergence of these populations.

Table 2. Tamura-Nei pairwise genetic distances (S.D) among Siberian jay mtDNA lineages.

Yenisei lineage Far-East lineage Fennoscandian lineage 0.0215 (0.0055) 0.0197 (0.0048) Far-East lineage 0.0116 (0.0035)

Pairwise genetic distances of the three mtDNA lineages of the Siberian jay across its distribution range are shown in Table 2. The Fennoscandian lineage is formed by the subspecies P. i. infaustus, Yenisei lineage by P. i. rogosovi and the Far-East lineage by the three morphological subspecies P. i.monjerensis, maritimus and yakutensis, which group to the same phylogenetic branch in the Neighbor-Joining tree (III). The three different mtDNA lineages have had separate and different sized refugia probably located at the same sites than the pine refugia (Fig. 4). P. i. infaustus may have colonised Fennoscandia from a refugium close to the Black Sea, and P. i. rogosovi northern Siberia from the refugium of the Pinus sylvestris at the River Yenisei. In addition, the River Yenisei is considered as a vicariant factor in Siberia (Cramp & Perrins 1994). The close genetic relationship of the three subspecies and the ragged shape of the observed distribution of the pairwise genetic differences in the Siberian jay in Far-East (III) indicate a common and large refugium for these subspecies during the Pleistocene glaciations (III). It is impossible to define the location of a refugium for the Far-East population, since several appear most likely. Fossil data shows that the Siberian jay has extended in range as far as the present-day Bangladesh area (Tyrberg 1998). The three Siberian jay mtDNA lineages could be dealt as separate ESUs.

30

Fig. 5. The consensus of Neighbor-Joining trees based on Tamura-Nei distances and 1000 bootstraps of the Siberian flying squirrel. ‘The circle’ of the NJ tree is formed by the Finnish samples and the two branches by Siberian samples.

0.01

BaikalAmutkan Nfzom Hot

Nizhnij NovgorodN. TunguskaYenisei

TunguskaN. TunguskaSakhalin

Lake Boloni

SakhalinSikhote

Alin

Sikhote

Alin

Plesheevo Lake

Tuzukhan

ski distr.

KostromaSikh

ote Alin

Anadyski Kzaj

Altai, Teleotskoe Lake

Altai, Teleotskoe LakeAltai, Teleotskoe LakeBolshoi, Shantar

Mosc. ZooKajaaniIisalmi

Troitskii distr.

31

Fig. 6. A minimum spanning network of the flying squirrel based on mtDNA sequences. Line between each tick mark represents a point mutation.

Despite the fact that only three morphological subspecies have been defined in the flying squirrel, several mitochondrial lineages exist in Siberia (Fig. 5-7). However, nine different ‘fur races’ have been described for the flying squirrel across its distribution range (Ognev 1940). Unfortunately, sampling has been too scarce to draw detailed conclusions regarding its evolutionary history. There are two Siberian branches in the Neighbor-Joining tree of the flying squirrel (Fig. 5). The haplotypes in Fig. 6 show several deep, old lineages on the one hand, and several rather new lineages, which differ from the main F1 haplotype only by some point mutations, on the other. This type of structure might be due to a history of cyclic fluctuations in population size and range, occasional migrations and colonizations of new habitats from several refugia during Pleistocene interglaciations. The minimum spanning network suggests that there have been several refugia in the flying squirrel during the Pleistocene. The deepest branch is the haplotype from the Lake Baikal area, which makes sense, since the original home of the Norway spruce and therefore, also the Siberian flying squirrel is expected to be in the Baikal area. AMOVA shows that 13.6 % of genetic variation in the flying squirrel in Russia is accounted by differences among five geographical areas: Moscow, Altai,

Kzashojazski K

zaj.

Nizhnij

. Novgo

rod

Miznoe Enisej

Khabazovski Kzaj

Anadyski Kzaj

Sakhalin

Altai Te

leotsk

oe

Moscow

R. Sheksna Rybinsk

Ivanovo

Sakhalin

Lake BoloniN. Tunguska

N. Tunguska

Pechovski

Vladimiz

Coastal Finland

Coastal Finland

Coastal Finland

Coastal Finland

Coastal Finland

Coastal Finland

Coastal Finland

Inland FinlandInland FinlandInland Finland

Inland Finland

Tung

uskaAnad

yski

Kzaj

Tver

Anadyski Kzaj

Kamchatka

Vadimiz

AnadyskiKzaj

Miznoe Enisej

Khabazovski Kzaj

N. Tunguska

Altai Teleotskoe

Sayanu Gintaza

Inland Finland

Enisej

32

Yenisei, Kamchatka and Sakhalin (Fig. 7). There are indications of admixture from several refugia in the flying squirrel in Siberia (Fig 7). The population genetic structure of the species studied is consistent with the multirefugia model.

Fig. 7. The minimum spanning network of the flying squirrel in Russia and Finland. Circled areas denote to the geographic groups in AMOVA: Moscow, Altai, Yenisei, Sakhalin and Kamchatka.

Recently deglaciated areas, such as Fennoscandia, would be expected to be characterised by lower levels of genetic variation as compared to unglaciated areas because they are usually founded by fewer numbers of individuals, or because of the influence of leading edge dispersal (Nei 1975, Hewitt 1996, 1999). Pairwise comparisons of populations (Fig. 8) can reveal such differences as well as relative differences in population growth rates. Nevertheless, sample sizes from Siberia were too small to perform pairwise comparisons among populations. Therefore, I studied relative differences in founder numbers and population growth rates in the willow tit and the great tit populations in Siberia published

33

by Kvist and co-authors (2001, 2003b). Population pairwise comparisons for sequence diversity indices (Table 3) did not indicate growth in any of the willow tit populations in

Fig. 8. Tukey-Kramer test results of pairwise comparisons of sequence diversity estimates among populations (Table 3.) of the willow tit populations in Kvist et al. (2001). Uppermost matrix shows asterisks denoting significant differences in nucleotide diversities, middlemost in theta per site and lowermost in pairwise genetic differences among populations studied. Experiment wise error rate was set at α = 0.05.

relation to the other populations studied (Fig. 8.). This result most probably stems from a large, long-term, effective population size in the willow tit (Kvist et al. 1998). The two marginal willow tit populations, Tovetorp and Latvia showed significantly smaller founder number as compared to other populations. The high founder number in the Alpian population may result from high original founder number or Pleistocene colonisation from the two spruce refugia, from the Alps and the Balkans. When performing similar comparisons for great tit populations across its distribution range to

OuluTovetorp

AlpsLatvia

Southern UralsMagadanAmur

KamchatkaNgano

Oulu

Tove

torp

Alp

sLa

tvia

Sout

hern

Ura

lsM

agad

anAm

urKa

mch

atka

Ngan

o

Sakhalin

Sakh

alin

Oulu

Alp

sLa

tvia

Amur

Ngan

o

OuluTovetorp

AlpsLatvia

Southern UralsMagadanAmur

KamchatkaNganoSakhalin

Sakh

alin

Sakh

alin

Tove

torp

Sout

hern

Ura

lsM

agad

anKa

mch

atka

OuluTovetorpAlpsLatvia

Southern UralsMagadanAmur

KamchatkaNgano

Sakhalin

Oulu

Tove

torp

Alps

Latv

iaSo

uthern

Ural

sM

agad

anAm

urKa

mch

atka

Ngan

o

34

that of Kvist and co-authors (2003b), I found the opposite result. None of the populations studied showed significant difference in founder number but some populations like Harjavalta in Finland and Estonia, showed an increase in size in relation to other populations. This is astonishing since I expected to find differences in founder numbers between populations in the great tit as in the willow tit. The great tit has expanded its range northwards during recent decades and its population has shown an increase in size as a result of human influences (Kvist et al. 2003b).

Table 3. Nucleotide diversities and thetas as per cents and average number of nucleotide differences (k) in the willow tit populations (sequences published by Kvist et al. (2001)).

Population N π (S.D.) % θ (S.D.) % k (S.D.) Oulu 13 0.620 (0.111) 1.036 (0.446) 3.667 (0.773) Tovetorp 12 0.397 (0.077) 0.616 (0.291) 2.348 (0.554) Alpit 8 0.767 (0.218) 1.044 (0.506) 5.833 (1.132) Latvia 8 0.387 (0.076) 0.457 (0.250) 2.381 (0.581) S. Urals 13 0.677 (0.104) 0.927 (0.404) 3.846 (0.648) Magadan 7 0.806 (0.158) 0.967 (0.489) 4.762 (1.317) Amur 4 0.592 (0.113) 0.646 (0.403) 3.500 (1.500) Kamsatka 7 0.419 (0.110) 0.414 (0.239) 2.800 (1.097) Nagano 7 0.612 (0.112) 0.622 (0.334) 4.762 (1.392) Sakhalin 4 0.733 (0.179) 0.738 (0.453) 4.333 (1.817)

4.1.1 Coalescence time estimates

The coalescence time for the dunlin, tit and Siberian jay populations have been estimated using mtDNA sequence data from pairwise genetic distances (Wilson et al. 1985, Wenink et al. 1993, Kvist 2000, I, II) although today, the tau of distribution of pairwise genetic differences is more often used (Rogers & Harpending 1992, Rogers 1995, Hundertmark et al. 2002, Liebers & Helbig 2002). The most recent common ancestor of all Siberian tit sequences would have occupied 230 000 years ago based on the tau (I). This differs from the coalescence time estimate derived from the pairwise genetic distance (I), due to the difference in time scales in two estimates. Tau and the phylogenetic tree (Rogers & Harpending 1992, Rogers 1995, Nichols 2001) measure time in generations whereas pairwise genetic distance is directly translated into years without taking generation time into consideration. Using tau for the Siberian jay provides an estimate of 2.6 – 4.3 million years as the time to the most recent common ancestor. The coalescence time estimate for the Siberian jay depends on the way tau is calculated. The larger estimate considers variation in two lineages whereas the smaller estimate is simply calculated based on three sequences from different lineages. Mismatch distributions are based on Tamura-Nei pairwise genetic distances and have been obtained by randomising sequences with 1000 bootstraps. The larger estimate also takes into consideration sample variation. The most recent common ancestor of all Siberian flying squirrels (Fig. 5 - 7) would have occurred 380 000 years ago. Most probably, the mutation rate estimate of control region I in the

35

Siberian flying squirrel is nearest to being correct. However, I think that mutation rate estimates should be corrected for generation time to gain a gene specific mutation rate. All mammals have the same number of cell divisions during their lifetime (Eckert et al. 1988). Unless mutation rate, which is obtained by calibrating substitution rate to some geographic vicariant phenomena or fossil record, is not corrected with generation time, it represents a species specific mutation rate.

4.2 Colonisation of Fennoscandia

The Holocene, being the latest interglacial period, began some 11 000 years ago with the receding continental ice over Fennoscandia. Scots pine and Norway spruce were among the first coniferous trees to arrive in Fennoscandia some 8 000 and 5500 years ago, respectively (Aartolahti 1966, Tolonen 1983, Hyvärinen 1987), during which time the first land vertebrates also colonised the area. In evolutionary terms, the Fennoscandian fauna is young and also typified by fast colonisation. This kind of population demographic history (a bottleneck followed by sudden expansion) can be seen as a star-like minimum spanning network, significantly negative Tajima’s D, and the distribution of the pairwise genetic differences, which follows that of an expanding population, and where the left shoulder of the curve is short and rising (Tajima 1989a, Rogers & Harpending 1992, Rogers 1995). The greenfinch (Carduelis chloris), the grey partridge (Perdix perdix), the great tit and the blue tit (Parus caeruleus) in the north, for example, show all of the previously mentioned characteristics in their mitochondrial genome (Merilä et al. 1997, Kvist et al. 1999a, b, Liukkonen-Anttila et al. 2002).

Significantly negative Tajima’s D values and distributions of pairwise genetic differences that follow expanding populations were characteristic for the Siberian tit and the Siberian jay in Fennoscandia (I, II, III). A significantly negative value usually indicates selection, but it may also be an outcome reflecting growth in population size (‘sudden expansion’) after a bottleneck (Tajima 1989a). However, changes in a population’s demographic history do not exclude selection. MtDNA is inherited together with the maternal sex chromosome, which may be more strongly selected than autosomal chromosomes. Also hitchhiking or backround selection may affect the pattern of genetic variation in mtDNA. Both avian species studied carried in their mtDNA signs of fast population growth in their histories (I, II, III). Tajima’s D did not deviate from expected neutrality for the Siberian flying squirrel (IV). The species’ mating system may be reason for the low level of mtDNA variation and non-significant Tajima’s D. Sometimes Tajima’s D may lack the power to detect selection (Fu 1997). More sensitive Fu’s FS gave significant deviation from expected neutrality for the local Siberian flying squirrel population in Kausala and also for the whole Fennoscandian population.

The high genetic differentiation (ΦST = 53 %) among the flying squirrel populations in Finland was a surprise. The Siberian flying squirrel was a little different with respect to its mtDNA: two closely related lineages, both of which, however, showed star-like structure, non-significant Tajima’s D and the observed pairwise genetic distribution which did not follow that of an expanding population (IV). The flying squirrel population has not experienced a marked increase during its colonization history. The topology of

36

the minimum spanning network, with two geographically distinct branches which cluster below and above the ancient shore of Littorina Lymnea stage of the Baltic Sea, suggested that the species experienced some kind of isolation resulting in two mtDNA lineages during the Late Pleistocene or its colonisation of Finland. Twenty-eight percent of genetic variation could be explained by differentiation among these two groups. It seems most likely that the Siberian flying squirrel colonised ‘inland’ Finland directly from the east following colonisation by the Norway spruce. Western, coastal Finland was probably colonised via the Karelian Isthmus in association with Norway spruce as the old sea bottom lifted. The large lake area in Central Finland, which was even larger at the time of the colonisation, prevented distribution of the F8 haplotype further to the west. The common shrew (Sorex araneus), the field vole (Migrotus agrestis) and the weevil (Otiorrhynchus scaber) show similar structuring in their genome supporting a two-way colonisation history already proposed in the 1950’s prior to the advent of genetic studies (Saura et al. 1976, Halkka et al. 1995, Jaarola & Tegelström 1995).

In Europe, the Pleistocene history of the red squirrel population appears different from previously mentioned species (IV). The red squirrel carries in its mtDNA signature of a large long-term effective population size (IV) similar to another boreal species, the willow tit (Kvist et al. 1998). The minimum spanning networks of both of these species are netlike and their distributions of pairwise genetic differences are bell-shaped suggesting that their population sizes were large long before they colonized Fennoscandia.

Results suggest that the Fennoscandian populations of the two avian species survived in a single refugium during the Late Pleistocene (I, II). Possible colonization sources of the Fennoscandian avian populations have been pine refugia either at Caspian Sea or to the south-west in the upper part of the Irtysh River valley. The most recent common ancestor of the flying squirrel could be dated to 166 000 years ago (IV). The Finnish population of the flying squirrel has undergone some kind of differentiation during Late Pleistocene or during its colonization history (IV). It is possible that there have been two close refugia for the Norway spruce in Russia in the same way as in Middle Europe (Lang, 1992). Possible locations of the two suggested refugia have been the upper part of the Irtysh River valley or the River Yenisei or between these two areas. A shared Pleistocene history of Finnish and British red squirrels was evident (IV).

4.3 Gene flow and dispersal studies

Gene flow estimates can be derived from FST. These estimates, which describe the number of effective migrants between populations per generation, are based on the assumption that populations are in equilibrium (Whitlock & McCauley 1998). Deviation from expected neutrality, estimated by e.g., Tajima’s D, indicates a deviation from equilibrium for the species studied. The flying squirrel fulfils the expectation of genetic equilibrium, whereas both of the avian species and the red squirrel show deviations. However, the present day population genetic structure is a result of a history of hundreds or thousands of generations. Slatkin & Barton (1989) have shown that FST is rather robust from deviations of the expected equilibrium, and it is not greatly affected by recent

37

demographic events. That is why I derived gene flow estimates from FST. I also used assignment studies (Pritchard et al. 2000) to estimate nuclear gene flow of the Siberian jay in Fennoscandia.

Significant differences in within population nucleotide diversities in the Siberian tit indicated some restrictions in gene flow (I). The Siberian tit population close to the Finnish-Russian border showed the highest levels of genetic variation as compared to other populations (I). This suggests a ‘source-sink’ metapopulation structure of the species, where Kuusamo serves as a source population. Decline and fragmentation of the Siberian tit population may be too recent to detect genetic effects like e.g., for the white-backed woodpecker (Dendrocopos leucotos) which is a highly endangered species in Fennoscandia (Ellegren et al. 1999). In Sweden, only three white-backed woodpecker populations are known to exist despite random mass invasions from Russia. A decline in population size during 10 generations, the species has not been affected regarding genetic structuring nor has it shown bottleneck signals in Fennoscandian and Baltic populations (Ellegren et al. 1999).

The mtDNA study of the Siberian jay showed that generally there were over four migrants per generation between populations (II). Significant population genetic structuring can be found in organellar genes if the effective gene flow is less than four individuals per generation (Birky et al. 1983). Denser sampling from larger amount of populations increased the rate of observed significant differentiations among study populations (III). Species distribution, high φST, FST, and assignment studies based on microsatellite alleles showed that the Siberian jay population is significantly genetically structured in Fennoscandia supporting the existence of a type of metapopulation which contains several central, well-connected populations and some distant, nearly isolated populations (III). Edge populations like Kristiinankaupunki and a suggested sink population – Blåkölen – had smaller level of genetic variation compared to other, more central populations. Males spread genes more effectively than females in the Siberian jay (III).

Genetic analyses suggest minimal effective gene flow in the flying squirrel, although ecological studies suggest that it is rather mobile species in terms of its dispersal ability (Hanski et al. 2000, Selonen et al. 2001, Selonen 2002, Selonen & Hanski 2003, IV). Small nucleotide diversities with small variances within populations support a leptocurtic colonization model, where a small portion of flying squirrel females have set up pocket populations ahead of the colonising front (IV). Some individuals, most often females, disperse distances of up to eight kilometers and they tend to select spruce-dominated directions (Selonen & Hanski 2001, 2002). Several female territories (having ~ 8.3 ha home ranges) can exist within one male territory (59.9 ha) (Hanski et al. 2000). To avoid incest females should disperse longer distances. Isolation-by-distance structure in nuclear microsatellites supports stepping-stone gene flow, where nuclear gene flow occurs only among adjacent populations (Selonen 2002). Local founder events and the formation of pocket populations, connected to present-day enforced isolation and genetic drift can explain the low level of within population genetic variation (Ibrahim et al. 1996). Intensive agriculture, industry and roadway traffic have led to the decline of suitable flying squirrel habitat particularly in the southwestern and coastal Finnish regions. Most probably, current gene flow is prohibited in this part of Finland. Five different areas based on significant differences in haplotype frequencies indicate short-term

38

demographic independence of females in these areas. The strong genetic subdivision among flying squirrel populations results from its colonisation history, territorial habits, and both past and present increases in patchiness regarding suitable flying squirrel habitat. Although the distribution of flying squirrel habitats is already naturally patchy, recent anthropogenically increased fragmentation has led to further loss of suitable habitats (Ympäristöministeriö 2001).

The Finnish red squirrel population exhibits greater genetic variation than the British population which may be due to a marginal distribution and a decrease in the population sizes in Britain (Barratt et al. 1999, IV). Gene flow between British populations is minimal or nonexistent (Barratt et al. 1999). Population divergence of the red squirrel in Finland was low, FST ≈ 5% indicating rather intensive gene flow (IV). The high amount of genetic variation in Oulu, as compared with other populations may indicate several colonisation directions: from east, south-east and from Sweden in the west.

Most passerine species have female biased natal and breeding dispersal, whereas most mammals exhibit male biased dispersal (Greenwood et al. 1980). Therefore, it was not unexpected to find panmictic population in the Siberian tit when assessing genetic structure using mtDNA (I). The Siberian jay actually shows marked structuring in mtDNA. Interestingly, Siberian jay males mediate gene flow more effectively when compared to Siberian jay females (III), which is opposite to the preliminary expectations of Greenwood et al. (1980). Also the flying squirrel shows contrasting results to the preliminary expectations when considering female biased gene flow. It may be that harsh winter conditions inflict stronger selection pressure on taiga species when compared with species belonging to same genus or taxa inhabiting more temperate regions.

Isolation due to habitat fragmentation will effectively reduce gene flow between populations (I, II, III, IV). A short natal dispersal combined with adult site-fidelity can facilitate geographical structuring. Most probably gene flow is totally prohibited between flying squirrel populations in coastal Finland, whereas inland populations may still be connected to one another by gene flow (IV). Both the Siberian jay and the flying squirrel are year-round, territorial species which may be one reason for their observed population genetic structures. The Siberian tit is territorial only during the breeding season. The Siberian tit may form mixed flocks during winters and is more mobile compared to the two other taiga species studied. The general finding suggests that species with more restrictive habitat requirements and dispersal potential exhibit stronger population genetic structure. The differences in genetic structuring in the three species studied appear clearly related to species’ differences in habitat use, ecology and colonization and/or evolutionary history.

4.4 Effective population size

Effective population size measures the amount of available genetic variation and determines a species ability to adaptively cope with natural selection. It defines the effective number of individuals making a contribution to the next generation and helps to predict fixation probability of favourable and deleterious alleles (Crow & Kimura 1970). Therefore, it determines the effect of both inbreeding and genetic drift (Falconer 1989,

39

Caballero 1994, Nomura 2002). Several different effective population size exists: inbreeding, variance, mutation, metapopulation, long-term and short-term effective population sizes (Crow & Kimura 1970, Barton & Whitlock 1997, Pannell & Charlesworth 1999). The effective size of a population depends on various demographic and ecological variables, such as mating system, deme size and dispersal (Nunney 1991, 1993). Effective population size can be determined from either ecological or, more indirectly, genetic data.

Table 4. Thetas (Nes = θ/ (2µ)) of the species studied. Overall short-term effective population size is derived from theta.

Species Population Theta (S.D) Fennoscandia 0.005 (0.002) Kuusamo 0.005 (0.002) Kemijärvi 0.002 (0.001) Rovaniemi 0.002 (0.001) Hetta 0.002 (0.002)

The Siberian tit

Sodankylä 0.002 (0.001) Fennoscandia 0.009 (0.003) Suomussalmi 0.003 (0.001) Kemijärvi 0.002 (0.001) Pomokaira 0.004 (0.002) Blåkölen 0.003 (0.001) Granlandet 0.003 (0.001) Arvidsjaur 0.003 (0.002) Åmsele 0.003 (0.002)

The Siberian jay

Kristiinankaupunki 0.002 (0.001) Finland 0.005 (0.002) three northernmost 0.000 (0.000) Vaasa 0.001 (0.001) Alavus 0.002 (0.001) Turku 0.001 (0.001) Nuuksio 0.002 (0.001) Kausala 0.001 (0.001) Anjalankoski 0.002 (0.005) Karjala 0.002 (0.002)

The Siberian flying squirrel

Savo 0.003 (0.002) Finland 0.080 (0.025) Oulu 0.076 (0.031) Helsinki 0.048 (0.021)

The red squirrel

Kuopio (Savo) 0.050 (0.020) Long-term and short-term effective population sizes are usually estimable from variation in neutral genes, and long-term effective population size refers to the number of generations having passed since two randomly picked genes had a common ancestor (Tajima 1983). Long-term effective population size is calculated from the mean pairwise

40

genetic distance of populations from neutral sequence data (Wilson et al. 1985). Short-term effective size of a population refers to the ideal size of the population assuming coalescence, ‘backward temporal perspective of gene genealogies, where lineages collapse or coalesce at nodes’ (Harding 1996). It seems a more proper measure of Ne for recently fragmented populations although Tajima (1989) has shown that a pairwise genetic difference is affected more strongly by population bottlenecks than theta. The short-term effective size of a population for sequence data is derived from theta per site, which is sensitive for population dynamic changes and especially growth in population size (Nei 1987). Overall thetas and therefore, overall short-term effective population sizes in all studied species are larger than in single subpopulations (Table 4), which is a result from past expansion and later fragmentation of populations. The only exceptions found were the red squirrel population in Oulu and the Siberian tit population in Kuusamo. Why have these populations maintained an ancient number of genealogical lineages? A similar trend is seen also in their nucleotide diversities (I, IV). Most probably these populations have been ‘central’ populations during and after their colonisation, and they have not experienced drastic bottlenecks as a result of present-day environmental destruction.

4.5 Spatio-temporal population dynamics

The population sizes of the Siberian tit, the Siberian jay and the Siberian flying squirrel have strongly declined since the 1950’s (Järvinen et al. 1977, Virkkala 1991, Väisänen et al. 1998). The decline in population sizes in both avian species parallels and even exceeds levels of habitat destruction (Järvinen et al. 1977, Lloyd 1999). It is possible to detect population demographic changes resulting from changes in population sizes by comparing different sequence diversity and microsatellite indices (I, II, III, IV, Selonen 2002, Fig. 9). Nucleotide diversity (π) decreases as a function of spatial and/or temporal isolation in the two avian species studied (I, II, III). It also shows a decreasing trend toward colonisation direction in the flying squirrel (IV). The microsatellite study of Selonen (2002) depicts a recent decline in two coastal populations: Anjalankoski and Luoto. The high amount of genetic variation in the Finnish populations associated with good habitat quality, population history and gene flow (I, II, III, IV, Selonen 2002).

The highest nucleotide diversity was found in Kuusamo, which is suboptimal habitat for the Siberian tit, and associates with high adult survival in this population (Orell et al. 1999, I). In more northern areas of Finland, where Siberian tit possesses lower genetic variation, lower adult survival has also been found, even though the ecological data gathered there is rather limited (Järvinen 1982, Virkkala 1990, Orell et al. 1999). MtDNA may carry mildly deleterious mutations in avian populations with significantly negative Tajima’s D, which was found for the Siberian tit population (Fry 1999, I). Inbreeding depression expressed in later life-history traits is caused by mutations of small effects (Dudash & Fenster 2000). In Siberian tit populations with lower genetic variation, lower fitness in later adulthood may result from inbreeding.

41

Fig. 9. Tukey-Kramer test results of population pairwise comparisons for the nucleotide diversities. No further conclusions can be derived from population demographic changes because most of the nucleotide diversities differ significantly among populations. Asterisks denote significant differences among populations at experiment wise error rate of α = 0.05.

Based on pairwise comparisons for different sequence diversity estimates, the Pomokaira population seems to suggest past population growth for the Siberian jay (III). Interestingly, no clear reason is suggested for this result, although obviously the area has been favourable, virgin habitat for the species in the past. Present day environmental destruction in the area, consisting of intensive logging and the filling of a large man-made lake in the 1970’s, are not reflected in the mtDNA diversity results. Nucleotide diversity shows a decreasing trend by increasing spatial and temporal isolation when comparing Kristiinankaupunki, Suomussalmi and Blåkölen (III). This trend must have arisen already during the colonisation history of the species in the beginning of the century. Recent population bottlenecks detected with microsatellites affirm this for Kristiinankaupunki, Suomussalmi, Blåkölen and Åmsele (III).

Low nucleotide diversities in some of the flying squirrel populations suggest a low founder number (Fig. 9, Table 5, IV) and/or association with a later decline in population size. The Luoto population does not contain any genetic variation at all in mtDNA (IV) and microsatellites show both older and recent population bottleneck signals there (Selonen 2002). Most likely Luoto has been founded by one haplotype, but it has, in addition, experienced a decline in population size (Selonen 2002) in its near history. The matrice (Fig. 9) of the population pairwise comparisons for nucleotide diversities indicates prevented migration among populations and also differences in founder numbers (Table 5). Both main mtDNA lineages co-exist in these populations possibly resulting from a secondary contact in the contact zone where different mtDNA lineages have met after colonization.

All species show trends of declining population size in mtDNA diversity and the flying squirrel and the Siberian jay in microsatellite allele data (I, II, III, IV, Selonen 2002). However, for all species studied, a low level of mtDNA diversity in any of the populations studied is more likely of historic origin than a result of present-day anthropogenic activity. Microsatellites seem to be efficient in showing decreasing trends in present-day population sizes.

A.ko

ski

3northernmosts

KarjalaSavo

VaasaAlavusTurkuNuuksioKausalaA.koski

3nor

ther

nmos

ts

Karja

laSa

vo

Vaas

aAl

avus

Turk

uNu

uksio

Kaus

ala

42

Table 5. Average number of nucleotide differences (k) in the flying squirrel populations.

Population k(S.D.) Kokkola, Pietarsaari, Luoto 0 Vaasa 0.303 (0.130) Alavus 0.795 (0.192) Turku 0.333 (0.379) Nuuksio 0.611 (0.239) Kausala 0.500 (0.210) Anjalankoski 0.857 (0.345) Karjala 0.679 (0.386) Savo 0.611 (0.158)

5 Concluding remarks and future prospects

In this thesis, I have contributed to the growing amount of molecular genetic studies directing attention to the relationship of phylogeography and conservation. My primary study objects are species original to Finnish fauna; the Siberian tit, the Siberian jay, and the Siberian flying squirrel (and the red squirrel as a reference species), whose populations appear vulnerable or threatened due to habitat loss and the fragmentation of virgin, coniferous forests. Already in naturally fragmented boreal taiga, each species studied revealed an increase in population genetic structuring associated with an increase in habitat association or restriction, and a decreased capacity for dispersal. Signatures of previous, as well as modern day population isolation, including declines in population size associated with human alterations of habitat, can be found in their genome.

The Siberian tit population close to the Finnish-Russian border showed the highest levels of genetic variation suggesting a ‘source-sink’ metapopulation structure of the species. The present day gene-flow has already been prohibited, since significant differentiation in genetic variation among populations is apparent. Possible indications of inbreeding effects were found in the Siberian tit populations. It would be interesting to compare local adaptation in the Siberian tit, using microsatellites and other ecological characteristics (besides adult survival), between populations showing high levels of genetic variation (Kuusamo) relative to those showing lower levels of variation. Microsatellite surveys could further reveal whether the Kuusamo population serves as a vector mediating gene-flow to other Siberian tit populations in Lapland. Microsatellites could also explain whether the high variation in Kuusamo is due to ongoing gene-flow attributable to populations associated with Russian virgin forests or certain long-lasting, favourable habitat conditions. Siberian tit samples from Yakutsk and Fennoscandia show remarkable similarity despite their extensive geographical distinction.

The most evident response in recent population decline was found in relation to microsatellite analyses in the Siberian jay, where complementary mtDNA results revealed more historical trends in population size and isolation. Mitochondrial and microsatellite data, together with assignment studies reveal that the Siberian jay population is significantly genetically structured and support the existence of a type of meta-population structuring in Fennoscandia. An increase in both temporal and spatial isolation can be seen as decreased levels of genetic variation and bottleneck signals. However, local

44

populations showed no signs of inbreeding. In addition to a Fennoscandian lineage, at least two different mitochondrial lineages were found to exist in Siberia. Genetic variation was found to be lower in the Fennoscandian than the Far-East populations. Three different mtDNA lineages have evolved in distinct and different sized Pleistocene refugia. Broader, more extensive phylogeographical studies in Siberia using both mtDNA and microsatellites would be interesting.

MtDNA analyses show surprisingly high levels of genetic structuring in the Siberian flying squirrel. Two different mitochondrial lineages in Finland appear to reflect vestiges of two different colonisation routes. This ancient, vestigial structuring is already worth conservation consideration. There may exist long-term adaptations between these areas. In addition, significant structuring within the two groups and detection of five ‘management units’ calls for immediate action on the local level and for the management of suitable habitats for these populations. However, these areas should not be considered as distinct, independent management units. Minimal effective gene flow found among populations contrasts with ecological studies, according to which the species is rather mobile. Flying squirrel females disperse according to a leptokurtic dispersal model where as the nuclear gene flow follows a stepping-stone model. Despite natural fragmentation and the possibility that the species’ may be well adapted to such characteristics of its environment, sustainable forest management and planning are needed at broader spatial scales to ensure and increase levels of present day and natural gene flow between populations. MtDNA and microsatellite analyses combined with ecological studies of local population characteristics may reveal possible adaptations and interesting interactions, for instance in dispersal propensity (short distance versus long distance dispersers and possible linkage to certain haplotypes), adult survival, nucleotide diversity, levels of heterozygosity or microsatellite length variability. It would be interesting and worthwhile to examine whether differences in fitness exist between populations showing high versus low levels of genetic variation, as found for Alavus compared to Luoto.

Species with greater genetic variation are less threatened and thus, of lesser importance in terms of conservation status in Finland. In Finland, and considering the species studied herein, the flying squirrel seems the most threatened because of its lack of genetic variation and strong population genetic structuring. Both of the avian species studied possess more genetic variation and they are also considered less threatened according to their conservation status (IUCN). The red squirrel, of all the species studied, showed the greatest genetic variation which is not surprising since its population in Finland is large, continuously distributed and also does not appear to be suffering from a decline. The red squirrel population is highly fragmented in Britain, where most local populations are fixed for one haplotype (Barratt et al. 1999).

Significant differences in nucleotide diversities indicated restrictions in gene flow among Fennoscandian populations in all species studied. A central Siberian jay population showed growth in relation to other populations in Finland when comparing different sequence diversity indices among populations. Some marginal willow tit populations showed significantly smaller founder number when compared to several other, more central, populations studied across the species’ Palearctic distribution range. The study objects showed differences in their population genetic structure across their whole distribution range consistent with the multirefugia model suggested by coniferous pollen data.

References

Aartolahti T (1966) Uber die Einwanderung und die Verhaufigung der Fichte in Finnland. Ann Bot Fenn 3: 368-379.

Alekseev MN (1997) Paleogeography and geochronology in the Russian easterb arctic during the second half of the Quaternary. Quat Internat 41/42: 11-15.

Amos W & Harwood J (1998) Factors affecting levels of genetic diversity in natural populations. Phil Trans R Soc Lond B 353: 177-186.

Andersen BG & Borns HW Jr (1997) The Ice Age World. WSOY. Porvoo. Aris-Brosou S & Excoffier L (1996) The impact of population expansion and mutation rate

heterogeneity on DNA sequence polymorphism. Mol Biol Evol 13: 494-504. Avise JC (1994) Molecular Markers, Natural History and Evolution. Chapman & Hall, New York. Avise JC (1998) The history and purview of phylogeography: a personal reflection. Mol Ecol 7:

371-379. Avise JC & Walker DE (1998) Pleistocene phylogeographic effects on avian populations and the

speciation process. Proc R Soc Lond B 265: 457-463. Ballard JWO & Kreitman M (1995) Is mitochondrial DNA a strictly neutral marker? Trends Ecol

Evol 12: 485-488. Barratt, E.M., Gurnell, J., Malarky, G., Deaville, R. & Bruford, M.W. 1999. Genetic structure of

fragmented populations of red squirrel (Sciurus vulgaris) in the UK. Mol Ecol 8S: 55-63. Barton NH & Whitlock MC (1997) The evolution of metapopulations. In Metapopulation biology.

Ecology, Genetics, and Evolution.(ed.) Hanski IA & Gilpin ME, pp. 183-210. Academic Press, San Diego.

Barton NH & Wilson I (1998) Genealogies and geography. In New Uses for New Phylogenies (ed.). Harvey PH, Leigh Brown AJ, Smith JM & Nee S. pp. 23-56. Oxford University Press, Oxford.

Berlin S & Ellegren H (2001) Clonal co-transmission of mitochondrial DNA and the female-specific avian W chromosome. Nature 413: 37-38.

Bermingham E & Moritz C (1998) Comparative phylogeography: concepts and applications. Mol Ecol 7: 367-369.

Bilton DT, Mirol PM, Mascheretti S, Fredga K, Zima J & Searle JB (1998) Mediterranean Europe as an area of endemism for small mammals rather than a source for northwards postglacial colonization. Proc R Soc Lond B 265: 1219-1226.

Birky CW Jr, Maruyama T & Fuerst P (1983) An approach to population and evolutionary genetic theory for genes in mitochondria and chloroplasts, and some results. Genetics 103: 513-527.

Boon WM, Kearvell JC, Daugherty CH & Chambers GK (2000) Molecular Systematics of New Zealand Cyanoramphus Parakeets: Conservation of Orange-fronted and Forbes' Parakeets. Bird Conserv. Int.

46 Borst P & Grivell LA (1981) Small is beautiful – portrait of a mitochondrial genome. Nature 290:

443–444. Brown WM, George M Jr & Wilson AC (1979) Rapid evolution of animal mitochondrial DNA.

Proc Natl Acad Sci USA 76: 1967-1971. Brown WM (1983) Evolution of Animal Mitochondrial DNA. In: Nei M & Koehn RK (ed.):

Evolution of Genes and Proteins. pp. 62-88. Sinauer Associates Inc Publishers, Sunderland, Massachusetts.

Brown GG, Gadaleta G, Pepe G, Saccone C & Sbisà (1986) Structural Conservation and Variation in the D-loop-containing Region of Vertebrate Mitochondrial DNA. J Mol Biol 192: 503-511.

Caballero A (1994) Developments in the prediction of effective population size. Heredity 73: 637-679.

Clark AM, Bowen BW & Branch LC (1999) Effects of natural habitat fragmentation on an endemic scrub lizard (Sceloporus woodi): an historical perspective based on a mitochondrial DNA gene genealogy. Mol Ecol 8: 1093-1104.

Clayton DA (1991) Nuclear gadgets in mitochondrial DNA replication and transcription. Trends Biol Sci 16: 107-111.

Cramp S & Perrins CM (1993) Handbook of the birds of Europe, the Middle East and North Africa. The birds of the western palearctic, Volume VII, pp. 187-195. Oxford University Press, Oxford.

Cramp S & Perrins CM (1994) Handbook of the birds of Europe, the Middle East and North Africa. The birds of the western palearctic, Volume VIII, pp. 31-42. Oxford University Press, Oxford.

Crow JF & Kimura M (1970) An Introduction to Population Genetics Theory (eds Harper, Row). New-York.

Darwin C (1859) The origin of species, first edition. Reprinted in 1987 by Richard Clay Ltd, Bungay, Suffolk.

Dayanandan S, Dole J, Bawa K & Kesseli R (1999) Population structure delineated with micrasatellite markers in fragmented populations of a tropical tree, Carapa guianensis (Meliaceae) 8:1585-1592.

Desjardins P & Morais R (1990) Sequence and gene organization of the chicken mitochondrial genome. A novel gene order in higher vertebrates. J Mol Biol 212: 599–634.

Desjardins P & Morais R (1991) Nucleotide sequence and evolution of coding and noncoding regions of a quail mitochondrial genome. J Mol Evol 32: 153–161.

Desrochers A, Hanski IK & Selonen V (2003) Siberian flying squirrel responses to high- and low-contrast forest edges. Lands Ecol 18: 543-552.

Di Rienzo A, Peterson AC, Garza JC, Valdes AM, Slatkin M & Freimer NB (1994) Mutational processes of simple-sequence repeat loci in human populations. Proc Natl Acad Sci USA 91: 3166-3170.

Doda JN, Wright CT & Clayton DA. 1981. Elongation of displacement-loop strands in human and mouse mitochondrial DNA is arrested near specific template sequences. Proc Natl Acad Sci USA 78: 6116-6120.

Dudash M & Fenster C (2000) Inbreeding and outbreeding depression in fragmented populations. In Genetics, Demography and Viability of Fragmented Populations. Young AC & Clarke GM (ed.). Cambridge University Press, Cambridge.

Eberhard JR, Wright TF & Bermingham E (2001) Duplication and concerted evolution of the Mitochondrial Control Region in the Parrot Genus Amazona. Mol Biol Evol 18: 1330-1342.

Eckert R, Randall D & Augustine G (1988) Animal physiology. Mechanisms and adaptations. W. H. Freeman and Company, New York.

Eeva T, Lehikoinen E & Veistola S (1989) Habitat preference and breeding performance in four hole-nesting passerines at the northern limit of their range. Ornis Fenn 66: 142-150.

Ekman J, Schlepkovych B & Tegelström H (1994) Offspring retention in the Siberian jay (Perisoreus infaustus): the prolonged brood care hypothesis. Behav Ecol 5: 245-253.

Ellegren H (2000) Microsatellite mutations in the germline, implications for evolutionary inference. Trends in Genet 16: 551-558.

Ellegren H, Carlson A & Stenberg I (1999) Genetic structure and variability of white-backed woodpecker (Dendrocopos leucotos) populations in northern Europe. Hereditas 130: 291–299.

Esseen P-A, Ehnström B, Ericson L & Sjöberg K (1997) Boreal forests. Ecol Bull 46: 16-47.

47 Estoup A, Andrne P & Cornuet J-M (2002) Homoplasy and mutation model at microsatellite loci

and their consequences for population genetics analysis. Mol Ecol 11: 1591-1602. Excoffier L, Smouse P & Quattro J (1992). Analysis of molecular variance inferred from metric

distances among DNA haplotypes: Application to human mitochondrial DNA restriction data. Genetics 131: 479-491.

Falconer DS (1989) Introduction to Quantitative Genetics, 3rd edn. Longman. Essex. Fedorov VB (1998) Phylogeography and Mitochondrial DNA Diversity in Arctic Lemmings. Ph.

D. thesis. University of Uppsala. Fleischer RC, McIntosh CE, Tarr CL (1998) Evolution on a volcanic conveyor belt: using

phylogeographic reconstructions and K-Ar-based ages of the Hawaiian Islands to estimate molecular evolutionary rates. Mol Ecol 7: 533-545.

Fowler EV, Houlden BA, Hoeben P & Timms P (2000) Genetic diversity and gene flow among southeastern Queensland koalas (Phascolarctos cinereus). Mol Ecol 9: 155-164.

Fry AJ (1999) Mildly deleterious mutations in avian mitochondrial DNA: Evidence from neutrality tests. Evolution 53: 1617–1620.

Fu Y-X (1997) Statistical tests of neutrality of mutations against population growth, hitchhiking and backround selection. Genetics 147: 915-925

Gaines MS, Diffendorfer JE, Tamarin RH & Whittam TS (1997) The effects of habitat fragmentation on the genetic structure of small mammal populations. J Hered 88: 294-304.

Galtier N & Boursot P (2000) A new method for locating changes in a tree reveals distinct nucleotide polymorphism vs divergence patterns in mouse mitochondrial control region. J Mol Evol 50: 224–231.

Gibbs HL, Prior KA, Weatherhead PJ & Johnson G (1997) Genetic structure of populations of the threatened eastern massasauga rattlesnake, Sistrurus c. catenatus: evidence from microsatellite DNA. Mol Ecol 6: 1123-1132.

Gibbs M, Dawson DA, McCamley C, Wardle AF, Armour JA & Burke T (1997) Chicken microsatellite markers isolated from libraties enriched for simple tandem repeats. Anim Gen 28: 401-417.

Greenwood PJ (1980) Mating systems, philopatry and dispersal in birds and mammals. Anim Behav 28: 1140-1162.

Gribbin J (1989) The end of the Ice ages? New Scientist 17th June, 1989, 48-52. Groombridge JJ, Jones CG, Bruford MW & Nichols RA (2000) 'Ghost' alleles of the Mauritius

kestrel. Nature 403: 616. Gurnell, J (1987) The natural history of squirrels. Christopher Helm, London. Gyllensten U, Wharton D, Josefsson A & Wilson AC (1991) Paternal inheritance of mitochondrial

DNA in mice. Nature 352: 255–257. Haartman von L, Hilden O, Linkola P, Suomalainen P & Tenovuo R (1963-72) Pohjolan linnut I-II.

Otava, Helsinki, Finland. Haftorn S (1973) A study of the Siberian tit Parus cinctus during the breeding season. Sterna 12:

91-155. Haddrath O & Baker AJ (2001) Complete mitochondrial DNA genome sequences of extinct birds:

ratite phylogenetics and the vicariance biogeography hypothesis. Proc R Soc Lond B Biol Sci 268: 939-945.

Hale ML, Lurz PWW, Shirley MDF, Rushton S, Fuller RM & Wolff K (2001) Impact of landscape management on the genetic structure of red squirrel populations. Science 293: 2246-2248.

Halkka L, Söderlund V, Skarén U & Heikkilä J (1987) Chromosomal polymorphism and racial evolution of Sorex araneus L. in Finland. Hereditas 106: 257-275.

Hanski IK & Simberloff D (1997) The metapopulation approach, its history, conceptual domain, and application to conservation. In Metapopulation Biology Ecology, Genetics and Evolution. Hanski IA & Gilpin ME. pp. 5-26. Academic Press, San Diego.

Hanski IK, Stevens PC, Ihalempiä P & Selonen V. (2000) Homerange size, movements, and nest-site use in the Siberian flying squirrel, Pteromys volans. J Mammal 81: 789-809.

Harding R (1996) New phylogenies: an introductory look at the coalescent. In New Uses for New Phylogenies. Ed. Harvey PH, Brown AJL, Smith JM & Nee S. pp. 15-22. Oxford University Press, Oxford.

48 Haring E, Riesing MJ, Pinsker W & Gamauf A (1999) Evolution of a pseudo-control region in the

mitochondrial genome of Palearctic buzzards (genus Buteo). J Zool Syst aEvol Res 37: 185-194.

Harrap S & Quinn D (1996) Tit, nuthatches & treekeepers. Christopher Helm, London. Harrison S & Taylor AD (1997) Empirical evidence for metapopulation dynamics. In

Metapopulation Biology Ecology, Genetics and Evolution. Hanski IA & Gilpin ME. pp. 27-42. Academic Press, San Diego.

Hays JD, Imbrie J & Shackleton NJ (1976) Variations in the Earth’s orbit: pacemaker of the ice ages. Science 194: 1121-1132.

Heikinheimo, O. 1915. Kaskiviljelyksen vaikutus Suomen metsiin. Keisarillisen senaatin kirjapaino, Helsinki.

Helle P (1985) Habitat selection of breeding birds in relation to forest succession in Northern Finland. Ornis Fenn 62: 113-123.

Hewitt GM (1996a) Some consequences of ice ages, and their role in divergence and speciation. Biol J Linn Soc 58: 247-276.

Hewitt GM (1996b) Post-glacial re-colonization of European biota. Biol J Linn Soc 68: 87-112. Hewitt GM (1999) Post-glacial recolonization of European Biota. Biol J Linn Soc 68: 87-112. Hewitt GM (2000) The genetic legacy of the Quaternary ice ages. Nature 405: 907-913. Hokkanen H, Törmälä T, Vuorinen H (1982) Decline of the flying squirrel Pteromys volans L.

populations in Finland. Biol Cons 23: 273-284. Houlden BA, Costello BH, Sharkey D et al. (1999) Phylogeographic differentiation in the

mitochondrial control region in the koala, Phascolarctos cinereus (Goldfuss 1817). Mol Ecol 8: 999-1011.

Hundertmark KJ, Shields GF, Udina IG, Bowyer RT, Danilkin AA and Schwartz CC (2002) Mitochondrial phylogeography of moose (Alces alces): late Pleistocene divergence and population expansion. Mol Phyl Evol 22: 375-387.

Hyvärinen H (1987) History of forests in northern Europe since the last glaciation. Ann Acad Science Fenn Ser A III. Geology-Geography 145: 7-18.

Ibrahim MI, Nichols RA and Hewitt GM (1996) Spatial patterns of genetic variation generated by different forms of dispersal during range expansion. Heredity 77: 282-291.

Jaarola M & Tegelström H (1995). Colonization history of north European field voles (Microtus agrestis) revealed by mitochondrial DNA. Mol Ecol 4: 299-310.

Jeffreys AJ, Tamaki K, MacLeod A, Monckton DG, Neil DL & Armour ANDL (1994) Complex gene conversion events in germline mutation at human minisatellites. Nature Genetics 6: 136-145.

Jin L, Macaubas C, Hallmayer J, Kimura A & Mignot E (1996) Mutation rate varies among alleles at a microsatellite locus: Phylogenetic evidence. Proc Natl Acad Sci USA 93: 15285-15288.

Järvinen A (1982) Ecology of the Siberian tit Parus cinctus in NW Finnish Lapland. Ornis Scand 13: 47-55.

Järvinen A (1987) A successful mixed breeding between Parus cinctus and Parus montanus in Finnish Lapland. Ornis Fenn 64: 158-159.

Järvinen O, Kuusela K & Väisänen RA (1977) Effects of modern forestry on the numbers of breeding birds in Finland in 1945-1975. Silva Fenn 11: 284-294.

Järvinen O, Kuusela K & Väisänen RA (1997) Metsien rakenteen muutoksen vaikutus pesimälinnustoomme viimeisten 30 vuoden aikana. Silva Fenn 11: 284-294.

Järvinen O & Väisänen RA (1978) Recent changes in forest bird populations in northern Finland. Ann Zool Fenn 15: 279-289.

Järvinen O & Väisänen RA (1979) Climatic changes, habitat changes, and competition: dynamics of geographical overlap in two pairs of congeneric bird species. Oikos 33: 261-271.

Kimura M & Crow JF (1964) The number of alleles that can be maintained in a finite population. Genetics 49: 725-735.

Kimura M & Ohta (1978) Stepwise mutation model and distribution of allelic frequencies in a finite population. Proc Natl Acad Sci USA 75: 2868-2872.

Klicka J & Zink RM (1997) The importance of recent Ice Ages in speciation: a failed paradigm. Science 277: 1666-1669.

49 Kocher TD, Thomas WK, Meyer A, Edwards SV, Pääbo S, Villablanca FX & Wilson AC (1989)

Dynamics of mitochondrial DNA evolution in animals: Amplification and sequencing with conserved primers. Proc Natl Acad Sci USA 86: 6196-6200.

Kremenetski CV, Liu K-B & MacDonald GM (1998) The late Quaternary dynamics of pines in northern Asia. In: Ecology and Biogeography of Pinus (Richardson, DM). University Press, Cambridge.

Kumar S, Tamura K & Nei M (1993) MEGA: Molecular Evolutionary Genetics Analysis, Vers. 1.02. Pennsylvania State University, University Park, PA, USA.

Kumazawa Y, Ota H, Nishida M & Ozawa T (1996) Gene rearrangements in snake mitochondrial genomes: highly concerted evolution of control-region-like sequences duplicated and inserted into a tRNA gene cluster. Mol Biol Evol 13: 1242-1254.

Kunst CB, Leeflang EP, Iber JC, Arnheim N & Warren ST (1997) The effect of FMR1 CGG repeat interruptions of mutation frequency as measured by sperm typing. J Med Gen 34: 627-631.

Kvist L, Ruokonen M, Thessing A, Lumme J & Orell M 1998. Mitochondrial control region polymorphism reveal high amount of gene flow in Fennoscandian willow tits (Parus montanus borealis). Hereditas 128: 133-143.

Kvist L, Ruokonen M, Lumme J & Orell M 1999a. The colonization history and present-day population structure of the European great tit (Parus major major). Heredity 82: 495-502.

Kvist L, Ruokonen M, Lumme J & Orell M. 1999b. Different population structures in northern and southern populations of the European blue tit (Parus caeruleus). J Evol Biol 12: 798-805.

Kvist L (2000) Phylogeny and phylogeography of European parids. Acad. Diss. Acta Universitatis Ouluensis. A 341. Finland.

Kvist L, Martens J, Ahola A & Orell M (2001) Phylogeography of a Palaearctic sedentary passerine, the willow tit (Parus montanusx). J Evol Biol 14: 930-941 .

Kvist L, Martens J, Hiroyoshi H, Nazarenko AA, Valchuk OP & Orell M (2003a) Evolution and genetic structure of the great tit (Parus major complex). Proc R Soc Lond B, 270: 1447-1454.

Kvist L, Martens J, Nazarenko AA & Orell M (2003b) Paternal leakage of mitochondrial DNA in the great tit (Parus major) Mol Biol Evol 20: 243-247.

L'Abbe D, Duhaime J-F, Lang BF & Morais R (1991) The transcription of DNA in Chicken Mitochondria from One Major Bidirectional Promoter. J Biol Chem 266: 10844-10850.

Lang, G. 1992. Some aspects of European late- and post-glacial flora history. Acta Bot Fenn 144: 1-17.

Levinson G & Gutman GA (1987) Slipped-strand mispairing: a major mechanism for DNA sequence evolution. Mol Biol Evol 4: 203-221.

Li Y-C, korol AB, Fahima T, Beiles A & Nevo E (2002) Microsatellites: genomic distribution, putative functions and mutational mechanisms: a review. Mol Ecol 11: 2453-2465.

Liebers, D & Helbig, AJ (2002) Phylogeography and colonization history of Lesser Black-backed Gulls (Larus fuscus) as revealed by mtDNA sequences. J Evol Biol 5: 1021-1033.

Lillandt B-G, Bensch S & Schantz von T (2001) Parentage determination in kin-structured populations: microsatellite analyses in the Siberian jay Perisoreus infaustus during a 25-year population study. Avian Science. 1: 3-14.

Lindgren F (1975) Iakttagelser rörande lavskrikan (Perisoreus infaustus) - huvudsakligen dess häckninsbiologi. Fauna och Flora 5: 198-210.

Lister A and Bahn P (2001). Mammutit Jääkauden jättiläiset. WSOY, Helsinki. Liukkonen-Anttila T, Uimaniemi L, Orell M & Lumme J (2002) Mitochondrial DNA variation and

the phylogeography of the grey partridge (Perdix perdix) in Europe: from Pleistocene history to implications for introductions. J Evol Biol 15: 971-982.

Lloyd S ed. (1999) The last of the last. The Old-growth Forests of Boreal Europe. Moscow. Lucchini W & Randi E (1998) Mitochondrial DNA sequence variation and phylogeographical

structure of rock partridge (Alectoris graeca) populations. Heredity 81: 528-536. Merikallio, (1958). Finnish birds. Their distribution and numbers. Soc Pro Fauna Fenn V:53-54. Merilä J, Björklund M & Baker AJ (1997) Historical demography and present day population

structure of the greenfinch, Carduelis chloris - an analysis of mtDNA control-region sequences. Evolution 51: 946-956.

50 Mindell DP, Sorenson MD & Dimcheff DE (1998) Multiple independent origins of mitochondrial

gene order in birds. Proc Natl Acad Sci, USA 95: 10693–10697. Moritz G (1994) Applications of mitochondrial DNA analysis in conservation: a critical review.

Mol Ecol 3: 401-411. Mönkkönen M, Reunanen P, Nikula A, Inkeröinen J & Forsman J (1997) Landscape characteristics

associated with the occurrence of the flying squirrel Pteromys volans in old-growth forests of northern Finland. Ecography 20: 634-642.

Morden CW & Loeffler W (1999) Fragmentation and genetic differentiation among subpopulations of the endangered Hawaiian mint Haplostachya haplostachya (Lamiaceae). Mol Ecol 8: 617-625.

Nei M, Maruyama T & Chakraborty R (1975) The bottleneck effect and genetic variability in populations. Evolution 29: 1-10.

Nei M (1987) Molecular Evolutionary Genetics. Columbia University Press, New York. Nichols R (2001) Gene trees and species trees are not the same. TREE 16: 358-364. Nichols RA, Bruford MW & Groombridge JJ (2001) Sustaining genetic variation in a small

population: evidence from the Mauritius kestrel. Mol Ecol 10: 593-602. Nomura, T (2002) Effective size of populations with heritable variation in fitness. Heredity 89:

413-416. Nunney, L (1991) The influence of mating system and overlapping generations on effective

population size. Proc R Soc Lond B 246: 71-76. Nunney, L (1993) The influence of mating system and overlapping generations on effective

population size. Evolution 47: 1329-1341. Ognev SI (1940) Mammals of the U.S.S.R. and adandcent countries. VI. - p. 429. Israel program

for scientific translations, Jerusalem. Orell M, Lahti K & Matero J (1999) High survival rate and site fidelity in the Siberian tit Parus

cinctus, a focal species of taiga. Ibis 141: 460-468. Orr MR & Smith TB (1998) Ecology and speciation. Trends Ecol Evol 13: 502-506. Oshida T, Masuda R & Yoshida MC (1996) Phylogenetic relationships among japanese species of

the family Sciuridae (Mammalia, Rodentia), inferred from nucleotide sequences of mitochondrial 12S ribosomal RNA genes. Zool Sci 13: 615-620.

Palo J (2003) Genetic diversity and phylogeography of landlocked seals. Acad Diss. University of Helsinki.

Pannell JR & Charlesworth B (1999) Neutral genetic diversity in a metapopulation with recurrent local extinction and recolonization. Evolution 53: 664-676.

Petes TD, Greenwell PW & Dominska M (1997) Stabilization of microsatellite sequences by variant repeats in the yeast Saccharomyces cerevisiae. Genetics 146: 491-498.

Pope LC, Estoup A & Moritz C (2000) Phylogeography and population structure of an ecotonal marsupial, Bettongia tropica, determined using mtDNA and microsatellites. Mol Ecol 9: 2041-2053.

Porter AH (1998) Refugees from lost habitat and reorganization of genetic population structure. Cons Biol 13: 850-859.

Primmer CR, Raudsepp T, Chowdhary BP, Moller AP & Ellegren H. (1997) Low frequency of microsatellites in the avian genome. Gen Res 7: 471-482.

Primmer CR, Saino N, Mφller AP & Ellegren H (1998) Unraveling the processes of microsatellite evolution through analysis of germ line mutations in barn swallows Hirundo rustica. Mol Biol Evol 15: 1047-1054.

Pritchard JK, Stephens M, Donnelly P (2000). Inference of population structure using multilocus genotype data. Genetics 155: 945-959.

Quinn TW & Wilson AC (1993) Sequence evolution in and around the mitochondrial control region in birds. J Mol Evol 37: 417-425.

Rassi, P., Alanen, A., Kanerva, T. & Mannerkoski, J. (2000): Suomen lajien uhanalaisuus 2000 Ympäristöministeriö, Helsinki. Uhanalaisten lajien seurantatyöryhmä, esipainos. http://www.vyh.fi/ luosuo/lumo/lasu/uhanal/uhanal.html

Reunanen P (1998) Liito-orava levinneisyytensä äärilaidalla Pohjois-Suomessa. Luonnontutkija 102: 29-30.

51 Reunanen P (2001) Landscape responses of the Siberian flying squirrel (Pteromys volans) in

Northern Finland The effect of scale on habitat patterns and species incidence. Acta Universitatis Ouluensis A 371.

Reunanen P & Nikula A (1998) Liito-oravan elinympäristöt and maiseman rakenne Koillismaalla. Metsähallituksen julkaisusarand: Sarja A, No 93. Vantaa.

Reunanen P, Mönkkönen M & Nikula A (2000) Managing boreal landscapes for flying squirrels. Cons Biol 14: 218-226.

Richardson DM (1998) Ecology and Biogeography of Pinus. University Press, Cambridge. Rogers AJ & Harpending H (1992) Population growth makes waves in the distribution of pairwise

genetic differences. Mol Biol Evol 9: 552-569. Rogers AJ (1995) Genetic evidence for a pleistocene population explosion. Evolution 49: 608-615. Rosquist G & Prentice HP (1999) Habitat fragmentation and the the structure of genetic diversity

within disjunct isolates of Anthericum ramosum L. (Anthericaceae) in Scandinavia. Biol J Linn Soc 69: 193-212.

Ruokonen M, Kvist L & Lumme J (2000) Close relatedness between mitochondrial DNA from seven Anser goose species. J Evol Biol 13: 532 -540.

Ruokonen M & Kvist L (2002) Structure and evolution of avian mitochondrial control region. Mol Phyl Evol 23: 422-432.

Saccheri I, Kuussaari M, Kankare M, Vikman P, Fortelius W & Hanski I (1998) Inbreeding and extinction in a butterfly metapopulation. Nature 392: 491-494.

Sambrook J, Fritsch EF & Maniatis T (1989) Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory Press, New York.

Saura A, Lokki J, Lankinen P & Suomalainen E (1976) Genetic polymorphism and evolution in parthenogenetic animals. III. Tetraploid Otiorhynchus scaber (Coleoptera: Curculionidae). Hereditas 82: 79-100.

Savolainen J (1992) Kuukkelimaita kulkemaan. Lintuviesti 1: 16-19. Segelbacher G & Storch I (2002). Capercaillie in the Alps: genetic evidence of metapopulation

structure and population decline. Mol Ecol 11: 1669-1677. Selonen V (2002) Spacing behaviour of the Siberian flying squirrel – effects of landscape structure.

Acad Diss. Helsinki. Selonen V & Hanski IK (2003) Movements of the flying squirrel (Pteromys volans) in corridors

and matrix habitat. Ecography 26: 641-651. Selonen V & Hanski IK (2004) Young flying squirrels (Pteromys volans) dispersing in fragmented

forests. Behav Ecol, in press. Selonen V, Hanski IK & Stevens P (2001) Space use of the Siberian flying squirrel Pteromys

volans in fragmented forest landscapes. Ecography, 24: 588-600. Seutin G, Ratcliffe LM & Boag PT (1995) Mitochondrial DNA homogeneity in the phenotypically

diverse redpoll finch complex (Aves: Carduelinae: Carduelis flammea-hornemanni). Evolution 49: 962-073.

Sherwin W & Moritz C (2000) Managing and monitoring genetic erosion. In Genetics, Demography and Viability of Fragmented Populations. Young AC & Clarke GM (ed.), pp. 9-34. Cambridge University Press, Cambridge.

Sklepkovych B (1997) Kinship and conflict resource competition in a proto-cooperative species: the Siberian jay. Acad. Diss. Stockholm, Sweden.

Slatkin M (1977) Gene flow and genetic drift in a species subject to frequent local extinctions. Theor Pop Biol 12: 253-262.

Slatkin M & Barton N (1989) A comparison of three indirect methods for estimating average levels of gene flow. Evolution 43: 1349-1368.

Smith GP (1976) Evolution of repeated DNA sequences by unequal crossover. Science 191: 528-535.

Smith TB, Holder K, Girmin D, O’Keefe K, Larison B & Chan Y (2000) Comparative avian phylogeography of Cameroon and Equatorial Guinea mountains: implications for conservation. Mol Ecol 9: 1505-1516.

Sokal RR & Rohlf FJ (1995) Biometry. The principles and practice of statistics in biological research. W. H. Freeman. New York.

52 Southern SO, Southern PJ & Dizon AE (1988) Molecular Characterization of a Cloned Dolphin

Mitochondrial Genome. J Mol Evol 28: 32-42. Spies TA & Franklin JF (1996) The diversity and management of old-growth forests. In: Szaro RC

& Johnston DW (eds) Biodiversity in managed landscapes. pp. 298-314. Oxford University Press, New York, USA.

Stewart JR & Lister AM (2001) Cryptic northern refugia and the origins of the modern biota. Trends Ecol Evol 16: 608-613.

Storz JF & Beaumont MA (2002) Testing for genetic evidence of population expansion and contraction: an empirical analysis of microsatellite DNA variation using a hierarchical Bayesian model. Evolution 56: 154–166.

Svendsen JI, Astakhov VI, Bolshiyanov DY, Demidov I, Dowdeswell JA, Gataullin V, Hjort C, Hubberten HW, Larsen E, Mangerud J, Melles M, Möller P, Saarnisto M & Siegert MJ (1999) Maximum extend of the Eurasian ice sheets in the Barents and Kara Sea region during the Weichselian. Boreas 28: 234-242.

Syrjänen K, Kalliola R, Puolasmaa A & Mattson J (1994) Landscape structure and forest dynamics in subcontinental Russian European taiga. Ann Zool Fenn 31: 19-34.

Taberlet P, Fumagalli L, Wust-Saucy AG & Cossons J-F (1998) Comparative phylogeography and postglacial colonization routes in Europe. Mol Ecol 7: 453-464.

Tajima F (1983) Evolutionary relationship of DNA sequences in finite populations. Genetics 105: 437-460.

Tajima F (1989a) Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123: 585-595.

Tajima F (1989b) The effect of change in population size on polymorphism. Genetics 123: 597-601.

Takahata N (1988) The coalescent in two partially isolated diffusion populations. Gen Res 52: 213–222.

Tamura K & Aotsuka T (1988) Rapid isolation method of animal mitochondrial DNA by the alkaline lysis procedure. Bioch Gen 26:815-819.

Tautz D (1989) Hypervariability of simple sequences as a general source for polymorphic DNA markers. Nucleic Acids Res 17: 6463-6471.

Templeton AR (1998) Nested clade analyses of phylogeographic data: testing hypotheses about gene flow and population history. Mol Ecol 7: 381–397.

Tiebout HM & Anderson RA (1997) A comparison of corridors and intrinsic connectivity to promote dispersal in transient successional landscapes. Cons Biol 11:620-627.

Tolonen K (1983) Kuusen levinneisyyshistoria Suomessa. Sorbifolia 14: 53-59. Tucker GM & Heath MF (1994) Birds in Europe: their conservation status. BirdLife International,

Cambridge. Tyrberg T (1998). Pleistocene birds of the Palearctic: a catalogue. Publications of the Nuttall

Ornithological Club no. 27. Cambridge, MA. Van Treuren R, Kuittinen H, Kärkkäinen K, Baena-Gonzalez and Savolainen O. 1997. Evolution of

Microsatellites in Arabis petraea and Arabis lyrata, Outcrossing relatives of Arabidopsis thaliana. Mol Biol Evol 14: 220-229.

Vawter L & Brown WM (1986) Nuclear and mitochondrial DNA comparisons reveal extreme rate variation in the molecular clock. Science 234: 194-196.

Veistola S & Lehikoinen E (1994) Lapintiaisen vähenemisen syistä. Linnut 6: 31. Virkkala R (1990) Ecology of the Siberian tit Parus cinctus in relation to habitat quality: effects of

forest management. Ornis Scand 21: 139-146. Virkkala R (1991) Spatial and temporal variation in bird communities and populations in north-

boreal coniferous forests: a multiscale approach. Oikos 62: 59-66. Virkkala R & Liehu H (1990) Habitat selection by the Siberian Tit Parus cinctus in virgin and

managed forests in northern Finland. Ornis Fenn 67: 1-12. Virkkala, R., Raandsärkkä, A., Väisänen, R. A., Vickholm, M and Virolainen, E. (1994)

Conservation value of nature reserves: do hole-nesting birds prefer protected forests in southern Finland? Ann Zool Fenn 31: 173-186.

53 Väisänen RA, Lammi E & Koskimies P (1998) Distribution, numbers and population changes of

Finnish birds. Otava. Helsinki. (in Finnish with English summary). Väli U (2002) Mitochondrial pseudo-control region in old world eagles (genus Aquila). Mol Ecol

11: 2189-2194. Wakeley J 1999 Nonequilibrium migration in human history. Genetics 153: 1863-1871. Walker CW, Vilá C, Landa A, Linden M & Ellegren H (2001). Genetic variation and population

structure in Scandinavian wolverine (Gulo gulo) populations. Mol Ecol 10: 53-63. Walsh SP, Metzger DA & Higuchi R (1991) Chelex 100 as medium for simple extraction of DNA

for PCR-based typing from forensic material. BioTechniques 10: 506-513. Watterson, GA (1975) On the number of segregating sites in genetical models without

recombination. Theor Pop Biol 7: 256-276. Weber JL & Wong C (1993) Mutation of human short tandem repeats. Hum Mol Gen 2: 1123-8. Weir BS & Cockerham CC (1984) Estimating F-statistics for the population structure. Evolution

38: 1358-1370. Wenink PW, Baker AJ & Tilanus MGJ (1993) Hyper-variable-control-region sequences reveal

global population structuring in a long-distance migrant shorebird, the dunlin (Calidris alpina). Proc Natl Acad Sci, USA 90: 94-98.

Wenink PW, Baker AJ & Tilanus MGJ (1994) Mitochondrial control-region sequences in two shorebird species, the turnstone and the dunlin, and their utility in population genetic studies. Mol Biol Evol 11: 22-31.

Wenink PW, Baker AJ & Tilanus MGJ (1996) Global mitochondrial DNA phylogeography of holarctic breeding dunlins (Calidris alpina). Evolution 50: 318-330.

Wettstein PJ, Strausbauch M, Lamb T, States J, Chakraborty R, Jin L & Riblet R (1995) Phylogeny of Six Sciurus aberti Subspecies Based on Nucleotide Sequences of Cytochrome b. Mol Phyl Evol 4: 150-162.

Whitlock MC & McCauley DE (1998) Indirect measures of gene flow and migration: FST≠1/(4Nm+1). Heredity 82: 117-125.

Willis KJ, Bennett KD & Birks JB (1998) The late Quaternary dynamics of pines in Europe. In: Ecology and Biogeography of Pinus (Richardson, DM). University Press, Cambridge.

Wilson AC, Cann RL, Carr SM, et al. 1985. Mitochondrial DNA and two perspectives on evolutionary genetics. Biol J Linn Soc 26: 375-400.

Wilson RC & Drury SA (1999) The Great Ice Age climate change and life. London and New York, England and USA.

Wright S (1931) Evolution in Mendelian populations. Genetics 16: 97-159. Ympäristöministeriö (2001) Liito-oravan (Pteromys volans) biologia and suojelu Suomessa.

Työryhmämuistio 459. (In Finnish with English attachments) Young A, Boyle T & Brown T (1996) The population genetic consequences of habitat

fragmentation for plants. Trends Ecol Evol 11: 413-418. Zink RM (1996) Comparative phylogeography in North American birds. Evolution 50: 308-317. Zouros E, Freeman KR, Ball AO & Pogson GH (1992) Direct evidence for extensive paternal

mitochondrial DNA inheritance in the marine mussel Mytilus. Nature 359: 412–414. Zuckerkandl E & Pauling L (1965) Evolutionary divergence abd convergence in proteins. In

Evolving Genes and Proteins (eds Bryson V & Vogel HJ) Academic Press, New York.