University of Groningen
Genetic structure of the moss genus polytrichum
van der Velde, Marco
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to
cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2000
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
van der Velde, M. (2000). Genetic structure of the moss genus polytrichum Groningen: s.n.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the
author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the
number of authors shown on this cover page is limited to 10 maximum.
Download date: 16-06-2017
SummarisinsDiscussion
Background ofthe Study
It is generally acceptedthat genetic variation is essentialfor evolutionary processes.
Mutation, selection,geneflow and randomdrift are the main factorsinfluencing the level
and structureof geneticvariation within and among natwal populations,and, therefore,
may have important evolutionary consequences.
In addition to these factors also the
geneticsystemof a species,such as the breedingsystem,has been shown to affect the
'fhis
geneticstructureof naturalplant populations(Hamrick& Godt, 1990).
thesisaimedat
getting insight into the factorsthat determinethe geneticstructureof bryophyte species,
because.due to the unique predominantly haploid liÍb cycle and the (assumed)
predominanceof asexualreproductionin bryophytes,thesefactors might differ between
bryophytesand vascularplant species(Longton, 1976; Mishler, 1988; During, 1990;
Ennos, 1990). For this reason the level and structure of genetic variation has been
investigatedfor five speciesof the moss genusPolytríchum(P. commune,P.formosum,P
pili.ferum,P. juniperinum utd P. longisetum)at three hierarchicallevels: among species,
amongpopulationswithin speciesand within populations.Thesestudieswere performed
by usingallozymeelectrophoresis
andmicrosatelliteanalysisasthe main tools.
Genetic Variation among Species
Geneticvariation amongPolytrichum specieshas been studiedbecause(l) phylogenetic
relationshipsamongthe examinedtaxa might elucidatedisputedtaxonomic relationships
within this genus,(ll) the level of genetic divergencebetweenthe studiedPolytrichum
speciesmight be indicativefor their evolutionaryrate,and (lii) differencesin the level of
geneticvariation betweenthe examinedspeciesmight reveal relationshipsbetweenthe
level of geneticvariationandboth lifè-historycharacteristics
aswell as the geneticsystem.
Phylogeny
For severalPolytrichunt species,for exampleP. communes.l. and P. juniperinum s.1.,
infra-specifictaxa have been described.However, using morphologicalcharactersonly.
bryologists have not (yet) reachedconsensuson the taxonomic status of these taxa
(Schiebl, 1982,1991;Wyatt & Derda,1997).For instance,P. communevaÍ. commune
andP. communevar.uliginosumhave
traditionallybeenrecognised
as varietieswithin the
speciesP. communes.l. Using allozymeelectrophoresis,
however,it is shovm in the
chapters2 & 4 that Nei's (1987)geneticidentity(1 :0.61) betweentheserwo taxa is
comparableto those generallyobservedfor congenericmoss species(see Table 2.5 of
chapter 2). Moreover,6 of the 14 analysedenzyme loci showed alleles that were
diagnosticfor each of thesetwo taxa, and no hybrid genotypes,i.e. genotypesshowing
diagnosticallelesof both ta"xasimultaneously,have ever been observed(chapter 2). An
across-species
analysisofmicrosatellites(chapter 3) that havebeendevelopedspecifically
for P. formosum (chapter 5) did also indicate thal P. communevar. commune utd, p.
communevar. uliginosum are geneticallyrather different. For thesereasons,also taking
into accountthe differencesin morphologyand habitatassociation,thesetwo taxa should
119
Chapter9
be regardedas two distinct species:P. commune and P. ttliginosum (chapter 2). In
additionto this study,high levelsof geneticdivergencehave also beenobservedbetween
other pairs of bryophytetaxa, for examplebetweenP. communeandP. jensenli (Derda&
Wyatt, 1999a),and betweenP. juniperínumandP. striclum(M. van der Velde, unpubl.
data;Derda& Wyatt, unpubl.ms.),that previouslyhavebeenregardedas conspecifictaxa
of P. communes.l. and P. juniperinums.l. (e.g.Crum & Anderson.1981).This suggest
that also to the taxa P. jensenii andP. stricíttn speciesrank could be assigned(Derda &
Wyatt,1999a,unpubl.ms.).
As severaldifferentclassifications
ofthe Polytrichaceae
havebeenproposed(Smith,
1971;KoponeneÍ al., 1977;Andersonet al., 1990),it is clearthat for tl.retaxonomyof this
family also on the genericlevel no unanimityexistsamong bryologists.Nevertheless.
many bryologistsÍbllow the classification
of Smith (1971),who proposedthe divisionof
Polylrichunts.l. into the generaPolytrichums.s.andPolylrichastrum.As P. .fbrmosumand
P. longisetumaretentativePolylrichastrurz
species(Smith,1971),the low levelsof Nei's
(1987) geneticidentity (I : 0.13-0.24)observedbetweenP. formo.sumand eachof the
species
P. commune,
P. uliginosum,
P.juniperinumandP. piliferum(chapters2 & 4), and
the closerelationshipbetweenP. .formosumutd P. longisetum(chapter 3) do supporta
divisionof tlregenusPolytríchums.l.into the generaPolynichums.s.and Polytrichctstrwn
(Smith,1971).Moreover,phylogenetic
treesconstmctedin the chapters2 & 4 shorv.in
addition to those published in other recent studies on the Polyrichaceaeusing
morphologicalcharactersas well as allozyme electrophoresis
and DNA sequencing
(Hyvónenet al., 1998;Derdaet al., 1999),that speciesbelongingto eitherof thesetwo
generaclusterin different groups.However,as the level of divergencebetweenthe three
clustersP. piliferum andP.juniperinum, P. communeandP. uliginosum,andP. fitrmosum
and P. longisetumis approximatelyequally high, this indicatesthat if the division of
Polytrichums.s.and Polytrichastrunzis acceptedthen the taronomic statusof the section
Juniperifoliaof the ger.rusPolytrichum sl.rouldalso be elevatedto the level of a separate
genusaswell (chapter4; Derdaet al.,1999).
ln additionto elucidatingtaxonomicrelationships
betu'eenbryophye taxa. studies
usingmolecularmarkersalsohavegivenbetterinsightinto the phylogeneticrelationships
betweendiploid bryophyte speciesand their tentativehaploid progenitors.For example.
many diploid species that, based on their morphology, have been thought to be
autodiploids,
haverecentlybeenshownto be allodiploidspecies(Wyattet al., 1992;Wyatl
et ul., 1993a,b;Cronberg,1996a;Wyatt & Odrzykoski,1998).In this study, both the
chromosomenumberof 2n:14 (the basic chromosomenumber for the Poiytrichaceaeis
regardedto be n:7; Ramsay,1983),as well as the Ílxed heterozygous
bandingpattems
observedfor both allozymeand microsatelliteloci indisputablyshow that P. longisetumís
also an allodiploid species(chapter 3). Comparisonof the diploid allozyme and
microsatellitebandingpattemsof P. longisetumwtththoseobservedfor the most closely
related extant Polytrichum speciesindicates that P. formosum, or one of its recent
ancestors,is one of the two haploid progenitor speciesof P. longisetum, and that the
secondprogenitorspeciesmay be extinct (chapter 3). These findings are especially
irrterestingÍion.r a taxonomic point of view, because(i) P. formosum has also been
mentionedas a possibleprogenitorspeciesof tluee other allodiploid Polytrichums.I.
P. ohioense,
species:
P. sexangulare
andP.pallidisetum(Derdaet al.1999), and (ll) these
threeallodiploidPolytrichums.l. speciestogetl.rer
with P. ./brmosuntand P. longisetum
120
C
Í
(
I
]
t
e
1
(
I
I
I
SummarisinsDiscussion
inosum (chapter 2). In
o beenobsewedbetween
andP.jensenii (Derda&
l, van der Velde,unpubl.
;ardedasconspecifictaxa
:son,1981).This suggest
rld be assigned(Derda &
(Smith.
ve beenproposed
Í for thetaxonomyof this
lryologists.Nevertheless,
proposed
the divisionof
and
rum.As P..f'ormosum
i), thelow levelsof Nei's
vmosumand eachof the
urn(chapters2 & 4), and
(chapter3) do supporta
r s.s.andPolytrichastrum
chapters2&4show,in
re Polytrichaceaeusing
s and DNA sequencing
ng to eitherof thesetwo
rrgence
betweenthe three
ínosum,andP.formosum
:s that if the division of
rmic statusof the section
to the level of a separate
r bryophyetaxa. studies
,hylogenetic
relationships
rogenitors.For example,
ve been thought to be
Wyat et al.. 1992;Wyatt
. In this study, both the
is
for the Polytrichaceae
'zygousbandingpattems
row thatP. longisetumis
: diploid allozyme and
ved for the most closely
n, or one of its recent
'ongisetum,
and that the
findingsare especially
)rmosumhas also been
diploid Polyrrichums.l.
r al. 1999),and (ii) these
tsum nd P. longiselum
have all been assignedto the genusPolytrichastrun by Smith (1971). This suggeststhat
the genusPolytrichastrun is comprisedof mainly allodiploid speciesand their haploid
progenitors.
Based on the results obtainedso far, it is expectedthat the application of these
molecularmarkersin the future, in addition to morphologicalcharacters,will provide us
with a much betterunderstanding
of the phylogeneticrelationshipsbetweenthe taxa of the
mossgenusPolytrichum.
EvoluÍionaryRate
Allozyme data have shown that in general bryophyte speciesare genetically highly
diverged,as Nei's (1987)geneticidentities(I) betweencongenericmoss speciesranged
from 1 = 0.21-0.63for severaldifferentmoss genera(seeTable 2.5 of chapter 2, and
chapter 4). Theseidentitiesare all lower than the averagegeneticidentity observedfor
congenericspeciesof severaldifferent vascularplant genera(l:0.79; Crawford, i983).
Moreover,the geneticidentity observedbetweenthe closelyrelatedtaxa P. communeand
P. uliginosum(1:0.61) which traditionallyhavebeenregardedas varietiesof one species
but in this thesishavebeenshownto be distinctspecies(chapter 2), is evenlower thanthe
averageidentity observedfor congenericvascularplants (1:0.79; CrauÍbrd, 1983). This
indicatesthat, in contrastto what previously has been assumedon basis of observed
differencesin morphology(Anderson,1963;Longton,19761'
Mishler, 1988 Wyatï et al.,
1989b;Ennos,1990),the evolutionaryrateofmosses,at leastat the proteinlevel,hasnot
been constrainedby lack of geneticvariation (chapter 4). Altematively, it might be that
the evolutionaryorigin of most of the examinedmoss speciesmentionedin Table 2.5
(chapter2) is much earlierthanthat of the examinedvascularplant speciesreviewedby
Crawford(1983).As such,thesemossspeciesmay havebeenreproductivelyisolatedand,
consequently,may have geneticallydivergedover a much longer time than the examined
vascularplant species.This may have resulted in these much lower levels of genetic
identity betweenthe studiedcongenericbryophytespecies,
The Polytrichr.rmspeciesexaminedin this studyare geneticallydivergedto a greater
extent (1 : 0.22) than most other congenericmoss speciesstudied so far (Table 2.5 of
chapter 2, chapter 4). The high ievel of divergencebetweenthese speciesis also
illustrated by the relatively low number of microsatellite primersetsthat have been
developedspecifically for P. Jbrmosum (chapter 5), which showed also successful
amplificationin relatedPolytrichumspecies(Table3.3 of chapter3). This contraststo the
higherlevelsof across-species
applicabilityof microsatellite
primersobservedfor other
relatedplant speciesgroups(chapter 3). As the Polyrichalesare generallyregardedas an
ancientor "primitive"groupof species(Crum,1.976;Wyatt, 1982),andevolutionaryolder
thanmost othergroupsof mosses(Mishler & Churchill, i 984), the origin of the examined
Polytrichum speciesmay date back to before the origin of most of the other examined
mossspecies.This possiblyexplainsthe much lower geneticidentitiesobservedbetween
the Polytrichun speciescomparedto those observedbetween other congenericmoss
species.However, part of the differencesin mean geneticidentity betweenthe different
moss genera may also be explained by differencesin the methodology of allozyme
electrophoresisused by the different investigators.This suggestionis supportedby the
121
Chapter 9
largedifferencebetweenthe geneticidentitydescribedin this study(l:0.22; chapter 4)
andthat reportedby Derdaet al. (1999),who observedan averagegeneticidentityof 1:
P..iuniperinum
P../itrmostun.
of P, commune,
mainly Americanpopulations
0.68bet'w'een
is
andP. piliJ'erunr,speciesthat havealsobeenexaminedin this study.A counter-argument
that the mean genetic identitiesobservedbetweencongenericspeciesof the genera
RacopilumandPlagiothecium(De Vries el al., 1983,1989;Hofman, 1991),showing.on
average,much higher identitiesthan observedfor Polytrichurz (chapter 2 & 4), were
determinedin the samelaboratoryusingthe sameproceduresand methodologyfor protein
asappliedin this study.
electrophoresis
Di//ërence,s
in Level oí GeneticVariutionamongSpecies
The Polytrichuz speciesexaminedin this study.and more specificallyP. commune.P.
uliginosttmandP. J'ormosun,show lower levelsof allozymevariation(Hr:0.025-0.166;
chapter 4) than thosereportedfor most other mossspeciesand vascularplants (Wyatt e/
et nl-. 1991).Nevertheless,
comparably
ttl., 1989b Hamrick & Godt, 1990;Stonebumer
lor.l'levelsof genediversityhavealsobeenobservedfor a numberof othermossspecies
(chapters 4 & 8; Hofman. 1991; Wyatt, 1992).As relatively low levels of genetic
variability have also beenreportedfor North Americanpopulationsof thesePolytrichum
species(Wyatt & Derda, 1997; Derda & Wyatt, 1999a.b),it can be concludcdthat
Polytrichum speciesseem to be among the least geneticallyvariable moss species
examinedso far (chapter4).
differences
in the meanlevelof genediversityexistamongcongeneric
Conspicuous
moss species(e.g. De Vries e/ al., 1983, 1989;Hofman, 1991;Wyatt, 1992; Wyatt &
Derda.1997;Cronberg,1998;Shaw,1999).Among the Polytrichunspeciesexaminedin
this study significantdifferencesin the mean level of gene diversity have also been
observed(É1s:0.022-0.
I 27; chapters4 & 8). For vascularplants,Hamrick& Godt ( I 990)
have shownthat, arnongother characteristics,
the breedingsystemhas a significanteÍlèct
on the level of geneticvariationin naturalpopulations.
As such,it was shownthat selÍing
plant speciesexhibit less geneticvariationthan outcrossingspecies(Hamrick & Godt.
1990).For bryophytes,severalauthorshavearguedthat monoecious(facultativelyselÍing)
speciesmay exhibit lower levels of genetic variation than dioecious (obligatory
outcrossing)
species(Wyatt et al., 1989b,Cronberg.1998),becauseof the high level of
self-fertilisationin monoeciousspecies,which is geneticallyequivalentto asexual
reproduction(seeGeneralIntroduction).
Hofman(1988,1991)and Cronberg(1998)have
investigatedwhetherdifferencesin the breedingsystemof bryophyeswere correlatedwith
diÍïèrences
in the levelof geneticvariationbetweencongeneric
speciesof the mossgenera
PlagiotheciuntandSphagrutnz,
respectively.In both studiesconsiderabledifferencesin the
level of geneticvariationwere observedbetweenmonoeciousand dioeciousspecies.
However,in contrastto the expectation,
monoeciousspeciesprovedmore variablethen
dioeciousspeciesin thesestudies.
As all examinedPolytrichziz speciesale dioecious,differencesin sexr-rality
as
describedabove. thereÍbre.can not accountÍbr the observeddifferencesin genetic
variation betweenthese species(chapters 4 & 8). However, P. píli/èrunt and P.
iuniperinumaregenerallyobservedin moredynamichabitats.wherethe turnoverof genets
122
iq hiol
.formo
mioht
or ase)
an imt
(chap
tumovr
studies
From t
show 1
evoluti
selecti
1999) t
underg
rate of
morphr
bryoph
found t
in adap
Shaw d
Gene
'l'he
di
random
(Wrigh
structu
alsobe
extent(
Ir
spores
limited
Newto
in.rpor
patem
that for
Additio
cushio
indicat
Conse
flow be
Pr
& Pócs
distanc
Discussion
Summarising
: 0.22:chapter4)
retic identityof 1 =
um.P..juniperinum
ls
counter-argument
cies of the genera
on
showing,
1991),
rpter2&4),were
rodologyfor protein
tlly P. commune,P.
cn (Hr:0.025-0.166;
ular plants(Wyatt e/
,rtheless,
comparably
rf othermossspecies
,w leveisof genetic
of thesePolytrichum
r be concludedthat
riable moss species
istamongcongeneric
yatt, 1992;Wyatt &
speciesexaminedin
:sity have also been
mrick& Godt(1990)
n a significanteffect
asshownthat selfing
s (Hamrick& Godt,
(facultativelyselfing)
ioecious(obligatory
: of the higli level of
luivalent to asexual
)ronberg(1998)have
;wereconelatedwith
esof themossgenera
Lbledifferencesin the
Lddioeciousspecies.
d more variablethen
ncesin sextnlity as
ifferencesin genetic
P. pili/enm and P.
thetumoverof genets
is high and sexual reproductionmay prevail, while P. commune,P. uliginosum and P.
.formosumare generally observedin more stable habitats,where asexualreproduction
might be more important.Therefore,the differencesin the predominanceof either sexual
or asexualreproduction,throughdifferencesin life spanand tumover of genets,might be
an important determinantof the level of geneticvariation in these Polytrichum species
of thesedifÍèrencesin life span and genet
(chapter 4). A priori the exact consequences
genetic
structureis not clear, but further experimental
tumover betweenspecieson their
give additionalinsight in this matter.
possibly
might
modelling
computer
studiesand
From the latter two sectionsit can be concludedthat bryophytes,in general.do neither
show particularly lower levels of geneticvariation than vascularplants nor show a low
evolutionaryrate. However. these data are basedon allozymes,which may be (partly)
selectivelyneutral(Tumer et al., 1979;Yamazaki,1984;Bijlsma et al., 19911Stensien'
1999)and, therefore,may not necessarilybe indicativefor the level of variation of genes
undergoingselection(Emos, 1990).The level of geneticvariation and the evolutionary
rate of bryophytesmay be lower for charactersthat have adaptivesignificance,such as
morphological characters,which are consideredto possesslow evolutionary rates in
bryophytes(Anderson,I 963;Crum, 1972;Longton, 1976). On the otherhand,it hasbeen
involved
levelsofgeneticvariationfor characters
foundthat mossesalsoshowsubstantial
(Shawet al., 1987;Ennos,1990;Shaw, 1990.1991;
to local environments
in adaptation
Shaw& Beer,1999).
Genetic Variation among Populations within Species
ï'he distribution of neutral geneticvariation among popuiationsis mainly the result of
randomgeneticdrift within populations,counteractedby gene flow betweenpopulations
(Wright, 195l). Therefore,the level of geneflow is an importantdeterminantof the genetic
structureamongpopulations.Additionally, historic events,as e.g. the glacial periods,have
also been shown to affect the geneticstructureamong many plant populationsto a large
extent(e.g.Taberletet a|.,1998).
In bryophytes,geneflow can be mediatedby asexualpropagation,male gametesand
spores.Generally,asexualdiasporesand male gametesare thought to have a much more
limiteddispersalrangethan spores(Wyatt, 1982,1994:Longton,I 976; Kimmerer,I 991;
Newton& Mishler,1994),and sexuallyproducedspores,therefore,areprobablythe most
important factor facilitating gene flow betweenbryophyte populations.By means of a
patemity analysisof sporophyesusing microsatellitemarkers.it is shown in chapter 7
that for P. Jbrmosummale gametedispersalis restrictedto maximally a few metrcs.
Additionally,the absenceof identicalmulti-locusgenotypesfor spatiallydisjunctmosscushions both within populations and from different populations (chapters 6 & 7)
indicatesthat asexualpropagationof gametophytesis also restrictedto a few metres.
Consequently,male gametesand asexualpropagationare indeedunlikely to facilitate gene
flow betweenpopulationsin this andpossibly(most)otherPolytrichumspecies.
Potentially,sporesare able to disperseover much larger distances(e.g. Van Zanten
& Pócs,l98i) and may, therefore,facilitategeneflow betweenpopulations.Such long
distanceeffective spore dispersalis practically impossibleto measurein the field, but,
t/-J
Chapter9
of an "islandmodei"of migration.it canbe estimatedindirectlyfrom
underthe assumption
the relationshipbetweenthe level of geneticdifferentiation(Fs'r)andthe level of geneflow
(Nz) betweenpopulations(Fsr=lll4Nm+11,Wright 1951). As the examinedPolytrichum
drift is twice as effectivein these
populations
aremainlycomposed
of haploidpopulations.
haploidmossthan in diploid plant populations:F5p1/[]//rr+11 (Whitlock & McCauley
1999).This meansthatthe levelof geneflow hasto be twice as high in haploidbryoph1'te
populationscomparedto diploid plant populationsin order to achievethe same level of
populationdifferentiation.
For most Polyn"ichumspecies(P. .formosum.P. commune,P. uliginosum and P.
pilifertm), which havebeenexaminedfor a largegeographicalrangewithin Europein this
(-FsrÍ 0.1) betweenpopulations.
the absence
study,the low levelsof geneticdifferentiation
of "isolationby distance"andthe lack of clusteringof regionalpopulationsshow thal there
is no cleargeneticstructuringwithin thesespecies(chapter4, 6 & 8). This indicatesthat
extensivegeneflow has beenthe most importantdeterminantof geneticshuctureamong
populationsin these species.In contrast. for P. juniperirum ligh levels of population
differentiation (Fsr : 0.34) and a clear genetic structurehave been observedamong
Europeanpopulations(chapters4 & 8). A clearbifurcationofpopulationsin this species
possiblyshowsthe impact of the recolonisationof Europefrom two diÍferent refugia aÍïer
thelatestglacialperiodin combinationwith the low levelsof geneflow betweenthesetwo
groupsof populationson the geneticstnrctureof this species(chapter8). Wyatt & Derda
(1997) and Derda & Wyatt (1999a.b)observedconsiderablyhigher levels of genetic
differentiationamong North American populationsof P. commune,P. piliferum and P.
.iuniperinumthan those reportedin this thesisfor Europeanpopulationsof these species
(chapter 8). These differencesin the level of population differentiation between the
continentsÍbr eachof thesespeciesmay be explainedby a dilferent impact of the glacial
periodson the geneticstructureof North Americancomparedto Europeanpopulationsof
tlresespecies(Wyatt et al., 1993b).
From the previousparagraphit is evidentthat extensivegene flow is an important
determinantof geneticstructuringwithin Polytrichum species.It appearsthat sporescan
effectively disperse,and prevent substantialgenetic differentiation by randont drift,
betweenpopulationsat least up to 1500 km apart (chapter 8). The fact that for
microsatellitesalpine populations of P. formo.sumwere shown to be substantially
differentiatedboth frorn each other and from lowland popuiations(chapter 8), suggests
that mountainspossiblyact as barriersof geneflow, and reducesporedispersalbetween
populationsto a certainextent.In addition,substantiallevels of geneticdifferentiation
betweenpopuiationsfrom different continentssuggestthat also extremelylong distances
and oceansmay limit the level of gene flow considerably(chapter 8). The latter is
corroboratedby a study on Plagiomniumellipticum (Wyatt et al., 1992),where relatively
higherlevelsof geneticdiflèrentiation
wereobservedbetweenpopulationsfrom the North
American and European continent, than between populations within each of these
continents. Nevertheless,the level of differentiation between continentally disjunct
bryophyte populations is generally much lower than that observed for continentally
disjunctvascularplant populations(Wyatt, 1994),showingthat dispersalof sporesis
facilitatinggeneflow amongcontinentsmore eÍ1èctivelythanseeddispersal.
124
hrr
Gene
Since
Mishk
gelletll
plants
popul
Iarge1
differe
larger
sexua
spatia
locus1
showi
horizo
growth
only at
of obs
(Hobb
infieq
demo
from :
may c
high I
Durin
formc
large
with
of th,
Hofn
to be
ofg
partl,
Cor
Alth
othe
vari
Stor
(fac
con
Gor
SummarisingDiscussion
imatedindirectlyfrom
. thelevelofgeneflow
examinedPolytrichum
ceaseffectivein these
lhitlock & McCauley
h in haploidbryophyte
ievethe samelevel of
P. uliginosumand P.
e within Europein this
theabsence
cpulations.
rlationsshowthat there
8). This indicatesthat
eneticstructureamong
h levelsof population
beenobservedamong
,ulations
in this species
r differentrefugiaafter
low betweenthesetr,'"'o
ter 8). Wyatt & Derda
3herlevels of genetic
e, P. piliferum and P.
ltionsof thesespecies
entiationbetweenthe
I impactof the glacial
ropeanpopulationsof
: flow is an important
lpearsthat sporescan
ron by random drift,
l). The fact that for
r to be substantially
(chapter8), suggests
ore dispersalbetween
teneticdifferentiation
remelylong distances
rter 8). The latter is
)92),whererelatively
ationsfrom the North
rithin each of these
rontinentallydisjunct
red for continentally
lispersal
of sporesis
persal.
Genetic Variation within Populations
Sincemossesare ableto reproduceasexuallyquite easily(Anderson,I 963; Longton, 1976.
Mishler, 1988; During, 1990),bryophye populationscould be expectedto possessa
genetic structureconsiderablydifferent from that of obligatorily sexually reproducing
plants (Murawski & Hamrick, 1990; Shapcott,1995; McFadden,1997).For example.
populationsof clonally reproducingbryophye speciesmay be comprisedof only a few
large genets(individuals).In contrast,high levels of genotypicdiversity (i.e. number of
different multi-locus genotypes= genets)were observedwithin populationsand withln
large moss-cushions
of P. formosum (chapters 6 & 1). This showsthat in P. formosum
sexualreproductionis a much more imporlant determinantof genetic structureat these
spatialscalesthan previouslythought(chapters6 & 7). On the otherhand,identicalmultilocus genotypeswere also frequently observed,but only for spatially clusteredshoots.
showing that in this speciesasexualreproductionoccurs mainly through branchingand
horizontal proliferation of gametophytesthrough undergroundrhizomes (i.e. vegetative
growth of genets;chapter 7). Consequently,asexualreproductionseemsto be important
only at a small,very local scalein P.formosum.
The importanceof sexualreproductionin P..formosumseemsto contradictthe rarity
of observationsof recruitmentof new genetsfrom sporesin naturalbryophy'tepopulations
(Hobbs& Pritchard,1987;Miles & Longton,1987, 1990;Innes, 1990).However,the
infrequent observationof recruitment from spores might be due to the difficulty of
demonstratingits occurrenceQllewton& Mishler, 1994).On the other hand,if recruitment
from sporesis indeedrare, extremelongevity of genets,due to extensiveclonal growth,
may over long time periodsallow for accumulationof genetsin populationsresulting in
high levelsof genofypicdiversitywithin thesepopulations(e.g.Sebens& Thome, 1985;
During & Van Tooren, 1987; Eriksson& Fróborg, 1996). Altematively, as the P.
populationexaminedin chapter7 was growingunderbeechesplantedin 1923,
.formosum
large scaledisturbanceof a then presentdiasporebank by this plantation,in combination
with small scaledisturbanceofthe forest soil every 5 year,due to cutting ofthe branches
of thesetrees, could also accountfor the relatively high genotypic diversity. Similarly,
Hofman(1991)and Cronberg(1996)arguedthat disturbance
of the soil. which is thought
to be importantfor diasporegermination(During, 1997),also resultedin increasedlevels
of genotypic diversity in Plagiothecium undulatum and Sphagnum capillifolium at
particuiarspotsin the population.
Concluding Remarks
AlthoughPolytrichumspeciesshow on averagelower levelsof geneticvariationthan most
other bryophlte species(this thesis;Wyatt & Derda, 1997),in general,mossesseem,
despitetheir predominantlyhaploid life cycle, not to exhibit lower levels of genetic
variation than diploid vascular plants (Wyatt et al., 1989b; Hamrick & Godt, 1990;
Stonebumer et al., 1991). In addition,the levels of genetic variation observedin
(facultatively) asexually reproducing plants, including bryophytes, are, in general,
comparable
plants(Ellstrand& Roose,1987;Hamrick&
to thoseof sexuallyreproducing
Godt, 1990;'Widén& Widén, 1994).This indicatesthat neithera predominantlyhaploid
tl)
Chapter9
ïffffh":J;i"iïltyJ.::&-*ïff-""Jïï:1Ë{::iïï:i.':"'ï."is'"}":-"
:ï
i:l
;:*Ïkp*;ïËËlï:1f
***
that
i:'",*ï,ïï
Iï:i::{
r' a{aliiol to studies
i1""er"
;'#ilï;;i;
^ïi'ï"i"Á
bryophltepopulations
Ë
qïï*'f:*"'ri:ïfr
*ï*',trt4ï'ïJffi:ffi
ffif
Ë"#i"ffiJ*::nïï:;ffi ,ffi
:'-ï'jï,
H,'"ïr"'i'h'.?'ï:tffi
*3:Hïl:ïï';***ï*"dÍ,ïltrfr
ïtrïffi
:'iïiÏ'rïïi
*nïu*rtl;*ï[1,f"fiï*:l
be greatlYfacilitatedbY
t26