How important is intraspecific geneticadmixture to the success of colonisingpopulations?Marc Rius1 and John A. Darling2

1 Ocean and Earth Science, University of Southampton, National Oceanography Centre, European Way, Southampton, SO14 3ZH,

UK2 National Exposure Research Laboratory, United States Environmental Protection Agency, Research Triangle Park, NC 27711, USA

Review

Glossary

Colonisation pressure: the number of species introduced to, or released in, a

single recipient location.

Inbreeding depression: the reduction of fitness caused by mating between

close relatives, often resulting from the unmasking of recessive deleterious

Genetic admixture of divergent intraspecific lineages isincreasingly suspected to have an important role in thesuccess of colonising populations. However, admixtureis not a universally beneficial genetic phenomenon. Se-lection is typically expected to favour locally adaptedgenotypes and can act against admixed individuals,suggesting that there are some conditions under whichadmixture will have negative impacts on populationfitness. Therefore, it remains unclear how often admix-ture acts as a true driver of colonisation success. Here,we review the population consequences of admixtureand discuss its costs and benefits across a broad spec-trum of ecological contexts. We critically evaluate theevidence for a causal role of admixture in successfulcolonisation, and consider that role more generally indriving population range expansion.

Mixing divergent genotypes: a common phenomenonwith complex implicationsIntraspecific genetic admixture (hereafter admixture)occurs when multiple divergent genetic lineages come intogene flow contact and interbreed. A growing number ofstudies show that admixture of previously allopatriclineages is a common phenomenon in nature [1–4], andthat such hybridisation generates novel allelic combina-tions that can be beneficial when certain conditions aremet. Potential benefits resulting from admixture includeshort-term increased population fitness (e.g., heterosis, [5])and increased adaptive potential [6]. These benefits paral-lel in many ways the observed benefits of interspecifichybridisation, which have been dealt with elsewhere(e.g., [7–10]).

Observation of admixture in colonising populations hasled to speculation that admixture has an important role indriving the success of those populations [2,11,12]. By con-trast, some recent studies have questioned the importanceof admixture (e.g., [13–15]), because nonadmixed or even

0169-5347/$ – see front matter

� 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tree.2014.02.003

Corresponding author: Rius, M. (M.Rius@soton.ac.uk).

severely bottlenecked populations can often perform at thesame level as admixed populations. In addition, there aresome theoretical grounds for questioning the assumptionthat admixture might provide an evolutionary advantage[9,16]. For instance, given that natural selection favourslocally adapted genotypes, admixture should often be mal-adaptive [6]. Therefore, it remains unclear whether admix-ture is a true driver of colonisation success (in its broadersense) or rather a consequence of its correlation withfactors known to be associated with colonisation success(e.g., increased propagule pressure through multiple intro-ductions, [17,18]).

Here, we review the population consequences of admix-ture and consider some of the evolutionary and ecologicalfactors determining the costs and benefits of admixturerelative to colonisation success. We explore several ques-tions that are raised by existing theoretical and empiricalevidence. For instance, is admixture generally a driver ofcolonisation success or is it more often a passenger, aphenomenon correlated with other factors that have great-er influence on the outcomes of colonisation events? Underwhat conditions is admixture likely to promote colonisationand range expansion? We discuss potential future direc-tions for research that can substantially improve our abili-ty to answer such questions.

Genetic consequences of admixtureThere are several costs and benefits associated with ad-mixture that can have both immediate and long-termconsequences for colonisation success. One of the mostwidely recognised short-term benefits of admixture is

alleles or the loss of heterozygote advantage (overdominance).

Outbreeding depression: a reduction of fitness in offspring of parents from

different populations, possibly due to disruption of co-adapted gene com-

plexes.

Propagule pressure: the number of viable colonisers introduced to a recipient

environment in a single introduction event (propagule size) and the number of

introduction events (propagule number).

Transgressive phenotype: a phenotype generated in segregating hybrid

populations that is extreme relative to that of their parents.

Trends in Ecology & Evolution, April 2014, Vol. 29, No. 4 233

Box 1. Intraspecific genetic admixture and adaptation

When colonisers enter a novel niche, they frequently experience

environmental conditions outside their parental mean and variance.

This results in environmental filtering that selects individuals [78]

and forces survivors to adjust their phenologies to the new climate

[79], as well as to reconfigure ecological interactions with the

receiving community [80]. Thus, colonisation often brings with it

novel selective pressures, and successful establishment might be

contingent on the ability to respond to those pressures. Therefore, it

is expected that populations locally adapted to parental selective

regimes have relatively low fitness when exposed to such recipient

environments. In such cases, influx of novel alleles is expected to be

beneficial, because it should increase the adaptive potential of the

founding population. This is because admixture provides increased

genetic variation on which selection can act. In particular, rapid

selection has been attributed to high additive genetic variance

conveyed through multiple introductions and admixture [54].

Although there might be limits on the conditions under which

increased genetic variation enhances the likelihood of colonising

populations becoming established (e.g., [80]), in principle admixture

should be particularly favourable in nonparental selective environ-

ments (such as transitional habitats not occupied by parental

populations; e.g., [75]). When a colonised environment is dissimilar

to the ecosystem that initially selected parental genotypes (i.e., low

environmental matching), the fitness benefits of admixture are

expected to be high [6] (Figure 1, main text). Thus, admixture can be

particularly beneficial during periods of range expansion when

novel territory that is not already occupied by differentiated

populations is newly colonised. This dynamic has been observed,

for instance, in colonising populations of the invasive plant

Verbascum thapsus. Observation of multiple invasive populations

revealed that some exhibited fitness-related traits that were

mismatched to their newly colonised habitats [4]. However,

admixture between introduced populations provided an opportunity

to erode this ‘mosaic of maladaptation’ by bringing the alleles

responsible for those traits into novel environments where their

fitness benefits can be realised.

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

heterosis or hybrid vigour (e.g., [5]), broadly defined as thephenotypic superiority of hybrids over their parents [19].Heterosis can also lift inbreeding depression (see Glossa-ry), thus reducing the negative effects of genetic bottle-necks by increasing the frequency of heterozygotes, whichcan mask deleterious mutations [20] or allow expression ofoverdominance [21]. Also critically important might be theemergence of novel genotypes not found in parental popu-lations [3,22]; in some cases, these new genotypic combina-tions can produce transgressive phenotypes well outsideparental norms [1,23].

These genetic phenomena can all have a variety ofpositive influences on fitness, ranging from benefits thatare independent of selective environment to those that arespecific to the novel environments encountered by colonis-ing populations. For instance, Drake [24] demonstratedthat heterosis could enhance postcolonisation populationfitness to such extent that it generates a ‘catapult effect’ inadmixed populations. The heterotic effect of admixture canbe limited to a single generation, because fitness of F1hybrids can differ considerably compared with latercrosses due to the effects of recombination [9]. However,if colonisers gain sufficient transient fitness benefitsthrough heterosis, this might buffer against demographicdangers of initial small population size and reduce thelong-term risk of extinction. Thus, heterosis might lead tosignificant advantages in terms of establishment likeli-hood and be sufficient to put founding populations on atrajectory towards successful colonisation and range ex-pansion [25,26].

The emergence of novel phenotypes can also have animportant role in postcolonisation adaptation and nichedivergence [27]. To the degree that range expansion corre-lates with novel environmental conditions and associatedselective pressures, the emergence of nonparental geno-types might provide opportunities for local adaptation thatwere previously inaccessible. Models of evolution at envi-ronmental range limits predict that local adaptation isunlikely to result from in situ emergence of beneficialalleles in marginal habitats [28]; this suggests that theappearance of novel allelic combinations through multipleintroductions and admixture provides unique opportu-nities for colonising populations to respond to selectionpressures in novel environments.

The primary long-term benefit of admixture is to in-crease overall population genetic variance, resulting inheightened capacity to respond to selective pressures. Ithas long been recognised that genetic diversity shouldincrease the likelihood of successful establishment by in-creasing the probability that at least some colonisinggenotypes are adapted to environmental conditions inrecipient habitats [29]. Some studies suggest that theavailability of genotypic variation is especially importantwhen colonised habitats are particularly novel or challeng-ing [29,30]. The role of admixture in increasing overallgenetic variance has been broadly discussed in terms ofovercoming the potentially deleterious influences of found-er effects [3,4], and several recent studies suggest thatenhanced genetic diversity associated with admixture doesindeed provide the raw material for local adaptation innewly colonised habitats (see below and Box 1).

234

Few studies have examined the relative importance ofshort-term (e.g., heterosis) versus long-term (e.g., adapta-tion) benefits of admixture. Facon et al. [5] analysed thepotential role of admixture in driving multiple invasions ofthe freshwater snail Melanoides tuberculata, and foundthat observed phenotypic changes in admixed individualswere unlikely to arise by selection, suggesting that, in thiscase, the fitness benefits of admixture are primarily heter-otic. A more recent study on genomic admixture in theweedy plant Silene vulgaris similarly indicated predomi-nantly short-term, heterotic fitness effects in determiningpopulation fitness [31]. Additional studies that examinethe relative roles of long- and short-term fitness benefitsshould help to better understand the causal linkages be-tween admixture and colonisation success (see ‘Statisticalevidence’ below).

Admixture can also have immediate negative impactson fitness. Most severely, crosses between divergent pa-rental lineages can result in reduced viability or fertility inthe case of serious genetic incompatibilities [23], althoughsuch costs are expected more frequently with interspecifichybridisation. More commonly, outbreeding depressioncan result from a selective disadvantage of intermediategenotypes or loss of advantageous parental traits [32],resulting in reduced population fitness. When the recipientenvironment is similar to the parental range environment,populations might be ‘pre-adapted’ to the new system [16]

Fitness

Low HighLo

wEn

vironmen

tal

matching

High

Low

High

(A)

(B)

(C)

(D)

Gene�c admixtureTRENDS in Ecology & Evolution

Figure 1. Conceptual diagram of evolutionary responses of both parental and

admixed populations to different environmental regimes. (A) Parental types

(unadmixed) in a novel environment. Fitness is low (inability to respond to novel

selection pressures). (B) Parental types (unadmixed) in a parental environment.

Fitness is high (pre-adaptation). (C) Admixed genotypes in a parental environment.

Fitness is low (disruption of pre-adapted gene complexes). (D) Admixed genotypes

in a novel environment. Fitness is high (increased diversity and response to

selection).

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

and, under such circumstances, selection should actagainst the hybridisation of ecologically divergent subpo-pulations (Figure 1). In effect, selection for parental typescan put constraints on the dilution of locally adapted genepools, such that admixture could have substantial fitnesscosts [33].

This clearly has implications for range expansions gen-erally. If populations are best adapted to the environmentat the range centre, then marginal or colonising popula-tions can experience nonparental selective pressures thatfavour admixture. In such circumstances, gene flow fromcore populations can have maladaptive effects [34,35];alleles that could increase fitness at the margins can incurfitness costs at the range centre (and vice versa), andselection against those alleles combined with gene flowfrom core to margin could hinder adaptation to marginalenvironments (high proportion of poorly adaptedmigrants). However, when marginal populations face asudden change in environmental and/or ecological condi-tions that were previously only found outside the speciesrange, rapid adaptation might occur [28]. Such conditionsare likely to be met under certain scenarios, including bothclimate change and when species colonise distant regions,because the new conditions are likely to differ greatly fromthose of source populations; in such cases, the increase ingenetic variation resulting from multiple introductionsand admixture can further enhance adaptive responses[36].

Nongenetic factors associated with admixtureEven in the absence of any genetic effects, it is likely thatthere is some causal relation between successful colonisa-tion and multiple introductions [37–40]. Admixturerequires the arrival of previously divergent evolutionarylineages in the same place, which implies multiple sourcesof colonists to the novel habitat. One implication of this isthat those regions where admixture is observed are also,

all else being equal, regions subject to heightened propa-gule pressure relative to regions that receive colonists fromonly single sources. Given that regions with high propagulepressure are those where multiple lineages are most likelyto come into contact and hybridise (e.g., locations wheremany vectors and pathways of introduction converge,[41,42]), one might expect a strong correlation betweenpropagule pressure and admixture. Lending support forthis hypothesis, Kajita et al. [43] recently demonstratedthat genetic diversity in an invasive ladybeetle was strong-ly positively correlated with historical records of propagulesize. Thus, the demographic benefits of propagule pressureare expected to be frequently confounded with the poten-tial genetic benefits (admixture and increased geneticdiversity due to multiple introductions), suggesting thatthe latter is a passenger as well as a driver of successfulcolonisation [29]. In general, there does exist a ‘null hy-pothesis’ to explain the frequent observation of admixturein introduced populations; admixture is not driving suc-cess, but rather coming along for the ride with the truedriver, propagule pressure. Unfortunately, as we explorein more detail below, it remains challenging to test thishypothesis.

Does intraspecific genetic admixture contribute tocolonisation success?Much of what we now know of the role of admixture incolonisation and population expansion derives from thestudy of biological invasions; indeed, the assertion thatadmixture can contribute to invasiveness has become com-monplace [3,6,44]. However, finding direct evidence for itscausal role in colonisation success has proven difficult.

Observational evidence

Introduced populations often exhibit levels of geneticdiversity that are attributed to multiple introductionsand admixture of differentiated sources [45,46]. In a pio-neering study, Kolbe et al. [47] observed heightened mi-tochondrial haplotype diversity in introduced populationsof the lizard Anolis sagrei, owing to introductions from atleast eight different sources. Although other studies foundthat coexistence of divergent lineages does not alwaysresult in admixture [14,15,48], additional investigationof A. sagrei identified numerous introduced populations inwhich admixed individuals were clearly derived frominterbreeding of different parental stocks [12]. Admixtureis particularly prevalent in species with low dispersalcapabilities and widespread ranges [38,49–52], a paradoxthat has typically been explained as a result of anthropo-genic transport.

Inferences of admixture are made by using severalavailable analytical tools (Box 2) that assess inheritancepatterns of either multiple genomes (e.g., nuclear andplastid) or multiple markers within a single genome. Forinstance, both Rosenthal et al. [39] and Keller et al. [11]analysed shifts in cytonuclear associations using nuclearmicrosatellites and chloroplast markers to demonstrateadmixture and introgression among introduced popula-tions of plant species, whereas Nolte et al. [53] analyseddistributions of mitochondrial haplotypes, single nucleo-tide polymorphisms, and microsatellite loci to identify

235

Box 2. Detection of intraspecific genetic admixture

There are several methods and analytical tools (reviewed in [81,82])

that can be used to detect admixture. In the simplest case, admixture

is inferred from the presence of independent evolutionary lineages in

a single introduced population, sometimes with evidence that those

lineages derive from geographically distinct sources (e.g., [83,84]).

Another way is to analyse gene genealogies. Expanding populations

typically have a star-shaped gene genealogy [85], with a higher

number of unique genotypes than expected in a stationary population

[86], although singletons can also be indicative of the admixture of

diverging genotypes. Mismatch distributions are another useful

method because they are sensitive to demographic events. If the

frequency distributions of pairwise differences are bimodal or multi-

modal, these results likely indicate admixture [69,83,87,88]. Other

analytical measures (e.g., Fu’s FS and Tajima’s D) can indicate the

presence of admixture and range expansions [69,89].

There are several pieces of software that can be used to infer,

visualise, simulate, and evaluate admixture and can be used with

sequence and microsatellite data, as well as genomic approaches

(admixture along the genome). BAPS [90] and STRUCTURE [91] use

Bayesian clustering analysis to identify individuals with admixed

genotypes (Figure I); principle components analysis [92] and dis-

criminant analysis of principal components [93] are used to infer

admixture through identifying individuals that are intermediate

between populations; DIYABC [94] performs model-based inference

of complex evolutionary scenarios and admixture can be determined

by contrasting competing scenarios (Figure II); HAPMIX [95] scans for

actual chromosomal recombinants using dense genetic data;

HybRIDS (http://www.norwichresearchpark.com/HybRIDS) visualises

recombination blocks in genome sequence data; and finally, HYBRI-

DLAB [96] and SPLATCHE2 [97] simulate genotypes between pairs of

hybridising populations and molecular diversity under complex

demographic scenarios, respectively.

Detecting admixture can be a challenging task when propagules are

transported with human-mediated transport. In such cases, it is likely

that genotypes are so extensively shuffled that the genetic signal of

the native range is eroded, and that novel admixed genotypes are

introduced into native populations.

European African Americans AfricanTRENDS in Ecology & Evolution

Figure I. STRUCTURE output (K = 2, predetermined genetic clusters, blue and

red) obtained from human populations of Africa, America, and Europe. Modified

from [98].

Introduced

Scenario 1 Scenario 2 Scenario 3

Introduced IntroducedSource 1 Source 2 Source 2 Source 2Source 1 Source 1TRENDS in Ecology & Evolution

Figure II. Set of competing scenarios used to infer the most likely origin of populations within the introduced range (Introduced) using two possible source populations

(Source 1 and Source 2). Scenarios 1 and 2 are unadmixed, whereas Scenario 3 incorporates admixture. Adapted from [38].

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

admixture in an invasive fish. The rapidly growing numberof studies such as these suggest that the observed geneticpatterns have some role in successful establishment and/orspread of introduced populations.

Mechanistic evidence

A growing number of studies demonstrate linkage betweenadmixture and phenotypic outcomes with potential rele-vance to colonisation success (Box 3). Such correlationshave been observed in non-indigenous plants [31,54],invertebrates [55,56], and vertebrates [12,53,57]. Convinc-ing correlation of admixture with phenotypic changes thatinfluence population fitness provides a putative mechanis-tic connection between admixture and successful establish-ment. Research has now shown that admixture can shapephenotypic variation in traits that could prove influentialin determining colonisation success, including morphology[57], life-history characteristics [56], fruit production [31],and several other fitness-related traits [54]. In at least one

236

case, admixture between previously allopatric lineages hasbeen shown to result in measurable phenotypic differencesthat apparently enabled species expansion into previouslyuncolonised habitats [53]. Assessment of trait variation ismore challenging than observation of admixture, oftenrequiring manipulation in common gardens or other ex-perimental systems. Although these efforts continue toestablish plausible links between admixture and colonisa-tion, it remains to be seen whether such links are the rulefor admixed populations.

Statistical evidence

Not all introduced populations are admixed and, even with-in the same study taxon, it is possible to observe successfulcolonisations and range expansions by both admixed andnonadmixed populations. For instance, Chapple et al. [13]questioned the importance of admixture in driving invasion,based on the observation that patterns of admixture failedto correlate with patterns of colonisation success among

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

introduced populations of the skink Lampropholis delicata.Similar observations have been made in the case of theglobally invasive marine crab Carcinus maenas, in whichone of the most successful invasive populations is also one ofthe most genetically depauperate [58,59].

These studies also reveal the potential confoundingeffects of other factors. In the case of C. maenas, a recentintroduction to Newfoundland (northwest Atlantic) wasshown to derive from within an admixture zone betweentwo other colonising populations in the same region [60]. Atfirst glance, this raises the possibility of some fitnessbenefits favouring admixed populations as sources for

Box 3. Potential role of intraspecific genetic admixture in colonis

Several recent studies have revealed that patterns of admixture

resulting from multiple introductions can have effects with potential

relevance to population fitness, offering further support for the

hypothesis that admixture drives successful establishment and

spread of populations.

Studies of highly admixed invasive populations of the Caribbean

lizard Anolis sagrei (Figure IA) have revealed significant phenotypic

differences in the number of toe-pad lamella and body size. The latter

has been shown to correlate strongly with the number of admixed

sources (Figure IB).

Another example is from the analysis of an experimental

population of the Asian ladybird beetle Harmonia axyridis

(Figure IIA), which has developed to mimic admixture between

non-indigenous North American populations and a strain intention-

ally introduced for biocontrol purposes. Researchers have revealed

associations of admixture with fitness-related traits (such as

fecundity, Figure IIB) that might have promoted the invasiveness

of European populations.

Similar results have been observed in the case of admixed native

populations arising in natural contact zones. For instance, evidence of

postglacial admixture was observed in populations of the European

aspen, Populus tremula (Figure IIIA), and admixture patterns ex-

plained the strength of selection for traits measured in common

garden experiments. Selection differentials [covariance between bud

set (a trait of known adaptive significance) and relative growth rate]

0

5

10

15

20

25

30

35(A) (B)

Fitn

ess i

ndex

(egg

s per

day

)

Figure II. Admixture and fecundity (A) Asian ladybird beetle Harmonia axyridis. Repr

types (America, Biocontrol, Admixed, and Europe). Vertical bars denote standard erro

Modified from [55].

particularly successful colonists of new territories. Howev-er, the most likely source region in the admixture zone wasalso proved to be the source of heaviest shipping traffic inNewfoundland, suggesting that propagule pressure can bethe more important driver of the observed range expan-sion. In a similar example, study of non-native rainbowtrout Oncorhynchus mykiss in Chile revealed substantialadmixture from multiple farm escapes, leading the authorsto suggest a role for admixture in facilitating establish-ment [61]. However, they also noted strong correlationbetween propagule pressure and the frequency of non-natives, indicating that escapes from aquaculture facilities

ation success

varied widely over a latitudinal gradient, showing marked elevation in

the contact zone (Figure IIIB).

2Number of popula�ons sampled

(A) (B)

Body

size

(cm

)3.9

4

4.1

3 4 5

TRENDS in Ecology & Evolution

Figure I. Admixture and adult size. (A) Caribbean lizard Anolis sagrei.

Reproduced, with permission, from Ianare Se vi. (B) Correlation between body

size and number of admixed sources. Modified from [57].

America Biocontrol Admixed Europe

A ABB B

TRENDS in Ecology & Evolution

oduced, with permission, from Bruce Marlin. (B) Fecundity of the different cross

r, and different letters above the bars indicate significantly different treatments.

237

56

–0.5

0

0.5

(A) (B)

1

58 60 62

La�tudeCO

V (w

,z)

64

TRENDS in Ecology & Evolution

Figure III. Admixture across latitude. (A) European aspen, Populus tremula. Reproduced, with permission, from Erlend Bjørtvedt. (B) Selection differentials [COV(w,z)]

across latitude. The admixture cline is shaded. Modified from [99].

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

likely drove the spread. Such correlations between propa-gule pressure and admixture can confound attempts toinfer a causal role for the latter in driving successfulcolonisations.

Unfortunately, it remains difficult to disentangle suchcorrelations. For one thing, studies in which species areintroduced to analogous ecosystems via dissimilar coloni-sation histories (e.g., single and multiple introductions) arerelatively difficult to come by (but see [13,47,59,62]). Evenwhen such systems can be identified, the challenge ofisolating for the effects of propagule pressure and admix-ture is likely to be a prodigious one. Other confoundingfactors also exist; for instance, admixture can also becorrelated with colonisation pressure (the species diversityintroduced to a recipient system [63]), adding anotherpotential driver to the mix.

One possible approach to this problem is to designcontrolled experiments in systems that allow independentmanipulation of propagule pressure and admixture. Forinstance, Hufbauer et al. [29] applied a factorial experi-mental design to explore the relative roles of the number offounders and genetic background (degree of inbreeding) insuccessful colonisation by the whitefly Bemisia tabaci,demonstrating that positive effects of propagule pressurecan indeed be driven by associated genetic factors. Similarexperimental approaches could be utilised to test directlyfor correlations between intraspecific admixture and colo-nisation success while controlling for effects of propagulepressure. Another approach would be to investigate sys-tems in which a large number of introductions have beenmade with sufficient historical knowledge to draw infer-ences regarding both propagule size and admixture levels.Zenni et al. [64] adopted such an approach to investigatethe role of propagule diversity on invasion success in Pinusspecies in Brazil, revealing that the number of sourcepopulations and number of introduced individuals wereboth strong predictors of colonisation success.

238

Overall, although mechanistic evidence is proving valu-able in this regard, the greatest remaining challenge is toestablish statistically robust correlations between degreeof recombination within populations and colonisation suc-cess [31]. To our knowledge, no study has yet attempteddirectly to answer the critical question: do admixed popu-lations colonise new habitats and expand their rangesmore frequently and more rapidly than non-admixed popu-lations? The studies described above illustrate the impor-tance of systems that allow estimation of both admixturelevels (either directly or by some proxy measurement) aswell as confounding factors such as propagule pressure fora large number of introduction events; they also highlightthe importance of assessing the relative rate of colonisationsuccess. Such information will be available in the case ofmany experimental manipulations and in certain rarecases in which historical records are sufficiently robust.Unfortunately, few studies have been able to explore theimportance of genetic factors in determining colonisationsuccess in systems where the rate of success can be reliablyestimated [65].

What can we learn about natural range expansions?Natural and human-assisted range expansions sharemuch in common, including many of the same facilitatorsand barriers to successful colonisation (Figure 2). Themost obvious difference is the role of human vectors inovercoming barriers to the arrival of propagules in newhabitats. The often-dramatic differences in the dispersalof colonists to new sites in the two cases [66] result inseveral differences in the degree of connectivity betweenperipheral and central populations of the species range.One would expect demographic, genetic, and environmen-tal connectivity between peripheral and central popula-tions to be typically greater in the case of natural rangeexpansions than in the case of sporadic human-mediatedtransport. In addition, demographic rescue effects should

• Anthropogenic vectors• Climate change*

• Propagule pressurei

• Geography and/or dispersal limita�on

• Environmental mismatch• Bio�c resistance

• Low dispersal capacity

• Disrup�on of pre-adapted gene complexes

• Low gene�c diversity (poor response to selec�on)

• Allee effects

Parental popula�on(s)

Arrival

Survival

Reproduc�on

Persistence

Spread

• Propagule pressureii

• Heterosis and/or overdominance• Cul�va�on

• Propagule pressureiii

• High dispersal capacity• Propagule pressureiv

• High gene�c diversity (inc. response to selec�on)

• Pre-adapta�on• Cul�va�on

• Cul�va�on

• Anthropogenic vectors

• Anthropogenic disturbance

Facilitators Barriers

Colonisa�on stages

• Hybrid breakdown and/or outbreeding depression

TRENDS in Ecology & Evolution

Figure 2. The stages of colonisation, indicating barriers to successful passage between stages (red, right) and facilitators (green, left). Facilitators shown in italics are

specifically associated with anthropogenic introductions, and * indicates that the factor can have natural or anthropogenic origins, or both. Propagule pressure indicates the

arrival of new colonists from the parental population(s), and has several potential roles: (i) high propagule pressure increases the likelihood that some proportion of

colonists survive initial introduction; (ii) propagule pressure increases the initial population density and reduces the likelihood of Allee effects; (iii) recolonisation from the

parental populations can help overcome low initial reproductive rates and founder effects; and (iv) establishment of multiple colonisation foci through recurrent

introductions can facilitate spread through metapopulation dynamics. For further details, see the main text.

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

be more pronounced, reducing the likelihood of stochasticextinctions of colonising populations. Finally, environ-mental matching between parental and newly colonisedhabitats is likely to be higher. Nevertheless, the degree ofconnectivity observed can vary widely. Even for naturalrange expansions, peripheral populations might be sub-stantially isolated from core populations and subject to theecological and evolutionary forces associated with thatisolation [34,35].

Admixture in natural range expansions is already awidely appreciated phenomenon. For instance, a largenumber of empirical studies have found admixture deriv-ing from multiple glacial refugia following postglacialrecolonisations or recent climate change. These patternshave been observed in various taxa [25,67–72]. In many ofthese cases, admixture of colonists from multiple refugiaresulted in significant increases in genetic diversity rela-tive to glacial relicts [25,68,72], although in some casesstrong founder effects associated with colonisation havelikely led to bottlenecks and limited diversity even inadmixed populations [69,70]. One wide-ranging study ofmultiple European tree species found that genetic diversi-ty generally peaked not in glacial refugia (the expectedcentres of diversity, at the core of the species range), butrather where multiple divergent lineages likely admixedafter recolonisation [73].

Despite these observations, there has been little re-search to assess the evolutionary implications of admix-ture in the context of natural range expansions. Thus,only a few studies have questioned whether admixturehas a causal role in the success of those colonising popu-lations. In one study, Schmitt and Mu ller [70] examinedadmixed postglacial populations of the butterfly Erebiamedusa and found fitness levels comparable to parentalpopulations. An investigation of northern Europeanpopulations of the moor frog Rana arvalis suggested thatincreased diversity in a contact zone between multiplepostglacial colonisations could have provided the rawmaterial for adaptive differentiation of colour morphs[68]. In a recent study, Keller and Taylor [31] experimen-tally tested for correlations between admixture and fit-ness in S. vulgaris and found that admixed introducedgenotypes in North America exhibited heightened fitnessrelative to unadmixed European populations, thus sup-porting the conclusion that admixture promotes colonisa-tion. Most interesting, however, were results withCentral European populations that derived from multipleglacial refugia and exhibited admixture. These popula-tions showed none of the fitness benefits observed in theintroduced populations, leading the authors to hypothe-sise a major role for short-term fitness benefits of admix-ture (e.g., heterosis) as opposed to long-term benefits,

239

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

such as response to selection, which would have resultedinstead in highly fit postglacial populations. Studies suchas this, which examine the relative fitness of naturallyadmixed populations (Box 3), are important to assess thegenerality of lessons learned from biological invasions.Particularly interesting is the question of whetheradmixed postglacial populations serve more frequentlyas sources of successful colonisations than do their unad-mixed parental populations. This prediction of the hy-pothesis that admixture has a causal role in colonisationsuccess should be testable by examining cases similar tothose of S. vulgaris, in which glaciation or other vicariantphenomena provide potential sources for colonising popu-lations with various degrees of both admixture and ge-netic diversity.

Concluding remarks, synthesis, and future directionsThe mixing of divergent genetic pools significantly altersthe evolutionary trajectories of colonising populations andcreates new opportunities for range expansion. Althoughwe show that biological invasions and natural rangeexpansions are similar in nature, the potential benefitsof admixture are most obvious in the former. In part, this islikely due to the sheer number of opportunities for admix-ture that arise because of anthropogenic shuffling of evo-lutionary lineages. However, it can also result from the factthat large differences in the adaptive landscapes of recipi-ent and source environments limit the fitness benefits oflocally adapted genotypes and magnify both short- andlong-term benefits of admixture. Although the co-occur-rence of divergent lineages in sympatry does not necessar-ily mean introgression [15,74,75], a growing number ofstudies have detected admixture directly and recognise itspotential to lead to fitness and/or phenotypic consequencesand adaptive responses in colonising populations[31,54,56,57]. If admixture does increase fitness, it cansubsequently lead to larger and more productive popula-tions with lower probabilities of extinction, and ultimatelyincrease the chance of range expansion. However, there is ascarcity of studies that test hypotheses concerning the roleof admixture on phenotypic change, population fitness andrange expansion.

There is a need to test rigorously hypotheses regardingthe role of admixture driving colonisation success andpopulation expansion to understand fully the relative im-portance of abiotic, biotic, and genetic factors. For example,it remains unclear whether admixture has a causal role inpopulation expansion in both natural and artificial rangeexpansions. The use of reciprocal transplant experimentsto assess population performance at different levels ofrecombination [colonisations with and without admixture(e.g., [13])] has the potential to advance understandingsubstantially. This, combined with the wider use of next-generation sequencing (e.g., [76,77]), has the potential todisentangle the drivers of population expansion. For ex-ample, whole-transcriptome sequencing can be used toobtain data directly derived from functional genomic ele-ments that are expressed in tissues and represent a sig-nificant proportion of adaptive variation. Information onthe sequence variation of admixed versus parental tran-scriptomes would allow for the inference of gene expression

240

patterns that can be relevant to environmental responseand adaptation.

Perhaps the largest remaining challenge is to test sta-tistically whether admixed populations are more likely tobecome successful colonisers than nonadmixed popula-tions. Such research will be particularly important indisentangling the effects of admixture and nongeneticfactors, especially propagule pressure, on colonisation suc-cess. Identifying appropriate model systems to addressthese questions might well be facilitated by the growingnumber of studies that explore patterns of admixtureamong introduced populations.

AcknowledgementsThe idea for this review sparked during the international workshop‘Molecular Tools for Monitoring Marine Invasive Species’, held in Lecce,Italy, in September 2012. We would like to thank the organisers of thisworkshop, most especially Stefano Piraino. We are grateful to threeanonymous reviewers for providing comments that greatly improved thefinal version of the manuscript. The manuscript has been subjected toAgency review and approved for publication. The views expressed arethose of the authors and do not necessarily reflect the policy positions ofthe Environmental Protection Agency.

References1 Bell, M.A. and Travis, M.P. (2007) Hybridization, transgressive

segregation, genetic covariation, and adaptive radiation. TrendsEcol. Evol. 20, 358–361

2 Lawson Handley, L.J. et al. (2011) Ecological genetics of invasive alienspecies. BioControl 56, 409–428

3 Roman, J. and Darling, J.A. (2007) Paradox lost: genetic diversity andthe success of aquatic invasions. Trends Ecol. Evol. 22, 454–464

4 Dlugosch, K.M. and Parker, I.M. (2008) Founding events in speciesinvasions: genetic variation, adaptive evolution, and the role ofmultiple introductions. Mol. Ecol. 17, 431–449

5 Facon, B. et al. (2005) Hybridization and invasiveness in the freshwatersnail Melanoides tuberculata: hybrid vigour is more important thanincrease in genetic variance. J. Evol. Biol. 18, 524–535

6 Verhoeven, K.J.F. et al. (2011) Population admixture, biologicalinvasions and the balance between local adaptation and inbreedingdepression. Proc. R. Soc. B: Biol. Sci. 278, 2–8

7 Rieseberg, L.H. et al. (2007) Hybridization and the colonization of novelhabitats by annual sunflowers. Genetica 129, 149–165

8 Ellstrand, N.C. and Schierenbeck, K.A. (2000) Hybridization as astimulus for the evolution of invasiveness in plants? Proc. Natl.Acad. Sci. U.S.A. 97, 7043–7050

9 Arnold, M.L. and Hodges, S.A. (1995) Are natural hybrids fit or unfitrelative to their parents? Trends Ecol. Evol. 10, 67–71

10 Seehausen, O. (2004) Hybridization and adaptive radiation. TrendsEcol. Evol. 19, 198–207

11 Keller, S.R. et al. (2012) Bayesian inference of a complex invasionhistory revealed by nuclear and chloroplast genetic diversity in thecolonizing plant, Silene latifolia. Mol. Ecol. 21, 4721–4734

12 Kolbe, J.J. et al. (2008) Admixture determines genetic diversity andpopulation differentiation in the biological invasion of a lizard species.Biol. Lett. 4, 434–437

13 Chapple, D.G. et al. (2013) Divergent introduction historiesamong invasive populations of the delicate skink (Lampropholisdelicata): has the importance of genetic admixture in the successof biological invasions been overemphasized? Divers. Distrib. 19,134–146

14 Dutech, C. et al. (2012) The chestnut blight fungus world tour:successive introduction events from diverse origins in an invasiveplant fungal pathogen. Mol. Ecol. 21, 3931–3946

15 Ordonez, V. et al. (2013) Mixed but not admixed: a spatial analysis ofgenetic variation of an invasive ascidian on natural and artificialsubstrates. Mar. Biol. 160, 1645–1660

16 Lenormand, T. (2002) Gene flow and the limits to natural selection.Trends Ecol. Evol. 17, 183–189

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

17 Lockwood, J.L. et al. (2005) The role of propagule pressure inexplaining species invasions. Trends Ecol. Evol. 20, 223–228

18 Simberloff, D. (2009) The role of propagule pressure in biologicalinvasions. Ann. Rev. Ecol. Evol. Syst. 40, 81–102

19 Lippman, Z.B. and Zamir, D. (2006) Heterosis: revisiting the magic.Trends Genet. 23, 60–66

20 Keller, L.F. and Waller, D.M. (2002) Inbreeding effects in wildpopulations. Trends Ecol. Evol. 17, 230–241

21 Fitzpatrick, B.M. and Shaffer, H.B. (2007) Hybrid vigor between nativeand introduced salamanders raises new challenges for conservation.Proc. Natl. Acad. Sci. U.S.A. 104, 15793–15798

22 Lavergne, S. and Molofsky, J. (2007) Increased genetic variation andevolutionary potential drive the success of an invasive grass. Proc.Natl. Acad. Sci. U.S.A. 104, 3883–3888

23 Johansen-Morris, A.D. and Latta, R.G. (2006) Fitness consequences ofhybridization between ecotypes of Avena barbata: hybrid breakdown,hybrid vigor, and transgressive segregation. Evolution 60, 1585–1595

24 Drake, J.M. (2006) Heterosis, the catapult effect and establishmentsuccess of a colonizing bird. Biol. Lett. 2, 304–307

25 Sakaguchi, S. et al. (2011) Lineage admixture during postglacialrange expansion is responsible for the increased gene diversity ofKalopanax septemlobus in a recently colonised territory. Heredity107, 338–348

26 Praebel, K. et al. (2013) Invasion genetics of vendace (Coregonus albula(L.)) in the Inari-Pasvik watercourse: revealing the origin andexpansion pattern of a rapid colonization event. Ecol. Evol. 3, 1400–1412

27 Rieseberg, L.H. et al. (1999) Transgressive segregation, adaptation andspeciation. Heredity 83, 363–372

28 Bridle, J.R. and Vines, T.H. (2007) Limits to evolution at rangemargins: when and why does adaptation fail? Trends Ecol. Evol. 22,140–147

29 Hufbauer, R.A. et al. (2013) Role of propagule pressure in colonizationsuccess: disentangling the relative importance of demographic, geneticand habitat effects. J. Evol. Biol. 26, 1691–1699

30 Ahlroth, P. et al. (2003) Founder population size and number of sourcepopulations enhance colonization success in waterstriders. Oecologia137, 617–620

31 Keller, S.R. and Taylor, D.R. (2010) Genomic admixture increasesfitness during a biological invasion. J. Evol. Biol. 23, 1720–1731

32 Lynch, M. (1991) The genetic interpretation of inbreeding depressionand outbreeding depression. Evolution 45, 622–629

33 Angert, A.L. et al. (2008) Using experimental evolution to investigategeographic range limits in monkeyflowers. Evolution 62, 2660–2675

34 Holt, R. (2003) On the evolutionary ecology of species’ ranges. Evol.Ecol. Res. 5, 159–178

35 Hardie, D.C. and Hutchings, J.A. (2010) Evolutionary ecology at theextremes of species’ ranges. Environ. Rev. 18, 1–20

36 Sexton, J.P. et al. (2009) Evolution and ecology of species range limits.Ann. Rev. Ecol. Evol. Syst. 40, 415–436

37 Blair, A.C. et al. (2012) Hybridization and invasion: an experimentaltest with diffuse knapweed (Centaurea diffusa Lam.). Evol. Appl. 5, 17–28

38 Rius, M. et al. (2012) Tracking invasion histories in the sea: facingcomplex scenarios using multilocus data. PLoS ONE 7, e35815

39 Rosenthal, D.M. et al. (2008) Evidence for multiple sources of invasionand intraspecific hybridization in Brachypodium sylvaticum (Hudson)Beauv. in North America. Mol. Ecol. 17, 4657–4669

40 Kolbe, J.J. et al. (2007) Multiple sources, admixture, and geneticvariation in introduced Anolis lizard populations. Conserv. Biol. 21,1612–1625

41 Carlton, J.T. (1987) Patterns of transoceanic marine biologicalinvasions in the Pacific Ocean. Bull. Mar. Sci. 41, 452–465

42 Suarez, A.V. et al. (2005) The role of opportunity in the unintentionalintroduction of nonnative ants. Proc. Natl. Acad. Sci. U.S.A. 102,17032–17035

43 Kajita, Y. et al. (2012) A population genetic signature of humanreleases in an invasive ladybeetle. Mol. Ecol. 21, 5473–5483

44 Duchen, P. et al. (2013) Demographic inference reveals African andEuropean admixture in the North American Drosophila melanogasterpopulation. Genetics 193, 291–301

45 Geller, J.B. et al. (2010) Genetic perspectives on marine biologicalinvasions. Ann. Rev. Mar. Sci. 2, 367–393

46 Chun, Y.J. et al. (2010) Gene flow and population admixture as theprimary post-invasion processes in common ragweed (Ambrosiaartemisiifolia) populations in France. New Phytol. 185, 1100–1107

47 Kolbe, J.J. et al. (2004) Genetic variation increases during biologicalinvasion by a Cuban lizard. Nature 431, 177–181

48 Gunn, B.F. et al. (2011) Independent origins of cultivated coconut(Cocos nucifera L.) in the old world tropics. PLoS ONE 6, e21143

49 Zhan, A. et al. (2012) Complex genetic patterns in closely relatedcolonizing invasive species. Ecol. Evol. 2, 1331–1346

50 Pineda, M.C. et al. (2011) The whereabouts of an ancient wanderer:global phylogeography of the solitary ascidian Styela plicata. PLoSONE 6, e25495

51 Lombaert, E. et al. (2010) Bridgehead effect in the worldwide invasionof the biocontrol harlequin ladybird. PLoS ONE 5, e9743

52 Goldstien, S.J. et al. (2011) Global phylogeography of the widelyintroduced north west Pacific ascidian Styela clava. PLoS ONE 6,e16755

53 Nolte, A.W. et al. (2005) An invasive lineage of sculpins, Cottus sp.(Pisces, Teleostei) in the Rhine with new habitat adaptations hasoriginated from hybridization between old phylogeographic groups.Proc. R. Soc. B: Biol. Sci. 272, 2379–2387

54 Chun, Y.J. et al. (2011) Adaptive divergence for a fitness-related traitamong invasive Ambrosia artemisiifolia populations in France. Mol.Ecol. 20, 1378–1388

55 Turgeon, J. et al. (2011) Experimental evidence for the phenotypicimpact of admixture between wild and biocontrol Asian ladybird(Harmonia axyridis) involved in the European invasion. J. Evol.Biol. 24, 1044–1052

56 Facon, B. et al. (2008) High genetic variance in life-history strategieswithin invasive populations by way of multiple introductions. Curr.Biol. 18, 363–367

57 Kolbe, J.J. et al. (2007) Differential admixture shapes morphologicalvariation among invasive populations of the lizard Anolis sagrei. Mol.Ecol. 16, 1579–1591

58 Tepolt, C.K. et al. (2009) European green crabs (Carcinus maenas) inthe northeastern Pacific: genetic evidence for high populationconnectivity and current-mediated expansion from a singleintroduced source population. Divers. Distrib. 15, 997–1009

59 Darling, J.A. et al. (2008) Genetic patterns across multipleintroductions of the globally invasive crab genus Carcinus. Mol.Ecol. 17, 4992–5007

60 Blakeslee, A.M.H. et al. (2010) A hitchhiker’s guide to the Maritimes:anthropogenic transport facilitates long-distance dispersal of aninvasive crab to Newfoundland. Divers. Distrib. 16, 879–891

61 Consuegra, S. et al. (2011) Winning the invasion roulette: escapes fromfish farms increase admixture and facilitate establishment of non-native rainbow trout. Evol. Appl. 4, 660–671

62 Rius, M. and Shenkar, N. (2012) Ascidian introductions through theSuez Canal: the case study of an Indo-Pacific species. Mar. Pollut. Bull.64, 2060–2068

63 Lockwood, J.L. et al. (2009) The more you introduce the more you get:the role of colonization pressure and propagule pressure in invasionecology. Divers. Distrib. 15, 904–910

64 Zenni, R.D. and Simberloff, D. (2013) Number of source populations asa potential driver of pine invasions in Brazil. Biol. Invasions 15, 1623–1639

65 Zenni, R.D. and Nunez, M.A. (2013) The elephant in the room: the roleof failed invasions in understanding invasion biology. Oikos 122, 801–815

66 Ricciardi, A. (2007) Are modern biological invasions an unprecedentedform of global change? Conserv. Biol. 21, 329–336

67 Krehenwinkel, H. and Tautz, D. (2013) Northern range expansion ofEuropean populations of the wasp spider Argiope bruennichi isassociated with global warming-correlated genetic admixture andpopulation-specific temperature adaptations. Mol. Ecol. 22, 2232–2248

68 Knopp, T. and Merila, J. (2009) The postglacial recolonization ofNorthern Europe by Rana arvalis as revealed by microsatellite andmitochondrial DNA analyses. Heredity 102, 174–181

69 Seifertova, M. et al. (2012) Multiple Pleistocene refugia and post-glacialcolonization in the European chub (Squalius cephalus) revealed bycombined use of nuclear and mitochondrial markers. J. Biogeogr. 39,1024–1040

241

Review Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

70 Schmitt, T. and Mu ller, P. (2007) Limited hybridization along a largecontact zone between two genetic lineages of the butterfly Erebiamedusa (Satyrinae, Lepidoptera) in Central Europe. J. Zool. Syst.Evol. Res. 45, 39–46

71 Vila, M. et al. (2005) The importance of time scale and multiplerefugia: incipient speciation and admixture of lineages in thebutterfly Erebia triaria (Nymphalidae). Mol. Phylogenet. Evol 36,249–260

72 Huertz, M. et al. (2004) Nuclear microsatellites reveal contrastingpatterns of genetic structure between western and southeasternEuropean populations of the common ash (Fraxinus excelsior L.).Evolution 58, 976–988

73 Petit, R.J. et al. (2003) Glacial refugia: hotspots but not melting pots ofgenetic diversity. Science 300, 1563–1565

74 Taylor, D.R. and Keller, S.R. (2007) Historical range expansiondetermines the phylogenetic diversity introduced duringcontemporary species invasion. Evolution 61, 334–345

75 Tishkoff, S.A. et al. (2009) The genetic structure and history of Africansand African Americans. Science 324, 1035–1044

76 Montana, G. and Pritchard, J.K. (2004) Statistical tests for admixturemapping with case-control and cases-only data. Am. J. Hum. Genet. 75,771–789

77 Schlebusch, C.M. et al. (2012) Genomic variation in seven Khoe-Sangroups reveals adaptation and complex African history. Science 338,374–379

78 Emerson, B.C. and Gillespie, R.G. (2008) Phylogenetic analysis ofcommunity assembly and structure over space and time. TrendsEcol. Evol. 23, 619–630

79 Wolkovich, E.M. and Cleland, E.E. (2010) The phenology of plantinvasions: a community ecology perspective. Front. Ecol. Environ. 9,287–294

80 Jones, E.I. and Gomulkiewicz, R. (2012) Biotic interactions, rapidevolution, and the establishment of introduced species. Am. Nat.179, E28–E36

81 Choisy, M. et al. (2004) Estimating admixture proportions withmicrosatellites: comparison of methods based on simulated data.Mol. Ecol. 13, 955–968

82 Francois, O. et al. (2010) Principal component analysis underpopulation genetic models of range expansion and admixture. Mol.Biol. Evol. 27, 1257–1268

83 Simon-Bouhet, B. et al. (2006) Multiple introductions promoterange expansion of the mollusc Cyclope neritea (Nassariidae) inFrance: evidence from mitochondrial sequence data. Mol. Ecol. 15,1699–1711

242

84 Roman, J. (2006) Diluting the founder effect: cryptic invasions expanda marine invader’s range. Proc. R. Soc. B: Biol. Sci. 273, 2453–2459

85 Slatkin, M. and Hudson, R.R. (1991) Pairwise comparisons ofmitochondrial DNA sequences in stable and exponentially growingpopulations. Genetics 129, 555–562

86 Navascues, M. et al. (2009) Characterization of demographicexpansions from pairwise comparisons of linked microsatellitehaplotypes. Genetics 181, 1013–1019

87 Gillis, N.K. et al. (2009) Higher genetic diversity in introduced than innative populations of the mussel Mytella charruana: evidence ofpopulation admixture at introduction sites. Divers. Distrib. 15, 784–795

88 Rollins, L.A. et al. (2011) Mitochondrial DNA offers unique insightsinto invasion history of the common starling. Mol. Ecol. 20, 2307–2317

89 Rand, D.M. (1996) Neutrality tests of molecular markers and theconnection between DNA polymorphism, demography, andconservation biology. Conserv. Biol. 10, 665–671

90 Corander, J. et al. (2004) BAPS 2: enhanced possibilities for theanalysis of genetic population structure. Bioinformatics 20, 2363–2369

91 Pritchard, J.K. et al. (2000) Inference of population structure usingmultilocus genotype data. Genetics 155, 945–959

92 Ma, J. and Amos, C.I. (2012) Principal components analysis ofpopulation admixture. PLoS ONE 7, e40115

93 Jombart, T. et al. (2010) Discriminant analysis of principalcomponents: a new method for the analysis of genetically structuredpopulations. BMC Genet. 11, 94

94 Cornuet, J-M. et al. (2010) Inference on population history and modelchecking using DNA sequence and microsatellite data with thesoftware DIYABC (v1.0). BMC Bioinformatics 11, 401

95 Price, A.L. et al. (2009) Sensitive detection of chromosomal segmentsof distinct ancestry in admixed populations. PLoS Genet. 5,e1000519

96 Nielsen, E.E. et al. (2006) HYBRIDLAB (version 1.0): a program forgenerating simulated hybrids from population samples. Mol. Ecol.Notes 6, 971–973

97 Ray, N. et al. (2010) SPLATCHE2: a spatially explicit simulationframework for complex demography, genetic admixture andrecombination. Bioinformatics 26, 2993–2994

98 Bryc, K. et al. (2010) Genome-wide patterns of population structure andadmixture in West Africans and African Americans. Proc. Natl. Acad.Sci. U.S.A. 107, 786–791

99 De Carvalho, D. et al. (2010) Admixture facilitates adaptation fromstanding variation in the European aspen (Populus tremula L.), awidespread forest tree. Mol. Ecol. 19, 1638–1650