Perspective
Genetics in conservation management: Revised recommendations for
the 50/500 rules, Red List criteria and population viability analyses
Richard Frankham a,b,⇑
, Corey J.A. Bradshaw c
, Barry W. Brook c
a
Department of Biological Sciences, Macquarie University, NSW 2109, Australia
b
Australian Museum, 6 College St, Sydney, NSW 2010, Australia
c
The Environment Institute and School of Earth and Environmental Sciences, The University of Adelaide, South Australia 5005, Australia
a r t i c l e i n f o
Article history:
Received 15 May 2013
Received in revised form 16 December 2013
Accepted 28 December 2013
Keywords:
Effective population size
Evolutionary potential
Fragmentation
Gene flow
Inbreeding depression
Minimum viable population size
a b s t r a c t
Conservation managers typically need to make prompt decisions based on limited information and
resources. Consequently, generalisations have essential roles in guiding interventions. Here, we (i) critique
information on some widely accepted generalisations and variables affecting them, (ii) assess
how adequately genetic factors are currently incorporated into population viability analysis (PVA) models
used to estimate minimum viable population sizes, and (iii) relate the above to population size thresholds
of the IUCN Red List criteria for threatened species that were derived from genetic considerations.
Evidence accumulated since 1980 shows that genetically effective population size (Ne) = 50 is inadequate
for preventing inbreeding depression over five generations in the wild, with Ne P 100 being required to
limit loss in total fitness to 610%. Further, even Ne = 500 is too low for retaining evolutionary potential for
fitness in perpetuity; a better approximation is Ne P 1000. Extrapolation from census population size (N)
to Ne depends on knowing the ratio of Ne/N, yet this information is unavailable for most wild populations.
Ratio averages ($0.1–0.2) from meta-analyses are sufficient, provided adjustments are made for dissimilar
life histories. Most PVA-based risk assessments ignore or inadequately model genetic factors. PVA
should routinely include realistic inbreeding depression, and genetic impacts on evolutionary potential
should be incorporated where appropriate. Genetic generalisations used in conservation, the treatment
of genetics in PVAs, and sections of the IUCN Red List criteria derived from genetic considerations, all
require revision to be more effective conservation tools.
Ó 2014 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2. Population size required to avoid inbreeding depression in the short term: Ne = 50? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.1. Overview and recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3. Population size required to maintain evolutionary potential in perpetuity: Ne = 500? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.1. Re-evaluating evolutionary potential. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.1.1. Lethals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.1.2. Mildly deleterious alleles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.1.3. Alleles subject to balancing selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.2. Overview and recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4. Extrapolating from effective (Ne) to census (N) population sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.1. Overview and recommendation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
5. Fragmented populations and connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
5.1. Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
6. How well is genetics incorporated into PVAs and MVPs? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
6.1. Inbreeding depression in PVAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
0006-3207/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biocon.2013.12.036
Abbreviations: d, dominance; MVP, minimum viable population size; Ne, effective population size; N, census population size; PVA, population viability analysis;
qe, equilibrium frequency; s, selection coefficient; VA, additive genetic variation; Vm, mutational variation for a quantitative character; u, mutation rate.
⇑ Corresponding author at: Department of Biological Sciences, Macquarie University, NSW 2109, Australia. Tel.: +61 2 9850 8186; fax: +61 2 9850 8245.
E-mail addresses: richard.frankham@mq.edu.au (R. Frankham), corey.bradshaw@adelaide.edu.au (C.J.A. Bradshaw), barry.brook@adelaide.edu.au (B.W. Brook).
Biological Conservation 170 (2014) 56–63
Contents lists available at ScienceDirect
Biological Conservation
journal homepage: www.elsevier.com/locate/biocon
6.2. Evolutionary potential in PVAs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
6.3. Minimum viable population size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
6.4. Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
7. Relationships between Ne = 100, Ne = 1000, MVP and IUCN Red List criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
8. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Appendix A. Supplementary material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
1. Introduction
Conservation biology is a crisis discipline requiring urgent management
for threatened species often with inadequate information
(Soulé, 1985). As most species have inadequate information on
which to base effective intervention decisions, conservation action
is frequently opportunistic, seeking compromise under competing
demands, and/or politically expediency (Pressey and Bottrill,
2008). For example, parcels of land offered for sale require that
decisions to purchase for conservation must be made promptly
(McDonald-Madden et al., 2008). Given limited resources and
sparse information for most threatened species, scientific generalisations
are often required. Some authors have criticised these
(Flather et al., 2011), but the alternative is usually unscientific conservation
decisions made at the political and bureaucratic levels,
especially in poorer countries and for non-charismatic species
(Brook et al., 2011).
The International Union for Conservation of Nature and Natural
Resources (IUCN) recognises the need to conserve biodiversity at
three level: genetic diversity, species and ecosystems (McNeely
et al., 1990), with genetic issues being involved in all three
(Frankham et al., 2010). We concentrate on the first two. Species
are usually driven to extinction by a combination of systematic human-associated
threats (habitat loss, over-exploitation, introduced
species, pollution and climate change) and stochastic events associated
with small population size (demographic, ecological and genetic
stochasticity, and catastrophes) (Shaffer, 1981), typically
interacting in a synergistic feedback (Brook et al., 2008) termed
the ‘extinction vortex’ (Gilpin and Soulé, 1986). In this Perspective
we focus on controversial aspects of genetic stochasticity (see
Glossary in Appendix A1), primarily encompassing inbreeding
depression, and reduced evolutionary potential (Frankham et al.,
2010; Jamieson and Allendorf, 2012).
Inbreeding and loss of genetic diversity are unavoidable in
small, closed, sexually reproducing populations, and accumulate
in a ratchet-like manner over generations for diploid randommating
populations, as follows (Wright, 1969):
Ht=H0 ¼ 1 À
1
2Ne
 t
¼ 1 À F ð1Þ
where Ht is heterozygosity at generation t (for neutral variation), H0
initial heterozygosity, Ne genetically effective population size and F
the inbreeding coefficient (with generation zero defined as having
F = 0). In naturally outbreeding species, this typically results in
inbreeding depression (unless they have already experienced it)
and reduced ability to evolve (Frankham et al., 2010).
To work, generalizations depend on different taxa responding
similarly, or at least groups of them doing so. While ecologists typically
emphasise species distinctiveness (e.g. Flather et al., 2011),
evolutionary and conservation geneticists usually focus on the
similarity of evolutionary processes across species with similar
breeding systems. For example, across most major taxa: (i)
inbreeding has consistently deleterious effects on fitness in wild
outbreeding diploid and polyploid species (Crnokrak and Roff,
1999), (ii) population mean genetic diversity, mean fitness and
population size are positively correlated (Frankham, 2012), (iii)
heritabilities (genetic variation as a proportion of phenotypic variation)
are lower for fitness than for quantitative traits peripherally
related to fitness (Mousseau and Roff, 1987; Falconer and Mackay,
1996), (iv) non-additive genetic variation is greater for fitness than
peripheral traits (Frankham et al., 2010), and (v) mutation rates for
quantitative characters are relatively similar (Houle et al., 1996).
Consequently, generalisations are often justifiable for genetic
issues in conservation biology (see also Appendix A2).
Our focus is on three genetic issues relating to generalisations.
First, an effective population size of at least 50 (Franklin, 1980;
Soulé, 1980) has been long recommended as a ‘rule’ for avoiding
inbreeding depression in the short term. Second, Ne = 500 has been
considered sufficient to retain evolutionary potential in perpetuity
(Franklin, 1980; Lande and Barrowclough, 1987). These two issues
had important roles in the development and implementation of the
IUCN Red List categorisation system for threatened species (Mace
et al., 2008), especially criterion C that relates to population size
(Appendix A3). Third, minimum viable population sizes (MVP;
Shaffer, 1981) provide estimates of the sizes required for species
to persist with high probability in the long-term. Given that over
30 years have elapsed since the classic Ne = 50 and Ne = 500 recommendations
were proposed, and their tenacity in conservation
management circles, we now ask whether current evidence
supports them, and how they might be modified. Jamieson and
Allendorf (2012) reviewed these issues, but we reach different conclusions
to them on several important issues.
We critique and make recommendations (Table 1) on (i) the
Ne = 50 and 500 rules, (ii) how best to translate Ne into census
population size (N), (iii) the genetic consequences of fragmented
populations, and (iv) the treatment of genetic issues in PVA and
their effect on estimation of MVPs. Finally, we (v) evaluate the
implications of these for the IUCN Red List categorisation system.
2. Population size required to avoid inbreeding depression in
the short term: Ne = 50?
Soulé (1980) and Franklin (1980) proposed Ne = 50 as sufficient
to prevent inbreeding depression in naturally outbreeding diploid
species in the short term, and most authors, including Jamieson
and Allendorf (2012) still endorse this value. However, specifying
the duration as ‘short term’ is too vague, because it can mean
different things in different disciplines. Since Soulé and Franklin
had $5 generations in mind (from discussions with RF), we recommend
that 5 generations be used (because genetic effects scale to
generations).
Franklin (1980) and Soulé (1980) based their Ne = 50 on the
opinion of animal breeders, plus limited data from domestic and
laboratory animals. However, inbreeding depression is generally
greater in stressful, wild environments than in benign, captive ones
(Fox and Reed, 2010), and domestic animals might have different
susceptibility to inbreeding depression compared to wild populations
because they have been subjected to artificial selection,
R. Frankham et al. / Biological Conservation 170 (2014) 56–63 57
adaptation to captive environments and frequent population-size
bottlenecks (Frankham, 2009; Mattila et al., 2012).
Subsequent to 1980, inbreeding depression has been reported
in many laboratory populations with Ne P 50 (Latter and Mulley,
1995; Bryant et al., 1999; Reed and Bryant, 2000; Woodworth
et al., 2002). Even housefly populations with Ne $ 90 for 5 generations
exhibited inbreeding depression for fitness (Bryant et al.,
1999). Current information predicts an average decline in total fitness
of 26% in the wild due to 5 generations at Ne = 50 (cumulative
F $ 5%) (see computations in Appendix A4) based on 12 diploid
lethal equivalents in wild environments from the meta-analysis
in O’Grady et al. (2006), while the 16 lethal equivalents value from
Grueber et al. (2010) yields a decline of 33%. Further, Newman and
Pilson (1997) found that when F increased by 5%, population
extinctions rose from 25% to 69% in Clarkia pulchella plants. These
values represent intolerably high inbreeding depression in the
wild.
Since most studies report linear relationships between F and
fitness (Frankham et al., 2010), F > 0 will usually cause adverse
fitness effects in natural outbreeding species. For conservation
management, we suggest specifying that a 10% decline in fitness
is tolerable (following similar logic to the captive breeding objective
of retaining 90% of genetic diversity for 100 years; Frankham
et al., 2010). Theory predicts that Ne = 142 is required for a population
to retain 90% of current fitness after inbreeding of F = 5%,
based on 12 diploid lethal equivalents. Given limited data on lethal
equivalents and the potential for modest purging (see Appendix
A5), we suggest that the Ne be raised to at least 100 and reevaluated
as more data become available.
2.1. Overview and recommendations
In summary, we recommend that the generalisation for avoiding
inbreeding depression in the short term should (i) specify an
explicit duration of 5 generations; (ii) recognise that any F > 0 will
usually have adverse fitness effects; (iii) be defined as an Ne that
leads to <10% cumulative decline in total fitness in the wild, and
(iv) stipulate that minimum Ne be increased to 100.
3. Population size required to maintain evolutionary potential
in perpetuity: Ne = 500?
Franklin (1980) argued that additive quantitative genetic variation
(VA), rather than single-locus variation determined the ability
to evolve. He concluded that Ne = 500 was sufficient to retain
evolutionary potential in perpetuity, based on the equilibrium between
adding genetic variation by mutation (Vm), and losing VA/
(2Ne) per generation by random genetic drift for a quantitative trait
that is either unaffected by selection or subject to stabilising selection
(typical of peripheral characters). He then solved for Ne using
empirical data on Vm and the heritability of the trait, as detailed in
Frankham et al. (2010). Lande and Barrowclough (1987) reached a
similar conclusion based on a model of mutation, drift and stabilising
selection on a peripheral character. However, more recent
alternative quantitative genetic theory has yielded Ne values
P1000 (Appendix A6). Jamieson and Allendorf (2012) queried
the association of Ne, evolutionary potential and extinction, despite
extensive theory and empirical evidence (field, laboratory and simulation
data) supporting it (see Appendix A7), but they generally
endorsed the Ne = 500 value. While Franklin (1980) considered
only evolutionary potential and quantitative genetic variation in
the context of Ne = 500 it cannot be entirely divorced from concerns
about inbreeding depression (Frankham et al., 2013; Appendix
A8).
Subsequent considerations of Ne to retain evolutionary potential
raised issues of the fitness of new mutations and the appropriate
value of Vm (see Appendix A9), and led us to ask whether
evolutionary potential should be determined for peripheral or fitness
traits.
3.1. Re-evaluating evolutionary potential
By definition, the evolutionary adaptation of populations by
natural selection involves changes in reproduction and/or survival
rates (reproduction fitness), even if the changes also affect other
characters. Fitness traits have fundamentally different features
from peripheral characters, especially in the predominant forms
of natural selection experienced, asymmetry of mutation effects,
non-additive genetic variation and the maintenance of genetic variation
(Table 2). As there is no agreed value of Ne required to retain
evolutionary potential for fitness in perpetuity, we compile several
lines of evidence to provide a preliminary solution.
Genetic variation for fitness in a closed, random-mating population
is maintained by the balance between mutation (adding it),
drift (removing it) and selection (either removing or retaining variation).
Much of the genetic variation is maintained by a balance
between addition of deleterious mutations, and their removal by
selection and drift (Charlesworth and Hughes, 2000; RodríguezRamilo
et al., 2004; but see Charlesworth et al., 2007), resulting
in equilibria with low frequencies of deleterious alleles that are
susceptible to loss by drift (see below).
Table 1
Topics addressed in this perspective.
Topic Current recommendation Revised recommendation
1. Avoid inbreeding depression Ne = 50 Ne P 100
2. Maintain evolutionary potential Ne = 500 Ne P 1000
3. Extrapolating from Ne to N Ne/N = 0.10–0.14 Use different Ne/N according to life-history of species, but current default
is 0.1–0.2
4. Fragmented populations and
connectivity
Evaluate on case basis Evaluate, but distinguish current and historical gene flow
5. Genetic factors in PVA Often none Routinely include inbreeding depression
(i) Inbreeding depression Lethal equivalents = 3.14 on juvenile mortality, if
included
Routinely apply realistic levels ($12 lethal equivalents on total fitness)
(ii) Evolutionary potential Ignored Include in long-term and environmental-change contexts
6. MVPs Persistence probability inconsistent Apply common standard and specify (suggest 99%)
Duration inconsistent Standardise and specify in generations (suggest 40)
Important factors often ignored Routinely include all systematic and stochastic factors (including
genetic)
7. IUCN Red List Criterion C Population size
Critically Endangered <250 <500
Endangered <2500 <5000
Vulnerable <10,000 <20,000
58 R. Frankham et al. / Biological Conservation 170 (2014) 56–63
We divide alleles at fitness-related loci into five categories (near
recessive lethals, mildly deleterious, ones experiencing balancing
selection, beneficial, and neutral in the current environment),
and determine what Ne is required to maintain approximately
the same genetic diversity as expected in an infinite population.
We concentrate on the first three, because the Franklin (1980)
and other theory described above applies to the small proportion
that is neutral in the current environment, while the few categories
that are beneficial in the environment to which the population is
adapted (Perfeito et al., 2013) are rapidly fixed or lost by drift, so
few are polymorphic at any given time. Of these categories, it is
the mildly deleterious alleles that are likely to be of predominant
importance in adapting to changed environments, as they are most
frequent, and because lethals and other highly deleterious alleles
are usually unconditionally deleterious (Lindsley and Zimm,
1992). Alleles that are deleterious in one environment can contribute
to adaptation because they are often beneficial in others: genotype
 environment interactions (Frankham et al., 2010), including
home-site advantage (Hereford, 2009) and adverse fitness effects
experience by captive-adapted populations in wild habitats
(Frankham, 2008).
3.1.1. Lethals
Species’ genomes typically contain rare lethal alleles at many
loci. For example, $5000 loci in Drosophila potentially have lethal
alleles whose heterozygotes have an average reduction in viability
of $2.5% (ds) per generation compared to wild-type homozygotes,
and new lethal mutations arise at a rate (u) of $3 Â 10À6
locusÀ1
generationÀ1
(Simmons and Crow, 1977). The balance between
deleterious mutations and selection against heterozygotes for such
alleles in an infinite diploid population results in an equilibrium
frequency (qe) of:
qe $ u=ds ð2Þ
where d is the dominance of the allele (proportionate deviation of
heterozygote fitness from the mean of the homozygotes) and s
the selection coefficient against the deleterious homozygote. Consequently,
lethals (s = 1) have low equilibrium frequencies at individual
loci ($1.2 Â 10À5
) in infinite populations. Their equilibrium
frequencies depend on Ne up to at least Ne = 1000 (Nei, 1968;
Hedrick, 2002; not independent of Ne as stated by Lande and
Barrowclough, 1987). Average lethal frequencies in experimental
Drosophila melanogaster populations maintained with Ne = 500 for
51–67 generations were substantially less than that in a recently
caught sample of the wild base population (Fig. A1), consistent with
the above theory.
3.1.2. Mildly deleterious alleles
Deleterious partial recessives (at many loci) with average heterozygous
disadvantages $1–5% on viability (Houle et al., 1996)
are predicted to have equilibrium frequencies of 2 Â 10À4
to 10À3
in infinite populations (assuming u $ 10À5
and the viability effects
are representative of total-fitness effects). Genetic diversity for
partially recessive alleles (d < 0.5) at single loci begin to asymptote
by Ne = 1000 when s P 0.01, based on theoretical analyses of
inbreeding depression (a function of the heterozygosity for such
alleles; Falconer and Mackay, 1996), and fixed deleterious allele
frequencies (Bataillon and Kirkpatrick, 2000).
Visible mutations (due to mildly deleterious alleles) in Drosophila
populations maintained with Ne of 25, 50, 100, 250 and 500 for
51–67 generations showed, on average, progressively increasing
frequencies of chromosomes containing unique visible mutations
as Ne was raised, but a recently wild-caught sample of the base
population had a 37% higher frequency than the Ne = 500 treatment
(Fig. A2), consistent with the above theory.
3.1.3. Alleles subject to balancing selection
Few loci appear to experience balancing selection and have
intermediate allele frequencies (Charlesworth and Hughes, 2000;
Bubb et al., 2006). In an infinite population, the equilibrium
frequency for a locus subject to heterozygote advantage (relative
fitnesses for genotypes A1A1, A1A2 and A2A2 of 1 À s, 1 and 1 À t,
respectively) is:
qe ¼
s
s þ t
ð3Þ
Such loci are less affected by drift than neutral loci when equilibrium
frequencies are 0.2–0.8, but are more affected outside this
range (Robertson 1962; Hedrick, 2012). Robertson concluded for
fitness loci subject to heterozygote advantage in nature that ‘‘the
effective population size must be of the order of 1000 before the
genetic variation found within it approaches the value for very
large populations . . .’’ Similarly, for frequency-dependent selection,
Roff (1998) argued on the basis of multi-locus simulations ‘‘that
the heritability increases with population size, typically approaching
its maximal value at a population size of 5000, when the mutation
rate is 10À4
, and double this when the mutation rate is 10À5
.’’
Maintenance of genetic variation by selection that varies temporally
or spatially only occurs under quite restricted conditions,
based on theoretical models (Prout, 2000). Further, empirical studies
indicate that such selection is implicated in maintaining variation
at a low proportion of loci (Hedrick, 2006). We found no
relevant theory relating genetic variation to Ne for such selection.
3.2. Overview and recommendations
Several independent lines of theoretical and empirical evidence
indicate that at least Ne = 1000 is required to maintain initial evolutionary
potential in perpetuity. We recommend the minimum Ne
required to maintain evolutionary potential equal to that of large
wild populations in perpetuity be revised to (i) specify maintenance
of genetic diversity for reproductive fitness, (ii) be derived
from theoretical models involving mutation, selection and drift,
plus empirical data, and (iii) specify Ne of at least 1000 in naturally
outbreeding species. Populations with Ne < 1000 are not doomed to
extinction in the short to medium term, but their ability to evolve
to cope with environmental change will erode with time and this
will reduce their long-term viability.
4. Extrapolating from effective (Ne) to census (N) population
sizes
Ne discussed above need to be converted into mean (adult) census
population sizes (N) per generation for conservation managers.
Information on the Ne/N ratio is required for this conversion, but
comprehensive estimates that encompass all relevant variables
are only available for $100 species (Frankham, 1995; Palstra and
Ruzzante, 2008; Palstra and Fraser, 2012). Accordingly, most
Table 2
Differences in characteristics of reproductive fitness traits versus peripheral quantitative
characters (after Frankham, 2009).
Characteristic Quantitative character
Fitness Peripheral
Inbreeding depression and heterosis Strong Weak
Asymmetry of selection response Strong $Symmetrical
Heritabilities $10–20% $30–70%
Non-additive genetic variation Present Little or none
Effect of mutation on mean Reduces Minimal change
Genotype  environment interactions Stronger Weaker
Natural selection Directional Stabilising
R. Frankham et al. / Biological Conservation 170 (2014) 56–63 59
conversions have been based on average Ne/N (0.10–0.14;
Frankham, 1995; Palstra and Ruzzante, 2008). Palstra and Fraser
(2012) reported a median ratio of 0.12, but favoured one of 0.23
from cases where appropriate Ne and N were strictly paired.
Jamieson and Allendorf (2012) stated that ‘‘the actual Ne/N value
in a particular population or species differs tremendously
depending upon demography and life history.’’ Below we consider
this issue and whether extrapolating from Ne to N should be
avoided. Following Mace and Lande (1991), we consider it imperative
that population size guidelines be provided for conservationists
working with species without Ne/N estimates.
Differences in Ne/N estimates are largely due to use of comprehensive
and non-comprehensive estimates (Frankham, 1995;
Leroy et al., 2013), sampling variation (Frankham, 1995; Palstra
and Fraser, 2012), methodological variation (Appendix A10), use
of different measures of N (adults, breeding pairs, and adults plus
juveniles combined), current N not necessarily reflecting long-term
N (Frankham, 1995), pairing Ne with diverse (often inappropriate)
N (Palstra and Fraser, 2012), and variation in life-history traits
(Waples et al., 2013). Care must also be taken to use N consistently,
as both mean N (most frequent) and harmonic mean N are used
(Frankham et al., 2010). However, there are constraints on this variation.
Frankham (1996) concluded that ‘‘the overall relationship
between heterozygosity and logN [correlation 0.81 across species]
would not be possible if Ne/N ratios varied widely among different
species and taxa.’’ Further, correlations of genetic diversity-logNe,
and genetic diversity-logN, are similar within species and across
species (Frankham, 2012), and there is little statistical support
for differences between major taxa (Frankham, 1995). Methodological
variation for comprehensive Ne/N estimates appears minor
(Appendix A10). Additionally, putative population-density effects
(Palstra and Ruzzante, 2008; Luikart et al., 2010) were not confirmed
by Palstra and Fraser (2012).
Palstra and Ruzzante (2008) reported Ne/N was much lower
($2 Â 10À5
) in highly fecund exploited fish compared to other
species (as predicted by theory: Waples, 2002; Hedrick, 2005),
but suggested that this might either be a methodological artefact
and/or a consequences of human exploitation. However, given that
both exploited (fish, oysters, and shrimp) and unexploited
(Drosophila and brown algae, Fucus serratus) species with high
fecundity also have low Ne/N (Frankham, 1995, 2012), the role of
exploitation is questionable. Further, methods other than the temporal
one used in their study also yielded low Ne/N in Drosophila
and water hyacinth (Eichhornia paniculata) (Frankham, 1995).
Consequently, low ratios for highly fecund species could be
realistic, but more information is required.
Since the meta-analysis by Palstra and Ruzzante (2008), the
ongoing use of a single mean value of Ne/N for all species is unsound.
Waples et al. (2013) reported that up to half the variance
in single generation Ne/N was explained by age at maturity and
adult lifespan, opening the possibility of providing more precise
extrapolations from Ne to N than possible previously. This must
be combined with information on fluctuations in population sizes
across generations to yield comprehensive Ne/N estimates.
4.1. Overview and recommendation
Conversion from Ne into N must both involve comprehensive
estimates of Ne and account for different Ne/N for species with diverse
life histories. Mean Ne/N is $0.1–0.2.
5. Fragmented populations and connectivity
So far, we have considered only single, approximately randommating
closed populations. However, free-ranging wild species
usually have fragmented distributions, with population structures
varying from effectively single random-mating populations
(fragmented spatially, but connected by gene flow), through
partially connected fragments and meta-populations, to completely
isolated subpopulations; each circumstance has different
genetic consequences in relation to genotype frequencies, genetic
differentiation, inbreeding, fitness, and extinction risks (Table A1;
Frankham et al., 2010).
Population fragmentation of species is recognised as a threatening
process in the IUCN Red List Categorisation System (IUCN,
2013). However, Jamieson and Allendorf (2012) noted that
unaccounted gene flow can reverse adverse genetic effects of small
population size. Alternatively, connectivity might be weaker than
generally recognised in species of conservation concern due to
ongoing habitat loss, declining population sizes, declines of many
plant pollinators (Brosi and Briggs, 2013), expansion of roads,
fences and other human-constructed barriers to gene flow (Crooks
and Sanjayan, 2006), increasing impacts of invasive species
(Gilbert and Levine, 2013), and lower fitness of immigrants than
residents (Hanski et al., 2000). How serious a genetic threat is
fragmentation?
While human impacts increase connectivity for some species
(especially invasives), most empirical evidence indicates limited
connectivity among subpopulations of many once-contiguous species
(see Table A1 where references are provided). First, >25% of
species have populations that differ genetically (FST [inbreeding
and drift due to population isolation] >0.2) (see Appendix A11).
Second, populations within species typically exhibit correlations
between genetic diversity and population size. Third, population
mean fitness is typically correlated with population size. Fourth,
species often show isolation by distance. Fifth, FST correlates negatively
with dispersal ability. Recent research also suggests that
marine organisms disperse less than previously assumed (Cowen
and Sponaugle, 2009; Jones et al., 2009). Tropical trees could be
an exception where fragmentation increases gene flow via pollen,
but this might only occur over a restricted geographic range (White
et al., 2002; Dick et al., 2008; Hamrick, 2010). Conversely, for 15/21
tropical forest tree species, low-density (generally fragmented)
plants showed lower multilocus outcrossing rates compared to
high-density (usually undisturbed forest) plants (Dick et al.,
2008), calling into question whether overall gene flow is generally
higher in these fragmented tropical tree species.
We conclude that population fragmentation is a serious genetic
threat to the persistence of populations of many species.
5.1. Recommendations
We recommend that genetic connectivity continue to be assessed
among populations within species to inform species management
and advise that current and historical connectivity be
distinguished. For previously connected populations that now have
limited gene flow and evidence of genetic erosion, we recommend
augmentation of gene flow be considered, provided the risk of outbreeding
depression is low (Frankham et al., 2011).
6. How well is genetics incorporated into PVAs and MVPs?
Minimum viable population (MVP) sizes required for long-term
population persistence are commonly determined using population
viability analysis (PVA) simulation models (Traill et al.,
2007). Ideally these models should include all systematic and stochastic
variables potentially affecting population viability. But how
much is genetics considered in most PVAs?
There is now extensive evidence that genetic factors influence
extinction risk (Frankham, 2005; Frankham et al., 2010; Allendorf
60 R. Frankham et al. / Biological Conservation 170 (2014) 56–63
et al., 2013). First, inbreeding depression substantially reduces
median times to extinction in PVAs for real species in natural environments
(Brook et al., 2002; O’Grady et al., 2006). Second, empirical
field studies of the Glanville fritillary butterfly (Melitaea cinxia)
and of two plant species revealed that inbreeding reduces population
viability in wild populations (Newman and Pilson, 1997;
Saccheri et al., 1998; Vilas et al., 2006). Third, most species are
not driven to extinction before genetic factors can influence them
(Spielman et al., 2004). Fourth, reduced genetic diversity for fitness
compromises the ability of populations to adapt to environmental
change (Appendix A6). Fifth, mutation accumulation might reduce
long-term population viability, especially in asexual and selfing
species (Frankham et al., 2010).
6.1. Inbreeding depression in PVAs
Jamieson and Allendorf (2012) concluded: ‘‘we believe that
inbreeding depression is fully accounted for in standard PVAs
and for estimates of MVP size resulting in the ‘50’ component of
the 50/500 rule.’’ However, ‘‘most published PVAs have not included
genetics’’ (Allendorf and Ryman, 2002), and even in the last
decade many still have not (Traill et al., 2007; Zeigler et al., 2013).
Further, most PVAs that include inbreeding depression apply only
3.14 diploid lethal equivalents to juvenile survival (the default setting
in the widely-used Vortex PVA software) (Traill et al., 2007).
This underestimates the total impact of inbreeding (Allendorf
et al., 2013), because the true value averages $12 for total fitness
(O’Grady et al., 2006). In addition, Ne/N in PVAs is typically higher
($0.30) than observed in unmanaged wild populations (mean 0.1–
0.2), because most software default settings, including those in
Vortex, assume Poisson distributions of family sizes (Frankham
et al., 2010 p. 506), whereas species usually exhibit much larger
variance (Frankham et al., 2010 p. 246), and thus lower Ne/N.
6.2. Evolutionary potential in PVAs
As long-term persistence of populations in changing environments
typically requires the ability to evolve, this capacity should
be included in long-term PVAs. While PVAs have not included
evolutionary potential (Jamieson and Allendorf, 2012), there are
few impediments to doing so. A practical and empirically tested
equation exists for predicting long-term genetic adaptation
(Appendix equation A2). Further, several studies have used individual-based
stochastic computer-simulation models to investigate
how adaptive evolutionary changes affect population persistence
of fish (Dunlop et al., 2007; Reed et al., 2011; Kuparinen and
Hutchings, 2012; Piou and Prévost, 2012).
6.3. Minimum viable population size
Minimum viable population size refers to the number of individuals
required for a high probability of population persistence
over the long term (Shaffer, 1981; Soulé, 1987). However, MVPs
have been determined for few species, so average values or statistical
models are all that can be used for the remainder.
MVPs estimates have used a variety of different probabilities of
persistence and durations in years, or generations (99% for
1000 years: Shaffer, 1981; 95% for several centuries: Soulé, 1987;
99% for 40 generations: Traill et al., 2007). To be biologically comparable,
duration should be in generations, because extinction risk
scales better to generations than years (O’Grady et al., 2008). We
suggest the standardised 99% probability of persistence for 40 generations
used in the Traill et al. (2007) meta-analysis be routinely
reported, even if other scales are also considered.
Based on the data in the Traill et al. (2007) meta-analysis, the
MVP for 99% persistence for 40 generations for a typical outbreeding
species is of the order of several thousand (N). Whether major taxa
differ in MVPs is unclear (Frankham et al., 2010 p. 518), but they
could be higher in species with higher fecundities (see Section 4).
6.4. Recommendations
We recommend that (i) inbreeding depression be routinely
included in PVA models for all sexually reproducing diploid or
polyploid species with normal chromosomal segregation, (ii)
inbreeding depression in PVAs be routinely applied to total fitness
(i.e. survival and fertility across all age/stage classes), (iii) realistic
inbreeding depression be applied for outbreeding species ($12 total
diploid lethal equivalents for wild environments), but lower
values for populations subjected to bottlenecks that have already
purged highly deleterious alleles, (iv) fewer lethal equivalents be
applied for species/populations with selfing or mixed-mating systems,
(v) evolutionary potential be included in PVA models of
long-term environmental change, or for estimates of indefinite persistence
of species, and (vi) MVPs be reported against a standard
99% probability of persistence for 40 generations using PVAs that
include all relevant systematic and stochastic variables.
7. Relationships between Ne = 100, Ne = 1000, MVP and IUCN Red
List criteria
MVPs identify populations that are relatively safe from all
threats. If available and based on adequate data, a species-specific
PVA is the most comprehensive guide to the likely persistence for a
species, or population. If this is not available, or a PVA yields results
of doubtful validity, or a species falls into the Data Deficient category
of the IUCN Red List, then Ne 6 100 indicates that it likely
faces serious genetic threats after 5 or more generations, whereas
Ne P 1000 indicates little or no genetic threats to its long-term
persistence: this usually translates into census population sizes
(N) similar to MVPs.
As IUCN Red List population size thresholds under Criterion C
derive from genetically effective population sizes (Appendix A3),
and two explicitly from Ne of 50 and 500 theories, it is logical to
double them to reflect the revisions we recommend above. Mace
and Lande (1991) assumed Ne/N = 0.2 to translate effective population
sizes into number of mature adults, but this may need reconsideration,
as described in Section 4 above. The incorporation of
genetic risks in the IUCN Red List categorisation system appear
illogical and would likely benefit from revision (Appendix A12).
We expect that a doubling of the above thresholds will increase
the number of species falling into each of the threatened categories,
but by how much needs to be determined by empirical evaluation.
The results of assessments on 28 taxa (Appendix A13)
showed that one third of the species that could potentially change
did move.
PVA can be used to assess risk of extinction for categorisation
under the IUCN Red List Criterion E, but most current values are
underestimates, as described in Section 6 above.
Many will have concerns that recommended targets for effective
population sizes are unattainable for most threatened species.
Although increasing N might be impractical in many wild populations,
Ne can also be augmented by increasing the Ne/N ratio, or
re-establishing gene flow among isolated sub-populations of a species
(Frankham et al., 2010). Substantial increases in Ne can usually
be made if totally isolated populations of a species that were previously
connected by gene flow are re-connected using artificial
gene flow, or habitat corridors (Frankham et al., 2010). In some
conservation circumstances, there are no practical remedies to
avoid genetic deterioration and the population or species needs
to be categorised as fragile with risk minimised by non-genetic
R. Frankham et al. / Biological Conservation 170 (2014) 56–63 61
means, as done in the case of the Wollemi pine (Wollemia nobilis)
(Frankham et al., 2010 p. 399).
8. Conclusions
Based on our review of recent theoretical and empirical evidence,
we conclude that the oft-cited Ne = 50 rule for avoiding
(minimising) inbreeding depression in the short-term (5 generations),
and Ne = 500 proposed for maintaining evolutionary potential
in perpetuity, need to be at least doubled, as do the genetically
derived IUCN Red List population-size thresholds for Criterion C.
Further, population viability analyses need to incorporate more
realistic genetic risks by routinely including inbreeding depression
on total fitness applied at levels relevant to the intended
environment (wild or captive populations). For PVAs that consider
long-term scenarios the effects of evolutionary potential should be
modelled (and this is feasible).
Acknowledgments
We thank G. Cooke, M. Eldridge, C. Fenster, I. Franklin, P. Hedrick,
W. Hill, A. Hoffmann, C. Lees, G. Mace, D. Spielman, M. Soulé,
Y. Willi and five anonymous reviewers for comments on the manuscript.
C.J.A.B. and B.W.B. are supported by Australian Research
Council Future Fellowships.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.biocon.2013.
12.036.
References
Allendorf, F.W., Ryman, N., 2002. The role of genetics in population viability
analysis. In: Beissinger, S.R., McCullough, D.R. (Eds.), Population viability
analysis. University of Chicago Press, Chicago, pp. 50–85.
Allendorf, F.W., Luikart, G., Aitken, S.N., 2013. Conservation and the Genetics of
Populations, second ed. Wiley-Blackwell, Oxford.
Bataillon, T., Kirkpatrick, M., 2000. Inbreeding depression due to mildly deleterious
mutations in finite populations: size does matter. Genet. Res. 75, 75–81.
Brook, B.W., Tonkyn, D.W., O’Grady, J.J., Frankham, R., 2002. Contribution of
inbreeding to extinction risk in threatened species. Conserv. Ecol. 6 (1), 16.
Brook, B.W., Sodhi, N.S., Bradshaw, C.J.A., 2008. Synergies among extinction drivers
under global change. Trends Ecol. Evol. 23, 453–460.
Brook, B.W., Bradshaw, C.J.A., Traill, L.W., Frankham, R., 2011. Minimum viable
population size: not magic, but necessary. Trends Ecol. Evol. 26, 619–620.
Brosi, B.J., Briggs, H.M., 2013. Single pollinator species losses reduce floral fidelity
and plant reproductive function. Proc. Natl. Acad. Sci. U.S.A. 110, 13044–13048.
Bryant, E.H., Backus, V.L., Clark, M.E., Reed, D.H., 1999. Experimental tests of captive
breeding for endangered species. Conserv. Biol. 13, 1487–1496.
Bubb, K.L., Bovee, D., Buckley, D., Haugen, E., Kibukawa, M., Paddock, M., Palmerieri,
A., Subramanian, S., Zhou, Y., Kaul, R., Green, P., Olson, M.V., 2006. Scan of
human genome reveals no new loci under ancient balancing selection. Genetics
173, 2165–2177.
Charlesworth, B., Hughes, K.A., 2000. The maintenance of genetic variation in lifehistory
traits. In: Singh, R.S., Krimbas, C.B. (Eds.), Evolutionary Genetics: From
Molecules to Morphology. Cambridge University Press, Cambridge, pp. 369–
392.
Charlesworth, B., Miyo, T., Borthwick, H., 2007. Selection response of means and
inbreeding depression for female fecundity in Drosophila melanogaster suggest
contributions from intermediate-frequency alleles to quantitative trait
variation. Genet. Res. 89, 85–91.
Cowen, R.K., Sponaugle, S., 2009. Larval dispersal and marine population
connectivity. Annu. Rev. Mar. Sci. 1, 443–466.
Crnokrak, P., Roff, D.A., 1999. Inbreeding depression in the wild. Heredity 83, 260–
270.
Crooks, K.R., Sanjayan, M. (Eds.), 2006. Connectivity Conservation. Cambridge
University Press, New York.
Dick, C.W., Hardy, O.J., Jones, F.A., Petit, R.J., 2008. Spatial scales of pollen and seedmediated
gene flow in tropical rain forest trees. Tropical Plant Biol. 1, 20–33.
Dunlop, E., Shuter, B.J., Dieckmann, U., 2007. Demographic and evolutionary
consequences of selective mortality: predictions from an eco-genetic model
for smallmouth bass. Trans. Am. Fish. Soc. 136, 749–765.
Falconer, D.S., Mackay, T.F.C., 1996. Introduction to Quantitative Genetics, fourth
ed. Longman, Harlow, England.
Flather, C.H., Hayward, G.D., Beissinger, S.R., Stephens, P.A., 2011. Minimum viable
populations: is there a ‘magic number’ for conservation practitioners? Trends
Ecol. Evol. 26, 307–316.
Fox, C.W., Reed, D.H., 2010. Inbreeding depression increases with environmental
stress: an experimental study and meta-analysis. Evolution 65, 246–258.
Frankham, R., 1995. Effective population size/adult population size ratios in
wildlife: a review. Genet. Res. 66, 95–107.
Frankham, R., 1996. Relationship of genetic variation to population size in wildlife.
Conserv. Biol. 10, 1500–1508.
Frankham, R., 2005. Genetics and extinction. Biol. Conserv. 126, 131–140.
Frankham, R., 2008. Genetic adaptation to captivity in species conservation
programs. Mol. Ecol. 17, 325–333.
Frankham, R., 2009. Genetic architecture of reproductive fitness and its
consequences. In: van der Werf, J., Graser, H.-U., Frankham, R., Gondro, C.
(Eds.), Adaptation and Fitness in Animal Populations: Evolutionary and
Breeding Perspectives on Genetic Resource Management. Springer, Dordrecht,
pp. 15–39.
Frankham, R., 2012. How closely does genetic diversity in finite populations
conform to predictions of neutral theory? Large deficits in regions of low
recombination. Heredity 108, 167–178.
Frankham, R., Ballou, J.D., Briscoe, D.A., 2010. Introduction to Conservation Genetics,
second ed. Cambridge University Press, Cambridge, U.K.
Frankham, R., Ballou, J.D., Eldridge, M.D.B., Lacy, R.C., Ralls, K., Dudash, M.R., Fenster,
C.B., 2011. Predicting the probability of outbreeding depression. Conserv. Biol.
25, 465–475.
Frankham, R., Brook, B.W., Bradshaw, C.J.A., Traill, L.W., Spielman, D., 2013. 50/500
rule and minimum viable populations: response to Jamieson and Allendorf.
Trends Ecol. Evol. 28, 187–188.
Franklin, I.R., 1980. Evolutionary change in small populations. In: Soulé, M.E.,
Wilcox, B.A. (Eds.), Conservation Biology: An Evolutionary-Ecological
Perspective. Sinauer, Sunderland, MA, pp. 135–149.
Gilbert, B., Levine, J.M., 2013. Plant invasions and extinction debts. Proc. Nat. Acad.
Sci. U.S.A. 110, 1744–1749.
Gilpin, M.E., Soulé, M.E., 1986. Minimum viable populations: processes of species
extinction. In: Soulé, M.E., Wilcox, B.A. (Eds.), Conservation Biology: An
Evolutionary-Ecological Perspective. Sinauer, Sunderland, MA, pp. 19–34.
Grueber, C.E., Laws, R.J., Nakagawa, S., Jamieson, I.G., 2010. Inbreeding depression
accumulation across life-history stages of the endangered takahe. Conserv. Biol.
24, 1617–1625.
Hamrick, J.L., 2010. Pollen and seed movement in disturbed tropical landscapes. In:
DeWoody, J.A., Bickham, J.W., Michler, C., Nichols, K., Rhodes, D., Woeste, K.
(Eds.), Molecular Approaches in natural Resource Conservation and
Management. Cambridge University Press, New York, pp. 190–211.
Hanski, I., Alho, J., Moilanen, A., 2000. Estimating the parameters of survival and
migration of individuals in metapopulations. Ecology 81, 239–251.
Hedrick, P.W., 2002. Lethals in finite populations. Evolution 56, 654–657.
Hedrick, P., 2005. Large variance in reproductive success and the Ne/N ratio.
Evolution 59, 1596–1599.
Hedrick, P.W., 2006. Genetic polymorphism in heterogeneous environments: the
age of genomics. Annu. Rev. Ecol. Evol. Syst. 37, 67–93.
Hedrick, P.W., 2012. What is the evidence for heterozygote advantage selection?
Trends Ecol. Evol. 27, 698–704.
Hereford, J., 2009. A quantitative survey of local adaptation and fitness trade-offs.
Amer. Nat. 173, 579–588.
Houle, D., Morikawa, B., Lynch, M., 1996. Comparing mutational variabilities.
Genetics 143, 1467–1483.
IUCN, 2013. IUCN Red List of Threatened Species. <http://www.redlist.org/>.
Jamieson, I.G., Allendorf, F.W., 2012. How does the 50/500 rule apply to MVPs?
Trends Ecol. Evol. 27, 578–584.
Jones, G.P., Almany, G.R., Russ, G.R., Sale, P.F., Steneck, R.S., van Oppen, M.J.H., Willis,
B.L., 2009. Larval retention and connectivity among populations of corals and
reef fishes: history, advances and challenges. Coral Reefs 28, 307–325.
Kuparinen, A., Hutchings, J.A., 2012. Consequences of fisheries-induced evolution
for population productivity and recovery potential. Proc. Roy. Soc. Lond. B 279,
2571–2579.
Lande, R., Barrowclough, G.F., 1987. Effective population size, genetic variation, and
their use in population management. In: Soulé, M.E. (Ed.), Viable Populations for
Conservation. Cambridge University Press, Cambridge, pp. 87–123.
Latter, B.D.H., Mulley, J.C., 1995. Genetic adaptation to captivity and inbreeding
depression in small laboratory populations of Drosophila melanogaster. Genetics
139, 255–266.
Leroy, G., Mary-Huard, T., Verrier, E., Danvy, S., Charvolin, E., 2013. Methods to
estimate effective population size using pedigree data: examples in dog, sheep,
cattle and horse. Genet. Sel. Evol. 45, 1.
Lindsley, D.L., Zimm, G.G., 1992. The Genome of Drosophila melanogaster. Academic
Press, San Diego.
Luikart, G., Ryman, N., Tallmon, D.A., Schwartz, M.K., Allendorf, F.W., 2010.
Estimation of census and effective population sizes: the increasing usefulness
of DNA-based approaches. Conserv. Genet. 11, 355–373.
Mace, G.M., Lande, R., 1991. Assessing extinction threats: towards a reevaluation of
IUCN threatened species categories. Conserv. Biol. 5, 148–157.
Mace, G.M., Collar, N.J., Gaston, K.J., Hilton-Taylor, C., Akçakaya, H.R., LeaderWilliams,
N., Milner-Gulland, E.J., Stewart, S.N., 2008. Quantification of
62 R. Frankham et al. / Biological Conservation 170 (2014) 56–63
extinction risk: IUCN’s system for classifying threatened species. Conserv. Biol.
22, 1424–1442.
Mattila, A.L.K., Duplouy, A., Kirjokangas, M., Lehtonen, R., Rastas, P., Hanski, I., 2012.
High genetic load in an old isolated butterfly population. Proc. Nat. Acad. Sci.
U.S.A. 109, E2496–E2505.
McDonald-Madden, E., Bode, M., Game, E.T., Grantham, H., Possingham, H.P., 2008.
The need for speed: informed land acquisitions for conservation in a dynamic
property market. Ecol. Lett. 11, 1169–1177.
McNeely, J.A., Miller, K.R., Reid, W.V., Mittermeier, R.A., Werner, T.B., 1990.
Conserving the World’s Biological Diversity. IUCN, World Resources Institute,
Conservation International, WWF-US and the World Bank, Washington, DC.
Mousseau, T.A., Roff, D.A., 1987. Natural selection and the heritability of fitness
components. Heredity 59, 181–197.
Nei, M., 1968. The frequency distribution of lethal chromosomes in finite
populations. Proc. Nat. Acad. Sci. U.S.A. 60, 517–524.
Newman, D., Pilson, D., 1997. Increased probability of extinction due to decreased
genetic effective population size: experimental populations of Clarkia pulchella.
Evolution 51, 354–362.
O’Grady, J.J., Brook, B.W., Reed, D.H., Ballou, J.D., Tonkyn, D.W., Frankham, R., 2006.
Realistic levels of inbreeding depression strongly affect extinction risk in wild
populations. Biol. Conserv. 133, 42–51.
O’Grady, J.J., Reed, D.H., Brook, B.W., Frankham, R., 2008. Extinction risk scales
better to generations than years. Anim. Conserv. 11, 442–451.
Palstra, F.P., Fraser, D.J., 2012. Effective/census population size ratio estimation: a
compendium and appraisal. Ecol. Evol. 2, 2357–2365.
Palstra, F.P., Ruzzante, D.E., 2008. Genetic estimates of contemporary effective
population size: what can they tell us about the importance of genetic
stochasticity for wild population persistence? Mol. Ecol. 17, 3428–3447.
Perfeito, L., Sousa, A., Bataillon, T., Gordo, I., 2013. Rates of fitness decline and
rebound suggest pervasive epistasis. Evolution. http://dx.doi.org/10.1111/
evo.12234.
Piou, C., Prévost, E., 2012. A demo-genetic individual-based model for Atlantic
salmon populations: model structure, parameterization and sensitivity. Ecol.
Model. 231, 37–52.
Pressey, R.L., Bottrill, M.C., 2008. Opportunism, threats, and the evolution of
systematic conservation planning. Conserv. Biol. 22, 1340–1345.
Prout, T., 2000. How well does opposing selection maintain variation? In: Singh,
R.S., Krimbas, C.B. (Eds.), Evolutionary Genetics: From Molecules to
Morphology. Cambridge University Press, New York, pp. 157–181.
Reed, D.H., Bryant, E.H., 2000. Experimental test of minimum viable population size.
Anim. Conserv. 3, 7–14.
Reed, T.E., Schindler, D.E., Hague, M.J., Patterson, D.A., Meir, E., Waples, R.S., Hinch,
S.G., 2011. Time to evolve? Potential evolutionary responses of Fraser River
sockeye salmon to climate change and effects on persistence. PLoS One 6 (6),
e20380.
Robertson, A., 1962. Selection for heterozygotes in small populations. Genetics 47,
1291–1300.
Rodríguez-Ramilo, S.T., Pérez-Figuero, A., Fernández, B., Fernández, J., Caballero, A.,
2004. Mutation-selection balance accounting for genetic variation for viability
in Drosophila melanogaster as deduced from an inbreeding and artificial
selection experiment. J. Evol. Biol. 17, 528–541.
Roff, D.A., 1998. The maintenance of phenotypic and genetic variation in threshold
traits by frequency-dependent selection. J. Evol. Biol. 11, 513–529.
Saccheri, I., Kuussaari, M., Kankare, M., Vikman, P., Fortelius, W., Hanski, I., 1998.
Inbreeding and extinction in a butterfly metapopulation. Nature 392, 491–494.
Shaffer, M.K., 1981. Minimum viable populations size for species conservation.
Bioscience 31, 131–134.
Simmons, M.J., Crow, J.F., 1977. Mutations affecting fitness in Drosophila
populations. Annu. Rev. Genet. 11, 49–78.
Soulé, M.E., 1980. Thresholds for survival: maintaining fitness and evolutionary
potential. In: Soulé, M.E., Wilcox, B.A. (Eds.), Conservation Biology: An
Evolutionary-Ecological Perspective. Sinauer, Sunderland, MA, pp. 151–169.
Soulé, M.E., 1985. What is conservation biology? Bioscience 35, 727–734.
Soulé, M.E. (Ed.), 1987. Viable Populations for Conservation. Cambridge University
Press, Cambridge.
Spielman, D., Brook, B.W., Frankham, R., 2004. Most species are not driven to
extinction before genetic factors impact them. Proc. Nat. Acad. Sci. U.S.A. 101,
15261–15264.
Traill, L.W., Bradshaw, C.J.A., Brook, B.W., 2007. Minimum viable population size: a
meta-analysis of 30 years of published estimates. Biol. Conserv. 139, 159–166.
Vilas, C., Miguel, E.S., Amaro, R., Garcia, C., 2006. Relative contribution of inbreeding
depression and eroded adaptive diversity to extinction risk in small populations
of shore campion. Conserv. Biol. 20, 229–238.
Waples, R.S., 2002. Evaluating the effect of stage-specific survivorship on the Ne/N
ratio. Mol. Ecol. 11, 1029–1037.
Waples, R.S., Luikart, G., Faulkner, J.R., Tallmon, D.A., 2013. Simple life-history traits
explain key effective population size ratios across diverse taxa. Proc. Roy. Soc.
Lond. B 280, doi.org/10.1098/rspb.2013.1339.
White, G.M., Boshier, D.H., Powell, W., 2002. Increased pollen flow counteracts
fragmentation in a tropical dry forest: An example from Swietenia humilis
Zaccarini. Proc. Nat. Acad. Sci. U.S.A. 99, 2038–2042.
Woodworth, L.M., Montgomery, M.E., Briscoe, D.A., Frankham, R., 2002. Rapid
genetic deterioration in captivity: causes and conservation implications.
Conserv. Genet. 3, 277–288.
Wright, S., 1969. Evolution and the Genetics of Populations. 2. The Theory of Gene
Frequencies. University of Chicago Press, Chicago, IL.
Zeigler, S.L., Che-Castaldo, J.P., Neel, M.C., 2013. Actual and potential use of
population viability analyses in recovery of plant species listed under the U.S.
Endangered Species Act. Conserv. Biol. http://dx.doi.org/10.1111/cobi.12130.
R. Frankham et al. / Biological Conservation 170 (2014) 56–63 63