Simultaneous inbreeding and outbreeding depression in
reintroduced Arabian oryx
INTRODUCTION
Animal and plant breeders have known for a long time
that inbreeding generates offspring with lower fitness than
their non-inbred counterparts (Darwin, 1868; Falconer &
Mackay, 1996), an observation termed inbreeding depression.
Inbreeding depression is thought to arise from the
accumulation of deleterious recessive mutations or the
existence of loci with heterozygote advantage
(Charlesworth & Charlesworth, 1987). More recently a
number of studies have demonstrated inbreeding depression
in non-domestic animals both in captivity (for
reviews, see Ralls, Brugger & Ballou, 1979; Ballou &
Ralls, 1982; Lacy, Petric & Warneke, 1993) and in the
wild (Greenwood, Harvey & Perrins, 1978; Van
Noordwijk & Scharloo, 1981; Ralls, Harvey & Lyles,
1986; Stockley et al., 1993; Bensch, Hasselquist & Von
Schantz, 1994; Jiménez et al., 1994; Keller et al., 1994;
Coltman, Bowen & Wright, 1998; Coulson et al., 1998;
Keller, 1998; Coltman et al., 1999; Slate et al., 2000),
although some studies of wild populations (Gibbs &
Grant, 1989; Hoogland, 1992; Grant & Grant, 1995) failed
to find inbreeding depression (see also reviews by
Frankham, 1995a; Crnokrak & Roff, 1999). Inbreeding
depression may be sufficiently intense to cause local population
extinction if population sizes are small (Frankham,
1995b; Saccheri et al., 1998; see also Frankham, 1998).
However, offspring fitness may not be maximized by
breeding with the most distantly related mate
(Templeton et al., 1986). Low fitness of offspring under
outbreeding, known as outbreeding depression, occurs
either by the disruption of complexes of interacting
genes that have jointly evolved under natural selection
or by mixing of genomes adapted to different environments
(Templeton, 1986; Waser, 1993). Outbreeding
depression is thought to be less common than inbreeding
depression. It is most frequently observed in plants
(for a review, see Waser, 1993) but has also been
detected in marine invertebrates (for a review, see
Knowlton & Jackson, 1993) and in captive callimico
(Callimico goeldii; Lacy et al., 1993). Most recently,
Coulson et al. (1998, 1999) found that different components
of fitness in free-living red deer (Cervus elaphus)
may be negatively or positively associated with
outbreeding.
Here we examine the impact of inbreeding and outbreeding
on juvenile survival of reintroduced Arabian
Animal Conservation (2000) 3, 241–248 © 2000 The Zoological Society of London Printed in the United Kingdom
T. C. Marshall1
and J. A. Spalton2
1
Institute of Cell, Animal and Population Biology, University of Edinburgh, West Mains Road, Edinburgh EH9 3JT, UK
2
Office of the Adviser for Conservation of the Environment, Diwan of Royal Court, P.O. Box 246, Muscat 113, Sultanate of Oman
(Received 4 October 1999; accepted 7 March 2000)
Abstract
In most species the offspring of closely related parents have reduced fitness compared with the offspring
of unrelated parents, a phenomenon known as inbreeding depression. However if parents are
very distantly related, their offspring may also have reduced fitness. This pattern, outbreeding depression,
has been most commonly observed in plants and only rarely in animals. Here we examine the
consequences of inbreeding and outbreeding on juvenile survival of reintroduced Arabian oryx (Oryx
leucoryx) in Oman, a population with a small number of founders drawn from a number of sources.
Using microsatellite-based measures of inbreeding and outbreeding, there was no apparent relationship
between inbreeding or outbreeding and survival when inbreeding and outbreeding were tested in
separate statistical models. However when inbreeding and outbreeding were tested in the same
statistical model, we found simultaneous inbreeding depression and outbreeding depression acting
on juvenile survival. Outbreeding depression may be more common in vertebrates than previously
supposed, and conservation strategies that seek to maximize the genetic diversity of managed populations
may risk mixing lineages that are sufficiently differentiated to cause outbreeding depression
among descendants.
All correspondence to : Dr Tristan Marshall. Tel: +44 7932 698
445; Fax: +44 131 650 6564; E-mail: tristan.marshall@ed.ac.uk.
oryx (Oryx leucoryx), a species regarded as a paradigm
of conservation by captive breeding and reintroduction
(Stanley Price, 1989). The captive populations that provided
oryx for the reintroduction in Oman had a small
number of founding animals. The reintroduced population
may therefore be vulnerable to inbreeding depression.
However the captive populations are themselves
thought to have been founded from several geographical
areas, and it is therefore possible that hybrids
between lineages may be vulnerable to outbreeding
depression. In this study we use microsatellite-based
measures of inbreeding and outbreeding to show that
reintroduced Arabian oryx in Oman suffer, simultaneously,
from both inbreeding depression and outbreeding
depression.
MATERIALS AND METHODS
Study population
The Arabian oryx was formerly distributed across most
of the Arabian peninsula, but hunting pressure, particularly
in the 1950s and 1960s, led to a dramatic decline
in its range and eventually to extinction in the wild in
1972 (Henderson, 1974). Captive breeding was initiated
at Phoenix Zoo, Arizona in 1963 using three oryx captured
in eastern Yemen in 1962, one captured in central
Oman in 1960, four captive oryx from Saudi Arabia and
one non-breeding captive animal from Kuwait. A second
breeding group was founded at Los Angeles Zoo in 1967
with a pair from the same Saudi Arabian group that contributed
to the Phoenix breeding programme. Between
1963 and 1977, the captive population in the USA
increased at 17.2% per year and, in 1978, the first oryx
were sent back to the Middle East (Stanley Price, 1989).
From 1980 onwards Arabian oryx were gradually
reintroduced to the Jiddat-al-Harasis, a stony desert
plateau in central Oman that is now a World Heritage
Site known as the Arabian Oryx Sanctuary (AOS). Five
releases took place in the AOS in the 1980s and early
1990s involving 34 captive-bred animals, mostly from
the USA but also some animals of Qatari ancestry
(Fig. 1). A three year drought and the presence of a
sterile dominant male slowed the rate of increase of the
242 T. C. MARSHALL AND J. A. SPALTON
Fig. 1. The sources of Arabian oryx reintroduced to the Arabian Oryx Sanctuary, Oman up to December 1993. ‘Operation
Oryx’ was a capture expedition mounted in the Eastern Aden Protectorate, now Yemen (Grimwood, 1962), and the animal
from London Zoo was captured in central Oman (Woolley, 1962). HM King Saud of Saudi Arabia provided two pairs of oryx
for the initial breeding herd in the USA at Phoenix Zoo, and it is thought that a third pair, later sent from Riyadh to Los Angeles
Zoo, was from the same source. King Saud’s oryx are thought to have originated from one or more capture expeditions. Al
Wukayr and Al Sulaimi were two separate private collections in Qatar thought to have been founded independently using wildcaught
oryx. A non-breeding female sent from Kuwait to Phoenix Zoo has been omitted for clarity. The Shaumari Wildlife
Reserve in eastern Jordan and the Bait Barakah Breeding Centre in Muscat, Oman are two additional locations at which oryx
reintroduced to the Arabian Oryx Sanctuary were bred. The number of oryx transferred is indicated in each case, and a bold
arrow is used to indicate the lineage that provided the largest number of oryx to found each group. Further details of the history
of Arabian oryx in the Middle East may be found in Stanley Price (1989), Marshall (1998) and Marshall et al. (1999).
reintroduced population during the early years. Following
rainfall in 1986, when supplementary feeding was discontinued,
the population increased at an average rate of
31% per year between 1987 and 1993 (Fig. 2).
Between January 1987 and May 1993, J.A.S. collected
detailed data to assess the factors influencing mortality
of the individually-monitored Arabian oryx population
in the Sanctuary (Spalton, 1995). A total of 173 calves
were born during this period. Mortality of calves during
their first year of life accounted for 53% of all recorded
deaths and was concentrated in the drought of 1991
(Fig. 3; Spalton, 1993). Blood samples were obtained
from live oryx whenever possible, and skin or horn samples
were collected post mortem. Oryx that died as juveniles
were significantly more likely to have been
sampled than oryx that survived (16 out of 27 versus 41
out of 146; χ2
= 8.66, d.f. = 1, P = 0.003), but we believe
that among oryx that died and oryx that survived, probability
of sampling was random with respect to inbreeding
and outbreeding, and therefore that statistical models
relating juvenile survival to inbreeding and outbreeding
should not be affected by the bias in sampling in favour
of oryx that died.
In a previous analysis, crude protein content of
Stipagrostis spp., the preferred food of Arabian oryx,
was calculated for each oryx on each day. Information
on the decline in crude protein content following last
rainfall (Spalton, 1999), obtained from a detailed vegetation
survey carried out in 1989, was combined with
long-term spatial rainfall records and daily data on the
location of each oryx. Percentage crude protein content
of Stipagrostis grasses available to the calf’s mother in
the 30 days before and 60 days after parturition was the
major factor influencing juvenile survival of the 173
Arabian oryx born between January 1987 and May 1993.
Higher levels of crude protein, corresponding to periods
immediately following rainfall, led to higher juvenile
survival (Spalton, 1995).
Measures of inbreeding and outbreeding
DNA samples were obtained from 57 oryx born between
1987 and 1993 (33% of 173). We typed all 57 oryx for
six microsatellite loci (Marshall et al., 1999). Using
these data we calculated individual heterozygosity (a
measure of inbreeding) and the microsatellite-based
measure mean d2
(a measure of outbreeding) across the
six loci for each oryx (Table 1).
Heterozygosity at each locus is a binary variable, with
the value 1 if alleles differ and the value 0 if they do
not. Individual heterozygosity is the mean across loci of
these heterozygosity values. More inbred animals have
lower individual heterozygosity because there is a higher
probability of both alleles at each locus having been
descended from a single common ancestor. Although in
our analysis individual heterozygosity could only take
five discrete values (the 57 oryx were heterozygous for
between zero and four loci out of six), we treated it as
a continuous variable to give optimum statistical power,
noting that the generalized linear models used in our
analysis (see Statistical analysis, below) are tolerant
of explanatory variables that deviate from a normal
distribution.
243Inbreeding and outbreeding depression
Fig. 2. Number of Arabian oryx in the reintroduced population
on the Jiddat-al-Harasis, Oman on 31st December each
year from 1980–1993, including animals in the pre-release
enclosure. Note that only 34 of the 41 captive-bred oryx
brought to the Jiddat-al-Harasis were actually released during
this period. [ss], wild-born oryx; [ ], surviving captive-bred
oryx; [s], new captive-bred oryx.
Table 1. Microsatellite loci used in analysis of inbreeding and outbreeding
in Arabian oryx
Locus Number of Average Maximum Average
alleles heterozygosity d2
d2
RPB3 2 0.088 1 0.088
OarFCB304 2 0.456 1 0.456
MAF50 4 0.456 4 0.956
MAF46 4 0.421 25 4.281
OarCP26 5 0.491 25 2.018
OarHH64 5 0.596 64 5.737
Primer references and typing protocols may be found in Marshall
et al. (1999). The figures given for the number of alleles, average
observed heterozygosity, maximum d2
and average d2
score refer to
the 57 oryx that were sampled in this study.
Fig. 3. Number of Arabian oryx calves born on the Jiddat-alHarasis,
Oman from 1987–1993. For each year calves that died
during their first year of life (s) are shown separately from
calves that survived (ss) this period. The data for 1993 include
only calves born before 8th May.
Mean d2
(Coulson et al., 1998) is related to individual
heterozygosity but incorporates allele length information.
The value of d2
at each locus is the squared
difference in allele lengths, measured in repeat units (one
repeat was 2 base pairs (bp) for all loci in this study).
Mean d2
is the mean across loci of these d2
values. More
outbred individuals have higher mean d2
scores because
on average at each locus a longer time has elapsed since
the two alleles at each locus shared a common ancestor,
and hence there has been more opportunity for the allele
lengths to diverge via stepwise mutation (Pemberton et
al., 1999). Mean d2
was treated as a continuous variable.
Although the maximum d2
varies considerably between
loci (Table 1), it is not appropriate to scale d2
values by
maximum d2
at each locus prior to taking the mean (this
procedure would constrain scaled d2
values to vary
between zero and one), because those loci with high
maximum d2
carry much more information about outbreeding
than loci with low maximum d2
.
Following Coulson et al. (1999) we also calculated
mean d2
outbreeding by taking the mean of d2
values for each
animal across heterozygous loci only. Because three
oryx were homozygous at all six loci, mean d2
outbreeding
could be calculated for only 54 out of the 57 sampled
oryx. The distributions of mean d2
and mean d2
outbreeding
scores were right-skewed, and, following Coltman et al.
(1998), scores were transformed as loge (x + 1) prior to
statistical analysis.
Individual heterozygosity was positively correlated
with mean d2
(r = 0.522, d.f. = 55, P < 0.001) but not
significantly correlated with mean d2
outbreeding (r = 0.210,
d.f. = 52, P = 0.131). Mean d2
and mean d2
outbreeding were
highly positively correlated (r = 0.956, d.f. = 52,
P < 0.001).
Statistical analysis
Logistic regression models (McCullagh & Nelder, 1989)
were developed using GENSTAT 5 (NAG, Oxford), with
a binomial response of 1 representing survival to one
year of age and 0 representing failure to survive to one
year. In all models we fitted a term representing the percentage
crude protein content of Stipagrostis spp.
(Spalton, 1995, 1999) prior to testing the genetic terms
individual heterozygosity, mean d2
and mean d2
outbreeding.
The significance of terms not included in the final model
was assessed by the change in deviance, distributed
approximately as χ2
, when they were added to the final
model, while the significance of terms included in the
final model was assessed by the change in deviance
when they were dropped from the final model.
RESULTS
We assessed the impact of inbreeding and outbreeding
on survival of 57 Arabian oryx calves for which individual
heterozygosity and mean d2
were available.
Controlling for crude protein content of Stipagrostis spp.
(X2
= 11.09, d.f. = 1, P = 0.001), we found that fitted
singly, neither individual heterozygosity (X2
= 2.29,
d.f. = 1, P = 0.130) nor mean d2
(X2
= 1.94, d.f. = 1,
P = 0.164) were associated with juvenile survival.
However when individual heterozygosity and mean d2
were fitted to the model jointly, both were significant
(individual heterozygosity: X2
= 7.14, d.f. = 1, P = 0.008;
mean d2
: X2
= 6.79, d.f. = 1, P = 0.009; both terms
together: X2
= 9.09, d.f. = 2, P = 0.011). Crude protein
and individual heterozygosity were positively associated
with juvenile survival and mean d2
was negatively associated
with juvenile survival (Fig. 4). Crude protein,
individual heterozygosity and mean d2
together
explained 27% of the total deviance in the model. The
robustness of the model was supported by a balanced
bootstrap analysis (1000 replicates): 95% confidence
intervals for crude protein, individual heterozygosity and
mean d2
coefficients in the logistic model did not include
zero. Furthermore restricting the analysis to those animals
born in less favourable environmental conditions
(when crude protein content of Stipagrostis spp. was less
than 6.5%: see Fig. 4(a)) made no difference to the
results (data not shown).
We examined the model further by replacing mean d2
with mean d2
outbreeding, a derivative of mean d2
that is not
correlated with individual heterozygosity because mean
d2
outbreeding is calculated only across heterozygous loci
(see Materials and Methods, above). Controlling for
crude protein content of Stipagrostis spp. (X2
= 10.30,
d.f. = 1, P = 0.001), we found that fitted singly, individual
heterozygosity was not associated with juvenile
survival (X2
= 2.43, d.f. = 1, P = 0.119) while mean
d2
outbreeding was negatively associated with juvenile
survival (X2
= 4.46, d.f. = 1, P = 0.035). When individual
heterozygosity and mean d2
outbreeding were fitted to the
model jointly, both were significant (individual heterozygosity:
X2
= 4.60, d.f. = 1, P = 0.032; mean
d2
outbreeding: X2
= 6.63, d.f. = 1, P = 0.010; both terms
together: X2
= 9.05, d.f. = 2, P = 0.011). Again crude
protein and individual heterozygosity were positively
associated with juvenile survival and mean d2
outbreeding
was negatively associated with juvenile survival. This
model illustrates clearly that the association between
mean d2
and survival is independent of the association
between individual heterozygosity and survival.
Following the methodology described by Coulson et al.
(1999) and Pemberton et al. (1999), we also tested the
model for dependence on particular loci by testing each
locus individually and by dropping each locus in turn from
the calculation of both individual heterozygosity and
mean d2
(Table 2). Of the individual loci tested, only d2
at locus OarCP26 showed a significant association with
survival (X2
= 4.52, d.f. = 1, P = 0.038). However the
influence of this locus alone cannot explain the strength
of the association found between overall mean d2
and survival
(X2
= 6.79, d.f. = 1, P = 0.009), implying that other
loci must have contributed to this association.
When each locus was dropped in turn from the analysis,
the model changed in two cases. First, individual
heterozygosity became non-significant when the locus
MAF50 was excluded (X2
= 3.24, d.f. = 1, P = 0.071),
and second, mean d2
became non-significant when the
244 T. C. MARSHALL AND J. A. SPALTON
locus OarCP26 was excluded (X2
= 2.98, d.f. = 1,
P = 0.084). However in both cases there remained a
trend (P < 0.1) in the predicted direction. We conclude
that the effects of individual heterozygosity and mean d2
across all six loci cannot be completely explained by the
influence of one locus alone.
DISCUSSION
Individual heterozygosity measures recent inbreeding,
and mean d2
, while incorporating a component of heterozygosity,
is thought to be most sensitive to outbreeding,
particularly between populations that have
245Inbreeding and outbreeding depression
Fig. 4. Association of crude protein, individual heterozygosity and mean d2
with juvenile survival. (a) and (b) relationship
between crude protein and probability of survival. (c) and (d) relationship between individual heterozygosity and probability
of survival. (e) and (f) relationship between mean d2
and probability of survival. Plots (a), (c) and (e) show the fitted survival
probability from the full model for each animal plotted against the explanatory variable, while plots (b), (d) and (f) show the
predicted survival probability for 21 selected values and approximate standard errors of the predictions (error bars), along with
a logistic curve illustrating the predicted relationship between survival probability and the explanatory variable. Predicted survival
probabilities were obtained by setting each of the two explanatory variables not under consideration to its mean value
across all individuals.
been separated sufficiently long for microsatellite allele
lengths to diverge (Pemberton et al., 1999). We therefore
interpret the positive association between individual
heterozygosity and juvenile survival as inbreeding
depression, and the negative association between mean
d2
and juvenile survival as outbreeding depression. The
fact that both very inbred and very outbred calves had
low survival explains why neither measure showed an
association with survival when fitted singly.
Only six loci were used in this study, and hence estimates
of inbreeding provided by heterozygosity and of
outbreeding provided by mean d2
are relatively crude.
However in common with several previous studies using
individual heterozygosity and mean d2
as measures of
inbreeding and outbreeding (Coltman et al., 1999;
Coulson et al., 1999; Pemberton et al., 1999), we have
tested the influence of individual loci, and believe that
the absence of strong influences of individual loci on fitness
favours the idea that individual heterozygosity and
mean d2
are measures of genome-wide inbreeding and
outbreeding.
Given the modest number of individuals and loci
typed, it is surprising that we were able to demonstrate
inbreeding depression and outbreeding depression, and
implies that these are strong influences on juvenile survival.
Although deviances are not strictly additive in a
logistic model, the change in deviance when individual
heterozygosity and mean d2
were both included in the
model of survival (X2
= 9.09) is of similar magnitude to
the change in deviance when crude protein was included
in the model (X2
= 11.09), suggesting that inbreeding
depression and outbreeding depression are together
almost as important an influence on juvenile survival as
the quality of grazing available to the oryx.
Inbreeding depression may be concealed by
outbreeding depression
Inbreeding depression is an almost routine observation
in many higher taxa including ruminants (Ralls et al.,
1979; Ballou & Ralls, 1982), and most studies use pedigree-based
analysis of inbreeding coefficients to look for
evidence of inbreeding depression (Crnokrak & Roff,
1999). A pedigree-based analysis in this same population
of Arabian oryx did not detect a significant association
between inbreeding coefficient and juvenile
survival among 122 calves, although there was a nonsignificant
trend (P < 0.1) for lower survival among more
inbred calves (T. C. M., unpublished results). The
absence of an association between inbreeding coefficient
and fitness was surprising given that 24% of Arabian
oryx in this population are closely (F = 0.25) or moderately
(F = 0.125) inbred (T. C. M., unpublished
results), giving a substantial sample of inbred matings
in which to detect inbreeding depression.
To our knowledge this is the first study to document
inbreeding depression concealed by simultaneous outbreeding
depression. We suggest that studies finding no
association between inbreeding and fitness should be
wary of concluding that inbreeding depression is absent,
and that it should become standard practice to test for
outbreeding depression as well as inbreeding depression.
Causes of outbreeding depression in Arabian oryx
Hybridization is a term typically used to indicate mating
between individuals belonging to distinct species.
Hybridization commonly results in hybrid progeny with
low fitness, if viable hybrids can be formed at all
(Harrison, 1993; see also Greig, 1979). In contrast, the
term outbreeding is most often used to refer to mating
of distantly related individuals within a given species.
Outbreeding depression therefore generally refers to the
reduced fitness of progeny of distantly related individuals
within a species. The implication is that the two individuals
are so distantly related that their genes are no
longer adapted to work with each other or with the environment
in which they find themselves (Templeton,
1986; Waser, 1993).
The founders of the Arabian oryx population studied
here are from at least five independent capture expeditions
(Fig. 1). While documentation on all except two
of these founder groups is extremely limited, it is
unlikely that all five populations were founded with animals
from the same part of the Arabian oryx’s former
range. The existence of outbreeding depression suggests
that two or more of the founder animals or groups came
from genetically differentiated populations.
Surprisingly allozyme and microsatellite data suggest
relatively low levels of genetic differentiation between
contemporary Arabian oryx populations (Vassart,
Granjon & Greth, 1991a; Marshall et al., 1999). This
may be because these populations are mixtures of lineages,
and samples representing the lineages prior to
mixing are not available. However previous karyotyp-
246 T. C. MARSHALL AND J. A. SPALTON
Table 2. An evaluation of the influence of individual loci on the model
of the relationship between individual heterozygosity, mean d2
and
juvenile survival
Locus Effect of locus fitted Effect of
individually removing locus
Heterozygosity d2
Heterozygosity Mean d2
RPB3 N/A N/A 0.011 0.014
OarFCB304 N/A N/A 0.045 0.025
MAF50 0.692 0.216 0.077 0.019
MAF46 0.613 0.346 0.007 0.016
OarCP26 0.110 0.038 0.017 0.090
OarHH64 0.386 0.319 0.009 0.015
In each cell the P-value for a model testing the influence of heterozygosity
or d2
/mean d2
at the given locus is shown. P-values shown
in bold are significant (P < 0.05). Crude protein was fitted in all models;
in models testing heterozygosity, d2
or mean d2
was fitted, while
in models testing d2
or mean d2
, heterozygosity was fitted. The effect
of RPB3 and OarFCB304 as individual loci could not be assessed
since these two loci were biallelic, and therefore heterozygosity and
d2
terms were completely correlated.
ing work on Arabian oryx (in which the standard chromosomal
complement is 2n = 58) detected a
Robertsonian translocation (a fusion between chromosomes
17 and 19) segregating in several populations in
the Middle East (Vassart et al., 1991b). In a limited survey
we found that two males, introduced to the AOS
population from Shaumari in Jordan, were heterozygous
(2n = 57) for this same translocation (J. A. S., unpublished
results). The pattern of occurrence in all herds is
consistent with a single source of the translocation in the
Al Wukayr population in Qatar, and this suggests some
degree of genetic differentiation between the Al Wukayr
lineage and other lineages (Fig. 1).
Implications for conservation
Our findings are important for two reasons. First, the
existence of outbreeding depression implies that prior to
extinction in the wild, there were genetically distinct but
phenotypically cryptic races of Arabian oryx. Founders
of the reintroduced population came from a vast span of
the Middle East, and in each geographical area, genes
may have become adapted to work with each other or
with the local environment (Templeton, 1986). Our
study suggests that intra-specific genetic differentiation
sufficient to cause outbreeding depression may be a more
widespread phenomenon in vertebrates than was previously
thought.
Second, well-managed captive-breeding programmes
aim to minimize inbreeding. However, in seeking to
avoid inbreeding depression and ensuring future adaptability
by maximizing the number of founders, populations
managed for conservation may be vulnerable to
outbreeding depression if founders are taken from genetically
differentiated populations. In captive callimico, in
which simultaneous inbreeding and outbreeding depression
has also been found, inbreeding was an order of
magnitude more powerful an influence on fitness than
outbreeding (Lacy et al., 1993). However in AOS
Arabian oryx, outbreeding depression is a much more
important influence on fitness: outbreeding explains as
much variation in juvenile survival as inbreeding.
In Omani Arabian oryx, the high rate of intrinsic population
growth (Fig. 2) suggests that the simultaneous
inbreeding and outbreeding depression acting on the
population is not currently a major threat to population
viability. We note that the level of inbreeding in the AOS
population of Arabian oryx (average F = 0.10) is considerably
lower than the level at which population viability
is likely to be threatened by inbreeding (Frankham,
1995b). Furthermore, subsequent experience has shown
us that human factors, particularly illegal hunting, represent
much greater threats to reintroduced Arabian oryx
(Spalton, Lawrence & Brend, 1999). However, in other
less fecund species, the joint action of inbreeding depression
and outbreeding depression may in itself represent
a serious obstacle to successful captive breeding and
reintroduction.
Acknowledgements
We thank Sultan Qaboos bin Said for funding of the
White Oryx Project, and the many people who have been
involved with the project over the last 20 years, especially
Mark Stanley Price and Ralph Daly. Josephine
Pemberton, Tim Coulson, Loeske Kruuk, Dave Coltman,
Jon Slate, Aubrey Manning, Georgina Mace and two
anonymous referees provided helpful comments on earlier
drafts of this paper. This work was carried out while
T. C. M. was funded by a BBSRC studentship.
REFERENCES
Ballou, J. & Ralls, K. (1982). Inbreeding and juvenile mortality
in small populations of ungulates: a detailed analysis. Biol.
Conserv. 24: 239–272.
Bensch, S., Hasselquist, D. & Von Schantz, T. (1994). Genetic
similarity between parents predicts hatching failure: nonincestuous
inbreeding in the great reed warbler? Evolution 48:
317–326.
Charlesworth, D. & Charlesworth, B. (1987). Inbreeding depression
and its evolutionary consequences. Annu. Rev. Ecol. Syst.
18: 237–268.
Coltman, D. W., Bowen, W. D. & Wright, J. M. (1998). Birth
weight and neonatal survival of harbour seal pups are positively
correlated with genetic variation measured by microsatellites.
Proc. R. Soc. Lond. ser. B 265: 803–809.
Coltman, D. W., Pilkington, J. G., Smith, J. A. & Pemberton,
J. M. (1999). Parasite-mediated selection against inbred Soay
sheep in a free-living island population. Evolution 53:
1259–1267.
Coulson, T. N., Pemberton, J. M., Albon, S. D., Beaumont,
M. A., Marshall, T. C., Slate, J., Guinness, F. E. & CluttonBrock,
T. H. (1998). Microsatellites reveal heterosis in red deer.
Proc. R. Soc. Lond. ser. B 265: 489–495.
Coulson, T. N., Albon, S. D., Slate, J. & Pemberton, J. M. (1999).
Microsatellite loci reveal sex-dependent responses to inbreeding
and outbreeding in red deer calves. Evolution 53: 1951–1960.
Crnokrak, P. & Roff, D. A. (1999). Inbreeding depression in the
wild. Heredity 83: 260–270.
Darwin, C. (1868). The variation of animals and plants under
domestication. London: John Murray.
Falconer, D. S. & Mackay, T. F. C. (1996). Introduction to quantitative
genetics. London: Longman.
Frankham, R. (1995a). Conservation genetics. Annu. Rev. Genet.
29: 305–327.
Frankham, R. (1995b). Inbreeding and extinction – a threshold
effect. Conserv. Biol. 9: 792–779.
Frankham, R. (1998). Inbreeding and extinction: island populations.
Conserv. Biol. 12: 665–675.
Gibbs, H. L. & Grant, P. R. (1989). Inbreeding in Darwin’s
medium ground finches (Geospiza fortis). Evolution 43:
1273–1284.
Grant, P. R. & Grant, B. R. (1995). The founding of a new population
of Darwin’s finches. Evolution 49: 229–240.
Greenwood, P. J., Harvey, P. H. & Perrins, C. J. (1978).
Inbreeding and dispersal in the great tit. Nature, Lond. 271:
52–54.
Greig, J. C. (1979). Principles of genetic conservation in relation
to wildlife management in Southern Africa. S. Afr. J. Wild. Res.
9: 57–78.
Grimwood, I. R. (1962). Operation Oryx. Oryx 6: 308–334.
Harrison, R. G. (Ed.) (1993). Hybrid zones and the evolutionary
process. New York: Oxford University Press.
Henderson, D. S. (1974). Were they the last Arabian oryx? Oryx
12: 347–350.
247Inbreeding and outbreeding depression
Hoogland, J. L. (1992). Levels of inbreeding among prarie dogs.
Am. Nat. 139: 591–602.
Jiménez, J. A., Hughes, K. A., Alaks, G., Graham, L. & Lacy,
R. C. (1994). An experimental study of inbreeding depression
in a natural habitat. Science 266: 271–273.
Keller, L. F. (1998). Inbreeding and its fitness effects in an insular
population of song sparrows (Melospiza melodia). Evolution
52: 240–250.
Keller, L. F., Arcese, P., Smith, J. N. M., Hochachka, W. M. &
Stearns, S. C. (1994). Selection against inbred song sparrows
during a natural population bottleneck. Nature, Lond. 372:
356–357.
Knowlton, N. & Jackson, J. B. C. (1993). Inbreeding and outbreeding
in marine invertebrates. In The natural history of
inbreeding and outbreeding: 200–249. Thornhill, N. W. (Ed.).
Chicago: University of Chicago Press.
Lacy, R. C., Petric, A. & Warneke, M. (1993). Inbreeding and
outbreeding in captive populations of wild animal species. In
The natural history of inbreeding and outbreeding: 352–374.
Thornhill, N. W. (Ed.). Chicago: University of Chicago Press.
Marshall, T. C. (1998). Inbreeding and fitness in wild ungulates.
PhD thesis: University of Edinburgh.
Marshall, T. C., Sunnucks, P., Spalton, J. A., Greth, A. &
Pemberton, J. M. (1999). Use of genetic data for conservation
management: the case of the Arabian oryx. Anim. Conserv. 2:
269–278.
McCullagh, P. & Nelder, J. A. (1989). Generalized linear models.
London: Chapman and Hall.
Pemberton, J. M., Coltman, D. W., Coulson, T. N. & Slate, J.
(1999). Using microsatellites to measure the fitness consequences
of inbreeding and outbreeding. In Microsatellites: evolution
and applications: 151–164. Goldstein, D. & Schlötterer,
C. (Eds). Oxford: Oxford University Press.
Ralls, K., Brugger, K. & Ballou, J. (1979). Inbreeding and juvenile
mortality in small populations of ungulates. Science 206:
1101–1103.
Ralls, K., Harvey, P. H. & Lyles, A. M. (1986). Inbreeding in
natural populations of birds and mammals. In Conservation biology
– the science of scarcity and diversity: 35–56. Soulé, M. E.
(Ed.). Sunderland, MA: Sinauer Associates.
Saccheri, I., Kuussaari, M., Kankare, M., Vikman, P., Fortelius,
W. & Hanski, I. (1998). Inbreeding and extinction in a butterfly
metapopulation. Nature, Lond. 392: 491–494.
Slate, J., Kruuk, L. E. B., Marshall, T. C., Pemberton, J. M. &
Clutton-Brock, T. H. (2000). Inbreeding depression influences
lifetime breeding success in a wild population of red deer
(Cervus elaphus). Proc. R. Soc. Lond. ser. B. (in press).
Spalton, A. (1993). A brief history of the reintroduction of the
Arabian oryx into Oman 1980–1992. Int. Zoo Yb. 32: 81–90.
Spalton, J. A. (1995). Effects of rainfall on the reproduction and
mortality of the Arabian oryx Oryx leucoryx (Pallas) in the
Sultanate of Oman. PhD thesis: University of Aberdeen.
Spalton, J. A. (1999). The food supply of Arabian oryx Oryx leucoryx
in the desert of Oman. J. Zool., Lond. 248: 433–441.
Spalton, J. A., Lawrence, M. W. & Brend, S. A. (1999). Arabian
oryx reintroduction in Oman: successes and setbacks. Oryx 33:
168–175.
Stanley Price, M. R. (1989). Animal re-introductions: the Arabian
oryx in Oman. Cambridge: Cambridge University Press.
Stockley, P., Searle, J. B., MacDonald, D. W. & Jones, C. S.
(1993). Female multiple mating behaviour in the common shrew
as a strategy to reduce inbreeding. Proc. R. Soc. Lond. ser. B
254: 173–179.
Templeton, A. R. (1986). Coadaptation and outbreeding depression.
In Conservation biology – the science of scarcity and diversity:
105–116. Soulé, M. E. (Ed.). Sunderland, MA: Sinauer
Associates.
Templeton, A. R., Hemmer, H., Mace, G., Seal, U. S., Shields,
W. M. & Woodruff, D. S. (1986). Local adaptation, coadaptation,
and population boundaries. Zoo Biol. 5: 115–125.
Van Noordwijk, A. J. & Scharloo, W. (1981). Inbreeding in an
island population of the great tit. Evolution 35: 674–688.
Vassart, M., Granjon, L. & Greth, A. (1991a). Genetic variability
in the Arabian oryx (Oryx leucoryx). Zoo Biol. 10: 399–408.
Vassart, M., Granjon, L., Greth, A., Asmodé, J. F. & Cribiu, E.
P. (1991b). Genetic research on captive Arabian Oryx in Saudi
Arabia. In Ungulates ‘91: 595–598. Spitz, F., Janeau, G.,
Gonzalez, G. & Aulagnier, S. (Eds). Toulouse, France:
S.F.E.P.M. – I.R.G.M.
Waser, N. M. (1993). Population structure, optimal outbreeding,
and assortative mating in angiosperms. In The natural history
of inbreeding and outbreeding: 173–199. Thornhill, N. W. (Ed.).
Chicago: University of Chicago Press.
Woolley, C. W. (1962). Letter. New Scientist 300: 368–369.
248 T. C. MARSHALL AND J. A. SPALTON