? 2005 The Society for the Study of Evolution. All rights reserved.
Evolution, 59(2), 2005, pp. 317–323
OUTBREEDING DEPRESSION, BUT NO INBREEDING DEPRESSION IN HAPLODIPLOID
AMBROSIA BEETLES WITH REGULAR SIBLING MATING
KATHARINA PEER1AND MICHAEL TABORSKY1,2
1Department of Behavioural Ecology, Institute of Zoology, University of Bern, Wohlenstrasse 50A,
CH-3032 Hinterkappelen, Switzerland
When mating partners are too closely related, increased homozygosity generally causes inbreeding depression, whereas
crossing between too distantly related individuals may disrupt local adaptations or coadaptations within the genome
and result in outbreeding depression. The optimal degree of inbreeding or outbreeding depends on population structure.
A long history of inbreeding is expected to reduce inbreeding depression due to purging of deleterious alleles, and
to promote outbreeding depression because of increased genetic variation between lineages. Ambrosia beetles (Xy-
leborini) are bark beetles with haplodiploid sex determination, strong local mate competition due to regular sibling
mating within the natal chamber, and heavily biased sex ratios. We experimentally mated females of Xylosandrus
germanus to brothers and unrelated males and measured offspring fitness. Inbred matings did not produce offspring
with reduced fitness in any of the examined life-history traits. In contrast, outcrossed offspring suffered from reduced
hatching rates. Reduction in inbreeding depression is usually attributed to purging of deleterious alleles, and the
absence of inbreeding depression in X. germanus may represent the highest degree of purging of all examined species
so far. Outbreeding depression within the same population has previously only been reported from plants. The causes
and consequences of our findings are discussed with respect to mating strategies, sex ratios, and speciation in this
In sexual reproduction the genetic similarity or dissimilarity between mates strongly affects offspring fitness.
Coleoptera, outcrossing, population structure, purging, reproductive isolation, Xyleborini.
Received February 25, 2004.Accepted November 10, 2004.
Inbreeding results in increased levels of homozygosity,
which usually lead to inbreeding depression (Keller and Wal-
ler 2002). The main mechanism that has been proposed to
explain the fitness reduction associated with inbreeding de-
pends on the genetic load of recessive deleterious alleles
(partial dominance hypothesis; Charlesworth and Charles-
worth 1999; Roff 2002). Increased homozygosity through
inbreeding results in increased expression of deleterious al-
leles, and thus inbreeding depression. Anything that would
reduce the genetic load would therefore also reduce inbreed-
ing depression, which results in two predictions. First, pro-
longed inbreeding should lead to purging of the genetic load
due to increased exposure of deleterious mutations in ho-
mozygotes (Lande and Schemske 1985; Charlesworth and
Charlesworth 1987; Waller 1993). Second, haplodiploid spe-
cies, in which deleterious mutations are regularly exposed to
selection in haploid males, should have a lower genetic load
than diploid species (Bruckner 1978; Crozier 1985; Werren
The first prediction has been confirmed by comparative
(Husband and Schemske 1996) and experimental studies
(Latta and Ritland 1994; McCall et al. 1994; Crnokrak and
Barrett 2002). However, in a variety of species and taxa with
regular inbreeding or selfing, inbreeding depression is still
observed (Ritland 1990; Demeester 1993; Doums et al. 1996;
Wedekind et al. 1998; Weeks et al. 1999; Haag et al. 2002).
The second prediction has received little attention until re-
cently, when comparative studies showed that haplodiploid
organisms experience less inbreeding depression than dip-
loids, although it may still be substantial (Antolin 1999; Hen-
ter 2003). Since both haplodiploidy and regular sibling mat-
ing should reduce deleterious load, one could conclude that
it should be lowest in haplodiploid species with chronic in-
breeding. However, theory predicts that haplodiploid species
benefit less from prolonged inbreeding than diploid species
(Werren 1993). In fact, genetic load of these females may
even increase slightly with continuous sibling mating. In
agreement with theoretical predictions, no difference was
found in the extent of inbreeding depression between pre-
sumably outcrossing and sib-mating species of haplodiploid
hymenoptera (Henter 2003). Therefore, inbreeding depres-
sion may be expected in haplodiploid organisms even in the
case of no male dispersal and exclusive sibling mating.
Although outbreeding is generally associated with elevated
fitness compared to inbreeding, this is only true up to certain
levels of parental dissimilarity. Above this level, a decrease
in fitness known as outbreeding depression can occur (e.g.,
Price and Waser 1979; Mitton 1993). Mechanisms respon-
sible for this effect may be either physiological, such as un-
derdominance (heterozygote disadvantage), breaking up of
coadapted gene complexes and epistatic interactions between
alleles, or environmental, such as disruption of local adap-
tations (Price and Waser 1979; Waser and Williams 2001;
Edmands 2002). In a number of plants, there is evidence for
both inbreeding and outbreeding depression (Waser 1993a;
Barrett and Harder 1996). Many studies have reported a neg-
ative relationship between parental divergence and offspring
fitness (see Edmands 2002), and there are species in which
fitness achieves an optimum at intermediate levels of simi-
larity (Waser 1993a; Waser et al. 2000). However, the critical
level of dissimilarity is still poorly understood (Edmands
2002). In sessile organisms such as plants, outbreeding de-
pression can occur over relatively small spatial scales, which
is probably related to restricted gene flow (Waser and Price
1994; Waser et al. 2000). Thus, when examining negative
effects of inbreeding versus outbreeding, one has to consider
K. PEER AND M. TABORSKY
from Bremgartenwald, one female per cross (inbreeding, within-
population, between-populations) was used. The males for the in-
breeding and within-population crosses originated from the same
broods, while those for the between-population crosses came from
Spilwald broods. From each Spilwald parental brood, one female
and one male were used for within-population crosses.
Design of experimental crosses. From each parental brood
the potential reduction in fitness above a certain level of
divergence, especially in highly structured populations with
The close link between inbreeding depression and the evo-
lution of mating patterns and sex ratios can be illustrated
particularly well in the haplodiploid ambrosia beetles (Xy-
leborini, Scolytinae). The Xyleborini are the only other hap-
lodiploid lineage in holometabolic insects apart from the Hy-
menoptera (Mable and Otto 1998), and probably all of the
approximately 1200 xyleborine species show high levels of
inbreeding (Kirkendall 1993). Males are unable to fly, sex
ratios are highly female biased, and mating was believed to
take place exclusively among siblings within the natal gallery
(Kirkendall 1993). However, recently it has been shown that
males may disperse short distances, and females may have
the opportunity to mate with unrelated males instead of their
brothers (Peer and Taborsky 2004). This may substantially
alter population structure, fitness consequences of inbreeding,
and optimal sex allocation (Greeff and Taylor 1997). The
extent of inbreeding depression is likely to influence female
mating decisions. If continuous sibling mating leads to a
reduction of the genetic load through purging, then gene flow
across lineages through males may counteract this process
(but see Whitlock et al. 2000).
In Xyleborini, populations can be assumed to be highly
structured due to their life history with limited male dispersal
ability and local mating. Therefore, both inbreeding and out-
breeding depression may be observed on a relatively small
spatial scale. The aim of this study was to determine the
fitness consequences of mating with brothers versus unrelated
males with different genetic distances. We experimentally
mated females to brothers, males from the same or different
populations, and measured fitness traits of the resulting dip-
loid female offspring. We expected to find moderate inbreed-
ing depression in offspring produced by sib-mated females
compared to outcrosses, and outbreeding depression that is
visible in between-population crosses.
MATERIALS AND METHODS
Xylosandrus germanus is one of nine xyleborine species in
Europe. It is an Asian species with a body length of about
2.2 mm, which has been introduced throughout the Holarctic
and was described in Europe for the first time in the 1950s
(Gauss 1960). Dispersal flights last from May to June, when
females colonize freshly fallen trees and start excavating a
brood chamber or gallery. They carry the spores of fungi
(ambrosia, Ambrosiella hartigii) in special structures (my-
cangia) and cultivate these fungi on the surface of their gal-
lery walls. After the ambrosia has started to grow, the females
oviposit and broods develop until the end of August. Al-
though mating generally takes place among siblings, some
males leave their natal gallery and search for other galleries
on the same log. The mated females overwinter in their gal-
lery and disperse the next spring (Gauss 1960; Heidenreich
1960, 1964). Female dispersal distance probably ranges up
to 100 m (estimate based on our own trapping data; Heiden-
We collected mated females during dispersal flight in June
2003 with ethanol-baited traps in two different forests near
Berne (Bremgartenwald and Spilwald, about 6 km apart).
These females were transferred to the laboratory and allowed
to excavate brood chambers and cultivate fungus in artificial
medium in test tubes (see Peer and Taborsky 2004). After
four weeks, when about half of the offspring had reached the
pupal or adult stage, we dissected the galleries, and pupae
and adults were removed. Since mating takes place within
the natal gallery soon after eclosion, this step was necessary
to obtain unmated females, which could then be paired ac-
cording to experimental protocol.
From each brood, we placed two female pupae per replicate
into separate small plastic boxes with moist filter paper, and
added a male or a male pupa. For females from Bremgar-
tenwald, the male was either from the same brood (inbreed-
ing), from a different brood from Bremgartenwald (within-
population cross), or from a brood from Spilwald (between-
population cross; Fig. 1). The males for the within-population
crosses originated from the same broods as those for the
inbreeding treatment to control for genetic differences be-
tween broods. This reciprocal design was not possible for
between-population crosses. Because of heavily biased sex
ratios, only two males were available for experimental cross-
ings for most broods, and therefore crosses between males
from Bremgartenwald and females from Spilwald could not
be conducted. However, we also performed within-popula-
tion crosses for females from Spilwald, and compared their
offspring fitness to that of within-population crosses from
Bremgartenwald females to avoid the possibility that poten-
tial population differences would be interpreted erroneously
as effects of the crossing treatment.
Since it is not possible to observe the development of
offspring in intact galleries, eggs were extracted for moni-
toring early life-history traits. For this reason, we used two
INBREEDING AND OUTBREEDING IN XYLOSANDRUS GERMANUS
females per brood in each replicate. One (randomly assigned)
was used to obtain data on early life-history traits of offspring
(destructive method) and the other was used to measure total
brood sizes produced in the presence of maternal care. In
total, we used 30 broods per treatment and population, which
resulted in 240 experimental matings.
Following eclosion of experimental males and females,
these were kept together in the experimental boxes for at
least two days to ensure successful copulation. Females were
then allowed to tunnel in artificial medium in test tubes.
Because they had been removed from their brood chambers
prematurely, the mycangia of these females were devoid of
fungal spores from their natal gallery. Therefore, we artifi-
cially inoculated the medium with ambrosia fungus from al-
ready established laboratory galleries. In consequence,
broods with fungal contamination or unsuccessful experi-
mental inoculation with ambrosia fungus had to be excluded
from the experiment, so that final sample sizes differed be-
Early Life-History Traits of Offspring
On day 11 after gallery initiation, we dissected half of the
experimental galleries and extracted the eggs. They were
placed on a surface of artificial medium contained in small
transparent plastic boxes. The eggs carried fungal mycelium
on their surface, so that ambrosia started to grow on which
the larvae could feed after hatching. Because of the absence
of brood care, there often was excessive ambrosia growth or
fungal contamination of the medium and accumulation of
debris. Therefore, offspring development and survival was
obviously reduced compared to undisturbed galleries. All
treatments suffered equally from this effect and thus could
be compared. We recorded the following data: number of
eggs extracted, number of eggs and larvae present on day 7,
number of larvae present on day 12, and number of pupae
present on day 21 after extraction. The timing of data re-
cording was determined during pilot observations of the du-
rations of egg, larval, and pupal stages. Between treatments,
we compared the proportions of (1) eggs hatched on day 7,
(2) larvae surviving from days 7 to 12, and (3) larvae un-
dergoing pupation from days 12 to 21. For a more integral
measure of fitness we multiplied hatching rate, larval sur-
vival, and pupation rate for the broods where all three mea-
sures were available to obtain total fitness.
Total Number of Offspring and Offspring Fecundity
The other half of the experimental galleries was left un-
touched for eight weeks, when brood development was com-
plete. We then dissected the galleries and counted the number
of male and female offspring. To estimate the environmental
quality within the brood chambers, ambrosia growth (non-
existent, bad, good) and occurrence of contamination with
other fungus than ambrosia (none, present) were recorded.
To measure F1fecundity, three female offspring from each
brood were randomly selected and again transferred to arti-
ficial medium in tubes. After 14 days, we dissected the re-
sulting galleries and counted the eggs. For analyses, the num-
ber of eggs was averaged over the three offspring females
from each brood. An additional fitness trait that may differ
between inbred and outbred offspring is the capability to
transfer ambrosia successfully and prevent contamination of
the brood with harmful fungi. Thus, we also scored ambrosia
growth and contamination for these offspring galleries, and
calculated the probability of successful ambrosia propagation
and of contamination of offspring galleries.
Before analyzing the treatment effect, all variables were
tested for differences between populations, only using data
from within-population crosses. Whenever there was a ten-
dency for a population difference in a trait (P ? 0.1), we
excluded between-population crosses from the final analyses
to prevent confounding effects of intrinsic differences be-
tween populations (the traits concerned were total brood sizes
and offspring fecundity). Only broods from Bremgartenwald
females were used for between-treatment comparisons.
We used a generalized linear models approach to analyze
fecundity (number of eggs) and brood sizes (number of fe-
male pupae and adult offspring) with Poisson regressions.
Hatching rates, larval survival, and pupation rates were an-
alyzed with a weighted logistic regression, which uses broods
as individual datapoints and assumes a binomial error dis-
tribution of the number of successes (e.g., hatched eggs) per
brood. Both Poisson and logistic regressions employ logit
link functions to linearize the data. The significance of effects
was tested by stepwise deletion from the final model. When-
ever the final model was overdispersed, we used F statistics
instead of ?2statistics to test for significance (Crawley 1993).
For the analysis of total fitness, the percentage data were
arcsine square-root transformed to obtain a normal distri-
bution of the residuals, and analyzed using an ANOVA. All
data were analyzed with R 1.8.0 (The R Developing Core
Team, 1993, Vienna).
Matriline (brood from which the parental female originat-
ed) was always included in the analyses to control for po-
tential genetic differences between broods. This accounts for
the differences between residual degrees of freedom and sam-
ple size. Because we found that ambrosia growth and fungal
contamination have strong effects on brood development and
individual offspring fitness, we also included these factors
and removed them from the final model if they were not
Differentiation among Populations and Broods
Parental females, whose clutches were used for the early
life-history trait measurements, laid on average 15.1 eggs (n
? 102). We found a highly significant effect of matriline
(Table 1), but not of population on the number of eggs laid
(F1,37? 2.8, P ? 0.22; n ? 102 clutches). None of the early
life-history traits of within-population crosses differed be-
tween the two populations used in the experiment (hatching
rates: F1,28? 0.005, n ? 30, P ? 0.94; larval survival: F1,20
? 0.17, P ? 0.68, n ? 22; pupation rate: F1,14? 1.604, P
? 0.21, n ? 16 clutches). Average brood size across all
treatments and populations was 24.3 (n ? 73 broods). When
comparing only within-population crosses, total brood sizes
K. PEER AND M. TABORSKY
surements. All significant sources of variation are shown in bold. Ambrosia growth and fungal contamination of the natal gallery were
retained in the final model as covariables if their effect was significant.
Influence of female lineage and treatment (inbreeding, within-population cross, between-population cross) on fitness mea-
Hatching rate (%)2
Larval survival (%)1
Pupation rate (%)1
1Using F-test because of overdispersion.
3Between-population crosses were excluded due to differences between populations.
from Spilwald tended to be larger than those from Brem-
gartenwald (population: F1,28? 2.90, P ? 0.1; ambrosia
growth: F1,28? 12.45, P ? 0.001; contamination: F18,25?
103.84, P ? 0.001, n ? 32 broods), and fecundity of F1
offspring differed significantly between populations (popu-
lation: F1,19? 11.73, P ? 0.001; contamination: F1,19?
13.89, P ? 0.001, n ? 22 broods). The probability of con-
tamination of F1galleries (F1,20? 1.08, P ? 0.3.) and suc-
cessful ambrosia transfer (F1,20? 0.83, P ? 0.36) did not
differ between populations. Thus, for all traits except total
brood sizes and offspring fecundity the between-population
crosses were included in the subsequent analyses of treatment
Effect of Treatment on Offspring Fitness
Egg hatching rates were significantly influenced by in-
breeding status (i.e., treatment) and matriline (see Table 1),
such that eggs resulting from inbred matings had higher
hatching success than eggs from either of the outbred matings
(medians were 100%, 89.0%, and 88.9% for inbreeding, with-
in-population crosses, and between population crosses, re-
spectively; Fig. 2). Polynomial contrasts revealed that hatch-
ing success of within-population crosses was lower than of
within-brood crosses (P ? 0.01), but was not significantly
further reduced in between-population crosses (P ? 0.44).
Both larval survival (medians 83.3%, 100%, and 87.5%) and
pupation rate (medians 19.5%, 38.1%, and 50.0%) did not
differ significantly between treatments, but matriline had a
large effect on pupation rate (see Table 1). Also, when com-
bining all three measures into total fitness, crossing had no
significant effect (ANOVA: F2,8? 1.14, P ? 0.37, n ? 40).
Using helmert contrasts, no significant effect was found when
comparing sibling versus (WP ? BP) or WP versus BP (P
? 0.27 and 0.57, respectively).
Brood sizes were largely influenced by matriline (see Table
1), ambrosia growth, and fungal contamination, but inbreed-
ing status (i.e., treatment) had no effect (mean brood sizes
were 22.35 and 21.0, respectively). Matriline and ambrosia
growth in the natal gallery significantly affected fecundity of
F1offspring. However, inbreeding status had no influence
(mean numbers of eggs laid were 18.2 and 13.1, respectively;
see Table 1). Similarly, neither the proportion of F1offspring
galleries with successful ambrosia growth nor the proportion
of galleries with contamination was affected by inbreeding
status, but matriline again had an effect on both (see Table 1).
Our results show that X. germanus does not suffer from
inbreeding depression in any of the examined fitness traits.
On the contrary, inbreeding resulted in higher offspring
hatching rates than either of the outcrossing treatments. In
haplodiploid insects and mites, inbreeding depression is low-
er than in diploids due to purging of recessive deleterious
alleles in male hemizygotes (Antolin 1999; Henter 2003).
This finding, which had been based mainly on data from
haplodiploid Hymenoptera, is taken one step further by the
complete absence of inbreeding depression in our study sys-
tem, which represents a phylogenetically independent group.
Ecologically imposed inbreeding may promote the evolution
of haplodiploidy, because inbreeding depression will be re-
duced through purging of deleterious alleles in haploid males,
and effective mutation rates are lower due to lower ploidy
levels (Werren 1993). Inbreeding has been argued to select
for longer haploid phases in a wide range of organisms and
reproductive systems (Mable and Otto 1998). Nevertheless,
in haplodiploid species, fitness of inbred individuals may also
be reduced by 20–40% compared to outbred individuals (An-
tolin 1999; Henter 2003). Until now it could not be dem-
onstrated that a history of inbreeding further enhances purg-
ing in haplodiploids due to the expression of deleterious al-
leles in homozygous diploid females. However, X. germanus
is probably the species with the highest degree of continuous
sibling mating in which inbreeding effects ever have been
examined. Theoretical work has shown that purging may not
occur until a critical threshold selfing/inbreeding rate is
reached (Lande et al. 1994).
The lack of inbreeding depression also has implications
for the mating system, which has been explored extensively
in plants (e.g., Lande and Schemske 1985; Waller 1993; Bar-
rett and Harder 1996), but not in animals. Local mate com-
INBREEDING AND OUTBREEDING IN XYLOSANDRUS GERMANUS
Datapoints always represent individual broods. (a–d) Early life-history traits, (e) resulting brood sizes, and (f) F1fecundity. Datapoints
aligned horizontally on the ordinate represent identical values. Except for (b), residual values are corrected for all other sources of
variation besides treatment. In (b) the raw data without correcting for matriline and the corresponding medians are shown to visualize
the effect of matriline. The lack of a significant difference in hatching rates between WP and BP may be due to small sizes of broods
with low hatching rates in BP. As a result these broods have a lower weight in a weighted logistic regression.
Effect of treatment (I, inbreeding; WP, within-population cross; BP, between-population cross) on different fitness measures.
K. PEER AND M. TABORSKY
petition leads to female-biased sex ratios, provided that the
benefits of sibling mating outweigh any costs of inbreeding
depression (Denver and Taylor 1995). The strength of the
sex-ratio bias is predicted to depend on the extent of in-
breeding depression, such that lower inbreeding depression
should lead to lower production of males (Greeff and Taylor
1997; De Jong et al. 1999). However, there are hardly any
data on the effects of inbreeding on individual fitness in hap-
lodiploid species with local mate competition (Antolin 1999;
Henter 2003). Xylosandrus germanus females produce only
5–10% males (Peer and Taborsky 2004), which is one of the
most extreme sex-ratio biases found in animals and may be
a result of low inbreeding costs.
Xylosandrus germanus suffers from outbreeding depression
in hatching rates even when crossed within the same popu-
lation. Genetic mechanisms of outbreeding depression (as
opposed to local adaptation) over moderate distances are
thought to occur mainly in highly inbred species (Waser
1993b). Restricted recombination can lead to intrinsic co-
adaptation of genes within a gene pool (Templeton 1986).
Disruption of such gene complexes results in reduced fitness,
which has been shown theoretically and experimentally with
parthenogenetic strains of Drosophila (Templeton et al. 1976,
1986). A number of plant species with restricted gene flow
have the highest fitness at intermediate levels of outbreeding
(e.g., Waser et al. 2000). This effect may occur within very
small spatial scales. In animals, there has been little evidence
for outbreeding depression within a single population so far.
However, X. germanus is subjected to outbreeding de-
pression at the smallest possible scale or genetic distance:
sibling mating, which in a haplodiploid population with con-
tinuous inbreeding is comparable to selfing, resulted in the
highest egg hatching rates. Similar to our results, in a pop-
ulation of a highly inbred fern selfing resulted in the highest
fitness (Schneller 1996). Recently, a negative correlation of
fitness with outbreeding (measured as d2, the genetic distance
between parental gametes) has been reported for an isolated,
highly inbred population of lizards (LeBas 2002). In X. ger-
manus, there is little dispersal by males, mating takes place
almost exclusively within the natal galleries, and a population
may consist of a large number of highly differentiated inbred
lines, which was suggested by the large effect of maternal
identity on offspring traits (Table 1). Intrinsic coadaptation
may lead to outbreeding depression whenever two different
Outbreeding depression is often only apparent in the F2
generation and later (Burton 1990; Fenster and Galloway
2000) or in fecundity reduction of the F1generation (Tem-
pleton 1986), when coadapted gene complexes are broken up
through recombination at meiosis. In contrast, our study re-
vealed outbreeding depression already at very early stages of
F1(hatching rates), but no effect of outbreeding on fertility
of F1females. If maternal effects are important and dependent
on offspring genotype, coadaptation may occur not only be-
tween genes within one individual, but also between mother-
offspring genes (Wolf 2000). Maternal genes may often ex-
plain as much as half the variance in offspring characters
early in life. Endosymbionts have been shown to play an
important role in the fertilization process of another xyle-
borine species (X. ferrugineus; Peleg and Norris 1972). It is
possible that cytoplasmatic components such as these en-
dosymbionts have additional effects within the eggs, resulting
in maternal effects. In the zygote, selection may act against
genes generating incompatibility with endosymbionts. In
such a scenario, reduced hatching rates of eggs producedfrom
outcrosses could be explained by nuclear genes, which are
not adapted to the zygotic environment. Finally, the presence
of Wolbachia in this species (R. K. K. Koivisto and K. Peer,
unpubl. data) may result in cytoplasmatic incompatibility
caused by different Wolbachia strains present in the popu-
lation, which could result in reduced egg-hatching rates.
It may seem surprising that outbreeding depression was
not detectable in total brood sizes and total fitness, since
reduced hatching rates should result in a lower number of
offspring reaching maturity. However, differences in mater-
nal care and environmental factors such as ambrosia growth
and fungal contamination may be more important at later
stages in the life cycle, so that they override the outbreeding
effect on hatching rates.
The findings of this study are highly relevant to understand
speciation in the Xyleborini. In this group of bark beetles,
haplodiploidy has apparently evolved in concert with close
inbreeding and was followed by rapid diversification, pos-
sibly enhanced by the habit of fungus (ambrosia) cultivation
(Normark et al. 1999). The tribe now consists of over 1200
species, compared to 40 species of its diploid sister group,
the Dryocoetini (Jordal et al. 2000). As a result of continued
inbreeding and reduced gene flow between lineages in the
Xyleborini, the lack of inbreeding depression and the pres-
ence of outbreeding depression may have enhanced behavior-
induced reproductive isolation. Above a certain threshold,
outbreeding depression can be seen as equivalent to repro-
ductive isolation (see also Coyne and Orr 1997; Waser et al.
2000). Further studies of genetic population architecture in
this group may shed more light on the evolution of mating
systems and speciation in general.
We thank L. Kirkendall for sharing his knowledge about
bark beetles with us and for remarks on a previous version
of the manuscript; R. Bergmu ¨ller and L. Keller fordiscussion;
C. Goodnight, N. Waser and an anonymous referee for com-
ments; and D. Heg for statistical advice.
Antolin, M. F. 1999. A genetic perspective on mating systems and
sex ratios of parasitoid wasps. Res. Popul. Ecol. 41:29–37.
Barrett, S. C. H., and L. D. Harder. 1996. Ecology and evolution
of plant mating. Trends Ecol. Evol. 11:73–79.
Bruckner, D. 1978. Why are there inbreeding effects in haplodiploid
systems? Evolution 32:456–458.
Burton, R. S. 1990. Hybrid breakdown in developmental time in
the copepod Tigriopus californicus. Evolution 44:1814–1822.
Charlesworth, B., and D. Charlesworth. 1999. The genetic basis of
inbreeding depression. Genet. Res. Camb. 74:329–340.
Charlesworth, D., and B. Charlesworth. 1987. Inbreeding depres-
sion and its evolutionary consequences. Annu. Rev. Ecol. Syst.
Coyne, J. A., and H. A. Orr. 1997. ‘‘Patterns of speciation in Dro-
sophila’’ revisited. Evolution 51:295–303.
Crawley, M. 1993. GLIM for ecologists. Blackwell, Oxford, U.K.
Crnokrak, P., and S. Barrett. 2002. Perspective: Purging the genetic
323 Download full-text
INBREEDING AND OUTBREEDING IN XYLOSANDRUS GERMANUS
load: A review of the experimental evidence. Evolution 56:
Crozier, R. H. 1985. Adaptive consequences of male haploidy. Pp.
201–222 in W. Helle and M. W. Sabelis, eds. Spider mites: Their
biology, natural enemies, and control. Elsevier, Amsterdam.
De Jong, T. J., P. G. L. Klinkhamer, and M. C. J. Rademaker. 1999.
How geitonogamous selfing affects sex allocation in hermaph-
rodite plants. J. Evol. Biol. 12:166–176.
Demeester, L. 1993. Inbreeding and outbreeding depression in
Daphnia. Oecologia 96:80–84.
Denver, K., and P. D. Taylor. 1995. An inclusive fitness model for
the sex-ratio in a partially sibmating population with inbreeding
cost. Evol. Ecol. 9:318–327.
Doums, C., F. Viard, A. F. Pernot, B. Delay, and P. Jarne. 1996.
Inbreeding depression, neutral polymorphism, and copulatory
behavior in freshwater snails: A self-fertilization syndrome.
Edmands, S. 2002. Does parental divergence predict reproductive
compatibility? Trends Ecol. Evol. 17:520–527.
Fenster, C. B., and L. F. Galloway. 2000. Population differentiation
in an annual legume: genetic architecture. Evolution 54:
Gauss, R. 1960. Ist Xylosandrus germanus Blandf. ein Prima ¨rscha ¨-
dling. Anz. Scha ¨dlingskd. Pflanzenschutz Umweltschutz 23:
Greeff, J. M., and P. D. Taylor. 1997. Effects of inbreeding de-
pression on relatedness and optimal sex ratios. Evol. Ecol. 11:
Haag, C. R., J. W. Hottinger, M. Riek, and D. Ebert. 2002. Strong
inbreeding depression in a Daphnia metapopulation. Evolution
Heidenreich, E. 1960. Prima ¨rbefall durch Xylosandrus germanus an
Jungeichen. Anz. fur Schadlingskd. Pflanzenschutz Umwelts-
———. 1964. O¨kologische Bedingungen fu ¨r Prima ¨rbefall durch
Xylosandrus germanus. J. Appl. Entomol. 54:131–140.
Henter, H. 2003. Inbreeding depression and haplodiploidy: Exper-
imental measures in a parasitoid and comparisons across diploid
and haplodiploid insect taxa. Evolution 57:1793–1803.
Husband, B. C., and D. W. Schemske. 1996. Evolution of the mag-
nitude and timing of inbreeding depression in plants. Evolution
Jordal, B. H., B. B. Normark, and B. D. Farrell. 2000. Evolutionary
radiation of an inbreeding haplodiploid beetle lineage (Curcu-
lionidae, Scolytinae). Biol. J. Linn. Soc. 71:483–499.
Keller, L. F., and D. M. Waller. 2002. Inbreeding effects in wild
populations. Trends Ecol. Evol. 17:230–241.
Kirkendall, L. R. 1993. Ecology and evolution of biased sex ratios
in bark and ambrosia beetles. Pp. 235–345 in D. L. Wrensch
and M. A. Ebbert, eds. Evolution and diversity of sex ratio in
insects and mites. Chapman and Hall, New York.
Lande, R., and D. W. Schemske. 1985. The evolution of self-fer-
tilization and inbreeding depression in plants. I. Genetic models.
Lande, R., D. W. Schemske, and S. T. Schultz. 1994. High inbreed-
ing depression, selective interference among loci, and the thresh-
old selfing rate for purging recessive lethal mutations. Evolution
Latta, R., and K. Ritland. 1994. The relationship between inbreeding
depression and prior inbreeding among populations of four Mi-
mulus taxa. Evolution 48:806–817.
LeBas, N. R. 2002. Mate choice, genetic incompatibility, and out-
breeding in the ornate dragon lizard, Ctenophorus ornatus. Evo-
Mable, B. K., and S. P. Otto. 1998. The evolution of life cycles
with haploid and diploid phases. BioEssays 20:453–462.
McCall, C., D. M. Waller, and T. Mitchell-Olds. 1994. Effects of
serial inbreeding on fitness components in Impatiens capensis.
Mitton, J. B. 1993. Theory and data pertinent to the relationship
between heterozygosity and fitness. Pp. 17–41 in N. W. Thorn-
hill, ed. The natural history of inbreeding and outbreeding. Univ.
of Chicago Press, Chicago.
Normark, B. B., B. H. Jordal, and B. D. Farrell. 1999. Origin of a
haplodiploid beetle lineage. Proc. R. Soc. Lond. B 266:
Peer, K., and M. Taborsky. 2004. Female ambrosia beetles adjust
their offspring sex ratio according to outbreeding opportunities
for their sons. J. Evol. Biol. 17:257–264.
Peleg, B., and D. M. Norris. 1972. Bacterial symbiote activation of
insect parthenogenetic reproduction. Nat. New Biol. 236:
Price, M. V., and N. M. Waser. 1979. Pollen dispersal and optimal
outcrossing in Delphinium nelsonii. Nature 277:294–297.
Ritland, K. 1990. Inferences about inbreeding depression based on
changes of the inbreeding coefficient. Evolution 44:1230–1241.
Roff, D. A. 2002. Inbreeding depression: Tests of the overdomi-
nance and partial dominance hypotheses. Evolution 56:768–775.
Schneller, J. J. 1996. Outbreeding depression in the fern Asplenium
ruta-muraria L.: Evidence from enzyme electrophoresis, meiotic
irregularities and reduced spore viability. Biol. J. Linn. Soc. 59:
Templeton, A. R. 1986. Coadaptation and outbreeding depression.
Pp. 105–116 in M. E. Soule ´, ed. Conservation biology: The
science of scarcity and diversity. Sinauer, Sunderland, MA.
Templeton, A. R., C. F. Sing, and B. Brokaw. 1976. The unit of
selection in Drosophila mercatorum. I. The interaction of selec-
tion and meiosis in parthenogenetic strains. Genetics 82:
Templeton, A. R., H. Hemmer, G. Mace, U. S. Seal, W. M. Shields,
and D. S. Woodruff. 1986. Local adaptation, coadaption, and
population-boundaries. Zoo Biol. 5:115–125.
Waller, D. M. 1993. The statics and dynamics of mating system
evolution. Pp. 97–117 in N. W. Thornhill, ed. The natural history
of inbreeding and outbreeding. Univ. of Chicago Press, Chicago.
Waser, N. M. 1993a. Sex, mating systems, inbreeding, and out-
breeding. Pp. 1–13 in N. W. Thornhill, ed. The natural history
of inbreeding and outbreeding. Univ. of Chicago Press, Chicago.
———. 1993b. Population structure, optimal outbreeding, and as-
sortative mating in angiosperms. Pp. 173–199 in N. W. Thorn-
hill, ed. The natural history of inbreeding and outbreeding. Univ.
of Chicago Press, Chicago.
Waser, N. M., and M. V. Price. 1994. Crossing distance effects in
Delphinium nelsonii: Outbreeding and inbreeding depression in
progeny fitness. Evolution 48:842–852.
Waser, N. M., and C. F. Williams. 2001. Inbreeding and outbreed-
ing. Pp. 84–96 in C. W. Fox, D. A. Roff, and D. J. Fairbairn,
eds. Evolutionary ecology: Concepts and case studies. Oxford
University Press, Oxford, U.K.
Waser, N. M., M. V. Price, and R. G. Shaw. 2000. Outbreeding
depression varies among cohorts of Ipomopsis aggregata planted
in nature. Evolution 54:485–491.
Wedekind, C., D. Strahm, and L. Scha ¨rer. 1998. Evidence for stra-
tegic egg production in a hermaphroditic cestode. Parasitology
Weeks, S. C., V. Marcus, and B. R. Crosser. 1999. Inbreeding
depression in a self-compatible, androdioecious crustacean, Eu-
limnadia texana. Evolution 53:472–483.
Werren, J. H. 1993. The evolution of inbreeding in haplodiploid
organisms. Pp. 42–49 in N. W. Thornhill, ed. The natural history
of inbreeding and outbreeding. Univ. of Chicago Press, Chicago.
Whitlock, M. C., P. K. Ingvarsson, and T. Hatfield. 2000. Local
drift load and the heterosis of interconnected populations. He-
Wolf, J. B. 2000. Gene interactions from maternal effects. Evolution
Corresponding Editor: C. Goodnight