Hybrid vigor between native and introduced salamanders raises new challenges for conservation
Hybridization between differentiated lineages can have many different consequences depending on fitness variation among hybrid offspring. When introduced organisms hybridize with natives, the ensuing evolutionary dynamics may substantially complicate conservation decisions. Understanding the fitness consequences of hybridization is an important first step in predicting its evolutionary outcome and conservation impact. Here, we measured natural selection caused by differential viability of hybrid larvae in wild populations where native California Tiger Salamanders (Ambystoma californiense) and introduced Barred Tiger Salamanders (Ambystoma tigrinum mavortium) have been hybridizing for 50-60 years. We found strong evidence of hybrid vigor; mixed-ancestry genotypes had higher survival rates than genotypes containing mostly native or mostly introduced alleles. Hybrid vigor may be caused by heterozygote advantage (overdominance) or recombinant hybrid vigor (due to epistasis or complementation). These genetic mechanisms are not mutually exclusive, and we find statistical support for both overdominant and recombinant contributions to hybrid vigor in larval tiger salamanders. Because recombinant homozygous genotypes can breed true, a single highly fit genotype with a mosaic of native and introduced alleles may eventually replace the historically pure California Tiger Salamander (listed as Threatened under the U.S. Endangered Species Act). The management implications of this outcome are complex: Genetically pure populations may not persist into the future, but average fitness and population viability of admixed California Tiger Salamanders may be enhanced. The ecological consequences for other native species are unknown.
Hybrid vigor between native and introduced
salamanders raises new challenges for conservation
Benjamin M. Fitzpatrick*
and H. Bradley Shaffer
*Ecology and Evolutionary Biology, University of Tennessee, Knoxville, TN 37996; and
Evolution and Ecology, University of California, Davis, CA 95616
Edited by John C. Avise, University of California, Irvine, CA, and approved August 10, 2007 (received for review May 22, 2007)
Hybridization between differentiated lineages can have many
different consequences depending on ﬁtness variation among
hybrid offspring. When introduced organisms hybridize with na-
tives, the ensuing evolutionary dynamics may substantially com-
plicate conservation decisions. Understanding the ﬁtness conse-
quences of hybridization is an important ﬁrst step in predicting its
evolutionary outcome and conservation impact. Here, we mea-
sured natural selection caused by differential viability of hybrid
larvae in wild populations where native California Tiger
Salamanders (Ambystoma californiense) and introduced Barred
Tiger Salamanders (Ambystoma tigrinum mavortium) have been
hybridizing for 50–60 years. We found strong evidence of hybrid
vigor; mixed-ancestry genotypes had higher survival rates than
genotypes containing mostly native or mostly introduced alleles.
Hybrid vigor may be caused by heterozygote advantage (over-
dominance) or recombinant hybrid vigor (due to epistasis or
complementation). These genetic mechanisms are not mutually
exclusive, and we ﬁnd statistical support for both overdominant
and recombinant contributions to hybrid vigor in larval tiger
salamanders. Because recombinant homozygous genotypes can
breed true, a single highly ﬁt genotype with a mosaic of native and
introduced alleles may eventually replace the historically pure
California Tiger Salamander (listed as Threatened under the U.S.
Endangered Species Act). The management implications of this
outcome are complex: Genetically pure populations may not per-
sist into the future, but average ﬁtness and population viability of
admixed California Tiger Salamanders may be enhanced. The
ecological consequences for other native species are unknown.
Ambystoma 兩 ﬁtness 兩 hybridization 兩 invasive species 兩 genetics
ybridization (interbreeding bet ween differentiated lin-
eages) occurs in almost all sexually reproducing groups of
organ isms (1–3). Consequences of hybridization include fusion
of prev iously distinct lineages, extinction or local extirpation of
one or both lineages, evolution of reproductive isolation via
reinforcement, and production of novel, highly fit hybrid phe-
not ypes. These outcomes depend on the distribution and inher-
it ance of fitness among hybrids and of ten have important
implications for both evolutionary and conservation biology.
In evolutionary biology, hybrid fitness and its genetic basis are
import ant for understanding the evolution of reproductive iso-
lation and the c onsequences of horizont al gene flow on the
diversit y and complexity of life (4–7). In c onservation biolog y,
hybrid fitness is a key to the mechanistic basis of inbreeding and
outbreeding depression, the potential loss of biodiversity due to
genetic swamping, and the evolution of invasiveness (8–11).
These considerations become particularly import ant when in-
troduced organisms successfully hybridize with natives.
Unlike cases of natural hybridization, some outcomes of
human-mediated hybridization are generally considered less
desirable than others. Low fitness of hybrids can reduce mean
population fitness, rendering populations more vulnerable to
extinction (8, 12). Novel hybrid genotypes may contribute to de
novo evolution of invasiveness (9), and admixed populations may
be considered less valuable than an authentic native gene pool
(10). On the other hand, admixture may rescue declining native
populations by alleviating inbreeding depression (13, 14) or
facilit ating adaptive evolution in modified or degraded habit ats
The long-term consequences of hybridization are strongly
influenced by the genetic basis of hybrid fitness. In the case of
hybrid vigor, genetic models fall into two classes: heteroz ygote
advant age and rec ombinant hybrid vigor (18–20). Heterozygote
advant age (overdominance) refers to beneficial interactions
bet ween heterospecific alleles of a single locus. Recombinant
hybrid vigor depends on multilocus genotypes and may be caused
by epistasis (beneficial interactions between heterospecific al-
leles f rom different loci) or by complementary effects of inde-
pendent advantageous alleles from each parental population (19,
20). For example, if dif ferent loci have fixed deleterious alleles
in each parental population, hybrids bearing the superior allele
at each locus will have higher fitness than either parent.
Unlike heterozygote advantage, where the most fit genot ype
cannot breed true, recombinant hybrid vigor favors the fixation
of a single recombinant genotype bringing together beneficial
alleles f rom each parent. This new, true breeding genotype may
be considered a modified version of one of the parent al lineages
or a new hybrid species. In conservation terms, fixation of a
rec ombinant genot ype may represent a minor evolutionary
modification or total genetic ‘‘extinction’’ of a native form (this
may depend more on one’s perspective than on the genetic
det ails). Perhaps more important, the new form may have
undesirable ec ological effects on native commun ities (9).
Recent, human-mediated hybridization between native Cali-
forn ia Tiger Salamanders (Ambystoma califor niense) and intro-
duced Barred Tiger Salamanders, (Ambystoma tigrinum mavor-
tium) constitutes an important case study of the evolutionary and
c onservation consequences of hybridization (21–23). Before the
introduction 50–60 years ago (by bait dealers interested in selling
salamander larvae to bass fishermen), the two lineages had been
geographically separated for ⬇5 my (24, 25). Despite such a long
period of divergence, they hybridize readily and have formed a
hybrid swarm in the Salinas Valley of California (21–23). Mul-
tiple introduction points, combined with subsequent dispersal
and movements of salamanders across the landscape (23) have
led to broad-scale admixture across 15–20% of the natural range
of the native California Tiger Salamander. The recent listing of
A. californiense as Threatened under the US Endangered Species
Act (26) recognized hybridization as a major conservation
In this study, we estimated viability of hybrid tiger salamander
larvae over the first few weeks after hatching. This period of high
mort ality represents an enormous opportunity for natural se-
Author contributions: B.M.F. and H.B.S. designed research; B.M.F. and H.B.S. performed
research; B.M.F. analyzed data; and B.M.F. and H.B.S. wrote the paper.
The authors declare no conﬂict of interest.
This article is a PNAS Direct Submission.
To whom correspondence should be addressed. E-mail: benﬁtz@utk.edu.
This article contains supporting information online at www.pnas.org/cgi/content/full/
© 2007 by The National Academy of Sciences of the USA
October 2, 2007
lection to operate on the diverse genot ypes of a hybrid popula-
tion (27). Observations were made in five wild breeding popu-
lations including the three major types of breeding habitats
found in the Salinas Valley: natural vernal pools, seasonal cattle
ponds, and perennial ponds (22). We categorized individual
hybrids using multilocus genotypes based on n ine physically
unlinked molecular markers. Each individual was characterized
as having native or introduced mtDNA and as being homozygous
native, heterozygous, or homoz ygous introduced at each nuclear
marker (22). Thus, our approach has the advantages of consid-
ering the fitness of genotypes defined more finely than simple
pedigree categories (F
, backcross, etc.) and of estimating fitness
in the wild across the range of larval habit ats found within the
hybrid swarm (28, 29). We focused on how early larval survival
varies among hybrid genotypes in the wild. Specifically, is the
pattern of survival consistent with hybrid dysfunction or hybrid
vigor, and does fitness variation depend more on additive,
dominant (heterozygous), or recombinant genetic effects?
Generalized linear models relating survival probability to mo-
lecular marker-based hybrid indices clearly support hybrid vigor
(Table 1 and Fig. 1). Survival was higher for indiv iduals with high
, and intermediate ancestry,
Negative coef ficients for
indicate that surviving larvae
tended to be of less pure and more mixed ancestry. A more
c omplex model (with
ter ms) did not fit the data
better than the reduced model shown in Table 1 [supporting
infor mation (SI) Table 2]. The significant ef fect of mixed
ancestry over and above the effect of heterozygosity suggests that
rec ombinant genotypes have enhanced survival that is not
predicted by heterozygote advantage alone.
Det ails of the multivariate outc ome differed among ponds
(Table 1), but this is attributable to dif ferences in in itial geno-
t ype frequencies. Ponds that started highly introduced became
more native and ponds that st arted highly native became more
introduced (Fig. 1). This is consistent with selection favoring
mixed ancestry over either pure native or pure introduced
genot ypes. Fig. 1 shows triangular regions of genotype space
defined by the interdependence of ancestry and heterozygosity.
Pure native or introduced genotypes, by definition, have no
heteroz ygous loci. Individuals with all heteroz ygous loci must
have evenly mixed ancestry, but mixed ancestry can also be
achieved when some loci are homozygous for native alleles and
others are homoz ygous for introduced alleles. Fig. 1 shows the
densit y of individuals across this triangular genotype space
before (hatchlings) and after (larvae) natural selection. In every
pond, there is a clear shift toward a higher frequency of
mixed-ancestry salamanders. That is, in all five ponds, the
multilocus ancestry index shifts toward a more intermediate
value and away f rom initially more extreme values. This was tr ue
regardless of whether the before-selection ancestry indices were
fairly evenly distributed (Pond F) or skewed toward native (Pond
G) or nonnative (BW1, CVP, JCL 2) ancestry values (Fig. 1).
Changes in single-locus allele frequencies were idiosyncratic
among markers and ponds (SI Table 3). However, most genot ype
f requencies (28/40 marker-pond combinations) showed a trend
in the direction of increased frequency of heterozygotes after
Table 1. Multivariate analyses of the difference between hatchlings and survivors
BW1 CVP Pond F Pond G JCL2
Coefﬁcient P Coefﬁcient P Coefﬁcient P Coefﬁcient P Coefﬁcient P
Intercept 1.151 ⫺0.598 2.777 1.188 3.376
1.911 0.002 13.744 0.000 4.246 0.000 ⫺4.641 0.002 2.816 0.267
3.508 0.000 2.076 0.000 5.190 0.000 3.079 0.000 5.924 0.000
⫺1.756 0.088 ⫺12.368 0.002 ⫺6.276 0.002 ⫺7.788 0.002 ⫺6.335 0.042
A positive coefﬁcient for
indicates selection favoring nonnative alleles, a positive coefﬁcient for
indicates selection favoring heterozygous hybrid
genotypes, and negative coefﬁcients for
indicate selection favoring mixed genotypes regardless of heterozygosity. These heuristic interpretations of the
coefﬁcients are correct when all else is equal; refer to Fig. 1 to see actual changes in genotype distributions.
Fig. 1. Approximate densities of multilocus genotypes in hatchlings (before
selection) and larvae (after selection) in ﬁve populations of hybrid tiger
salamanders were estimated by using a two-dimensional density-estimation
procedure with a bivariate normal kernel [function kde2d in R (30, 31)]. Gray
triangles represent potential genotype space described by ancestry (the frac-
tion of introduced alleles in an individual’s marker genotype) and heterozy-
gosity (fraction of markers heterozygous for native and introduced alleles).
Darker areas are those with larger numbers of observed individuals.
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0704791104 Fitzpatrick and Shaffer
selection, supporting the interpretation of hybrid vigor (SI Table
3). One possible exception is the GNAT1 marker, for which the
dominance c oefficient was negative (albeit not statistically sig-
n ificant) in all five ponds, suggesting low fitness of heterozygotes
for this genomic region despite the overall pattern of hybrid vigor
(SI Table 3).
Our study of early-larval survival in hybrid tiger salamanders
reveals higher fitness of hybrid genotypes with evenly mixed
ancestry than genotypes composed predominantly of alleles
f rom one parental lineage. Such strong evidence of hybrid vigor
was unexpected based on previous studies suggesting constraints
on admixture, inferred from high levels of linkage disequilibrium
and habitat-dependent invasion success (21, 22). Our statistical
analyses suggest that hybrid survival is influenced by both
heteroz ygote advantage and recombinant hybrid vigor. Correct
interpret ation of the underlying genetics is crucial to predicting
the evolutionary outcome of admixture between native Califor-
n ia Tiger Salamanders and introduced Barred Tiger
Salamanders. From a conservation perspective, hybrid vigor may
be viewed as good or bad for the California Tiger Salamander,
depending on whether one chooses to emphasize fitness of ext ant
populations or genetic purit y as a primary conservation goal.
The Genetic Basis of Hybrid Vigor. When considering the genetic
basis of hybrid vigor, it is important to recall that our markers
represent a very sparse sample of the genome. Our ability to
derive infor mation from this set of markers is enhanced by
extensive linkage disequilibrium due to admixture in hybrid
populations (22). Admixture linkage disequilibrium decreases
the likelihood that a particular marker–trait association repre-
sents close physical linkage between the marker and a quanti-
t ative trait locus (32). By the same token, admixture linkage
disequilibrium increases the ac curacy of our marker-based esti-
mates of genomic ancestry (
) and heterozygosity (
the sampled loci are correlated w ith many other regions of the
genome (33, 34).
Our analyses show a strong positive relationship between
heteroz ygosity and probability of survival (Table 1 and Fig. 1).
This result is consistent with heterozygote advantage (overdomi-
nance for fitness) contributing to the genetic basis of hybrid
vigor. The coefficient on the ‘‘admixture’’ ter m (
) is consis-
tently negative, implying that recombinant hybrid vigor also
plays a role. Recombinant hybrid v igor may be caused by
synergistic interactions bet ween heterospecific alleles at dif fer-
ent loci (epistasis) or complement ary effects of superior intro-
duced alleles at some loci and superior native alleles at other loci
(19, 20). That is,
is better viewed as an indicator of overall
genomic admixture rather than additive ⫻ additive interaction
Hybrid vigor in other systems has often been attributed to
heteroz ygote advantage (35–39). However, heterozygote advan-
t age is lost each generation because of segregation among
gametes. If the most-fit genotype is heterozygous at many loci,
it cannot breed tr ue and will rarely, if ever, reappear in gener-
ations beyond the F
(40). In contrast, if hybrid vigor is caused
by complementary or epistatic genes, the most-fit genotype is
homoz ygous for alleles derived from each lineage at different
loci (19). Such recombinant homozygotes appear only in later
generations, and they can breed true. Tr ue breeding recombi-
nant genotypes can therefore establish new lineages with ge-
nomes that are mosaics of the two ancestral genomes (41–44).
If both heteroz ygote advantage and recombinant hybrid vigor
are operating in the Ambystoma hybrid zone, then one likely
outc ome is a population of tiger salamanders that is fixed for
native alleles at some loci, advant ageous introduced alleles at
other loci, and segregating native and introduced alleles as
balanced polymorphisms at still other loci. However, the statis-
tical pattern of high mean fitness of highly heterozygous geno-
t ypes may be transient, a temporary consequence of the current
levels of admixture and linkage disequilibrium that may abate as
the population genetic background changes from a highly vari-
able hybrid swarm to a more stable mosaic of native and
introduced alleles in dif ferent parts of the genome. As the
population bec omes fixed for a single admixed genetic back-
ground, the heterozygote advantage at a given locus may disap-
pear (SI Text).
Variation, link age disequilibrium, and divergent allele fre-
quencies bet ween habitats have been maintained over 12–24
generations of admixture between A. californiense and A. t.
mavor tium in the Salinas Valley (22, 23). However, hybrid vigor
tends to maint ain genetic variation, decrease link age disequi-
librium bet ween conspecific alleles, and homogenize allele fre-
quencies across habit ats. A ll of these processes are illustrated in
Fig. 1. Three processes are likely working against the tendency
of hybrid vigor to bring about a st able distribution of mixed-
ancestry genotypes. First, genetic drif t contributes to variance
among breeding ponds (45). Second, higher frequencies of
introduced alleles are maintained in perennial than in seasonal
breeding ponds (22). Finally, immigration from pure native
populations outside of the hybrid swarm tends to increase native
allele frequencies and contribute to a broad-scale geog raphic
gradient in allele frequencies (23).
The high frequencies of introduced alleles in perennial ponds
remains one of the strongest genetic patterns in the tiger
salamander hybrid system; it may be maint ained by habit at
choice or selection in later life history stages. Fitzpatrick and
Shaf fer (22) hypothesized that introduced genotypes gain an
advant age in perennial water bodies by extending their larval
period or even breeding in the larval (paedomorphic) condition.
Pure native Californ ia Tiger Salamanders must met amorphose
to reproduce (46), but Barred Tiger Salamanders regularly forgo
met amorphosis in perennial ponds and breed as paedomorphs
(47, 48). Paedomorphs often reach sexual maturity earlier than
met amorphs, produce larger clutches, and may breed earlier in
a given season; any of these life-history factors may provide
introduced genotypes w ith an advantage in perennial ponds (49,
50). The perenn ial pond populations studied here (BW1 and
JCL2) show strong viability selection favoring hybrids during the
early part of the larval period (Table 1 and Fig. 1). Ant agonism
bet ween hybrid vigor for larval survival in all ponds and later
selection favoring introduced genotypes in perenn ial ponds may
ex plain the persistent pattern of high introduced allele frequen-
cies in perennial ponds and relatively even native and introduced
allele frequencies in seasonal ponds throughout the range of the
hybrid swarm (23).
Conservation Consequences of Hybrid Vigor. Hybridization between
an introduced form (A. t. mavor tium) and a declin ing native
protected by the Endangered Species Act (A. califor niense) raises
several difficult conservation issues (8, 10, 23, 51). These include,
(i) what are the consequences of hybridization for population
viability? (ii) what are the consequences for other native species
and the communities in which they oc cur? (iii) which individuals
or populations of salamanders should be protected? and (iv)
should any individuals or populations be eradicated? The an-
swers to iii and iv depend, in part, on the answers to i and ii but
may also depend on practical considerations and perspectives on
the merit of genetic purit y as a conservation goal. Our results
have important implications for population viability and genetic
purit y of Californ ia Tiger Salamanders and for potential impacts
on other organisms native to California ponds and grasslands.
Hybrid vigor is probably not detrimental to population via-
bilit y. However, some authors have raised the concern that
negative effects of introgressive hybridization on fitness may not
Fitzpatrick and Shaffer PNAS
October 2, 2007
be evident except during infrequent ‘‘ecological cr unches’’ (10,
52). Such episodic changes in selection do oc cur (53–55), and
may be widespread on the California landscape, given its tre-
mendous among-year variance in rainfall and pond hydroperiod.
Introgression could potentially contribute to loss of a crucial
adapt ation during an interlude between episodes of selection,
leaving the population more vulnerable to extinction when the
next crunch arrives. In addition, our characterization of hybrid
fitness is inc omplete, focusing only on early larval mortalit y.
However, native and introduced tiger salamanders have been
mixing in California for ⬎50 years, and the hybrid swarm has
survived and expanded (23) through droughts, El Nin˜os, and
subst antial anthropogenic environmental change. The ef fect of
admixture on population v iability in Californ ia Tiger
Salamanders remains an important open question, but it does not
appear to us, at this time, to be a major threat.
Hybrid vigor implies that the native gene pool is not resist ant
to invasion of introduced alleles. Thus, a mixture of gene pools
is expected, with fixation of advantageous introduced alleles and
preservation of advantageous native alleles. This constitutes a
loss of biodiversit y, because alleles and genotypes unique to A.
califor niense are being lost. However undesirable it may be, this
evolutionary transformation should not engender the same
ethical concerns raised by true demog raphic extinction. In fact,
the aesthetic appeal of a genetically authentic native gene pool
may be in conflict with the best interests of the animals if hybrid
vigor enhances population viability. Whether or not population
viability is enhanced in this case depends on the relationship
bet ween average fitness and density regulation (56).
If hybrid vigor results in higher densities, larger body sizes, or
more rapid growth of larvae in hybrid populations, the impacts
of hybridization will almost certainly extend to other members of
the ecological community. Tiger salamander larvae are vora-
cious c onsumers of anuran tadpoles and aquatic invertebrates.
Endangered, threatened, and sensitive prey that share habitats
with tiger salamanders in the hybrid zone include California
Red-Legged Frogs (Rana draytonii), Western Spadefoots (Spea
hammondii), and Vernal Pool Fairy Shrimp (Branchinecta lyn-
chi). Thus, hybrid vigor in tiger salamander larvae presents a
credible but unproven threat to other pond-breeding organisms.
Our results support the view that hybridization can alter the
evolutionary process by c ontributing novel genetic advantages to
admixed populations (1, 15, 41). This view also suggests that
hybridization between native and introduced lineages could
ac celerate the evolution of invasiveness and intensify the eco-
logical impact of a biological invasion (9). Understanding the
ef fect of hybridization on the ecological interactions of tiger
salamander populations is a priorit y for future work in this
Natural selection among hybrid genotypes was estimated by
using a cohort analysis in which we c ompared the distributions
of genot ypes in a single group of offspring at t wo different points
in time. This prov ides a direct measure of the response to natural
selection within a single generation (57). Furthermore, by fo-
cusing on diagnostic codominant molecular markers, we mea-
sured genetic changes that can be readily explained in ter ms of
additive, dominant, and epistatic or complementary effects of
dif ferences between the native and introduced salamanders.
Sampling. We studied a subset of the breeding sites described in
ref. 22. Pond F and CVP are natural vernal pools that fill during
winter rains and dry by early to midsummer. They represent the
natural, unmodified breeding habitat for native California Tiger
Salamanders (46). Pond G is a man-made seasonal cattle pond
with a hydroperiod similar to vernal pools. JCL2 and BW1 are
perenn ial ponds that normally hold water through the summer,
although both had dried completely during the summer before
We quantified multilocus genotypes during two early life-
history stages. Our before-selection samples consisted of eggs
c ollected during the laying season, whereas our after-selection
samples were drawn early in the larval stage. Each pond was
outfitted w ith artificial oviposition sites consisting of 25 bamboo
st akes (1 m long and 0.5 cm in diameter) planted into the pond
bottom in water 30 cm deep and four 1.2-m-long coils of
12-gauge galvanized steel wire in water 40–50 cm deep. Ponds
were visited weekly from December 2002 through February
2003. On each visit, all salamander eggs (if present) were
removed from all st akes and wires to guarantee that all eggs
c ollected were laid during the preceding week. Eggs were
c ounted and placed in a bucket, from which a random sample of
50 eggs was collected. Remain ing eggs were scattered in shallow
water where they could readily adhere to submerged vegetation
and rocks (natural oviposition sites). Collected eggs were held in
cups contain ing 20% Holtfretter’s solution (58) and allowed to
hatch. Hatchlings were k illed and stored at ⫺80C.
Young larvae were captured by haphazardly sein ing ponds.
Collections were made on Febr uary 6 and March 26, 2003.
Reproduction in Pond F did not begin until the week of February
28, when there was a single bout of egg laying. The artificial
oviposition sites in Pond G were destroyed in early Febr uary by
cattle. Therefore, these two ponds are represented by single
samples of larvae. In the other three ponds, we found no
st atistically significant (
⫽ 0.05) differences between the two
samples of larvae (data not shown) and therefore pooled them
for all hatchling vs. larvae comparisons below.
Molecular Methods. Each individual was assayed for the nine
molecular markers used by Fitzpatrick and Shaffer (22). These
are single nucleotide differences that distinguish native A.
califor niense from introduced A. t. mavortium alleles at mtDNA
and eight mapped nuclear genes. Each individual was charac-
terized as having native or introduced mtDNA and as being
homoz ygous native, heterozygous, or homoz ygous introduced at
each nuclear marker. Primer sequences and PCR c onditions can
be found in ref. 22 (also see refs. 21 and 59). Diagnostic alleles
were identified by restriction enzyme digestion and agarose
electrophoresis of PCR products, as described by Fitzpatrick and
Shaf fer (22).
Data Analyses. We used both single-marker and multilocus sta-
tistical analyses to evaluate variation in survival among geno-
t ypes. For single locus selection analyses, we used standard
marker–trait regressions (60, 61);
1 ⫺ P
⫽ intercept ⫹ b
⫹ error, 
where P is the probabilit y that a dat a point is a surviving larva,
⫽⫺1, 0, or 1 for homozygous native, heteroz ygous, and
homoz ygous introduced genotypes, respectively, and X
1 for homozygous and heterozygous genotypes, respectively. The
fitted c oefficients b
represent additive and dominance
ef fects. Regressions were fitted in the prog ram R 2.4.1 (31) by
using the glm() function with a binomial error distribution and
observations weighted by stratum (see below). We assume fitted
models represent indirect responses to selection on loci with
unk nown linkage relationships to our markers. The residual
error includes variance due to incomplete linkage between
markers and selected sites as well as environmental and sampling
For multilocus selection analyses, we used the quantit ative
genetics approach described by Lynch (18). This model identifies
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0704791104 Fitzpatrick and Shaffer
additive, dominance, and epistatic effects on hybrid fitness as
ter ms in a linear statistical model. In our case,
1 ⫺ P
⫽ intercept ⫹
⫹ error, 
where individual genotypes are described by
, a linear ancestry
index ranging from ⫺1 (all native alleles) to ⫹1 (all introduced
, a heterozygosity index ranging from ⫺1 (all
markers homozygous) to ⫹1 (all markers heteroz ygous). The
fitted coef ficients
describe simple additive and dom-
inance effects, whereas
, and (
describe additive ⫻
additive, dominance ⫻ dominance, and additive ⫻ dominance
epist asis, respectively (18). This model was fitted to the data
f rom each pond by using the glm() function with a binomial error
distribution and observations weighted by stratum, as for the
single-locus case (see below).
Two additional issues needed to be addressed in our statistical
analyses. First, an interesting problem was created by the tem-
porally extended breeding season in three of our study ponds
(BW1, G, and JCL2). In these ponds, breeding consisted of three
to five major episodes corresponding to significant rain storms;
each episode potentially involved a different set of adults with
dif ferent genotype frequencies. Thus, our before-selection sam-
ples were stratified simple random samples. Each weekly sample
(stratum sample) was weighted by stratum size when estimating
before-selection genotype frequencies and their moments (62).
We used the tot al number of eggs laid on our artificial ovipo-
sition sites each week as that week’s stratum size (we assume this
number is proportional to the total number of eggs produced in
the pond that week). Each before-selection observation (hatch-
ling) was weighted ac cording to stratum size [(stratum size ⫼
stratum sample size) ⫻ (tot al sample size ⫼ sum of stratum
sizes)] and those weights passed to the glm() function. Under this
weighting scheme, the after-selection individuals each get a
weight of 1.0 because they come from a single stratum, and the
averages of the individual weights before and after selection are
Sec ond, traditional estimates of confidence intervals for re-
gression c oefficients do not account for error in the independent
were estimated with error because
our markers are samples of the genome (
⫺ 1, and
⫺ 1, where P
is the fraction of an individual’s marker
alleles derived f rom A. t . mavor tium, and P
is the fraction of
markers heterozygous for native and introduced alleles). There-
fore, we used a bootstrap and randomization procedure to
inc orporate this sampling variance into hypothesis tests regard-
ing the fitted c oefficients described in Eq. 2. For each of 10,000
replicates, we performed a two-step Monte Carlo procedure.
First, we bootstrapped across nuclear markers. Bootstrapping
across markers addresses error in individual estimates of
due to sampling only a f raction of each salamander’s genome.
This is in the same spirit as bootstrapping across characters in
phylogenetic analysis (63) and results in each individual having
associated with it a distribution rather than a single point
estimate of each
. MtDNA was treated as a fixed effect; it was
included once and only once in every replicate, because there is
only a single mitochondrial allele to sample. Excluding mtDNA
slightly affected parameter estimates but not their signs or P
values. Including mtDNA in the pool of markers to resample is
inappropriate because it does not contribute to
. The second
step in each replicate was to randomize observations to simulate
the null hypothesis that hatchlings and surviving larvae were
random samples from a single distribution of genotypes. We
rec orded the fraction of replicates giving coefficients larger
) and smaller ( f
) than the estimate from the original
dat a and estimated two-tailed P values as 2 ⫻ minimum(f
). The procedure was applied separately to each of our five
The full Lynch (18) model (Eq. 2 ) may overfit some dat a sets
(particularly those with highly skewed allele frequencies), lead-
ing to problems with colinearity of variables and high variances
around coefficient estimates. Therefore, we also fitted a reduced
1 ⫺ P
⫽ intercept ⫹
This model focuses on fitness variation as a function of
individual ancestry (
), individual heterozygosit y (
), and as a
function of individual admixture (
) decoupled from ancestry
and heterozygosit y (
equals 1 in both pure native and pure
introduced genotypes, and it equals 0 in individuals with evenly
mixed ancestry, i.e., half native and half introduced alleles
regardless of heterozygosity). The coefficient
ef fects of mixed ancestry not depending on heterozygosity but
does not distinguish true epistasis from c omplementary effects.
We compared the likelihoods of the full and reduced models by
means of Akaike’s Information Criterion (AIC) (64).
We thank A. Chang, M. Fujit a, O. J. Abramyan, P. C. Trenham, and
W. K. Savage for assistance in the field and laboratory; the H.B.S.
laboratory for discussion; and many ranchers for access to their prop-
erties. Our work was funded by Environmental Protection Agency
Grants U 91572401 and R 828896, the U.S. Department of Agriculture
(Exotic/Invasive Pests and Diseases Research Program Grant 04XN022
and National Science Foundation Grants DEB 0516475 and DEB
0213155), CAL FED Grant 01-N43, and the Un iversity of California,
Davis, CA, Agricultural Experiment Station.
1. Arnold ML (2006) Evolution Through Genetic Exchange (Oxford Univ Press,
2. Dowling TE, Secor CL (1997) Annu Rev Ecol Syst 28:593–619.
3. Mallet J (2005) Trends Ecol Evol 20:229–237.
4. Barton NH (2001) Mol Ecol 10:551–568.
5. Rivera MC, Lake JA (2004) Nature 431:152–155.
6. Simonson AB, Servin JA, Skophammer RG, Herbold CW, Rivera MC, Lake
JA (2005) Proc Natl Acad Sci USA 102:6608–6613.
7. Arnold ML, Meyer A (2006) Zoology 109:261–276.
8. Rhymer JM, Simberloff D (1996) Annu Rev Ecol Syst 27:83–109.
9. Ellstrand NC, Schierenbeck KA (2000) Proc Natl Acad Sci USA 97:7043–7050.
10. Allendorf FW, Leary RF, Spruell P, Wenburg JK (2001) Trends Ecol Evol
11. Allendorf FW, Luikart G (2007) Conservation and the Genetics of Populations
(Blackwell, Malden, MA).
12. Reisenbichler RR, Rubin SP (1999) ICES J Mar Sci 56:459–466.
13. Mansfield KG, Land ED (2002) J Wildl Dis 38:693–698.
14. Hedrick PW (1995) Conservation Biol 9:996–1007.
15. Lewontin RC, Birch LC (1966) Evolution (Lawrence, Kans) 20:315–336.
16. Grant BR, Grant PR (1996) Ecology 77:500–509.
17. Anderson E (1948) Evolution (Lawrence, Kans) 2:1–9.
18. Lynch M (1991) Evolution (Lawrence, Kans) 45:622–629.
19. Rieseberg LH, Archer M A, Wayne RK (1999) Heredity 83:363–372.
20. Lippman ZB, Zamir D (2007) Trends Genet 23:60–66.
21. Riley SPD, Shaffer HB, Voss SR, Fitzpatrick BM (2003) Ecol App 13:1263–
22. Fitzpatrick BM, Shaffer HB (2004) Evolution (Lawrence, K ans) 58:1282–1293.
23. Fitzpatrick BM, Shaffer HB (2007) Ecol App 17:598–608.
24. Shaffer HB, McKnight ML (1996) Evolution (Lawrence, Kans) 50:417–433.
25. Shaffer HB, Pauly GB, Oliver JC, Trenham PC (2004) Mol Ecol 13:3033–3049.
26. US Fish and Wildlife Service (2004) Federal Register 69:47211–47248.
27. Anderson JD, Hassinger DD, Dalrymple GH (1971) Ecology 52:1107–1112.
28. Lexer C, Randell RA, Rieseberg LH (2003) Ecology 84:1688–1699.
29. Rieseberg LH, Linder CR (1999) Ecology 80:361–370.
30. Venables WN, Ripley BD (2002) Modern Applied Statistics with S-Plus, 4th Ed
(Springer, New York).
Fitzpatrick and Shaffer PNAS
October 2, 2007
31. R Development Core Team (2006) R: A Language and Environment for
Statistical Computing (R Foundation for Statistical Computing, Vienna,
32. Musani SK, Halbert ND, Redden DT, Allison DB, Derr JN (2006) Genetics
33. Parra EJ, Marcini A, Akey J, Martinson J, Batzer MA, Cooper R, Forrester T,
Allison DB, Deka R, Ferrell RE, Shriver MD (1998) Am J Hum Genet
34. Redden DT, Divers J, Vaughan L, Tiwari HK, Beasley MM, Fernandez JR,
Kimberly RP, Feng R, Padilla M, Lui N, et al. (2006) PLoS Genet 2:e137.
35. Mitton JB (1997) Selection in Natural Populations (Oxford Univ Press, New
36. David P (1998) Heredity 80:531–537.
37. Lynch M, Walsh B (1998) Genetics and Analysis of Quantitative Traits (Sinauer,
38. Hansson B, Westerberg L (2002) Mol Ecol 11:1267–1274.
39. Pujolar JM, Maes GE, Vancoillie C, Volckaert FAM (2005) Evolution (Law-
rence, Kans) 59:189–199.
40. Lewontin RC (1974) The Genetic Basis of Evolutionary Change (Columbia Un iv
Press, New York).
41. Rieseberg LH, Raymond O, Rosenthal DM, Lai Z, Liv ingstone K, Nakazato
T, Durphey JL, Schwartzbach AE, Donovan LA, Lexer C (2003) Science
42. James JK, Abbot RJ (2005) Evolution (Lawrence, Kans) 59:2533–2547.
43. Gompert Z, Fordyce JA, Forister ML, Shapiro AM, Nice CC (2006) Science
44. Mavarez J, Salazar CA, Bermingham E, Salcedo C, Jiggins CD, Linares M
(2006) Nature 441:868–871.
45. Whitlock MC (2004) in Ecology, Genetics and Evolution of Metapopulations, eds
Hanski I, Gaggiotti O (Elsevier, Burlington, MA), pp 153–173.
46. Shaffer HB, Trenham PC (2005) in Amphibian Declines: The Conservation
Status of United States Species, ed Lannoo MJ (Univ of California Press,
Berkeley, CA), pp 605–608.
47. Petranka JW (1998) Salamanders of the United States and Canada (Smithsonian
Institution Press, Washington, DC).
48. Collins JP (1981) Copeia 1981:666–675.
49. Whiteman HH (1994) Q Rev Biol 69:205–221.
50. Wilbur HM, Collins JP (1973) Science 182:1305–1314.
51. Daniels MJ, Corbett L (2003) Wildl Res 30:213–218.
52. Moran P (2002) Ecol Freshw Fish 11:30–55.
53. Bumpus H (1899) Biological Lectures 1898:209–228 (Marine Biological Lab-
oratory, Woods Hole, M A).
54. Grant BR, Grant PR (1993) Proc Roy Soc London B 251:111–117.
55. Lande R, Arnold SJ (1983) Evolution (Lawrence, Kans) 37:1210–1226.
56. Christiansen FB (1975) Am Nat 109:11–16.
57. Endler JA (1986) Natural Selection in the Wild (Princeton Un iv Press,
58. Asashima M, Malacinski GM, Smith SC (1989) in Developmental Biology of the
Axolotl, eds Ar mstrong JB, Malacinski GM (Oxford Univ Press, New York), pp
59. Voss SR, Smith JJ, Gardiner DM, Parichy DM (2001) Genetics 158:735–
60. Liu BH (1998) Statistical Genomics (CRC, New York).
61. Clayton D (2001) in Handbook of Statistical Genetics, eds Balding DJ, Bishop
M, Cannings C (Wiley, New York), pp 519–540.
62. Heyadat AS, Sinha BK (1991) Design and Inference in Finite Population
Sampling (Wiley, New York).
63. Felsenstein J (1985) Evolution (Lawrence, Kans) 39:783–791.
64. Burnham KP, Anderson DR (2004) Sociol Methods Res 33:261–304.
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0704791104 Fitzpatrick and Shaffer