Hybridization Between Canada Lynx and Bobcats: Genetic Results and Management Implications

Abstract

Hybridization between taxonomically similar species is an often-overlooked mechanism limiting the recovery of threatened and endangered species. We present molecular genetic data for the first time demonstrating that Canada lynx and bobcats hybridize in the wild. We verify that two microsatellite loci Lc106 and Lc110 have non-overlapping allele ranges between Canada lynx and bobcats, and that three putative lynx from Minnesota contain DNA from both bobcats and lynx. Additionally, we use a published test for the 16S rRNA region of mitochondrial DNA (mtDNA) to determine the maternal species; all hybrids had lynx mothers. Fifteen per cent (3/20) of our putative lynx samples were hybrids, although these data are not from a representative sampling effort. Hybridization may be an under-appreciated factor limiting the distribution and recovery of lynx. The presence of hybrids is thus a new factor in the population management of both species with potential implications for hunting and trapping of bobcats.

Full text

Hybridization between Canada lynx and bobcats: Genetic results and
management implications
Michael K. Schwartz
1
*, Kristine L. Pilgrim
1
, Kevin S. McKelvey
1
, Edward
L. Lindquist
2
, James J. Claar
3
, Steve Loch
2
& Leonard F. Ruggiero
1
1
USDA/USFS Rocky Mountain Research Station, 800 E. Beckwith, Missoula, MT 59801, USA;
2
USDA/
USFS Superior National Forest, 8901 Grand Avenue Place, Duluth, MN 55808, USA;
3
USDA/USFS
Northern Regional Office, 200 E. Broadway, Missoula, MT 59801, USA (*Author for correspondence: fax:
+1-406-543-2663; e-mail: mkschwartz@fs.fed.us)
Received 4 June 2003; accepted 15 August 2003
Key words: Bobcat, DNA, hybridization, Lynx, management, micro satellite, non-invasive genetic sampling
Abstract
Hybridization between taxonomically similar species is an often-overlooked mechanism limiting the
recovery of threatened and endangered species. We present molecular genetic data for the first time
demonstrating that Canada lynx and bobcats hybridize in the wild. We verify that two microsatellite loci
Lc106 and Lc110 have non-overlapping allele ranges between Canada lynx and bobcats, and that three
putative lynx from Minnesota contain DNA from both bobcats and lynx. Additionally, we use a published
test for the 16S rRNA region of mitochondrial DNA (mtDNA) to determine the maternal species; all
hybrids had lynx mothers. Fifteen per cent (3/20) of our ‘putative lynx’ samples were hybrids, although
these data are not from a representative sampling effort. Hybridization may be an under-appreciated factor
limiting the distribution and recovery of lynx. The presence of hybrids is thus a new factor in the population
management of both species with potential implications for hunting and trapping of bobcats.
Introduction
Hybridization between taxonomically similar spe-
cies has been proposed as a mechanism for limiting
species’ geographic ranges (Barton 2001). Rare
species that come into contact with more common
species may be particularly sensitive to this pro-
cess. While hybridization is common in plant
species, there are fewer examples in the Animal
Kingdom, and the process has been often over-
looked as a significant factor influencing the evo-
lution, range, and distribution of many animals
(Dowling and Secor 1996; Rhymer and Simberloff
1996). To understand the conservation and man-
agement implications of hybridization it is
important to distinguish whether the causes of
hybridization are natural or anthropogenic in
origin (Allendorf et al. 2001).
Natural hybridization can influence the evolu-
tionary trajectory of a species, providing new ge-
netic material for evolutionary forces to act upon.
Determining how to manage cases of natural
hybridization may depend, in part, on the kind of
the hybridization (Allendorf et al. 2001). For in-
stance, managers may treat hybrids that have
arisen from novel contact (e.g., via range expan-
sion) between distinct parental species different
from hybrids formed thousands of years ago now
proceeding on their own evolutionary tracks as
separate entities. In general, it has been recom-
mended that hybrids arising from natural causes
be given protected status when needed (Allendorf
et al. 2001).
On the other hand, hybridization is often
facilitated by anthropogenic disturbances on the
landscape. Specifically, the globalization of species
Conservation Genetics 5: 349–355, 2004.
Ó2004 Kluwer Academic Publishers. Printed in the Netherlands.
349

through habitat changes as well as deliberate and
accidental translocations have increased contact
between previously isolated species. In these cases
it has been recommended that hybrids be given less
protection under current US law (Allendorf et al.
2001).
Management of hybrids also depends on the
genetic consequences of hybridization. The two
extreme consequences of hybridization are: (1) the
widespread genetic introgression or complete
admixture of taxa, and (2) hybridization without
genetic introgression (see Allendorf et al. 2001 for
an extensive review on categorization of hybrid-
ization). Widespread introgression and complete
admixture can occur when the hybrids are repro-
ductively fertile and hybrid matings are not avoi-
ded, and thus mate with either the parental types
or other hybrids (e.g., Goodman et al. 1999).
Hybridization without introgression often occurs
when the hybrids have post-zygotic reproductive
isolating mechanisms (Mayr 1972) rendering them
effectively sterile. For example, hybridization
without introgression has been documented in bull
trout (Salvelinus confluentus) in Montana where
this species extensively hybridizes with brook trout
(S. fontinalis), but seldom produces offspring be-
yond the F
1
generation (Leary et al. 1993; Spruell
et al. 2001).
The production of hybrid offspring is problem-
atic for the preservation of rare species. Production
of sterile offspring is particularly destructive if fe-
males of the rare species are involved in hybrid
coupling because small populations are often lim-
ited by the number of reproductively fertile females.
Production of fertile hybrid offspring is also prob-
lematic because the rare species must compete and
potentially further hybridize both with the more
common species and hybrids.
The Canada lynx (Lynx canadensis) is a wide
ranging felid (Ward and Krebs 1985; Slough and
Mowat 1996; Mowat et al. 2000; Schwartz et al.
2002). The primary core habitat of the lynx is the
boreal forest of Canada and Alaska. The southern
distribution of native lynx extends into the
northern US Rockies, the north Cascades of
Washington State, northeastern Minnesota, and
western Maine (McKelvey et al. 2000). Lynx are
also located in Colorado where a population was
introduced in 1999. The lynx was recently listed as
‘Threatened’ by the United States Fish and Wild-
life Service under the US Endangered Species Act
in the contiguous United States (Federal Register
2000).
Bobcats (Lynx rufus), a distinct species from
lynx (Werdelin 1981; Johnson and O’Brien 1997),
are widespread throughout the conterminous
United States and reach their northern extent in
southern Canada. Lynx and bobcats do not typi-
cally occur in the same habitats. However, while
bobcats are generally precluded from areas of
heavy snow cover, the two will occasionally co-
exist and are likely competitors (Aubry et al. 2000;
Buskirk et al. 2000).
In Minnesota, the peripheries of lynx and
bobcat ranges overlap. Lynx were historically
trapped in Minnesota, yet were either present
in low numbers or absent from the state between
1993 and 2000 (McKelvey et al. 2000). Photo evi-
dence in 2001 and non-invasive genetic sampling in
2002 confirmed their existence in four counties
in northeastern and north-central Minnesota
(Figure 1). However, lynx remain concentrated in
the northeastern corner of Minnesota despite sim-
ilar forest habitat continuing to the southwest.
Bobcats are infrequently trapped in the northeast-
ern corner of the state, but are common in areas
further south (e.g., 0.63%of the bobcats trapped in
Minnesota between 1991 and 2002 came from the
two northeast counties).
Surveys commencing in 2002 in Minnesota
documented lynx presence in the state. During
these surveys, many non-invasive genetic samples
(i.e., hair and feces) were collected and tested to
identify species (Mills et al. 2001). Additionally,
three putative lynx were killed (via train, highway,
and trapping mortalities in December 2001,
November 2002, and December 2002, respec-
tively). Two of the mortalities had slightly abnor-
mal morphology. Specifically, they had large feet
and a mostly black tail band, similar to lynx, while
also having short ear tufts and compact bodies
associated with bobcats. Here we confirm the hy-
brid origin of these animals and discuss the
implications of these data. This represents the first
verified hybridization between Canada lynx and
the common bobcat in the wild.
Methods
As part of an effort to document the presence and
distribution of lynx in Minnesota we collected
350

non-invasive genetic samples by following snow
tracks associated with putative lynx. When ‘lynx’
tracks were found, they were followed and hair
from daybeds and scats were collected (E. Lind-
quist, unpublished data). These surveys resulted in
40 non-invasively collected scats and hair samples.
Hair and scats were immediately sent to the Rocky
Mountain Research Station’s Wildlife Ecology
unit located in Missoula, Montana. Once in the
laboratory, we extracted DNA from each hair
sample using the QIAGEN DNEASY Tissue Kit
(QIAGEN Inc., Germany) and manufacturer pro-
tocols, and from each scat using the QIAMP DNA
Stool Minikit (QIAGEN Inc., Germany).
We established a ‘hybrid assay’ using differ-
ences in the nuclear genome between lynx and
bobcats. This approach was similar to the felid
species identification method of Ernest et al. (2000)
where species were identified by non-overlapping
allele distributions. If two species have non-over-
lapping allele distributions at a locus, then an F
1
hybrid can be identified by the presence of an allele
from both distributions. Our approach had two
stages. First, we amplified all samples at micro-
satellite markers Lc106 and Lc110 (Carmichael
et al. 2001) and visualized the resultant products
on a LICOR DNA analyzer (Lincoln, Nebraska,
USA). Our PCR reactions for tissues were con-
ducted at volumes of 10 ll containing 50–100 ng
of purified genomic DNA, 10 mM Tris–HCl (pH
8.3), 50 mM KCl, 2.0 mM MgCl
2
, 0.2 mM of each
dNTP, 0.2 lM of each primer and 0.5 U Taq
DNA polymerase. When we used hair and scat
samples we modified this protocol by using 2.5 ll
of DNA preparation, along with 2 lg/ml BSA and
1UTaq DNA polymerase. The thermal profile for
the PCR reaction involving DNA from tissue was
94 °C for 5 min, followed by 30 cycles of: 94 °C
for 30 s, 56 °C for 30 s, and 72 °C for 30 s. PCR
profiles for hair and scat samples were the same,
except we increased all steps to 1 min and cycled
the reaction 45 times.
Both Lc106 and Lc110 have been reported to
have non-overlapping allele frequency distribu-
tions in lynx and bobcats. Locus Lc106 has been
reported to have alleles between 98 and 110 base
pairs (bp) in lynx and either 88 or 90 bp in bob-
cats; Marker Lc110 was reported to have allele
frequencies between 91 and 103 bp in lynx and to
be fixed at 80 bp in bobcats (Carmichael et al.
2000). To ensure that the reported allele frequency
distributions were geographically consistent, we
evaluated these microsatellites on 108 lynx tissue
samples from Alaska, Washington, Wyoming,
Montana, Ontario, British Columbia, Yukon
Territories, and Northwest Territories (Schwartz
et al. 2002, 2003) and 79 bobcat tissue samples
from Minnesota, Wisconsin, Wyoming, Colorado,
Oregon and Florida. Thirty-eight of the lynx
samples were from areas just north of Minnesota
in Ontario where bobcats had historically been
absent, and 23 of the bobcat samples were from
Minnesota counties adjacent to counties where
lynx were reported at the time of the survey. The
second stage of the hybrid assay was implemen-
tation of 16S rRNA mtDNA species identification
test (Mills et al. 2001) to determine the direction of
Figure 1. Current distribution of lynx in Minnesota, based on
non-invasive genetic sampling surveys (circles), and locations of
the Canada lynx–bobcat hybrids (asterisks).
351

hybridization (i.e., if mtDNA was consistent with
lynx, the maternal parent was a lynx; if consistent
with bobcat then the maternal parent was a bob-
cat).
We were also interested in providing an index
of the rate of hybridization between lynx and
bobcats in our ad hoc surveys. While this index is
from a non-representative sample, it provides
some indication as to the rate of hybridization
expected given a more representative sampling
strategy. To obtain this index we needed to esti-
mate the number of unique individuals represented
by the 40 non-invasive genetic samples and the
three putative lynx incidentally killed. Thus, in
addition to the two microsatellites used in the
hybrid assay we amplified DNA with four addi-
tional microsatellite markers described by Carmi-
chael et al. (2000; Lc109,Lc111,Lc118,Lc120)
using our aforementioned protocols with the
exception that markers Lc109 and Lc111 were
annealed at 54 °C. Probability of identity among
siblings (P
(ID)Sib
; Evett and Weir 1998; Waits et al.
2001) using these six microsatellites and the lynx
samples collected from Ontario immediately north
of our population of interest was 0.005. Using only
a four-locus genotype (see below), P
(ID)Sib
was
0.03, sufficiently low to detect unique individuals
based on the criteria of Mills et al. (2000) and
Waits et al. (2001).
Because the DNA from hair is subject to allelic
dropout and other genotyping errors (Taberlet
et al. 1996, 1999; Gagneux et al. 1997; Goossens
et al. 1998; Waits and Leberg 2000; Creel et al.
2003) we attempted to analyze each hair or scat
sample three times at each locus. Samples that
produced scorable products on less than four of six
loci were removed from subsequent analyses (18
samples culled from the database), similar to the
protocols of Mowat and Paetkau (2002) and Pae-
tkau (2003). We observed four cases of allelic
dropout based on three repetitions (twice at Lc120,
once at Lc106, and once at Lc110). In these cases,
we ran each loci an additional two times and ob-
served consistent heterozygous genotypes. All
other loci were consistent across the three runs.
Individual genotypes were considered to be
from unique individuals if they differed by more
than one allele. However, individuals that differed
by only one allele had all replicates of each locus
re-scored by an independent observer.
Results
Extensive testing revealed that allele frequencies
for loci Lc106 and Lc110, though more variable
than reported by Carmichael et al. (2000) at locus
Lc110 in bobcats, do not overlap (Figures 2 and 3,
Figure 2. Microsatellite gel image showing lynx, bobcats, and the three hybrids identified in this study. Outer lanes of the gels contain
size standards.
352

Table 1); thus our hybrid test is robust. We de-
tected a minimum of 20 different genotypes that
were consistent with lynx based on mtDNA (Mills
et al. 2001); 17 were associated with the 40 non-
invasively collected scats and hairs from Minne-
sota (Table 2) and three corresponded to the
incidental kills. One hair sample from these 17
individuals (collected in February 2002), and the
two tissue samples described as lynx with abnor-
mal morphology were hybrids (Figure 2, Table 1).
Thus three of 20 individuals were hybrids. Because
all three hybrids were identified as lynx based on
mtDNA tests, hybridization was between male
bobcats and female lynx. At this time we cannot
determine whether these hybrids were the result of
one or multiple litters from our genetic results,
although their geographic (Figure 1) and temporal
separation suggest multiple litters.
Discussion
We show here the first confirmed substantiation
that wild female lynx mate with wild male bobcats.
There have been some anecdotal reports of hybrids
in the trapping and fur farm communities. We
contacted felid fur farms that claim to have at-
tempted to breed lynx and bobcats in captivity.
Only one asserted success in generating Canada
lynx–bobcat hybrids, but stated that subsequent
attempts to backcross lynx and bobcats failed.
Other fur farms were often unsure of the species of
Figure 3. Allele frequencies for lynx and bobcats at loci Lc106
and Lc110. Allele size ranges do not overlap between species at
either locus.
Table 1. Genetic variability for microsatellite loci Lc106 and Lc110 for the samples of bobcats (n = 79) and lynx (n = 108) used to
assess the validity of our hybrid test
Bobcat Lynx Hybrid 1 Hybrid 2 Hybrid 3
Range H
e
H
o
ARange H
e
H
o
A
Lc106 88–90 0.48 0.52 2 96–110 0.79 0.77 8 88/100 88/98 88/98
Lc110 72–81 0.80 0.77 9 94–108 0.81 0.70 8 75/98 76/94 76/98
Allele ranges are described in base pairs. H
e
is the expected heterozygosity calculated using Nei (1987), H
o
is observed heterozygosity, and Ais
the number of alleles. Additionally, the specific genotypes associated with the three hybrid samples are presented.
Table 2. Descriptive statistics of the 17 unique lynx (excluding
hybrids) identified by non-invasive genetic sampling in Minne-
sota
Locus H
e
H
o
A
Lc106 0.78 0.75 6
Lc109 0.76 0.63 5
Lc110 0.77 0.88 6
Lc111 0.73 0.76 5
Lc118 0.79 0.88 5
Lc120 0.67 0.56 4
Averages 0.75 0.74 5.17
H
e
is the expected heterozygosity calculated using Nei (1987), H
o
is
observed heterozygosity, and Ais the number of alleles.
353

lynx (Lynx canadensis,Lynx lynx,orLynx pardi-
nus) they had attempted to breed with bobcats.
Although the test we present may detect any
degree of hybridization, it is only fully diagnostic
for F
1
hybrids. If hybridization leads to sterile
offspring, we diagnosed all hybrids in our samples.
However, if hybrids are fertile, our test will pro-
vide an underestimate of the number of Canada
lynx–bobcat hybrids. As more markers are tested
and better reference collections from lynx and
bobcats established, we can hope to develop tests
(e.g., Spruell et al. 2001; Tranah et al. 2003) to
identify future generation hybrids, if they exist. We
are currently evaluating additional genetic samples
identified as lynx and bobcats based on mtDNA to
better determine the extent of hybridization and
whether hybrid coupling occurs between male lynx
and female bobcats.
While we recognize that our sample is not
representative of lynx in northeastern Minnesota,
it is interesting that three out of 20 identified
individuals (15%) were hybrids. We believe that
hybridization is an under-appreciated factor that
potentially limits the distribution and recovery of
lynx. If the F
1
hybrids are always sterile, the threat
to the lynx population is from lost recruitment
opportunities (Rhymer and Simberloff 1996; Al-
lendorf et al. 2001). On the other hand, the pro-
duction of fertile F
1
hybrids may eventually lead to
hybrid swarms (Avise et al. 1984; Forbes and Al-
lendorf 1991), also reducing the range and persis-
tence of pure lynx.
These data have at least two legal implications.
First, bobcat trapping is legal in counties that
currently contain lynx, while it is illegal to trap
lynx anywhere in the conterminous United States.
The United States Fish and Wildlife Service, the
agency responsible for conserving, protecting, and
enhancing the nation’s fish and wildlife and their
habitats, does not have an official hybrid policy
(Allendorf et al. 2001), thus it is unclear if the
bobcat-lynx hybrid is protected under the Endan-
gered Species Act. If protection is afforded, bobcat
trapping in areas with known lynx could be
problematic because both lynx and lynx–bobcat
hybrids can be incidentally taken from extant
populations. A second legal implication is the
identification of a potential threat to lynx recov-
ery. The United States Fish and Wildlife Service is
currently reevaluating the conservation needs of
lynx (Federal Register 2003). Any factors that may
favor bobcats in lynx habitat may lead to the
production of hybrids and thus be potentially
harmful to lynx recovery.
Overall, this paper presents evidence that
Canada lynx and bobcats hybridize in the wild.
Future efforts need to be undertaken to describe
the extent, rate, and nature of hybridization be-
tween these species, and to understand the eco-
logical context in which hybridization occurs.
Acknowledgements
We thank John Erb, the Minnesota DNR (Divi-
sion of Wildlife), Dave Kuehn, Mike Kennedy,
Tim Lee, and Larry Bickel for their help with
sample collection. Yvette Ortega, Rudy King,
Roman Biek and Fred Allendorf provided helpful
comments on earlier drafts of this manuscript.
Cheryl Copeland supplied helpful GIS advice.
Cory Engkjer and Carla Burgess provided valu-
able laboratory assistance. All samples were col-
lected under valid state and national permits.
References
Allendorf FW, Leary RF, Spruell P, Wenburg JK (2001) The
problems with hybrids: Setting conservation guidelines.
Trends Ecol. Evol.,16, 613–622.
Aubry KB, Koehler GM, Squires JR (2000) Ecology of Canada
lynx in Southern Boreal Forests. In: Ecology and Conserva-
tion of Lynx in the United States (eds. Ruggiero LF et al.),
pp. 373–396. University Press of Colorado, Boulder, Colo-
rado, USA.
Avise JC, Bermingham E, Kessler LG, Saunders NC (1984)
Characterization of mitochondrial DNA variability in a
hybrid swarm between subspecies of bluegill sunfish (Lep-
omis macrochirus). Evolution,385, 931–941.
Barton N (2001) The role of hybridization in evolution. Mol.
Ecol., 10, 551–568.
Buskirk SW, Ruggiero LF, Krebs CJ (2000) Habitat fragmen-
tation and interspecific competition: Implications for lynx
conservation. In: Ecology and Conservation of Lynx in the
United States (eds. Ruggiero LF et al.), pp. 83–100. Uni-
versity Press of Colorado, Boulder, Colorado, USA.
Carmichael LE, Clark W, Strobeck C (2000) Development and
characterization of microsatellite loci from lynx (Lynx
canadensis) and their use in other fields. Mol. Ecol., 9, 2197–
2198.
Creel S, Spong G, Sands JL, Rotella J, Zeigle J, Joe L, Murphy
KM, Smith D (2003) Population size estimation in Yellow-
stone wolves with error-prone noninvasive microsatellite
genotypes. Mol. Ecol.,12, 2003–2009.
Dowling TE, Secor CL (1996) The role of hybridization and
introgression in the diversification of animals. Annu. Rev.
Ecol. Syst.,28, 593–619.
354

Ernest HB, Penedo MCT, May BP, Syvanen M, Boyce WM
(2000) Molecular tracking of mountain lions in the Yosemite
Valley region in California: Genetic analysis using micro-
satellites and faecal DNA. Mol. Ecol.,9, 433–441.
Evett IW, Weir BS (1998) Interpreting DNA Evidence:Statis-
tical Genetics for Forensic Scientists. Sinauer Associates,
Sunderland, MA, USA.
Federal Register (2000) Endangered and Threatened Wildlife
and Plants; Determination of Threatened Status for the
Contiguous US Distinct Population Segments of the Canada
Lynx and Related Rule; Final Rule. Vol. 65, pp. 16051–
16086.
Federal Register (2003) Endangered and Threatened Wildlife
and Plants; Reopening of Comment Period for Final Rule to
list the contiguous United States Distinct Population Seg-
ment of the Canada lynx. Vol. 68, pp. 12611–12612.
Forbes SH, Allendorf FW (1991) Associations between mito-
chondrial and nuclear genotypes in cutthroat trout hybrid
swarms. Evolution,45, 1332–1349.
Gagneux P, Boesch C, Woodruff DS (1997) Microsatellite
scoring errors associated with noninvasive genotyping based
on nuclear DNA amplified from shed hair. Mol. Ecol.,6,
861–868.
Goodman SJ, Barton NH, Swanson G, Abernethy K, Pem-
berton JM (1999) Introgression through rare hybridization:
A genetic study of a hybrid zone between red and sika
deer (Genus Cervus) in Argyll, Scotland. Genetics,152, 355–
371.
Goossens B, Waits LP, Taberlet P (1998) Plucked hair samples
as a source of DNA: Reliability of dinucleotide microsatellite
genotyping. Mol. Ecol., 7, 1237–1241.
Johnson W, O’Brien SJ (1997) Phylogenetic reconstruction of
the Felidae using 16S rRNA and NADH-5 mitochondrial
genes. J. Mol. Evol., 44, S98–S116.
Leary RF, Allendorf FW, Forbes SH (1993) Conservation
genetics of bull trout in the Columbia and Klamath River
Drainages. Conserv. Biol.,7, 856–865.
Mayr E (1972) Sexual selection and natural selection. In: Sexual
Selection and the Descent of Man (ed. Campbell B.), pp. 87–
104. Heinemann, London, England.
McKelvey KS, Aubry KB, Ortega YK (2000) History and
distribution of lynx in the contiguous United States. In:
Ecology and Conservation of Lynx in the United States (eds.
Ruggiero LF et al.), pp. 207–264. University Press of Colo-
rado, Boulder, Colorado, USA.
Mills LS, Citta JJ, Lair KP, Schwartz MK, Tallmon DA (2000)
Estimating animal abundance using noninvasive DNA
sampling: Promise and pitfalls. Ecol. Appl.,10, 283–294.
Mills LS, Pilgrim K, Schwartz MK, McKelvey K (2001) Iden-
tifying lynx and other North American felids based on
MtDNA analysis. Conserv. Genet., 1, 285–289.
Mowat G, Paetkau D (2002) Estimating marten Martes amer-
icana population size using hair capture and genetic tagging.
Wildlife Biol.,8, 201–209.
Mowat G, Poole KG, O’Donoghue M (2000) Ecology of lynx
in Northern Canada and Alaska. In: Ecology and Conser-
vation of Lynx in the United States (eds. Ruggiero LF et al.),
pp. 265–306. University Press of Colorado, Boulder, Colo-
rado, USA.
Nei M (1987) Molecular Evolutionary Genetics. Columbia
University Press, New York, USA.
Paetkau D (2003) An empirical exploration of data quality in
DNA-based population inventories. Mol. Ecol.,12, 1375–
1387.
Rhymer JM, Simberloff D (1996) Extinction by hybridization
and introgression. Annu. Rev. Ecol. Syst.,27, 83–109.
Schwartz MK, Mills LS, McKelvey KS, Ruggiero LF, Allen-
dorf FW (2002) DNA reveals high dispersal synchronizing
the population dynamics of Canada lynx. Nature,415, 520–
522.
Schwartz MK, Mills LS, Ortega Y, Ruggiero LF, Allendorf
FW (2003) Landscape location affects genetic variation of
Canada lynx. Mol. Ecol.,12, 1807–1816.
Slough BG, Mowat G (1996) Population dynamics of lynx in a
refuge and interactions between harvested and unharvested
populations. J. Wildlife Manage.,60, 946–961.
Spruell P, Bartron M, Kanda N, Allendorf FW (2001) Detec-
tion of hybrids between bull trout (Salvelinus confluentus)
and brook trout (Salvelinus fontinalis) using PCR primers
complementary to interspersed nuclear elements. Copia,4,
1093–1099.
Taberlet P, Griffin S, Goossens B, Questiau S, Manceau V,
Escaravage N, Waits LP, Bouvet J (1996) Reliable geno-
typing of samples with very low DNA quantities using PCR.
Nucl. Acids Res.,26, 3189–3194.
Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic
sampling: Look before you leap. Trends Ecol. Evol.,14, 323–
327.
Tranah GJ, Bagley M, Agresti JJ, May B (2003) Development
of codominant markers for identifying species hybrids.
Conserv. Genet.,4, 537–541.
Waits JL, Leberg PL (2000) Biases associated with population
estimation using molecular tagging. Animal Conserv., 3, 191–
199.
Waits LP, Luikart G, Taberlet P (2001) Estimating the proba-
bility of identity among genotypes in natural populations:
Cautions and guidelines. Mol. Ecol.,10, 249–256.
Ward RMP, Krebs CJ (1985) Behavioral responses of lynx to
declining snowshoe hare abundance. Can. J. Zool.,63, 2817–
2824.
Werdelin L (1981) Evolution of lynxes. Ann. Zool. Fenn.,18,
37–71.
355