Genetics, evolution and phylogeography of moose

Abstract

Early studies of genetic variation in moose (Alces alces) indicated little variation. Recent studies have indicated higher levels of variation in nuclear markers; nonetheless, genetic heterogeneity of moose is relatively low compared with other mammals. Similarly, variation in mitochondrial DNA of moose is limited worldwide, indicating low historic effective population size and a common ancestry for moose within the last 60,000 years. That ancestor most likely lived in central Asia. Moose likely exhibit low levels of heterogeneity because of population bottlenecks in the late Pleistocene caused by latitudinal shifts in habitat from recurrent climate reversals. A northward movement of boreal forest associated with the end of the last ice age facilitated the northward advance of Asian populations and colonization of the New World, which occurred as a single entry by relatively few moose immediately prior to the last flooding of the Bering land bridge. Despite suffering serial population bottlenecks historically, moose have exhibited a notable ability to adapt to a changing environment, indicating that limited neutral genetic variation may not indicate limited adaptive genetic variation. We conclude that morphological variation among moose worldwide occurred within a few thousand years and indicates that moose underwent episodes of rapid and occasionally convergent evolution. Genetic change in moose populations over very short time scales (tens or hundreds of years) is possible under harvest management regimes and those changes may not be beneficial to moose in the long term. Modeling exercises have demonstrated that harvest strategies can have negative consequences on neutral genetic variation as well as alleles underpinning fitness traits. Biologists should consider such outcomes when evaluating management options.

Full text

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
103
GENETICS, EVOLUTION, AND PHYLOGEOGRAPHY OF MOOSE
Kris J. Hundertmark1 and R. Terry Bowyer2
1Institute of Arctic Biology and Department of Biology and Wildlife, University of Alaska Fairbanks,
Fairbanks, AK 99775, USA; 2Department of Biological Sciences, Idaho State University, Pocatello,
ID 83209, USA
ABSTRACT: Early studies of genetic variation in moose (Alces alces) indicated little variation.
Recent studies have indicated higher levels of variation in nuclear markers; nonetheless, genetic
heterogeneity of moose is relatively low compared with other mammals. Similarly, variation in
mitochondrial DNA of moose is limited worldwide, indicating low historic effective population size
and a common ancestry for moose within the last 60,000 years. That ancestor most likely lived in
central Asia. Moose likely exhibit low levels of heterogeneity because of population bottlenecks
in the late Pleistocene caused by latitudinal shifts in habitat from recurrent climate reversals. A
northward movement of boreal forest associated with the end of the last ice age facilitated the
northward advance of Asian populations and colonization of the New World, which occurred as
a single entry by relatively few moose immediately prior to the last flooding of the Bering land bridge.
Despite suffering serial population bottlenecks historically, moose have exhibited a notable ability
to adapt to a changing environment, indicating that limited neutral genetic variation may not indicate
limited adaptive genetic variation. We conclude that morphological variation among moose
worldwide occurred within a few thousand years and indicates that moose underwent episodes of
rapid and occasionally convergent evolution. Genetic change in moose populations over very short
time scales (tens or hundreds of years) is possible under harvest management regimes and those
changes may not be beneficial to moose in the long term. Modeling exercises have demonstrated
that harvest strategies can have negative consequences on neutral genetic variation as well as
alleles underpinning fitness traits. Biologists should consider such outcomes when evaluating
management options.
ALCES VOL. 40: 103-122 (2004)
Key words: adaptation, Alces, convergent evolution, moose, mtDNA, phylogeography, Pleistocene,
range expansion
Genetics have long had a central role in
biological investigations, and provide ana-
lytical tools that are applicable across a
broad spectrum of investigation. For in-
stance, genetic analysis can provide insights
into such diverse investigations as evolu-
tionary histories of species (Avise et al.
1987, 2000), interactions and relationships
among populations (Blundell et al. 2002) or
individuals (Quellar et al. 1993), evaluation
of the success of specific management ac-
tions (Vernesi et al. 2002), population and
behavioral ecology (Scribner and Chesser
2001), and food habits (Symondson 2002).
Recent advances in collection and analysis
of genetic data have facilitated more re-
fined approaches to evolutionary and popu-
lation genetic questions, and our under-
standing of moose biology has benefited as
a result of those advances.
Evolution and taxonomy of moose (Alces
alces) have been reviewed previously
(Peterson 1955; Groves and Grubb 1987;
Geist, 1987a,b, 1998; Sher 1987; Lister 1993;
Guthrie 1995; Bubenik 1998; Bowyer et al.
2003) and have encompassed aspects of
behavior, morphology, paleontology, and
genetics, but no review has dealt specifi-

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
104
cally with genetics. In this review, our goal
is to provide an overview of older studies
while focusing on recent advances in genet-
ics and phylogeography (see definition in
Appendix 1) of moose and the insights they
provide.
The broad scope of genetic and evolu-
tionary investigations in species biology
would make a complete review of all studies
disjointed. Yet, approximately half of all
published studies of moose genetics have
been published since the comprehensive
treatise “Ecology and Management of the
North American Moose” (Franzmann and
Schwartz 1998) was compiled; thus, we
were compelled to present the most com-
plete review possible. In an effort to present
a cogent summary of all relevant studies,
we have divided this review into 3 parts: (1)
assessing genetic diversity, where we re-
view the different types of markers exam-
ined in studies of moose genetics and the
conclusions drawn from those studies; (2)
moose evolution and phylogeography, where
we examine the evolutionary descent of
moose and processes that have shaped the
genetic variation and structure observed
today; and (3) genetic effects of harvest,
which reviews a small but important body of
work composed of management-based
modeling that examined effects of various
harvest regimes on population and genetic
measures.
ASSESSING GENETIC DIVERSITY
One potential difficulty in discussing
genetic analyses is the use of specialized
terminology. To avoid uncertainty and en-
hance understanding, we provide a brief
glossary of terms used in this review (Ap-
pendix 1). Terms defined in the glossary
are highlighted in bold in the text at their first
usage.
The Allozyme Era
First reports of genetic investigations of
moose were published by Braend (1962,
cited by Gyllensten et al. 1980), Nadler et
al. (1967), and Shubin (1969, cited by
Gyllensten et al. 1980), wherein those au-
thors examined electrophoretic variation in
proteins from blood serum; no variation in
those genetic markers occurred in Scandi-
navia, North America, and central Russia,
respectively. The first study to report ge-
netic variability was Ryman et al. (1977),
who examined 1,384 moose from 3 areas of
Sweden for polymorphism at 23 allozyme
loci. That study reported only 1 locus to be
polymorphic, however, and only in 1 region.
Although they suspected that allele fre-
quencies varied geographically within the 1
variable region, the difference was not sta-
tistically significant. Those authors con-
cluded that genetic drift associated with a
severe population bottleneck (reduction in
population size) in Sweden in the 19th cen-
tury was a probable cause of the observed
lack of diversity. Wilhelmson et al. (1978),
examining variation in serum proteins, noted
no differences between Canadian and Eu-
ropean moose. From that evidence, they
concluded that moose populations on sepa-
rate continents had not undergone signifi-
cant genetic drift despite being separated
for thousands of years, implying that effec-
tive population sizes of moose populations
historically had been large.
Wilhelmson et al. (1978) also proposed
that historic population bottlenecks in Swe-
den had not been severe enough to have had
an effect on genetic diversity of moose.
Gyllensten et al. (1980) conducted exten-
sive screening of a transferrin locus from
moose across Fennoscandia and detected a
single polymorphism occurring in Norway,
Sweden, and Finland. Nonetheless, the poly-
morphism was present in only 6 of 16
populations and the uncommon allele never
exceeded 6% in any population. The au-
thors presented differences in frequency of
the uncommon allele as evidence of differ-

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
105
entiation of populations geographically, sup-
porting observations of Ryman et al. (1977).
Reliance on the occurrence of a single rare
polymorphism to demonstrate population
subdivision, however, is tenuous at best.
Those early studies and others created
an impression among some biologists that
certain species, including moose, possessed
little genetic variation across the genome.
Hypotheses explaining this in evolutionary
terms were proposed. Selander and
Kaufman (1973) proposed the environmen-
tal-grain hypothesis, which stated that large,
highly mobile animals exhibited less genetic
variability than small, sedentary species.
That hypothesis further stated that highly
mobile mammals would exhibit greater ho-
mogeneity across large areas than more
sedentary forms. Other hypotheses pro-
posed that r-strategists were less variable
than K-strategists (Harrington 1985), ge-
netic heterogeneity was greater in species
inhabiting broad arrays of habitats com-
pared with habitat specialists (Nevo 1978),
or that northern cervids inhabiting boreal
forests were less variable than their rela-
tives to the south (Smith et al. 1990).
Analyses of 23 allozyme loci in > 700
individuals representing 18 moose
populations in Scandinavia were required to
reveal extensive genetic variation in moose
(Ryman et al. 1980). Those authors refuted
the environmental-grain hypothesis, con-
cluding that large mammals in general, and
large cervids in particular, are not naturally
monomorphic; previous studies of moose
had examined too few loci or individuals to
detect variation. Nonetheless, those au-
thors noted that genetic variability in moose
was somewhat less than that observed in
many other species of mammals (Nevo
1978), but that genetic drift due to small
historic population size was a more likely
explanation than any specific evolutionary
strategy. Thus, genetic drift was once again
proposed as being an important factor in
determining the structure of genetic diver-
sity. In another comprehensive study,
Chesser et al. (1982) examined 1,169 indi-
viduals from 4 regions in Sweden for a
single polymorphic locus and reported vari-
ation in allele frequencies among those re-
gions and, perhaps more importantly, sig-
nificant variation within 1 of those regions.
Detecting variation at geographic scales
small enough to be considered within a
single population illustrated that structure of
moose populations existed at scales smaller
than previously imagined, and that vagility
was not inconsistent with genetic structur-
ing.The most recent study of allozyme vari-
ation in moose reported extensive variation
in a moose population in Alaska
(Hundertmark et al. 1992). The level of
genetic diversity observed was greater than
that reported elsewhere in moose and was
similar to levels observed in white-tailed
deer, a species known for extensive allozyme
diversity (Smith et al. 1984). Hundertmark
et al. (1992) hypothesized that lesser levels
of variability described in moose from Scan-
dinavia and other regions of North America
were attributable to glacial history. All
other moose populations studied to that point
occurred in previously glaciated terrain that
was colonized by moose after retreat of
Pleistocene ice sheets. Colonization of
previously glaciated areas could have re-
sulted in serial founder events that reduced
genetic diversity (Sage and Wolff 1986).
Hundertmark et al. (1992) argued that
Alaska could have served as a refugium for
moose in which genetic diversity could have
been maintained because of a large effec-
tive population size.
Assessing Variation at the Sequence
Level
The first investigation of highly poly-
morphic molecular markers in moose docu-
mented 4 alleles in a microsatellite-like

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
106
locus in 17 individuals from Sweden (Ellegren
et al. 1991). Eight genotypes were re-
ported, which represented a heretofore un-
thinkable level of polymorphism. Those
authors investigated the inheritance of that
locus in a 2-generation pedigree and deter-
mined that the alleles exhibited Mendelian
inheritance. That study and others like it
laid the foundation for the explosion of
interest in population genetics and
phylogenetics based on molecular markers
and the polymerase chain reaction (PCR;
Mullis et al. 1986).
The advent of PCR represented one of
the truly significant advances in the history
of molecular biology and genetics. The
process allows in-vitro amplification of DNA
from miniscule amounts of starting material
(theoretically as little as 1 molecule of DNA)
to provide sufficient quantities for analysis.
No longer were researchers required to
sacrifice animals to acquire sufficient quan-
tities and types of tissues for genetic analy-
ses because any nucleated cell held the
complete genetic complement of the indi-
vidual. PCR offered unsurpassed access to
the genome, and researchers soon applied
that to study genetics of moose.
Mikko and Andersson (1995) conducted
the first analysis of functional loci in an
analysis of variation in the major histocom-
patibility complex (MHC) in moose from
Sweden and Canada. The MHC is a family
of genes important in immune system func-
tion, and low levels of diversity of MHC
alleles have been interpreted as indicators
of lost evolutionary potential and increased
susceptibility to pathogens (Hedrick 1994).
Mikko and Andersson (1995) noted very
low levels of MHC variation in both Swed-
ish and Canadian moose. Moreover, those
authors documented similarity among alleles
between continents and inferred the exist-
ence of a bottleneck in an ancient moose
lineage prior to divergence of European and
Canadian lineages. Mikko and Andersson
(1995) applied a molecular clock to DNA
sequence variation in the control region
of mitochondrial DNA (mtDNA) to date
the time of divergence of Swedish and
Canadian moose, which they estimated at
165,000-350,000 years ago.
The time since divergence of European
moose also was analyzed by Ellegren et al.
(1996). They assessed variation in
minisatellite loci of Swedish moose and
concluded that normal evolutionary proc-
esses could have generated the amount of
variation observed within 10,000-50,000
years after a severe bottleneck. Their
implication, therefore, was that the estimate
of divergence provided by Mikko and
Andersson (1995) was too old by perhaps 1
order of magnitude. Surprisingly, the two
data sets are not inconsistent; indeed, they
show similar levels of variation considering
differences in evolutionary rates among
marker types. The differences in estimates
for date of divergence relate more to the
evolutionary rate estimates used than to
differences in genetic variability.
Microsatellite loci were first described
for moose by Wilson et al. (1997) and Røed
and Midthjell (1998). Broders et al. (1999)
demonstrated the utility of microsatellites
for assessing population structure in moose
by assessing consequences of founder
events in Canada. Heterozygosity in 3
populations founded by few individuals de-
creased from 14-30% compared with the
source population. Broders et al. (1999), in
assessing variability of moose on the island
of Newfoundland, Canada, demonstrated
that 2 consecutive founder events reduced
heterozygosity by 46%. Although they could
not discern any decrease in fitness as a
result of the decrease in diversity, the au-
thors questioned the long-term viability of
those moose populations. Nonetheless, lev-
els of diversity in neutral microsatellite loci
were not indicative of diversity in functional
loci in moose (Wilson et al. 2003).

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
107
The Future
Microsatellites have replaced allozymes
as the most widely used molecular marker
for assessing nuclear genetic diversity, and
will be the choice of geneticists for the
foreseeable future (Bruford et al. 1996).
There are problems, however, with analysis
of microsatellites because the ways in which
they mutate into new forms are not entirely
understood (Hancock 1999). In the future,
a new type of analysis called single nucle-
otide polymorphism (SNP, pronounced
“snip”) may replace microsatellites for some
applications (Fries and Durstewitz 2001,
Brumfield et al. 2003). This new technol-
ogy allows a single nucleotide site to be
queried for presence of a particular nucle-
otide and presence or absence can be con-
verted to a binary code. Current technology
(so-called “real-time PCR” and “DNA
chips”) allows for fast and accurate exami-
nation of many individuals and SNPs, but
we must await the development and char-
acterization of marker loci before broad
application of this new family of molecular
markers can be considered.
MOOSE EVOLUTION AND
PHYLOGEOGRAPHY
Origins of Modern Moose
Moose (Alces alces) are a young spe-
cies in the evolutionary scheme of large
mammals. The genus Alces first appears in
the fossil record 2 million years ago
(Thouveny and Bonifay 1984) and fossils
attributable to A. alces are first recorded
approximately 100,000 years ago (Lister
1993). Those dates are very recent consid-
ering that the subfamily Odocoileinae, to
which moose belong, diverged from other
deer lineages 9-12 million years ago
(Miyamoto et al. 1990).
Paleontological evidence indicated Eu-
rope as the place of origin of the genus
Alces (Lister 1993). The genus never was
diverse, with only one species present in the
fossil record at any particular time. Yet, the
species assumed to be the precursor to A.
alces, the broad-fronted moose (A.
latifrons) was distributed across Eurasia
and into northwestern North America for a
time before becoming extinct in Beringia at
the end of the Pleistocene (Guthrie 1995).
Thus, the widespread distribution of modern
moose and its immediate ancestor indicate
a degree of evolutionary success despite a
paucity of species diversity.
Up to 8 subspecies of moose are recog-
nized worldwide (Fig. 1); 4 in Eurasia and 4
in North America (Peterson 1955). That
number is open to question, however. Geist
(1987a, 1998) contends that there are 2
predominant types of moose in the world:
American and European, following the con-
vention of Flerov (1952). To the former
type he assigns all North American moose
as well as eastern Asian subspecies A. a.
burturlini and A. a. pfizenmayeri. He
based his opinion primarily on morphology
and noted similar geographic divisions in
taxonomy among reindeer and caribou
(Rangifer tarandus) and red deer and North
American elk (Cervus elaphus; Geist 1998).
He further contended that those morpho-
logical types should correspond to subspe-
cies designations. Therefore, Geist (1998)
recognized A. a. alces of Europe and A. a.
americana in eastern Asia and North
America. He also suggested that A. a.
americanus has precedence under
nomenclatural conventions as the proper
name for the east Asian-North American
subspecies. He referred to A. a.
cameloides in northern China, Mongolia,
and southeastern Russia as part of a primi-
tive fauna native to that region and recog-
nized that subspecies as a valid taxon al-
though he also refers to it as an American-
type moose (Geist 1998:230).
The 2-types hypothesis is supported to
some degree by karyotype (Boeskorov 1996,
1997) and some data on mtDNA (Mikko

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
108
Fig. 1. Approximate ranges of 8 subspecies of moose worldwide. A. a. a. = A. a. alces, A. a. p. = A.
a. pfizenmayeri, A. a. c. = A. a. cameloides, A. a. b. = A. a. burturlini, A. a. g. = A. a. gigas, A. a.
an. = A. a. andersoni, A. a. s. = A. a. shirasi, A. a. am. = A. a. americana, * = introduced population
in Newfoundland.
and Andersson 1995). Most Eurasian moose
have a karyotype of 2N = 68, whereas
North American moose have 2N = 70. That
difference derives from a Robertsonian
translocation of 2 acrocentric chromosomes
into a single metacentric chromosome or
vice versa. Although that chromosomal
polymorphism originally was thought to sepa-
rate Eurasian and North American moose
(Groves and Grubb 1987), the 2N = 70 form
was recently discovered in eastern Asia
(Boeskorov 1996, 1997). Similarly, a length
mutation (insertion-deletion, or indel) within
the control region of mtDNA originally was
described as discriminating between North
American and European moose (Mikko and
Andersson 1995), but subsequent investiga-
tions documented that indel in moose from
Eastern Asia (Hundertmark et al. 2002b,
Udina et al. 2002). Although the precise
geographic distributions of those
polymorphisms in karyotype and mtDNA

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
109
length are not well described, they seem to
correspond geographically with a zone of
intergradation in east-central Asia, similar
to that proposed for American and Eurasian
types of moose (Flerov 1952, Geist 1998).
More work is needed to determine the ex-
tent of that geographic correspondence and
to determine if it coincides with subspecies
boundaries. Moreover, the question of re-
productive viability of the two chromosomal
races must be addressed. Indeed, Boeskorov
(1997) has proposed that the chromosomal
races are different species and Groves and
Grubb (1987) have identified them as “semi-
species.” We caution, however, that chro-
mosome numbers may be a poor designator
of species among large mammals (Bowyer
et al. 2000).
Hundertmark et al. (2002b) tested the
2-types hypothesis by examining the distri-
bution of genetic variance of mtDNA among
and within different hypothetical population
structures. Those authors sampled moose
throughout their worldwide distribution and
arranged them into either 2 groups (corre-
sponding to the 2-types hypothesis) or 3
groups corresponding to continent of origin
(Asia, Europe, and North America). They
then examined the distribution of genetic
variance within and among or between
groups, predicting that the correct structure
would minimize within-group variation and
maximize among-group variation. Percent-
age of total variation observed among the 3
groups was slightly greater than variation
between the 2 groups (61.5% vs. 58.1%),
and variation among populations within
groups was minimized in the 3-group com-
parison (21.6% vs. 28.8% in the 2-group
comparison). Therefore, Hundertmark et
al. (2002b) concluded that mtDNA data
provided no support for a 2-type over a 3-
type hypothesis. That finding can be visu-
alized by a phylogenetic tree constructed
from haplotypes originally reported by
Hundertmark et al. (2002b, 2003), which
shows North American moose as distinct
from both Asian and European forms
(Fig. 2).
Ancient Bottlenecks and the Mother of
All Moose
Moose worldwide exhibit little varia-
tion in a fragment of the mitochondrial cyto-
chrome-b gene (Hundertmark et al. 2002a).
Cytochrome b is useful for constructing
mammalian phylogenies (Irwin et al. 1991)
and a paucity of variation in moose indi-
cated a recent common ancestry likely due
to a severe bottleneck that affected all
extant lineages (Hundertmark et al. 2002a),
which is in agreement with the findings of
Mikko and Andersson (1995). Analysis of
variation within the mitochondrial control
region, which evolves at a much faster rate
than cytochrome b (Lopez et al. 1997), was
Fig. 2. Unrooted phylogenetic tree of moose
populations and subspecies worldwide, with
the exception of A. a. pfizenmayeri, using FST
as the distance measure [data from
Hundertmark et al. (2002b, 2003)]. Note the
distinct positions of the 3 Eurasian subspe-
cies and North American moose, which do not
support an hypothesis of 2 or 3 races of moose
worldwide.

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
110
Fig. 3. A phylogenetic tree of haplotypes of the
mitochondrial control region of moose. Sym-
bols indicate region of origin, with black sym-
bols indicating Asian origin. Distinct clades
or phylogroups are indicated on the right.
From Hundertmark et al. (2002b).
necessary to reveal significant levels of
variation in moose and subsequently geo-
graphic patterns were revealed.
Control region haplotypes of moose
can be divided into 3 clades, or haplogroups
(Fig. 3). The basal haplogroup (i.e., the
group that diverges first from the base of
the tree) is entirely Asian, which suggests
that those are the oldest moose haplotypes.
The two other haplogroups are primarily
European and primarily North American,
although some Asian haplotypes occur in
both. The distribution of haplogroups on a
worldwide scale illustrates the age of con-
tinental assemblages of haplotypes (Fig. 4).
The Yakutia area contains moose from all 3
haplogroups. The Russian Far East con-
tains both European and Asian haplogroups,
but not North American, and both Europe
and North America contain only 1
haplogroup each. Therefore, Yakutia can
be identified as the area from which all
extant moose lineages were derived, i.e., it
is the oldest extant moose population that
has been sampled.
Yakutia probably was the center of a
single moose population during the last ice
age, or at least was the location of the only
population to provide descendants of mod-
ern moose. Moose would have been re-
stricted in their distribution because the
cooler climate in Asia at that time would
have resulted in a shift of boreal forest
habitat to the south. That habitat could have
shifted only so far southward, because of
prominent mountains running east-west,
which would have formed an effective bar-
rier to further movement to the south (Hewitt
1996).
During the last ice age, there were 2
periods of maximum glacial advance, termed
the lower and upper pleniglacials. Those
episodes occurred at approximately 62,000
and 20,000 years ago, respectively (Fulton
et al. 1986). Boreal forest habitats in Asia
would have shifted to the south during those
cooling episodes and would have been com-
pressed against the northern slopes of moun-
Fig. 4. Distribution of mitochondrial haplogroups
worldwide. Note that moose from the Yakutia
area have the most diverse composition and
that moose from North America do not share
haplogroups with moose from Russian Far
East.

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
111
59,000 and 14,000 years ago. When expan-
sion times of moose are overlaid on a profile
of global temperature change for the last ice
age (Jouzel et al. 1987), population expan-
sion is correlated with warming trends fol-
lowing the pleniglacials (Fig. 6). Conse-
quently, the evolution and geographic distri-
bution of moose seems to have been af-
fected substantially by climate change, par-
ticularly climate reversals associated with
the late Pleistocene and early Holocene.
Coming to America
Cronin (1992) analyzed subspecific vari-
ation in North American cervids using re-
striction fragment length polymorphisms
(RFLP) of mtDNA. Despite analyzing 32
moose sampled from different regions of
North America, he documented no varia-
tion among haplotypes from that continent.
In comparison to other North American
cervids, the lack of variation among subspe-
cies of moose was interpreted as an indica-
Fig. 5. Glacial coverage during the last glacial
maximum, superimposed on a map of present-
day sea level. Note that the Bering land bridge
between North America and Asia would have
been exposed during the glacial maximum due
to lower sea levels. Names of major ice sheets
are provided.
tain ranges. During subsequent warming
intervals, moose habitat would have spread
to the north, allowing moose populations to
expand (Guthrie 1995). Unlike North
America, much of Eurasia was not glaci-
ated during those periods (Arkhipov et al.
1986), providing potential dispersal routes
across the continent (Fig. 5). The process
of latitudinal shifts of range associated with
episodes of climate change results in de-
creases of existing genetic diversity (Hewitt
1996) and leaves characteristic signatures
in the genome of modern moose.
Moose in Eurasia underwent 2 distinct,
recent population expansions (Hundertmark
et al. 2002b). Any other historic population
processes preceding those expansions are
not detectable because low population sizes
eliminated much of the genetic variation
present in the pre-bottleneck populations,
and hence no signature from those times
exists in the present genome. By applying a
molecular clock to those expansion data,
Hundertmark et al. (2002b) estimated that
the expansions occurred approximately
160 140 120 100 80 60 40 20 0
2
0
-2
-4
-6
-8
-10
Thousands of years ago
Temperature di
f
f
erence
(
C
º
)
Fig. 6. Representation of mean global tempera-
tures during the last 160,000 years relative to
mean temperature in 1900. Negative tempera-
ture differences indicate periods colder than
today. The 2 glacial maxima of the last ice age
are indicated by arrows. Estimated dates of
moose population expansion are indicated by
dashed lines and correspond to periods of
warming associated with glacial retreat and
northward advance of the boreal forest in
Eurasia. Temperature profile adapted from
http://gcrio.ciesin.org/CONSEQUNCES/win-
ter96/article1-fig3.html.

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
112
ity in composition of mtDNA haplotypes.
As noted previously, North American sub-
species are distinct from European and
Asian subspecies (Fig. 2) and none of the
Asian haplotypes in the North American
haplogroup were found in the Russian Far
East. All haplotypes found in the Russian
Far East (Magadan Oblast) are restricted to
haplogroup 2 (Fig. 3), which contains all
European haplotypes. A likely colonization
scenario entails closely related moose from
central Asia colonizing both Europe and
North America.
Lack of genetic similarity between
moose in Alaska and the Russian Far East
is inconsistent with the scenario proposed
by Hundertmark et al. (2002b) concerning a
colonizing wave of moose traveling from
Asia to North America through Beringia. A
single wave of moose colonizing North
America through Beringia would have left
genetically similar populations on either side
of the Bering Strait after the Bering land
bridge flooded. Yet, ages of subfossil re-
mains of moose from North America indi-
cate clearly that A. alces was present in
Alaska prior to anywhere else on the conti-
nent (Guthrie 1990, Hundertmark et al. 2003),
unequivocally supporting an entry through
Beringia. One hypothesis that accounts for
this seeming paradox is that 1 or both of
those populations underwent population bot-
tlenecks shortly after the colonization of
North America, and have only recently
reestablished populations in those areas,
leading to the presence of different genetic
lineages in each. Moose in the Russian Far
East show evidence of an expansion ap-
proximately 1,200 years ago (Hundertmark
et al. 2002b) and continue to expand their
range toward the Bering Sea (Zheleznov
1993).
Similarly, Alaskan moose show a sur-
prising lack of mitochondrial diversity com-
pared with moose elsewhere on the conti-
nent (Hundertmark et al. 2003), which is
tion of a recent common ancestry, consist-
ent with colonization of the continent after
the retreat of ice sheets from the last ice
age (Cronin 1992). Similarly, no variation
was detected within a fragment of cyto-
chrome b in North American moose com-
pared with slight variation in Eurasia
(Hundertmark et al. 2002a).
Moose populations in North America
were established as a result of a single entry
into the continent, and that entry occurred
during the population expansion of moose
14,000 years ago at the end of the last ice
age, just before the closure of the Bering
land bridge (Guthrie 1995, Hundertmark et
al. 2002b). A recent entry into North
America is the only conclusion that is con-
sistent with limited variation in the mtDNA
control region both within North America
and between North America and Eurasia.
If moose had existed in 4 separate refugia in
North America during the last ice age, as
suggested by Peterson (1955), or had en-
tered North America from Asia more than
once, a more distant common ancestor
would be indicated. The timing of the entry
is supported not only by genetic data but
also by the distribution of suitable moose
habitat at the end of the last ice age. The
cold, dry grassland habitat that prevailed in
Beringia for most of the last ice age was
unsuitable for moose and was replaced by
boreal forest only within the last 14,000
years (Guthrie 1995).
Evidence from mtDNA variation also
indicates that North American moose did
not originate in Beringia, as some have
speculated (Cronin 1992, Geist 1998), or
recolonize the Russian Far East from North
America (Coady 1982). Moose in North
America are not closely related to moose on
the western side of the Bering Strait (Rus-
sian Far East). If those moose were once
part of the same population recently sepa-
rated by the flooding of the Bering land
bridge, we would still expect to find similar-

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
113
indicative of a bottleneck notwithstanding
the moderate levels of allozyme diversity
reported for Alaskan moose by Hundertmark
et al. (1992). Diversity of mtDNA can be
reduced to a much greater degree by a
bottleneck than diversity of nuclear DNA
because mtDNA is 4 times more sensitive
to genetic drift due to its haploid, uniparental
mode of inheritance (Birky et al. 1983).
Simultaneous bottlenecks in moose from
both sides of the Bering Strait suggest a
widespread causal factor. Recent studies
have found evidence of significant biotic
effects of climate reversals in Beringia af-
ter flooding of the Bering land bridge (Elias
2000, Mason et al. 2001, Anderson et al.
2002). Those effects offer an intriguing
mechanism for bottlenecks in Beringian
moose populations.
The greatest variation in mtDNA in
North American moose occurs within the
range of A. a. andersoni (Hundertmark et
al. 2003). Alces a. shirasi from Colorado
exhibited no diversity and A. a. gigas and A.
a. americana exhibited very little diversity.
If the paucity of mitochondrial diversity in
Alaska is due to a bottleneck and recent
expansion, those data would be consistent
with the serial-founder-events hypothesis
of North American colonization
(Hundertmark et al. 1992).
The pattern of colonization of North
America undoubtedly was influenced by
the retreating glaciers and may have had
some effect on genetic structure
(Hundertmark et al. 2003). Based on the
reconstruction of the retreat of the
Laurentide ice sheet by Dyke and Prest
(1987), we offer the following scenario of
colonization. At the last glacial maximum,
the Cordilleran and Laurentide ice sheets
created an effective barrier between east-
ern Beringia (Alaska) and other parts of the
continent (Fig. 5). As glaciers retreated, a
corridor opened on the eastern slopes of the
Rocky Mountains allowing passage to the
south. By 10,000 years ago, western Canada
was ice-free but central and eastern Canada
remained covered by the Laurentide ice
sheet and large proglacial lakes (Fig. 7).
Passage to eastern Canada north of the
Great Lakes was impossible at this time and
the only dispersal corridor was south of the
lakes. By 8,400 years ago, moose arriving
in the eastern continent could have dis-
persed westward north of the lakes, skirting
the southern shores of the proglacial lakes
to the north, and come into secondary con-
tact with moose from the west once the
proglacial lakes receded. Moose in the
eastern part of the continent (A. a.
americana) would have been on the end of
a series of founder events, explaining their
low mitochondrial variability and the pres-
ence of a contact zone between A. a.
americana and the much more variable A.
a. andersoni in Ontario between the Great
Lakes and Hudson Bay.
Morphological Adaptation
Moose of the Pleistocene and those that
entered North America at the beginning of
the Holocene were significantly larger than
those living today, a trait shared with other
northern ruminants (Guthrie 1984). Guthrie
(1984) proposed that the reduction in body
size was an adaptation to changes in sea-
sonal forage availability that occurred as a
result of climate amelioration at the end of
the last ice age. The ability of moose to
respond to a rapidly changing environment
belies the relatively low levels of genetic
variation documented by the studies we
have reviewed and demonstrates that evo-
lutionary potential is not easily predicted
solely by genetic variability but ultimately is
determined by the presence of adaptive
genetic variation and heritability of traits
that improve fitness (Lynch 1996).
A general reduction in body size is not
the only change to occur since the coloniza-
tion of North America. Moose in

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
114
Fig. 7. Coverage of North America by the
Cordilleran and Laurentide ice sheets and
proglacial lakes at 14,000, 10,000, and 8,400
years ago (adapted from Dyke and Prest 1987).
North America exhibit many differences in
behavior and morphology. Alaskan moose
are perhaps the most divergent; they exhibit
a degree of sociality not observed else-
where (Molvar and Bowyer 1994) and have
more distinctive body markings, also indica-
tive of increased sociality (Bowyer et al.
1991). Molvar and Bowyer (1994) sug-
gested that moose in Alaska have evolved
sociality recently as a response to living in
open environments. Adaptation to open
environments also applies to their mating
system, which is harem-based. Harem
mating is adaptive in open environments
(Hirth 1977) where a male can protect a
harem from competitors. Moose elsewhere
in North America exhibit a tending-bond
system of mating, which is adaptive for
forested environments.
Moose in Alaska and Siberia exhibit the
largest body size of moose in North America
and Eurasia, respectively. The similar ap-
pearance of moose occurring on either side
of the Bering Strait has caused some inves-
tigators to consider them the same subspe-
cies (e.g., Telfer 1984). As moose from
those 2 regions are not closely related
(Hundertmark et al. 2002a,b), their similar-
ity in size must result from convergent evo-
lution. Both subspecies have adapted to
open, northern habitats by increasing body
size. Adaptation to open habitats was dem-
onstrated with a multivariate analysis of
antler size among moose inhabiting differ-
ent areas and habitats in Alaska (Bowyer et
al. 2002). Those moose inhabiting open
habitats (tundra) tended to have larger ant-
lers overall than those living in boreal forest
(taiga). Similarly, moose occupying moun-
tainous habitat in the southern portions of
their range in North America (A. a. shirasi)
and Asia (A. a. cameloides) are similar;
exhibiting small body and antler size (Geist
1998). Bubenik (1998) explained that simi-
larity by proposing a second entry into North
America by Asian moose—an entry that
bypassed Beringia by traveling along the
southern coast of Alaska. Neither genetic
nor fossil data support that hypothesis
(Guthrie 1990, Hundertmark et al. 2002a,b,
2003).
GENETIC EFFECTS OF
MANAGEMENT
To this point we have discussed distri-
bution of genetic and morphological diver-
sity over space and time in the context of
genetic drift and selection for locally adap-
tive traits. Those patterns are developed
over relatively long periods of time and are
integral to the process of evolution. Genetic

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
115
change also can occur over short periods
due to human influences, notably harvest
management, and those changes may have
unintended and undesirable consequences
on individuals and populations (Harris et al.
2002) and therefore are important to recog-
nize.
Although it might be assumed that a
well-designed harvest plan acts in a random
fashion on genetic makeup of individuals, in
reality even a managed harvest can be a
highly selective force with measurable con-
sequences in just a few generations
(Coltman et al. 2003). Ryman et al. (1981)
modeled the effect of different harvest strat-
egies on 2 factors critical in determining the
extent of genetic drift and inbreeding in
populations: effective population size and
generation interval, respectively. Hunt-
ing strategies were defined by different
probabilities of harvest for both juveniles
and adult females, and resulted in stable
populations with decreased generation in-
tervals and effective sizes compared with
unhunted populations. Moreover, temporal
changes in genetic diversity differed for
different harvest strategies but always de-
creased. Thus, Ryman et al. (1981) demon-
strated that improperly designed harvest
regimes can affect genetic characteristics
of populations and by extension may have
an influence on evolutionary potential. Con-
versely, Cronin et al. (2001) detected no
differences in numbers of alleles or levels
of heterozygosity for 5 microsatellite loci
among 3 moose populations in Quebec, 1
heavily hunted, 1 lightly hunted, and 1 not
hunted.
Another critical factor in management
is the effect of harvest on genetic loci
underlying characters having a direct effect
on reproductive fitness, e.g., antler size in
moose. Controlling harvest by defining
legal males according to antler size is com-
mon in management of North American elk
(Thelen 1991) and is a strategy employed in
moose management in British Columbia,
Canada, and Alaska, USA (Child 1983,
Schwartz et al. 1992). In an effort to
evaluate genetic effects of the selective
harvest system in Alaska, Hundertmark et
al. (1993, 1998) modeled moose populations
subject to harvest strategies employing dif-
ferent definitions of legal males. They
concluded that selective harvest systems
could result in allele frequency changes at
loci coding for antler characteristics
(Hundertmark et al. 1993) and that the
position of a population relative to nutri-
tional carrying capacity of the habitat af-
fected the rate of change in allele frequen-
cies (Hundertmark et al. 1998). Those
results indicated that limiting harvest to
moose with large antlers could cause a
genetically based decrease in antler size
over time. Such a reduction in adaptive
genetic variation runs counter to general
conservation goals (Lynch 1996). A stun-
ning example of the effect of harvest on
fitness traits was recently reported by
Coltman et al. (2003), who documented
significant genetic effects on horn size and
body mass in bighorn sheep (Ovis
canadensis) as a result of selective harvest
of males with large horns.
Detecting potential changes in genetic
composition of moose as they respond to
various anthropogenic influences, whether
related to management or to changes in the
environment, is a difficult task. Nonethe-
less, it is an important area of investigation
and deserves attention. Modeling exercises
and studies of genetic variation have not
addressed interrelationships of moose
populations at fine scales. The effects of
management and habitat alteration on proc-
esses involved in maintenance of connec-
tivity of moose populations, such as male-
mediated gene flow via yearling dispersal
are extremely important and should be de-
scribed. Moose are highly adaptable ani-
mals, but the intensive management of moose

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
116
populations and the environmental factors
influencing their habitat (e.g., wildfire) may
have unintended and significant conse-
quences on the moose genome through a
change in selective forces. Proper conser-
vation of this species requires that we rec-
ognize and avoid that possibility.
ACKNOWLEDGEMENTS
We thank the organizers of the 5th Inter-
national Moose Symposium for the opportu-
nity to synthesize and present this informa-
tion. We wish to acknowledge funding sup-
port from Federal Aid in Wildlife Restora-
tion through the Alaska Department of Fish
and Game as well as the Institute of Arctic
Biology, University of Alaska Fairbanks. S.
Côté and an anonymous reviewer provided
constructive comments for improving the
manuscript.
REFERENCES
ANDERSON, P. M., A. V. LOZHKIN, and L. B.
BRUBAKER. 2002. Implications of a
24,000-yr palynological record for a
Younger Dryas cooling and for boreal
forest development in northeastern Si-
beria. Quaternary Research 57:325-
333.
ARKHIPOV, S. A., V. G. BESPALY, M. A.
FAUSTOVA, O. Y. GLUSHKOVA, L. L.
ISAEVA, and A. A. VELICHKO. 1986. Ice
sheet reconstructions. Pages 269-292
in V. Šibrava, D. Q. Bowen, and G. M.
Richmond, editors. Quaternary
Glaciations in the Northern Hemisphere.
Pergamon Press, Oxford, U.K.
AVISE, J. C. 2000. Phylogeography: the
History and Formation of Species.
Harvard University Press, Cambridge,
Massachusetts, USA.
_____, J. ARNOLD, R. M. BALL, JR., E.
BERMINGHAM, T. LAMB, J. E. NEIGEL, C.
A. REEB, and N. C. SAUNDERS. 1987.
Intraspecific phylogeography: the mito-
chondrial DNA bridge between popula-
tion genetics and systematics. Annual
Review of Ecology and Systematics
18:489-522.
BIRKY, C. W., T. MARUYAMA, and P. FUERST.
1983. An approach to population and
evolutionary genetic theory for genes in
mitochondria and chloroplasts, and some
results. Genetics 103:513-527.
BLUNDELL, G. M., M. BEN-DAVID, P. GROVES,
R. T. BOWYER, and E. GEFFEN. 2002.
Characteristics of sex-biased dispersal
and gene flow in coastal river otters:
implications for natural recolonization
of extirpated populations. Molecular
Ecology 11:289-303.
BOESKOROV, G. G. 1996. Karyotype of
moose (Alces alces L.) from north-
eastern Asia. Proceedings of the Rus-
sian Academy of Sciences 329:506-
508. (In Russian).
_____. 1997. Chromosomal differences in
moose. Genetika 33:974-978. (In Rus-
sian).
BOWYER, R. T., D. M. LESLIE, JR., and J. L.
RACHLOW. 2000. Dall’s and Stone’s
sheep. Pages 491-516 in S. Demarais
and P. R. Krausman, editors. Ecology
and Management of Large Mammals in
North America. Prentice Hall, Upper
Saddle River, New Jersey, USA.
_____, J. L. RACHLOW, V. VAN
BALLENBERGHE, and R. D. GUTHRIE. 1991.
Evolution of a rump patch in moose: an
hypothesis. Alces 27:12-23.
_____, K. M. STEWART, B. M. PIERCE, K. J.
HUNDERTMARK, and W. C. GASAWAY.
2002. Geographical variation in antler
morphology of Alaskan moose: puta-
tive effects of habitat and genetics.
Alces 38: 155-165.
_____, V. VAN BALLENBERGHE, and J.G.
KIE. 2003. Moose (Alces alces). Pages
931 - 964 in G.A. Feldhamer, B.
Thompson, and J. Chapman, editors.
Wild Mammals of North America: Biol-
ogy, Management, and Economics. Sec-

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
117
ond edition. The Johns Hopkins Uni-
versity Press, Baltimore, Maryland,
USA.
BRAEND, M. 1962. Studies on blood and
serum groups in the elk (Alces alces).
New York Academy of Sciences 97:296-
305.
BRODERS, H. G., S. P. MAHONEY, W. A.
MONTEVECCHI, and W. S. DAVIDSON.
1999. Population genetic structure and
the effect of founder events on the
genetic variability of moose, Alces
alces, in Canada. Molecular Ecology
8:1309-1315.
BRUFORD, M. W., D. J. CHEESMAN, T. COOTE,
H. A. A. GREEN, S. A. HAINES, C.
O’RYAN, and T. R. WILLIAMS. 1996.
Microsatellites and their application to
conservation genetics. Pages 278-297
in T. B. Smith and R. K. Wayne, edi-
tors. Molecular Genetic Approaches in
Conservation. Oxford University Press,
Oxford, U.K.
BRUMFIELD, R. T., P. BEERLI, D. A.
NICKERSON, and S. V. EDWARDS. 2003.
The utility of single nucleotide
polymorphisms in inferences of popula-
tion history. Trends in Ecology and
Evolution 18:249-256.
BUBENIK, A. B. 1998. Evolution, taxonomy
and morphophysiology. Pages 77-123
in A. W. Franzmann and C. C. Schwartz,
editors. Ecology and Management of
the North American Moose.
Smithsonian Institution Press, Wash-
ington, D.C., USA.
CHESSER, R. K., C. REUTERWALL, and N.
RYMAN. 1982. Genetic differentiation
of Scandinavian moose Alces alces
populations over short geographic dis-
tances. Oikos 39:125-130.
CHILD, K. N. 1983. Selective harvest of
moose in the Omineca: some prelimi-
nary results. Alces 19:162-177.
COADY, J. W. 1982. Moose. Pages 902-922
in J. A. Chapman and G. A. Feldhamer,
editors. Wild Mammals of North
America: Biology, Management, and
Economics. Johns Hopkins University
Press, Baltimore, Maryland, USA.
COLTMAN, D. W., P. O’DONOGHUE, J. T.
JORGENSON, J. T. HOGG, C. STROBECK,
and M. FESTA-BIANCHET. 2003. Unde-
sirable evolutionary consequences of
trophy hunting. Nature 426:655-658.
CRONIN, M. A. 1992. Intraspecific variation
in mitochondrial DNA of North Ameri-
can cervids. Journal of Mammalogy
73:70-82.
_____, J. C. PATTON, R. COURTOIS, and M.
CRÊTE. 2001. Genetic variation of
microsatellite DNA in moose in Québec.
Alces 37:175-187.
DYKE, A. S., and V. K. PREST. 1987. Late
Wisconsinan and Holocene history of
the Laurentide ice sheet. Geographie
Physique et Quaternaire 41:237-263.
ELIAS, S. A. 2000. Late Pleistocene cli-
mates of Beringia, based on analysis of
fossil beetles. Quaternary Research
53:229-235.
ELLEGREN, H., L. ANDERSSON, and K. WALLIN.
1991. DNA polymorphism in moose
(Alces alces) revealed by polynucle-
otide probe (TC)n. Journal of Heredity
82:429-431.
_____, S. MIKKO, K. WALLIN, and L.
ANDERSSON. 1996. Limited polymor-
phism at major histocompatibility com-
plex (MHC) loci in the Swedish moose
A. alces. Molecular Ecology 5:3-9.
FLEROV, K. K. 1952. Fauna of the USSR:
Mammals. Vol. 1 no 2. Musk deer and
deer. The Academy of Sciences of the
USSR, Leningrad. (English translation
by the Israel Program for Scientific
Translation, 1960, S. Monson, Jerusa-
lem).
FRANZMANN, A. W., and C. C. SCHWARTZ,
editors. 1998. Ecology and Manage-
ment of the North American Moose.
Smithsonian Institution Press, Wash-

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
118
ington, D.C., USA.
FRIES, R., and G. DURSTEWITZ. 2001.
DigitalDNA signatures: SNPs for ani-
mal tagging. Nature Biotechnology
19:508.
FULTON, R. J., M. M. FENTON, and N. W.
RUTTER. 1986. Summary of Quater-
nary stratigraphy and history, western
Canada. Pages 229-242 in V. Šibrava,
D. Q. Bowen, and G. M. Richmond,
editors. Quaternary Glaciations in the
Northern Hemisphere. Pergamon Press,
Oxford, U.K.
GEIST, V. 1987a. On the evolution and
adaptations of Alces. Swedish Wildlife
Research Supplement 1:11-23.
_____. 1987b. On speciation in Ice Age
mammals, with special reference to
cervids and caprids. Canadian Journal
of Zoology 65:1067-1084.
_____. 1998. Deer of the World: Their
Evolution, Behavior, and Ecology.
Stackpole Books, Mechanicsburg, Penn-
sylvania, USA.
GROVES, C. P., and P. GRUBB. 1987. Rela-
tionships of living deer. Pages 3-59 in
C. M. Wemmer, editor. Biology and
Management of the Cervidae.
Smithsonian Institution Press, Wash-
ington, D.C., USA.
GUTHRIE, R. D. 1984. Alaska megabucks,
megabulls, and megarams: the issue of
Pleistocene gigantism. Special Publi-
cations of the Carnegie Museum of
Natural History 8:482-509.
_____. 1990. New dates in Alaskan
Quaternary moose, Cervalces-Alces –
archaeological, evolutionary and eco-
logical implications. Current Research
in the Pleistocene 7:111-112.
_____. 1995. Mammalian evolution in
response to the Pleistocene-Holocene
transition and the break-up of the mam-
moth steppe: two case studies. Acta
Zoologica Cracoviensia 38:139-154.
GYLLENSTEN, U., C. REUTERWALL, N. RYMAN,
and G. STAHL. 1980. Geographical
variation in transferrin allele frequen-
cies in three deer species from Scandi-
navia. Hereditas 92:237-241.
HANCOCK, J. M. 1999. Microsatellites and
other simple sequences: genomic con-
text and mutational mechanisms. Pages
1-9 in D. B. Goldstin and C. Schlötterer,
editors. Microsatellites: Evolution and
Applications. Oxford University Press,
Oxford, U.K.
HARRINGTON, R. 1985. Evolution and distri-
bution of the Cervidae. Biology of deer
production. The Royal Society of New
Zealand, Bulletin 22:3-11.
HARRIS, R. B., W. A. WALL, and F. W.
ALLENDORF. 2002. Genetic conse-
quences of hunting: what do we know
and what should we do? Wildlife Soci-
ety Bulletin 30:634-643.
HEDRICK, P. W. 1994. Evolutionary genet-
ics of the major histocompatibility com-
plex. American Naturalist 143:945-
964.
HEWITT, G. M. 1996. Some genetic conse-
quences of ice ages, and their role in
divergence and speciation. Biological
Journal of the Linnaen Society 58:247-
276.
HIRTH, D. H. 1977. Social behavior of
white-tailed deer in relation to habitat.
Wildlife Monographs 53.
HUNDERTMARK, K. J., R. T. BOWYER, G. F.
SHIELDS, and C. C. SCHWARTZ. 2003.
Mitochondrial phylogeography of moose
(Alces alces) in North America. Jour-
nal of Mammalogy 84:718-728.
_____, P. E. JOHNS, and M. H. SMITH.
1992. Genetic diversity of moose from
the Kenai Peninsula, Alaska. Alces
28:15-20.
_____, G. F. SHIELDS, R. T. BOWYER, and C.
C. SCHWARTZ. 2002a. Genetic relation-
ships deduced from cytochrome-b se-
quences among moose. Alces 38:113-
122.

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
119
_____, _____, I. G. UDINA, R. T. BOWYER,
A. A. DANILKIN, and C. C. SCHWARTZ.
2002b. Mitochondrial phylogeography
of moose (Alces alces): late Pleistocene
divergence and population expansion.
Molecular Phylogenetics and Evolution
22:375-387.
_____, T. H. THELEN, and R. T. BOWYER.
1998. Effects of population density and
selective harvest on antler phenotype in
simulated moose populations. Alces
34:375-383.
_____, _____, and C. C. SCHWARTZ. 1993.
Genetic and population effects of se-
lective harvest systems in moose: a
modeling approach. Alces 29:225-234.
IRWIN, D. M., T. D. KOCHER, and A. C.
WILSON. 1991. Evolution of the cyto-
chrome b gene in mammals. Journal of
Molecular Evolution. 32: 128 - 144.
JOUZEL, J., C. LORIUS, J. R. PETIT, C. GENTHON,
N. I. BARKOV, V. M. KOTLYAKOV, and V.
M. PETROV. 1987. Vostok ice core: a
continuous isotope temperature record
over the last climate cycle (160,000
years). Nature 329:403-408.
LISTER, A. M. 1993. Evolution of mam-
moths and moose: the Holarctic per-
spective. Pages 178-204 in A. D.
Barnosky, editor. Quaternary Mam-
mals of North America. Cambridge
University Press, Cambridge, U.K.
LOPEZ, J. V., M. CULVER, J. C. STEPHENS, W.
E. JOHNSON, and S. J. O’BRIEN. 1997.
Rates of nuclear and cytoplasmic mito-
chondrial DNA sequence divergence in
mammals. Molecular Biology and Evo-
lution 14:277-286.
LYNCH, M. 1996. A quantitative-genetic
perspective on conservation issues.
Pages 471-501 in J. C. Avise and J. L.
Hamrick, editors. Conservation Genet-
ics: Case Studies from Nature.
Chapman & Hall, New York, New
York, USA.
MASON, O. K., P. M. BOWERS, and D. M.
HOPKINS. 2001. The early Milankovitch
thermal maximum and humans: ad-
verse conditions for the Denali complex
of eastern Beringia. Quaternary Sci-
ence Reviews 20:525-548.
MIKKO, S., and L. ANDERSSON. 1995. Low
major histocompatibility complex class
II diversity in European and North
American moose. Proceedings of the
National Academy of Sciences, USA
92:4259-4263.
MIYAMOTO, M. M., F. KRAUS, and O. A.
RYDER. 1990. Phylogeny and evolution
of antlered deer determined from mito-
chondrial DNA sequences. Proceed-
ings of the National Academy of Sci-
ences, USA 87:6127-6131.
MOLVAR, E. M., and R. T. BOWYER. 1994.
Costs and benefits of group living in a
recently social ungulate: the Alaskan
moose. Journal of Mammalogy 75:621-
630.
MULLIS, K., F. FALOONA, S. SCHARF, R. SAKAI,
G. HORN, and H. ERLICH. 1986. Specific
enzymatic amplification of DNA in vitro:
the polymerase chain reaction. Cold
Spring Harbor Symposia in Quantita-
tive Biology 51:263-273.
NADLER, C. F., C. E. HUGHES, K. E. HARRIS,
and N. W. ADLER. 1967. Electrophore-
sis of the serum proteins and transferrins
of Alces alces (elk), Rangifer tarandus
(reindeer), and Ovis dalli (Dall sheep)
from North America. Comparative
Biochemistry and Physiology 23:149-
157.
NEVO, E. 1978. Genetic variation in natural
populations: patterns and theory. Theo-
retical Population Biology 13:121-177.
PETERSON, R. L. 1955. North American
Moose. University of Toronto Press,
Toronto, Ontario, Canada.
QUELLAR, D. C., J. E. STRASSMANN, and C.
R. HUGHES. 1993. Microsatellites and
kinship. Trends in Ecology and Evolu-
tion 8:285-288.

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
120
RØED, K. H., and L. MIDTHJELL. 1998.
Microsatellites in reindeer, Rangifer
tarandus, and their use in other cervids.
Molecular Ecology 7:1773-1776.
RYMAN, N., R. BACCUS, C. REUTERWALL, and
M. H. SMITH. 1981. Effective popula-
tion size, generation interval, and poten-
tial loss of genetic variability in game
species under different hunting regimes.
Oikos 36:257-266.
_____, G. BECKMAN, G. BRUUN-PETERSEN,
and C. REUTERWALL. 1977. Variability
of red cell enzymes and genetic implica-
tions of management policies in
Scandinavian moose (Alces alces).
Hereditas 85:157-162.
_____, C. REUTERWALL, K. NYGREN, and T.
NYGREN. 1980. Genetic variation and
differentiation in Scandinavian moose
(Alces alces): are large mammals
monomorphic. Evolution 34:1037-1049.
SAGE, R. D., and J. O. WOLFF. 1986.
Pleistocene glaciations, fluctuating
ranges, and low genetic variability in a
large mammal (Ovis dalli). Evolution
40: 1092 - 1095.
SCHWARTZ, C. C., K. J. HUNDERTMARK, and
T. H. SPRAKER. 1992. An evaluation of
selective bull moose harvest on the Kenai
Peninsula, Alaska. Alces 28:1-13.
SCRIBNER, K. T., and R. K. CHESSER. 2001.
Group-structured genetic models in
analysis of the population and behavioral
ecology of poikilothermic vertebrates.
Journal of Heredity 92:180-189.
SELANDER, R. K., and D. W. KAUFMAN.
1973. Genic variability and strategies
of adaptation in animals. Proceedings
of the National Academy of Sciences
of the USA 70:1875-1877.
SHER, A. V. 1987. History and evolution of
moose. Swedish Wildlife Research
Supplement 1:71-97.
SHUBIN, P. N. 1969. The genetics of
transferrins in the reindeer and in the
European elk. Genetika 5:37-41. (In
Russian).
SMITH, M. H., R. BACCUS, H. O. HILLESTAD,
and M. N. MANLOVE. 1984. Population
genetics. Pages 119-128 in L. K. Halls,
editor. White-tailed Deer: Ecology and
Management. Stackpole Books,
Harrisburg, Pennsylvania, USA.
_____, K. T. SCRIBNER, L. H. CARPENTER,
and R. H. GARROTT. 1990. Genetic char-
acteristics of Colorado mule deer
(Odocoileus hemionus) and compari-
sons with other cervids. Southwestern
Naturalist 35: 1 - 8.
SYMONDSON, W. O. 2002. Molecular iden-
tification of prey in predator diets.
Molecular Ecology 11:627-641.
TELFER, E. S. 1984. Circumpolar distribu-
tion and habitat requirements of moose
(Alces alces). Pages 145-182 in R.
Olson, F. Geddes, and R. Hastings, edi-
tors. Northern Ecology and Resource
Management. University of Alberta
Press, Edmonton, Canada.
THELEN, T. H. 1991. Effects of harvest on
antlers of simulated populations of elk.
Journal of Wildlife Management 55:243-
249.
THOUVENY, N., and E. BONIFAY. 1984. New
chronological data on European
Plio-Pleistocene faunas and hominid
occupation sites. Nature 308:355-358.
UDINA, I. G., A. A. DANILKIN, and G. G.
BOESKOROV. 2002. Genetic diversity of
moose (Alces alces L.) in Eurasia.
Genetika 38:951-957. (In Russian).
VERNESI, C., E. PECCHIOLO, D. CARAMELLI, R.
TIEDEMANN, E. RANDI, and G. BERTORELLE.
2002. The genetic structure of natural
and reintroduced roe deer (Capreolus
capreolus) populations in the Alps and
central Italy, with reference to the mito-
chondrial DNA phylogeography of Eu-
rope. Molecular Ecology 11:1285-1297.
WILHELMSON, M., R. K. JUNEJA, and S.
BENGTSSON. 1978. Lack of polymor-
phism in certain blood proteins and en-

ALCES VOL. 40, 2004 MOOSE GENETICS - HUNDERTMARK AND BOWYER
121
zymes of European and Canadian
moose. Naturaliste Canadien 105:445-
449.
WILSON, G. A., C. STROBECK, L. WU, and J.
W. COFFIN. 1997. Characterization of
microsatellite loci in caribou Rangifer
tarandus, and their use in other
artiodactyls. Molecular Ecology 6:697-
699.
WILSON, P. J., S. GREWAL, A. RODGERS, R.
REMPEL, J. SAQUET, H. HRISTIENKO, F.
BURROWS, R. PETERSON, and B. N.
WHITE. 2003. Genetic variation and
population structure of moose (Alces
alces) at neutral and functional DNA
loci. Canadian Journal of Zoology
81:670-683.
ZHELEZNOV, N. K. 1993. Historic and
current distribution of moose in the
northeast USSR. Alces 29:213-218.

HUNDERTMARK AND BOWYER - MOOSE GENETICS ALCES VOL. 40, 2004
122
Term Definition
Adaptive genetic variation Genotypic variation at loci that control traits important to fitness, such as morphology,
physiology, and behavior.
Allozyme A gene product (protein) that is distinguished by its migratory characteristics in a gel
exposed to an electric field (electrophoresis). Differences among alleles (different variants
of the same gene) at allozyme loci relate to amino acid composition and secondary
structure of the protein. Only mutations that create proteins with different migratory
characteristics are detectable.
Control region A portion of mtDNA that evolves (incorporates mutations) at a very fast rate, which makes
it a valuable marker for examining intraspecific genetic variation. Occasionally referred to
as the D-loop.
DNA sequence The ultimate level of analysis of genetic material. This technique deduces the identity and
order of nucleotides in a fragment of DNA. Mutations (nucleotide substitutions) are
detectable whether or not they create different gene products.
Effective population size The size of a standardized population that has the same degree of genetic drift as the
population being studied. An ideal population is a closed population of constant size with
non-overlapping generations and no variance in reproductive success. The smaller the
effective size of a population, the faster it will lose genetic diversity through drift
regardless of actual population size. Effective population size (Ne) is almost always a
fraction of the true population size (N).
Generation interval Mean age of all parents.
Genetic drift Changes in allele frequencies across generations due to random sampling error associated
with less than infinite population size.
Haplotype The haploid equivalent of genotype. Genetic type of an individual when haploid DNA,
such as mtDNA, is analyzed.
Heritability The proportion of variance in the expression of a trait, such as antler size or body size, that
is due to genetic effects (as opposed to environmental effects), i.e., the degree to which a
trait can be passed on to the next generation.
Microsatellite Segments of DNA composed of sequence units varying from 2-4 nucleotides in length that
are tandemly repeated. Size variation (number of repeated units) defines alleles.
Microsatellites are Mendelian in their inheritance, i.e., they are diploid, specific to a site on
a chromosome, and occur in either a homozygous or heterozygous genotype.
Minisatellite Similar to microsatellites except that the repeat units vary from 16-64 nucleotides in length
and occur at many sites throughout the chromosomes, thus exhibiting more than 2 alleles
per individual and creating complex genotypes of gel banding patterns that resemble bar
codes. This is the technique that pioneered genetic fingerprinting.
Mitochondrial DNA (mtDNA) A circular DNA molecule occurring in the mitochondrion. Unlike chromosomes, which
occur in pairs (diploid), mtDNA occurs as a single copy (haploid) because it is inherited
only from the maternal line. Therefore, it is not subject to recombination and changes only
via mutation. That property makes it particularly useful for tracing lineages through time.
Molecular clock The assumption that the average rate of mutation for a particular DNA sequence is
constant over evolutionary time. If a molecular clock can be assumed, the amount of
genetic divergence between populations or species can be converted to time since
divergence.
Phylogeography The study of genetic lineages in a spatial and temporal context, revealing historic
population processes and evolutionary histories.
RFLP Restriction Fragment Length Polymorphism – a section of DNA of known length is
digested with restriction enzymes (endonucleases), which cleave DNA at sites with
specific target sequences of 4-6 nucleotides. The digested fragments are separated by
length (number of nucleotides) using gel electrophoresis and haplotypes are characterized
by numbers and sizes of fragments.
Appendix 1. Glossary of specialized terms used in this review.