Finally, a species level phylogeny was produced for the
genus Microtus(Jaarola et al. 2004), a diverse group in
the subfamily Arvicolinae that is distributed throughout
the Holarctic (Musser and Carleton 1993; Chaline et al.
1999). This phylogeny was derived using nucleotide se-
quences from the mitochondrial cytochrome bgene and 48
of the 65 species of Microtus. Several clades received good
support and appeared in both maximum likelihood and
maximum parsimony analyses. These clades include mem-
bers of the subgenus Microtus(containing the social voles
of the socialisspecies group), the subgenus Terricola(Pale-
arctic), Nearctic species, and an Asian group also supported
by a separate molecular study (Conroy and Cook 2000).
Deeper nodes in the phylogeny received poor support, mak-
ing the diagnosis of relationships among the major clades
more difficult.
Relationships among genera and species of Sciuridae
The squirrel family Sciuridae is the second most diverse
group of rodents, consisting of 273 species and 50 genera,
traditionally subdivided into the subfamilies Pteromyinae
(flying squirrels) and Sciurinae, consisting of tree squirrels
and ground squirrels (Hoffmann et al. 1993; Thorington
et al. 2002). Most members of the family are diurnal and
are diverse in terms of behavior and ecology, including the
formation of complex social groups among various species
of ground dwelling sciurids, and the occupation of a vari-
ety of habitats, ranging from deserts to grasslands and for-
ests (Dobson 1985; Armitage 1999a). Several genera (e.g.,
Spermophilusand Marmota) have a Holarctic distribution,
whereas others like Cynomysspeciated in North America.
As indicated by Thorington et al. (1997), the two major sys-
tematic issues are relationships among tribes and evidence
for monophyly of the recognized subfamilies. For instance,
there is strong morphological evidence for the monophyly
of flying squirrels (Thorington 1984). Nevertheless, immu-
nology data (Hight et al. 1974) suggest paraphyly for the
subfamily Pteromyinae, with gliding evolving independently
in Old World and New World forms. In addition, relation-
ships among Old and New World groups of ground squir-
rels and tree squirrels in the subfamily Sciurinae are un-
clear. Several recent molecular studies (Harrison et al. 2003;
Mercer and Roth 2003; Steppan et al. 2004; Herron et al.
2004) have contributed greatly to the resolution of the sci-
urid phylogeny (see fig. 2.3 for a summary of major clades).
In terms of the two subfamilies, the Pteromyinae is mono-
phyletic, suggesting a single origin for the gliding mecha-
nism, and the Sciurinae is paraphyletic, with flying squirrels
sister to New World tree squirrels. Although the tribes rep-
resenting Holarctic (Marmotini) and Afro-Asian (Xerini)
ground squirrels are not sister-groups, the monophyly of
each is supported (Mercer and Roth 2003; Steppan et al.
2004). In addition to resolving relationships among higher
taxa of sciurids, several studies provide detailed assessments
of relationships among ground squirrel genera (Steppan
et al. 1999; Harrison et al. 2003; Herron et al. 2004). Three
of these genera, Cynomys, Marmota,and Spermophilus,
are of considerable interest to behavioral ecologists. Mono-
phyly for both Marmotaand Cynomysis supported (Har-
rison et al. 2003; Herron et al. 2004), whereas the genus
Spermophilusappears paraphyletic, with the genera Am-
mospermophilus, Cynomys,and Marmotagrouping closer
to subsets of Spermophilus.
As previously indicated, the phylogeny for sciurids pro-
vides an interpretive framework for reconstructing the bio-
geographic history of the group. In addition, this same phy-
logeny has been used to test specific hypotheses about the
ancestry of sciurids in general. For instance, Steppan et al.
(2004) used a phylogenetic approach to determine whether
the ancestral sciurid was arboreal or terrestrial. They de-
fined three character states for degree of arboreality: arbo-
real (tree dwelling), intermediately arboreal (terrestrial with
time in trees), and terrestrial (ground dwelling). Optimiza-
tion of these traits on the existing phylogeny provided over-
all support for an arboreal ancestor. Using their phylogeny
for Marmota,Steppan et al. (1999) tested the hypothesis
that the establishment of large colonies and increased so-
ciality evolved once, and found that such complex social
systems originated two or more times. Finally, species-
level studies of the genus Cynomysfound a strong asso-
ciation between sister-group relationships and geographic
distribution (Harrison et al. 2003; Herron et al. 2004).
The following sister-group relationships in Cynomyswere
found: (1) (Cynomys ludovicianus Cynomys mexicanus);
(2) (Cynomys gunnisoni Cynomys leucurus) Cynomys
parvidens.Testing Evolutionary HypothesesA robust phylogeny provides an interpretive framework
that allows for inferences to be drawn regarding the evolu-
tion of various traits, including morphological, behavioral,
ecological, physiological, and so on. A phylogeny allows
one to test for relationships between the phenotype, envi-
ronment, and phylogeny (Harvey and Pagel 1991; Crespi
1996; Irwin 1996; Burda et al. 2000). Thus one can elu-
cidate shared traits resulting from similar environmental
constraints among phyogenetically distantly related line-
ages versus traits arising as a result of shared evolutionary
history.
Comparative methods and phylogenies provide valuable
tools for studying the ecology and behavior of social mam-Rodent Evolution, Phylogenetics, and Biogeography 19