two factors: the degree to which unrelated groups show
parallel changes in morphology, and the relationship be-
tween the age of particular lineages and their current dis-
tributions. As suggested by the preceding quote, a natural
classification of rodents requires knowledge of phylogenetic
relationships based on both morphological and molecular
characters. The starting point for establishing such a clas-
sification requires a reassessment of interfamilial relation-
ships among rodents. Recent cladistically based morpho-
logical analyses that consider both fossil and extant forms
(Luckett and Hartenberger 1985; Meng 1990; Marivaux
et al. 2004) and molecular analyses (Nedbal et al. 1996;
Adkins et al. 2001, 2003; DeBry and Sagel 2001; Huchon
et al. 2001, 2002; Montegelard et al. 2002; DeBry 2003;
Jansa and Weksler 2004) have provided strong support for
several rodent clades as well as several hypotheses that can
be tested with the addition of taxa and characters. Although
phylogenies derived from molecules and morphology are
not totally congruent (fig. 2.4), they do provide strong sup-
port for several monophyletic groups, including: (1) Hu-
chon et al.’s (1999) Ctenohystrica, containing a sister-
group relationship between the Ctenodactyliae and the
Hystricognathi (Phiomorpha 1 Caviomorpha); (2) the Sci-
uroidea, consisting of the families Sciuridae and Aplodon-
tidae; (3) Montgelard et al.’s (2002) Anomaluromorpha,
represented by Anomaluridae and Pedetidae; (4) the Myo-
donta, containing the superfamilies Dipodoidea and Mu-
roidea; and (5) a larger clade comprised of Dipodoidea/
Muroidea, Anomaluridae /Pedetidae, Castoridae, and Ge-
omyoidea (Geomyidae and Heteromyidae). These two phy-
logenies differ in the placement of the families Castoridae,
Geomyidae, and Myoxidae. They also support the notion
that characteristics of both the zygomasseteric system and
structure of the lower jaw have evolved independently sev-
eral times throughout the rodent radiations.Relationships among the Hystricognathi
The suborder Hystricognathi represents a monophyletic
group (Luckett and Hartenberger 1993; Martin 1993; Ned-
bal et al. 1996; Huchon et al. 2000; Adkins et al. 2001; Hu-
chon and Douzery 2001; Murphy et al. 2001a) containing
17 of the 28 extant families of rodents and one-tenth of the
species (229 of 2021 species), and it shares a sister-group
relationship with the family Ctenodactylidae (fig. 2.2; Hu-
chon et al. 2000; Adkins et al. 2001). In terms of ecology
and behavior, members of this suborder display a diver-
sity of mating systems (monogamous, polygynous, polyan-
drous, and promiscuous), maintain lifestyles ranging from
solitary to group living (e.g., the eusocial naked mole-rat,
Heterocephalus glaber), and occupy a broad range of habi-
tats. Given the interest in understanding the influence that
environmental and historical constraints have on these eco-
logical and behavioral attributes, a well-resolved phylog-
eny for hystricognath rodents provides an ideal interpretive
framework for testing evolutionary hypotheses (Ebensper-
ger and Cofré 2000). Therefore, we will attempt to synthe-
size what is known and not known about the phylogeny of
hystricognaths.
Families of hystricognaths can be subdivided into three
major groups (fig. 2.2): (1) Bathy-Phiomorpha (Wood
1965), representing three families (Bathyergidae, Thryono-
myidae, and Petromuridae) endemic to sub-Saharan Africa;
(2) Hystricomorpha (Wood 1965), consisting of Old World
porcupines distributed in Africa and parts of Asia; and
(3) Caviomorpha (Lavocat 1973), a South American group
containing 13 of the 17 rodent families. Within the Bathy-
Phiomorpha, there is strong support for a monophyletic
Thryonomyoidea, comprised of the families Thryonomyi-
dae and Petromuridae (Lavocat 1973; Nedbal et al. 1994;
Huchon et al. 2001; Rowe 2002), and several studies sup-
port a monophyletic Caviomorpha (Lavocat and Parent
1985; Huchon et al. 2001; Rowe 2002). Although families
of caviomorphs are subdivided into four superfamilies,
Cavioidea, Chinchilloidea, Erethizontoidea, and Octodon-
toidea (Patterson and Wood 1982), the contents of and
relationships among these superfamilies have not been ade-
quately addressed. In addition, the placement of Hystrico-16 Chapter Two
Figure 2.4 Phylogeny of rodent families (molecules versus morphology).
A. Molecular-based phylogeny representing a compilation of information from
several sources (Nedbal et al. 1996; Adkins et al. 2001; Montgelard et al. 2002;
DeBry 2003). B. Results of a cladistic analysis of morphological data by Marivaux
et al. (2004).