molecular evolution within rodents and among eutherian
orders are not homogeneous, and the calibration of a mo-
lecular clock based on a single rate tends to bias divergence
times, especially for groups like rodents that evolve consid-
erably faster than other eutherian orders (Bromham et al.
2000). Therefore, a more accurate estimate is one that cor-
rects for apparent rate heterogeneity across lineages. Two
recent studies (Adkins et al. 2001; Huchon et al. 2002) at-
tempted to correct for rate heterogeneity and derive dates
based on multiple calibration points, and obtained diver-
gence times for rodents more in line with the fossil record.
For instance, Huchon et al. (2002) provided an estimate
of 55.8 mya for the origin of most rodent lineages, and
this estimate is congruent with fossil evidence suggesting
a Paleocene /Eocene radiation for rodents (Huchon et al.
2002). Likewise, the Mus/Rattusdivergence provided by
Adkins et al. (2001) was 23 mya, a date older than the fos-
sil record but congruent with earlier molecular dates (e.g.,
Sarich 1985).
Biogeographic Hypotheses and Phylogenetics
A combination of phylogenetics and molecular dating can
be used to examine broad-scale biogeographic events coin-
ciding with the diversification and distribution of rodents.
There are several intriguing controversies surrounding ro-
dent biogeography and the invasion of island continents
that lend themselves to renewed investigations using a phy-
logenetic approach.
Origin of South American rodents
Existing rodent fauna of South America represent two sepa-
rate invasions: an ancient African invasion of hystricognath
ancestors and a more recent Nearctic invasion of murid an-
cestors of the subfamily Sigmodontinae. The fossil records
and levels of diversity seen for both South American groups
have stimulated considerable debate regarding the timing of
these invasions and the dispersal pathways utilized.
The 13 families of South American hystricognath rodents
are collectively placed in the Caviomorpha (Wood 1965),
and their closest relatives, sometimes referred to as Phio-
morpha (Lavocat 1973), reside primarily in Africa (families
Bathyergidae, Petromuridae, and Thryonomyidae), with
the Old World porcupines (Hystricidae) occurring in both
Africa and Asia. The earliest caviomorph rodent fossils in
South America appear approximately 31–37 mya (Wyss
et al. 1993, 1994), and the oldest fossil phiomorphs date to
34 –37 mya in Africa (Hartenberger 1998). By the Desea-
dan (21–27 mya) representatives of all major superfamilies
and seven distinct families were present in South America
(Patterson and Wood 1982). These dates are well after the
Late Cretaceous separation of South America from Africa,
suggesting a trans-Atlantic, waif dispersal across an ocean
expanse of up to 1700 km (fig. 2.2). As with other cases of
island invasions via waif dispersal, considerable debate has
focused on whether South American caviomorph rodents
are the result of a single African invasion or are derived in-
dependently from multiple invasions. Earlier studies em-
ploying both immunological (Sarich and Cronin 1980) and
morphological (Bugge 1985; Woods and Hermanson 1985)
data suggested that caviomorphs might be the result of
two to three independent invasions. To the contrary, more
recent molecular data and morphological comparisons
strongly support monophyly of caviomorphs and a single
colonization event of South America (Luckett 1985; Ned-
bal et al. 1994; Adkins et al. 2001; Huchon and Douzery
2001). Recent estimates based on molecular clocks place
the origin of caviomorph rodents in South America at 40 –
60 mya (Huchon and Douzery 2001; Rowe 2002), sup-
porting colonization via waif dispersal during the late Pa-
leocene to Middle Eocene.
The Sigmodontinae represents a New World subfamily
of the Muridae that some taxonomic treatments divide into
North American and South American groups (Hooper and
Musser 1964; Carleton and Musser 1984; Musser and Car-
leton 1993). The earliest sigmodontines in South America
appear between 2 and 3 mya (Simpson 1950), and presum-
ably represent an adaptive radiation of a North American
stock that entered South America subsequent to the forma-
tion of the Panamanian land bridge. The major controversy12 Chapter Two
Figure 2.2 Biogeographic scenario for the evolution and dispersal of Hystri-
cognathirodents based on Huchon and Douzery (2001) and Rowe (2002). The
origin of Caviomorphafrom an African ancestor is well supported, as well as
the sister-group relationship between the Ctenodactylidaeand hystricognaths.
The placement of Hystricomorphais unclear.