Fauna) relative to contemporaneous faunas in North America. Among six species, there is no significant difference in tooth size
1. Introduction
Palaeogeography, Palaeoclimatology, Palaeoebetween Panama and North America. A second test comparing the artiodactyl Paratoceras wardi between Panama and Texas
shows a significant difference in tooth size, with P. wardi from Panama having slightly larger teeth (104% to 112% larger) than
P. wardi from Texas. This difference in size, however, is small compared to that predicted from insular evolution and is more
consistent with the amount of variation seen in continent-wide species distributions. Results from both tests are consistent with
the hypothesis that southern Central America had a dry-land connection to North America during the middle Miocene. Based
on these data, there is no support for the alternate hypothesis of an extensive archipelago in southern Central America during
this time.
D 2005 Elsevier B.V. All rights reserved.
Keywords: Panama; Land mammal; Body size; Miocene; Paleobiogeography; Cucaracha FormationWas southern Central America an archipelago or a peninsula in the
middle Miocene? A test using land-mammal body size
Michael Xavier Kirby a,*, Bruce MacFadden b
aCenter for Tropical Paleoecology and Archaeology, Smithsonian Tropical Research Institute, Box 2072, Balboa, Panama
bFlorida Museum of Natural History, University of Florida, Gainesville, FL 32611, USA
Received 21 December 2004; received in revised form 2 June 2005; accepted 3 June 2005
Abstract
There has been considerable discussion about the complex geological history of southern Central America in the late
Cenozoic prior to the closing of the Isthmus of Panama during the Pliocene. It is particularly unclear how far nuclear Central
America extended southward as a continuous land connection with North America versus the alternate paleogeographic
reconstruction of an extensive island-arc system during this time. In modern faunas, terrestrial mammal body sizes have
fundamentally different patterns within regions of continuous land-mass versus island populations as a result of the bisland
rule.Q Variation in body size is relatively limited in species with continent-wide distributions, whereas, relative to mainland
populations, those on islands can have as much as a four-fold increase, or decrease, in body size. Using tooth molar dimensions
as a proxy for body mass, tooth size is compared from a middle Miocene land-mammal fauna from Panama (Gaillard Cut Local0031-0182/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.palaeo.2005.06.002
* Corresponding author.
E-mail address: kirbym@ancon.si.edu (M.X. Kirby).cology 228 (2005) 193?202
www.elsevier.com/locate/palaeoThe paleogeography of southern Central America
has changed profoundly over the past 20 million
years (Ma), from a volcanic arc separated from
South America by a wide seaway, to an isthmus
that connected North and South America only 3 Ma
(Duque-Caro, 1990; Coates et al., 1992, 2003,
2004; Coates and Obando, 1996). The formation
of the Isthmus of Panama was important because
it facilitated the mixing of terrestrial faunas between
the two continents (Webb, 1976; Stehli and Webb,
1985), as well as physically separating a once con-
tinuous marine province into separate and distinct
Pacific and Caribbean communities (Woodring,
1965, 1966; Vermeij, 1978; Vermeij and Petuch,
1986; Jackson et al., 1993, 1996). Although several
studies have clarified the geologic history of the
formation of the isthmus (Emiliani et al., 1972;
Keigwin, 1978; Duque-Caro, 1990; Coates et al.,
1992, 2003, 2004; Coates and Obando, 1996; Col-
lins and Coates, 1999), the paleogeography of
southern Central America before the isthmus has
remained unclear. Evidence from marine rocks and
volcanic terranes suggest that the volcanic arc dur-
ing the Miocene consisted of an archipelago sepa-
mammals to swim across? If the former hypothesis
is correct, in which there was continuous dispersal
M.X. Kirby, B. MacFadden / Palaeogeography, Palaeo194Fig. 1. The two paleogeographic models that have been proposed
for Central America during the middle Miocene. (A) Archipelago
model (Coates and Obando, 1996). (B) Peninsula model (Whitmore
and Stewart, 1965). Emergent land is represented by gray.between North and Central America as a result of a
peninsula, then we may expect little to no isolation
and, therefore, little to no divergent evolution in
body size between population centers. If the latter
hypothesis is correct, then we may expect a greater
degree of isolation and, therefore, divergent evolu-
tion in body size of the populations reaching islands
in Central America via swimming or some other
sweepstakes event, relative to dispersal across dry
land. The purpose of this study is to test these two
hypotheses by comparing differences in tooth size,
here used as a proxy for body size, between Mio-
cene land mammals from Panama and their conspe-
cifics from North America. Testing for differences
in tooth size allows us to constrain the strength of a
peninsular connection to North America.
2. Cast of characters: Miocene land-mammals
from Panama
Fossil land mammals are poorly known from
Central America. One of the few reported localities
containing fossil land mammals is the bGaillard Cut
Local FaunaQ (L.F.) in the middle Miocene Cucar-
acha Formation (Ferrusqu??a-Villafranca, 1978; Rich
and Rich, 1983; Tedford et al., 1987, 2004), which
lies within the Panama Canal Basin in central Pan-
ama (Fig. 2). Whitmore and Stewart (1965) pre-
sented a preliminary faunal list for the Gaillard
Cut L.F., and MacFadden (in press) formally de-rated by shallow straits (Fig. 1A; Coates and
Obando, 1996). Alternatively, evidence of terrestrial
mammals with only North American affinities sug-
gests a dry-land connection with North America by
16 Ma (Fig. 1B), effectively making the Central
American volcanic arc a peninsula of North Amer-
ica during this time (Whitmore and Stewart, 1965).
It remains unclear, however, just how strong this
connection was before formation of the isthmus.
Specifically, did North American mammals (1) dis-
perse over dry land between North and Central
America, or (2) were there narrow straits that
allowed some, but not all, North American land
climatology, Palaeoecology 228 (2005) 193?202scribed the assemblage, which now includes 10
species comprising four orders and eight families
in body size is plastic and can vary within certain
Fig. 2. Location of the Gaillard Cut Local Fauna (open circle) within the Panama Canal Basin in central Panama.
M.X. Kirby, B. MacFadden / Palaeogeography, Palaeoclimatology, Palaeoecology 228 (2005) 193?202 195of mammals (Table 1). The Gaillard Cut L.F. is of
great paleobiogeographic significance because it is
located 2000 km south of the closest similar aged
fauna in central Mexico (Ferrusqu??a-Villafranca,
1978), and about 500 km northwest of middle
Miocene faunas from Colombia (Kay et al., 1997).
Despite the relative proximity to South America, the
Gaillard Cut L.F. is of completely North American
taxonomic affinity (Table 1; Fig. 3), suggesting a
seaway barrier to the east of the Panama Canal
Basin. This fauna is particularly important because
it indicates that a diverse continental-like communi-
ty was already well established by the middle Mio-
cene in Panama (Whitmore and Stewart, 1965;
MacFadden and Higgins, 2004). These land mam-
mals, therefore, represent the southernmost North
Table 1
Land mammal taxa from the Gaillard Cut Local Fauna, Cucaracha Forma
Order Family Genus and s
Rodentia ? Texomys ste
Carnivora Canidae Tomarctus b
Carnivora Amphicyonidae or Hemicyonidae ?
Artiodactyla Tayassuidae Cynorca sp
Artiodactyla Oreodontidae Merycochoe
Artiodactyla Protoceratidae Paratoceras
Perissodactyla Equidae Anchitherium
Perissodactyla Equidae Archaeohipp
Perissodactyla Rhinocerotidae Menoceras
Perissodactyla Rhinocerotidae Floridacera
Source: Whitmore and Stewart (1965), Slaughter (1981), MacFadden and
a New discovery.limits given changes in the ecology of local popula-
tions. In this section, we briefly review the degree to
which variation occurs within a recognizable speciesAmerican fauna before the great American biotic
interchange about 3 Ma.
3. Mammalian body size variation on islands and
continents
Body size is an extremely important trait of any
species and is highly predictive of ecology, adapta-
tion, and life-history characteristics (e.g., Eisenberg,
1981; Peters, 1983). Likewise, phenotypic variabilityof mammals, both on islands relative to mainland
tion, Panama
pecies Common name Biogeographic affinity
warti Geomyoid rodent North America
revirostris Dog North America
Bear dog North America
.a Peccary North America
rus matthewi Oreodont North America
wardi Protoceratid North America
clarencei Horse North America
us sp. Horse North America
barbouri Rhinoceros North America
s whitei Rhinoceros North America
Higgins (2004), MacFadden (in press), this study.
alaeoM.X. Kirby, B. MacFadden / Palaeogeography, P196counterparts, and within species that are widespread
on continents.
3.1. Island mammals
Numerous studies have demonstrated that popula-
tions inhabiting islands rapidly diverge in body size
relative to mainland populations of the same, or close-
ly related sibling, species. Commonly known as the
bisland ruleQ (Van Valen, 1973), within an island
community of terrestrial mammals, species with
large body sizes will evolve smaller body sizes rela-
tive to their mainland counterparts (dwarfism or nan-
ism); whereas species with small body sizes will
evolve larger body sizes (gigantism) (Foster, 1964;
Gould and MacFadden, 2004). The explanations for
dramatic and rapid size changes in insular mammal
populations are complex and appear to generally re-
late to changes and limitations in resource availability
(Lomolino, 1985; McNab, 1994), or to release from
Fig. 3. Paleobiogeographic distribution of land mammals from the Gaillard
their conspecifics in North America. Data from Whitmore and Stewart (19
Paleobiology Database (http://paleodb.org/).climatology, Palaeoecology 228 (2005) 193?202competition and predation that allows evolution to-
ward an optimum body size for energy acquisition
(Case, 1978; Lomolino, 1985; Damuth, 1993). Nu-
merous examples of these phenomena are reported
from extinct Miocene through Pleistocene mammalian
faunas of the Mediterranean islands. At one end of the
spectrum, the fox-sized Miocene insectivore Deino-
galerix koenigswaldi from the Gargano peninsula of
Italy (which was an island during the Neogene) is the
largest insectivore ever reported, extinct or living
(Freudenthal, 1972). In contrast, the tiny elephant
Elephas falconeri from the Pleistocene of Sicily is
about four times smaller in skull size than its conge-
ner, and presumed ancestor, E. namadicus from the
adjacent mainland (Sondaar, 1977). Size reduction in
extinct proboscideans is in fact pervasive in island
faunas. In addition to the Mediterranean during the
Pleistocene, similar examples of bdwarfQ probosci-
deans are known from Stegodon of Indonesia (Hooi-
jer, 1962) and the pygmy mammoth (Mammuthus
Cut Local Fauna, Cucaracha Formation, Panama Canal Basin, and
65), MacFadden and Higgins (2004), MacFadden (in press) and the
ences in mammalian body size is predicated on the
assumption that this character is a known quantity.
alaeoexilis) of the Channel Islands off the coast of Cali-
fornia (Johnson, 1978).
An excellent example of insular dwarfism is also
seen in the modern three-toed sloth Bradypus from the
Bocas del Toro archipelago of Panama (Anderson and
Handley, 2002). This series of islands has been se-
quentially isolated at various times over the past
10,000 years since the latest Pleistocene. Relative to
mainland populations of Bradypus, the degree of
dwarfism is proportional to the amount of time the
particular island populations have been isolated (i.e.,
populations from older islands have significantly
smaller body sizes relative to populations from youn-
ger islands). Likewise, the recent discovery of Homo
floresiensis from the island of Flores in Indonesia
offers an intriguing example of dwarfism from within
our own genus (Brown et al., 2004).
3.2. Continental mammals
When there are few barriers to dispersal and con-
sequently ready gene flow between populations within
a species, body size is known to vary, but considerably
less than what is seen for insular mammals relative to
their mainland counterparts. Yablokov (1974) pre-
sented an exhaustive analysis of variation of soft and
hard anatomical characters in modern mammals. Lin-
ear measurements (e.g., body lengths), normally vary
within a coefficient of variation (V) of b10%; whereas,
body mass can have significantly more variation with-
in species, with V?s frequently approaching 30%, or in
some cases more in highly dimorphic species of mam-
mals. As a general rule, Mayr (1969) stated that V?s of
taxonomic relevance (i.e., within a given species), are
usually less than 30%. Simpson et al. (1960) found
that the great majority of linear anatomical dimensions
have V?s between 4% and 10%. Several studies
reported in MacFadden (1989) have shown that for
morphologically recognizable bmorphospeciesQ of fos-
sil mammals (e.g., horses), linear measurable charac-
ters, such as tooth lengths and widths, will
demonstrate V?s between 5% and 10%.
3.3. Predictive test of insularity versus continental
mammals
M.X. Kirby, B. MacFadden / Palaeogeography, PThe foregoing discussion indicates that linear tooth
dimensions measured from the Gaillard Cut L.F. asIn the fossil record, however, true body size is
rarely preserved. Thus, inferences about body size
must be made from characters that readily preserve
in the fossil record. For extinct mammals, teeth
provide one of the most frequently preserved clues
to estimating general body size. Studies have shown
that within taxonomic groups, such as fossil horses,
molar dimensions are highly correlated (R2N0.90)
to body mass (MacFadden, 1986). We, therefore,
infer that teeth can be used as a reliable proxy for
relative body size (Damuth and MacFadden, 1991).
In this study, premolar and molar lengths were usedcompared to similar-aged faunas from mainland North
America can be used as a proxy for understanding
whether Panama was an archipelago or a southern
peninsular extension of North America during the
middle Miocene. On the one hand, if we find signif-
icant differences between the Panamanian and North
American samples, such that the Panamanian mam-
mals are significantly smaller, then the archipelago
hypothesis is supported. Alternatively, if there is no
significant difference in tooth size, then the hypothesis
of a contiguous mainland connection with ready gene
flow is supported.
4. Methods
Samples were analyzed for six of the 10 taxa
represented in the Gaillard Cut L.F. as compared to
their North American counterparts; specifically, the
canid Tomarctus brevirostris, oreodont Meryco-
choerus matthewi, protoceratid Paratoceras wardi,
horse Anchitherium clarenci, and rhinos Menoceras
barbouri and Floridaceras whitei (Table 1). Four
other taxa from the Gaillard Cut L.F. were not includ-
ed in this analysis because they were not identifiable
to species (indeterminant amphicyonid or hemicyonid,
peccary Cynorca sp., horse Archaeohippus sp.) or
they were endemic to Panama (Texomys stewarti).
4.1. Tooth dimensions as a proxy for body size
The above discussion concerning possible differ-
climatology, Palaeoecology 228 (2005) 193?202 197as a proxy to estimate differences in relative body
size between the six mammalian species from Pan-
ture. In the second analysis, we compared tooth size
within a single species (P. wardi) from Panama and
10
20
30
40
50
0
Panama
North America
Tr
an
sv
er
se
 w
id
th
 (m
m)
oster
M. barbouri
A. clarencei
P. wardi T. brevirostris
M. matthewi
F. whitei
anam
M.X. Kirby, B. MacFadden / Palaeogeography, Palaeoclimatology, Palaeoecology 228 (2005) 193?202198ama and their counterparts in continental North
America.
4.2. Measurements, comparisons, and statistical tests
We performed two separate analyses on the land-
mammal teeth from the Cucaracha Formation. First,
we compared tooth size between all six species from
0
0 10 2
Anteriop
Fig. 4. Bivariate plot comparing tooth size of six species between PPanama and North America. Teeth were measured
along their greatest transverse width and anteriopos-
terior length with dial calipers. In the case of tapering
teeth, such as in artiodactyls, the measurement was
taken from the base. Tooth size of some of the North
American conspecifics was derived from the litera-
Table 2
Comparison of tooth size of six species from Panama and North America
Taxon Tooth
position
Panama
Count Apl (mm) Trn (mm)
Tomarctus brevirostris m1 1 22.5 10.9
Merycochoerus matthewi m3 1 47.2 18.9
Paratoceras wardi m3 9 19.0F0.8 11.2F1.0
Anchitherium clarencei M1? 1 20.5 23.4
Menoceras barbouri RI1 1 35.0 40.0
Floridaceras whitei p4 1 35.7 24.3
Reported as meanF1 standard deviation.
North American data from Patton and Taylor (1973) and MacFadden (in
Apl=Anteroposterior length.
Trn=Transverse width.Texas, for which we have adequate data. In both
analyses, we used paired sign tests to test for signif-
icant differences in tooth size (Sokal and Rohlf,
1995).30 40 50
ior length (mm)
a and North America. Error bars represent one standard deviation.5. Results
Comparison between the six species from Panama
and North America shows no overall significant dif-
ference in tooth size (paired sign test, p N0.6874; Fig.
4; Table 2). One Panamanian taxon (F. whitei) was 91
North America Apl percent
difference
Trn percent
differenceCount Apl (mm) Trn (mm)
5 22.8F1.2 9.7F0.6 99 113
3 47.2F1.7 19.9F1.0 100 95
12 17.0F0.6 10.8F0.5 112 104
5 20.8 23.3 99 100
1 33.3 37.0 105 108
1 37.9 26.8 94 91
press).
the difference appears about equal for the other teeth
0
5
10
15
Panama p3
Texas p3
Panama p4
Texas p4
Panama m1
Texas m1
Panama m2
Texas m2
Panama m3
Texas m3
Panama m1-m3
Texas m1-m3
Tr
an
sv
er
se
 w
id
th
 (m
m)
ster
tween
M.X. Kirby, B. MacFadden / Palaeogeography, Palaeoclimatology, Palaeoecology 228 (2005) 193?202 199to 94% smaller than its North American conspecific.
Three taxa from Panama (T. brevirostris, P. wardi and
M. barbouri) were as much as 113% larger than their
North American conspecifics. Two taxa (M. matthewi,
and A. clarencei) were too close in size to differenti-
ate. The second test, the comparison of P. wardi
between Panama and Texas, shows a significant dif-
ference in tooth sizes, with P. wardi from Panama
0 10 20
Anteropo
Fig. 5. Bivariate plot comparing tooth size of Paratoceras wardi behaving significantly larger teeth ( pb0.001; Fig. 5;
Table 3). All five lower premolars and molars (p3,
p4, m1, m2, m3), as well as the anteroposterior length
of m1?m3, are larger in P. wardi from Panama. There
is a greater difference in anteroposterior length than
Table 3
Comparison of tooth size of Paratoceras wardi from Panama and Texas
Tooth position Panama Texas
Count Apl (mm) Trn (mm) Count
p3 5 13.1F1.0 5.4F0.6 3
p4 3 12.5F2.9 7.3F0.4 8
m1 8 11.8F1.0 9.8F0.6 11
m2 9 13.3F0.3 11.6F0.9 12
m3 9 19.0F0.8 11.2F1.0 12
m1?m3 7 43.4F1.3 11.5F1.0 11
Reported as meanF1 standard deviation.
North American data from Patton and Taylor (1973).
Apl=Anteroposterior length.
Trn=Transverse width.(Fig. 5; Table 3). Based on these data, P. wardi from
Panama is at most 112% larger than P. wardi from
Texas.
6. Discussionthere is in transverse width for m3 and m1?m3, but
30 40 50
ior length (mm)
Panama and Texas. Error bars represent 95% confidence intervals.The island rule predicts that populations of large
mammals isolated on islands from their mainland
counterparts will evolve smaller body sizes, whereas
populations of small mammals similarly isolated will
Apl percent
difference
Trn percent
differenceApl (mm) Trn (mm)
12.5F0.3 5.1F0.1 104 107
11.5F0.3 6.7F0.4 108 108
11.1F0.3 9.4F0.6 106 104
12.1F0.3 10.6F0.9 110 110
17.0F0.7 10.8F0.5 112 104
39.6F1.0 10.8F0.5 110 106
7. Conclusions
alaeoevolve larger body sizes (Foster, 1964; Van Valen,
1973; Case, 1978; Lomolino, 1985). The threshold
between larger and smaller body-size evolution has
been in dispute, but probably lies around 1 kg in body
size (Lomolino, 1985; Damuth, 1993). This body size
reflects an optimum body size for energy acquisition,
such that, if on islands a species? usual competitors
and predators are absent, then the species should tend
to evolve toward the optimum body size of 1 kg
(Damuth, 1993).
The present study uses tooth size as a proxy for
body size to test for the effect of insular evolution on
an island among six species of land mammals. The
data offer no support for insular evolution on an
island, but they are consistent with evolution on a
peninsula where populations had easy dispersal and
intermixing with populations in North America. The
first test shows that there is no overall difference in
tooth size among the six species. Some species from
Panama are slightly larger than North American con-
specifics, whereas some are smaller. However, none
of these slight differences (b113%) in tooth size are as
large as what would be expected after insular evolu-
tion on an island. The second test shows that P. wardi
from Panama had significantly larger teeth, but the
difference is only at most 112% larger. Furthermore,
this difference is in the opposite direction from what
the island rule would predict for a large mammal, such
as P. wardi. P. wardi has only been described from
one other location: The upper part of the Fleming
Formation (early Bartovian) in San Jacinto and Walk-
er counties, Texas (Patton and Taylor, 1973). It is not
understood why P. wardi from Panama would have
evolved a significantly larger body size compared to
its conspecific in Texas.
The paleogeography of southern Central America
before formation of the Panamanian isthmus has
remained poorly known, despite advances in under-
standing the timing and formation of the Isthmus. The
results from this study are consistent with the view
that southern Central America existed as a peninsula
during the middle Miocene. Nevertheless, there is
geologic evidence that narrow straits once crossed
southern Central America before and after this time.
Paleoenvironmental analysis of the underlying Cule-
bra Formation (older than the Gaillard Cut L.F.)
M.X. Kirby, B. MacFadden / Palaeogeography, P200shows that a narrow strait connected the Pacific
Ocean with the Caribbean Sea across the PanamaBefore the formation of the Isthmus of Panama in
the Pliocene, southern Central America either existed
as an island-arc archipelago or as a long peninsula that
was connected to North America. We tested these
alternate paleogeographic models by comparing the
size of teeth (as a proxy for body size) of middle
Miocene land mammals from the Cucaracha Forma-
tion in Panama with their conspecifics in North Amer-
ica. Of the six species examined, there is no
significant difference in tooth size. A further test of
the artiodactyl P. wardi showed a significant differ-
ence in tooth size, with P. wardi from Panama having
as much as 112% larger teeth than their conspecific
from Texas. However, this difference in tooth size is
opposite, as well as smaller, to that predicted from the
bisland rule,Q where insular evolution on an island
should have produced up to a four-fold smaller body
size relative to conspecifics on the mainland. Results
from both tests are inconsistent with insular evolution
on an island-arc and we, therefore, conclude that the
land mammals from Panama had ready gene flow withCanal Basin (Kirby, submitted for publication). This
strait had a maximum width of 60 km during deposi-
tion of the lower Culebra Formation, based on the
present distribution of pre-Miocene volcanic terranes
(Kirby, submitted for publication). The upper Culebra
and Cucaracha formations, the latter containing the
land mammals of the Gaillard Cut L.F., represent a
prograding delta that filled in this strait in part or
completely during the middle Miocene (Kirby, sub-
mitted for publication). A second, even younger strait
across the Panama Canal Basin formed again about 6
Ma, based on benthic foraminifera showing Pacific
affinities in the Caribbean Chagres Formation (Collins
et al., 1996). During these two episodes, when a strait
formed across southern Central America, insular evo-
lution may have occurred in land-mammal popula-
tions cut off from continental populations. But
during the middle Miocene, our study found no evi-
dence for straits or other significant barriers to gene
flow separating the Gaillard Cut L.F. from related
faunas living in North America.
climatology, Palaeoecology 228 (2005) 193?202populations in North America across a continuous
peninsula during the middle Miocene.
America. Univ. Chicago Press, Chicago, pp. 21?56.
Coates, A.G., Jackson, J.B.C., Collins, L.S., Cronin, T.M., Dow-
M.X. Kirby, B. MacFadden / Palaeogeography, Palaeoclimatology, Palaeoecology 228 (2005) 193?202 201sett, H.J., Bybell, L.M., Jung, P., Obando, J.A., 1992. Closure
of the Isthmus of Panama: the near-shore marine record of
Costa Rica and western Panama. Geol. Soc. Amer. Bull. 104,
814?828.
Coates, A.G., Aubry, M.P., Berggren, W.A., Collins, L.S., 2003.
Early Neogene history of the Central American arc from
Bocas del Toro, western Panama. Geol. Soc. Amer. Bull. 115,
271?287.
Coates, A.G., Collins, L.S., Aubry, M.P., Berggren, W.A., 2004.
The geology of the Darien, Panama, and the late Miocene?
Pliocene collision of the Panama arc with northwestern south
America. Geol. Soc. Amer. Bull. 116, 1327?1344.
Collins, L.S., Coates, A.G. (Eds.), 1999. A paleobiotic survey of
Caribbean faunas from the Neogene of the Isthmus of Panama,
Bull. Am. Paleontol., vol. 357.
Collins, L.S., Coates, A.G., Berggren, W.A., Aubry, M.P., Zhang, J.,
1996. The late Miocene Panama isthmian strait. Geology 24,
687?690.
Damuth, J., 1993. Cope?s rule, the island rule and the scaling ofAcknowledgements
We are grateful to A. Coates and J. Jackson for
supporting this research. We thank the Autoridad del
Canal de Panama (ACP) for granting access to local-
ities, F.C. de Sierra and J. Villa of the General de
Recursos Minerales, Republic of Panama, for granting
the necessary permits, and P. Franceshi of the ACP for
providing logistical support in the field. We thank R.J.
Emry and D. Bohaska of the United States National
Museum for access to the Gaillard Cut L.F. Funded by
a Smithsonian Institution postdoctoral fellowship to
MXK and the Smithsonian Walcott Fund. This is
University of Florida Contribution to Paleobiology
574.
References
Anderson, R.P., Handley, C.O., 2002. Dwarfism in insular sloths:
biogeography, selection, and evolutionary rate. Evolution 56,
1045?1058.
Brown, P., Sutikna, T., Morwood, M.J., Soejono, R.P., Wayhu
Saptomo, E., Due, Rokus Awe, 2004. A new small-bodied
hominin from the late Pleistocene of Flores, Indonesia. Nature
431, 1055?1061.
Case, T.J., 1978. A general explanation for insular body size trends
in terrestrial vertebrates. Ecology 59, 1?18.
Coates, A.G., Obando, J.A., 1996. The geologic evolution of the
central American Isthmus. In: Jackson, J.B.C., Budd, A.F.,
Coates, A.G. (Eds.), Evolution and Environment in Tropicalmammalian population density. Nature 365, 748?750.Damuth, J., MacFadden, B.J. (Eds.), 1991. Body Size in Mamma-
lian Paleobiology: Estimation and Biological Implications.
Cambridge Univ. Press, New York.
Duque-Caro, H., 1990. Neogene stratigraphy, paleoceanography
and paleobiogeography in northwest south America and the
evolution of the Panama seaway. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 77, 203?234.
Eisenberg, J.F., 1981. The Mammalian Radiations: An Analysis of
Trends in Evolution, Adaptations, and Behavior. Univ. Chicago
Press, Chicago.
Emiliani, C., Gartner, S., Lidz, B., 1972. Neogene sedimentation on
the Blake Plateau and the emergence of the Central American
isthmus. Palaeogeogr. Palaeoclimatol. Palaeoecol. 11, 1?10.
Ferrusqu??a-Villafranca, I., 1978. Distribution of Cenozoic vertebrate
faunas in middle America and problems of migration between
North and South America. Inst. Geol., Univ. Nac. Autonoma
Me?xico 101, 193?329.
Foster, J.B., 1964. Evolution of mammals on islands. Nature 202,
234?235.
Freudenthal, M., 1972. Deinogalerix koenigswaldi nov. gen., nov.
spec., a giant insectivore from the Neogene of Italy. Scr. Geol.
14, 1?19.
Gould, G.C., MacFadden, B.J., 2004. Gigantism, dwarfism, and
Cope?s rule: nothing in evolution makes sense without a phy-
logeny. Bull. Am. Mus. Nat. Hist. 285, 219?237.
Hooijer, D., 1962. Stegodon trigonocephalus florensis [sic] Hooijer
and Stegodon timorensis [sic] Sartono from the Pleistocene of
Flores and Timor. Proc. Koninkl. Nederl. Akademi van
wetensch., Amsterdam, ser. B 75, 12?33.
Jackson, J.B.C., Jung, P., Coates, A.G., Collins, L.S., 1993. Diver-
sity and extinction of tropical American mollusks and emer-
gence of the Isthmus of Panama. Science 260, 1624?1626.
Jackson, J.B.C., Jung, P., Fortunato, H., 1996. Paciphilia revisited:
transisthmian evolution of the Strombina group (Gastropoda:
Columbellidae). In: Jackson, J.B.C., et al., (Eds.), Evolution and
Environment in Tropical America. Univ. Chicago Press, Chi-
cago, pp. 234?270.
Johnson, D.L., 1978. The origin of island mammoths and the
Quaternary land bridge history of the northern Channel Islands,
California. Quat. Res. 10, 204?225.
Kay, R.F., Madden, R.H., Cifelli, R.L., Flynn, J.J. (Eds.), Vertebrate
Paleontology in the Neotropics: The Miocene Fauna of La Venta,
Colombia. Smithsonian Institution Press, Washington, DC.
Keigwin, L.D., 1978. Pliocene closing of the Isthmus of Panama,
based on biostratigraphic evidence from nearby Pacific Ocean
and Caribbean cores. Geology 6, 630?634.
Kirby, M.X., submitted for publication. Oligocene to Miocene
paleoenvironments of the Panama Canal Basin: Paleogeographic
implications to the Central American Peninsula. Geol. Soc.
Amer. Bull.
Lomolino, M.V., 1985. Body size of mammals on islands: the island
rule reexamined. Am. Nat. 125, 310?316.
MacFadden, B.J., 1986. Fossil horses from bEohippusQ (Hyra-
cotherium) to Equus: scaling, Cope?s law, and the evolution
of body size. Paleobiology 12, 355?370.
MacFadden, B.J., 1989. Dental character variation in paleopopula-
tions and morphospecies of fossil horses and extant analogues.
In: Prothero, D.R., Schoch, R.M. (Eds.), The Evolution of Peri-
ssodactyls. Clarendon (Oxford) Press, New York, pp. 128?141.
MacFadden, B.J., in press. North American Miocene land mammals
from Panama. Jour. Vert. Paleontol.
MacFadden, B.J., Higgins, P., 2004. Ancient ecology of 15-million-
year-old browsing mammals within C3 plant communities from
Panama. Oecologia 140, 169?182.
Mayr, E., 1969. Principles of Systematic Zoology. McGraw-Hill
Pub. Co., New York.
McNab, B.K., 1994. Resource use and the survival of land and fresh-
water vertebrates on oceanic islands. Am. Nat. 144, 643?660.
Patton, T.H., Taylor, B.E., 1973. The Protoceratinae (Mammalia,
Tylopoda, Protoceratidae) and the systematics of the Protocer-
atidae. Bull. Am. Mus. Nat. Hist. 150, 347?414.
Peters, R.H., 1983. The Ecological Implications of Body Size.
Cambridge Univ. Press, Cambridge.
Rich, P.V., Rich, T.H., 1983. The central American dispersal
route: biotic history and paleogeography. In: Janzen, D.H.
(Ed.), Costa Rican Natural History. Univ. Chicago Press, Chi-
cago, pp. 12?34.
Simpson, G.G., Roe, A., Lewontin, R.C., 1960. Quantitative Zool-
Tedford, R.H., Skinner, M.F., Fields, R.W., Rensberger, J.M.,
Whistler, D.P., Galusha, T., Taylor, B.E., Macdonald, J.R.,
Wedd, S.D., 1987. Faunal succession and biochronology of
the Arikareean through Hemphillian interval (late Oligocene
through earliest Pliocene epochs) in North America. In: Wood-
burne, M.O. (Ed.), Cenozoic Mammals of North America:
Geochronology and Biostratigraphy. Univ. California Press,
Berkeley, pp. 153?210.
Tedford, R.H., Albright III, L.B., Barnosky, A., Ferrusqu??a-Villa-
franca, I., Hunt Jr., R.M., Storer, J.E., Swisher III, C.C., Voorhies,
M.R., Webb, S.D., Whistler, D.P., 2004. Mammalian biochronol-
ogy of the Arikareean through Hemphillian interval (late Oligo-
cene through early Pliocene epochs). In: Woodburne, M.O. (Ed.),
Late Cretaceous and Cenozoic Mammals of North America.
Columbia Univ. Press, New York, pp. 169?231.
Van Valen, L., 1973. Pattern and the balance of nature. Evol. Theory
1, 31?49.
Vermeij, G.J., 1978. Biogeography and Adaptation: Patterns of
Marine Life. Harvard Univ. Press, Cambridge.
Vermeij, G.J., Petuch, E.J., 1986. Differential extinction in tropical
American molluscs: endemism, architecture, and the Panama
land bridge. Malacologia 27, 29?41.
Webb, S.D., 1976. Mammalian faunal dynamics of the great Amer-
M.X. Kirby, B. MacFadden / Palaeogeography, Palaeoclimatology, Palaeoecology 228 (2005) 193?202202Slaughter, B.H., 1981. A new genus of geomyoid rodent from
the Miocene of Texas and Panama. J. Vertebr. Paleontol. 1,
111?115.
Sokal, R.R., Rohlf, F.J., 1995. Biometry. W.H. Freeman and Com-
pany, San Francisco.
Sondaar, P.Y., 1977. Insularity and its effect on mammal
evolution. In: Hecht, M.K., Goody, P.C., Hecht, B.M. (Eds.),
Major Patterns in Vertebrate Evolution. Plenum Press, New
York, pp. 671?707.
Stehli, F.G., Webb, S.D., 1985. The Great American Biotic Inter-
change. Plenum Press, New York.ican interchange. Paleobiology 2, 220?234.
Whitmore Jr., F.C., Stewart, R.H., 1965. Miocene mammals and
Central American seaways. Science 148, 180?185.
Woodring, W.P., 1965. Endemism in middle Miocene Caribbean
molluscan faunas. Science 148, 961?963.
Woodring, W.P., 1966. The Panama land bridge as a sea barrier.
Proc. Am. Philol. Assoc. 110, 425?433.
Yablokov, A.V., 1974. Variability of Mammals. U.S. Dept. Com-
merce, Nat. Tech. Information Service.ogy. Harcourt, Brace, and Company, New York.