ARTICLE

ONTOGENY AND DIVERSITY OF THE OLDEST CAPYBARAS (RODENTIA:
HYDROCHOERIDAE; LATE MIOCENE OF ARGENTINA)

CECILIA M. DESCHAMPS,*,1,3 A. ITATÍ OLIVARES,2,4 EMMA CAROLINA VIEYTES,2,4 and
MARÍA GUIOMAR VUCETICH1,4

1División Paleontología Vertebrados, Facultad de Ciencias Naturales y Museo, Universidad Nacional de La Plata,
Paseo del Bosque s/n, 1900 La Plata, Argentina, ceci@fcnym.unlp.edu.ar;

2Sección Mastozoología, División Zoología Vertebrados, Facultad de Ciencias Naturales y Museo,
Universidad Nacional de La Plata, Paseo del Bosque s/n, 1900 La Plata, Argentina;

3CIC (Comisión de Investigaciones Científicas, Provincia de Buenos Aires);
4CONICET (Consejo Nacional de Investigaciones Científicas y Técnicas)

ABSTRACT—According to our previous model for interpreting the fossil record of capybaras the cheek teeth grow in
width and length throughout life; flexids (especially h.s.i. and h.t.i.) deepen allometrically resulting in diverse occlusal
morphologies during ontogeny; in the more derived species the ‘onset’ of flexid development is pre-displaced, and the
relative depth and development rate of the flexids increase. Consequently, we interpreted the different occlusal mor-
phologies found in the fossil record as different ontogenetic stages, leading to a drastic diminution of latest Miocene-
Pliocene taxonomic diversity. The analysis of the geologically oldest capybaras, which are from the Arroyo Chasicó
Formation (Chasicoan SALMA, late Miocene), is added. This study suggests a single species occurs in this formation that
cannot be separated at the genus level from Huayquerian species. In the Arroyo Chasicó species, which is older and
theoretically more primitive than that of the Huayquerian, the flexids are shallower as predicted. The analysis supports
our model of capybara dental ontogeny and evolution and encourages revision of the whole family according to this
model.

INTRODUCTION

Living capybaras are notable among rodents for their large
size (100–130 cm body length) and semiaquatic habits. They usu-
ally live in groups of about 20 individuals, but colonies up to 100
have been reported (Nowak and Paradiso, 1983). The Hydro-
choeridae are first recorded in the Arroyo Chasicó Formation of
central Argentina (Chasicoan South American Land Mammal
Age, late Miocene; Fig. 1A–D). Capybara remains are relatively
common in some late Miocene and Pliocene localities, they de-
cline in the Pleistocene, and today the family is represented by a
single genus (Hydrochoerus) with two species (Woods and
Kilpatrick, 2005).

As for most mammals, fossil capybara taxonomy is based
mainly on tooth morphology. Owing to the particularly fragmen-
tary fossil record of capybaras, especially those of Huayquerian
age, many taxa are represented by a single specimen, sometimes
an isolated tooth. This fact, together with the common practice
of using typological criteria for classification, has resulted in a
long list of nominal taxa.

In a previous article (Vucetich et al., 2005), we proposed new
criteria based on molar ontogenetic growth patterns that permit
a better understanding of the fossil record. In short, we proposed
that: (1) euhypsodont (� hypselodont) teeth of capybaras grow
throughout life in all dimensions, with the peculiar characteristic
that the morphology of the occlusal surface becomes more com-
plex with age (i.e., the base of the tooth enlarges and keeps
folding throughout life); (2) flex/ids grow in length at different

rates (with positive allometry, isometrically, or apparently inde-
pendently of tooth size), resulting in slightly different occlusal
morphologies in relation to tooth size (� ontogenetic stage);
consequently, (3) several occlusal morphologies previously
thought to pertain to different species or even genera are in fact
ontogenetic stages of a single species. Capybara remains from
classical localities of Huayquerian age (uppermost Miocene) of
central Argentina were recently analyzed within this framework
(Vucetich et al., 2005) and their taxonomy was revised accord-
ingly (Table 1).

In this paper we study the oldest capybaras, those found in the
Arroyo Chasicó Formation (Buenos Aires Province, Argentina),
in order to test whether they follow the proposed model of on-
togenetic change and to revise their taxonomy and estimates of
diversity.

Institutional Abbreviations—GHUNLPam, Universidad Na-
cional de La Pampa, Geología Histórica; MD-CH, Museo Dar-
win, Punta Alta, Arroyo Chasicó Collection; MLP, Museo de La
Plata, La Plata; MMH, Museo Municipal de Ciencias Naturales
de Monte Hermoso, Monte Hermoso; MMP, Museo Municipal
de Ciencias Naturales ‘L. Scaglia’, Mar del Plata; MPEF, Museo
Paleontológico Egidio Feruglio, Trelew; SPV-FHC, Sección Pa-
leontología Vertebrados, Facultad de Humanidades y Ciencias,
Montevideo, Uruguay. All of these museums except the last are
in Argentina.

METHODS

Nomenclature and Measurements

Tooth nomenclature (Fig. 2A–C) and measurements (Fig. 2D–
F) follow Vucetich and colleagues (2005). Abbreviations refer to*Corresponding author.

Journal of Vertebrate Paleontology 27(3):683–692, September 2007
© 2007 by the Society of Vertebrate Paleontology

683

JVP Reprint Policy
SVP policy on distribution of the author's pdf file gives the following rights to the author:
		
(1)  Subject to payment of the required, one-time, non-refundable fee as determined by the Society, the right to post the unaltered pdf file as provided by JVP on a single, author-controlled, private or institutional web site for public access.

(2)  Without fee payment, the right to include on the author's Web page (a) a link to the JVP web site where members and subscribers may download a pdf copy of the article; (b) a link whereby interested persons may email the author and request a pdf copy for their own personal use but not for re-distribution.

(3)  Without fee payment, the right to send a copy of the article in pdf form to persons on their reprint mailing list, and to persons who request a copy as per (2) above, provided that the pdf is accompanied by the clear statement that: (a) Society of Vertebrate Paleontology is the copyright holder of the article and pdf; (b) the pdf is provided to the recipient for her/his personal use only and is not to be redistributed or disseminated, except for educational use within a school, college, or university setting, and then only if accompanied by a clear statement of these conditions. Such an author-distributed version must be identical to the final published version.



the Spanish names in order to conform to previous publications
on the Hydrochoeridae (e.g., Frailey, 1986).

Linear measurements of p4–m3 were taken with a stereomi-
croscope with micrometer eyepiece and included: AP, antero-
posterior length; AWa, width of prism Ia; AWb, width of prism
Ib; HFEL, length of fundamental external flexid (h.f.e.); HPEL,
length of primary external flexid (h.p.e.); HSEL, length of sec-
ondary external flexid (h.s.e.); HSIL, length of secondary inter-
nal flexid (h.s.i.); HTIL, length of tertiary internal flexid (h.t.i.);
MW, middle width; and PW, posterior width (Fig. 2D–F).

The lengths of the flexids with respect to the tooth width were
measured linearly on the occlusal surface and expressed as a
percent of total tooth width. It is described in the text as the
‘depth’ of the flexid. Short flexids are ’shallow’ and long flexids
are ‘deep.’

Quantitative Analysis

Only m1 and m2 were considered for the quantitative analysis
because they vary less among individuals than p4 and because
they are the best represented cheek teeth in the sample of Car-
diatherium patagonicum on which the proposal of our model was
originally based. Isolated m1s and m2s are difficult to differen-
tiate; consequently, they were analyzed together.

The linear measurements of m1 and m2 (Fig. 2) were log-
transformed and examined through principal components analy-
sis (PCA) based on a correlation matrix (Bookstein et al., 1985;
Legendre and Legendre, 1998). AP was used as a size estimate
following Vucetich et al. (2005). In order to test whether C.
chasicoense follows the proposed model of growth pattern of
h.t.i. and h.s.i., HSIL and HTIL versus AP of m1–m2 (Fig. 2)
were compared with allometric change of these flexids of the
Huayquerian species of the genus analyzed by Vucetich and col-
leagues (2005).

GEOLOGICAL SETTING

All the remains of Chasicoan capybaras were recovered from
the Las Barrancas Member of the Arroyo Chasicó Formation
(see below), in southwest Buenos Aires Province, in front of the
Bajada de los Toros locality (Fig. 1A–C). This locality is about
500 m downstream from the ‘Norma Alicia’ bridge over Arroyo
Chasicó (IGM 1:50.000 ‘Estancia Los Chañares’ chart N° 3963-
10-3; 63° 00� 00� W, 38° 36� 30� S). The Arroyo Chasicó Forma-
tion was formalized by Pascual (1961:63) based on the ‘Chasi-
coense’ of Kraglievich (1930) in reference to the sediments crop-
ping out along the margins of Arroyo Chasicó. Fidalgo and Porro
(in Bondesio et al., 1980; see also Fidalgo et al., 1987) described

FIGURE 1. A, location map with the Argentine Chasicoan and Huayquerian (late Miocene) localities bearing capybaras; B, detailed map of the
Arroyo Chasicó area with Bajada de los Toros locality; C, photograph of the right margin of Bajada de los Toros locality, white arrow points to the
fossil-bearing level; D, chronological chart of the late Miocene in southern South America.

JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 27, NO. 3, 2007684



two members for this formation, the lower Vivero Member and
the upper Las Barrancas Member (Fig. 1C). Overlying the Ar-
royo Chasicó Formation are the Pleistocene conglomerates of
the Bajada de los Toros Formation and the eolian sands of the
Estancia La Aurora Formation (Fidalgo et al., 1979). Zárate and
colleagues (2003) and Blasi and colleagues (2004) described four
lithofacies (S, F, FSC, and P) for the Arroyo Chasicó Formation.
At the base of the profile (S) are very fine sands with escorias
interpreted as impact glasses. Overlying are sandy silts (F) with
many reworked escorias, laminated mudstones (FSC), and sandy
silts and mudstones (P). The whole Arroyo Chasicó Formation
represents a fluvial depositional environment of channel bars
and areas of flood-plain and overflow with intervals of soil for-
mation (Zárate, 2005). The hydrochoerids studied here were re-

covered from mudstones and sandy siltstones composing litho-
facies association 3 (�facies FSC and P) of Arroyo Chasicó
Formation (Zárate pers. comm., July 2005). These sediments
were deposited under generally low energy conditions in a
swampy environment (Zárate et al., 2007).

The Arroyo Chasicó Formation—stratotype of the Chasicoan
Stage—represents the beginning of a new sedimentary cycle gen-
erally recognized as ‘Araucano’ (Doering, 1882; Ameghino,
1889) or an equivalent name (e.g. Araucanense, Estratos Arau-
canos; Rovereto, 1914), and later as ‘Edad de las planicies aus-
trales’ (Pascual and Bondesio, 1982). This cycle began with the
withdrawal of the ‘Mar Paranense,’ which flooded a widespread
surface from northern Patagonia (central Argentina) to northern
South America (Ramos, 1999) during the middle to late

FIGURE 2. Nomenclature (A, B, C) and measurements (D, E, F) of Cardiatherium chasicoense teeth. A and D, right p4; B and E, right m1 or m2;
C and F, right M1 or M2. Abbreviations: c.1e.–c.3e., 1st–3rd external column; c.1i.–c.3i., 1st–3rd internal column; h.1e.–h.3e., 1st–3rd external flexid;
h.1i.–h.5i., 1st–5th internal flexid; h.f.e., fundamental external flexid; H.F.I., fundamental internal flexus; H.P.E., primary external flexus; h.p.i.,
primary internal flexid; h.s.e., secondary external flexid; H.S.E., secondary external flexus; h.s.i., secondary internal flexid; h.s.i.a., secondary anterior
internal flexid; h.s.i.p., secondary posterior internal flexid; h.t.i., tertiary internal flexid; pr.I, prism I; pr.II, prism II; pr.s.a., anterior secondary prism.
Measurements: AP, anteroposterior length; AW, anterior width; AWa, prism Ia width; AWb, prism Ib width; HFEL, h.f.e. length; HPEL, H.P.E.
length; HSEL, H.S.E. length; HSIL, h.s.i. length; HTIL, h.t.i. length; MW, middle width; PW, posterior width.

TABLE 1. Revised taxonomy of the Hyayquerian (late Miocene) Hydrochoeridae based on lower cheek tooth morphology.

Systematics sensu Mones, 1991 Stratigraphic and geographic provenance Revision sensu Vucetich et al., 2005

Anchimys leydi Entre Rı́os Province; ‘conglomerado osı́fero,’ Cardiatherium paranense
Anchimys marshi Ituzaingó Formation (Mones, 1991 and references therein)
Cardiatherium paranense
Cardiatherium sp. A
Kiyutherium denticulatum
Kiyutherium scillatoyanei
Procardiatherium crassum
Procardiatherium simplicidens
?Kiyutherium rosendoi Catamarca Province; Puerta de Corral ?

Quemado (Bondesio, 1985)

Kiyutherium orientalis San José, Uruguay; Kiyú Formation (Francis and Mones, 1965b) Cardiatherium orientalis

Kiyutherium aff. orientalis A La Pampa Province; Cerro Azul Formation (Pascual and Bondesio,
1982)

Kiyutherium aff. orientalis B Rı́o Negro Province; Rı́o Negro Formation, capa d (Angulo and
Casamiquela, 1982)

Cardiatherium patagonicum*

Cardiatherium isseli Rı́o Negro Province; Rı́o Negro Formation, unknown levels
(Rovereto, 1914).

Cardiatherium isseli

Cardiatherium talicei San José, Uruguay; San José Formation (Francis and Mones,
1965a)

Cardiatherium talicei

*The holotype of C. patagonicum comes from Puerto Madryn Formation, Penı́nsula, Valdés, Chubut Province.

DESCHAMPS ET AL.—THE OLDEST CAPYBARAS 685



Miocene. The regression is coeval with the Quechua diastrophic
Phase of the Andean orogeny (Pascual and Bondesio, 1985), and
a global climatic amelioration (Janis, 1993). According to Pas-
cual (1984) this withdrawal began around 10.8 Ma. Radiometric
analysis of escorias of the basal levels of the Arroyo Chasicó

Formation at the type locality yielded a 40Ar/39Ar date of 9.23 ±
0.09 Ma (Schultz et al., 2004). The cuspidate palustrine facies
(FSC) at this area would have been deposited after ca. 9 Ma
(Zárate, 2005).

A biochronologic and biostratigraphic scheme for the late

FIGURE 3. Right mandible of Cardiatherium chasicoense, new comb, in occlusal (A, B) and lateral (C, D) views. A, C, holotype MMP 300-M; B,
D, MMH-CH 85-4-40. Scale bar equals 2 cm.

JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 27, NO. 3, 2007686



Miocene of central Argentina recently has been proposed on the
basis of an anagenetic sequence of octodontid rodents from the
Vivero Member of the Arroyo Chasicó Formation and different
localities of the Cerro Azul Formation, La Pampa Province
(Verzi et al., 2004, in press; Montalvo et al., 2005). In this
scheme, the oldest biozone is the Chasichimys bonaerense Zone
from the Vivero Member, followed by the Chasichimys scagliai
Zone in Laguna Chillhué (La Pampa), and, finally, the biozones
of the Late Huayquerian based on the genus Xenodontomys.
The Chasicoan capybaras were found in the upper levels of the
Las Barrancas Member (� FSC facies) overlying the Chasichi-
mys bonaerense Zone. C. orientalis was found in the Chasichimys
scagliai Zone (Fig. 1D). In the upper biozones of the Huayque-
rian, no capybaras have been found.

SYSTEMATIC PALEONTOLOGY

Order RODENTIA Bowdich, 1821
Suborder HYSTRICOGNATHI Tullberg, 1899

Family HYDROCHOERIDAE Gill, 1872
Genus CARDIATHERIUM Ameghino, 1883

Emended Diagnosis—Hydrochoerids with prisms of cheek
teeth undivided in adults; flexids variably developed according to
ontogeny; M3 composed of 6–11 prisms.

Type Species—Cardiatherium doeringi Ameghino, 1883.
Included Species—C. paranense (including C. doeringi), C.

patagonicum, C. orientalis, and C. chasicoense (see Table 1).
Age and Distribution—Late Miocene (Chasicoan and

Huayquerian SALMAs) of Argentina, Brazil, Peru, Uruguay,
Venezuela.

CARDIATHERIUM CHASICOENSE new combination
(Figs. 3–6, Table 2)

Procardiatherium chasicoense Pascual and Bondesio, 1968:238,
pl. 1 (figs. A–J).

?Cardiatherium dubium Pascual and Bondesio, 1968:244, 247,
249–250, pl. 1 (fig. K).

Procardiatherium dubium (Pascual and Bondesio) Mones,
1984:5.

Procardiatherium sp. Bondesio, Laza, Scillato-Yané, Tonni, and
Vucetich, 1980:110; Mones, 1991:25.

Holotype—MMP 300-M, fragment of right mandible with p4–
m3 (Fig. 3A, C).

Revised Diagnosis—p4 with h.p.i up to 50% of total width (up
to 70% in C. patagonicum, and 60% in C. orientalis and C.
paranense); h.2i. and h.3i. nearly similar in depth (this character
is rarely present in C. paranense and C. patagonicum); m1–2 with
h.s.i. only up to 17% of the total width and h.t.i. up to 31% (up
to 34% and 50%, respectively, in C. paranense, and 50% and
75% in C. patagonicum); anterior prism of m3 cordiform (heart-
shaped); P4–M2 with H.P.E. less than 50% of the transverse
diameter of anterior lobe, making the posterior lobe of Pr.I much
smaller than the anterior one and than those of C. patagonicum
(the most derived species of the genus); Pr.II somewhat labially
and posteriorly projected, unlike in C. patagonicum in which it is
anteroposteriorly thin; M3 with only six laminae (7–8 in C.
paranense and C. orientalis, 10–11 in C. patagonicum).

Hypodigm—The holotype; MLP 60-VI-18-37, left p4 (holo-
type of ?Cardiatherium dubium); MMP 305a-M, right M1/M2;
MMP 305b-M, left M1/M2; MMP 305c-M, right P4; MMP 306-M,
fragment of right m3; MMP 306 bis-M, fragment of right m1/m2;
MMP 307-M, right m3; MMP 308-M, right m1/m2; MMP 319-M,
right M1/M2, and left incisor; MMP 574-M, fragment of right M3;
MMP 583-M, right p4; MLP 55-IV-28-15, fragment of left m2,
fragment of right mandible with partial m2 and complete m3;
MLP 60-VI-18-39, right p4; MLP 55-IV-28-23, right P4?; MLP
60-VI-18-40, left M3; MLP 60-VI-18-45, fragment of left M3;
MLP 68-VIII-29-1, fragment of left mandible with p4–m2. Ad-
ditions to the hypodigm: MMP 979-M, left mandible with dam-
aged p4–m2; MMH-CH 85-4-40, almost complete right mandible
with i1, p4, m1, and m3; MMH-CH 88-6-71, pair of mandibles of a

FIGURE 4. Right (or left, reversed) p4 (A–J) and m1/2 (K–T) of Cardiatherium chasicoense (A–I, K–Q), C. orientalis (R), C. paranense (S), and
C. patagonicum (J, T). A, MMH-CH 83-3-60; B, MD-CH-06-235-1; C, MMP 300-M (holotype); D, MMH-CH 88-6-92e (reversed); E, MMH-CH
86-9-71b (reversed); F, MMH-CH 85-4-40; G, MMH-CH 86-9-118c (reversed); H, MMH-CH 88-6-71 (reversed); I, MLP 60-VI-18-37 (reversed, type
of Procardiatherium dubium); J, MPEF 740/9 (reversed); K, MMH-CH 86-9-71c; L, MMP 300-M (m1, holotype); M, MMP 300-M (m2, holotype);
N, MMH-CH 88-6-43; O, MLP 68-VIII-29-1 (reversed, m2); P, MMH-CH 85-4-40; Q, MMH-CH 87-7-104a (reversed); R, SPV-FHC-27-XI-64-20
(holotype of Kiyutherium orientalis); S, MLP 40-XI-15-1 (neotype); T, MPEF 740/24. Scale bar equals 5 mm.

DESCHAMPS ET AL.—THE OLDEST CAPYBARAS 687



single individual with left i1–m2 and right m1; MMH-CH 86-9-
71b, left p4.

Description

New materials from the Arroyo Chasicó Formation allow us to
expand the original description of the species made by Pascual
and Bondesio (1968).

Mandible—The mandible coincides mostly with the original
description, but two new mandibles (MMH-CH 85-4-40, Fig. 3B,
D; MMH-CH 88-6-71), more complete than the holotype, allow
a better description of some characters. Pascual and Bondesio
(1968) described the masseteric fossa of the holotype as wide and
uniformly extended anteroposteriorly, lacking a clear separation
between the anterior and posterior fossae, as in ‘P. simplicidens,’
‘Eucardiodon marshii,’ and ‘Anchimys leidyi,’ all of these now
regarded as juvenile specimens of C. paranense (Vucetich et al.,
2005). In the new mandibles (which are also larger than the
holotype) the masseteric fossa is clearly divided into anterior and
a posterior fossae. On the external surface, there is a triangular
surface between both (Fig. 3B). This morphology of the masse-
teric area is also seen in large specimens of Cardiatherium
paranense. Hence, the shape of the masseteric fossa of the ho-
lotype described by Pascual and Bondesio (1968) is a conse-
quence of its juvenile condition (see Fig. 3A, C). The juvenile
condition of the holotype can also be seen in the position of the
posterior margin of the symphysis; it is positioned anterior of p4
in the holotype, but is below the midpoint of p4 in larger (adult)
specimens such as MMH-CH 85-4-40 (Fig. 3C). In addition, the
mandible projects superiorly anterior of p4 in the holotype, the
anterior portion of the symphysis being much higher than the
level of the occlusal surfaces of the cheek teeth (Fig. 3C); in
adults, this portion of the mandible descends immediately in
front of p4, and the anterior portion of the symphysis is level with
the cheek teeth (Fig. 3D; see Vucetich et al., 2005).

Teeth—The size range of the available material is not as great
as in C. paranensis and C. patagonicum, but four new small speci-
mens, two right p4 (MMH-CH 83-3-60 and MD 06-235-1; Fig.
4A–B), one right m1/m2 (MMH-CH 86-9-71c; Fig. 4K), and one
right M3 (MD-CH-06-107/1; Fig. 6N) allow a better understand-
ing of ontogenetic change in cheek tooth morphology as well as
better comparisons with other species of Cardiatherium.

The p4 (Fig. 4A–I) has three prisms, the pr.s.a. being the most
variable. All the flexids are shallower than in the other species of
the genus (Fig. 4A–I vs. 4J); h.p.i. penetrates up to 50% the total
width in C. chasicoense, less than in C. patagonicum (70%) or C.
paranense and C. orientalis (60%). The h.4i. is variably devel-
oped from deep to shallow or even absent, especially in small
specimens (Fig. 4E–G). The h.3i. is similar in depth to the h.2i;
consequently, c.3i. is at the middle of an almost symmetric Pr.I.
The h.2e. can be transverse (especially in small specimens), ob-
lique or directed straight forward (Fig. 4G). Rarely, a small
h.sn.i. can be present (Fig. 4H).

The m1–2 (Fig. 4K–Q) are similar in shape. The h.s.i. and h.t.i.
are very shallow as compared to C. orientalis (Fig. 4R), C.
paranensis (Fig. 4S), and C. patagonicum (Fig. 4T). Both h.s.i.
and h.t.i. display little variation with increasing AP (Fig. 8), un-
like the other species in which these flexids increase markedly
with size (Vucetich et al., 2005). Even in the largest specimens of
C. chasicoense, the h.s.i. reaches only up to 17 %, and the h.t.i.
31% of the total width, whereas these values are 33% and 52%
in C. paranense, 19% and 52% in C. orientalis, and 46% and 78%
in C. patagonicum, respectively.

In m3 (Fig. 5A–F), pr.I is cordiform, similar to that of m1–2,
and has a short h.s.i.; this contrasts with the condition in C.
orientalis (Fig. 5G), C. paranensis (Fig. 5H–L) and C. patagoni-
cum (Fig. 5M–N) in which pr.I of m3 is semilunar with an h.s.i.
that deepens with size (� age). The h.t.i. extends only up to one

third of the tooth width, maintaining the triangular shape of
pr.IIa. The h.p.i. deepens with size, and in large specimens (Fig.
5D–F) it bifurcates.

In P4 (Fig. 6A–D) the H.P.E. penetrates up to 38% the total
width, whereas in C. orientalis (Fig. 6E) it is up to 55%, and C.
patagonicum (Fig. 6F) it is about 60%. In addition, this flexus
extends backward, dividing the Pr.I into a large anterior portion
and a very small posterior one, rather like a neck that joins Pr.II.
In C. patagonicum the H.P.E. is transverse to the AP axis, and
thus the posterior part is proportionally larger than in C. chasi-
coense; in C. orientalis it is intermediate (Fig. 6E–F). Pr.II is
anteroposteriorly thicker in C. chasicoense than in C. patagoni-
cum (Fig. 6A–D, F) and the portion behind the H.S.E. is some-
what labially and posteriorly projected as in C. orientalis (Fig.
6E) but shallower. In C. patagonicum H.S.E. is even deeper and
Pr.II is very curved, almost semilunar.

In M1 and M2 (Fig. 6G–L) both prisms are anteroposteriorly
narrower than in P4, and in Pr.II the labial and posterior margins
extend less than in P4. Both H.P.E. and H.S.E. are as in P4. C.
patagonicum has deeper H.S.E. and H.P.E. and Pr.II is narrow
and semilunar, as in P4. The upper M1 and M2 of C. paranense
and C. orientalis are still under study.

M3 (Fig. 6N–O) has six laminae, the first one with H.P.E. and
the last one short; C. orientalis has 7–8 laminae, and C. patagoni-
cum has ten (Fig. 6P). The M3 referred to C. paranense has been
lost (see Mones, 1991:27). The flexi deepen with age but even in
large specimens (Fig. 6O) they are not as deep as in C. patagoni-
cum (Fig. 6P). Each lamina is cordiform and anteroposteriorly
thicker than in the other species of the genus.

FIGURE 5. Right m3 (or left reversed) of Cardiatherium chasicoense
(A–F), C. orientalis (G), C. paranense (H–L), and C. patagonicum (M,
N); right m1–3 of Eocardia (O); and right m2–3 of Dolichotis (P). An-
terior to toward the top of the page. A, MMH-CH 88-6-41; B, MMP
300-M (holotype); C, MMH-CH 85-4-40; D, MMH-CH 88-6-71 (re-
versed); E, MMH-CH 88-6-92b (reversed); F, MLP 55-IV-28-15; G, SPV-
FHC-27-XI-64-20 (holotype of Kiyutherium orientalis); H, MLP 73-I-10-
7; I, MLP 61-VI-8-1 (reversed); J, MLP 78-II-27-1; K, MLP 61-VI-8-2; L,
MLP 40-XI-15-1 (neotype of C. paranense); M, MPEF 740/20; N, MPEF
740/7; O, MLP 15-217; P, MLP 247. Scale bar equals 5 mm.

JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 27, NO. 3, 2007688



Quantitative Analysis

Multivariate Analysis—A principal components analysis
(PCA) of eight linear measurements of m1/m2 of C. patagoni-
cum, C. paranense, C. orientalis and C. chasicoense was per-
formed for the size range in which variables increase linearly
(Fig. 7). The first principal component (PC) axis reflects mainly
size because variables had high positive loadings (Fig. 7, Table 2;
Lessa and Stein, 1992). In PC2, h.t.i. length (HTIL), loads high-
est, clearly discriminating C. chasicoense and C. patagonicum
from C. orientalis and C. paranense. This distribution along PC2
matches a chronological distribution of the species from oldest
(positive scores) to youngest (negative scores).

Regression Analysis—Bearing in mind the proposal of Vuce-
tich et al. (2005), we compared the scaling of HTIL and HSIL vs.
AP. In C. chasicoense these variables did not meet the assump-
tions of the regression analysis (e.g. homoscedasticity) because
the sample is biased toward large size (Table 2). Consequently,
we plotted the raw data and obtained a trendline through least-
squares. We compared the results with the regression lines of the
other species of the genus (Fig. 8).

The HTIL of C. chasicoense seems to behave differently from
the other species with respect to increasing size. In the smallest
specimen, HTIL shows values similar to those of the other spe-
cies at similar size, but in the rest, HTIL is comparatively shorter
than in the other species. Thus in C. chasicoense, h.t.i shows little
change with increasing size, at least within the range of the avail-
able sample. Likewise, HSIL shows little change with increasing
size, but values are within the range of those of C. paranense, C.
orientalis and C. patagonicum of similar size (Fig. 8).

DISCUSSION

Our studies show that all the hydrochoerids found in the Ar-
royo Chasicó Formation belong to a single species. ‘?C. dubium’
was based on a single specimen, and the characters used to dis-
tinguish it from ‘P.’ chasicoense (h.4i., h.2e. directed straight for-
ward) are, in fact, also present, but differently combined, within
the new sample (see Description). The original sample of ‘P.’
chasicoense was too small (five specimens including the holotype
of ?C. dubium; Pascual and Bondesio, 1968) to notice the p4
variability. On the other hand, and in the present state of the
knowledge, this species cannot be separated at the genus level
from younger species of Huayquerian Age (uppermost
Miocene). Some characters used to define the genus Procardia-
therium (e.g., h.3i. similar in depth to h.2i. of p4, and conse-
quently, c.3i. at the middle of an almost symmetric Pr.I.; see
Ameghino, 1885) are, in fact, observed in young specimens of
Cardiatherium paranense and in some specimens of C. patagoni-
cum (Vucetich et al., 2005; fig. 5D, G). In the case of the holo-
type of Procardiatherium chasicoense, the shape and depth of the
masseteric fossa reflect only differences in ontogenetic stage (it is
shallow in juveniles and deep in adults) rather than different
character states. This is supported by other juvenile characters
(Vucetich et al., 2005) such as the relatively anterior position of
the posterior part of the symphysis with respect to p4 (Fig. 3A,
B), and the upward projection of the mandibular diastema (Fig.
3C, D). Although C. chasicoense displays some characters that
differ from other late Miocene species (cordiform lobes, shal-
lower flexi/ids, cordiform Pr.I of m3, fewer prisms in M3) it is
more parsimonious to consider it a primitive species of the same
genus (see below) than generically distinct. In addition, the
cheek tooth morphology differs more dramatically between late
Miocene and Pliocene-Recent species (extreme depth of flexi/ids
dividing lobes, and multiplication of laminae of the M3) than
between C. chasicoense and other species of Cardiatherium. The

FIGURE 6. Occlusal view of right (or left reversed) P4 (A–F), M1/M2 (G–M), and M3 (N–P) of Cardiatherium chasicoense (A–D, G–L, N–O),
C. orientalis (E), and C. patagonicum (F, M, P). A, MMH-CH 88-6-92f; B, MMH-CH 88-6-39; C, MMH-CH 87-7-49 (reversed); D, MMH-CH 86-9-71f
(reversed); E, GHUNLPam 14452 (reversed); F, MPEF 740/22 (reversed); G, MLP 71-II-26-1; H, MMH-CH 87-7-104b; I, MMP 319; J, MMH-CH
88-6-92a; K, MLP 76-VI-12-98c (reversed); L, MMP 305a; M, MPEF 740/23; N, MD-CH-06-107/1; O, MMP 979-M (reversed); P, MPEF 740/18. Scale
bar equals 5 mm.

FIGURE 7. Plot of scores and factor loadings of PC1 and PC2 of
m1/m2 linear measurements. Polygons represent the ranges of variation
of species analyzed. Loadings > 0.7 are in bold. Abbreviations as in
Fig. 2.

DESCHAMPS ET AL.—THE OLDEST CAPYBARAS 689



discovery of additional specimens, especially skull and/or post-
cranianal remains, will permit testing this proposal against new
evidence.

Considering the general tendency in cheek teeth development
of euhypsodont cavioids from simple and bilobed (lower
Miocene Eocardia and the living Dolichotis; Fig. 5O, P) to mul-
tilaminar (Hydrochoerus), the morphology present in C. chasi-

coense turns out to be more primitive than that of C. paranense,
C. orientalis and C. patagonicum. All the flexi/ids of C. chasi-
coense are shallower than those of the other species of the genus.
The h.t.i. is especially enlightening in this sense. This flexid, a
novelty for the Superfamily Cavioidea (Vucetich et al., 2005), is
the one that grows most in the remaining species of the genus. In
the contrary, it stops growing or at least slows its growth in C.

TABLE 2. Dental measurements of Cardiatherium chasicoense, new. comb.

Cardiatherium
chasicoense

p4

AP AWa HSIL AWb HFEL MW HTIL PW

MMH-CH 83-3-60 6.42 2.30 0.03 2.40 2.50 3.20 3.80
MD 06-235-1 7.04 7.04 2.60 0.05 2.70 2.80 3.60 4.08
MMP 300-M Type 11.04 3.84 2.72 4.16 4.64 6.24 6.08
MMH-CH 88-6-92c 11.20 4.64 0.80 4.72 3.20 6.08 6.24
MLP 68-VIII-29-1 11.36
MLP 60-VI-18-39 11.52
MMH-CH 88-6-92e 11.84 5.12 0.048 5.20 4.32 6.24 7.20
MMH-CH 86-9-71b 12.16 4.80 1.12 4.96 4.16 6.40 7.68
MMH-CH 88-6-45 12.16 4.64 0.24 4.80 4.00 6.24 6.56
MMH-CH 85-4-40 12.16 4.80 0.32 4.96 4.16 6.08 6.88
MLP 60-VI-18-38b 12.32 4.16 0.48 5.28 4.00 6.24 7.20
MMH-CH 86-0-118c 12.64 3.84 0.64 4.00 4.16 5.92 6.56
MLP 60-VI-18-38a 12.96 4.96 0.32 5.28 4.48 6.40 6.40
MMH-CH 88-6-71 13.44 4.32 16.80 4.48 3.52 5.92 6.96
MLP 76-VI-12-98a 4.64 4.32 3.84 5.76 6.56
MLP 76-VI-12-98b 13.60 4.64 0.64 4.96 3.52 5.60 7.20
MLP 60-VI-18-37 13.76 4.80 1.44 5.28 3.36 6.88 7.36

m1 or m2

MMH-CH 86-9-71c 6.08 3.44 0.64 3.44 2.96 3.36 1.60
MMP 300-M m1 Type 9.60 7.20 1.28 6.40 4.00 5.60 1.60 6.72
MMP 300-M m2 Type 9.60 6.72 1.60 6.40 4.32 5.76 1.76 6.88
MMP 308-M 9.92 6.72 0.72 6.24 5.28 5.60 1.28 7.04
MMH-CH 88-6-71 m1 10.24 7.04 1.36 6.56 6.08 5.60 2.08 7.36
MMH-CH 88-6-43 10.40 7.52 1.60 7.36 6.72 6.24 2.56 7.60
MMH-CH 88-6-92d 10.40 1.12 5.92 1.84 7.52
MMH-CH 87-7-104a 11.52 7.68 1.28 7.36 6.08 6.80 1.84 7.52
MMH-CH 88-6-71 m2 10.56 7.20 1.44 6.56 5.60 5.60 1.60 7.04
MMH-CH 88-6-40 10.56 7.52 0.96 6.56 5.60 5.76 2.08 7.20
MMH-CH 88-6-92g 10.72 7.52 1.12 7.36 6.88 6.40 2.56 7.84
MLP 68-VIII-29-1 m1 10.72 6.72 0.96 6.40 3.84 5.60 2.24 6.56
MLP 68-VIII-29-1 m2 10.72 6.56 1.12 6.40 4.32 5.28 2.88 6.40
MMH-CH 85-4-67 10.88 7.36 0.64 6.88 5.76 6.24 1.92 7.60
MMH-CH 85-4-40 10.88 7.68 1.52 7.20 6.24 6.56 2.72 7.52
MMH-CH 87-7-68 7.36 1.44 7.36 7.04 6.24 2.40
MMP 306 b-M 5.92 1.60

m3

MMP 300-M Type 6.24 2.24 7.04 6.08 8.00 3.84
MMH-CH 85-4-40 14.08 7.52 1.92 8.16 7.36 8.64 3.68 8.96
MMH-CH 88-6-41 14.08 6.08 1.92 7.52 6.72 8.32 3.52 8.96
MMH-CH 88-6-71 14.88 7.04 1.76 7.36 6.88 8.00 3.68 9.44
MLP 55-IV-28-15 14.88 7.68 2.08 8.00 7.20 8.48 3.52 8.80
MMH-CH 88-6-92b 15.52 7.68 1.44 8.32 7.68 8.48 4.32 9.44
MMP 285-M 7.04 1.44 7.68 4.00
MMP 307-M 1.28 7.52 7.04 8.00 3.20 8.32
MMP 306-M 8.48 3.48 9.12

P4

AP AW HPEL HSEL PW

MMP 319-M 8.00 7.20 2.72 1.44 7.68
MMP 305 c-M 8.64 7.20 2.40 1.12 7.52

M1 or 2

MMP 319-M 8.16 7.68 2.88 1.60 8.32
MMP 305 b-M 8.48 6.72 2.72 1.28 7.68
MMP 305 a-M 9.28 7.20 3.04 1.44 8.00

M3

MD-CH-06-107/1 12.75 5.10
MLP 60-VI-18-40 19.36 8.00
MLP 71-II-26-1 19.20 7.20
MMP 979-M 20.16 7.36

JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 27, NO. 3, 2007690



chasicoense. Such a growth pattern is expected based on our
model of ontogenetic change for a species older than those pre-
viously analyzed. The primitive condition of the cheek teeth of
C. chasicoense is also seen in the shape of the anterior prism of
m3. This prism is similar to those of m1–2 (Fig. 5A–F vs. Fig.
4L–Q; see also Fig. 3A, B), equivalent to the condition seen in
Eocardia and Dolichotis (Fig. 5O, P). Whereas in the other spe-
cies of the genus this prism is semilunar (Fig. 5G–N vs. Fig.
4R–T), different from that of m1–2.

Modern capybaras inhabit areas around ponds, lakes, rivers,
marshes and swamps, using water primarily as refuge (Nowak
and Paradiso, 1983). The geographic distribution of late Miocene
capybaras is scattered, restricted to certain localities. All the
lithological fossil bearing units have been interpreted as depos-
ited in water-related settings. The unit that produces C. chasi-
coense is a swampy environment (Zárate et al., in press). Simi-
larly, C. orientalis from Laguna Chillhué (La Pampa Province) is
found in lacustrine levels of the Cerro Azul Formation (Mon-
talvo et al., 2005). The ‘conglomerado osífero’ at the base of the
Ituzaingó Formation (Entre Ríos Province) in which C.
paranense is found is a discontinuous fluvial deposit formed by
sands and conglomerates that filled the previously irregular to-
pography of the Paraná Formation (Herbst, 2000). C. patagoni-
cum is found in levels deposited in a freshwater channel, not far
from the sea (Cione et al., 2005:36). All these environments
would have been quite propitious for capybaras. This fact may
explain their absence from other rich fossil-bearing sites of this
age of La Pampa and southwest Buenos Aires provinces; these
depositional environments are mostly eolian, probably unsuit-
able for capybaras.

The revision of the family with our new approach (Vucetich et
al., 2005) sheds light on the supposed high diversity of capybaras;
instead of demonstrating the high specific diversity suggested by
typological criteria used in the classical systematics, we interpret
capybaras as exhibiting high morphological and ontogenetic in-
traspecific diversity. This results in the recognition of only one
species at each locality and thus low local taxonomic diversity.
(analyses in progress suggest this is also valid for the ‘conglo-
merado osífero’; see Vucetich et al., 2005). According to our
proposal, the species of Cardiatherium probably represent dif-
ferent evolutionary stages from the most primitive—C. chasi-
coense—to the most derived—C. patagonicum—with C.
paranense and C. orientalis being intermediate; this trend con-
tinues in the Pliocene with another genus (i.e., Phugatherium =
Chapalmatherium; Vucetich et al., 2005). The relationships
among these taxa have yet to be analyzed within the context of
the whole family, however. Although there is not a single strati-
graphic sequence with different levels to test this, it is now rec-
ognized that species do not occur together in the same levels (or
locality). The species within the new systematic framework may
turn out quite useful for biocorrelation, especially in some late
Miocene localities of Argentina, which are far from being com-
pletely understood from a biostratigraphic point of view. In this
sense, the capybaras suggest that the Las Barrancas Member of
the Arroyo Chasicó Formation is older than the sediments of the
Cerro Azul Formation cropping out at Laguna Chillhué, which,
according to the record of octodontoid rodents, could be consid-
ered coeval (see Verzi, 1999; Fig. 1D).

We consider that our proposal of tooth growth pattern (Vuce-
tich et al., 2005) provides a useful basic framework to study
Chasicoan and Huayquerian capybaras. Both the related cardio-
myines of the late Miocene—their putative sister group (Vuce-
tich et al., 2005)—and their Pliocene descendants will be studied
within this framework in order to understand the systematics and
the evolutionary patterns of the whole group.

ACKNOWLEDGMENTS

We thank Dr. D. Verzi for fruitful comments; V. Di Martino
(director of the MMH), A. Dondas (MMP), M. Reguero (MLP),
and T. Manera (MD) for access to material under their care. C.
Montalvo and J. Urrutia kindly lent us unpublished materials.
UNLP 11/N442 and CONICET PIP 5242 grants supported this
research. Two anonymous reviewers provided thorough com-
ments for improvements to the manuscript.

LITERATURE CITED

Ameghino, F. 1883. Sobre una nueva colección de mamíferos fósiles
recogidos por el Profesor Pedro Scalabrini en las barrancas del
Paraná. Boletín de la Academia Nacional de Ciencias en Córdoba
5:257–306.

Ameghino, F. 1885. Nuevos restos de mamíferos fósiles oligocenos
recogidos por el Profesor Pedro Scalabrini y pertenecientes al Mu-
seo Provincial de la ciudad de Paraná. Boletín de la Academia Na-
cional de Ciencias en Córdoba 8:5–207.

Ameghino, F. 1889. Contribución al conocimiento de los mamíferos fó-
siles de la República Argentina. Actas de la Academia Nacional de
Ciencias en Córdoba 6:1–1027; atlas, 98 pls.

Angulo, R. J., and R. M. Casamiquela. 1982. Estudio estratigráfico de las
unidades aflorantes en los acantilados de la costa norte del Golfo de
San Matías (Río Negro y extremo austral de Buenos Aires) entre los
meridianos 62° 30� y 64° 30� W. Mundo Ameghiniano 2:20–73.

Blasi, A., M. A. Zárate, and P. Schultz. 2004. Sedimentación de la For-
mación Arroyo Chasicó (Mioceno tardío), Buenos Aires, Argentina.
X Reunión Argentina de Sedimentología, Resúmenes: 34–35.

Bondesio, P. 1985. Un nuevo roedor Cardiatheriinae (Hydrochoeridae)
del Departamento de Belén, Provincia de Catamarca: inferencias
paleoambientales: probable edad de los sedimentos portadores.
Ameghiniana 21:280–284.

Bondesio, P., J. H. Laza, G. J. Scillato Yané, E. P. Tonni, and M. G.

FIGURE 8. Model II regressions (of log-transformed data) of tertiary
internal flexid length (HTIL; above) and secondary internal flexid length
(HSIL; below) on antero-posterior length (AP) for Huayquerian hydro-
choerids (modified from Vucetich et al., 2005). Raw data (diamonds) for
Cardiatherium chasicoense are included.

DESCHAMPS ET AL.—THE OLDEST CAPYBARAS 691



Vucetich. 1980. Estado actual del conocimiento de los vertebrados
de la Formación Arroyo Chasicó (Plioceno temprano) de la Provin-
cia de Buenos Aires. II Congreso Argentino de Paleontología y
Bioestratigrafía y I Congreso Latinoamericano Paleontología
(1978), Actas 3:101–127.

Bookstein, F. L., B. Chernoff, R. L. Elder, J. M. Humphries, G. R. Smith,
and R. E. Strauss. 1985. Morphometrics in evolutionary biology.
Special Publication Academy of Natural Sciences Philadelphia 15:
1–277.

Bowdich, T. E. 1821. An analysis of the natural classifications of Mam-
malia for the use of students and travellers. J. Smith, Paris, 115 pp.

Cione, A. L., M. M. Azpelicueta, J. R. Casciotta, and M. T. Dozo. 2005.
Tropical freshwater teleosts from Miocene beds of eastern Patago-
nia, southern Argentina. Geobios 38:29–42.

Doering, A. 1882. Geología; pp. 299–530 in Informe oficial de la
comisión científica agregada al Estado Mayor General de la Expe-
dición al Río Negro (Patagonia) realizada en los meses de abril,
mayo y junio de 1879 bajo las órdenes del General D. Julio A. Roca.
Imp. Ostwald, Buenos Aires.

Fidalgo, F., J. H. Laza, N. Porro, and E. P. Tonni. 1979. Algunas carac-
terísticas de la Formación Arroyo Chasicó y sus relaciones geológi-
cas. VII Congreso Geológico Argentino (1978), Actas 1:213–225.

Fidalgo, F., E. P. Tonni, N. Porro, and J. H. Laza. 1987. Geología del área
de la Laguna Chasicó (Partido de Villarino, Provincia de Buenos
Aires) y aspectos bioestratigráficos relacionados. Revista de la Aso-
ciación Geológica Argentina 42:407–416.

Francis, J. C., and A. Mones. 1965a. Sobre el hallazgo de Cardiatherium
talicei n. sp. (Rodentia, Hydrochoeridae) en la playa Kiyú, Departa-
mento de San José, República Oriental del Uruguay. Kraglieviana
1:3–44.

Francis, J. C., and A. Mones. 1965b. Sobre el hallazgo de Kiyutherium
orientalis n.g., n.sp. (Rodentia, Hydrochoeridae) en la Formación
Kiyú, de Barrancas de San Gregorio, Departamento de San José,
República Oriental del Uruguay. Kraglieviana 1:45–54.

Frailey, C. D. 1986. Late Miocene and Holocene mammals, exclusive of
the Notoungulata, of the Rio Acre region, Western Amazonia. Con-
tribution in Sciences, Natural History Museum of Los Angeles
County 374:1–46.

Gill, T. 1872. Arrangement of the families of mammals with analytical
tables. Smithsonian Miscellaneous Collections 11:1–98.

Herbst, R. 2000. La Formación Ituzaingó (Plioceno). Estratigrafía y dis-
tribución; pp. 181–190 in F. G. Aceñolaza and R. Herbst (eds.), El
Neógeno de Argentina. Instituto Superior de Correlación Geológica
INSUGEO, San Miguel de Tucumán.

Janis, C. M. 1993. Tertiary mammal evolution in the context of changing
climates, vegetation, and tectonic events. Annual Review of Ecology
and Systematics 24:467–500.

Kraglievich, L. 1930. La Formación Friaseana del río Frías, río Fénix,
Laguna Blanca, etc., y su fauna de mamíferos. Physis 10:127–161.

Legendre, P., and L. Legendre. 1998. Numerical ecology (Second Ed.).
Elsevier, Amsterdam, 853 pp.

Lessa, E. P., and B. R. Stein. 1992. Morphological constraints in the
digging apparatus of pocket gophers (Mammals: Geomyidae). Bio-
logical Journal of the Linnean Society, London 47:439–453.

Mones, A. 1984. Estudios sobre la Familia Hydrochoeridae, XIV: re-
visión sistemática. Senckenbergiana Biologica 65:1–17.

Mones, A. 1991. Monografía de la familia Hydrochoeridae (Mammalia,
Rodentia): sistemática–paleontología–bibliografía. Courier For-
schungsinstitut Senckenberg 134:1–235.

Montalvo, C. I., D. H. Verzi, and C. M. Deschamps. 2005. Biochronology
and biostratigraphy of the upper Miocene of central Argentina: evi-
dence from rodents and taphonomy. II International Meeting
TAPHOS 2005 and IV Reunión de Tafonomía y Fosilización. Ab-
stracts:119–120.

Nowak, R. M., and J. L. Paradiso. 1983. Walker’s Mammals of the World,
Vol. II. The John Hopkins University Press, Baltimore and London,
1362+15 pp.

Pascual, R. 1961. Un nuevo Cardiomyinae (Rodentia, Caviidae) de la
Formación Arroyo Chasicó (Plioceno inferior) de la Provincia de
Buenos Aires. Ameghiniana 2:61–71.

Pascual, R. 1984. Late Tertiary mammals of southern South America as
indicators of climatic deterioration. Quaternary of South America
and Antarctic Peninsula 2:1–30. A. Balkema, Rotterdam.

Pascual, R., and P. Bondesio. 1968. Los Cardiatheriinae (Rodentia, Ca-
viomorpha) de la Formación Arroyo Chasicó (Plioceno inferior) de
la provincia de Buenos Aires. Ameghiniana 5:237–251.

Pascual, R., and P. Bondesio. 1982. Un roedor Cardiatheriinae (Hydro-
choeridae) de la Edad Huayqueriense (Mioceno tardío) de La
Pampa: sumario de los ambientes terrestres en la Argentina durante
el Mioceno. Ameghiniana 19:19–35.

Pascual, R., and P. Bondesio. 1985. Mamíferos terrestres del Mioceno
Medio-Tardío de las Cuencas de los ríos Colorado y Negro (Argen-
tina): evolución ambiental. Ameghiniana 22:133–145.

Ramos, V. 1999. Las provincias geológicas del territorio argentino; pp.
41–96 in R. Caminos (ed.), Geología Argentina. Anales Servicio
Geológico Minero Argentino 29, Instituto de Geología y Recursos
Minerales, Buenos Aires.

Rovereto, C. 1914. Los estratos araucanos y sus fósiles. Anales del Mu-
seo Nacional de Historia Natural 25:1–247.

Schultz, P., M. A. Zárate, B. Hames, C. Koeberl, T. Bunch, D. Storzer,
P. Renne, and J. Wittke. 2004. The Quaternary impact record from
the Pampas, Argentina. Earth and Planetary Science Letters 219:
221–238.

Tullberg, T. 1899. Über das System der Nagethiere, eine phylogenetische
Studie. Nova Acta Regiae Societatis Scientarium Upsalensis 3(18):
1–514.

Verzi, D. H. 1999. The dental evidence on the differentiation of the
ctenomyine rodents (Caviomorpha, Octodontidae, Ctenomyinae).
Acta Theriologica 44:263–282.

Verzi, D. H., C. M. Deschamps, and C. I. Montalvo. 2004. Bioestrati-
grafía y biocronología del Mioceno Tardío de la Argentina central.
Ameghiniana 41(4, Supplement):21R.

Verzi, D. H., C. I. Montalvo, and C. M. Deschamps. In press. Biostratig-
raphy and biochronology of the Late Miocene of central Argentina:
evidence from rodents and taphonomy. Geobios.

Vucetich, M. G., C. M. Deschamps, A. I. Olivares, and M. T. Dozo. 2005.
Capybaras, size, shape, and time: a model kit. Acta Palaeontologica
Polonica 50:259–272.

Woods, C. A., and W. Kilpatrick. 2005. Infraorder Hystricognathi; pp.
1538-1600 in D. E. Wilson and D. M. Reeder (eds.), Mammal Spe-
cies of the World: A Taxonomic and Geographic Reference, Third
Ed., Vol 2. The Johns Hopkins University Press, Baltimore, Mary-
land.

Zárate, M. A. 2005. El Cenozoico tardío continental de la Provincia de
Buenos Aires; pp. 139–158 in R. E. De Barrio, R. O. Etcheverry,
M. F. Caballé, and E. Llambías (eds.), Geología y Recursos Miner-
ales de la Provincia de Buenos Aires. Relatorio XVI Congreso
Geológico Argentino, Quick Press, La Plata.

Zárate, M. A., A. Blasi, P. Schultz, C. Heil, and J. King. 2003. Litofacies
y paleoambientes de la Formación Arroyo Chasicó (Buenos Aires):
implicancias bioestratigráficas. Ameghiniana 40(4, Supplement):
97–98R.

Zárate M. A., P. H. Schultz, A. Blasi, C. Heil, J. King, and W. Hames.
2007. Geology and geochronology of the type Chasicoan (late
Miocene) mammal-bearing deposits of Buenos Aires Province (Ar-
gentina). Journal of South American Earth Sciences 23:81–90.

Submitted December 18, 2006; accepted February 15, 2007.

JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 27, NO. 3, 2007692