See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/249551916

A new 'South American ungulate' (Mammalia: Litopterna) from the Eocene of

the Antarctic Peninsula

Article  in  Geological Society London Special Publications · January 2006

DOI: 10.1144/GSL.SP.2006.258.01.12

CITATIONS READS

45 1,001

4 authors:

Mariano Bond Marcelo A. Reguero

Universidad Nacional de La Plata Instituto Antártico Argentino
94 PUBLICATIONS   3,050 CITATIONS    279 PUBLICATIONS   4,464 CITATIONS   

SEE PROFILE SEE PROFILE

Sergio Fabián Vizcaíno Sergio A. Marenssi

Universidad Nacional de La Plata National Scientific and Technical Research Council
234 PUBLICATIONS   6,759 CITATIONS    147 PUBLICATIONS   4,113 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Body size, allometry View project

Evolution and diversity of Antarctic Eocene whales View project

All content following this page was uploaded by Marcelo A. Reguero on 07 January 2014.

The user has requested enhancement of the downloaded file.

12_SP258 Bond (to_d) 25/5/06 9:18 am Page 163

A new ‘South American ungulate’ (Mammalia: Litopterna) from

the Eocene of the Antarctic Peninsula
M. BOND1, M. A. REGUERO1, S. F. VIZCAÍNO1 & S. A. MARENSSI2
1División Paleontología Vertebrados, Museo de La Plata, Paseo del Bosque s/n, 1900 La
Plata, Argentina (e-mail: regui@fcnym.unlp.edu.ar)
2Instituto Antártico Argentino, Cerrito 1248, 1010 Buenos Aires, Argentina

Abstract: Notolophus arquinotiensis, a new genus and species of the family Sparnotheri-
odontidae (Mammalia, Litopterna), is represented by several isolated teeth from the
shallow-marine sediments of the La Meseta Formation (late Early–Late Eocene) of
Seymour Island, Antarctic Peninsula, which have also yielded the youngest known sudame-
ricids and marsupials. The new taxon belongs to the extinct order of ‘South American
native ungulate’ Litopterna characterized by the convergence of the later forms with the
equids and camelids. Notolophus arquinotiensis shows closest relationships with Victor-
lemoinea from the Itaboraian (middle Palaeocene) of Brazil and Riochican–Vacan (late
Palaeocene–early Eocene) of Patagonia, Argentina. Although still poorly documented, this
new taxon shows that the early Palaeogene Antarctic faunas might provide key data
concerning the problems of the origin, diversity and basal phylogeny of some of the ‘South
American ungulates’ (Litopterna). This new taxon shows the importance of Antarctica in
the early evolution of the ungulates and illustrates our poor state of knowledge.

Initial palaeontological work in early 1980 on Litopterna is considered a natural group of

Seymour Island produced a modest assemblage South American native ungulates. Miocene–
of terrestrial fossil mammals (marsupials and Pleistocene forms show a notable convergence
South American ungulates). During the with equids (Proterotheriidae) and camelids
1989–1990 season, geologists of the Instituto (Macraucheniidae). One of the most unusual of
Antártico Argentino, while mapping Eocene the litopterns was the Pleistocene camel-like
marine rocks in Seymour Island, discovered Macrauchenia with large size and proboscis. The
small- and medium-sized land mammals, includ- early Palaeogene forms (Palaeocene–Eocene)
ing two representatives of the South American show morphological resemblances with the
native ungulates, Litopterna and Astrapotheria ‘ancestral ungulates’, the ‘condylarths’. Spar-
(Marenssi et al. 1994 and see also Hooker 1992). notheriodontids were medium- to large-sized
The Antarctic litoptern was referred by Bond ungulates. The family is known in the middle
et al. (1990) to the eolitoptern sparnotheriodon- Palaeocene Itaboraian South American Land
tid genus Victorlemoinea. Renewed field efforts Mammal Age (SALMA) of Brazil and the late
on Seymour Island (1992–2000) greatly Palaeocene Riochican SALMA of Patagonia,
enhanced the original collection and the sites are and survived through at least the Late Eocene
known to contain a high number of sparnotheri- (Divisaderan SALMA) of Mendoza, Argentina.
odontids, as well as many other taxa previously The species of sparnotheriodontids are classified
unknown from the area (Reguero et al. 2002). in three genera and are listed in the Table 1.
This new material allows us to reinterpret the The fossil record of the family Sparnotheri-
teeth initially attributed to Victorlemoinea. odontidae in South America is rather sparse;

Table 1. Sparnotheriodontid species formally recognized

Species Geographic location Age Source

Victorlemoinea labyrinthica Cañadón Vaca, Chubut Riochican Ameghino 1901

Victorlemoinea prototypica Itaboraí, Brazil Itaboraian Paula Couto 1952
Sparnotheriodon epsilonoides Cañadón Vaca, Chubut Vacan Soria 1980a
Phoradiadus divortiensis Divisadero Largo, Mendoza Divisaderan Simpson et al. 1962
Sparnotheriodontidae?
Heteroglyphis dewoletzky Cerro del Humo, Chubut Mustersan Roth 1899

Species of Sparnotheriodontidae known in South America (Argentina and Brazil). Data from Soria (2001).

From: FRANCIS, J. E., PIRRIE, D. & CRAME, J. A. (eds) 2006. Cretaceous–Tertiary High-Latitude
Palaeoenvironments, James Ross Basin, Antarctica. Geological Society, London,
Special Publications, 258, 163–176. 0305–8719/06/$15 © The Geological Society of London 2006.
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 164

164 M. BOND ET AL.

sparnotheriodontids are seldom common in any Table 2. Dimensions of sparnotheriodontid teeth

given locality, even during the late Palaeocene from Seymour Island, Antarctica. See abbreviations in
(Riochican SALMA), when the group reached the text
its climax.
Specimen L W
The taxonomy of Sparnotheriodontidae (mm) (mm)
remains contentious and is currently based
solely on teeth. This family has long been the MLP 90-I-20-1 20.00 c. 20.00
subject of discussion over its systematic MLP 90-I-20-3 15.80 12.70
position, generic content and nomenclatural M LP 90-I-20-5 c. 10.20 c. 100.0
priorities. The main causes of these problems MLP 91-II-4-1 21.70 12.60
are the uniformity of its dental morphology MLP 91-II-4-5 10.90 6.70
(taxonomic differences are often minor and MLP 92-II-2-135 –.00 –.00
MLP 94-III-15-3 10.80 8.00
easily confused with intraspecific variation) and
MLP 95-I-10-6 25.60 c. 25.00
the poor quality of type specimens. Originally MLP 96-I-5-9 12.45 10.40
considered by Ameghino (1901) as a member MLP 96-I-5-10 17.20 13.80
of the meniscotheriid condylarths, Simpson MLP 01-I-1-1 31.00 16.80
(1945, 1948) regarded Victorlemoinea as a prim- MLP 04-III-3-1 17.50 13.50
itive Macraucheniidae (Litopterna). Later, this
genus was included in the enigmatic family
Sparnotheriodontidae (Soria 1980b, 2001;
Cifelli 1983a, b, 1993). Morphological evidence
suggests that sparnotheriodontids are most Argentino and División Paleontología de Verte-
closely related to other primitive litopterns brados, Museo de La Plata localities designated
such as the eolitoptern Anisolambdidae (Hoff- by ‘IAA’ and ‘DPV’, respectively. Dry sieving
stetter & Soria 1986; Soria 2001; Anisolambdi- and surface crawling were the primary tech-
nae of Cifelli 1983b). However, other niques for specimen collection.
morphological studies, based on tarsals, argue All measurements are reported in mm
that the Sparnotheriodontidae belongs to the (Table 2). Terminology and measurements for
Didolodontoidea, a group included in the para- litoptern teeth follow Nessov et al. (1998) and
phyletic Condylarthra (Cifelli 1983a, b, 1993). Soria (2001).
As the association of tarsal and dental elements
that supports this last statement is not clear we Institutional abbreviations
follow here Soria (2001), treating the
Sparnotheriodontidae as eolitopterns closely AMNH, American Museum of Natural History,
related to the Anisolambdidae. New York, USA; DGM, Divisao de Geologia e
The new sparnotheriodontid sample is Mineralogia do Departamento Nacional da
important for a number of reasons: (1) it is Producao Mineral, Rio do Janeiro, Brazil;
valuable for systematic evaluation of previously MACN, Museo Argentino de Ciencias Natu-
collected specimens; (2) it can be used to test rales ‘Bernardino Rivadavia’, Buenos Aires,
previous hypotheses about the age of the Argentina; MLP, Museo de La Plata, La Plata,
terrestrial mammal-bearing horizons of La Argentina; MNRJ, Museu Nacional do Rio de
Meseta Formation; and (3) it provides for a Janeiro, Brazil.
more complete assessment of the bio-
geographic associations of the La Meseta Material
terrestrial fauna.
Comparisons to other sparnotheriodontid taxa
Material and methods were made using the following specimens:
Victorlemoinea prototypica, MNRJ 1470-V
Comparisons were made with specimens in the (holotype), right M3, MNRJ 1471-V (paratype),
Vertebrate Palaeontology collections of the left M3, MNRJ 1472V, left M3, MNRJ 1477V,
Museo de La Plata (MLP), Museo Argentino de right M1 or M2 (DP4?), MNRJ 1481V, right p3,
Ciencias Naturales ‘Bernardino Rivadavia’ MNRJ 1484, left M1, MNRJ 1487V, left p3,
(MACN), Museo Nacional of Rio de Janeiro MNRJ 1402V, right m2, MNRJ 1488-V, left p3,
(MNRJ) and the American Museum of Natural DGM 268-M, left dp3-m1?, AMNH 49816, left
History (AMNH). All Seymour Island speci- M3; Victorlemoinea sp., AMNH 28465, left m1
mens are housed in the Vertebrate Palaeontol- or m2; AMNH 28466, left M1 or M2, AMNH
ogy collection of the MLP. All listed specimens 28467, right m3; AMNH 28468, left M1; AMNH
were collected from Instituto Antártico 28508, right p2?; AMNH 28515, right upper
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 165

‘SOUTH AMERICAN UNGULATE’ FROM ANTARCTICA 165

premolar; AMNH 27895, right M3; MLP 61- lophoselenodont and bicrescentic, dp4 fully
VIII-3-163, fragmentary right upper molar; molarized, with talonid and trigonid subequal
Victorlemoinea labyrinthica, MACN A-10671 (Sparnotheriodon) or trigonid somewhat
(type), left P4-M1?; Victorlemoinea emarginata, smaller than the talonid (Notolophus), with
MACN A-10672 (type), right M1–M2; ?Victor- well-developed labial (ectoflexid) and lingual
lemoinea longidens, MACN A-10670 (type), (meta and entoflexid) flexids. Trigonid with a
right m1–m2?; Sparnotheriodon epsilonoides, very well developed paralophid, its lingual end
MACN 18225 (holotype), incomplete lower jaw with a cuspid (paraconid? or ?neoparaconid)
with left and right i1–m3; Victorlemoinea sp., rapidly coalescent with wear. Metaconid high,
MLP 66-V-12-1, right M3, MLP 66-V-12-2, right but especially conspicuous on m1–m3 (e.g.
DP4-M2; Phoradiadus divortiensis, MACN Sparnotheriodon); lingual wall of the meta-
18061 (type), right M2–M3; MLP 87-III-20-7, conid flattened, with descending crest enclosing
left p4?; MLP 87-III-20-16, fragmentary rostrum part of the talonid basin that is more conspicu-
with left and right I3–P3; MLP 87-III-20-17, ous on p3–p4. Talonid with cristid obliqua
right P4-M3; MLP 87-III-20-39, left m3; MLP connected to the lingual end of the metalophid
87-III-20-71, right P1–P3; MLP 87-III-20-72, (metaconid). Entoconid very small (Sparnothe-
very damaged skull and lower jaw with right riodon) to well developed (Phoradiadus) and
P2–M3 and m2–m3 preserved; Heteroglyphis coalescent at the base with the hypoconulid
dewoletzky, MLP 12-1462 (type) left upper (e.g. Sparnotheriodon). The m3 is larger than
molariform. the m1 and m2, with talonid of m3 subequal to
the trigonid or longer and narrower than the
Systematic palaeontology trigonid with a posteriorly projecting
hypoconulid (e.g. Notolophus).
Class MAMMALIA Linnaeus, 1758 I1–I3 with lingual cingulum well developed.
Grandorder UNGULATA Linnaeus, 1766 I3 equal or larger than the I1–I2. C1 very well
Order LITOPTERNA Ameghino, 1889 developed, robust, similar to those of the
Suborder EOLITOPTERNA Soria, 2001 Isotemnidae notoungulates, with sharp anterior
Family SPARNOTHERIODONTIDAE Soria, and posterior crests (Phoradiadus). P1–P4 with
1980a labial and lingual cingula, variably developed,
continuous or not. P1 simple, enlarged antero-
Emended diagnosis (after Soria 1980a) posteriorly, with a single labial cusp, single
rooted but bilobed lingually. P2–P4 increasingly
Medium-sized (e.g. Phoradiadus) to large-sized complex and expanded transversally. P2–P3
(e.g. Sparnotheriodon) litopterns. Complete and with a labial (paracone) cusp showing no or very
closed dental series, i3/3, c1/1, p4/4, m3/3; teeth little differentiation of the metacone and a well-
brachyodont, lophobunoselenodont to lophose- developed anterior parastyle. P2 with a very
lenodonts; i1–i3 relatively robust, foliform or small protocone with anterior and posterior
spatuliform, increasing in size posteriorly crests enclosing a basined trigon. The P3 is more
(i1<i2<i3). The i3 is non-caniniform (e.g. complex, with a well-developed protocone, high
Sparnotheriodon, Phoradiadus), lingual and and enclosing with the paraloph and metaloph
labial cingula variably developed. The c1 are a central fossette in the trigon basin. Paraloph
enlarged and conical (similar to those of the connected to the ectoloph, with one or two
notoungulate Isotemnidae), with anterior and cuspules trending lingually to the trigon basin
posterior crests, normally obliterated by wear, from the ectoloph. Protostyle variably
especially the anterior one; with labial and developed in the anterolingual cingulum. The
lingual cingula. Lower cheek-teeth (p1–m3) P4 is molariform, with a metacone well differen-
with labial and lingual cingula variably tiated, protocone very well developed and a
developed, but normally the lingual cingulum is crescentic metaconule. Lingual cingulum
weaker than the labial. The p1 is not molari- continuous or interrupted, always with well-
form, simple and single rooted, elongated developed pre- and post-cingulum, sometimes
anteroposteriorly with a single main cuspid, with a double cingulum. The M1–M2 with a
prolonged by an anterior and a posterior crest strongly lophoselenodont ectoloph. Parastyle
or very short talonid. The p2 is more complex, and mesostyle very well developed, with strong
with trigonid crescentic and short talonid labial columns projected labially or anterolabi-
(Phoradiadus) or trigonid and talonid subequal ally. Metastyle fairly to little developed.
and bicrescentic (Sparnotheriodon). Paracone and metacone selenodont, with labial
The p3–p4 are molarized and, with the columns little developed or absent; projecting
m1–m3, all are morphologically very similar, lingually into the trigon and variably developed
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 166

166 M. BOND ET AL.

there are cuspules, forming one or two short follow Soria (2001) in his use of the Sparno-
crests. Protocone bunoid, connected by a short theriodontidae.
crest to the paraconule and to the hypocone,
closing the internal valley, but with a shallow Notolophus gen. nov.
lingual sulcus. Hypocone smaller than the
protocone and connected by a short crest to the Type species
metaconule (e.g. Victorlemoinea, Phoradiadus)
or directly to it (Notolophus). Paraconule and Notolophus arquinotiensis, sp. nov.
metaconule subcrescentic, sometimes connected
to the ectoloph by very short and low crests. In Diagnosis
some cases (Notolophus) the paraconule is no
longer recognizable as an independent cusp, Same as for the type species.
present as a short paraloph connected to the
anterior cingulum. Post-metaconule crista Etymology
present but variably developed. Labial
cingulum not very strong, sometimes restricted Notos, is derived from the greek νοτοσ, south,
to the posterior portion; lingual cingulum in reference to the geographical area where the
variably developed. Precingulum, with a very taxon was found; and λοφοσ, lophs, crests.
well developed protostyle, sometimes
connected to the paraloph (Notolophus). Post- Notolophus arquinotiensis sp. nov. (Figs 2a, b,
cingulum encloses a low fossette. Pre- and 4a, c & 5a, b)
post-cingulum present as a low extra cingulum,
forming a ‘double cingulum’ that occurs also in Holotype
the Anisolambdidae litopterns. The M3 is
similar to the M1–M2, but with the hypocone MLP 95-I-10-6, left M3 incomplete (the buccal
absent. Of the deciduous molars known, the part of paracone and metacone is missing)
DP4 is fully molarized, with prominent (Fig. 2a). La Meseta Formation, Submeseta
mesostyle, hypocone, paraloph, metaloph, Member (TELM 7), DPV 16/84 locality. This
postcingulum fossette, accessory cusps project- molar was briefly described and figured by
ing lingually from the ectoloph as a ‘double Vizcaíno et al. (1997).
post-cingulum’. As so far known, the recognized
taxa in this family posses enamel with vertically Hypodigm
oriented Hunter–Schreger bands.
Holotype plus MLP 90-I-20-1, left upper molari-
Comments form (M1 or M2?), Cucullaea I Member
(TELM 5), IAA 1/90. MLP 91-II-4-1, right p4,
Soria (1980a) established the Sparnotheriodonti- Cucullaea I Member (TELM 4), DPV 2/84
dae as an undetermined notoungulate monotypic locality. MLP 95-I-10-7, fragmentary left upper
family based on Sparnotheriodon epsilonoides molariform, Cucullaea I Member (TELM 5),
from the Vacan subage (late Palaeocene–early MLP 01-I-1-1, right m3, Cucullaea I Member
Eocene) of Patagonia. Subsequently, Soria (TELM 5), IAA 1/90 locality. MLP 04-III-3-1,
(2001) characterized the family and included incomplete right p4, Cucullaea I Member
with it the Anisolambdidae (regarded by Cifelli (TELM 5), IAA 1/95.
as a subfamily of Proterotheriidae) in a new
suborder, Eolitopterna. Referred specimens
Cifelli (1993) defined the Sparnotheriodonti-
dae by several advanced characters, including a MLP 90-I-20-3, right I3?, Cucullaea I Member
lophoid metaconule and an expanded post- (TELM 5), IAA 1/90 locality. MLP 90-I-20-5,
cingulum, but included in this family the left upper premolar incomplete (P2 or P3?),
Indaleciinae, a group of very small ungulates Cucullaea I Member (TELM 5), IAA 1/90
traditionally considered as Adianthidae locality. MLP 91-II-4-5, right upper premolar
litopterns (Cifelli & Soria 1983) or as a family, (P1), Cucullaea I Member (TELM 5), IAA 1/90
Indaleciidae, of the Order Notopterna (Soria locality. MLP 92-II-2–135, fragment of a molari-
1989). Bonaparte & Morales (1997) followed form (lower?), Campamento Member (TELM
Cifelli (1993) in the grouping of Victorlemoinea 3), IAA 1/92. MLP 94-III-15-3, left lower
and Indalecia, but considered them all incisive, Cucullaea I Member (TELM 5), IAA
litopterns. Here, we exclude the indaleciids from 1/90 locality. MLP 96-I-5-5, left upper incisivi-
the Sparnotheriodontidae and, as stated earlier, form (I1?), Cucullaea I Member (TELM 5),
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 167

‘SOUTH AMERICAN UNGULATE’ FROM ANTARCTICA 167

IAA 2/95 locality. MLP 96-I-5-9, left lower inci- Additional specimens are known from other
siviform or first premolar?, Cucullaea I Member localities (Fig. 1) in lower levels (Cucullaea I
(TELM 5), IAA 3/96 locality. and Campamento Members) of the La Meseta
Formation.
Type locality
Stratigraphy and age
Museo de La Plata locality DPV 16/84, Seymour
Island, Antarctic Peninsula (Fig. 1). GPS data: La Meseta Formation (late Early Eocene–Late
64°14
04.672S and 56°39
56.378W. Sr isotope Eocene), Campamento (Early Eocene), Cucul-
dating from this horizon yields an age of approx- laea I (Middle Eocene) and Submeseta (Late
imately 34.2 Ma (Dingle & Lavelle 1998). Eocene) members.

Fig. 1. Map of Seymour (Marambio) Island (Antarctic Peninsula) showing the IAA and DPV localities
mentioned in the text.
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 168

168 M. BOND ET AL.

Fig. 2. Occlusal views of upper molars of sparnotheriodontids from Antarctica and Patagonia. (a) Notolophus
arquinotiensis, gen. et sp. nov. MLP 95-I-10-6, left M3, holotype; (b) Notolophus arquinotiensis, gen. et sp. nov.
MLP 90-I-20-1, left M1 or M2; (c) Victorlemoinea labyrinthica, MACN A-10871, left M1 (based also on M2 of
the same individual); and (d) Victorlemoinea sp. MLP 66-V-12-2, right M1 (reversed). The scale bar equals 5
mm (drawing by A. Viñas).

Etymology Sparnotheriodon epsilonoides. Upper molars

with a very strong and well-developed ectoloph,
The specific epithet, arquinotiensis, is in labial cingulum very weak and restricted to the
reference to the Ihering’s (1927) Archinotis posterior portion of the ectoloph, between the
continent. metacone and the short metastyle. Short but
strong lingual projections, one from the posterior
Short diagnosis part of the paracone and other from the anterior
part of the metacone. Protocone elongated and
A sparnotheriodontid larger than Phoradiadus projected anteriorly by a paraloph in which no
and nearly equal as Sparnotheriodon. Differs paraconule is visible as separate cusp. Hypocone
from the other known taxa in having the upper little developed connected by a short crest to the
molars with a protocone projected anteriorly by protocone; the hypocone is vestigial or absent on
a short paraloph which connected to the proto- M3. Metaconule lophoid, nearly straight and
style in the second anterior cingulum (precingu- very anteriorly extended, directly connected to
lum). Metaconule lophoid anteriorly extended the hypocone, without the short intermediate
and connected directly to the poorly developed crest connecting these cusps as in Victorlemoinea
hypocone, without intermediate crest as in the or Phoradiadus. Post-metaconule crista low and
other taxa known. In the M3 the hypocone is post-cingulum enclosing a small basin that is not
very weak or absent, also the paraloph and the so developed as in Victorlemoinea or Phora-
protostyle connects directly to the protocone. diadus. In the M3, the metaconule is connected
Lower molariforms with the trigonid smaller with the first post-cingulum. The second anterior
than the talonid. The m3 has a well-developed cingulum (or precingulum) possess a prominent
bunoid entoconid and a posteriorly projecting protostyle, which is connected by a short
hypoconulid. posterolabial isthmus to the paraloph. In the M3
the protostyle directly connects to the anterior
Differential diagnosis portion of the protocone, the paraloph being
vestigial or absent.
Sparnotheriodontid much larger than Phora- Lower molariforms with trigonid somewhat
diadus divortiensis, and nearly as large as smaller than the talonid. Trigonid and talonid
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 169

‘SOUTH AMERICAN UNGULATE’ FROM ANTARCTICA 169

basins not so narrow and Phoradiadus and but the enamel structure of the isotemnids is
Sparnotheriodon. The trigonid exhibits a very completely different lacking the vertically
well developed paralophid, with an engrossed oriented Hunter–Schreger bands (see Fortelius
lingual end that could represents a paraconid or 1985). By comparison with the upper and lower
?neoparaconid. Posterior premolars with a incisors known in Phoradiadus and Sparnothe-
conspicuous descending crest posterior to the riodon we refer tentatively these specimens as
metaconid; small entoconid connected with the probable upper incisors (I3?) of Notolophus
hypoconulid. The m3 with a talonid larger and arquinotiensis, but recognize that they are more
more elongated than the trigonid, with a robust than the I3 of these species. It cannot be
projecting hypoconulid in a rudimentary ‘third ruled out that the teeth aforementioned could
lobe’; well-developed bunoid entoconid antero- represent upper canines, but since in sparno-
posteriorly enlarged and connected to the theriodontids, like Phoradiadus divortiensis, the
hypoconulid. Labial cingulum variably canines (upper and lower) are pointed and with
developed; lingual cingulum low and continuous sharp edges, we therefore, identify these teeth
to absent. tentatively as I3. MLP 96-I-5-5 is a very worn
compressed mesiodistal incisiviform with an
Description ellipsoid coronal figure, with no trace of a labial
cingulum; it is very probably an anterior incisor,
As stated earlier, our knowledge of previously perhaps the I1. By comparison with the anterior
known Sparnotheriodontidae is meager. Taking lower dentition known in Sparnotheriodon
this in account, the unassociated and, some- epsilonoides, MLP 94-III-15-3 is considered as a
times, fragmentary nature of the Antarctic probable lower incisiviform (right i3?); it is a
sparnotheriodontid specimens precludes an simple tooth, very worn, with a principal labial
adequate interpretation. Most of the Antarctic cusp and a short anterior crest; there is also a
ungulate teeth undoubtedly can be assigned to lingual cingulum connected with the anterior
Sparnotheriodontidae, and with a high degree of crest and it has a middle lingual cuspule. The
confidence to Notolophus arquinotiensis. specimen MLP 96-I-5-9 very probably repre-
Notwithstanding, some of them are difficult to sents a first lower premolar (left p?1); this tooth,
interpret, not in taxonomical reference but in its although very worn occlusally, shows an antero-
proper position in the dental series given the posteriorly enlarged and wide trigonid, with a
aforementioned scanty knowledge of the principal labial cusp area and a very short
complete dental anatomy of this group. talonid, somewhat different then to the more
The molariform teeth, upper and lower, can elongated p1 of Sparnotheriodon epsilonoides.
be referred with a high degree of confidence to This tooth is single rooted with a very oblique
this new taxon because all upper molars known root. No teeth were found that could be referred
have the same derived features. The lower confidently as the canines, upper or lower, of
molariform teeth match well in size with the Notolophus arquinotiensis.
upper ones and are therefore referred to the MLP 91-II-4-5 is a very simple tooth, single
same taxon. rooted, with a flattened crown by wear. It has a
MLP 90-I-20-3 (Fig. 5) and MLP 96-I-5-10 are principal labial cusp (paracone) with a short
incisiviforms rather than caniniforms and match anterolabial crest interpreted as a parastyle, and
in size with the other teeth assigned to Notolo- a shorter posterior crest (metastyle?). Labially,
phus arquinotiensis. They are robust, with a the principal cusp has a convex surface and an
straight root and a labial single cusp with a anterior shallow fold which delimitates the
convex labial wall (in MLP 90-I-20-3 the wear parastyle from the paracone. A strong lingual
has obliterated this cusp). There are very well cingulum is connected to the parastyle and
developed labial and lingual cingula. The metastyle; this lingual cingulum has a well
enamel is thick with strongly marked alternat- defined cuspule which is connected to the
ing bands of the vertically oriented paracone by a short posterolabially directed
Hunter–Schreger bands. The morphology of crest. This tooth is interpreted here as a P1.
these teeth is very different from the incisors MLP 90-I-20-5, by comparison with the upper
known of Pyrotheria and Astrapotheria (see premolars of Phoradiadus, represents an upper
Simpson 1967) (astrapotheres do not have premolar, possibly a left P3. The specimen is not
upper incisors), which also have vertically complete, but has a well-developed protostyle, a
oriented Hunter–Schregger enamel bands (see bunoid protocone, apparently lacks the
Fortelius 1985) as in the Sparnotheriodontidae. hypocone and short lingual crests project from
Also, these teeth resemble the I3 of some the ectoloph, and the posterior fossette formed
notoungulate families such as the Isotemnidae, by the metaloph and posterior cingulum has
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 170

170 M. BOND ET AL.

been obliterated by wear. Posterior to the

cingulum there is an extra cingulum.
Sparnotheriodontid molars are quite uniform
in form, and those of N. arquinotiensis share the
same general pattern with other Patagonian
sparnotheriodontids, but their proportions and
especially the morphology of the upper molars
is quite characteristic.
The holotype, MLP 95-I-10-6 (Fig. 2a), is of
roughly rectangular outline, and the anterior
and medium part of the ectoloph is missing. The
preserved ectoloph shows a lophoselenoid
metacone with a flattened labial area and a short
metastyle which descends posteriorly; there is
also preserved part of a low labial cingulum, but
which may or may not have been continuous.
The shallow internal basin or principal valley is
formed between the protocone and the
ectoloph, and exhibits two short, low crests
projecting from the ectoloph. The protocone is
large, anteroposteriorly elongated and con-
nected to a very well developed anterolingual
cusp. This lingually displaced cusp is interpreted
here as an enlarged protostyle cingular cusp, Fig. 3. Occlusal view of right M3 (MLP 66-V-12-1) of
although we do not rule out the possibility that Victorlemoinea sp. The scale bar equals 5 mm
it could also be a displaced paraconule fused (drawing by A. Viñas).
with the protostylar cusp. Nevertheless, its
position and the relationships with the second
precingulum are more indicative of an enlarged
protostyle. The protocone possess a posterior ectoloph. The hypocone is small and connected
crest that connects to the post-cingulum. The to the protocone by a short crest, with a very
metaconule is strongly lophoid and projected shallow sulcus between the protocone and
mesiodistally to the internal valley, post-meta- hypocone. The metaconule is lophoid, strongly
conular crista well developed and directed projected mesiodistally as in the M3, and is
labially connecting the metaconule to the connected directly to the hypocone without the
metacone area. No hypocone exists, and the short intermediate lingually projected crest that
metaconule connects directly with the posterior connects the metaconule and hypocone in
projection of the protocone. The post-cingulum, Victorlemoinea (e.g. V. labyrinthica) and Phora-
connected to the protocone and metaconule, is diadus, but which is very short and nearly absent
expanded and encloses a small fossette; this in MLP 66-V-12-2 (Fig. 2d) identified as Victor-
basined post-cingulum is proportionally more lemoinea sp. from the Vacan (early Casamay-
developed in Victorlemoinea (Fig. 3) and Phora- oran) of Patagonia. Post-metaconular crista is
diadus than in Notolophus (Fig. 2). Pre- and similar in form and direction as in the M3,
post-cingula with a very low extra cingulum. The although it is lower and not so well developed.
lingual cingulum is very low and restricted to the Lingual cingulum apparently restricted to the
anterior part of the protocone. anteriormost part of the protocone. Anterior
MLP 90-I-20-1 is very probably a left M1 or and posterior cingula with low extracingula,
M2 (Fig. 2b), although it could represent a conforming the double cingulum of the
molariform DP4. It is very similar to the Sparnotheriodontidae.
above described M3, but has a complete Two lower molariforms, MLP 91-II-4-1
ectoloph. No labial columns are present on the (Fig. 4c) and MLP 04-III-3-1, are tentatively
paracone and metacone, and, except in the assigned to the ‘molarized’ premolars of this
middle, which is slightly convex, the walls of the species, and they probably represent two right
paracone and metacone are flattened to slightly p4, or a p4 and a p3, respectively. They are fully
concave. The parastyle is conspicuous, but the molariform with the trigonid crescent relatively
mesostyle represents the strongest element of shorter than that of the talonid and not so
the ectoloph with a very wide base. The labial labially projected. The paralophid is very well
cingulum is restricted to the posterior part of the developed and lingually projected as a small
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 171

‘SOUTH AMERICAN UNGULATE’ FROM ANTARCTICA 171

Fig. 4. Occlusal views of lower molars of sparnotheriodontids from Antarctica and Patagonia. (a) Notolophus
arquinotiensis, gen. et sp. nov. MLP 01-I-1-1, right m3; (b) Sparnotheriodon epsilonoides, MACN 18225, right
m3. The scale bar equals 5 mm (drawing by A. Viñas). (c) Notolophus arquinotiensis, gen. et sp. nov.
MLP 91-II-4-1, occlusal view of right p4. The scale bar equals 5 mm (drawing by A. Viñas).

cuspid (paraconid or ?neoparaconid). Meta- (paraconid or ?neoparaconid) as in the premo-

conid with a very sharp descending crest, similar lars described above, but (at least in this state of
to that observed in the p3–p4 of Sparnotheri- wear) with no trace of an independent cusp. The
odon elipsonoides and Phoradiadus divortiensis. metaconid is the highest cusp and has a rela-
The entoconid is reduced and coalescent with a tively wide descending posterior crest. The
very short hypoconulid. Labial fold (ectoflexid) talonid is more elongated anteroposteriorly
and lingual folds (meta and entoflexid) very well than the trigonid, with a posteriorly projected
developed. The ectoflexid is deeper and more hypoconulid separated by a labial fold forming
penetrating than the lingual flexids, with the a short and rudimentary ‘third lobe’. The ento-
entoflexid more open than the metaflexid. Well- conid is bunoid, projects anteriorly and is
developed anterior and posterior cingula extend connected to the hypoconulid; it is more
labiolingually and may or not be connected to developed and inflated than in Sparnotheriodon
the labial and lingual cingula. The labial epsilonoides and similar to ?V. longidens, but the
cingulum is present in these two specimens, but entoconid is not so differentiated from the
it is continuous (MLP 04-III-3-1) or is restricted hypolophid as in Phoradiadus divortiensis. The
to the base of the labial fold (ectoflexid) (MLP ectoflexid is more open, deep and penetrating
91-II-4-1). The lingual cingulum is low but than the lingual folds, which are relatively
continuous (MLP 04-III-3-1) or absent (MLP shallow. The anterior cingulum is well
91-II-4-1). developed and extends transversely with the
A nearly complete right m3, MLP 01-I-1-1 lingual portion higher and directed to the
(Fig. 4a), has a trigonid shorter than the more paralophid; it is not connected to the labial
elongated talonid. The trigonid shows the cingulum that extends from the hypoconulid
lingual portion of the paralophid engrossed lobe to the posterior part of the protoconid
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 172

172 M. BOND ET AL.

column. Some cuspules occur in the ectoflexid

valley. The lingual cingulum is apparently
restricted to the trigonid, extending from the
paralophid to the anterior portion of the meta-
conid.

Discussion
Notolophus arquinotiensis is one of the most
abundant taxa among the terrestrial mammals
from the La Meseta Formation. N. arquinotien-
sis is currently represented by a small number
of specimens collected at six localities in
Seymour Island (Fig. 1). Its tooth anatomy, as
described above, is distinctive and allows a clear
differentiation from other Palaeocene and
Eocene sparnotheriodontids.
Only three sparnotheriodontid genera are so
far known in South America (Table 1). A fourth
genus, Heteroglyphis, from the Mustersan Age
(late Eocene) was included tentatively within
the family by Soria (2001), although restudy of
the type and only known specimen suggests that
Heteroglyphis dewoletzky, Roth 1899 belongs to
the Anisolambdinae or Anisolambdidae eoli-
topterns. The specimens discussed here were
initially referred to Victorlemoinea (Bond et al.
1990). The genus Victorlemoinea was erected by
Ameghino (1901), who recognized two species:
V. labyrinthica, the genotypical one (Fig. 2c) and
V. emarginata, both based on upper molariform
teeth (see Simpson 1948) from the Casamay-
oran SALMA (possibly Vacan ‘subage’) of
Patagonia. From the same area and age,
Simpson (1948) doubtfully referred Victor-
lemoinea to the species Anisolambda longidens
Ameghino, 1901, based on lower teeth. Later,
Paula Couto (1952) referred a fourth species to
Victorlemoinea: V. prototypica from the Itabora-
ian SALMA (middle Palaeocene) of Brazil and
based on upper and lower teeth.
Notolophus arquinotiensis (Fig. 2) is different
from V. labyrinthica: V. emarginata and V. proto-
typica being somewhat larger than V. laby-
rinthica, and definitely larger than V. emarginata
and V. prototypica. The peculiar connection of
the protocone–paraloph with the enlarged
protostyle is clearly distinct from the morphol-
ogy observed in the species of Victorlemoinea.
It is interesting to note that upper molars from
the Early Casamayoran SALMA (Vacan
subage), referred here as Victorlemoinea sp.,
MLP 66-V-12-2, have a similar size to those of
the type of Victorlemoinea labyrinthica, but
differ in having a smaller hypocone and a Fig. 5. Notolophus arquinotiensis, gen. et sp. nov.
shorter crest connecting the metaconule with MLP 90-I-20-3, right I3?. (a) Labial view and (b)
this cusp. These molars, similar to those figured occlusal view. The scale bar equals 5 mm (drawing by
by Simpson (1948) (i.e. AMNH 28466), also A. Viñas).
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 173

‘SOUTH AMERICAN UNGULATE’ FROM ANTARCTICA 173

from the Vacan subage (Casamayoran age), browse, stripping off twigs and saplings from
approach the condition observed in Notolophus evergreen trees even during winter months
arquinotiensis, but clearly differ by the paraloph (Vizcaíno et al. 1998b). No post-cranial infor-
which in MLP 66-V-12-2 is not united to the mation is available for the Antarctic ungulates,
protostyle as in V. labyrinthica. but information from the nearest relatives (all
?Victorlemoinea longidens is based on lower of them fossils) can be used to infer the locomo-
premolars and molars not clearly associated. tor adaptation to the cursoriality. Cifelli
The lower premolars are different from other (1983a, b) associated teeth and astragalus and
known Sparnotheriodontidae, and do not have calcaneum to the Itaboraian (Palaeocene)
vertically oriented Hunter–Schreger bands; its species Victorlemoinea prototypica of Brazil.
morphology is more reminiscent of a notoungu- The faunal evidence, mainly that provided by
late Isotemnidae than a litoptern, and we do not the marine invertebrates (Stilwell & Zins-
consider this premolar as those of a sparnothe- meister 1992), indicates the deposition of the
riodontid. The incomplete right lower molars Submeseta Member, where the holotype was
(m1–m2), although of smaller size than those of recovered, was in cool-temperate conditions,
Notolophus, have an enlarged entoconid and a unlike the underlying Cucullaea I Member. A
weak lingual cingulum, which are characters sharp decrease of diversity near the contact
also observed in the lower molars of Notolo- between the upper members of La Meseta
phus, but they differ in the more narrow and Formation (Cucullaea II and Submeseta) may
penetrating meta and entoflexids of ?V. longi- be correlated with the climatic cooling event
dens. Also, it is very possible that ?V. longidens which culminated at the time of deposition of
could represent the lower teeth of Victor- the uppermost part of the La Meseta Formation
lemoinea labyrinthica. (Gazdzicki et al. 1992). The presence of Notolo-
Sparnotheriodon epsilonoides is only known phus, together with a ground-dwelling bird
from its lower teeth and mandible (Soria 1980a), (ratite) and Nothofagus leaves from the same
so no direct comparison can be made between horizon, suggest that the terrestrial environment
it and MLP 95-I-10-6. However, the lower during the time of deposition of at least part of
molars of the hypodigm of Notolophus (MLP the Submeseta Member was apparently not
91-II-4-1 and MLP 01-I-1-1) are clearly lophose- dissimilar to that reconstructed by Reguero et
lenodonts and match very well in size and al. (2002) for the Cucullaea I Member with
general anatomy with those of Sparnotheriodon. Nothofagus forests and mountainous cordillera.
Recent work on the faunal similarities of the Notolophus had a more bilophodont than
La Meseta fauna indicate a strong biogeograph- bunodont dentition, and their molariforms teeth
ical connection with the southern tip of South had strong enamel ridges extending between the
America (Patagonia) (Goin et al. 1999; Reguero cusps. These enamel ridges serve as shearing
et al. 2002), and the identification of archaic surfaces, and the formation of dentine ‘lakes’
marsupial prepidolopids and derorhynchids at along the ridges produce double-edged shearing
Seymour Island reinforces that link. Similarly, blades. These mainly performed a shearing
the recovery of sudamericid gondwanatheres action, slicing leaves into quite large pieces like
from Seymour Island and the recognition of a modern tapir that feeds almost entirely on
strong morphological correspondence between leaves of forest trees. The body size of the
the Seymour gondwanathere and Sudamerica Antarctic sparnotheriodontid (395–400 kg) indi-
ameghinoi also demonstrate a late Palaeocene cates that it was the largest terrestrial herbivore
connection with Patagonia (Reguero et al. 2002). living in Antarctica at this time (Vizcaíno et al.
The rare occurrences of sparnotheriodontids 1998b). Evidently, the large size of this herbi-
in an otherwise very well recorded faunal vore favoured the exploitation of leaves because
context of the Palaeocene of Patagonia and a longer time in the gut for bacterial fermenta-
Brazil leads to the assumption that they could tion is required to obtain sufficient nutrients
be extreme ecological specialists. They show a from leaves. Based on dental morphology,
number of dental characteristics that may be sparnotheriodontids were probably hindgut
adaptations to forested habitats, and the striking fermenters like non-ruminant artiodactyls and
dental features of the Antarctic taxon are perissodactyls (Fortelius 1985; Rensberger &
brachyodonty and the particular structure of the Pfretzschner 1992). Astrapotheres and
enamel (vertically oriented Hunter–Schreger sparnotheriodontids also have teeth with
bands) (Reguero et al. 2002). Janis (1984) vertical Hunter–Schreger bands. Fortelius
pointed out that brachyodonty is associated (1985) indicated that a number of lophodont
with browsing herbivores that are adapted to ungulates have evolved vertically oriented
forest habitats. In particular, Notolophus could Hunter–Schreger bands, a modification that
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 174

174 M. BOND ET AL.

Fig. 6. Reconstruction of the archaic litoptern Notolophus and the opossum-like marsupial Antarctodolops on
the eastern shore of the Antarctic Peninsula during the Middle Eocene. In the background Nothofagus forest
and mountains (drawing by A. Viñas).

involves the mode of prism decussation and gration event or a cluster of taxa arriving on the
three-dimensional arrangement of the bands. Antarctic Peninsula at different times by chance
This condition has been interpreted as an adap- routes.
tation to resist cracking when the enamel edges A more precise reconstruction of the palaeo-
are loaded in a direction away from the support- ecology of Notolophus would be possible if
ing dentine (Boyde & Fortelius 1986). In cranial and post-cranial remains were known.
Notolophus, as in the rest of the representatives Clearly, much remains to be learned about this
of the family, the ectoloph forms a thin, vertical, rare Antarctic litoptern, questions that only
blade-like ectoloph with a strong mesostyle. future discoveries of additional material can
Notolophus arquinotiensis is a large answer.
sparnotheriodontid (Fig. 6), larger and different
in morphology than the last ones of the Conclusion
Divisaderan SALMA (late Eocene), and more
similar in size to some remains of the Vacan The new taxon reported here is the first well-
Subage (Casamayoran Age; early Eocene). The documented Antarctic ‘South American
material of Notolophus from Seymour Island ungulate’, and it belongs to an archaic and
exhibits no change of size through the Campa- uncommon lineage whose ultimate ancestry
mento Member (TELM 3) to the Submeseta may be Laurasiatic ‘condylarths’. Notolophus
Member (TELM 7) of the La Meseta For- arquinotiensis definitively confirms the occur-
mation, indicating that the individuals of rence of an archaic ungulate population in
Notolophus arquinotiensis were of very large Antarctica and supports the role of the conti-
size existing over a large timespan. Also, related nent as a probable centre of eutherian evolution
forms in the Vacan Subage (early Eocene) may (Vizcaíno et al. 1998a). Notolophus arquinotien-
tempt one to propose an immigration event for sis has close affinities with Victorlemoinea, indi-
the sparnotheriodontids in Antarctica near the cating at least a very close common ancestor,
Vacan Subage or Riochican Age (late Pale- probably a ‘condylarth’ despite its strikingly
ocene). However, other taxa (e.g. the marsupi- molariform P3–4/p3–4.
als) could indicate an earlier migration datum,
but additional taxa from the La Meseta Forma- We express our gratitude to Dr J. E. Martin and an
tion are required to demonstrate either an anonymous reviewer for critical review of the manu-
impoverished fauna of a previous, single immi- script. We thank the personnel and authorities of the
12_SP258 Bond (to_d) 25/5/06 9:18 am Page 175

‘SOUTH AMERICAN UNGULATE’ FROM ANTARCTICA 175

Instituto Antártico Argentino, especially S. Santillana (Seymour Island, Medial Eocene). Journal of
and E. Yermolin for their logistic support during field- Mammalian Evolution, 6, 335–365.
work at locality IAA 1/90 in the Antarctic Peninsula; HOFFSTETTER, R. & SORIA, M.F. 1986. Neodolodus
and Mr. J. J. Moly for his fieldwork in Antarctica. We colombianus gen. et sp. nov., un nouveau Condy-
also acknowledge the following people for access to larthre (Mammalia) dans le Miocene de Colombie.
fossils housed in their respective institutions: M. Comptes Rendus de l’Académie des Sciences, Paris,
Norell and M. Novacek (AMNH), and J. Bonaparte série II, 17, 1619–1622.
and A. Kramarz (MACN). Fieldwork at Seymour HOOKER, J.J. 1992. An additional record of a placen-
Island and museum research was supported by the tal mammal (Order Astrapotheria) from the
Instituto Antártico Argentino (IAA), Consejo Eocene of Western Antarctica. Antarctic Science,
Nacional de Investigaciones Científico y Técnicas 4, 107–108.
(CONICET) and the National Geographic (grant to IHERING, H.V. 1927. Die Geschichte des Atlantischen
S. A. Marenssi). We gratefully acknowledge A. Viñas Ozeans. Gustav Fischer, Jena.
for his fine artwork in Figures 2–5. JANIS, C.M. 1984. The use of fossil ungulate communi-
ties as indicators of climate and environment. In:
References BRENCHLEY, P. (ed.) Fossils and Climates. Wiley,
Chichester, 85–104.
AMEGHINO, F. 1901. Notices préliminaires sur des MARENSSI, S.A., REGUERO, M.A., SANTILLANA, S.N. &
ongulés nouveaux des terraines crétacés de Patag- VIZCAÍNO, S.F. 1994. Eocene land mammals from
onie. Boletín de la Academia Nacional de Ciencias Seymour Island, Antarctica: Palaeobiogeographi-
de Córdoba, 16, 349–426. cal implications. Antarctic Science, 6, 3–15.
BONAPARTE, J.F. & MORALES, J. 1997. Un primitivo NESSOV, L.A., ARCHIBALD, J.D. & KIELAN-JAWOROSKA,
Notonychopidae (Litopterna) del Paleoceno Z. 1998. Ungulate-like mammals from the late
inferior de Punta Peligro, Chubut, Argentina. Cretaceous of Uzbekistan and a phylogenetic
Estudios Geológicos, 53, 263–274. analysis of Ungulatomorpha. Bulletin of Carnegie
BOND, M., PASCUAL, R., REGUERO, M.A., SANTILLANA, Museum of Natural History, 34, 40–88.
S.N. & MARENSSI, S.A. 1990. Los primeros ungula- PAULA COUTO, C. DE. 1952. Fossil mammals from the
dos extinguidos sudamericanos de la Antártida. beginning of the Cenozoic in Brazil. Condylarthra,
Ameghiniana, 16, 240. Litopterna, Xenungulata and Astrapotheria.
BOYDE, A. & FORTELIUS, M. 1986. Development, Bulletin of the American Museum of Natural
structure and function of rhinoceros enamel. History, 99, 355–394.
Zoological Journal of the Linnean Society, 87, REGUERO, M.A., MARENSSI, S.A. & SANTILLANA, S.N.
181–214. 2002. Antarctic Peninsula and South America
CIFELLI, R.L. 1983a. Eutherian tarsals from the late (Patagonia) Paleogene terrestrial faunas and
Paleocene of Brazil. American Museum Novitates, environments: biogeographic relationships.
1761, 1–31. Palaeogeography, Palaeoclimatology, Palaeoecol-
CIFELLI, R.L. 1983b. The origin and affinities of the ogy, 179, 189–210.
South American Condylarthra and early Tertiary RENSBERGER, J.M. & PFRETZSCHNER, H.U. 1992.
Litopterna (Mammalia). American Museum Novi- Enamel structure in Astrapotheres and its
tates, 2772, 1–49. functional implications. Scanning Microscopy, 6,
CIFELLI, R.L. 1993. The phylogeny of the native South 495–510.
American ungulates. In: SZALAY, F.S., NOVACEK, ROTH, S. 1899. Aviso preliminar sobre mamíferos
M.J. & MCKENNA, M.C. (eds) Mammal Phylogeny, mesozoicos encontrados en Patagonia. Revista del
Volume 2. Placentals. Springer, New York, 195–216. Museo de La Plata, 9, 381–388.
CIFELLI, R.L. & SORIA, M.F. 1983. Systematics of the SIMPSON, G.G. 1945. The principles of classification
Adianthidae (Litopterna, Mammalia). American and a classification of mammals. Bulletin of the
Museum Novitates, 2771, 1–25. American Museum of Natural History, 85, 1–350.
DINGLE, R. & LAVELLE, M. 1998. Late Cretaceous– SIMPSON, G.G. 1948. The beginning of the Age of the
Cenozoic climatic variations of the northern Mammals in South America. Part I. Bulletin of the
Antarctic Peninsula: new geochemical evidence American Museum of Natural History, 91, 1–232.
and review. Palaeogeography, Palaeoclimatology, SIMPSON, G.G. 1967. The beginning of the Age of the
Palaeoecology, 107, 79–101. Mammals in South America. Part II. Bulletin of the
FORTELIUS, M. 1985. Ungulate cheek teeth: develop- American Museum of Natural History, 137, 1–259.
mental, functional and evolutionary interrelations. SIMPSON, G.G., HINOPRIO, J.L. & PATTERSON, B. 1962.
Acta Zoologica Fennica, 180, 1–76. The mammalian fauna of the Divisadero Largo
GAZDZICKI, A.J., GRUSZCZYNSKI, M., HOFFMAN, A., Formation, Mendoza, Argentina. Bulletin of the
MALKOWSKI, K., MARENSSI, S.A., HALAS, S. & Museum of Comparative Zoology, 127, 239–293.
TATUR, A. 1992. Stable carbon and oxygen isotope SORIA, M.F. 1980a. Una nueva y problemática forma
record in the Paleogene La Meseta Formation, de ungulado del Casamayorense. II Congreso
Seymour Island, Antarctica. Antarctic Science, 4, Argentino de Paleontología y Bioestratigrafía y I
461–468. Congreso Latinoamericano de Paleontología,
GOIN, F.J., CASE, J.A., WOODBURNE, M.O., VIZCAÍNO, Buenos Aires, 1978, 2, 193–203.
S. F. & REGUERO, M.A. 1999. New discoveries of SORIA, M.F. 1980b. Las afinidades de Phoradiadus
‘oppossum-like’ marsupials from Antarctica divortiensis Simpson, Minoprio and Patterson,
 12_SP258 Bond (to_d) 25/5/06 9:18 am Page 176

176 M. BOND ET AL.

1962. Circular Informativa Asociación Paleon- mammals from Antarctica, its significance on the
tológica Argentina, 4, 20. evolution of the terrestrial environment of the
SORIA, M.F. 1989. Notopterna: un nuevo orden de Antarctic Peninsula during the late Eocene.
mamíferos ungulados eógenos de América del Sur. Journal of Paleontology, 71, 348–350.
Parte I: Los Amilnedwardsiidae. Ameghiniana, 25, VIZCAÍNO, S.F., PASCUAL, R., REGUERO, M.A. & GOIN,
245–258. F.J. 1998a. Antarctica as a background for
SORIA, M.F. 2001. Los Proterotheriidae (Litopterna, mammalian evolution. In: CASADÍO, S. (ed.)
Mammalia), sistemática, origen y filogenia. Mono- Paleógeno de América del Sur y de la Península
grafías del Museo Argentino de Ciencias Natu- Antártica. Asociación Paleontológica Argentina,
rales, 1, 1–167. Publicación Especial, 5, 199–209.
STILWELL, J.D. & ZINSMEISTER, W.J. 1992. Molluscan VIZCAÍNO, S.F., REGUERO, M.A., GOIN, F.J., TAMBUSSI,
Systematics and Biostratigraphy, Lower Tertiary La C.P. & NORIEGA, J.I. 1998b. An approach to the
Meseta Formation, Seymour Island, Antarctic structure of the Eocene terrestrial vertebrate
Peninsula. American Geophysical Union, Antarc- community from Antarctic Peninsula. In:
tic Research Series, 55. CASADÍO, S. (ed.) Paleógeno de América del Sur y
VIZCAÍNO, S.F., BOND, M., REGUERO, M.A. & PASCUAL, de la Península Antártica. Asociación Paleontológ-
R. 1997. The youngest record of fossil land ica Argentina, Publicación Especial, 5, 177–183.

View publication stats