The first application of the mesowear method to endemic South American ungulates (Notoungulata)

https://doi.org/10.1016/j.palaeo.2008.08.007Get rights and content

Abstract

Notoungulates were the most diverse and abundant clade of endemic South American ungulates, present in nearly all Tertiary faunas. Hypsodont dentitions evolved in several notoungulate lineages by the early Oligocene, and by the middle Miocene, four families were characterized by hypselodont (ever-growing) cheek teeth. The presence of hypsodont and hypselodont dentations in notoungulates has generally been interpreted as indicating grazing and/or open habitat feeding, but few studies have attempted to test this assumption using other methods of dietary inference. The present study uses mesowear analysis to explore the correlation between crown height and diet in the three most common notoungulates of the late Oligocene (Deseadan SALMA) fauna of Salla, Bolivia: a mesotheriid (Trachytherus alloxus), an ‘archaeohyracid’ (Archaeohyrax sp. nov.), and an interatheriid (Federicoanaya sallaensis).

Mesowear analysis followed standard protocols. Specimens were digitally photographed in labial view so that cusp shape and occlusal relief could be scored. Cusp shape was scored subjectively as sharp, round, or blunt. Occlusal relief (measured digitally) was calculated as cusp height divided by intercusp distance and was categorized as ‘high’ or ‘low’ using three scenarios. The percentage of individuals of each species displaying sharp, blunt, and high cusps was calculated, normalized using the arcsine transformation, and used as variables in subsequent analyses to infer dietary preference. Statistical analyses included discriminant function (DFA) and hierarchical cluster (HCA). Mesowear scores were also explored. Analyses were executed using a comparative mesowear dataset of 64 modern ungulates of known diet (classified as browsers, grazers, or mixed feeders) as well as a subset of 27 ‘typical’ ungulates.

Trachytherus was classified as a grazer in most (5/6) DFAs; it primarily grouped with grazing ungulates (e.g., Alcelaphus, Connochaetes, Damaliscus, Equus) in both HCAs. Archaeohyrax was classified as a grazer in the DFA of ‘typical’ ungulates and as a mixed feeder or grazer in the expanded dataset DFA; it mostly clustered with grazers in the ‘typical’ HCA and with grazers (Alcelaphus buselaphus, Connochaetes, Hippotragus equinus) and mixed feeders (Axis spp., Rucervus) in equal proportions in the expanded HCA. Federicoanaya was classified as a mixed feeder in all DFAs; it grouped with grazers and mixed feeders in the ‘typical’ DFA and with all dietary types in the expanded HCA including two grazers (Hippotragus equinus and Ourebia), a mixed feeder (Tetracerus), and a browser, (Litocranius). Mesowear scores for the three species generally overlapped those for extant grazers, extending into the upper range of mixed feeders.

These analyses suggest: (1) all three notoungulates included some grass in their diet and/or fed at least part of the time in open habitats; (2) these notoungulates were partitioning available dietary resources by foraging in different areas and/or varying the composition of their diets; and (3) Trachytherus had the most abrasive diet of the three notoungulates, and Federicoanaya had the least abrasive. Based on these analyses and the diversity of other mammals present at Salla, the fauna probably represents a semi-open woodland community.

Introduction

Notoungulates may be the most diverse mammalian clade devoid of living representatives. They ranged in size from less than 1 kg (e.g., the interatheriid Punapithecus; López and Bond, 1995, Hitz et al., 2006) to more than 1,000 kg (e.g., Toxodon; Fariña et al., 1998); their dentitions varied from brachydont and complete (e.g., Homalodotherium) to extremely hypsodont and highly reduced (e.g., Mesotherium); and their postcrania suggest diverse forms of terrestrial locomotion ranging from cursoriality (e.g., some interatheriids; Stirton, 1953, Cifelli, 1985, Croft and Anderson, 2008) to fossoriality (e.g., mesotheriids; Shockey et al., 2007). They occurred throughout South America from the Paleocene to the Pleistocene and some 150 genera in 13 families have been described (Cifelli, 1985, McKenna and Bell, 1997, Croft, 1999).

Within the great diversity of notoungulates, a common theme is evident: the group had a remarkable propensity to evolve hypsodont (high-crowned) dentitions. This trend occurred independently in at least four clades, resulting each time in species with predominantly or wholly hypselodont (i.e., ever-growing) dentitions (Simpson, 1980, Cifelli, 1985, Bond, 1986, Croft, 1999, Croft et al., 2003, Croft et al., In press). Only among rodents have hypselodont dentitions appeared independently as many times. (The generally accepted monophyly of xenarthrans – sloths, armadillos, glyptodonts – suggests hypselodonty was inherited from a common ancestor in this group; Engelmann, 1985, Rose and Emry, 1993, Delsuc et al., 2001, Gaudin, 2003.) Unlike in rodents, cheek tooth hypselodonty became a characteristic feature in notoungulates, present in all middle Miocene and younger species; it is a remarkable example of parallel evolution. Also noteworthy is the ‘precocious’ appearance of hypsodonty in notoungulates (and various other South American mammals) by the earliest Oligocene; similar changes did not characterize the mammalian herbivores of other continents until roughly 10–15 million yeas later (Patterson and Pascual, 1968, Simpson, 1980, Pascual and Ortiz Jaureguizar, 1990, Flynn and Wyss, 1998, Flynn et al., 2003, Bargo et al., 2006, Croft et al., In press). Clearly, the factors affecting the evolution of hypsodonty (and hypselodonty) in notoungulates must be clarified in order to understand: (1) the adaptive significance of hypsodonty among mammals in general; and (2) the early prevalence of this feature among South American mammals in particular.

The correlation between hypsodonty and diet/habitat has been discussed at length and has primarily focused on ungulates. Rather than attempting to summarize this extensive body of literature in a few paragraphs, we refer the reader to a handful of recent papers that provide useful summaries and discussions of key aspects of the hypsodonty issue in extant mammals: Janis and Fortelius (1988), an overview of hypsodonty and other wear-related dental adaptations in mammals; Williams and Kay (2001), a study that uses phylogenetically independent contrasts to assess hypsodonty in ungulates and rodents; Janis (2007), an overview of the dietary history of ungulates; Clauss et al. (2007), a review of adaptations in diverse groups of browsers and grazers; and Mendoza and Palmqvist (2008), an investigation of the correlation between hypsodonty and other variables in modern ungulates. For the purposes of the present study, we highlight two important aspects of this body of research.

First, investigations of hypsodonty in extant mammals have primarily focused on ungulates, particularly cetartiodactyls and perissodactyls. Only a single study (Williams and Kay, 2001) has examined the adaptive significance of hypsodonty in non-ungulates (rodents). Other than the notoungulate studies listed below, none has attempted to explore the correlation between diet and hypsodonty in extinct groups other than cetartiodactyls and perissodactyls using a method of dietary inference that is more or less independent of hypsodonty such as carbon isotopes, microwear, or mesowear (but see Vizcaíno et al., 2006, for a study of occlusal surface area in relation to diet in some extinct xenarthrans, an exclusively hypselodont group). The value of hypsodonty for inferring diet and/or habitat preference in other groups of mammals therefore remains relatively unexplored.

Second, no investigation of the adaptive role of hypsodonty has distinguished between hypsodont (high-crowned) and hypselodont (open-rooted) species. Such a delineation is unnecessary for extant ungulates (because all have rooted teeth), but should be taken in account for groups that include both very hypsodont and hypselodont species (e.g., rodents). Given a hypsodont and a hypselodont species with equal hypsodonty indices (i.e., relative crown height), the capacity of a hypselodont species to indefinitely offset tooth wear via additional growth makes a hypselodont species functionally different from a merely hypsodont one. Not taking such morphofunctional differences into account might obscure important differences in biology.

Despite the caveats noted above, it is apparent that hypsodonty is correlated both with grazing (i.e., eating grasses) and with open habitat feeding (i.e., feeding in areas of low vegetation) in extant mammals. Of these two factors, habitat appears to be more highly correlated, at least among ungulates (Mendoza and Palmqvist, 2008). Based on analogy with living ungulates, notoungulates generally have been interpreted as grazers and/or open habitat feeders (e.g., Patterson and Pascual, 1968, Simpson, 1980, Cifelli, 1985, Bond, 1986, Kay and Madden, 1997, Shockey, 1997, Flynn et al., 2003, Croft et al., 2004). Nevertheless, recent investigations of notoungulate diets based on carbon isotopes (MacFadden, 2005) and enamel microwear (Townsend and Croft, 2008a) have indicated that some hypselodont notoungulates may have been browsers (i.e., consumers of soft leaves, twigs, and/or fruit) and/or may have fed predominantly in closed habitats. It thus appears that hypselodonty (in notoungulates) may be less correlated with diet and habitat than hypsodonty is (in extant ungulates). This also might be true for other groups of mammals that have not been closely scrutinized.

Our study uses mesowear analysis to test whether functionally hypselodont notoungulates (i.e., notoungulates in which tooth roots formed only in very old individuals, if at all) were grazing and/or feeding predominantly in open environments. Mesowear analysis is a method that gauges the relative roles of attrition (tooth on tooth wear) and abrasion (food on tooth wear) in shaping tooth cusp morphology over an animal's lifetime (Fortelius and Solounias, 2000). Broadly speaking, abrasion dominates in mammals feeding on grasses and other low vegetation, resulting in low, rounded or flat tooth cusps; attrition dominates in mammals feeding on softer leaves and twigs, resulting in high, sharp cusps. Either process can dominate in frugivores, with low, blunt cusps forming when many hard seeds are consumed. The term mesowear was coined to indicate that it reflects the diet of an animal over a period of time greater than that of enamel microwear (formed over days to weeks) but less than what could be termed ‘macrowear,’ wear dictated by population- or species-specific craniodental morphology (developed over many generations) (Fortelius and Solounias, 2000). It primarily has been used to infer diet in extinct species based on correlations between tooth cusp shape and diet in extant mammals. Mesowear has been examined in three major clades of extant ungulates (cetartiodactyls, perissodactyls, hyracoids) and has been used to infer diet in a variety of extinct groups (primarily cetartiodactyls) including antilocaprids (pronghorns), bovids (antelopes, cattle, etc.), cervids (deer), equids (horses), giraffids (giraffes), merycoidodontids (oreodonts), palaeomerycids (extinct horned ruminants), moschids (musk deer), and tragulids (chevrotains and mouse deer) (e.g., Kaiser et al., 2000, Franz-Odendaal and Kaiser, 2003, Kaiser and Fortelius, 2003, Kaiser and Solounias, 2003, Franz-Odendaal and Solounias, 2004, Semprebon et al., 2004, Mihlbachler and Solounias, 2006, Rivals and Semprebon, 2006, Kaiser and Rössner, 2007, Merceron et al., 2007, Rivals et al., 2007a, Rivals et al., 2007b, Schubert, 2007, Semprebon and Rivals, 2007, Rivals et al., 2008).

Mesowear analysis has not been applied previously to endemic South American ungulates such as notoungulates. As detailed below, we chose to study notoungulates from Salla, Bolivia, owing to the ready availability of large samples of well-preserved notoungulate specimens from this locality.

Section snippets

Salla Fauna

The Salla Fauna derives from rich late Oligocene strata exposed as badlands near the town of Sahalla (also known as Salla) in northwest Bolivia, approx. 100 km southeast of La Paz (17° 8–14' S, 67° 37–40' W; Fig. 1). The locality lies within the Eastern Cordillera of the Andes, and the badlands are exposed at elevations of ca. 3500–4000 m. The formation has yet to receive an official geologic designation, and has been referred to simply as the Salla Beds (MacFadden et al., 1985). The 540 m

Study taxa

Mesowear analysis requires a sample of at least 10 individuals, preferably more (Fortelius and Solounias, 2000). We therefore restricted our analysis to the three most common ungulates from Salla: a mesotheriid notoungulate (Trachytherus), an ‘archaeohyracid’ notoungulate (Archaeohyrax), and an interatheriid notoungulate (Federicoanaya) (Fig. 2). Each species is represented in our dataset by 16–21 specimens (see below).

The Salla mesotheriid was recently recognized as a new species of

Discriminant Function Analyses (DFA)

The DFA of the expanded modern ungulate dataset resulted in two canonical discriminant functions, the first of which was significant (p > 0.001, Wilks' lambda). Percentage of individuals with blunt cusps had the greatest influence on this function (loading =  0.791); percentage of individuals with high occlusal relief had an effect nearly as great, but positive (loading = 0.753). Percentage of individuals with sharp cusps was correlated with occlusal relief but with a lower magnitude (loading = 0.594).

Inferred diets of Salla notoungulates

Taken together, these analyses suggest: (1) all three notoungulates included some grass in their diet and/or fed at least part of the time in open habitats; (2) these notoungulates were partitioning available dietary resources by foraging in different areas and/or varying the composition of their diets; and (3) Trachytherus had the most abrasive diet of the three notoungulates, and Federicoanaya had the least abrasive.

Specifically, Trachytherus appears to have been a committed grazer or to have

Conclusions

South American notoungulates represent an exceptional radiation of small to large-bodied herbivores that characterized South American faunas for more than 50 million years. The parallel acquisition of hypselodonty in several notoungulate lineages makes this clade an excellent model for testing hypotheses of the relationship between tooth crown height and diet/habitat in extant and extinct mammals. Such testing requires well-supported dietary interpretations for extinct taxa based on several

Acknowledgments

We thank F. Rivals and G. Semprebon for providing us with unpublished mesowear data and for helpful comments; B. Shockey for providing the photograph of Trachytherus in Fig. 2; B. MacFadden and R. Hulbert for access to fossil specimens at UF; and T. Matson for access to the modern ungulate specimens at the Cleveland Museum of Natural History. We appreciate the efforts of anonymous reviewers, whose suggestions improved this manuscript. This research was supported by the Case Western Reserve

References (108)

  • SemprebonG.M. et al.

    Was grass more prevalent in the pronghorn past? An assessment of the dietary adaptations of Miocene to Recent Antilocapridae (Mammalia: Artiodactyla)

    Palaeogeography, Palaeoclimatology, Palaeoecology

    (2007)
  • BaezA.M. et al.

    Bufonid toads from the late Oligocene beds of Salla, Bolivia

    Journal of Vertebrate Paleontology

    (2004)
  • BargoM.S. et al.

    Hypsodonty in Pleistocene ground sloths

    Acta Palaeontologica Polonica

    (2006)
  • BilletG. et al.

    Late Oligocene mesotheriids (Mammalia, Notoungulata) from Salla and Lacayani (Bolivia): implications for basal mesotheriid phylogeny and distribution

    Zoological Journal of the Linnean Society

    (2008)
  • Billet, G., Reguero, M. and Mamani Quispe, B., in press. Archaeohyracids (Notoungulata, Mammalia) from the late...
  • BondM.

    Los ungulados fósiles de Argentina: evolución y paleoambientes

    Actas, IV Congreso Argentino de Paleontología y Bioestratigrafía, Mendoza

    (1986)
  • CarliniA.A. et al.

    Cingulata del Oligoceno tardio de Salla, Bolivia, Programa y Resúmenes

  • CerdeñoE. et al.

    Los Mesotheriinae (Mesotheriidae, Notoungulata) del Mioceno Superior de La Pampa, Argentina

    Revista Española de Paleontología

    (2001)
  • CifelliR.L.

    South American ungulate evolution and extinction

  • CifelliR.L. et al.

    Notes on Deseadan Macraucheniidae

    Ameghiniana

    (1983)
  • CifelliR.L. et al.

    Systematics of the Adianthidae (Litopterna, Mammalia)

    American Museum Novitates

    (1983)
  • ClaussM. et al.

    The maximum attainable body size of herbivorous mammals: morphophysiological constraints on foregut, and adaptations of hindgut fermenters

    Oecologia

    (2003)
  • ClaussM. et al.

    The morphophysiological adaptations of browsing and grazing mammals

  • CroftD.A.

    Placentals: South American ungulates

  • CroftD.A.

    Cenozoic environmental change in South American as indicated by mammalian body size distributions (cenograms)

    Diversity and Distributions

    (2001)
  • CroftD.A.

    The middle Miocene (Laventan) Quebrada Honda Fauna, southern Bolivia, and a description of its notoungulates

    Palaeontology

    (2007)
  • CroftD.A. et al.

    A new middle Miocene hegetotheriid (Notoungulata: Typotheria) and a phylogeny of the Hegetotheriidae

    Journal of Vertebrate Paleontology

    (2006)
  • CroftD.A. et al.

    Locomotion in the extinct notoungulate Protypotherium

    Palaeontologia Electronica

    (2008)
  • CroftD.A. et al.

    Large archaeohyracids (Typotheria, Notoungulata) from central Chile and Patagonia including a revision of Archaeotypotherium

    Fieldiana: Geology (New Series)

    (2003)
  • CroftD.A. et al.

    Notoungulata and Litopterna of the early Miocene Chucal Fauna, northern Chile

    Fieldiana: Geology (New Series)

    (2004)
  • Croft, D.A., Flynn, J.J. and Wyss, A.R., In press. The Tinguiririca Fauna of Chile and the early stages of...
  • DelsucF. et al.

    The evolution of armadillos, anteaters and sloths depicted by nuclear and mitochondrial phylogenies: implications for the status of the enigmatic fossil Eurotamandua

    Proceedings of the Royal Society of London Series B - Biological Sciences

    (2001)
  • EisenbergJ.F. et al.
  • ElissamburuA.

    Morphometric and morphofunctional analysis of the appendicular skeleton of Paedotherium (Mammalia, Notoungulata)

    Ameghiniana

    (2004)
  • EngelmannG.F.

    The phylogeny of the Xenarthra

  • EngelmannG.F.

    A new Deseadan sloth (Mammalia: Xenarthra) from Salla, Bolivia, and its implications for the primitive condition of the dentition in edentates

    Journal of Vertebrate Paleontology

    (1987)
  • FariñaR.A. et al.

    Body mass estimations in Lujanian (late Pleistocene–Early Holocene of South America) mammal megafauna

    Mastozoología Neotropical

    (1998)
  • ForteliusM. et al.

    Functional characterization of ungulate molars using the abrasion-attrition wear gradient: a new method for reconstructing paleodiets

    American Museum Novitates

    (2000)
  • Franz-OdendaalT. et al.

    Differential mesowear in the maxillary and mandibular cheek dentition of some ruminants (Artiodactyla)

    Annales Zoologici Fennici

    (2003)
  • Franz-OdendaalT.A. et al.

    Comparative dietary evaluations of an extinct giraffid (Sivatherium hendeyi) (Mammalia, Giraffidae, Sivatheriinae) from Langebaanweg, South Africa (early Pliocene)

    Geodiversitas,

    (2004)
  • GaudinT.J.

    Phylogeny of the Xenarthra

    Senckenbergiana Biologica

    (2003)
  • GagnonM. et al.

    Dietary preferences in extant African Bovidae

    Journal of Mammalogy

    (2000)
  • Hitz, R., 1997. Contributions to South American Mammalian Paleontology: New Interatheres (Notoungulata) from Chile and...
  • HitzR. et al.

    New basal Interatheriidae (Typotheria, Notoungulata, Mammalia) from the Paleogene of central Chile

    American Museum Novitates

    (2006)
  • HitzR.B. et al.

    New interatheres (Mammalia, Notoungulata) from the late Oligocene Salla Beds of Bolivia

    Journal of Paleontology

    (2008)
  • HoffstetterR.

    Un gisement de mammifères déséadiens (Oligocène infèrieur) en Bolivie

    Comptes Rendus de la Académie des Sciences, Paris, Série D

    (1968)
  • HoffstetterR.

    Un primate de l'Oligocène infèrieurran sudamerica: Branisella boliviana gen. et sp. nov

    Comptes Rendus de la Académie des Sciences, Paris, Sèrie D

    (1969)
  • HoffstetterR. et al.

    Paraborhyaena boliviana et Andinogale sallensis, deux Marsupiaux (Borhyaenidae) nouveaux de Déséadien (Oligocène inférieur) de Salla (Bolivie)

    Comptes Rendus des Seances de l'Académie des Sciences, Serie 3: Sciences de la Vie

    (1983)
  • JanisC.

    The evolutionary strategy of the Equidae and the origins of rumen and cecal digestion

    Evolution

    (1976)
  • JanisC.

    An evolutionary history of browsing and grazing ungulates

  • Cited by (52)

    • Mesowear study of ungulates from the early Pleistocene site of ‘Ubeidiya (Israel) and the implications for early Homo dispersal from Africa

      2018, Quaternary International
      Citation Excerpt :

      However, this comparative database included taxa with problematic diets (such as duikers, water chevrotains, and hyraxes), and the overall classification had a low reliability (Fortelius and Solounias, 2000). We compiled a comparative dataset categorizing extant euungulate herbivores into three diets only: grazers, browsers and mixed feeders, as in Croft and Weinstein (2008). Our dietary dataset ultimately included 73 modern taxa, with categorizations from Fortelius and Solounias (2000), Croft and Weinstein (2008), and Cantalapiedra et al. (2014).

    • Tooth wear and diets of extant and fossil xenarthrans (Mammalia, Xenarthra) – Applying a new mesowear approach

      2017, Palaeogeography, Palaeoclimatology, Palaeoecology
      Citation Excerpt :

      Holmesina occidentalis has slightly (but not significantly) less blunt mean angle (173°) than Holmesina majus (180°), but these are based on only two specimens for each. Mesowear, especially the relief (relative height of cusp and ridges) on worn tooth surfaces, has been shown to serve as a robust proxy for diet abrasiveness (browser to grazer gradient) of various herbivorous mammal groups, such as ungulates (e.g. Fortelius and Solounias, 2000; Clauss et al., 2007; Croft and Weinstein, 2008; Kaiser et al., 2013; Mihlbachler et al., 2011), marsupials (Butler et al., 2014), rodents and leporines (Ulbricht et al., 2015) and proboscideans (Saarinen et al., 2015). A new methodology based on measuring the relief as angles from between the worn surface features enables a wider spectrum of worn tooth morphologies to be analysed (see Saarinen et al., 2015).

    View all citing articles on Scopus
    View full text