ORIGINAL PAPER The Concept of a Pedolateral Pes Revisited: The Giant Sloths Megatherium and Eremotherium (Xenarthra, Folivora, Megatheriinae) as a Case Study Néstor Toledo1,2 & Gerardo De Iuliis3,4 & Sergio F. Vizcaíno1,2 & M. Susana Bargo1,5 # Springer Science+Business Media, LLC 2017 Abstract The concept of a pedolateral pes in many extinct sloths began effectively with Owen’s mid-nineteenth century descriptions of Glossotherium and Megatherium. Pedolaterality denotes a pes that is habitually inverted, with the digital plane oriented nearly vertically so that weight is borne largely by the lateral digits (mainly metatarsal V) and the plantar surface faces almost entirely medially. Subsequent researchers were strongly influenced by Owen’s interpreta- tions. Astragalar morphology, with the medial and lateral por- tions of its trochlea forming, respectively, a peg-shaped odontoid process and a discoid facet, came to be viewed as a proxy for pedolaterality and, eventually, horizontal rotation around a nearly vertical axis as the main movement of the pes. Such motion necessitates a nearly vertical orientation for the odontoid process. However, analysis of the pes of the Pleistocene megatheriines Megatherium and Eremotherium, the astragalus of which conforms to the type usually interpreted in the literature as indicative of pedolaterality, sug- gests that the pes was not strongly inverted. Rather, the digital plane was about 35o to the horizontal plane, so that weight was borne largely by metatarsal V, but also by metatarsal IV and possibly the ungual phalanx of digit III. The astragalus was positioned so that the odontoid process was oriented obliquely to the vertical axis. With this element so positioned, mediolateral rotation in the horizontal plane was minor, and the main movement of the pes produced dorsiflexion and plantar flexion in nearly the parasagittal plane, the usual movement of the pes in terrestrial mammals. Keywords Xenarthra . Folivora . extinct sloth . foot anatomy . foot inversion Introduction Sloths (Folivora), together with anteaters (Vermilingua), and armadillos (including glyptodonts) and pampatheres (Cingulata), form Xenarthra, one of the most prominent and conspicuous clades of South and Central American placental mammals. The present diversity of sloths comprises only two genera of medium-sized, suspensory forms (two-toed sloths Choloepus and three-toed sloths Bradypus: see Mendel 1985; Reid 1997; Chiarello 2008; Nyakatura et al. 2010) that are distantly related (Patterson and Pascual 1968; Webb 1985; Gaudin 2004). However, their fossil record reveals a consid- erable diversity including more than 90 genera documented since the early Oligocene (see McDonald and De Iuliis 2008; Pujos et al. 2012 for a review). Extinct sloths included * Néstor Toledo ntoledo@fcnym.unlp.edu.ar Gerardo De Iuliis gerry.deiuliis@utoronto.ca Sergio F. Vizcaíno vizcaino@fcnym.unlp.edu.ar M. Susana Bargo msbargo@fcnym.unlp.edu.ar 1 División Paleontología Vertebrados, Unidades de Investigación, Anexo Museo, Facultad de Ciencias Naturales y Museo, Av. 60 y 122, 1900 La Plata, Buenos Aires, Argentina 2 Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Buenos Aires, Argentina 3 Department of Ecology and Evolutionary Biology, University of Toronto, 25 Harbord Street, Toronto, Ontario M5S 3G5, Canada 4 Department of Palaeobiology, Royal Ontario Museum, 100 Queen’s Park Circle, Toronto, Ontario M5S 2C6, Canada 5 Comisión de Investigaciones Científicas de la Provincia de Buenos Aires (CIC), La Plata, Buenos Aires, Argentina J Mammal Evol DOI 10.1007/s10914-017-9410-0 http://orcid.org/0000-0002-6833-3165 mailto:ntoledo@fcnym.unlp.edu.ar http://crossmark.crossref.org/dialog/?doi=10.1007/s10914-017-9410-0&domain=pdf medium-sized climbing forms (White 1993, 1997; Bargo et al. 2012; Toledo et al. 2014; Toledo 2016) to fully terrestrial giants (Casinos 1996; de Toledo 1996; Fariña et al. 1998; Blanco and Czerwogonora 2003; De Esteban-Trivigno et al. 2008), forms able to dig (Zárate et al. 1998; Bargo et al. 2000; Vizcaíno et al. 2001), and even semiaquatic taxa (Muizon et al. 2004; Amson et al. 2014, 2015a, b). Most accepted phylogenetic hypotheses (Fig. 1) recognize five main sloth clades (Gaudin 1995, 2004; Pujos et al. 2012; Amson et al. 2016): Bradypus (Bradypodidae), a lineage with no known fossil record; Megalonychidae (Choloepus and a diversity of extinct forms ranging from small- to large-sized taxa); Nothrotheriidae (comprising a diversity of medium- to large- sized forms); Megatheriidae (including medium-sized forms to giants such as the Pleistocene Eremotherium and Megatherium); and Mylodontidae (large to giant forms, such as Scelidotherium, Glossotherium - well suited for digging- Paramylodon, Mylodon, and the gigantic Lestodon). One of the most intriguing features of the pes of extant and most extinct sloths is the nature of the inward torsion of its elements (see McDonald 1977, 2003, 2007, 2012; Hirschfeld 1985;Webb 1989;McDonald and De Iuliis 2008; Argot 2008; see also Amson and Nyakatura 2017 this special issue). As well, many of their tarsals are considerably different morpho- logically from those of most mammals, and no modern anal- ogies with any other mammal have been proposed (De Iuliis 1996). In extant sloths, inward rotation of the pes (so that the plantar surface faces medially) has been described in relation to their slow-climbing, arboreal habits (see Mendel 1985). In most extinct sloths such rotation resulted in a configuration in which weight was supported mainly by the calcaneum and fifth metatarsal (Mt V) (McDonald 2012), with the re- maining digits having little or no support function. Termed Bpedolateral^ by most authors (see McDonald 1977, 2003, 2007, 2012; Hirschfeld 1985; Webb 1989; Amson et al. 2015a) and Binvertigrade^ by de Toledo (1996), this condition appears to evolve in a complex way, given that the lineages exhibit different degrees of pes modification, and that the most well-known assemblage among older sloths, the early Miocene Santacrucian forms of Patagonia (see White 1993, 1997; Bargo et al. 2012; Toledo 2016), which include many mid-sized members of most sloths lineages, possess a gener- alized, conservative pes without marked digit reduction (Toledo et al. 2015a, b; Fig. 2). However, one of the ear- liest taxa exhibiting the derived astragalar morphology is Octodontotherium from the late Oligocene (Deseadan) of Patagonia, which McDonald (2007) considered the earliest record of pedolaterality. Hirschfeld (1985) argued that fea- tures related to the pedolateral condition are traceable to anteaters, but McDonald (2007, 2012) claimed that pedolaterality was not derived from a common ancestor and that different sloth lineages apparently acquired the condition independently. The concept of pedolaterality may be traced to Owen’s 1842 description of Glossotherium, for which he stated that the pes, when articulated with the crus, rested on the substrate by its outer or lateral margin, and that the discoid facet (lateral crest of the astragalar trochlea) was almost horizontal and hence orthogonal to the odontoid facet (medial crest of the astragalar trochlea; Fig. 3). In his monograph on Megatherium americanum, Owen (1859) described the astragalus as assum- ing a similar orientation as inGlossotherium and that, similarly, Fig. 1 Phylogenetic relationships of Folivora as depicted following hypotheses provided by a, Amson et al. (2016), and b, Gaudin (2004). Megatheriines analyzed in this work are in bold. Extant sloths are underlined J Mammal Evol the tibia articulated with the side of the astragalus and that the pes rested on its outer surface. De Iuliis (1996) proposed that Owen’s interpretation was incorrect and suggested that Owen may have been influenced by the orientation of the fibular facet, which Owen (1859) described as facing dorsally if the astragalus were oriented with the pes in a plantigrade position. Owen further proposed that the odontoid process, oriented nearly vertically, played the role of a kind of pivot for move- ment of the pes, and illustrated the pes with the astragalar odontoid process directed almost entirely vertically. Owen’s ideas and illustrations strongly influenced later researchers, and the presence of a well-developed odontoid process has generally been a proxy for inferring a pes orientation with plan- tar surface facing medially and weight borne largely or entirely by the lateral surface of the pes, influencing the depiction of most extinct sloths as large, slow herbivores (de Toledo 1996). Although Owen did not suggest that rotational pivoting in the horizontal plane (i.e., mediolateral rotation) around the odontoid process was the main movement in the pes of Megatherium (see De Iuliis 1996) rather than dorsiflexion and plantar flexion, these ideas were presented, expanded, and applied to other groups of extinct sloths by, for example, Stock (1917, 1925), Hirschfeld (1985), McDonald (2003, 2007, 2012), and Amson et al. (2015a). Thus, inversion of the pes has been correlated with modi- fication of the ankle joint from an astragalus with a general- ized mammalian configuration (see Polly 2007) towards a highly derived odontoid-discoid morphology, and a change in movement at the cruroastragalar joint from mainly dorsiflexion and plantar flexion to mainly or entirely mediolateral rotation. Once postulated, the pedolateral pes has also then, in turn, been used to interpret other anatomical functions or features: for example, Amson et al. (2015a) relat- ed pedolaterality to knee abduction and hence linked to fem- oral condylar asymmetry and patellar groove asymmetry, among other features (see below). Perhaps the clearest expression of the link between a well- developed odontoid process (that is, with the process and the odontoid facet that it bears meeting the discoid facet at an an- gle of nearly 90o) and pedolaterality is provided byMcDonald (2012: 211): BIn the megalonychids, the angle between the medial and lateral trochlea is close to 0° [altered to 180o by, Fig. 2 Comparative anatomy in dorsal view of right pes of a, Hapalops (an early Miocene Santacrucian sloth; see Toledo et al. 2015a, b) and of b, Megatherium(based on MLP 2-73), showing the conservative morphology of the Santacrucian form as opposed to the highly modified arrangement of the Pleistocene genus. Abbreviations as follows: ast, astragalus; calc, calcaneum; cub, cu- boid; ecto, ectocuneiform; ento, entocuneiform; fmid-pr ph, fused middle and proximal phalanges; meso, mesocuneiform; mid ph, middle phalanx; Mt I, first metatarsal; Mt II, second metatarsal; Mt III, third metatarsal; Mt IV, fourth metatarsal; Mt V, fifth meta- tarsal; nav, navicular; pfmeso-ento, probable fused meso- entocuneiform; ph, vestigial phalanges; pr ph, proximal phalanx; pvMt II, probable vestigial second metatarsal; ung ph, ungual pha- lanx. Scale bar = 5 cm J Mammal Evol for example, Amson et al. (2015a)] while in the most derived taxa such as the mylodontids, Lestodon . . . and Paramylodon . . . or megatheres such as Megatherium and Eremotherium . . . , the angle between the medial and lateral trochlea approaches 90°. The angle between the two trochleae in other sloths such as the nothrotheres (e.g., Nothrotheriops, . . .) and scelidotheriines (e.g., Catonyx, . . .) will vary between these two extremes. If one considers the extremes in the angle between the two trochlear surfaces resulting from the modifi- cation of the medial trochlea, a totally unmodified astragalus with an angle of 0° between the two trochleae has 100% dorsoplantar flexion and 0% mediolateral rotation, while an astragalus with a 90° angle between the two trochleae has 0% dorso- plantar movement and 100% mediolateral rotation. Consequently, there is a direct relationship that with each degree increase in the angle between the medial and lateral trochlea, there is a concurrent loss of dorso-plantar movement with an equivalent gain in mediolateral rotation by the same amount. Therefore, in some clades, including some nothrotheres and scelidotheriines, with an angle between the medial and lateral trochlea that is intermediate between these two extremes, there will be a similar intermediate range of movement of the pes relative to the tibia, consisting of both a dorso-plantar extension-flexion as well as some mediolateral rotation.^ Intermediate degrees of pedolaterality and dorsiflexion- plantar flexion versus mediolateral rotation of the pes have been suggested (see e.g., McDonald 2012; Amson et al. 2015a, 2016) for sloths in which the odontoid process is not as well developed (that is, the angle between odontoid and discoid facets is between 0o (or 180o) and 90o. However, our observations on some sloths (see below) suggest that this gen- erally accepted view (that is, as expressed by McDonald) is not accurate. Fig. 3 Megatheriinae right astragalus (a-e; Eremotherium laurillardi, ROM 22068) and tibia-fibula (f; Eremotherium laurillardi, USNM 457115). a, anterior view (proximal or tibial side towards top; lateral or fibular side towards left); b, dorsal (tibial or proximal) view (anterior towards bottom, lateral towards left); c, medial view (plantar towards bottom, anterior towards left); d, lateral (fibular) view (proximal towards top, anterior towards right); e, dorsolateral view commonly and mistakenly depicted as lateral, with odontoid process oriented nearly vertically; f, distal (astragalar) view of right tibia- fibula (anterior towards top, lateral towards left) showing cochlea tibiae. Abbreviations: aff, astragalar facet of fibula; cub, cuboidal facet of astragalus; dis, discoid facet; dit, discoid groove of cochlea tibiae; efa, ectal facet of astragalus; ffa, fibular facet of astragalus; nav, navicular facet of astragalus; odo, odontoid facet on odontoid process; odt, odontoid facet of cochlea tibiae; sus, sustentacular facet of astragalus. Scale bar = 5 cm J Mammal Evol This contribution considers the criteria for recognizing a pedolateral pes posture, including the movement at the cruroastragalar joint (ankle joint), in terms of the morphology and architectural pattern of the pes as well as functional mor- phology of the ankle joint in the Pleistocene megatheriines Megatherium and Eremotherium, one member of which, as noted above, was among the first examples cited as possessing a pedolateral pes. Such analysis provides insight on the corre- lation between anatomy, posture and function of the pes in these giant extinct sloths, and tests the traditionally accepted concepts of pedolaterality. The questions that require comprehensive consideration and that are specifically addressed in this contribution are: 1) Does the presence of a well-developed odontoid process necessarily indicate full pedolaterality (i.e., in which the plantar surface of the pes faces nearly or entirely medially)? 2) Does the presence of a well-developed odontoid process necessarily indicate that the astragalus moved, at the cruroastragalar joint, mainly or entirely mediolaterally (i.e., rotated) in the horizontal or frontal plane to the com- plete or near exclusion of the typical dorsiflexion and plantar flexion movements in a nearly parasagittal plane of generalized terrestrial mammals at this joint? Material and Methods Institutions American Museum of Natural History, New York, USA (AMNH); Field Museum of Natural History, Chicago, USA (FMNH); Museo Argentino de Ciencias Naturales BBernardino Rivadavia,^ Ciudad Autónoma de Buenos Aires, Argentina (MACN); Museo de La Plata, La Plata, Argentina (MLP); Royal Ontario Museum, Toronto, Canada (ROM); Smithsonian Institution - United States National Museum, Washington, D.C., USA (USNM); Yale Peabody Museum, Vertebrate Paleontology Princeton University, New Haven, USA (YPM-VPPU). Materials Megatherium americanum: FMNH P13662, MACN-PV 54, MACN-PV 12815aa, MACN-PV 514, MLP 2–29, MLP 2– 30, MLP 2–31, MLP 2–207, MLP 2–73, MLP 27-VII-1-1, MLP 44-XII-28-1, MLP 71-II-18-1, ROM 10439. Eremotherium sp.: AMNH 55725, FMNH 207869; E. laurillardi: FMNH P26970, ROM 21928, ROM 21932, ROM 21941, ROM 21953, ROM 21965, ROM 22008, ROM 22068, ROM 221973, ROM 23003, ROM 28856, ROM 28860, ROM 28861, ROM 28863, ROM 28905, ROM 30768; USNM 457115. Prepotherium potens: MACN-A 4694, YPM-VPPU 15345, YPM-VPPU 15521, YPM-VPPU 15568, YPM- VPPU 15568. Several Megatherium and Eremotherium specimens were studied in order to assess the reliability of the correlation be- tween pes morphology and posture. For comparative pur- poses, specimens of the early Miocene megatheriid from Patagonia, Prepotherium, were included. To study function of the ankle joint, specimens with both astragalus and tibia- fibula were articulated to gain insight on the angle and range of mobility at the cruroastragalar joint. When present, other pes elements (e.g., calcaneum, navicular, cuboid, cuneiforms, metatarsals, and phalanges) were articulated as well to analyze correlation between pedal architecture and posture. Anatomical terminology and orientation of limb elements In this work we will follow the anatomical reference frame- work from De Iuliis and Pulerá (2010). To avoid misconcep- tions, Bdigit^ is considered as involving the metapodials and fingers, which are in turn composed by phalanges (see De Iuliis and Pulerá 2010), and hence Bdigital plane^ is used to denote the plane defined by the array formed by metatarsals and phalanges, independent of the loss or fusion of elements. Concerning spatial orientation of pes elements, we considered here a generalized plantigrade mammalian pes as the anatom- ical reference frame for orientation (see Polly 2007). Thus, the following standard anatomical orientations are used in figures: anterior, posterior, dorsal, ventral (effectively plantar), medial, and lateral. The surface of the astragalus articulating with the distal surface of the tibia is considered part of its proximal surface. The distal surface of the astragalus is the surface, facing anteriorly, with which the more distal tarsals articulate. The surface that is toward the substrate is the plantar (or ventral) surface. Thus, a figure depicting the tibia and astragalus articulated in a life orientation and observed from the front of the animal is considered an anterior view, even though the distal surface of the astragalus is depicted. The surface of the tarsals, metatarsals, and phalanges that faces dorsally is considered the dorsal surface of these elements. The following assumptions are made regarding hind limb posture in sloths: 1. The femur was approximately vertical, aligned with the parasagittal plane, as in most mammals (see Polly 2007). 2. The tibia and fibula were approximately vertical, aligned with the parasagittal plane, as in most mammals (see Polly 2007). 3. Some degree of knee abduction was considered, but pre- liminary observations of the knee joint anatomy suggest that the tibia was not strongly deviated from the J Mammal Evol parasagittal plane. Although femoral condylar asymmetry has been proposed as an indicator of knee abduction for extinct sloths (see Amson et al. 2015a, and indirectly as indicative of pedolaterality), it is important to note that knee abduction (with the femur and tibia in articulation and forming an angle in the transverse plane with the knee facing anteriorly; i.e., equivalent to the pathological con- dition genu varum in humans) is not the same as lateral excursion of the knee. The latter implies that the entire hind limb deviates laterally from a parasagittal plane due to mediolateral rotation at the femoroacetabular joint, with the knee directed anterolaterally, a position exhibited to varying degrees in many mammals (see Jenkins 1971: Fig. 4). A larger medial femoral condyle is observed in ground-dwelling extant xenarthrans (see White 1993; Toledo et al. 2015a, b), while fully arboreal ones exhibit subequal femoral condyles, in spite of frequently displaying lateral excursion of the knee. This is also de- scribed for extant marsupials by Argot (2002) and for ex- tant rodents by Candela and Picasso (2008). In addition, Milne et al. (2012) proposed that body size has a strong effect on femoral morphology, with the largest xenarthrans (terrestrial sloths and glyptodonts) exhibiting more strong- ly asymmetrical femoral condyles, and indicating a stance with a more extended knee posture. Finally, many terres- trial mammals exhibit asymmetric femoral condyles and asymmetric patellar trochlear ridges without exhibiting knee abduction or pedolaterality at all (e.g., armadillos, carnivorans, pangolins, artiodactyls, perissodactyls, and elephants, among others; pers. obs., but, see also examples in Lessertisseur and Saban 1971). Indeed, such features are readily observable in Equus, in which there is no question of pedolaterality. Femoral condylar asymme- try induces, also, a differential sliding of the tibial plateau, promoting axial rotation of the crus around its proximodistal axis during knee flexion and extension (see Lovejoy 2007; Milne et al. 2012), although functional interpretations of this feature are still in discussion. In sum- mary, despite isolated efforts to understand features of the knee related with posture and locomotion in extinct sloths (compare discussions of Amson et al. 2015a and Toledo et al. 2015a, b), the subject requires further research. Here, we do not consider any inference or assumption about putative upright posture and locomotion in extinct sloths (see for example Blanco and Czerwogonora 2003). The results of this contribution are not considered with respect to bipedal vs. quadrupedal posture and locomotion in extinct sloths. Results Overall anatomy of pes In Eremotherium and Megatherium, the pes exhibits a highly derived arrangement with respect to the condition observed in early Miocene Santacrucian sloths and other stem megatheriids (see Toledo et al. 2015a, b; Amson et al. 2016). The astragalus bears distinct odontoid (medial trochle- ar crest) and discoid (lateral trochlear crest) facets that meet at a right angle in anterior view (Fig. 3a). The sessile astragalar head bears anteriorly the concave and convex facet for the navicular, with the concave portion approximately centered with the midline of the astragalar trochlea, and the convex portion more nearly plantar to the odontoid process. Plantarly, the head bears a convex facet for the cuboid (Fig. 3). In contrast to the Miocene genus Prepotherium and to other Pliocene megatheriines such as Pliomegatherium (Brandoni 2006) and Pyramiodontherium (Brandoni et al. 2004) where the angle between the fibular facet and astragalar trochlea is slightly acute (less than 90°), the angle between the fibular facet and the astragalar trochlea is slightly obtuse (more than 90°), as in the Miocene Megathericulus (De Iuliis et al. 2008). The calcaneum of both Megatherium and Eremotherium is elongated and wide, but its tuber narrows posteriorly (at its proximal end, following the terminology used by Amson et al. 2015a); it appears to have rested on the substrate along its entire length. The ectal facet, the main articular surface for the astragalus, is convex along its major axis, which is oriented obliquely from mediodorsal to Fig. 4 Right femur, tibia-fibula and astragalus, in disarticulation, to reveal the orientation of the astragalus of megatheriines analyzed in this work: a, orientation followed in this work, with the tibia-fibula and femur approximately vertical and hence a non-inverted astragalus; b, orientation of the tibia-fibula and femur required if astragalus is inverted as suggested by previous literature so that the odontoid process is oriented nearly vertically. Scale bar = 5 cm J Mammal Evol lateroplantar. The sustentacular facet is similarly oriented, as in other extinct sloths, and contiguous lateroplantarly with that fo r the cubo id . The ec tocune i fo rm is s t rong ly anteroposteriorly (or proximodistally) compressed. The ento- and mesocuneiform are usually fused into a single complex. Digits I and II are almost entirely absent in both genera (a reduced Mt II is observed in some specimens). Digit III is the only complete digit of the pes with fused proximal and middle phalanges and a large ungual phalanx bearing a curved ungual core. The lateral metatarsals are large but the corre- sponding phalanges are only vestigial (Fig. 2). Orientation of the cruroastragalar joint Orientation of the mediolateral major axis of the astragalar facets of both the tibia (cochlea tibiae or tibial cochlea) and fibula (forming together the astragalar mortise) is slightly oblique (about 12° in Megatherium americanum, MACN- PV 54) relative to the major axis of the tibial plateau in both Megatherium and Eremotherium. The concave median ridge of the cochlea tibiae, separating the astragalar facets, is orient- ed slightly posteromedially to anterolaterally. Thus, the axis of the medial surface of the cochlea tibiae, articulating with the odontoid facet of the astragalus (and thus the odontoid process itself), is tilted slightly anteriorly with respect to the transverse plane of the tibia (Fig. 3f). This implies that the astragalus (and hence the proximodistal axis of the pes as a whole) was ori- ented slightly laterally with respect to the vertical (or parasagittal) plane of the tibia. Regarding tilting of the ankle in the transverse plane, when the astragalus is articulated with the tibia-fibula (usually fused), the cruroastragalar joint axis does not exhibit a marked vertical inclination (i.e., it is only slightly oblique to the frontal plane) unless the tibia is strikingly abducted at the knee joint (i.e., genu varum) and the anatomical evidence of both the tibia-fibula and femur support our reconstruction – that is, the tibia could only have undergone slight abduction at its proximal end without dislocation of the knee joint (Fig. 4). The medial and lateral facets of the cochlea tibiae do not exhibit marked differences in depth (e.g., MACN-PV 54, MLP 71-II-18-1, USNM 457115), indicating that the cochlea tibiae did not deviate markedly from the plane of the tibial plateau. In turn, this indicates that the astragalar trochlea was not markedly inclined either, similar to the early Miocene Santacrucian megatheriid Prepotherium potens (YPM-VPPU 15345 and 15568), and in contrast to Owen’s (1859) depiction that was supported in later literature (Kraglievich 1928; Gazin 1957; Paula Couto 1978; Hirschfeld 1985; McDonald 2007, 2012). The distal facet of the fibula, converse to the condition observed in P. potens (see Toledo et al. 2015a, b), is oblique to the fibular diaphyseal axis (Fig. 4). In effect, then, when in anatomical position within the astragalar mortise, the astraga- lus was tilted no more than a few degrees, so that the most proximal part of the odontoid process of the astragalus projected proximally just slightly beyond the most lateral sur- face of the discoid facet (Fig. 4a). Inclination of the digital plane InMegatherium and Eremotherium, when the astragalus is in the anatomical position above defined, the navicular facet does not lie entirely dorsal to the cuboidal facet: the medial portion of the former lies approximately medial to the latter. This is because the long axis of the navicular facet is oriented dorsolaterally to plantomedially at approximately 35o to the horizontal axis (or frontal plane). This orientation, in turn, establishes the orientation of the long axis of the ectocuneiform, so that the distal articular surface of the latter, for Mt III, is similarly inclined. The long axis of Mt III is thus likewise inclined so its keeled distal surface is also oblique, rather than dorsoplantarly aligned. This orientation is carried through the fused phalanges proximal to the large ungual pha- lanx of the third digit. The bony ungual core and it keratinous sheath in life were therefore directed plantomedially. Mt IV articulated laterally with Mt V, dorsomedially with Mt III, and proximally with cuboid, and Mt III articulated proximally with the entocuneiform. The spatial arrangement of these articulations were such that the plane of the digital array at the cuboid and ectocuneiform was oblique, at approx- imately 35o, to the horizontal axis. However, the proximodistal orientation of Mt IV was such that its distal end was more plantar in position than its proximal end, as shown by the fused Mt IV and Mt V that occurs in some specimens (e.g., Eremotherium laurillardi, ROM 28905). Mt III was capable of limited movement so the position of its distal end varied slightly. The effect of these proximal to distal differences produced a less oblique digital array distally. In addition, the navicular facet (on the astragalar head) does not lie directly dorsal to the cuboid facet of the calcaneum, but medial and slightly dorsal to it. The navicular articulated with its major axis inclined ventromedially (or plantomedially). Therefore, the digital plane was not disposed vertically as depicted in Owen (1859), but obliquely. The inclination of the digits was not constant proximodistally along the digital array: Mt V and Mt IV rested on the substrate on their lateroplantar surface, whereas the ungual phalanx of digit III rested on its lateral surface. The contact surface with the ground (bony sole) was con- stituted by the ventral (plantar) aspect of the calcaneum, lateroplantar surface ofMt V, lateroplantar surface of the distal part ofMt IV, and the reduced phalanges of the fourth and fifth digits. The lateral surface, or at least part of it, of the ungual phalanx of digit III also appears to have contacted the sub- strate, as discussed by Toledo et al. (2015a, b, 2016), and as indicated also by ichnological evidence (see Aramayo et al. 2015). The oblique orientation of digit III thus served to keep J Mammal Evol part of the large keratinous claw (possibly its tip) free of the substrate. Functional morphology of the cruroastragalar joint Experimental manipulation of the astragalus within the crural mortise was performed in specimens with both the astragalus and tibia-fibula preserved (e.g. M. americanum, MLP 2–30, MACN-PV 54; E. laurillardi, USNM 457115). The displace- ment observed was mainly in the vertical plane (or at least in a plane subparallel to the tibia-fibular diaphyseal axis - see Fig. 5). Minor mediolateral displacement was noted, involving medial displacement during plantar flexion and lateral dis- placement during dorsiflexion (Fig. 6). Thus, in its proper anatomical position, the odontoid process could not have functioned as a subvertical pivot imparting a rotational motion to the pes around a vertical axis; for such motion to have been possible, the odontoid process would have to have been ori- ented with its long axis nearly vertically. These findings indicate that the main movement at the ankle was dorsiflexion-plantar flexion in a more or less parasagittal plane, rather than a rotational movement in the frontal plane (Fig. 5). Therefore, the main ankle movement is similar to that described as the generalized mammal condition, including humans (see Gray 1918; Polly 2007), as was also described for Santacrucian sloths (Toledo et al. 2015a, b). Orientation of the anatomical sole As previous analyses by De Iuliis (1996) indicated, in the megatheriines studied herein, and in accordance with the oblique orientation of the digital plane at the cuboid-ectocuneiform surfaces, the anatomical sole of the pes faced medioventrally (or medioplantarly) in both Eremotherium and Megatherium; in part this is due to the decreased degree of obliquity more distally along the dig- ital array. The only elements raised from the substrate were the navicular (beneath which the presence of a medial soft tissue plantar pad was proposed by Toledo et al. 2016), the cuboid, meso-ectocuneiform, Mt III, and the fused proximal and mid- dle phalanges of the third digit. Discussion As noted in the Introduction, previous literature defined pedolaterality on a number of criteria. The degree of modifi- cation of the astragalus, characterized by the transformation of the medial ridge of the astragalar trochlea into a peg-shaped odontoid process bearing an odontoid facet, whereas the lat- eral ridge arches as a broad discoid facet (De Iuliis 1994; McDonald 2003, 2007, 2012; McDonald and De Iuliis 2008; Amson et al. 2016), has been advanced as an indicator of a pedolateral pes by several authors (Scott 1903–1904; Stock 1917, 1925; Hirschfeld 1985; McDonald 2003, 2007, 2012; Argot 2008). Further, the development of an odontoid-discoid condition has been related to a lateral inclination of the cruroastragalar joint axis resulting in not only a verticalization of the digital array (i.e., a highly and permanently inverted pes), but also a change in ankle movement from dorsiflexion-plantar flexion mainly in the parasagittal plane to rotation in the hori- zontal (or frontal) plane. In this regard, several authors have recognized a pedolateral pes from the presence of odontoid and discoid facets, including Stock (1925), Webb (1989), and McDonald (2007, 2012). A pedolateral pes has been recognized follow- ing this criterion in Mylodontidae, Megatheriidae, and Nothrotheriidae, whereas the condition in Megalonychidae has been described as lacking the traits related to such a modification (McDonald 2007). Thus, Bprimitive^ sloths with a more generalized astragalus show only slight pedolateralization, while Bderived^ sloths with pronounced odontoid and discoid facets (e.g., mylodontines and lestodontines) naturally exhibit a higher degree of pedolateralization. On this criterion and taking into account that the astragalus of extant sloths lacks an odontoid process, it Fig. 5 Experimental movement of right cruroastragalar joint of megatheriines, showing that movement occurs mainly in the same plane as the tibia-fibula diaphyseal axis, as may be noted by the degree of deviation of the navicular facet of the astragalus between dorsiflexed and plantar flexed positions (modified from De Iuliis 1996). a, b anterior view; c, d lateral view. The astragalus is a and c, in dorsiflexed position, b and d, in plantar flexed position. Scale bar = 10 cm J Mammal Evol also follows that they are not pedolateral (McDonald 2007). According to Hirschfeld (1985) and McDonald (2003, 2007), Bderived^ astragalar features linked to a pedolateral stance are: 1- presence of an odontoid process; 2- discoid and odontoid facets approximately orthogonal to each other; 3- discoid facet flattened posteriorly; and 4- discoid facet posteriorly extended with respect to the odontoid facet. Variation of the pedolateral pattern, according to McDonald (2007), involves differences in the proportion of the calcaneum resting on the surface, and the angle between facets on Mt V for the cuboid and Mt IV. The cuboid served as point of transmission of weight from the calcaneum toward digit V (Hirschfeld 1985; McDonald 2007, 2012). McDonald (2012: 213) also considered that concurrent Bwith the in- creased amount of mediolateral rotation of the pes, there is an increase in the degree of curvature of the lateral margin of the astragalus,^ and that the Bgreatest amount of curvature of the lateral margin of the astragalus is seen in the megatheres.^ Hirschfeld (1985) proposed that the extant anteaters Myrmecophaga and Tamandua could be used as a model of basal mammalian foot architecture and ancestral type for ground sloths. Features of the more well-represented early Miocene Santacrucian sloths (e.g.,Hapalops, Analcimorphus) are congruent with Hirschfeld’s hypothesis and also with the traditional depiction of a generalized mammalian pes sensu Polly (2007): a pentadactyl, plantigrade low-arched pes with- out significant reduction and/or fusion of elements (Scott 1903–1904; Bargo et al. 2012; Toledo et al. 2015a, b). According to Hirschfeld (1985), modifications from this gen- eralized condition towards a derived, specialized foot pattern of Pleistocene terrestrial sloths involves: 1- loss or reduction of inner digits; 2- increase of size and robustness of outer digits (which does not occur in extant sloths); 3- dorso-medial rotation of the arch of the foot (which oc- curs in most, including extant, sloths) with transference of weight bearing towards the lateral surface (but not in ex- tant sloths); 4- posterior extension of the calcaneum (as in most sloths and also in other plantigrade mammals such as humans); 5- dorsal development of the medial astragalar condyle (not applicable to extant sloths and the orientation dependent on the reconstructed position of the astragalus); and 6- an unspecified Bbasic reorganization of the structural re- lationship of various tarsal and metatarsal elements^ (Hirschfeld 1985: 58). However, the results of the analyses of the present con- tribution indicate that, at least for the megatheriines stud- ied herein, there is no obvious correlation between the presence of an odontoid-discoid condition and the verticalization of the cruroastragalar joint axis and digital array (without application of marked knee abduction so pronounced that it would cause dislocation). Indeed, we have found no evidence that relates the transformation of the medial trochlea into an odontoid process and a change of the direction of movement of the pes from mainly dorsiflexion-plantar flexion into mediolateral rotation, in contrast to Hirschfeld’s (1985) and McDonald’s (2007 and 2012) suggestions. Moreover, the criterion for determin- ing direction of the cruroastragalar rotation axis is the inclination of the plane between the astragalar trochlear plane and the cochlea tibiae. Findings on specimens ana- lyzed in this work are coherent with prior unpublished analyses performed by one of us in his dissertation (De Iuliis 1996). Therefore, and converse to the descriptions based on Owen (1859), the pes of megatheres was not fully inverted as a whole. When the femur and tibia are aligned approx- imately in the parasagittal plane, the position and orienta- tion of both the astragalus and calcaneum are, in the spec- imens analyzed, similar to those of other mammals, as in- dicated also by previous literature (see McDonald 2003, 2007) and that Toledo et al. (2015a, b) recognized in early Miocene Santacrucian sloths. Only the digital plane is slightly inclined, as in early Miocene sloths and, for exam- ple, in the human pes (whereas it is orthogonal in extant sloths) and perhaps different to what Amson et al. (2016: 9) characterized as metatarsals Bstacked partly dorsoven- trally.^ Thus, Owen’s depiction of a vertical digital array in Megatherium is not appropriate. The relative positions of the astragalar head and cuboidal facets of both the astrag- alus and calcaneum determine the inclination of the digital array. Therefore, inclination of the digital plane (and hence inversion of digits) is related more closely to the relative positions of the navicular and cuboid (and so to the relative position of the astragalar head with respect to the calcaneum, as McDonald (2003) stated that the rotation of the foot occurs) than to the degree of development of odontoid and discoid facets. As well, the idea that a strongly curved lateral margin of the astragalus, most markedly so in megatheriines, that was noted by McDonald (2012) as another indication of mediolateral rotation of the pes, requires, based on the results of this study, further scrutiny. It is worth noting that both specimens ofMegatheriummounted at present in the MLP and Paris galleries reconstruct the pes in a posture cor- roborated by the findings presented here (Figs. 7 and 8). Thus, it is not that factors distal to the cruroastragalar joint have been ignored in understanding the inverted pes in many terrestrial sloths, but that the extent of their role has been overshadowed in deference to the importance of astragalar morphology. This is clearly illustrated by McDonald’s (2012: 209) statement that while Bthe rotation of the foot oc- curs as a functional complex resulting in the modification of J Mammal Evol many bones in the pes, the astragalus is the one bone that shows the highest degree of departure from the primitive mammalian condition and the most distinctive changes in morphology.^ However, as demonstrated in this contribution (and summarized below), it would seem that the astragalus in megatheriines, despite possessing among the more notably developed odontoid and discoid facets among sloths, func- tions much like the astragalus in a typical plantigrade mam- mal, with most movement occurring in a nearly parasagittal plane. In other words, the odontoid and discoid morphology does not have much bearing on an orientation of the astragalus that is notably different than usual among mammals, and thus with producing the marked inversion that has come to be considered characteristic of many terrestrial sloths. Perhaps this should not come as a surprise: as is generally well known, inversion and eversion movements of the pes in humans are not produced mainly at the cruroastragalar joint, but primarily at the joints among the astragalus, calcaneum, and navicular, and the subtalar joints (Gray 1918; Polly 2007). With regard to functional morphology, movement at the cruroastragalar joint occurs mostly in the parasagittal plane, producing dorsiflexion and plantar flexion. Slight rotation around the vertical axis also occurs so that the pes moves slightly medially in plantar flexion and slightly laterally in dorsiflexion. Such deviation from a strictly parasagittal plane, however, is not atypical in many mammals. For example, manipulation of the equine astragalus and tibia demonstrates such motion (albeit in the opposite direction) during dorsiflexion and plantar flexion. The combined movement described by the pes inMegatherium and Eremotherium thus involves mainly plantar extension (with slight medial excur- sion) and mainly dorsiflexion (with slight lateral excursion), as in other mammals (see Lewis 1980). Nevertheless, in no specimen studied here does the pes appear to have moved mainly in the horizontal plane and hence did not rotate horizontally around the odontoid process, in contrast to pro- posals made in previous literature (Hirschfeld 1985; McDonald 2003, 2007, 2012). In this context, changes in the actions of the pes and the digital extensor and flexor muscu- lature (with respect to the generalized mammalian function) proposed by Hirschfeld (1985) and McDonald (2003, 2007, 2012) may require revision for megatheriines. The characteristics of the megatheriine astragalus that de- part most markedly from those of typical plantigrade mam- mals are the odontoid process (and its facet) and the discoid facet. As noted, these characteristics have been explained as due mainly to the development of an inverted pes and a mediolateral rotation of the pes. However, this posture and function do not seem to be adequate explanations for the form of these features, at least not in megatheriines. In some ways, these articular surfaces are so notably distinct because of the disparity in the size of the medial and lateral surfaces of the astragalar trochlea. This seems to have been caused by a re- duction in the extent of the medial surface, which has thus produced the peg-shaped form of the odontoid process, rather than an increase in size of the lateral surface. The deep groove between these facets is atypical of the astragalus of the pes of most other plantigrade mammals, but is typical of many cur- sorial mammals such as equids and cervids (but in which the medial and lateral facets of the astragalus are nearly equal in size). The function of this morphology in such mammals is generally cited as being the enhancement of stabilization of the cruroastragalar joint to restrict movement in the parasagittal plane, thereby enhancing efficiency in rapid loco- motion (see Polly 2007). Clearly, whereas speed was unlikely a concern among sloths, stabilization of the cruroastragalar joint must have been important, although this does not explain the morphology of the odontoid process (i.e., the marked disparity between the lateral and medial surfaces of the astragalar trochlea). Fig. 6 Experimental movement of left cruroastragalar joint in Eremotherium, USNM457115, in anterior view, from dorsiflexed (left) to plantar flexed (right) positions. Black circles show relative position of the same anatomical point, dotted line describes the total displacement, and the white BL^ depicts relative vertical and horizontal displacement J Mammal Evol Concluding Remarks Summarizing, the only criteria proposed by Hirschfeld (1985) for mylodontids that are exhibited in themegatheriines studied in the current analysis are: 1- loss or reduction of inner digits (also occurring in extant but not in early Miocene sloths); 2- increase in size and robustness of the lateral digits (which does not occur in extant sloths, in which digit V is reduced or absent and the remaining digits are more or less equally developed), and of only their metapodial segments, whereas their phalangeal elements are vestigial or absent; 3- inclination of the digital array (equivalent to Hirschfeld’s 1985 Bdorsomedial rotation of the arch of the pes^), shared bymost, including extant, sloths, with transference of weight bearing towards the lateral side (not applicable to extant sloths). thus, the results of the analyses conducted here suggest that the response to both questions posed in the Introduction (does the presence of a well-developed odontoid process necessarily indicate 1- full pedolaterality, and 2- that the astragalus moved –rotated– at the cruroastragalar joint, mainly or entirely mediolaterally in the horizontal or frontal plane) is no. The traditionally accepted views of pedolaterality suggest an adaptive link between the development of an odontoid process and pedolaterality. That is, the function of having an odontoid process in its most extreme expressions is that it permits a pedolateral posture (with plantar surface facing nearly or entirely medially) and rotation of the pes in the horizontal plane. Our results suggest that such a tight causal relationship may not be warranted. We do not claim that there is no link. However, if such a causal relationship exists, it does it does not appear to be supported by the conditions described for Megatherium and Eremotherium. This, in turn, may mean that we need to recon- sider the functional significance of a well-developed odontoid process. Indeed, it is proposed here that the key anatomical criterion for defining pedolaterality in megatheriines is inclination of the digital plane, which can be inferred from the relative positions of the navicular and cuboid. It is necessary here to expand the research to include not only the more derived extinct terrestrial sloths, but also extant sloths, in which verticalization of the digital plane is verified without development of an odontoid- discoid astragalar condition, and even anteaters. Studies in progress by the current authors will expand the survey to other taxa and lineages in order to shed light on the evolution of the pedal architecture of sloths and include muscular and ligamen- tous reconstruction to assess pes function. Fig. 8 Articulated specimen of Megatherium americanum exhibited in theMuséumNational d’histoire Naturelle (MNHN), Paris, France. Note the position of astragalus respect to the tibia- fibula and the low inclined digital plane, similar to that exhibited by the MLP specimen of Fig. 7 Fig. 7 Dorsal (left) and lateral (right) views of left pes of a mounted specimen of Megatherium americanum at the Museo de La Plata, La Plata, Argentina (MLP 2–73). The perspective in the lateral view appears to show the digital plane as vertically oriented but compare with the dorsal view and Fig. 2b for an appreciation of the position of the digits. Scale bar = 5 cm J Mammal Evol Acknowledgements The authors thank collection managers and chairs of institutions housing specimens analyzed in this work (AMNH, FMNH, USNM, MACN, MLP, ROM, YPM) for kindly granting access to mate- rials in their care; to J. Nyakatura, J.C. Fernicola, R. Kay, G.H. Cassini, and A. Racco for providing valuable suggestions; and to Eli Amson and one anonymous reviewer for greatly improving the quality of this work. This contributionwas presented during the ICVM11meeting in Bethesda (Washington, D.C.) in July 2016, as part of the symposium Morphology and Evolution of the Xenarthra organized by M. Susana Bargo and John A. Nyakatura. Attendance at the congress was partially granted by UNLP Viajes y Estadías 2016 to SFV and NT, and CIC 1827/15 (Comisión de Investigaciones Científicas de la provincia de Buenos Aires) to MSB. This is a contribution to the projects UNLP 11/N750 (Universidad Nacional de La Plata) and PICT 2013-0386 and 0389 (Agencia Nacional de Promoción Científica y Técnica). Data sharing is not appli- cable to this article as no datasets were generated or analyzed during the current study. Compliance with ethical standards Conflicts of interest None References Amson E, Argot C, McDonald HG, Muizon C de (2015a) Osteology and functional morphology of the hind limb of the marine sloth Thalassocnus (Mammalia, Tardigrada). J Mammal Evol 22:355- 419. https://doi.org/10.1007/s10914-014-9274-5 Amson E, Argot C, McDonald HG, Muizon C de (2015b) Osteology and functional morphology of the axial postcranium of the marine sloth Thalassocnus (Mammalia, Tardigrada) with paleobiological impli- cations. J Mammal Evol 22: 473–518 Amson E, Muizon C de, Gaudin TJ (2016) A reappraisal of the phylog- eny of the Megatheria (Mammalia: Tardigrada), with an emphasis on the relationships of the Thalassocninae, the marine sloths. Zool J Linn Soc. https://doi.org/10.1111/zoj.12450 Amson E, Muizon C de, Laurin M, Argot C, de Buffrénil V (2014) Gradual adaptation of bone structure to aquatic lifestyle in extinct sloths from Peru. Proc Roy Soc B-Biol Sci 281(1782): 1–6 Amson E, Nyakatura JA (2017) The postcranial musculoskeletal system of xenarthrans: insights from over two centuries of research and future directions. J Mammal Evol. https://doi.org/10.1007/s10914- 017-9408-7 Aramayo SA, Manera de Bianco T, Bastianelli NV, Melchor RN (2015) Pehuen Co: updated taxonomic review of a late Pleistocene ichnological site in Argentina. Palaeogeogr Palaeoclimatol Palaeoecol 439: 144–165 Argot C (2002) Functional-adaptive anatomy of the hindlimb in the Didelphidae, and the paleobiology of the Paleocene marsupials Mayulestes ferox and Pucadelphys andinus. J Morphol 253: 76–108 Argot C (2008) Changing views in paleontology: the story of a giant (Megatherium, Xenarthra). In: Sargis EJ, Dagosto M (eds) Mammalian Evolutionary Morphology: A Tribute to Frederick S. Szalay. Springer Science, Dordrecht, pp 37–50 Bargo MS, Toledo N, Vizcaíno SF (2012) Paleobiology of the Santacrucian sloths and anteaters (Xenarthra, Pilosa). In: Vizcaíno SF, Kay RF, Bargo MS (eds) Early Miocene Paleobiology in Patagonia: High-latitude Paleocommunities of the Santa Cruz Formation. Cambridge University Press, Cambridge, pp 216–242 Bargo MS, Vizcaíno SF, Archuby FM, Blanco RE (2000) Limb bones proportions, strength and digging in some Lujanian (late Pleistocene-early Holocene) mylodontid ground sloths (Mammalia. Xenarthra). J Vertebr Paleontol 20: 601–610 Blanco RE, Czerwonogora A (2003) The gait of Megatherium Cuvier 1796 (Mammalia, Xenarthra, Megatheriidae). Senckenberg Biol 83: 61–68 Brandoni D (2006) A review of Pliomegatherium Kraglievich 1930 (Xenarthra: Phyllophaga: Megatheriidae). N Jb Geol Paläontol Mh 4: 2012–2224 Brandoni D, Carlini AA, Pujos F, Scillato-Yané GJ (2004) The pes of Pyramiodontherium bergi (Xenarthra, Tardigrada): the most com- plete pes of a Tertiary Megatheriinae. Geodiversitas 26: 643–659 Candela AM, Picasso MBJ (2008) Functional anatomy of the limbs of Erethizontidae (Rodentia, Caviomorpha): indicators of locomotor behavior in Miocene porcupines J Morphol 269: 552–593 Casinos A (1996) Bipedalism and quadrupedalism in Megatherium: an attempt at biomechanical reconstruction. Lethaia 29: 87–96 Chiarello AG (2008) Sloth ecology: an overview of field studies. In: Vizcaíno SF, Loughry WJ (eds) The Biology of the Xenarthra. University Press of Florida, Gainesville, pp 269–280 De Esteban-Trivigno S, Mendoza M, De Renzi M (2008) Body mass estimation in Xenarthra: a predictive equation suitable for all qua- drupedal terrestrial placentals? J Morphol 269: 1276–1293 De Iuliis G (1994) Relationships of the Megatheriinae, Nothrotheriinae, and Planopsinae: some skeletal characteristics and their importance for phylogeny. J Vertebr Paleontol 14: 577–591 De Iuliis G (1996) A systematic review of the Megatheriinae (Mammalia: Xenarthra: Megatheriidae). Ph.D. dissertation, University of Toronto, Toronto, 756 pp De Iuliis G, Brandoni D, Scillato-Yané GJ (2008) New remains of Megathericulus patagonicus Ameghino, 1904 (Xenarthra, Megatheriidae): information on primitive features of megatheriines. J Vertebr Paleontol 28(1): 181–196 De Iuliis G, Pulera D (2010) The Dissection of Vertebrates: A Laboratory Manual. 2nd ed. Academic Press, Amsterdam, 332 pp de Toledo PM (1996) Locomotor patterns within Pleistocene sloths. Unpublished Ph.D Dissertation. University of Colorado, Boulder, 316 pp Fariña RA, Vizcaíno SF, Bargo MS (1998) Body mass estimation in Lujanian (late Pleistocene-early Holocene of South America) mam- mal megafauna. Mastozool Neotropical 51:87–108 Gaudin TJ (1995) The ear region of edentates and the phylogeny of the Tardigrada (Mammalia, Xenarthra). J Vertebr Paleontol 15: 672–705 Gaudin TJ (2004) Phylogenetic relationships among sloths (Mammalia, Xenarthra, Tardigrada): the craniodental evidence. Zool J Linn Soc 140: 255–305 Gazin CL (1957) Exploration of the remains of giant ground sloths in Panama. Smithsonian Annu Report 1956: 341–354 Gray H (1918) Anatomy of the Human Body. Bartebly, Philadelphia Hirschfeld SE (1985) Ground sloths from the Friasian La Venta fauna, with additions to the Pre-Friasian Coyaima fauna of Colombia, South America. Univ Calif Publ Geol Sci 128: 1–91 Jenkins FA Jr (1971) Limb posture and locomotion in the Virginia opos- sum (Didelphis marsupialis) and in other non-cursorial mammals. J Zool Lond 165: 303–3015 Kraglievich L (1928) Apuntes para la geología y paleontología de la República Oriental del Uruguay. Revista de la Sociedad Amigos de la Arqueología 2: 197–203 Lessertisseur J, Saban R (1971) Squelette apendiculaire. In: Grassé PP (ed) Traitè de Zoologie, Tome XVI, 3rd Fasc. Mason et Cie, Paris, pp 709-1078 Lewis OJ (1980) The joints of the evolving foot. Part I. The ankle joint. J Anat 130 (3): 527-543 Lovejoy CO (2007) The natural history of human gait and posture Part 3. The knee. Gait & Posture 25: 325–341 McDonald HG (1977) Description of the osteology of the extinct gravigrade edentate, megalonyx, with observations on its ontogeny, phylogeny and functional anatomy. M.S. Thesis, Department of Zoology University of Florida, Gainesville, 328 pp J Mammal Evol https://doi.org/10.1007/s10914-014-9274-5 https://doi.org/10.1111/zoj.12450 https://doi.org/10.1007/s10914-017-9408-7 https://doi.org/10.1007/s10914-017-9408-7 McDonald HG (2003) Xenarthran skeletal anatomy: primitive or de- rived? (Mammalia, Xenarthra). Senckenberg biol 83: 5–17 McDonald HG (2007) Biomechanical inferences of locomotion in ground sloths: integrating morphological and track data. In: Lucas SG, Spielmann JA, MG Lockley (eds) Cenozoic Vertebrate Tracks and Traces. New Mexico Museum of Natural History and Science Bulletin 42: 201–208 McDonald HG (2012) Evolution of the pedolateral foot in ground sloths: patterns of change in the astragalus. J Mammal Evol 19:209–215 McDonald HG, De Iuliis G (2008) Fossil history of sloths. In: Vizcaíno SF, Loughry WJ (eds) The Biology of the Xenarthra. University Press of Florida, Gainesville, pp 24–36 Mendel FC (1985) Adaptations for suspensory behavior in the limbs of two-toed sloths. In: Montgomery GG (ed) The Ecology and Evolution of Armadillos, Sloths and Vermilinguas. Smithsonian Institution Press, Washington, D.C., pp 151–162 Milne N, Toledo N, Vizcaíno SF (2012) Allometric and group differences in the xenarthran femur. J Mammal Evol 19: 199-208 Muizon C de, McDonald HG, Salas R, Urbina M (2004) The youngest species of the aquatic sloth Thalassocnus and a reassessment of the relationships of the nothrothere sloths (Mammalia, Xenarthra). J Vertebr Paleontol 24: 387–397 Nyakatura JA, Petrovitch A, Fischer MS (2010) Limb kinematics of the two-toed sloth (Choloepus didactylus, Xenarthra) and its implica- tions for the evolution of the sloth locomotor apparatus. Zoology 113: 221–234 Owen R (1842) Description of the Skeleton of an Extinct Gigantic Sloth, Mylodon robustus Owen, with Observations on the Osteology, Natural Affinities, and Probable Habits of the Megatherioid Quadrupeds in General. R & J Taylor, London, 176 pp Owen R (1859) On theMegatherium (Megatherium americanum, Cuvier and Blumenbach). Part V. Bones of the posterior extremities. Phil Trans Roy Soc Lond B 149: 809–829 Patterson B, Pascual R (1968) The fossil mammal fauna of South America. Q Rev Biol 43(4): 409–451 Paula Couto C de (1978) Mamíferos fósseis do Pleistoceno do Espírito Santo. An Acad Bras Cienc 50: 365–379 Polly PD (2007) Limbs in mammalian evolution. In: Hall BK (ed) Fins into Limbs: Evolution, Development, and Transformation. University of Chicago Press, Chicago, pp 245–268 Pujos F, Gaudin TJ, De Iuliis G, Cartelle C (2012) Recent advances on variability, morpho-functional adaptations, dental terminology, and evolution of sloths. J Mammal Evol 19: 159–169 Reid F (1997) A Field Guide to the Mammals of Central America and Southeast Mexico. Oxford University Press, Oxford, 398 pp ScottWB (1903–1904)Mammalia of the Santa Cruz beds. I. Edentata. In: Scott WB (ed) Reports of the Princeton University Expeditions to Patagonia 1896–1899. Princeton University Press, Princeton, pp 1– 364 Stock C (1917) Structure of the pes inMylodon harlani. Univ Calif Publ 10 (16): 267-286 Stock C (1925) Cenozoic gravigrade edentates of western North America with special reference to the Pleistocene Megalonychinae and Mylodontidae of Rancho La Brea. Carnegie Inst Wash 331:1–206 Toledo N (2016) Paleobiological integration of Santacrucian sloths (early Miocene of Patagonia). Ameghiniana 53:100–141 Toledo N, Bargo MS, Vizcaíno SF (2015a) Muscular reconstruction and functional morphology of the hind limb of Santacrucian (early Miocene) sloths (Xenarthra, Folivora) of Patagonia. Anat Rec 298: 842–864 ToledoN, Cassini GH,Vizcaíno SF, BargoMS (2014)Mass estimation of Santacrucian sloths from the earlyMiocene Santa Cruz Formation of Patagonia, Argentina. Acta Palaeontol Pol 59:267–280 Toledo N, Racco A, Bargo MS, Vizcaíno SF, Fernicola JC (2015b) Pedolateralización, apoyo del pie y soporte del peso corporal en perezosos fósiles (Xenarthra, Folivora). Reunión de Comunicaciones de la Asociación Paleontológica Argentina. Mar del Plata, Buenos Aires, Actas: 26 Toledo N, Racco A, Bargo MS, Vizcaíno SF, Fernicola JC (2016) Pedolateralization, foot anatomy, and weight support in extinct sloths (Xenarthra, Folivora). 11th International Congress of Vertebrate Morphology. Washington, USA, Meeting Abstracts: 208 Vizcaíno SF, Zárate M., Bargo MS, Dondas A (2001) Pleistocene bur- rows in the Mar del Plata area (Buenos Aires Province, Argentina) and their probable builders. Acta Palaeontol Pol 46:157–169 Webb SD (1985) The interrelationships of tree sloths and ground sloths. In: Montgomery GG (ed) The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C., pp 105–112 Webb SD (1989) Osteology and relationships of Thinobadistes segnis, the first mylodont sloth in North America. In: Redford KH, Eisenberg JF (eds) Advances in Neotropical Mammalogy. The Sandhill Crane Press, Inc., Gainesville, pp 469–532 White JL (1993) Indicators of locomotor habits in xenarthrans: evidence for locomotor heterogeneity among fossil sloths. J Vertebr Paleontol 13:230–242 White JL (1997) Locomotor adaptations in Miocene xenarthrans. In: Kay RF, Madden RH, Cifelli RL, Flynn JJ (eds) Vertebrate Paleontology in the Neotropics. The Miocene Fauna of La Venta, Colombia. Smithsonian Institution Press, Washington, D.C., pp 246–264 Zarate MA, Bargo MS, Vizcaíno SF, Dondas A, Scaglia O (1998) Estructuras biogénicas en el Cenozoico tardío de Mar del Plata (Argentina) atribuibles a grandes mamíferos. Asociación Argentina de Sedimentología Revista 5 (2): 95–103 J Mammal Evol The... Abstract Introduction Material and Methods Institutions Materials Anatomical terminology and orientation of limb elements Results Overall anatomy of pes Orientation of the cruroastragalar joint Inclination of the digital plane Functional morphology of the cruroastragalar joint Discussion Concluding Remarks References