Upper Limb Fossils of Homo naledi from the Lesedi Chamber , Rising Star System , South Africa

Homo naledi skeletal material described from the Dinaledi Chamber, Rising Star System, in the Cradle of Humankind, South Africa, includes upper limb material with remarkably ape-like morphology occurring in the context of a distinctly modern human-like lower limb, foot, and hand. Here we describe upper limb specimens from a new fossil hominin site within the Rising Star System, the Lesedi Chamber (Site U.W. 102), including an intact clavicle and a well-preserved proximal humerus and proximal ulna. Craniodental remains, in association with the postcranial elements described here, have been attributed to H. naledi. The upper limb material from the Lesedi Chamber is gracile and resembles the upper limb material from the Dinaledi Chamber in overall morphology. The primitive humeral morphology of the Dinaledi material is replicated within the Lesedi material. The fossils from the Lesedi Chamber material also preserve additional humeral morphology not represented in the Dinaledi Chamber, providing new information on humeral form in this species. Three-dimensional geometric morphometric analysis of the Lesedi proximal humerus demonstrates affinities with the humeri of Pan and Australopithecus. The complete clavicle from Lesedi is similarly primitive, supporting previous interpretations of the H. naledi shoulder as being more superiorly positioned than modern humans, and closer to the hypothesized australopith condition. The ulna is gracile, with a mediolaterally narrow olecranon process and anterior-facing trochlear notch similar to the derived state suggested for other fossil hominins. These new findings support the interpretation of overhead reaching and climbing behaviors having continued relevance in the locomotor repertoire of Homo naledi and provide a clearer picture of the upper limb morphology of this species. PaleoAnthropology 2019: 311−349. © 2019 PaleoAnthropology Society. All rights reserved. ISSN 1545-0031 doi:10.4207/PA.2019.ART134 ELEN M. FEUERRIEGEL Primate Evolutionary Biomechanics Laboratory, Department of Anthropology, Box 353100, University of Washington, Seattle, WA 98195-3100, USA; and, Evolutionary Studies Institute and Centre for Excellence in Palaeosciences, University of the Witwatersrand, Private Bag 3, Wits 2050, SOUTH AFRICA; efeuer@uw.edu JEAN-LUC VOISIN Aix Marseille Université, CNRS, EFS, ADES, Marseille, FRANCE; jeanlucvoisin2004@yahoo.fr STEVEN E. CHURCHILL Department of Evolutionary Anthropology, Box 90383, Duke University, Durham, NC 27708, USA; and, Evolutionary Studies Institute and Centre for Excellence in Palaeosciences, University of the Witwatersrand, Private Bag 3, Wits 2050, SOUTH AFRICA; churchy@duke.edu MARTIN HAEUSLER Institute of Evolutionary Medicine, University of Zürich, Winterhurerstr. 190, CH-8057 Zürich, SWITZERLAND; Martin.Haeusler@iem.uzh.ch SANDRA MATHEWS Institute of Evolutionary Medicine, University of Zürich, Winterhurerstr. 190, CH-8057 Zürich; and Paul Scherrer Institut, CH-5232 Villigen, SWITZERLAND; sandra.mathews@psi.ch PETER SCHMID Anthropological Institute and Museum, University of Zürich, Winterthurerstr. 190, CH-8057, SWITZERLAND; and, Evolutionary Studies Institute and Centre for Excellence in Palaeosciences, University of the Witwatersrand, Private Bag 3, Wits 2050, SOUTH AFRICA; smidi@aim.uzh.ch JOHN HAWKS Department of Anthropology, University of Wisconsin-Madison, Madison, WI 53593, USA; and, Evolutionary Studies Institute and Centre for Excellence in Palaeosciences, University of the Witwatersrand, Private Bag 3, Wits 2050, SOUTH AFRICA; jhawks@wisc.edu LEE R. BERGER Evolutionary Studies Institute and Centre for Excellence in Palaeosciences, University of the Witwatersrand, Private Bag 3, Wits 2050, SOUTH AFRICA; Lee.Berger@wits.ac.za submitted: 3 December 2018; accepted 5 March 2019 312 • PaleoAnthropology 2019 cally human-like hand (Arsuaga et al. 1997; Carretero et al. 1997; Churchill and Trinkaus 1990; Niewoehner et al. 1997; Trinkaus 1977; Trinkaus 1983; Trinkaus and Villemeur 1991; Vandermeersch and Trinkaus 1995). Even Homo floresiensis (initially recovered only from Late Pleistocene contexts (Sutikna et al., 2016), though recent discoveries from Mata Menge might represent a Middle Pleistocene occurrence of this species (van den Bergh et al. 2016)), appears to have been a committed biped with limited evidence for climbing as a major component of its locomotor repertoire, despite maintaining a primitive shoulder configuration (for more details, see Larson 2007). Contravening this upper limb pattern for Middle Pleistocene Homo is Homo naledi. Previous work on the upper limb material attributed to H. naledi from the Dinaledi Chamber in the Rising Star cave system (Gauteng Province, South Africa) suggested that H. naledi exhibited a suite of markedly primitive morphologies (Feuerriegel et al. 2017), and these traits were indicative of continued involvement of the upper limb in locomotion (Kivell et al. 2015). Dated to 335–236 ka (Dirks et al. 2017), H. naledi appears to represent an exception to an otherwise robust pattern of non-involvement of the upper limb in locomotion among Middle Pleistocene members of the genus Homo. Functional interpretations of the Dinaledi upper limb fossils have been, however, hindered by the fragmentary nature of the material. The discovery of more complete upper limb elements from a new chamber in the Rising Star cave system, the Lesedi Chamber, provides new information about the forelimb anatomy of this recently-discovered primitive member of the genus Homo. This chamber is within the same cave system as, and in close geographical proximity to, the H. naledi material described by Berger et al. (2015) from the Dinaledi Chamber. Craniodental comparisons to the Dinaledi Chamber materials (Hawks et al. 2017) referred this new material from Lesedi to H. naledi. The Lesedi Chamber is designated as Site U.W. 102 (hereafter Site 102), and its assemblage comprises 131 hominin specimens from at least 3 individuals across both adult and immature developmental stages. This includes 16 identifiable fragments of claviculae, scapulae, humeri, radii, and ulnae, of which 13 preserve diagnostic anatomy (see Figures 1–5 below; see Tables 1–3 below). The most complete of these elements are attributable to the large adult individual, designated LES 1, from collection area 102a (Hawks et al. 2017). Here we present detailed comparisons of the Lesedi Chamber upper limb material with upper limb material from the Dinaledi Chamber, and with extant and fossil comparative samples. ANATOMICAL DESCRIPTIONS OF DIAGNOSTIC FOSSILS The following descriptions pertain to the diagnostic clavicular, scapular, humeral, radial, and ulnar specimens within the Site 102 assemblage, including their preservation, with reference to comparable specimens from the Site 101/Dinaledi Chamber assemblage described in Feuerriegel et al. (2017) (see Figures 1–5 below; see Tables 1–3 beINTRODUCTION T modern human upper limb is unique among primates in functioning almost exclusively as an organ of manipulation rather than prehension related to locomotion. The commitment of the human forelimb to manipulation is reflected in its musculoskeletal morphology; however, the point in our evolutionary history where the human upper limb went from a structure used predominantly in locomotion to one used predominantly in manipulation is unclear. Primitive upper limb features in certain ancestral species are interpreted by some authors to indicate a continued reliance on the upper limb for climbing and suspension, and interpreted by others as adaptively inconsequential primitive retentions in habitual terrestrial bipeds. New fossil material and new analyses are providing greater resolution on the meaning of evolutionary changes in upper limb morphology in the hominin lineage. Previous work on early members of the genus Homo (fossils dated to approximately 2–1 Ma, including OH 7, OH 48, OH 62, KNM-ER 3735, KNM-WT 15000, and the Dmanisi hominins) has indicated the beginning of a temporal trend within the genus Homo for decreasing involvement of the upper limb in locomotive behaviors and an increasing commitment to manipulation and prehensile movements of the hand (e.g., Jashashvili 2005; Leakey et al. 1989; Lordkipanidze et al. 2007; Napier 1965; Oxnard 1969; Voisin 2008). This Early Pleistocene material is variable in its expression of primitive and derived upper limb morphology, with some traits suggestive of continued use of the upper limb in locomotion and others reflecting an emphasis on manipulation. By the Middle Pleistocene, the complete decoupling of the arm from the constraints of locomotion in Homo is characterized by, among other things, a lowering of scapular position on the thorax from the high position seen in australopiths, a lateral orientation to the scapular glenoid fossa, relatively long claviculae, a proximally-oriented ulnar olecranon process, and changes in manual proportions (Drapeau 2008; Larson 2007, 2013). Precisely when the modern human shoulder and upper limb configuration first arose remains an open question and there are competing models for the trajectory of upper limb (shoulder, elbow, and hand) configuration in the genus Homo (e.g., see Larson 2007; Roach et al. 2013; Roach and Richmond 2015). Nevertheless, it is generally accepted that by at least 800 ka a basically modern human-like upper limb arrangement was in place with the emergence of Homo antecessor (Carretero et al. 1997; Carretero et al. 1999; Lorenzo et al. 1999; Marzke 1997; Susman 1998). While they exhibit some variation away from modern humans in terms of morphological particulars (variations that may reflect performance differences in certain manipulatory behaviors but not capability differences; Niewoehner 2000, 2001), Homo heidelbergensis and Homo neanderthalensis also display a human-like shoulder and upper limb configuration, and total freedom of the upper limb from the constrai