Journal of Human Evolution 81 (2015) 13e28

Contents lists available at ScienceDirect

Journal of Human Evolution

journal homepage: www.elsevier.com/locate/jhevol

The Neandertals of northeastern Iberia: New remains from the Cova del Gegant (Sitges, Barcelona)

* Rolf Quam a, b, c, , Montserrat Sanz d, Joan Daura d, Kate Robson Brown e, Rebeca García-Gonzalez f, Laura Rodríguez f, g, Heidi Dawson e, Rosa Flor Rodríguez d, Sandra Gomez d, Lucía Villaescusa d, Angel Rubio d, h, Almudena Yagüe d, María Cruz Ortega Martínez b, Josep Maria Fullola i,Joao~ Zilhao~ i, j, Juan Luis Arsuaga b, k a Department of Anthropology, Binghamton University (SUNY), Binghamton, NY 13902-6000, USA b Centro UCM-ISCIII de Investigacion sobre la Evolucion y Comportamiento Humanos, Avda. Monforte de Lemos, 5, 28029 Madrid, Spain c Division of Anthropology, American Museum of Natural History, Central Park West at 79th St., New York, NY 10024-5192, USA d Grup de Recerca del Quaternari-Seminari d'Estudis i Recerques Prehistoriques, Dept. Prehistoria, H. Antiga i Arqueologia, Facultat de Geografia i Historia, Universitat de Barcelona, C/Montalegre, 6, 08001 Barcelona, Spain e Department of Archaeology and Anthropology, University of Bristol, 43 Woodland Road, Bristol BS8 1UU, UK f Departamento de Ciencias Historicas y Geografía, Universidad de Burgos, Facultad de Humanidades y Educacion, 09001 Burgos, Spain g Centro Nacional de Investigacion sobre la Evolucion Humana (CENIEH), Paseo Sierra de Atapuerca s/n, 09002 Burgos, Spain h Laboratorio de Antropología, Depto de Medicina Legal, Toxicología y Antropología Física, Facultad de Medicina, Universidad de Granada, Av de Madrid, 11, 18012 Granada, Spain i Seminari d'Estudis i Recerques Prehistoriques, Dept. Prehistoria, H. Antiga i Arqueologia, Facultat de Geografia i Historia, Universitat de Barcelona, C/Montalegre, 6, 08001 Barcelona, Spain j Institucio Catalana de Recerca i Estudis Avançats (ICREA), Passeig Lluís Companys, 23, 08010 Barcelona, Spain k Departamento de Paleontología, Facultad de Ciencias Geologicas, Universidad Complutense de Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain article info abstract

Article history: The present study describes a new juvenile hominin mandible and teeth and a new juvenile humerus Received 21 June 2014 from level V of the GP2 gallery of Cova del Gegant (Spain). The mandible (Gegant-5) preserves a portion Accepted 5 February 2015 of the right mandibular corpus from the M1 distally to the socket for the dc mesially, and the age at Available online 9 March 2015 death is estimated as 4.5e5.0 years. Gegant-5 shows a single mental foramen located under the dm1/ dm2 interdental septum, a relatively posterior placement compared with recent hominins of a similar Keywords: developmental age. The mental foramen in Gegant-5 is also placed within the lower half of the Homo neanderthalensis mandibular corpus, as in the previously described late adolescent/adult mandible (Gegant-1) from this Tooth Mandible same Middle Paleolithic site. The Gegant-5 canine shows pronounced marginal ridges, a distal accessory Humerus ridge, and a pronounced distolingual tubercle. The P3 shows a lingually-displaced protoconid cusp tip Upper Pleistocene and a distal accessory ridge. The P4 shows a slightly asymmetrical crown outline, a continuous trans- Iberian Peninsula verse crest, a mesially placed metaconid cusp tip, a slight distal accessory ridge, and an accessory lingual cusp. The M1 shows a Y5 pattern of cusp contact and a well-developed and deep anterior fovea bounded posteriorly by a continuous midtrigonid crest. Gegant-4 is the distal portion of a left humerus from a juvenile estimated to be between 5 and 7 years old at death. The specimen shows thick cortical bone. Although fragmentary, the constellation of morphological and metric features indicates Neandertal af- finities for these specimens. Their spatial proximity at the site and similar ages at death suggest these remains may represent a single individual. The addition of these new specimens brings the total number of Neandertal remains from the Cova del Gegant to five, and this site documents the clearest evidence for Neandertal fossils associated with Middle Paleolithic stone tools in this region of the Iberian Peninsula. © 2015 Elsevier Ltd. All rights reserved.

* Corresponding author. E-mail address: [email protected] (R. Quam). http://dx.doi.org/10.1016/j.jhevol.2015.02.002 0047-2484/© 2015 Elsevier Ltd. All rights reserved. 14 R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28

Introduction most important karstic systems of NE Iberia, is composed of Jurassic and Cretaceous limestone and dolomite. The site is currently During the last two decades, numerous archaeological sites in accessible both from the sea and through a natural vertical shaft. It the northeast (NE) of the Iberian Peninsula have documented consists of a principal chamber (GP), now eroded by wave action, Neandertal occupations at rock shelters, open-air sites, and caves and its inner part (GP1 and GP2), where a small conduit (GLT) leads during Marine Isotope Stages (MIS) 5e3. Recent research at Abric to the adjacent Cova Llarga. Two galleries branch off of the right Romaní (Vallverdú et al., 2005; Camps and Higham, 2012), Roca side of GP, one more interiorly (GL2) and the other nearer to the sea dels Bous (Martínez Moreno et al., 2010), and Cova Gran de Santa (GL1). Linya (Mora et al., 2011) has contributed to debates on the Middle to At least eight site formation episodes from the Upper Pleisto- Upper Paleolithic transition focusing on radiocarbon dating (Camps cene (Episodes 0e3) to the Holocene (Episodes 4e7) have been and Higham, 2012; Maroto et al., 2012; Vaquero and Carbonell, recognized in the Cova del Gegant stratigraphic sequence, alter- 2012), lithic technology (Thiebaut et al., 2012) and its variability nating between continental sediment deposition and periods of (Mora et al., 2008), and stratigraphic sequences (Martínez Moreno marine erosion followed by the accumulation of beach deposits et al., 2010). However, Neandertal fossils remain scant, especially (Daura et al., 2010). This framework makes it possible to establish when compared to the southern Mediterranean coast (Arsuaga correlations between the deposits that yielded the Neandertal re- et al., 1989, 2001, 2007, 2012; Garralda, 2005; Walker et al., 2008, mains and those located elsewhere in the cave (Table 1). The 2010, 2011a, 2011b, 2012). The relative rarity of hominin remains Gegant-1 mandible and the Gegant-2 incisor come from the same can be explained by (i) the small number of archaeological sites gallery (GL1) and likely derive from the same layer (XV). Thus, they excavated, (ii) site function, (iii) the low intensity of occupations, may have been in close spatial proximity (Daura and Sanz, (iv) behavioral patterns, and/or (v) sampling bias. 2011e2012), and correspond to Episode 2 or 3. The archaeological Hitherto, Neandertal remains from this northeastern area were sequence in this part of the site has been dated to between limited to the Cova de Mollet isolated tooth, the Banyoles mandible, 49.4 ± 1.8 ka and 60.0 ± 3.9 ka (two and one sigma errors, and the Cova del Gegant mandible and isolated tooth. Although the respectively; Daura et al., 2010). The Gegant-2 incisor falls within stratigraphic provenance of the hominin molar from Cova Mollet is this time range, while the Gegant-1 mandible has been directly uncertain (Maroto et al., 1987; Cortada and Maroto, 1990), a recent dated by U-series to 52.3 ± 2.3 ka (2 sigma error). study, based on the degree of fossilization and the sediment The Neandertal remains described here (Gegant-4 and Gegant- adhering to the tooth, places it in Layer 5 from the site, which 5) were recovered in layer V of the GP2 gallery, which contains possibly dates to MIS 7 (ca. 215 ka; Maroto et al., 2012). The most intact deposits in the rear part of the main conduit. This layer accepted date for the Banyoles mandible is ca. 45 ka obtained for overlies the rocky floor of the cave, and it is possible that it once the encasing travertine by Julia and Bischoff (1991), but there is filled most of the principal chamber (GP); it consists of dark ongoing disagreement as to both its true chronological ascription brown to black lutites and sands with faunal remains, a few stone (Grün et al., 2006) and its taxonomic status (Alcazar de Velazco tools, and a large number of coprolites. In the chronostratigraphic et al., 2011). At Cova del Gegant (Fig. 1), the Neandertal fossils framework proposed for Cova del Gegant (Daura et al., 2010), came from sediments that also contained Mousterian stone tools these new hominin remains belong to Episode 1. The Gegant-3 and Pleistocene faunal remains (Daura et al., 2005; Rodríguez et al., central incisor germ also likely comes from this same level and 2011). episode but in GL2. A small sample was removed from the The Cova del Gegant specimens were recovered during a thor- Gegant-4 humerus for DNA analysis and radiocarbon dating at ough revision of the faunal material excavated in the gallery closer Oxford (P-27870), but no results were obtained due to low to the sea (GL1; Fig. 1) in the 1950s (Arxiu Historic Municipal de collagen yield. Nevertheless, Episode 1 at the site has been dated Sitges, AHSI) and in the 1970s (Paleontological Collections, Natural to 55.7 ± 4.8 ka by the Optically Stimulated Luminescence (OSL) History Museum of Barcelona, MGB). These remains correspond to technique (Daura et al., 2010), suggesting that all the hominin an adolescent/adult hominin mandible (Gegant-1; Daura et al., remains from the Cova del Gegant are roughly similar in 2005, 2010; Arsuaga et al., 2011) and an isolated lower lateral chronology. incisor (Gegant-2; Rodríguez et al., 2011; Sanz, 2013). Martínez Although there is clear evidence of hominin activity at the site Moreno (1990) mentions a central incisor germ (Gegant-3) from a in the form of hearths (GP2) (Sanz, 2013), hominin presence was different gallery (GL2; Fig. 1), but this specimen remains unpub- likely short-term and sporadic, as evidenced by the small num- lished. Although fragmentary, all of the hominin fossils were sug- ber of stone tools (GP2, GL1, and GL2) and the low level of an- gested to show Neandertal affinities. Their approximate thropic impact (e.g., cut marks and burning) on the faunal provenience within the site is fairly clear, and considerable effort remains. In addition, no evidence of on-site stone tool production has been made to provide a precise chronological control for the has been documented, and the tool assemblage is entirely associated deposits in order to place these specimens within an composed of finished tools that were transported to and dis- appropriate paleontological context (Daura et al., 2010; Daura and carded at the cave. It is plausible that hominin activity would Sanz, 2011e2012). Here, we present the anatomical description have been primarily concentrated at the entrance, which has and comparative study of two new Neandertal remains discovered been heavily modified by marine erosion. In contrast to the during the 2009 and 2010 field seasons: a partial humerus (Gegant- limited evidence for hominin activity, the presence of hyena re- 4) and a juvenile mandible (Gegant-5) from a different area of the mains, partly digested bones, and coprolites suggest that the site site (GP2; Fig. 1). Both fossils were recovered during the ongoing primarily functioned as a hyena den. The damage and breakage archeological excavation of the site, and their stratigraphic prove- patterns in the faunal remains are in keeping with those nance is known. described in fossil hyena assemblages. Long bones have been turned into cylinders as a result of ravaging, with diaphyses be- The Cova del Gegant site ing more abundant than epiphyses. At Cova del Gegant, carni- vores are plausibly the main agents responsible for the The Cova del Gegant (Sitges, Barcelona) is located on the accumulation of ungulate remains (Mora, 1988; Daura, 2008; seaward edge of the Garraf Massif (Fig. 1), some 40 km south of Daura and Sanz, 2011e2012; Sanz, 2013; Samper Carro and Barcelona (146027.3300E, 4113024.7500N). This massif, one of the Martínez Moreno, 2014). R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28 15

Figure 1. Location of the Cova del Gegant, including panoramic views of Cova del Gegant and other caves from Punta de les Coves in 1928 (1) and in 2012 (2). Site plan of the Cova del Gegant (3) indicating the position of the different galleries discussed in the text and the location of the Neandertal remains. Schematic reconstruction (4) of stratigraphic profiles in different sectors of the cave with radiometric dates and the position of the Neandertal remains indicated.

Materials and methods margin along its length, but few morphological details of the external or internal corpus can be discerned and reliable corpus Preservation and description of the Cova del Gegant fossils dimensions can be recorded only at the level of the mental fora- men. The dm2 and M1 are fully erupted and in situ in their alveolar The Gegant-5 mandibular specimen (Fig. 2) preserves a portion sockets. The latter is well preserved, but the crown of the dm2 is of the right corpus from the M1 distally to the socket for the dc nearly entirely missing, with only a small portion of the enamel mesially. The alveolar margin is better preserved than the basal preserved distally, in contact with the mesial face of the M1. The

Table 1 Human fossils from the Cova del Gegant.

Specimen Field label Season Provenience Age at death Preservation Source

Gegant-1 AHSI 567 1954 Layer XVa (GL1)a Adolescent/Adult Edentulous mandibular corpus Daura et al. (2005) Gegant-2 MGB V2828 1974e1975 Layer XVa (GL1)a 8e10 years Lower lateral incisor Rodríguez et al. (2011) Gegant-3 UAB CG85-GB 1985 Layer V (GL2) Subadult Central incisor (Germ) Martínez-Moreno (1990) Gegant-4 UB CG09-H23-Vf-2311 2009 Layer V (GP2) 5e7 years Fragmentary left distal humerus Present study

Gegant-5 UB CG10-H24-Vf-3131 2010 Layer V (GP2) 4.5e5.0 years Fragmentary mandible and left C1-M1 Present study a Tentatively assigned stratigraphic level. 16 R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28

Comparative morphology

The morphology and metric characteristics of the Gegant-5 mandible and teeth are compared with middle and upper Pleisto- cene members of the Neandertal lineage and Homo sapiens. The mandibular dimensions have been compared with juvenile Nean- dertal and H. sapiens individuals measured by the authors or compiled from the literature. The dental dimensions and morphology are compared mainly with the large sample of teeth from the middle Pleistocene site of the Sima de los Huesos (SH) in the Sierra de Atapuerca in northern Spain, Upper Pleistocene Ne- andertals from Europe and southwest Asia, and recent H. sapiens. Description of the dental morphology generally follows those fea- tures defined by the Arizona State University Dental Anthropology System (ASUDAS; Turner et al., 1991). In addition, a few features not included in the ASUDAS, but defined more recently, are also considered (Bailey, 2002, 2006; Martinon-Torres et al., 2012). The Figure 2. The Gegant-5 mandible with the dm2 and M1 in place in their alveolar sockets. Note the canine still housed in its crypt within the mandibular corpus. Scale main metric dimensions of the crown and roots were recorded to bar ¼ 3 cm. the nearest 0.1 mm using sliding calipers. In addition, the areas of the individual molar cusps and the crown area were measured from scaled occlusal photographs following previously defined tech- remaining teeth (C eP ) are still housed in their crypts within the 1 4 niques (Wood and Abbott, 1983; Wood et al., 1983; Bailey, 2004). mandibular corpus. They were removed for study during cleaning High resolution serial CT scan images were used to create virtual and restoration of the specimen. 2D and 3D representations of the Gegant-4 humerus. The Gegant-4 Gegant-4 is the distal fragment of a left humerus (Fig. 3), with a specimen was CT scanned using a Skyscan 1172 system located at maximum preserved length of 82 mm. Proximally, the fragment the University of Bristol, set at 74 kV and 133 mA with an Al/Cu filter extends well into the diaphysis and includes the nutrient foramen, to provide images with an isometric voxel resolution of 35 mm. which is directed distally. Diaphyseal surfaces are well preserved, Analysis of the resulting reconstructed image files was achieved and the lateral supracondylar ridge is marked. No deltoid tuber- using the software CTAn (Version 1.9.2.5, Skyscan, Kontich, osity is visible, and the distal section of the radial groove is lightly Belgium) and Avizo v.6. Cross-sectional geometry of the diaphysis marked. The distal epiphysis (capitulum, trochlea, and medial and was analyzed at the approximately 35% location (measured from lateral epicondyles) is not present. The distal posterior bone sur- the distal end), in a slice at a right-angle to the bone's long axis. The face is largely intact, and the rounded cross-section of the lateral following values were measured: periosteal perimeter (PP), total border suggests that, at least in this region, only the superficial area (TA), cortical bone area (CA), % cortical area (%CA), and mean bony tissue has been lost. The medial border is quite strongly cortical thickness (CTh). TA was measured by selecting the entire projecting medially, and the epicondyle may have been unfused. slice as the region of interest. CA measurements were repeated The distal anterior bone surface has suffered more damage, three times and the mean calculated to account for any minor possibly caused by scavengers, resulting in most of the coronoid discrepancies when selecting the region of interest. CTh was fossa being lost. calculated as the mean of cortical thickness taken in the medial, lateral, anterior, and posterior sections of the bone. The cross- sectional properties in the Gegant-4 specimen were compared with samples of Pleistocene European fossils and recent hominis to place the specimen within a broader comparative context.

Age at death estimation

Despite some initial controversy, many studies have shown that Neandertals and modern humans differ in some aspects of their dental anatomy that are crucial to determining a dental age at death. Among the most important differences are those regarding the relative dental developmental sequence (Tompkins, 1996a; Bayle et al., 2009), the distribution of perikymata on the crown surface (Guatelli-Steinberg et al., 2007; Guatelli-Steinberg and Reid, 2008), and cuspal enamel thickness (Olejniczak et al., 2008) and extension rate (Smith et al., 2007a). Thus, dental age at death es- timates in Neandertals should generally not be based on modern human standards (Smith et al., 2010). Similarly, there is growing evidence that the tempo and mode of skeletal ontogeny have varied in hominin evolution, which sug- gests that it may not always be possible to extrapolate directly from a modern human reference to a Neandertal (Smith and Tompkins, 1995). The debate as to whether Neandertal maturational sched- ules proceeded faster, slower, or at a similar pace to modern Figure 3. The Gegant-4 humerus in medial (left), anterior (center), posterior (right), humans is ongoing, and it is possible that hominin life history and inferior (lower) views. Scale bar ¼ 2 cm. evolution was, and may continue to be, a modular process (Heim, R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28 17

1982; Tompkins and Trinkaus, 1987; Madre-Dupouy, 1992; Tillier, distance from the dentine horn tip to the position of the first 1999; Leigh and Blomquist, 2007; Cowgill, 2010). formed perikymata. However, it was not possible to observe the Dental developmental sequence To assess the dental develop- first perikymata in the CT scan. For this reason, the position of this mental pattern of the Gegant-5 mandible, high resolution serial CT perikymata was established by visual comparison of the CT image scan images were used to create virtual 3D representations of the and the SEM image. After taking this into account, the enamel individual teeth. The specimen was CT scanned using a YXLON formation time was estimated by two different means, based on the Compact (YXLON International X-Ray GmbH, Hamburg, Germany) thickness. industrial multi-slice computed-tomography (CT) scanner, located First, a minimum cuspal formation time was calculated as cuspal at the Universidad de Burgos in Spain. The mandible was aligned thickness divided by a mean secretion rate in the cuspal region. along the long axis of the right mandible corpus. Scanner energy Given that mean secretion rates do not vary between Neandertals was set at 160 kV and 4 mA and the field of view was 69.9 mm. and modern humans (Dean et al., 2001; Macchiarelli et al., 2006; Individual slices were obtained as a 1024 1024 matrix of 32 bit Smith et al., 2007a), we relied on secretion rates reported in Float format for processing. The final pixel size was 0.051 mm modern humans for anterior and posterior teeth (Schwartz et al., with an inter-slice distance of 0.2 mm. The Mimics™ (Materialise, 2001; Smith et al., 2007b). Second, a maximum cuspal formation Belgium) software program was used to visualize the CT images time was derived relying on a regression equation relating cuspal and make the virtual reconstructions. We have used thickness and formation time (Dean et al., 2001). In both cases, semiautomatic segmentation in order to define Hounsfield values cuspal thickness was multiplied by a factor of 1.15 in order to take for the dentine, enamel, bone, and air. into account prism decussation (Risnes, 1986). To assess the degree of similarity or difference between the Lateral enamel formation time can be calculated as the total dental developmental sequence in the Gegant mandible and that of number of perikymata multiplied by the periodicity. Perikymata modern humans, we have relied on a Bayesian approach (Braga and were imaged with an Environmental Scanning Microscope (JEOL Heuze, 2007), which provides a probability that the development JSM-6460LV) in low vacuum mode, in secondary emissions mode, pattern in the Gegant individual could be found within a modern and an accelerating voltage of 15 kV. Teeth were placed with the human population. The developmental stages of the Gegant per- buccal surface orthogonal to the electron beam inside the micro- manent dentition were scored following the systems developed by scope chamber. Several micrographs were taken at 55 magnifi- Moorrees et al. (1963) and Demirjian et al. (1973), while the dm2 cation from the earliest formed enamel at the cusp to the latest was scored based on the system established by Liversidge and formed enamel at the cervix. To count the perikymata, a photo- Molleson (2004). Although dental development is under tight ge- montage was made from these micrographs with Photoshop CS5™ netic control, some variability in both tooth formation and eruption software. is present (Tompkins, 1996b). In light of this, the Demirjian dental The Gegant periodicity cannot be determined directly, since we developmental sequence obtained for the Gegant-5 permanent do not have sectioned teeth from this individual. Neandertal peri- mandibular teeth was also compared to a pooled sample of modern odicities range from 6 to 9 days, with a mean and modal value of 7 human children of both sexes and diverse geographic and temporal or 8 (Smith et al., 2010). This range is broader in modern humans periods (n ¼ 100). (6e12 days) but the mean and modal value is the same (7e8 days; This pooled sample was drawn from three sources. A subset Dean and Reid, 2001; Reid and Dean, 2006). Moreover, these pe- was chosen from each source based on having at least one tooth riodicities are inversely related with perikymata number (Reid and that showed the same score as that seen in at least one tooth Ferrell, 2006). We addressed the question of periodicity in Gegant- from the Gegant-5 mandible. The first subset was derived from 5 in two ways. First, assuming an equivalent crown formation time the data included in the Electronic Encyclopedia on Maxillo- in both Neandertals and modern humans, the linear regression Facial, Dental and Skeletal Development CD-ROM (Demirjian, formula derived from Reid and Ferrell (2006) for canines was used 1996). These data come from a longitudinal study of Montreal to predict the periodicity in Gegant. Since the periodicity is the French-Canadian children conducted in the 1960s and 1970s. A same in all teeth belonging to the same individual (FitzGerald, subset of girls (n ¼ 40) and boys (n ¼ 40) aged from 6 to 10 years 1998), we use this predicted periodicity for the P3 as well. Sec- was selected. The second subset consists of cross-sectional stan- ond, periodicity in the Gegant individual was estimated relying on dardized orthopantomographs of ten living girls of known chro- the mean, mode, and range of periodicities reported in Neandertals nological age ranging between 3 and 8 years old from Burgos (Smith et al., 2010). Thus, crown formation times in the Gegant (Spain). The last subset is a sample of ten children from a Me- individual are calculated based on a range of periodicities (from 6 to dieval archaeological population excavated from the Dominican 10 days). Monastery of San Pablo (Adan- Alvarez, 2003) that are now Assessment for the age at death of Gegant-5 was carried out housed at the Laboratory of Human Evolution at the Universidad based on incremental dental enamel characteristics. This is an ac- de Burgos. CT image data for these ten individuals were obtained curate, species-specific method that is independent of the modern with similar parameters as those for the Gegant mandible, except human dental pattern (Bromage and Dean, 1986; Beynon et al., for a larger field of view (111.1e187.5 mm) to encompass the 1998; Smith et al., 2006), but it requires estimating some param- entire mandible. Virtual reconstruction of the individual teeth eters (e.g., periodicity and the age of onset of enamel formation), was performed in the same way as in the Gegant mandible. The which may lead to a degree of uncertainty in the assessment of age development of the dm2 was not included in the comparative at death. With the exception of the deciduous teeth and the M1, analysis due to lack of data. which start to form enamel matrix in utero (Schour, 1936; Crown formation time Crown formation time was calculated only FitzGerald, 1998), estimating the (post-natal) enamel initiation for the C1 and P3. The P4 was still housed within the body of the time for the other teeth is necessary to assess the dental age at mandible when the analyses were performed, and the M1 shows death. The initiation time can be established histologically using some slight wear on the mesiobuccal cusp, which makes it difficult the earliest formed M1 cusp as a reference, and we have relied on to measure the cuspal enamel thickness and number of perikymata. published histological data for enamel initiation in the C1 and P3 in Crown formation time in the C1 and P3 was calculated as the time to modern humans and Neandertals (Dean et al., 1993; Reid et al., form both cuspal enamel plus the lateral or imbricational enamel. 1998; Reid and Dean, 2000; Guatelli-Steinberg et al., 2005; Gar- Cuspal enamel thickness can be measured on CT scans as the cía-Gonzalez et al., 2011). 18 R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28

Humerus ontogeny To facilitate comparative analysis, the age at the posterior dentition but a delay in the P3, relative to the in- death of the Gegant-4 humerus was estimated relying on a variety cisors. This relative advance in the formation of the molars was of approaches. While the maximum length and distal width of the subsequently documented in other immature Neandertals, humerus are routinely used to estimate age at death in modern including Chateauneuf^ 2 (Colombo et al., 2013). Although no in- human fetal and juvenile individuals (Maresh, 1970; Fazekas and cisors are preserved in the Gegant-5 individual, the posterior Kosa, 1978; Scheuer and Black, 2000), the incomplete dentition (P4 and M1) is slightly advanced relative to the C1, and preservation of the Gegant-4 specimen means neither of these the dental sequence is matched in the La Quina H18 Neandertal measures can be taken directly. However, it is possible to (Smith et al., 2010). estimate maximum humeral length based on anatomical criteria, Long-period line count of perikymata in the Gegant-5 P3 (98) taking into account features such as the location of the nutrient (Table 3) is closest to the modern South African mean but does also foramen relative to the olecranon fossa, as well as other fall within the range of Neandertal variation. In contrast, the peri- diaphyseal dimensions and morphological features, including the kymata count in the C1 (141; Fig. 5) is closest to the Neandertal absence of the distal epiphysis, the morphology of the lateral mean (159), a sample that generally shows fewer perikymata than border, and the gracility of the deltoid insertion and radial groove. in H. sapiens (Table 3). The perikymata count in the Gegant-5 C1 is To further contribute to the assessment of the age at death and outside the lower limits of the range of variation in both Inuits and morphology of the Gegant-4 specimen, cross-sectional properties Europeans but does fall within the range of variation seen in at the 35% level were compared with a small modern human modern South Africans. Neandertal canines are generally charac- ontogenetic series generated for this study. Eight well-preserved terized by a more uniform spacing of the perikymata over the entire modern human humeri aged between 2 and 10 years were buccal surface of the tooth crown. In contrast, and despite some selected for mCT scanning from the medieval skeletal collection of variation, H. sapiens canines tend to show a “packing” of the peri- the Priory of St. Peter and St. Paul, Taunton, UK (Dawson and Robson kymata closer to the cervicoenamel junction (CEJ) (Guatelli- Brown, 2012). These sub-adult skeletons were aged by dental for- Steinberg et al., 2007; Guatelli-Steinberg and Reid, 2008). The mation and eruption, and epiphyseal fusion. The tooth formation distribution of perikymata counts per decile in the Gegant lower stages of Moorrees et al. (1963), as modified by Smith (1991b),were canine (beginning at the incisal edge for the first decile and ending used along with the dental eruption chart of Schour and Massler at the cervical edge for the last decile) is: (1944). Epiphyseal fusion and ossification age stages were ob- 11e11e9e8e10e16e20e21e20e15. Based on this distribution, tained from Scheuer and Black (2000). A comparative data set of 65.2% of perikymata are located in the cervical half of the tooth. bone parameters was generated following the same protocol as This value falls well within the range of variation of modern human employed for the fossil specimen. samples and is only slightly higher than that shown by Neandertals. Therefore, although the total number of perikymata of the Gegant- The Cova del Gegant-5 mandible 5 canine is close to the Neandertal mean, their distribution on the tooth surface more closely matches that in modern human Age at death and sex estimation samples. Taking into account cuspal enamel formation time and a range e The M1 shows only very slight wear resulting in a flattening of of values (6 10 days) for the periodicity (Table 4), crown formation the protoconid cusp tip but no dentine exposure. This indicates times for the Gegant teeth range from 3.0 to 4.6 years (C1)to the tooth had been in functional occlusion for a short period of 2.5e3.5 years (P3). The maximum estimate in Gegant is very close time. The dm2 is missing most of the enamel cap, and wear cannot to the mean in the Medieval Danish sample for the C1 and falls be assessed on this tooth. The canine and premolars are still between the South African and European samples for the P3 unerupted and housed within the body of the mandible (Fig. 4). (Table 5), while the minimum estimate falls below the range of The formation stages of the permanent teeth (Table 2) suggest an variation in all the comparative samples. Nevertheless, a 6-day age at death for the Gegant individual between 5.5 and 6.8 years, periodicity has been reported in a few Neandertals (Smith et al., based on modern human standards (Anderson et al., 1976; 2010) and also represents the lower end of the modern human Liversidge et al., 2006). However, the Demirjian developmental range of variation (Reid and Dean, 2006). Relying on the mean and sequence displayed by the Gegant permanent teeth (DeEeDeF) is modal periodicity (8 days) in modern humans yields crown for- not represented by any individual (0 out of 100) within our mation times of 3.8 and 3.0 years for the C1 and P3, respectively. modern reference sample (p ¼ 0.00). The greatest discrepancy in When the mean periodicity (7 days) of Neandertals is used, the the Gegant developmental sequence appears to be a relatively estimated crown formation times are 3.4 and 2.7 years for the C1 advanced stage of mineralization in the P3. Tompkins (1996b) has and P3, respectively. In the latter case, the closest fit to both of the noted that Neandertals show an advance in the mineralization of Gegant teeth is shown by Neandertals.

Figure 4. Virtual reconstruction of Gegant-5 (a) showing the placement of the P3 and P4 within the mandibular corpus (b) and the internal morphology of the dentition (c). Scale bar ¼ 2 cm. R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28 19

Table 2 Formation stages and metric dimensions of the Gegant-5 teeth.a

Formation stage MD (mm) BL (mm) CI MCA (mm2) Crown height (mm) Total preserved crown/root height (mm) Tooth Moorrees Liversidge/Molleson Demirjian

C1 Crc-Ri D 8.0 9.1 113.8 51.7 11.4 12.0 P3 Ri E 7.9 8.4 106.3 54.5 8.5 10.1 P4 Crc D 7.8 8.6 110.3 57.3 b b M1 R1/2-R3/4 F 11.4 10.4 91.2 97.0 7.3 18.2 dm2 H2 e 10.0 est. a Crown Index (CI) ¼ (BL 100)/MD. Measured Crown Area (MCA) following Wood and Abbott (1983). b Corrected for slight wear.

Table 3 dimorphic tooth in the dental arcade (Garn et al., 1977; Bermúdez Perikymata counts in the Gegant-5 teeth compared with Neandertals and recent de Castro et al., 2001). The overall size of the Gegant-5 canine is H. sapiens.a modest and does not provide a clear indication as to sex. The b Specimen/sample C1 P3 presence of a distal accessory ridge would be more consistent with mean ± s.d. mean ± s.d. a male classification. This is one of the most sexually dimorphic range (n) range (n) features of the dentition, appearing more often in males than fe- Gegant-5 141 98 males across ten modern human samples (Scott, 1977). Neverthe- Neandertals 159 ± 20 109 ± 16 e e less, this provides, at most, tentative support for a male 135 198 (10) 88 130 (5) fi Inuit 198 ± 16 classi cation of Gegant-5. 172e215 (10) Newcastle 199 ± 22 137 ± 17 Mandibular morphology (Fig. 2) 164e249 (13) 109e182 (19) South African 163 ± 21 101 ± 12 125e203 (24) 83e125 (16) A single mental foramen is present at the level of the dm1/dm2 and is located in the lower half of the corpus. Corpus height a Comparative data from Guatelli-Steinberg and Reid (2008). b Measured on the buccal cusp. (22.6 mm) and thickness (12.8 mm) can be measured at the mental foramen (Table 7). Comparison with fossil and recent juvenile mandibles shows Neandertals to be generally characterized by Table 6 shows the age at death estimates in the Gegant-5 indi- taller and thicker mandibles for their developmental age (Fig. 6), vidual based on the C1 and P3. Age at death is estimated using and the Gegant-5 specimen generally falls above the fossil and modern human values for the enamel initiation time, since no data recent H. sapiens specimens of similar developmental age. Never- are available for Neandertal lower dentitions. Relying on a 7 or 8 theless, the resulting robusticity index (56.6) is modest. There is a day periodicity, the age at death in the Gegant individual ranges tendency for the robusticity index to decline with developmental from 3.8 to 5.4 years (C1) and from 4.4 to 4.8 years (P3). Thus, a final age (Table 7), and in this respect Gegant-5 surpasses only Combe age at death estimate of approximately 4.5e5.0 years seems Grenal 1 (49.6), which is slightly older. It is also only slightly higher reasonable. than the late adolescent/adult Gegant-1 mandible (49.4 est.). Sex determination in Gegant-5 is complicated by its fragmen- In Gegant-5, the mental foramen is located below the dm1/dm2 tary state of preservation and juvenile status. Sex-related features interdental septum. Recent humans of a similar developmental age are largely restricted to the canine, since this is the most sexually as Gegant-5 mainly show the mental foramen placed under the dm1 (64.3%), although more than a quarter of individuals (26.8%) show a similar placement under the dm1/dm2 as in Gegant-5 (Coquegniot and Minugh-Purvis, 2003). In contrast, 37.5% of Ne- andertals show a mental foramen placed under the dm1/dm2, while an additional 12.5% show a placement under the dm2. Thus, although there is some overlap between samples, Neandertals do generally show a more posteriorly placed mental foramen with respect to the tooth row, even at the young age of Gegant-5. In Gegant-5, the mental foramen is located 8.4 mm above the basal margin. The vertical placement of the mental foramen within the mandibular corpus can be assessed by comparing the distance from the basal margin with the corpus height at the level of the mental foramen. Neandertals are generally characterized by a mental foramen that is placed closer to the basal margin, within the lower half of the mandibular corpus (Daura et al., 2005). The resulting index in Gegant-5 (37.2) falls between that calculated for the Cova Negra CN7755 mandible (41.5) and the late adolescent/ adult Gegant-1 mandible (35.8), indicating a low placement for the mental foramen.

Second deciduous molar (Fig. 2)

Figure 5. SEM montage of the Gegant-5 C1 (a) and P3 (b) showing the perikymata. The two teeth were imaged separately at a magnification of 25 (C1) and 19 (P3) in lower The Gegant-5 dm2 is in situ in the mandible but is missing nearly vacuum mode and with an accelerating voltage of 20 KV. Scale bars ¼ 1 mm. the entire tooth crown, with only a small portion of enamel 20 R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28

Table 4 Crown formation time estimation in the Gegant-5 teeth.a

Tooth Cuspal thickness Minimum cuspal Maximum cuspal Number of CFT (6 days) CFT (7 days) CFT (8 days) CFT (9 days) CFT (10 days) (mm) formation time (days)b formation time (days)c perikymata

C1 750 227 290.6 141 1104.8 1245.8 1386.8 1527.8 1668.8 P3 950 265.8 353.2 98 897.5 995.5 1093.5 1191.5 1289.5 a CFT (Crown Formation Time) ¼ [(Maximum formation time þ minimum formation time)/2] þ (No. of perikymata periodicity). CFT estimates assume a periodicity ranging from 6 to 10 days. b Following Smith et al. (2007c). c Following Dean et al. (2001).

Table 5 tentatively estimated at c. 10 mm based on the preserved enamel Crown formation time (days) in Neandertals and recent H. sapiens. portion and the complete dentine surface. This is nearly identical to Tooth Neandertals South African Newcastle Medieval Danish the MD dimension in the Cova Negra CN7755 individual (10.1 mm; Mean ± s.d. (n) Mean ± s.d. (n) Mean ± s.d. (n) Mean ± s.d. (n) Arsuaga et al., 1989), and falls within one standard deviation below a b b b ± ¼ C1 1315 ± 15 (2) 1494 ± 56 (25) 1866 ± 73 (13) 1670 ± 60 (67) the mean in Neandertals (10.5 0.6; n 30; Mallegni and Trinkaus, a c c P3 989 ± 7 (2) 1122 ± 42 (16) 1437 ± 68 (19) 1997). It is also close to the mean values in recent human males ± ¼ ± ¼ a Calculated from Smith et al. (2007c, 2010). (9.9 0.5; n 69) and females (9.7 0.5; n 64; Mallegni and b Guatelli-Steinberg et al. (2005). Trinkaus, 1997). c Reid et al. (1998).

Canine (Fig. 7) preserved along the distal face of the tooth. Thus, nothing can be said about its morphological details. The pulp chamber does not The right canine crown is unworn and well preserved. The appear expanded in CT imaging and the specimen does not show buccal face shows a pronounced hypoplastic line located 9.6 mm taurodontism. The MD dimension of the crown (Table 2)is from the cusp tip. This would correspond to 2.3e4.3 years of age

Table 6 a Age at death estimate for the Cova del Gegant-5 mandible based on C1 and P3 formation times.

Tooth Initiation Age at death Source time (days) 6 day periodicity 7 day periodicity 8 day periodicity 9 day periodicity 10 day periodicity (years) (years) (years) (years) (years)

C1 123 3.4 3.8 4.1 4.5 4.9 Antoine (2001) 126 3.4 3.8 4.1 4.5 4.9 García-Gonzalez et al. (2011) 138 3.4 3.8 4.2 4.6 5.0 Dean et al. (1993) 146 3.4 3.8 4.2 4.6 5.0 Antoine (2001) 201 3.6 4.0 4.4 4.7 5.1 Reid et al. (1998) 570 4.6 5.0 5.4 5.7 6.1 Antoine (2001)

P3 609 4.1 4.4 4.7 4.9 5.2 Dean et al. (1993) 675 4.3 4.6 4.8 5.1 5.4 Reid et al. (1998)

a Based on a periodicity ranging from 6 to 10 days and taking into account different ages for enamel initiation.

Table 7 Main metric dimensions of the Gegant-5 mandible.

Specimen/Sample Age at death (yrs.) Corpus height at Corpus thickness at Robusticity Index at Source mental foramen (mm) mental foramen (mm) mental foramena

Gegant-5 4.5e5.0 22.6 12.8 56.6 Present study

Neandertals Palomas 49 c. 2.0 19.2 11.4 59.4 Walker et al. (2010) Barakai c. 3.0 20.1 14.2 70.6 Mallegni and Trinkaus (1997) Archi 1 c. 3.0 20.0 12.0 60.0 Mallegni and Trinkaus (1997) Roc de Marsal c. 3.0 17.0 12.7 74.7 Madre-Dupouy (1992) Il Molare 1 c. 3.5 20.9 12.2 58.4 Mallegni and Trinkaus (1997) Palomas 7 c. 4.0 21.3 12.7 59.6 Walker et al. (2010) La Chaise 13 c. 4.0 20.5 12.5 61.0 Mallegni and Trinkaus (1997) Devil's Tower c. 4.0 22.8 13.6 59.6 Mallegni and Trinkaus (1997) Cova Negra (CN 7755) c. 5.0 20.0 13.3 66.5 Arsuaga et al. (1989) Combe Grenal 1 c. 7.0 27.4 13.6 49.6 Garralda and Vandermeersch (2000)

Modern Humans Le Figuier c. 3.0 18.0 10.7 59.4 Billy (1979) La Madeleine 4 c. 3.0 19.0 9.4 49.5 Heim (1991) Lagar Velho c. 4.5 20.5 11.5 56.1 Trinkaus (2002) Skhul 1 c. 4.5 16.4 11.0 67.1 Mallegni and Trinkaus (1997) Qafzeh 4 c. 6.0 26.3 14.2 54.0 Original specimen Qafzeh 10 c. 6.0 24.2 13.3 55.0 Original specimen Recent children (n ¼ 20) 2.0e5.0 17.2 ± 1.8 10.3 ± 1.0 60.4 ± 6.9 Madre-Dupouy (1992)

a Calculated as (corpus thickness/corpus height) 100. R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28 21

and is perhaps related to the age at weaning. The crown shows a high, pointed, centrally placed cusp tip, but the crown is slightly asymmetrical, with the mesial shoulder being higher than the distal. The mesial and distal marginal ridges are well developed, and a clear distal accessory ridge (ASUDAS Grade 4) is present on the lingual face of the crown. The lingual face shows a marked expression of several morphological features. Although the mesial marginal ridge is more pronounced along its length, the distal marginal ridge shows a very prominent bulge approximately midway down from the tip. This feature is continuous with the distal marginal ridge but in occlusal view is separated from the lingual face of the tooth by a clear furrow. Just mesial to this feature, at the base of the crown, there is a very small lingual tubercle with a free apex. The pattern of crown asymmetry and development of the mesial and distal marginal ridges in the Gegant-5 canine is similar to that reported to characterize both the Atapuerca SH teeth (Martinon- Torres et al., 2012) and the Neandertals (Bailey, 2006). In contrast, the distal accessory ridge is found in only 25% of the Atapuerca SH lower canines (Martinon-Torres et al., 2012) but is much more common (approximately 85%) in Neandertals (Bailey, 2006). The expression of a bulge in the distal marginal ridge is considerably weaker in the Atapuerca SH sample and Neandertals compared with the Gegant-5 canine, which shows an exaggerated expression of this feature. Several of the Atapuerca SH specimens show a slight bulge along the lower portion of the distal marginal ridge, but it never reaches the expression seen in the Gegant-5 canine (Martinon-Torres et al., 2012). Similarly, a weak bulge is present in the Neandertal lower canine from Arcy-sur-Cure (Bailey and Hublin, 2006). Metrically, the MD (8.0 mm) and BL (9.1 mm) dimensions of the Gegant-5 canine fall within one standard deviation above the Atapuerca SH and Neandertal means, while a greater difference is Figure 6. Scatterplots of developmental age versus corpus height (top) and thickness noted compared with recent H. sapiens (Table 8). The measured (bottom) at the level of the mental foramen. Neandertals, and the Gegant-5 specimen, crown area (MCA) in the Gegant-5 canine (51.7 mm2) is also generally show a taller and thicker mandibular corpus for their developmental age modest, falling only slightly above the mean in the Atapuerca SH than do H. sapiens. sample (mean ± s.d. ¼ 49.7 ± 7.4 mm2; range ¼ 37.1e65.7 mm2; n ¼ 14) (Bermúdez de Castro et al., 2001). Data for Neandertals are scarce, but the Gegant-5 canine is smaller than that of Amud 1 (58.4 mm2) and Petit-Puymoyen 3 (52.9 mm2). It is, however, outside the range of variation in a pooled-sex contemporary H. sapiens sample (mean ± s.d. ¼ 37.3 ± 4.8 mm2; range ¼ 25.9e49.3 mm2; n ¼ 216; Bermúdez de Castro et al., 2001).

Third premolar (Fig. 8)

The unworn P3 crown is well preserved, and the marked hy- poplasias present on the canine are not visible on the buccal face of the P3. In occlusal view, this tooth is only slightly asymmetrical (Grade 1; Bailey, 2006), with the distal portion of the lingual con- tour being convex and the mesial portion being weakly concave. The dominant buccal cusp is joined to a much smaller lingual cusp by a continuous transverse crest. The buccal cusp tip is displaced lingually and is located in the center of the tooth crown, and the buccal face of this cusp makes up a large portion of the tooth in occlusal view. Neither mesial nor distal buccal grooves are present on the buccal face. However, a pronounced distal accessory crest (Grade 2; Bailey, 2006) is present just distal to the tip of the pro- toconid, running vertically from the occlusal edge into the talonid basin distal to the transverse crest. There is a single small lingual cusp located in a medial position, opposite the buccal cusp. No accessory lingual cusps are present, but a mesial lingual groove is observed. Figure 7. The Gegant-5 lower canine. Note the well-developed distal marginal ridge The slight asymmetry and central placement of the dominant and the clear distal accessory ridge. Scale bar ¼ 1 cm. buccal cusp tip seen in the Gegant-5 P3 also generally characterizes 22 R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28 Source on-Torres et al. (2012) Bermúdez de Castro (1993) Martin Bermúdez de Castro (1993) ) n s.d. ( 0.5 (402) 0.5 (38) 0.6 (34) ± ± ± ± BL (mm) Mean 1 M ) n s.d. ( 0.7 (402) 10.7 0.5 (39)0.7 (33) 10.5 10.8 ± ± ± ± MD (mm) Mean ) n s.d. ( ± 0.5 (213) 11.1 0.6 (35)0.8 (32) 11.3 11.5 Figure 8. The Gegant-5 P3. Note the well-developed distal accessory crest. Scale ± ± ± BL (mm) bar ¼ 1 cm. Mean 4 P

) the Atapuerca SH sample as well as Neandertals (Gomez-Robles n et al., 2008; Martinon-Torres et al., 2012). A distal accessory s.d. ( crest is found in approximately 50% of the Atapuerca SH speci- ± 0.4 (25)0.5 (29) 8.6 8.9 0.5 (213) 8.3 ± ± ± mens and is nearly ubiquitous (90%) in Neandertals but is also MD (mm)

Mean found in high frequencies in fossil H. sapiens (Bailey, 2006; Martinon-Torres et al., 2012). The lingual cusp in the Gegant-5

) P is generally smaller than in the Atapuerca SH sample but the n 3 absence of accessory lingual cusps is a point of similarity. Nearly s.d. ( half of the Atapuerca SH sample (52.9%) shows a single lingual ± 0.5 (32)0.7 (29) 7.2 7.5 0.5 (264) 7.0 ± ± ± BL (mm) cusp, although accessory cusps do occur at a higher frequency Mean among Neandertals (Martinon-Torres et al., 2012). Similarly, a 3

P mesial lingual groove is present in nearly two-thirds of Neandertal )

n P3s(Bailey, 2006). The MD dimension (7.9 mm) in the Gegant-5 P3 is very similar s.d. (

± to both the Atapuerca SH and Neandertal means (Table 8), while 0.4 (33)0.5 (28) 9.0 8.8 0.5 (264) 7.8 ± ± ± the BL dimension (8.4 mm) is somewhat smaller, perhaps a result MD (mm)

Mean of the relatively small lingual cusp. The MCA in Gegant-5 (54.5 mm2) is close to the Atapuerca SH and Neandertal means )

n but outside the range of variation in recent H. sapiens (Table 9). s.d. ( 0.7 (29)0.8 (20) 7.9 7.7 0.6 (87) 6.8 ± Fourth premolar (Fig. 9) ± ± ± BL (mm)

Mean The unworn P4 crown is well preserved. In occlusal view, the 1 crown is slightly asymmetrical (Grade 1; Bailey, 2006) due to the )

n talonid development in the distal portion. The main buccal and lingual cusps are joined by a continuous transverse crest (Grade 2; s.d. ( 0.4 (29)0.5 (20) 8.7 8.8 0.4 (87) 7.6 ± Bailey, 2002) that delimits an anterior fovea mesially and a larger ± ± ±

MD (mm) and deeper talonid basin distally. Along the buccal margin, just 6.6 Mean distal to the main buccal cusp tip, a slight raised crest extends into the central basin, corresponding to the distal accessory ridge (Grade 1; Bailey, 2006). Distal to this is a small cuspule without any crest-like extension. No mesial accessory ridge is present. The lingual margin shows a dominant metaconid located mesial to the H. sapiens protoconid tip. No mesial lingual groove is present, but an accessory cusp is present distal to the metaconid (ASUDAS Grade Specimen/Sample C Gegant-5Atapuerca (SH)Neandertals 7.6 8.0 7.8 9.1 7.9 8.4 7.8 8.6 11.4 10.4 Present Study Recent e Table 8 Dimensions of the Gegant-5 teeth compared with Pleistocene and recent humans. 2 3). R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28 23

Table 9 2 Measured crown areas (mm ) in the Gegant-5 P3 and P4 and some comparative samples.

2 2 Specimen MCA P3 (mm ) MCA P4 (mm ) Ratio Source MCA P3/P4 Gegant-5 54.5 57.3 95.1 Present study Atapuerca (SH) mean ± s.d. 53.1 ± 6.6 50.6 ± 7.3 105.4 ± 5.5 Bermudez de Castro et al. (2001) Atapuerca (SH) range (n) 40.5e64.6 (14) 35.5-63.4 (14) 97.3e114.7 (14) Neandertal mean ± s.d. 53.2 ± 6.4 56.9 ± 10.3 104.0 ± 14.0 Hershkovitz et al. (2011) Neandertal range (n) 43.5e64.3 (11) 43.2e67.7 (8) 95.7e128.8 (5) Recent humans (pooled sex) mean ± s.d. 38.4 ± 4.3 44.0 ± 5.2 87.3a Bermudez de Castro et al. (2001) Recent humans (pooled sex) range (n) 27.9e50.4 (224) 31.4e57.4 (204)

a Calculated based on mean values.

Although the phylogenetic polarity of some of these traits is a the distal root, which was still in the process of formation, is matter of debate (Bailey and Lynch, 2005; Martinon-Torres et al., exposed. The occlusal surface shows five main cusps and neither a 2006), the asymmetrical crown, continuous transverse crest, C6 nor C7 are present. The metaconid and hypoconid clearly make distal accessory crest, mesial placement of the metaconid tip, and contact in the central basin, and the Gegant-5 M1 thus shows a Y5 accessory lingual cusps (all seen in the Gegant-5 P4) occur at high pattern of cusp contact. Anteriorly, the specimen shows a deep, frequencies as individual features and in combination in both the wide anterior fovea delimited posteriorly by a continuous middle Atapuerca SH sample and in Neandertals (Bailey, 2006; Martinon- trigonid crest (Grade 2; Bailey, 2002) separating this feature from Torres et al., 2012). Thus, the morphology of the Gegant-5 P4 the central basin. Neither a distal trigonid crest nor a deflecting clearly aligns it with the Neandertal evolutionary lineage. wrinkle is present. The mesial marginal ridge does not show any The MD dimension of the Gegant-5 tooth falls slightly above the accessory tubercles, and there is no expression of a protostylid. The mean values for both the Atapuerca SH and Neandertal samples, pulp chamber is clearly expanded in the Gegant-5 M1 (Fig. 4c), and while the BL dimension is at or below the mean values in these this tooth would be classified as mesotaurodont (Index ¼ 1.35) same samples (Table 8). In contrast, both dimensions fall above the according to Shaw (1928). recent human means. The MCA in the Gegant-5 P4 is also modest, Many of the morphological features expressed in the Gegant-5 falling within one standard deviation above the Atapuerca SH mean M1 either occur in similar frequencies in Neandertals and modern and close to the mean value in Neandertals (Table 9). Comparison of humans or do not vary along taxonomic lines. Nevertheless, the the size of the P3 versus the P4 reveals that the Gegant-5 specimen presence of a continuous midtrigonid crest, like in the Gegant-5 M1, does not show the derived relative expansion of the P3 that char- is nearly ubiquitous in Neandertals (Bailey, 2002, 2006) and is also acterizes the Atapuerca SH and Neandertal samples. found at a high frequency in the Atapuerca SH sample (Martinon- Torres et al., 2012). In contrast, this feature is rare or absent in First molar (Fig. 10) fossil and recent H. sapiens (Bailey, 2006). Taurodontism of the pulp chamber has been reported to occur at high frequencies in Nean- The crown of the first molar is well preserved and shows only dertal dentitions (Kallay, 1963), and this feature is variably present the slightest trace of wear on the protoconid cusp tip. Posteriorly, in recent human populations (Shaw, 1928). The MD and BL dimensions in the Gegant-5 tooth are very close to the mean values for the Atapuerca SH and Neandertal samples, and not far removed from recent H. sapiens (Table 8). Comparative data for the MCA and individual cusp areas are scarce, but the MCA in Gegant (97.0 mm2) is similar to the Atapuerca SH mean value (98.7 ± 8.9 mm2; n ¼ 22; Bermúdez de Castro et al., 2001)aswellas Bolomor HCB-02 (Table 10). However, the Gegant-5 M1 is small

Figure 9. The Gegant-5 P4. Note the slight crown asymmetry, continuous transverse Figure 10. The Gegant-5 M1 in occlusal (left) and distal (right) views. Note the well- crest, mesially placed metaconid tip, slight distal accessory crest, and lingual accessory defined anterior fovea bounded posteriorly by a continuous midtrigonid crest and tubercle. Scale bar ¼ 1 cm. the incomplete formation of the root. Scale bar ¼ 1 cm. 24 R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28

Table 10

Measured crown area and relative cusp areas in the Gegant-5 M1 and some comparative specimens and samples.

Specimen MCA Relative Relative Relative Relative Relative Source protoconid metaconid hypoconid entoconid hypoconulid

Gegant-5 97.0 23.3 21.7 21.4 18.8 14.8 Present study Valdegoba 1 108.2 26.2 23.8 18.0 18.1 14.0 Original specimen Bolomor HCB-02 97.5 26.0 24.2 20.1 19.6 10.0 Arsuaga et al. (2012)

Middle Pleistocene mean ± s.d. (n ¼ 6) 117.6 ± 15.9 26.4 ± 2.9 21.8 ± 2.0 20.0 ± 1.7 19.2 ± 2.5 12.6 ± 1.3 Trefný (2005) Krapina mean ± s.d. (n ¼ 6) 114.1 ± 15.5 26.2 ± 1.0 21.0 ± 1.7 20.3 ± 1.2 20.8 ± 1.0 11.6 ± 1.3 Trefný (2005) Neandertals mean ± s.d. (n ¼ 6) 108.8 ± 8.4 26.8 ± 2.0 21.4 ± 1.2 19.6 ± 1.7 19.7 ± 2.3 12.5 ± 1.2 Trefný (2005) Living humans (pooled sex) mean ± s.d. (n ¼ 33) 94.4 ± 7.5 26.1 ± 1.5 21.3 ± 1.5 20.0 ± 1.1 21.4 ± 1.1 11.1 ± 1.9 Trefný (2005)

relative to other (non-SH) middle Pleistocene humans as well as Lagar Velho (53%) but more similar to Skhul I (67%; Trinkaus et al., Neandertals, including the specimen from Valdegoba. The sizes of 2002) and lower than the Lower Pleistocene juvenile individual the individual cusps, from largest to smallest, conform to the from the Gran Dolina ATD6-121 (82.7%; Bermúdez de Castro et al., following sequence: protoconid > metaconid ¼ 2012). The value in Gegant-4 is lower than in adult Lower and hypoconid > entoconid > hypoconulid. Comparing the relative sizes Middle Pleistocene fossils and Neandertals but more similar to of the individual cusps in Gegant-5 with other fossil samples adult Upper Pleistocene H. sapiens individuals (Table 12). Adult (Table 10) reveals a smaller protoconid and larger hypoconulid in Neandertal long bones are characterized by a pronounced hyper- the Gegant-5 tooth. Gegant-5 also differs from two other Nean- trophy of the cortical bone (Trinkaus, 1983), and this feature is dertal specimens from the Iberian Peninsula (Valdegoba 1 and known to appear relatively early in ontogeny (Ruff et al., 1994; Bolomor HCB-02) in showing a smaller protoconid and metaconid. Cowgill, 2010). For example, in the Dederiyeh 1 Neandertal infant, whose age at death has been estimated at around 2 years based on The Cova del Gegant-4 humerus dental formation, the cortical bone thickness is similar to the mean of modern human children aged 5e6 years (Sawada et al., 2004). It Age at death is not surprising, then, that the cortical thickness in the Gegant-4 humerus appears elevated for its age, compared with recent Based on the preserved portion of the shaft as well as the humans. presence of some anatomical structures (e.g., nutrient foramen, superior margin of the olecranon fossa, etc.), a humeral diaphyseal Discussion length of 175e180 mm is estimated. This corresponds to an age at death of 5e5.5 years according to modern human standards The combination of morphological and metric features in the (Scheuer and Black, 2000). However, the absence of the distal Gegant-5 individual clearly indicates Neandertal affinities for the epiphysis, morphology of the lateral border, and gracility of the specimen. Although little can be said about the bony morphology of deltoid insertion and radial groove suggest an age at death of the mandible, the mental foramen is placed posteriorly with around 7 years using standards for modern humans (Scheuer and respect to the tooth row and is located low in the mandibular Black, 2000). Taken together, an age at death of approximately corpus. In both of these features, it is similar to the late adolescent/ 5e7 years is suggested. However, the robust diaphysis, triangular adult Neandertal mandible from the same site and to adult Nean- cross-section, and thick cortical bone seen in the Gegant-4 hu- dertal mandibles more generally. The dentition shows a suite of merus (Figs. 3 and 11) are not features which would normally be morphological features that are most consistent with a Neandertal associated with a similarly aged modern human child. classification. Among these is the presence of a distal accessory crest in the C1,P3, and P4. In addition, the asymmetry, continuous Comparative morphology transverse crest, mesial placement of the lingual cusp, and presence of an accessory lingual cusp in the P4 are nearly ubiquitous features Gegant-4 has a percent cortical area of 69.3% at 35% of shaft in Neandertals as well as their middle Pleistocene precursors from length, a higher value than any of the modern human specimens in Atapuerca (SH). Finally, the presence of a midtrigonid crest and our ontogenetic series (Table 11). This value is also higher than that taurodont pulp chamber in the M1 are also most consistent with a reported for the Upper Paleolithic modern human specimen from Neandertal classification. Indeed, in most morphological details the Atapuerca (SH) population closely resembles Neandertals (Martinon-Torres et al., 2012), and the dental similarity with the Gegant-5 teeth indicates this individual is a member of the Nean- dertal evolutionary lineage. Metrically, all of the Gegant-5 teeth are of modest size, falling near the Atapuerca (SH) and Neandertal means in most di- mensions. The Atapuerca SH sample and Neandertals have been argued to show an increase in the size of the P3 relative to that of the P4. This relative increase in the size of the P3 in the Neandertal evolutionary lineage has been linked to the expansion of the anterior dentition more generally (Bermúdez de Castro, 1993; Bermudez de Castro and Nicolas, 1996), and Neandertal lineage specimens tend to show a P3:P4 size ratio >100. The Gegant-5 in- dividual does not show this derived Neandertal feature, with a Figure 11. Cortical thickness in three recent H. sapiens juvenile humeri (a) compared predominance of the P4. Interestingly, a similar pattern is seen in with the Gegant-4 specimen (b). All specimens are to the same scale. the Valdegoba 1 Neandertal, also from the Iberian Peninsula. R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28 25

Table 11 Cross sectional parameters of the Gegant-4 humerus at 35% of total bone length compared with an ontogenetic series of modern humans.a

Specimen Age at AeP diameter MeL diameter Mean cortical Periosteal Total area Cortical area (mm2) % Cortical area (%) death (years) (mm) (mm) thickness (mm) perimeter (mm) (mm2)

Gegant-4 5e7 13.0 15.3 3.3 48.6 142.3 98.7 69.3 SK303 2 9.5 10.1 1.1 39.4 101.2 40.6 40.2 SK1806 3 9.8 10.4 1.5 38.9 101.8 41.2 40.4 SK3050 4 10.4 11.8 1.9 40.2 107.3 54.0 50.3 SK2020 6 11.3 11.9 2.1 42.3 115.6 64.2 55.5 SK1205 7 12.3 12.8 2.2 43.4 124.5 69.0 55.4 SK1206 8 11.7 13.3 2.3 44.1 127.6 66.1 51.8 SK2080 9 11.9 13.4 2.2 44.3 128.4 62.2 48.3 SK2023 10 13.7 16.2 2.3 52.2 173.2 85.2 49.2

a Modern human specimens from the medieval archaeological collection of the Priory of St Peter and St Paul, Taunton, UK. Mean cortical thickness is the average of cortical thickness on the medial, lateral, anterior, and posterior margins.

Table 12 Cortical thickness in the Gegant-4 humerus at 35% of total bone length compared with Pleistocene and recent humans.a

Specimen/sample Age at death (years) Total area (mm2) Cortical area (mm2) % Cortical area (%)

Gegant-4 5e7 142.3 98.7 69.3 ATD6-148 Adult 232.7 203.4 90.0 Atapuerca-Sima de los Huesos (n ¼ 5) Adult 318.82 ± 35.98 254.96 ± 45.43 79.5 ± 8.0 Neandertals (n ¼ 10) Adult 300.16 ± 52.72 240.70 ± 45.43 80.0 ± 5.0 Early Upper Paleolithic H. sapiens Adult 269.5 ± 36.1 200.9 ± 33.5 74.3 ± 4.8 Late Upper Paleolithic H. sapiens Adult 261.8 ± 36.9 189.6 ± 28.6 72.6 ± 8.2

a Comparative data from Bermúdez de Castro et al. (2012).

Changes in the relative sizes of the individual tooth cusps have studies suggesting that the thick cortical bone characteristic of adult recently been shown to be a useful taxonomic tool in the M1 for Neandertal humeri is also present at an early ontogenetic age (Ruff Pleistocene Homo (Quam et al., 2009), and Neandertals have been et al., 1994; Sawada et al., 2004; Arsuaga et al., 2007; Cowgill, 2010). characterized as derived in their M1 cusp proportions (Bailey, Whether this is caused by mechanical loading, genetic control, or 2004). In contrast, few if any differences are apparent between metabolic differences between populations is currently unclear. Neandertals and modern humans in the M1 cusp proportions Some authors have suggested that increased cortical robusticity in (Trefný, 2005). Two Neandertal specimens from the Iberian adult Neandertals and archaic H. sapiens, relative to recent modern Peninsula do differ from other Neandertals in showing a somewhat humans, could reflect higher levels of habitual activity (Ben-Itzhak larger metaconid, but the metaconid in the Gegant-5 specimen et al., 1988; Trinkaus and Churchill, 1999; Rhodes and Knüsel, does not seem to follow this same pattern. Whether the variation in 2005). Similarly, in a study of humeral growth in an archaeolog- M1 cusp proportions in Neandertals consistently follows any tem- ical modern human population, cortical area growth was found to poral or geographic patterns can only be assessed within a larger lag behind growth in length (Sumner and Andriacchi, 1996). These sample of Pleistocene Homo teeth. authors found that by age 10 the bone had reached about 70% of The age at death estimate of 4.5e5.0 years in the Gegant-5 in- adult length, but only about 45% of adult cortical area. In addition, dividual is younger than that derived based on modern human increase in length ceased between 15 and 20 years of age, but in- standards (5.5e6.8 years), but is consistent with the suggestion creases in cortical area did not plateau until the early 20s. This might that dental development in Neandertals was more rapid than in imply that increased mechanical loading after growth in height had modern humans (Ramirez-Rozzi and Bermúdez de Castro, 2004; ceased underpins cortical thickening, a suggestion echoed by some Smith et al., 2007c). Smith et al. (2007c) suggest than one crucial studies of living human cohorts (Clark et al., 2007). Nevertheless, developmental difference between Neandertals and modern the appearance of thickened cortices in Neandertal specimens at humans may be an earlier onset of enamel mineralization in the relatively young ages suggests some degree of genetic control is also former. However, further study is required to confirm this sugges- present (Ruff et al., 1994). tion, and modern humans do show some variation in this param- eter as well (Dean et al., 1993; Tompkins, 1996a; Reid et al., 1998; Minimum number of individuals (MNI) Liversidge, 2008). At 4.5e5.0 years of age, the M1 in the Gegant individual was The presence of five hominin specimens in the Cova del Gegant already erupted, although only very slight wear is visible on the sample suggests that some of the remains could be associated. The mesiobuccal cusp. This result is consistent with other studies that approximately similar ages at death for both the Gegant-4 and have claimed an advanced molar development in Neandertals Gegant-5 specimens raises the possibility that these two bones (Wolpoff, 1979; Smith et al., 2010), and an early M1 eruption has represent the same individual, and indeed they were found in close important implications since many life history traits correlate spatial proximity (~170 cm apart and at the same depth), in the strongly across primates with the emergence of the M1 (Smith, same geological level (layer Vf) and area (GP2) of the site. Based on 1991a; Macho, 2001). However, hypotheses about developmental age at death incompatibilities, chronological differences, and patterns in Neandertals should ideally integrate both dental and spatial separation at the site, the remaining specimens seem to skeletal data. represent isolated individuals. Thus, the MNI for the Gegant sample Although Neandertal humeri from children of similar age as the would be four if Gegant-4 and Gegant-5 are associated or five if Gegant-4 specimen are rare, the present results agree with previous they are not. 26 R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28

Conclusion Gegant (Spain). In: Condemi, S., Weniger, G.C. (Eds.), Continuity and Disconti- nuity in the Peopling of Europe: One Hundred Fifty Years of Neanderthal Study. Springer, Dordrecht, pp. 213e217. The Cova del Gegant remains show clear Neandertal affinities Arsuaga, J., Fernandez Peris, J., Gracia, A., Quam, R., Carretero, J., Gonzalez, V., and currently are the only known fossil assemblage from NE Iberia Blasco, R., Cuartero, F., Sanudo,~ P., 2012. Fossil human remains from Bolomor e where this diagnostic morphology has been identified in a secure, Cave (Valencia, Spain). J. Hum. Evol. 62, 629 639. Bailey, S., 2002. A closer look at Neanderthal postcanine dental morphology: the well-dated stratigraphic context. The two new specimens reported mandibular dentition. Anat. Rec. (New Anat.) 269, 148e156. here, a juvenile mandible (Gegant-5) and humerus (Gegant-4), Bailey, S., 2004. A morphometric analysis of maxillary molar crowns of Middle-Late provide evidence for rapid dental development and early post- Pleistocene hominins. J. Hum. Evol. 47, 183e198. Bailey, S., 2006. Beyond shovel-shaped incisors: Neandertal dental morphology in a cranial cortical thickening in Neandertals. The similar age at death comparative context. Periodicum Biologorum 108, 253e267. estimations suggest that the remains may derive from a single in- Bailey, S., Hublin, J., 2006. Dental remains from the Grotte du Renne at Arcy-sur- dividual. The specimens provide new information on the anatom- Cure (Yonne). J. Hum. Evol. 50, 485e508. Bailey, S., Lynch, J., 2005. Diagnostic differences in mandibular P4 shape between ical variation and ontogenetic processes which characterized Neandertals and anatomically modern humans. Am. J. Phys. Anthropol. 126, Neandertals and join a growing sample of Pleistocene hominin 268e277. fossils from the Iberian Peninsula documenting the course of hu- Barroso Ruiz, C., de Lumley, H., 2006. La Grotte du Boquete de Zafarraya (Malaga, man evolution during the European Upper Pleistocene (Quam et al., Andalousie). Junta de Andalucia. Consejería de Cultura, Sevilla. Bayle, P., Braga, J., Mazurier, A., Macchiarelli, R., 2009. Dental developmental pattern 2001; Zilhao and Trinkaus, 2002; Arsuaga et al., 2002, 2007, 2012; of the Neanderthal child from Roc de Marsal: a high-resolution 3D analysis. Daura et al., 2005; Garralda, 2005; Barroso Ruiz and de Lumley, J. Hum. Evol. 56, 66e75. 2006; Rosas et al., 2006; Trinkaus et al., 2007; Walker et al., Ben-Itzhak, S., Smith, P., Bloom, R., 1988. Radiographic study of the humerus in Neandertals and Homo sapiens sapiens. Am. J. Phys. Anthropol. 77, 231e242. 2008; Willman et al., 2012). Bermúdez de Castro, J., 1993. The Atapuerca dental remains. New evidence (1987e1991 excavations) and interpretations. J. Hum. Evol. 24, 339e371. Acknowledgments Bermudez de Castro, J., Nicolas, M.E., 1996. Changes in the lower premolar size sequence during hominid evolution. Phylogenetic implications. Hum. Evol. 11, 205e215. This paper is an output of the research project Humans, carní- Bermúdez de Castro, J.M., Sarmiento, S., Cunha, E., Rosas, A., Bastir, M., 2001. Dental size variation in the Atapuerca-SH Middle Pleistocene hominids. J. Hum. Evol. vors i medi natural durant el Plistoce al massís del Garraf-Ordal i e 41, 195 209. curs baix del riu Llobregat, part of the Research Project El Plistoce Bermúdez de Castro, J.M., Carretero, J.M., García-Gonzalez, R., Rodríguez-García, L., Superior i l'Holoce a Catalunya, supported by the 2014SGR-108 Martinon-Torres, M., Rosell, J., Blasco, R., Martín-Frances, L., Modesto, M., (Generalitat de Catalunya) and HAR2011-26193 (MICINN) projects. Carbonell, E., 2012. Early Pleistocene human humeri from the Gran Dolina-TD6 site (Sierra de Atapuerca, Spain). Am. J. Phys. Anthropol. 147, 604e617. Fieldwork was sponsored by Servei d'Arqueologia i Paleontologia Beynon, A., Clayton, C., Ramírez Rozzi, F., Reid, D., 1998. Radiographic and histo- (2014/100639) (Generalitat de Catalunya) and Ajuntament de logical methodologies in estimating the chronology of crown development in Sitges. A portion of this research was financed by the Ministerio de modern humans and great apes: a review, with some applications for studies on juvenile hominids. J. Hum. Evol. 35, 351e370. Ciencia e Innovacion of the Government of Spain (Grant number: Billy, G., 1979. L'enfant Madgalenien de la Grotte du Figuier (Ardeche). L'Anthro- CGL2009-12703-C03-03). The authors thank the various curators pologie (Paris) 83, 223e252. and institutions that have kindly provided access to original fossil Braga, J., Heuze, Y., 2007. Quantifying variation in human dental development se- quences: an EVO-DEVO perspective. In: Bailey, S., Hublin, J. (Eds.), Dental Per- specimens housed under their care. CT scanning of the Gegant-5 spectives on Human Evolution. State of the Art Research in Dental mandible was carried out in collaboration with Jose Miguel Carre- Paleoanthropology. Springer, Dordrecht, pp. 247e262. tero at the Laboratorio de la Evolucion Humana at the Universidad Bromage, T., Dean, C., 1986. Re-evaluation of the age at death of immature fossil e de Burgos (Spain) and financed by a grant from the Junta de Castilla hominids. Nature 317, 525 527. Camps, M., Higham, T., 2012. Chronology of the Middle to Upper Paleolithic tran- yLeon (Project No. BU005A09). J. Daura was supported by a post- sition at Abric Romaní, Catalunya. J. Hum. Evol. 62, 89e103. doctoral grant (Juan de la Cierva Subprogram JCI-2011-09543). L. Clark, E., Ness, A., Tobias, J., 2007. Gender differences in the ratio between humerus fi width and length are established prior to puberty. Osteoporosis Int. 18, Rodríguez has enjoyed nancial support from the Fundacion Ata- e 463 470. puerca and the European Social Fund of the Junta de Castilla y Leon Colombo, A., Coqueugniot, H., Dutailly, B., Desbarats, P., Tillier, A.-M., 2013. Nou- (CPIN Project No. 03-461AA-692.01). velles donnees sur l'edi fication des molaires et l'estimation de l'age^ dentaire des enfants par imagerie medicale et 3D: apports et perspectives. Bull. Mem. Soc. Anthropol. Paris 25, 127e146. References Coquegniot, H., Minugh-Purvis, N., 2003. Ontogenetic patterning and phylogenetic significance of mental foramen number and position in the evolution of Upper Adan- Alvarez, G., 2003. Memoria de la Actuacion Arqueologica en el Antiguo Pleistocene Homo sapiens. In: Thompson, J., Krovitz, G., Nelson, A. (Eds.), Pat- Monasterio de San Pablo, Burgos. Junta de Castilla y Leon, Valladolid. terns of Growth and Development in the Genus Homo. Cambridge University Alcazar de Velazco, A., Arsuaga, J., Martínez, I., Bonmatí, A., 2011. Revision de la Press, Cambridge, pp. 343e360. mandibula humana de Banolas,~ Gerona, Espana.~ Bol. R. Soc. Esp. Hist. Nat. Sec. Cortada, T., Maroto, J., 1990. La dent humana paleolítica de la cova de Mollet I Geol. 105, 99e108. (Serinya). Quaderns del Centre d'Estudis Comarcals de Banolas~ 1988e1989, Anderson, D., Thompson, G., Popovich, F., 1976. Age of attainment of mineralization pp. 135e148. stages of the permanent dentition. J. Forensic Sci. 21, 191e200. Cowgill, L., 2010. The ontogeny of Holocene and Late Pleistocene human postcranial Antoine, D., 2001. Evaluating the periodicity of incremental structures in dental strength. Am. J. Phys. Anthropol. 141, 16e37. enamel as a means of studying growth in children from past populations. Ph.D. Daura, J., 2008. Caracteritzacio arqueologica i paleontologica dels jaciments plis- dissertation, University College London, London. tocens del massís del Garraf-Ordal i curs baix del riu Llobregat. Ph.D. Thesis, Arsuaga, J.L., Gracia, A., Martínez, I., Bermúdez de Castro, J.M., Rosas, A., Universitat de Barcelona. Villaverde, V., Fumanal, M.P., 1989. The human remains from Cova Negra Daura, J., Sanz, M., 2011e2012. Procedencia estratigrafica de los restos humanos (Valencia, Spain) and their place in European Pleistocene human evolution. neandertales de la cova del Gegant (Sitges, Barcelona). Mainake XXXIII, 221e236. J. Hum. Evol. 18, 55e92. Daura, J., Sanz, M., Subira, M., Quam, R., Fullola, J., Arsuaga, J.L., 2005. A Neandertal Arsuaga, J.L., Martìnez, I., Villaverde, V., Lorenzo, C., Quam, R., Carretero, J.M., mandible from the Cova del Gegant (Sitges, Barcelona, Spain). J. Hum. Evol. 49, Gracia, A., 2001. Fosiles humanos del Paìs Valenciano. In: Villaverde, V. (Ed.), De 56e70. Neandertales a Cromanones.~ El Inicio del Poblamiento Humano en las Tierras Daura, J., Sanz, M., Pike, A., Zilhao,~ J., Subira, M., Fornos, J., Fullola, J., Julia, R., 2010. Valenicanas. Universitat de Valencia, Valencia, pp. 265e322. Stratigraphic context and direct dating of the Neandertal mandible from Cova Arsuaga, J.L., Villaverde, V., Quam, R., Gracia, A., Lorenzo, C., Martínez, I., del Gegant (Sitges, Barcelona). J. Hum. Evol. 59, 109e122. Carretero, J.M., 2002. The Gravettian occipital bone from the site of Malladetes Dawson, H., Robson Brown, K., 2012. Childhood tuberculosis: a probable case from (Barx, Valencia, Spain). J. Hum. Evol. 43, 381e393. late mediaeval Somerset, England. Int. J. Paleopathol. 2, 31e35. Arsuaga, J.L., Villaverde, V., Quam, R., Martínez, I., Carretero, J.M., Lorenzo, C., Dean, M.C., Reid, D., 2001. Perikymata spacing and distribution on hominid anterior Gracia, A., 2007. New Neandertal remains from Cova Negra (Valencia, Spain). teeth. Am. J. Phys. Anthropol. 116, 209e215. J. Hum. Evol. 52, 31e58. Dean, M.C., Beynon, A., Reid, D., Whittaker, D., 1993. A longitudinal study of tooth Arsuaga, J., Quam, R., Daura, J., Sanz, M., Subira, M., Dalen, L., Gotherstr€ om,€ A., 2011. growth in a single individual based on long- and short-period incremental Neandertal mtDNA from a Late Pleistocene human mandible from the Cova del markings in dentine and enamel. Int. J. Osteoarchaeol. 3, 249e264. R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28 27

Dean, C., Leakey, M., Reid, D., Schrenk, F., Schwartz, G., Stringer, C., Walker, A., 2001. Moorrees, C., Fanning, E., Hunt, E., 1963. Age variation of formation stages for ten Growth processes in teeth distinguish modern humans from Homo erectus and permanent teeth. J. Dent. Res. 42, 1490e1502. earlier hominins. Nature 414, 628e631. Mora, R., 1988. El Paleolítico Medio en Catalunya. Universitat de Barcelona, Demirjian, A., 1996. Electronic encyclopedia on maxillo-facial, dental and skeletal Barcelona. development. Universite de Montreal, Montreal. Mora, R., Martínez Moreno, J., de la Torre, I., Casanova, J., 2008. Variabilidad Tecnica Demirjian, A., Goldstein, H., Tanner, J., 1973. A new system of dental age assesment. del Paleolítico Medio en el Sudoeste de Europa. Universidad Autonoma de Hum. Biol. 45, 211e227. Barcelona, Barcelona. Fazekas, I., Kosa, F., 1978. Forensic Fetal Osteology. Akademiai Kaido, Budapest. Mora, R., Benito-Calvo, A., Martínez Moreno, J., Gonzalez Marcen, P., de la Torre, I., FitzGerald, C., 1998. Do enamel microsturctures have regular time dependency? 2011. Chrono-stratigraphy of the Upper Pleistocene and Holocene archaeolog- Conclusions from the literature and a large-scale study. J. Hum. Evol. 35, ical sequence in Cova Gran (South-Eastern Pre-Pyrenees, Iberian Peninsula). 371e386. J. Quatern. Sci. 26, 635e644. García-Gonzalez, R., Carretero, J., Rodríguez, L., Arsuaga, J., 2011. Histological Olejniczak, A., Smith, T., Feeney, R., Macchiarelli, R., Mazurier, A., Bondioli, L., reconstruction of dental development in six individuals from San Pablo medi- Rosas, A., Fortea, J., de la Rasilla, M., García-Tabernero, A., Radovcic, J., eval site (Burgos, Spain). In: Kohler,€ M., Jordana, X., Moratalla, N. (Eds.), Skinner, M., Toussaint, M., Hublin, J., 2008. Dental tissue proportions and I International Symposium on Paleohistology, p. 36. enamel thickness in Neandertal and modern human molars. J. Hum. Evol. 55, Garn, S., Cole, P., Wainwright, R., Guire, K., 1977. Sex discriminatory effectiveness 12e23. using combinations of permanent teeth. J. Dent. Res. 56, 697. Quam, R., Arsuaga, J.L., Bermúdez de Castro, J.M., Díez, J.C., Lorenzo, C., Garralda, M., 2005. Los Neandertales en la Península Iberica. Munibe 57, 289e314. Carretero, J.M., García, N., Ortega, A., 2001. Human remains from Valdegoba Garralda, M., Vandermeersch, B., 2000. Les Neandertaliens de la Grotte de Combe- Cave (Huermeces, Burgos, Spain). J. Hum. Evol. 41, 385e435. Grenal (Domme, France). Paleo 12, 213e259. Quam, R., Bailey, S., Wood, B., 2009. Evolution of M1 crown size and cusp pro- Gomez-Robles, A., Martinon-Torres, M., Bermúdez de Castro, J., Prado, L., portions in the genus Homo. J. Anat. 214, 655e670. Sarmiento, S., Arsuaga, J., 2008. Geometric morphometric analysis of the crown Ramirez-Rozzi, F., Bermúdez de Castro, J., 2004. Surprisingly rapid growth in Ne- morphology of the lower first premolar of hominins, with special attention to anderthals. Nature 428, 936e939. Pleistocene Homo. J. Hum. Evol. 55, 627e638. Reid, D., Dean, M., 2000. Brief communication: the timing of linear hypoplasias on Grün, R., Maroto, J., Eggins, S., Stringer, C., Roberston, S., Taylor, L., Mortimer, G., human anterior teeth. Am. J. Phys. Anthropol. 113, 135e139. McCulloch, M., 2006. ESR and U-series analyses of enamel and dentine frag- Reid, D., Dean, M., 2006. Variation in modern human enamel formation times. ments of the Banolas~ mandible. J. Hum. Evol. 50, 347e358. J. Hum. Evol. 50, 329e346. Guatelli-Steinberg, D., Reid, D., 2008. What molars contribute to an emerging un- Reid, D., Ferrell, R., 2006. The relationship between number of striae of Retzius and derstanding of lateral enamel formation in Neandertals vs. modern humans. their periodicity in imbricational enamel formation. J. Hum. Evol. 50, 195e202. J. Hum. Evol. 54, 236e250. Reid, D., Beynon, A., Ramírez Rozzi, F., 1998. Histological reconstrucion of dental Guatelli-Steinberg, D., Reid, D., Bishop, T., Larsen, C., 2005. Anterior tooth growth development in four individuals from a medieval site in Picardie, France. periods in Neandertals were comparable to those of modern humans. Proc. J. Hum. Evol. 35, 463e477. Natl. Acad. Sci. 102, 14197e14202. Rhodes, J.A., Knüsel, C.J., 2005. Activity-related skeletal change in medieval humeri: Guatelli-Steinberg, D., Reid, D., Bishop, T., 2007. Did the lateral enamel of Neandertal cross-sectional and architectural alterations. Am. J. Phys. Anthropol. 128, anterior teeth grow differently from that of modern humans? J. Hum. Evol. 52, 536e546. 72e84. Risnes, S., 1986. Enamel apposition rate and the prism periodicity in human teeth. Heim, J.L., 1982. Les Enfants Neandertaliens de La Ferrassie. Masson, Paris. Scand. J. Dent. Res. 94, 394e404. Heim, J.L., 1991. L'enfant Magdalenien de la Madeleine. L'Anthropologie (Paris) 95, Rodríguez, L., García, R., Sanz, M., Daura, J., Quam, R., Fullola, J., Arsuaga, J., 2011. 611e638. A Neanderthal lower incisor from Cova del Gegant (Sitges, Barcelona, Spain). Hershkovitz, I., Smith, P., Sarig, R., Quam, R., Rodriguez, L., García, R., Arsuaga, J., Bol. R. Soc. Esp. Hist. Nat. Sec. Geol. 105, 25e30. Barkai, R., Gopher, A., 2011. Middle Pleistocene dental remains from Qesem Cave Rosas, A., Martinez-Maza, C., Bastir, M., Garcia-Tabernero, A., Lalueza-Fox, C., (Israel). Am. J. Phys. Anthropol. 144, 575e592. Huguet, R., Ortiz, J.E., Julia, R., Soler, V., de Torres, T., Martinez, E., Canaveras, J.C., Julia, R., Bischoff, J., 1991. Radiometric dating of Quaternary deposits and the Sanchez-Moral, S., Cuezva, S., Lario, J., Santamaria, D., de la Rasilla, M., Fortea, J., hominid mandible of Lake Banyolas, Spain. J. Archaeol. Sci. 18, 707e722. 2006. Paleobiology and comparative morphology of a late Neandertal sample Kallay, J., 1963. A radiographic study of the Neanderthal teeth from Krapina, Croatia. from El Sidron, Asturias, Spain. Proc. Natl. Acad. Sci. 103, 19266e19271. In: Brothwell, D. (Ed.), Dental Anthropology. Pergamon Press, Oxford, pp. 75e86. Ruff, C., Walker, A., Trinkaus, E., 1994. Postcranial robusticity in Homo, III: ontogeny. Leigh, S., Blomquist, G., 2007. Life history. In: Campbell, C., Fuentes, A., Am. J. Phys. Anthropol. 93, 35e54. MacKinnon, A., Panger, M., Bearder, S.K. (Eds.), Primates in Perspective. Oxford Samper Carro, S., Martínez Moreno, J., 2014. Who let the hyenas out? Taphonomic University Press, pp. 396e407. analysis of the faunal assemblage from GL-1 of Cova del Gegant (Sitges, Spain). Liversidge, H., 2008. Timing of human mandibular third molar formation. Ann. Quatern. Int. 330, 19e35. Hum. Biol. 35, 294e321. Sanz, M., 2013. Patrons d'acumulacio de restes de fauna del Plistoce superior al Liversidge, H., Molleson, T., 2004. Variation in crown and root formation and nord-est peninsular (area del Massís del Garraf-Ordal). Ph.D. Thesis, Universitat eruption of human deciduous teeth. Am. J. Phys. Anthropol. 123, 172e180. de Barcelona. Liversidge, H., Chaillet, N., Mornstad,€ H., Nystrom,€ M., Rowlings, K., Taylor, J., Sawada, J., Kondo, O., Nara, T., Dodo, Y., Akazawa, T., 2004. Bone histomorphology of Willens, G., 2006. Timing of Demirjian's tooth formation stages. Ann. Hum. Biol. the Dederiyeh Neanderthal child. Anthropol. Sci. 112, 247e256. 33, 454e470. Scheuer, L., Black, S., 2000. Developmental Juvenile Osteology. Academic Press, San Macchiarelli, R., Bondioli, L., Debenath, A., Mazurier, A., Tournepiche, J., Birch, W., Diego. Dean, M., 2006. How Neanderthal molar teeth grew. Nature 444, 748e751. Schour, I., 1936. Neonatal line in enamel and dentin of human deciduous teeth and Macho, G., 2001. Primate molar crown formation times and life history evolution first permanent molar. J. Am. Dent. Assoc. 23, 1946e1955. revisited. Am. J. Primatol. 55, 189e201. Schour, I., Massler, M., 1944. Development of the Human Dentition. American Madre-Dupouy, M., 1992. L'enfant du Roc de Marsal: Etude Analytique et Dental Association, Chicago. Comparative. CNRS, Paris. Schwartz, G., Reid, D., Dean, C., 2001. Developmental aspects of sexual dimorphism Mallegni, F., Trinkaus, E., 1997. A reconsideration of the Archi 1 Neandertal in hominoid canines. Int. J. Primatol. 22, 837e860. mandible. J. Hum. Evol. 33, 651e668. Scott, G., 1977. Classification, sex dimorphism, association and population variation Maresh, M.M., 1970. Measurements from roentgenograms. Hum. Growth Develop. of the canine distal accessory ridge. Hum. Biol. 49, 453e469. 157e200. Shaw, J., 1928. Taurodont teeth in South African races. J. Anat. 62, 276e298. Maroto, J., Soler, N., Mir, A., 1987. La cueva de Mollet I (Serinya, Gerona). Cypsela 6, Smith, B.H.,1991a. Dental development and the evolution of life history in Hominidae. 101e110. Am. J. Phys. Anthropol. 86, 157e174. Maroto, J., Vaquero, M., Arrizabalaga, A., Baena, J., Baquedano, E., Jorda, J., Julia, R., Smith, B.H., 1991b. Standards of human tooth formation and dental age assessment. Montes, R., van der Plicht, J., Rasines, P., Wood, R., 2012. Current issues in late In: Kelley, M., Larsen, C. (Eds.), Advances in Dental Anthropology. Wiley-Liss, Middle Paleolithic chronology: new assessments form Northern Iberia. Quatern. New York, pp. 143e168. Int. 247, 15e25. Smith, B.H., Tompkins, R., 1995. Toward a life history of the Hominidae. Ann. Rev. Martínez Moreno, J., 1990. Informe tecnico de los restos oseos de la cova del Gegant Anthropol. 24, 257e279. (Sitges, Garraf). Servei d'Arqueologia i Paleontologia, Centre d'Informacioi Smith, T., Reid, D., Siriani, J., 2006. The accuracy of histological assessments of Documentacio del Patrimoni Cultural de la Generalitat de Catalunya, Barcelona. dental development and age at death. J. Anat. 208, 125e138. Martínez Moreno, J., Mora, R., de la Torre, I., 2010. The Middle-to-Upper Paleolithic Smith, T., Reid, D., Dean, M., Olejniczak, A., Ferrell, R., Martin, L., 2007a. New per- transition in Cova Gran (Catalunya, Spain) and the extinction of Neandertals in spectives on chimpanzee and human molar crown development. In: Bailey, S., the Iberian Peninsula. J. Hum. Evol. 58, 211e226. Hublin, J. (Eds.), Dental Perspectives on Human Evolution: State of the Art Martinon-Torres, M., Bastir, M., Bermúdez de Castro, J., Gomez, A., Sarmiento, S., Research in Dental Paleoanthropology. Springer, Dordrecht, pp. 177e192. Muela, A., Arsuaga, J.L., 2006. Hominin lower second premolar morphology: Smith, T., Tafforeau, P., Reid, D., Grün, R., Eggins, S., Boutakiout, M., Hublin, J., 2007b. evolutionary inferences through geometric morphometric analysis. J. Hum. Earliest evidence of modern human life history in North African early Homo Evol. 50, 523e533. sapiens. Proc. Natl. Acad. Sci. 104, 6128e6133. Martinon-Torres, M., Bermúdez de Castro, J., Gomez-Robles, A., Prado-Simon, L., Smith, T., Toussaint, M., Reid, D., Olejniczak, A., Hublin, J., 2007c. Rapid dental Arsuaga, J., 2012. Morphological description and comparison of the dental re- development in a Middle Paleolithic Belgian Neandertal. Proc. Natl. Acad. Sci. mains from Atapuerca-Sima de los Huesos site (Spain). J. Hum. Evol. 62, 7e58. 104, 20220e20225. 28 R. Quam et al. / Journal of Human Evolution 81 (2015) 13e28

Smith, T., Tafforeau, P., Reid, D., Pouech, J., Lazzari, V., Zermeno, J., Guatelli- Vaquero, M., 2005. Short human occupations in the Middle Palaeolithic level i Steinberg, D., Olejniczak, A., Hoffman, A., Radovcic, J., Masrour, M., Toussaint, M., of the Abric Romaní rock-shelter (Capellades, Barcelona, Spain). J. Hum. Evol. Stringer, C., Hublin, J., 2010. Dental evidence for ontogenetic differences be- 48, 157e174. tween modern humans and Neanderthals. Proc. Natl. Acad. Sci. 107, Vaquero, M., Carbonell, E., 2012. Some clarifications on the Middle-Upper Paleo- 20923e20928. lithic transition in Abric Romaní: Reply to Camps and Higham (2012). J. Hum. Sumner, D., Andriacchi, T., 1996. Adaptation to differential loading: comparison of Evol. 63, 711e717. growth-related changes in cross-sectional properties of the human femur and Vinas,~ A., Villalta, J., 1975. El deposito cuaternario de la “Cova del Gegant”. Speleon V humerus. Bone 19, 121e126. Symposium de Espeleología 19e33. Thiebaut, C., Mourre, V., Chalard, P., Cologne, D., Doudenneau, A., Deschamps, M., Walker, M., Gibert, J., Lopez, M., Lombardi, A., Perez-P erez, A., Zapata, J., Ortega, J., Sacco-Sonador, A., 2012. Lithic technology of the final Mousterian on both sides Higham, T., Pike, A., Schwenninger, J., Zilhao,~ J., Trinkaus, E., 2008. Late Nean- of the Pyrenees. Quatern. Int. 247, 182e198. dertals in southeastern Iberia: Sima de las Palomas del Cabezo Gordo, Murcia, Tillier, A.-M., 1999. Les Enfants Mousteriens de Qafzeh: Interpretation Phylogene- Spain. Proc. Natl. Acad. Sci. 105, 20631e20636. tique et Paleoauxologique. CNRS Editions, Paris. Walker, M.J., Lombardi, A.V., Zapata, J., Trinkaus, E., 2010. Neandertal mandibles Tompkins, R., 1996a. Human population variability in relative dental development. from the Sima de las Palomas del Cabezo Gordo, Murcia, southeastern Spain. Am. J. Phys. Anthropol. 99, 79e102. Am. J. Phys. Anthropol. 142, 261e272. Tompkins, R., 1996b. Relative dental development of Upper Pleistocene hominids Walker, M.J., Ortega, J., Lopez, M.V., Parmova, K., Trinkaus, E., 2011a. Neandertal compared to human population variation. Am. J. Phys. Anthropol. 99, 103e118. postcranial remains from the Sima de las Palomas del Cabezo Gordo, Murcia, Tompkins, R., Trinkaus, E., 1987. La Ferrasie 6 and the development of Neanderthal southeastern Spain. Am. J. Phys. Anthropol. 144, 505e515. pubic morphology. Am. J. Phys. Anthropol. 73, 233e239. Walker, M.J., Ortega, J., Parmova, K., Lopez, M.V., Trinkaus, E., 2011b. Morphology, Trefný, P., 2005. Size reduction of the Le Moustier 1 molars e a 2D analysis. In: body proportions, and postcranial hypertrophy of a female Neandertal from the Ullrich, H. (Ed.), The Neandertal Adolescent Le Moustier 1. New Aspects, New Sima de las Palomas, southeastern Spain. Proc. Natl. Acad. Sci. 108, Results. Staatliche Museen zu Berlin, Berlin, pp. 187e196. 10087e10091. Trinkaus, E., 1983. The Shanidar Neandertals. Academic Press, New York. Walker, M.J., Lopez-Martínez, M.V., Ortega-Rodriganez,~ J., Haber-Uriarte, M., Lopez- Trinkaus, E., 2002. The mandibular morphology. In: Zilhao,~ J., Trinkaus, E. (Eds.), Jimenez, A., Aviles-Fern andez, A., Polo-Camacho, J.L., Campillo-Boj, M., García- Portrait of the Artist as a Child. The Gravettian Human Skeleton from the Abrigo Torres, J., Carrion García, J.S., 2012. The excavation of buried articulated Nean- do Lagar Velho and its Archeological Context. Instituto Portugues^ de Arqueo- derthal skeletons at Sima de las Palomas (Murcia, SE Spain). Quatern. Int. 259, logía, Lisbon, pp. 312e325. 7e21. Trinkaus, E., Churchill, S.E., 1999. Diaphyseal cross-sectional geometry of Near Willman, J.C., Maki, J., Bayle, P., Trinkaus, E., Zilhao,~ J., 2012. Middle Paleolithic Eastern Middle Palaeolithic humans: the humerus. J. Archaeol. Sci. 26, 173e184. human remains from the Gruta Da Oliveira (Torres Novas), Portugal. Am. J. Phys. Trinkaus, E., Ruff, C., Esteves, F., Santos-Coelho, J., Silva, M., Mendonça, M., 2002. The Anthropol. 149, 39e51. upper limb remains. In: Zilhao, J., Trinkaus, E. (Eds.), Portrait of the Artist as a Wolpoff, M., 1979. The Krapina dental remains. Am. J. Phys. Anthropol. 50, 67e117. Child. The Gravettian Human Skeleton from the Abrigo do Lagar Velho and its Wood, B., Abbott, S., 1983. Analysis of the dental morphology of Plio-Pleistocene Archeological Context. Instituto Portugues^ de Arqueología, Lisboa, pp. 466e488. hominids I. Mandibular molars: crown area measurements and morpholog- Trinkaus, E., Maki, J., Zilhao,~ J., 2007. Middle Paleolithic human remains from the ical traits. J. Anat. 136, 197e219. Gruta da Oliveira (Torres Novas), Portugal. Am. J. Phys. Anthropol. 134, 263e273. Wood, B., Abbott, S., Graham, S., 1983. Analysis of the dental morphology of Plio- Turner, C., Nichol, C., Scott, G., 1991. Scoring procedures for key morphological traits Pleistocene hominids II. Mandibular molars e study of cusp areas, fissure of the permanent dentition: the Arizona State University Dental Anthropology patterns and cross sectional shape of the crown. J. Anat. 137, 287e314. System. In: Kelley, M., Larsen, C. (Eds.), Advances in Dental Anthropology. Zilhao, J., Trinkaus, E., 2002. Portrait of the Artist as a Child. The Gravettian Human Wiley-Liss, New York, pp. 13e31. Skeleton from the Abrigo do Lagar Velho and its Archaeological Context, Tra- Vallverdú, J., Allue, E., Bischoff, J., Caceres, I., Carbonell, E., Cebria, A., García- balhos de Arqueologia. Instituto Portugues de Arqueologia, Lisbon. Anton, M., Huguet, R., Ibanez,~ N., Martínez, K., Pasto, I., Rosell, J., Saladie, P.,