Temporal labyrinths of eastern Eurasian Pleistocene humans

Discussion

These data and comparisons confirm what has been documented for the western Old World, that there is a general similarity in temporal labyrinthine proportions through most of the genus Homo (1, 7). The exception is the western and central Eurasian Neandertals (1⇓–3), almost all of whom exhibit a derived morphology of their labyrinths. As with many individual features, for which the Neandertals are overwhelmingly separate from modern humans but for which their distributions overlap given sufficient sample sizes (16), the ranges of variation of these samples overlap in their labyrinthine aspects. This minimal distributional overlap is evident in the positions of the Le Moustier 1 labyrinth and those of a few recent humans.

In eastern Asia, the Early and Middle Pleistocene Lantian 1 and Hexian 1 crania follow the general Homo pattern, and they join Sangiran 2 and 4 in establishing this labyrinthine morphology as ancestral for Homo across Eurasia. The Late Pleistocene Liujiang 1 has an equally modern labyrinthine configuration, in agreement with its overall modern human morphology (17). In this context, the labyrinthine morphology of the Xujiayao 15 temporal bone conforms to the derived Neandertal pattern, the first documentation of this pattern among Pleistocene Homo outside of the Neandertal geographical range.

Despite the similarity of the Xujiayao 15 labyrinth to those of the Neandertals, its external temporal morphology (Fig. 4 and SI Appendix, Fig. S2) bears little affinity to that of the Neandertals (18, 19). It shares ancestral Homo features with the Neandertals, including a large juxtamastoid eminence and an inferomedially sloping lateral mastoid process. It differs in its other aspects. The squamous portion is high and rounded. The zygomatic arch extends posteriorly above the auditory porous, continuing into a supramastoid crest that is horizontal and ends temporally at the parietal notch. The porous is ovoid, but its long axis is vertical. The laterally rugose mastoid process lacks an anterior mastoid tubercle. The parietomastoid suture slopes inferiorly to asterion, and the transverse sinus crossed to the temporal bone across the posterior parietomastoid suture. Some of these aspects of Xujiayao 15 can be found on individual Neandertal specimens, but the constellation of discrete traits of the bone is outside of the well-documented range of variation in Neandertal temporal bone external morphology.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

Lateral views of the Xujiayao 15 left temporal bone (Upper) and the La Quina 27 Neandertal right temporal bone (reversed, Lower).

The same morphological consideration applies to the Xujiayao human remains generally (SI Appendix, Table S7). A couple of their features are common among the Neandertals but not unique to them (the Xujiayao 1 bilevel nasal floor and strongly shoveled maxillary central incisor (I¹) and canine (C¹), and the Xujiayao 14 enlarged superior medial pterygoid tubercle) (20⇓–22), but the overall configurations of the Xujiayao remains are unlike those of most Neandertals (20⇓⇓⇓⇓–25). The Xujiayao 1 inferior nasal margin is broad and rounded, and its M¹ has marked cingular development and peripherally placed cusps. The Xujiayao 1 and 13 maxillary first molars (M¹s) possess subrectangular contours. The Xujiayao 6 and 12 occipital bones are angled at the nuchal torus and lack suprainiac fossae. The Xujiayao 14 mandibular ramus is wide and gonially everted, and it has an open mandibular foramen and a laterally placed mandibular notch crest. The same pattern holds for other, late Middle and early Late Pleistocene archaic human remains from eastern Asia (14, 26⇓–28). What emerges from these East Asian later Pleistocene archaic humans, therefore, is a complex mix of features, many of them generally ancestral for Homo and a few features that occur frequently in the Neandertals but need not be diagnostic of them. The Xujiayao 15 labyrinth therefore provides the first secure evidence of a distinctly derived Neandertal configuration in eastern Asia.

The geographically closest presence of a similar labyrinthine configuration is the fragmentary Obi-Rakhmat 1 cranium from Uzbekistan (13), a specimen assigned to the Neandertals on the basis of its dental morphology (29) despite an apparent mix of morphological features in other aspects of the remains (13). The next geographically closest, morphologically “Neandertal” labyrinths are from the eastern Mediterranean littoral (Dederiyeh 1 and Tabun 1).

The broader implications of the finding, a “Neandertal” labyrinthine morphology in an otherwise distinctly “non-Neandertal” sample of late archaic humans from eastern Eurasia, remain unclear. It is tempting to use it as evidence of population contact (gene flow) between central and western Eurasian Neandertals and these eastern archaic humans. Indeed, this labyrinthine morphology has been used repeatedly as a taxonomic marker of the Neandertals (1, 3, 11⇓–13). However, there have also been suggestions (1) that labyrinth shape may be related to other aspects of Neandertal biology, albeit with reservations, and only secondarily a consequence of their phylogenetic history.

Spoor et al. (1) suggested a series of Neandertal postcranial features that might account for at least the dimensions of their anterior and posterior semicircular canals, as related to overall agility in locomotion (3, 30). None of these postcranial features is known for the Xujiayao humans, but they can be assessed for the Neandertals. However, for them to be relevant to the Neandertal (and Xujiayao 15) labyrinthine proportions, they need to be features that distinguish the Neandertals from both modern humans and the earlier Pleistocene, non-Neandertal, humans who lack this unusual labyrinthine configuration.

There is a series of Neandertal features that may be related to issues of balance and agility (1). Neandertals have relatively abbreviated limbs (31), yet similarly proportioned modern arctic humans do not appear to differ from other modern humans in their labyrinths (1). Neandertals have robust necks, especially with respect to cervical spinous processes (19, 32), but their occipital nuchal areas are modest in size for Pleistocene Homo (19, 33, 34). Moreover, their necks were not particularly short (19, 32, 35). Their clavicles are moderately long, but they scale to body mass in the same manner as other humans (36). Some, but not all, Neandertals have broad pelves (37), but so apparently do all archaic Homo (38⇓–40). Neandertal (and earlier Homo) femoral diaphyseal shape has suggested more mediolateral and less anteroposterior loading during locomotion; however, properly scaled, there is no difference through Pleistocene Homo in anteroposterior femoral strength, only mediolateral variation apparently related to pelvic breadth (41, 42). Related are issues of head balance and momentum in running with a prognathic and platycephalic cranium (43); however, the Neandertals had shorter faces than their Pleistocene Homo predecessors (44, 45), and their relative neurocranial heights are similar to all earlier Homo and overlap the variation of early modern humans (46). Given the contrast in labyrinth morphology between earlier Pleistocene Homo and the Neandertals, it is therefore difficult to account for the latter’s (and Xujiayao 15’s) labyrinth configuration from postcranial and overall cranial proportions.

It has also been suggested that labyrinthine variation may be influenced by cranial base configuration, especially as related to posterior cranial fossa proportions (1), and this interpretation has been supported by some covariation with cranial base proportions among recent humans (47). Neandertals have relatively small posterior cranial fossae, yet the same proportions are in at least one early modern human with a modern labyrinth configuration (Cro-Magnon 1) (48). It is possible that other neurocranial considerations, related to cerebral expansion on an archaic Homo cranial base (49), given similar endocranial capacity and encephalization across all Late Pleistocene humans (50), are related to the labyrinthine morphology, but that remains to be evaluated. The endocranial capacity of Xujiayao 15 is unknown, but another individual from the same sample (Xujiayao 6) has an estimated cranial capacity ∼1,700 cc, among the larger Late Pleistocene crania; the earlier East Asian crania providing labyrinthine details have substantially smaller neurocrania (Hexian 1: 1,025 cc and Lantian 1: ∼780 cc) (51, 52), making it unlikely that the petrous angular similarities between Xujiayao 15 and Hexian 1 are due to brain size.

It is therefore unclear whether the Neandertal labyrinthine configuration, and by extension that of Xujiayao 15, can be attributed to other aspects of their biology. Moreover, the overall cranial morphology of the Xujiayao sample is incompletely known and their postcrania are unknown.

These considerations of the Xujiayao 15 labyrinth also raise questions regarding the use of individual features (whether morphological or molecular), in the absence of relatively complete paleontological remains, to identify the presence of Neandertals in regions outside of their well-documented core area of western Eurasia (west of ∼45° E, versus ∼70° E for Obi-Rakhmat and ∼114° E for Xujiayao). Indeed, many of the features so used are generally ancestral for later Homo (if poorly documented other than among the Neandertals), of unknown phylogenetic polarity, of uncertain primacy in morphological integration, and/or isolated features in otherwise non-Neandertal remains. The Xujiayao 15 labyrinthine morhology, and other features documented for eastern Eurasian and the earlier Pleistocene Homo (e.g., 16, 19, 20, 28, 36, 39, 42, 53), suggest that many of the purported “Neandertal” features were widespread during the Pleistocene, albeit often occurring in higher frequencies among the Neandertals.

Regardless of the ultimate populational and morphofunctional implications of the Xujiayao 15 and the other East Asian labyrinthine configurations, they serve to further confirm the general modern human pattern as the ancestral Homo one and to document that the apparently derived Neandertal configuration was not unique to western Eurasia. Given its presence in Europe, southwest Asia, central Asia, and now eastern Asia, it may well have been present across Eurasia during the later Pleistocene among archaic humans. As such, whether a discrete feature in itself or a secondary consequence of other aspects of morphological variation, it becomes less of a “Neandertal” marker and of broader relevance to the paleobiology of later Pleistocene Homo.