Conclusions: implications of the Liang Bua excavations for hominin evolution and biogeography

Excavations at Liang Bua, on the Indonesian island of Flores, have yielded a stratified sequence of stone artifacts and faunal remains spanning the last 95 k.yr., which includes the skeletal remains of two human species, Homo sapiens in the Holocene and Homo floresiensis in the Pleistocene. This paper summarizes and focuses on some of the evidence for Homo floresiensis in context, as presented in this Special Issue edition of the Journal of Human Evolution and elsewhere. Attempts to dismiss the Pleistocene hominins (and the type specimen LB1 in particular) as pathological pygmy humans are not compatible with detailed analyses of the skull, teeth, brain endocast, and postcranium. We initially concluded that H. floresiensis may have evolved by insular dwarfing of a larger-bodied hominin species over 880 k.yr. or more. However, recovery of additional specimens and the numerous primitive morphological traits seen throughout the skeleton suggest instead that it is more likely to be a late representative of a small-bodied lineage that exited Africa before the emergence of Homo erectus sensu lato. Homo floresiensis is clearly not an australopithecine, but does retain many aspects of anatomy (and perhaps behavior) that are probably plesiomorphic for the genus Homo. We also discuss some of the other implications of this tiny, endemic species for early hominin dispersal and evolution (e.g., for the “Out of Africa 1” paradigm and more specifically for colonizing Southeast Asia), and we present options for future research in the region.     

Body size and its consequences: Allometry and the lower limb length of Liang Bua 1 (Homo floresiensis)

Bivariate femoral length allometry in recent humans, Pan, and Gorilla is investigated with special reference to the diminutive Liang Bua (LB) 1 specimen (the holotype of Homo floresiensis) and six early Pleistocene femora referred to the genus Homo. Relative to predicted body mass, Pan and Gorilla femora show strong negative length allometry while recent human femora evince isometry to positive allometry, depending on sample composition and line-fitting technique employed. The allometric trajectories of Pan and Homo show convergence near the small body size range of LB 1, such that LB 1 manifests a low percentage deviation (d_yx of Smith [1980]) from the Pan allometric trajectory and falls well within the 95% confidence limits around the Pan individuals (but also outside the 95% confidence limits for recent Homo). In contrast, the six early Pleistocene Homo femora, belonging to larger individuals, show much greater d_yx values from both Pan and Gorilla and fall well above the 95% confidence limits for these taxa. All but one of these Pleistocene Homo specimens falls within the 95% confidence limits of the recent human sample. Similar results are obtained when femoral length is regressed on femoral head diameter in unlogged bivariate space. Regardless of the ultimate taxonomic status of LB 1, these findings are consistent with a prediction made by us (Franciscus and Holliday, 1992) that hominins in the small body size range of A.L. 288-1 (“Lucy”), including members of the genus Homo, will tend to possess short, ape-like lower limbs as a function of body size scaling.     

Descriptions of the upper limb skeleton of Homo floresiensis

Several bones of the upper extremity were recovered during excavations of Late Pleistocene deposits at Liang Bua, Flores, and these have been attributed to Homo floresiensis. At present, these upper limb remains have been assigned to six different individuals - LB1, LB2, LB3, LB4, LB5, and LB6. Several of these bones are complete or nearly so, but some are quite fragmentary. All skeletal remains recovered from Liang Bua were extremely fragile, but have now been stabilized and hardened in the laboratory in Jakarta. They are now curated in museum-quality containers at the National Research and Development Centre for Archaeology in Jakarta, Indonesia. These skeletal remains are described and illustrated photographically. The upper limb presents a unique mosaic of derived (human-like) and primitive morphologies, the combination of which is never found in either healthy or pathological modern humans.     

LB1 and LB6 Homo floresiensis are not modern human (Homo sapiens) cretins

Excavations in the late Pleistocene deposits at Liang Bua cave, Flores, have uncovered the skeletal remains of several small-bodied and small-brained hominins in association with stone artefacts and the bones of Stegodon. Due to their combination of plesiomorphic, unique and derived traits, they were ascribed to a new species, Homo floresiensis, which, along with Stegodon, appears to have become extinct ∼17 ka (thousand years ago). However, recently it has been argued that several characteristics of H. floresiensis were consistent with dwarfism and evidence of delayed development in modern human (Homo sapiens) myxoedematous endemic (ME) cretins. This research compares the skeletal and dental morphology in H. floresiensis with the clinical and osteological indicators of cretinism, and the traits that have been argued to be associated with ME cretinism in LB1 and LB6. Contrary to published claims, morphological and statistical comparisons did not identify the distinctive skeletal and dental indicators of cretinism in LB1 or LB6 H. floresiensis. Brain mass, skeletal proportions, epiphyseal union, orofacial morphology, dental development, size of the pituitary fossa and development of the paranasal sinuses, vault bone thickness and dimensions of the hands and feet all distinguish H. floresiensis from modern humans with ME cretinism. The research team responsible for the diagnosis of ME cretinism had not examined the original H. floresiensis skeletal materials, and perhaps, as a result, their research confused taphonomic damage with evidence of disease, and thus contained critical errors of fact and interpretation. Behavioural scenarios attempting to explain the presence of cretinous H. sapiens in the Liang Bua Pleistocene deposits, but not unaffected H. sapiens, are both unnecessary and not supported by the available archaeological and geochronological evidence from Flores.     

Preface: research at Liang Bua, Flores, Indonesia

Excavations at Liang Bua, Flores, Indonesia, have yielded evidence for an endemic human species, Homo floresiensis, a population that occupied the cave between ∼95-17 ka. This discovery has major implications for early hominin evolution and dispersal in Africa and Asia, attracting worldwide interest. This preface describes the rationale for the excavations in historical, geographical, and wider research contexts, as well as the methods used. It also introduces the other papers on aspects of Liang Bua research that feature in this edition of the Journal of Human Evolution.     

Size, shape, and asymmetry in fossil hominins: the status of the LB1 cranium based on 3D morphometric analyses

The unique set of morphological characteristics of the Liang Bua hominins (Homo floresiensis) has been attributed to explanations as diverse as insular dwarfism and pathological microcephaly. This study examined the relationship between cranial size and shape across a range of hominin and African ape species to test whether or not cranial morphology of LB1 is consistent with the basic pattern of static allometry present in these various taxa. Correlations between size and 3D cranial shape were explored using principal components analysis in shape space and in Procrustes form space. Additionally, patterns of static allometry within both modern humans and Plio-Pleistocene hominins were used to simulate the expected cranial shapes of each group at the size of LB1. These hypothetical specimens were compared to LB1 both visually and statistically. Results of most analyses indicated that LB1 best fits predictions for a small specimen of fossil Homo but not for a small modern human. This was especially true for analyses of neurocranial landmarks. Results from the whole cranium were less clear about the specific affinities of LB1, but, importantly, demonstrated that aspects of facial morphology associated with smaller size converge on modern human morphology. This suggests that facial similarities between LB1 and anatomically modern humans may not be indicative of a close relationship. Landmark data collected from this study were also used to test the degree of cranial asymmetry in LB1. These comparisons indicated that the cranium is fairly asymmetrical, but within the range of asymmetry exhibited by modern humans and all extant African ape species. Compared to other fossil specimens, the degree of asymmetry in LB1 is moderate and readily explained by the taphonomic processes to which all fossils are subject. Taken together, these findings suggest that H. floresiensis was most likely the diminutive descendant of a species of archaic Homo, although the details of this evolutionary history remain obscure.     

Homo floresiensis Contextualized: A Geometric Morphometric Comparative Analysis of Fossil and Pathological Human Samples

The origin of hominins found on the remote Indonesian island of Flores remains highly contentious. These specimens may represent a new hominin species, Homo floresiensis, descended from a local population of Homo erectus or from an earlier (pre-H. erectus) migration of a small-bodied and small-brained hominin out of Africa. Alternatively, some workers suggest that some or all of the specimens recovered from Liang Bua are pathological members of a small-bodied modern human population. Pathological conditions proposed to explain their documented anatomical features include microcephaly, myxoedematous endemic hypothyroidism (“cretinism”) and Laron syndrome (primary growth hormone insensitivity). This study evaluates evolutionary and pathological hypotheses through comparative analysis of cranial morphology. Geometric morphometric analyses of landmark data show that the sole Flores cranium (LB1) is clearly distinct from healthy modern humans and from those exhibiting hypothyroidism and Laron syndrome. Modern human microcephalic specimens converge, to some extent, on crania of extinct species of Homo. However in the features that distinguish these two groups, LB1 consistently groups with fossil hominins and is most similar to H. erectus. Our study provides further support for recognizing the Flores hominins as a distinct species, H. floresiensis, whose affinities lie with archaic Homo.     

Phylogenetic analysis of the calvaria of Homo floresiensis

Because until 2006 the Liang Bua human fossil remains were not available to the entire paleoanthropological community, the taxonomic position of Homo floresiensis was only a matter of opinion in publications. From the beginning, two schools of thought prevailed, and this situation persists today. One purports that the Liang Bua human series belongs to a local modern human (Homo sapiens sapiens) with anatomical particularities or pathologies that may be due to insular isolation/endogamy. The second argues in favour of the existence of a new species that, depending on the authors, is either a descendant of local Homo erectus, or belongs to a much more basal taxon, closer to archaic Homo or to australopithecines. Because there are no postcranial remains confidently attributed to Homo erectus in the fossil record, and because the Homo erectus type specimen is a single and partial calvaria, a cladistic analysis was undertaken using both nonmetric morphological features and metrics of the calvariae of human fossil specimens including LB1 to test if it belongs to this taxon. Our results indicate that LB1 is included in the Homo erectus clade.     

The Liang Bua faunal remains: a 95 k.yr. sequence from Flores, East Indonesia

Excavations at Liang Bua, a limestone cave on the island of Flores, East Indonesia, have yielded a well-dated archaeological and faunal sequence spanning the last 95 k.yr., major climatic fluctuations, and two human species - H. floresiensis from 95 to 17 k.yr.¹, and modern humans from 11 k.yr. to the present. The faunal assemblage comprises well-preserved mammal, bird, reptile and mollusc remains, including examples of island gigantism in small mammals and the dwarfing of large taxa. Together with evidence from Early-Middle Pleistocene sites in the Soa Basin, it confirms the long-term isolation, impoverishment, and phylogenetic continuity of the Flores faunal community. The accumulation of Stegodon and Komodo dragon remains at the site in the Pleistocene is attributed to Homo floresiensis, while predatory birds, including an extinct species of owl, were largely responsible for the accumulation of the small vertebrates. The disappearance from the sequence of the two large-bodied, endemic mammals, Stegodon florensis insularis and Homo floresiensis, was associated with a volcanic eruption at 17 ka and precedes the earliest evidence for modern humans, who initiated use of mollusc and shell working, and began to introduce a range of exotic animals to the island. Faunal introductions during the Holocene included the Sulawesi warty pig (Sus celebensis) at about 7 ka, followed by the Eurasian pig (Sus scrofa), Long-tailed macaque, Javanese porcupine, and Masked palm civet at about 4 ka, and cattle, deer, and horse - possibly by the Portuguese within historic times. The Holocene sequence at the site also documents local faunal extinctions - a result of accelerating human population growth, habitat loss, and over-exploitation.     

Homo floresiensis: a cladistic analysis

The announcement of a new species, Homo floresiensis, a primitive hominin that survived until relatively recent times is an enormous challenge to paradigms of human evolution. Until this announcement, the dominant paradigm stipulated that: 1) only more derived hominins had emerged from Africa, and 2) H. sapiens was the only hominin since the demise of Homo erectus and Homo neanderthalensis. Resistance to H. floresiensis has been intense, and debate centers on two sets of competing hypotheses: 1) that it is a primitive hominin, and 2) that it is a modern human, either a pygmoid form or a pathological individual. Despite a range of analytical techniques having been applied to the question, no resolution has been reached. Here, we use cladistic analysis, a tool that has not, until now, been applied to the problem, to establish the phylogenetic position of the species. Our results produce two equally parsimonious phylogenetic trees. The first suggests that H. floresiensis is an early hominin that emerged after Homo rudolfensis (1.86 Ma) but before H. habilis (1.66 Ma, or after 1.9 Ma if the earlier chronology for H. habilis is retained). The second tree indicates H. floresiensis branched after Homo habilis.     

A SINGLE LINEAGE IN EARLY PLEISTOCENE HOMO: SIZE VARIATION CONTINUITY IN EARLY PLEISTOCENE HOMO CRANIA FROM EAST AFRICA AND GEORGIA

The relationship between Homo habilis and early African Homo erectus has been contentious because H. habilis was hypothesized to be an evolutionary stage between Australopithecus and H. erectus, more than a half‐century ago. Recent work re‐dating key African early Homo localities and the discovery of new fossils in East Africa and Georgia provide the opportunity for a productive re‐evaluation of this topic. Here, we test the hypothesis that the cranial sample from East Africa and Georgia represents a single evolutionary lineage of Homo spanning the approximately 1.9–1.5 Mya time period, consisting of specimens attributed to H. habilis and H. erectus. To address issues of small sample sizes in each time period, and uneven representation of cranial data, we developed a novel nonparametric randomization technique based on the variance in an index of pairwise difference from a broad set of fossil comparisons. We fail to reject the hypothesis of a single lineage this period by identifying a strong, time‐dependent pattern of variation throughout the sequence. These results suggest the need for a reappraisal of fossil evidence from other regions within this time period and highlight the critical nature of the Plio‐Pleistocene boundary for understanding the early evolution of the genus Homo.     

Unique Dental Morphology of Homo floresiensis and Its Evolutionary Implications

Homo floresiensis is an extinct, diminutive hominin species discovered in the Late Pleistocene deposits of Liang Bua cave, Flores, eastern Indonesia. The nature and evolutionary origins of H. floresiensis’ unique physical characters have been intensively debated. Based on extensive comparisons using linear metric analyses, crown contour analyses, and other trait-by-trait morphological comparisons, we report here that the dental remains from multiple individuals indicate that H. floresiensis had primitive canine-premolar and advanced molar morphologies, a combination of dental traits unknown in any other hominin species. The primitive aspects are comparable to H. erectus from the Early Pleistocene, whereas some of the molar morphologies are more progressive even compared to those of modern humans. This evidence contradicts the earlier claim of an entirely modern human-like dental morphology of H. floresiensis, while at the same time does not support the hypothesis that H. floresiensis originated from a much older H. habilis or Australopithecus-like small-brained hominin species currently unknown in the Asian fossil record. These results are however consistent with the alternative hypothesis that H. floresiensis derived from an earlier Asian Homo erectus population and experienced substantial body and brain size dwarfism in an isolated insular setting. The dentition of H. floresiensis is not a simple, scaled-down version of earlier hominins.     

Body size and body shape in early hominins - implications of the Gona Pelvis

Discovery of the first complete Early Pleistocene hominin pelvis, Gona BSN49/P27, attributed to Homo erectus, raises a number of issues regarding early hominin body size and shape variation. Here, acetabular breadth, femoral head breadth, and body mass calculated from femoral head breadth are compared in 37 early hominin (6.0-0.26 Ma) specimens, including BSN49/P27. Acetabular and estimated femoral head sizes in the Gona specimen fall close to the means for non-Homo specimens (Orrorin tugenesis, Australopithecus africanus, Paranthropus robustus), and well below the ranges of all previously described Early and Middle Pleistocene Homo specimens. The Gona specimen has an estimated body mass of 33.2 kg, close to the mean for the non-Homo sample (34.1 kg, range 24-51.5 kg, n = 19) and far outside the range for any previously known Homo specimen (mean = 70.5 kg; range 52-82 kg, n = 17). Inclusion of the Gona specimen within H. erectus increases inferred sexual dimorphism in body mass in this taxon to a level greater than that observed here for any other hominin taxon, and increases variation in body mass within H. erectus females to a level much greater than that observed for any living primate species. This raises questions regarding the taxonomic attribution of the Gona specimen. When considered within the context of overall variation in body breadth among early hominins, the mediolaterally very wide Gona pelvis fits within the distribution of other lower latitude Early and Middle Pleistocene specimens, and below that of higher latitude specimens. Thus, ecogeographic variation in body breadth was present among earlier hominins as it is in living humans. The increased M-L pelvic breadth in all earlier hominins relative to modern humans is related to an increase in ellipticity of the birth canal, possibly as a result of a non-rotational birth mechanism that was common to both australopithecines and archaic Homo.     

Continuities in stone flaking technology at Liang Bua, Flores, Indonesia

This study examines trends in stone tool reduction technology at Liang Bua, Flores, Indonesia, where excavations have revealed a stratified artifact sequence spanning 95 k.yr. The reduction sequence practiced throughout the Pleistocene was straightforward and unchanging. Large flakes were produced off-site and carried into the cave where they were reduced centripetally and bifacially by four techniques: freehand, burination, truncation, and bipolar. The locus of technological complexity at Liang Bua was not in knapping products, but in the way techniques were integrated. This reduction sequence persisted across the Pleistocene/Holocene boundary with a minor shift favoring unifacial flaking after 11 ka. Other stone-related changes occurred at the same time, including the first appearance of edge-glossed flakes, a change in raw material selection, and more frequent fire-induced damage to stone artifacts. Later in the Holocene, technological complexity was generated by “adding-on” rectangular-sectioned stone adzes to the reduction sequence. The Pleistocene pattern is directly associated with Homo floresiensis skeletal remains and the Holocene changes correlate with the appearance of Homo sapiens. The one reduction sequence continues across this hominin replacement.     

Where are inion and endinion? Variations of the exo- and endocranial morphology of the occipital bone during hominin evolution

The occipital bone is frequently investigated in paleoanthropological studies because it has several features that help to differentiate various fossil hominin species. Among these features is the separation between inion and endinion, which has been proposed to be an autapomorphic trait in (Asian) Homo erectus. Methodologies are developed here to quantify for the first time the location of these anatomical points, and to interpret their variation due to the complex interactions between exocranial and endocranial size and shape of the occipital and nuchal planes, as well as the occipital lobes and cerebellum. On the basis of our analysis, neither ‘the separation between inion and endinion’ nor ‘endinion below inion’ can be considered as an autapomorphic trait in H. erectus, since this feature is a condition shared by extant African great apes and fossil hominins. Moreover, our results show that the exo- and endocranial anatomy of the occipital bone differs between hominins (except Paranthropus boisei specimens and KNM-ER 1805) and great apes. For example, chimpanzees and bonobos are characterized by a very high position of inion and their occipital bone shows an antero-posterior compression. However, these features are partly correlated with their small size when compared with hominins. Asian H. erectus specimens have a thick occipital torus, but do not differ from other robust specimens, neither in this feature nor in the analysed exo- and endocranial proportions of the occipital bone. Finally, the apparent brain size reduction during the Late Pleistocene and variation between the sexes in anatomically modern humans (AMH) reflect that specimens with smaller brains have a relatively larger posterior height of the cerebellum. However, this trend is not the sole explanation for the ‘vertical shift’ of endinion above inion that appears occasionally and exclusively in AMH.     

Mandibular molar root morphology in Neanderthals and Late Pleistocene and recent Homo sapiens

Neanderthals have a distinctive suite of dental features, including large anterior crown and root dimensions and molars with enlarged pulp cavities. Yet, there is little known about variation in molar root morphology in Neanderthals and other recent and fossil members of Homo. Here, we provide the first comprehensive metric analysis of permanent mandibular molar root morphology in Middle and Late Pleistocene Homo neanderthalensis, and Late Pleistocene (Aterian) and recent Homo sapiens. We specifically address the question of whether root form can be used to distinguish between these groups and assess whether any variation in root form can be related to differences in tooth function. We apply a microtomographic imaging approach to visualise and quantify the external and internal dental morphologies of both isolated molars and molars embedded in the mandible (n = 127). Univariate and multivariate analyses reveal both similarities (root length and pulp volume) and differences (occurrence of pyramidal roots and dental tissue volume proportion) in molar root morphology among penecontemporaneous Neanderthals and Aterian H. sapiens. In contrast, the molars of recent H. sapiens are markedly smaller than both Pleistocene H. sapiens and Neanderthals, but share with the former the dentine volume reduction and a smaller root-to-crown volume compared with Neanderthals. Furthermore, we found the first molar to have the largest average root surface area in recent H. sapiens and Neanderthals, although in the latter the difference between M₁ and M₂ is small. In contrast, Aterian H. sapiens root surface areas peak at M₂. Since root surface area is linked to masticatory function, this suggests a distinct occlusal loading regime in Neanderthals compared with both recent and Pleistocene H. sapiens.     

Dental microwear texture analysis of hominins recovered by the Olduvai Landscape Paleoanthropology Project, 1995-2007

Dental microwear analysis has proven to be a valuable tool for the reconstruction of aspects of diet in early hominins. That said, sample sizes for some groups are small, decreasing our confidence that results are representative of a given taxon and making it difficult to assess within-species variation. Here we present microwear texture data for several new specimens of Homo habilis and Paranthropus boisei from Olduvai Gorge, bringing sample sizes for these species in line with those published for most other early hominins. These data are added to those published to date, and microwear textures of the enlarged sample of H. habilis (n = 10) and P. boisei (n = 9) are compared with one another and with those of other early hominins. New results confirm that P. boisei does not have microwear patterns expected of a hard-object specialist. Further, the separate texture complexity analyses of early Homo species suggest that Homo erectus ate a broader range of foods, at least in terms of hardness, than did H. habilis, P. boisei, or the “gracile” australopiths studied. Finally, differences in scale of maximum complexity and perhaps textural fill volume between H. habilis and H. erectus are noted, suggesting further possible differences between these species in diet.     

Nacurrie 1: Mark of ancient Java, or a caring mother’s hands, in terminal Pleistocene Australia?

There has been a protracted debate over the evidence for intentional cranial modification in the terminal Pleistocene Australian crania from Kow Swamp and Coobool Creek. Resolution of this debate is crucial to interpretations of the significance of morphological variation within terminal Pleistocene-early Holocene Australian skeletal materials and claims of a regional evolutionary sequence linking Javan Homo erectus and Australian Homo sapiens. However, morphological comparisons of terminal Pleistocene and recent Australian crania are complicated by the significantly greater average body mass in the former. Raw and size-adjusted metric comparisons of the terminal Pleistocene skeleton from Nacurrie, south-eastern Australia, with modified and unmodified H. sapiens and H. erectus, identified a suite of traits in the frontal, parietal, and occipital bones associated with intentional modification of a neonate’s skull. These traits are also present in some of the crania from Kow Swamp and Coobool Creek, which are in close geographic proximity to Nacurrie, but not in unmodified H. sapiens or Javan H. erectus. Frontal bone morphology in H. erectus was distinct from all of the Australian H. sapiens samples. During the first six months of life, Nacurrie’s vault may have been shaped by his mother’s hands, rather than though the application of fixed bandages. Whether this behaviour persisted only for several generations, or hundreds of years, remains unknown. The reasons behind the shaping of Nacurrie’s head, aesthetics or otherwise, and why this cultural practice was adopted and subsequently discontinued, will always remain a matter of speculation.