Human Evolution

Human evolution is the biological and behavioral history of the lineage that includes modern humans, Homo sapiens, and their extinct relatives following its divergence from the lineage leading to chimpanzees and bonobos. Humans are primates and members of the great-ape family; they share ancestors with other living apes rather than descending from any living ape species. Human origins involved branching lineages, the coexistence of different species, and the gradual emergence of characteristics such as habitual upright walking, enlarged brains, and increasingly complex technologies and social behavior. These changes did not occur together or form an inevitable progression toward modern humans. (humanorigins.si.edu)

Fossils, archaeological remains, and genetic evidence establish an African origin for the human lineage and for Homo sapiens, whose early fossil record extends back approximately 300,000 years. Subsequent migrations carried our species across the world, while interbreeding with Neanderthals, Denisovans, and other ancient populations contributed to its ancestry. Research combines paleontology, archaeology, comparative anatomy, and population genetics to reconstruct this history and distinguish well-supported findings from uncertainties about ancestry, species boundaries, and population relationships. (nature.com)

Understanding human evolution explains humanity’s place within evolution and the historical origins of its distinctive combination of anatomy, behavior, and culture. Its study encompasses early hominins, the diversification of Homo, environmental change, global dispersal, and continuing interactions between biological inheritance and cultural practices. (humanorigins.si.edu)

Evolutionary Framework and Humanity’s Place among Primates

Human evolution encompasses the biological history of the human branch after its divergence from the lineage leading to chimpanzees and bonobos, including extinct relatives rather than only direct ancestors. This branch is broader than the genus Homo: it includes genera such as Ardipithecus, Australopithecus, and Paranthropus. Fossils attributed to Homo extend back approximately 2.8 million years, whereas Homo sapiens, the species to which all living humans belong, is documented from about 300,000 years ago. Carl Linnaeus named H. sapiens in 1758; that act of classification is distinct from identifying when the species evolved. The human lineage, the genus Homo, and H. sapiens therefore denote different, nested scopes of evolutionary history. (australian.museum)

In biological classification, humans belong to the mammalian order Primates, which also includes lemurs, lorises, tarsiers, monkeys, and other apes. Within Primates, apes form the superfamily Hominoidea: gibbons and siamangs belong to Hylobatidae, while humans, orangutans, gorillas, chimpanzees, and bonobos belong to the great-ape family Hominidae. A hominid is a member of Hominidae and thus need not be human or especially close to the human branch. In customary paleoanthropological usage, a hominin belongs to the human branch after its separation from the chimpanzee–bonobo lineage. Older publications frequently used hominid for this narrower group, so terminology must be interpreted in its historical context. These classifications express ancestry, not a ranking of intelligence or anatomical complexity. (australian.museum)

Chimpanzees (Pan troglodytes) and bonobos (Pan paniscus) are humanity’s closest living relatives. Their lineage and the human lineage descend from a shared ancestral population, with the divergence commonly placed approximately 6–8 million years ago; this is an estimated interval, not a precisely dated event. Living chimpanzees and bonobos are not surviving ancestral stages through which humans passed. Each lineage has undergone its own evolutionary changes since separation, and the common ancestor cannot simply be assumed to have possessed the anatomy or behavior of a modern chimpanzee. Comparisons among living apes help identify shared inherited features, but reconstructing that ancestor also requires fossil evidence. (humanorigins.si.edu)

The mechanisms governing human evolution are the same as those operating in other organisms. Heritable mutations alter DNA and generate new variants; they do not arise because a population needs a particular adaptation. Natural selection changes the representation of heritable variants when their carriers differ consistently in reproductive success under particular conditions. Genetic drift changes allele frequencies through chance sampling between generations, with especially strong effects in small populations; a variant can disappear or become fixed without conferring an advantage. Gene flow transfers variants between populations when migrants reproduce or populations interbreed. Mutation supplies new variation, while selection, drift, and gene flow influence its distribution, often acting simultaneously rather than as competing explanations. (pressbooks.umn.edu)

Human evolutionary relationships are consequently represented by branching trees, not a ladder progressing from “ape” to “human.” Splitting lineages can coexist, and extinction can eliminate entire branches: a fossil species is not necessarily a direct ancestor merely because it is older than H. sapiens. Traits also accumulated in a mosaic rather than as one coordinated transformation. Upright walking appeared well before the major expansion of brain size, and early hominins combined terrestrial bipedalism with retained climbing abilities. Between at least 6 and 3 million years ago, locomotor anatomy included combinations of apelike and humanlike features. Such combinations are evidence of differing evolutionary histories of individual traits, not incomplete steps toward a predetermined endpoint. (humanorigins.si.edu)

Fossils, Archaeological Evidence, and Dating

Skeletal fossils reveal anatomy: skulls preserve evidence of braincase size and facial form, while limb bones and joints constrain reconstructions of locomotion. Teeth provide evidence of development, chewing adaptations, and wear. Archaeological remains—including stone tools and modified animal bones—document activities rather than simply bodily structure. Footprints record actual movement: the Laetoli trails in Tanzania, discovered in 1978 by Paul Abell during Mary Leakey’s field project, preserve hominin walking in wet volcanic ash approximately 3.6 million years ago. Such evidence is complementary, but proximity alone does not identify a toolmaker or demonstrate that a fossil and an artifact were deposited simultaneously; their sedimentary context must establish the association. (humanorigins.si.edu)

Excavation preserves that context by documenting finds’ positions, sedimentary units, and relationships before removal. Stratigraphy establishes relative order: in an undisturbed sequence, lower deposits generally precede upper ones, although erosion, burrowing, and redeposition can complicate the sequence. Taphonomy examines the processes between death or trace formation and discovery. Scavenging, decomposition, abrasion, mineral dissolution, and mineral infilling determine what survives and whether remains retain their original positions. Teeth and mineralized skeletal parts survive more commonly than soft tissues; consequently, the record disproportionately represents certain anatomical structures. Geographical sampling is also uneven: preservation, burial, and subsequent exposure make some sedimentary basins and caves productive, while other occupied landscapes leave little recoverable evidence. Absence of fossils therefore cannot automatically establish absence of a population. (humanorigins.si.edu)

Radiometric dating uses radioactive decay, but different clocks date different materials and events. Radiocarbon measures remaining $^{14}\mathrm{C}$ in once-living material, such as charcoal or preserved bone collagen, to estimate elapsed time since carbon exchange ceased. Its half-life is approximately 5,730 years, and its practical range usually extends to about 50,000 years—not the millions of years represented by early hominin fossils. Modern-carbon contamination can make an old sample appear substantially younger. Because atmospheric radiocarbon concentrations varied, laboratory ages require calibration against independently dated records. Paula Reimer and colleagues’ 2020 IntCal20 Northern Hemisphere calibration curve extends to 55,000 calendar years before present; radiocarbon “present” conventionally means 1950. (pubs.usgs.gov)

Potassium–argon dating measures the accumulation of $^{40}\mathrm{Ar}$ produced by $^{40}\mathrm{K}$ decay in potassium-bearing minerals after cooling and argon retention. It is particularly useful for volcanic deposits hundreds of thousands to millions of years old, although suitable rocks permit a wider range. Argon–argon dating irradiates a sample to convert $^{39}\mathrm{K}$ into $^{39}\mathrm{Ar}$, allowing argon-isotope measurements against an irradiated reference standard. Single-crystal analysis helps detect older crystals incorporated into younger eruptions. These methods commonly date volcanic ash above and below fossil-bearing deposits, rather than the bones themselves: the underlying ash supplies an older bound and the overlying ash a younger bound, provided their stratigraphic relationships are intact. (pubs.usgs.gov)

Uranium-series methods, especially uranium–thorium dating, measure disequilibrium within a decay chain as $^{230}\mathrm{Th}$ accumulates from uranium. They date carbonate formation in cave flowstones, stalagmites, and related deposits, typically from a few centuries to approximately 500,000 years. Carbonate beneath or covering archaeological sediment can bracket its age; contamination and subsequent uranium or thorium movement can invalidate the clock. Luminescence dating instead measures radiation-induced charge trapped in minerals. Optically stimulated luminescence usually estimates when sediment grains were last sufficiently exposed to daylight before burial; thermoluminescence can date the last strong heating of flint. Applicable ranges extend from recent centuries to several hundred thousand years, depending on mineral properties, environmental radiation, and signal saturation. (historicengland.org.uk)

A direct date measures material belonging to the specimen or artifact; an associated date measures another object or deposit whose relationship must be demonstrated. In 2017, Daniel Richter and colleagues reported a weighted thermoluminescence age of $315 \pm 34$ thousand years for heated flints associated with human fossils at Jebel Irhoud, Morocco—not a direct thermoluminescence date on the bones. Dates should specify the sampled material, dated event, method, and uncertainty, including the confidence level represented by an interval. Stratigraphic uncertainty remains separate from laboratory precision. Because preservation and discovery are incomplete, the earliest securely dated occurrence establishes that a lineage or behavior existed by that time, not that it originated then. (nature.com)

Genetic Reconstruction of Human Ancestry

Comparative genomics reconstructs relationships by aligning corresponding regions of human and other primate genomes, identifying shared changes, and estimating genealogical relationships across loci. The Chimpanzee Sequencing and Analysis Consortium’s 2005 draft comparison catalogued approximately 35 million single-nucleotide differences and five million insertion or deletion events relative to humans. Its reported mean nucleotide difference of 1.23% concerned compared nucleotide positions, not every category of genomic difference. Comparisons also use other apes as outgroups to distinguish ancestral from derived variants; sequence similarity alone neither identifies the anatomical effects of particular changes nor supplies a complete evolutionary chronology. (nature.com)

Different genomic compartments preserve different inheritance histories. Mitochondrial DNA, carried in mitochondria, is normally maternally inherited and effectively behaves as a single, non-recombining genealogy. The non-recombining portion of the Y chromosome traces paternal transmission; its recombining regions are excluded from such lineage reconstructions. Autosomes are inherited from both parents and reshuffled by recombination, so their segments contain numerous genealogies rather than one family tree. Consequently, mitochondrial and Y-chromosomal histories can differ from each other and from genome-wide history because of lineage loss, unequal reproductive success, migration, and sampling. Genome-wide evidence is therefore essential for distinguishing a marker’s history from that of a population. (nature.com)

A molecular clock converts accumulated sequence changes into elapsed time using an estimated mutation or substitution rate. Calibration may use fossil constraints, directly observed mutations in parent–offspring pedigrees, or dated ancient sequences. In 2012, Augustine Kong and colleagues sequenced 78 Icelandic parent–offspring trios and estimated an average mutation rate of $1.20 \times 10^{-8}$ per nucleotide per generation; mutation counts also increased with paternal age. Converting generations into years therefore requires assumptions about ancestral generation lengths and reproductive ages. Under coalescent theory, two sampled gene copies trace backward to a shared ancestral copy. This genetic coalescence can predate the separation of their populations because both copies already varied within the ancestral population. Neither a gene-coalescence date nor a population-divergence estimate is equivalent to the first fossil appearance of a species. (nature.com)

Ancient DNA provides sequences from past individuals rather than reconstructing ancestry solely from living populations. Researchers extract short fragments from bones, teeth, or other preserved material, attach sequencing adapters, and perform DNA sequencing, sometimes enriching selected regions before sequencing. Authentication combines fragment-length distributions, characteristic chemical damage, and contamination estimates. Cytosine deamination produces excess apparent cytosine-to-thymine substitutions near fragment ends, a useful signal of antiquity but not sufficient authentication by itself. Dedicated clean laboratories, extraction and library blanks, and computational contamination tests help distinguish endogenous DNA from modern human and microbial material. Preservation depends strongly on temperature and burial chemistry: warm environments generally accelerate degradation, while even well-preserved specimens may yield predominantly microbial DNA. These constraints make the geographical and chronological coverage of ancient genomes uneven. (nature.com)

Demographic reconstruction fits models to allele frequencies, heterozygosity, linked variants, and the distribution of inferred coalescence times. In 2011, Heng Li and Richard Durbin introduced the pairwise sequentially Markovian coalescent method, using variation along a single diploid genome to estimate changes in effective population size over approximately 10,000 to one million years. This quantity describes genetic drift and coalescence, not a direct census of people. Ancestral subdivision and changing gene flow can produce signals resembling population growth or decline even without equivalent changes in total population size. Sequence reads and measured variants are observations; inferred bottlenecks, migration histories, and population sizes remain conditional on mutation rates, sampling, and demographic assumptions. (nature.com)

The mitochondrial and Y-chromosomal most recent common ancestors were ancestors of particular surviving genetic lineages, not humanity’s only contemporaries. In a 2013 analysis, G. David Poznik and colleagues estimated Y-chromosomal coalescence at 120,000–156,000 years ago and mitochondrial coalescence at 99,000–148,000 years ago under their sampling and calibration. Such estimates can change when additional divergent lineages are sampled. Other people living in those generations could contribute autosomal ancestry even if their uninterrupted maternal or paternal lines later disappeared. The two common ancestors need not have lived simultaneously, met, or constituted a founding couple. (pmc.ncbi.nlm.nih.gov)

Environmental Settings and Ecological Change

The environmental record relevant to human origins spans the late Miocene, approximately 11.63–5.333 million years ago; the Pliocene, 5.333–2.58 million years ago; the Pleistocene, 2.58 million–11,700 years ago; and the Holocene, beginning 11,700 years ago. These divisions provide a consistent geological chronology, not boundaries between evolutionary stages. Across this interval, marine oxygen-isotope records document long-term cooling and increasingly pronounced fluctuations in temperature and continental ice volume. African environments, however, did not respond uniformly: regional rainfall and vegetation depended on monsoon circulation, topography, and other processes as well as global glacial conditions. (stratigraphy.org)

African hominins occupied landscapes containing different combinations of woodland, forest, grassland, river corridors, and lake-margin habitats, rather than one universal open savanna. The proportions and arrangement of these environments changed across space and time. A 2011 study by Thure Cerling and colleagues used carbon-isotope measurements from approximately 1,300 ancient soils in the Awash and Omo–Turkana basins to reconstruct woody cover over six million years. Most sampled fossil-site environments had less than approximately 40% woody cover, indicating widespread open landscapes but not uniformly treeless grassland. Conversely, a 2009 investigation of mammals associated with Ardipithecus ramidus at Aramis, Ethiopia, inferred predominantly woodland conditions from animal anatomy, dental wear, and isotopes. Local habitat reconstructions and basin-wide averages therefore describe different spatial scales and need not coincide. (nature.com)

Paleoclimatology combines independent proxies because no single measurement reconstructs an entire environment. Lake sediments record changes in water depth, oxygenation, and sediment supply: at Chew Bahir in southern Ethiopia, clay-rich sediments associated with poorly ventilated lake-bottom water help identify wetter intervals. Stable carbon isotopes in ancient soils distinguish contributions from vegetation using different photosynthetic pathways—especially woody plants using the $C_3$ pathway and tropical grasses using $C_4$—and can support estimates of woody cover. Tooth-enamel carbon isotopes instead principally record foods consumed by an animal, not vegetation cover directly. Associated mammals provide additional evidence through their locomotion, feeding adaptations, and ecological requirements. Marine foraminiferal oxygen isotopes supply a broader climatic context, but their signal combines ocean temperature and ice volume rather than directly measuring rainfall at an African fossil locality. (nature.com)

Habitat-specific hypotheses propose that particular conditions favored particular adaptations; the savanna hypothesis, for example, linked upright walking to expanding open habitats. Richard “Rick” Potts’s variability selection hypothesis instead proposes that repeated environmental shifts favored capacities usable across different conditions, such as flexible resource use and behavior. Its proposed mechanism is differential survival and reproduction when specialists encounter changing resources, whereas more versatile organisms retain access to alternatives. Neither environmental change itself nor chronological coincidence demonstrates this mechanism. Tests must establish the timing, magnitude, and ecological consequences of change and determine whether the proposed adaptation actually improved performance under variable conditions. Woodland-associated early hominins also prevent open grassland from being treated as a necessary setting for the initial emergence of bipedal adaptations. (humanorigins.si.edu)

A concrete landscape-scale test comes from the region around Olorgesailie, Kenya. In a 2020 study, Potts and colleagues reconstructed environmental change from a sediment core approximately 24 kilometres from the archaeological sequence. Beginning around 400,000 years ago, tectonic, hydrological, and ecological changes disrupted a comparatively stable resource base, producing larger fluctuations in freshwater availability, grassland composition, and woody cover. Nearby deposits record major herbivore turnover and the replacement of Acheulean by Middle Stone Age technologies between approximately 500,000 and 300,000 years ago. The researchers proposed a resource-based explanation for greater behavioral flexibility, but the association does not isolate climate as the sole cause: tectonics and local drainage changes also altered the landscape. (digitalcommons.uri.edu)

Rainfall changes could alter population connectivity by opening or closing routes between suitable habitats. Reconstructions of Saharan paleolakes and rivers identify potential humid corridors during Marine Isotope Stage 5, approximately 130,000–80,000 years ago. Such waterways and vegetated terrain could facilitate dispersal and encounters between populations; renewed aridity could fragment their ranges and restrict opportunities for gene flow. A reconstructed corridor establishes ecological possibility, not proof that a particular population traversed it. Habitat loss could also increase extinction risk when populations could neither relocate nor exploit alternative resources, although disappearance from a local fossil sequence must not automatically be equated with species-wide extinction. (pmc.ncbi.nlm.nih.gov)

Earliest Proposed Hominins

Sahelanthropus tchadensis is known from Toros-Menalla in the Djurab Desert of northern Chad and dates to approximately 7–6 million years ago. A Franco-Chadian team led by Michel Brunet discovered its remains in 2001; the species was described in 2002. The principal specimen, the cranium TM 266-01-060-1, nicknamed “Toumaï,” combines a small braincase and prominent browridge with relatively small canines and a shortened middle face. Its reduced canines and the position and orientation of the foramen magnum—the opening through which the spinal cord enters the skull—were important grounds for assigning it to the human lineage. A more underside-facing opening suggests that the head balanced on a relatively upright trunk, although it does not by itself establish the frequency or mechanics of two-legged walking. (humanorigins.si.edu)

Postcranial remains attributed to Sahelanthropus include a partial femur and two ulnae recovered at the same locality, rather than an articulated skeleton associated with the cranium. In 2022, Guillaume Daver and colleagues interpreted the femur as evidence of habitual bipedalism and the ulnae as evidence of substantial climbing activity. These functional interpretations have been challenged. A 2026 analysis by Scott Williams and colleagues found that the limb bones resembled chimpanzees in overall size and shape but identified features supporting bipedal hip and knee function, including a femoral tubercle associated with attachment of the iliofemoral ligament. These studies concern incomplete bones: neither the precise walking gait nor the relationship of Sahelanthropus to other early lineages can be read directly from a complete locomotor skeleton. (doi.org)

Orrorin tugenensis, approximately 6 million years old, was discovered in 2000 in the Lukeino Formation of Kenya’s Tugen Hills by a team led by Brigitte Senut and Martin Pickford. Its remains include teeth, jaw fragments, and portions of limb bones; no complete cranium is known. Small teeth with relatively thick enamel contribute to its proposed hominin identification, but the strongest locomotor evidence comes from the proximal femur. The femoral neck and distribution of cortical bone have been interpreted as adaptations to loading during upright walking. Brian Richmond and William Jungers’s 2008 comparative analysis found that the femur most strongly resembled those of australopithecines, supporting bipedalism without supporting the original suggestion that Orrorin was especially close to Homo. Other remains indicate that climbing remained part of its locomotor repertoire. (humanorigins.si.edu)

Ardipithecus comprises two named species from Ethiopia. Ar. kadabba, dated to approximately 5.8–5.2 million years ago, was first discovered by Yohannes Haile-Selassie in the Middle Awash in 1997. Initially classified as a subspecies of Ar. ramidus, it received separate species status in 2004 following analysis of additional teeth, including differences in canine–premolar wear. Its limited postcranial record includes a toe bone interpreted as permitting bipedal push-off. However, that bone is younger than most of the dental material and was found approximately 10 miles away, limiting the certainty of its association and the reconstruction of habitual locomotion. (humanorigins.si.edu)

Ar. ramidus, approximately 4.4 million years old, is represented at Middle Awash and Gona, including the partial skeleton ARA-VP-6/500, or “Ardi,” discovered at Aramis in 1994 and extensively described in 2009. Its brain volume was approximately 300–350 cubic centimetres. The reconstructed pelvis combines a shortened upper portion interpreted as facilitating upright walking with features retained for movement in trees. The foot has a divergent, grasping great toe, alongside features interpreted as providing support during terrestrial propulsion. This combination does not reproduce either modern human walking anatomy or the entire locomotor pattern of living chimpanzees; comparative studies continue to debate the extent of its African-ape-like climbing adaptations. Crushed pelvic remains, missing skeletal regions, and uncertainty about which similarities reflect shared ancestry rather than independently evolved adaptations limit its placement in a detailed phylogenetic tree. Neither chronological succession nor a combination of reduced canines and possible bipedalism establishes that these named species form a confirmed chain of direct ancestors. (humanorigins.si.edu)

Australopithecines and Paranthropus

*Australopithecus* fossils document several African hominin species rather than a single successive ancestral population. A. anamensis is known from Kenya and Ethiopia approximately 4.2–3.8 million years ago; A. afarensis occupied eastern Africa, including Ethiopia, Kenya, and Tanzania, approximately 3.85–2.95 million years ago. In southern Africa, A. africanus is conventionally placed approximately 3.3–2.1 million years ago, although individual cave assemblages have dating uncertainties. A. sediba, discovered at Malapa, South Africa, in 2008 and described in 2010, dates to approximately 1.98 million years ago. These geographical and temporal distributions do not establish a direct ancestral chain: a 3.8-million-year-old A. anamensis cranium reported in 2019 indicates overlap with A. afarensis for at least 100,000 years, complicating a model in which the entire former species simply transformed into the latter. (humanorigins.si.edu)

Australopithecine anatomy combines habitual bipedal locomotion with features compatible with climbing. In A. anamensis, the tibia and ankle joint indicate weight-bearing during upright walking, whereas long forearms and wrist anatomy suggest continued arboreal activity. A. afarensis possessed a pelvis and lower limbs adapted to walking but retained long, strong arms and curved fingers. Its brain volume was generally below 500 cubic centimetres. A. africanus similarly combines bipedal adaptations in the pelvis, femur, and foot with climbing-related shoulder and hand features. These combinations show that terrestrial walking preceded major enlargement of the brain; they do not require either exclusively ground-dwelling behavior or a gait identical in every respect to that of living humans. (humanorigins.si.edu)

Lucy, the approximately 3.2-million-year-old A. afarensis partial skeleton found at Hadar, Ethiopia, in 1974 by Donald Johanson’s expedition, supplies associated bones through which body proportions and locomotion can be assessed together. The Laetoli trackways in Tanzania provide a different kind of evidence: actual walking traces preserved in approximately 3.6-million-year-old volcanic ash. In 1978, Paul Abell, working with Mary Leakey’s team, identified the celebrated trail containing about 70 hominin footprints over 27 metres. The aligned great toe and pattern of heel-to-toe weight transfer support upright walking. Their attribution to A. afarensis rests on the contemporary fossil record and anatomical compatibility, not on a skeleton physically associated with a trackmaker. (humanorigins.si.edu)

The Taung Child illustrates the importance of developmental evidence. Recovered in South Africa in 1924 and described by Raymond Dart as A. africanus in 1925, it preserves a juvenile face, jaw, teeth, and a natural cast of the braincase interior. Dental development indicates an age at death of approximately three years. Its forward-positioned foramen magnum—the opening through which the spinal cord enters the skull—supported an interpretation of upright posture despite the small braincase. Because it is immature, its cranial proportions must be distinguished from adult species characteristics; its teeth also provide evidence about maturation rather than merely adult chewing anatomy. (humanorigins.si.edu)

*Paranthropus*, often called the robust australopithecines, is distinguished principally by chewing specializations. P. aethiopicus inhabited eastern Africa approximately 2.7–2.3 million years ago, followed by P. boisei approximately 2.3–1.2 million years ago; P. robustus is known from South African sites. Large premolars and molars, thick enamel, substantial jaws, and flaring cheekbones characterize the group. In strongly crested skulls, the sagittal crest provided attachment for enlarged temporalis muscles. “Robust” therefore chiefly describes cranial and dental construction, not an evolutionary transition toward a uniformly larger body. P. boisei had especially enlarged cheek teeth, whereas P. aethiopicus retained a more projecting face and small braincase. (humanorigins.si.edu)

These structures indicate mechanical capacities, not a uniquely determined menu. Although P. boisei acquired the nickname “Nutcracker Man,” dental microwear does not show the heavy pitting expected from habitual hard-object consumption. Carbon isotope measurements instead indicate a diet dominated by $C_4$-derived resources in the sampled eastern African populations, consistent with substantial consumption of grasses or sedges, without identifying exact species or plant parts. P. robustus exhibits different dietary signals despite broadly similar chewing anatomy. Such differences, together with the anatomical mosaic of A. sediba and temporal overlap among taxa, constrain phylogenetic reconstruction: resemblance, specialization, and chronological succession alone cannot establish direct ancestry. (pmc.ncbi.nlm.nih.gov)

The Emergence and Early Diversification of Homo

The earliest widely accepted fossils attributed to Homo include LD 350-1, a partial lower jaw with teeth from Ledi-Geraru in Ethiopia’s Afar region, dated to 2.80–2.75 million years ago. Described by Brian Villmoare and colleagues in 2015, it combines australopithecine-like features with dental and mandibular characteristics found in later Homo. Because it preserves neither a braincase nor an associated skeleton, it cannot establish the brain size or body proportions of the earliest members of the genus. Its assignment rests on a combination of anatomical traits, rather than a single defining innovation. More generally, distinguishing early Homo from australopithecines involves tooth proportions, jaw architecture, facial shape, and cranial anatomy; these characteristics did not necessarily change together. The specimen documents the genus’s presence by this date, not its exact evolutionary origin. (pubmed.ncbi.nlm.nih.gov)

*Homo habilis*, generally dated to approximately 2.4–1.4 million years ago, was named in 1964 by Louis Leakey, Philip Tobias, and John Napier from fossils at Olduvai Gorge, Tanzania. Its name, meaning “handy man,” reflected the proposed association with stone tools found there. Compared with many australopithecines, specimens assigned to this species generally have a larger braincase and reduced facial projection, but retain substantial variation and some primitive anatomy, including relatively long arms. The approximately 1.8-million-year-old type specimen, OH 7, preserves a partial juvenile skull, lower jaw, and hand bones. A virtual reconstruction published by Fred Spoor and colleagues in 2015 revealed a long, narrow dental arcade resembling that of Australopithecus afarensis, rather than the more parabolic arrangement of later humans. (humanorigins.si.edu)

*Homo rudolfensis* is principally represented by cranial and jaw material from eastern Africa, especially the Koobi Fora region of Kenya. Its reference cranium, KNM-ER 1470, has an estimated cranial capacity of about 775 cubic centimetres and a large, relatively flat face; associated dental evidence indicates larger premolars and molars than in typical H. habilis. By contrast, KNM-ER 1813, commonly assigned to H. habilis, has a smaller face and a braincase of approximately 500 cubic centimetres. These differences have been interpreted either as variation within one species or as evidence of distinct species. Facial and mandibular fossils described by Meave Leakey and colleagues in 2012, dated to 1.95–1.78 million years ago, strengthened the case for a distinct 1470-like group. Nevertheless, incomplete specimens and the absence of securely identified H. rudolfensis postcranial remains constrain comparisons of body size and proportions. (humanorigins.si.edu)

Brain enlargement therefore provides neither a simple species boundary nor a complete account of the genus’s emergence. The reconstructed OH 7 braincase yielded an estimated volume of 729–824 cubic centimetres, overlapping estimates for H. rudolfensis and early H. erectus. Jaw shape could distinguish specimens whose brain sizes overlapped. Similarly, reduced teeth, larger bodies, and longer legs were not acquired as one coordinated package. Susan Antón, Richard Potts, and Leslie Aiello’s 2014 synthesis emphasized this mosaic pattern and interpreted early Homo as showing increased dietary and habitat flexibility. Such ecological interpretations concern patterns across fossil and archaeological evidence; they do not demonstrate identical diets or locomotor capacities in every species assigned to the genus. (nature.com)

Early Homo also overlapped with other hominins rather than immediately replacing them. Discoveries published in 2025 document both Homo and Australopithecus at Ledi-Geraru between approximately 2.78 and 2.59 million years ago. Around 1.9 million years ago, eastern Africa supported H. habilis, H. rudolfensis, early H. erectus, and Paranthropus boisei. Such overlap undermines a straightforward sequence in which each named species simply transformed into the next. Establishing ancestry requires phylogenetic analysis of shared anatomical characters as well as chronology; being older, geographically nearby, or anatomically intermediate does not by itself identify a direct ancestor. (pmc.ncbi.nlm.nih.gov)

The taxonomic emergence of Homo must also be separated from the origin of stone-tool manufacture. Sonia Harmand and colleagues reported stone artefacts from Lomekwi 3 in Kenya in 2015, dating them to approximately 3.3 million years ago—about half a million years before LD 350-1. Their makers remain unidentified. Consequently, tool manufacture cannot serve as an exclusive diagnostic criterion for Homo, and finding fossils near artefacts does not alone establish which hominin produced them. The technological record and the anatomical classification of the genus are related but distinct lines of evidence. (nature.com)

Homo erectus and Early Dispersal beyond Africa

*Homo erectus* is generally recognized by a combination of a long, low braincase, prominent brow ridges, and a relatively large face, together with changes in limb proportions associated with terrestrial locomotion. Its classification varies: H. erectus in the broad sense includes African and Asian specimens, whereas some researchers assign early African fossils to *Homo ergaster* and reserve H. erectus more narrowly for Asian material. These alternatives concern the interpretation of anatomical variation, not simply different names for geographical populations. Differences in skull thickness, robusticity, and proportions must be assessed against variation within populations and changes through time; geographical separation alone does not establish distinct species. (nhm.ac.uk)

The Nariokotome skeleton, KNM-WT 15000, provides unusually extensive evidence for body form. Discovered by Kamoya Kimeu in 1984 west of Lake Turkana, Kenya, and dated to approximately 1.6 million years ago, it belonged to a juvenile male commonly called “Turkana Boy.” Reconstructions place his height at about 1.6 metres and body mass at approximately 48 kilograms at death. Dental microstructure indicates an age of roughly eight or nine years, although estimates of eventual adult stature depend on assumptions about growth. His elongated legs and relatively shorter arms document proportions more like those of living humans than those of australopithecines. These features are consistent with efficient terrestrial travel, but do not imply that every aspect of his growth or anatomy was modern. His cranial capacity was approximately 880 cubic centimetres. (humanorigins.si.edu)

Dmanisi, in the Republic of Georgia, documents early Homo outside Africa approximately 1.8 million years ago. Archaeological and geological investigations published in 2011 identified occupations between about 1.85 and 1.78 million years ago, rather than a single brief visit. The fossils are often included within a broad H. erectus, although their precise taxonomic placement remains debated. Postcranial remains combine small body size and primitive upper-limb features with more humanlike lower-limb proportions. The exceptionally complete Skull 5 has a cranial capacity of only about 546 cubic centimetres. Dmanisi therefore shows that dispersal did not require the relatively tall stature or larger brain represented by Nariokotome, and cautions against treating these characteristics as prerequisites for leaving Africa. (pmc.ncbi.nlm.nih.gov)

Java and China provide geographically distant records whose chronologies have repeatedly been revised. A 2020 study using uranium–lead and fission-track dating of volcanic zircons placed the first known Sangiran hominins most probably around 1.3 million years ago, and younger than 1.5 million years, rather than accepting older estimates without qualification. In China, a 2026 study assigned the fossil-bearing layer at Yunxian an age of $1.77 \pm 0.08$ million years, with a total uncertainty of approximately 0.11 million years; this would place the crania close in time to Dmanisi, although their anatomical affinities remain under investigation. The later Zhoukoudian record near Beijing includes fossil-bearing lower deposits dated by aluminium–beryllium burial dating to approximately 770,000 years ago. These are dates of associated deposits, not direct measurements of fossil bone. (cir.nii.ac.jp)

Persistence was strongly regional. At Ngandong, Java, Yan Rizal and colleagues combined 52 radiometric age estimates with Bayesian modelling to date the H. erectus bone bed to approximately 117,000–108,000 years ago. Published online in December 2019, this chronology places a late Javanese population far beyond the early African and Georgian records. It dates the accumulation of the fossil-bearing deposit and establishes a late known occurrence, not the exact moment of extinction. The long interval and broad distribution encompassed by H. erectus should consequently not be mistaken for anatomical uniformity or uninterrupted occupation of every region. (nature.com)

Early dispersal is better investigated as population expansion through changing ecological opportunities than as one identifiable migration event. Landscape models examine how freshwater, neighbouring habitats, and topographic corridors could support movement and settlement; lower sea levels periodically increased terrestrial connections toward Southeast Asia. Such models propose mechanisms rather than reconstructing a proven itinerary. Repeated occupations at Dmanisi likewise document local population activity, not the number of departures from Africa. The available evidence permits multiple expansions, interruptions, and regional persistence, without determining a single route or founding population. (disperse-project.org)

Middle and Late Pleistocene Human Diversity

Middle Pleistocene fossils from Africa and Eurasia combine large braincases with robust faces, prominent brow ridges, and other features that vary among specimens. Examples include Bodo in Ethiopia, Kabwe in Zambia, Arago in France, and Dali in China. Their taxonomic classification remains unsettled. Otto Schoetensack named Homo heidelbergensis in 1908 from the Mauer mandible in Germany; subsequent researchers extended the name to substantially different African and Eurasian remains. In its broad usage, or sensu lato, it encompasses fossils sometimes interpreted as ancestors of both Neanderthals and modern humans; narrower usages restrict it chiefly to European material. Because its defining specimen is a jaw, connecting it to isolated skulls requires anatomical comparisons rather than direct association. The label therefore does not identify a universally agreed ancestral population. Kabwe illustrates the chronological complexity: a study published in 2020 estimated its age at $299 \pm 25$ thousand years, considerably younger than some earlier interpretations. (onlinelibrary.wiley.com)

The European record documents the gradual establishment of Neanderthal characteristics rather than an abrupt replacement of one anatomical type by another. At Sima de los Huesos in Spain, approximately 430,000-year-old fossils already display Neanderthal affinities; nuclear DNA recovered from two specimens and reported in 2016 places them closer to Neanderthals than to Denisovans. Later Homo neanderthalensis occupied Europe and southwestern and central Asia until approximately 40,000 years ago. Characteristic features include an elongated, low cranial vault, substantial brow ridges, a projecting middle face, and a robust skeleton with relatively short forearms and lower legs. Compact body proportions reduced surface area relative to body volume and are consistent with heat conservation. These skeletal traits are not measures of intelligence or evidence for particular social practices: behavioral interpretations require independent archaeological evidence. (eva.mpg.de)

Denisovans were first recognized through molecular evidence, not a diagnostic complete skeleton. Research published in 2010 sequenced DNA from a finger-bone fragment from Denisova Cave in Siberia, identifying a population related to, but genetically distinct from, Neanderthals. Its initial fossil record consisted principally of small fragments and teeth. Molecular identification has subsequently connected more informative remains to this lineage: in 2025, mitochondrial DNA from dental calculus linked the Harbin cranium in northeastern China, more than 146,000 years old, to Denisovans. Nevertheless, a genetically identified population does not automatically correspond to a traditional fossil species. Genetic relationships describe ancestry among sampled individuals, whereas fossil classifications depend on anatomical diagnoses; establishing their correspondence requires specimens preserving both kinds of evidence. (nature.com)

Island Southeast Asia preserves additional combinations of traits. Homo floresiensis, discovered at Liang Bua on Flores in 2003, had a very small body and brain, relatively large feet, and no projecting chin. Its skeletal remains date to approximately 100,000–60,000 years ago, whereas associated stone artifacts extend to about 50,000 years ago; those are distinct evidence ranges, not a securely established extinction date. On Luzon in the Philippines, Florent Détroit and colleagues described Homo luzonensis in 2019 from teeth and limb-bone fragments from Callao Cave, with reported minimum ages of approximately 50,000 and 67,000 years. Its dental and finger-and-toe anatomy combines features otherwise found in different hominin groups. These fragmentary remains do not establish its overall stature or a definitive relationship to the Flores population. (humanorigins.si.edu)

In South Africa, Homo naledi provides another example of late survival without an entirely modern anatomical pattern. Excavations associated with Lee Berger’s research team recovered more than 1,550 specimens representing at least 15 individuals from the Dinaledi Chamber of Rising Star Cave in 2013–2014. Dating published in 2017 constrained this assemblage to approximately 335,000–236,000 years ago, despite anatomical features initially suggesting a much greater age. Such regional records document overlapping lineages, not a succession in which every population acquired the same traits in the same order. They also have unequal resolution: European skeletal assemblages, Siberian molecular evidence, and isolated tropical-island remains permit different kinds of comparison. Similar anatomy, chronological overlap, and genetic affinity must therefore be distinguished when reconstructing evolutionary relationships; none alone supplies a definitive family tree for all these populations. (humanorigins.si.edu)

The African Emergence of Homo sapiens

Fossil Homo sapiens is generally recognized through a combination of features rather than a single diagnostic character: a relatively reduced, retracted face, a projecting bony chin, and a high, more globular braincase, usually accompanied by reduced brow ridges. These features concern shape and proportions, not simply brain volume. Their expression varies, and early representatives need not possess the complete anatomical pattern found in living humans. Consequently, the boundary between ancestral populations and early H. sapiens depends partly on how researchers define the species and weight different anatomical characters. (pmc.ncbi.nlm.nih.gov)

At Jebel Irhoud in Morocco, fossils described by Jean-Jacques Hublin, Abdelouahed Ben-Ncer, and colleagues in 2017 combine facial and dental features resembling those of H. sapiens with an elongated, less globular braincase. Daniel Richter and colleagues obtained a weighted mean age of $315 \pm 34$ thousand years from thermoluminescence measurements of heated flints associated with the remains. This dates the heating of the artifacts, rather than directly dating the bones; a recalculated uranium-series/electron-spin-resonance age of $286 \pm 32$ thousand years for a tooth provides independent support. The fossils document an early stage of the lineage in northwestern Africa, not a demonstrated birthplace of the species. Their anatomical combination illustrates mosaic evolution: facial reduction was established before the fully rounded cranial form characteristic of recent humans. (nature.com)

Eastern African evidence provides distinct anatomical and chronological comparisons. Omo I, from the Omo-Kibish Formation in southern Ethiopia, possesses a more clearly modern cranial form and chin. Its frequently cited age of approximately 197,000 years was revised by Céline Vidal and colleagues in 2022. They chemically matched an overlying volcanic ash layer to an eruption of Shala volcano dated to $233 \pm 22$ thousand years ago. Because the fossil-bearing sediments lie below that ash, this establishes a minimum age, not an exact fossil age. The Herto remains from Ethiopia’s Middle Awash region, described by Tim White and colleagues in 2003, are conventionally placed around 160,000–154,000 years ago. They combine broadly sapiens-like anatomy with substantial cranial robusticity; their dating depends on associated volcanic deposits and stratigraphic correlations rather than direct measurements of the skulls. (pmc.ncbi.nlm.nih.gov)

The geographical spread and differing anatomical combinations of these fossils are compatible with an African population history involving regional differentiation and intermittent contact. In such models, shifting ecological barriers separate populations, allowing genetic drift and local evolutionary change, while renewed migration restores gene flow. Nevertheless, fossil resemblance does not directly measure interbreeding, and the sparse record cannot establish which regions contributed particular traits or whether every early population contributed ancestry to living people. Eleanor Scerri and colleagues articulated this interconnected-population interpretation in 2018 as an alternative to locating the entire emergence process within one small, permanently isolated population. (pmc.ncbi.nlm.nih.gov)

Genomic evidence likewise permits more complex histories than a simple sequence of population splits. In 2023, Aaron Ragsdale and colleagues compared demographic models using variation and associations between variants in African genomes, including newly sequenced genomes from 44 Nama individuals. Their favored “weakly structured stem” models describe two or more ancestral populations that remained genetically connected over hundreds of thousands of years. Within those models, the earliest divergence among the sampled contemporary populations occurred approximately 120,000–135,000 years ago—considerably later than the Jebel Irhoud fossils. These are model-dependent population-history estimates, not dates for the origin of H. sapiens, and alternative histories, including additional ancestral populations, remain possible. (nature.com)

African population-structure models should be distinguished from historical worldwide multiregional models. The latter proposed long-term regional continuity from earlier Homo populations across Africa and Eurasia, maintained by interregional gene flow. An interconnected African origin instead addresses differentiation and exchange among populations within Africa during the emergence of H. sapiens; it does not require equivalent evolutionary continuity across Eurasia. Ragsdale and colleagues explicitly distinguish their weakly differentiated ancestral populations from models requiring prolonged reproductive isolation followed by admixture with deeply divergent African archaic populations. (discovery.ucl.ac.uk)

Global Expansion of Homo sapiens

Early appearances outside Africa do not necessarily identify the populations ancestral to most living non-Africans. Israel Hershkovitz and colleagues reported in 2018 that the Misliya Cave upper jaw from Israel dated to approximately 177,000–194,000 years ago; fossils from Skhul and Qafzeh document further occupation of the Levant around 90,000–120,000 years ago. These finds establish early range expansions, but not uninterrupted settlement or direct genealogical continuity. Genomic studies generally associate most present-day non-African ancestry with a later expansion, commonly placed broadly around 50,000–70,000 years ago. Its timing and the possible contribution of earlier dispersals remain model-dependent. Luca Pagani and colleagues’ 2016 analysis, for example, inferred a small earlier contribution to Papuan ancestry while attributing the great majority of non-African ancestry to the later expansion. (pubmed.ncbi.nlm.nih.gov)

Southwestern Asia provided connections between Africa and Eurasia, rather than a permanently open migration corridor. Proposed exits include a northern passage through northeastern Africa and the Levant and a southern passage across the Red Sea into Arabia. Climate simulations identify intervals when rainfall could support movement along either route; archaeological sequences in Arabia independently show repeated hominin occupations associated with wetter conditions. However, stone tools without diagnostic human remains cannot establish that every occupation represents Homo sapiens. Evidence therefore supports changing opportunities for dispersal, not a single demonstrated coastal itinerary followed by all expanding populations. (nature.com)

By approximately 45,000 years ago, H. sapiens occupied widely separated Eurasian regions. Research published in 2024 identified human bone fragments from Ranis, Germany, using morphology, proteins and mitochondrial DNA, with direct radiocarbon dating demonstrating occupation at this early date. In eastern Asia, the approximately 40,000-year-old Tianyuan individual near Beijing provides both skeletal and genomic evidence: analyses published in 2013 and 2017 found closer genetic relationships to Asian populations than to European populations. These are dated regional occurrences, not exact arrival dates. Nor did every early population persist: genomes from Ranis and Zlatý kůň, analysed in a study published in 2025, represent a population with no detected contribution to later sampled populations. (nature.com)

Settlement of Sahul, the landmass joining Australia, New Guinea and Tasmania during lower sea levels, required passage through southeastern Asia and maritime crossings through Wallacea. Lower seas exposed continental shelves but did not eliminate the deep-water channels between Sunda and Sahul; reconstructed routes include a crossing approaching 100 kilometres. Occupation is generally placed within approximately 50,000–65,000 years ago, although its earliest timing remains contested. Chris Clarkson and colleagues’ 2017 excavations at Madjedbebe in northern Australia yielded artifacts in sediments dated by luminescence to approximately 65,000 years ago. This dates the archaeological context rather than a human fossil, and uncertainties about initial settlement should not be collapsed into one universally accepted date or route. (nature.com)

The Americas were populated through Beringia, the exposed region connecting northeastern Asia and northwestern North America during glacial low sea levels. Occupation south of the continental ice sheets is established before the Clovis archaeological tradition: Loren Davis and colleagues reported in 2019 that occupation at Cooper’s Ferry, Idaho, began approximately 16,560–15,280 years ago. Earlier evidence requires separate assessment. White Sands footprints in New Mexico were dated to approximately 21,000–23,000 years ago, with independent pollen radiocarbon and quartz luminescence measurements published in 2023 supporting their antiquity. The Pacific coastal route and an inland ice-free corridor remain distinct hypotheses; a 2022 study placed the corridor’s full opening around 13,800 years ago, too late for these earlier occupations. Coastal movement is consequently a major candidate, although submerged shorelines limit direct archaeological testing. (nature.com)

Expansion also changed population structure. In a founder effect, a small colonizing group carries only part of its source population’s genetic variation; repeated founding events and genetic drift can reduce diversity along an expanding range. Models incorporating subsequent gene flow reproduce the broad decline in genetic diversity with distance from Africa without requiring complete isolation. Later migrations further altered these patterns: ancient genomes document substantial mixing associated with Near Eastern farming expansions and large-scale population movements across Bronze Age Eurasia. Present-day geographical distributions therefore cannot be read directly as maps of the first settlers. (pmc.ncbi.nlm.nih.gov)

Interbreeding and the Legacy of Archaic Populations

Introgression is the incorporation of genetic material from one population into another through interbreeding followed by reproduction across subsequent generations. Evidence comes from both excess sharing of alleles with archaic humans and identifiable stretches of archaic ancestry within modern genomes. In 2010, Richard Green and colleagues reported that Neanderthals shared more genetic variants with sampled Eurasians than with sampled sub-Saharan Africans, supporting interbreeding after the ancestors of those Eurasians left Africa. That year, David Reich and colleagues showed that a genome from Denisova Cave in Siberia represented a distinct archaic population that had also contributed ancestry to modern humans, particularly Melanesians. These findings distinguish genetic contributions from Neanderthals and Denisovans rather than treating all archaic ancestry as a single component. (pmc.ncbi.nlm.nih.gov)

Many present-day people with predominantly non-African ancestry carry approximately 1–2% Neanderthal ancestry, although estimates vary with the individuals sampled and the analytical method. Some studies estimate that East Asian population averages are about 12–20% higher than European averages—a relative difference, not an additional 12–20 percentage points. Population averages also conceal differences among individuals and among the genomic segments they inherited. Present-day geography does not directly identify where interbreeding occurred: later gene flow, including migrations back into Africa, redistributed Neanderthal-derived sequences. Lu Chen and colleagues demonstrated in 2020 that detectable Neanderthal ancestry in sampled African individuals was partly attributable to Eurasian back-migration. Earlier movement of modern-human-related DNA into Neanderthals additionally complicates the interpretation of sequences shared between the groups. (nature.com)

Ancient genomes constrain the timing of encounters because recombination progressively breaks inherited archaic DNA into shorter segments. Arev P. Sümer and colleagues analysed six individuals from Ranis, Germany, and the Zlatý kůň individual from Czechia, connecting their Neanderthal ancestry to the major admixture episode shared with other sampled non-Africans. Their study appeared online on December 12, 2024, and in the February 20, 2025 issue of Nature. It placed this episode approximately 45,000–49,000 years ago, rather than identifying a single mating or an exact geographical location. Additional encounters are evident in three approximately 45,000-year-old individuals from Bacho Kiro Cave, Bulgaria: a 2021 analysis found about 3.0–3.8% Neanderthal ancestry and unusually long segments, indicating Neanderthal ancestors fewer than six or approximately seven generations earlier. These recent family connections were additional to the older ancestry shared across non-African populations. (nature.com)

Interbreeding also occurred between archaic populations themselves. In 2018, Viviane Slon and colleagues identified Denisova 11, a female represented by a bone fragment, as the first-generation offspring of a Neanderthal mother and a Denisovan father. Her father also carried evidence of earlier Neanderthal ancestry, demonstrating encounters beyond the mating that produced this individual. Unlike population-level statistical reconstructions, this genome directly identifies the ancestry of two parents. It does not, by itself, establish how frequently such pairings occurred or whether this individual reproduced. (pmc.ncbi.nlm.nih.gov)

Denisovan-related ancestry reflects multiple contributions rather than descent from one uniform source population. Sharon Browning and colleagues identified two distinguishable components in East Asians in 2018. Guy Jacobs and colleagues subsequently reported two deeply divergent Denisovan-related ancestries in Papuans, alongside another contribution associated with East Asians. These source populations were inferred from surviving DNA segments, not each represented by a sequenced fossil genome; “Denisovan-related” therefore need not mean genetically identical to the sampled Siberian individual. Ancestry proportions also reflect later mixing: a 2021 study found approximately 30–40% more Denisovan ancestry in sampled Ayta Magbukon individuals from the Philippines than in sampled Australians or Papuans, and documented dilution through admixture with populations carrying less Denisovan ancestry. (sciencedirect.com)

Introgressed variants subsequently experienced genetic drift and natural selection. Drift can remove or retain variants by chance, while selection can reduce harmful variants or increase beneficial ones. Recombination separates archaic variants from neighbouring sequences, allowing their frequencies to change independently. Neanderthal ancestry is unevenly distributed: large genomic “deserts” contain little surviving ancestry, and depletion near functional regions and on the X chromosome supports selective removal of some introgressed material. Ancient genomes indicate that several depleted regions were already present in early Eurasian populations. These patterns concern the fate of particular sequences, not uniform retention or rejection of an entire archaic genome. (nature.com)

Bipedalism and the Transformation of the Skeleton

Bipedalism encompasses several locomotor patterns rather than a single anatomical condition. Occasional, or facultative, bipedalism occurs when an animal normally using other forms of locomotion stands or travels on two legs; living chimpanzees do this without possessing the full skeletal adaptations for sustained upright walking. Habitual bipedalism means that two-legged locomotion is regularly used, but does not necessarily exclude substantial climbing. Efficient long-distance terrestrial walking additionally depends on limb proportions, joint mechanics, and foot function. These distinctions matter because early hominins could be habitual bipeds while retaining arboreal capabilities; the longer legs documented in Homo erectus by approximately 1.9 million years ago supported longer strides and more extensive terrestrial travel. (humanorigins.si.edu)

At the skull base, a more anteriorly positioned foramen magnum—the opening through which the spinal cord passes—places the head more directly above the vertebral column. This arrangement is evidence relevant to upright posture, although it cannot independently establish how frequently an animal walked bipedally. Farther down, lumbar lordosis, the forward curvature of the lower spine, helps position the trunk’s mass above the hips rather than in front of them. Shorter iliac blades leave greater freedom for the lower lumbar region, linking pelvic transformation with the capacity to maintain an upright trunk over extended lower limbs. (humanorigins.si.edu)

The pelvis also changes the mechanics of balance. Compared with the tall iliac blades of living African apes, human ilia are shorter and wrap around the sides of the body. This reorientation changes the leverage of the gluteus medius and minimus: during the single-support phase of walking, they counteract the tendency of the pelvis to drop toward the unsupported side. The femur slopes inward from hip to knee, producing a valgus alignment that brings the supporting knee closer to the body’s midline. Together, pelvic muscle action and lower-limb alignment reduce the difficulty of balancing alternately on one leg. These are coordinated functional changes, not independent diagnostic switches between “ape” and “human” anatomy. (pmc.ncbi.nlm.nih.gov)

The foot combines weight-bearing stability with propulsion. An adducted great toe, aligned with the other toes rather than divergent for grasping, contributes to push-off. The plantar tissues and arches help the foot function as a supportive lever with elastic properties. Fossil bones reveal joint geometry and potential movement, whereas footprint trails record actual foot placement, step spacing, and interaction with the ground. At Laetoli, Tanzania, tracks approximately 3.66 million years old provide direct evidence of bipedal progression. However, footprint depth depends on substrate properties as well as movement: experimental analyses published in 2010 and 2016 reached different conclusions about how closely the trackmakers’ limb posture resembled modern human walking. A short trackway also cannot establish daily walking distances or the absence of climbing. (pmc.ncbi.nlm.nih.gov)

Energetic economy is one measurable consequence of the modern walking pattern. In a 2007 study, Michael Sockol, David Raichlen, and Herman Pontzer found that human walking was approximately 75% less energetically costly than chimpanzee quadrupedal and bipedal walking, associating the difference with longer hindlimbs and more extended hip posture. This comparison supports an energetic mechanism, but living chimpanzees are not unchanged models of ancestral hominins, so it does not establish why bipedalism originally evolved. Upright locomotion also frees the hands for carrying food, infants, and objects; these opportunities are functional consequences and possible selective benefits, not proof of a single initiating cause. (pubmed.ncbi.nlm.nih.gov)

Bipedal adaptations preceded the major expansion of the brain. Habitual upright walking was established among hominins by roughly four million years ago, while substantial increases in brain volume occurred later, particularly within Homo. Smithsonian compilations identify especially pronounced enlargement between approximately 800,000 and 200,000 years ago. The chronological separation rules out a large, modern-sized brain as a prerequisite for the initial transformation of the locomotor skeleton. (humanorigins.si.edu)

Brain Evolution, Development, and Social Life

Human brain evolution involved changes in size, regional proportions, organization, and developmental timing, rather than uniform enlargement. A 2020 reconstruction of Australopithecus afarensis estimated a mean adult endocranial volume of approximately $445\ \text{cm}^3$ in its sample. For comparison, the Smithsonian reports average adult brain masses of 384 g in chimpanzees and 1,352 g in modern humans; mass and endocranial volume are related but distinct measurements. Absolute size must also be distinguished from encephalization, enlargement relative to body size, because larger bodies themselves require additional neural resources. Neither measure directly establishes cognitive ability: regional organization, connectivity, developmental experience, and the demands of particular tasks also matter. (pmc.ncbi.nlm.nih.gov)

Fossil evidence comes principally from endocasts, natural sedimentary casts or artificial and digitally reconstructed models of the cranial cavity. These reveal approximate volume, overall shape, asymmetries, and sometimes impressions of grooves on the cerebral cortex. They do not preserve individual neurons, synapses, or detailed circuitry, and cannot directly reconstruct intelligence or language. In 2021, Marcia Ponce de León and colleagues interpreted frontal-lobe impressions in early African Homo and the Dmanisi fossils as retaining an ape-like organization, whereas African specimens younger than 1.5 million years showed a more humanlike pattern. Size and shape also changed separately within Homo sapiens: Simon Neubauer, Jean-Jacques Hublin, and Philipp Gunz’s 2018 analysis of 20 fossil endocasts found modern-range volumes by approximately 300,000 years ago, but present-day globular shape only between about 100,000 and 35,000 years ago. (humanorigins.si.edu)

Prolonged development provides time for experience to modify neural connections and for children to acquire complex skills. It did not necessarily originate alongside large brains. Philipp Gunz and colleagues’ 2020 comparison of infant and adult A. afarensis endocasts suggested prolonged brain growth despite ape-like cortical organization. Comparative tissue research likewise shows that chimpanzees share extended juvenile synapse formation and delayed maturation of prefrontal neurons, although later human cortical development includes myelination continuing into early adulthood. Childhood learning therefore draws on an inherited primate developmental foundation as well as subsequently evolved changes. Young chimpanzees’ acquisition of nut-cracking and other tool skills illustrates how prolonged dependence creates opportunities for learning from experienced individuals. (pmc.ncbi.nlm.nih.gov)

Growing brains impose substantial demands on metabolism. In 2014, Christopher Kuzawa and colleagues combined existing positron-emission tomography and magnetic-resonance imaging data to estimate brain glucose consumption across development. At its childhood peak, glucose uptake was equivalent to approximately 66% of whole-body resting metabolic requirements and 43% of daily energy requirements. These are energy-equivalent comparisons, not measurements showing that the brain consumes 66% of every nutrient. Peak cerebral glucose demand coincided with particularly slow body-weight growth, supporting a developmental trade-off between neural expenditure and somatic growth. Human children also remain dependent on provisioned food after weaning, linking these energetic costs to caregiving and resource transfers. (pmc.ncbi.nlm.nih.gov)

Archaeological and skeletal evidence constrain, but rarely specify, ancient social arrangements. At Kanjera, Kenya, approximately two-million-year-old tools include stone transported from sources up to 12 km away, alongside butchered animal remains. Such concentrations document resource transport and processing at repeatedly used places; they do not by themselves identify recipients or rules of food sharing. At Dmanisi, the approximately 1.77-million-year-old individual represented by skull D3444 survived substantial tooth loss long enough for the jaw to remodel. This has been interpreted as evidence of assistance, but the fossil does not directly record who provided care or what assistance involved. (humanorigins.si.edu)

Cooperative breeding models propose that assistance from fathers and other caregivers helped sustain dependent offspring while increasing opportunities for social learning. Judith Burkart and colleagues’ 2014 standardized experiments with 24 groups from 15 primate species found that care by individuals other than the mother best predicted unsolicited helping. This comparative association supports a proposed connection between shared childcare and cooperative motivation; it does not establish when either arose in hominins or demonstrate a cause of brain enlargement. In developmental models, tolerant access to several experienced individuals can improve learning without requiring an immediate increase in cognitive capacity, while provisioning can help meet the costs of prolonged maturation. (nature.com)

Diet, Food Processing, and Material Technology

Dietary reconstruction combines evidence with different biological timescales. Tooth shape and enamel thickness indicate mechanical capabilities, rather than a menu; microscopic pits and scratches record comparatively recent use, including abrasion by food and grit. Stable carbon isotopes in enamel distinguish resources derived from $C_3$ plants, including most trees and shrubs, from those derived from $C_4$ grasses and some sedges. However, a $C_4$ signal may reflect either plant consumption or eating animals that consumed those plants. Isotopic measurements integrate diet during tissue formation, whereas microwear can reflect a much shorter interval; disagreement between them need not indicate analytical failure. (pmc.ncbi.nlm.nih.gov)

Animal remains and plant residues provide evidence of particular processing activities. Cut marks can indicate skinning, dismemberment, or flesh removal, while percussion damage can indicate marrow extraction. Such traces establish carcass processing more readily than whether an animal was hunted or scavenged. At Nyayanga, Kenya, Thomas Plummer and colleagues reported in 2023 Oldowan assemblages dated to approximately 3.03–2.58 million years ago, including cut-marked hippopotamid bones and tools bearing wear consistent with processing animal and plant tissues. Associated Paranthropus teeth do not identify the toolmakers. Starch residues on approximately 780,000-year-old percussive tools from Gesher Benot Ya’aqov, Israel, provide a separate line of evidence for processing starch-rich plants, whose contributions are easily underestimated in bone-dominated assemblages. (ora.ox.ac.uk)

At Lomekwi 3, west of Lake Turkana in Kenya, Sonia Harmand and colleagues described approximately 3.3-million-year-old stone artifacts in 2015. The assemblage combines core reduction with battering activities, showing deliberate exploitation of stone fracture properties before securely documented Homo. No fossil securely identifies its makers. Oldowan technology emphasizes hammerstones, cores, and sharp flakes removed by percussion; flakes, not merely conspicuous “choppers,” supplied useful cutting edges. Acheulean assemblages include large cutting tools such as handaxes, shaped through repeated removals from both faces, and are securely documented by approximately 1.76 million years ago. These names designate technological traditions, not biological species or universally successive evolutionary stages. (nature.com)

Prepared-core methods, including the Levallois technique, organize a core’s surfaces and convexities before removing flakes of controlled form. Their development was geographically uneven: early European examples occur alongside other reduction methods rather than replacing them immediately. Composite tools added another manufacturing problem—joining separately made components. At Sibudu, South Africa, adhesive residues concentrated along the backs or bases of stone implements support attachment to handles or shafts; some residues contain conifer resin, and ochre was sometimes incorporated. These artifacts preserve only parts of a system that also required organic components and joining materials. (sciencedirect.com)

Fire evidence requires distinctions between natural burning, fire use, sustained management, habitual use, and cooking. At Wonderwerk Cave, South Africa, microscopic plant ash and burned bone in approximately one-million-year-old deposits support burning within the cave, but do not establish when fire-making began or how regularly food was cooked. European evidence for habitual fire use becomes substantially more frequent around 400,000–300,000 years ago. In 2022, Irit Zohar and colleagues interpreted fish teeth from Gesher Benot Ya’aqov, approximately 780,000 years old, as evidence of cooking: enamel crystallographic changes indicated heating below $500\,^\circ\mathrm{C}$, while spatial associations with burned flint and the composition of the fish assemblages supported human processing. This is a contextual inference, not a direct observation of a cooking procedure. (doi.org)

Raw-material provenance and manufacturing traces allow archaeologists to reconstruct acquisition, transport, shaping, use, and discard. At Olorgesailie, Kenya, early Middle Stone Age assemblages document long-distance transport of stone, showing that an artifact’s excavation location need not be its raw-material source. The durable stone record nevertheless represents only part of ancient material culture. Waterlogged sediments at Kalambo Falls, Zambia, preserved two logs joined by an intentionally cut notch in deposits dated to approximately $476{,}000 \pm 23{,}000$ years ago. Such exceptional preservation exposes woodworking that would normally disappear through decay, leaving stone implements without the organic objects they helped manufacture. (research.si.edu)

Language, Symbolism, and Cumulative Culture

Communication transfers information through signals, including calls, gestures, and facial expressions; it is widespread among animals. Speech is the production of articulated vocal sounds, whereas language is a structured system in which conventional signs and rules allow speakers or signers to construct meanings. Signed languages demonstrate that language does not require vocalization. Symbolic representation uses an object, mark, or action to stand for something through a shared convention, rather than merely resembling it. Cumulative culture concerns the retention and modification of socially transmitted knowledge across successive learners; neither communication nor a persistent learned tradition alone demonstrates such accumulation. (nature.com)

Evidence for speech capacity is not equivalent to evidence for language. The hyoid bone supports muscles involved in tongue and laryngeal movement; the approximately 60,000-year-old Neanderthal hyoid from Kebara Cave, Israel, resembles modern human examples. Ruggero D’Anastasio and colleagues’ 2013 analysis also found comparable internal structure and modelled mechanical behaviour. Nevertheless, a hyoid cannot reconstruct the complete vocal tract, neural control, or grammar. Genetic evidence is similarly indirect: Johannes Krause and colleagues reported in 2007 that Neanderthals shared two derived amino-acid substitutions in FOXP2 with modern humans. This gene participates in developmental networks relevant to speech and language, but is neither exclusively linguistic nor sufficient to produce language. Anatomical and genetic findings therefore constrain possible capacities without establishing when language originated. (journals.plos.org)

Pigments, engravings, and ornaments require different interpretive tests in archaeology. Ochre can provide colour but also serve practical purposes, including use in adhesive mixtures; its presence alone does not establish symbolism. Microscopic examination can distinguish deliberate incisions from abrasion, but deliberate marking need not encode a shared meaning. Repeated motifs, prepared surfaces, patterned placement, and recurrence across securely associated objects strengthen interpretations of a convention. In 2002, Christopher Henshilwood and colleagues described two geometrically engraved ochre pieces from Blombos Cave, South Africa, in layers dated to a mean of approximately 77,000 years by thermoluminescence of burnt stone artifacts. The contextual date establishes the antiquity of the engravings, not the meaning their makers assigned to them. (humanorigins.si.edu)

Personal ornaments provide evidence for display, potentially conveying affiliation or identity. El Mehdi Sehasseh and colleagues’ 2021 study documented 33 marine-shell beads from Bizmoune Cave, Morocco, many from deposits at least 142,000 years old; modification, wear, and pigment traces supported their interpretation as ornaments. Figurative art instead represents recognizable subjects. In 2024, Adhi Agus Oktaviana and colleagues reported a minimum age of 51,200 years for a Sulawesi cave painting at Leang Karampuang depicting a pig and three human-like figures. Uranium-series measurements dated carbonate overlying the pigment, so the painting itself must be older. The arrangement was interpreted as a narrative scene, although neither the depicted story nor its associated spoken language can be recovered. (pubmed.ncbi.nlm.nih.gov)

Cumulative transmission involves innovations being learned, retained, and subsequently improved or elaborated. Observation, imitation, practice, teaching, and verbal explanation can transmit different components of a skill; teaching is helpful but not universally necessary. Retention creates a “ratchet” effect when learners build on earlier achievements instead of repeatedly reinventing them. Experiments using multistage tasks and transmission chains test these mechanisms by tracking changes between successive participants. Animal traditions and evidence for limited accumulation in some nonhuman species caution against equating every learned behaviour with uniquely modern cognition. Archaeological recurrence establishes continuity more readily than it establishes the precise learning mechanisms or cumulative sequence responsible. (pmc.ncbi.nlm.nih.gov)

These innovations do not appear everywhere as a synchronized package. Sally McBrearty and Alison Brooks’ 2000 synthesis showed that behaviours formerly grouped into a European “revolution” around 40,000–50,000 years ago had earlier, geographically dispersed African precedents. Different regional histories of ornaments, engravings, and representational images must also be interpreted through preservation and sampling: speech, body decoration, and perishable symbolic objects often leave no durable trace. An innovation’s first surviving occurrence consequently cannot be treated as the birth date of the cognitive capacity that enabled it. (pubmed.ncbi.nlm.nih.gov)

Recent Evolution and Gene–Culture Interactions

Biological evolution continued after the emergence of Homo sapiens: mutation supplied new variants, natural selection altered their frequencies through differences in survival and reproduction, genetic drift produced random changes, and migration redistributed variants through gene flow. Changing population sizes also reshaped genetic variation. A study published in 2013 of 6,515 human exomes—the protein-coding portions of genomes—estimated that approximately 73% of the observed protein-coding single-nucleotide variants had originated within the preceding 5,000 years, largely reflecting recent population growth. Such estimates concern the variants in that dataset, not the age of human genes generally. Ancient-genome studies further demonstrate why an increasing allele frequency cannot automatically be attributed to selection: immigration and population replacement can produce similar changes. (nature.com)

Dairying illustrates gene–culture coevolution, in which socially transmitted practices alter the conditions acting on inherited variation. Lactase, encoded by LCT, breaks down the milk sugar lactose; its production normally declines after childhood. Lactase persistence maintains substantial enzyme production into adulthood. Associated variants occur in regulatory DNA within the neighboring MCM6 gene, rather than changing lactase’s protein sequence. The −13910T variant is widespread in northern Europe, whereas independently evolved variants contribute to persistence in African and Middle Eastern populations. In a 2007 study, Sarah Tishkoff and colleagues examined 470 people from Tanzania, Kenya, and Sudan and identified three associated variants—−14010C, −13915G, and −13907G—which enhanced transcription from the LCT promoter in laboratory assays. Different regulatory mutations therefore produced a similar digestive phenotype. (pmc.ncbi.nlm.nih.gov)

The relationship between milk consumption and selection was not an immediate or uniform response to dairying. Richard Evershed and colleagues’ 2022 analysis of approximately 7,000 pottery-fat residues from more than 550 European archaeological sites documented widespread milk exploitation across the preceding 9,000 years. Milk use substantially preceded high lactase-persistence frequencies, and geographical variation in its intensity did not explain the inferred selection pattern better than uniform selection beginning in the Neolithic. The researchers proposed that famine and increased pathogen exposure intensified the disadvantages of consuming milk without persistence; these are explanatory hypotheses supported by model comparisons, not directly observed prehistoric causes. Cultural processing also mattered: fermentation and cheese-making could reduce lactose exposure without requiring genetic change. (nature.com)

High-altitude adaptation connects selection with ancestry acquired from archaic humans. EPAS1 encodes HIF-2α, a transcription factor involved in responses to low oxygen. In 2010, Cynthia Beall and colleagues identified EPAS1 variants associated with relatively low hemoglobin concentrations in Tibetan highlanders; among participants living at 4,200 metres, major-allele homozygotes averaged 0.8 grams per decilitre less hemoglobin than heterozygotes after adjustment for sex. In 2014, Emilia Huerta-Sánchez and colleagues sequenced the surrounding region in 40 Tibetan and 40 Han individuals. The selected Tibetan haplotype closely resembled Denisovan DNA, supporting its introduction through interbreeding with Denisovans or a related population. Selection subsequently increased this inherited sequence in highlanders; EPAS1 is one component of adaptation, not a complete explanation of high-altitude physiology. (pubmed.ncbi.nlm.nih.gov)

Pigmentation provides geographically variable and partly convergent examples. Variants in SLC24A5 and SLC45A2 contribute substantially to lighter pigmentation in European populations, while different variants contribute in East Asia. A 2016 functional study linked the OCA2 His615Arg substitution, enriched in East Asian populations, to reduced melanin production, using human associations and experiments in zebrafish and mice. Similar visible outcomes thus need not share identical genetic histories. Pigmentation is polygenic, and the contributions of particular variants differ among populations; these local changes do not establish a hierarchy of evolutionary advancement. (pubmed.ncbi.nlm.nih.gov)

Inherited adaptation must also be distinguished from physiological acclimatization and cultural responses. Increased ventilation after ascent to altitude is an adjustment within an individual’s lifetime, whereas selection changes inherited variant frequencies across generations. Likewise, processing milk changes dietary exposure, while lactase persistence changes enzyme regulation. These mechanisms can operate together without being interchangeable: a cultural practice does not by itself demonstrate genetic adaptation, and a physiological response does not establish evolutionary change. (pmc.ncbi.nlm.nih.gov)

Research History, Scientific Limits, and Research Practice

In The Descent of Man, and Selection in Relation to Sex (1871), Charles Darwin applied common descent and evolutionary mechanisms to humans, drawing on comparative anatomy, development, and behavior rather than an extensive ancestral fossil record. African discoveries subsequently supplied evidence unavailable to him. In February 1925, Raymond Dart described Australopithecus africanus from the Taung juvenile skull, recovered in South Africa in 1924. Its small braincase and forward-positioned foramen magnum combined ape-like proportions with evidence of upright posture, challenging expectations that brain enlargement preceded other human characteristics. Later research increasingly integrated fossils with archaeology and molecular comparisons; the 2010 draft Neanderthal genome, for example, provided evidence of interbreeding that skeletal resemblance alone could not establish. (darwinproject.ac.uk)

These developments undermined representations of human ancestry as a single sequence of increasingly humanlike stages. Anatomical traits could evolve at different times, while branching populations could subsequently exchange genes. Nevertheless, identifying branches remains difficult: fossils often preserve only teeth or isolated skeletal regions, and deformation can alter diagnostic measurements. Differences between specimens may reflect age, sex, geographical variation, or distinct species, making fossil taxonomy partly dependent on the criteria adopted. Ancient DNA introduces another uneven sampling process because molecular survival depends strongly on temperature and preservation conditions. Missing genomes therefore do not establish that a population lacked descendants or genetic contact with other groups. (pmc.ncbi.nlm.nih.gov)

Field surveys and excavation establish geographical distributions and associations among remains, sediments, and artifacts; laboratory methods answer different questions. Computed tomography permits examination of internal anatomy and virtual reconstruction without physically dismantling a fossil. In 2005, Christoph Zollikofer and colleagues digitally reconstructed the distorted Sahelanthropus tchadensis cranium, correcting displacement and deformation to investigate its original shape. Proteomics can extend molecular investigation beyond specimens retaining usable DNA: a 2020 study recovered enamel protein sequences from approximately 800,000-year-old Homo antecessor remains at Atapuerca, Spain, and supported a close sister relationship to the lineage containing modern humans, Neanderthals, and Denisovans. Protein sequences provide fewer variable positions than whole genomes, so their resolution is not equivalent to genome-wide evidence. (humanorigins.si.edu)

Statistical reconstruction also has limits independent of specimen preservation. Different histories of population size, subdivision, and migration can produce similar genetic observations. In particular, population structure can generate patterns that, under a model assuming an unstructured population, resemble expansion or contraction. More sequence data may narrow uncertainty within an assumed model without establishing that the model is correct. Reproducible analysis therefore involves documenting sample selection, sequence processing, model assumptions, and software settings, alongside releasing data and analysis code where appropriate. Simulations and comparisons with alternative models test whether an inferred history reproduces observations beyond those used to estimate its parameters. (arxiv.org)

Human-remains research additionally involves stewardship of material that cannot be replaced. DNA and protein extraction may consume bone or tooth tissue, requiring the scientific value of an analysis to be considered alongside damage and future research possibilities. A 2021 guidelines paper, developed by researchers representing 24 countries, documented practices including advance research plans, compliance with applicable permissions, minimizing destructive sampling, data availability, and engagement with stakeholders from the beginning of a project. Its case studies described collaboration with source-country researchers and institutions, custodians, and descendant communities where relevant. Such collaboration encompasses study design, sampling decisions, interpretation, and communication rather than simply obtaining access to specimens. (pmc.ncbi.nlm.nih.gov)

References

1. About Human Evolution - The Australian Museumaustralian.museum
2. Early Homo at 2.8 Ma from Ledi-Geraru, Afar, Ethiopia - Arizona State Universityasu.elsevierpure.com
3. Homo sapiens - The Smithsonian Institution's Human Origins Programhumanorigins.si.edu
4. Apes: Hominoids - The Psychological Well-Being of Nonhuman Primatesncbi.nlm.nih.gov
5. Genetics - The Smithsonian Institution's Human Origins Programhumanorigins.si.edu
6. Sharing a common ancestor - The Australian Museumaustralian.museum
7. Mechanisms of Evolution – Environmental Biologyopen.lib.umn.edu
8. Genetic Drift - National Human Genome Research Institutegenome.gov
9. Introduction to Human Evolution - The Smithsonian Institution's Human Origins Programhumanorigins.si.edu
10. Walking Upright - The Smithsonian Institution's Human Origins Programhumanorigins.si.edu
11. Human Evolution Evidencehumanorigins.si.edu
12. Laetoli Footprint Trailshumanorigins.si.edu
13. Datinghumanorigins.si.edu
14. Taphonomy—Death & Decaynps.gov
15. About Fossilsnps.gov
16. Geologic Time: Radiometric Time Scalepubs.usgs.gov
17. The IntCal20 Northern Hemisphere Radiocarbon Age Calibration Curve (0–55 cal kBP)cambridge.org
18. The age of the hominin fossils from Jebel Irhoud, Morocco, and the origins of the Middle Stone Agenature.com
19. Initial sequence of the chimpanzee genome and comparison with the human genomenature.com
20. Mitochondrial genome variation and the origin of modern humansnature.com
21. Reconstructing human origins in the genomic eranature.com
22. Sequencing Y Chromosomes Resolves Discrepancy in Time to Common Ancestor of Males versus Femalespmc.ncbi.nlm.nih.gov
23. Rate of de novo mutations and the importance of father’s age to disease risknature.com
24. Inferring human population size and separation history from multiple genome sequencesnature.com
25. Ancient DNA analysisnature.com
26. Ancient DNA damagepubmed.ncbi.nlm.nih.gov
27. Reconstructing ancient genomes and epigenomesnature.com
28. Inference of human population history from individual whole-genome sequencesnature.com
29. On the importance of being structured: instantaneous coalescence rates and human evolution—lessons for ancestral population size inference?nature.com
30. Sequencing Y chromosomes resolves discrepancy in time to common ancestor of males versus females.escholarship.org
31. International Chronostratigraphic Chart, December 2024stratigraphy.org
32. Climate Effects on Human Evolutionhumanorigins.si.edu
33. Paleo-ENSO influence on African environments and early modern humanspmc.ncbi.nlm.nih.gov
34. Woody cover and hominin environments in the past 6 million yearsnature.com
35. Pleistocene climate variability in eastern Africa influenced hominin evolutionnature.com
36. Increased Ecological Resource Variability during a Critical Transition in Hominin Evolutiondigitalcommons.uri.edu
37. Ancient watercourses and biogeography of the Sahara explain the peopling of the desertpmc.ncbi.nlm.nih.gov
38. Sahelanthropus tchadensishumanorigins.si.edu
39. Geology and palaeontology of the Upper Miocene Toros-Menalla hominid locality, Chadnature.com
40. A new hominid from the Upper Miocene of Chad, Central Africanature.com
41. Nature and relationships of Sahelanthropus tchadensisdoi.org
42. Postcranial evidence of late Miocene hominin bipedalism in Chadnature.com
43. Earliest evidence of hominin bipedalism in Sahelanthropus tchadensispubmed.ncbi.nlm.nih.gov
44. Orrorin tugenensishumanorigins.si.edu
45. Orrorin tugenensis femoral morphology and the evolution of hominin bipedalismpubmed.ncbi.nlm.nih.gov
46. Ardipithecus kadabbahumanorigins.si.edu
47. Ardipithecus ramidushumanorigins.si.edu
48. The Ardipithecus Ramidus Skull and Its Implications for Hominid Originsoaks.kent.edu
49. Australopithecus anamensishumanorigins.si.edu
50. Australopithecus afarensishumanorigins.si.edu
51. Australopithecus africanushumanorigins.si.edu
52. Australopithecus sedibahumanorigins.si.edu
53. AL 288-1humanorigins.si.edu
54. Taung Childhumanorigins.si.edu
55. Paranthropus aethiopicushumanorigins.si.edu
56. Paranthropus boiseihumanorigins.si.edu
57. Paranthropus robustushumanorigins.si.edu
58. Diet of Paranthropus boisei in the early Pleistocene of East Africapmc.ncbi.nlm.nih.gov
59. Isotopic evidence of early hominin dietspmc.ncbi.nlm.nih.gov
60. Paleoanthropology. Early Homo at 2.8 Ma from Ledi-Geraru, Afar, Ethiopiapubmed.ncbi.nlm.nih.gov
61. Homo habilishumanorigins.si.edu
62. Reconstructed Homo habilis type OH 7 suggests deep-rooted species diversity in early Homonature.com
63. Homo rudolfensishumanorigins.si.edu
64. New fossils from Koobi Fora in northern Kenya confirm taxonomic diversity in early Homonature.com
65. Human evolution. Evolution of early Homo: an integrated biological perspectivepubmed.ncbi.nlm.nih.gov
66. Key human traits evolved at widely different timeshumanorigins.si.edu
67. New discoveries of Australopithecus and Homo from Ledi-Geraru, Ethiopiapmc.ncbi.nlm.nih.gov
68. Homo erectushumanorigins.si.edu
69. 3-million-year-old stone tools from Lomekwi 3, West Turkana, Kenyanature.com
70. Homo erectus, our ancient ancestornhm.ac.uk
71. KNM-WT 15000humanorigins.si.edu
72. Earliest human occupations at Dmanisi (Georgian Caucasus) dated to 1.85–1.78 Mapmc.ncbi.nlm.nih.gov
73. Postcranial evidence from early Homo from Dmanisi, Georgianature.com
74. Age control of the first appearance datum for Javanese Homo erectus in the Sangiran areacir.nii.ac.jp
75. The oldest in situ Homo erectus crania in eastern Asia: The Yunxian site dates to ~1.77 Mapmc.ncbi.nlm.nih.gov
76. Age of Zhoukoudian Homo erectus determined with 26Al/10Be burial datingnature.com
77. Last appearance of Homo erectus at Ngandong, Java, 117,000–108,000 years agonature.com
78. Evolution and dispersal of the genus Homo: A landscape approachdisperse-project.org
79. Evolution of Homo in the Middle and Late Pleistocenedoi.org
80. Nuclear DNA sequences from the Middle Pleistocene Sima de los Huesos homininseva.mpg.de
81. Homo neanderthalensishumanorigins.si.edu
82. Genetic history of an archaic hominin group from Denisova Cave in Siberianature.com
83. Homo floresiensishumanorigins.si.edu
84. A new species of Homo from the Late Pleistocene of the Philippinesnature.com
85. Previously unknown human species found in Asia raises questions about early hominin dispersals from Africanature.com
86. Homo naledihumanorigins.si.edu
87. Did Our Species Evolve in Subdivided Populations across Africa, and Why Does It Matter?pmc.ncbi.nlm.nih.gov
88. New fossils from Jebel Irhoud, Morocco and the pan-African origin of Homo sapiensnature.com
89. Age of the oldest known Homo sapiens from eastern Africanature.com
90. Pleistocene Homo sapiens from Middle Awash, Ethiopiain-africa.org
91. A weakly structured stem for human origins in Africanature.com
92. A weakly structured stem for human origins in Africapmc.ncbi.nlm.nih.gov
93. Did Our Species Evolve in Subdivided Populations across Africa, and Why Does It Matter?discovery.ucl.ac.uk
94. The earliest modern humans outside Africapubmed.ncbi.nlm.nih.gov
95. The earliest modern humans outside of Africaedisciplinas.usp.br
96. The great human expansionpmc.ncbi.nlm.nih.gov
97. Genomic analyses inform on migration events during the peopling of Eurasiapmc.ncbi.nlm.nih.gov
98. Climatic windows for human migration out of Africa in the past 300,000 yearsnature.com
99. Multiple hominin dispersals into Southwest Asia over the past 400,000 yearsnature.com
100. Homo sapiens reached the higher latitudes of Europe by 45,000 years agonature.com
101. DNA analysis of an early modern human from Tianyuan Cave, Chinapmc.ncbi.nlm.nih.gov
102. 40,000-Year-Old Individual from Asia Provides Insight into Early Population Structure in Eurasiapmc.ncbi.nlm.nih.gov
103. Earliest modern human genomes constrain timing of Neanderthal admixturenature.com
104. Early human settlement of Sahul was not an accidentnature.com
105. Stochastic models support rapid peopling of Late Pleistocene Sahulnature.com
106. A Draft Sequence of the Neandertal Genomepmc.ncbi.nlm.nih.gov
107. Quantifying the contribution of Neanderthal introgression to the heritability of complex traitsnature.com
108. Multiple episodes of interbreeding between Neanderthal and modern humansnature.com
109. Identifying and Interpreting Apparent Neanderthal Ancestry in African Individualspmc.ncbi.nlm.nih.gov
110. Initial Upper Palaeolithic humans in Europe had recent Neanderthal ancestrynature.com
111. The genome of the offspring of a Neandertal mother and a Denisovan fatherpmc.ncbi.nlm.nih.gov
112. Multiple Deeply Divergent Denisovan Ancestries in Papuansdoi.org
113. Philippine Ayta possess the highest level of Denisovan ancestry in the worldpmc.ncbi.nlm.nih.gov
114. The genomic landscape of Neanderthal ancestry in present-day humansnature.com
115. How Do We Know These Are Human Fossils?humanorigins.si.edu
116. The evolution of the human pelvis: changing adaptations to bipedalism, obstetrics and thermoregulationpmc.ncbi.nlm.nih.gov
117. Evolution of the human hip. Part 1: the osseous frameworkpmc.ncbi.nlm.nih.gov
118. Laetoli Footprints Preserve Earliest Direct Evidence of Human-Like Bipedal Biomechanicspmc.ncbi.nlm.nih.gov
119. Laetoli footprints reveal bipedal gait biomechanics different from those of modern humans and chimpanzeespmc.ncbi.nlm.nih.gov
120. Chimpanzee locomotor energetics and the origin of human bipedalism.pubmed.ncbi.nlm.nih.gov
121. Brainshumanorigins.si.edu
122. Australopithecus afarensis endocasts suggest ape-like brain organization and prolonged brain growthpmc.ncbi.nlm.nih.gov
123. The primitive brain of early Homopubmed.ncbi.nlm.nih.gov
124. The evolution of modern human brain shapepubmed.ncbi.nlm.nih.gov
125. Synaptogenesis and development of pyramidal neuron dendritic morphology in the chimpanzee neocortex resembles humanspmc.ncbi.nlm.nih.gov
126. Metabolic costs and evolutionary implications of human brain developmentpmc.ncbi.nlm.nih.gov
127. Social Lifehumanorigins.si.edu
128. D3444humanorigins.si.edu
129. The evolutionary origin of human hyper-cooperationnature.com
130. Cooperative breeding and human cognitive evolutionbennington.edu
131. Expanded geographic distribution and dietary strategies of the earliest Oldowan hominins and Paranthropusora.ox.ac.uk
132. Starch-rich plant foods 780,000 y ago: Evidence from Acheulian percussive stone toolspmc.ncbi.nlm.nih.gov
133. Early Stone Age Toolshumanorigins.si.edu
134. The Still Bay and Howiesons Poort at Sibudu and Blombos: Understanding Middle Stone Age Technologiespmc.ncbi.nlm.nih.gov
135. On the earliest evidence for habitual use of fire in Europepubmed.ncbi.nlm.nih.gov
136. Evidence for the cooking of fish 780,000 years ago at Gesher Benot Ya’aqov, Israelnature.com
137. Gesture links language and cognition for spoken and signed languagesnature.com
138. The evolution of early symbolic behavior in Homo sapienspmc.ncbi.nlm.nih.gov
139. What is cumulative cultural evolution?pmc.ncbi.nlm.nih.gov
140. Micro-Biomechanics of the Kebara 2 Hyoid and Its Implications for Speech in Neanderthalsjournals.plos.org
141. The derived FOXP2 variant of modern humans was shared with Neandertalspubmed.ncbi.nlm.nih.gov
142. The eloquent ape: genes, brains and the evolution of languagenature.com
143. Language & Symbolshumanorigins.si.edu
144. Emergence of modern human behavior: Middle Stone Age engravings from South Africapubmed.ncbi.nlm.nih.gov
145. Early Middle Stone Age personal ornaments from Bizmoune Cave, Essaouira, Moroccopubmed.ncbi.nlm.nih.gov
146. Early Middle Stone Age personal ornaments from Bizmoune Cave, Essaouira, Moroccovliz.be
147. Narrative cave art in Indonesia by 51,200 years agonature.com
148. Human Teaching and Cumulative Cultural Evolutionpmc.ncbi.nlm.nih.gov
149. Analysis of 6,515 exomes reveals the recent origin of most human protein-coding variantsnature.com
150. The contribution of gene flow, selection, and genetic drift to five thousand years of human allele frequency changepmc.ncbi.nlm.nih.gov
151. Evolution of lactase persistence: an example of human niche constructionpmc.ncbi.nlm.nih.gov
152. Convergent adaptation of human lactase persistence in Africa and Europepubmed.ncbi.nlm.nih.gov
153. Dairying, diseases and the evolution of lactase persistence in Europenature.com
154. Natural selection on EPAS1 (HIF2alpha) associated with low hemoglobin concentration in Tibetan highlanderspubmed.ncbi.nlm.nih.gov
155. Altitude adaptation in Tibetans caused by introgression of Denisovan-like DNAnature.com
156. Genetic evidence for the convergent evolution of light skin in Europeans and East Asianspubmed.ncbi.nlm.nih.gov
157. A Genetic Mechanism for Convergent Skin Lightening during Recent Human Evolutionpmc.ncbi.nlm.nih.gov
158. Natural selection on EPAS1 (HIF2α) associated with low hemoglobin concentration in Tibetan highlanderspmc.ncbi.nlm.nih.gov
159. Descentdarwinproject.ac.uk
160. Australopithecus africanus The Man-Ape of South Africanature.com
161. Dart and the Taung juvenile: making sense of a century-old record of hominin evolution in Africapmc.ncbi.nlm.nih.gov
162. Virtual cranial reconstruction of Sahelanthropus tchadensisnature.com
163. The dental proteome of Homo antecessornature.com
164. Demographic inference using genetic data from a single individual: separating population size variation from population structurearxiv.org
165. Ethics of DNA Research on Human Remains: Five Globally Applicable Guidelinespmc.ncbi.nlm.nih.gov