One of the most fossil-rich and well-known Pliocence sites in paleoanthropology, the Hadar Research area encompasses over 100 km2a located approximately 300 km northeast of Addis Adaba, Ethiopia. Hadar is named for the nearby Kada Hadar tributary that feeds into the the Awash River. The thin strip of greenbelt surrounding the Awash River is the only greenery seen among the otherwise arid badlands of rolling hills, desert scrub brush, and acacia trees.
The geologic record of Hadar may hold the keys to some of the most critical periods in human evolution. Fossils found at Hadar date between 3.4 and 2.3 million years ago, and document the existence of at least 2 fossils hominins that may sit in the direct evolutionary line to modern humans: Australopithecus afarensis and early Homo.
A large number of fossils are found in Afar due in large part to the fact that the Afar region is tectonically active. Hundreds of fossil localites have been identified at Hadar since since excavations beganin the late 1960. In 1976, Donald Johanson identified a 3 million year old (ma) fossilized knee that exhibited human-like morphology. Realizing the scientific importance of Hadar, Johanson and Dr. Maurice Taieb established the International Afar Research Expedition. By 1977, over 250 Australopithecus afarensis specimens had been discovered in the Hadar area. Between 1978 and 1990, research at Hadar stopped to allow for proper description of previously discovered fossil material. New work began in 1990 under the the Hadar Research Project.
Of the nearly 400 hominin fossils collected at Hadar, all but 1 represent A. afarensis. The taxonomic assignment of a 2.3 Ma maxilla discovered in 1994 remains in contention. Some researchers tentatively identified the specimen as Homo habilis, though others argue that the fossil’s geologic antiquity suggests the specimen should be assigned to an unidentified Homo species or to that of A. afraensis. Crude stone flakes discovered in association with the Homo specimen also suggest that Hadar may be home to some of the earliest evidence for stone tool use.
Australopithecus fossils have been discovered in South and East Africa locality, but over 90% of all A. afarensis material comes from Hadar. The importance of Hadar in the study of human evolution and human origins continues to grow as more scientific explorations take place. Fossils discovered at the site (A. afarensis and H. habalis) may represent major radiation events in human evolutions, including the proliferation of A. afarensis and the first appearance of Homo. Future discoveries are sure to increase our own understand of how we came to be human.
Two species possibly occupied Hadar beginning approximately 3.4 million years ago, and spanning almost a million years of geologic time.
A. afarensis
Hadar has been a fossil hot spot since Australopithecus afarensis was discovery in 1973. Since then, nearly 400 fossil specimens have been found, ranging in age between 3.4 Ma and 3 Ma. The first A. afarensis specimen, known as AL 129 (“Johanson’s Knee”) was
discovered by Donald Johanson while surveying the site in the early 1970’s. Subsequent invegitations, conducted by the International Afar Research Expedition (IARE), have resulted in a myarid of A. afarensis fossils, including a 3 Ma cache of ~13 individuals. The most famous A. afarensis specimen consists of a nearly complete skeleton known as AL 288-1, or “Lucy” that dates to ~3.2 Ma.
On November 24, 1974, Johanson and his research partner Tom Gray were out surveying and mapping possible fossils sites at hadar when they noticed a fragmented arm bone on the ground. Prompted by this discovery, the team began excavations at the locality known as AL 288. Later that night, the team returned to camp and were stunned to discovered they had unearthed one of the most primitive and complete fossil hominin skeletons every found. Over 40% of the skeleton, now identified as AL 288-1, was recovered, with nearly every bone in the body represented.
A.L. 288-1 was identified as an adault female, based largely on the shape of her pelvis, and the nearly complete femur indicated that she stood 107 cm (3’6’) tall. She was assigned to A .afarensis, a species first discovered at Laetoli, Tanzania in 1934. The genus name Australopithecus means “southern ape” and the species epithet afarensis refers to the local Afar region and culture. That night, while the Beatles song “Lucy in the Sky with Diamonds” played, AL 228 was lovingly referred to as “Lucy” for the first time. Lucy is also known as “Dinkenesh”, which means “you are beautiful” in the local Amharic language of Ethiopia. Today, Lucy is permanently housed at The Ethiopian National Museum in Addis Ababa, Ethiopia.
Because of the abundance of skeletal fragments from one individual, anthropologists are able reconstruct Lucy’s anatomy and hypothesize about her behavior as it might apply to all of A. afarensis. For example, Lucy’sremains include a nearly complete os coxae (hip bone), along with a fragmented femur and tibia. When held in anatomical position, the femur angles toward the midline of the body, forming a distinct carrying angle between the hip joint and the knee (known the bicondylar angle. This results in what is known as a valgus knee, which is an adaptation that centers the body weight over the center of gravity and aids in maintaining balance when standing and moving upright. Comparatively, a quadrupedal chimpanzee’s knees are straight, or almost perpendicular to the ground. Since the modern anatomy of Lucy’s lower limb suggested that, while on the ground, Lucy walked bipedally, her discovery served to confirm the bipedal nature of A. afarensis, and pushed back the estimated evolution of bipedalism from 1.8 Ma to at least 3.2 Ma, if not older.
A little less well-known, but no less remarkable, was the discovery of the AL 333 cache, an assemblage of A. afarensis fossils commonly referred to as the “First Family”. On November 1, 1975, a team lead by Johanson recovered several fossil lying on the surface of the hillside. The team tirelessly sifted through the food deposits over the next 3 years, which resulted in the identification of over 200 A. afarensis fossils from numberous individuals. The AL 333 locality produced bones from nearly every part of the body: phalanges, metatarsals, cranial fragments, teeth, ribs, vertebrae, arm, and leg bones. When analyzed, anthropologists found the cache consisted of at least 9 adults, 2 juveniles, and 2 infant individuals.
Anthropologists suspect that the AL 333 group may have been caught in a flash flood, which would account for such a diverse assemblage of individuals in one place. However, the AL 333 remains exhibit inconsistent inter-group preservation. For example, while some specimens show prefossilization damage such as weathering (indicating surface deposition), other in situ fossils remained relatively articulated which suggests rapid burial prior to soft tissue disintegration. There is no evidence that carnivores were responsible for the AL 333 assemblage, since non-hominin faunal is not associated with the hominin remains. In addition, AL 333 does not include more individuals of one age than another, so the assemblage is less likely to be the results of natural attrition. If the hominins were buried at the same time, then the assemblage may constitue a single hominin familial unit.
Like AL 333, Lucy, and AL 129, most hominin specimens from Hadar are surface finds, and all but one have been identified as A. afarensis. AL 444 (~3 Ma) represents the most complete and largest A. afarensis cranium found to date; AL 417 is a 3.25 Ma intact female face with mandible, and AL 438 preserves a male mandible, frontal bone, and partial upper limb skeletons (~3 Ma).
The diversity of A. afarensis specimens available from Hadar helped to answer questions about the ontogeny of early hominins, including relative brain size, sexual dimorphism, diet, and social structures. In turn, this information helps scientists develop a greater understanding of hominin evolution. The ~400 Ka A. afarensis was present at Hadar infers period of evolutionary stasis, during which time the species retained a mosaic of primiative and derived features. For example, although A. afarensis was clearly bipedal when on the ground, the species also has a relatively small cranial capacity that is comparable to a modern chimpanizee (~415 cc), pronounced subnasal prognathism, a well-developed temporal-nuchal crest, and large canines. A. afarensis also exhibits curved phalanges, a trait common in arboreal primates. These features are reduced in australopiths compared to chimpanzee, but the retention of these features may suggest that A. afarensis continued regularly practice arboreal locomotion in addition to some form of bipedalism.
H. habilis
On November 2, 1994, Ali Yesef and Maumin Allahendu located bone and tooth fragments, several stone flakes, and non-hominin mammalian bones at the bottom of a gully at locality AL 666 in the Makaamitalu drainage basin. Among these finds was a fragmented maxilla that articulated together perfectly when reassembled. Dated to ~2.3 Ma, the maxilla was believed initially represent Australopithecus afarensis. Today, AL 666 has been cautiously assigned to the species Homo habilis, based on presence of several derived traits. This designation does not come without controversy, however, and many researchers prefer to assign AL 666 to either A. afarensis or to an unspecified Homo species based on the specimen’s antiquity and locality.
When discovered, the maxilla had only been recently exposed to the surface, and was relatively well preserved. Further excavation recovered isolated dentition that, upon examinination, could be easily rearticulated into the maxilla. In total, almost of AL66’s canines and cheek teeth were recovered. The relatively large size of the maxilla and associated teeth suggest that AL 666 was most likely male. The eruption of the third molar and wear patterns present indicate the individual was probably a mature adult.
The contentious assignment of AL 666 to Homo habilis is based largely on its morphological differences from Australopithecus, Paranthropus, and later Homo species. Compared to the rectangular U-shaped maxilla and tapered midface of australopiths, the AL 666 midface is relatively squarer, and the dental arcade is relatively more parabolic. Although the overall maxillary length is similar to australopith, the AL 666 maxilla is broader and deeper than the condition seen in more primitive hominins (though not as deep as later Homo). Other derived features include moderate subnasal prognathism, slightly flatter incisors, more narrow first molars, and thinner molar enamel. AL 666’s derived features are homolgous to later Homo species, such as H. ergaster, though these conditions, such as subnasal prognathism, often appear relatively reduced in later Homo.
Many anthropologists argue, however, that AL 666 cannot be considered a member of H. habilis since too many of the maxilla’s traits fall outside of the range of variation seen in H. habilis. For example, AL 666’s dental arch and fourth premolar are relatively primitive, and measure slightly larger than H. habilis fossils from Olduvai Gorge and Koobi Fora. Counterarguments suggest that the seemingly extensive differences between AL 666 and other H. habilis may be a simple matter of intra-species variation, with AL 666 represnting the more primitive condition. Since there is a temporal gap of nearly 400 Ka between AL 666 and the next oldest H. habilis specimen at ~1.9 Ma (KNM ER 1813), then primitive trait retention would not be unexpected. The recovery of more Homo specimens of the same geological age as AL 666 are needed to better understand the variation between AL 666, A. afarensis, and H. habilis.
No artifacts have been associated with A. afarensis at Hadar.
The archaeological record at Hadar begins at locality AL 666, which dates to 2.33 Ma. Approximately 20 lithics, consisting of flakes and bifacial choppers were discovered on the surface in close association with a fragmented hominin mandible in 1994. The surface flakes were remarkably well-preserved and still had sharp edges, indicating recent surface exposure.
In total, 34 artifacts were found at the AL 666 locality, including 14 lithics recovered from the Kada Hadar member, the uppermost strata of the Hadar geologic formation. All the artifacts (as well as the hominin maxilla) were found in close proximity to one another, clustered within a 90 x 80 cm area in a layer no more than 9 cm thick.
Many of the AL 666 artifacts exhibit simple striking platforms and preserved cortical material, all fetures consistent with Oldowan tools, the earliest form of stone tool technology. While the assemblage included at least 3 bi-facial choppers, most of the material constituted flakes typically used for food processing. However, the flakes show no evidence of use or retouching. The lithics were most likely manufactured from rounded chert and basalt river cobbles. In at least 1 instance, excavators rearticulated a flake onto its original core.
Even though the Hadar flakes appeared unused, the AL 666 assemblage also contained several large mammalian bones, which suggest the site might have been used for butchering. In addition, small striae found on a bovid scapula have been interpreted as cut marks made when the muscle was cut from the bone by a stone tool.
The lithic and mammalian assemblage at Hadar indicates that stone tools were manufactured at Hadar by at least 2.3 Ma. The 2.6 Ma stone tools discovered at nearby Gona currently represent the oldest known tool assemblages. The AL 666 lithics, however, represents the oldest associated between hominin and faunal remains. It’s not possible to determine if maxilla found at the site represents an individual who actually produced the stone tools, but the presences of these lithics does imply that hominins had a precision grip by 2.3 Ma. It also implies some form of culture, since stone tool production was probably a learned behavior.
The Hadar hominins lived in a taxonomically rich habitat influenced by oscillations between wet and dry environmental episodes between 4-2 Ma. Along with the early hominins, Hadar’s Plio-Pleistocene inhabitants thrived in a variety of ecological zones that included tropical forest, wetlands, and dry grasslands. A few species are present throughout the entire Plio-Pleistocene, while others species appear only during excessive wet or dry periods. Tracking changes in the frequency of stenotopic species can assist in a holistic reconstruction of ancient habitats, including speciation events and climatic fluctuations either on a local or regional level. Folivores and frugavores were attracted to the open woodlands, bushlands and wooded grasslands present during the relatively low seasonality wet periods that existed approximately 3.42 – 3.26 Ma. Animals whose diets consisted of rich grasses, leaves and branches include various antelope species (e.g., impalas, dwarf antelope, duikers, and extinct forms similar to modern wildebeests and waterbucks), giraffes, gazelles, and the extinct three-toed horse. Large mammals include extinct elephant species, black and white rhinos, and several extinct forms of suids (i.e., pigs). Present during periods of flooding were aquatic animals such mollusks, crabs, and cane rats that live in thick reeds near standing water, along with crocodiles, otters, and hippopos typically found in deep fresh water. Carnivores such as hyenas, jackals, and predatory cats (e.g., servals and an extinct form similar to the saber-toothed cat) likely moved into the area following their prey. Among the many birds, reptiles and rodents, also present were dietary specialists such as the root-eating porcupines, and fruit-eating primates such as the colobus monkey, extinct baboons, and A. afarensis.
Between 3.2 and 2.9 Ma, Hadar’s climate was relatively dry, initiating an increase in arid-adapted populations that roamed open grasslands, such the antelope, gazelle, buffalo, bison, and other forms of wild cattle. The dry conditions also opened up new ecological niches for species not previously seen at Hadar, such as burrowing animals like rabbits, hares, badgers, mongoose, and an extinct rodent similar to the modern mole rat. In addition to the hyenas and saber-toothed cats seen in previous depositional periods, new predators like the panther were introduced to Hadar during this period. Just prior to 2.9 Ma, colobus monkeys and baboons return to Hadar.
Comparing the frequency, or change in frequencies, of species present at certain times can help to identify climate shifts. For example, a decrease in the abundance of arboreal feeders, such as the colubus monkey and Parapapio (an extinct baboon species), coupled with an increase in terrestrial grazer populations between 4 and 2.9 Ma may signal a general shift from wet wooded areas to drier, more open habitats such as grasslands. Indeed, by ~3 Ma species typical of wooded habitats with low seasonality (e.g, impalas and colobus monkeys) are almost completely absent from the fossil record, as are wetland fauna such as waterbucks (Reduncins). Meanwhile, populations of arid-adapted grazers and mixed feeders such as antelopes and wildebeests (Alcelaphini and Antelopini respectively) increase substantially over the same period.
Despite these environmental changes, several eurytopic species are occupied Hadar during all depositional periods, including an extinct African elephant (Elephas recki), the carnivorous striped hyenas (Crocuta dietrichi), and the mixed feeder giraffe (Sivatherium marusium). Among those species present prior to 2.9 Ma are several wild pig species (Kolpochoerus afarensis and Notochoerus euilus), wild cattle (Ugandax coryndoni), the three-toed horse (Eurygnathohippus hasumense), an extinct large-bodied baboon-like primate (Theropithecus darti), and A. afarensis.
Approximately 3.42 Ma, A. afarensis encompasses over 30% of the total large mammalian faunal sample and then drops in frequency to approximately 15% by 3.3 Ma. The A. afarensis population continues to decrease, making up only about 5% of the faunal sample (based on minimum number of individuals) between 3.26 and 3.2 Ma. From the time of Lucy at 3.2 Ma, until the species disappears from Hadar around 2.9 Ma, the A. afarensis population comprises less than 10% of faunal assemblage. The consistent presence of A. afarensis fossils in deposits indicating all types of habitats suggests that the species was eurytopic, meaning that A. afarensis was adapted to a wide variety of environmental conditions. Faunal and paleoecological evidence suggests that the A. afarensis population drops during wet periods such as the SH depositional period as a result of flooding, and that population increases may related to a drying-out period, such as during the Denen Dora period. Scientists have not been able to ascertain a clear reason for the disappearance of A. afarensis at Hadar after 2.9 Ma due to gaps in the geologic record.
Nearly all mammalian species present at Hadar prior to 2.9 are replaced by other species, possibly a result of migrations, extinctions, or speciation events trigged by severe climatic or geological shifts that effected food supplies (i.e. changes in vegetation). Hadar began to ardify around 2.9 Ma, and by 2.0 Ma more arid-adapted and fresh-grass grazers were present at Hadar than at any other time. A high percentage of edaphic grass feeders migrated into Hadar for the first time, including antelopes, camels, and several extinct species of gazelles and wild cattle. If the AL 666 maxilla does represent early Homo, then Homo habilis appears at Hadar for the first time beginning around 2.3 Ma.
The Hadar site is named for deposits exposed along the Kada Hadar and Ounda Hadar tributaries of the Awash River. These deposits are notable for the Pliocene fossil record contained within them. The site is located in the Afar depression, an area where three rifts come together: the East African Rift, the Red Sea Rift, and the Gulf of Aden rift. As a result of this junction, there is considerable tectonic and volcanic activity in the area.
The geological formation at Hadar, represented in the diagram on the left, is comprised mainly of silty clays and sands, deposited by rivers and, when a lake was present, along shores and deltas. The formation is divided into four members. Over the Basal Member sits the volcanic Sidi Hakoma Tuff, dated to 3.40±0.03 Mya. The Sidi Hakoma Member lies above the Sidi Hakoma Tuff and ends at Triple Tuff 4 (3.22±0.01 Mya). A layer of Kadada Mournou Basalts (3.28±0.04 Mya) gives further resolution to the Sidi Hakoma member, which contains the hominin localities A.L. 137 and A.L. 417. Next, the Denen Dora Member lies between Triple Tuff 4 and the Kada Hadar Tuff (3.18±0.01 Mya). The Denen Dora Member contains hominin locality A.L. 333. Finally, the Kada Hadar Member is the best studied section of the Hadar Formation. It contains Bouroukie Tuff 2 (2.92±0.03 Mya) and is topped by Bouroukie Tuff 3 (2.33±0.07 Mya).
The Kada Hadar Member contains notable hominin fossil localities A.L. 288, A.L. 438, A.L. 444, and A.L. 666 (with associated artifacts). During the period when this member was deposited, a lake expanded across the Hadar study area a total of three separate times. These events are partially responsible for the sandy fluvial deposits in the Kadar Hadar Member. The flooding and receding cycles of this lake also affected the environment in which hominins such as A. afarensis were living.
Between 3.45 and 2.9 Ma, Hadar’s environment remained a mosaic of open and closed woodland and bushland habitats, with various expansions and regressions of dry open grasslands and wet closed forests (i.e., woodlands). On at least three occasions, a large lake expanded (i.e. transgressed) across the area, leaving behind mollusks and reed pollen in the fossil record. The changing habitats influenced by climatic and geologic conditions, also attracted a variety of animals that migrated in and out of the area as environmental conditions changed, contributing to the subsistence of several stenotopic species.
Hadar’s current vegetation consists of plants adapted to hot, arid environments such as acacia trees and desert scrub brush, but 4 – 3.5 Ma, Hadar was covered in open and closed wooded areas, including a large open flat floodplain covered with lush grasses and an expansive lake surrounded by a riparian forest. The currently badlands topography results from an ancient river system that sustained a wider range of floral and faunal species than those that currently exist at Hadar. The trees and shrubs found in Hadar’s paleoenvironment were adapted for warm, wet conditions, while the grasses and underground storage organs (i.e., roots and tubers) present during later depositional periods, around 2.9 Ma, were adapted for more arid and hotter climates. For the majority of the Hadar formation, most of the grasses were probably sustained from water sources such as lakes and wetlands. After 2.9 Ma, the previously low occurrence of C4 grasses.
During most of the Miocene (23.8 – 5.3 Ma), much of Africa was covered in thick tropical forests, but at the Miocene-Pliocene boundary around 5 Ma, the general climate of the earth was moving toward a cooling event. In Africa, this meant a reduction in heavily wooded areas, such as tropical forests, and the spread of more open woodlands, savannas, and grasslands. In the early Pliocene, Hadar was still relatively closed and wet, meaning that the environment consisted of heavily wooded areas and received large amounts of rainfall each year, probably resulting from strong monoons that form regularly off the east coast of Africa each year. In fact, between 3.42 and 3.35 Ma, Hadar received a higher than average amount of rainfall resulting in less seasonal weather and the expansion of wet habitats such as humid woodlands, wetlands, and lake margins.
During the mid Pliocene, between 3.4 and 3.22 Ma, Hadar’s habitat was a mosaic of mostly dry and riparian woodlands, shrublands, and a few open grasslands. At some point during this period, between 3.35 and 3.26 Ma, the wet climate aided in the development of a major wetland event, including at least two major lake transgressions and regression in the Hadar floodplain. Paleolake Hadar formed amidst a dry open woodland and savanna habitat, and its marshy borders were full of reeds, sedges, and mollusks.
Hadar’s climate continued to fluctuate between wet and arid periods, until the end of the Denen Dora depositional period (~3.2 Ma) when Hadar’s environment began a general drying trend, resulting in more seasonality (i.e., a rainy season), and more dry open woodlands and wooded grasslands. By 3 Ma, Hadar’s once tropical rainfall was reduced to only 31 – 35 inches (80-90 cm) per annum, whereas today’s average tropical rainforest receives approximately 100 inches of rain per year. Between 2.9 and 2.33 Ma, Hadar appeared as a flat open plain covered with savanna grasslands with small patches of trees around small isolated water sources. Fresh grasses and USO were the major sources of vegetation attracting many arid-adapted grazing animals such as antelopes and other bovines, which in turn attracted large numbers of carnivores.
After the final regression of Paleolake Hadar, the major water source became the Awash River system, and riverine forest and wetlands became more prominent by 3.22 – 3.18 Ma. The A. afarensis population increases dramatically during this time, including the famous “Lucy” individual. Animals adapted for wetland environments disappear from the fossils record shortly after 2.9 Ma, indicating that Hadar’s climate became increasingly arid. By the time Hadar entered another wet period resulting in a large wetland across the area, the fauna common during previous depositional periods was nearly complete replaced by new species.
Climate changes and major shifts in habitats may have influenced migration patterns, as evidenced by a near complete absence in Busidima fossil record of large bodied mammals common in the Hadar Formation (> 2.9 Ma) within the Busidima fossil record (<2.35 Ma), including the replacement of A. afarensis with early Homo. No A. afarensis fossils dating younger than 2.5 Ma have been found at Hadar. Prior to 2.5 Ma, A. afarensis is found in all depositional period between the upper Basal and the upper Kada Hadar members, indicating that A. afarensis was euytopic rather than a habitat specialist, an advantageous characteristic at Hadar given the climatic fluctuations.
While other animals affected by environmental changes would migrate in and out of the area, A. afarensis could stay in one place and exploit the changing habitat. In fact, A. afarensis fossil have been identified within a variety of ecological niches, from open wet woodlands to relatively drier open shrublands. On the other hand, the early Homo specimen exisited during a time of open grassland, when Hadar was known to be more arid. However, the exact association between the disappearance of A. afarensis, the advent of Homo, and Hadar’s changing habitats remains unknown as a large section of sediments are missing from the geologic record that could otherwise help reconstruct Hadar’s habitat during that time (2.9 – 2.35 Ma).
Aronson JL, Hailemichael M, and Savin SM. 2008. Hominid environments at Hadar from paleosol studies in a framework of Ethiopian climate change. Journal of Human Evolution 55:532-550.
Aronson JL, Schmitt TJ, Walter RC, Taieb M, Tiercelin JJ, Johanson DC, Naeser CW and Nairn AEM. 1977. New geochronologic and paleomagnetic data for the hominid-bearing Hadar formation of Ethiopia. Nature 267:323 - 327
Campisano, C.J. & Feibel, C.S. Depositional environments and stratigraphic summary
of the Hadar Formation at Hadar, Afar Depression, Ethiopia. In: The Geological Context
of Human Evolution in the Horn of Africa (J. Quade and J.G. Wynn, Eds.). Geological
Society of America Special Paper 446, pp. 179-201
Campisano CJ, and Feibel CS. 2007. Connecting local environmental sequences to global
climate patterns: Evidence from the hominin-bearing Hadar Formation, Ethiopia. Journal
of Human Evolution 53:515-527
Campisano CJ and Feibel CS. 2008. Depositional environments and stratigraphic summary
of the Hadar Formation at Hadar, Afar Depression, Ethiopia. In: The Geological Context
of Human Evolution in the Horn of Africa (J. Quade and J.G. Wynn, Eds.). Geological
Society of America Special Paper 446:179-201
Johanson DC and Edey MA. 1981. Lucy: The Beginnings of Humankind. New York: Simon an Schuster.
Johanson D and Edgar B. 2006. From Lucy to Language: Revised, Updated, and Expanded. New York: Simon and Schuster.
Johanson D and Taieb M. 1976. Plio-Pleistocene homoninid discoveries in Hadar, Ethiopia. Nature 260:293-297.
Klein RG. 2009. The Human Career: Human Biological and Cultural Origins, Third Edition. University Of Chicago Press. Reed KE. 2007. Paleoecological patterns at the Hadar hominin site, Afar Regional State, Ethiopia. Journal of Human Evolution 54: 743-768.
Walter RC. 1994. Age of Lucy and the First Family: Single-crystal 40Ar/39Ar dating of the Denen Dora and lower Kada Hadar members of the Hadar Formation, Ethiopia. Geology 22:6-10
