Life and Death at Lake Mungo

Lake Mungo

Archaeologically significant part of the Willandra Lakes World Heritage Area

Located 987 km west of Sydney via the Hume and Sturt Highways and 128 km north of Balranald, Mungo National Park is a 27847-ha archaeological and geomorphological site of world importance. Lake Mungo is one of 17 dry lakes which constitute the Willandra Lakes World Heritage Area, declared in 1981. The stark, silent, desolate and sometimes eerie landscape of sand, sparse but resurgent vegetation, and spiny, hard, pitted, crinkled and fluted dunes and ridges can look more like a moonscape.

25 000 to 45 000 years ago the lake covered 135 square kilometres and was about 10 m deep. It was one of a series of freshwater lakes along Willandra Creek, which was then a major branch of the Lachlan River. The lakes dried up about 14 000 years ago. They are, however, an extraordinarily rich source of fossils. Indeed the quality and quantity of evidence pertaining to the landforms, animal life and environmental conditions during the last ice age are of the highest calibre, in part due to the alkaline rather than acidic quality of the soils.

The remains of extinct creatures - Tasmanian tigers, giant, short-faced kangaroos and a strange oxen-sized animal called a zygomaturus - have been found. Crucially, carbon dating has indicated that Aborigines inhabited the area 40 000 years ago, making it the site of the oldest known human occupation in Australia. From the lake they gathered mussels, Murray cod and golden perch. They also hunted wallabies and rat kangaroos and collected emu eggs.

Findings of ochre in the area, dating back 32 000 years, constitute the earliest evidence in the Pacific Basin of the deliberate selection of pigments. As there was no local source it has been deduced that the material was carried there for aesthetic purposes. Moreover, a 28 000-30 000-year-old burial site reveals that the body was covered in red ochre. A 26 000-year-old grave contains the earliest known human example of cremation. After the ritual incineration the bones were smashed and deposited in a hole by the pyre. These practices clearly suggest the presence of spiritual considerations.

Convex flake tools made from local material dating back 20 000 years have been found, while sandstone grinders from 10 000 BP (before the present) or earlier suggest the inhabitants adapted to the arid conditions which later prevailed by grinding wild grass seeds, making them among the first people in the world to grind flour. The sandstone came from at least 100 km away, suggesting patterns of seasonal migration. A number of the finds indicate practices parallel with recent Tasmanian Aborigines.

Prior to being declared a National Park in 1979 this land was part of Mungo sheep station, created when the Gol-Gol station was subdivided in the 1920s for returned soldiers. It was named by the Cameron Brothers after a picture they saw of St Mungo's Church in Scotland. The park still contains a 45-m woolshed, built by Chinese labour of local pine logs in 1869. There are other buildings, including a former homestead, relating to the sheep station. Squatters first arrived with their sheep in 1840. Considerable conflict ensued with the indigenous tribes - the Barkindji, Ngiyampaa and Mutthi Mutthi, descendants of the area's ancient inhabitants. However, many were decimated by European diseases and forced to live on a mission at Balranald. Today they are involved in the management of the park and their wishes concerning the handling of their dead ancestors are now respected.

Today the vegetated dry lake basins are situated within a dunefield stabilised by mallee-type vegetation. Tall, steep escarpments abut the western perimeters of the lakes with crescent-shaped dunes called lunettes to the east, formed by quartz sands and pelletised clay, blown from the lake by the westerly winds. The most famous example of a lunette in the park is the the 'Walls of China' which rises to 30 m above the plain and runs for some 30 km around the old lake's eastern shore.

The area has been relatively free of clearing and pasture improvement although introduced animals destroyed the native vegetation, particularly along the Walls of China. Ironically it is this stripping of the flora which exposed the dune's top soil to erosion and hence uncovered the archaeological finds. Today the pre-European vegetation is returning. Birdlife is increasing, particularly pink cockatoos and chats and the striking mulga parrot. There are also kangaroos, emus and plenty of lizards.

Budget DVD Sale

539202_Free Ground Shipping 88x31

Click here for Jane's Oceania Home Page Newsletter for September/October 2011, November/December 2011, January/February 2012, and including our Special Christmas/New Year Edition!

Now - $50 Off the WINDLITE™ FZ FLEECE at! Limited Time Only!! Shop Mizuno Team Sports! Never Settle!

The first migrants to the Australian continent encountered a favourable environment. Once they penetrated inland, most of the fauna would have been unfamiliar, but would have presented little threat to the new arrivals. There are few carnivorous predators in Australian fauna. Those that existed then in mainland Australia were the native cat (Dasyurus), the Tasmanian 'tiger' or 'wolf' (Thylacinus cynocephalus), the Tasmanian 'devil' (Sarcophilus) and the marsupial 'lion' (thylacoleo), a leopard-sized carnivore, which may have been a predator or may have eaten only carrion. There would also have been crocodiles in the tropical rivers and some poisonous snakes, fish and insects, but these would have been familiar dangers to migrants from Asia.

In addition, a number of giant animals and birds (megafauna), which are now extinct, existed over much of Australia until the last phase of the ice age, which ended some 10 000 years ago. These included huge flightless birds and giant animals, such as a donkey-sized wombat, kangaroos 3 metres tall and the marsupial Diprotodon, which was the size of the rhinoceros. Some of these animals were relatively slow-moving herbivores and would have fallen easy prey to hunters. However, their extinction may have been caused by climatic change rather than by human overkill, or been a combination of both.

The tropical environment in coastal and riverine northern Australia would have provided the human colonists with largely familiar fish, shellfish, birds and plant foods. The arid-adapted flora further south would have been quite new, together with the marked seasonality of rainfall. Australia is also the world's driest inhabited continent; on over 75 per cent of its surface, rainfall is exceeded by potential annual evaporation. This was not always the case, for great changes in the Australian climate occurred during the ice age. One of the key areas where both climatic changes and early human occupation are documented is the semiarid belt of western new south Wales. As most of the earliest Australian skeletal remains, together wit very early habitational debris, come from this region, it seems appropriate for our survey of the early human fossil record to begin here.

Lake Mungo

Mungo III excavation (J. M. Bowler & A. G. Thorne in The Origin of the Australians, 127Š138;
Australian Institute of Aboriginal Studies, Canberra, 1976). Permission to use image kindly approved
by the Three Traditional Tribal Groups (Elders Council) of the Willandra Lakes World Heritage Area.

In 1968, Jim bowler, a geomorphologist at the Australian National University, was studying the nature of sediments to establish the pattern of climatic change over the past 10,000 years. His work was focused on the Willandra Lakes in western New south Wales, a series of interconnected lake basins carrying the waters of a tributary of the Lachlan River to the Murray (figure 3.1), and the following account is based on his research findings. The Willandra 'lakes' have been dry for the past 15 000 years, but once they had a surface area of more than 1000 square kilometers of fresh water. During the Pleistocne, there were long periods when western New South Wales, and Australia as a whole, had much more standing water than occurs today, and the Willandra Lakes were full of water, mainly because of the much lower evaporation rate caused by lower temperatures. The present barren landscape of the Willandra Lakes would have been very different 50 000 years ago, with lakes full of fresh water and teeming with large fish. The now dry bed of lake Mungo would have been 20 kilometres long and 10 kilometres wide, with a depth of some 15 metres. On its eastern side, sand dunes provided sheltered campsites by the lake shore.

Most research in the fossil lake system was concentrated on Lake Mungo, which has suffered extensive erosion of its lunette, the crescent-shaped dune formed on the lake shore, exposing 600 metres high. It is visible from several kilometres distant as a long, low, white hill among the flat, brown plains. Erosion has sculpted the lunette into such spectacular shapes that it was named 'the Walls of china' in the 1860s, possibly by Chinese workers from local sheep stations.

Budget DVD Sale

Lake Mungo

The earliest sediments in the area, called the 'Golgol sediments', were laid down perhaps 140 000 years ago, when the lake was drying. From then, until about 60 000 years ago, the lake was dry and the lake floor was covered with soil and vegetation, much as it is today. No evidence of human presence has ever been found in the Golgol sediments, and it seems that people did not begin to camp at Lake Mungo until the latest full-water phase. This commenced about 60 000 years ago and, by about 50 000, the lake was full, as a result of increased runoff from the Lachlan River catchment flowing down Willandra Crek and filling the Willandra Lakes. This full-water period has been named the 'Lower Mungo lacustral phase'. Freshwater shellfish and other aquatic fauna inhabited the lake, and many large trees grew around its margins; outlines of their branching roots have been fossilised and preserved by calcium carbonate. Waves driven by the westerly wind created a crescent-shaped sandy beach (a lunette) on the eastern lee shore. This dune consists of the Zanci, Arumpo, Mungo and golgol units, named after local pastoral properties.  

About 40 000 years ago, the water level fell (but never completely dried up0 in both Lake Mungo and OUter Lake Arumpo. Expanses of saline mud were exposed, which blew away to form clay-rich dunes. This episode has been called the 'Upper Mungo aeolian phase'. Increased salinity in the lakes may have weakened fish, which became easy prey for those camping around the shores. This period was short-lived and by 35 000 the lake was again full of fresh water. A period of fluctuating water levels followed, marked by alternating phases of beach and clay dune deposition, now named the Arumpo unit. From 345 000 to 22 000 the water level varied, but was never as deep as before. Then, around 22 000 years ago, the 'Zanci drying phase' began - the lakes gradually dried, and Lake Mungo disappeared about 19 000. The wind built up dunes of gypsum and saline clay on lee shores, with a period of maximum dune-building between 20 000 and 19 000. There was a buried resurgence of water between about 19 000 and 17 500 at lake Mulurulu in the northern end of the Willandra system. Thereafter, water never again flowed into the Willandra Lakes, and people and animals moved to use the more permanent channels of the Murray, Darling and other rivers as their new lifelines.


Among the stark residuals and shifting sands of the massive eroding dunes, Bowler came across the first exciting hint of early human presence at Mungo. Eroding out of a small midden in the upper part of the Mungo sediment, he found some stone artefacts and mussel shells bearing an encrustation of carbonate. The presence of large freshwater mussel shells in the dune and their association with artefacts is difficult to explain except by invoking human transport. Radiocarbon dating of these shells gave an age of 32 750 +- 1250 BP.

Bowler later noticed some burnt, carbonate-encrusted bones protruding from a low hummock on the dunes that clearly belonged to the Mungo sediments. He marked the site and left it intact for future archaeological excavation. A group from the Australian National University inspected the site in March 1969, and immediately suggested that the bones were human. The bones were contained in a quarter of a metre square calcrete block, which was only 15 centimetres thick and was fragmenting; many wind-eroded, broken pieces were scattered around. The features of the site were plotted, photographed and fully recorded; the loose bones were numbered and collected. Then the central carbonate block was undercut and removed in order to take the whole block back to the laboratory for closer analysis. The most secure container available was Professor John Mulvaney's suitcase, which he nobly emptied out to transport the precious finds safely back to Canberra.

Lake Mungo

The archaeologists returned to the site fully prepared for a major excavation just one week later, only to find that a freak rainstorm had totally changed the scene. If thy had left the Mungo skeleton there, it might have been washed away. But the storm had revealed hitherto hidden stone tools in the same area, and 200 carbonate-encrusted stone tools were collected. Also exposed were fifteen patches of black deposit. They were roughly circular or oval in shape 60 to 90 centimetres in diameter, and 5 to 10 centimetres deep. The black deposits contained charcoal, burnt animal and fish bones, freshwater mussel shells, emu eggshells and, in four places, stone artefacts. The black deposits seem to have been hearths, and their contents the discarded remains of human meals and tools associated with food collection and preparation. The human bones had lain about 15 metres away from the nearest hearth, and the whole area appears to have been a camp site on the lake shore. Here the ancient inhabitants camped, and roasted and ate their food. They used stone tools and burnt their dead on the sandy dune and beach a few yards from the shingle of the high freshwater lake shore. The people also collected raw material for tool-making and ochre pigment, consolidated earth made up of clay and hydrated oxide of iron. The ochre did not occur in the Mungo area, but must have been brought there from at least 200 kilometres away (the Barrier Range to the northwest).

Back in the laboratory, physical anthropologist Alan Thorne had begun the painstaking removal of the concrete-hard carbonate crust from 'Mungo Man', as the bones had been named. Reconstruction of the skull was a massive task, since it was broken into 175 small fragments. When Mungo Man finally emerged from this process, 'he' proved to be a 'she'. Mungo I, or Willandra Lakes Homind (WLH) 1 as she is now called, was a young adult female of slender build and small stature; she was 148 centimetres tall (4 feet 10 inches). Her head is very round in shape and her eyebrow ridges are small compared with the heavy, beetling brows of some archaic Australian skulls. Nontheless, she is one of the oldest human beings so far discovered in Australia, radiocarbon dating of burnt bone from WLH 1 gave an age of 24 710 +1270/-1100 BP (lab. no. ANU-618B, acid insoluble residues). Later accelerator mass spectrometer dates on the humic acid fraction in further samples of WLH 1 burnt bone gave dates of 25 120 +_ 1380 BP (NZA-230) and 24745+- 2400 BP (NZA-246). These dates seemed to confirm the age of the remains, which Bowler had estimated from their stratigraphic position (figure 3.3). However these dates on bone fragments are now considered unreliable, and WLH 1 is regarded as 'undated'.

Lake Mulakemungo sand

Careful analysis of the surface and fractures of the bones tells us that the corpse was first cremated, then the burnt skeleton, especially the face, was thoroughly smashed, and finally the ash and smashed bones were gathered together and deposited in a small depression beneath or adjacent to the cooled funeral pyre. This method of disposal of the dead was still in use among Australian Aborigines in historic times in eastern Australia and Tasmania. The WLH 1 site is possibly the oldest evidence of ritual cremation in the world. It is interesting that it is a woman who was cremated. Although no conclusions can be drawn from a sample of one, it at least shows that long, long ago women were considered worthy of complex burial rites. What emotions inspired those rites we will never know, but this cremation shows a concern for the deceased that is the essence of humanity. The presence of pellets of red ochre suggests the use of this pigment for ritual, art or decoration, and another nearby site has shown that ochre was used even earlier than WLH 1 in funerary rites. This is the site of another burial, but this time of a male, who was placed in a grave with his body thickly coated with red ochre. This Mungo III (now WLH 3) burial was discovered 500 metres east of the WLH 1 cremation site (figure 3.3). (the name WLH 2 is reserved for a very fragmentary, burnt, probably male hominid at the WLH 1 site.)

In 1974 bowler was examining an eroded area of the Mungo stratigraphic deposit when the slanting rays of the late afternoon sun highlighted a small, white object protruding 2 centimetres above the surface. It was part of a human skull. Heavy rain had made erode out of the lunette, and rapid excavation was essential because of the great fragility of the bones and the possibility of further rainstorms. Again, archaeologists from the Australian National University rushed to Mungo, and thee was elation when excavation gradually revealed not just fragments of bone or a skull, but a whole skeleton (plate 3). Finds of complete skeletons are extremely rare, and archaeologist Wilfred Shawcross recalls the excitement of the find. 'Two to three people worked flat out for two days. All the time you felt it couldn't go on; but it did. A neck appeared, then a rib. Normally you are lucky to get a skull; in Africa they are lucky to get a jaw. But this was a whole skeleton.

Sthenurus jawbone in situ, Cloggs Cave, Victoria. The remains of this
species of kangaroo lay at a depth of 2 metres in a layer older than 21 0000 years.

Excavation revealed a fully adult man, on his side in a shallow grave, with hands clasped. The bones and surrounding sand were stained pink; the pink colour, derived from ochre powder scattered over the corpse, clearly defined the size and shape of the grave. The burial lies within the lower Mungo sediments, and its age was estimated as about 30 000 years old. Since they first dated the corpse, bowler and others have significantly revised and extended their research, which was published in October 1998 in a special edition of Archaeology in Oceania entitled Willandra Lakes: People and Palaeoenvironments. Moreover, attempts to obtain direct dates on WLH 3 have finally succeeded with far-reaching results. Meanwhile, Stephen Webb's recent study of 130 Willandra Lakes hominids revealed that WLH 3 had most unusual tooth wear on his molars. Webb suggests these distinctive striations 'could be the result of the stripping of some sort of plant fibre with sharp inclusions (quartzose grains or phytoliths). It is tempting to suggest that these were figures for fishing nets, baskets or dilly bags. This fits very well with the evidence describe below from the study of fish remains that people were using net to catch fish in Lake Mungo in that time period. 

This waisted axe from Papua New Guinea was one of twenty-four flaked axes found
on the 80-metre Huon Terraces. It was discovered in a layer of volcanic ash dated to
about 40 000 years old, and is the earliest stone axe adapted for hafting yet found in the world.

WLH 3, who was probably about fifty at the time of death, had severe osteoarthritis in his right elbow, initially suggestive of 'spear-thrower elbow', but Webb concludes that is cause is uncertain, and it may have been an infection exacerbated by spar-throwing, with or without a spear-thrower. (this may therefore be used as evidence for the great antiquity of the use of spears, but not of spear-throwers, the antiquity of which is still uncertain.) WLH 3 had also lost his two lower canine teeth simultaneously when he was much younger. (Likewise, WLH 22 had lost his two lower central incisors many years before his death.) this pattern of loss is most unusual and may indicate that the teeth were removed in the rite of tooth avulsion. Among Holocene Aborigines, including Tasmanian, the removal of two front teeth formed a common part of initiation rites of young men, but it was two top teeth rather than lower ones which were removed. If these are actually cases of tooth avulsion, in a slightly different form from that practised more recently, this is the oldest instance of the practice anywhere in the world. It would also be further evidence for the great antiquity of Aboriginal religious beliefs.

The significance for this ochred burial is that it shows that such burial rituals go back at least as far in Australia as in other parts of the world, such as France, where ochred burials have been found in Grimaldi Cave at a similar time. In fact, at Mungo, red pigment was in use even earlier, for lumps of ochre and stone martefacts were found deep below the ashes of a fire lit 32 000 years ago. As the ochre did not occur naturally at Mungo, it must have been deliberately carried there from some distance away. Similar lumps of pigment, some of them showing signs of use, have been found in Pleistocene levels in their widely separated sites, such as Kenniff Cave in Queensland, Cloggs Cave in Victoria, Miriwun and Devil's Lair in Western Australia, and several Arnhem Land rock-shelters. Ochre has no utilitarian functions, such as medicinal use; it is simply a pigment used (at least in the recent past) to decorate rock walls, artefacts, dancers' bodies in ceremonies, and corpses during some burial rites. 

The manufacture and use of stone tools.

many camp sites, hearths and middens (prehistoric refuse heaps) have now been excavated in the Willandra Lakes region, and more than a hundred radiocarbon dates obtained. Many of these dated occupation sites are between 10 000 and 30 000 years old, but a considerable number also belong to the last 5000 years. No trace of human presence has been found in the Golgol sediments, so it looks as if humans only came to camp by these inland lakes when they last filled with water around 60 000 years ago. From then until about 25 000 years ago, occupation was intensive and the rich resources of the freshwater environment were fully exploited. Analysis of more than a hundred radiocarbon dates from hearths and shell middens in the region has revealed its continuous use during the last 40 000 years, with seventeen dates in excess of 30 000 years from around Lakes Mungo, Garnpung and Arumpo, and the Prungle Lakes.

The stratification and evidence from Cloggs Cave, Victoria

Continuing multidisciplinary work in the Lake Mungo area by archaeologists, geomorphologists, palaeontologists, and others has added considerably to these discoveries, although regrettably little of this work is yet published. Mungo has also become a National Park, with excellent displays set up in the Visitors Centre. In 1980 the Willandra Lakes region was entered in the World Heritage List as a place of outstanding universal value on the basis of both its natural and cultural heritage. Major excavations of potential human habitation sites were carried out in the 1970s. In 1973 Mulvaney opened up a large trench in the southern end of the Mungo lunette, not far from the WLH 1 cremation site. Below the light-coloured sediments of the Zanci unit lay the deep-brown 2-metre thick Mungo unit, in which a hearth was located. This gave a radiocarbon date of 31 000 BP +2250/-1750 BP. Sealed beneath this hearth were small pellets of ochre, which must have been carried to the site. And 1.8 metres below the hearth, in a gravel layer at the base of the Mungo sand unit, were a number of water-rolled stone flakes. A small sam0ple of shells associated with these gave a radiocarbon date of more than 40 000 BP.

Time chart of the last 3 million years of human occupation

The problem with the radiocarbon dating method is that 40 000 years is its theoretical limit; the carbon-14 in any samples older than that is indistinguishable from background. At this order of magnitude, contamination is a major problem and even the best charcoal samples are only 70 per cent pure carbon. The relationship between carbon-14 activity and age is logarithmic; beyond 30 000 years, a small difference in carbon-14 activity can produce a large difference in age. This means that the reliability of radiocarbon ages beyond 30 000 BP is open to question, and they may well be too young. In this situation, Mulvaney's date of 40 000 BP for shell associated with artefacts at lake Mungo was set on one side, in spite of the sample's impeccable stratigraphic context, but now, in view of the similar or older dates from the Northern Territory, new guinea and elsewhere which have been obtained more recently by other dating methods, it seems eminently credible that humans were camping by Lake Mungo 40 000 or more years ago. Early dates of other nearby sites support this vies. The shell from a thin, stratified midden of mussle shell, ash and charcoal on lake Outer Arumpo, excavated in 1975, was dated to about 36 000 years, although dates on the soil sediments gave a very different result. This midden was exposed in a gully wall, buried under 5 metres of lake shore sediments. A remarkable find at Lake Mungo, 5 kilometres northeast of the WLH 1 and WLH 3 sites, was a group of five fireplaces 26 000 or more years old. The oldest fireplace was a typical Aboriginal oven - a shallow depression that was filled with ash and charcoal, with several lumps of baked clay on top. This oven was dated to 30 780 +- 520 BP.

Diagrammatic cross section through the Keilor site, Victoria. (After Bowler 1976)

A surprise furnished by these fireplaces was the evidence they gave concerning palaeomagnetism" the phenomenon of fossil magnetism of archaeological material such as baked earth and clay from ancient fireplaces. The first Australian evidence for deviations in the earth's magnetic field was found in the early 1970s. The research revealed that 30 000 years ago magnetic north had swung right round 120 degrees to the southeast. This magnetic 'reversal' or 'excursion' lasted for about 2500 years, after which the direction of magnetisation reverted to normal again. This major magnetic reversal is now known as the Mungo excursion, and is one of the youngest and best-documented examples of such changes in polar magnetic direction. When such reversals in magnetism are found in other archaeological sites, they can be dated by comparison with the Mungo and other reversals. A series of thermoluminescence dates has been done on baked clay in Aboriginal fireplaces at Mungo. The thermoluminscence dates come out consistently older than the radiocarbon dates, and it appears that all radiocarbon dates on charcoal, at or earlier than about 30 000 BP, may be about 4000 years too young. Dates on freshwater mussel shell are the most reliable, but finely disseminated charcoal is often contaminated by younger organic material moving down through the sandy sediments, giving younger radiocarbon ages than samples of shells in the same layer.

Comparison of the Keilor skull with more robust skulls from Cohuna,
Talgai and Mossgiel. (All are male.) The forehead of the Cohuna skull has
probably been flattened by artificial deformation. (After Macintosh 1965)

Similar cooking methods seem to have persisted in Aboriginal society for over 30 000 years. Two types of Aboriginal fireplace were in use in the nineteenth century in the Willandra Lakes region: 'hearths', and 'ovens'. Hearths are small areas of blackened earth resulting from an open fire. They were probably used for roasting small animals and do not contain cooking stones. Ovens consist of a shallow depression or pit containing a band of ash and charcoal and cooking stones or lumps of baked clay. The use of such ovens in the region was described by the explore Edward Eyre in his Journals of Expeditions of Discovery, published in 1845 (vol.2, p.289):

The native oven is made by digging a circular hole in the ground, of a size corresponding to the quantity of food to be cooked. It is then lined with stones in the bottom or clay balls where stones are unavailable and a strong ire made over them so as to hear them thoroughly, and dry the hole. As soon as the stones are judged to be sufficiently hot, the fire is removed, and a few of the stones taken, and put inside the animal to be roasted if it be a large one. A few leaves or a handful of grass, are then sprinkled over the stones in the bottom of the oven, on which the animal is deposited, generally whole, with hot stones ... laid on top of it. It is covered with grass, or leaves, and then thickly coated over with earth, which effectually prevents the heat from escaping.


The artefacts recovered from the Willandra Lakes area give us some idea of the technological level that these Mungo people had attained. Very few bone tools have been found, but it is difficult to know whether this is because they were not present or because they have rotted away. Since the alkaline soils have preserved both large and small animal bones, it is unlikely that bone tools were numerous. Only three bone tools were found in the first decade of archaeological work at the Willandra Lakes. These are three pointed bone implements, all less than 10 centimeres long. Two ere found at the WLH 1 site, on the Alls of China, and one on the Lake Mulurulu lunette. One has been worked to a sharp point at both ends. It is possible that such bipoints were used as lures to catch the large Murray cod, bones of which were found at the same WLH 1 site. In the nineteenth century, Aborigines often caught Murray cod by attaching a fishing line to the middle of a bipointed bone and then pulling it rapidly through the water so that the bone looked like a small darting fish. These bones lures, or fish gorges, were called 'muduk' by Aborigines at the Murray River. This name has been adopted by some archaeologists for gone points, but is inappropriate since it is not certain hat all, or even any, of the prehistoric bone bipoints and unipoints were used as lures.

The Kow Swamp 5 skull compared with modern artificially deformed and underformed skulls.
Left: Midline cranial contours of Kow Swamp 5 (dashed line) and a modern Murray Valley male Aborigine.
Right: Midline cranial contours of an artificially deformed Arawe male (dashed line) and an underformed male from northern New Britain. (After Brown 198a)

A recent study of use-wear (wear produced on the working edge of a tool from use) on Pleistocene bone tools has identified the function of bone points as the piercing of dry skins in 'sewing', the scraping of skin, spearing of mammals and use as clothes toggles. In historic times, the uses were similar: spear tips, awls for piercing holes in animal skins being sewing together into cloaks, nose pegs, and pins for fastening cloaks. Some tools were far more plentiful than bone ones in the Willandra Lakes region. Most are made from silicified quartzite, a hard, fine-grained stone available locally. The major tool types are choppers and flakes. Choppers are large, heavy tools made from lumps of rock, and they have a flaked cutting edge. They are used for heavy woodworking, such as chopping down trees. A flake is a piece of stone formed when a lump of rock is struck with a hammer-stone. The force of the percussion blow detaches a flake of stone that has sharp edges and can be used to cur or scrape flesh, sinew or fur. Sturdy steep-edged flakes were also used for woodworking, such as scraping, sawing, incising and chiselling.

Cloggs Cave, Buchan, Victoria. The small black overhang on the
right protects the rock-shelter and the high cleft leads into the inner chamber.

These smaller tools are traditionally called scrapers, although they were not necessarily used for scraping. They may be made on a flake or on a core, a lump of rock from which flakes have been struck. Some rather specialised scrapers found at Mungo have deeply notched, concave working edges suitable for use as spokeshaves for smoothing wooden shafts such as spear shafts. The large tools made from lumps or nodules of rock are generally termed core tools. Some of them have a flat base, an overhanging, step-flaked edge and a high, domed shape like a horse's hoof, hence they have been called horsehoof cores. The tool kit from the Mungo cremation site has been described by Rhys Jones of the Australian Nation al University as the 'Australian core tool and scraper tradition'. This term has now been adopted Australia-wide for the early Australian stone industry. Its main characteristics are the presence of large cores of core tools, steep-edged, chunky, high-backed scrapers and concave, notched and 'nosed' working edges. Flatter, convex-edged and round scrapers also occur, which may have been used to make skins pliable for use as cloaks.

Mangrove log raft, Western Australia. A Worora youth paddles this raft on George water,
Glenelg River district. (In the Basedow Collection, National Museum of Australia, Canberra.)

These tools were used for the manufacture and maintenance of wooden tools rather than to extract food from the environment. In general, this industry is typologically similar to late Pleistocene and early Holocene tool kits found at the base of such sites as Kenniff Cave and Mushroom Rock in Queensland, Capetree and Burrill Lake in New south Wales, Ingaladdi and Malangangerr (but lacking the ground-edge axes) in the Northern Territory, and New Guinea highland sites such as Kafiavana. The tools also resemble both Plaistocene and Holocene industries of Tasmania, although the upper levels of some Pleistocene sites there , such as Kutikina, Nunamira and bone Cave, are dominated by tiny thumbnail scrapers. 

Kow Swamp skull 5 (male), Victoria, showing the massive and archaic features
of this group of robust, early Australians. The burial is about 13 000 years old.

Where river or beach cobbles were available, they were often made into chopping tools, termed 'pebble tools', but with the same function as the Mungo-type horsehoof cores. Similar industries based on flakes and pebble tools have been found at about 40 000 BP in southeast Asia, Sulawesi and at sites such as Tabon Cave in the Philippines and Niah Cave in Borneo. The Australian core tool and scraper tradition, although often amorphous and generalised, is distinctive when compared with stone tools elsewhere, but it also bears a general resemblance to the Middle Palaeolithic Mousterian industries of the Old World. There are no signs of the Upper Palaeolithic Mousterian industries of the Old world. There are no signs of the upper Palaeolithic revolution, except perhaps in the Holocene small tool tradition. Why are Australian Pleistocene industries so different? Was the tool kit of the earliest migrants into Australia a package from a source area? if so, where?

Kow Swamp skeleton 14 (male), during excavation. Quartz stone
artefacts and freshwater mussel shells were included in the grave fill. (A. Thorne)

These major questions remain to be answered by further research both in Australia and Asia by comparing the cultural content of the earliest dated cultural sites from both Greater southeast Asia and Greater Australia with the following conclusion:

There certainly appear to be resemblances between the stone artifacts at the different sites from both regions. It seems possible to suggest that these assemblages can be construed as being part of a common tradition, with local variation due to different environments, site usage and raw material availability. In the light of earlier opinion, it is interesting to note the general low frequency of pebble tools and other large core tools, with the exception of the Huon Terrace waisted tools. Otherwise these industries may be charcterized as somewhat amorphous, comprising ad hoc flake tools. There are however some types which occur in several sites, in both the Sunda and Sahul provinces, namely small steep edge-scrapers and thumbnail scrapers.

Necklace from the Lake Nitchie burial, New South Wales. The Lake Nitchie man wore a necklace of 178
pierced Tasmanian devil teeth, taken from at least forty-seven different animals. Each tooth is
pierced by a hole that was ground and gouged out. (The Australian Museum)

Over the millennia, new tools were invented to suit new uses in response to environmental change.  The drying-up of the Willandra Lakes was accompanied by an eventual change from a reliance on freshwater resources to an economy based on the exploitation of wild grass seed. The small, hard grass seeds were crushed to make flour by means of large, flat millstones or mortars. Considerable human activity continued in the region after the lakes had dried up, but it was not related to lake exploitation. Seed grindstones have been distinguished by Michael Smith from other types of grindstones on the basis of his extensive archaeological work in Central Australia. He identified the following varieties as seed grinders:

1. Millstones: flat surface slabs with shallow grooves worn on the grindstone face used as lower grinding surfaces for the wet milling of seeds.

2. Mullers: the upper grindstones used in the same process.

3. Mortars: flat surface blocks with a shallow oval or circular depression ground in on e or both faces, used for preliminary pounding and crushing of hard seeds.

Robust skull from Cossack, Western Australia, about 6500 years old
(L. Freedman and M. Lofgren, with permission from Academic Press, London)

Until recently it was believed that seed-grinding technology was introduced into the area about 17 000 years ago as a response to the decrease in aquatic resources from the drying of the Willandra Lakes. This theory was based on the apparent association of grindstones with sites of 17 000 years or younger, and their absence on older sites. However, re-examination of the archaeological evidence by Harry Allen and Jane Balme has shown that: (a) some of the grindstones are not seed-grinding stones; (b) there is an effective absence of seed-grinding stones in situ on the Pleistocene sites; (c) it is most likely that seed-grinders were introduced into the region in the early Holocene; and (d) seed-grinding was not a direct response to increasing aridity. However, a flat grindstone bearing use-polish and siliceous starchy residues has been found in a 30 000 BP layer of the Cuddie Springs site in New south Wales. It seems likely that the starch grains are from grass or wattle seeds. Whether or not archaeological evidence shows the occasional use of grass seeds in the Pleistocene, the extensive and intensive use of this food seems to belong to a much later period, of only a few thousand years ago.


The Willandra Lakes have provided evidence for the Pleistocene exploitation of a freshwater environment more than 40 000 years ago. As well as 'base' camps containing remains of many different creatures, 'dinnertime' camps have been found containing the remains of a lunch. The top Hut III midden is probably the remains of a single meal eaten by a small group around 40 000 years ago. Another such site, on the Lake Tandou lunette, has only the remains of 500 yabbies, a small freshwater crayfish. This dinnertime camp is dates to 30 000 years ago. And one Tandou site has been identified as a frog kill site!

Gracile skull from Lake Mungo, New South Wales, compared with Kow Swamp 1.
Three views of WLH 1 cranium A: left lateral' B: frontal; C: vertical; D: Kow Swamp 1. The thin bone,
rounded forehead and lack of brow ridges of WLH 1 are characteristic of gracile early Australians.

The number, size and species of fish remains in sites have been identified by comparing their otoliths, or ear bones, with those of modern fish in the same region. Seventy per cent of fish caught in the Pleistocene Willandra Lakes were golden perch (Petroplites ambiguus). The large numbers of perch at the sites, which dated between 22 000 and 26 000 BP and were each believed to result from a single event, came from tightly restricted size ranges, which strongly suggests the use of gill nets at some sites and traps at others. Fishing with fixed gill nets is a highly selective process: it tends to catch fish of the same species and age. Nets were probably set at the time of a spring spawning run, when the fish migrate up the rivers in large numbers. Golden perch are difficult to catch by other means, such as spearing, lines or poisoning, and if such methods had been employed, thee would be a far greater age range among the fish remains.

On the Darling River in the nineteenth century, Aborigines used to set nets 100 metres long about 20 metres offshore to catch golden perch. These nets were made from bulrush fibre (Typha) and had wide mesh in which to catch the fishes' gills. Other fishing methods were probably used for larger fish. Some bones have been found in middens of huge Murray cod (Maccullochella macquariensis), which were estimated to weigh as much as 15 kilograms. These were probably speared or caught with line and lure.

The diet of the hunter-gatherers at Lake Mungo was varied and rich in protein. As well as fish and mussels, they ate the rate kangaroo, the western native cat, the brown-haired wallaby, the hairy-nosed wombat and various other small animals and bird. Remains of these creatures have been found in ancient fireplaces, together with numerous broken emu shells. Their presence indicates that people were camping at Lake Mungo in the spring, when emu eggs hatch. In the heat of summer, people would have stayed close to the plentiful fresh water and shellfish of the lakes. In the cooler winter, they probably spread out away from the lakes onto the arid plains and hunted land animals, thus conserving the lake's food supplies for the harsh summers. Such a pattern of exploitation and seasonal movement is characteristic of Aborigines in arid regions, and was observed in the Willandra Lakes region in the nineteenth century.

Recent research by archaeologist Harvey Johnston of the New south Wales National Parks and Wildlife Service has shown that the Willandra shell middens are small, and the magnitude of importance of Pleistocene shellfish collecting thee may have been overestimated. Plants and land animals may have played a more important role than previously thought, for Loy's analysis of residues on some stone flake tools from the southern end of Lake Mungo has shown they were used for stripping meat from bone and for cleaning tubers, probably the native sweet potato, Ipomea polpha. Very few bones of extinct animals have been found in the Mungo region, and none in association with remains of human occupation. The few bones of the giant kangaroo, Procoptodon, found so far do not suggest big game hunting.


The Mungo evidence documents the most distant dispersal in the world of Homo sapiens sapiens and their achievement in adapting so successfully to a freshwater but semiarid environment. One of the most remarkable finds is the evidence that sophisticated fishing methods employing nets and traps were used in Australia as long as 30,000 years ago. An important implication is that the people fishing at Lake Mungo had the knowledge to make cordage out of plant fibres by this time period. The use-wear on the molars of Willandra lakes hominid 3 (WLH 3), dated to more than 30 000 years ago, supports this view. It is not now too far-fetched to suggest that their ancestors may have brought knowledge of plaiting fibres into ropes, string and even large mats for sails with them when they crossed the water barriers of Wallacea into the Australian continent more than 60 000 years ago. The Mungo sites also provide early evidence of intellectual life. The deliberate transport and use of coloured pigment more than 40 000 years ago parallels its contemporary use in Europe, indicating very early development of an aesthetic sense. This was to flower into the rich decorative and ritual art for which Aboriginal Australia is renowned.

The elaborate ice age cremation of a young woman is possible the earliest evidence for this rite in the world, and shows not only complex ritual concepts and respect for the dead, but also respect for women. Woman have always held an important place in Aboriginal society as gatherers of most of the staple food for the community. At Mungo, shellfish were clearly one of the most important foods, available all year round, and gathering shellfish was traditionally women's work. The archaeological evidence from the Willandra Lakes also reveals great cultural continuity in Aboriginal society from the Pleistocene to the present day. The ritual, symbolic and aesthetic concepts of modern Aboriginal society have their roots in the remote past. Methods of disposal of the dead such as cremation and inhumation have been used from earliest times right through to the ethnographic present. Not only is this complex culture found on the shores of Lake Mungo important in understanding the development of Homo sapiens in world prehistory, but it endows Aboriginal society with the dignity and respect it has often been denied.

There has been much debate concerning the gracility of the Mungo human remains, which contrast so strongly with the heavy build of most other Pleistocene and modern Aborigines. Indeed, it was very fortunate that the pelvis and femurs (thigh bones) of Mungo 3 survived, because they clearly showed that the remains, although so gracile, are definitely male. The differences between these gracile Mungo people and the robust Murray Valley population cannot therefore be explained just by sexual dimorphism, the differences in size and shape between males and females which occur naturally in any population. This issue has been re-examined with the result that ancient Australian Aboriginals outsized their recent counterparts, second, that shape differences can be distinguished between fossil and recent Australians; third, that 'crania order than 15 000 BP are either unusually small or large ("gracile" or "robust"), and these extremes somehow settled down to the intermediate size demonstrated by the fossil crania dated between 15 000 and 6000 BP'; fourth, that 'the much-vaunted division of Australian Pleistocene human crania into "robusts" and "graciles" boils down to a contrast among the Willandra Lakes specimens'; fifth, that this extreme variability among the Willandra Lakes crania cannot be explained simply by sexual dimorphism.

Bulbeck puts forward a model of biological evolution and adaptation within Australia, arguing that 'the smaller crania re-dated the larger crania and that cranial size and robusticity increased as a biological adaptation to climatic deterioration leading up to the Late Glacial Maximum'

Colin Pardoe has long been studying Aboriginal physical variability and its causes, and completed his PhD at the Australian National University in 1984 on 'Prehistoric human morphological variation in Australia'. Subsequently he has focuses on the remarkable biological and cultural diversity of southeastern Australia, particularly the river Murray with its cemeteries, which seem to have begun perhaps 13 000 years ago. Pardoe sees no reason to imagine two founding populations but emphasises evolution through adaptation:

The unique demographic make-up of peoples along the River Murray, coupled with the restraint of a linear environment, has been a major factor in their morphological evolution. River peoples differed from those in the hinterland in their population size and density, social structure, their age, sex and kin structures, patterns of mortality and morbidity, migration and gene flow patterns. Evidence from cranial, non-metric traits shows that such a system must result in regional similarity along a huge stretch of river and in extreme diversity within the region. Kow Swamp and Coobool Creek are not archaic, atypical members of this group. The represent the immense diversity engendered by their unique historical patterns of social organisation, demography and environment.

Australia - Aboriginal Sites

Australian Aboriginal Anthropology

Australian Aboriginal Music

Australia Postcards and Picture Galleries
Australia Home Page
Pacific Islands Radio
Please enter your email address for your free Pacific Islands Radio Newsletter!


Hosted By Topica















Jane's Oceania Home Page Newsletter
to get the latest news, information and Web site updates!
Please enter your email address below,
then click the 'Join' button
for your free Newsletter!
 Join newsletter! 
Search Jane's Oceania Travel Page Search the Internet

Oceania Time Zones
Oceania Weather

Back to Jane Resture's Oceania Page

Back to Jane's Oceania Home Page

(E-mail: -- Rev. 9th April 2015)
eXTReMe Tracker
eXTReMe Tracker