ARTICLE IN PRESS

Quaternary Science Reviews 24 (2005) 2167–2172

Viewpoint

What can the data on late survival of Australian megafauna

tell us about the cause of their extinction?
Christopher N. Johnson
School of Tropical Biology, James Cook University, Townsville Qld 4811, Australia
Received 8 October 2004; accepted 17 June 2005

Abstract

[Roberts et al. 2001a. New ages for the last Australian megafauna: continent-wide extinction about 46,000 years ago. Science 292,
1888–1892] concluded that the extinction of Australia’s late Pleistocene megafauna was an abrupt event that took place about 46 ka
ago throughout the continent. By placing the extinctions soon after the arrival of people in Australia but before climate changes
associated with the Last Glacial Maximum, this study implicated human impact as cause. However, the study was controversial
because it excluded evidence, mainly from archaeological sites containing disarticulated megafauna remains, for survival of some
megafauna well past 46 ka ago. Here, I ask how our interpretation of the extinctions would be changed if this evidence were
accepted. Contrary to climate-change models of extinction, the young megafauna sites are not concentrated in mesic refuges around
the coast. These sites do suggest that relatively small-bodied megafauna species were the last to disappear, as predicted by a version
of the overkill hypothesis. More work is needed to test the evidence for late survival of megafauna species in Australia, but at present
this evidence supports overkill, not climate change, as the cause of the extinctions.
r 2005 Elsevier Ltd. All rights reserved.

1. Introduction claims have been made for an earlier human presence

but these remain controversial). This was roughly mid-
Close to the end of the Pleistocene, Australia and way through the last glacial cycle, and well before the
New Guinea lost perhaps 60 species of mammals, birds onset, between 30 and 20 ka, of extreme drought
and reptiles, including all of the largest forms. What conditions around the LGM (Barrows et al., 2002;
caused these extinctions has been debated for over a Hesse et al., 2004). Evidence that the extinctions
century. One view is that people were responsible, either happened at this time would strongly implicate human
because they hunted the megafauna to extinction or arrival; a concentration of extinctions in the interval
because their widespread use of fire destroyed the 30–20 ka ago would support climate change as the cause.
habitat of megafauna species. The major alternative is Roberts et al. (2001a) obtained ages for megafauna
the idea that climate change—specifically, Australia’s remains in 27 sites from Australia and one from New
entry into a period of deep aridity around the Last Guinea, choosing sites in relatively young geomorpho-
Glacial Maximum (LGM)—caused the extinctions, with logical contexts in order to focus on the final occur-
little or no contribution from human impact. rences of extinct species. They used optical dating of
Determining the timing of these extinctions is critical sediments surrounding fossils to determine their burial
to inferring their cause. Recent archaeological dating ages, and distinguished sites with articulated skeletal
has confirmed that people were widespread in Australia remains from those in which remains were disarticu-
by 45 ka (Gillespie, 2002; O’Connell and Allen, 2004; lated. They based their conclusions only on sites with
articulated remains to avoid the problem that re-
Tel: +61 7 4781 4141; fax: +61 7 4725 1570. working of sediments may have caused remains to be
E-mail address: christopher.johnson@jcu.edu.au. re-deposited in sediments younger than the fossils

0277-3791/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.quascirev.2005.06.006
 ARTICLE IN PRESS
2168 C.N. Johnson / Quaternary Science Reviews 24 (2005) 2167–2172

themselves. The youngest age obtained by Roberts et al. ages less than 46 ka. Their critics object that this
for articulated megafauna remains was 46 ka. Statistical criterion for selection of sites ignored other evidence
modelling of the distribution of ages gave an estimate of that should have been used to determine whether
46.4 ka for the extinction event (see also Brook and sediments containing megafauna remains had been re-
Bowman, 2002). This has been supported by dating of worked, and defended the integrity of the Cuddie
megafauna from Wet Cave at Naracoorte in South Springs stratigraphy in particular (Field and Fullagar,
Australia showing disappearance of megafauna slightly 2001; see rejoinders by Roberts et al., 2001b and
before 45.3 ka (Pate et al., 2002), and is consistent with a Gillespie and David, 2001, and subsequent work by
study by Miller et al. (1999), who used a combination of Trueman et al., 2005).
methods to age eggshell of the extinct giant bird Megafauna remains in archaeological sites are very
Genyornis newtoni in the arid Lake Eyre region of likely to be disarticulated because of handling by people;
south-central Australia and showed that it was common to exclude sites with disarticulated remains is, therefore,
in the area between 130 and 50 ka, but had disappeared effectively to exclude all archaeological sites. However,
by 45 ka. such sites might be expected to provide the best
information on the duration of overlap of people and
megafauna. Most archaeological sites in Australia are
2. The controversy over megafauna ages younger than 45 ka, so their exclusion from the analysis
closes off a potentially rich source of data on late
Roberts et al. were criticised because they had survival of megafauna. Brook and Bowman (2002) re-
excluded evidence for megafauna survival well past analysed Roberts et al.’s data and found that sites with
46 ka (Field and Fullagar, 2001; Wroe and Field, 2001; disarticulated remains were significantly younger (med-
Wroe et al., 2004). This evidence comes mainly from ian age ¼ 37.5 ka) than sites with articulated remains
archaeological sites, and at one of these—Cuddie (median ¼ 63 ka).
Springs—there is substantial stratigraphic overlap of Table 1 lists sites at which now-extinct megafauna
cultural material and megafauna, indicating at least have been dated to less than 46 ka. The list excludes sites
6000 years of coexistence (Field and Dodson, 1999). where late survival was originally claimed but where
Roberts et al. obtained ages younger than 46 ka for subsequent work has shown the dating to be erroneous
megafauna-bearing sediments at Cuddie Springs, but (note that this explains the absence of some sites, such as
discarded them because the remains were disarticulated, Lancefield Swamp (Joe Dortch, pers. commun.), that
as they did for 11 sites in all, including another four with have figured prominently in the extinction debate in the

Table 1
Sites at with evidence possibly indicating survival of megafauna after 46 ka in Australia and New Guinea

Site Latest age (ka)a Speciesb Reference

Mooki River 42 T. carnifex, Palorchestes azael, Phascolonus gigas, Diprotodon Roberts et al. (2001a, b)
optatum
Devil’s Lairc,d 30.7 Procoptodon brownoreum, Balme et al. (1978)
26.5 Protemnodon brehus
Tight Entrance Cave 33 Simosthenurus occidentalis Roberts et al (2001a, b)
Cuddie Springsc 30 Diprotodon optatum, Sthenurus andersoni, Procoptodon sp., Trueman et al. (2005)
Genyornis newtoni
Lake George 28.7 Procoptodon goliah Sanson et al. (1980)
Lake Tandouc 29.6 Procoptodon sp. (probably goliah) Hope et al. (1983)
Cloggs Cavec 28 Simosthenurus occidentalis Flood (1974)
Lake Menindeec 31 Propleopus oscillans, Phasocolonus gigas, Sthenurus sp., Tedford (1955), Tindale (1964)
Procoptodon sp., Protemnodon sp., Diprotodon sp.
Lime Springsc 24 Diprotodon sp., Protemnodon sp., Procoptodon sp., Stehenurus sp. Gorecki et al. (1984)
Kelangurr Cave 16 Maokopia ronaldi, Protemnodon hopei Roberts et al (2001a, b)
Seton rock shelterc 20 Procoptodon gilli Hope et al. (1977)
Main Cave, Montagu 13 Simosthenurus occidentalis, Zaglossus sp. Goede and Bada (1985)
Nombe Rock shelterc 14.5 Protemnodon nombe Flannery et al. (1983)
11.5 Protemnodon tumbuna
a
Radiocarbon dates calibrated using the University of Cologne Radiocarbon Laboratory on-line age converter (www.calpal-online.de/). Where a
stratigraphic age range was given, the youngest age is shown.
b
Not including species considered to be large late Pleistocene forms of living species (Macropus titan, Dendrolagus noibano etc).
c
Sites with human occupation.
d
Refers only to layers above those also dated by Roberts et al.’s (2001a, b).
 ARTICLE IN PRESS
C.N. Johnson / Quaternary Science Reviews 24 (2005) 2167–2172 2169

past). All ages apply to disarticulated remains. There are Both features meant that a small rate of killing could
currently 14 such sites, and at 10 of these there is have driven populations to extinction, but might have
evidence of human occupation, including almost all the done so over a long period of time. In this model, the
younger ones, with stratigraphic overlap with mega- species with the greatest demographic sensitivity to over-
fauna remains at five (Devil’s Lair, Lake Menindee, harvest—generally, the largest animals—would have
Cuddie Springs, Lime Springs and Nombe Rock Shelter; gone extinct soonest, while smaller-bodied species with
overlap has also been reported at Keilor in Victoria higher growth rates would have persisted longer. So, this
(Duncan, 2001) and Lembudu Cave, Aru Islands hypothesis predicts that (1) extinctions may have been
(O’Connor et al., 2002), but without precise dates). spread over a long time, and if so, large-bodied species
Two questions are raised by these young ages. First, would have gone extinct earlier than small-bodied
are they reliable? And second, what would it mean for species, but (2) there should have been no spatial
our understanding of the late Pleistocene extinctions if pattern in the timing of extinctions, because most of
they were correct? So far, the controversy has focused them would have happened well after the human
on the first question. Thus, the debate between Field and population occupied the whole of the continent.
Fullagar (2001), Wroe and Field (2001) and Trueman et
al. (2005) on the one hand, and Roberts et al. (2001b) 3.3. Extinction due to increased aridity
and Gillespie and David (2001) on the other, turned on
the evidence for and against re-working of sediments at It has often been argued that increased aridity close to
Cuddie Springs. Eventually we will need a systematic the end of the Pleistocene may at least have contributed
evaluation of the evidence for re-working of sediments to the megafauna extinctions. Horton’s (1984,2000)
at all of the sites listed in Table 1, and where possible model of this process is the most explicit. He argued
validation of ages by direct dating of bone. This may that as aridity expanded from the centre of the continent
well show that many of the ages listed in Table 1 are too towards the coasts, habitats suitable for megafauna
young, but in this paper I sidestep the problem of the species retreated and were compressed and fragmented
reliability of the post-46 ka ages and tackle the second of around the margins of the continent. The area of habitat
the two questions stated above. My goal is to test available to megafauna species would have been further
whether the published data indicating late survival of restricted by their need to remain within range of
megafauna would provide evidence for or against drinking water. Compression of populations into small
particular hypotheses of cause, if those data were areas of refuge habitat would have caused local
accurate. extinctions, and the repetition of these events at many
locations eventually resulted in the total extinction of
species.
3. What if the post 46 ka ages are right? This hypothesis predicts that local extinctions would
have happened first in the centre of the continent but
Three hypotheses of the cause of megafauna extinc- that some species would have persisted near the coast in
tions are sufficiently explicit to make clear predictions habitat refuges for some time after they disappeared
on their timing and pattern, as follows. from the arid centre, and that the losses would have
been spread over a period of several thousand years in
3.1. Overhunting by blitzkrieg the approach to the LGM.

Flannery (1990,1994) argued that Australia’s late

Pleistocene megafauna species were easy for people to 4. Fit of predictions to data on post-46 ka survival of
kill because they were both naı̈ve to hunters and slow- megafauna
moving, and were wiped out in a brief episode of
massive over-harvest. The hypothesis therefore predicts Fig. 1 plots all ages less than 46 ka that have been
that, on a millennial time scale, extinction would have claimed for the extinct Pleistocene megafauna, against
been effectively (1) instantaneous and (2) simultaneous estimates of body mass for those species. These show a
across the continent. clear relationship between body mass and the prob-
ability of late survival—of the species that went extinct,
3.2. Over-hunting by attrition the smaller ones evidently persisted longest. This effect
produces a significant positive relationship between
Murray (1991) and Johnson (2002) assumed that body mass and age (F 1;20 ¼ 9:92, P ¼ 0:005). The
people hunted megafauna but did not specialize on relationship is triangular rather than linear, and is
large-animal hunting, but that large-bodied species were defined by the presence of a lower bound: small species
susceptible to human hunting because of their low can have early or recent ages, whereas ages for large
population growth rates and low population density. species are predominantly early. I tested the significance
 ARTICLE IN PRESS
2170 C.N. Johnson / Quaternary Science Reviews 24 (2005) 2167–2172

50 on the other hand mesic coastal habitats might have

45 been more drastically altered by fire and the pace of
40 extinction may have been quicker in those habitats.
Age (000’s years)

35
Susceptibility to fire-induced habitat changes would
J
presumably have depended heavily on details of the
30
ecology of individual species, with habitat specialists
25
more vulnerable. Since large-bodied species are often
20 more general in their habitat requirements than small
15 species, it would seem difficult to explain the pattern
10 shown in Fig. 2 as an effect of fire-induced habitat
5 change. Therefore, these results give a poor fit to the fire
10 100 1000 10000 hypothesis.
Body mass (kg)

Fig. 1. Relationship between body mass and dates on remains of 5. A reconciliation of the conflicting ages
species of extinct Pleistocene megafauna from Australia and New
Guinea, for remains o46 ka. Data on body mass taken from Johnson
and Prideaux (2004). Data sources are listed in Table 1; megafauna
Roberts et al. (2001a) claimed that their dating study
occurrences were included only where identification was to species. documented continent-wide extinction at 46 ka. If even
some of the ages in Table 1 were eventually accepted as
reliable, then it would seem that this conclusion was
50 incorrect. But the conclusion that articulated remains
45 disappear from the fossil record by 46 ka appears sound,
Age (000’s of years)

40
and agrees with the disappearance of Genyornis egg-
35
shells documented by Miller et al. (1999). It seems that
30
25
something did change 46 ka ago. In fact, it may be
20 possible to reconcile the Roberts et al. and Miller et al.
15 studies with the evidence from archaeological sites for
10 post-46 ka survival, as follows.
5 Articulated megafauna remains are found in two
0 kinds of sites: deep caves that acted as pitfall traps, and
0 100 200 300 400 500 600
deposits around the edges of lakes, swamps or rivers.
Distance from coast (km) The latter kind of site preserves remains of animals that
Fig. 2. Relationship between distance to coastlines and youngest dates were initially trapped in mud and submerged in shallow
on extinct Pleistocene megafauna at sites in Australia, for sites dated water so that their remains were not disturbed by
o46 ka; species and sites as listed in Table 1. scavengers. Both kinds of sites acted as passive traps,
and would have preserved animals in proportion to their
abundance in the surrounding environment. The prob-
of this lower bound by calculating the regression using ability that an animal will be preserved as a fossil in one
only the youngest age for each species (F 1;14 ¼ 5:75, of these situations and eventually recovered by a
Po0:05), and this was used to find the position of the palaeontologist is very low, so that low-density popula-
lower bound in Fig. 1. Fig. 2 shows that there is no tions of megafauna might well be invisible in the fossil
tendency for sites with more recent megafauna remains record from such sites. Therefore, abrupt disappearance
to be found nearer to the coast. The regressions on of articulated remains could represent either the abrupt
distance from the coastline of the youngest dates for extinction of those species, or sudden declines in their
sites with megafauna was not significant (F 1;10 ¼ 0:06, abundance to levels that determined a very low
P ¼ 0:89). probability of recovery of fossils. This reasoning also
Some authors have argued that human use of fire may applies to the Genyornis eggshells dated by Miller et al.
have contributed to megafauna extinctions (Jones, 1968; (1999). They dated shell fragments from non-archae-
Merrilees, 1968; Miller et al., 1999), but this hypothesis ological aeolian deposits; these collections would have
does not make such clear predictions on the spatial accumulated passively with low probabilities of pre-
pattern and timing of extinctions. Presumably, if fire- servation and recovery, and they provide low power to
induced habitat changes had caused the extinctions the detect populations at low abundance. The signal in
process might have been asynchronous throughout the Miller et al.’s data can also be interpreted either as an
continent. One could argue that the drier habitats of extinction event or a sudden decline in abundance.
inland Australia might have been the first to be affected In contrast, megafauna remains in archaeological sites
so that extinctions would have begun earliest there; but are very likely to have been actively accumulated, by
 ARTICLE IN PRESS
C.N. Johnson / Quaternary Science Reviews 24 (2005) 2167–2172 2171

people. Archaeological collections may therefore have extinctions and increases in aridity. Whatever view one
much greater power to detect species present in the takes of the post 46 ka ages, it seems clear that
environment at low abundance, if such species were megafauna populations declined dramatically over the
being actively sought out by hunters. So, the ages whole continent between about 50 and 46 ka ago, and
provided by Roberts et al. (2001a, b) and Miller et al. direct human impact is the factor most likely to have
(1999) on the one hand, and the dates from archae- caused this.
ological sites indicating more recent survival on the
other, are both consistent with the following scenario:
before 50 ka, large vertebrates were abundant in the Acknowledgements
Australian environment, but something drove their
abundances down to drastically low levels by 46 ka, I thank David Bowman, Barry Brook, Gavin
such that they became effectively invisible in the non- Prideaux, Bert Roberts and Stephen Wroe for helpful
archaeological fossil record. Some species persisted at comments on the manuscript.
low abundances, perhaps in restricted geographic
ranges, after this event, with final extinction of remnant
populations spread over a comparatively long period; References
these later events might be visible only in the archae-
Balme, J., Merrilees, D., Porter, J.K., 1978. Late Quaternary mammal
ological record.
remains, spanning about 30,000 years, from excavations in Devil’s
This scenario is consistent with evidence from the Lair, Western Australia. Journal of the Royal Society of Western
Pleistocene archaeological record as a whole that, if Australia 61, 33–65.
megafauna survived past 46 ka ago, they were very rare. Barrows, T.T., Stone, J.O., Fifield, L.K., Cresswell, R.E., 2002. The
There are over 160 Pleistocene archaeological sites in timing of the last glacial maximum in Australia. Quaternary
Australia and New Guinea (Smith and Sharp, 1993), but Science Reviews 21, 159–173.
Brook, B.W., Bowman, D.M.J.S., 2002. Explaining the Pleistocene
very few contain megafauna remains and most of these megafaunal extinctions: models, chronologies, and assumptions.
contain very small numbers of megafauna fragments. Proceedings of the National Academy of Science of the USA 99,
The scenario outlined above also agrees with a recent 14624–14627.
modelling study of the effects of human hunting of Brook, B.W., Bowman, D.M.J.S, 2004. The uncertain blitzkrieg of
Pleistocene megafauna. Journal of Biogeography 31, 517–523.
Pleistocene megafauna, by Brook and Bowman (2004).
Duncan, J., 2001. Megafauna at Keilor and the timing of their
They found that the imposition of human predation on extinction. Australian Archaeology 53, 16–22.
megafauna populations resulted either in declines to Field, J., Dodson, J., 1999. Late Pleistocene megafauna and
extinction in less than 1000 years, or in declines followed archaeology from Cuddie Springs, south-eastern Australia. Pro-
by persistence at very low abundance. Populations ceedings of the Prehistoric Society 65, 275–301.
driven to low abundance by the initial impact of hunting Field, J., Fullagar, R., 2001. Archaeology and Australian megafauna.
Science 294, 7a www.sciencemag.org/cgi/content/full/294/5540/7a.
would be very susceptible to any subsequent variation in Flannery, T.F., 1990. Pleistocene faunal loss: implications of the
hunting pressure, and would be at high risk of ultimately aftershock for Australia’s past and future. Archaeology in Oceania
going extinct. These two phases of response could 25, 45–67.
possibly be reflected in the evidence for general and Flannery, T.F., 1994. The Future Eaters. Reed, Melbourne.
sudden population declines at 46 ka, and prolonged Flannery, T.F., Mountain, M.J., Aplin, K., 1983. Quaternary
kangaroos (Macropodidae: Marsupialia) from Nombe Rock
persistence at low density for many species that Shelter, Papua New Guinea, with comments on the nature of
eventually went extinct. megafaunal extinction in the New Guinea Highlands. Proceedings
of the Linnean Society of New South Wales 107, 75–97.
Flood, J., 1974. Pleistocene man at Cloggs Cave: his tool kit and
environment. Mankind 9, 175–188.
6. Conclusion
Gillespie, R., David, B., 2001. The importance, or impotence, of
Cuddie Springs. Australasian Science 22, 42–43.
Acceptance of published ages of less than 46 ka for Gillespie, R., 2002. Dating the first Australians. Radiocarbon 44,
megafauna in Australia would change our interpretation 455–472.
of the cause of extinction of the megafauna, but only by Goede, A., Bada, J.L., 1985. Electron spin resonance dating of
replacing blitzkrieg with a more general interpretation Quaternary bone material from Tasmanian caves—a comparison
with ages determined by aspartic acid recemization and C14.
of how over-hunting by humans caused extinctions. Australian Journal of Earth Sciences 32, 155–162.
Human hunting remains the most likely cause because Gorecki, P., Horton, D.R., Stern, N., Wright, R.V.S., 1984.
the order in which species appear to have gone extinct Coexistence of humans and megafauna in Australia: improved
correlated with their relative susceptibility to over- stratigraphic evidence. Archaeology in Oceania 19, 117–119.
harvest. Taken as a whole, the studies indicating post- Hesse, P.R., Magee, J.W., van der Kaavs, S., 2004. Late quaternary
climates of the Australian avid zone: a review. Quaternary
46 ka survival provide strong evidence against climate International 118–119, 87–102.
change as a cause of the extinctions, because there was Hope, J.H., Lampert, R.J., Edmondson, E., Smith, M.J., van Tets,
no correlation in space or time between the pattern of G.F., 1977. Late Pleistocene faunal remains from Seton rock
 ARTICLE IN PRESS
2172 C.N. Johnson / Quaternary Science Reviews 24 (2005) 2167–2172

shelter, Kangaroo Island, South Australia. Journal of Biogeogra- (Eds.), Bridging Wallace’s Line: The Environmental and Cultural
phy 4, 363–385. History and Dynamics of the Southeast Asian–Australasian region.
Hope, J.H., Dare-Edwards, A., McIntyre, M.L., 1983. Middens and Catena Verlag, Cremlingen, Germany, pp. 279–306.
megafauna: stratigraphy and dating of Lake Tandou Lunette, Pate, F.D., McDowell, M.C., Wells, R.T., Smith, A.M., 2002.
Western New South Wales. Archaeology in Oceania 18, 38–45. Last recorded evidence for megafauna at Wet Cave, Naracoorte,
Horton, D., 1984. Red kangaroos: last of the Australian megafauna. South Australia 45,000 years ago. Australian Archaeology 54,
In: Martin, P.S., Klein, R.G. (Eds.), Quaternary Extinctions: A 53–55.
Prehistoric Revolution. The University of Arizona Press, Tucson, Roberts, R.G., Flannery, T.F., Ayliffe, L.K., Yoshida, H., Olley, J.M.,
pp. 639–680. Prideaux, G.J., Laslett, G.M., Baynes, A., Smith, M.A., Jones, R.,
Horton, D., 2000. The Pure State of Nature. Allen & Unwin, Sydney. Smith, B.L., 2001a. New ages for the last Australian megafauna:
Johnson, C.N., 2002. Determinants of loss of mammal species during continent-wide extinction about 46,000 years ago. Science 292,
the late Quaternary ‘megafauna’ extinctions: life history and 1888–1892.
ecology, but not body size. Proceedings of the Royal Society of Roberts, R.G., Yoshida, H., Flannery, T.F., Ayliffe, L.K., Olley, J.M.,
London B 269, 2221–2227. Prideaux, G.J., Laslett, G.M., Baynes, A., Smith, M.A., Jones, R.,
Johnson, C.N., Prideaux, G.J., 2004. Extinctions of herbivorous Smith, B.L., 2001b. Archaeology and Australian megafauna: reply.
mammals in Australia’s late Pleistocene in relation to their feeding Science 294, 7a www.sciencemag.org/cgi/content/full/294/5540/7a.
ecology: no evidence for environmental change as cause of Sanson, G.D., Riley, S.J., Williams, M.A., 1980. A late Quaternary
extinction. Austral Ecology 29, 553–557. Procoptodon fossil from Lake George, New South Wales. Search
Jones, R., 1968. The geographical background to the arrival of man in 11, 39–40.
Australia and Tasmania. Archaeology and Physical Anthropology Smith, M.A., Sharp, N.D., 1993. A revised bibliography of Pleistocene
in Oceania 3, 186–215. archaeological sites in Australia, New Guinea and island Melane-
Merrilees, D., 1968. Man the destroyer: late Quaternary changes in the sia. In: Smith, M.A., Spriggs, M., Fankhauser, B. (Eds.), Sahul in
Australian marsupial fauna. Journal of the Royal Society of Review. The Australian National University, Department of
Western Australia 51, 1–24. Prehistory, Research School of Pacific Studies, Canberra,
Miller, G.H., Magee, J.W., Johnson, B.J., Fogel, M.L., Spooner, N.A., pp. 283–312.
McCulloch, M.T., Ayliffe, L.K., 1999. Pleistocene extinction of Tedford, R.H., 1955. Report on the extinct mammalian remains at
Genyornis newtoni: human impact on Australian megafauna. Lake Menindee, New South Wales. Records of the South
Science 283, 205–208. Australian Museum 11, 299–305.
Murray, P., 1991. The Pleistocene megafauna of Australia. In: Vickers- Tindale, N.B., 1964. Radiocarbon dates of interest to Australian
Rich, P., Monaghan, J.M., Baird, R.F., Rich, T.H. (Eds.), archaeologists. Australian Journal of Science 27, 24.
Vertebrate palaeontology of Australasia. Pioneer Design Studio Trueman, C.N.G., Field, J.H., Dortch, J.H., Charles, B., Wroe, S.,
& Monash University, Melbourne, pp. 1071–1164. 2005. Prolonged coexistence of humans and megafauna in
O’Connell, J.F., Allen, J., 2004. Dating the colonization of Sahul Pleistocene Australia. Proceedings of the National Academy of
(Pleistocene Australia-New Guinea): a review of recent research. Science of the USA 102, 8381–8385.
Journal of Archaeological Science 31, 835–853. Wroe, S., Field, J., 2001. Mystery of megafauna extinctions remains.
O’Connor, S., Aplin, K., Spriggs, M., Veth, P., Ayliffe, L.K., 2002. Australasian Science, 21–25.
From savanna to rainforest: changing environments and human Wroe, S., Field, J., Fullagar, R., Jermiin, L.S., 2004. Megafaunal
occupation at Liang Lembudu, Aru Islands, Maluku (Indonesia). extinction in the Late Quaternary and the global overkill
In: Kershaw, A.P., Tapper, N.J., David, B., Bishop, P., Penny, D. hypothesis. Alcheringa 28, 291–330.