Content uploaded by Robert G. Bednarik
Author content
All content in this area was uploaded by Robert G. Bednarik on Jul 23, 2021
Content may be subject to copyright.
L’art pléistocène dans le monde / Pleistocene art of the world / Arte pleistocénico del Mundo
Forensics in Australian cave art research
Robert G. BEDNARIK and Yann-Pierre MONTELLE
Abstract: In illustrating the practical application of the tenets and techniques of
forensic science in rock art research, the authors relate the work of specific projects
conducted in Australian limestone caves. Australia has the second-largest known
concentration of cave art in the world, which is being studied by the Parietal Markings
Project, established in 1981 and responsible for the discovery and investigation of
over forty decorated caves. Several of the sites have yielded evidence attributing the
petroglyphs to the Pleistocene, and it is assumed, by extrapolation from established
data, that much if not most Australian cave art is of that period. Forensic methods
have been applied to this corpus for decades and some of the recent work is
presented by the authors, illustrating methods of observation, analysis and replication
procedures to explore production processes.
The Parietal Markings Project
This project is concerned with the wall markings found in limestone caves around the
world, both of anthropic and natural origins, and was commenced in 1981 (Bednarik
1982). Most of its work over almost 30 years has been conducted in Australia, where
rock art in deep limestone caves is a comparatively rare phenomenon. Nevertheless,
with currently forty-eight known sites (not counting sites where the rock art is limited
to the entrances of caves; e.g. David & David 1988), the Australian corpus is the
second-largest body of cave art in the world. Of these sites, forty-one were
discovered by the Parietal Markings Project (PMP), which in total investigated over
300 caves in Australia and a similar number in five other continents collectively. With
few exceptions (Lane & Richards 1966; Gallus 1968, 1971; Hallam 1971; Morse
1984; Cosgrove & Jones 1989), the well over fifty publications addressing the cave
art of Australia are the work PMP researchers. In contrast to the cave art occurrences
in other parts of the world, Australian cave art consists largely of petroglyphs and is
entirely non-figurative. Pictograms, almost always in the form of hand stencils, occur
at only six of the sites.
There is limited evidence that some of the Australian cave art dates from the
Holocene (e.g. Bednarik 1998), and more extensive, albeit mostly circumstantial
evidence, that it is of Pleistocene ages. The latter includes many strands of evidence,
including geological, radiometric, stylistic and the contextual relationship with
features demanding such age, e.g. where the palaeoart pre-dates megafaunal
scratch marks.
The work of the PMP has focused on several specific topics, among them the
discrimination between natural and humanly made rock markings in caves (Bednarik
1980, 1986a, 1991), and the development of forensic methods to analyse cave
markings. Indeed, mastering the distinction of anthropic and non-anthropic wall
markings is itself very much dependent upon forensic science. It often hinges on
intensive examination, in much the same way as traceology has to attempt the
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
2
discrimination between markings that were made, for instance, with steel or stone
points. Among the natural markings in caves, those of animal claws are particularly
prominent in Australian caves, just as those of cave bears are in Europe (Bednarik
1993), but plant root marks (of mycorrhizal origin), clastic movement marks and other
taphonomic markings occur also. Without safely identifying natural cave marks there
can be no effective study of cave art, as the frequent misidentifications by
archaeologists show (e.g. Bednarik 1986b, 1994, 2002).
Description of Australian cave petroglyphs
A simplified taxonomy of Australian cave petroglyphs leads to the identification of six
basic classes:
A. Finger flutings: They occur on formerly soft calcite deposits which in all but two
sites are of a secondary, i.e. reprecipitated carbonate (moonmilk, Montmilch or
Mondmilch). Consisting of a microscopic, snow-like lattice of calcite crystals, it can
absorb up to 60% water and is initially just as soft as snow. These white cave
deposits were extensively marked by pre-Historic people (in France, Spain,
U.S.A., Dominican Republic, Papua-New Guinea and Australia), and they survived
in some cases through carbonatisation or desiccation (Bednarik 1999). Only about
sixty such sites are known worldwide, and the majority of them occur within 40 km
of Mount Gambier in South Australia.
B. Karake genre: These petroglyphs are deeply abraded (up to 40mm deep) and
probably often pounded. Motif types are dominated by circles and cell-like
arrangements of curvilinear enclosures. The circles are usually under 50cm but
may range up to about a metre in diameter, while the panels of mazes may extend
over several metres. Motifs also include parallel lines, arcuate designs,
“convergent lines motifs” (including the “trident” but also with two, four or five
“toes” which are not necessarily connected at the point of convergence), rare wave
lines, circles with internal design (vertical barring or lozenge lattice), and radial and
dot arrangements. This motif range has many parallels in other Australian rock
arts, which are frequently considered to be of Pleistocene age (Bednarik 2010)
and it is very similar to that of pre-iconic art globally. Several of these sites have
provided good evidence for such Pleistocene antiquity for this genre.
C. Tool marks: There is no indication that these are utilitarian and, in contrast to the
Karake motifs which are found on walls only, they are as likely to occur on ceilings.
They may form groups of sub-parallel lines or occur as apparently unstructured
assemblages, but occasionally they form apparent patterns such as lattices. The
tool material used in their production has been identified at two of the sites (Nung-
kol and Mandurah Caves), and forensic analysis has provided much information
about production sequences (e.g. Bednarik 1992a).
D. Deep pits: Traces of a widespread activity in which a soft rock, such as a cave
wall, has been extensively marked by a non-utilitarian but quite specific percussion
activity that resulted in panels featuring deep gashes, including the highly
distinctive, pocket-shaped “alveoli”. This phenomenon is not restricted to caves
and has not been properly examined, described or even recognised at open sites.
E. Shallow engravings: They are incised with usually single strokes of a pointed tool,
and are frequently responses to earlier designs of which they are sometimes
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
3
copies. The “shallow engravings” occur at very few cave sites and are separated
from the preceding Karake style by a substantial layer of cutaneous calcite
precipitate in Malangine Cave (Bednarik 1984).
F. Recent petroglyphs: Occur at only two of the cave sites, and only at the entrances.
Despite several credible age estimations the chronology of Australian cave
petroglyphs remains largely unresolved. Apart from the clear sequence at some sites
of classes A-B-E/F, chronological relationships remain under investigation. Class C
and D markings may relate to any phase, or to none of the others, but class C has
never been observed to precede class A, and there is corroborating evidence (such
as past fluctuations in floor level) suggesting that C postdates A. Nevertheless, some
of the finger flutings are certainly of Holocene age, in fact there are known
occurrences even of modern finger markings in five Australian caves. In particular,
the finger flutings in Prung-kart Cave near Millicent are thought to be of mid-
Holocene age, on the basis of laminae-derived radiocarbon dates (Bednarik 1998,
1999). The relative chronological placement of the chert mining remains uncertain,
except that at all art sites where it occurs it coincides with finger flutings. But this may
still be coincidence, and the mining evidence also occurs at three caves without the
finger marks (Bednarik 1992b, 1995).
Some of the Australian cave petroglyph sites have been subjected to detailed
archaeological studies: Orchestra Shell Cave (Hallam 1971), Koonalda Cave (Gallus
1971; Wright 1971), Malangine and Koongine Caves (Frankel 1986, 1989) and New
Guinea 2 Cave (by P. Ossa). Most of the archaeological data are not directly relevant
to the art as the sites were frequented at various times; the art cannot be
convincingly related to any of the occupation phases, and may in fact relate to none
of them. In some cases the occupation evidence is probably much more recent than
the art, e.g. in Orchestra Shell Cave, where the occupation stratum is in a deposit
that formed after a floor subsidence occurred, whereas the art antedates the time of
that collapse (Bednarik 1978/88).
Forensic work in Australian caves
Specific forensic methods employed early on by the PMP have included the
determination of the types of tools used in making wall markings, including the
recognition of the signatures (Fig. 1) of specific rock types (e.g. Bednarik 1987-88,
1992a); the determination of the approximate ages of the producers of finger flutings
(e.g. Bednarik 1986b, 2008); the detection of geomorphological events related to the
rock art, such as tectonic adjustments, roof falls, subsidences, inundations, speleo-
weathering processes or biospheric weathering (e.g. Bednarik 1989, 1999); the
reconstruction of superimposition sequences of finger flutings (Fig. 2) (e.g. Bednarik
1984, 1985, 1986b); the determination of the order tool applications (Fig. 3) and their
direction to reconstruct wall marking events (e.g. Bednarik 1987-88, 1992a, 2006);
and a variety of replicative experiments to test hypotheses. Another highly relevant
aspect is any evidence of other human activities in the caves, which mayor may have
coincided with the rock art production. For a number of reasons it is crucial to
establish whether such relationships can be documented.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
4
Fig. 1. Analysed superimposition of several different tool point applications, Nung-kol Cave,
South Australia (1985).
Fig. 2. Analysed superimposition sequence of finger flutings in Malangine Cave, South
Australia (1983).
Fig. 3. Tool marks in Koonalda Cave, South Australia.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
5
Specific anthropic activities demonstrated to have occurred in Australian caves
include the mining of sedimentary silica deposits, their use as campsites and living
quarters, and the exploration of deep passages. The PMP focused especially on the
chert mining evidence in Karlie-ngoinpool and Gran Gran Caves (Fig. 4) and the
chalcedony mining in Koonalda Cave (e.g. Bednarik 1986a, 1990, 1992b, 1995),
which in some cases involved the application of specialised techniques and tools,
established through forensic evidence provided by tool marks and actual impressions
of tool points (Fig. 5). Pleistocene silica mining is known from one cave site each in
Hungary and France (Bednarik 1986a; 1990), and from two alluvial sites in Egypt
(Vermeersch et al. 1986). In Australia, extensive traces of subterranean silica mining
have been located in nine caves so far, and in six of them they occur close to
petroglyphs. The reconstruction of the activities that led to these traces by forensic
methods has provided a basis of distinguishing five basic mining methods at the
Australian pre-Historic silica mines (Fig. 6).
Fig. 4. Chert mining traces with petroglyphs in Karlie-ngoinpool Cave, South Australia.
Fig. 5. Impression of the point of a wooden stake among extensive chert mining traces in
Gran Gran Cave, South Australia.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
6
Fig. 6. Methods of chert mining in caves as determined by forensic study (1985).
A recent example of forensics
In the last four years, the authors have focused on one site, Ngrang Cave, used here
to illustrate some generic principles of forensic work with cave art. Such work,
generally intended to establish precisely “what happened at a given site”, involves
various levels of analysis (see also Montelle 2009):
1. The macro-level: overall setting of the cave, its speleogenesis and establishing
how the present evacuational and convacuational spaces developed through time.
2. The medium level: the site formation processes that contributed to the present
state of the immediate environment of the cave art (i.e. within a few metres of it).
3. The micro-level: the precise details of the features which the previous levels of
investigation have identified as relevant, such as wall markings, weathering
details, speleothems, or any form of tectonic, fluvial, phreatic, vadose or biological
traces.
4. The microscopic level: the magnified examination of tools, markings, residues,
traces and so forth, details which are not visible to the unaided eye.
5. Replication: this refers to the experimental work of reproducing observed
outcomes or traces for the purpose of testing specific hypotheses concerning
specific observations.
This sequence is crucial to the optimal understanding of the circumstances of the
production of the rock art in question, and in combination with an understanding of
taphonomic effects and the relative or absolute chronology of events it forms the
basis of any scientific appreciation or analysis. In the absence of knowledgeable
owners or custodians of the rock art, no other approach can possibly lead to valid
interpretation.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
7
In the case of Ngrang Cave, we are dealing with a fluvial tunnel cave formed by a
Pleistocene subterranean stream that very likely drained into the nearby Glenelg
River. After the general lowering of the region’s aquifer levels during glacial periods
the horizontal tunnel, just a few metres below the surface plain, began to collapse in
some places along its course. This tunnel is now accessible in one place, from where
it can be followed for about 30m. Anthropic wall markings (Fig. 7) occur in much of
the low passage, but we have focused on one specific location, a series of 52 tool-
made cavities (“deep pits” or “alveoli”) forming a single panel, up to 25cm deep, how
they were made and how they relate to their surroundings (Fig. 8).
Fig. 7. Anthropic wall markings in the low passage of Ngrang Cave, South Australia.
Fig. 8. Part of the deep pits panel in Ngrang Cave.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
8
The present entrance of Ngrang Cave is at a roof collapse proceeding E to W, whose
rim retreat rate exceeds the build-up of the cone deposit inside it, facilitating
continued access to the tunnel (Fig. 9). Because the quantity of the collapsed rock is
known from the tunnel’s morphology and size, the contents of the cone slope cane
be estimated: 32 m3 limestone (c. 78 t) and 14 m3 sediment (c. 25 t), assuming
airspaces of about 2 m3. Pleistocene sediment from the exterior descends down the
slope and contributes to creating the cone deposit’s stratigraphy. Within it occur
dense lenses of evidence of occupation, comprising charcoal, bone fragments, emu
eggshell fragments and chert tools. As the rim of the collapse retreats with every new
rock fall, the lower limestone strata (the tunnel roof) become similarly unstable and
are claimed by gravity. One such event has claimed the northern end of the panel of
deep cupules, when a projecting wall portion bearing seven of the deep pits broke off
and fell to the ground relatively recently (Fig. 10).
Fig. 9. The gradual retreat of the cave entrance of Ngrang Cave towards west, producing an
elongate doline.
Fig. 10. Wall portion with seven deep pits that fell to the ground since their manufacture,
Ngrang Cave.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
9
Thus the history of the site can be broadly reconstructed through its geomorphology
and tectonic adjustments over time. This provides a relative timeframe within which
the production of the extraction pits on the wall ledge must be situated. The
preservation and surface texture of the fallen block with deep pits differ considerably
from the rest of the panel, which demonstrates the great effects of taphonomy: the
wall is subjected to the regime of capillary and ambient substrate moisture (subjected
to slight variations depending of surface geometry), while the fallen block is cut off
from the rock’s internal hydrology.
Fig. 11. One of the deep pits in Ngrang Cave, showing clear tool traces.
Placing the extraction pits (Fig. 11) into a relative chronological framework proved to
be more demanding. Their contemporaneity with any one occupation episode cannot
be readily demonstrated; in fact they cannot even be related with any of the
numerous other tool marks deeper in the cave. They are not related to any
speleothem, and only one of them has been truncated by a subsequent mass-
exfoliation event (the debris could not be located). This leaves only one index of
relative age, surface retreat by speleo-weathering, but it can be applied in three
different ways. First, the limestone contains tiny marine fossil casts that tend to
weather significantly less than the matrix, and at the opening of one of the holes,
such a cast extended 9mm above the present surface. This would suggest a
considerable surface retreat since the hole was made, but it needs to be assumed
that the cast was not protruding some of the distance at that time. Secondly, the
holes contain numerous tool marks, and their relative degree of weathering is a
measure of antiquity. However, several factors complicate the utility of this index:
weathering rates differ depending on the depth within the hole; the morphology of the
tool gauges at the time they were made is unknown and must be assumed; and the
type of tool used is not known. These considerations render it desirable to conduct
replicative experiments to better deal with these issues.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
10
The third indicator of the age of the holes is perhaps the most reliable, and the
easiest variable to quantify. The rims of the holes are considerably more rounded
today than they can be assumed to have been at the time of manufacture. There can
be no doubt that the more acute rims become progressively rounded with time. To
determine their initial morphology, replication is again required.
Fig. 12. Experiment in Ngrang Cave in progress, showing application of kangaroo long-bone
by indirect percussion to manufacture replication deep pit. Note pre-Historic pits in the left
background.
Since 2007 a series of experiments has been conducted to establish (a) what tools
and what extraction method(s) were most likely to have been used in making the pits;
and (b) what would the rims of the fresh pits have looked like. For this purpose flat
surfaces on blocks of the same limestone forming the cave walls were selected and a
dozen holes were created under controlled conditions, using three types of tools: a
broken leg bone of a kangaroo with a hammerstone (Fig. 12); a wooden pointed
chisel, also using indirect percussion; and large stone picks, hand-held in direct
percussion. This led to the realisation that the hollow bone was the most effective in
terms of speed of operation and in replicating the grooves seen in the extraction pits.
Its effectiveness is in part due to its functioning in the manner of a core drill, as the
internal channel fills with limestone dust. This does not necessarily prove that such
bones were used, but a key technological factor is that the pits have mostly sidewalls
parallel with their central axis, in fact in several cases they flare out with increasing
depth to a dimeter greater than that close to the rim. This suggests strongly a very
deliberate effort to create holes of the greatest possible ratio of opening diameter to
depth. Most are two or three times as deep as they are wide. Creating this distinctive,
alveolar shape demands a very focused effort and the use of an extraction tool
making this procedure practically possible. The broken end of a long-bone is clearly
the most likely candidate of tool used.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
11
Outlook and summary
The project of applying forensic methods to cave markings commenced almost 30
years ago in Australia. Progress since then may seem modest, especially as no
attempt has been made to sensationalise findings or over-interpret data.
Nevertheless, such analytical methods were first applied in Australian caves, and
indeed to rock art generally, and this work has in more recent years inspired various
developments. Among them are the work by Kevin Sharpe and Leslie Van Gelder,
who is here with us. The PMP project has in recent years focused on a few specific
caves near Mt Gambier and investigating specific key issues, because quite
obviously the mass of information waiting to be recorded and processed in this field
is simply monumental and overwhelming. We suggest that it will involve virtually
centuries of forensic investigations. In this field we shall need to very patiently learn
to walk before we can expect to run.
REFERENCES
BEDNARIK, R. G., 1980. On natural cave markings. Unpubl. MS, AURA Archive.
BEDNARIK, R. G., 1982. “The Parietal Markings Project”. Bulletin of the Archaeological and
Anthropological Society of Victoria 3: 3-4.
BEDNARIK, R. G., 1984. “Die Bedeutung der paläolithischen Fingerlinientradition”. Anthropologie 23:
73-79.
BEDNARIK, R. G., 1985. “Parietal finger markings in Australia”. Bollettino del Centro Camuno di Studi
Preistorici 22: 83-88.
BEDNARIK, R. G., 1986a. “Cave use by Australian Pleistocene man”. Proceedings of the University of
Bristol Spelaeological Society 17: 227-245.
BEDNARIK, R. G., 1986b. “Parietal finger markings in Europe and Australia”. Rock Art Research 3:
30-61, 159-70.
BEDNARIK, R. G., 1987-88. “The cave art of Western Australia”. The Artefact 12: 1-16.
BEDNARIK, R. G., 1989. “Perspectives of Koongine Cave and scientistic Archaeology”. Australian
Archaeology 29: 9-16.
BEDNARIK, R. G., 1990. “About Pleistocene chert mining”. Sahara 3: 113-115.
BEDNARIK, R. G., 1991. “On natural cave markings”. Helictite 29(2): 27-41.
BEDNARIK, R. G., 1992a. “Base pour des études de pointe des débuts de l’art”. L’Anthropologie
96(2/3): 609-611.
BEDNARIK, R. G., 1992b. “Early subterranean chert mining”. The Artefact 15: 11-24.
BEDNARIK, R. G., 1993. “Wall markings of the cave bear”. Studies in Speleology 9: 51-70.
BEDNARIK, R. G., 1994. “The discrimination of rock markings”. Rock Art Research 11: 23-44.
BEDNARIK, R. G., 1995. “Untertag-Bergbau im Pleistozän”. Quartär 45/46: 161-175.
BEDNARIK R. G., 1998. “Direct dating results from Australian cave petroglyphs”. Geoarchaeology
13(4): 411-418.
BEDNARIK, R. G., 1999. “The speleothem medium of finger flutings and its isotopic geochemistry”.
The Artefact 22: 49-64.
BEDNARIK, R. G., 2006. “The methodology of examining very early engravings”. Rock Art Research
23: 125-128.
BEDNARIK, R. G., 2008. “Children as Pleistocene artists”. Rock Art Research 25: 173-182.
BEDNARIK, R. G., 2010. “Australian rock art of the Pleistocene”. Rock Art Research 27: 95-120.
COSGROVE, R. and R. Jones, 1989. “Judds Cavern: a subterranean Aboriginal painting site,
southern Tasmania”. Rock Art Research 6: 96-104.
BEDNARIK R.G. & MONTELLE Y.P., “Forensics in Australian cave art research”
Congrès de l’IFRAO, septembre 2010 – Symposium : Application techniques police scientifique… (Pré-Actes)
IFRAO Congress, September 2010 – Symposium: Applications of forensic techniques… (Pre-Acts)
12
DAVID, B. and M. David, 1988. “Rock pictures of the Chillagoe-Mungana limestone belt, north
Queensland”. Rock Art Research 5: 147-156.
FRANKEL, D., 1986. “Excavations in the lower southeast of South Australia: November, 1985”.
Australian Archaeology 22: 75-87.
FRANKEL, D., 1989. “Koongine Cave Excavations 1986-87: investigating spatial patterning”.
Australian Archaeology 22: 3-13.
GALLUS, A., 1968. “Parietal art in Koonalda Cave, Nullarbor Plain, South Australia”. Helictite 6: 43-49.
GALLUS, A., 1971. “Results of the exploration of Koonalda Cave, 1956-1968”. In R. V. S. Wright (ed),
Archaeology of the Gallus Site, Koonalda Cave, pp. 87-133. Canberra: Australian Institute of
Aboriginal Studies.
HALLAM, S. J., 1971. “Roof markings in the ‘Orchestra Shell’ Cave, Wanneroo, near Perth, Western
Australia”. Mankind 8: 90-103.
LANE, E. A. and A. M. RICHARDS, 1966. “Hand paintings in caves”. Helictite 4: 33-50.
MONTELLE, Y.-P., 2009. “Application of forensic methods to rock art investigations — a proposal”.
Rock Art Research 26: 7-13.
MORSE, K., 1984. “First record of painted Aboriginal rock art in a south-western Australian limestone
cave”. Records of the Western Australian Museum 11: 197-199.
VERMEERSCH, P. M., E. PAULISSEN, C. GIJSELINGS and J. JANSSEN, 1986. “Middle Palaeolithic
chert exploitation pits near Qena (Upper Egypt)”. Palëorient 12: 61-65.
WRIGHT, R. V. S. (ed.), 1971. Archaeology of the Gallus Site, Koonalda Cave. Canberra: Australian
Institute of Aboriginal Studies.