Content uploaded by Joe Dortch
Author content
All content in this area was uploaded by Joe Dortch on Jun 21, 2019
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=raaa20
Australian Archaeology
ISSN: 0312-2417 (Print) 2470-0363 (Online) Journal homepage: https://www.tandfonline.com/loi/raaa20
Late Quaternary Aboriginal hunter-gatherer
occupation of the Greater Swan Region, south-
western Australia
Joe Dortch & Charles Dortch
To cite this article: Joe Dortch & Charles Dortch (2019): Late Quaternary Aboriginal hunter-
gatherer occupation of the Greater Swan Region, south-western Australia, Australian Archaeology,
DOI: 10.1080/03122417.2019.1594556
To link to this article: https://doi.org/10.1080/03122417.2019.1594556
Published online: 10 May 2019.
Submit your article to this journal
Article views: 64
View Crossmark data
ARTICLE
Late Quaternary Aboriginal hunter-gatherer occupation of the Greater
Swan Region, south-western Australia
Joe Dortch
a
and Charles Dortch
b
a
Centre for Rock Art Research and Management, University of Western Australia, Perth, Australia;
b
24 Howard Street, Fremantle,
Australia
ABSTRACT
The Greater Swan Region is an archaeological study area on the Indian Ocean coast in
south-western Australia. It covers the central Swan Coastal Plain and the adjacent part of
the formerly emergent Rottnest Shelf. Assessment of Aboriginal hunter-gatherer activity
throughout this late Quaternary coastal landscape is based on dated records of open-air
sites within their environmental settings, informed by historical accounts of mainland sub-
sistence activities. Artefacts flaked from Eocene fossiliferous chert and other stone, identified,
in situ, in dated dune soils on the mainland and in palaeosols and Tamala Limestone succes-
sions on Rottnest Island, show that during regressive sea levels human groups were distrib-
uted across a sand plain reaching from the Darling Scarp 70km westward across the
emergent Rottnest Shelf. On the mainland, open-air site settings near freshwater sources
and terrestrial and estuarine habitats give insight into adaptive strategies of Aboriginal occu-
piers. A similar occupation pattern is proposed for the emergent shelf. Early Holocene
records of terrestrial plant species and freshwater microfauna from Barker Swamp on
Rottnest Island imply that freshwater sources and terrestrial habitats were comparable to
those of the mainland. Further archaeological investigations on the Rottnest Shelf should
include extant and former freshwater swamps on Rottnest Island.
ARTICLE HISTORY
Received 22 April 2018
Accepted 11 March 2019
KEYWORDS
Aboriginal occupation;
aeolianite; palaeosols; late
quaternary; Greater swan
region
Introduction
This paper assesses late Quaternary Aboriginal
hunter-gatherer occupation patterns throughout the
‘Greater Swan Region’(GSR) –an archaeological
study area on the Indian Ocean coast of south-west-
ern Australia covering the central Swan Coastal
Plain and the adjacent part of the formerly emergent
Rottnest Shelf (Figure 1). The assessment is based
on dated occupation sites within their environmen-
tal settings, informed by historic European and
Aboriginal accounts of mainland Aboriginal coastal
economies. The study takes in the range of physical
settings historically occupied by coastal Aboriginal
groups and evaluates previous settings in areas now
offshore. The GSR covers a 60 km wide front from
the Darling Escarpment 70 km westward to the con-
tinental shelf edge (Figure 1; Dortch and Dortch
2012). Its mainland portion –now part of the Perth
Metropolitan Region –largely overlaps Hallam’s ori-
ginal survey transect (Hallam 1987). During regres-
sive sea levels, Aboriginal populations had access
across a coastal sandplain more than twice its 30 km
width today. Dated evidence for this access consists
of flaked stone artefacts identified in situ in dune
soils on the mainland and palaeosols on Rottnest
Island and is supported by undated finds of artefacts
on the mainland and on Rottnest and Garden
Islands.
Late Quaternary physical setting
The mainland portion of the study area is a repre-
sentative sector of the Swan Coastal Plain of which
its exceptional feature is the Swan River Estuary
(Figure 1). Three dune systems are defined across
this sand plain, marked from west to east by an
increase in biologically diverse terrestrial habitats
(McArthur 1991). The youngest of these systems –
the Quindalup Dunes –consists of white shelly
sands of late Holocene to modern age, formed dir-
ectly on the sea coast. Next is the Spearwood Dune
System, developed during the late Quaternary, and
featuring flaked stone artefacts, in situ, in dune soils
overlying Tamala Limestone, a major Upper
Pleistocene lithostratigraphic unit developed along
the littoral and submerged continental shelves on
the Indian and Southern Ocean coasts of south-
western Australia (Playford et al. 1976,2013). The
third and oldest set of dune soils is the Bassendean
Dune System, where many of the study area’s
CONTACT Joe Dortch joe.dortch@uwa.edu.au Centre for Rock Art Research and Management, University of Western Australia, M257, Perth,
WA 6009, Australia
ß2019 Australian Archaeological Association
AUSTRALIAN ARCHAEOLOGY
https://doi.org/10.1080/03122417.2019.1594556
pre-European, former occupation sites are recorded
(Hallam 1987). The soils in the Bassendean system
are deeply weathered quartz sands in which the car-
bonate component has been leached away. Between
the Bassendean sands and the Darling Scarp are
alluvial and colluvial deposits, the latter termed the
Pinjarra Plain (McArthur 1991). The Swan River
Estuary, passing through all these land divisions, is
estuarine upstream to the foot of the Darling Scarp.
The offshore portion of the GSR is part of the
Rottnest Shelf, centred on Rottnest Island (Collins
1988; Nicholas et al. 2013). Remote survey and
physical sampling indicate that during Pleistocene
low sea levels the shelf was carbonate sand plain
featuring on its seaward edge a dune barrier, and
eastward a succession of dune ridges aligned along
former shorelines. Today’s limestone reefs and
islands are relicts of the coastal dune barrier
(Brooke et al. 2010,2014; Nicholas et al. 2013,2014;
Playford 1997). A few kilometres north of the
Rottnest pre-island locality, the ancestral Swan River
Estuary traversed the emergent shelf before entering
the Perth Canyon incised into the shelf edge
(Playford 1983:14). Rottnest, Carnac and Garden
Islands were isolated with rising sea level reaching
present height about 6.5 ka (Churchill 1959;
Playford 1983,1988,1997). There is no evidence
that Rottnest Island was occupied by mainland
Aboriginal groups following its formation about
6.5 ka (see below). Rottnest Island is known to
Nyoongar Aboriginal people today as Wadjemup –a
name transcribed in the first decade of European
colonisation (Lyon 1833:64). A Nyoongar traditional
story related to colonist George Fletcher Moore in
this same decade tells how the offshore islands
became separated from the mainland (Moore
1978 [1884]).
Fresh water and other resources
Some 10 km inland from the coast, the onshore sand
plain is marked by freshwater lakes and wetlands,
rich in aquatic foods (Hallam 1987). Taking into
account periodic east-west shifts in the positions of
different plant associations, mainly in response to
changes in precipitation, pollen sequences covering
the past 40 ka in the central Swan Coastal Plain indi-
cate that eucalypt, banksia and casuarina woodlands
and melaleuca-fringed wetlands were present during
the late Quaternary similar to those of today
(Newsome and Pickett 1993; Pickett 1997). The
emergent Rottnest Shelf was presumably vegetated
and interspersed with wetlands and freshwater lakes,
a probable example being the terminal
Pleistocene–early Holocene lagoon delineated by the
10 m contour in the centre of Cockburn Sound
115° 30’ 116° 00’
115 °30’ 116° 00’
32° 00’
32° 00’
King
George
Sound
King
George
Sound
Rottnest IslandRottnest Island
Cockburn
Sound
Cockburn
Sound
Carnac Island
Carnac Island
Garden Island
Garden Island
Point Peron
Point Peron
Swan RiverSwan River
Darling
Scarp
Darling
Scarp
Rockingham-
Becher
Plain
Rockingham-
Becher
Plain
Fremantle
Fremantle
Minim Cove
Minim Cove
Upper
Swan
Rottnest Shelf
Rottnest Shelf
50 m contour
Depth interval:
0 - 5 m
5 - 10 m
10 - 50 m
50 - 100 m
100 - 150 m
>150 m
Houtman
Abrolhos
Houtman
Abrolhos
Perth
Basin
Perth
Basin
Leeuwin
Block
Leeuwin
Block
GSRGSR
Shark BayShark Bay
King
George
Sound
King
George
Sound
36° S 32° S 28°S
36° S 32° S 28°S
116° E 120° E 124° E
0 400 km
116° E 120° E 124°
0 20 km
Datum: GDA94
PerthPerth
Lake
Monger
Lake
Monger
Fiona Stanley
Hospital
Fiona Stanley
Hospital
Beeliar
Wetlands
Beeliar
Wetlands
Perth
Airport
Perth
Airport
Guildford
Guildford
Pt WaylenPt Waylen
Swan
Coastal
Plain
Swan
Coastal
Plain
QD SD BD PP
AD
AD
Hope
Road
Helena
River
Figure 1. Greater Swan Region, showing localities and sites. Inset: South-western Australia. QD: Quindalup Dunes; SD:
Spearwood Dunes; BD: Bassendean Dunes; PP: Pinjarra Plain; AD: Alluvial Deposits.
2 J. DORTCH AND C. DORTCH
Little Armstrong Bay
salt lakes
Charlotte Point
Fish Hook Bay
Bathurst Point
Barker Swamp
Fairbridge Bluff
N
0 4 km
32 00' S
32 00' S
115 30' E
115 30' E
City of York Bay
-10 m
- 5 m
Figure 2. Rottnest Island with artefact find sites and other places mentioned in the text.
AUSTRALIAN ARCHAEOLOGY 3
(Figure 1; Churchill 1959: Figure 2). In its ‘pre-
island’stage and as a newly formed island, Rottnest
offered fresh water in woodland and shrubland set-
tings, as shown by pollen and freshwater microfauna
identified in dated cores of Holocene age at Barker
Swamp (Figure 2; Backhouse 1993; Gouramanis
et al. 2012). The island’s present-day salt lakes are
karst features that during glacio-eustatic low sea lev-
els were probably freshwater sources (Dortch and
Hesp 1994:31; Playford 1997:786, 1983:10).
Site interpretation
The material evidence for pre-European Aboriginal
occupation in the GSR consists almost entirely of
flaked stone artefacts, mainly chert, quartz, and
other siliceous rocks. Other evidence includes iso-
lated burials and stone manuports. The region’s 674
recorded mainland archaeological sites include 609
concentrations of artefacts, both surface and sub-
surface (Department of Planning, Lands and
Heritage 2018), which are interpreted as former
open-air campsites (e.g. Hallam 1972). None of
these sites has food remains or structures (e.g.
hearths or fish traps) connected with past subsist-
ence. Past Aboriginal subsistence patterns are
inferred from site distributions relative to exploit-
able habitats, with reference to historic accounts of
regional Nyoongar hunter-gatherer subsistence
activities (Anderson 1984; Dortch 2002; Hallam
1972,1975,1987; Meagher 1974; Smith 1999). These
accounts show that traditional Nyoongar economy
was based on the systematic exploitation of largely
seasonally available resources in coastal and adjacent
inland districts, particularly in and around wetlands.
These districts formed contiguous, permeable dialect
group territories making up the Nyoongar language
region or cultural bloc (Dortch 2002). Local bio-
diversity and the adaptability of the Nyoongar econ-
omy meant year-round availability of most foods,
with seasonal abundances and occasional shortages.
As noted above the environmental opportunities
supporting this economy possibly held throughout
much of the period of human occupation.
Eocene fossiliferous chert
For 20 ka or longer, Eocene fossiliferous chert was
much used for artefacts along the Indian Ocean
coast of south-western Australia, i.e. the Perth Basin
and adjacent Leeuwin Block (Figure 1; Dortch 2004;
Glover 1984: Figure 1; Playford et al. 1976). This
distinctive chert, which features marine microfossils
of Eocene age, has yet to be identified in any sour-
ces (Glover 1984). The stone ceased to be used
1–2 ka after sea level reached its present position
c.6.5 ka (Ferguson 1980; Pearce 1978). Geochemical
differences detected in fossiliferous chert artefacts
sampled from 78 archaeological sites broadly dis-
persed along this 700 km littoral support the view
that fossiliferous chert outcrops were once widely
accessible (Glover and Lee 1984). This supposition
is now questioned by claims that (1) the geochem-
ical composition of some of the original 78 samples
does not vary as claimed; (2) nowhere on the Indian
Ocean coast were fossiliferous chert sources access-
ible to tool makers; and (3) the chert artefact
assemblages on this coast probably derive from
sources of very similar stone on the Nullarbor Plain
c.1200 km to the east (O’Leary et al. 2017).
In the absence of conclusive evidence for any
particular source, we judge localised sources as
more likely. Collecting chert from outcrops formerly
exposed a few kilometres offshore the present coast
(or prior to dune formation on the littoral, Pearce
1978) is the simplest explanation for the presence of
abundant assemblages at archaeological sites all
along the western littoral. Irrespective of the source
–whether <20 km to the west, 1200 km to the east,
or both –artefacts flaked from fossiliferous chert
are a distinctive feature of late Pleistocene and early
Holocene sites on the Swan Coastal Plain (Clarke
and Dortch 1977; Dortch and Dortch 2012;
Ferguson 1980; Hallam 1987; Pearce 1978), in the
Leeuwin-Naturaliste Region (Dortch 1979;2004;
Ferguson 1981) and along the Southern Ocean coast
(Dortch and Gardner 1976; Dortch and Morse 1984;
Smith 1993). The material is also represented by a
single flake in the Houtman Abrolhos, last con-
nected to the mainland during the late Pleistocene
(Figure 1; Marwick 2002). The decreasing occur-
rence and then disappearance of fossiliferous chert
artefacts in mainland sites dated from the middle to
late Holocene supports the argument that this dis-
tinctive stone was regionally available only when the
shelf was emergent.
Late Quaternary sites in mainland
geomorphic zones
Here we review the oldest dated artefact find sites
within the Spearwood and Bassendean Dune
Systems and the Pinjarra Plain in the mainland por-
tion of the GSR (Table 1,Figure 1). As a result of
poor preservation and survey bias, the Quindalup
Dune System generally lacks dated pre-European
archaeological evidence (Hallam 1987: Table 1). All
dates presented below are given as calibrated ages
calculated using the OxCal SH13 curve (see Table 1
for the original uncalibrated ages).
4 J. DORTCH AND C. DORTCH
Table 1. Dates for stone artefacts in situ in dune sands and palaeosols in the Greater Swan Region, south-western Australia (Figure 1).
Site Stratigraphic setting
Laboratory
code Method
Nature of
sample Age (BP)
Calibrated
age (ka) Archaeological evidence Environment Reference
Mainland
Minim Cove Artefact-bearing Spearwood
dune soil on eroded
Tamala Limestone
SUA 454
14
C Charcoal 9,930 ± 130 11.4 ± 0.5 Quartz and fossiliferous
chert artefacts, including
16 in dated profile
Spearwood System dune
overlooking Swan River
Clarke and Dortch 1977
Mandurah Road
(Rockingham-Becher Plain)
Estuarine / marine shelly
sand cut into artefact-
bearing Spearwood
dunes
GX 12906
14
C Bivalve shell
(Katelysia
sp.)
8,365 ± 375
(C13 corrected)
9.3 ± 0.9 One quartz artefact and
one fossiliferous chert
artefact
Spearwood System dunes
within 1 km of coastal
plain lakes
Searle et al. 1988: Table
2, Figure 8
Fiona Stanley Hospital Bassendean dune contain-
ing artefacts (lowermost
same depth as Wk
24525)
Wk 24524
14
C Charcoal 8,467 ± 284 9.4 ± 0.75 Quartz, fossiliferous chert,
mylonite and granite
artefacts, including two
artefacts in dated profile
Bassendean System dunes
within 1 km of Beeliar
wetlands
Dortch et al. 2009;
Dortch and
Dortch 2012
Wk 24894
14
C - AMS Charcoal 13,826 ±60 16.7 ±0.25
Wk 24525
14
C - AMS Charcoal 31,201 ±385 35.1 ±0.75
Perth Airport: Adelaide St Bassendean dune contain-
ing artefacts W4754-
47755 from trench
TPX02; remainder from
trench TPX08
W4754 TL Quartz sand NA 12.0±1.4 Quartz, fossiliferous chert,
and granite artefacts,
including 1,247 artefacts
in dated trenches
Bassendean System dunes
within 1 km of wetlands
Mattner et al. 2014
W4755 TL NA 19.6 ±2.6
W4756 TL NA 24.4 ±2.9
W4757 TL NA 22.4 ±2.6
W4758 TL NA 35.0 ±2.8
W4759 TL NA 28.4 ±2.8
Rottnest Island
Charlotte Point Artefact-bearing palaeosol
intercalated with
aeolianite
GU8.5 SG OSL Calcareous
sand
NA 12.7±1.25 Fossiliferous chert artefact Palaeosol in Tamala
Limestone ridge over-
looking coastal plain
Ward et al. 2016: Table 1
GU8.6 SG OSL NA 12.1±1.01
Wk 37948
14
C Charcoal Not reported 10.3±0.06
Little Armstrong Bay Upper part of artefact-bear-
ing palaeosol interca-
lated with aeolianite:
GU8.4 same depth as
chert tool
GU8.3 Single grain OSL Calcareous
sand with
quartz
grains
NA 12.6±1.1 Fossiliferous chert artefact Palaeosol in Tamala
Limestone ridge over-
looking coastal plain
Ward et al. 2016: Table 1
GU8.4 Single grain OSL NA 16.8±1.59
CURL-16668
14
C - AMS Emu eggshell Not reported 28.2±0.18
Bathurst Point Surface of breccia calcrete
cut block in which chert
artefact is embedded
P 2.2 U-series Calcium car-
bonate
cement
NA 17.1±0.85 Fossiliferous chert artefact Tamala Limestone ridge
overlooking coastal plain
Ward et al. 2016: Table 1
Base of breccia calcrete cut
block in which chert
artefact is embedded
R11/K0034 Single grain OSL Breccia calcrete
with quartz
grains
NA 42.0±2.7 Ward et al. 2016: Table 1
City of York Bay Palaeosol containing feld-
spar pebble (gastrolith or
manuport)
R5/K0022 Single grain OSL Sand NA 49.1±3.3 Feldspar pebble (manuport) Palaeosol in Tamala
Limestone ridge over-
looking coastal plain
Dortch and Hesp
1994:25–26; Ward
et al. 2016: Table 1
Fish Hook Bay Charcoal concentration in
palaeosol in sea cave,
associated with quartz
pebble (gastrolith or
manuport)
SUA 3030
14
C Charcoal 18,660±250 22.5±0.6 Quartz pebble (manuport) Palaeosol in Tamala
Limestone ridge over-
looking coastal plain
Dortch and Hesp
1994:27–28
AMS: Accelerator Mass Spectrometry.
Samples in each site are listed in stratigraphic order from uppermost to lowermost. Radiocarbon ages were calibrated using the OxCal SH13 curve (Bronk Ramsey 2009); errors are at 2 sigma; radiocarbon years are not applicable
(NA) for OSL, TL and U-series dates.
AUSTRALIAN ARCHAEOLOGY 5
Spearwood Dune System
Minim Cove
A radiocarbon date implying a minimum age of
11.4 ka for fossiliferous chert artefacts in situ in the
Spearwood Dunes comes from Minim Cove,
Mosman Park (Clarke and Dortch 1977). In 1975,
archaeologists from the Western Australian Museum
(WAM) excavated a 1.5 m
2
test pit in a two-metre
deep deposit of undisturbed, uniformly mottled, yel-
low quartz sand infilling an irregular calcarenite
(aeolianite) surface of pinnacles and solution pipes.
These limestone formations are covered by a thin
veneer of secondarily cemented calcium carbonate,
suggesting that the enveloping quartz sand unit
in-filled a deflated Tamala Limestone exposure typ-
ically featuring pinnacles and solution pipes. Two
fossiliferous chert chips cemented to vertical faces
within this calcium carbonate veneer substantiate
the excavators’view that the artefact-bearing dune
soil accumulated on the calcarenite and is not a
leached residual (Clarke and Dortch 1977:38). The
charcoal sample dated 11.4 ka is 40 cm above 14 fos-
siliferous chert flakes and two quartz flakes, suggest-
ing occupation began earlier than the reported date.
Mandurah road
Three fossiliferous chert artefacts were located in
situ within Spearwood Dunes in road cuttings along
the eastern shoulder of the north-south running
Mandurah Road, east of Lakes Cooloongup and
Walyungup (C. Dortch, unpublished field notes). A
sample of marine molluscs (cockle shells: Katelysia
sp.) in situ in shelly marine beach sands abutting
this Spearwood dune soil –which rests on Tamala
Limestone –indicates a calibrated age of 9.3 ka
showing that post-glacial sea level rise had brought
the marine shoreline to this position at that time
(Searle et al. 1988: Table 2). This date suggests an
early Holocene or terminal Pleistocene age for the
chert artefacts buried within the older Spearwood
Dune soil. Searches in the deflated dunes around
the two lakes and in the prograded beach ridges
covering c.5 km between the lakes and the present
coast (i.e. the Rockingham-Becher Plain: Searle et al.
1988) revealed no stone artefacts.
Point Peron
Immediately west of the Rockingham-Becher Plain
is the Tamala Limestone headland of Point Peron,
where three fossiliferous chert artefacts lie within
orange-brown soils of Spearwood type in-filling
solution pipes and pinnacles exposed along the
shoreline cliffs. A fourth fossiliferous chert artefact
is embedded in a horizontal cemented carbonate
unit in the Tamala Limestone at this site.
Regrettably, this once datable find is now buried
beneath a concrete slab poured sometime after its
identification. No stratigraphic association can be
established between the three fossiliferous chert arte-
facts mentioned above and a sand sample from the
same site with a thermoluminescence (TL) date of
c.78 ka (Price et al. 2001: Table 2). The four fossil-
iferous chert flakes from Point Peron, along with
artefacts of this stone exposed in railway cuttings in
Spearwood dunes overlooking the sea at Mosman
Park (Clarke and Dortch 1977) are the sole finds on
the mainland seacoast of the GSR.
Bassendean Sands
Fiona Stanley Hospital
The two younger of three radiocarbon dates on
charcoal samples collected from a 1 m
2
test-pit in
the Bassendean Dune System provide a minimum
age for a single stone tool (Figure 3) found in situ,
depth c.150 cm below the original ground surface, in
a well-sorted quartz sand dune overlooking wetlands
(Dortch et al. 2009). The retouched tool is made of
mylonite, a chert type outcropping in the Darling
Scarp minimally 17 km east of the find site. The
youngest date, c.9.4 ka, is from 40 cm higher up in
the deposit; the second date, c.16.7 ka, is 20 cm
higher than the tool’s find position, and the third
date, c.35.1 ka, is the same depth and position as
the tool and suggests its true age. A large fossilifer-
ous chert flake (length 8 cm –the largest recorded
on the Swan Coastal Plain) and 48 other fossilifer-
ous chert and quartz flakes and cores were identified
Figure 3. Four views of a mylonite adze flake, Bassendean
Dunes (Fiona Stanley Hospital). The top view shows front of
heavily used, adze working edge. Central lower view indi-
cates that the truncated end consists of a single negative
flake scar, with edge-damage consistent with subsequent
use as a tool (Illus. Franc¸ois Mazieres).
6 J. DORTCH AND C. DORTCH
in grader cuttings in this dune and at lower eleva-
tions within 100 m of the test excavation. The
nearby presence of undated fossiliferous chert arte-
facts implies a late Quaternary age for this dune
soil; like the mylonite tool, they indicate group
movements or interaction within the GSR.
The mylonite retouched tool is interpreted as an
adze flake (i.e. a hafted tool), based on morphology
and use-flaking. It has been truncated by a single
blow against its bulbar face. Its lower right-hand
corner shows edge-wear consistent with its having
been used as a tool after this event (Figure 3). This
specimen could conceivably contribute to the con-
tinental record of considerable antiquity for the
hafting of stone tools (e.g. Hamm et al. 2016;
McDonald et al. 2018; Mulvaney and
Kamminga 1999).
Perth Airport
Some of the most prolific mainland sites recorded
are in Bassendean Sands near wetlands in the Perth
Airport estate, covering c.3,000 ha in the eastern
part of the study area (Figure 1). Some 50 sites have
been identified since surveys began in the 1970s
(Anderson 1984). The artefact assemblages include
fossiliferous chert artefacts, late Holocene tool types
(e.g. microliths) and flaked historic glass, indicating
continuity in site use (Anderson 1984:19, 24). At
one site, ‘Adelaide Street’, 1,247 stone artefacts were
recovered from 100 cubic metres of excavated sedi-
ment (Mattner et al. 2014). Artefacts were concen-
trated in several locations, suggesting former activity
areas. Similar concentrations have been reported
from other excavations in the airport grounds,
though no other dates are yet available. Finding no
charcoal, the excavators at Adelaide Street under-
took TL dating of six bulk sand samples, inferring
an age range for occupation from c.35 to 12 ka
(Table 1). A dating reversal between the lowermost
two samples in trench TPX08 is possibly explained
by variations in background dosage at lower depths.
The sequence of remaining TL dates in broad strati-
graphic order suggests the preservation of a chrono-
stratigraphic sequence, among the oldest recorded in
the GSR.
Pinjarra plain
This innermost geomorphic zone in the study area
consists of alluvial clay-rich soils (Hallam
1987:19–20; McArthur 1991:26). Bowdler et al.
(1991:24) label this geomorphic zone ‘Alluvial
Deposits’, and affirm its archaeological research
potential. Dated charcoal associated with artefacts at
the Helena River site (Schwede 1983) could be as
old as 29 ka, while the Upper Swan site –15 km
north of the GSR study area –has produced a date
of 38 ka (Pearce and Barbetti 1981). Despite justifi-
able concerns about the reliability of the associations
of artefacts and dated charcoal at Upper Swan
(Bowdler et al. 1991:24; Mulvaney and Kamminga
1999:138), the dates from this site are in broad
stratigraphic order, and some of the flaked quartz,
quartzite and chert artefacts from the same deposi-
tional units can be re-fitted, indicating some degree
of stratigraphic integrity.
Mainland summary
The mainland distribution of dated open-air, arte-
fact find spots relative to wetlands and other
exploitable habitats offers insight into the adaptive
strategies of hunter-gatherer groups from late
Quaternary times to the historic era (Table 1).
Except possibly for Minim Cove, all the sites
reviewed above were occupied repeatedly during
this prolonged time period.
Late Quaternary stone artefact find sites on
the Rottnest Shelf
The archaeological evidence recorded on the
Rottnest Shelf consists of isolated fossiliferous chert
and calcrete artefacts on Rottnest and Garden
Islands, and one quartz flake identified on the latter
(Figure 1; Dortch 1991; Dortch and Dortch 2012;
Dortch and Hesp 1994; Dortch and Morse 1984).
Rottnest Island
Upper Pleistocene geochronology
The absolute chronology for Rottnest Island’s con-
stituent Tamala Limestone (c.140 to 10 ka) is based
on about 40 mostly optically stimulated lumines-
cence (OSL) dates from several widely-distributed
localities along the island littoral (Brooke et al.
2014; Ward et al. 2016), which accord with earlier
TL and other datings (Price et al. 2001). The oldest
OSL date (c.140 ka, for aeolianite at Fairbridge
Bluff) is consistent with isotopic and TL dates giving
a last interglacial age for the Rottnest Limestone
(Price et al. 2001; Stirling et al. 1995,1998; Szabo
1979), which is a fossil coral reef intercalated with
Tamala Limestone at this site (Playford 1988,1997).
During the early to middle Holocene, transgressive
seas cut cliffs and headlands into the newly forming
island’s limestone hills and ridges.
All the late Pleistocene luminescence and isotopic
dates in Table 1, and most of those in Ward et al.
2016: Table 1 are significantly younger than the
nearly 100 age estimates produced by whole-rock
amino acid racemisation (AAR) assay of the same
AUSTRALIAN ARCHAEOLOGY 7
units within the island’s Tamala Limestone succes-
sions (Hearty 2003). Acceptance of the whole-rock
AAR chronology for these successions, including the
two oldest of the three sites noted below in which
chert artefacts are in situ within intercalated palaeo-
sols (Charlotte Point, Little Armstrong Bay and
Bathurst Point), would connote absolute ages for
human presence on Rottnest Island far earlier than
the oldest dated Australian archaeological sites (e.g.
O’Connell and Allen 2015). Brooke et al. (2014:120)
critically review discrepancies between whole-rock
AAR and luminescence (mainly OSL) chronologies
for the Tamala Limestone on Rottnest Island and
the adjacent mainland, as do Playford et al.
(2013:102–103) for similarly contrasting chronolo-
gies at Shark Bay c.700 km to the north.
Flaked stone artefacts
Six fossiliferous chert artefacts and four calcrete
artefacts are recorded from the island. Three of the
fossiliferous chert artefacts were identified in situ in
dated, late Quaternary palaeosols intercalated within
Tamala Limestone successions at Charlotte Point,
Little Armstrong Bay and Bathurst Point (Table 1).
Ward et al. (2016: Table 1; Figure 2) provide strati-
graphic sections for these three sites, showing prove-
nances of the artefact finds and the sand, calcrete
and other samples mostly dated by OSL measure-
ments, both single-grain (SG) and single aliquot
(SA) assay.
Charlotte Point and Little Armstrong Bay
OSL dates for two artefact-bearing palaeosol units at
Charlotte Point and Little Armstrong Bay (Figures 4
and 5) range from 16.8 to 12.1 ka (Table 1). A
radiocarbon date 10.3 ka for charcoal in the palaeo-
sol at Charlotte Point may represent plant growth
or roots burning within the sediment before
cementation.
The artefact-bearing palaeosol at Little Armstrong
Bay has two OSL dates –12.6 and 16.8 ka –the
younger for a sample from just above, and the older
from the same depth as, the chert tool shown in
situ in Figure 6. This palaeosol yielded the first ver-
tebrate fossil find identified on Rottnest Island: a
fragment of emu eggshell (Dromaius novaehollan-
diae) in situ at the same depth as the chert tool.
Radiocarbon dated to c.28.2 ka, this eggshell
Figure 4. Charlotte Point, Rottnest Island: artefact-bearing
palaeosol intercalated within aeolianite units. Chert artefact
find spot is behind person’s lower legs. Photo Ron Priemus,
Rottnest Island Authority (RIA).
Figure 5. Sea-cut Tamala Limestone cliff face at artefact find
site at Little Armstrong Bay, Rottnest Island (cf. Ward et al.
2016:Figure 2C). The person at the cliff base is pointing at
the artefact-bearing palaeosol (Figure 6). The palaeosol is
bounded top and bottom by aeolianite units. Mainland coast
is visible on the horizon. Photo Mark Ward, RIA.
Figure 6. Eocene fossiliferous chert tool in situ in palaeosol
section at Little Armstrong Bay (Figure 5). Modern graffiti is
incised in the section face. Photo Reg Yarran, RIA.
8 J. DORTCH AND C. DORTCH
fragment may be redeposited from an older deposit
(Ward et al. 2016:23; Table 1). In 2016, fine sieving
c.0.8 m
3
of undisturbed sediment dug from the
palaeosol during construction of a beach access
stairway revealed no artefacts or macrofaunal
remains.
Bathurst Point
The Tamala Limestone succession exposed in the
sea cliff at Bathurst Point has two dated strati-
graphic sections: East and West, 60 m apart (Ward
et al. 2016: Figure 2(B); Table 1). The basal unit in
each is the older aeolianite, which in the West sec-
tion has a TL date c.67 ka and an OSL date c.77 ka
(Brooke et al. 2014: Figures 3F, 4F; Price et al. 2001:
Table 2). At the top of the older aeolianite in each
section is a massive breccia calcrete pavement that
in the East section is OSL dated c.42 ka (Figure 7).
In this section, this unit is capped by a 0.5–1cm
lens of carbonate cement, which has a U-series date
of c.17.1 ka, based on a weighted average age of six
sub-samples from its surface. These samples are
from the top of an artefact-bearing, 8 8 cm block
cut from the uppermost part of the brecciated
calcrete unit (Ward et al. 2016: Figure 3;18, 21).
Provisionally classifiable as a small flake core
(Figure 8), the lower part (presumed striking plat-
form) of this artefact is marginally within the dated
thin carbonate lens, and is c.2 cm above the base of
the block, a sample from which yielded the above-
noted c.42 ka OSL date. As the artefact cannot con-
fidently be associated with this date, its absolute age
is that of the U-series date. Ward et al. (2016:21–23)
regard the c. 32.9 ka OSL date for the rhyzolithic
palaeosol (Figure 7: unit 3) as anomalously old.
However, given the ubiquitous evidence for erosion,
re-deposition and re-cementation at this site (Ward
et al. 2016:25), this date is not necessarily discrepant
with the c.17.1 ka U-series date for the carbonate
lens overlapping the base of the Eocene fossiliferous
chert artefact in Figure 8. This date and the
c.16.8 ka OSL date for the chert tool from Little
Armstrong Bay provide a minimum c.17 ka age for
the earliest recorded human presence on Rottnest
Island.
The upper part of the Tamala Limestone succes-
sion in both East and West sections –in which no
cultural material has been identified –has relatively
minor dating anomalies (Ward et al. 2016). Apart
from the fossiliferous chert flaked piece (Figure 8),
the only other ‘pre-island’artefact finds from
Bathurst Point are two calcrete flakes in situ in the
undated, unconsolidated, calcareous palaeosol
(stratigraphic unit 4) overlying the limestone succes-
sion in the East section (Figure 7).
Other stone artefact finds
Undated fossiliferous chert artefacts from Rottnest
Island include two retouched flakes (Figure 9(B,D))
from an eroding palaeosol at the summit of the cliff
overlooking Fish Hook Bay at the island’s West
End, and a notched flake (Figure 9(A)), which
Figure 7. Bathurst Point East section (Eocene fossiliferous
chert artefact find site), Rottnest Island, photographed from
offshore (Ward et al. 2016: Figure 2(B)). Person is standing
on breccia calcrete pavement 50 cm behind the find position
of the Eocene fossiliferous chert artefact shown in Figure 9.
Photo Ron Priemus, RIA. Numbered units in reverse strati-
graphic order, as follows: (1) ‘Older aeolianite’(continuing
below sea level). (2) Breccia calcrete capped by carbonate
lens. An OSL date for the breccia calcrete is c.42 ka; as sam-
ple position is immediately below the artefact, this date can-
not be related to it, though it provides an absolute age for
the top of the ‘older aeolianite’. (3) Rhizolithic palaeosol
(seen as a 50 cm high bench behind person’s legs): sediment
sample from the top of this bench is OSL dated c. 32.9 ka
(see text). (4) Unconsolidated calcareous palaeosol (esti-
mated age more than 6.5 ka). This unit, featuring numerous,
weakly cemented Leptopius sp. weevil pupal cases and in
which two calcrete flakes have been recorded in situ, is the
partly vegetated white unit at the person’s shoulder level.
(5) Disturbed, vegetated dune.
Figure 8. Eocene fossiliferous chert flaked piece in situ in
surface of block cut from breccia calcrete (unit 2 in Figure
7), Bathurst Point, East Section. The surface of this block has
a U-series age estimate of c.17.1 ka. Photo Alice Beale,
Western Australian Museum (WAM).
AUSTRALIAN ARCHAEOLOGY 9
almost certainly had fallen from the artefact-bearing
palaeosol at Little Armstrong Bay (Dortch and Hesp
1994: Figure 3). These three fossiliferous chert tools
and four calcrete flakes (including the two from
Bathurst Point, mentioned above) pre-date 6.5 ka,
the estimated time of the island’s separation from
the mainland (Playford 1988,1997). The two cal-
crete flakes from Bathurst Point East (Figure 7:
stratigraphic unit 4) and a third calcrete flake from
a deflated dune overlying Tamala Limestone near
Fish Hook Bay (Dortch and Hesp 1994: Figure 2:2)
indicate human presence of the pre-island
Wadjemup locality during the early Holocene.
Lastly, Dortch and Hesp (1994) reported a feldspar
pebble, identified as a gastrolith or manuport, from
the City of York Bay palaeosol, now dated 49 ka
(Table 1).
Garden and Carnac Islands
The pre-island cultural record for Garden Island,
20 km south of Rottnest, consists of three fossilifer-
ous chert flakes, five calcrete flakes and a single
quartz flake, all found on wind-eroded, Tamala
Limestone cliff summits on the island’s western
shore. The quartz flake was cemented by secondary
carbonate to a limestone exposure (Dortch and
Morse 1984:39). It is the sole cultural item from the
Rottnest Shelf showing a connection to the Darling
Escarpment or districts further east, the nearest
sources of naturally occurring quartz (Glover 1984).
The southern end of Garden Island is 2.5 km from
the mainland. Despite this relatively short distance,
there is as yet no evidence that Aboriginal people
visited the island following its formation as such. In
several searches, no stone artefacts have been identi-
fied on Carnac Island, a 19 ha nature reserve 4 km
north of Garden Island.
Shelf summary
The Rottnest Shelf cultural record tallies 19 flaked
stone artefacts on two islands. The late to terminal
Pleistocene age range of the three dated fossiliferous
chert artefact contexts on Rottnest Island and the
20 km distance between it and Garden Island
implies human presence over much of the emergent
shelf before 17 ka, and as recently as 6.5 ka –the
estimated age for the islands’formation and the
shelf’s submergence.
Discussion
The archaeological potential of the GSR lies in
mainland and offshore localities where excavation
could disclose associated cultural and biotic material
implying subsistence activities or past environmental
conditions, or yield information about occupation
patterns relative to resources (Table 1). Here we
assess the relative potential of the various geo-
morphic zones in the GSR.
Central Swan Coastal Plain
The largest and most prolific middle to late
Holocene stone artefact sites in the central Swan
Coastal Plain are the Perth Airport estate, and
undated fossiliferous chert and quartz artefact
assemblages at Lake Monger and the Beeliar
Figure 9. Eocene fossiliferous chert tools from Rottnest
Island find sites, giving year of find. (A) Notched flake, Little
Armstrong Bay, 1992. (B) Retouched flake, Fish Hook Bay,
1984. (C) Retouched flake, Little Armstrong Bay, 2010 (see
also Figure 6). (D) Retouched flake, Fish Hook Bay, 1992
(Illus. Franc¸ois Mazieres).
10 J. DORTCH AND C. DORTCH
wetlands (Figure 1), all located near abundant fresh-
water resources 7–20 km inland, with relatively few
sites recorded closer to the present seacoast. As an
example of the archaeological value of this region, a
site at Hope Road, in the Beeliar Wetlands, contains
fossiliferous chert artefacts underlying artefacts
made from quartz, and at the surface, flaked bottle
glass and other European materials (Dortch and
Hook 2017). This undated succession implies
repeated wetland occupation from mid-Holocene
times until the historic period. However, open-air
sites in these heavily leached sands are unlikely to
yield vertebrate or other faunal remains. There are
no published reports of excavations yielding arte-
facts in locations more than 1 km from water sour-
ces, suggesting the significance of water in any
predictive model for the region.
Test pits in caves or overhangs in the Tamala
Limestone cliffs bordering the Swan River Estuary
would offer some chance for identifying biotic
remains in association with stone artefacts. The cen-
tral Swan Coastal Plain seems to lack limestone
caves with deep floor deposits as in nearby coastal
regions (Dortch 2004; Monks et al. 2016). No arte-
facts have been recorded in situ in palaeosols
exposed in Tamala Limestone successions along the
Swan River Estuary or on the mainland coast within
the GSR. Twelve of the youngest OSL dates for
these successions exceed 100 ka (Brooke et al. 2014:
Tables 2, 4) and five TL dates range c.72–83 ka
(Price et al. 2001: Table 2).
Midden deposits may be an unlikely site type in
the region. Dortch et al. (1984) note the abundance
of marine bivalves in mid-Holocene shell beds at
several locations along the course of the Swan River
Estuary but identified no midden deposits here.
Rottnest shelf
The archaeological record of the Rottnest Shelf is
most accessible on Rottnest Island, which has arch-
aeological potential beyond that of rare finds of
artefacts in situ within limestone successions. ‘Pre-
island’subsistence opportunities are indicated by
the eight formerly freshwater swamps, which suggest
that the Rottnest ‘pre-island’locality offered attract-
ive conditions for human occupiers (Backhouse
1993; Gouramanis et al. 2012). The sediments
revealed in cores around these former swamps are
peat and calcilutite indicating fresh water conditions
by 7.5 ka giving way to brackish water at 6.6 ka
(Backhouse 1993). Despite more arid conditions
during full glacial periods, the presence of fresh
water at the Rottnest locality during the late
Quaternary is also implied by calcrete units in the
Tamala Limestone cliffs at many island locations,
e.g. Bathurst Point (Playford 1997:794; Semeniuk
1986; Ward et al. 2016:23).
Other potential excavation sites on Rottnest
Island are sea caves in which unconsolidated weath-
ered quartz sands and palaeosols are exposed, e.g. a
3.5 m high, brown palaeosol present in a sea cave at
Fish Hook Bay, featuring two charcoal lenses in its
vertical section, one of which is radiocarbon dated
22.6 ka. The lower right-hand portion of the
Charlotte Point stratigraphical section (Figure 4)
shows, several metres below the artefact-bearing
unit, two palaeosols intercalated with aeolianite
units. This image clearly evokes the multiplicity of
palaeosols in Tamala Limestone successions on
Rottnest Island.
Artefact rarity
Dortch and Hesp (1994:29–30) review three factors
that seem significant in accounting for the scarcity
of ‘pre-island’cultural finds on Rottnest Island.
Firstly, the locality’s elevated position near the edge
of the emergent continental shelf may have been
less suitable for human occupation than more low-
lying and well-watered areas. Secondly, sites on the
Rottnest coast would have been destroyed as the
newly formed island was reduced 40% in area
through continued middle Holocene sea level rise
(Dortch and Hesp 1994; Playford 1997). Site loss
has probably continued, given that nearly all of the
recorded artefacts noted above are from actively
eroding sea cliffs. Thirdly, the island’s dense scrub
and dune cover restrict archaeological visibility.
These two factors may help explain the similarly
sparse artefact record on Garden Island.
Sea floor appraisal
In the 1990s, dive searches on parts of the sub-
merged shelf failed to identify any cultural evidence.
However, systematic recording of former landscape
features on the sea floor has the potential to reveal
past terrestrial settings, e.g. calcrete root or solution
pipes (‘pinnacles’) and other Tamala Limestone
landscape features are identifiable at varying depths
(Dortch and Dortch 2012: Figure 8; Dortch and
Hesp 1994: Figure 7). Searches of carbonate shell
and sand tailings dredged from the floor of
Cockburn Sound in industrial operations revealed
lumps of reddish cemented soil and pieces of pre-
served wood indicative of former terrestrial settings
buried beneath marine shelly sand (Dortch 1991).
Estuary histories
In the earlier stages of its inundation through late
Quaternary sea level rise, the Rottnest Shelf should
AUSTRALIAN ARCHAEOLOGY 11
have been a viable living space for successive
Aboriginal populations. Little can be said about the
range of terrestrial and aquatic habitats present
there during this long transition, though one seem-
ingly exploitable habitat continuing to exist was the
ancestral Swan River Estuary, however much altered
by shifts in its hydrology and configuration, and in
the reordering of its biota. Although nothing is
established about this ancestral estuary as the shelf
was progressively submerged, palaeontological inves-
tigations give insight into the estuary’s hydrological
and biological development since the middle
Holocene (Kendrick 1977; Yassini and Kendrick
1988). Two marine shell beds, one at Guilford
radiocarbon-dated 7.1 ka and a similar one associ-
ated with marine microfauna dated 6.4 ka at Point
Waylen, are matched by undated, similar shell beds
at localities upstream and downstream from Point
Waylen. The Guildford molluscan assemblage indi-
cates that during the early to middle Holocene the
Swan River Estuary experienced less winter flooding
than today, and was more open to the sea
(Kendrick 1977:97). The estuary then had a greater
diversity of marine invertebrate fauna than it has
had since. At Point Waylen, molluscan and micro-
faunal suites suggest a change post-4 ka towards
higher rainfall, reduced faunal diversity and regres-
sion of 1 m (Yassini and Kendrick 1988:112).
The estuary in this post-4 ka phase still offered a
wide variety of crustaceans, fish, and waterfowl,
though its complement of edible molluscs was
greatly diminished. These studies of marine shell
beds from the Swan River Estuary, and ones of
similar age at Peel Inlet 60 km south, along with
marine faunal records from estuaries on the Indian
and Southern Ocean coasts of south-western
Australia, support the view that resource-rich estua-
ries were present on the outer parts of the continen-
tal shelves throughout late Quaternary phases of sea
level rise (Dortch 1999:30–32; Dortch et al.
1984:93). In considering these regional estuarine his-
tories, Hodgkin and Kendrick (1984:91) state: ‘the
estuaries and their biota …had a degree of continu-
ity [throughout] …many rises and falls of
Pleistocene sea level’.
Lack of faunal remains
Throughout the GSR, the only macrofaunal speci-
men having even provisional cultural associations is
the emu eggshell fragment from the chert artefact-
bearing palaeosol at Little Armstrong Bay (Figures 2
and 6). The absence of specific archaeological indi-
cations for subsistence activities in the GSR is not
surprising. At present, nearly the whole of the pre-
sent archaeological record for estuarine fishing on
both coasts in south-western Australia –confirmed
by oral traditions and diverse historic accounts to
have been a crucially important economic activity –
is limited to a 130 km reach of the Southern Ocean
coast. Even here, with some 30 stone structures
interpreted as tidal weirs on the shores of Broke
and Wilson Inlets, Oyster Harbour (King George
Sound) and several smaller estuaries, the sole faunal
record for estuarine fishing is from Katelysia Rock
Shelter at Wilson Inlet (Dortch 1999). Throughout
south-western Australia, the exceptional records of
faunal remains are those excavated from Tamala
Limestone cave and rock shelter occupation deposits
(Dortch 2004; Monks et al. 2016). In the GSR this
absence of direct evidence for past subsistence is
even more noticeable today than was observed in
the first field surveys in the region a half-century
ago (Hallam 1972).
Conclusion
The Greater Swan Region is at the heart of a dis-
tinct linguistic and cultural block that provides
invaluable information about past Aboriginal sub-
sistence and traditional land usage. Rottnest Island,
in particular, has provided 10 artefacts, including
three dated c.13–17 ka, representing Pleistocene use
of the former Greater Swan Region during marine
regression. Together with pollen cores from Barker
Swamp, the Rottnest record indicates potential for
reconstruction of occupation patterns at this time.
The totality of Pleistocene finds across the GSR
indicates occupation patterns centred on persistent
wetlands, supporting the view that former wetland
areas will be the most archaeologically prospective
locations on the submerged parts of the coastal
plain. As yet, no sites on the Greater Swan Region
provide abundant biotic remains that would give
direct insight into Aboriginal land usage, largely
because of post-depositional effects of erosion and
poor preservation of biotic remains in sands.
Appraisal of late Quaternary human adaptation
throughout the Greater Swan Region study area will
depend upon understanding the territorially and
seasonally adaptive arrangements of mainland
Aboriginal occupation, with reference to occupa-
tional records –both archaeological data and histor-
ical accounts –from here and other coastal parts of
south-western Australia and elsewhere along the
continental littoral. The record of stone artefacts
identified on offshore islands broadens the perspec-
tive of Late Quaternary Aboriginal activities in this
region. Further archaeological study of these activ-
ities on Rottnest Island should include survey of
palaeosols along the littoral and residual dunes
12 J. DORTCH AND C. DORTCH
overlying aeolianite successions, and test excavation
around extant and former fresh-water swamps.
Acknowledgements
Fieldwork on Rottnest Island (Wadjemup) was supported
by Nyoongar family groups, including the Corunna,
Bodney, Bropho, Hume, Jacobs, McGuire and Wilkes
families, and by the Rottnest Island Authority (RIA),
including Aboriginal Heritage Officers Ezra Jacobs and
Reg Yarran, and RIA staff Wendy Bradshaw, Ron
Priemus, Dave Tunnecliffe, Patsy Vizents, Harriet Wyatt,
Mark Ward and Hannah Eames. Mack McCarthy and
Ross Anderson of the Western Australian Museum
Maritime Archaeology Department transported us to
Carnac Island. Moya Smith and Ross Chadwick, WAM
Anthropology and Archaeology Department, enabled
access to collections; Alice Beale, WAM, photographed
the artefact in Figure 8. Franc¸ois Mazieres prepared
Figure 3. Fieldwork on mainland sites, including Fiona
Stanley Hospital, was supported by Nyoongar families of
the Perth region. We thank all these individuals, families
and institutions for their essential help.
Disclosure statement
The authors report no potential conflict of interest.
References
Anderson, J. 1984 Between plateau and plain: flexible
responses to varied environments in Southwestern
Australia. Occasional Papers in Prehistory 4. Canberra:
Research School of Pacific Studies, Australian National
University.
Backhouse, J. 1993 Holocene vegetation and climate
record from Barker Swamp, Rottnest Island, Western
Australia. Journal of the Royal Society of Western
Australia 76(2):53–61.
Bowdler, S.B., L. Strawbridge and M. Schwede 1991
Archaeological mitigation in the Perth Metropolitan
Region. Australian Archaeology 32:21–25.
Bronk Ramsey, C. 2009 Bayesian analysis of radiocarbon
dates. Radiocarbon 51(01):337–360.
Brooke, B., J. Creasey and M. Sexton 2010 Broad-scale
geomorphology and benthic habitats of the Perth
coastal plain and Rottnest Shelf, Western Australia,
identified in a merged topographic and bathymetric
digital relief model. International Journal of Remote
Sensing 31(23):6223–6237.
Brooke, B.P., J.M. Olley, T. Pietsch, P.E. Playford, P.W.
Haines, C.V. Murray-Wallace and C.D. Woodroffe
2014 Chronology of Quaternary aeolianite deposits and
the drowned coastline of southwestern Western
Australia –a re-appraisal. Quaternary Science Reviews
93:106–124.
Churchill, D.M. 1959 Late Quaternary eustatic changes in
the Swan River district. Journal of the Royal Society of
Western Australia 42(2):53–55.
Clarke, J. and C.E. Dortch 1977 A 10,000 year BP radio-
carbon date for archaeological finds within a soil of the
Spearwood Dune System, Mosman Park, WA. Search
8(1–2):36–38.
Collins, L.B. 1988 Sediments and history of the Rottnest
Shelf, southwest Australia: a swell-dominated, non-
tropical carbonate margin. Sedimentary Geology
60(1–4):15–49.
Department of Planning, Lands and Heritage 2018:
Aboriginal heritage site data. Retrieved 18 March 2018
from <https://www.daa.wa.gov.au/heritage/place-
search/>
Dortch, C.E. 1979 Devil’s Lair, an example of prolonged
cave use in south-western Australia. World Archaeology
10(3):258–279.
Dortch, C.E. 1991 Rottnest and Garden Island prehistory
and the archaeological potential of the adjacent contin-
ental shelf, Western Australia. Australian Archaeology
33:38–43.
Dortch, C.E. 1999 Archaeological assessment of
Aboriginal estuarine fishing on the Southern Ocean
coast of Western Australia. In J. Hall and I. J.
McNiven (eds), Australian coastal archaeology,
pp.25–35. Canberra: ANH Publications, Department of
Archaeology and Natural History, Research School of
Pacific and Asian Studies, The Australian National
University.
Dortch, C.E. 2002 Modelling past Aboriginal hunter-gath-
erer socio-economic and territorial organisation in
Western Australia’s lower South-west. Archaeology in
Oceania 37(1):1–21.
Dortch, J. 2004 Palaeo-environmental change and the per-
sistence of human occupation in South-western
Australian Forests. British Archaeological Reports
International Series 1288. Archaeopress, Oxford.
Dortch, C.E. and J. Dortch 2012 Archaeological evidence
for early human presence in the western reaches of the
Greater Swan Region, Western Australia. Fremantle
Studies 7:51–76.
Dortch, J., C. Dortch, D. Cuthbert and S. Fisher 2009
Report of archaeological excavations and related ethno-
graphic consultations during Stage A forward works,
Fiona Stanley Health Precinct. Unpublished report pre-
pared for Appian Group and Department of Housing
and Works.
Dortch, C.E. and G. Gardner 1976 Archaeological sites in
the Northcliffe district, Western Australia. Records of
the Western Australian Museum 4(3):257–294.
Dortch, C.E. and P.A. Hesp 1994 Rottnest Island artifacts
and palaeosols in the context of Greater Swan Region
prehistory. Journal of the Royal Society of Western
Australia 77(1):23–32.
Dortch, J. and F. Hook 2017 Archaeology in the age of
alternative facts: archaeological investigations and Roe
8. In Gaynor, A., P. Newman and P. Jennings (eds),
Never again: reflections on environmental responsibility
after Roe 8, pp.107–116. Perth: UWA Press.
Dortch, C.E., G.W. Kendrick and K. Morse 1984
Aboriginal mollusc exploitation in southwestern
Australia. Archaeology in Oceania 19(3):81–104.
Dortch, C.E. and K. Morse 1984 Prehistoric stone arte-
facts on some offshore islands in Western Australia.
Australian Archaeology 19:31–47.
Ferguson, W.C. 1980 Fossiliferous chert in Southwestern
Australia after the Holocene transgression: a behav-
ioural hypothesis. The Artefact 5(3–4):155–169.
Ferguson, W.C. 1981 Archaeological investigations at the
Quininup Brook site complex, Western Australia.
Records of the Western Australian Museum 8(4):
609–637.
Glover, J.E. 1984 The geological sources of stone for arte-
facts in the Perth Basin and nearby areas. Australian
Aboriginal Studies 19841:17–25.
AUSTRALIAN ARCHAEOLOGY 13
Glover, J.E. and R. Lee 1984 Geochemistry and proven-
ance of Eocene chert artifacts, south-western Australia.
Archaeology in Oceania 19(1):16–20.
Gouramanis, C., J. Dodson, D. Wilkins, P. De Deckker
and B.M. Chase 2012 Holocene palaeoclimate and sea
level fluctuation recorded from the coastal Barker
Swamp, Rottnest Island, south-western Western
Australia. Quaternary Science Reviews 54:40–57.
Hallam, S.J. 1972 An archaeological survey of the Perth
area, Western Australia: a progress report on art and
artefacts, dates and demography. Australian Institute of
Aboriginal Studies Newsletter 3:11–19.
Hallam, S.J. 1975 Fire and hearth: a study of Aboriginal
usage and European usurpation in south-western
Australia. Canberra: Australian Institute of Aboriginal
Studies.
Hallam, S.J. 1987 Coastal does not equal littoral.
Australian Archaeology 25:10–29.
Hamm, G., P. Mitchell, L.J. Arnold, G.J. Prideaux, D.
Questiaux, N.A. Spooner, V.A. Levchenko, E.C. Foley,
T.H. Worthy, B. Stephenson, V. Coulthard, C.
Coulthard, S. Wilton and D. Johnston 2016 Cultural
innovation and megafauna interaction in the early
settlement of arid Australia. Nature 539(7628):280–283.
Hearty, P.J. 2003 Stratigraphy and timing of eolianite
deposition on Rottnest Island, Western Australia.
Quaternary Research 60(02):211–222.
Hodgkin, E.P. and G.W. Kendrick 1984 The changing
aquatic environment 7000 BP to 1983 in the estuaries
of south western Australia. In E. P. Hodgkin (ed.),
Estuarine environments of the Southern Hemisphere,
pp.85–95. Bulletin 16.1. Perth: Department of
Conservation and Environment.
Kendrick, G.W. 1977 Middle Holocene marine molluscs
from near Guildford, Western Australia, and evidence
for climatic change. Journal of the Royal Society of
Western Australia 59(4):97–104.
Lyon, R.M. 1833 A glance at the manners and language
of the Aboriginal inhabitants of Western Australia,
with a short vocabulary. Perth Gazette and Western
Australian Journal 13:51–52.
Marwick, B. 2002 An Eocene fossiliferous chert artefact
from Beacon Island: first evidence for prehistoric occu-
pation in the Houtman Abrolhos, Western Australia.
Records of the Western Australian Museum 20(4):
461–464.
Mattner, J., O. Lenihan, F. Wiig and K. Edwards 2014
Excavations at archaeological site DAA #17502
“Adelaide Street”in High Wycombe, Perth.
Unpublished report prepared for Perth Airport Pty Ltd.
McArthur, W.M. 1991 Reference soils of south-western
Australia. Western Australia, Perth: Department of
Agriculture and Food.
McDonald, J., W. Reynen, F. Petchey, K. Ditchfield, C.
Byrne, D. Vannieuwenhuyse, M. Leopold and P. Veth
2018 Karnatukul (Serpent’s Glen): a new chronology
for the first occupation of Australian deserts by 50,000
years ago. PLoS One 13(9):e0202511.
Meagher, S.J. 1974 The food resources of the Aborigines
of the south-west of Western Australia. Records of the
Western Australian Museum 3(1):14–65.
Monks, C., J. Dortch, G. Jacobsen and A. Baynes 2016
Pleistocene occupation of Yellabidde Cave in the north-
ern Swan Coastal Plain, southwestern Australia.
Australian Archaeology 82(3):275–279.
Moore, G.F. 1978 [1884] A descriptive vocabulary of the
language in common use amongst the Aborigines of
Western Australia. (Facsimile ed.). Perth: University of
Western Australia Press.
Mulvaney, D.J. and J. Kamminga 1999 Prehistory of
Australia. St Leonards: Allen and Unwin.
Newsome, J. and E.J. Pickett 1993 Palynology and palaeo-
climatic implications of two Holocene sequences
from south-western Australia. Palaeogeography,
Palaeoclimatology, Palaeoecology 101(3–4):245–261.
Nicholas, W. A., I. Borissova, L. Radke, M. Tran, G.
Bernardel, D. Jorgensen, J. Siwabessy, A. Carroll and T.
Whiteway 2013 Seabed environments and shallow geol-
ogy of the Vlaming Sub-Basin, Western Australia: mar-
ine data for the investigation of the geological storage
of CO2. GA0334 post-survey report. Record 2013/09.
Canberra: Geoscience Australia.
Nicholas, W. A., F. J. F. Howard, A. G. Carroll, P. J. W.
Siwabessy, M. Tran, L. Radke, K. Picard and R.
Przeslawski 2014 Seabed environments and shallow sub-
surface geology of the Vlaming Sub-basin, offshore Perth
Basin: Summary results from marine survey. GA0334.
Record 2014/49. Canberra: Geoscience Australia.
O’Connell, J. and J. Allen 2015 The process, biotic
impact, and global implications of the human coloniza-
tion of Sahul about 47,000 years ago. Journal of
Archaeological Science 56:73–84.
O’Leary, M.J., I. Ward, M.M. Key, M.S. Burkart, C.
Rawson and N. Evans 2017 Challenging the ‘offshore
hypothesis’for fossiliferous chert artefacts in south-
western Australia and consideration of inland trade
routes. Quaternary Science Reviews 156:36–46.
Pearce, R.H. 1978 Changes in artefact assemblages during
the last 8 000 years at Walyunga, Western Australia.
Journal of the Royal Society of Western Australia 61(1):
1–10.
Pearce, R.H. and M. Barbetti 1981 A 38,000-year old site
at Upper Swan, Western Australia. Archaeology in
Oceania 16(3):173–178.
Pickett, E. J. 1997 The late Pleistocene and Holocene
vegetation history of three lacustrine sequences from
the Swan Coastal Plain, Southwestern Australia.
Unpublished PhD thesis, Geography Department, The
University of Western Australia.
Playford, P.E. 1983 Geological research on Rottnest
Island. Journal of the Royal Society of Western Australia
66(1–2):10–15.
Playford, P.E. 1988 Guidebook to the geology of
Rottnest Island. Perth: Geological Society of Australia,
WA Division, and Geological Survey of Western
Australia.
Playford, P.E. 1997 Geology and hydrogeology of
Rottnest Island. In Vacher, H.L. and T. Quinn (eds),
Geology and hydrogeology of carbonate islands,
pp.783–810. Developments in sedimentology 54.
Amsterdam: Elsevier.
Playford, P.E., A.E. Cockbain, P.F. Berry, A.P. Roberts,
P.W. Haines and B.P. Brooke 2013 The geology of
Shark Bay. Bulletin 146. Perth: Geological Survey of
Western Australia.
Playford, P.E., A.E. Cockbain and G.H. Low 1976 Geology
of the Perth Basin Western Australia. Bulletin 124.
Perth: Geological Survey of Western Australia.
Price, D.M., B.P. Brooke and C.D. Woodroffe 2001
Thermoluminescence dating of aeolianites from Lord
14 J. DORTCH AND C. DORTCH
Howe Island and south-west Western Australia.
Quaternary Science Reviews 20(5–9):841–846.
Schwede, M. 1983 Supertrench –phase two: a report on
excavation results. In M Smith (ed), Archaeology at
ANZAAS 1983, pp. 52–62. Perth: Western Australian
Museum
Searle, D.J., V. Semeniuk and P.J. Woods 1988
Geomorphology, stratigraphy and Holocene history of
the Rockingham-Becher Plain, South-western Australia.
Journal of the Royal Society of Western Australia 70:
89–109.
Semeniuk, V. 1986 Holocene climate history of coastal
southwestern Australia using calcrete as an indicator.
Palaeogeography, Palaeoclimatology and Palaeoecology
53(2–4):289–308.
Smith, M. V. 1993 Recherche a L’Esperance: a prehistory
of the Esperance region, South-western Australia.
Unpublished PhD thesis, Department of Archaeology,
University of Western Australia, Perth.
Smith, M.V. 1999 Southwest Australian coastal econo-
mies:anewreview.InJ.HallandI.J.McNiven
(eds), Australian coastal archaeology, pp.15–24.
Canberra: ANH Publications, Department of
Archaeology and Natural History, Research School of
Pacific and Asian Studies, The Australian National
University.
Stirling, C.H., T.M. Esat, K. Lambeck and M.T.
McCulloch 1998 Timing and duration of the Last
Interglacial: evidence for a restricted interval of wide-
spread coral reef growth. Earth and Planetary Science
Letters 160(3–4):745–762.
Stirling, C.H., T.M. Esat, M.T. McCulloch and K.
Lambeck 1995 High-precision U-series dating of corals
from Western Australia and implications for the timing
and duration of the Last Interglacial. Earth and
Planetary Science Letters 135(1–4):115–130.
Szabo, B.J. 1979 Uranium-series age of coral reef growth
on Rottnest Island, Western Australia. Marine Geology
29(1–4):11–15.
Ward, I., T.J. Pietsch, E.J. Rhodes, G.H. Miller, J.
Hellstrom and C.E. Dortch 2016 Chronostratigraphic
context for artefact-bearing palaeosols in late
Pleistocene Tamala Limestone, Rottnest Island,
Western Australia. Journal of the Royal Society of
Western Australia 99(1):17–26.
Yassini, I. and G.W. Kendrick 1988 Middle Holocene
ostracodes, foraminifers and environments of beds at
Point Waylen, Swan River Estuary, southwestern
Australia. Alcheringa 12(2):107–121.
AUSTRALIAN ARCHAEOLOGY 15