THE SOUTH-WEST OF WESTERN AUSTRALIA

ABORIGINAL FISH TRAP — ALBANY

THE SOUTH-WEST OF WESTERN AUSTRALIA

PREFACE

Although Western Australia has an area comparable with that of India,
ninety per cent of her population, which is little more than a million, live in
the south-west corner of the State, in an area similar to that of the British
Isles; and seventy per cent of the residents there live in metropolitan Perth.
It is in the south-west, isolated by oceans and infertile, arid country, where
most of the State’s forests are found, where most of its intensive agriculture
is practised, and where most of the population find their recreation. Here the
Aboriginal inhabitants first came into extensive contact with Europeans, and
here man has had his greatest impact on the environment of the State.

To mark the occasion of the 45th Congress of the Australian and New
Zealand Association for the Advancement of Science, held in Perth during
August 1973, the Council of the Royal Society of Western Australia decided to
devote part of the Journal to a series of reviews concerning this southwestern
corner. It was felt that the papers would be of interest to delegates attend-
ing the Congress, and also to residents of the State who may wish for access
to information not readily obtained from more detailed papers, which are
scattered through the specialist literature.

There is a further reason for believing that this is an appropriate time
to review information concerning the south-west. There is a rapid increase
in the rate of accumulation of knowledge about the State. Extensive exploration
for minerals is in progress, and there is increasing pressure from the general
public to come to terms with the environment. The number of scientists and
other specialists is increasing. The Western Australian Institute of Technology
has recently been established as a major tertiary institution, and in 1975,
when the first students enter the new Murdoch University, the University of
Western Australia will cease to be one of the world’s most isolated universities,
and any lingering traces of intellectual isolation will disappear.

The Society wishes to express its gratitude to the authors of the review
papers; to Professor Webb (Honorary Organizing Secretary of the 45th ANZAAS
Congress) for his encouragement and co-operation; and to the Government
Printing Office for generous assistance in the production of the Journal.

A. J. McComb,

Honorary Editor,

Botany Department,

University of Western Australia.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July, 1973.

1

CONTENTS

Page

The geology of southwestern Australia — a review.

By M. H. Johnstone, D. C. Lowry and P. G. Quilty 5

Landforms and soils of southwestern Australia.

By M. J. Mulcahy 16

Species diversity in the southwestern flora.

By N. G. Marchant 23

Animal and plant speciation studies in Western Australia.

By H. E. Paterson and S. H. James 31

Aboriginal man in southwestern Australia.

By D. Merrilees, W. C. Dix, S. H. Hallam, W. H. Douglas and R. M.

Berndt 44

European man in southwestern Australia.

By G. C. Bolton and D. Hutchison 56

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July, 1973.

3

1. — The geology of southwestern Australia — a review

By M. H. Johnstone'. D. C. Lowry' and P. G. Quilty'

Abstract

The gneisses and granites which constitute
the bulk of the Archaean Yilgarn Block give a
surprisingly consistent age of 2.667 ± 27 m.y.
However, metasedimentary belts infolded into
this gneissic terrain contain boulders dated at
3.000 m.y., so the younger figure probably repre-
sents the time of stabilization of the radiogenic
elements of the shield. Flanking the shield are
belts of Proterozoic metamorphics which have
been dated from 1300 m.y. to 670 m.y.

The first important Phanerozoic sedimentation
consists of Early Permian glacially-derived rocks
deposited in downwarps to the east and west of
the main shield. The western trough continued
to receive marine and lacustrine sediments
throughout the Early Permian. In the southern
Perth Basin, graben development began in the
Late Permian with deposition of thick Upper
Permian Coal Measures and coarse fluvial
Triassic sandstone. In the north, the Upper
Permian is represented by paralic to continental
sandstone and the Lower Triassic by marine
shale. In this northern area coarse sandstone
marking the onset of graben tectonics does not
appear until the Late Triassic. This tectonism
and depositional style was renewed in the Late
Jurassic to Early Neocomian after a period of
comparative stability in the Middle Jurassic.

The second tectonic phase also saw the develop-
ment of a graben across the southern margin of
Australia which again was filled with coarse
detritus. Both of these graben were filled with
detritus from the shield by an extensive river
system whose relics are the chains of salt lakes
seen today. In the Late Neocomian. spreading
between India and Antarctica commenced and
India moved away from Western Australia along
the Wallaby-Pertli transform. Marine shelf sedi-
mentation then commenced on the west coast
and spread into the southern graben in the
Aptian when a widespread transgression occurred
in the Eucla and Ollicer Basins. In Eocene
times, the spreading between An tralia and
Antarctica commenced and. for the first time,
warm waters from the Indian Ocean could enter
the southern basins. Sedimentation in the
Paleocene, Eocene and Miocene was largely car-
bonate but minor terrigenous material derived
from the shield was deposited in the Perth Basin
in the Paleoeene-Early Eocene and near Albany
in the Late Eocene.

Introduction

This paper is an abbreviated geological history
of the southwestern portion of Australia. It high-
lights those phases of geological evolution which
have helped to shape the landforms, soils, and
the varied environments for the development of
its present day unique flora and fauna.

Such a paper cannot be both all-embracing and
at the same time, definitive. The later facets of
the geological history of the southwestern corner
of the continent which led to the isolation of
this area and the development of its biological
uniqueness are stressed to the detriment of the
equally interesting history of the buildup of the
Archaean nucleus which forms the core of the

'West Australian Petroleum Ptv Limited. G.P.O. Box
Cl 580. Perth 6001.

Australian continent. The latter has been well
covered in Special Publication No. 3 of the
Geological Society of Australia '1971.)

A pioneer study of the Perth basin was made
by Campbell (1910». who examined its northern
part and made two major contributions in recog-
nising the existence of a major fault (the Darling
Fault) and also in inferring a glacial origin for
the Early Permian sediments. Jutson <1934‘
produced a monumental work on Western
Australian geomorphology and discussed in some
detail the salt lake system. McWhae et al..
( 1958 > produced a review of Western Australian
stratigraphy and this work is still the standard
reference for the Phanerozoic.

Tectonic elements

The Archaean Yilgarn Block (Fig. D is the
major nucleus of the present Australian contin-
ent. but is one of many such nuclei of the
Gondwanaland supercontinent (Fig. 2AL The
dominant trend of gneissic foliation and
orientation of infolded metasediments in the
block is north-northwest.

Along the southern margin of the block,
metamorphics of the Fraser Range and Albany-
Esperance Blocks trend northeast-southwest and
east-west respectively, almost at right angles to
the grain of the abruptly truncated shield.

The variation of mineral association and
metamorphic grade in a westerly direction in the
Albany-Esperance Block is matched exactly by
similar east-west trending rocks in the vicinity
of the Windmill Islands in Eastern Antarctica
(Oliver, 1972). providing one of the strongest
pieces of evidence for the geological fit of Antarc-
tica and Australia in the reconstruction of
Gondwanaland.

Along the western margin of the Yilgarn
Block, the Archaean Shield is separated from
Proterozoic high grade metamorphic rocks
(garnet granulites with a meridional trend) by
the Darling Fault — a major crustal feature 1000
km long, trending north-south and having up
to 15000 m of vertical movement since the
Permian and with a history since at least the
Proterozoic. To the north, the strike of the
Proterozoic rocks becomes more northeast-
southwest and they trend around the northern
margin of the Yilgarn Block. Overlying the
Archaean on the western margin of the block
are several sequences of unmetamorphosed
sediments which dip west into the present day
Perth Basin (Billeranga and Moora Groups.
Cardup Group L The Billeranga and Moora
Groups (Fig. 1 > may be unmetamorphosed
margin facies which are age equivalents of the
garnet granulites in the trough farther west.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July. 1973.

5

During the Phanerozoic, the Precambrian
crystalline shield acted as a stable emergent
craton and sedimentation was restricted to
marginal downwarps. Along the western margin
of the shield a downwarp developed early in the
Palaeozoic and was reactivated in the Permo-

Carboniferous. Between the Late Triassic and
the Early Cretaceous (Neocomian) a deep graben
developed in the centre of this downwarp to
accommodate the 15 , 000 + m of mainly continen-
tal sediments of the Perth Basin. Although the
north-south Darling Fault forms the eastern

CAINOZOIC
MESOZOIC
PALAEOZOIC
PROTEROZOIC

ARCHAEAN (METASEDS. STIPPLED)
NEOCOMIAN OCEANIC CRUST
EOCENE OCEANIC CRUST
BATHYMETRY IN KILOMETRES
PROMINENT LINEATION
AGE IN 10 9 YEARS

KILOMETRES

AFTER TECTONIC MAP OF AUSTRALIA W7 1

Figure 1 .—Geological sketch map of southwestern Australia.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July, 1973.

6

boundary between this graben and the shield
(Jones and Pearson, 1972,), an equally promin-
ent fault alignment in the graben is NNW,
paralleling the grain of the Archaean Shield and
that of a prominent transform fault in the
Indian Ocean postulated by Falvey (1972).

Along the southern margin of the shield, a
graben paralleling the lineation of the Albany-
Esperance Block is inferred to have developed
at least as early as the Late Jurassic but possibly
even contemporaneously with the earlier Triassic
formation of the Perth Basin. Little is known
of the early history of this graben, since no
remnant of it has yet been discovered off the
southern coast of Western Australia, and much
of its history is inferred from the Elliston Trough
and the Robe-Penola Trough in South Australia.
The whole of the western portion of this graben
probably now lies on the Antarctic plate. A
marginal marine Eocene sequence laps the
southern coast and, probably, a thin veneer of
Miocene limestones occurs offshore, overlying the
crystalline basement beneath the southern shelf.

The eastern margin of the Yilgarn Block has
never been downfaulted. Gentle epeirogenic
warping formed the Eucla and Officer Basins
where several relatively thin, flatlying sequences
of sediments were deposited. In the latter,
Ordovician, Permian, and Early Cretaceous
sediments were laid down and the later Eucla
Basin contains Cretaceous and Tertiary
sequences.

The Precambrian crystalline shield
and associated sediments

The dominant geological element in south-
western Australia is the Archaean Yilgarn
Block which forms the nucleus of the Western
Australian Shield. This extends from near the
south coast for 900 km to the north and has
an east-west width of 700 km. Around this
nucleus, belts of younger Precambrian rocks
have accreted and its western margin is now
marked by the deep Mesozoic graben of the
Perth Basin. The eastern and southern margins
of the shield have been onlapped by Cretaceous
and Tertiary sediments.

Apart from the dissected western margin, the
Yilgarn Block has an ancient, subdued land
surface, and most of the older rocks are obscured
by the widespread cover of laterite and its
associated soil horizons which form extensive
sandplains. The shield is composed mainly of
gneisses and granites, with minor infolded belts
of metasediments with a general north-
northwest strike. These metasediments reveal
different grades of regional metamorphism. The
westernmost belt — the Jimperding belt (Prider,
1944), which has undergone the deepest dissec-
tion — consists of sillimanite-zone and kyanite-
zone rocks. The other major belts, notably those
of Southern Cross and Kalgoorlie, show degrees
of metamorphic grade which lessen in an
easterly direction (Fig. 1). The Jimperding and
Southern Cross belts contain thin bedded,
shallow water sediments whereas those of the
Kalgoorlie region contain a thick eugeosynclinal
sequence containing pillow basalts. The sedi-
mentary and volcanic rock suites and the
chemical composition and areal distribution of

the various volcanic and hypabyssal rocks of the
Eastern Goldfields are similar to those of
present day island arcs and subduction zones.
Possibly these Archaean greenstone belts repre-
sent the earliest zones of thick sedimentation
when the “continental” crust was little more
than a basaltic differentiate from the mantle
(White et al., 1971).

The original crust on which these ancient
sediments were laid down has not yet been posi-
tively identified or dated radiometrically. The
belts of metasediments are folded into the
gneissic complex of the shield which probably
represents a granitization of some of the earlier
crustal differentiate. Alternatively, the green-
stone belts became closed to the loss of radio-
genic daughter elements earlier than the gneissic
terrain of the catazonal basement rocks. Thus
the geologically older gneisses may give younger
radiometric ages (Windley and Bridgwater,
1971). The age of granitization is remarkably
uniform over the Yilgarn Block, giving a Rb/Sr
age of 2,667 ± 27 m.y. Granitic boulders within
conglomerates in the Kalgoorlie greenstone belt
have given an age of 3,000 m.y. which is con-
sistent with the above theory (Compston and
Arriens, 1968). It is interesting to note that
the age of a prominent metasomatic event in
the Kalgoorlie area (2,670 zb 30 m.y.) agrees
closely with the widespread age of the gneissic
parts of the shield (Arriens, 1971).

A final major event in the history of the
Archaean shield was the emplacement of large
east-west trending basic dykes near Norseman, of
which the Jimberlana Dyke is the best known
(Campbell et al., 1970; and McCall and Leish-
man, 1971). These, and the associated gold
mineralization at Kalgoorlie, are dated at 2,400
± 40 m.y. (Fig. 1).

To the southeast of the Yilgarn Block, and
striking approximately at right angles to the
north-northwest trend of the main shield is a
zone of augen gneisses with associated amphi-
bolites and granulites which were welded onto
its southern margin approximately 1,300 m.y.
These rocks appear to merge into the east-
west trending gneisses, granites, and metasedi-
ments of the Albany-Esperance Block which are
dated at 1,150 ± 40 m.y. (Compston and Arriens,
1968) and which form the entire southern
boundary of the shield.

Movement on the Darling Fault formed the
Perth Basin — a graben with as much as 15,000
m of Late Palaeozoic and Mesozoic sediments.
The Precambrian high grade metamorphics
which floor this graben are much younger than
the adjacent shield. To the north of Geraldton,
the outcropping garnet granulites of the North-
ampton Block (Fig. 1) have been dated at
1,040 ± 50 m.y. whereas similar rocks from the
Leeuwin-Naturaliste Block in the southwestern
corner of the State give an isochron dating of
670 ± 25 m.y. (Compston and Arriens, 1968).
The age of the younger granulites is also
registered in oveprinted micas and pegmatites
along the western margin of the Yilgarn Block.

Also along the western margin of the Yilgarn
Block, two sequences of relatively unaltered
sed'ments rest unconformably on the Archaean.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

7

Figure 2A.— Reconstruction of the Gondwanaland continents (after several authors) showing major Precambrian
cratonic elements and later Precambrian lineaments. Radiometric ages are shown in 10 9 years. Note that the promi-
nent “Pan-African Event’’ of 0.55 x 10 9 years is primarily an overprint of a later tectonic event on earlier Proterozoic
belts. The zones along which Gondwanaland broke in the Neocomian and in the Eocene are shown.

ELLISTON(POLDA) TROUGH
(LATE JURASSIC CONTINENTAL
SEDIMENTS)

ROBE-PENOLA TROUGH
JURASSIC NEOCOMIAN
CONTINENTAL SEDIMENTS)

INFERNO

TECTONICS OF MARGINS OF
SOUTHWESTERN AUSTRALIA DURING THE
LATE JURASSIC AND EARLY NEOCOMIAN

200

KILOMETRES

LATE JURASSIC
CONTINENTAL SEDIMENTS

l J AREA OF SEDIMENTATION

l 1 LATE NEOCOMIAN OCEANIC CRUST

-2- BATHYMETRY IN KILOMETRES
— 8— ISOPACHS( PERTH BASIN)

INFERRED DRAINAGE SYSTEM

Figure 2B. — Late Jurassic to Neocomian sedimentation marginal to southwestern Australia and the inland drain-
age system.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

8

In the vicinity of Perth, the Cardup Group can
be dated as younger than the widespread peg-
matite formation and dolerite intrusions of 700-
750 m.y. (and probably no younger than 500-590
m.y.) ; and 350 km north of Perth, the Billeranga
Group (Fig. 1) has been tentatively dated at
1,400 m.y. (Compston and Arriens, 1968).

In summary, it can be seen that the western
and southern margins of the Archaean Yilgarn
Block are marked by zones of Proterozoic high
grade metamorphics, granites, and some un-
metamorphosed sediments, the depositional and
structural trends of which bear no relation to
those of the Archaean nucleus. As can be
demonstrated in other parts of Gondwanaland,
these Proterozoic orogenic zones provide the
lines of weakness along which the intitial graben
formation (“rifting”) and final drifting apart
occurred (Fig. 2).

Older Palaeozoic

In the northern Perth Basin, more than
1050 m of cross bedded, fine to coarse fluviatile
sandstones of probable Ordovician to Early
Silurian age crop out (Konecki et al., 1958). The
Tumblagooda Sandstone may be a piedmont
deposit associated with early movements on the
northern portion of the Darling Fault for,
although the entire sequence appears to be
fluviatile or deposited in extremely shallow
water, geophysical evidence indicates that the
unit may be up to 3000 m thick.

In the Officer Basin during the Ordovician,
very extensive basalt flows covered folded Pro-
terozoic sediments. Subsequently in the Ordo-
vician, Devonian or Carboniferous, there was
widespread deposition of two fine grained sand-
stone units (Lennis Sandstone and Wanna Beds;
see Lowry et al., 1972). The units are inter-
preted as shallow marine sediments deposited
under the influence of strong tidal currents in
a sea that probably extended from the Canning
Basin around the southern side of the Musgrave
Block into South Australia.

Permo-Carboniferous

In the northern Perth Basin, the older
Palaeozoic fluviatile deposits are overlain un-
conformably by a sequence of poorly bedded,
poorly sorted, sandy siltstones containing abund-
ant boulders (usually up to 50 cm in diameter,
but exceptionally up to 6 m) of a great variety
of Precambrian igneous, metamorphic, and sedi-
mentary rocks. The variety of provenance,
variety of size, and incompatability of these
boulders with their fine-grained host sediment
leads to their interpretation as icerafted detritus
dropped into an epeiric sea from icebergs.
Although no more than 130 m can be measured
in any one section, it is thought that they may
attain a total thickness of 350 m. Their age is
Late Carboniferous to Early Permian (Sakmar-
ian). Up to 2400 m of similar sediments were
deposited in the Carnarvon Basin to the north,
and similar, but thinner, deposits are known
down the length of the Perth Basin and in the
Collie, Canning, and Officer Basins, so the
glaciation was widespread in Western Australia

(Fig. 3). Sediments associated with this glacia-
tion are known from Eastern Australia, and the
other Gondwanaland continents, notably India,
Africa and Antarctica.

In the northern Perth Basin, the glacial
sequence is capped by 530 m of black marine
shale (containing the Sakmarian goniatite
Metalegoceras jacksoni Etheridge Jr.) which
culminates in shellbanks rich in brachiopods,
crinoids, and bryozoans. This, in turn, is capped
by 330 m of fluviatile and coal swamp deposits,
indicating an amelioration of the harsh, glacial
climate. Three hundred metres of marine silt-
stone complete the sequence (Johnstone and
Willmott, 1966). Mild faulting and erosion of
this sequence preceded the deposition of a Late
Permian sandstone which precedes thick (300-
1000 m) Early Triassic marine shale.

Analysis of the structure and stratigraphy of
the sequences indicates that, whereas the Early
Permian glacial and later marine siltstones were
linked with the sea by a gulf which connected
with the Carnarvon Basin to the east of the
Northampton Precambrian Block, the seas of the
Late Permian and Early Triassic came into the
basin from the west of the Northampton Block.
Thus the minor faulting in the Late Permian of
the Dongara area (Hosemann, 1971) actually
dates the initiation of a set of north -north west —
south-southeast trending faults and rifts which
produced the deep graben of the Perth Basin
from Triassic to Neocomian times. This fault
trend marks a crustal weakness which also con-
trolled the inferred large transform fault which
permitted India to move away from Western
Australia during the Neocomian (Falvey, 1972).

The lower part of the Permian sequence
(including minor glacial deposits at the base) is
thin in the southern part of the Perth Basin, and
in the Collie and Wilga Basins which were prob-
ably part of the same basin of deposition at the
time. However, the accumulation of more than
2000 m of Late Permian coal measures in the
southern Perth Basin suggests that rifting began
in this area in the Late Permian. There is little
evidence of marine conditions reaching this
part of the basin during any part of the Permian.

There is no evidence of Permo-Carboniferous
to Triassic sediments along the south coast. In
the Officer Basin, the Paterson Formation and
the Wilkinson Range Beds are thin (60-100 m),
flat-lying, glacially derived, boulder-bearing
sandstones of Early Permian age. These lie
unconformably on the older Palaeozoic and are
overlain disconformably by the Early Cretaceous
Bejah Beds.

Triassic to Early Neocomian

In the northern Perth Basin, the quiet deposi-
tion of the Early Triassic marine shales and the
Middle Triassic deltaic sediments in a gently
subsiding trough was interrupted in the Late
Triassic by intense uplift of the margins of the
trough and the dumping in the rapidly subsid-
ing graben of more than 2000 m of coarse
grained fluvial sandstones (Jones and Pearson,
1972). In the southern Perth Basin, coarse
dominantly fluviatile sandstone was deposited
throughout the Triassic. Deposition continued

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

9

10

uninterrupted into the Early Jurassic when up
to 2400 m of sandstone and claystone grading
upwards into coal swamp deposits were laid
down. The climate of the Early Jurassic was
thus favourable for the development of a dense
coal swamp vegetation. In the Middle Jurassic,
a minor marine incursion from the northwest
deposited a widespread blanket about 250 m
thick of shallow water marine shale, sandstone
and limestone containing ammonites, pelecypods
and gastropods of Bajocian age (Arkell and
Piayford, 1954).

Major movements of the trough and its
borderlands in the Middle to Late Jurassic
caused further deposition of the coarse fluvial
sediments similar to those in the Triassic. Up
to 4250 m of coarse sandstones were deposited
during this major development of the Perth
Basin rift.

In the Neocomian, the locus of intense dov.n-
faulting moved westwards into the offshore
Viaming Sub-basin where up to 6000 m of fluvial
sand and estuarine shale and sand were
deposited immediately before (and possibly dur-
ing) the violent jostling of the fault blocks in
the Perth Basin graben which preceded the
active initiation of the transform fault which
moved India away from the western coastal
basins of Australia. The stage was then set for
the next phase in the evolution of the west
coast basins.

No sediments of Triassic to Neocomian age are
known from the south coast area or the Eucla
Basin. However, geophysical surveys and drilling
in the Elliston (or Polda) Trough and the
Duntroon Basin on the eastern side of the Eucla
Basin, show that graben formation commenced
in this area as early as Late Jurassic (Smith
and Kamerling, 1969). This graben formation
continued eastward into the Robe-Penola
Trough in the Late Jurassic and had reached
the eastern Otway, Bass, and Gippsland Basins
by the Neocomian (Griffiths, 1971). These
graben were the precursors of the spreading
which separated Antarctica from Australia.
Thus, although there is no published evidence
for rocks of this age off the southern coast of
Western Australia, it is logical to assume that
the graben between southwestern Australia and
Antarctica had formed by Late Jurassic times,
and possibly even by Middle Jurassic or earlier.

N eocomian - Maestrichtian
Tectonism in the Perth Basin reached a
climax in the Neocomian (Jones and Pearson,
1972). The sequence of events included major
subsidence of the Viaming Sub-basin (up to
6000 m) ; a period of arching and faulting caus-
ing local uplift and erosion of as much as 2500 m
of the recently deposited soft sediments; erup-
tion of the Bunbury Basalt at the southern end
of the basin, and subsidence of the continental
margin causing a marine transgression. These
events mark the change from continental
“rifting” to “drifting”. With generation of sea
floor between India and Australia the tectonic
stresses were relieved, and the Darling Fault and

most others ceased to move. Late subsidence
within the graben was due to compaction of
underlying sediments and general sagging of the
continental margin. The precise mechanics of
the split are still conjectural; however, it is
likely that the Perth Basin is a rift controlled
by an ancient crustal weakness that developed
into a transform fault (Falvey, 1972).

Sediments deposited later in the Neocomian
and Aptian include a thick marine shale west
of Perth (South Perth Shale) and shallow
marine and paralic glauconitic sandstone
(Leederville Sandstone) (Fig. 4). Deposition at
the southern end of the main Perth Basin was
continental, but it presumably extended around
the west side of the Leeuwin-Naturaliste Block,
where marine sediments of this age have been
recovered in deep sea' cores (Burckle et cil., 1967).

The present salt lake system (Jutson, 1934),
a relic of ancient drainage systems, is indicated
on Figure 2B and has been discussed in several
papers by Bettenay and Mulcahy (see their 1972
papers for full reference list).

The system seems to be a relic of the Late
Jurassic-Early Cretaceous drainage, which was
most active in filling the basins at that time,
but which has had only a relatively minor sedi-
mentational history since. Voluminous terri-
genous sediments of Late Jurassic age (Perth
Basin only) and Early Cretaceous (Perth and
Eucla Basins) attest to well developed drainage
systems to supply them. No sediments younger
than Early Cretaceous seem voluminous enough
to justify such a drainage system. Late Eocene
sediments in Rollos Bore, Coolgardie (Balme
and Churchill, 1959), show that the system is
pre-Eocene. Although the present salt lake
system is a relic of the most active (Late Juras-
sic to Neocomian) phase, it could also mark the
drainage system which supplied sand to the
Early Triassic.

In the Albian-Cenomanian and Senonian, thin
greensands and a chalk bed were deposited north
of Perth (Osborne Formation, Molecap Green-
sand, Gingin Chalk, and Poison Hill Greensand;
see McWhae et al., 1958). Maestrichtian sand-
stone and calcilutite occur offshore in Warnbro
No. 1.

In the Eucla Basin, a marine transgression
began in the Neocomian-Aptian (Ingram, 1968;
more probably Aptian, A. Williams, vers . comm.).
The beds of carbonaceous and glauconitic sand-
stone and shale (Madura Formation; see Lowry,
1970) lie on a deeply dissected granite surface
in the southern portion of the basin, near the
edge of the continental slope. The valleys,
which are as much as 500 m deep, may have
formed in the Late Jurassic and Neocomian by
up-arching of the flank of a graben that de-
veloped along the impending Australia-Antarc-
tica rupture. During the Aptian, much of the
Australian continent was submerged and sea
covered parts of the Canning, Officer, and Great
Artesian Basins (Skwarko, 1967). In the central
Eucla Basin, deposition of greensand continued,
perhaps intermittently, into the Senonian.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

12

Figure 4.— Major cycles of deposition around southwestern Australia after the unconformity within the Neocomian.

The epeirogenic subsidence of the Eucla Basin
is possibly part of a general subsidence of a
newly developed southern continental margin
following minor generation of sea floor between
Antarctica and Australia. This interpretation is
consistent with the palaeomagnetic data of
Wellman et al., (1969) and Weissel and Hayes’
(1972) interpretation of sea floor magnetic
anomalies. Although the latter interpretation is
taken to indicate that the major motion of Aust-
ralia away from Antarctica was initiated in the
Eocene, stratigraphic evidence from the Eucla
and Otway Basins indicates that a pronounced
marine gulf extended across the southern part
of the Continent during the Late Cretaceous.
Thus Australia became an isolated continental
mass by the uppermost Cretaceous, with only a
tenuous link joining Tasmania to Antarctica
across a transform fault.

Late Cretaceous faunas in the Gingin area and
from the offshore Perth area contain abundant
Globotruncana which Bandy (1967) would take
to be an indication of warm water conditions.
The presence of carbonate-rich sediments also
supports this view. These warm water indicators
are absent from the Eucla Basin but this absence
could be due to facies control or water circulation
rather than purely climate. It is unlikely that
warm waters from the Indian Ocean would cir-
culate freely into the narrow gulf separating
Australia and Antarctica until at least the Middle
Eocene when the southern tip of the Tasmanian
peninsula had cleared Antarctica and thus per-
mitted the warm Indian Ocean waters to flow
through the widening gulf into the South Pacific
Ocean. However, once the warm currents could
penetrate the widening gulf, warm water forms
from the Indian Ocean could migrate in an
easterly direction along Australia’s south coast.
Hence the Eocene faunas of Australia’s south
coast should show marked similarities.

Rates of sedimentation and runoff had de-
creased markedly by this time and the drainage
system probably became very subdued. This
pattern of minimal influx of terrigenous material
from the defunct river systems would have con-
tinued into the Tertiary.

Paleocene — Early Eocene

The only sediments of this age known are
from the vicinity of Perth (Fig. 4) and have
been documented by Quilty (in press). The
maximum thickness of the Kings Park Formation
near Perth is approximately 500 m of shale and
sandstone with some carbonate content. The
sediments can be differentiated into a northern
sandy marine facies and a more southerly
marine shale facies. The former may be related
to the ancestral Swan River where it enters the
Perth Basin near Walyunga National Park. The
shale facies is likely to have been deposited by
smaller streams (Canning and Helena Rivers)
flowing into a deep embayment — probably an
old submarine canyon eroded during the Late
Cretaceous and/or Early Paleocene into Creta-
ceous terrigenous sediments.

The drainage at this time is very minor com-
pared with the Early Cretaceous but probably
it ran in the same stream channels.

Very little can be said of the climate at this
time as foraminiferal faunas are almost cosmo-
politan. Globorotalia rex Martin, G. dolabrata
Jenkins and other keeled Globorotaliae may in-
dicate warm water conditions (Bandy, 1964).

Middle and Late Eocene

Marine sediments of Middle and Late Eocene
age occur in the Eucla Basin and along the
south coast between Esperance and Albany
(Cockbain, 1967, 1968, Lowry, 1970, Quilty, 1969)
— see Figure 4.

In the Eucla Basin, both Middle and Late
Eocene are present but the Plantagenet Group
to the west is so far known to contain only Late
Eocene faunas.

The Eucla Basin sediments (Hampton Sand-
stone, Wilson Bluff Limestone, Toolinna Lime-
stone) consist of up to 300 m of biogenic car-
bonate sediments containing the warm water
bivalve Spondylus. The thin sandstone at the
base suggests minor terrigenous material being
supplied by the poor drainage, probably from the
northwest.

The Plantagenet Group consists of up to 100 m
of fine sandstone, spongolite and minor lime-
stone. Near Esperance, the limestone contains
the large warm water foraminifer Asterocyclina
and the tropical alga Neomeris (Cockbain, 1967,
1969).

Thus the Middle and Late Eocene sediments
attest to warm water sedimentation and also
the presence of minor southerly drainage.

Early and Middle Miocene

Sediments of Early and Middle Miocene age
occur in both the Perth and Eucla Basins (Fig.
4).

In the Perth Basin, friable limestones with
some dolomite and chert (Stark Bay Formation
of Quilty, in press) occur offshore from Perth
and attain a thickness of some 200-250 m.
There is no terrigenous content. They contain
abundant bryozoans and foraminifera including
Lepidocyclina and keeled Globorotalia, both
warm water indicators.

The age of the sediments is latest Early
Miocene and earliest Middle Miocene.

Sedimentation in the Eucla Basin took place
over a longer period, beginning before and ceas-
ing after that in the Perth Basin. The age
limits of the Eucla Basin Miocene are not as
well established as those in the Perth Basin.
The older Miocene sequence (Abrakurrie Lime-
stone) seems to be Early Miocene (Longfordian
of southeastern Australia) and underlies the
Nullarbor Limestone and Colville Sandstone dis-
conformably. The Nullarbor Limestone and
Colville Sandstone are coeval and laterally
equivalent. Both contain Middle Miocene warm
water benthonic foraminifera which may in-
dicate sediments slightly younger than the
Middle Miocene of the Perth Basin. The Col-
ville Sandstone occurs on the northern rim of
the Eucla Basin and is probably derived from
Palaeozoic sandstones by marine erosion.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

13

Pliocene

Pliocene marine sediments occur in small,
scattered, poorly-known lenses to the north,
east and south of Perth. Darragh and Kendrick
(1971) reported ages based on mollusc faunas
and Kendrick (pers. comm.) has since sub-
stantiated a Pliocene age from Redcliffe (a Perth
suburb) on the basis of the pelagic gastropod
Hartungia typica typica Bronn.

The knowledge of the Pliocene is so far too
imperfect to make any comments on palaeoe-
cology.

Pleistocene — Recent

A wide variety of Quaternary units is developed
around the coastal margin. On the west, the
sediments of the central Perth Basin have been
eroded to form a coastal plain with alluvium
inland and a series of dune systems, lakes, inter-
dunal swamps, and relict estuaries nearer the
coast (McArthur and Bettenay, 1960). The
calcareous dune sands show varying degrees of
lithification and leaching, according to age. The
lithified dune systems and intercalated marine
lenses are known as the “Coastal Limestone”.
These form prominent hills along the coast and
the backbone of several offshore islands and
reefs. The total age range of the “Coastal
Limestone” is unknown but the unit is still form-
ing along the coast and probably on the seabed.
The greatest known age is 100,000 ± 20,000
years for reef limestone on Rottnest Island
(Teichert, 1967) .

Palaeontological age control is very poor but
Kendrick (pers. comm.) notes the presence of
early Pleistocene marine molluscs at Jandakot.
He also (Kendrick, 1960) has discussed the sig-
nificance of a Late Pleistocene mollusc fauna in
the “Coastal Limestone” at Peppermint Grove.
Details of these various Quaternary rock and
soil units are given in Dr. Seddon’s excellent
review of the Swan Coastal Plain (Seddon, 1972).

Similar dune systems occur along the south
coast and, in the Eucla Basin, the Roe Calcaren-
ite is tentatively referred to the Pleistocene. The
formation is marine with a warm water fauna
which may indicate an interglacial age for the
unit if it is Pleistocene.

Other young sediments in the Eucla Basin in-
clude kankar and dune sands discussed in some
detail by Lowry (1970).

Lateritization

Details of laterites in the southwest of Aust-
ralia will be dealt with in a subsequent article in
this volume, but some comments will be made
here on stratigraphic evidence for the time of
lateritization. In eastern arid areas the laterite
is Miocene or Oligocene in age. The evidence is
that the Late Eocene Eundynie Group around
Norseman, the Plantagenet Group around Esper-
ance and the Early Cretaceous Bejah Formation
of the Officer Basin are all lateritized whereas
the adjoining Middle Miocene Colville Sandstone
of the Eucla Basin is not.

In the extreme southwest, Pleistocene alluvium
and dune sands have been lateritized and laterite
appears to be forming at the present day on the
coastal plain where there is a temperate climate
with a strongly seasonal rainfall of 750 to 1000
mm.

In coastal areas north of Perth, Late Creta-
ceous sediments are lateritized but there are no
datings to indicate a minimum age. However,
the laterite is considerably dissected, and a
Tertiary age is likely. Farther north, in the
Carnarvon Basin, lateritization has affected rocks
of Late Eocene and older age but not Middle
Miocene and younger.

Thus there was a major period of lateritization
in the Oligocene and/or Miocene. Earlier
periods have been assumed by other workers, but
there is no geological evidence for them. Since
the Middle Miocene, lateritization has occurred
(presumably intermittently) only in coastal areas
of the southwest that have a moderate and
strongly seasonal rainfall.

Crustal movements in the Cainozoic

Whereas the Eocene sediments along the south
coast of Western Australia accumulated in very
shallow water and are now only a few metres
above their level of deposition, the old beach
levels associated with this cycle of sedimentation
are now at about 300 m at Mt Ragged (Lowry,
1970) and the Stirling Range (H. Schumann,
pers. comm.). Also, marine sediments of Eocene
age now at 300 m at Lake Cowan near Norse-
man, and in the Kennedy Range in the Carnar-
von Basin, near the northern margin of the
Yilgarn Block. Thus it appears that the whole
shield area (apart from the present coastal
margins) has been uplifted by approximately
300 m since the Eocene, but the relative im-
portance of marginal warping and eustatic sea
level changes cannot be determined.

Perhaps the most perplexing structural unit
near the south coast of Western Australia is
the Stirling Range. This feature, which lies just
south of the southern margin of the Archaean
Shield, thrusts metasediments of the Albany-
Esperance Block to a height of 1200 m above
sea level (about 1000 m above the level of the
flat shield to the north). The abruptness and
straightness of the north front of the range
suggest that this is a very young feature, but
the presence of the Late Eocene marine bench
at 300 m and the concordance of this with other
Eocene levels indicates that the Stirling Range
was in existence prior to the Late Eocene.
Perhaps its elevation was a final adjustment of
the contental margin before block faulting
ceased with Eocene drift.

Acknowledgements . — The authors gratefully acknow-
ledge the assistance of West Australian Petroleum Pty.
Ltd., In providing technical facilities for the prepara-
tion of this paper, and, In particular, Mr. H. D. Jones
for the drafting.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

14

References

Arkell, W. J. and P. E. Playford (1954). — The Bajocian
ammonites of Western Australia. Philosophi-
cal Transactions of the Royal Society of
London series B 237 : 547-604.

Arriens, P. A. (1971). — The Archaean geochronology of
Australia. Special Publications of the Geolo-
gical Society of Australia No. 3: 11-23.

Balme, B. E. and D. M. Churchill (1959). — Tertiary sedi-
ments at Coolgardie. Western Australia.
Journal of the Royal Society of Western
Australia 32: 37-43.

Bandy, O. L. (1964). — Cenozoic planktonic foraminiferal
zonation. Micropaleontology 10. 1-17.

(1967). — Cretaceous planktonic foraminiferal

zonation, ibid. 13: 1-31.

Bettenay, E. and M. J. Mulcahy (1972). — Soil and land-
scape studies in Western Australia. (2)
Valley form and surface features of the
South-West Drainage Division. Journal of
the Geological Society of Australia 18: 359-
369.

Burckle, L, H., T. Saito and M. Ewing (1967).— A Cre-
taceous (Turonian) core from the Natural-
iste Plateau southeast Indian Ocean. Deep
Sea Research 14: 421-426.

Campbell, W. D. (1910). — The Irwin River coalfield and
the adjacent districts from Arrino to North-
ampton. Bulletin of the Geological Survey
of Western Australia No. 38.

Campbell, I. H., G. J. H. McCall, and D. S. Tyrwhitt
(1970). — The Jimberlana Norite. Western
Australia — a smaller analogue of the Great
Dyke of Rhodesia. Geological Magazine 107
( 1 ): 1 - 12 .

Cockbain, A. E. (1967). — Astcrocyclina from the Plan-
tagenet Beds near Esperance, W.A. Austra-
lian Journal of Science 30: 68.

(1968). — The stratigraphy of the Plan-

tagenet Group. Western Australia. Annual
Report of the Geological Survey of Western
Australia for 1967: 61. 62.

(1969). — Dasycladacean algae from the

Werillup Formation, Esperance. ibid. 1968;
52. 53.

Compston, W. and P. A. Arriens (1968). — The Pre-
cambrian geochronology of Australia. Cana-
dian Journal of Earth Sciences 5: 561-583.

Darragh, T. A. and G. W. Kendrick ( 1971 ) .—Zenatiopsis
ultima sp. nov., terminal species of the
Zenatiopsis lineage (Bivalvia: Mactridae).

Proceedings of the Royal Society of Victoria
84: 87-92.

Falvey, D. A. (1972). — Sea floor spreading in the Wharton
Basin (Northeast Indian Ocean) and breakup
of Eastern Gondwanaland. The APEA Jour-
nal. 12 (2): 86-88.

Griffiths, J. R. (1971). — Continental margin tectonics
and the evolution of South East Australia.
ibid. 11 (1) : 75-79.

Hosemann, P. (1971). — The stratigraphy of the Basal
Triassic Sandstone. North Perth Basin,
Western Australia, ibid 11 (1): 59-63.

Ingram, B. S. ( 1968).— Stratigraphical palynology of
Cretaceous rocks from bores in the Eucla
Basin. Western Australia. Annual Report of
the Geological Survey of Western Australia
for 1967: 64-67.

Johnstone, M. H. and S. P. Willmott (1966). — The strati-
graphy of the Permian of the northern
Perth Basin, Western Australia. The APEA
Journal (1966): 100-104.

Jones, D. K., and G. R. Pearson (1972). — The tectonic
elements of the Perth Basin, ibid. 12 (1):
17-22.

Jutson, J. T. (1934). — The Physiography (Geomorpho-
logy) of Western Australia. Bulletin of the
Geological Survey of Western Australia No.
95.

Kendrick, G. W. (1960). — The fossil Mollusca of the
Peppermint Grove Limestone. Swan River
district of Western Australia. West Austra-
lian Naturalist 7: 53-66.

Konecki, M. C., J. M. Dickins. and T. Quinlan (1958). —
The geology of the coastal area between the
lower Gascoyne and Murchison Rivers. West-
ern Australia. Report of the Bureau of Mi 71-
era/ Resources. Geology and Geophysics. Aus-
tralia No. 37.

Lowry, D. C. (1970). — Geology of the Western Austra-
lian part of the Eucla Basin. Bulletin of
the Geological Survey of Western Australia
No. 122 (Published 1972).

M. J. Jackson. W. J. E. van de Graaff and

P. J. Kennewell ( 1972) .—Preliminary results
of the geological mapping in the Officer
Basin. Western Australia. 1971. Annual Re-
port of the Geological Survey of Western
Australia for 1971: 50-56.

McArthur. W. M.. and E. Bettenay (I960).— The devel-
opment and distribution of the soils of the
Swan Coastal Plain. Western Australia.
Australian Commonwealth Scientific and
Industrial Research Organisation Soil Pub-
lication No. 16.

McCall. G. J. H., and J. Leishman (1971). — Clues to the
origin of Archaean eugeosynclinal perido-
tites and the nature of serpentinisation.
Special Publications of the Geological Society
of Australia No. 3: 281-299.

McWhae. J. R. H.. P. E. Playford. A. W. Lindner. B.

F. Glenister. and B. E. Balme (1958). — The
stratigraphy of Western Australia. Journal
of the Geological Society of Australia 4:
1-161.

Mulcahy. M. J.. and E. Bettenay (1972).— Soil and
landscape studies in Western Australia, ibid.
18: 349-357.

Oliver. R. L. (1972). — Some aspects of Antarctic-Aus-
tralian geological relationships. Proceedings
of the SCAR/IUGS symposium on Antarctic
Geology and solid earth geophysics. Oslo. 1970.

Prider, R. T. (1944). — The geology and petrology of
part of the Toodyay District. Journal arid
Proceedings of the Royal Society of Western
Australia 28: 83-137.

Quilty. P. G. (1969). — Upper Eocene planktonic Fora-
miniferida from Albany. Western Australia.
Journal of the Royal Society of Western
Australia 52: 41-58.

press. — Cainozoic stratigraphy of the
Perth Area. ibid.

Seddon, G. (1972). — “Sense of Place" (Universitv of
Western Australia Press: Perth).

Skwarko, S. K. (1967). — Mesozoic Mollusca from Aus-
tralia and New Guinea. Bulletin of the
Bureau of Mineral Resources. Geology and
Geophysics. Australia No. 75.

Smith. R.. and P. Kamerling ( 1969).— Geological frame-
work of the Great Australian Bight. The
APEA Journal 9: 60-66.

Teichert, C. (1967). — Age of Coastal Limestone, Western
Australia. Australian Journal of Science 30:
68. 69.

Weissel. J. K.. and D. E. Hayes ( 1972) .—Magnetic
anomalies in the Southeast Indian Ocean.
American Geophysical Union Antarctic Re-
search Series. 19: 165-196.

Wellman. P.. M. W. McElhinny, and I. McDougall
(1969). —On the polar-wander path for Aus-
tralia during the Cenozoic. Geophysical Jour-
nal Royal Astronomical Society. 18: 371-395.

White. A. J. R., P. Jakes, and D. M. Christie ( 1971 ) .

Composition of greenstones and the hypo-
thesis of sea-floor spreading in the Archaean.
Special Publications of Geological Society
of Australia. No. 3: 47-56.

Windley, B. F.. and D. Bridgwater (1971).— The evolu-
tion of Archaean low- and high-grade ter-
rains. ibid. No. 3: 33-46.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July. 1973.

15

2. — Landforms and soils of southwestern Australia

By M. J. Mulcahy 1

Abstract

The landforms of southwestern Australia are
outlined as a basis for a description of the dis-
tribution of the extremely weathered materials
and superficial deposits on which the soils are
formed. While soils on younger landscape ele-
ments change predictably with climatic change,
a major factor controlling soil distribution is the
degree of erosional modification of the leached
and deeply weathered profiles associated with the
older landforms which are extensively preserved,
both in higher rainfall areas and the arid
interior. The significance of the soil patterns
and their characteristics are briefly discussed.

Introduction

The area dealt with in this paper lies to the
south of the latitude of Geraldton, and extends
from the west coast eastwards to include Kal-
goorlie and Esperance. It forms part of the
Great Plateau of Western Australia (Jutson,
1914) consisting of a stable Archaean shield,
characterized by low relief, extending over the
Mesozoic rocks of the Perth basin in the north
west, all flanked by narrow coastal plains of
younger sediments.

It is part of the landscape described by Hills
(1961) an “ageless and undatable old land”.
Woolnough (1927) associated the widespread
deep weathering with peneplanation, and
Prescott (1931) pointed out the occurrence of
leached, acid, lateritic soil materials extending
from the humid coastal areas to the now arid
interior.

Early soil maps such as that of Prescott (loc.
cit .) and later Stephens (1961) influenced by
the Russian school of pedologists, emphasized
climatic zonation, and south-western Australia
tended to be shown with soil boundaries parallel
to the isohyets. Teakle’s regional soil classifica-
tion, published in 1938, and subject to the same
influence, left out of account the “azonal”
lateritic soils, dealing mainly with the more
fertile soils of the younger landscape elements.
This was good sense at that time since the tech-
nology necessary for agricultural development of
the lateritic soils, including the use of minor
elements, was only just then becoming available,
and the use of soils in the higher rainfall areas
as a bauxite ore was still some years away. A
later compilation in the form of the Atlas of
Australian Soils (Northcote et al., 1967), based
on much more detailed information, indicates
a strong relationship between soil distribution
and drainage pattern as well as climate, while
Mulcahy et al. (1972) have shown that geo-
logical structure and the extent of drainage
rejuvenation are also important.

'Division of Land Resources Management, CSIRO,
Wembley, Western Australia.

Stephens (1946), in his classical paper on
pedogenesis following the dissection of lateritic
regions by downcutting streams rejuvenated
after uplift, suggests a fairly simple picture of
lateritic materials preserved on peneplain
remnants and removed from the slopes below.
Play ford (1954), on the other hand, pointing
out that lateritic materials are frequently found
at many levels in the one landscape, and assum-
ing laterite formation during some single period
in the past, concluded that this must therefore
have been postuplift.

Subsequent investigations in soil-landform
relationships, on which this account is based,
show that the situation is probably rather more
complex than the early workers believed. It
appears that laterite profiles, including both
surface ferruginous horizons and underlying
pallid zones (Walther, 1915), are to be found
extensively in a wide range of topographic situa-
tions, ranging from piedmont deposits of the
Swan Coastal Plain (McArthur and Bettenay,
1960), and older alluvial terraces in well-incised
valleys such as that of the Avon River (Mulcahy
and Hingston, 1961), to the most extensive
divides of the Great Plateau. While the laterites
of some of the younger landforms, particularly
those of the Swan Coastal Plain, may be re-
garded as forming today in the sense of meeting
the classical environmental requirements for
laterite formation postulated by Prescott and
Pendleton (1952), those of the older landforms
clearly cannot, being either in unsuitable topo-
graphic situations or in semi-arid and arid
climatic conditions, or both. Further, in all but
highest rainfall areas the extremely leached and
weathered pallid zones now contain an appre-
ciable store of soluble salts (Dimmock et al. in
prep.) so that leaching conditions are clearly
less effective than in the past.

Physiography

Figure 1 shows some of the main physiographic
features of south western Australia, to which the
soil pattern may be related.

The Darling Scarp marks the western margin
of the shield and Great Plateau, beyond which
the Precambrian rocks are buried by a consider-
able thickness of sediments of the Perth Basin
and Swan Coastal Plain. To the south the
shield slopes gently into the Southern Ocean,
with a discontinuous, thin veneer of Tertiary and
Recent sediments. The section (Fig. 1, ABC)
shows clearly the relatively high relief and
elevation of the Darling Range, regarded by
King (1962) as a marginal upwarping of the
shield rocks. It is separated from the gradually
rising plateau levels of the interior by a belt of

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

16

lower country showing partial coincidence with
an important zone of seismic activity, the “Yan-
danooka/Cape Riche Lineament” (Everingham,
1968).

A little further inland is another important,
but not so obvious physiographic feature, the
Meckering Line (Mulcahy, 1967), which marks
a striking change in the drainage pattern.
Inland of it the drainage is sparse, open and
sluggish, with chains of salt lakes in the main
trunk valleys. A large proportion of the west-
ward-flowing lake chains joins the headwaters of
the Swan-Avon system, off the mouth of which is
a prominent submarine canyon on the continen-
tal shelf (Von der Borch, 1968). Most of the
remainder reaches the Moore and the Blackwood
Rivers (Bettenay and Mulcahy, 1972). A major
continental divide separates these systems from a
generally eastward trending one draining to the
Nullarbor Plain and Great Victoria Desert in the
interior. In most years under the present cli-
matic regime the system as a whole does not
flow, acting as a sump in which the salts accum-
ulate in the playa lakes. In exceptionally wet
years, however, it becomes functional, and flow-
ing water flushes out the accumulated salts into
the downstream drainage lines (Fig. 1).

Downstream of the Meckering Line in the
west, the drainage lines form a closer network,
have steeper gradients, are more sharply incised,
and thus form a much more effective drainage
system functional in all normal winters. Depth
of incision of the streams increases progressively,
with a change from shallow flat floored valleys
to deep V-shaped valleys where they issue from
the Darling Scarp.

Thus the zone marginal to the Meckering Line,
particularly in the Darling Range, is one of con-
siderable relief relative to that inland of it.
Nevertheless, many of the more extensive divides
in the marginal zone are upland areas of low
relief, with features such as lakes and swamps
associated with sluggish drainage lines compar-
able with those inland. Their broad, flat-floored
valleys characteristically have grey sandy valley
fills, and some are known to have sedimentary
sequences of considerable thickness, such
as the Kirup Conglomerate (Hobson and Mathe-
son, 1949), or the Permian sediments of the
Collie Basin (Lord, 1952). While it is tempting
to regard these features collectively as dismem-
bered remnants of the old drainage systems now
preserved only in the interior, such an interpre-
tation is undoubtedly too simple. It could not
account, for example, for the Kirup Conglomer-
ate which, with its smoothly rounded cobbles in
a finer sandy and clayey matrix, is obviously a
deposit resulting from a high energy means of
transport. A conclusion must await further in-
vestigations in a field of enquiry as yet virtually
untouched.

Laterites, superficial deposits and soils

Distribution of Icitcritic materials
Figure 2 illustrates the broad regional distri-
bution of lateritic and other soil materials in
south western Australia, based largely on North-
cote ct al. ( 1967 > .

The map distinguishes between the low level
laterites of the Swan Coastal Plain, the “iron-
stone gravels” or bauxitic laterites of the Darling
Range and adjacent high rainfall areas to the
south, and the extensive “sand and gravel plains”
inland and to the north. All may be classed as
laterites in that they have surface horizons of
accumulation of iron oxides, frequently, but not
invariably overlying deeply weathered kaolinized
country rock, the pallid zone (Waltlier, 1915).
The Darling Range laterites, together with
the sandplains, are associated with Jutson’s
(1914) Old Plateau and Woolnough’s (1918)
Darling Peneplain, but the Section ABC of
Fig. 2 shows the considerable relief of the
lateritic surfaces of the Darling Range —
only inland of the Meckering Line does
the low relief traditionally postulated for the
lateritized peneplain become apparent. In both
the Darling Range and the sandplain areas the
lateritic pallid zones are most widespread and
deepest. They may be up to 30 metres in
thickness, beneath the floors of the trunk valleys
of the inland drainage systems, while they are
much shallower beneath the sandplains of the
divides (Bettenay et al.. 1964). Downstream of
the Meckering Line deepest weathering is prob-
ably associated with the older valley forms of
the low relief upland divides.

On the Swan Coastal Plain, west of the
Darling Scarp, deep pallid zones are common on
the older piedmonts, but associated ferruginous
duricrusts are less frequent, except near the foot
of the scarp on slightly elevated spurs corres-
ponding with the Ridge Hill Shelf described by
Prider (1948).

It is evident (Fig. 2) that areas free or
almost entirely free of lateritic materials are
restricted, being confined to the deeply en-
trenched valleys of the downstream sections
of streams as they issue from the Darling Range,
or further inland, to the entrenchment of
streams associated with the Yandanooka/Cape
Riche Lineament. High points above the general
level, such as the Stirling Range, and a number
of others too small to show because of scale, are
also almost completely free of laterite.

The principal remaining area of the map
(Fig. 2) shown as “Dissected Laterites” is one
of considerable complexity in terms of land-
forms and soils. The small divides and valley side
spurs tend to be capped with lateritic materials,
and bounded by prominent, erosionally-active
scarps or breakaways, the long pediment slopes
below them being underlain by pallid zones of
the truncated laterites. These features are best
developed immediately west of the Meckering
Line, where available relief is greatest, while
rainfall is still low. less than 500 mm per an-
num. Towards the coast, and in higher rain-
falls, these typically arid-zone landforms are
replaced by more gently inflected concavo-con-
vex slopes associated with less active erosion.
Only adjacent to sharply incised streams and
on the steeper valley sides are fresh rock mate-
rials locally exposed.

( 2)— 1 7390

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2. July. 1973.

17

Superficial deposits

Many of the lateritic materials mapped in
Fig. 2 are detrital. Those of the high relief
areas of the Darling Range often consist of
fragments of ironstone, and sometimes fresh
rock, recemented with skins of iron oxide (Mul-
cahy, 1961). These materials may extend down-
slope as colluvial sheets overlying a variety of
substrates, including relatively fresh country
rock (Mulcahy et al., 1972).

The sandplain materials too have been shown
to be colluvial deposits of local origin derived
largely from the ferruginous duricrust of pre-
existing laterites (Mulcahy, 1961; Brewer and
Bettenay, 1973). Further, they are multiple
deposits, not all of the same age; some, the most
extensive, carrying undifferentiated sandy yellow
earth soils, are clearly very young, while others
are older, and have developed segregations of
iron oxide which harden on exposure. Both
forms, the undifferentiated younger deposits
overlying the older, may be found capping classi-
cal lateritic residuals bounded by breakaways
(Mulcahy, 1964). It follows that the surfaces of
such residuals cannot be uncritically accepted as
any kind of time marker or datum.

Colluvial and alluvial deposits derived from
fresh rock outcrop are important though limited
in extent. They are confined principally to the
valley sides and floors, and to limited areas on
the Swan Coastal Plain, where they give rise to
the more fertile soils in comparison with those
derived from lateritic materials. Fresh colluvial
deposits are also occasionally found on upland
divides where rock outcrops are exposed by the
stripping of the pre-existing laterites to form the
sandplain deposits downslope.

Aeolian activity and dune formation are not
common on the sandplain surfaces, except per-
haps in the arid interior, in the east of the
area considered. There is some aeolian acti-
vity associated with the drainage lines. Bet-
tenay (1962) has described the formation of
“lake parna”, a silty, calcareous and saline
deposit blown out of the salt lake chains, which
may account for the extensive calcareous soils
shown in Fig. 2. They may, on the other

hand, be due in part to the more extensive oc-
currence of basic and ultrabasic rocks in the
eastern goldfields. Aeolian sands blown out

of stream channels have been reported by
Northcote et al. (1967) from the higher rain-
fall areas. Windblown coastal beach dunes
are almost continuous, particularly on the west
coast, and show age sequences related to Pleisto-
cene sea levels (Fairbridge, 1954; McArthur and
Bettenay, 1960). Aeolianite limestone dune and
beach rocks, the result of leaching of the origi-
nal calcareous sands and consolidation of the
lower layers by deposition of carbonates, are
prominent in the coastline, offshore reefs and
coastal islands (Fairbridge, 1953).

Soils

Only a brief outline of the nature of the soils,
related to the foregoing account of the geo-
morphology and superficial deposits, can be

given here. A considerable amount of further
information within this framework may be
extracted from the published Atlas of Australian
Soils (Northcote et al., 1967), with careful study
of the map legend and the accompanying
Memoir. Reference to areas of detailed study
on which this account is based is also given
where relevant.

Relatively little has been published on the
Darling Range soils, apart from a broad scale
study by Mulcahy et al. (1972). Smith (1951a,
1951b) has published accounts of the rather
similar patterns in the Donnybrook Sunkland
and towards the south coast. In the latter of
these, on the Frankland-Gordon river valley, he
drew attention to the sequential change of valley
form from the broad valleys of the inland areas
to the more sharply incised forms downstream.
The dominant soils are, of course, the lateritic
gravels (KS-Uc4.1 and KS-Uc4.2)*, consisting
of up to 5 metres or more of ironstone gravels
in a yellow sandy matrix and the related lateritic
podsolics (Dy3.61) with ironstone gravels in a
sandy surface overlying a mottled yellow-brown
clay subsoil. These materials frequently overlie
a pallid zone up to 30 metres or more in thick-
ness. Massive ironstone pavements are common
on ridgetops and occasionally on slopes. It is
worth emphasizing that, apart from more exten-
sive divides, there is some considerable relief
(Fig. 2), with slopes up to 8 U . In general, the
gravels tend to become finer downslope, some-
times grading into sandy yellow earths (Gn2.21)
in the lowest positions. The mid-slope gravels
are those currently being mined as bauxite.
The broader valleys of the more extensive
divides carry grey sands over ironstone gravels
(Uc2.3) or solonetzic profiles (Dg3.81), both
overlying deep pallid zones. Further downstream
the sides of the more incised valleys have a
range of soils including red and yellow podsolics
(Dr2.21 and Dy3.21) and red and yellow earths
(Gn2.14, and Gn2.21).

The complex soil pattern of the zone of
dissected laterites has been described by
Mulcahy and Hingston (1961), and Bettenay
and Hingston (1964). West of the Meckering
Line the truncated laterites and limited ex-
posures of fresher rock give rise to a range of
solodic and podsolic soils (Dy3.82 and Dy3.81)
and some red-brown earths with generally
neutral reaction trends (Dr2.22). Further in-
land, with longer, gentler slopes and lower rain-
fall, reaction trends of the red-brown earths
become alkaline (Dr.2.33).

The soils of the predominantly sandplain
areas and the associated broad valleys, forming
the larger part of the agricultural area of
Western Australia, have been described by
Bettenay and Hingston (loc. cit.) for an area
representative of the shield and by Churchward
(1970) where they are extensively developed
over the Jurassic sediments, often sandstones,

*Notation in parentheses after names of great soil
groups refers to the classification of Australian
soils by Northcote 1971.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2, July, 1973.

18

flanking the shield to the north. The sandplain
soils are predominantly sandy yellow earths
(Gn2.21) with some sands over ironstone gravels
(Uc5.22). The limited areas of fresher rock
outcrop on the valley sides, and the alluvial
deposits of the valley floors carry sodic brown
soils (Dr2.33) with calcareous subsoils. Both
upland and valley floor are, however, underlain
by lateritic pallid zones, deepest in the latter
situation, where the ground water is similar in
composition to sea water, though it may exceed
it in the concentration of salts. Thus the
relatively fresh materials of the soils of the
valley floors, which are calcareous and alkaline,
overlie acid and saline substrates. Bettenay
et al. (1964) calculate that 90% of the salts
stored in the landscape are in these valley
ground waters, and only a small proportion in
the rather more obvious playa lakes.

The aeolian lake parnas (Bettenay, 1962) give
rise to the silty, saline calcareous earths (Gel. 12
and Gel. 22) adjoining the salt lakes, usually
on the eastern or south eastern < downwind >
margins. The greater inland extent of these
soils (Fig. 2) may be due to the greater extent
of the lakes as a source area, or alternatively
to a greater relative abundance of fine textured
basic and ultrabasic rocks.

The Swan Coastal Plain provides two import-
ant age sequences of soils covering a period
extending as far back as the Mindel-Riss inter-
glacial (McArthur and Bettenay, I960) or per-
haps beyond it. One is developed on the
coalescing piedmonts near the foot of the scarp,
and the other on the wind blown sands of the
coastal areas.

The older, and most widespread, alluvial
deposits have been lateritized, but also have
been extensively stripped, so that the dominant
soil is a meadow podsolic (Dy5.81> consisting
of a sandy surface over a poorly structured
subsoil clay of low permeability developed in
the lateritic pallid zone. Younger deposits, in
the form of the terraces incised in these older
materials, or of alluvial fans laid over them,
carry a sequence of red and yellow podsolics
(Dr2.81 and Dy2.21) and of undifferentiated soils
on the relatively fresh youngest deposits. Thus
McArthur and Bettenay (loc. cit.) were able to
establish the age relationships of the soils by
the frequently observed and invariable order
of superposition of the deposits.

The youngest beach dune systems are at the
present coastline, consisting of a highly cal-
careous shell sand (Ucl.l). Inland they are
succeeded by slightly podsolized yellow sands
<Uc4.2), almost entirely quartz, but with some
local areas where heavy minerals are abundant.
The yellow sands are invariably underlain by
aeolianite rock which, however, extends beyond
them far to the west, underlying the present
day dunes at the coast and on the offshore
islands, so that the system must have been at
one time far more extensive. The oldest sys-
tem furthest inland has now lost its dune mor-
phology, and is of low relief compared with the
younger systems nearer to the coast. Its soils

are extremely leached and podsolized white
quartz sands with B horizons of iron and
organic matter accumulation (Uc2.3). They
are important intake areas for the coastal plain
aquifers and are underlain by fresh ground
waters at shallow depths.

Discussion and conclusion

The account of soils and landforms given here
is necessarily brief and incomplete, partly be-
cause a comprehensive treatment is impossible
in a short paper, and partly because of gaps in
our knowledge. Similarly, a discussion of the
significance of the data presented must also
be limited, since it would involve consideration
of a large number of aspects, ranging from
fundamental questions of soil and landscape
development to practical and applied problems
as diverse as minerals exploration, agricultural
technology, or land use planning. Some of the
more important are briefly considered below.

Landscape development

Preoccupation with questions of the age and
development of laterite is a common and under-
standable characteristic of discussions of this
topic in Western Australia. One or perhaps
two periods of formation in the past are as-
sumed <e.g., Prider, 1966), and thus the possi-
bility of using a lateritized surface as a time
marker is raised.

The evidence reviewed here shows that well-
developed, deep laterite profiles may be found
on surfaces as young as (early?) Pleistocene on
the Swan Coastal Plain, where the process may
still be continuing. They are most widespread
on the older landscape elements of the Old
Plateau, which Johnstone et al. (1973) consid-
ered to have been established in its present form
by the mid-Cretaceous. The older laterites,
except in the highest rainfall areas, now con-
tain appreciable quantities of soluble salts in
their pallid zones indicating that the necessary
leaching conditions for their development are
no longer operative. These facts, together with
the detrital and transported nature of many
lateritic materials, makes their use as a strati-
graphic marker suspect, though they are, of
course, likely to be preserved where erosional
forces are least effective, i.e., on gentler slopes
and in well vegetated higher rainfall areas.

Age of landscape is much more likely to be
indicated by a combination of landscape charac-
teristics rather than the occurrence of a single
characteristic such as laterite. These would
include low relief, widespread deep weathering,
ineffective drainage systems, and widespread re-
tention within the landscape of weathering pro-
ducts and sediments derived from them. These
are the conditions found most extensively in-
land of the Meckering Line.

Tectonics are, of course, an important factor
in landscape development. In this connection
the correspondence of entrenchment of the
drainage and greater relief with the Yanda-
nooka/Cape Riche lineament (Fig. 2) is note-
worthy. and is reflected in the soil pattern with

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

19

the exposure of fresher rocks as soil parent
materials and the truncation and dissection of
the lateritic profiles.

Ecological implications

Southwestern Australia is probably unique in
its great extent of deeply weathered and
leached soil materials, and consequent extremely
low levels of natural soil fertility. An account
of the natural vegetation adapted to these con-
ditions is given elsewhere in this volume. Isola-
tion by climatic barriers, particularly that of
aridity, is often invoked to account for some
of the unique features and speciation in the
Western Australian fauna and flora. An addi-
tional factor may be the formidable barrier
presented by the great extent of infertile sand-
plain soils on the major continental and regional
drainage divides shown in Fig. 2.

Minerals exploration

The great extent of deep weathering and of
superficial deposits, particularly the sandplains,
not directly related to the underlying geology,
obviously makes location of ore bodies difficult.
In many areas the nature of the surface mate-
rials is very poorly understood; for example, the
mallisols of the eastern goldfields, which may
be either aeolian deposits blown from the lakes,
or formed directly from ultrabasic rocks.

The old drainage lines of the interior and
possible remnants of them in marginal areas
are also of importance in this context, since
they drain catchments now deeply weathered,
including those in which ore bodies have been
discovered. Deep leads are well known in the
eastern goldfields (e.g., Campbell, 1906), and
have been described from the extreme south
west at Greenbushes, where some of the tin
mined is alluvial (Hobson and Matheson, 1949).
The study of these and other fragmented rem-
nants of old drainage and sedimentary systems
shown in Fig. 1 may lead to further discoveries
of economic value.

Agricultural development

Early pastoral and agricultural development
was confined to the more fertile non-lateritic
soils of the valley floors and flanking rock out-
crops, where, too, water supplies and natural
grazing for stock were more readily available.

With the identification of the nutrient defi-
ciencies, including the minor elements zinc,
copper, cobalt and molybdenum, which limit
crop and pasture establishment on the lateritic
soils, and with the growth of a fertilizer tech-
nology, agricultural development has been, until
recently, proceeding at an extremely rapid rate
on these soils. This has been possible, however,
only where rainfall is adequate, so that inland
of about the 250 mm isohyet, where sandplains
are extensive, neither farming nor grazing is
possible, due to the low nutritive value of the
native vegetation and the frequent occurrence
of poisonous legumes. Only further inland
again, in the vicinity of the eastward draining
salt lake systems, does pastoral activity become
possible.

Hydrology

The store of soluble salts in the widespread
lateritic pallid zones (Dimmock et al. in prep.,
Bettenay et al., 1964) presents a peculiar hydro-
logic problem. It is continually being replen-
ished by atmospheric accession in rainfall at
rates of the order of 200 kg/ha NaCl in high
rainfall areas of the Darling Range, falling
off rapidly inland (Teakle, 1937, Hingston, 1958).
Its retention within the landscape is apparently
a matter of an extremely delicate balance. This
is easily disturbed by clearing of the perennial
native vegetation with its deep rooting systems
and growth habit persisting into the dry sum-
mer season and its replacement by short lived
shallow rooting, winter growing, annual crops
and pastures. As a result evaporative losses of
water from surface soils decrease and ground
water recharge increases, with increased dis-
charge of water and salts in streams. Agri-
cultural development on sandplain areas in-
land of the Meckering Line has thus increased
the frequency with which the salts from the
lake chains are flushed into the headwaters
of the major rivers, so increasing the salt con-
tent of the latter. Where drainage systems are
ineffective the salts accumulate in the valleys,
so that agricultural development has led to
soil salinity in those situations (Lightfoot et al
1964). Peck and Hurle (in press) in a study
of water and salt balances of farmed and
forested areas show that catchments with aver-
age annual rainfalls less than 1000 mm/annum,
if extensively cleared, become unsuitable for
damming for water supply purposes due to the
salinity of the water. Their analysis also shows
that where the salt balance is so disturbed the
achievement of a new equilibrium at an accept-
ably low level of salinity may take times of the
order of tens, or more commonly, hundreds of
years.

Land use planning

Development of different forms of land use
in southwestern Australia has proceeded very
largely on an ad hoc basis. Gold, and later
other minerals, were mined where most readily
accessible, and when technological development
and growth of world demands made them trans-
portable and marketable.

Agricultural development has proceeded simi-
larly, where the current state of farming tech-
nology permitted, and in response to the fluctu-
ating demands of overseas markets. While
farming has built up soil fertility by importing
phosphate and through the use of legume nitro-
gen, it has been at the cost of considerable
effects on water supply and quality. Only
forestry, with its traditional attachment to the
policies of sustained yield, under management
by a single State authority, has developed on
a conservative basis. Even so, the native hard-
wood forest today compares badly with the few
isolated areas within it still untouched by the
timber fallers and sawmillers. But the extent
to which the forest has been preserved is the
extent to which the delicate hydrologic balance,

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

20

on which the water supplies of the south-west
depend, has been conserved.

As the population and industrial and mining
development increases in the south west of West-
ern Australia, with its relatively equable climate,
competing demands on the agricultural, forestry,
mineral, public amenity and water supply re-
sources especially in the Darling Range, will
increase. Of these, the last is most likely to
be limiting, so that its conservation is of highest
priority.

A continuing objective of agricultural and
forestry research should be the restoration of
the hydrologic balance through the strategic
deployment of land-use systems more efficient
in the use of water, since its misuse and wastage
through inability to utilize the abundant winter
rainfall is the root cause of water supply and
salinity problems.

Acknowledgement.— I am grateful to Mr. W. M.
McArthur for compilation of the figures, and for valu-
able criticism of the text.

References

Bettenay, E. (1962). — The salt lake systems and their
associated aeolian features in the semi-
arid regions of Western Australia. Journal
of Soil Science 13: 10-17.

and Hingston, F. J. ( 1964) .—Development

and distribution of soils in the Merredin
area, Western Australia. Australian Journal of
Soil Research 2: 173-186.

and Mulcahy, M. J. (1972).— Soil and land-
scape studies in Western Australia. (2)
Valley form and surface features of the
south west drainage division. Journal of the
Geological Society of Australia 18: 359-369.

, Blackmore, A. V. and Hingston, F. J.

(1964). — Aspects of the hydrologic cycle and
related salinity in the Belka Valley, Western
Australia. Australian Journal of Soil Re-
search 2: 187-210.

Brewer, R., and Bettenay, E. (1973). — Further evidence
concerning the origin of the Western Aus-
tralian sand plains. Journal of the Geologi-
cal Society of Australia 19 (4): 533-541.

Campbell, W. D. (1906). — The geology and mineral re-
sources of the Norseman District. Bulletin,
Geological Survey of Western Australia No.
21 .

Churchward, H. M. (1970). — Erosional modification of a
lateritized landscape over sedimentary rocks.
Its effect on soil distribution. Australian
Journal of Soil Research 8: 1-19.

Churchward, H. M. and Bettenay, E. (in prep.). — The
physiographic significance of conglomeratic
sediments and associated laterites in valleys
of the Darling Plateau, near Harvey, West-
ern Australia.

Dimmock, G. M., Mulcahy, M. J. and Bettenay, E. (in
prep.). — Salt content of lateritic profiles in
the Darling Range, Western Australia.

Everingham, I. B. (1968). — Seismicity of Western Aus-
tralia. Bulletin Bureau of Mineral Resources,
Geology and Geophysics, Australia No. 132.

Fairbridge, R. W. (1953). — “Australian Stratigraphy.”
(University of Western Australia Press:
Perth.)

(1954).— Quaternary eustatic data for West-
ern Australia and adjacent States. Proc. 2nd
Pan Indian Ocean Sci. Congr., Perth, p. 64-
84.

Hills, E. S. (1961). — Morphotectonics and the geomor-
phological sciences with special reference
to Australia. Quarterly Journal of the Geolo-
gical Society of London 117: 77-89.

Hingston. F. J. (1958). — The major ions in Western
Australian rainwaters. CSIRO Australia. Divi-
sion of Soils, Divisional Report No. 1/58.

Hobson, R A. and Matheson, R. S. (1949). — Greenbushes
Mineral Field. Bulletin, Geological Survey of
Western Australia No. 102.

Johnstone, M. H., Lowry, D. C. and Quilty, P. G.

(1973). — The Geology of South western Aus-
tralia — A Review (in press).

Jutson, J. T. (1914). — An outline of the physiographical
geology (physiography) of Western Austra-
lia. Bulletin, Geological Survey of Western
Australia No. 61.

King, L. C. (1962). — “The Morphology of the Earth”.
(Oliver and Boyd: Edinburgh.)

Lightfoot, L. C.. Smith, S. T. and Malcolm, C. V.

(1964). — Salt land survey, 1962. Journal of
Agriculture of Western Australia 5: 396-410.

Lord, J. H. (1952). — Collie mineral field. Bulletin, Geo-
logical Survey of Western Australia No. 105.

McArthur, W. M. (in prep.). — Soils and some aspects of
land use in the Manjimup-Pemberton area.

and Bettenay, E. (1960). — The development

and distribution of the soils of the Swan
Coastal Plain, Western Australia. CSIRO
Australia, Soil Publication No. 16.

Mulcahy, M. J. (1961). — Soil distribution in relation to
landscape development. Zeitschrift fur Geo-
morphologie 5: 211-225.

(1964). — Laterite residuals and sandplains.

Australian Journal of Science 27: 54-55.

(1967). — Landscapes, laterites and soils in

south western Australia: in Jennings, J. N.
and Mabbutt, J. A. (eds.), “Landform
Studies from Australia and New Guinea”,
p. 211-230. (Australian National University
Press: Canberra.)

and Bettenay, E. (1972). Soil and land-
scape studies in Western Australia (1) The
major drainage divisions. Journal of the
Geological Society of Australia 18: 349-357.

, Churchward, H. M. and Dimmock, G. M.

(1972). — Landforms and soils on an uplifted
peneplain in the Darling Range, Western
Australia. Australian Journal of Soil Re-
search 10: 1-14.

and Hingston, F. J. (1961). — The develop-
ment and distribution of the soils of the
York-Quairading area, Western Australia, in
relation to landscape evolution. CSIRO Aus-
tralia, Soil Publication No. 17.

Northcote, K. H. (1971). — “A Factual Key for the Recog-
nition of Australian Soils. 3rd Ed. (Rellim
Technical Publications: Glenside, South Aus-
tralia).

, Bettenay, E., Churchward, H. M. and

McArthur, W. M. (1967). — "Atlas of Austra-
lian Soils, Sheet 5, Perth-Albany-Esperance
Area. With Explanatory Data”. (CSIRO Aus-
tralia, Melbourne University Press: Mel-

bourne.)

Peck, A. J. and Hurle, D. H. (1973). — Chloride balance
of some farmed and forested catchments
in south western Australia. Water Resources
Research (in press).

Playford, P. E. (1954). — Observations on laterite in West-
ern Australia. Australian Journal of Science
17: 11-14.

Prescott, J. A. (1931). — The soils of Australia in rela-
tion to vegetation and climate. CSIR Aus-
tralia. Bulletin No. 52.

and Pendleton, R. L. (1952). — Laterite and

lateritic soils. Commonwealth Bureau of Soil
Science, Technical Communication No. 47.

Prider, R. T. (1948).— The geology of the Darling Scarp
at Ridge Hill. Journal of the Royal Society
of Western Australia 32: 105-129.

(1966). — The lateritized surface of Western

Australia. Australian Journal of Science 28'
443-451.

omu-n,

( iyoia;.— sons oi me Margaret and Lower
Blackwood Rivers, Western Australia. CSIR
Australia. Bulletin No. 262.

- (1951b).— Pedogenesis in the Frankland
River Valley, Western Australia. CSIR Aus-
tralia. Bulletin No. 265.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

21

Stephens, C. G. (1946). — Pedogene.is following the dis-
section of lateritic regions in southern
Australia. CSIR Australia, Bulletin No. 206.

(1961). — The soil landscapes of Australia.

CSIRO Australia Soil Publication No. 18.

Teakle, L. J. H. (1937). — The salt (sodium chloride) con-
tent of rainwater. Journal of Agriculture,
Western Australia 14: 115-123.

(1938). — Regional classification of the soils

of Western Australia. Journal of the Royal
Society of Western Australia 24: 123-195.

Von der Borch, C. C. (1968). — Southern Australian sub-
marine canyons: their distribution and

ages. Marine Geology 6; 267-279.

Walther, J. (1915). — Laterit in West Australein Zeitung
Deutsches Geologisches Gesellschaft 67B:
113-140.

Woolnough, W. G. ( 1918). — The Darling peneplain of West-
ern Australia. Journal and Proceedings of
the Royal Society of New South Wales 52:
385-396.

W. G. (1927).— Pres, address Pt. 1: The

chemical criteria of peneplanation; Pt. II:
The duricrust of Australia. Journal and Pro-
ceedings of the Royal Society of New South
Wales 61: 17-53.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

22

Figure 1 .

PHYSIOGRAPHY OF SOUTH WESTERN AUSTRALIA.

Maior divides including Continental Divide
(Mulcahy & Bettenay 1972)

Meckermg line (Mulcahy 1967)

Salt lake chains

Maior faults

Oivide features including grey sands, lakes
and swamps

DIVIDE FEATURES INCLUOE

1 Grey sandy valley floors with swamps (Mulcahy et al 1972)

2 Old valley fills indicated by cobble traces (Churchward and Bettenay m preparation)

3 Grey sands overlying Permian sediments (Lord 1952)

4 Kirup conglomerate (Hobson and Matheson 1949)

THE SOILS OF SOUTH WESTERN AUSTRALIA.

-30

-32

116

118

I

120

— I —
122

-34

116

I

118'

l

120 '

i

122 -
I

30 -

32-

34 -

Major divides, including Continental Divide
(Mulcahy & Bettenay 1972)

Meckering line (Mulcahy 1967)

Salt lake chains

Darling Range latentes

lateritic sandplains

Low level laterite of the Swan Coastal Plain

Oissected latentes

Soils on fresh rock and associated young deposits

Calcareous malhsols.

Coastal dune systems

Printed with permission of the
Royal Society of Western Australia.

AGRICULTURAL EDUCATION,
EDUCATION DEPARTMENT
OF WESTERN AUSTRALIA

65874/5/77 -2M

3. — Species diversity in the southwestern flora

By N. G. Marchant 1

Abstract

The South West Botanical Province is noted
for its floristic richness and high degree of
endemism.. Over 30% of the total species of
Epacridaceae , Myrtaceae and Proteaceae are
recorded in the Province. The percentage of
species restricted to the area, calculated from
available data, is 68%.

The flora is clearly related to that of eastern
Australia with only tenuous links with the flora
of South Africa and Malagasy.

The richness has arisen through the relative
isolation of the area, major climatic fluctua-
tions. and the diversification of the flora on
sandy and gravelly soils with a marked summer
drought.

Introduction

The South West Botanical Province of West-
ern Australia is the southwestern corner of the
State, where there is a reliable winter rainfall
and marked summer drought. The boundary
has been variously defined by Diels ( 1906 > and
Gardner (1942; 1956) as a line extending from
Shark Bay (latitude 26 S, longitude 114 W) in
the north to Israelite Bay (latitude 33 S. longi-
tude 124 > in the south. The crescentic area
west of this boundary is approximately two
hundred thousand square kilometres.

Nowhere is the boundary between the South
West Province and the Eremaean Province well-
defined. Gardner (1942) relied upon a 175 mm
winter (May to October') isohyet to substantiate
his field observations on the distribution of what
he considered South-Western and Eremaean
species. Diels ( 1906 > used the 300 mm annual
isohyet. a line slightly to the west of Gardner’s.
Burbidge (I960) also accepted this line, using
the 10” isohyet. These are presented in Figure 1.

Floristically the South West Province is de-
limited by the inland extent of many heath
components such as Conostylis . Anigozanthos
(Haemodoraceae) and the Restionaceae and
and Epacridaceae. The eastern limits of low
heaths on sand or gravel, mallee and mallee
heaths and certain woodlands also delimit the
South West Province. Further east are the
Mulga. Salmon Gum woodlands and shrublands
dominated by species of Acacia and species of
Chenopodiaceae and Myoporaceae.

Floristic diversity

Few studies have been made on the lower
plants of Western Australia. Bibby and Smith
(1954) list 36 lichens for the whole state and
an estimate of the numbers of Bryophytes can
only be made from a few localised studies.

'Western Australian Herbarium, Department of Agricul-
ture. South Perth.

Smith <1962) lists the number of Bryophytes
on granitic slopes in the well -watered Poron-
gurup Range as 14. Willis < 1 954 > has listed
some 33 mosses. 20 of which occur in the dry
Goldfields region.

The number of ferns and fern allies for West-
ern Australia is only 49 (Smith. 1966'. Of these.
26 occur in the South West Province, most
being associated with moist microhabitats on
granitic monadnocks. To this list must be added
five or six south-west species of Isoetes.

Only 11 Gymnosperms are recorded for the
province. This includes one widespread Macro-
zamia . a shrubby species of Podocarpus with
morphological affinities with the eastern species
P. spinulosus (Dallimore and Jackson. 1966' and
six species of Callitris as well as three species
of the endemic Actinostrobus.

Beard (1969) records at total of 3.611 Gymno-
sperms and Angiosperms in the South West
Province. This record was based on data, ex-
tracted from species folders in the W.A.
Herbarium, which were published as a ‘‘Descrip-
tive Calalogue of West Australian Plants”
(Beard. 1965 ». These data are limited because
so much of the Western Australian flora still
remains to be studied.

In comparison with most parts of the world
the recorded number of ferns. Gymnosperms and
Angiosperms in the South West Province is
relatively high, though a survey of recent
literature reveals other areas which have greater
numbers of species per square kilometre. These
data are presented in Table 1. arranged in de-
creasing species density. Species numbers have
been obtained from Adamson and Salter <1950>.
Allan (1961). Beadle et al. (1972). Black <1929>.
Burbidge and Gray <1970). Chippendale <1971'.
Curtis (1969) and Willis <1962).

Table 1

The numbers of plant species in the floras of different regions

Area
sq. km

Estimated
no. of
species

Species
per sq. km

Cape Peninsula, South Africa

471

2.622

5-567

Australian Capital Territory

2 359

1,037

0-440

Sydney District

28 500

2,000

0-070

Tasmania

68 330

1,200

0-018

South West Province

220 150

3,637

0-017

Victoria

227 620

2,500

0-011

New Zealand

268 670

1,457

0-005

South Australia

984 380

2,500

0-003

Northern Territory

1 347 525

2,736

0-002

The incredible diversity of the Cape Peninsula
flora is no doubt partly due to its diverse

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July. 1973.

23

Figure 1. — The South West Botanical Province as defined by Diels (1906) and Gardner (1956).

physical features, which modify the climate so
that various locations receive from 50 to over
200 cm annual rainfall. The higher elevations
also receive summer mists, which must alleviate
the summer drought to some extent. Of the
species total recorded by Adamson and Salter
(1950>, 846 are monocots. In comparison, Beard
(1965» lists less than 700 monocots for the South
West Province, a lower proportion than that of
the Cape Peninsula, due to the absence of many
bulbous plants in the former.

In comparison to the data available on the
number of species in other Australian states the
South West Province compares favourably. The
richness of the flora and its high degree of
endemic species have been commented on by

many authors. Burbidge (I960) in a detailed
analysis of Australian genera listed 462 genera
in 96 families in the South West province. Of this
total 111 genera are restricted to the Province.
Only five families are endemic. These are
Cephalotaceae, (1 sp.); Eremosynaceae, (1 sp.);
Emblingiaceae, (1 sp.); Ecdeiocoleaceae, (1 sp.);
Anarthriaceae, (5 spp.). The last two families
are morphologically and anatomically related to
the Restionaceae but are regarded as separate
families by Cutler and Airy Shaw (1965).

Among the families which are well represented
by a number of genera in the province are the
Restionaceae (16 genera), Liliaceae (21),
Xanthorrhoeaceae (9), Myrtaceae (26), Epacri-
daceae (15), Leguminosae (31), Rutaceae (13),

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2. July, 1973.

24

Goodeniaceae (10) and Asteraceae (30) (largely
from Burbidge, 1960).

At the species level, Gardner (1942) lists the
families well developed in the province as
Tremandraceae, Proteaceae, Leguminosae, Ruta-
ceae, Rhamnaceae and Verbenaceae.

An indication of the number of species in some
of the larger families of the province and the
percentage of the total species estimated for
Australia is presented in Table 2. Data have
been extracted from Burbidge (1963) and Beard
(1965).

Table 2

The numbers of species in selected families of plants from the
South West Province and the whole of Australia

Estimated

total

Australian

species

Estimated

S.W.

province

species

Percent, of
Australian
species
in SAY.

Centrolepidaceae

35

23

66

Dilleniaceae

107

63

59

Epacridaceae

499

170

34

Goodeniaceae

309

128

41

Myrtaceae

1,918

608

32

Proteaceae

1,211

408

34

ltestionaceae

82

57

70

Stylidiaceae

144

102

71

The three large families Epacridaceae, Myrta-
ceae and Proteaceae, which are regarded by
Good (1964) as large Australian families, have
between them over 3,600 species, over one-third
of the total number of species in the Australian
flora estimated by Good (loc. cit.). The South
West Province total of these three families is
just under 1,200 species, which is one third of
the total number of species of the families and
one third of the species total of all families re-
corded for the province.

In contrast to the richness in the south-west
of the families presented in Table 2 are the
distribution patterns of the other large Austra-
lian families listed by Good (loc. cit.). For
example in the Poaceae, Gardner (1952) records
only 82 species native to the Province in compari-
son to 278 native Western Australian grasses.
Figures for the Orchidaceae (George, 1971)
reveal the same degree of paucity in the South
West Province, only 128 species out of an Aust-
ralian total of approximately 600. The numbers
of species of Australia’s largest genera Acacia
and Eucalyptus show the same pattern. Only
161 Acacia species are recorded and 79 eucalypts
out of a total of very approximately 600 species
for each genus. The southwestern representa-
tives of the two genera respectively contribute
only 27% and 13% of the Australian total.

Endemism

Good (1964) comments on the difficulties in
comparing species richness and degrees of
endemism. No attempt will be made here to
present the facts using the methods outlined
by Exell and Williams (vide Good loc. cit.).

The high degree of endemism in the flora of
southwestern Australia has been commented on
by Hooker (1860), East (1912) and Gardner
(1942; 1959). The number of species restricted
to the South West Province has been estimated
at 75% of the southwestern flora (Gardner, 1959)
and as 86-87% by Beard G969). The latter
paper, based on the descriptive catalogue men-
tioned above (Beard, 1965), and a survey of the
literature, presents figures for the number of
both widespread and restricted species for the
South West Province, the Eremaean Province
and the Northern Province. The percentages of
endemism recorded in the analysis are not
clearly comprehensible, as there is some confu-
sion between the use of the terms “endemic” and
“solely southwestern” as well as errors in cal-
culation. On the basis of figures listed by Beard,
the number of species restricted to the South
West Province is 2,472 and the total number of
species in the province is 3,611. This means
that the percentage of species restricted to the
province is 68%.

In comparison with this percentage of
endemism Good (1964) suggests that the Cape
Peninsula, South Africa, has more than 90%
endemism. Malagasy is also quoted as perhaps
having 85% of the species being restricted to
the island. It must be remembered however
that the larger an area the higher the propor-
tion of endemic species. Australia for example
probably has a greater proportion of endemic
species than any other major geographical
region.

Figures for species endemism also exist for
Tasmania (Curtis 1969) as 17% and for Dar-
win and the Gulf region of the Northern Terri-
tory, 57% (Chippendale 1971), though this lat-
ter figure apparently refers only to the percent-
age of endemism of that region in comparision
to the Northern Territory and not the neigh-
boring States.

There is no doubt that the South West Prov-
ince does has a high degree of endemism in
comparison to other regions of comparable size.
It . is also worth noting that a considerable area
of the South West Province is covered by forests
and woodlands which are dominated by a single
or very few species of Eucalyptus. For example,
prime forests of Jarrah ( Eucalyptus marginata
Donn ex Sm.) cover over 17,000 sq. km yet har-
bour only a few hundred associated species. The
incredible richness and uniqueness of the local
flora is to be found on the heaths or so called
sandplains of sandy or lateritic soil which oc-
cupy considerable areas, particularly in the
northern and eastern parts of the province.
Diels (1906) and Gardner (1942) comment on
this richness at the “cusps” of the province.
Diels (loc. cit.) quotes the total number of re-
corded species and the percentage endemism in
each of the six botanical districts he designated
within the Province. Even though his total
figures are low and it is not clear if he uses the
term “endemism” to refer to the species restricted
to the district or whole Province, the results
are interesting. Endemism is quoted as 6% for
the extreme south-west corner, 37% for the most

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

25

northerly district and 33% for the most easterly.
The districts in between range from 20 to 33%.

An indication of the species richness of heath
areas of the province is given in Table 3, which
presents recent information on the species total
in three reserves.

Table 3

The numbers of species in selected National Parks in
southwestern Australia

Area

Xo. of

(sij. km.)

species

Tutanning (32' 30'S

1 1 7°E.)

18-2

400

Stirling Range (34 S.

118°E.)

1 093

550

Fitzgerald River (32 : '

30'S. 117°E.)

2 444

700

With more field work the figures for the
species of the Stirling Range and Fitzgerald
River reserve will rise. Already it has been
found that 10% of the plants of the Fitzgerald
River reserve are restricted to that area. Most
of these local endemics are restricted to the
slopes of the Barren and Eyre Ranges which
rise to over 500 m. It is worth noting from the
above data that the species density of Tutan-
ning, the most intensively studied reserve, is 22
species per square kilometre. This figure is
probably typical of many small areas of heath
in the south-west, particularly along the south-
ern coast on some monadnocks such as Mt.
Manypeaks and Peak Charles and some moun-
tain ranges such as the Stirling Range, the
Barrens and Mt. Ragged. A similar situation
exists in the Grampians, Victoria, with approxi-
mately 750 species of angiosperms and ferns, 30
of them restricted to that location in Victoria
and some of these having Western Australian
affinity (Willis, 1962).

Disjunctions and links with other floras.

The closest relationship of the South West
Province judging by the number of shared genera
and species is with southeastern Australia.
Beard (1969) records 491 species which occur
in the south-west and in eastern Australia. Of
these 211 are widespread in Western Australia
while 280 are restricted in Western Australia
to the South West Province and also occur in
the south-east of Australia.

Green (1964) mentions that several hundred
species occur in the south-west and in the
eastern states but not in the intervening regions
of Western Australia and western South Aus-
tralia. He selected 35 of these with a minimum
intervening distance in their distribution of
1200 km and a maximum of 2500 km. A recent
unverified collection of the species quoted with
the greater disjunction, (Stylidium perpusillum
Hook f.), reduces the intervening distance to
1600 km. In the same paper, Green lists 48
pairs of species, the members of which are re-
garded as morphologically related and far re-
moved geographically. Some of these are con-
sidered by Green to be true vicariads which

have been separated by unfavourable climatic
conditions in the intervening region and have
diverged morphologically.

The explanation for some of the disjunct
species listed by Green is regarded as long
distance wind dispersal. To support this sug-
gestion the author refers to the high proportion
of minute-seeded species such as Orchidaceae,
Stylidium and Levenhookia in his lists. Recent
work on the Orchidaceae at the Western Aus-
tralian Herbarium has revealed other species
which could be added to list of southwestern
and southeastern Australian species. One of
these, Thelymitra mathewsii Cheeseman, occurs
in a few localities at the northern tip of New
Zealand (George, 1971). This is only a few
degrees of latitude south of the Western Aus-
tralian occurrence, and approximately 5000 km
to the east.

Disjunctions between southwestern and south-
eastern Australian species with large propagules
require an explanation involving geological and
climatic history. Disjunctions within the South
West Province itself, particularly between the
northern and southeastern sections and between
the south-west and the granitic monadnocks of
the interior, can also be explained on the basis
of past climatic fluctuations. This subject will be
dealt with later.

A floristic relationship between southern and
east Africa and Western Australia is often
quoted in the literature on plant geography and
continental drift. This relationship has been
discussed by Gardner (1942), Specht (1958),
Burbidge (I960) and Good (1964). Specht loc.
cit) . lists four genera which range from East
Africa and Malagasy to Australia. These are
Adansonia, Diplopeltis, Keraudrenia and
Rulingia. These genera except Adansonia are
well represented in the South West Province.
The record for Diplopeltis from Malagasy has
recently been refuted (George and Erdtman,
1969) though Hoogland (1949) adds the genus
Hibbertia, which is very well represented in the
south-west, to those genera which occur in
Malagasy and Australia.

Gardner (1942) and Burbidge (1960) discuss
in detail the distribution of the families Centro-
lepidaceae, Restionaceae and Proteaceae and
their links between southwestern Australia,
southern Africa and southern South America.
These links are well developed only at the
family level. Weimarck (1941) listed only four
genera which occur in the Cape Peninsula and
in the South West Province. These are Restio,
Leptocarpus and Hypolaena ( Restionaceae ) and
Caesia (Liliaceae). As pointed out by Burbidge
(1960) there is no certainty that the respective
taxa in the two regions are congeneric. Certainly
very little is known at the present time of the
generic limits within the local Restionaceae.

The case of the distribution of the Proteaceae
is frequently regarded as a keystone in phyto-
geographical interpretation, rivalling the posi-
tion of Nothofagus. The study of the distribu-
tion of genera (Rao, 1971) reveals that none of
those represented in Western Australia occur

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

26

outside Australia and that there is a greater
relationship between Australia and South
America than between South Africa and
Australia.

The concept of a close relationship between
the flora of southern South Africa and the South
West Province has arisen largely because of
the incredible morphological similarity of some
genera, particularly in Proteaceae and Papi-
lionaceae, in the respective regions. This simi-
larity, as well as the parallel floristic diversity
and localised richness of the two areas, suggests
not recent contact but a parallel in environ-
mental history.

Thus the relationships of the southwestern
flora are clearly with that of eastern Australia
and, in common with that region, the north-
ern regions of Australia and the neighbouring
islands. Through the relationship of the flora
of eastern Australia to that of New Guinea,
New Caledonia, New Zealand and South
America, the western flora is more related to
these as well, than it is to that of southern
Africa and Malagasy.

Discussion

A considerable amount of controversy has
raged over the location of the origin of the
angiosperms. The most widely accepted view at
present is that they arose in the area which
now lies between Assam and Fiji, most prob-
ably in what is now south-east Asia <Takhta-
jan. 1969).

Hooker (I860') suggested that the South West
Province was a cradle of angiosperm develop-
ment but later workers such as Crocker and
Wood (1947) dispute this, pointing out that
the area is merely a centre of diversity. The
figures presented earlier showed this diversity
in some of the families represented in the
south-west. Cytological evidence on the origin
of the Myrtaceae and Epacridaceae (Smith-
White, 1959) points to an eastern Australian
origin for most of the genera within these
families. Rao (19711 also suggests an eastern
origin for the Proteaceae. Thus the South
West Province is only a centre of diversity in
a few families which are usually considered as
typical “Australian elements”. The origin of
the Australian flora then, lies either in the
eastern Australian region or, as is probably more
likely, in the area to the north of that region.

Smith- White (1959) believes that the angio-
sperms developed outside Australia and that the
orders and families were established before the
predicted massive invasion of Australia after
the mid-Cretaceous. Beadle (1966) also follows
this view.

It is significant to note that the suggested
period of the drifting apart of the pre-Creta-
ceous great southern continent was the Neoco-
mian, the same period as that of the earliest
fully-authenticated Angiosperm remains (Takh-
tajan, 1969). It is possible that the flowering
plants were well developed at the time of this
separation but it is doubtful if many present
day families had appeared. Thus the very

tenuous botanical relationship between the South
West Province and Africa could possibly be due
to continental drift. As well, if we accept that
the final breaking up of the southern continental
mass into Australia, Antarctica and South
America occurred at a much later date then
we would expect the greater degree of botani-
cal relationship between Australia and South
America, as has been found.

Burbidge <1960) discusses the affinities of
the floras of the southern continents, point-
ing out that the reconstruction of “Pangaea”
by Carey < 1958 > fails to explain the strong
generic affinities between South America and
Australia and the lack of them with South
Africa. More recent reconstructions of the
supposed southern continent by Griffiths (1972)
and the paper dealing with geological history in
this volume, show how it is possible for Aus-
tralia to have more affinities with South America
than South Africa. However it must be pointed
out again that Africa is supposed to have
separated from “Gondwanaland” before most
modern plant families had appeared.

Far more important considerations than con-
tinental drift and links with other floras are
those dealing with the entry of angiosperms into
Australia, and for the present study, a discus-
sion of the factors which have led to diversity
in the South West Province.

Knowledge of the Australian Angiosperm flora
in the early Tertiary is considerably better than
that of the earlier periods. The Eocene and
Miocene are presumed to have been periods of
adequate, reliable rainfall over southern Aus-
tralia. This enabled the development of a so-
called pan-Australian mesophytic flora which
included Cinnamomum , Nothofagus . Podocar-
vus . Casuarina, Banksia and perhaps Eucalyptus
(Crocker & Wood, 1947; Wood, 1959). In south-
western Australia sediments of Eocene origin
have revealed evidence of Araucaria, Banksia.
Nothofagus and Gleiclienia (McWhae et al.
1958). From this evidence it is presumed that
the climate of southern Australia was warm and
wet. There is no evidence of climatic conditions
in the mid-north of Western Australia, though
warm water marine faunas are recorded during
the Tertiary for the Carnarvon basin < Brown
et al., 1969). If the present climatic zones were
applicable in the Miocene then it is possible
that the area which is now North West Cape
and the Hamersley Range experienced a Medi-
terranean type climate.

The Miocene period has been noted as an
important one in Australian plant geography by
Wood (1959). It marked the end of a long
period of peneplanation. it was the beginning
of earth movements in the south-east of the
continent, and it was the beginning of the
partial isolation of the South West Province
from the rest of southern Australia.

In the Miocene to the Pliocene a marine trans-
gression deposited the limestones which now
constitute the Nullarbor Plain. This became an
edaphic barrier between the east and west
though, as will be suggested later, not a com-
pletely effective one.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

27

Smith-White ( vide Burbidge, 1960) suggests a
pre-Miocene origin for many genera in the
Rutaceae and Myrtaceae. It is likely that if
this is the case, the diversification and develop-
ment of endemism in the South West Province
commenced in this period. Perhaps the Eocene
transgression, which may have extended over
much of the southern part of the Western Shield
marked the beginning of the diversification. At
this time the higher mountains and ranges along
the south coast of Western Australia would have
been islands and as such would have acted as
refugia. It is in refugia such as these that much
evolutionary change could occur, especially
during the subsequent migration and inter-
mingling of these refugial floras on newly
liberated soils of the Oligocene and Miocene.

Another important factor contributing to the
development of the flora in the Miocene was
probably the development of laterite in the
humid climate. Lateritization of the whole
plateau commenced in the Eocene or Miocene
but subsequent aridity restricted formation to
the wetter south-west. The shield could have
been colonised by laterite-tolerant species from
the south or north-west to give a widespread
flora adapted to the post-Miocene drier condi-
tions. This flora was to a certain extent cut
off from the eastern parts of the continent.

Later weathering of the laterities on the
central part of the shield would have broken
the continuous flora: two large remnants of
these which exist today are the northern and
southern sandplains, which have numerous dis-
junct species, i.e., only occurring in these regions
or in isolated areas in between.

The scanty evidence of the post-Miocene floras
has been discussed by Burbidge (I960). It is
possible that the Cinnamomum type vegetation
moved northward in response to a cooling of the
climate leaving behind some of the present flora
which had adapted to the changed conditions.

Specht and Rayson (1957) suggest that the
present day sclerophyll genera developed in a
warmer, wetter climate. The growth of these
plants occurs at a time of great water stress in
the late summer. This is interpreted as being
out of phase with the present climatic condi-
tions and a result of their origin in a more
humid climate. It is possible that the leafy shoots
which initiate and differentiate small, rigid cells
in a period of stress would have an evolutionary
advantage under arid conditions.

A number of authors, including Wood (1959),
Smith-White (1954), Burbidge (1960) and Green
(1964), stress the importance of the Miocene
inundation and the subsequent development of
limestone and arid land to the north, as a major
barrier to plant migration between the east and
west. Recent surveys of the area to the north of
the Nullarbor Plains have revealed a surprising
number of genera and species which are normally
considered as southwestern. These include Acacia
leptopetala Benth., Daviesia ulicifolia Andr.,
Leptospermum roei Benth. and species of Micro-
myrtus, Thryptomene and Logania.

It is evident from the existence of pockets of
southwestern flora at North West Cape (lat.

22 S, long. 114 E) and Roe Plains, especially at
Twilight Cove (approx, lat. 32°S, long. 216°E)
as well as others of southwestern species on
granitic monadnocks as far east as Kalgoorlie,
that the present day vegetation of the South
West Province was of much greater extent in the
past. Evidence for climatic fluctuations in the
recent past (4000 to 3000 BC) have been pro-
vided by Churchill (1968). Under pluvial con-
ditions it can be imagined that the area north
of the Nullarbor Plain could provide an effective
corridor between the southwest and southeastern
Australia. The importance of this corridor for
birds has been presented by Ford (1971).

Crocker and Wood (1947) postulated a mid-
Recent period of aridity which, with subsequent
amelioration of the climate causing contraction
and expansion of the vegetation, had profound
effect on the ecological and species complexity.
During a period of severe southern Australian
aridity the South West Province would have
suffered greater decimation of the vegetation
than the southeastern Australian region. The
southeast in contrast to the west has a greater
latitudinal and altitudinal variation. In the
south-west three refugial areas could have been
important. One of these is the Hamersley Range
(approx, lat. 22 C S, long 118°W). The refugial
nature of the present day Hamersley Range flora
has been commented on by Burbidge (1959). If
the arid period in southern Australia postulated
by Crocker and Wood (l.c.) was due to a relative
southwards movement of the weather zones in
the mid-Recent as has been proposed by Keble
(1947), Gentilli (1961) and Specht (1958), then
it is likely that more of northern Australia came
under the influence of monsoonal weather pat-
terns. The Hamersley region could not only
have provided a refugium for some southwestern
species but also allowed some mixing of these
with species of northern origin.

Two areas of refugial nature are postulated in
the south-west. One is the deeper valleys of the
Darling Scarp which because of uplift of rain-
bearing westerly winds would always have a
higher rainfall than the neighbouring plateau
and coastal plain. From these refugia the
recolonization of the laterites to the east of the
scarp and the coastal plain could have taken
place.

The remaining refugia could have been pro-
vided by the monadnocks and ranges, particularly
along the south coast. These were probably the
most important areas because of the variety of
habitats available and the relative reliability of
rainfall. If the mean path of the high pressure
wind systems was much lower than at present,
the on-shore winds would have brought more or
less reliable rains to the south coast as they do
in summer today. Such refugia as the Stirling
Range and Barrens could have developed a
species diversity and degree of endemism similar
to that of the present Cape Peninsula.

The South West Province is not a centre of
origin bf the Australia flora but a centre of
great diversity of some of these elements and of
some relict groups. Continental drift occurred
too early to have a profound effect on relation-
ships with southern Africa. The diversity is

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2. July. 1973.

28

due to the long standing stability of the western
plateau with diversity beginning in the Miocene
and continuing with limited migration between
east and west and north and south. The whole
Southwestern corner of Australia developed into
a virtual island which underwent drastic
climatic change which must have favoured
species adapted to sandy and gravelly soils. The
South West Province can be regarded as a
relatively isolated area with a latitudinal spread
of only 9° and a limited altitudinal variation.
This “island” has always been well away from
any migration routes. Eastern Australia, in
contrast to the south-west, has a latitudinal
range of over 35° with a wide altitudinal
variation. According to Burbidge (1960) this
part of Australia is also a well developed
migration route parallel and related to other
routes from New Guinea to New Caledonia and
New Zealand.

References

Adamson, R. S. and Salter, T. M. (1950). — “ Flora of the
Cape Peninsula” . Juta, Cape Town.

Allan, H. H. (1961). — “ Flora of New Zealand”. Govern-
ment Printer, Wellington.

Beadle, N. C. W. (1966). — Soil phosphate and its role in
molding segments of the Australian flora
and vegetation, with special reference to
xeromorphy and sclerophylly. Ecology 47 (6):
992-1007.

Beadle, N. C. W., Evans, O. D. and Carolin, R. C.

(1972). — “ Flora of the Sydney Region”. Reed,
Sydney.

Beard, J. S. (1965). — “ Descriptive catalogue of West
Australian Plants”. Society for Growing
Australian Plants, Sydney.

Beard, J. S. (19o9). — Endemism in the Western Aus-
tralian flora at the species level. Journal of
the Royal Society of Western Australia 52
(1): 18-20.

Bibby, P. and Smith G. G. (1954). — A list of Lichens
of Western Australia. Journal of the Royal
Society of Western Australia 39: 28.

Black, J. M. (1929). — “ Flora of South Australia” . Gov-
ernment Printer, Adelaide.

Brown, D. A., Campbell, K. S. W. and Crook, K. A. W.

(1969). — “The Geological Evolution of Aus-
tralia and New Zealand”. Pergamon, Lon-
don.

Burbidge, N. T. (1959).— Notes on plants and plant habi-
tats observed in the Abydos-Woodstock area,
Pilbara district, Western Australia. C.S.I.R.O.
Division of Plant Industry Technical Paper
12 .

Burbidge, N. T. (I960).— The Phytogeography of the
Australian region. Australian Journal of
Botany 8 (2): 75-212.

Burbidge, N. T. (1963). — “Dictionary of Australian Plant
Genera”. Angus and Robertson, Sydney.

Burbidge, N. T. and Gray, M. ( 1970) .—“Flora of the
Australian Capital Territory.” A.N.U. Press,
Canberra.

Carey, S. W. (1958). — A tectonic approach to continental
drift. Geology Department Symposium, Uni-
versity of Tasmania, March 1956 pp. 117-355.

Chippendale, G. M. (1971).— Check list of Northern Ter-
ritory Plants, Proceedings of the Linnaean
Society of New South Wales 96 (4): 207-267.

Churchill, D. M. (1968).— The distribution and prehis-
tory of Eucalyptus diversicolor F. Muell., E.
marginata Donn ex Sm. and E. calophylla
R. Br. in relation to rainfall. Australian
Journal of Botany. 16: 125-151.

Crocker, R. L. and Wood, J. G. (1947). — Some historical
influences on the development of the South
Australian vegetation communities and their
bearing on concept and classification in
Ecology. Transactions of the Royal Society
of South Australia. 71: 91-136.

Curtis, W. M. (1969). — The vegetation of Tasmania. Tas-
manian Year Book 1969: 55-59.

Cutler. D. F. and Airy Shaw, H. K. ( 1965) .— Anarthria-
ceae and Ecdeiocoleaceae : two new Monco-
tyledonous families, separated from the
Restionaceae. Kew Bulletin 19: 489-499.

Dallimore, W. and Jackson, A. B. (1966).— "A Hand-
book of the Coniferae and Ginkgoaceae” .
Arnold, London.

Diels. L. ( 1906 ) . — “Die Vegetation der Erde VII Die
Pflanzenwelt von West Austr alien” . Engel-
mann, Leipzig.

East, J. J. (1912). — The flora of Western Australia in
'Cyclopedia of Western Australia” . Hussey
and Gillingham, Adelaide.

Ford, J. R. ( 1971).— Distribution and taxonomy of Birds
in the Great Victoria Desert. Emu 71 m-
27-36.

C. A. (1973 ). — “ Enumeratio Plantarum Austra-
liae Occidentalis” . Government Printer.
Perth.

C. A. (1942).— The vegetation of Western Aus-
tralia with particular reference to the
cnmate and soils. Journal of the Royal
Society of Western Australia. 28: 11-87.

C. A. ( 1952 ).— "Flora of Western Australia .
Vol 1, pt. 1 Gramineae”. Government
Printer, Perth.

C. A. (1956) in Gardner, C. A. and Bennetts,
H. W .-—"The Toxic Plants of Western Aus-
tralia”. West Australian Newspapers, Perth.
C. A. (1959). — The vegetation of Western Aus-
tralia. Biogeography and Ecology in Aus-
tralia. Monographiae Biologicae VIII, 274-282.
1961).— Quaternary climates of the Australian
region. Annals of the New York Academy
of Science 95: 465-501.

George, A. S. (1971). — A check list of the Orchidaceae
of Western Australia. Nuytsia 1 (2): 166-196.

George, A. S. and Erdtman, G. (1969). — A revision of
the genus Diplopeltis Endl. Grana Palyno-
logica 9: 92-109.

Good, R. (1964 ). — “The geography of the Flowering
Plants”. Longmans Green and Co., London.

Green, J. W. (1964). Discontinuous and presumed
vicarious plant species in southern Aus-
tralia. Journal of the Royal Society of West-
ern Australia 47 (1): 25-32.

Griffiths, J. R. ( 1972). — Australia and Gondwanaland. Aus-
tralian Natural History 17 (8): 249-253.

Hoogland, R. D. (1951). — Dilleniaceae, in Flora Mele-
siana Series 1 (4): 50.

Hooker, J. D. (1860 ). — “The Botany of the Antarctic
Voyage of H.M. Discovery Ships ‘Erebus’ and
‘Terror’. Introductory essay to the flora of
Tasmania” . Lovell Reeve, London.

Keble, R. A. (1947). — Notes on Australian Quaternary
climates and migrations. Memoirs of the
National Museum of Victoria 15: 28-81.

McWhae, J. R. H., Playford, P. E. .Lindner, A. W., Glenis-
ter, B. F. and Balme, B. E (1958).— The
Stratigraphy of Western Australia. Journal
of the Geological Society of Australia 4 -
1-155.

Rao, C. V. (1971).— “Proteaceae”, C.S.I.R., New Delhi.

Raven, P. H (1972) — An introduction to continental
drift. Australia Natural History 17 (8)-

245-248.

Smith, G. G. (1962). — The flora of granite rocks of the
Porongurup Range, South Western Aus-
tralia. Journal of the Royal Society of West-
ern Australia 45 (1): 18-23.

Smith, G. G. (1966). — A census of Pteridophyta of
Western Australia. Journal of the Royal
Society of Western Australia 49 (1)- 1-12

Smith-White, S. ( 1959).— Cytological evolution in the
Australian flora. Cold Springs Harbour Sym-
posia in Quantitative Biology 24: 273-89.

Specht, R. L. (1958 ). — “Records of American- Australian
scientific expedition to Arnhem Land”
Vol 3 No. 14.

Gardner,

Gardner,

Gardner,
Gardner,
Gardner,
Gentilli (

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

29

Specht. R. L. and Rayson, P. (1957).— Dark Island
Heath I. Definition of the ecosystem. Aus-
tralian Journal of Botany 5: 52-85.
Takhtajan. A. (1969). — "Flowering Plants Origin and
Dispersal" . English Edition, Oliver and Boyd,
Edinburgh.

Weimarck, H. < 1941 ).— Phytogeographic groups, centres
and intervals within the Cape flora. Kung-
lica Fysiographiska Sallskapets Handlingar

52.

Willis, J. ( 1954) .—Mosses new to W.A. Victorian Natur-
alist 71: 8-12.

Willis, J. (1962).— Land flora of Victoria. Victorian Year
Book 1962.

Wood, J. W. (1959).— The phytogeography of Australia
in relation to radiation of Eucalyptus,
Acacia, etc.. Biography and Ecology in Aus-
tralia. Monographiae Biologicae VIII- 291-
302.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

30

4. — Animal and plant speciation studies in Western Australia

By H. E. Paterson 1 and S. H. James-

Abstract

Speciation studies involving; Western Austra-
lian. and especially southwestern, animals and
plants are reviewed. The paper presents differ-
ent approaches to the problems of integrating
cytogenetics, ecology and taxonomy, and fall's
into two parts.

Current studies on animals in Western Aus-
tralia illustrate some of the consequences of
the evolutionary process of speciation. as well
as some aspects of the mechanism bringing it
about.

Cy toevolutionary responses in ten groups in
the Western Australian llora are reviewed.
Cytogenetic innovations in evidence include
polyploidy, apomixis. complex hybridity, ancu-
ploidy, lethal systems and B chromosomes, and
in general, these are interpreted as conserva-
tive devices, conserving either heterozygosity,
adapted gene combinations, or both.

These studies support Mayr’s arguments for
a discipline covering the speciation process and
its consequences.

Intrcduclimi

Speciation studies amongst Western Austra-
lian animals and plants are not extensive, but
they illustrate differing points of view in a
number of areas. Firstly, the application of a
species definition to taxonomic practices varies
between zoologists and botanists; for example,
the existence of infraspecific polyploidy in plant
species must often, perhaps invariably, combine
genetically distinct though morphologically
similar biological species into a single taxonomic
species. Comparable genetic barriers in animal
species would normally be taken to delimit
taxonomic species. Secondly, there is a con-
tinuing discussion concerning the importance of
postmating isolation and its reinforcement in
tlie speciation process. In this paper, the theory
of allopatric speciation involving an initial
attainment of premating isolation is vigorously,
and profitably, applied to animal speciation
studies. On the other hand, while recognizing
that the theory of allopatric speciation may well
be of great utility in interpreting the evolu-
tionary origin of many plant species, the plant
speciation studies reviewed here have been
selected, and in many cases the work actually
directed, with a view to seeking out situations
in which alternative processes of speciation may
be demonstrable. Thus, these latter studies do
do not concentrate on an integration of
genetically-determined ecological attributes with
geohistorical circumstances, but attempt to in-
vestigate situations where endophenotypic attri-
butes, the attributes of which postulating isola-
tion is made, are of adaptive utility at the
infraspecific level.

•Zoology Department, University of Western Australia.
Necilands, W.A.

-Botany Department. University of Western Australia.
Necilands. W.A.

Animal species studies

by H. E. Pater; on

The species concept

In 1957 Ernst Mayr proposed that a Science
of Species could be justified as a legitimate field
of study. It was conceived as including not
only the study of the process of speciation itself,
but the study of many of its sequelae as well.
Such a field of study would clearly be most
closely related to population genetics because it
is based on the genetical concept of the species,
which, in turn, has the idea of a species-specific
gene pool at its centre. The studies reported on
here are mostly still in progress. They are
diverse in approach and cover a wide range of
animal species, but they all fall within the
scope of Mayr s Science ol Species. Bv present-
ing them together it is hoped that ' they will
provide support for May r's suggestion by reveal-
ing something of the breadth, interest and even
utility ol such a discipline.

At the outset it is necessary to outline
clearly what is meant by the genetical
concept ol the species, because the species
in genetics does not always correspond
to ffny species currently recognised in taxonomy.
This usually means that insufficient time has
elapsed for the necessary adjustments to be
made. A gene pool can only exist in organisms
which reproduce sexually, though it could be
argued that it is applicable to certain prokar-
yotic organisms characterised by at least one
of Haldane’s "alternatives to sex" which would
enable recombination to occur (Haldane. 1955;
Sanderson. 1971'. In sexually reproducing
organisms all reproduct ively mature members
of a species can contribute to. and share in its
gene pool. Furthermore, the gene pool of this
species is isolated from the gene pools of other
species due to the functioning of genetically
determined isolating mechanisms. For these
reasons the gene pool of a species is able to be
distinct in genetical properties and structure
from those of all other species.

These ideas can be expressed in other ways.
For example, the essential characteristic of an
animal species is that it comprises a number of
individuals each of which is able to recognise
reproductively mature mating partners of the
same species. This is achieved by means of a
genetically -determined, co-adapted behavioural
signalling system. Members of a particular
species, because of this system, do not usually
respond effectively to the analagous signals of
members of other species under natural condi-
tions.

Journal of the Royal Society of Western Australia. Vol. 5(i Parts 1 and 2. July. 1973

31

The explosive diversification of forms of
organisms which apparently followed the evolu-
tion of eukaryotic cells in late Precambrian
times cannot be attributed to the newly evolved
methods of generating genetic diversity alone.
Recombination will generate variation, but will
not on its own lead to the radiation of forms
which we observe. Radiation will, however, be
initiated if recombination is restricted to occur-
ring within discrete gene pools. This raises the
key question how did the species as an evolu-
tionary phenomenon first arise? No general
answer to this question will be attempted be-
cause it is very likely that species arose on
several independent occasions following the in-
dependent acquisition of sexual reproduction by
several different eukaryotic Protista (Margulis,
1970). In the following discussion only the
animal lineage will be considered. Most animals
are dioecious or can be shown to have had
dioecious ancestors. This suggests that the
whole animal kingdom evolved from an early
dioecious protistan ancestor. Because we are
discussing the origin of the species as an evolu-
tionary phenomenon, such an ancestral form
must have lacked an isolating mechanism as
such. However, being dioecious it must have had
a genetically-determined mechanism which en-
abled the one mating type to recognise the
other. Credibly, this system could have provided
the basis for the first isolating mechanism
which brought the first two animal species into
existence, because the principles of allopatric
speciation long advocated by Mayr (1942; 1963)
are as applicable to this basic system as to more
advanced ones. In simple terms the mechanism
behind allopatric speciation may be as follows.
Once an extrinsic barrier has split the original
“gene pool” into two, the genetic structure of
each will begin to deviate in response to selec-
tion from its distinct environment. Some of the
gene substitutions brought about in this way
will have a pleiotropic effect on the mate selec-
tion mechanism. In this way the mechanisms
of the two populations will progressively diverge
to the point when the signals of members of the
one population are no longer effectively recog-
nised by members of the other. At this stage
speciation will have been achieved. An example
of a gene with pleiotropic effects of the sort
invoked is yellow in Drosophila melanogaster
Meigen. When compared with “wild type” flies
homozygotes for this allele are more resistant
to starvation (Kalmus, 1941) and the males have
a modified courtship pattern (Bastock 1956).
Following effective long term isolation, the mate
recognition systems of the two sub-populations
may, thus, also function as premating isolating
mechanisms, which effectively protect the in-
tegrity of their gene pools. This in turn enables
the two populations to continue to diverge
adaptively even though their ranges may come
to overlap.

Details of the speciation scheme outlined
above will be disputed by some, but if it happens
to be correct it will be noted to have the follow-
ing implications. Species come into existence as
by-products of adaptive evolution. They are

not products of selection for diversity, though
selection may decide whether a newly evolved
species will survive for long. Diversity is not
selected for as such, and it is ultimately depend-
ant on factors which favour speciation. How-
ever, selection does decide the specific pattern of
diversity which comes into existence in a particu-
lar environment at a particular time.

In Western Australia the work of A. R. Main
and his students on Leptodactylid frogs is the
pioneer work on animal species studies. This
work is still continuing and has recently been
reviewed (Main, 1970), which provides an indica-
tion of how young a field this is within this
State.

Studies on the origin of species

The term “speciation studies” is often used
loosely to mean almost any type of species study.
Here it is used in a stricter sense to mean the
study of the mechanism by which isolating
mechanisms, which protect the gene pool of a
species from introgression, come into existence.

Earlier work (Main, Lee and Littlejohn, 1958)
showed a rich frog fauna in southwestern Aust-
ralia, an area almost free of geographical barriers
of the kind which have proved important in
initiating speciation. These authors compared
the faunae of southwestern and southeastern
Australia, using as particular criteria male call
structure and genetic divergence as judged from
the results of in vitro crosses. They concluded
that during the Pleistocene three distinct pluvial
periods, separated by interpluvials, had allowed
the entry into Western Australia of faunal
elements from the east. The arid interpluvial
periods effectively led to the splitting of popula-
tions into western and eastern sub-populations,
thus providing the conditions for subspeciation
and speciation to occur. Main (1970) has modi-
fied details of the original model, suggesting that
the last Pleistocene pluvial, and one of the two
earlier connections, were only wet enough to
allow frogs from seasonally arid environments
to invade from the east. During the third
pluvial, however, both wetland and semi-arid
country frogs were able to penetrate westwards.
This modified model accounts for why there is
only one western species representing certain
eastern wetland (Bassian) species, while numbers
of eastern semi-arid (Eyrean) species each
have three western representative species. These
ideas have proved applicable to spiders of the
tribe Aganippini (Main, 1962), horseflies <Taba-
nidae) (Mackerras, 1962) and, if allowance is
made for the biological features of the group, to
birds (Serventy and Whittell, 1951 ).

The genus Calliphora (Diptera) has radiated
enormously within Australia, so much so that
most of the world’s species are endemic here.
These flies offer excellent opportunities for test-
ing Main’s ideas on speciation at a new level of
sophistication. Most species are easily cultured,
and pre- and postmating isolation between mem-
bers of western and corresponding eastern popu-
lations can be studied with considerable preci-
sion. The eastern and western populations can
also be compared with great precision at the
cytological level because these flies possess

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

32

excellent polytene chromosomes (Thomson, 1969).
Their large size make them very suitable for
biochemical studies using gel electrophoresis, for
example. Such a study is at present in progress.
N. Monzu is exploiting some of the techniques
mentioned to compare Western Australian
species with close relatives in eastern Australia
(Table 1).

Table 1

Western Australian species of the genus Calliphora with
closely related species in the eastern states of Australia

Western form

Eastern form

Calliphora nociva Hardy

Calliphora augur (F.)

Calliphora albifrontalis Malloch

('alliphora stygia (F.)

Calliphora varifrons Malloch

Calliphora hilli Patton

R. W. George* has in recent years paid atten-
tion to the rather neglected field of speciation
in the sea (George and Main, 1967; George,
1969). The organisms with which he has worked
are the rock lobsters, Palynuridae. By consider-
ing and taking into account the biological
properties of present day species of the genus
Jasus, for example, he was able to propose
credible models of speciation in the sea, which
do no violence to the well-accepted ideas of how
speciation occurs on land. In this case specia-
tion of a group of six contemporary species
which occur in the waters of the southern hemi-
sphere was explained in the following broad
terms: George accepts that their close similarity
implies a common ancestral species which
formerly had a circumpolar distribution. Because
global temperatures were formerly higher on the
average than now, the ancestral species was
thought to have occupied an area occupied by
the subantarctic zone today. With climatic
cooling this temperate zone must have moved
to lower latitudes, forcing the species north-
wards until it could colonize appropriate regions
of the southern continents and several islands
(New Zealand, Tristan da Cunha, St. Paul’s
Island and Juan Fernandez). In this way the
original single circumpolar species was effec-
tively fragmented into isolated populations, each
of which would then have been subjected to a
different set of selection pressures. The isola-
tion could have been maintained by specific
local conditions resulting from local currents
due to the configuration and position of the
island or continent with respect to the prevail-
ing major currents.

In a comparable way speciation in other
genera was accounted for. In general, a study
of the biological attributes of present day
species was considered to provide the key to
elucidating the pattern of speciation proposed
for each genus.

Earlier work by Littlejohn and Main and his
students on the isolating mechanisms operating
to maintain the integrity of the gene pools of
the frogs Crinia insignifera Moore and C. pscud-

* Western Australian Museum.

insignifera Main has been repeated in consider-
able detail by M. Bull. These two species meet
along the Darling Scarp. C. insignifera occupies
the coastal plain and C. pseudinsignifera extends
over large areas of the interior of southwestern
Australia. Bull has demonstrated clearly that
post-mating isolating mechanisms between these
species are absent as judged by the in vitro
cross method. Reproductive isolation between
them is therefore largely at the premating level
In transects across the rather sharp line of
contact between the species no indication of re-
inforcement of the premating mechanisms, such
as the male calling characteristics, has been
revealed. It seems that in this case the pre-
mating isolating mechanisms did not come into
existence through natural selection acting to
reduce gene flow between two partially diverged
populations as postulated by Wallace, Fisher and
Dobzliansky, among others.

The study of isolating mechanisms

It is generally accepted that postmating
isolating mechanisms arise as pleiotropic by-
products of adaptive evolution in isolated popu-
lations. This is because it is not easy to see how
they could be evolved directly by natural selec-
tion. Less agreement exists over the origin of
premating isolating mechanisms, because less
difficulty exists in proposing models which ac-
count for their evolution by selection acting
against inferior hybrids, as an alternative to
their production in the same manner as post-
mating mechanisms. Many cases of "reinforce-
ment” of premating isolating mechanisms have
been reported (Levins, 1970), but it should be
appreciated that the mere demonstration that
selection strengthens premating isolation can in
no way be regarded as compelling evidence in
support of the thesis that premating isolating
mechanisms evolve in this manner. This is be-
cause no satisfactory mechanism has been pro-
posed to explain how genes, selected at a para-
patric interface between two populations for
their property of reinforcing premating isolation,
can spread through the body of each popula-
tion outside the zone of contact. Because premat-
ing isolating mechanisms are properties of species
as a whole, models of speciation must account
for this fact. This is an old objection (Moore.
1957) but one which is still avoided by many
advocates of the hypothesis of the selective
origin of premating isolating mechanisms.

S. J. Miles, R. Irving-Bell and I are particu-
larly interested in reproductive isolation between
the four members of the Culex pipiens L. com-
plex which occur in Western Australia, C. fati-
gans Wd„ C. molestus Forsk&l. C. australicus
Dobr. & Drumm. and C. globocoxitus Dobr. here
treated as distinct biological species. C. australi-
cus has not yet been cultured successfully, but
the other three species are readily cultured and
can be crossed in the laboratory. Except in
the crosses between C. globocoxitus and C. fati-
gans. there is little evidence of postmating
isolation between them. In cages the premating
isolation is detectable but is by no means com-
plete. Yet in nature very little hybridization

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2. July, 1973.

33

has been detected, and then only between C.
molestus and C. fatigcms, which often share the
same breeding place. This is of interest because
C. molestus was introduced from Europe, C.
fatigans was introduced, possibly from the orien-
tal region where it almost certainly originated,
while the remaining two species are confined to
the Australasian region. Thus C. molestus and
C. fatigans do not naturally coexist with each
other, or with the two endemic species. Yet
they are all very effectively reproductively iso-
lated when brought into coexistence due to
man’s interference.

Yvonne Henderson is applying the pioneering
studies of Margaret Bastock, Aubrey Manning
and Herman Spieth to a comparative study of
the courtship behaviour of members of the
Drosophila melanogaster species sub-group, re-
cently reviewed by Bock and Wheeler (Bock
and Wheeler, 1972).

N. Monzu’s studies of the genus Calliphora
include the measurement of reproductive isola-
tion between sympatric and allopratic forms.
Studies of pre- and postmating isolation between
isolated demes of the frog species, Crinia insig-
nifera, Moore, by J. Blackwell, have failed to
substantiate the suspected existence of a devia-
tion between demes breeding on sand and demes
breeding on clay (Main, 1970).

An aspect of the study of isolating mechanisms
is the detection of unsuspected sibling species.
Using the methods of polytene chromosome cyto-
logy C. A. Green has been investigating the taxon
Anopheles annulipes Walker which has a wide dis-
tribution within Australia and its neighbouring
islands. Thus far he has detected four sibling
species which he has temporarily designated as
Species A, B, C and D. The species are readily
identified by their X-chromosomes, which have
been involved in extensive rearrangement dur-
ing the course of evolution. Species A and B
are sympatric over considerable areas of West-
ern Australia (at least between Kalbarri and
Geraldton). Species A and species C co-exist in
Southern Queensland. Species D extends south-
wards from the Ord River possibly as far as
Geraldton, though the southern record requires
confirmation by means of crossing experiments
with an Ord River stock. So far no hybrids
have been detected in nature. Studies in this
complex are hampered by the need to main-
tain them in culture by means of artificial
mating methods. Crosses are performed in the
same way.

Julian Ford and S. A. Parker have recently
shown that the Australian Wedgebill comprises
two sibling species Psophodes cristatus (Gould)
in the east, and P. occidentalis (Matthews) in
the west. The ranges of these two bird species
overlap in the region of Oodnadatta, South Aus-
tralia. The fact that the two gene pools exist
was revealed by the use of song analysis (Ford
and Parker, 1973).

R. J. Mahon has made use of the gel electro-
phoresis of heart muscle enzymes to demonstrate
that the common blue and orange rock crabs,
hitherto referred to as Leptograpsus varicgatus

(F), actually belong to two distinct species.
Until this demonstration L. variegatus has been
generally regarded as a single polymorphic
species.

The study of the genetic structure of species

The study of the genetic structure of species
poses particular problems which have been
considerably alleviated in recent years with the
introduction of gel electrophoresis as a tool
with which to score populations for variation
at loci determining certain enzymes and other
proteins.

Crinia insignifera Moore is a very small
Leptodactylid frog found on the coastal plain of
south-west Western Australia. The species is
broken up into a large number of unusually dis-
crete demes each occurring near a winter breed-
ing site. Because of the small size of these frogs
their vagility is considered to be rather low.
This situation has been exploited by J. Blackwell
to study genetic divergence between demes
within a species. Here again use has been made
of gel electrophoresis to study variation at loci
determining enzymes. Using eight enzyme sys-
tems she will have data on at least 10 loci. This
information will enable her to study the diver-
gence of demes, and to estimate how effectively
each population is isolated from its nearest
neighbours. To facilitate the scoring of the gels
she has studied the genetics of each system
utilized. Studies such as this are much needed
as checks on population genetic theory, which
has moved well ahead of the studies which are
actually feasible.

J. den Hollander is studying the structure of
populations of Musca domestica L. using loci
involved in determining malencss, and by study-
ing variation involving the sex chromosomes.
Most Australian housefly populations are
characterized by the absence of a Y-chromo-
some, in sharp contrast to most populations of
the species which have been studied elsewhere
(Rubini, 1964; Milani, 1964). By using marked
chromosomes den Hollander has shown that in
the Western Australian flies so far examined
there is a male-determining locus only on
chromosome III. This is in contrast with
Wagoner’s (1969) report on a strain from Bris-
bane which carried male-determining loci on
chromosomes II, III and V, and one from Can-
berra which showed holandric inheritance of a
chromosome II (Kerr, 1961). The strain from
Canberra was XY in the males and XX in the
females, but Wagoner’s Brisbane strain was XX
in both sexes. In this study den Hollander has
only found strains without a Y-chromosome:
these stocks came from Sydney, Canberra. Mel-
bourne, Adelaide, Perth, Albany, Kalgoorlie and
Kununurra. The results presented in Table 2
have been obtained from crosses between the
Perth stock with the others mentioned. Controls
and stocks maintain a sex ratio very close to 1.

With further elaboration it is hoped that
methods will be developed with which popula-
tions of Musca domestica from anywhere in the
world can be characterized.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

34

Table 2

Sex ratios of crosses between various stocks of
Musca domestica

Cross

<5

off-spring

$

off -spring

Perth $ $ x Canberra $

510

115

Canberra $ $ x Perth

262

229

Perth $ $ x Melbourne c£ <d

420

31

Melbourne $ $ x Perth

553

503

Perth $ $ x Ord S cd

668

322

Ord $ $ x Perth $

362

388

Melbourne ? ? x Canberra cd c£

659

713

Canberra $ $ x Melbourne cd 3

296

292

Ecological studies

Darwin pointed out that the ecological inter-
actions of closely related species are of particular
importance in understanding evolution. For
this reason the studies, discussed above, which
deal with the Culex pipiens complex, the species
of the genus Calliphora and the crabs of the
genus Leptograpsus, are all concerned with this
subject.

In the Culex pipiens complex there are quite
marked differences between the four West Aust-
ralian species. The two native species are not
closely associated with man. They breed in
natural pools, usually low in organic content,
and they rarely feed on man. C. australicus will
not breed in cages, no doubt because of its
ecological requirements for mating-swarm for-
mation. C. globocoxitus will mate in small cages
and, therefore, it is very probable that its re-
quirements for swarm formation in nature are
quite distinct from those of C. australicus. This
contrast in mating behaviour between the two
sympatric endemic Australian species is remin-
iscent of the differences between the two en-
demic members of the complex in Europe,
C. pipiens, which resembles C. australicus in
requirements and C. molestus which is compar-
able with C. globocoxitus. C. australicus has
been recorded as feeding on rabbits and birds
(Lee et al., 1954). Neither endemic species
appears to bite man in nature. C. fatigans and
C. molestus both bite man and poultry. The
latter species can survive without blood meals
(i.e. it is autogenous) since females in their first
gontrophic can ripen their ovaries on reserves
laid down during the larval stages. Autogeny
has a relatively simple genetic basis, but never-
theless, it has a striking effect on the ecology
of the species by freeing it from dependence on
vertebrate hosts. This enables it to penetrate
the far north of Europe by invading sheltered
underground cellars, underground railway tun-
nels, etc., provided breeding places are available,
and despite the fact that it cannot hibernate.
This in turn might be expected to lead to a
greater degree of inbreeding than is found in
the anautogenous members of the complex.
These ecological topics are being explored by
R. Irving-Bell.

N. Monzu’s study of blowflies of the genus
Calliphora includes an attempt at defining the
niche of each common carcass-breeding species.
Much work has been done on blowfly ecology,
but most of it has been done in the laboratory
and relatively little in the field. Yet studies on
interspecific competition, for example, are only
realistic if they take into account such factors
as the behaviour patterns which determine
choice of oviposition site and the seasonal peaks
of abundance of the species. Other factors to
be considered are adaptations such as ovovivi-
parity and fecundity, the behaviour of larvae in
relation to parasites and predators, geographical
distribution and interactions with exotic blow-
flies which have intruded into the Australian
scene.

The sibling species of the rock crab genus
Leptograpsus show slight but palpable ecological
differences which were first detected by Shield
(1956), but recently studied in greater detail by
Mahon. The blue species is more resistant to
desiccation than is the orange, and as a con-
sequence it car move and feed at a greater
distance from the splash zone to which the
orange species is more or less limited. The two
species show slight differences in breeding season
as judged by the numbers of females carrying
eggs attached to pleopods on their abdomens,
differences in habitat preference as shown by
clumping in tiansects along the reefs, and in
geographical distribution.

The role of symbiosis in the Culex pipiens
complex of mosquitoes

Early workers on the Culex pipiens complex
found that crosses between certain colonies of
Culex molestus from various parts of Europe
were incompatible in one or both directions
(Marshall and Staley, 1937). This phenomenon
was later sudied in detail by Laven (1953)
who demonstrated that the incompatibility in-
volved mosquito genotypes and cytoplasmic “fac-
tors”. This has suggested to several authors
( e.g ., Rai, 1967 and Irving-Bell and myself),
that the factors may in fact be symbiotic micro-
organisms analgous to the sigma factors in
Drosophila melanogaster (L’Heritier, 1970). This
idea has been resisted by Laven (e.g., 1967). He
has considered the factors to be RNA (Laven,
1967) and more recently his colleague Jost has
suggested that they are DNA particles (Jost,
1970). However, Laven’s basis for rejecting the
symbiont hypothesis does not appear very im-
pressive. The early work of Hertig & Wolbach
(1924) and Hertig (1936) suggested to Yen
and Barr (1971) and independently, to Irving-
Bell and myself (Irving-Bell & Paterson, 1973)
the hypothesis that the cytolplasmic factors
of Laven are rickettsiae. Yen and Barr and
we have confirmed the work of various earlier
workers (Hertig, 1935; de Zulueta, 1964 and
Byers & Wilkes, 1970) by detecting rickettsia-
like symbionts in the ovaries of members of the
complex. The species in which they have been
found by us are Culex molestus and C. fatigans
from Australia and C. pipiens pipiens L. from
England. We have failed to find them in the

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

35

endemic Australian species, C. globocoxitus and
C. australicus, but have detected virus-like par-
ticles in their place in these two species.

Our attempts at relating the micro-organisms
to the phenomenon of cytoplasmic incompati-
bility have been held up by our inability to
detect incompatible populations in C. fatigans
or C. molestus in Australia, despite their re-
ported existence 'Dobrotworsky, 1955). However,
Irving-Bell has recognised in C. fatigans and C.
molestus, an apparently new form of degenera-
tion in the oocytes which is distinct from that
found with natural abortion This type of
degeneration coincides with a large increase
in the numbers of symbionts, and she is now
investigating the connection between the two
phenomena. The symbiotic relations which
have been described above are evidently of long
standing, since the one appears to be older than
the species C. Viviens, C. fatigans and C. moles-
tus, and the other evidently predates the diverg-
ence of C. globocoxitus and C. australicus . Ex-
periments now in progress are investigating
whether the two types of symbiont can be ob-
tained in the same cytoplasm, and the inter-
action of the various genotypes with the various
symbionts.

Concluding remarks

The studies reported on here will give some
impression of the scope of Species Studies as
defined above. All the studies mentioned were
designed from a purely scientific point of view,
but it will be noted that those involving pest
animals are of considerable practical value as
well. All are essentially studies of single species.
In other words, they are studies at the popula-
tion level of complexity. Species Studies are
also an essential first stage in the study of
ecosystems. This is, of course because the organ-
isation which is apparent at the community
level results from selection acting at the level
of the individual within a species. The
ecosystem comprises a physical environment
occupied by a number of species. Ultimately its
structure depends on the process of speciation
although, more immediately, it may depend on
the migration of species. In turn, the coad-
justment of a migrant species and the ecosystem
provides the selective pressure which lead to
speciation under the system outlined in the
introduction to this section of the paper.

The significance of Species Studies might seem
obvious, and yet they are rarely pursued in the
way which Mayr had in mind. That much
remains to be done is shown by the fact that
disagreement is still widespread over so funda-
mental a matter as the process of speciation.
We find population biologists of high standing
quite unable to agree on the answer to a ques-
tion such as: “Are species the product of direct
selection for diversity, or are they by-products
of adaptive evolution?". And yet each alter-
native carries with it far reaching and contrast-
ing evolutionary implications.

Cytogenetic aspects of the speciation process
in plants

by S. H. James

Persvective

The more characteristic elements of the Aus-
tralian flora have been investigated cytologically,
mainly by Smith-White, and his findings and
those of other cytologists have been integrated
with a geological history of the continent to
produce a rather satisfying, yet suggestive,
picture of the evolution of our flora. In sum-
mary, the hypothesis developed is as follows
(Smith-White, 1959). The angiosperms origin-
ated elsewhere than in Australia. Before their
entry into Australia they achieved geographical
expansion over the major continents and under-
went adaptive diversification. The major orders
and families were established whilst the Aus-
tralian continent remained geographically
isolated and unavailable. Isolation of Australia
was temporarily broken down during the middle
Cretaceous, and a massive angiosperm invasion
occurred, leading to the replacement of the
older more primitive flora. The invading angio-
sperms included only those lineages which were
near the “bridgeheads". Subsequent isolation
of the continent has allowed the immigrant
flora to indulge in a second evolutionary cycle
which has resulted in the establishment of the
present day autochthonous groups. These
groups (e.g. Boronieae of the Rutaceae, Chamae-
laucoideae-Myrtaceae, Proteoideae and Grevil-
lioideae-Proteaceae, and the Section Crypto-
stomae of Casuarma-Casuarinaceae) show
remarkable variation in chromosome number;
they contain considerably more gross cytological
variation than their extra-Australian relatives.
Generally, this variation is shown by differences
between the chromosome numbers characteriz-
ing genera, but within the genera the numbers
are constant. The diversity in generic chromo-
some numbers originated during the early phase
of the Australian evolutionary cycle, just as the
diversity in numbers between orders and families
developed in the earlier extra-Australian
primary radiation, and in a pan-Australian
flora. Dissection of this pan-Australian flora
into its eastern and western components oc-
curred in the Miocene, after the establishment
of present day genera, and this dissection, estab-
lished first by marine intrusion, has been main-
tained by the Great Australian Desert (Crocker
and Wood, 1947), and progressive evolution
within these isolates has led to the development
of their own suites of species within common
genera. Continentally, recent phytogeographical
events include a partial breakdown in isolation
allowing the entry of modern Indo-Malayan
elements; changes of climate and topography,
particularly the relatively recent (Quaternary)
establishment of alpine conditions in Eastern
Australia, which has allowed the establishment
by long distance dispersal and by intrusion from
the north along the montane tableland route
of a “Southern" element showing affinities with
New Zealand and South America; and a general

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

36

contraction, in Western Australia, of the auto-
chthonous elements to the south west corner,
associated with a general increase in aridity
(Crocker, 1959; Churchill, 1961).

The recent immigrants, especially the her-
baceous ones, are of particular interest. They
may be expected to show, and do show, a third
order cycle of adaptive diversification. By study-
ing the distribution of cytogeographical races
and their genetic systems, it should be possible
to gain information which, in time, may contri-
bute to a coherent picture of the dynamic
phytogeography of the Quaternary. Significant
work in this field, for mainly eastern taxa, has
been done by Hayman (I960', Briggs (1962) and
Peacock (1962).

In his essay on the cytological evolution in
the Australian flora. Smith-White (1959) de-
veloped an hypothesis to account for the
chromosome number variation discussed. He
suggested that adaptive diversification of the
original immigrant lineages was mediated by a
multiplicity of newly available habitats, and by
reduced biological competition. Consequently,
the immigrant flora underwent an evolutionary
cycle which "favoured genetic experiments, in-
cluding experiments in structural changes and
in chromosome number”. This hypothesis im-
plies that the changes in chromosome structure
and number involved are basically innovative in
nature. On the other hand, it is frequently
argued (Darlington, 1958; Lewis and John,
1963) that such changes find their adaptive
utility as conservative devices.

Much of the current cytological research on
the Western Australian flora, reveiwed below, is
directed towards an understanding of the gen-
etic systems operating in particular, mainly
herbaceous, plant groups to determine how, and
with what consequences, these groups have res-
ponded to the vicissitudes of a new environment.
A major aim is to evaluate the role of chromo-
some changes as innovative and conservative
devices, and to determine whether in fact, par-
ticular* types of genetic architectures, at the
population level, are associated with particular
types of cyto-evolutionary responses. In this
way, it may be possible to account for the cyto-
logical diversity characterising the endemic
Australian tribes.

Calectasia cyanea — ancient and unchanging

Green (1964) has listed the species which
occur in both Eastern and Western Australia.
Of these, many are equipped with small seeds or
other such devices that might allow for long
distance dispersal. A few species showing this
disjunct distributional pattern, however, are not
so equipped, and their occurrence poses an im-
portant problem. What is the basis of their
conservatism? Anway (1969) investigated the
variation pattern and genetic system of Calec-
tasia cyanea R.Br., the blue tinsel lily, and noted
there to be little morphological and phytochemi-
cal variation between populations within its dis-
tributional range. However, the difference in
habit exhibited between the extensively rhizo-

matous plants of the eastern populations and
the shortly rhizomed plants of Western Austra-
lia was taken as a key character in differentiat-
ing these taxa at varietal rank. The general
exophenotypic constancy, however, was shown to
be associated with a conservative genetic system
in which Anway was able to demonstrate a close
relationship between the frequency of chiasma
formation in the long arm of chromosome no. 9.
which constitutes some 20^- of the total genetic
material, and the percentage frequency of sterile
pollen. It appears that all C. cyanea plants are
self pollinating and heterozygous for differenti-
ated chromosome 9 long arms, and that recom-
bination between these “supergenes" results in
the synthesis of gene combinations lethal in the
haploid phase. As Anway concludes: "The lack
of . . (genetic) . . flexibility has prevented sig-
nificant evolutionary change, and the species is
now restricted to those limited areas which
provide the ecology to which it is adjusted.
Calectasia is a true relic, apparently committed
to ultimate extinction by the adoption of an
ultra-conservative genetic system."

Isotoma petraea — recent complex hybridity

Isotoma petraea F. Muell. is another species
which has evolved a system . of permanent hy-
bridity. It is a largely autogamic species which
occurs in discrete populations on granite out-
crops and other rocky areas throughout the
Ereman Province of Australia. In the south-
western corner . of its distributional range in
Western Australia, complex hybridity has
evolved; the populations are made up of plants
heterozygous for multiple interchanges . made
permanent by the operation of its autogamic
breeding habit and a system of balanced lethal
genes. A detailed hypothesis concerning the
evolutionary origin of this genetic system from
structurally homozygous progenitors has been
developed (James, 1965; 1971).

Beltran (1971) has investigated the generation
of hybrid vigour in interpopulational crosses in
I. petraea. He has shown that the degree of
heterosis achieved in such crosses depends in
large measure upon the genetic systems of the
parents involved. Interpopulational hybrids
between structural homozygotes will show sub-
stantial heterosis if the parental populations
exhibit little outcrossing, and little or no
heterosis if the parental populations exhibit
substantial amounts of outcrossing. Interpopula-
tional hybrids between complex heterozygotes
exhibit either no or negative heterosis. These
results suggest that the I. petraea population
system may be viewed as a series of lineages
inbred to varying degrees; amongst the structu-
ral homozygotes, crossing generates heterosis,
the degree and significance of which depending
upon the relative degree of inbreeding character-
ising the parents, but all the structurally
homozygous populations constitute a single
coadapted gene pool in that the union of gametes
drawn from different populations produces a
hybrid as vigorous as or more vigorous than the

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July. 1973.

37

parents. On the other hand, amongst the com-
plex hybrid populations, the absence of heterosis
in interpopulation hybrids conforms with the
hypothesis that the complex hybrids are already
conserving high levels of genetic hybridity, and
the negative heterosis observed in some inter-
populational hybrids suggests that coadaptation
no longer exists between the complex hybrid
populations.

Laxmannia — responses to autogamy
Laxmannia , a member of the tribe John-
sonieae of the Liliaceae is one genus which
promises rewards for detailed biosystematic
studies (Keighery, unpublished). Possibly the
most striking morphological feature in this genus
is the development of shortly to strongly pedun-
culate inflorescences in most species, with sessile
inflorescences in L. sessiliflora Dene. This sug-
gests that natural selection has favoured a more
effective presentation of flowers and, therefore,
cross pollination. More importantly, L. sessili-
flora has an open pollinating mechanism while
the rest of the species so far examined have a
floral mechanism in which open flowers are pre-
sented for cross pollination, but closure of the
flowers in the evenings causes the anthers, still
with pollen attached, to be pressed on to the
stigma, to effect self pollination. However, a
comparison of seed set following seifs and crosses
shows that the products of self pollination in
these latter species are to a large extent elimi-
nated by lethals operating post zygotically. This
seems to be a rather inefficient system, involv-
ing the wastage of many ovules, a situation par-
ticularly important in Laxmannia where, it is
interesting to note, the number of ovules per
flower is three in L. sessiliflora and increases to
15 to 21 in species with lethal systems. Addi-
tionally, in L. ramosa Lindl., certain popula-
tions on the coastal plain around Perth exhibit
high frequencies of interchange heterozygotes as
well as the automatic self-pollination system and
the elimination of selfed products by lethal sys-
tems. Again, there is a conjunction of inter-
change exploitation and the generation of lethal
systems in association with a change in the
breeding system towards autogamy, but the
system has not achieved the rigidity of that
exhibited by Isotoma pctraca. Laxmannia squar-
rosa Lindl. is a highly variable species, and with
L. grancliflora Lindl., L. sessilis Lindl. and sev-
eral other forms, constitutes a complex of some
taxonomic confusion. These species exhibit the
automatic self-pollination mechanism described
above, post-zygotic elimination of selfed pro-
ducts, and a variety of other cytogenetic devices
including chiasma localization, polyploidy, aneu-
ploid reduction, and a complex but as yet un-
resolved pattern of genetic coadaptation between
forms. It is clear that an understanding of the
genetic architecture and strategies of this group
will greatly facilitate the construction of a
meaningful taxonomy.

Stylidium — lethals and aneuploidy

Stylidium. is one of the more remarkable com-
ponents of the Western Australian flora. It is
by far the largest genus of the Stylidiaceae,
a family restricted to Australia (including Tas-
mania), New Zealand and Antarctic South
America, plus a few species penetrating back
into the Indo-Malayan region. Of the 140 species
currently recognized, 102 occur in the south-
west of Western Australia, 30 occur in northern
Australia and beyond, while 8 occur only in the
eastern states. A comparison of the cytological
information presently available indicates that
the base chromosome number is x 15, a num-
ber constant, except for polyploidy, amongst all
the eastern states and New Zealand records.
However, associaed with the radiation of species
that has occurred in the south-west of West-
ern Australia, is a massive propensity for aneu-
ploid reduction. In Western Australia, chromo-
some numbers ranging from n 16 down to
n 5 have been recorded, with polyploidy oc-
curring on 15, 14 and 13 (James, 1973; Banyard,
1973). Aneuploid reduction series are in evi-
dence in most sections of the genus as recog-
nised by Mildbraed (1908), with possibly several
independent reduction series occurring in some
of the larger sections. Carlquist (1968) has com-
mented upon the bio-systematic interest of the
Stylidiaceae and upon the lack of published
information. He speculated upon the factors
which may have contributed to its striking
speciation in Western Australia, and suggested
that adaptation to pollination vectors and to
soil mosaic patterns may have been of prime
importance. The information on chromosome
number variation, which was not available to
Carlquist, is indicative, however, of important
cytogenetic situations.

The genetic systems and biosystematic rela-
tionships in several species groups in Stylidium
have been in\ estigated. Farrell (1973) has
shown that Stylidium calcaratum R.Br. and
S. ecorne * the latter formerly considered to be
a variety of the first, differ in chromosome
number (n 11 and n 13 respectively). S. ecorne
having the higher number, may be considered
primitive to S. calcaratum. and this directional
reading of the evolutionary sequence conforms
with the proposition that the well developed
floral spur and the adaptation to more xeric
habitats characteristic of 5. calcaratum are de-
rived conditions. Seed set following selfing in
both species is as good as that following intra-
populational crosses. However, interpopulational
crosses in S. ecorne yield high seed sets, but in
S. calcaratum reduced seed sets, relative to in-
trapopulational crosses, are found. Thus, the
evolution of S. calcaratum from S. ccome-like
ancestors has involved not only the acquisition of
a perfected floral spur and an increased ecologi-
cal amplitude, but also an aneuploid reduction of
chromosome number, and a loss of coadaptation
between the calcaratiun populations.

*‘S ccornc' is ;i manuscript name to be formally pub-
lished In Farrell (1973), the taxon so named Is
equivalent to Stylidium calcaratum var. ccornc F.
Mucll, ex Erickson and Willis.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July. 1973.

38

Banyard (1973) has made comparable studies
on S. elongaium Benth. (n 13 and 26),
S. confluens* (n— 14) and S. crassifolium R.Br.
(n 14 and 28), three distinct species considered
by Carlquist (1968) to form a continuous varia-
tional cline. In all three species, at the diploid
level at least, comparison of seed set following
selfing and intrapopuiational crossing demon-
strate the elimination of most selfed products
by the operation of post-zygotic lethals. Inter-
populational crosses show that there is again, in
all species, a breakdown in coadaptation. Best
seed set is obtained in interpopulational crosses
involving parents drawn from populations some
15 to 40 miles apart. When the interparental
distance is 100 to 150 miles, the seed set is re-
duced to levels comparable to that obtained in
intrapopuiational crosses. Thus, it would seem
that in these species, individual plants are
heterozygous for several to many recessive lethal
genes, and these lethals are shared by most or all
members of a population. The arrays of lethals
vary from population to population and popula-
tions sufficiently distant have fewer lethals in
common. However, increasing distance between
populations is associated with increasing genetic
divergence, and this may be fostered either by
exophenotypic adaptation to differing environ-
mental circumstances, or by random, perhaps
neutral, endophenotypic variation.

Tetraploid laces occur in both S. elongatum
and S. crassifolium. In the latter, the tetra-
ploids occur towards the centre of the species
range; the species is characteristically distribut-
ed in this region as very small localized popula-
tions which must perforce be relatively inbred.
Since the ancestral diploid populations contain
arrays of recessive lethal genes within their gene
pools, it seems certain that the role of tetra-
ploidy here is to conserve hybridity. A similar
role may be attributed to tetraploidy in
S. elongatum, but in this case, the species
occurs as small scattered populations in yellow
sandplain, in contiguous association with diploid
S. elongatum to the west and diploid 5. confluens
to the south and east. Thus, it may well be that
the adaptive utility of tetraploidy here is found
in the genetic barrier it establishes between
these three ecologically-differentiated conti-
guous forms.

Stylidium brunonianum Benth. (n— 9) is an
extremely variable species. Typically, local pop-
ulations are made up of one or a few forms
differing from each other in various morpho-
logical and ecological attributes. Each biotype
appears to have its own suite of characteristics.
One infraspecific taxon, sub sp. minor, seems of
valid rank, but within sub sp. brunonianum the
variation pattern cannot be partitioned into a
satisfactory infraspecific taxonomy. The limits
and relationships of S. striatum Lindl. S. rigidi-
folium Mildbr. and S. brunonianum sub sp.
brunonianum are most confused. Ling (un-
published and Stone (unpublished) have in-

*‘S. confluens’ is a manuscript name to be formally
published in Banyard (1973). The taxon so named
is presently embraced in Stylidium crassifolium
R.Br. subsp. crassifolium sensu Carlquist.

vestigated the genetic systems found in this
group. Again, post-zygotic lethals operate to
eliminate the products of self pollination, and
inter populational crosses reveal a breakdown in
genetic coadaptation between populations as in
S. elongatum, S. confluens, S. crassifolium and
S. calcaratum. However, in S. brunonianum, the
loss of coadaptation with interparental distance
is much more rapid than in the other species
mentioned; the optimum interparental distance
being in the order of miles rather than tens of
miles. S. brunonianum shows further develop-
ments in its genetic system in that many popu-
lations exhibit high frequencies of male-sterile
plants in which the anthers abort soon after
pollen mother cell meiosis, and the stigma
develops precociously within unopened buds.
In addition a high proportion of the male-
fertile plants exhibit substantial degrees of
pollen sterility, although no cytological aberra-
tions have been associated with this. In one
population examined, Stone (unpublished) found
a multimodal distribution of pollen sterilities
among plants, with a pronounced peak around
the 50-60 % sterility level. This suggests the
existence of synthetic gametic lethals, in which
recombinant products from an independently
assorting bifactorial system are lethal.

The extent to which lethal systems have been
developed in Stylidium appears to be quite
novel, if not unique, and its seems safe to pre-
dict that causal association between lethal
system exploitation and the aneuploid specia-
tion method involved will accumulate in the
future. Although the details of the speciation
methods operative in this genus have yet to be
resolved, it is incontestable that the phenomena
reviewed above have anything but profound
effects on the biosystematics and evolutionary
potentials of the group.

Dampiera linearis — polyploidy and
B-chromosomes

Dampiera linearis R.Br. is a species of the
Goodeniaceae to which attention was initially
drawn by Peacock’s (1963) observation of intra-
specific polyploidy. Bousfield (1970) has shown
that in this species there is a wide-ranging
hexaploid race in the southwestern corner of
Western Australia, a tetraploid race on the
lower western coastal plain and two disjunct
areas of diploids, one near Albany, the second
confined to the Wicher Range below Busselton.
The diploids are cross incompatible with the
tetraploids and hexaploids, and all forms exhibit
a self-incompatibility system based on an inter-
action between pollen tubes and stylar tissue.
Detailed analyses of the Wicher Range area
has shown that the diploids are confined to the
elevated lateritic caps of the old land surface,
while the tetraploids occur on the alluvial soils
of the erosion valleys incising the plateau and
on the coastal plains. In addition, the diploid
populations on the ecotonal slopes contain many
plants with from one to five B chromosomes,
indicating yet another cytoevolutionary response!
Bousfield has clearly demonstrated B chromo-
some/A chromosome interactions, including

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2. July. 1973.

39

physical associations between B’s and A’s during
meiotic prophase, and a polarized nondisjunction
of A chromosome bivalents at anaphase- 1 which
is directed such that the nondisjoining bivalents
become associated with nuclei containing high
numbers of B chromosomes. A survey of the
population concerned showed that plants with-
out B chromosomes were either highly pollen
fertile or had substantial pollen sterility, while
plants with one or more B chromosomes were
never associated with the “substantial” pollen
sterility levels. There is, then, a striking lack
of substantially pollen-sterile plants with one or
more B chromosomes in the population. From
this, it is inferred that there are plants which
are potentially capable of exhibiting substantial
pollen sterility, and these are postulated as
“interpopulational hybrids”, but which have had
their fertility restored by a B chromosome effect,
and this effect is postulated as being a system
of bivalent co-orientation which reduces the
segregational fractionation of the divergent
genomes combined in these interpopulational
hybrids.

Thus, the biology and the cytogenetics of the
situation .have been integrated into an evolu-
tionary hypothesis which proposes the follow-
ing. The primitive diploids on the lateritic caps
are unable to penetrate the coastal environment.
However, genetic recombination at the diploid
level in peripheral ecotonal situations has al-
lowed the construction of genotypes which are
better adapted to the novel ecology of the
coastal plain, and systems contributing to the
conservation of such newly fabricated genetic
combinations are subjected, therefore, to strong
positive selection pressure. The B chromosome
system is the conservatice device selected in this
case. A consequence of B chromosome activity
here, however, is A bivalent nondisjunction
in interpopulational hybrids. It is proposed also,
that this aberration may lead to the production
of totally unreduced gametes and hence provide
a mechanism for the origin of polyploids adaptsd
to the coastal plain invironment. This inter-
pretation for the evolutionary origin of tetra-
ploidy in D. linearis is strengthened by the find-
ing of a single tetraploid plant carrying some
six B chromosomes.

Although the interpretations of the observed
phenomena are somewhat speculative at this
stage, Dampiera linearis offers an exquisite op-
portunity to study in detail the evolutionary
process of biological spediation. Present infor-
mation suggests, then, that the invasion of the
coastal plain by the derived tetraploids is not
due to an increased ecological amplitude of
tetraploidy per se, but that tetraploidy provides
a mechanism of totally conserving genotypes
assembled at the diploid level, the whole pro-
cess being mediated by the unique properties of
the B chromosome system.

Eremophila glabra-polyploidy and ecological
preference

In Eremophila glabra (R.Br.) Ostenf., Ey and
Barlow (1972) have demonstrated a relation-
ship between chromosome number and ecolo-

gical preference. Although their intensive work
was centered around the gulfs area in South
Australia, the species is widespread in Western
Australia. In this group, the diploid progeni-
tors occur as localized populations in temperate
refugia. The tetraploids are more widespread,
but still relatively restricted compared to the
hexaploids, and occur in temperate to semi-
arid areas, while the hexaploids are of wide-
spread distribution throughout the arid region.
Thus progressive adaptation to the arid environ-
ment is associated with increasing euploid levels.
Similar patterns of geographical replacement of
euploid races in other species of Eremophila
have been documented by Barlow (1971) on a
less detailed scale, with “strong indications that
the ancient land surfaces of Western Australia
are a reservoir of relic diploid populations of
Eremophila , \ Barlow discusses the possible roles
of polyploidy in the genus, and suggests that
“the conservative genetic effects of polyploidy
in reducing segregation in newly arisen, highly
adaptive hetrozygous outbreeding biotypes are
... of far greater evolutionary value than its
supposed role of restoring fertility to partly
sterile hybrids.” In this he is in accord with
the conclusions reached by Bousfield (1970)
concerning the role of polyploidy in Dampiera
linearis. A similar interpretation of the role of
polyploidy is considered by Banyard (1973) con-
cerning the tetraploid race of Stylidium elonga-
tum.

Cassia — polyploidy , hybridization and apomixis

In Cassia (Caesalpiniaceae) the species of the
series Subverrucosae of the section Psilorhegma
exhibit a very complicated biosystematic situa-
tion. Randell (1970) has extensively surveyed
this group throughout the Eremean Province,
and although her work was largely concerned
with central Australian populations, probably
similar situations obtain in Western Australia.
The following interpretation was developed.
Polyploid forms are widespread throughout the
Eremaea. The diploid species are confined to
mountain refugia such as the Macdonnell, Flin-
ders and Kimberley Ranges, and are apparently
fairly strongly isolated from each other in that
no evidence of hybridization at the diploid level
has been obtained. However, infraspecific
polyploid races occur and interspecific hybri-
dization at the tetraploid level, and between
tetraploids and diploids, apparently occurs
with considerable frequency, generating taxo-
nomically confusing hybrid swarms containing
several chromosome numbers. In addition,
facultative pseudogamous adventitious polyem-
bryony occurs in the polyploid forms, leading
to a competition between the single sexual and
up to nine apomictic embryos within the single
embryo sac. There is some evidence that the
components necessary to assemble this .genetic
system, namely mechanisms for generating dip-
loid gametes and inducing adventitious polyem-
bryony, may be found at the diploid level.

Once assembled, this remarkable system pro-
vides for many things. Sexual reproduction and
hybridization between disparate forms generates

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

40

hybrids which may themselves have an adaptive
advantage in heterotic vigour, and from which
the sexual process may assemble a limitless
array of novel genotypes. Any genotype is
potentially conservable by the apomictic mech-
anism, but the pseudogamous and facultative
nature of this mechanism requires sexual
experimentation to proceed. The products of
sexual fertilization are subjected to a competi-
tive screening against adventitious embryos of
maternal genotype during seed formation. Thus
ovules carrying “inferior quality” sexually pro-
duced embryos are not wasted, but instead they
carry embryos of an already tested genotype.
These Cassia population systems seem to have
achieved a remarkable balance between innova-
tive sexuality and conservative apomixis.

Other groups — possibilities

Several other groups have been investigated
cytotaxonomically, and here interesting situa-
tions which may well respond to detailed cyto-
genetic analysis are in evidence. For example,
intra- and interspecific polyploidy have been
observed by Brittan (unpublished) during his
wide-ranging experimental taxonomic study of
the genus Thysanotus Polyploid cytotypes have
been found in 13 of the 27 species so far investi-
gated cytologically. There appears to be some
correlation between the occurrence of polyploidy
and a change in the pollination mechanism
towards inbreeding. In some of the polyploid
forms the anthers dehisce along their full length
by longitudinal slits instead of by terminal pores
as is usual in the genus. Under insect-free
glasshouse conditions it is found that capsules
are regularly developed by non-manipulated
flowers as a reult of self pollination associated
with the freely dehiscing type of anther, whereas
capsules only form in the pore dehiscing types
when manual self pollination has been carried
out. This association of polyploidy and self
pollination occurs in T. patersonii R.Br., the
genus’ most widespread species in Australia
which exhibits polyploidy at the tetraploid,
hexaploid and octoploid levels, in T. tenuis Lindl.
which is of relatively restricted occurrence and
uniformly hexaploid, and in T. multiflorus R.Br.
Interestingly, in T. multiflorus, tetraploids
exhibiting pore dehiscence occur as well, and
Brittan (1962) has suggested that this form is
of amphidiploid origin. It seems that in this
genus, further cytogenetic work may be able to
discriminate between differing roles for poly-
plody, even at an infraspecific level.

Secondly, Drosera is well developed in Western
Australia, having some 34 species currently
recognised. Chromosomally, the genus is quite
constant on a base number of x 10 except in
the section Lamprolepis, the pygmy sundews,
where aneuploid reduction to n 5 has been
established (Marchant, unpublished). The
pygmy sundews with the lower numbers tend
to occur in discrete localized populations whereas
the wider ranging species maintain the primitive
chromosome number. It will be of interest to
learn of any other endophenotypic differences,

as well as exophenotypic differences, which
differentiates the section Lamprolepis from the
rest of Drosera.

Similar suggestive observations have been
made by Bennett (1972) concerning Hybanthus ,
a shrubby genus of the Violaceae. Here it seems
that a primitive aneuploid reduction series,
8 >6 >4, is represented in present day taxa. In
the section Variables, the widespread Western
Australian taxa, H. floribundus (Lindl.) F. Muell.
subsp. floribundas (n— 6, 12 and 24), H. caly-
cinus (DC. ex Ging.) F. Muell. (n=6 and 12)
and H. epacridoides (C. A. Gardn.) Melch.
(n— 12) include polyploid cytotypes, while
four diploid taxa have restricted distributions.
This pattern is somewhat confounded, however,
by the presence of the polyploid H. bilobus C.
A. Gardn. (n = 12 and 24) which is of relatively
restricted occurrence, and the diploid H. mono -
petalus (Roem.et Schult.) Domin (n = 4) which
is widespread in the eastern states. Bennet
(loc. cit) also notes that pollen fertility is
characteristically quite high in the widespread
H. floribundus and H. calycinus, and characteris-
tically rather low in the other Western Austra-
lian taxa. It would seem again, that different
genetic strategies are associated with different
degrees of success, and detailed comparative
studies on the cytogenetics of Hybanthus species
may well prove rewarding.

In like manner, comparative cytogenetic
studies involving V erticordia, Darwinia and
other membrs of the Chamaelaucoideae and
members of the cytologically invariant Lepto-
spermoideae may indicate a causal basis for the
strikingly different cytological responses in these
two tribes of the Myrtaceae. Clearly, such an
approach can be extended to any group. Indeed,
the Western Australian flora provides a formid-
able store of materials for research into the
processes of biological diversification and specia-
tion.

Concluding remarks

Isotoma, Laxmannia and Stylidium, and prob-
ably Calactasia, exhibit lethal systems which
eliminate the relatively homozygous products of
self-fertilization. The polyploidy in Eremophila,
Cassia, Dampiera, Stylidium and Thysanotus
may be interpreted as a system allowing the
maintenance of high levels of hybridity within
populations, though other bases for its adaptive
utility are possible and even preferred. Never-
theless, it would appear that natural selection
has favoured systems that facilitate a pursuit
of hybridity during the differentiation of these
groups in the semi-arid to arid Western Aus-
tralian environment. Genetic hybridity is
necessary for the construction of new adapted
gene arrays by recombination, and the evolution-
ary success of lineages fostering hybridity may
well be achieved through default of the rela-
tively homozygous lineages. On the other hand,
heterozygosity per se may somehow code for a
more efficient, flexible and adaptive phenotype
than can homozygosity. The occurrence of
heterosis in interpopulational hybrids amongst

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

41

naturally occurring structural homozygotes in
Isotoma petraea and the dependence of the
extent of the heterotic effect on the breeding
system of the parents suggests value in hetero-
zygosity per se. In the Isotoma complex hybrids
and in Laxmannia and Stylidium hybridity has
been fixed by the lethal systems, and these
have led to a marked differentiation between
populations resulting in what here has been
called a loss in coadaptation. The intrinsic bar-
riers to reproduction so established between
populations in Stylidium and Laxmannia may
well provide the beginnings upon which specia-
tion may be built.

The self incompatibility system in Dampiera
linearis would seem to be a much more efficient
and less divisive system of excluding self ferti-
lization products. So far, little work on the
extent of self incompatability systems in the
Western Australian flora has been done. An-
other system which prevents self fertilization
is dioecy, and McComb (1968, 1969) has con-
cluded that the evolution of dioecy and poly-
ploidy are independent responses associated with
the maintenance of high levels of heterozy-
gosity in Isotoma fluviatilis (R.Br.) F. Muell. in
New South Wales. McComb (1966) has also
analysed the frequencies of the various sex
forms in the floras of the South-west Province
of Western Australia and of the British Isles,
and has compared these with each other and
with the analysis of the South Australian flora
made by Parsons (1958*. It was shown, inter
alia, that the proportions of species with herma-
phrodite flowers in South Australia and Western
Australia are comparable to each other (88.9%
and 90.0% respectively) and significantly
greater than in the British Isles (80.7%).
Additionally, it was demonstrated that the
average number of species in the Western
Australian hermaphrodite genera is 9.4 while
in the British Isles hermaphrodite genera it is
2.8. The average number of species per genus
for non-hermaphrodite genera shows much less
disparity, 4.1 for Western Australia and 3.1 for
the British Isles. Thus, the Western Australian
flora has proportionately fewer sexually differ-
entiated species and, among the hermaphrodites,
significantly larger genera than has the British
Isles. These differences may reflect the differ-
ent states of development of plant systematics
in the two regions, or it may reflect biologically
important differences. Xhe following specula-
tion is irresistible. The Western Australian flora
is younger and evolutionarily more dynamic
than that of the British Isles; the adaptive
diversification of its components is associated
with the exploitation of a variety of devices con-
serving hybridity or adaptive gene arrays, or
both, and some of these conservative mechanisms
are more divisive than others. Divisive mech-
anisms must result in the fragmentation of the
gene pool, and in speciation. The resources of
genetic variation available within each pro-
duct p species would be relatively restricted
compared with that available to the whole. As
the Western Australian evolutionary dynamism
dissipates, those lineages which adopted the less

restrictive conservative systems may be expected
to have the better chances of survival. Lineages
which have adopted dioecy to maintain hybri-
dity may fare much better than hermaphro-
dites encumbered with lethal systems and,
through natural selection, the spectrum of sex
forms in this flora may eventually approach
that of the British Isles.

Acknowledgements. — We sincerely thank our students
and colleagues who have permitted us to outline much
of their unpublished information and who. through
their lively and enthusiastic argument and discussion,
and through their own researches, have much influenced
our ideas. We are grateful for the support from
grants provided by the Australian Research Grants
Committee and by the University of Western Aus-
tralia for research in these fields.

References

Anway, J. C. (1969). — The evolution and taxonomy of
Calectasia cyanea R.Br. ( Xanthorrlioeaceae)
in terms of its present-day variation and
cytogenetics. Australian Journal of Botany
17. 147-59.

Banyard. B. J. ( 1973) .— Biosystematics of the Stylidium
crassifolium group. (In preparation )

Barlow. B. A. (1971). — Cytogeograpliy of the genus
Eremophila. Australian Journal of Botany
19: 295-310.

Bastock, M. (1956).— A gene mutation which changes
a behaviour pattern. Evolution 10: 421-439.
Beltran, I. C. (1971) .—"Embryology . balanced lethal sys-
tems and heterosis in Isotoma petraea”
PhD. Thesis. University of Western Aus-
tralia.

Bennett. Eleanor. M. (1972).- A revision of the Austra-
lian species of Hybantlius Jacquin (Viola-
ceae). Nuytsia 3: 218-41.

Bock. I. R. and Wheeler. M. R. (1972).— The Droso-
phila mclanogastcr species group. Univer-
sity of Texas Publication 7213. 1-102.
Bousfield. L. R. (1970). — "Chromosome races i?i Dam-
piera linearis R.Br." Ph.D. Thesis. Univer-
versity of Western An tralia.

Briggs, Barbara G. (1962). -Interspecific hybridization
in the Ranunculus lappaccus group. Evolu-
tion 16. 372-90.

Brittan. N. H. (1962). Variation, classification and
evolution in flowering plant —with parti-
cular reference to Th ysauotus. Journal of
the Royal Society of Western Australia 45.
1 - 11 .

Byers. J. R. and Wilkes. A. (1970). — A riekettsia-like
micro-organism in Dahlbominus fuseipen-
nis (Zett.) ( Hymenoptera. Eulophidae);
Observations on its occurrence and ultra-
structure. Canadian Journal of Zoology 48:
959-964.

Carlquist, S. ( 1969) .— Studies in Stylidiaceae: new taxa,
field observations, evolutionary tendencies.
A I iso 7: 13-64.

Churchill, D. (1961). — "The tertiary and quaternary vege-
tation and climate in relation to the living
flora in south Western Australia." Ph.D.
Thesis. University of Western Australia.
Crocker. R. L. (1959).— Past climatic fluctuations and
their influence upon Australian vegetation.
Monographiac Biologicae 8: 283-90.

Crocker. R. L., and J. G. Wood (1947).- Some historical
Influences on the development of the South
Australian vegetation communities and their
bearing on concepts and classification in
ecology. Transactions of the Royal Society
of South Australia 71: 91-136.

Darlington. C. D. ( 1958) . — "The Evolution of Genetic
Systems" . Allen and Unwin. London,
de Zulueta, J. (1965).— The Rickettsia Wolbaehia pipi-
entis in C. pi pirns and C. fatigans.
WllO/VC/ 125.65: 69-70. (Mimeographed docu-
ment. ) .

Dobrotwortsky. N. V. (1955).— The Culu.r pipiens groups
in South-Eastern Australia. IV. Crossbreed-
ing experiments within the Culex pipiens
group. Proceedings of the Linn can Society of
New South Wales 80: 33-43.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July. 1973.

42

Ey, T. M. and B. A. Barlow (1972). Distribution of
chromosome races in the Eremophila glabra
complex. Search 3 : 337-8.

Farrell. Pauline ( 1973).— Biosystematic studies on two
ephemeral species of SLylidium. (In prepara-
tion).

Ford. J. and Parker. S. A. (1973).- A second species of
Wedgebill. Emu. (In press)

George, R. W. (1969).- Natural distribution and specula-
tion of marine animals. Journal of the Royal
Society of Western Australia 52: 33-40.

George. R. W. and Main, A. R. (1967).— The evolution
of spiny lobsters (Palinuridae) : A study of
evolution in the marine environment.
Evolution 21: 803-820.

Green, J. W. (1964). — Discontinuous and presumed
vicarious plant species in southern Austra-
lia. Journal oj the Royal Society of West-
ern Australia 47: 25-32.

Hayman, D. L. (I960). The distribution of the chromo-
some races of Themcda australis in south-
ern Australia. Australian Journal of Botany
8. 58-68.

Hertig, M (1936).— The rickettsia, Wolbachia pipientis
(Gen. et Sp.N.) and associated inclusions of
the mosquito, Culex pipiens. Parasitology
28: 453-490.

Hertig, M. and Wolbach, S. B. ( 1924).— Studies on
rickettsia-like micro-organisms in insects.
Journal of Medical Research 44: 329-374.

Irving-Bell, R. and Paterson, H. E. (1973). — Intracellu-
lar symbionts and infertility in mosquitoes
of the Culex pipieris complex. Science (In
pres).

James, S. H. (1965). — Complex hybridity in Isotoma

pclraca. 1. The occurrence of interchange
heterozygosity, autogamy and a balanced
lethal system. Heredity 20: 341-53.

James, S. H. (1971). — Complex hybridity in Isotoma

petraea. II. Components and operation of a
possible evolutionary mechanism. Heredity
35: 53-78.

James, S. H. (1973). — Chromosome numbers and cyto-
evolution in Stylidiaceae. (In preparation.)

Jost, E. (1970). — Untersuchungen zur Inkompatibilitat
in Culex pipiens Komplex. Wilhelm Roux
Archiv fur Entwicklungsmechanik der Organ-
ismen 166: 173-188.

Kalmus, H. (1941). — Physiology and ecology of cutical
colour in insects. Nature, London 148: 428-
431.

Kerr, R. W. (1961). — Inheritance of DDT resistance in-
volving the Y-chromosome in the housefly
< Musca domestica) . Australian Journal of
Biological Sciences 14: 605-6)9.

Laven, H. (1953). — Reziprok Unterschiedliche Kreuzbar-
keit von Stechmucken (Cu'icidae), and ihre
Deutung als plasmatische Vererbung. Zeitz-
chrift fur Vererbung slehre 85: 118-136.

Laven, H. ( 1967) .— Speciation and Evolution in Culux
pipiens. In: Wright J. W. and Pal R.: “ Gene-
tics of Insect Vectors of Disease”. Elsevier,
Amsterdam.

Lee. D. J., Clinton, K. J. and O’Gower, A. K. (1954). —
The blood sources of some Australian mos-
quitoes. Australian Journal of Biological
Sciences 7: 282-301.

Levin, D. A. (1970). — Reinforcement of reproductive
isolation: Plants versus Animals. American
Naturalist 104 (940): 571-581.

Lewis, K. R. and B. John (1963). — ‘‘Chomosome Marker” .
J. and H. Churchill Ltd., London.

L’Heritier, P. (1970). — Drosophila viruses and their role
as evolutionary factors. Evoluntionary Bio-
logy 4: 185-207.

Mackerras, I. M. (1962). — Speciation in Australian Taba-
nidae. In Leeper, G. W. (ed.) “ The evolu-
tion of living organisms” . Melbourne
University Press, Melbourne.

Main, A. R. (1970). — Ecology, Systematics and Evolu-
tion of Australian Frogs. Advances in Ecolo-
gical Research 5: 37-86.

Main, A. R., Lee, A. K. and Littlejohn, M. J. (1958). —
Evolution in three genera of Australian
frogs. Evo'ution 12: 224-233.

Main, B. Y. (1962). Adaptive responses and speciation
in the spider genus Aganippe Cambridge.
In Leeper, G. W. (ed.) “ The evolution of
living organisms” . Melbourne University
Press, Melbourne.

Marshall, J. F. and Staley, J. (1937).— Some notes re-
garding the morphological and biological
differentiation of Culex pipiens Linnaeus and
Culex rnolestus Forskal (Diptera, Culicidae).
Proceedings of the Royal Entomological
Society of London (A) 12: 17-26.

Mayr, E. ( 1942 ) . — “Systematics and the origin of species”.
Columbia University Press, New York.

Mayr, E. Editor ( 1957). — ‘‘The Species Problem”.

American Association for the Advancement
of Science, Washington.

Mayr, E. (1963). “ Animal species and their evolution”.

Columbia University Press, New York.

McComb, J. A. (1966). The sex forms of species in the
flora of the south-west of Western Austra-
lia. Australian Journal of Botany 14: 303-
16.

McComb, J. A. (1968). The occurrence of unisexuality
and polyploidy in Isotoma fluviatilis. Aus-
tralian Journal of Botany 16: 525-37.

McComb, J. A. (1969).— The genetic basis of sexuality in
Isotoma fluviatilis. Australian Journal of
Botany 17: 515-26.

Milani, R. (1964). — Citologia della mosca domestica

Musca domestica L.) Quadernidi La Ricerca
Scientifica 25: 111-116.

Mildbraed, J. ( 1908 ) .—Stylidiaceae. In Engler, A. “ Das
Pflanzenreich" IV, 278. Engelmonn, Leipzig.

Moore, J. A. (1957).— An embryologist’s view of the
species concept. In Mayr, E. Edit. “ The
Species Problem”. American Association for
the Advancement of Science, Washington.

Parsons, P. A. ( 1958) .— Evolution of sex in flowering

plants of South Australia. Nature 181:
1673-4.

Peacock, W. J. ( 1962 ) .—Chromosome races in Goodenia
bellidifolia Sm. Proceedings of the Linnean
Society of New South Wales 87 : 388-97.

Peacock, W. J. ( 1963) .—Chromosome numbers and cyto-
evolution in the Goodeniaceae. Proceedings
of the Linnean Society of New South Wales
88: 8-27.

Randell, Barbara R. ( 1970).— Adaptations in the genetic
system of Australian arid zone Cassia
species (Leguminosae, Cae' alpinioideae) .
Australian Journal of Botany 18: 77-97.

Rai, K. S. (1967).— Techniques for the study of cyto-
genetics and genetics of vectors. In Wright,
J. W. and Pal, R. ‘‘Genetics of Insect vectors
of Disease”. Elsevier. Amsterdam.

Rubini, P. G. ( 1964 ) .— Polimorfismo cromosomico in
Musca domestica L. Bollettino di Zoologica
31: 679-694.

Sanderson, K. E. ( 1971 ) .—Genetic homology in the
Enterobacteriaceae. Advances in Genetics
16: 35-51.

Serventy, D. L. and Wittell, H. M. (1951 ). — ‘‘Handbook
of the birds of Western Australia” . 2nd
Edition. Lamb Publications, Perth.

Shield, J. W. ( 1956) .—Population studies in the littoral
et Rottnest Island. Journal of the Royal
Society oj Western Australia 42: 89.

Smith-White, S. ( 1959) .— Cytological evolution in the
Australian flora. Cold Spring Harbour Sym-
posia in Quantitative Biology 24: 273-89.

Thomson, J. A. (1969). — The interpretation of puff pat-
terns in polytene chromosomes. Currents in
Modern Biology 2: 333-338.

Wagoner, D. E. ( 1969 ) .— Presence of male determining
factors found in three autosomes in the
housefly, Musca domestica. Nature, London
223: 187-188.

Yen, J. H. and Barr, A. R. ((1971). — New hypothesis of
the cause of cytoplasmic incompatibility in
Culux pipiens L. Nature, London 232: 657-
658.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2. July. 1973.

43

5. — Aboriginal man in southwestern Australia

By D. Merrilees 1 , W. C. Dix\ S. J. Hallam 2 3 * , W. H. Douglas' 5 and R. M. Berndt 2

Abstract

Firm evidence from excavations demonstrates
that Aborigines have been present in south-
western Australia for about 25.000 years. How-
ever. findings from excavated and non-excavated
sites could appreciably extend this time-depth
to show a much earlier human occupation.
Many of these sites are of major scientific im-
portance. Some are listed and located on an
accompanying map. One criterion for classify-
ing them is their original nature or use: e.g.,
ceremonial sites, stone arrangements, hunting
devices, rock engravings, quarries, and surface
scatter or artefacts. They are now subject to
legislative protection. A topographical archaeolo-
gical approach based on field survey with select-
ive excavation, supplemented by ethno-historical
evidence, provides an ecological perspective on
the interaction of southwestern Aboriginal popu-
lations. through time, with their environment.
A special focus here is on the Perth area, parti-
cularly in relation to surface artefact assem-
blages.

The focus then shifts to the contemporary
and near-contemporary scene. Various languages
and dialects were spoken in this area, but
intensive alien contact led to the emergence of
one, Njungar (Nyungar), as a general south-
west language. Its phonological grammatical
and lexico-semantic characteristics are outlined.
The final paper sketches traditional south-west
Aboriginal socio-cultural life at the beginning
of European settlement. In the wake of that
settlement came rapid depopulation and, within
fifty years or so, destruction of the traditional
systems. Out of this physical, psycho-social
and cultural wreckage, have come the mixed
Njungar people of Aboriginal descent who now
seek an Aboriginal identity of their own.

Introduction

These five papers are devoted to a considera-
tion of what can be called the south-west
Aboriginal heritage. Three sorts of data are
involved. 1) The archaeological approach takes
as its point of departure surviving Aboriginal
sites through which the past can be explored.
It is building up a picture of Aboriginal settle-
ment that can be related to the situation in
other parts of this continent. 2) Because
systematic anthropological research was un-
developed during the early settlement of the
south-west, reliance has had to be placed on the
often incomplete and subjective evidence of the
earlier recorders. It is therefore impossible, at
this point in time, to obtain a deep understand-
ing of traditional Aboriginal life in this area.
These early sources, however, do afford valuable
insights. 3) On the other hand, it has been
possible to reconstruct aspects of the socio-
cultural past from the often fragile memories
of elderly Aborigines. And with this, is direct

’Western Australian Museum.

2Department of Anthropology, University of Western
Australia.

3 Director, United Aborigines Mission Language Depart-

ment, Kalgoorlie, Western Australia.

anthropological research into the contemporary
situation, among persons of Aboriginal descent
who have more associations with the wider
Australian society than they have with their
Aboriginal forbears.

The range of accumulating data on the south-
west Aborigines, from the far-distant to the
near-past and to the present-day, even though
so much information is missing, adds up to
something approximating an overall statement.
And in summarized form, this is what is pre-
sented here. Traditionally, these Aborigines be-
longed to a non-circumcising area: this fact
alone placed them, anthropologically speaking,
in a minority position vis-a-vis other Australian
Aborigines. They also possessed some unique or
distinctive characteristics which separated them
conceptually from the others. In the process of
intensive contact, virtually all of these identi-
fying aspects were eroded. Those which survive
identify them simply as being of Aboriginal
descent in a generalized sense and not distinc-
tively as south-west Aborigines: that has been
irretrievably lost.

Early human occupation of southwestern
Australia

by D. Merrilees

Sites indicating early occupation

Excavation localities and other sites providing
evidence of early occupation of the southwestern
part of the continent are included in Figure 1.
Notes on these sites are given below, with refer-
ence where possible to published information,
even if only brief, or an indication of studies
planned or in progress. Some of these sites are
mentioned below in other contexts by other
authors. Occupation sites represented by surface
scatters of artefacts not readily dateable are not
referred to here. Hallam (1972b) gives a rela-
tive chronology for such artefact assemblages in
the Perth area.

5. Wilgie Mia. Aboriginal ochre mine in use from
late Holocene to modern time. (I. M. Crawford, personal
communication.)

9. Coolarburloo Pool. Artefact apparently in same
geological formation as remains of large extinct mar-
supial and australite, though not at same site. (Merri-
lees, 1968b: 15) Pleistocene?

36. Hastings Cave. (Lundelius, 1960: as ‘Drovers Cave’.)
Later excavations have shown charcoal concentrations
interpreted as hearths and an artefact; mid-Holocene
on radiocarbon dating. (A. Baynes, personal communi-
cation.)

47. Caladenia Cave. Artefacts with associated faunal
remains in excavation. (R. Roe and D. Merrilees, in
progress.) Holecene?

48. Rock shelter near Gingin. Artefacts and faunal
remains. (Roe 1971.) Holocene?

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

44

Figure 1.— Sites of excavations and other sources of Information on Aboriginal occupation of the south west of
Western Australia from prehistoric time to early contact with European Invaders. See texts by Merrilees. Dlx
and Hal lam. The map also attempts to give a first Impression of vegetation at the time of European con-
tact and of the physiographic nature of the region. The dissected remnant of the long Darling Fault scarp sharply
demarcates an eastern plateau from a low lying plain to the west. In the far south west another fault system
demarcates this plain from a raised coastal block. The rivers shown are (from N to S) the Murchison. Swan-

Avon and Blackwood.

49. Dunstan's Quarry. Douglas. Kendrick and Mer-
rilees (1966) report site as fossil carnivore den. but
presence of mussel shell and subsequent recognition
of limestone flake as an artefact (A 17553 In Western
Australian Museum collections) suggest an archaeolo-
gical component. Holocene?

49. Orchestra Shell Cave. (Hallam, 1971b.) Roof
markings, with artefacts and faunal remains In excava-
tion. Holocene on radiocarbon dates.

49. Murray’s Cave. (Hallam. 1971b.) Artefact and
faunal remains In excavation. Holocene on radiocarbon
date.

49. Yonderup Cave, Yanchep. Excavations by D. S.
Davidson. unpublished. Some faunal remains in
Museum collection. Holocene?

56. 65. Examples (only) of fossil lake sites In ’wheat
belt’. Artefacts and faunal remains reported by Bet-
tenay (1962). Pleistocene? Reconnaissance studies sub-
sequently made by J. M. Bowler.

57. Frieze Cave. (Hallam. 1972b.) Excavation with
artefacts, ochre. Late Holocene.

66. Cowaramup Point. Artefacts and faunal remains
in well lithifled fossil soil. (C. E. Dortch, personal com-
munication.) Pleistocene?

67. Mammoth Cave. (Merrilees. 1968b.) The main
source of information on the Pleistocene fauna of the
south-west, with many species now totally or locally
extinct and some Indication of a climate differing from
the present (Merrilees. 1968a). Some bones charred,
and Investigation in progress of possibility that this
charring results from man-made flres (M. Archer.
I. M. Crawford and D. Merrilees). Radiocarbon date
greater than 37.000 yr. B.P. reported (Lundellus. 1960).
but there Is uncertainty concerning the specimens to
which this date refers.

68. Devil's Lair. (Dortch and Merrilees. 1972; 1973.)
Deep deposit of ’cave earth’ with artefacts, human
teeth, and abundant faunal remains. Cave occupied at
times by human beings. Self-consistent set of radio-
carbon dates Indicates man present 25.000 years ago.
and bottom of deposit not yet reached. Studies con-
tinuing.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July. 1973.

45

68 Strong's Cave. Human remains at shallow depth
(Merrllees, 1968b). Holocene? Other caves in Cape
Leeuwin - Cape Naturaliste region have yielded similar
human skeletal material.

69 Coast south of Scott River. Artefacts and human
skeletal and other faunal remains in blown-out dunes
(Butler. 1969). Holocene?

79. 80 Guaralya and Wonbema rock holes. Balla-
donia district. Artefacts associated with faunal re-
mains, some of extinct species, some in well lithified
deposit (Merrilees. 1968b: 14). Pleistocene?

Antiquity of man in the south-west

The association of artefacts with extinct
animals in a well-lithified deposit at Wonbema.
a similar probable association at Coolarburloo
Pool, the occurrence of artefacts in a well-
lithified buried soil at Cowaramup Point, arte-
facts at the ‘wheat belt’ extinct lake sites, and
some of the surface occurrences of artefacts
mentioned by Hallam (1972b), all suggest con-
siderable antiquity for the human beings con-
cerned. but no firm age estimates are available.
An age estimate beyond the limit of radiocarbon
dating for the sample concerned from Mammoth
Cave is of uncertain application to specimens
recovered from the cave, and has not yet been
related unequivocally to human presence. Thus
the oldest human occupation of the south-w r est
of the continent reliably established so far is
that represented in Devil’s Lair at about 25.000
years B.P. The bottom of the Devil’s Lair deposit
has not been reached and might reveal occupa-
tion more than 30.000 years ago (Dortch and
Merrilees, 1973).

Acknowledgements. — I am grateful to Mrs. P. Kaill for
preparing Figure 1. to officers of the Western Austra-
lian Department of Agriculture for their advice on
vegetation, and to Mrs. S. J. Hallam for criticisims of
a draft of this note.

Aboriginal Art: Ceremonial and other sites
in southwestern Australia

by W. C. Dix

Registration of Aboriginal sites
In 1960, the Anthropological Society of
Western Australia began to prepare a list of
known or reported Aboriginal sites. The Western
Australian Museum undertook to maintain and
expand that list, and is continuing to do so. In
1962. a Panel was formed to advise the Govern-
ment on matters relating to sites. The Panel
also prepared a draft of legislation which, after
extensive revision, was eventually proclaimed as
the Aboriginal Heritage Act in December 1972.
This Act states that it is an offence to excavate,
destroy, damage, conceal or in any way alter an
Aboriginal site, even if it happens to be on
privately owned land.

As part of the effort to protect such sites, I
was appointed in 1970 Registrar of Aboriginal
Sites, with supporting staff, having headquarters
at the Western Australian Museum. Since then,
we have embarked on a programme of verifying
reports and have been exploring various methods
of physically protecting the sites. In many cases,
Reserves and ‘Protected Areas’ centring on
particular places have been established.

Aboriginal sites

Some of the more significant sites listed in our
index are noted on the map (Figure 1), and an
indication is given below of the nature of the
sites. Few have been the subject of detailed
published reports. Figure 1 does not show sites
from which artefacts have been collected from
surface scatter unless these have been published.
Locations of stores for sacred ritual objects, and
sites reported in confidence, are not indicated.
A number of sites which have been excavated
are discussed above by D. Merrilees.

a Sacred or Ceremonial Sites: Sites now in disuse
are located (in Figure 1) at 16. 18. 33. The only cur-
rently significant site shown is the complex at Weebo,
14. Focused on a banded siltstone, now called ‘Weebo
Stone', the site has attracted considerable publicity in
recent years, and the location is generally well known.

b Stone Arrangement (not including stone struc-
tures): Authenticated arrangements are located at 4.
6. 8. 13, 15. 17. 20. 21. 27. 28. 30. 31. 32. 34, 38, 41. 43.
46. 58. 59. 63. 70, 74, 75, 76. 77.

c Hunting Devices: Many sites may have had devices
for hunting, and these include some stone structures,
pits. etc. The purpose of some is quite obvious, and
some are known through an information link with
early settlers. Fish traps are located at 54, 72. 73. Large
animal traps at 71, and small animal or reptile traps
at 59, 75.

d Rock Engravings, etc.: Very little engraving is in
evidence, probably due to the lack of suitable material,
bv comparison with the north-west. Some known sites
are at 1. 2. 3. 10. 49. 64. 78.

e Paintings: 2. 3. 4. 8. 11. 12. 19. 24. 25. 26. 40.
42. 44. 45. 60. 61. 62. 78.

/ Quarries: The well known ochre quarry at Wilgie
Mia. 5. has an extensive archaeological deposit. Other
ochre quarries are at 16. 29. 39. Significant quarries for
stone are known at 7. 22.

g Surjace Scatter of Artefacts: Examples include
those located at 5. 8. 23. 36. 37. 50. 51. 52. 53, 55.

References : Some of the sites shown on Figure 1.
have been referred to. and in some cases described,
in the following references:

Butler (1958)

Campbell (1914)

Cawthorn (1963)

Crawford (1963)

David: on (1936)

Davidson (1952)

Davies (1961)

Glauert (1952)

Gould (1968)

Gould (1969)

Glover and Cockbain (1971
Hallam ( 1971a. b.)

Hallam ( 1972a. b.c.)

Hallam (1973)

Le Soeuf (1907)

McCarthy (1962)

Serventy (1952)

Serventy and White (1958)
Uren (1940)

Woodward (1914)

50. 51. 52. 53. 55
24. 25. 26
8

5. 61. 62
12. 24. 25. 26

3. 5. 11. 12. 24. 25. 26. 44. 60. 62

I. 2. 10. 13
38

33, 34
34

24. 25. 36. 48. 50. 55. 69

59. 49

57. 64. 49

51

60

II. 12. 26
57. 60. 61. 62
28

27

5

Ecology and demography in southwestern
Australia

by Sylvia J. Hallam

Pleistocene to Present

Merrilees describes above the earliest decisive
evidence for Aboriginal populations in south-
western Australia, as yielded mainly by excava-
tions, and in particular by those in Devil’s Lair
(Dortch and Merrilees, 1972; 1973). which show
the wide range of activities of Pleistocene Abo-
rigines in the south-west — working bone; work-

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2. July, 1973.

46

ing wood; using steep scrapers, adze-fashion;
hafting flaked tools and bone (not only at a
time similar to Gould’s early hafted material
from Puntutjarpa [Gould, 19711, but also at a
date twice as remote [Dortch and Merrilees,
19731) ; utilizing, quarrying and trading a variety
of raw materials, including an Eocene fossi-
liferous chert similar to the ‘flint’ quarried
during the Pleistocene at Koonalda (and later
from the cliffs along the Nullarbor coast, and
traded hundreds of miles eastward ( Wright,
19711), and also to the chert from surface as-
semblages throughout hundreds of miles of the
west coastal plain (Glover and Cockbain, 1971).
Wilgie Mia attests to later large-scale mining
and trading; while Orchestra Shell Cave and
Frieze Cave provide successive exemplifications
and modifications of artistic, mythic and ritual
traditions in which serpents, fire, and dark
crevices and caves had their part. Stone
arrangements and art sites further illus-
trate that the links between Aboriginal groups
and their terrain were mediated through the
symbolic as well as the economic aspects of their
lore and usages. What other evidence can con-
tribute to analysis of the interacting transfor-
mations of Aboriginal life and land between the
Pleistocene and European contact?

The role of topographic archaeology

The British, and above all the Cambridge,
archaeological tradition has always put stress on
field survey as well as excavation, although
American archaeologists have only recently
realized the importance of ecological and settle-
ment studies. Like Stukeley in the eighteenth
century, O. G. S. Crawford (1953), Fox (1923),
Phillips (1964; 1970), Hoskins (1955), Hallam
(1970), Fowler (1972), etc., show a continuing
concern with distributions and settlement pat-
terns, with the changing reactions and cumula-
tive impact of human groups to and upon
changing regional landscapes — a concern, in
brief, with dynamic ecological systems, in which
the exploitative, technological and symbolic
activities of human societies were major com-
ponents. Field archaeology, or more properly
‘topographic archaeology’ (Clark, 1964), com-
prises the investigation of ‘all traces of former
human activity’ in the landscape. In any Aus-
tralian region this would include, for example,
fords, wells, water sources and their surround
of much-burnt ground; yam-diggings, pit-traps,
fish- weirs; tracks; belts and nodes of used and
fired countryside, open parklike grazing and
dense secondary regrowth; artefact-scatters
indicating ephemeral stopovers of small hunting
groups; the denser artefact-concentrations on
camping-spots frequented occasionally, often
seasonally or semi-permanently, by smaller or
larger aggregates of folk over various timespans
and for various purposes (fishing, fowling, taking
frogs, turtles, lizards, small mammals, or hunting
larger herbivores, digging various roots), leaving
the debris of weapon manufacture, woodwork-
ing, grinding, etc.; bare sand areas cleared in
getting fuel and shelter and kept clear by con-
stant usage; devegetated areas, cleared by fire.

sand-blows and mobile dunes — e.g. (see Fig-
ure 1) Williams Bay, 71a (Bermingham et al.
1971); Scott River, 69 (Butler, 1969); Moore
River, 37 (Hallam, 1972a); as well as sites of
art and ritual which patterned Aboriginal ac-
tivities.

Field surveys may fall into one of several
possible categories (Green, 1967): (1) Extensive
reconnaissance surveys in relatively little in-
vestigated areas, drawing together scattered
material, published and unpublished, are an
essential preliminary to further investigation
(e.g., the Anthropological Society of Western
Australia, 1960; Crawford, 1963; the Aboriginal
site list prepared by Miss Sarah Meagher of
the Western Australian Museum in 1967-8, and
amplified recently by the Registrar of Aboriginal
sites, Mr. W. Dix). (2) Survey in conjunction
with excavation may make clear the range and
context of a certain type of site (c/. those
around the Pleistocene lakes of N.S.W. [Bowler
1971; Bowler et al. 1972; Barbetti and Allen
19721). (3) Problem-oriented surveys concen-

trate on one particular type of site over a wide
area (e.g., Dix’s study of stone arrangements
(see above 1; the art studies of Davidson [1952],
Crawford, Bruce Wright, McCaskill, etc.). (4)
Intensive surveys ‘designed to extract all possible
information from each site found’ and recogniz-
ing the necessity for ‘as complete a random
sample as possible’ must necessarily be relatively
localized, but should preferably be sufficiently
extensive to include the normal range of a local
group and investigate shifts in the demarcation
of culture areas. Such systematic field studies
have been most effectively adumbrated and
pursued in the east and north of Australia (c/.
McBryde, 1962; 1973, etc.; Lampert, 1971a,b;
Stockton, 1970; 1972; and Dortch, 1972) but

have been relatively lacking in the South-West
(c/., however, Butler, 1958; Akerman, 1969;
Bignell, 1971); and valuable local amateur
studies are in progress of both the archaeological
and the folk evidence of Aboriginal occupation
and of the relationship of early European to
Aboriginal settlement ( e.g . by Gardner near
Northcliffe, and Mrs. Roe in the Gingin-Moore
River area).

Topographic studies in southwestern
Australia

Dr. John Glover* investigated artefact-scatters
on the coastal plain in order to define the dis-
tributions and proportions of a lithologically-
peculiar, exotic raw material (see above), which
he found to occur in exposures (e.g. in Figure
1, near 24-5, 36, 37, 48, a concentration of sites
around 53-5, and 69) in dune blows mostly over-
looking interdunal swamps and lakes (Glover
and Cockbain, 1971). No site east of the Darling
Fault has yielded more than a few flakes of
this Bryozoan chert. It is most abundant, on
the whole, on sites near the west of the coastal
plain. This suggests a derivation for the Perth
Basin material, perhaps from sources now off-
shore since the rise in sea-level which ended

♦Department of Geology, University of Western Australia.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

47

around 3000 BC [Churchill, 1959; I960]

in caves or river channels penetrating through
Pleistocene deposits into the underlying Eocene.
Churchill (1959; 1968) discusses evidence that
the Swan River cut a channel down to around
50 m below present mean sea-level, and was still
at around — 20 m by 7000 BC. Assemblages
composed almost exclusively of this fossiliferous
chert are also typologically distinctive, com-
prising flakes, chips, flaked pieces, and a great
variety of scrapers — nosed, concave and convex,
including many with a steep edge angle and
adze-type utilization ( e.g . 37; Hallam 1972a).
At 37, the fact that there are only two backed
blades in the group of artefact scatters suggests
that this assemblage, and the similar component
in other coastal plain assemblages, is early (i.e.
prior to the general floruit of backed-blade
assemblages from around 2000 BC \cf. Gould,
1971; Hallam, 1972a,bl).

European glass, sherds, clay pipes, etc., on
some sites show that a few continued to be
occupied from the early (‘pre-blade’) phase,
through into the middle (‘backed-blade’) and
late (‘post-blade’) phases, and a very few into
the final brief phase after European contact.
The analysis of such multi-phase occupations is
aided by study of one-phase sites: for example,
the early assemblages described above, or the
many relatively amorphous late assemblages
with a high proportion of quartz chips. Differ-
ences due to date must be distinguished from
those due to group size, span and frequency of
occupation, or exploitative function, as indicated
by extent, density of material, presence or ab-
sence of grinding equipment; and from regional
differences in facies between sites of the same
phase. Each assemblage can thus be assigned
to a phase or series of phases; and the distribu-
tion of numbers and sizes of sites in relation to
ecological zones can be analyzed for each phase.
Changes over time indicate changes in overall
population density, patterning into splitting and
aggregating groups, and stress on the different
resources which ethno-historical evidence shows
to characterize the different zones.

The Perth area project

The Australian Institute of Aboriginal Studies
and the University of Western Australia are
supporting a survey comprising a transect from
the coast eastward to beyond the Avon, centred
on the Swan estuary (Hallam, 1971b; 1972a, b,c;
1973). It is too soon to presage full results, but
certain trends are suggested by a pilot survey of
about 120 sites clustered into about 70 groups
and over 10,000 artefacts.

Increasing numbers of sites per unit timespan
on the coastal plain show increasing Aboriginal
usage and populations. The ‘early’ coastal
groupings using Eocene chert were confined to
the (then wider) plain west of the Darling
Scarp; later and ‘contact’ groups, on the
evidence both of material distributions and of
early European observers, ranged farther east
into the good grazing land along the scarp foot
and into the Darling Range; with intensive late

usage of the resources of interdunal lakes and
swamps.

By contrast, the typologically early (c/. Lam-
pert, 1971a, b) chopper/steep scraper assemblages
of the Avon Valley and eastward (e.g. 10 sites
in the Grass Valley-Quellington area, — c.55) are
richer in numbers, amount of material, and
variety of types, relative to later assemblages,
than on the coastal plain. They differ in
material (largely doloritic) and typology (with
their many high-backed ‘core’ scrapers) from
the coastal assemblages, though sharing steep
edge and in-biting use wear; and extend west
to the foot of the Darling Scarp (e.g. 48a, 50).
The line of demarcation between coastal and
inland-oriented groups would seem to have
shifted eastward with the rise of the sea level,
by about 3000-2000 BC. There are relatively
fewer and sparser ‘middle’ and ‘late’ assemblages
inland. I have suggested (1972b; 1973) that the
less wooded inland areas offered initially more
abundant grazing, supporting high animal and
early human populations; that firing will have
improved the openness and grazing potential of
the ‘open woodland’ on the inland margin of
the forest zone, and also on its coastal fringe.
During the drier phase which ensued from be-
fore 2000 BC (Churchill, 1968; Gould, 1971),
the inland zones proved incapable of supporting
as steep a rise of population as the coast, being
more sensitive to aridity and possibly also to
increased salinity as a result of deforestation.
Dr. C. A. Parker* (personal communication)
suggests that depletion of nitrogen could be
another effect of burning. Burning would, how-
ever, continue to develop rather than deplete
the grazing resources of the better-watered
alluvial piedmont zone, which Stirling was to
find open and park-like in 1829; while the more
varied fish, fowl, reptile, small mammal, root
and water resources of the estuaries, inlets,
lakes, and swamps of the coastal plain continued
to support steeply increased usage and popula-
tions right up to European contact.

Any regional survey must take into account
the archaeological and ethno-historical evidence
of the symbolic (artistic-mythic-ritual) as well
as the technological and economic aspects of the
knowledge and skills, the cognitive lore, by which
Aboriginal groups grasped and developed the
potentialities of their home terrain, creating an
ecological system which alien newcomers were
to take over (Hallam, 1972c; 1973). Immediately
following, Douglas discusses language, which
was the mould for these cognitive patterns, and
Professor Berndt discusses Aboriginal socio-
cultural patterns in traditional terms and under
European impact.

The language of southwestern Australia

by W. H. Douglas

Introduction to the ‘ Nyungar ’ language

Variant manifestations of the South-West
language have been referred to in the historical

♦Department of Soil Science and Plant Nutrition,
University of Western Australia.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2, July. 1973.

48

records. Several of the variants are still known
today by older speakers of what may be named
generically ‘Nyungar*, which is also the word
for ‘man’ in the language of the south-west.
G. O’Grady (in O’Grady and Voegelin, 1966:
especially p. 130 ff.) has listed four of these
dialects as Wadjuk, Balardong, Wardandi, and
Minang and indicates that the principle sources
of information on these variants are Grey
(1840), Moore (1842), Salvado (1850; 1886), and
Bates (1914) among others (see Oates and Oates,
1970). Later investigators include O’Grady and
Hale in 1958 and 1960. A description of the
language as it is spoken today was prepared by
Douglas in 1968.

Other known variants of the language include
Bibbulman, Kaniyang, Mirnong, Tjapanmay
and Kwetjman, all of which names were spelt
in diverse ways in the older records (see map
in Figure 3). This Nyungar group (Douglas
1968: part 1) contrasted in vocabulary and
grammatical structure with Watjari in the north
(the boundary today being roughly marked by
the Geraldton to Mt. Magnet railway line) ; with
the Western Desert language to the north-east;
and with Ngatju or Marlpa in the east (see
von Brandenstein 1970). Descendants of the
original speakers of the South-Western language
number approximately 8,000 today. Many of
these, however, have a very restricted knowledge
of genuine Nyungar, but speak what the author
has labelled ‘Neo-Nyungar’ (Douglas 1968: 8),
a hybrid speech which contains vocabulary items
and idiomatic constructions from the original
language placed into English grammatical con-
structions.

The phonological dimension

Orthography : Investigators in the past have
used a variety of spelling devices to record this
previously unwritten language. In the present
description the following symbols, arranged ac-
cording to point and manner of articulation,
have been used:

Consonants

Vowels

b dj d /d g

i u

m nj n /n rj

ij i /i

e o

r /r

w y

a

A stroke / preceding a letter indicates retroflexion.

The consonant phonemes: There is a series
of voiceless unaspirated stops occurring at the
labial, dental, alveolar, cerebral and velar points
of articulation. The cerebral is retroflexed. A
series of nasals occurs at the same five points
of articulation. There are three contrastive

laterals — dental, alveolar and cerebral. The two
central continuants are the alveolar, which is
allophonically flapped and trilled; and the
cerebral, which is retroflexed. The semi-vowels
are the labio-velar /w/ and the alveolar /y/.

The vowel phonemes: The language at present
has a five-vowel system. There is contrast be-
tween front and back with the high and mid
phonemes, the front vowels being unrounded,
the back rounded. The low vowel is central
and open.

Prosodic features: Syllable stress is predictable
and non-phonemic. Primary stress occurs on
the first syllable of multi-syllabic words. E.g.,
gudjal ‘two 5 , nindalinj ‘scorpion’. In reduplicated
forms there is equal stress on the first syllable
of each component stem: e.g., gudjal- gudjal
‘four’.

The general syllable structure is CV or CVC
(with C representing ‘consonant’ and V ‘vowel’) :
e.g., 7)Ur)U/lala ‘black cockatoo’, dadj ‘meat’. V and
VC may occur initially, followed by CV or CVC:
e.g., aliwa ‘beware’, idjinj ‘to lay (an egg)’.
There is also a CVNC syllable in which N
represents a nasal which assimilates to the same
point of articulation as the final consonant:
e.g., t) arjg ‘mother’/‘sun’, rju/n/d ' chest body
part)’.

The grammatical dimension

Nyungar grammatical structure is simple when
compared with northern members of the Pama-
Nyungan family, in which there are highly
complex verbal systems.

The verb does not carry tense, but has the
choice of two aspects — ‘completed action’ and
Uncompleted action’. Tense is indicated by the
use of a series of time words, phrases or de-
pendent clauses, when required.

The noun and pronoun rarely take more than
one suffix at a time. Transitive subject marker
and direct object marker are used only to avoid
ambiguity or to supply a subtle emphasis. For
example:

yog-il mam baminj

woman-subject male hitting

— ‘The woman is hitting a man.’

(Emphasis on Subject.)

mam yog-inj baminj

male woman-object hitting

— ‘The man is hitting the woman.’

(Emphasis on Object.)

yog gudjal geb ba/raninj

woman two water fetching

— ‘The two women are fetching water.’

(Subject apparent.)

Other inflexional suffixes include the ‘locative’
-ag (as in but nag ‘on the tree/log’); ‘instrument’
-ag/-al (as in bal bu/n godjag baminj. ‘He (was)
hitting the tree with an axe.’ [godj ‘axe’]).
‘Reason’ is indicated by - a (as in bal ga/rarj
njuna-r). ‘He is angry because of you.’ Injun
‘you’]).

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

49

The basic clause level constructions are the
transitive and intransitive command and state-
ment type clauses and the equational (including
the stative) clause type. Expansions of these
clauses include the addition of vocative, time,
manner, instrument, reason (cause, purpose)
and indirect object or benefactive which may
be manifested as single words, relator axis
phrases or as dependent clauses.

The lexico- semantic dimension
Words: Vocabulary range in old Nyungar
could have been quite extensive. Although
material culture was not highly developed,
knowledge of nature — especially of edible and
non-edible plants, animals and fish — was rich
in detail. Hunting terminology was also ex-
tensive, as were the vocabularies in connection
with religious culture, social organization and
law.

Phrase words: Body parts feature largely in
certain noun phrases which have become des-
criptive idioms. For example:

gad wa/ra (lit. ‘head-bad’) ‘stupid’.
dwa-rjg bu/d (lit. ‘ear-less’) ‘unreasonable’,
‘ignorant’.

ma/da gidj (lit. ‘leg-spear’) ‘bony-legged’,
‘skinny’ ( derogatory ) .

gobu/l wi/d (lit. ‘stomach-empty’) ‘hungry’.
Clause words: Many metaphors, similes, idioms
and other figures of speech have survived the
clash with the alien language and some of them,
as has been mentioned, have found their way
into the English of the South-West. Among
these are such expressions as:

ba/da-rjinj yo-qga muginj — ‘hopping like a
kangaroo’ (used in various contexts).
rjjidinj gwiya/r muginj — ‘cold as a frog’ (also
used in various contexts).
ge/d-ge/d gu/linj (lit. ‘running swiftly’) —
‘darting here and there’, ‘purposeless’.
geba rjaninj (lit. ‘water-drinking’) — now
used for ‘liquor drinking’.

Discourse analysis of any exhaustive nature
has been impossible because of the limited
amount of text-material now available, but
traditional narratives which have been collected
show the Nyungar ability to use all the subtleties
of story-telling. Dramatic presentation, mystery
and humour are not lacking; nor are cleverness
of characterization and development of plot.

Acknowledgement . — We acknowledge the assistance of
Dr. Susan Kaldor of the Department of Anthro-
pology, University of Western Australia.

Aborigines of southwestern Australia: The
past and the present

by Ronald M. Berndt

Pre-European population
For information on traditional Aboriginal life
in the south-west of this State, we must rely
on early records which are not detailed and
are anthropologically unsatisfactory. The pre-
European population could have approximated
6,000 persons — if we take the boundary as a

line drawn diagonally from Esperance to Mul-
lewa. This boundary marks off the non-circum-
cising zone (see Map on Figure 2), that is, the
southwestern corner of the State. If, however,
we include adjacent tribes farther inland (within
the circumcising area), the population could not
have been much less than 7,500. Moore (1884:
115) reported that in 1840 there were about
3,000 Aborigines in the Swan River Colony alone.
Radcliffe-Brown’s (1930a) estimate of 12,500
Aborigines in the south-west at the time of
first settlement is possibly too high (see Makin,
1970: Chapter 5).

The southwestern people felt the full force of
the disastrous impact of European settlement.
Against that, their traditional life could not
survive — not as a living, coherent entity.

Traditional social organization

The ‘true’ south-west Aborigines, then, did
not practise circumcision. However, a merging
of social units occurred quite early during the
contact period, when members of different tribal
groups were obliged to live on mixed-tribal set-
tlements. The map on Figure 2 shows the posi-
tioning of 13 ‘tribal’ divisions (after Tindale,
1940) south-west of the circumcision boundary.
In 1967, Douglas (1968) located 11 of these;
5, and perhaps 6, of them fall within the cir-
cumcisional area: see map on Figure 3. Tindale’s
list depends on old sources, Douglas’s on the
memory of elderly persons of Aboriginal des-
cent. Douglas’s interest is in language, and these
‘tribal’ names are really labels and/or dialectal
divisions which in toto can be classified today
as Njungar, or Nyungar, a word meaning ‘man’
or ‘person’.

The internal organization of south-west tribal
units was quite diverse. According to Radcliffe-
Brown (1930b: 216-19, 220-21, on the basis of
Moore, 1842; Salvado, 1851 and 1886; Bates,
1914 and 1923; among others), there were four
patterns for this relatively small region (see
map on Figure 4). Area ‘C’ had matrilineal
moieties named manitjmat and wardangmat
(‘white cockatoo’ and ‘crow’), with at least four
cxogamous matrilineal divisions (or clans)
grouped under each moiety. Their names had
‘totemic’ associations. Moore (1842: 4) speaks
of them as ‘family names’: four principal ones,
‘resolved again into many local or subdenomina-
tions’, several grouped under ‘one leg’, others
under another ‘leg’ and so on. However, this
requires further discussion. Ritual affiliation
was through the father. The pattern as des-
cribed by Bates (1923) suggests local patrilineal
descent group centring on totemic sites, with
mythic connections and correlated with specific
stretches of country. Thus, a person belonged
to the moiety and totemic clan of his (or her)
mother, but also to the local group of his (or
her) father. Within the father’s land division,
a person’s conception (or birth) totem, a par-
ticular natural species, was mythically defined
vis-a-vis a centre which was, in turn, the focus
of ritual. Area ‘B’ had a similar social organiza-
tion, except that the named moieties were patri-

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

50

/

/

lineal. Area ‘D’ had two alternating ‘endoga-
mous moieties’ named birangumat and djuamat
(‘kingfisher’ and ‘bee-eater’). These, however,
seem to be alternating generation levels similar
to the traditional Western Desert type (see R.
and C. Berndt, 1964/68: 56-8). The main focus
was on patrilineal local descent units. Area ‘A’
seems close to Area ‘D’, with the addition of
named totemic groups (probably patri-local
descent units).

Generally, the south-west Aborigines were
deeply attached to their country through mytho-
ritual ties. As in other parts of Aboriginal Aus-
tralia, the local descent group was concerned
with religious matters, while the socio-economic
unit (a mixed-membership group) moved over
limited stretches of territory, hunting and food-
collecting. Communication was kept open be-
tween members of different territories or dis-
tricts (subtribal or otherwise). These took
structural shape in the mandjar or ‘fair’, when
Aborigines met to barter a wide range of goods.
People living in the Perth area exchanged com-
modities with those from the Murray River on
the south, and also with others from the north.

Cultural background

Environmental and climatic conditions in
the south-west made it necessary for Abo-
rigines to protect themselves against cold
winters. They constructed bark-covered huts
which Hammond (1933: 25) described as

watertight. Also, they made buka (cloaks),
such as were used in southeastern Aus-
tralia: three or more kangaroo skins, specially
treated, were sewn together with sinew or rush
and worn with the fur side inward. Distinctive
to this area were the kangaroo skin bags used
by women: the goto, used generally, and the
gundir, for carrying a small child. And the
kadjo ‘hammer’ was a unique implement, broad
and blunt at one end and sharp-edged at the
other: it was affixed to a short thick stick by
means of prepared tudibi, Xanthorrlioea gum.

Initiation was fairly simple, and the major
operation involved piercing the nasal septum.
The novice was red-ochred during the rites and
was given a hairstring, cloak and weapons
(Bates, 1923: 236-7). Hammond (1933: 63) men-
tions that, although there was no circumcision
rite in this area, the name of a place near
Albany meant ‘circumcision site’. He reports the

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July. 1973.

51

• Keekatharra

• Cue

PULI NY A *Mt. Magnet

belief that this operation would have spoilt the
natural erection of the penis — but a modified
form of the rite, discontinued in the early 1870’s,
was the removal of the tip of the foreskin,
Salvado (1850) noted that the Aborigines ‘hide
carefully from strangers their customs and, in
particular, their beliefs’. From the evidence, it
would appear that increase rites were held by
local groups, and that a reasonably large body
of mythology existed, some of which was ex-
pressed ritually. Salvado mentioned Motogen
(which he translates as god), as a creative being.
He also referred to myths about the Sun and
Moon, the Morning Star and the great Rainbow
Snake. Moore (1842: 103) wrote of ‘a large

Figure 4.— Social organizational categories, after Rad-
cliffe-Brown, 1930 (map facing p. 42). A— without
moieties and sections, but named patri-totemic units.
B —patrilineal moieties. C.— matrilineal moieties and
totemic clans, with patri-local descent units. D.— named
pairs of sections or alternating generation levels.

Journal of the Royal Society of Western

winged serpent’ called Waugal, living in ‘deep
dark waters’, who was especially inimical to
females but was feared generally as being instru-
mental in bringing sickness. A similar mythic
snake emerged at Mt. Eliza and, crawling its
way to the sea, created the Swan River. Mt.
Eliza and the high ground where the State
Parliament House now stands were called
Ga-ra-katta, the site sacred to this snake (see
Moore, 1884, and Makin, 1970).

The moon (miga) was a man, the sun
( nganga ) a female. A cave at York had a
circular figure cut into the rock face, with hand
stencils: this was said to have been ‘visited’ by
Moon (Moore, 1842: 35). Many stars, too, were
mythic beings (or mythic beings had ‘turned
into’ stars). Bulgut, a wife of Tdadam or
Dedam, was one. Dedam had a sister (of the
same name) whom he speared when she allowed
his two children to stray: they were represented
by stars, and so was the spear. And the star
Dj ingun was a wife of Wurdytch or Wurdoitch.
Julagoling, the planet Venus, was an attractive
young woman who carried out sorcery. Ham-
mond (1933: 64-5) gives a story of a large tree
located east of Northam which was used by
mythic eagles who stole human babies to feed
their young: the tree was eventually burnt down
in revenge.

There was also quite a wide range of dances.
However, their significance has not been noted
by early recorders. The dtowalguorryn, for ex-
ample, was common among people living in the
eastern sector of this area; the yallor came from
the north; the kanggarak from the south; the
yuyltunmitch (direction unspecified) ; and the
yenma from the north and north-east. The
yenma is undoubtedly the well-known Western
Desert inma, a general term for an ordinary
ceremony. The yallor was performed mainly by
men, only occasionally by women. They drama-
tized hunting scenes and the actions of various
animals, birds and reptiles, along with the feats
of sorcerers. Men were elaborately decorated,
wearing ornamental sticks in their hair. It is
not clear whether these were cult totemic, but
the dancers were ‘surrounded by groups of ad-
miring spectators’. Salvado speaks of songs be-
ing ‘handed down with a kind of traditional
veneration’. Hammond (1933: 49-53; 63-5)

refers to inter-tribal ceremonies during which
betrothals were arranged. He gives a diagram
showing how participants and onlookers were
spatially arranged within a large circle; and
he illustrates a serrated-edged bullroarer.

It would appear that women had more to say
in tribal matters than has generally been em-
phasized. The custom of monyo, conferring the
status of moyran (or ‘grandmother’) on a
woman, gave her authority to arbitrate in
quarrels and during armed disputes.

Consistent with the interests of early observers
vis-a-vis ‘savage peoples’, the material on magic
and on death is more extensive. A sorcerer, for
instance, was said to be possessed of boylya
(power) which enabled him to fly through the
sky, consume his victim’s flesh, and use quartz
crystal (magic stone of the shark). He wa s able

Australia. Vol. 56 Parts 1 and 2. July, 1973.

52

to raise or calm the wind, and cause rain to
fall (Moore, 1842: 18-19). Salvado says that
sorcery was under the guardian spirit called
Cienga. The sorcerer was also a native doctor
and was able to cure his patients ( walbyn , to
cure by enchantment) .

Beliefs surrounding death were focused on
releasing the deceased’s spirit or soul (kadjin)
from its physical vehicle, the body. There were
various forms of earth burial: gotyt among the
‘mountain tribes’, dyuar for the ‘lowland tribes’.
Usually, the corpse was buried in its cloak. A
native doctor was present and listened for the
sound of the spirit’s flight from the body. At
that time, it informed him of the name of the
person responsible for its death. A small hut
of reeds or boughs was constructed over the
grave, and a fire lit at its entrance to make the
place more comfortable and home-like for the
spirit. There it was believed to remain until its
death was avenged (see also Nicolay, 1886: 8-10).
Bates (1923: 238-40; 1927?) speaks of the kanya
(soul of the newly dead) going first to the
tabu-ed moojarr or moodurt tree (Nuytsia flori-
bunda or ‘Christmas’ tree), where it rested on
its way to Kurannup, the home of the Bipelmen
dead located beyond the western sea: here, their
old skins were discarded and they appeared
‘white’. The ancestors were spoken of as
netingar, and lived on the island of souls. The
first European settlers, because of their light
skin pigmentation, were believed to have come
from that place: they were called djanga, ‘the
dead’.

Dispersal and the Njungar

The coming of ‘the dead’ meant death to
local Aborigines. The sorry story of Aboriginal-
European contact in Western Australia, especi-
ally in the south-west, between 1829-1897, has
been documented by Sir Paul Hasluck (1942).
Hammond (1933: 67-72), too, spells out the dis-
aster which befell them. At first, the settlers
were afraid of the Aborigines. They drove them
away, destroyed their camps, burnt their huts,
made it difficult for them to obtain indigenous
food, and employed them as menial workers.
Aborigines were threatened, shot at, and in many
cases killed. Their traditional territories were
no longer their own. The ‘battle of Pinjarra’ in
1834 was simply one of the more notorious ex-
amples which have become part of this State’s
history (see Neville, 1936: 10-46). Through the
early years of settlement, relations between
Aborigines and Europeans steadily deteriorated.
Shootings, assaults and theft continued. A
deputation of Aboriginal leaders waited on the
Lieutenant Governor of the time (Captain
Irwin) in an attempt to resolve the situation —
but without success. The establishment of
Rottnest Island prison in 1839, the removal of
children from their parents, and the increasing
‘mixed-blood’ population, all contributed to the
destruction of traditional culture. In the 1850’s,
‘Ticket-of-leave men and parties of convicts in
the bush mixed with the natives, supplied them
with drink, and there were often hideous orgies.

The dispossessed blacks had become paupers and
mendicants and deterioration, already begun,
proceeded apace’ (Neville, 1926: 40).

The measles epidemic of the early 1880’s was
another ‘killer’, and 50 years after settlement
(as Hammond puts it, 1933: 70) ‘the South-
West was left with scarcely a true-blooded
aboriginal in it’. Traditional life had by then
disappeared. By that time, too, in the early
1890’s, Aborigines from other areas were drifting
increasingly into the settled areas and inter-
mixing with the remaining local inhabitants.
An official Western Australian handbook (Hart,
1893: 162-71) claimed that, ‘Native labour being
cheap, the sheep farmer who might otherwise
be unable to work his station is able to do so
with profit’; but, as for Aborigines in the settled
districts, ‘their numbers diminish every year’.
Neville (1926: 46) underlines this point: ‘. . . here
in the South-West . . . between 1829 and 1901
. . . a people estimated to number 13,000 were
reduced to 1,419, of whom 45 percent were half-
caste’.

The result was to leave the entire south-west
with primarily a part-Aboriginal population —
few of them directly descended from the original
local people, most of considerably mixed Aborig-
inal affinity, and all possessing little or nothing
of their traditional heritage. Hasluck (1939) and
Neville (1948: 3-13; 1951: 274-90; 1947), among
others, have provided us with surveys of local
conditions. In 1948, Neville wrote (1948: 5):
‘Years ago I witnessed the passing of the last
of the Bibbulmen [a tribal name which he used
for the whole of the south-west] . The few full-
bloods now in the south-west have come from
outside this district’.

It left, too, the majority of these people of
Aboriginal descent living on settlement reserves,
or occupying squalid camps on fringes of country
towns — under the ‘supervision’ of the then
Western Australian Department of Native Wel-
fare. They worked spasmodically for Europeans
and were, for all general purposes, culturally but
not socially European-Australians. Because a
tribal background was now irrelevant as far as
its content and any of its details were concerned,
they saw themselves as being Njungar — different
from the non-Aborigines around them.

The story of discrimination and prejudice in
relation to these people cannot be discussed here.
However, over the last few years conditions have,
at long last, been changing radically: more op-
portunities are now open to them. In 1967,
Douglas estimated the part-Aboriginal popula-
tion as 8,000 in the larger south-west region, not
bounded by the non-circumcision line. In 1971,
the Central Division (i.e., Perth, Moora and Kel-
lerberrin) had a total population of 5,128. Of
these, 58 were ‘full-bloods’ (from non-south-
west areas), and 2,694 were children under 16
years (31 being ‘full-blood’). The Southern Di-
vision (i.e., Narrogin, Bunbury, Gnowangerup,
Albany) had a total population of 2,855. Of
these, 60 were ‘full-bloods’ (from non-south-
west areas), and 1,692 were children under 16
years (31 being ‘full-blood’). The best recent

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

53

studies of these people are by Makin (1970) and
McKeich (1971). For general discussions, see
Rowley (1970) and Biskup (1972).

Not all the south-west people of Aboriginal
descent would regard themselves as Njungar.
But, Njungar or not, they are New Aborigines —
people who seek a social identity of their own
in contrast to other Australians. And part of
what they seek relates to obtaining some know-
ledge of their traditional past. For the south-
west, unfortunately, only a very little of that
knowledge survives.

References

Akerman, K. (1969).— Walyunga: an Aboriginal site near
Perth, Western Australia, Ilchinkinja 3: 12-
16, figs. 1-62. Western Australian Anthropo-
logical Students’ Association. Perth.
Anthropological Society of Western Australia (1960). —
A preliminary report of a survey being car-
ried out by the Anthropological Society of
Western Australia relevant to the preserva-
tion of Australian Aboriginal sites in this
state. Roneoed, Perth.

Barbetti, M. and H. Allen (1972). — Prehistoric man at
Lake Mungo by 32,000 years B.P., Nature
240: 46-8.

Bates, D. M. (1914). — Social Organization of some West-
ern Australian Tribes. Report of the Aus-
tralian Association for the Advancement of
Science, Melbourne. XIV : 387-400.

(1923). — Tribus du Sud-Ouest de l’Aus-

tralie, Revue d’Ethnographie. IV : 225-40.
(1927?). — The Tree of Souls, West Austra-
lian.

Bermingham, A, D. R. Packham and R. G. Vines
( 1971) . — Search 2: 434-5.

Berndt, R. M. ed. (1969). — “Thinking About Australian
Aboriginal Welfare, with particular refer-
ence to Western Australia.” Department of
Anthropology, University of Western Aus-
t ra n a Pert;'

Berndt, R. M. and C. H. Berndt (1964/1968). — “The
World of the First Australians” . Ure Smith,
Sydney.

Bettenay, E. (1962). — The salt lake systems and their
associated aeolian features in thq. semi-arid
regions of Western Australia, Journal of
Soil Science 13: 10-17.

Bignell, M. (1971) .—“First the Spring. A history of the
shire of Kcfjonup, Western Australia”. Uni-
versity of Western Australia Press, Perth.
Biskup, P. ( 1972) . — “Not Slaves, not citizens. The Abori-
ginal Problem in Western Australia, 1898-
1954”. University of Queensland Press, St.
Lucia. ,

Bowler J. M. (1971). — Pleistocene salinities and climatic
change: evidence of lakes and lunettes in
Southeastern Australia. In “Aboriginal Man
and Environment in Australia” (D. J. Mul-
vaney and J. Golson, eds.). Australian
National University Press, Canberra.

Bowler, J. M., A. G. Thorne and H. Polach (1972).—
Pleistocene Man in Australia: age and signi-
ficance of the Mungo skeleton, Nature 240:

Brandenstein 6 "c. G. von (1970). —Linguistic Salvage Work
on the Eastern Goldfields and the South
Coast of Western Australia, Bulletin of the
International Committee on Urgent Anthro-
pological and Ethnological Research, No. 12.
Vienna. , . . .

Butler, W. H. (1958).— Some previously unrecorded arte-
fact sites near Perth, Western Australia, The
Western Australian Naturalist 6: 133-b.
(1969).— Remains of Sarcophilus the Tas-
manian” devil (Marsupialia, Dasyundae)
from coastal dunes south of Scott River,
Western Australia, The Western Australian
Naturalist 11 : 87-88.

Campbell, W. D. (1914).— Description of a rock shelter
with Aboriginal markings, and an indurated
cast of a footprint at the Greenough River,
and a cave dwelling at Narandagy Spring,
Lockier River, The Journal of the Natural
History and Science Society of Western Aus-
tralia 5: 9-14.

Cawthorn, P. (1963).— An Aboriginal Site at Wiluna,
The Western Australian Naturalist 8: 151-2.

Churchill, D. M. (1959).— Late Quaternary eustatic
changes in the Swan River district, Journal
and Proceedings of the Royal Society of
Western Australia 42: .53-5.

(1960). — Late Quaternary changes in the

vegetation on Rottnest Island, The Western
Australian Naturalist 7: 160-6.

(1968). — The distribution and prehistory of

Eucalyptus diversicolor . ... E. marginata
.... and E. calophylla ... in relation to
rainfall, The Australian Journal of Botany
16: 125-51.

Clark, J. G. D. (1964). — Review of Phillips 1964. Pro-
ceedings of the Prehistoric Society 30: 435.

Commissioner of Native Welfare (1971). — Annual Report
for the year ended June 30, 1971. Govern-
ment Printer, Perth.

Crawford, I. M. ( 1963) .—Aboriginal sites in Western
Australia. In “National Parks and Nature
Reserves in Western Australia” . West Aus-
tralian Subcommittee of the Australian
Academy of Science Committees. Perth.

Crawford, O. G. S. ( 1953) .— ' “Archaeology in the Field”.
Phoenix House, London.

Davidson, D. S. (1936).— Aboriginal Australian and Tas-
manian rock carvings and paintings,
Memoirs of the American Philosophical
Society 5: 1-151.

(1952).— Notes on the Pictographs and Petro-

glyphs of Western Australia and a discus-
sion of their affinities with appearances else-
where on the continuent, Proceedings of the
American Philosophical Society 96: 76-117.

Davies S. J. (1961).— Some Aboriginal Sites in the
Murchison District of Western Australia, The
Western Australian Naturalist 8: 1-5.

Dortch, C. E. (1972).— Archaeological work in the Ord
Reservoir area, East Kimberley, Australian
Institute of Aboriginal Studies Newsletter
3 (4): 13-18.

Dortch, C. E. and D. Merrilees (1972).— A salvage exca-
vation in Devil’s Lair, Western Australia,
Journal of the Royal Society of Western
Australia 54: 103-13.

(1973). — Human occupation of Devil’s Lair,

Western Australia, during the Pleistocene,
Archaeology and Physical Anthropology in
Oceania 8: in press.

Douglas, A M., G. W. Kendrick and D. Merrilees
(1966).— A fossil bone deposit near Perth,
Western Australia, interpreted as a carni-
vore’s den after feeding tests on living
Sarcophilus (Marsupialia, Dasyuridae), Jour-
nal of the Royal Society of Western Austra-
lia 49: 88-90.

Douglas, W. H. ( 1968) . — “The Aboriginal languages of
the South-West of Australia”. Australian
Aboriginal Studies No. 14, Linguistic Series
No. 4. Australian Institute of Aboriginal
Studies, Canberra.

Fowler P. J. ed. ( 1972).—' “Archaeology 'ind the Land-
’ scape. Essays for L. V. Grinsell”. Baker,
London.

Fox C (1923).— “The Archaeology of the Cambridge
Region: a Topographical Study ...” (Re-
issued with Appendix IV, 1948.) University
Press, Cambridge.

Glauert, L. (1952).— An Aboriginal relic near Pithara,
The Western Australian Naturalist 3: 143-4.

Glover J. E. and A. . Cockbain (1971) .—Transported
Aboriginal artefact material, Perth Basin,
Western Australia, Nature 234: 545-6.

Gould R. A. (1969). — “Yiwara. Foragers of the Austra-
lian Desert”. Collins, London/Sydney.

(1971).— The archaeologist as ethnographer: a

case from the Western Desert of Austra-
lia, World Arcnaeology 3: (2): 143-77.

Gould, R. A. and E. B Gould (1968).— Kunturu, an
Aboriginal sacied site on Lake Moore, West-
ern Australia. American Museum Novitates
2327.

Green, R. C. (1967).— Settlement patterns: four case
studies from Polynesia. In “Archaeology at
the Eleventh Pacific Congress” (W. G. Sol-
heim, ed.): 101-32.

Grey, G. (1840). — “ A Vocabulary of the dialects of
South-Western Australia”. London.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

54

Hallam, S. J. (1970). — Settlement round the Wash. In
"The Fenland of Roman Times: Studies of
a major area of peasant colonisation with
a Gazetteer covering all known sites and
finds".. (By P. Salway, S. J. Hallam and J.
I'A. Bromich. et al.\ C. W. Phillips, ed.)
Royal Geographical Society Research Series:
No. 5; 22-113, maps and Gazetteer.

(1971a). — Research in Anthropology in West-
ern Australia: origins and prehistory, West-
erly: 45-9.

( 1971b).— Roof markings in the “Orchestra

Shell" Cave, near Perth, Western Austra-
lia, Mankind 8: 90-103.

*( 1972a). —An archaeological survey project,

the Perth area. Western Australia. Report
No. 4. Mimeographed report to the Austra-
lian Institute of Aboriginal Studies.

(1972b).— An archaeological survey of the

Perth area, Western Australia: a progress
report on art and artefacts, dates and
demography. Australian Institute of Abori-
ginal Studies newsletter 3 (5): 11-19.

( 1972c).— Comment on A. Marshack “Cog-
nitive aspects of Upper Palaeolithic engrav-
ing”, Current Anthropology 13: 464-5.

(1973).—' “Man and Landscape in Western

Australia: a discussion of the role of fire
in the ecology of Aboriginal and' European
usage." Australian Institute of Aboriginal
Studies, Canberra. Forthcoming.

Hammond, J. E. (1933) .— 1 “ Win fan’s People. The story
of the South-West Australian Aborigines".
Imperial Printing Co., Perth.

Hart, F. (1893). — “Western Australia in 1893". Govern-
ment Printer, Perth.

Hasluck, A. ( 1961). — Yagan the Patriot, Proceedings of
the Western Australian Historical Society
V, Part VII: 33-48.

Hasluck, P. (1939). — "Our Southern Half-Caste Natives
and their Conditions” . Perth.

(1942). — “ Black Australians. A survey of na-
tive policy in Western Australia, 1829-1897".
Melbourne University Press, Melbourne.

Hoskins, W. G. (1955). — "The Making of the English
Landscape” . Hodder and Stoughton, London.

Lampert, R. J. (1971a). — Coastal Aborigines of South-
eastern Australia. In "Aboriginal Man and
Environment in Australia” (D. J. Mulvaney
and J. Golson, eds.). Australian National
University Press, Canberra.

(1971b).— 1 “But$$U Lake and Currarong:

coastal sites in New South Wales”. Terra
Australis 1. Department of Prehistory, Aus-
tralian National University, Canberra.

Le Souef, W. H. D. (1907 ).—"Wild Life in Australia”.
Whitcombe and Tombs, Christchurch.

Lundelius, E. L. (I960).— Post Pleistocene faunal succes-
sion in Western Australia and its climatic
interpretation, Report of the International
Geological Congress XXI Session, Norden
Part IV: 142-53.

Makin, C. F. ( 1970) . — "A Socio-Economic Anthropologi-
cal survey of people of Aboriginal descent
in the metropolitan region of Perth, W.A.”
Ph.D. thesis in Anthropology, University of
Western Australia.

Mathews, R. H. ( 1904) .— Etnnological Notes on the Abori-
ginal Tribes of Western Australia. Queens-
land Geographical Journal 1 : 45-72

McBryde, I. ( 1964).— Archaeological field survey work
in Northern New South Wales, Oceania 33:
12-17.

(1973). — “ Aboriginal prehistory in Neiv

England. An archaeological survey of north-
eastern New South Wales". University of
Sydney Press, Sydney.

McCarthy, F. D. ( 1962) ■ "Australian Aboriginal Rock
Art”. 2nd Ed. Australian Museum, Sydney.

McCarthy, F. D. (1970) .—“Aboriginal antiquities in Aus-
tralia. Their nature and preservation". Aus-
tralian Aboriginal Studies No. 22. Australian
Institute of Aboriginal Studies, Canberra.

McKeich, R. ( 1971) .—Problems of Part-Aboriginal Edu-
cation with special reference to the South-
West region of Western Australia. Ph.D.
thesis in Anthropology, University of West-
ern Australia.

Merrilees, D. (1968a). — Fossil bandicoots (Marsupialia,
Peramelidae) from Mammoth Cave, Western
Australia, and their climatic implications,
Journal of the Royal Society of Western
Australia 50: 121-8.

Merrilees, D. (1968b). — Man the destroyer: late Quatern-
ary changes .in the Australian marsupial
fauna, Journal of the Royal Society of West-
ern Australia 51: 1-24.

Moore, G. F. (18^2). — "A descriptive Vocabulary of the
language in common use amongst the Abori-
gines of Western Australia ..." Orr, London.

(1884). — “Diary of Ten Years Eventful Life

of an Early Settler in Western Australia ” .
Walbrook, London.

Neville, A. O. (1936). — Relations between Settlers and
Aborigines in Western Australia, The West-
ern Australian Historical Society II, part
XIX: 10-46.

(1947). — “Australia’s Coloured Minority”.

Currawong Publishing Co., Sydney.

(1948). — Contributory causes of Aboriginal

depopulation in Western Australia, Mankind
4 (1) : 3-13.

(1951). — The Half-Caste in Australia, Man-
kind 4: (7): 274-90.

Nicolay, C. G. (1886). — Notes on the Aborigines of West-
ern Australia, Colonial and Indian Exhibi-
tion. London: Clowes: 3-17.

Nind. I. S. (1832). — Description" of the Natives of King
George’s Sound (Swan River Colony) and
adjoining country, Journal of the Royal
Geographical Society of London I.

Oates, W. J. and L. F. Oates, eds. (1970). — “A Revised
Linguistic Survey of Australia” . Australian
Aboriginal Studies No. 33. Australian
Institute of Aboriginal Studies, Canberra.

O’Grady, G. N.. C. F. and F. M. Voegelin (1966).—
Languages of the World: Indo-Pacifie Fas-
cicle Six. In Anthropological Linguistics 8
( 2 ).

Phillips, C. W. (1963). — "Field Archaeology. Some notes
for beginners issued by the Ordnance 0 Sur-
vey". 4th Ed. Her Majesty’s Stationery Office,
London.

Radcliff e-Brown, A. R. (1930a). — In Official Year Book of
the Commonwealth of Australia , No. 23: 687-
96. Government Printer, Melbourne.

(1930b). — The Social Organization of Aus-
tralian Tribes, Oceania, I (1): 34-63; I (2):
206-46.

Roe, R. (1971). — Trial excavation in a small cave, Gingin,
The Western Australian Naturalist II: 183-4.

Rowley, C. D. (1970). — “ The Destruction of Aboriginal
Society. Aboriginal Policy and Practice”,
Vol. 1. Australian National University Press,
Canberra.

Salvado, R. (1850). — “Historia del origin, padecimientos,
progresos y porvenir de las Misiones Catolicas
de Nueva Holanda fundadas y sostenidas” .
Barcelona.

(1886). — In E. M. Curr, “The Australian

Race”. Vol. I: 318-21. Government Printer,
Melbourne.

Serventy, V. N. (1952). — Cave paintings near York and
Hyden, The Western Australian Naturalist 3:
121-30.

Serventy, V. N. and S. R. White (1958).— Stone arrange-
ments at Canna, Western Australia, The
Western Australian Naturalist 6: 85-92.

Stockton, E. D. (1970). — An archaeological survey of the
Blue Mountains, Mankind 7: 295-301.

— (1972). — A central coast survey, Australian

Institute of Aboriginal Studies Newsletter 3
(5): 20-4.

Tindale, N. B. (1940). — Distribution of Australian Abori-
ginal Tribes: a field Survey, Transactions of
the Royal Society of South Australia 64
(1): 140-231, with fplding map.

Uren, M. (1940). — “Sailormen’s Ghosts”, Robertson and
Mullens, Melbourne.

Woodward, H. P. (1914). — A geological reconnaissance
of a portion of the Murchison goldfield,
Geological Survey of Western Australia
Bulletin 57.

Wright, B. J. (1968). — “Rock Art of the Pilbara Region,
North-west Australia” . Occasional Papers in
Aboriginal Studies No. 11. Australian Insti-
tute of Aboriginal Studies, Canberra.

(1969). — Aboriginal occupation sites in the

Murchison River Gorge. Mimeographed re-
port to the Australian Institute of Abori-
ginal Studies.

Wright, R. V. S. (1971). — “Archaeology of the Gallus
Site, ‘ Koonalda Cave”. Australian Aboriginal
Studies No. 26. Australian Institute of Abo-
riginal Studies, Canberra.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

55

6. — European man in southwestern Australia

by G. C. Bolton 1 and D. Hutchison 2

Abstract

European man's impact on the south-west of
Western Australia may be treated in three
periods :

(1) 1829-1850, the period of initial impact. This
was characterized by the carrying over of
British and European preconceptions in the
appraisal of an unfamiliar terrain. Want of
labour, capital, and colonial experience led to
the simplest forms of adaptation of building
and agrarian techniques in the new environ-
ment.

(2) 1850-1890, a period of greater prosperity and
population growth. The task was envisaged
as “subduing" nature and imposing where
possible British models of building, landscaping,
and agricultural adaptation. Lack of means still
led to what could later be considered conserva-
tive and exploitative forms of land use.

(3) 1890-1973, initiated by a period of mineral
discovery stimulating economic takeoff. This
period is marked by increasing urbanization
and metropolitan influence on the hinterland,
and recently by the uneven growth of a greater
official and public recognition of the need for
policies of conservation and regeneration.

Introduction

Historians, economists, creative writers, biolo-
gists, naturalists, and agricultural scientists have
each contributed to our understanding of the
impact of western man on the environment of
Western Australia. Unfortunately each discip-
line has tended to work on its own, sometimes
apparently ignorant of advances made in other
fields, almost always without the full advantages
of an interdisciplinary approach. Of the three
standard histories of Western Australia, Kim-
berly (1897), Battye <1924), and Crowley (I960),
Battye shows the least, and Crowley the most
appreciation of the environmental and economic
factors influencing the spread and character of
settlement. All wrote before the rise of the
current concern with ecology, and none shows
the subtle environmental awareness of a recent
work such as W. K. Hancock, The Monaro
(1972). Consequently an essay in synthesis can
attempt little more than to record the various
contributions made by contemporary observers
and later scholars, and to suggest an immediate
and obvious need for future research.

The era of European contact may be divided
into a period of pioneering, 1829-50, when the
first appreciations and assessments of the ter-
rain took place; a period of consolidation, 1850-
90, following the stimulus of convict transporta-
tion and the spread and diversification of rural
activity; and the modern period, since 1890,
when growth reflected the multiplier effect of the
gold-rushes.

'Murdoch University, Western Australia.

2 Western Australian Museum, Francis Street, Perth.

1829-50: Pioneering

There are as yet no authoritative surveys of
the historical demography of Western Australia,
although a major project is shortly to be in-
itiated by Professor R. T. Appleyard, under
whose supervision current work is in progress at
the Department of Economic History at the
University of Western Australia. Statham is
engaged on a study of the first twenty years of
settlement, as a result of which it may be pos-
sible to test Kimberly’s assertion fop. cit., p.
39) that ‘Substantially, Western Australia had
for its pioneers more highly educated men of
good society than perhaps any other British
dependency’. “Musters” or census returns were
compiled regularly, from which it would appear
that although over 4,000 settlers were attracted
to the Swan River Colony by the too-optimistic
estimates of Captain Stirling and others, only
1,132 remained in 1830 and 2,311 in 1840. Num-
bers improved to 5,886 in 1850, in which year
convict transportation, rejected by other Aus-
tralian colonies, was introduced as a stimulus to
Western Australia’s economy (Gertzel, 1949).
Until then, it may be doubted if the white
inhabitants of the south-west significantly out-
numbered the Aborigines.

The spread of settlement was spasmodic and
uneven. The first comers in 1829 sought land
which was accessible to sea or river transport
and which, so far as possible, met English
criteria for desirable farming country. Along
the Swan River above Guildford a number of
estates still flourishing as vineyards were estab-
lished before 1840. Much of the coastal sand-
plain was either alienated to absentees or un-
suitable because of what was discovered much
later to be deficiency in trace elements. The
colony was too isolated and under-populated to
generate economic growth (Staples, 1961). The
pressure for land led to the early exploration
and settlement of the Avon Valley during the
1830s and the Victoria Plains during the 1840s
< Deacon, 1947; Erickson, 1971). Itinerant shep-
herds shifted their flocks frequently, following
feed and surface water. Meanwhile Albany on
King George’s Sound, since December 1826 the
site of a convict garrison, became after 1831
another nucleus for free settlement. The Bun-
bury and Vasse districts, sparsely settled from
1830, received some stimulus from the ill-fated
‘Wakefieldian’ settlement at Australind in 1840-
42 (Shann, 1926; Staples, 1948). Geraldton,
founded in 1848, became the base for the occu-
pation of the Champion Bay district (Kelly,
1958). These four centres of settlement — Ger-
aldton, Fremantle/Perth, Bunbury, and Albany
— were linked by a coastal shipping service

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

56

whose shipwrecks were frequent. Land trans-
port was inhibited by the tracts of dense timber
and scrub, or of sandy soil, which separated
each nucleus of settlement.

Shortages of capital, labour, and time made
for a modest architecture. An authoritative sur-
vey is in progess for the early colonial period
by Dr John White, Department of Architecture,
University of Western Australia. Cameron
(1968) following Irwin (1835) notes that a num-
oer of the 1829 settlers brought out prefabricated
timber houses, which warped in the heat and
excessively wet season of 1830. Most early houses
were of wattle-and-daub — basketwork plastered
with clay — or “ramjam”, a mixture of sand and
clay rammed into consistency and bolted at
intervals. More substantial buildings were later
constructed of locally-made brick or (in the
Fremantle area) of limestone. Jarrah and
sheoak shingles gradually replaced thatch as a
roofing material (Bunbury, 1930). Verandahs
were not common inland, but were a feature
of Perth and Fremantle houses (Ogle, 1839).
Most public buildings retained a modified
Georgian idiom until the 1850s, sometimes with
unfortunate effects, as with the heavy Grecian
portico of the first Anglican Church, more often
with a simple elegance, such as the Old Court
House of 1836, or the Chapel of the Children
of Mary of 1846 (Oldham, 1961). The cities of
Perth and Australind were both laid out on the
formal, rectangular pattern preferred by early
19th century town planners (cf. Adelaide, Mel-
bourne). Bold (1939) regretted that opportunity
was not taken to lay out the city around the
contours of the many swamps and lakes then
existing north of the Swan, but Roe, the first
Surveyor-General, seems to have considered that
the uneven and dissected terrain presented suf-
ficient difficulties without embarking on town-
planning experiments (J. B. Roe, 1927).

This raises the question of how the first
settlers perceived their environment. Their first
expectations were shaped by European experi-
ence. Cameron (1970) points out that as a naval
man Stirling judged the potential of the Swan
River in terms of trade, strategy, and anchorage:

He found difficulty in translating what he had seen
into meaningful terms, and had resorted to compari-
sons of the Swan River with Virginia and the Plains
of Lombardy. Both these comparisons no doubt raised
entirely different ideas in the minds of his readers
than he intended. Their frame of reference was some-
what different from his.

As the hinterland was penerated some settlers
viewed the new environment with nostalgia for
the past, others with an appreciation of its
differences. Following Dale’s expedition of 1830
to the Avon Valley, Clarkson and Hardey named
the district “Yorkshire” from a fancied resem-
blance to their native county (Deacon, 1947).
On the contrary James Henty, on a separate
expedition about the same time, saw the Aus-
tralian bush as sui generis:

The grandeur of the scene among the valleys sur-
rounded by increasingly tall white gums and the solemn
silence prevailing in the bush, totally unaccompanied
by any signs of civilization, imparts ideas that it is
impossible to reflect on without awe and reverence and

which those who have not experienced it can scarcely
appreciate (Bassett, 1954).

This sort of reaction was none too common. Most
observers, even after a dozen years of settle-
ment, still used the English landscape as a
standard of comparison, while some such as
Wollaston (1948) needed time to realize that
appreciation of the Western Australian bush
required the newcomer to divest himself of pre-
conceptions. He found the trees stunted and
sombre, and most praised those Australian trees
such as the peppermint which reminded him of
European foliage. Some, however, such as
Georgiana Molloy, were captivated by the
Western Australian wildflowers:

It is to be regretted that the flowers of this country
are so uncommon in England, as were they imitated
in that beautiful work of art, artificial flowers, they
would create a great rage. They are so well calculated
from their size for that work, also the brilliancy of
colour could well be imitated in the manufacture of
porcelain and china . . . (A. Hasluck, 1955)

Because there was no assured water supply and
little labour, even the most nostalgic settlers
could not hope to recreate an English landscape
in Western Australia. An article in the Perth
Gazette (9 March 1833) listed fruit-trees and
bushes introduced and already “flourishing”.
Most were Mediterranean or tropical: grape
vines, two varieties of fig, peach, almond, several
varieties of apple, pear, strawberries, aloes,
several varieties of plum, four varieties of olive,
mulberry, pineapple, plantain, sugar-cane, flax-
lily and Cape gooseberries. Alfred Stone, builder
of the first cottage in St. George’s Terrace,
planted a geometric layout of olives and vines
behind his house and a selection of “natives”
in front, including Nuytsia floribunda and vari-
ous “mahogany” trees (Oldham, 1961). The
Benedictines at Subiaco also planted olives. It
is known that Norfolk Island pines and Moreton
Bay fig were planted in Perth at this period,
and the cypress was probably introduced before
1860.

In a recent Flinders University honours thesis
Moon has appraised the attitudes to the rural
environment of South Australian colonists. It
is probable that a similar detailed study of the
attitudes of Western Australian colonists would
lead to the same general conclusions.

. . . . (the grasres and the trees) . . . were mostly
utilized in their natural state, little attempt being made
to improve, or conserve their value as a resource . . .
the trees were generally tried, found wanting, and
consequently cleared away as much as possible.

The early nineteenth century Englishman,
reacting against the immediate horrors of in-
dustrialization, created the myth of a rural Eden
which was certainly far from the reality of the
pre-industrialized rural life. The early colonists
would have found the strange Australian en-
vironment difficult enough to appreciate even
if their attitudes had not been partly shaped
by such a myth. Their attitude, as Moon has
shown for the South Australians, was too often
one which later generations would see as ex-
ploitative. It is an attitude which still persists
amongst many'. In any case, with so much
virgin bush around them, the early settlers
would not have seen any need for reserves. The

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

57

first reserves were set up under the Land Regu-
lations of 1872 and 1887, though they lacked
legislative security (Australian Academy of
Science, 1962).

There can be little doubt that the Aborigines
also exploited the environment beforehand and
Merrilees <1968) and Hallam (1971) have dis-
cussed the significant effect of the use of fire
by them. Hancock (1972) also discusses similar
studies of the effect of fire in south-eastern
Australia. However, there seems to have been
too little research on the comparative effects of
naturally-occurring fires, Aboriginal fires and
the fires of the European settlers. Hallam draws
attention to the relative infrequency of fire due
to lightning strike. However, the Western Aus-
tralian flora displays many adaptations to fire
and these adaptations may have required a
longer time span than that now suggested for
Aboriginal occupation. Further research may
reveal that this time span was in fact sufficient
for man-made fire to be effective. Some of
these adaptations may have evolved in response
to increasing aridity, and it may be impossible
to separate the effect of fire and climate.
Gardner <1957) argues for the primacy of fire
as a factor and suggests greater frequency of
“natural fires” caused by lightning strike, but
on more subjective evidence.

Man’s use of fire may have affected prin-
cipally the distribution of species, both of plants
and animals. It would be interesting to know
in more detail the different results of the vary-
ing practices of the European and the Aborigine.
It is possible that the European use of fire has
been more destructive, because of the altered
pattern and greater frequency of burning, set-
tlement in fixed locations, and because of the
more intense pressure of a greater population.
Hatch (1959) and Wallace (1966) have discussed
the effect of post-settlement fires in the jarrah
forests, while Gardner (1957) examined the
effect of fire on the whole flora. It is hoped
that current research by forestry scientists will
greatly improve our appreciation of the role of
fire in environmental management. It may be
too late, however, to study in detail the relative
effects of Aboriginal and European practices.

Fire is only one of the factors that changed
significantly when the Europeans arrived. It
appears that the grazing habits of sheep, for
example, very quickly altered the distribution of
native grasses. There appear to be fewer local
contemporary observations than Hancock could
draw on for his Monaro study. Cattle and goats
have, in their own ways, been destructive of the
environment, particularly in a region where
fresh-water sources are relatively thinly distri-
buted. Introduced animals have to concentrate
near these sources, whereas the native animals
have adapted to be independent, or less depend-
ent, on direct water supplies.

Popularly the introduction of the fox and the
rabbit have been recognized as environmental
tragedies. However, except for a colony of rab-
bits on Carnac Island (Moore, 1931) neither was
introduced into Western Australia before the
20th century. Because of sentiment there has

been far less popular recognition of the great
impact of the cat and the dog on marsupial
fauna. The descendants of European dogs “gone
bush” were also probably more often responsible
than dingoes for attacks on sheep. At the same
time native species of animals and birds were
assumed, sometimes quite falsely, to be vermin,
and this led to unwarranted destruction.

Introduced pathogens are part of the total
impact on the environment, although in the
earlier years of the colony the long ocean voyage
may have been an effective quarantine against
the introduction of plant and animal disease. It
has nevertheless been speculated, for example
that ‘jarrah die-back disease’ may have been
introduced with plants brought by the colonists
<Newhook and Podger, 1972). Scab was inter-
mittent among sheep from its first introduction
from Van Diemen’s Land in 1831 to its gradual
eradication around 1900 (Deacon, 1950: Cran-
field, 1959). Another early setback to pastoral
settlement was the presence of native poison
plants, leguminous pea-flowered bushes which to
an unaccustomed European eye might look like
good grazing. Drummond in 1840-41 identified
the York Road poison ( Gastrolobium calycinam ) ,
and other gastrolobia were later brought under
suspicion (Erickson, 1959).

Farming practices were at first of necessity
poor. Bunbury in 1837 censured the Avon Valley
settlers for slovenly farming, deploring especially
the habit of broadcasting seed among the old
stubble of last year’s crop. From an early period
agricultural associations were formed at Perth
and York to encourage improvement, and
Hutchison is at present undertaking a study of
19th century farming techniques which should
lead to a more accurate, and possibly more
favourable, assessment of the performance of
Western Australian settlers by contemporary
standards.

1859-90: Consolidation

The coming of convict transportation in 1850
gave the expected stimulus to Western Aus-
tralia’s growth. From 5,886 the Anglo-Aus-
tralian population rose to 25,084 in 1870 (Gertzel,
1949). With the ending of transportation a
period of recession and near-stagnation followed,
but from 1882 the growth of population again
accelerated, to reach 48,502 by the attainment of
self-government in 1890. Increased contact with
the outside world brought to the colony diseases
from which it appears to have been previously
immune. Research now nearing completion by
E. J. P. Joske should throw light on the first
decades of the colony’s medical history (M.A.
thesis, Department of History, University of
Western Australia). An outbreak of measles in
1860-61 attacked Aborigines and young children
with particular severity as did another in 1883.
Infant mortality was high throughout this
period; between 1860 and 1870 nearly 45 per
cent of all deaths were of children under five
(Knight, 1871). Diphtheria, first reported in
1865, was prevalent thereafter. Typhoid was
reported several times between 1872 and 1900,

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

58

though it was not until 1889 that the common
‘colonial fever’ was identified by Dr Adam
Jameson as a form of typhoid ( Inquirer , 3 April
1889). Sanitation was primitive; it was not
until 1888 that even the city of Perth could boast
a system of nightsoil collection, and though
improvements followed fairly rapidly in the city,
conditions were less good in country and gold-
fields districts.

European penetration inland was extensive
but very sparse during the period between 1850
and 1890. The activities of sandalwood cutters
from about 1845 resulted in the pioneering of a
number of tracks which almost certainly fol-
lowed Aboriginal routes, and would in turn be
followed and developed during the great period
of goldrush expansion in the 1890’s. Improving
wool prices and the availability of convict shep-
herds encouraged the eastward thrust of the
pastoral frontier, and by the mid- 1860’s much of
the present wheat-belt was held under grazing
leasehold; but because of impermanent water
and transport difficulties, very few improvements
resulted. From the 1860s pastoral expansion was
directed to the Gascoyne, north-west, and
Kimberleys. One effect of this spread of settle-
ment was the destruction of habitat and near-
extinction of some local fauna. The noisy scrub-
bird was believed extinct after 1889. but has
recently been located at Two People Bay, east
of Albany.

Rowley (1942) and Mouritz (1964) have dis-
cerned four major regions of closer settlement
in the period between 1860 and 1890. The Irwin-
Greenough district and the Avon Valley were
the habitat of the “yeomanry”, grain farmers on
small acreages. The Swan coastal plain was
mainly devoted to “kitchen gardens” and vine-
yards. In the fertile valleys of the south-west,
such as the Preston, the Capel, and the Black-
wood, the ‘‘bush grazing” of sheep and cattle
was supplemented by occasional fruit-growing
for local markets. Meanwhile from about 1850
a demand for local hardwoods, “Western Aus-
tralian mahogany”, led to the clearing of some
stands of jarrah in the south-west (Stewart,
1948; Robertson, 1958).

At first sight, the effect of these developments
on the Western Australian environment seems
meagre. Only 132,000 acres were cleared by
1890, understandably in view of the density of
the timber in much of the south-west. The
typical homestead was sited in a clear patch of
land surrounded by a huge area of bush. Fenc-
ing was slow to spread. Although some pad-
docks were enclosed in the Toodyay area as
early as the 1850s, fences were not common in
the Williams district until after the establish-
ment of a road board in 1877; while in the
impoverished Irwin-Greenough district even
after fencing became compulsory by the Land
Act of 1887 the usual expedient was wattle
fencing which rotted quickly, as there was no
money for any other (Erickson, unpublished;
Chate, 1953; McAleer, 1956). In the Hay River
district sheep were folded in hurdles: ‘Their stay
in one fold position was largely determined by
the deposit of manure within and the condition

of grazing without’ (Sten, 1943). In the Harvey
district problems of soil and water led to a cycle
of grazing between coastal homesteads and out-
stations towards the Darling scarp (Staples,
1947). There is indirect but suggestive evidence
that grazing practices during the 19th century
led to environmental deterioration. Deacon
(1950) noted the belief of farmers in the Manji-
mup district that frequent fires had promoted
the growth of scrub and undergrowth. Giles
<1950» noted the presence of severe soil erosion
and evidence of over-cropping in the long-settled
districts of the Avon Valley. Sten cites a Hay
River property occupied by one pioneering
family for 72 years after 1855, during which time
no more than 160 acres were cleared. By 1927
the native grasses were largely eaten out;
attempts to introduce prairie grass, timothy,
white clover, and other exotics had largely
failed; the land was neither rested, ploughed,
nor fertilized. Of the owner for much of this
period Sten wrote: ‘He makes no claim to have
improved his property outside the homestead
and the dividing fences; he even possibly feels a
little contemptuous of those who have expended
so much capital and have perhaps less ready
cash than himself. His aim was to acquire land
free of debt . . . .’ It is probable that such an
attitude towards investment was common. Re-
liance was placed on hard work and the recuper-
ative properties of the land. Economic necessity
fostered this attitude. One pioneer family in
the Kojonup district began farming in 1851 with
five sons, two daughters, ‘17 6 in cash after our
journey, two mares, one cart and three dogs on
which we depended to catch kangaroos for our
own use and the skins to provide us with what
else we required. We also had two spades, three
grub hoes and two sickles’ iBignell. 1971’).

Such small beginnings offered little protection
against adversity. In the Harvey district the
absence of good communications obliged farm-
ers to resort mostly to subsistence agriculture.
There was little margin of resilience against
natural hazards. Thus in the Irwin-Greenough
district a promising settlement of small farmers
on tillage blocks was severely checked by out-
breaks of rust in wheat from 1865. Contempor-
ary observe <e.g. Trollope. 1873; the Venn Com-
mission, 1887-90) criticized the backwardness of
these farmers, but in fact most were in debt to
the Gcraldton merchants and storekeepers who
would make no advances for improvements.
Fallowing was not practised because of wind
erosion. Technology improved slowly. Reapers
and binders were not commonly in use until the
1890s, and in many parts of the colony flails and
scythes were still in use (McAleer. 1956; Kelly,
1958; Cranfield, 1959). Although from the
187Cs guano deposits off the Western Australian
coast were systematically exploited, local farm-
ers did not consider it highly as a fertilizer, and
made little attempt to seek out alternatives.
Water conservation was not extensively prac-
tised. so that in the York district in the dry
season of 1877. farmers had to cart water up to
fifteen miles, though the severity of that drought
and of others during the late 19th century has

Journal of the Royal Society of Western Australia. Vol. 515 Paris 1 and 2, July. 1973.

59

been subsequently exaggerated. Some landhold-
ers such as Bishop Salvado practised extensive
well-sinking (Dom William, 1961), at times
exploiting earlier Aboriginal watering-places. In
assessing the “conservative” performance of
Western Australian farmers, it should be remem-
bered that farmers in England were also very
slow to adopt new machinery, even though they
had readier access to manufacturers. Many
colonists had not built up enough cash capital,
and the scale of operation of many farms may
not have been large enough to warrant mech-
anization. This question requires closer research.

The modest circumstances of most farmers
limited the possibilities of vernacular architec-
ture. Many settlers began with a bark hut,
graduating later to either a white-washed abode
house, “with mud walls a foot thick”, or to a
pit-sawn timber cottage. Roofing materials were
either rush thatch or jarrah shingles. Ver-
andahs became relatively common (Hillman. and
Norrish, 1938). A significant change came in
1879 with the introduction of galvanized iron.
Its use for housing and roofing followed almost
immediately in the Perth-Fremantle area. Little
opportunity existed for planned landscaping,
though mention must be made of the enlight-
ened policies of Maitland Brown, resident magis-
trate of Geraldton in the early 1870s, who
confronted by the encroachment of sandhills on
the town area, employed the out-of-work Green-
ough farmers in a systematic and extensive
plantation of native trees all round the outskirts
of Geraldton (Farrelly and Maley, 1927). Mean-
while public buildings were erected on a more
opulent scale, especially during the period of
convict transportation. In Fremantle the use of
limestone and an occasional hint of Cape Dutch
influence gave individual character to such
buildings as the Asylum (now the Fremantle
Museum). In Perth the favoured medium was
a brick provincial Gothic, often reflecting a
strong Tudor or ‘Flemish’ influence, as with the
Perth Town Hall, Wesley Church, and the Cloist-
ers (Oldham, 1961). Private housing also re-
flected the growing opulence of some citizens,
not only among the town residences of notables
along Adelaide Terrace and Mount Street, but
even in one or two country houses, such as
Faversham House at York, where an attempt
was made to reproduce the idiom of the English
villa.

The diffusion of English culture and the
stimulation of the Western Australian economy
both depended on improved communications.
Convict road-building gave Perth an adequate
link with the Avon Valley and Albany, but rail-
ways were essential. The first public line
from Geraldton to the base-metal centre of
Northampton, was not opened until 1879. During
the 1880s Fremantle and Perth were linked to
the Avon Valley towns. Private entrepreneurs
working on the land-grant principle were able
to plan more boldly, building a line to Albany
by 1889 and projecting another to Geraldton
(Bolton, 1958; Manford, Ph.D. thesis, submitted
1973). The coming of the railway would provide
primary producers with access to markets and

facilitate the introduction of wire fencing, gal-
vanized iron, fertilizer, and other products which
would transform the face of the Western Aus-
tralian landscape.

1890-1973: Modern period

Western Australia was transformed by a
series of gold discoveries beginning in a small
way in 1885 and culminating in the Coolgardie-
Kalgoorlie-Boulder finds of 1892-93. Demo-
graphically the “ancient colonists” were swamp-
ed by newcomers, mainly from Victoria and
South Australia. These men and their families
accounted for most of the increase from 48,500
in 1890 to 180,000 in 1900. Because of the conse-
quent natural increase, reinforced by British
migration, the population went on rising to
331,000 in 1920 and 432,000 in 1930. Depression
and war then checked the rate of increase, but
in 1946 the figure stood at half a million, and by
1973 exceeded a million — a more rapid growth
than that of the rest of post-war Australia,
though stimulated by the same ambitious pro-
gramme of European migration.

During the whole of this 20th century increase
it was remarkable that the area under white
settlement in Western Australia did not expand
significantly, and in some areas contracted
somewhat. Gold pushed the eastward frontier
of settlement forward during the 1890s to a line
east of Sandstone, Kurnalpi, and Esperance, but
except for the Kalgoorlie-Boulder regional
metropolis and a few outposts such as Norseman,
Menzies, Leonora and Laverton, the continuity
of settlement depended on other industries: wool
(sparsely), agriculture <to the 10-inch isohyet)
and, more recently, base metals such as iron ore
and nickel. It was nevertheless the gold-rush of
the 1890s which prompted a bold policy of public
works.. The Coolgardie pipeline completed in
1903 brought an assured water supply to the
Eastern Goldfields and the intervening farming
districts. The railways linking the newly settled
districts to Perth provided access to markets
and to imports from outside Western Australia.
The completion of Fremantle Harbour in 1897
encouraged regular shipping services, and the
advent of a federated Commonwealth in 1901
ensured that Perth, open to outside competition,
would produce few of the manufactures required
for its own hinterland. Thus the pattern was
established under which Western Australia’s role
was seen mainly as that of an exporter of prim-
ary produce, industrially backward and techno-
logically dependent on outside skills.

Between 1900 and 1930 the three major rural
areas were G) the timber and dairying districts
of the lower south-west (2) the mixed farming
districts of the Avon Valley and the upper Great
Southern and ( 3) the wheat-belt. It was the
wheat-belt that dominated the planning of the
Western Australian economy. The introduction
of “dry” wheat facilitated the occupation of the
200 miles east of the Avon Valley. The “wheat-
belt” extended from within five miles of the
Murchison south-east to Southern Cross, then
south to Hyden and Gnowangerup, for much of

Journal of the Royal Society of Western Australia, Vol. 56 Paris 1 and 2, July, 1973

60

its length paralleling the 10-inch isohyet and
the rabbit-proof fences erected in the early years
of the century to stem the invasion from the
Eastern States. Light railways and the ready
availability of public finance stimulated agricul-
tural settlement, even though the lack of surface
water and of dams exacerbated the drought
seasons of 1911 and 1914. Between 1900 and
1930 the area cleared of timber increased from
3.440,000 to 11.8 million acres, and the area
under crop from 201,000 to 4.8 million acres. In
1930-31 the wheat harvest reached 53 million
bushels, a figure unsurpassed for thirty years.
Despite the establishment of a Department of
Agriculture in 1898, this growth was largely
without scientific foundations. Most settlers
identified soil types by the characteristic vege-
tation, e.g. morrell gum soil, heavy gimlet soil,
York and jam soil (Schapper, 1955). Soil erosion
was discerned as a problem in the Geraldton
coastal districts as early as 1901. The planting
of rye grass was recommended as a palliative,
but no mention was made of altered farming
methods as a means of preventing erosion (Jour.
Dept- of Agric. 1901). Scientific soil analysis
was first attempted by Teakle in 1929 in an
attempt to solve the problem of poor crop yields
in the Salmon Gums area, but his findings of
excess salinity were at first disputed by “prac-
tical” farmers and politicians. It was only with
the collapse of wheat prices in the depression of
the early 1930s, followed by the extensive aban-
donment of marginal farms, that value came to
be placed on the research of agricultural scien-
tists. The same applied to the dairying industry
of the lower south-west where the group settle-
ment scheme of the 1920s was based on the
fallacious assumption that country supporting
tall timber must be capable of pasturing
dairy cattle tended by recently-arrived British
migrants. Here too the failure rate was high
(Hunt, 1958).

The 1930s saw the first break-throughs in the
scientific appreciation of Western Australia’s
agricultural potential. Underwood <1935; 1972)
showed that the ‘Denmark wasting disease’ in
cattle, and other symptoms of wastage and
infertility in livestock pastured in the coastal
districts, were due to a deficiency in trace
elements such as cobalt. Copper deficiency was
found to cause neonatal ataxia in lambs (Ben-
netts, 1937). Teakle (1938) published the first
comprehensive soil survey of agricultural regions
in Western Australia, and Gardner (1942)
definitely identified the three main vegetative
forms of the wheat-belt: sand-heath, savannah
woodland (York and jam gum country), and
eastern sclerophyllous woodland (mallee and
salmon-gum country). Meanwhile in an early
study of farming practices Roberts (1942)
showed that attempts at mechanization — speci-
fically through the use of tractors — often failed
because operation was neither efficient nor
economically planned. This lay some stress on
the quality of farm management, a factor to
which greater attention would be devoted after
the Second World War.

The years from 1930 to 1945 laid great burdens
on the farmers. Apart from the depression,
followed by the shortages of labour and materi-
als in the Second World War, there were natural
hazards: plagues of emus in 1927-29 and 1932,
grasshoppers for several years in the mid- 1930s,
dry seasons in 1936, 1938. and especially 1940.
Though unwelcome, the spread of foxes was
thought to check the incursions of rabbits.
Quarantine measures mostly prevented the
introduction of exotic stock diseases, such as the
tick fever common in the Kimberleys, but in
1923-24 there was a short-lived outbreak of
rinderpest near Fremantle. However ‘toxic
paralysis’, a form of botulism in sheep first re-
ported in 1927, was responsible for the deaths of
10,000 sheep in the summer of 1932-3. The
introduction of subterranean clover was hailed
in the 1920s as an important contribution to
pasture improvements. However it came under
suspicion during the late 1930s and early 1940s
as responsible for a serious fall in lambing per-
centages. This was later shown to be due to
the oestrogenic action of the plant 'Bennetts,
Underwood, and Shier. 1946: Bull, 1972). All
these problems were identified and countered by
agricultural scientists, but they were not always
made use of by the farmers, who with their
bitter experiences of debt were naturally wary
about investment in improvements, or even
about departure from established practice.

Among those farmers who survived, these
pressures led to changing policies. Wheat gave
way partially or wholly to sheep in some dis-
tricts. Farms were enlarged and amalgamated.
With returning prosperity after 1945 these larger
units were better able to make effective use of
tractors, combine harvesters, and other forms of
mechanization. In the drier wheatgrowing dis-
tricts (e.g. Burracoppin) the average size of
farms doubled between 1940 and 1960 to from
1,500 to 3,000 acres. Productivity was estab-
lished to be lower on smaller holdings. The
rotation of crops was modified so that cropping
for wheat tended to occur less frequently on any
one piece of land. Despite improved farming
practices wheat yields failed to increase signi-
ficantly, and while controlled experiments failed
to produce irrefutable proof of declining soil
fertility, the circumstantial evidence was in-
creasingly strong. Soil erosion was identified in
long-settled districts such as the Avon Valley.
Nevertheless in this and in other regions of
higher rainfall a concerted policy of pasture
improvement was followed to intensify sheep
grazing under the stimulus of buoyant wool
prices in the late 1940s and early 1950s
(Schapper, 1955; Mouritz, 1964).

One response to improving conditions was the
renewed expansion of rural settlement. During
the 1950s and 1960s the previously unexploited
“light lands” to the east and south of the old
wheat belt were brought under cultivation.
Official policy aimed at the yearly alienation of
a million new acres for agriculture — usually at
the expense of the pastoralists — and stimulated
clearing by the incentive of tax concessions.
The area under wheat reached a maximum of

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2. July. 1973.

61

7.3 million acres in 1968-69, and the harvest of
that year attained a record of 112.4 million
bushels. A rural recession in 1969-70 provoked
some doubts about the continuing wisdom of
these trends. ‘The time is fast approaching
when the only virgin bush left in Western
Australia will be that which has been deliber-
ately reserved for conservation* (Ride, 1968L
Hogstron (1968) estimated that in 1965 there
remained 11.5 million acres of land in Western
Australia suitable for allocation to agriculture —
an area sufficient only to meet the estimated
demand from families already settled in Western
Australia. Despite this growth, many rural
towns were stagnant or decaying after 1945. As
motor transport replaced the old pattern of rail-
ways largely fed by horse-drawn traffic, the need
passed for towns as service centres at intervals
of every 15 or 20 miles. Instead, growth tended
to concentrate in a few ports or railway junc-
tions. such as Geraldton, Bunbury, Albany,
Northam, Narrogin and Merredin. There was a
slight corrective to this trend because of modern
mining developments in the south-west. The
extensive working of bauxite in the Darling
Range after 1959 led to the choice of Pinjarra
as the site for an alumina refinery, and the
working of mineral sands in the south-west
during the same period fed the secondary
industry of Bunbuy. Both these developments,
however, gave concern to conservationists be-
cause of their influence on the environment, and
neither had any major effect on the prevailing
trend of population. Following a common
Australian experience, the bulk of Western
Australians concentrated in the towns, and
especially in the Perth-Fremantle metropolitan
area.

In the city of Perth the sprawl of suburban
growth had been facilitated since the 1890s by
the easy availability of timber weatherboards
and machine-made bricks for house construction.
Like Sydney, but unlike Brisbane and Hobart,
brick was mere highly esteemed as a building
material in Perth. Iron roofing was common in
workingclass suburbs, but its tendency to rust
led the better off to prefer tiles, at first imported
from Marseilles, after the First World War
manufactured locally. Streets were planned
almost entirely as straight lines or rectangles,
although in the eastern suburbs the lines often
ran diagonally on a north-west south-east axis
following the boundaries of the original Swan
River land grants devised to afford as many
settlers as possible a river frontage. Curved
crescents were introduced between 1911 and 1914
as an innovation in the laying out of Dalkeith
and the Mount Lawley Number 3 Estate, but
were not widely imitated outside the "prestige
areas” (Hope. 1968 >. The "silvertails” congre-
gated in Peppermint Grove. West Perth, and
subsequently Dalkeith Nedlands. suburbs com-
manding access to Melville Water, King’s Park,
and similar amenities, while the sandy eastern
suburbs, further removed from the ocean and
the sea breezes, and in general somewhat flatter
than the region between Perth and Fremantle,
became predominantly low-cost working-class

residential areas. The provision of suburban
railways was totally neglected, except where
existing main lines were used to provide a local
service; however during the 1890s light lines
were built from Midland to serve the orchards
and the timber-leases of the Darling Range, and
this was a great stimulus to picknicking. The
beaches, an alternative source of recreation,
were only developed in the Cottesloe-Leighton
area served by the suburban railways, until after
the First World War the acceptance of mixed
bathing and the growing supply of motor-cars
and buses facilitated the opening of City Beach,
Scarborough, and North Beach.

More than anything else, the provision of an
adequate piped water supply transformed the
Perth environment. Seddon (1970 and 19721
has shown how nearly all suburban householders
eradicated the local flora when they built houses,
and instead used their hoses to cultivate trim
lawns flanked by beds of largely imported
annuals, rose-bushes, frangipani, hibiscus, and
other non-natives. Even King’s Park, so often
extolled as "a thousand acres of natural bush
near the heart of the city” remained in that
condition only because the trustees, such as Sir
John Forrest, lacked the financial means to
redeem such a large area from "the drab
monotony of the Australian bush” (private com-
munication from Dr P. R. Wycherlyi*; and in
any case it soon ceased to be a sample of the
pristine flora of the coastal sandplain. It was
not until well into the post-World War II period
that public taste, perhaps in response to occas-
ional summer water shortages, came to favour
the encouragement of native trees and bushes in
the suburban environment: just as it was not
until then that variants on the conventional red-
brick and tile were sometimes voluntarily chosen
as acceptable building materials. One notice-
able side-effect of Perth’s domestic architecture
was its influence on the planning of country
farm-houses. With return of prosperity after
the Second World War farmers who could afford
to rebuild their homesteads forsook the old
vernacular architecture, with its overtones of
makeshift, and built houses in the bush undis-
tinguishable from the ordinary suburban bunga-
low.

Rapid post-war metropolitan growth provoked
a concern for environmental planning. Al-
though Perth had been under a Town Planning
Act since 1928 some factors affecting living
conditions were little understood. Few foresaw
that the establishment of an industrial complex
at Kwinana after 1952 might affect the atmos-
phere of suburbs dependent on the south-west-
erly sea breeze for summer relief. Transport
policy, despite the integration of bus services
under the Metropolitan Transport Trust in 1958,
failed to solve two major problems. These were
the increasing use of private cars and the de-
liberate concentration during the 1960s and
early 1970s of the administrative headquarters
of almost the entire business and financial world,
together with most State and federal govern-
ment offices, along one over-crowde d mile of

* King's Park and Botanic Garden, Perth.

Journal of the Royal Society of Western Australia. Vol. 56 Parts 1 and 2. July, 1973

62

high-rise blocks around Saint George’s Terrace.
A closely controlled policy of land zoning
checked inordinate suburban sprawl, but only at
the cost of high land prices. Differentiation
between the status of suburbs continued to grow,
the highest value being placed on those com-
manding views and access to the river or the
sea.

Conscious nevertheless that by most world
standards Western Australia was a favoured
environment, the community took remedial
action. At the official level the destruction of
some native fauna was mitigated by the Game
Acts of 1874, 1892, 1900 and 1912, and the Land
Act of 1898 specifically provided for the creation
of reserves for the protection of indigenous fauna
and flora: but these early initiatives were not
adequately followed up. At the voluntary level
some studies of the effect of man on the environ-
ment came to be made by such societies as the
Royal Society of Western Australia and the
Naturalists Club, which although founded as
early as 1914 and 1924 respectively, came to
shift the focus of their interests in the postwar
period (Australian Academy of Science, 1962).
In 1959 a concern for the State’s early buildings
led, none too soon, to the foundation of the
National Trust, which carried out useful work in
classifying those deserving of preservation. In
the same year the State Government took a new
initiative by setting up the Swan River Conser-
vation Board, an authority with overriding pow-
ers to control the problem of river pollution
which had first been noted as far back as 1870.
In 1970 the State’s legislators agreed to create a
Department of Environmental Protection, and in
1971 went further with an Act creating two
statutory authorities, the Environmental Pro-
tection Authority and Council, with power to
report on the implications of proposed new
industrial and commercial developments. Con-
troversy was aroused in 1972 when the Authority
reported against a proposal to site the Pacminex
alumina refinery in the upper Swan, but as
alternatives proved to be available, it was hard
to argue that the community lost through a
concern with environmental factors.

The problems remained formidable. Western
Australia — or at least its south-west — consisted
of a hinterland increasingly seen as a “big
man’s” agricultural frontier in which closer
settlement, if it came at all. would result only
from the chances of mining development or from
the deliberately subsidized transplantation of
selected industries. The inhabitants of its main
centre of population would continue overwhelm-
ingly to seek work and recreation on a narrow
coastal sand-plain with a delicate ecology; yet on
this basis most experts forecast a population of
one million by 1990, and the Lord Mayor of
Perth (on evidence as yet unpublished) consid-
ered the optimum population to be between 2
and 2.5 million. It would not be easy to recon-
cile the demands of economic growth with a
retention of the quality of life which made
Perth, despite its isolation, a fine city to live in.

References

Ackland, R. B. (1962).— The history of Wongan Hills.

Journal of the Historical Society of Western
Australia. 6 (1) : 60.

Albertus, Mother M. (1966).— The Irwin story. Early
Days. 6 (5) : 7.

Australian Academy of Science (1962). — “National Parks
and Nature Reserves in Western Australia:
by the Western Australian Subcommittee
... on National Parks”.

Bassett, M. (1954). — “ The Hentys: A Colonial Tapestry”.
Melbourne.

Battye, J. S. (1924). — ‘‘Western Australia: A History”.
Oxford.

Bennetts, H. S. (1928). — Carrion poisoning of sheep
(botulism). Australian Veterinary Journal. 4:
105.

Bennetts. H. S. and Chapman, F. E. (1937). — Copper
deficiency in sheep in Western Australia.
Australian Veterinary Journal. 13: 19.

Bennetts, H. S.. Underwood. E. J. and Shier, F. L.

(1946). — A specific breeding problem of sheep
on subterranean clover pastures in West-
ern Australia. Australian Veterinary Journal,
22 : 2.

Bignell, M. ( 1971 ) .— ‘First the Spring: A History of the
Shire of Kojonup. Western Australia" . Perth.

Bolton, G. C. (1958). — “ Alexander Forrest: His Life and
Times”. Melbourne.

Bolton, G. C. (1972). — " A Fine Country to Starve In”.
Perth.

Bryan. C. (1937). — The earliest medical history. Journal
of the Historical Society of Western Aus-
tralia. 2 (12): 47.

Bull. L. B. (1972). — Notes on the development of veteri-
nary science in Australia. Records of the
Australian Academy of Science. 2 (2): 7.

Cameron. J. McD. R. (1968). — Some comments on the
early rural dwelling. Australind. 2: 20.

Cameron. J. McD. R. ( 1970) .—Perception and settle-
ment. Australind. 3: 49.

Chate. A. H. (1953). — The Williams. Journal of the His-
torical Society of Western Australia. 4 (5):
20.

Cranfield, R. E. (1959). — The Lefroys at Walebing. Jour-
nal of the Historical Society of Western
Australia. 5 (5): 42.

Crowley, F. K. (1956). — Problems in local history. Jour-
nal of the Historical Society of Western
Australia. 5 (2): 19.

Crowley, F. K. (1960). — " Australia's Western Third”.
London.

Crowley. F. K. (1971). “ Forrest : vol. I. 1847-1891 ”.

Brisbane.

Deacon, J. E. (1947). — " The historical developxncnt of
the Avon Valley. 1830-1850". M.A. Thesis,
University of Western Australia.

Deacon. J. E. (1950). — Pioneering in the South-West
(Manjinnip). Journal of the Historical
Society of Western Australia. 4 (3): 54.

Deacon. J. E. (1951). — Early years in Manjimup. Jour-
nal of the Historical Society of Western
Australia. 4 (4): 34.

Dollar. G. J. (1968). — Ebb and flow of a boom town:
the story of Bullfinch. Journal of the His-
torical Society of Western Australia. 6 (8):
57.

Eastcott. W. G. (1938). — Pioneers of Wagerup. Early
Days. 3 (1): 31.

Erickson. R. ( 1951 ) . — “Orchids of the West". Perth.

Erickson, R. ( 1964) .—Early days at Bolgart. Early Days.
6 (3): 46.

Erickson. R. (1969). — "The Drummonds of Hau'thornden.
Osborne Park.

Farrelly, M. and Maley. H. K. (1927). — The Greenough
district. Journal of the Historical Society of
Western Austrulia. I (2): 34.

Field. T. P. (1957). — "Swanland”. Lexington.

Gardner. C. A. (1957). — The fire factor in relation to
the vegetation of Western Australia. W.A.
Naturalist. 5 (7): 166.

Gertzel, C. J. (1949). — "The conx'ict system in Western
Australia. 1850-1870” . B.A. Hons. Dissertation,
University of Western Australia.

Giles, D. K. (1950). — "Some aspects of land use in a
part of the Avon Valley”. M.Ag.Sci. Thesis,
University of Western Australia.

Harris, M. (1943). — Migration to Western Australia.
Early Days. 3 (6); 26.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973.

63

Hasluck, A (1955). — "Portrait with Background” . Mel-
bourne.

Hasluck, P. M. C. (1942 ). — “ Black Australians”. Mel-
bourne.

Hatch, A. B. (1959). — The effect of frequent burning on
the jarrah (Eucalyptus marginata) forest
soils of Western Australia. Journal of the
Royal Society of Western Australia. 42 (4):
97.

Hogstrom, A. W. (1968 ). — “ The availability of uncleared
land for development in the agricultural
areas of Western Australia” . M.Ag. Sci.
Thesis, University of Western Australia.

Hope, P. G. S. (1968).— Reminiscences of a surveyor with
particular reference to Dalkeith and Point
Walter. Early Days. 6 (7): 23.

Hunt, I. L. (1958). — Group settlement in Western Aus-
tralia. University Studies. 3 (2): 5.

Irwin, F. C. (1835 ). — "‘The State and Position of West-
ern Australia” . London.

Kelly, G. J. (1958 ). — “A history of the Champion Bay
district”. M.A. Thesis, University of Western
Australia.

Kerr, A. M. (1955).— “A study in regional development:
the South-Wesc region of Western Austra-
lia”. Ph.D. Thesis, University of Western
Australia.

Kimberley, W. B. ( 1897) .—‘‘History of Western Aus-
tralia”. Melbourne.

Knight, W. H. ( 1871 ) .—"‘Census of the colony of West-
tern Australia taken on the 31st March
1870 . . . . ” Perth.

Main, B. (1969 ). — “ Between Wodfil and Tor”. Perth.

McAleer, M. (1955).— Some aspects of the development
of the northern wheatbelt. Journal of the
Historical Society of Western Australia. 5
(1) : 45.

Merrilees, D. (1958).— Man the destroyer. Journal of
the Royal Society of Western Australia. 51
(1): 1.

Molyneux, I. (1972).— The Serpentine farm. The Archi-
tect. 72 (1): 1

Moore, F. ( 1931 ) .—Reminiscences. Journal of the His-
torical Society of Western Australia. 1 (9):
65.

Mouritz, N. E. (1964).— “An examination of patterns
and trends in the size of farms in the
South West of Western Australia with impli-
cations for the future”. M.Ag.Sci. Thesis,
University of Western Australia.

Murray-Gordan, A. E. ( 1952).— Pioneering days in the
eastern wheatbelt. Journal of the Historical
Society of Western Australia. 4 (4): 76.

Newhook, F. J. and Podger, F. D. (1972). — The role of
Phytophthora cinnamoni in Australian and
New Zealand Forests. Annual Review of
Phytopathology. 10: 229.

Norrish, T. and Hillman, J. F. (1943).— Old Ettakup.
Early Days. 3 (2): 21.

Ogle, N ( 1839 ) . — ‘‘The Colony of Western Australia” .
London.

Oldham, J. and R. ( 1958) .—Western Australia’s colonial
architecture. Journal of the Royal Society
of Western Australia. 5 (6): 31.

Oldham, J. and R. (1962 ).— “Western Heritage”. Perth.

Ride, W. D. L. (1968 ). — “ Conservation through reserva-
tion”. (unpublished paper).

Rintoul, J. ( 1964 ) . — “Esperance Yesterday and Today”.
Perth.

Roberts, R. P. (1942). — " Costs of tractor operation in
the Western Australian wheat belt, 1938-39”.
M.Ag.Sci. Thesis, University of Western
Australia.

Robertson, J. R. (1957). — The We tern Australian tim-
ber industry. University Studies. 3 (1): 51.

Roe, (Mrs.) J. B. (1927).— Some old time memories.

Journal of Historical Society of Western
Australia. 1 (2): 4.

Rowley, S. I. ( 1945) .—'"Land settlement in Western Aus-
tralia 1890 to 1942”. Research Report to
Minister of Post-War Reconstruction, Uni-
versity of We tern Australia.

Schapper, H. P. (1953).— “A survey of dairy farming in
the South-West of Western Australia”.
M.Ag.Sci. Thesis, University of Western Aus-
tralia.

Schapper, H. P. (1955).— “A survey of sheep and wheat
farming in the eastern wheat belt of West-
ern Australia”. Ph.D. Thesis, University of
Western Australia.

Scherer, A. ( 1968 ) .— “History of the Upper Blackwood”.
Perth.

Seddon, G. ( 1970) .— ' “Swan River Landscape”. Perth.

Seddon, G. ( 1972) .—‘‘Sense of Place”. Perth.

Serventy, D. L. and Whittell. H. W. ( 1967 > .-"‘Birds of
Western Australia” . 4th ed. Osborne Park.

Shann, E. O. G. ( 1926) .— 1 "Cattle Chosen”. Oxford.

Smith G ( 1953) .—Pioneer botanists. Journal of the
Historical Society of Western Australia. 4
( 5 ) : 59.

Staples A C. ( 1948 ) . — “The historical development of
the Harvey district up to 1895”. M.A. The=is
University of We-tern Australia.

Staples A. C. (1961). -The first ten years in the Swan
P ' River Colony: some old errors corrected.

Journal of the Historical Society of West-
ern Australia. 5 (7): 83.

qtpn T (1943). — ‘‘Agricultural development in the Hay
’ River area of West Australia” M.A. Thesis.

University of Western Australia.

Stephen R. (1956).— The Mingenew story. Journal of
the Historical Society of Western Austra-
lia. 5 (4): 59.

Stewart, A. ( 1948) .—Western Sussex (Busselton). tarty
Days. 3 (10): 32.

Thomson, J. A. (1971).— The Muir family .... from 1844.
Early Days. 8 (2): 19.

Underwood, E. J. and Filmer, J. F. ( 1935) .—Enzootic
marasmas: the determination of the biologi-
cally potent element (cobalt) in Limonite.
Australian Veterinary Journal. 11: 84.

Underwood, E. J. (1972). -The cobalt story. Records of
th.p Australian Academy of Science. 2 (2).

Webb M J (1968). — “ Planning and development in met-
ropolitan Perth to 1953” (presented to the
10th Congress of the Australian Planning
Institute, 18 August 1968). (unpublished
paper) .

William Dom (1961).— A survey of the correspondence
between Bishop Salvado and John Forrest.
Journal of the Historical Society of West-
ern Australia. 5 (8): 63.

Williams, D. (1959).— The Goomalling di trict. Journal
of the Historical Society of Western Aus-
tralia. 5 (6): 1.

Wollaston, J. R. ( 1948) .—“Picton Journal”. Perth.

Journal of the Royal Society of Western Australia, Vol. 56 Parts 1 and 2, July, 1973

64