Journal of the Royal Society of Western Australia

JOURNAL OF

THE

ROYAL

SOCIETY

OF

WESTERN

AUSTRALIA

Volume .6? » Part 3 • 1987

ISSN 0035-922X

THE

ROYAL SOCIETY
OF

WESTERN AUSTRALIA

PATRON

Her Majesty the Queen

VICE-PATRON

His Excellency Professor Gordon Reid, Governor of Western Australia

President

Vice-Presidents

Past President

Joint Hon. Secretaries

Hon. Treasurer

Hon. Librarian

Hon. Editor

COUNCIL 1986-1987

J. S. P. Beard, M.A., B.Sc., D. Phil.

J. T. Tippett. B.Sc., Ph.D.

J. S. Pate, Ph.D.D.Sc., FAA. FRS.

S. J. Hailam. M. A.

V. Hobbs, B.Sc. (Hons.), Ph.D.

K. W. Dixon. B.Sc. (Hons.), Ph.D.

W. A. Cowling. B..Agric.Sc. (Hons.), Ph.D.
H. E. Balme, M.A., Grad, Dip. Lib. Stud.
B. Dc!I, B.Sc. (Hons.), Ph.D.

J. Backhouse, B.Sc., M.Sc.

A. E. Cockbain, B.Sc., Ph.D.

S. J. Curry, M.A.

E. P. Hopkins, B.Sc., Dip.For.Ph.D.

L. E. Koch. M.Sc., Ph.D.

J. D. Majer, B.Sc., D.I.C.. Cert. Ed., Ph.D.
K. McNamara. B.Sc., (Hons.), Ph.D.

J. Webb. B.Sc.. Ph.D., Dip.Ed.

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1 987, p. 89-94.

Grazing pressure by the tammar {Macropus eugenii Desm.)
on the vegetation of Garden Island, Western Australia,
and the potential impact on food reserves
of a controlled burning regime

by David T. Bell, Janine C. Moredoundl and William A. Loneragan

Department of Botany, University of Western Australia
Nedlands, W.A. 6009

Manuscript received II February 1986: accepted 22 July 1986

Abstract

Tammar (Macropus eugenii Desm.) populations occur in a restricted number of Western Australia
mainland locations and two offshore islands. From faecal pellet epidermal remnants and analysis of
stomach contents, eleven plant species were documented as dietary species of the tammar
population of Garden Island. Of special preference were the dominant shrub Acacia rosteltifera,
introduced herbaceous species Asphadelus fistulosus and Asparagus asparagoides, and the native
grass Stipa flavescens. Grazing damage to these species was generally restricted to localized sites thus
assuring their continued survival on the island. Exclosure studies demonstrated a marked impact of
grazing on Asparagus asparagoides. Geranium mode and Stipa flavescens. Two small ground herbs,
Parietaria deoilis and Galium murale showed increases in cover in areas where larger species were
removed by grazing. The tammar was generally attracted to areas of new growth regenerating after
disturbance. This suggests that a planned fire management programme incorporating a system of
fire breaks, fire protection access routes and low fuel buffers should meet the perceived objectives of
protecting the naval installations, maintenance of the food resource and hence, continued survival
of the tammar populations on the island. The remaining areas of long-unburnt vegetation should be
retained to provide cover for the tammars and to conserve the native vegetation of the island,
especially fire sensitive species.

Introduction

The tammar (Macropus eugenii Desm.) is a herbivore
from the family Macropodidae. Its present range in
Western Australia is restricted and includes only Garden
Island, West Wallabi Island, and a few isolated pockets
on the mainland in the south-west. In the early 1970s the
number of tammars on Garden Island was ekimated at
2 000 (Bakker 1973) where as current sampling suggests
the population is around 1700 (A. Bradley, pers.
comm.). This small wallaby is also found in South
Australia and more panicularly on Kangaroo Island
(Kelsall 1965). Its distribution patterns in south-west
Australia have been associated with a range of site
factors, but of special importance are dense thickets of
vegetation which provide protection from predators and
associated grass species thought to be a preferred food
resource (Christensen 1981).

Other studies on the tammar have concentrated on
physiological factors (Kelsall 1965. Kinnear et ai 1968,
Bakker et al. 1 982). Any reference to the dietary habits
of the tammar in these studies was based on observation
of grazing damage and extrapolations from other
marsupial studies. Christensen (1981) studied the
biology of the tammar in relation to fire and listed
several major characteristics which might contribute to
the adaptation of the tammar to a fire-prone habitat.
These include: (1) seasonal breeding, (2) lack of
repression of juveniles, (3) group territoriality, (4) wide
dispersal, (5) absence of panic during fire, and (6) some
fidelity to home range. It is, however, the direct effects

of fire on the plant communities of Garden Island which
will ultimately affect this population since fire
protection is an important requirement of the recently
established naval facilities on this island.

Currently the W.A. Bushfires Board, W.A.
Department of Conservation and Land Management,
the Department of Defence (Navy) and the University of
Western Australia Departments of Zloology and Botany
are cooperating in a research effort to understand the
ecology of the tammar and to develop an appropriate
fire management plan that will both protect life and
property and ensure the short- and long-term health and
survival of this endangered species. The objective of this
study was to determine the plant species important in
the tammar diet and to make preliminary comments on
the potential impact that a fire management plan might
have on the population.

Methods

Garden Island (1I5°40’E, 32“I6’S) is a near-shore
island centred approximately 35 km southwest of Perth
and southward of Rottnest and Camac Islands. It
measures 9.5 km from north to south and 2 km at its
widest point. The islands are estimated to have been
separated from the Western Australian mainland for
6 000-7 000 (Main 1961). The major plant formations
are dense stands of shrub scrub dominated by Acacia
rostellifera and low forests dominated by Callitris
preissii and Melaleuca lanceolata. For a more detailed

1-53510

89

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

discussion of the major landforms and the vegetation
associations of the island, refer to McArthur and Bartle
(1981).

Exclosure studies

Four tammar grazing exclosures measuring 10 x 25 m
with 1.5 m high fences of 2 cm wire mesh were
established in a range of plant communities and regions
of Garden Island in May 1983 (Fig. I). Selection of the
exclosure sites was based on evidence of local grazing by
tammars, time since the last fire, and proximity to the
Zoology Department’s population sampling sites. The
Beacon Head Site, on the most northern portion of the
island, was situated in an area classified as Acacia
rostellifera Open Scrub; the Cliff Point Site near the
central western coastline in an area of Melaleuca
lanceolata Open Heath; the Denham Road Site in the

GARDEN ISLAND

N

I J

Figure 1 . — Map of Garden Island showing locations of grazing exclosures
(•) and major vegetation types (modified from McArthur and Bartle 1981).

90

centre of the southern half of the island in Melaleuca
lanceolata — Acacia rostellifera Low Open Forest, and
the Quarry Road Site near the southeastern tip of the
island in Melaleuca lanceolata Open Scrub. The Beacon
Head Site was burnt in December 1982. The Cliff Point
and Denham Road Sites probably last burned in the
disasterous fire of January 1956 which burned most of
the island north of Careening Bay (Baird 1958). The
Quarry Road Site lies in a region which has probably
remained fire-free for more than 40 years.

Percentage cover values for all vascular species were
visually estimated from permanently marked transects
of ten 1 m^ quadrats. One transect was situated inside
and one immediately adjacent but outside the exclosure
at each site. Cover values for selected species were
estimated in August 1983 and a full survey of all species
was made in September 1985. From these estimates
mean percentage cover values for species in each
transect were determined.

Tammar diet studies

The dietary preferences of the tammar were
determined by analysing the epidermal remnants from
faecal pellets and from the stomach contents of two
road-killed animals. A total of 40 faecal pellets were
collected; 5 from outside each of the tammar grazing
exclosure areas and 20 from a range of habitats generally
distant from the Naval installations. The emphasis on
native areas was made to reduce likely grazing by
tammars of the irrigated lawns and domestic gardens
within the settlement area which is known to occur
(McArthur and Bartle 1981). One series of samples was
obtained prior to the onset of winter rains (April, 1983)
and a second in late winter (August, 1983). The road-
killed animals were obtained during April 1983 from an
area just north of the North Gate of the Naval
installation. The stomach contents were removed
immediately upon collection and frozen for later
analysis.

The identification of epidermi in the faecal pellet and
stomach content materials followed the methods of
Halford et al. (1984a) which were modified from the
original techniques reported by Storr (1961) and Jain
(1976X Comparisons of faecal pellet epidermal remnants
to epidermal samples prepared from plant tissue
provided information on tammar diet preferences.
Frequencies of the proportions of the epidermal
fragments from microscopic analysis provided an
estimate of dietary preference (Halford et al. 1 984b).

Direct grazing observation and re-establishment response

Plants throughout the island were checked for direct
signs of grazing and. where tammars could be observed
grazing or browsing particular shrubs, these species were
recorded. Plant species were also categorized for re-
establishment strategy. Disturbance to a plant
community can result from fire damage, animal grazing
or the mechanical removal of vegetation such as occurs
in firebreaks or alongside roads. Post-fire modes of re-
establishment have been widely documented (Specht et
al. 1958, Keeley and Zcdler 1978, Keeley and Keeley
1981, Malanson and O’Leary 1982, Bell et al. 1984). In
contrast there is a paucity of information concerning re-
establishment strategies following grazing damage and
mechanical disturbance. In relation to fire, species can
be described as ephemerals, obligate seeders and
sprouters (Bell et al. 1984), although re-establishment
strategies may vary within a particular species as well as

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

between species. Mode of re-eslablishmeni may also
vary depending upon the intensity of the fire or the
degree of damage sustained from non-pyric
disturbances. In the main, fire prone environments are
high in the proportion of resprouting species (Siddiqi el
al. 1976. Bell el al. 1984). Similarly, areas subjected to
repeated mechanical disturbance might favour
resprouters, although obligate seeders with a
bradysporous habit may also survive clearing operations
(Griffin and Hopkins 1981). Similar information is not
available for grazed plants, but it might be assumed that
resprouting species would, in the long term, have a
greater chance of survival than obligate seeding species.
The collection of re-esiablishmcnt data allowed the
prediction of changes which might occur following an
imposed controlled-burning regime.

Results

Exclosure studies

Twenty-nine species of vascular plants were identified
in and adjacent to the four lammar cxclosure study sites
(Table 1 ). Only Acanlhocarpus preissii and Phyllanthus
calycinus occurred in every' transect and only eight
species occurred in more than half of the study transects.
Of these common species, those showing major
differences between the inside and the outside of the
exclosurcs were Acanlhocarpus preissii. Asparagus
asparagoides. Geranium molle and Stipa JJaveschis,
which were much more common inside, arid Galium
murale, Parietaria dehilis and Phyllanthus calycinus.
which showed consistently higher cover values outside
the grazing exclosures. The most obvious effect of the
grazing exclosure occurred at the recently burned
Beacon Head site where Asparagus asparagoides

averaged 78.9% under protection and only 1.8% where
exposed to grazing.

Comparisons of mean cover values of the common
species after about four months grazing protection
(1983) and after more than two years protection (1985)
showed an increase in the difference between protected
and unprotected samples for those species with greater
cover inside the exposures (Fig. 2).

Faecal and stomach content studies

Eleven different species of vascular plants were
identified in the tammar faecal pellet material and
stomach contents (Fig. 3). The most common were
Asphodelus fistulosus, Acacia rostellifera and Asparagus
asparagoides. The August faecal pellet samples included
a greater variety of species (11 species) compared to
either the April faecal pellet sample (5 species) or the
April stomach contents sample (8 species).

Re-establishment following grazing or fire

Direct obsciwaiions of tammar grazing proved
difficult although seven plant species were confirmed as
dietary species. These seven species, however, had
already been identified from faecal pellets or stomach
content analyses. Of significance was that the tammars
seemed to prefer young shoots of the resprouting species
or seedlings.

Of the forty-seven species identified from the 1983
and 1985 samplings, only ten are known to re-establish
from existing rootstocks following intense fires with the
remainder either ephemerals (16 species) or obligate
seeders (20 species) (Table 2). Of this total sample, 26%
(12 species) were introduced, and, of the eleven dietary
species, 27% (3 species) were introduced.

Table 1.

Mean percentage cover values for transects inside and outside the Garden Island Tammar exclosure study sites for data measured September 1985.

Species

Acacia rostellifera
Acanlhocarpus preissii
Anagallis spp.*

Aspara^is asparagoides
Asph odelus /isiulosus
Carduus pycnocephalus
Clematis microphylla
Crassula cnlorata
Daucus glochidiaius
Eremoptula glabra
Galium murale
Geranium molle
Guichenotia ledifolia
Ilardenbergia compioniana
Leucopogon insularis
Leucopogon parvfJorus
Oxalis pes-caprae
Poranlhera microphylla
Parietaria debilis
Phyllanthus calycinus
Rhagodia haccata
Senecio lautus
Solanum svmonii
Sonchus oleraceus
Spyndium glohulosum
Stipa Jlaxescens
Thnmasia cognala
Trachyandra di varicata
Zaniedeschia aelhiopica

Beacon

Head

In Out

— 3.1

5.8 1.9

5.2 1.4

78.9 1.8

— 1.4

— 1.1

— 0.1

0.7 22.4

4.6 48.8

— 0.4

— 0.1

— 0.1

0.5 6.5

— 0.7

3.6 0.2

Cliff

Point

In Out

0.1 0.3

12.5 7.4

— 1.7

12.7 0.2

1.3 10.1

2.7 —

7.3 0.4

6.8 28.2

23.7 0.1

1.9 —

9.5 —

0.9 8.8

0.5 —

10.2 0.1

0.2 —

— 0.5

Denham

Road

In Out

2.0 43.7

24.0 13.7

1.1 —

— 0.2

— o.s

— 0.2

16.1 —

6.2 56.1

0.4 1.1

0.3 —

0.1 —

5.8 0.1

2.6 0.2

Quarry

Road

In Out

86.0 62.4

— 0.1

36.1 0.1

— 0.1

— 0.4

1.0 0.8

6.9 —

8.1 —

0.1 2.1

0.1 —

— 0.1

3.4 4.2

0.2 7.8

0.4 —

0.4 —

0.3 —

0.5 —

— I.O

— 1.3

3.2 0.6

*Includes both Anagallis arvensis and A. foemina

91

2-53510

MEAN PERCENTAGE COVER MEAN PERCENTAGE COVER

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

35 _

Acanthocarpu8

preissli

Asparagus Geranium

asparagoides molle

1983 1985 1983 1985

OUTSIDE

1 983

1985

20 _

Stipa

flavescens

1 5 .

10 _

rTL-N-

1983 1985

Par ietaria Galium

1983 1985 1983 1985

Island, direct grazing evidence was not apparent. Kelsall
(1965) noted that Ihc tammar preferentially grazed this
species in the dr\^ season. In this study no seasonal
preference for Acacia rostellifera occurred between the
late Autumn first sampling period and the late Winter
second sampling when water would be readily available.
During summer the thick central parenchymatous tissue
of the phyllods of this species could be an important
source of moisture. Preference for legume species could
also relate to greater foliar nitrogen levels (Halford et al.
1984b).

Previous research established that grasses were a
dietary preference in mainland populations of the
tammar (Christensen 1981). Stipa flavescens has a
relatively wide but discontinuous distribution over
Garden Island occurring in higher densities in scrub
communities where Acanthocarpus preissii was either
absent or present in low densities. In these regions Stipa
flavescens tussocks often showed evidence of grazing
damage. The limited evidence for this species in the
faecal and stomach content analyses, however, could be
mainly an artifact of the laboratory technique. The
nitric/chromic acid maceration technique entirely
digests the epidermi of grasses and the digestion
processes of the tammar could duplicate this process.
Cautionary procedures regarding grass epidermal
remnants have been pointed out previously (Storr 1961.
Halford 1984a).

The Beacon Head exclosure site located in a small
experimental burn area show'ed vigorous regrowth of
Asparagus asparagoides. The major cover differences
between areas inside and outside the fences and the
common occurrence of this plant species in both the
faecal and stomach content samples implicate
asparagoides as a favoured dietary component. Evidence
of near complete digestion of epidermi of young leaves

Figure 2. — Mean percentage cover values from the four exclosurc sites for
selected species sampled inside and outside the cxclosure during
August [983 and .September 1985.

Discussion

The tammar wallabies on Garden Island appear to be
versatile feeders. Preferred plants generally included
young shoots from resprouiing species {Stipa flavescens.
Asparagus asparagoides). seedlings {Sohnum symomi.
Thomasia cognata) or short-li\ed ephemeral species
{Asphodehis fhlulosus). Results obtained from the
examination of faecal and stomach material indicated
that the introduced ephemeral. Asphodelus flistidosus.
dominated the diet of the tammar. This species is now
particularly widespread over the island, although
differences in cover values were not particularly
apparent in the exclosurc studies due to a tendency of
this species to be concentrated in disturbed areas such as
along road verges and in fire breaks. At the Cliff Head
site in the grazed area surrounding the exclosure, many
individuals appeared to have invaded the area following
an initial period of grazing. Grazing observations
suggested that the tammar preferred areas where recent
mowing or ploughing stimulated the production of new
shoots or seedlings.

The presence of Acacia rostellifera in both faecal and
stomach material indicated a strong preference for this
species. Other authors have reported various species of
Acacia in the diet of the tammar and other Macropods
(Storr 1961, Kelsall 1965, Christensen 1981. Halford et
al. 1984b). Probably because of the high density and
widespread distribution of Acacia rostellifera on Garden

PERCENTAGE OF FRAGMENTS IN SAMPLE
0 10 20 30 40 50 60

Figure 3. — Proportions of fragments separated by species from the April
and August, 1983 faecal pellet samples and the April. 1983 stomach
content samples.

92

Journal of the Royal Society of Western Australia, Vol. 69, Part 3. 1987.

of this species in the preparation of the voucher slides
suggests that fragments noted from the faecal pellet and
stomach content analyses probably underestimate the
actual proportions ingested.

Grazing damage to Rhagodia baccata on Garden
Island has been previously documented (McArthur
1957, Kelsall 1965. McArthur and Bartle 1981). The
absence of this species in the dietary preferences noted
in this study, however, is probably due to the particular
location of the exclosure sites in respect to the
occurrence of Rhagodia baccata. Other inconsistencies
arising belw-ecn the direct observation of grazing
pressure and the quantitative evidence may also be
explained by the distribution and density patterns of
individual species and/or relatively low number of
pellets analysed. For example, CaUitm preissii, which
did not occur in the transects, was identified from the
stomach contents of the road-killed animals.

In the twelve years since completion of the new naval
facilities the tammar population has remained relatively
constant: the difference of 300 animals between the pre-
1973 and current estimates could be attributed to the
general difficulty of accurately sampling animal
populations. Given the feeding preference for young
shoots and seedlings, combined with the known

Characteristics of Garden Island vascular plants relating to

behaviour of the tammar during fires (Christensen
1981), a controlled burning regime on Garden Island
could benefit the population of this rare marsupial by
increasing the areas of regenerating vegetation.
However, the large proportion of species requiring
seedling re-establishment indicates that the vegetation of
Garden Island does not normally have a regime of
frequent natural burns and, therefore, it would be
irresponsible to suggest that large areas of the native
vegetation be burned to provide greater feeding areas for
the tammar. In fact, the greater availability of preferred
food following large area burns could result in
unacceptably large population increases in the tammar
population. Frequent fires could also increase the
already considerable invasions of introduced species
into the native plant communities of the island as has
been noted elsewhere in small 'island-like* bushland
reserves within metropolitan Perth (Baird 1977,
Loneragan ei al. 1984. Wycherley 1984). The present
proportion of introduced species on the island (26%) is
similar to that included in the Star Sw'amp Bushland
Reserve (25%) (Bell et al. 1979) even though the
mainland reserve includes a greater diversity of habitats,
communities and species. What could prove beneficial
to both the tammars and the human population on
Garden Island would be a series of control burns to

2 .

', re-establishmeni strategy, tammar diet, and floral afTiIiation.

Acacia cochlearis
Acacia rostellifera
Acanthocarpus preissii
Anagallis arvensis
A namllisfopmi nn
Aspara^s asparagoides
. isphodelus fisiulosus
Beyeria viscosa
Boronia alata
Callitris preissii
Carduus pycnocephalus
Carpohrotus aequilaterus
Cenlaurium erythraea
Clemaiis niicrophylla
Crassula colorata
Crassula pedicellosa
Diploleana dampiera
Daucus glochidiatus
Eremopnila glabra
Exocarpus sparteus
Galium murale
Geranium molle
Guichenotia ledifolia
Hardenbergia comptoniana
Xfelaleuca huegelii
Melaleuca lancenlata
Lastapelalum opposiiifolium
Leucopogon insularis
Leucopogon parviflorus
Olearia axillaris
Oxalis pes-caprae
Parietaria debilis
Poranihera microphylla
Phytlamhus calycinus
Rhagodia baccata
Scaevola crassifolia
Scirpus nodosus
Senecio lautus
Solatium svmonii
Sonchus oleraceus
Spyridium globulosum
Stipa flavescens
Thomasia cognata
Trachyandra divaricata
Trachymene caerulea
Trachymene pilosa
Zantedeschia aeihiopica

Observed

grazing

damage

Re-establishment

strategy

Faecal

pellet

Stomach

content

Floral

affiliation

Resprouter

Native

Yes

Resprouter

Yes

Yes

Native

—

Resprouter

—

—

Native

—

Ephemeral

—

—

Introduced

—

Ephemeral

—

—

Introduced

Yes

Resprouter

Yes

Yes

Introduced

Yes

Seeder

Yes

Yes

Introduced

—

Seeder

—

—

Native

—

Seeder

—

—

Native

—

Seeder

Yes

Yes

Native

—

Ephemeral

—

—

Introduced

—

Seeder

—

—

Native

—

Ephemeral

—

—

Introduced

—

Resprouter

—

—

Native

—

Ephemeral

—

—

Native

—

Ephemeral

—

—

Native

—

Seeder

—

Native

—

Ephemeral

—

—

Native

—

Seeder

Yes

Yes

Native

—

Resprouter

—

—

Native

—

Ephemeral

—

Introduced

—

Ephemeral

—

Introduced

—

Seeder

—

—

Native

—

Resprouier

—

Native

—

Seeder

—

Native

—

Seeder

—

Native

—

Seeder

Yes

—

Native

—

Seeder

—

Native

—

Seeder

—

Native

—

Seeder

—

Native

—

Ephemeral

—

Introduced

—

Ephemeral

—

Native

—

Seeder

Native

Yes

Resprouter

Yes

Yes

Native

—

Seeder

—

Native

—

Seeder

—

—

Native

—

Resprouter

—

—

Native

—

Ephemeral

—

Native

Yes

Seeder

Yes

Yes

Native

—

Ephemeral

—

Introduced

—

Seeder

—

Native

Yes

Resprouter

Yes

Native

Yes

Seeder

Yes

Native

—

Seeder

Yes

Yes

Introduced

—

Ephemeral

—

Native

—

Ephemeral

—

—

Native

Ephemeral

—

—

Introduced

93

Journal of the Royal Society of Western Australia, Vol. 69. Part 3. 1987.

create low flammable fuel regions around the Naval
facilities and to break up the length of the island into
units to prevent large areas burning as occurred during
the summer of 1 956 (Baird 1958).

Garden Island has been largely free of wild fires
during the history of European settlement on the
mainland. Indications from growth ring analyses from a
small stand of Callitris preissii located in the northern
end of the island which escaped the fire of 1956 showed
that the trees were not much older than 50 years
(Pearman 1971). This stand would now be around 65
years old. .Although growth is generally slow, fuel build
up during fire free periods can be considerable.
McArthur (1957) reported that after 18 years of
accumulation litter depths w'ere of the order of 5 cm in
the Callitris Forest, 3 cm in Acacia rostellifera-M'wQd
Scrub, and 1 cm in Melaleuca heugelii Scrub. His
description of the vegetation at the time 4yr prior to the
1956 fire also indicated that Stipa flavescens occurred in
most of the plant communities and would have
contributed to the flammability of the litter.

Small spot fires accidentally lit by island visitors have
occurred since the 1956 fire but documentation of these
fires is unavailable. Controlled burning trials carried out
by the W. A. Bush Fires Board in conjunction with the
Department of Botany in April 1982 and December
1982 in the Beacon Head region were the first
documented fires for 27 years. Estimates following these
small experimental control burns indicated that only
about 50% of the lit area actually burnt, and indeed, on
both occasions some difficulty was experienced in
keeping the fires alight. These controlled burns were
carried out under mild temperatures and in the absence
of strong winds; in marked contrast to the conditions
under which the 1956 fire burned when temperatures in
the two weeks preceeding the wild fire averaged 38"C
and strong south-westerly winds had orevailed.

The ability to impose a controlled burning regime in a
region which normally only rarely receives a fire should
not be the only management decision for the regions of
native vegetation on Garden Island. As was noted by
Krinitskii (1974), ‘Man's help should be thoroughly
worked out: he should not lightly and arrogantly recarve
nature'. The protection of the Naval installations from
wild fires could be achieved by a system of fire breaks
and reduced fuel-load buffers, rather than the burning of
large tracts of the native plant communities of the island
which has the additional disadvantage of destroying
most of the cover for the tammars. If the fire-breaks
were of a firm-base construction (e.g. limestone) rather
than ploughed annually, there would be minimal
disturbance and, hence, greater probability of
controlling weed invasion. Changes in the tammar
population numbers that result from the increased areas
of preferred food resources could then indicate further
considerations for the management of the island.

Acknowledgements — Funds for the research were provided from Ihe
Garden Island Management Research Project. We wish to acknowledge the
assistance of Professor S. D. Bradshaw and Dr. A. J. Bradley of the
University of Western Australia Department of Zoology. The position of
Senior Lecturer in Plant Ecology to Dr. Bell is supported by Alcoa of
Australia Ltd. and Western Collieries Ltd. The .study was part of an
Honours in Botany research project of Ms. Moredoundl.

References

Baird, A. M. (1958). — Notes on the regeneration of vegetation of Garden
Island afterthe 1956 fire. J. Roy. Soc. IV.A.. 14: 102-107.

Baird. A. M. (1977). — Regeneration after fire m King’s Park, Perth.
Western Australia. / Roy Soc. H'.A.. 60: 1-22.

Bakker. H. R. (1973). — W'ater and electrolyte metabolism of the tammar.
Unpubl. Ph.D. Thesis, University of Western Australia.

Bakker. H. R., Bradshaw. S. D. and Main, A. R. (1982). — Water and
electrolyte metabolism of the tammar wallaby. Macropus eugenii.
Physwl. ZooL 55:209-219.

Bell. D. T., Loncragan. W. A. and Dodd, J. (1979). — Preliminary analysis
of the vegetation of Star Swamp. Western .Australia. li'.A.
Herbarium Res. Soles. 2: 1-21.

Bell. D. T.. Hopkins, A. J. M. and Pate, J. S. (1984). — Fire in the kwongan.
Pages 178-204. In Pate. J. S. and Beard. J. S. Kwongan (Eds). Plant
Life of the Sandptain. University of Western Australia Press,
Nedlands. Western Australia.

Christensen. P. (1981). — The biology of Bettongia penicUlata Gray and
Macropus eugemi Desm. in relation of fire. \V..4. For. Dep. Bull.

91.

Griffin. E. A- and Hopkins, A. J. M. (1981). — The short term effects of
brush harvesting on the kwongan vegetation at Eneabba. Western
Australia. B'w. Aust. Dep. Fish. & H'ildl. Rept. So. 45,

Halford. D. A.. Bell. D. T. and Loneragan. W. A. (1984a). — Epidermal
characteristics of some Western Ausialian wandoo-woodland
species for studies of herbivore diets. / Roy. Soc. W.A., 66: 1 1 1-
118.

Halford, D. A.. Bell. D. T, and Loneragan. W. A. (1984b). — Diet of the
western grey kangaroo {.\facropus fuliginosus Desm.) in a mixed
pasture-woodland habitat of Western Australia. J. Row Soc. W.A.,
66:119-128.

Jain. K. K. (1976). — Hydrogen peroxide and acetic acid for preparing
epidermal peels from conifer leaves. Stain Tech., 51: 202-204.

Keeley. J. E. and Keeley, S. C. (1981). — Post-fire regeneration of southern
California chaparral. Amer. J. Bot , 68; 524-530.

Keeley. J. E. and Zedler. P, H. (1978). — Regeneration of chaparral shrubs
after fire: A comparison of sprouting and seeding strategies. Amer.
.Midi Nat . 99: 142-161.

Kelsall. J. P. ( 1 965). — Insular variability in the tammar {Macropus eugenii)
of Western Australia. Unpubl. Ph.D. Thesis. University of
Western Australia.

Kinnear. J. E., Purohit. G. and Main. .A. R. (1968). — Ability of the tammar
wallaby {Macropu.s ingenit. Marsupialia) to drink sea water. Comp.
Biochem. Physiol. 25: 761-782.

Krinitskii, V. V. (1974). — Protected areas in the world’s industrially
advanced regions; importance, progress and problems. Pages 61-
67. In Ellioil. H. (Ed.). Second World C'onference on National
Parks. 1-U. C. N.. Morges. Switzerland.

Loneragan, W. A., McMillan. P.. Townley, L. R. and Watson. L. E.
(1984). — Star Swamp Bushland Reserve. Proposals for its
Development and Management. Report to the .Australian Heritage
Commission. Canberra.

Main, A. R. (1961). — The occurrence of Macropodidae on islands and its
climatic and ecological implications. /. Roy. Soc. IF A.. 44: 84-89.

Malanson. G. P. and O’Leary. J. F. (1982). — Post-fire regeneration
strategies of Californian coastal sage shrubs. Oecologia, 53: 355-
358.

McArthur, W. M, (1957). — Plant ecology of the coastal islands near
Fremantle, Western .Australia. ./ Roy. Sck\ 40: 46-64.

McArthur. W. M. and Bartle, G. .A. (1981). — The landforms, soils and
vegetation as a basis of management studies on garden Island.
CSIRO Div. Land Resour. Manage. Ser\. Bull. No. 7.

Pearman. G. I. (1971). An exploratory investigation of the growth rings of
Callitris presissii trees from Garden Island and Naval Base. West
. lust. Nalur.. 12: 12-17.

Siddiqi. M. Y., Carolin, R. C. and Myerscough. P. J. (1976). — Studies in
the ecology of coastal heath in New South Wales. Australia. Part 3.
Regrowth of vegetation after fire. Proc. Linn. Soc. M'.. 101: 53-
63.

Specht. R. L.. Rayson. P. and Jackman. M. E. (1958). — Dark Island heath
(Ninety-Mile Plain, .South .Australia). VI. Pyric succession:
Changes in composition, coverage, dry weight, and mineral
nutrient content. Aust. ./- Bot.. 6: 59-88.

Storr, G. M. (1961). — Microscopic analysis of faeces, and technique for
ascertaining the diet of herbivorous mammals, .-iiisf. J. Biol. Sci..
14: 157-164.

Wycherley, P. (1984). — People, fire and weeds: can the vicious spiral be
broken? In Moore. S. (Ed.), .\fanagement of Small Bush .Areas in
the Perth Metropolitan Region. Proc. of a Seminar held by Dept.
Fisheries and Wildlife 20th Sept. 1983.

94

Journal of the Royal Society orWeslcrn Australia. Vol. 69. Part 3. 1987. p. 95-1 12.

Wetlands of the Darling System — A geomorphic approach to

habitat classification

by C. A. Scmcniuk

21 Cilcnmerc Road, Warwick. W.A. 6024
Maniisaipi nrciwei 17 June I^S6: acccpicd IS November 1986

Abstract

A classiflcalion utilising the 2 primary components of wetlands, the “wetness" and "landform"
components, is proposed here. The water component distinguishes wetlands from other terrestrial
habitats and also influences biological response by its presence, permanence or longevity, depth,
chemistry, and movement. The landform determines wetland size, shape and depth. Using
subdivisions of cross-sectional landform geometry there arc recognised: basins, channels, and Hats.
Within the category of water longevity, there arc recognised: permanent inundation, seasonal
inundation, and seasonal waterlogging. Combining these “wetness" and landform allribulcs
provides 7 categories of common wetlands: I. permanently inundated basin lake: 2. seasonally
inundated basin sunipla/u/: 3. seasonally waterlogged basin dampland: 4. permanently
inundated channel river, 5. seasonally inundated channel creek: 6. seasonally inundated flat
floodplain: and 7. seasonally waterlogged flat - pafusplain. Water and landform

descriplors/modifiers are used to further augment the nomenclature of the primary units. Modifiers
for water include salinity and its consistency. Modifiers for landform include size and shape. Since
there arc only seven primary wetland types, the classification provides a practicable number of
categories for mapping. The addition of more precise or detailed information as
modifiers/descriptors increases the abilily to discriminate individual wetlands from each other.

Introduction

This paper presents a geomorphic classification of
wetlands in the Darling System (Swan Coastal Plain and
Darling Plateau; Fig. 1). an area occurring in the
subhumid and humid region of soulhw'cstern .Australia
and encompassing coastal plains and dissected plateau.
Many classifications of wetlands to date in Western
Australia and elsewhere have been based on vegetation
and water quality but it is considered that a geomorphic
classiflcalion best provides the initial framework to
understanding the various types of wetlands, their
distribution and their relationship to biota. The
rationale of the geomorphic approach is that ultimately
wetlands arc related fundamentally to landform
development and water maintenance.

The classification presented here may be used for
recognising the varied wetland types that occur in the
Darling System. Information such as this is needed to
establish a geographic, stratigraphic, hydrological and
biological pattern in the distribution of wetland types, to
determine regional and local significance of wetlands, to
determine conscr\ation strategies.and to manage
wetlands.

It has long been recognised that wetlands range from:
permanent lakes, small to large seasonal lakes, small to
large areas of seasonally water-logged soils, fluviatile
systems, estuarine systems, and marine systems, and
that these categories can be inicrgradational. Figure 2
illustrates the intergradalional relationship between, and
attributes of. the various wetland systems, recognising
that there are land-based, marine-based and
intermediate (estuarine) categories. This paper deals
with the land-based wetlands. Estuarine systems and
marine systems will be the subject of a later study. The
approach of this paper is to provide a review of

international literature on wetland classification and
nomenclature followed by a review of local studies that
deal with wetlands of the Darling System. This is in turn
followed by a description of the classification adopted
here.

Review of international literature on classification and
terms for land-based wetlands

Definition of wetland

In the more recent literature the definition of wetlands
encompasses lakes, water-saturated basins, estuaries and
fluvial systems (UNESCO 1971. Bayly and Williams
1973, Cowardin et ai 1979, Department of
Conservation and Environment 1980, Adam et ai
1985). However in the older literature the concept of
wetland was more rigidly confined to encompass only
lakes and water-saturated basins.

A variety of wetland definitions from the international
literature is presented here to indicate the wide range of
concepts of what constitutes a wetland. The Ramsar
Convention defined weilands as “areas of marsh, fen,
peailand or water, whether natural or artificial,
perrnanent or temporary, with water that is static or
flowing, fresh, brackish or salt, including areas of marine
water the depth of which at low tide docs not exceed
6m" (UNESCO 1971). Zoliai and Pollel (1983) define
wetlands as areas where wet soils arc prevalent, having a
water table near or above the mineral soil for most of the
thawed season, supporting a hydrophilic vegetation, and
pools of open water (-^^2m deep). This includes shallow
open water. It does not include areas that become
temporarily flooded, but remain relatively well drained
for most of the growing season. Hill (1978) identifies

3-53510

95

Journal of the Royal Society of Western Australia, Vol. 69. Part 3. 1987.

Figure 1. — Location of the study area, the System Six Region in
southwestern Australia.

wetlands by the presence of vegetation typically adapted
to life in areas inundated or saturated by water with the
appropriate duration and frequency to promote that
vegetation, Cowardin el aL (1979) define wetlands as
lands transitional between terrestrial and aquatic
systenjs where the water table is usually at or near the
surface or the land is covered by shallow water.
Cowardin el ai (1979) define wetlands as having one or
more of the following attributes: I) at least periodically
the land supports hydrophytes, 2) the substrate is
predominantly undrained hydric soil 3) the substrate is
nonsoil and is saturated with water or covered by
shallow water at some lime during the growing season of
each year. Bates and Jackson (1980) define wetland as a
general term for a group of wet habitats and include
areas that are permanently wet and/or intermittently
water covered, such as coastal marshes, tidal swamps
and flats, and associated pools, sloughs and bayous.

A consensus of Western Australian workers define
wetlands as:

“Areas of seasonally, intermittently or permanently
waterlogged soils or inundated land, whether
natural or otherwise, fresh or saline, e.g. waterlogged

soils, ponds, billabongs, lakes, swamps, tidal fiats,
estuaries, rivers and their tributaries’' (Wetlands
Advisory Committee 1977).

The definition is adequately encompassing and is
adopted herein.

II 'etland classificalions

Wetlands have been examined and/or classified from
a number of disciplines: biologically (e.g. mires, swamps;
Ivanov 1981, Gore 1983): physically (e.g. riverine lakes,
glacial lakes, tectonic lakes; Hutchinson 1957);
biologically/chcmically (e.g. fens, bogs: Tansley 1939,
Gorham 1957, Reeves 1968, Moore and Bellamy 1974,
Ruttner 1975, Gore 1983); chemically (e.g.

minerotrophic, ombrotrophic etc.; Kulczynski, 1949;
Hakanson and Jansson 1980, Wetzel 1983);

ontogenetically: and others. It is not proposed to present
a historical review of all the terms nor is it intended to
explore the meanings of terms w'hich arc not widely
used. Rather this review documents current
terminology, discusses the relevance of those terms to
the present classification, and assesses their applicability
to the wetlands in the Darling System.

I'erms for types of wetlands

Currently inland water bodies can be separated into 4
types; rivers, lakes, mirelands and marshlands. Rivers
refer to channelled surface water. Lakes, mirelands, and
marshlands arc “basin-like" or flat wetlands which are
differentiated on the basis of plant colonisation. The
recent distinction between running water (Lolic
environment) and standing water (Lenlic environments)
basically identifies the broader categories of riverine

BROAD CATEGORIES OF
WETLAND SYSTEMS

GEOMORPHIC/HYDROLOGIC

SYSTEM

SALINITY

CLOSED

Basins

and

Flats

OPEN

Riverine

Deltaic

and

Estuarine

Fresh

o

Mixed or
Alternating

Marine

Saline

Stippled area marks the categories of land-based wetlands
that are the subject of this paper

Figure 2. — Conceptual summary of types of broad wetland categories and
their inter-relationships in terms of geomorphic/hydrologic setting and
their salinity.

96

Journal of the Royal Society of Western Australia, Vol. 69. Part 3, 1987.

versus ‘'basin-lypc" of wetlands. The fluvial systems
(encompassing rivers, streams, creeks and all other
forms of established surfacx* water flow) generally in the
past were not considered to be wetlands. Since the
documentation and analysis of fluvial systems have been
separated traditionally from that of other types of wet
habitats, the approaches to their classification have been
variable (Schumm 1977. Leopold el al. 1964, Twidale
1968, Fairbridgc 1968, Bloom 1978).

Fluvial Welland systems

Fluvial systems already have been described and
categorised according to geomorphic principles (Gilluly
and Waters 1951, Schumm 1977. Leopold el al. 1964,
Chorlcy 1969, Bloom 1978. Strahler 1978, Cant 1982.
Lecdcr 1982). Both land and water criteria have been
combined to identify and nominate the features of a
fluvial system. The recognised geomorphic features of
fluvial systems include channels. Hood basins and
floodplains, levees, crevasses, sw'amps, and ox-bow lakes
(op cit).

Fluvial channel patterns have been subdivided by
workers using the criteria of I) genetic relationship to
landform. 2) channel size and position in relationship to
branching pattern. 3) sinuosity, 4) water permanence
and 5) sediment load. The genetic classification used
terms such as subsequent, consequent, obsequcnl and
insequent. and the terms "first order stream”, "second
order stream”, etc were used to categorise relationship of
channel to branching pattern. Several classes of channel
are recognised on the basis of sinuosity (i.e. the ratio of
channel length to valley length): these include straight,
transitional, regular, irregular, tortuous, sinuous,
meandering, straight-and-braided, and anastomosing
(Schumm 1977, Leopold el al. 1964, Miall 1977. Cant
1982, Smith 1983); some of these terms arc
synonymous. C'hanncis occupied by water for different
periods of time are referred to as permanent,
intermittent, and ephemeral. Predominant sediment
transport mode also has been used to classify channels
(Schumm 1977) resulting in a similar set of classes as
those derived from sinuosity criteria, but with the
additional information on channel stability i.e. whether
the channel is stable, eroding, or depositional.

Basin wetland systems and flats

Research into “basin-type” inland water bodies or
wetlands, has been largely concentrated in the areas of
the Temperate, Boreal and Arctic zones of the Northern
Hemisphere, particularly Europe, where precipitation is
greater than evaporation and/or drainage. The wetlands
of these regions arc permanently inundated or
waterlogged areas and the terms derived from these
areas tend to describe only this limited range of
wetlands. Examples of wetland terms from these regions
are (Gore 1983): lake and mircland (English), moor, sec
(German), myr (Swedish), suo (Finnish), boloto
(Russian), veen (Dutch), myri (Icelandic) and muskeg
(Canadian).

Basins were divided, on the presence or absence of
emergent vegetation, into 2 water depth types, i.e. those
deeper than 2m and those shallower than 2m. The
current terms used to distinguish these types arc lake
and mireland. Lake has been defined as an “inland body
of standing water occupying a depression in the earth's
surface, generally of appreciable size (larger than a pond)
and too deep to permit vegetation (excluding

subaqueous vegetation) to take root completely across
the expanse of water: the water may be fresh or saline”
(Bates and Jackson 1980). Mircland refers to any area
which remains waterlogged and is vegetated (Ivanov
1981).

Lakes have been classified for specific purposes,
according to dissolved mineral content, or thermal
properties, but the most widely used classification is
according to origin (Hutchinson 1957. Reeves 1968,
Hakanson 1980, and Jansson 1983, Wetzel 1983). Thus
tectonic, glacial, volcanic, solution lakes etc., have been
distinguished.

Mirelands have been subdivided into mires and
swamps. Mires are permanently waterlogged areas which
must include a minimum thickness of peat (Gorham
1957, Moore and Bellamy 1974, Ivanov 1981). They are
further divided into two types: bog and fen. A bog is
domed, raised above the level of the surface of the
surrounding terrain, often with a steep marginal bank
(rand). Bogs arc oligolrophic. because they rely on
atmospheric precipitation, and as a result support a
typical vegetation of Sphagnum mosses. A fen occurs in
hollows or depressions, is minerotrophic and as a result
supports a variety of vegetation including mosses,
graminoids, or trees. Swamps arc permanently
waterlogged seasonally inundated, or permanently
inundated, vegetated areas. Swamps have alternatively
been identified or defined as peat-forming, and non-peat
forming (Holmes 1944, Golct and Larson 1974, Ivanov
1981). Swamps arc subdivided into numerous different
types according to the physiognomy of plant cover
(Hofsteller 1983, Anderson 1983, Gore 1983. Denny
1985, and others), e.g. conifer swamps, hardwood
swamps, reed swamps, and Ti-lrec swamps.

Wetlands other than mirelands, lakes and rivers, arc
fewer in number in the Northern European climatic
zones, but they exhibit enough common similarities in
terms of biotic and water parameters to be classified
collectively as marshlands. Marshlands have been
defined as seasonally inundated or waterlogged, with or
without a well developed peat, and supporting a
graminoid or herb plant cover (Gorham 1953, Riggcri
1964-1966, Zollai and Bollet 1983. Golct and Larson
1974, Campbell 1983, Hofsletler 1983. Denny 1985).
The term has also been applied to vegetated peripheral
flats of estuaries. Marshlands may be "basin-like” or
flats. The types of marshlands arc: 1) marshes, 2)
swamps, 3) meadows, and 4) wetlands. The range and
inconsistency in terms have been brought about by the
variability in marshlands occurring outside Europe.
Authors describing areas which satisfy only part of the
marshland definition (e.g. the seasonal waterlogging of
soils) but have different vegetation structure, often use
an alternative term or a modifying term to avoid
confusion.

Marshes arc subdivided on basis of saline and fresh
water, geographical distribution (e.g. coastal and inland),
and water depth during the growing season (Martin el al.

1 953, Golct and Larson 1 974). The term swamp, used in
the sense of. and as a synonym of marsh (Zoltai I 983,
Junk 1983, Hofstctler 1983. Howard-Williams and
Gaudcl 1985) as distinct from a subdivision of mircland
is defined as an area intermittently or permanently
covered with water, supporting woody plants and
essentially lacking peat development. In fact, marsh,
swamp, and wetland are sometimes used
interchangeably and the general term wetland has been
used to describe all land/watcr interfaces.

97

Journal of Ihe Royal Society of Western Australia. Vol. 69. Part 3. 1987.

(k’mral classijlcadons

A variety of classillcalions have been produced by a
number of workers. Some have been comprehensive and
generally exhaustive (Cowardin c/ al. 1979) whereas
some have concentrated mainly on limnologic wetlands
or some specialised aspect of wetlands (Marlin ct al.
1953. Hutchinson 1957. Bayly and Williams 1973,
Ivanov 1981. Wcl/el 1983, Rutiner 1963). Two of these
classifications are outlined below.

Marlin ci al. (1953) produced a classillcation of
wetlands based on I) geographical position 2) water
quality 3) period of Hooding 4) depth of water during the
growing season and 5) vegetation type within the
wetland. The freshwater wetlands were then divided into
meadows, marshes, swamps, open water and Hals, and
were ditTercniiated on the criteria of water depth and/or
vegetation cover. Similar wetland categories were
proposed for inland and coastal areas and again for
saline wetlands. An example from Marlin ci al. ( 1953) is
as follows: “Fresh Inland Meadow: No standing water,
but saturated within a decimeter of the soil surface.
Shallow depressions. Communities dominated by
graminoids and herbs". Within the classification by
Martin ct al. (1953). hovvever. criteria were not applied
consistently. In the case of “wooded swamps" the period
of Hooding is omitted whereas in the categories of “salt
Hal", “salt meadow" and "irregularly Hooded salt
marshes", it becomes the distinguishing factor. The use
of salinity is another problem with the classiHcaiion.
Salinity was noted to be cither fresh or saline. There is
no class of wetland which is gradational or dynamic. The
cross sectional shape of the wetlands was not
differentiated in each class. The terminology docs not
allow for wetlands with variegated vegetation pattern.
Also, the w'alcr levels are loo specific and in cases, where
this is the only differentiating factor, it is loo precise for
use in Australia.

The classification by C'ow'ardin ct al. (1979) is a
hierarchial one based on recognition of classes, orders
and progressively lower levels of taxonomic
differentiation. The classification al the onset denotes 5
systems of wetlands that “have certain homogeneous
natural attributes": these arc the marine, estuarine,
riverine, lacustrine and paluslrinc systems. Each system
is composed of subsystems of littoral and limnetic
habitats. The systems are divided into classes and then
dominance types, and finally modifying terms arc
added. The class to which a wetland belongs is
determined by the dominant life form of the vegetation
(or in the absence of vegetation, the allributcs of the
substrate): the dominance type refers to the dominant
plant or animal species present: the modifiers refer to
soil and water attributes of the wetland. The
classification was designed so that it could be applied at
all levels of data collection. As the information on a
wetland increases, the classification may be refined so
that two objectives were satisfied: the wetland could be
classified immediately within a regional framework, and
secondly the wetland could be finally described and
differentiated on important individual characteristics.

A number of classifications for wetlands also have
been devised by workers in Australia. These
classifications are intended mainly for local use
(Goodrick 1970. Cowling 1977, Jacobs 1983). Many
generally appear to ulili/e existing overseas schemes and
amend them where appropriate for the local system (e.g.
Riggcrt 1964-66. sec later). More recently F^aijmans ct al.
(1985) provided an overview of Australian wetlands and

proposed 6 categories of wetlands (lakes, swamps, land
subject to inundation, channels, tidal Hats and coastal
water bodies). Their approach however is loo broad for
this study and their categories are based on a variety of
mixed criteria.

Application of international classifications to the Darling
System

Wetlands of the Darling System are quite variable in
shape, size, vegetation, soil, peal, water quality and
water maintenance. Wetlands with permanently or
seasonally inundated soils have been predominantly
described in the lilcralure, but those with seasonally,
waterlogged soils generally have been neglected.
According to the current international literature most of
the wetlands of the Darling System would fall into the
categories of 1 ) channels. 2) lakes. 3) mirclands (fens and
swamps) and 4) marshlands. There arc no bogs as found
in Northern Europe. Marshlands, however, are the most
dominant but there are types of marshlands which do
not conform to the definition presented above e..g. there
are open bodies of water which arc fringed by a range of
vegetation types, and there arc areas of seasonally
waterlogged soils supporting variable vegetation.

L’se of riverine terms

The genetic and quantitative classifications for rivers
and streams (Twidale 1968, Fairbridge 1968) are
considered inappropriate for channels of the Darling
System because they do not provide precise information
on the important features of shape, size and
permanence, factors that are considered to be important
in the identification of the wetland habitat. The terms of
these classifications furthermore have neither ecological
nor descriptive gcomorphic implication. The
classification involving sediment load relates to stream
function and also docs not convey the necessary
de.scriptive information useful to identify the various
channel wetlands. The terms used to describe sinuosity
arc appropriate to describe wetland channels because
they convey geometric impressions of the channel
system particularly if used in conjunction with size
terms. However it is necessary only to use a limited
range of these terms to convey channel sinuosity: these
arc: straight, anastomosing, irregular and sinuous.

i'sc of vegetation in classification

Vegetation (using features of Horistics or structure, or
both) has been used to classify wetlands (Martin ct al.
1953, Goodrick 1970. Cowardin ct al. 1979. Briggs
1981. Campbell 1983, Sjors 1983, Pisano 1983, and
others). In some cases vegetation has even been used to
unravel! the edaphic features of w'cliands. For instance,
Ivanov (1981) deduces hydrology, stratigraphy, origin
and development from the use of restricted plant
habitats in mire ecology and uses species presence and
structure in relation to micro-relief to determine the
structure of mire formations. Vegetation has been used
to differentiate types of mires and swamps (e.g. spruce
swamp, pine bog).

It is considered here, however, that vegetation should
not be used as a primarv- criterion to classify wetlands of
the Darling System. * The vegetation structure is
dependent upon hydrological and gcomorphological
factors such as water maintenance, water quality, micro-
relief and soils, so that the primary classification of
wetlands solely based on vegetation structure may not

98

Journal of the Royal Society of Western Australia. Vol. 69. Part 3, 1987.

allow any differeniiaiion of other wetland properties.
Also, because of the relatively limited species pool that
dominantly contribute to the vegetation of wetlands, the
use of vegetation as a primary criterion would result in
many different wetlands being classified into one group.
Consider for example paperbark forests and woodlands
which occur on deltas, along river banks, on flood plains,
along the edge of basins, in blow outs, in interdune
depressions and fringing the shores of estuaries: the use
of this vegetation in classification would not bring out
the various and different primary wetland categories.

Use of wafer quality in classification

Wetlands have been subdivided variably on the basis
of different water types, such as ombrolrophic and
mineroirophic, or aspects of salinity (Tansley 1939,
Martin etal. 1953. Bayly and Williams 1973, Moore and
Bellamy 1974. Gore 1983) and in some cases it forms the
primary criterion for classification. More usually the
subdivision of water categories merely provides a
secondary subdivision of wetland types.

Since most of the wetlands in the Darling System arc
maintained by groundwater, it must be concluded that
they are to some extent mineroirophic. However, the
input of rain during the wet season shifts wetlands to the
ombrolrophic end of the spectrum. Although this simple
division based on dissolved mineral content cannot be
used successfully, the division according to salinity or to
the presence of one or more minerals may be attempted.
The use of w'atcr quality is considered impodant at
lower hierarchial levels but not at a primary level.
Accordingly salinity terms such as freshwater, brackish
water, subsaline, hyposaline, saline water and
hypersaline water become applicable (Davis and
DeWiest 1966. Logan et al. 1974. Cow^ardin et al. 1979,
Dreva 1982. Hammer et al. 1983). An excellent review
of the problems of terminology of categories of water
quality is provided by Hammer (1986). The consistency
of water quality also is considered here to be relevant in
classification at lower hierarchial levels, because of the
seasonal variation in water quality of many wetlands of
the Darling System.

Use of peat /stratigraphy in classification

The sccondao' classification of mirelands into fens
and bogs is based on the presence/absence of peat and
essentially underscores botanical and shallow
stratigraphic aspects of wetlands. Taken to conclusion
the stratigraphic approach would result in wetlands
being categorised on the origin of their surficial soils
such as peal, carbonate mud, diatomile. gypsum
deposits, etc. Welland shallow stratigraphy in the
Darling System is primarily determined by the
vegetation type, the gcomcrphic unit in which the
wetland is situated, and the developmental history of the
wetland. This shallow' stratigraphy is often complex and
necessitates detailed analysis. Thus although it is very
important in understanding wetland formation,
stratigraphic information is considered to be
inappropriate at the higher hierarchial levels of
classification.

An additional complication relates to the age of peat
units. Peal and peaty sand occur in many of the w'cllands
in the Darling System and are forming under present
conditions: they are young and often less than a metre
thick. In some places, however peal has been buried to
some depth and is essentially a “fossil" deposit. The

buried peal horizons may be relics of a different climatic
or wetland regime and should form no part of a
classification dealing with modern processes.

Use of internal morphology in classification

Morphological criteria such as internal relative relief
is used by a number of authors for basin wetlands
(Moore and Bellamy 1974, Ivanov 1981, Gore 1983) but
its use has tended to be subtly introduced at all stages of
classification. Where not directly mentioned, the
wetland internal morphology often is implied in the use
of vegetation structure and the water quality. For
instance with bogs and fens, use is made of vegetation
structure to infer the varieties of depressions, hollows,
basins, flats or slopes. In essence the use of internal
morphology in the international literature is merely a
variation of the use of vegetation criteria.

Use of external morphology in classification

The morphology of fluvial systems has been
successfully classified by a number of authors and
according to these approaches fluvial wetlands of the
Darling System could be classified as to channel shape
and water permanence. The external shape of basin-
wetland landform, however, generally is not directly
considered in the international literature even though it
is one underlying factor for the presence and extent of
the wetland itself and therefore should be a primary
factor in the classification of w'ctlands. In the Darling
System the morphological components of the landforms
produce definite, recognisable types of w^cllands, and
small scale scdimcntological features of wetlands can
also be used for more detailed categorisation. The larger
morphological components are considered to be crucial
in the primary classification of wetlands from a
geomorphic viewpoint as will be discussed later in the
paper.

Discussion

The review of international literature shows that
classifications and nomenclature of w'eilands have been
based on amost every edaphic or ecological aspect of the
sy stem: water supply, w ater chemistry, type of landform,
morphological structures within the wetland, shape of
wetlands, vegetation cover, and occurrence/types of
peat. There have been integrated approaches such as the
genelico-gcotopic classification of Ivanov (1981), the
w--etland classes of Zoltai el al. (1983), and the five major
systems of Cow'ardin el al. (1979). There also have been
numerous classifications based on a single feature, the
mo.st common being vegetation.

The primary^ classification of wetlands into lake,
mireland and marshland is inconsistently based on the
duration and depth of water supply and shows no clear
demarcation between wetlands with different water
levels and longevity. Three types of water longevity are
explicitly recognised by authors: 1) permanent

inundation, 2) permanent waterlogging, 3) seasonal
inundation or w-atcrlogging, but the terms lake, mireland
and marshland clearly do not mirror these categories: for
example:

• Permanently inundated areas may be termed lake or

swamp

• Permanently waterlogged areas may be termed mire,

swamp, marsh

99

Journal of the Royal Society of Western Australia. Vol. 69. Part 3. 1987.

• Seasonally inundated or waterlogged areas are termed
wetland, marsh, meadow, swamp.

There is no term other than marsh to refer to seasonally
waterlogged areas, as opposed to seasonally inundated
surfaces.

It is also dear that most classifications of wetland rely
heavily on categorisation using vegetation (which is
presumed to reflect water longevity and quality) at a
primary or secondary level. The classification of
wetlands where landform is linked to water
permanence/longcvity generally is not an approach
adopted in the literature. For instance, wetlands divided
into lake or mireland then are subdivided according to
biotopes (i.e. the presence or absence of peat and. on the
physiognomy of the plant cover). The third major factor
in categorising wetlands appears to be the use of water
quality. This may be based simply on aspects such as
minerotrophism. or specific salt content, or acidity.

Further subdivisions or classifications ot wetlands
have been based on the genetic morphologic
relationships of the entire wetland systems at a regional
scale, or on vegetation-related internal morphological
components of the wetland and the resulting structure of
the vegetation, or the presence of specific vegetation
(Moore and Bellamy 1974. Cowardin c/ al. 1979. Ivanov
1981, Gore 1983. Ruuhijarvi 1983. Sjors 1983).

The terms available in the literature that are useful in
classifying wetlands at a primary level therefore are:

the term wetland itself,
the term lake.

riverine geomorphic nomenelature such as river,
creek, channel sinuosity terms, etc.

A wide variety of wetland categories or terms
therefore are considered inapplicable for the Darling
System because they arc genetic, or they have imprecise
definition- or have strong vegetation connotation, or
soil/stratigraphy connotation, or they should be at lower
stage levels of hierarchy in classification, or because they
do not fully estend across the range of wetlands available
in the Darling Systems: these are:

1. genetic chemical categories such as ombroirophic.
minerotrophic etc.:

2. morphogenetic categories such as volcanic lake,
glacial lakes:

3. floristic categories such as swamp, meadow, marsh,
muskeg:

4. chemicakfloristic or soil-floristic categories such as
mire, swamp, moor, fen. bog etc.:

5. geologic /geographic base categories such as paludal,
continental:

6. the terms lake, swamp, marsh as used in a
sedimentoiogic sense.

Previous local classification of wetlands

There have been previous classifications of the
wetlands of the Swan Coastal Plain and Darling Plateau
of the Darling Svstem, notably by Serventy, Owen and
Pirrott (cited bv Clarke el al. 1971), Riggeit ( 1 964- 1 966).
Tingay and tingay (1976) the Wetlands Advisory
Committee (1977). and Allen (1980). All these
classifications devised to date have useful purposes since
each was developed for a specific task. These systems of
classification are briefly discussed below.

Classification adopted by Serventy, Owen and Pirrott

Serventy et al. (1971) (cited by Clarke el al. 1971) in
an unpublished document probably provided the first
classification of w^ellands in the Darling System. Three
genetic types of lakes or swamps were distinguished: 1)
isolated portions of the ocean: they cited examples such
as Preston-Clifton Lakes; 2) relic abandoned river
courses of which Herdsman and Perr>' Lakes are cited as
examples and 3) chance depressions.

Classification adopned by Riggerl

As part of a study into wetlands of Western Australia
bv the Department of Fisheries & Fauna, Riggert (1964-
1966) classified and evaluated wetlands of Western
Australia (including the Darling System) (see Fig 1). The
sludv was oriented to evaluating the utilisation ot
wetlands by waterfowl. Riggeit postulated that the
presence or absence of waterfowl provided an indication
of the physical stale of a wetland. The wetlands were
classified on criteria developed by Martin et at. (1953)
into 22 types. These were A. INLASD FRESH AREAS 1 .
Seasonally flooded basins or flats 2. Flooded
.Agricultural Land 3. Inland fresh meadow's 4. Inland
shallow fresh marshes 5. Inland deep fresh marshes 6.
Inland open fresh water 7. Permanent Open Water
(Reservoirs) 8. Shrub swamps 9. Wooded swamps and
10. Bogs: B. ISIAND SALINE ARP:AS 1 1. Inland saline
flats 12. Inland saline marshes and 13. Inland open
saline water: C. COASTAL LRhSH AREAS 14. Coastal
shallow fresh marshes 15. Coastal deep fresh marshes
amd 16. (Coastal open fresh water: and D. CO.iSTAL
SALINE AREAS 17. Coastal salt flats 18. Coastal salt
meadows 19. Irregularly flooded salt marshes 20.
Regularly flooded salt ma'rshes 21. Sounds and bays 22.
Mangrove Sw'amps.

Examples of these eategories of w'etlands in Western
Australia were described in an inventory approach by
Riggert noting size, depth, total surface area and
vegetation. The wetlands were also categorised and
evaluated on the basis of utilisation by waterfowl, in
terms of numbers of waterfowl per year, and types of
utilisation (e.g. breeding, feeding, migration and
loafing).

The approach by Riggert (1964-1966) indicated that
there was much variability in wetland types. However, it
did not distinguish between the mans types ot wetlands
that exist in the Darling System that can be separated on
the basis of geometrv', vegetation, degree of water
permanence and other faunal groups apart from
avifauna. For instance, riverine w'cllands were not
distinguished.

Classification adopted by Tingay and Tingay

As part of a study into wetlands of the Darling System
prepared for the Environmental Protection Authority,
Tingay and Tingay (1976) classified and compiled an
inventory of wetlands in the southwest of Western
Australia. They adopted the classification of Hutchinson
(1957) and Bayly and W'illiams ( 1 973), wherein wetlands
in the first instance are classified into lentic and lotic.
Tingay and Tingay (1976) utilised schemes that
subdivided the lentic categories into 1) lakes: tectonic,
volcanic, landslide, glacial, solution, fluviatile, wind
action or coastal types, and 2) shallow water bodies:
underground water, springs, water associated with
terrestrial vegetation, puddles, rock pools and ponds.

100

Journal of the Royal Society of Western Australia, Vol. 69. Part 3. 1987.

either permanent or temporary. Lotic categories were
sub-divided as permanent, temporary or episodic, and
also on features such as unidirectional flow, fluctuation
in flow rales, linear morphology, etc.

Tingay and Tingay applied the classification to the
Darling System and noted that lenlic wetlands dominate
the Bassendcan, Quindalup. Spearwood and Pinjarra
geomorphic systems of Mc.A.nhiir and Bcttenay (1960),
and that each wetland type within a geomorphic system
may have similarity in terms of origin, nature and biota.
Tingay and Tingay classified the lakes at Yanchep for
instance, as lentic solution lakes, and the wetlands of the
Bassendean system as Icntic wind action dune water
table lakes. They further subdivided the lenlic wetlands
of the coastal plain on the basis of soil elements and
series (i.e. they adopted soil associations as
environmental indicators) and they devised an acronym
symbol system to distinguish them. The lotic wetlands
were subdivided by Tingay and Tingay on a geographic
basis into Moore, Swan, Peel. Lcschcnault, Capel and
Hardy types with no further subdivision.

The approach of Tingay and Tingay (1976)
constituted a useful categorisation of wetlands for
inventory' purposes. The categories provided a checklist
of features to be determined for each wetland and
highlighted the variability of wetlands. However, the
system adopted is too genetic in the first instance, as well
as too complicated both for classification purposes and
for mapping. It involves a worker having to determine a
wide range of factors such as soil mosaics and landform
origin before a classification is possible. Yet a simpler
classification is possible based on readily-availabic
features gathered from maps and short field surveys. It
should be noted, however, that many of important
features of wetlands used by Tingay and Tingay arc
utilised in the classification developed herein.

Classification adopted by Wetlands Advisory Committee
'(1977)

The Wetlands Advisory Committee (1977) in a report
to the Environmental Protection Authority also devised
a classification of wetlands in the Darling System. The
main subdivisions adopted by the committee were: 1)
Lentic (non-flowing). 2) Lotic (flow'ing). 3) Estuarine,
and 4) Artificial. Thereafter the wetlands were
subdivided on criteria of size, salinity, permanence, and
degree of vegetation cover. These criteria were
considered important to determine the potential value of
a wetland for particular uses (e.g. waterfowl drought
refuge; or use for aquatic recreation). This approach
produced a total of 15 potential types of lentic wetlands,
which together w'ith the lotic, estuarine and artificial
types identified a total of 18 wetland types throughout
the Darling System region.

Wetlands classified by the Wetlands Advisory
Committee were allocated* symbols (similar to Tingay
and Tingay 1976) to distinguish them. Thus Lake
Jandabup at Wanneroo was designated LE.fLp.se
(lenlic. fresh, large, permanent, semi-closed vegetation
cover), the Swan River upper reaches were designated
LO (lotic): the Swan River at Fremantle was designated
E (estuarine) and ornamental lakes in Kings Park were
designated A (artificial).

The classification of the Wetlands Advisory
Committee (1977) provided a useful categorisation of
various types of wetlands. It addressed many of the
major attributes of wetlands and clearly showed their
variability and complexity in the Darling System.

However the classification did not utilise wetland shape,
only distinguished between two sizes of wetland (small
and large), did not separate types of Lotic systems and
did not distinguish between various “basin-type”
wetlands or identify wetland flats.

Classification by Allen

As part of a geohydrology study of the Swan Coastal
Plain. Allen (1976, 1981) identified wetlands as lakes
and swamps and further categorised them into six types
on the basis of age, inferred origin, and geographic
location. Allen identified the following: 1) Bambum
type. 2) Gnangara type, 3) Forrestdale type, 4)
Joondalup type. 5) Gwelup type and 6) Cooloongup
type. These categories I -6 were listed in order of inferred
decreasing age.

Other studies

In addition to the works cited above in which
classification of wetlands was a primary or important
motive, there arc other works in which .some form of
wetland classification or wetland vegetation
classification was utilised. These works include Evans
and Sherlock (1950), McComb and McComb (1967),
Seddon (1972). Bell et al. (1979), Watson and Bell
(1981); Muir (1983), Pen (1983), Speck and Baird
(1984). Congdon and McComb (1976) and Baird (1984).
Each of these works utilised established international
classification schemes. Thus the Yanchep wetland
system was termed a sw'amp and fen formation
(NlcComb and McComb 1967), Star Swamp and Yeal
Sw'amp were termed swamp and swampy flat,
respectively (Watson and Bell 1981, Baird 1984), and
Seddon (1972) utilised terms such as wetland, swamp,
lake.

Discussion

All classifications devised to date have not provided
an approach that enables categorisation of the variety of
wetlands across the whole Darling System. Only 3 of the
classifications have attempted to be comprehensive.
These arc the works of Riggcrt (1964-66), Tingay and
Tingay (1976). and the Welland Advisory Committee
(1977). However, even the most detailed of these
classifications use an essentially similar primary
subdivision of lenlic and lotic types (Tingay and Tingay
1976. Wetlands Advisory Committee 1977). The
distinction between still-s’tanding water and flowing
water (which is the basis of the lentic and lotic
subdivision and essentially allempls to separate fluvial
from lacustrine w'ctlands) however becomes indefinite in
some systems, e.g. some channels are slow flowing, and
in fact, have slower flow rales than some lakes, that are
pan of a groundwater drainage system. The approach to
classifying lakes by Tingay and Tingay (1976) is too
genetic to be of use and secondly only considers part of
the range of wetlands available, since many wetlands
may have no free-standing water at any lime of year. A
classification that separates the wetland into numerous
types as proposed by Riggcn (1964-1966) serves as a
useful function in identifying wetlands for use of
avifauna but docs not provide the adequate distinction
for additional purposes between the many varied
wetland types existing in the Darling System.

All the above classifications to date however have not
adequately addressed the identification and

101

Journal of the Royal Society of Western Australia. Vol. 69. Part 3. 1987.

nomenclature of waterlogged soils, or types of water
saluration/inundaiion or the full range of cross-sectional
and plan geometry of wetlands, or the range of sizes of
wetlands.

Classification — this study

Philosophical approach lo classification

Ideally classifications should initially be non-geneiic
and then with additional information the non-genetic
categories should relate to genetic categories.
Landform/gcomorphic and water/wetness criteria are
non-genetic and the proposed classification of wetlands
is based fundamentally on the two major features which
determine the existence of wetlands i.c. the water and
landform components. The water component is the
major feature that distinguishes the wetland habitat
from other terrestrial habitats and also the component
which inlluences biological response by its presence,
depth, chemistry, and movement. The landform
essentially is the “water container* and thus it
determines the size, shape and depth of a wetland. Any
wetland classification thus should reflect variability or
attributes of these two components (Figure 3). In
combination, the variables or attributes of these two
components result in a wide spectrum of possible
wetland types. As a prerequisite to examining these
combinations, a fuller explanation of the types of
subdivision of water and landform is presented below.

The water component

The component of water in a wetland may be viewed
from 4 inter-related aspects: 1) its persistence or

longevity, 2) its quality. 3) the consistency of water
quality, and 4) the mechanism by which water maintains
the wetland.

The longevity of water residing in a wetland, i.c. its
permanence or” intcrmittency. is directly related to the
precipitation and evaporation, mechanisms of water
supply, the permeability of underlying sediments, and
the shape of the wetland. Three types of longevity are
distinguished: 1) permanent inundation 2) seasonal or
intermittent inundation 3) seasonal or intermittent
waterlogging (without inundation).

Water quality referring to the dissolved solids, rather
than pH or coloration, may be subdivided into
categories of: fresh, brackish (or mixosaline). saline and
hypersalinc. However, there is inconsistent use of terms
and definitions for categories such as brackish, saline,
hypersaline (Davis and Dewiest 1966. Drever 1982
Cowardin ei al. 1979, Logan ei al. 1974, Hammer et ai
1983. Hammer 1986). The category terms and
boundaries adopted in this paper arc presented in Table
1. It also should be pointed out that many w'ctlands vary
in salinity during the year due to variable input ol water
sources and evapo-transpiration. Wetlands that are
seasonally variable in salinity are categorised by the
salinitv stale in which the wetland exists for the major
part of each year e.g. a wetland that ranges from
freshwater for most of the year, to brackish during the
season of reduced water supply, w'ould be classified as
freshwater. However, a term is introduced to denote
whether salinity is constant or variable. Water quality
that is consistent throughout the year (i.e. it remains
totally within a given salinity field, e.g. the freshwater
field or saline water field) is termed stasohalinef water
quality that markedly fluctuates throughout the year is

termed poikilohaline*. For instance a wetland that
ailernaies seasonally from freshwater to saline is
described herein as poikilohaline. In areas of seasonal
inundation and seasonal waterlogging. the
determination of salinity throughout the year will^
necessitate sampling the shallow groundwater for part of
the year.

Table 1

Classification of water salinity based on total dissolved solids

Salinity mg/1

Water Category*

less than 1,000

Fresh

1.000—3,000

Subhaline

3.000—20,000

Hyposaline

20.000—50.000

Mesosaline

50.000—100.000

Hypersaline

100,000 and greater

Brine

♦The terms and boundaries for fresh, subhaline, hyposaline mesosaline and
hypersalinc arc from Hammer 1986: the term “brine” is delineated by
Davis and Dewiest 1966.

The mechanisms by w-hich water maintains the
wetland arc variable and include direct precipitation,
groundwater seepage, surface inflow, ponding etc.
However the mechanisms of water maintenance will not
be considered further here.

Landform component

The landform component of a wetland can be
categorised on the basis of cross sectional geometry, plan
geometry and scale. The cross sectional geometry'
subdivides a wetland into basins, flats and channels.
Basins and flats have no external surface drainage
system. Basins range from flat bottomed lo concave,
from steep lo gentle sloping sides, from shallow to deep
and arc represented by many ot the common and
ubiquitous lakes and “swamps" of the Swan Coastal
Plain. They have clearly defined centres or depressions
but may have shaip or broad margins. Cross sectional
shapes of basins include: broad u-shape, steep u-shape.
saucer, and slightly concave depressions. It is on the
basis of a topographic depression that an individual
basin is recognised. This factor is particularly important
when a number of basins may form a nearly coalescing
network: the local central depressions define each basin
even though the littoral zones of adjacent basins may
merge or overlap. Flats arc marked by little or no
marginal relief, and have diffuse lateral boundaries.
Flats in the Darling System commonly occur on the
tributary interfluves of creeks and rivers, as overbank
floodplains, and on broad alluvial flats and slopes in
front of foothills of the Darling Scarp.

Channels refer to any incised water course, or reach,
including those connecting basin-type bodies of water.
Channels have clearly defined margins and may be
bedrock confined or alluv'ial and this distinction may
result in deeply incised channels as opposed to shallowly-
incised channels. The alluvial channels vary according to
their hydraulic characteristics which arc determined by
their slabilitv. channel and valley slope gradients, the
mode of transport, and composition of their load
(suspended vs. bedload).

*siaso aticr Greek Stasimos ( -constanl)

♦poikilo aher Greek Poikiios ( variable)

102

WETLAND COMPONENTS FOR USE IN CLASSIFICATION

Permanently

inundated

Seasonally

inundated

Seasonally

waterlogged

WET I I LAND I

WATER PERMANENCE

CROSS-SECTIONAL SHAPE

1

Basin

Channel

Flat

CRITERIA USED TO
DEVELOP

PRIMARY WETLAND
CATEGORIES

Fresh

Subsaline

Hyposaline

Mesosaline

Hypersaline

Brine

Poikilohaline

Stasohaline

3

WATER SALINITY

SIZE

CONSISTENCY OF
WATER SALINITY

PLAN SHAPE

Megascale

Macroscale

Mesoscale

Microscale

Leptoscale

Linear —

Elongate
Irregular
Ovoid

Round —

Straight —

Sinuous
Anastomosing
Irregular —

CRITERIA USED TO

DEVELOP

SECONDARY

WETLAND

CATEGORIES

Telluric —
Meteoric —
Marine —

WATER MAINTENANCE
( ORIGIN )

STRATIGRAPHY

ORIGIN

CRITERIA NOT USED
IN THIS PAPER

Figure 3. — Components or attributes of wetlands and their terminology used in the proposed classification.

Channel terms include rivers, streams, brooks, creeks,
drainage lines or troughs. Only the terms rivers and
creeks are used in this classification.

In plan form, encompassing the limnetic and littoral
zones (Hutchinson 1957, Cowardin el al. 1979), wetland
shapes arc easily discernible and may be described as
linear, elongate, irregular, ovoid or round, for basins;
and straight, sinuous, anastomosing, or irregular for
channels (Fig. 4).

Wetlands may be further categorised according to
scale. For basins and flats the categories of geomorphic
scale (modifed from Semcniuk 1986) are:

Megascale: Very large scale wetlands larger than a
frame of reference 1 0km x 1 0km

Macroscale; Large scale wetlands encompassed by a
frame of reference 1 000m x 1000m to 10km x
10km

Mesoscale; Medium scale wetlands encompassed by
a frame of reference 500m x 500m to 1000m x
1000m

Microscale: Small scale wetlands encompassed by a
frame of reference 100m x 1 00m to 500m x
500m

In the case of channels, a definitive width to length
relationship is used:

Macroscale: Large scale channels I km and greater
wide, by several to tens of kilometres long.

Mesoscale; Medium scale channels hundreds of
metres wide, by thousands of metres long.

Microscale: Small scale wetlands tens of metres
wide, hundreds of metres long.

Leptoscale: Fine scale channels several metres wide,
tens of metres long.

The suggested order of importance to classification for
the components of landform is presented in Figure 3.

I'he proposed classification

The classification of wetlands proposed here has been
developed by combining the various components of
water and landform. Using primary subdivisions of
cross-sectional landform geometry there are recognised:

• basins

• channels

• Hats

Within the category of basin, using primary subdivision
of water there are recognised:

• permanently inundated types

• seasonally inundated types

• seasonally waterlogged types.

Within the category of channels there arc:

• permanently inundated types

• seasonally, or intermittently, inundated types.

103

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

Within the category of flats there arc;

• seasonally, or intermittently, inundated types

• seasonally, or intermittently, waterlogged types.

This categorisation then allows recognition of 7 main

wetland types:

Water

Landform

Longevity

Basin

Channel

Flat

Permanent

inundation

Permanently

inundated

basin

Permanently

inundated

channel

—

Seasonal (or

intermittent)

inundation

Seasonally

inundated

basin

Seasonally

inundated

channel

Seasonally

inundated

flat

Seasonal (or

intermittent)

waterlogging

Seasonally

waterlogged

basin

—

Seasonally

waterlogged

flat

These basic 7 categories require nomenclature so that
they may form the primary units of wetland
differentiation. Permanently inundated flats and
seasonally waterlogged channels arc not common
wetlands.

1.

2 .

3.

4.

5.

6 .
7.

The proposed terms for the basic wetland units arc:

Permanently inundated basin
Seasonally inundated basin
Seasonally waterlogged basin
Permanently inundated channel
Seasonally inundated channel
Seasonally inundated flat
Seasonally waterlogged Hat

= LAKE

= SUMPLAND

= DAMPLAND

-RIVER

-CREEK

-FLOODPLAIN

-PALUSPLAIN

Some of the terms above arc established and well
defined previously in the literature and hence have been
rc-utilised. Others (i.e. sumpland, dampland, palusplain)
have been coined in this paper. The rationale for. the
definition of, and the origin of the terms arc described in
Table 2. A comparison of the proposed wetland terms
with other previously established terms is provided in
Table 3. It should be noted that the terms river and creek
are used to denote permanence and intcrmittcncy of
water flow, respectively; the size of the river or creek is
indicated by the use of the scale modifier.

Areas such as damplands and palusplains which are
rarely or infrequently inundated do not necessarily
qualify to be termed sumplands and floodplains. The
characteristic of seasonal waterlogging should be a
prevailing, recurring feature and this is the deciding
factor in determining the categorisation of a wetlarid by
the longevity and type of its “wetness". The occasional
or infrequent flooding of terrain can take place anywhere
during torrential rainfall or sheet flooding but such
phenomena do not categorise the temporarily inundated
land surface as a wetland. .An example illustrating how a
basin cross sectional geometr>' interacts with a varying
water level to develop 3 categories of wetland (i.e. lake,
sumpland and dampland) is presented in Figure 5.

Further uses of water and landform
descriptors/modifiers are implemented to elaborate and
ornament the nomenclature of the primary units. For
example, the Swan Coastal Plain contains numerous

PLAN GEOMETRY OF WETLANDS

examples of freshw'ater large scale ovoid, permanently
inundated basins, that remain fresh throughout the year;
these would be simply and obviously termed freshwater,
stasohaline macroscale, ovoid lakes. Lake Coogee for
instance is a hyposaline. stasohaline, macroscale, ovoid
lake; Lake Joondalup is a freshwater, stasohaline
macroscale lake; Lake Pinjar is a freshwater, stasohaline
megascale sumpland.

The full range of descriplors/modifiers are listed in
Figure 3. A variety of wetland types developed by
various combinations of water and landform features are
listed in Table 4 together with some examples of each.
Figure 6 illustrates a typical range in geometry and size
of various wetlands in the Darling System.

104

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

Table 2

Definition and origin of terms

Wetland term

Definition

Defined by

Origin of term

Usage in this paper

Lake

Permanently inundated
basin of variable size
and shape

Mill (1900-1910)
Monkhouse ( 1 965)
Bates & Jackson ( 1 980)
Fairbridge (1968)
Ruttner(1953)

Established term, from Latin
(acus, a hollow.

The usage in this paper does not
distinguish between shallow lakes and
deep lakes.

Sumpland

Seasonally inundated
basin of variable size
and shape

This paper

After “sump" meaning site of
water retention or ponding or
accumulation; The term is
fortuitously similar to “sumpf'
the German term for swamp

As defined.

Dampland

Seasonally waterlogged
basin of variable size
and shape

This paper

After “damp” meaning moist or
wet. Thus it refers to a dampness
or waterlogging of soils of some
basin wetlands

As defined.

River

Permanently inundated
channel of variable size
and shape

Swayne (1956)
Trowbridge (1962)
Morisawa(1968)

Established term from Latin of
most mm, a stream (Shipley
1982)

This usage conforms with the concept of
other authors that a river is defined as
channelled water flow, but is different to
most authors in its necessity for
permanence of water. The permanence of
water, also generally implies a channel of
large rather than small size.

Creek

Seasonally inundated
channel of variable size
and shape

Whittow(I984)
Monkhouse (1965)
Trowbridge (1962)

Bates and Jackson
(1980)

Established term

This usage generally conforms with that of
Australia and southwestern U.S.A.

Floodplain

Seasonally inundated
flat

Mill (1900-1910)
Monkhouse (1965)
Moore ( 1 949)

Established term

This differs from other authors in that
inundation of the plain need not be linked
to a river; in general however a floodplain
is associated with a river or creek.

Palusplain

Seasonally waterlogged
flat

This paper

After latin palus meaning
“marshy”; thus the term refers to
flats which are similar to
dampland basins.

As defined.

Stasohaline

Water of relatively
constant salinity

remaining in a given
salinity field

This paper

After staso (Greek) meaning
constant

As defined.

Poikilohaline

Water of variable
salinity fluctuating from
one salinity field to
another

Originally defined Dahl
(1956)

After poikilo (Greek) meaning
variable

As defined.

Waterlogged

Area in which water
stands near, or at the
land surface

Established term

Usage conforms with Golet and Larson
(1974), Martin et al., (1953) and most
other authors.

The use of vegetation in describing and classifying
wetlands has always been accepted by wetland workers.
Vegetation has been viewed in previous work in terms
of: 1) structure. 2) life form, 3) species dominance, 4)
percentage cover of area, and 5) zonation. Each of these
parameters significantly increase understanding of an
individual wetland, and its uniqueness or similarity to
other wetlands. The methods for describing vegetation
have been thoroughly documented by numerous authors,
as have interrelationships with physical chemical and
biological variables. Vegetation also becomes important
when correlating or establishing ecological linkage for
chains or series of wetlands in the same physiographic
settings, but because it is a dynamic, responsive and
mutable feature, it is suggested that the use of vegetation

should only be employed in a later stage description of
individual wetlands as opposed to initial stage regional
descriptions of wetlands. It is suggested that the use of
vegetation be applied in a tertiary or quaternary
modifier capacity. As such, it is suggested that
vegetation be used as an adjectival modifier to the
wetland classification by utilising floristic and structural
terms with a primary w'ctland class. However it should
be pointed out that catcgorisation/classificalion of
w'etland vegetation becomes a difficult task where
vegetation is not a simple extensive unit but rather
concentrically zoned, or in the form of complex mosaics.

The use by many limnologists and biologists
(Hutchinson 1957, Wetzel 1983, Ruttner 1953, Bayly

105

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

Fable 3

Comparison of wetland terms used in this paper with established classifications

This paper

Martin

elal.{\957>)

Cowardin
et al. (1979)

Golet &
Larson (1974)

Paijmans
ei fl/. (1985)

General

European

General
N. American

Lake

Open fresh water
Deep fresh marshes
Open saline water

Lacustrine

Open water
Shrub swamp
Deep marsh

Lakes
Swamp
Coastal water
bodies

Lake

Swamp

Lake

Swamp

Sumpland

Wooded swamp
Seasonally flooded
Basins

Shallow fresh marshes
Deep fresh marshes
Saline marshes
Open saline water

Palustrine

Deep marsh
Shallow marsh
Shrub swamp
Wooded swamp
Open water

Lakes

Swamp

Marsh

Marsh

Meadow

Dampland

Fresh meadows
Wooded swamp

Palustrine

Meadow

—

—

Meadow

River

—

Riverine

—

River and

creek

channels

River

Stream

Creek

Brook

River

Stream

Creek

Brook

Creek

—

Riverine

—

River and creek
channels

—

.Arroyo

Floodplain

Shrub swamp
Wooded swamp

—

Seasonally flooded
flats

Land subject to
inundation

Floodplain

Roodplain
Seasonally
Flooded flat

Palusplain

Wooded swamp
Saline flat
Salt meadow?

Palustrine

—

—

—

—

BASIN WETLANDS IN RELATIONSHIP TO WATER LEVEL FLUCTUATION

LAKE SUMPLAND DAMPLAND

Figure 5.— The development (based on permanence of water) of the 3 basin wetland categories, which in this example are related to
landform position with respect to fluctuating water level.

106

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

and Williams 1973, Cowardin et al. 1979) of the terms
limnetic and littoral to denote physical regions of a basin
wetland is important in this classification. By denoting
these separate zones, a wetland can be viewed in its
entirety rather than as two or more separate wetlands
e.g. the gradation from inundated surface to waterlogged
soil may be seen as a continuous system. Thus the
exposed marginal zone of a lake need not be viewed as a
sumpland. but rather as a littoral zone.

Application of the proposed classification to wetlands of
the Darling System

Several areas in different physiographic regions have
been selected to indicate the practical use of the
proposed classification and to display the range of
wetlands occurring throughout the Darling System in
their variability of geometry, size, and salinity. Figure 7
presents maps from four d'lffcrcnt geomorphic settings
on the Swan Coastal Plain (Mc.Arthur and Beltenay
1960) and illustrates a range of wetland types in each
selected area. Figure 8 presents hydrographs of selected
wetlands from these areas to illustrate their hydrologic
properties. The areas selected are:

1 . Gnangara area

2. Ballajura area

3. Coogee area

4. Pt. Becher area

fable 4

Classification of some typical wetlands in the Darling system

Wetland

Primary

Category

Full Classification *

Lake Joondalup

lake

macroscale, elongate
stasohaline, lake.

fresh.

Lake Bambun

lake

mesoscale. round subsaline
stasohaline. lake.

Lake Coogee

lake

mesoscale, elongate hyposaline
stasohaline. lake.

Lake Bibra

lake

macroscale. irregular
poikilohaline, lake

fresh.

Lake Cooloongup

lake

macroscale, ovoid hyposaline
poikilohaline, lake.

Carine Swamp

sumpland

mesoscale. ovoid

stasphaline. sumpland.

fresh.

Lake Pinjar

sumpland

macroscale. ovoid

stasphaline. sumpland.

fresh.

Bollard Bullrush

Swamp

sumpland

macroscale. round

poikilohaline. sumpland.

fresh.

Melaleuca Park

sumpland

microscale. ovoid

poikilohaline, sumpland.

fresh.

Stable Swamp

sumpland

microscale. linear

stasohaline. sumpland.

fresh.

Lake Bindiar

dampland

mesoscale. round

stasohaline. dampland.

fresh.

Lake Adams

dampland

macroscale. ovoid

poikilohaline. dampland.

fresh.

Yalbanberup Pool

river

mesoscale. sinuous hypersaline,
poikilohaline. river.

Ellen Brook

creek

microscale. sinuous

poikilohaline. creek.

fresh.

Collie River

tributory at Schotts

creek

microscale. straight

stasohaline, creek

fresh.

* Information for this classification based on 3 years of seasonal surveys.

The Gnangara area (Fig. 7) set in the Bassendean
Dune system of McArthur and Bettenay (1960) is
dominated by numerous unnamed microscale, round
stasohaline, freshwater, sumplands and damplands.
Lake Gnangara itself was an isolated macroscale, round,
stasohaline freshwater lake; in recent years it has
changed to being a sumpland. Snake Sw'amp is a
mesoscale. ovoid, stasohaline, freshwater dampland.
Badjerup Lake is a mesoscale. ovoid, stasohaline,
freshwater sumpland.

In contrast, the Ballajura area set in the Bassendean
Dune system and Pinjarra Plain alluvial system of
McArthur and Bettenay (1960), contains wetlands that
are mesoscale, straight and sinuous stasohaline creeks, a
mesoscale. freshwater, poikilohaline river, mesoscale to
macroscale floodplains which arc poikilohaline. and
microscale, freshwater, stasohaline sumplands and
damplands.

The Lake Coogee area set in the Spearwood Dune
System of McArthur and Bettenay (I960) has the
following wetlands: Lake Coogee North is a mesoscale,
elongate, poikilohaline. mesosaline sumpland; Lake
Coogee South is a mesoscale elongate, stasohaline,
hyposaline lake: the Henderson wetland complex is
comprised of mesoscale, elongate, poikilohaline,
freshwater sumplands; and Brownman*s Swamp is a
mesoscale ovoid, poikilohaline freshwater sumpland.

The Pi. Bccher area, set in the Quindalup Dune
system of McArthur and Bettenay 1960, is one in which
microscale, ovoid, linear to irregular shaped,
poikilohaline freshwater sumplands and damplands
predominate.

Discussion

Wetlands arc inherently complex ecological habitats
but their analysis is simplified somewhat by a
classification which brings into prominence the
important wetland components of water and landform.
Since there are only seven basic geomorphic wetland
types, to which are added modifiers/descriplors of scale,
plan geometry, and water properties, the classification
enables one to distinguish a practicable number of
wetland types. The basic wetland categories can then be
ornamented/embellished to illustrate further variability
with the addition of more detailed field information. But
even without the more detailed studies a given wetland
still can be readily classified into the primary categories
with a minimum of field surveys. Thus the approach
provided here has the advantages when additional
detailed information becomes available of increasing the
discrimination of individual wetlands from each other
and from their surrounding areas; with less precise
information available, categorisation at a broader level
into a regional physiographic setting is also feasible.

The proposed classification also can provide useful
mapping units since the various wetland types may be
readily identified and mapped as categories. The
regional differences and similarities between wetlands
also would emerge more clearly and may be linked to
other types of mapping parameters, e.g. contours,
climate, geology. The "diw" wetland types in the Darling
System, i.e. the damplands. which in the past were in
some cases excluded, or included as wetlands on the
basis of mixed criteria, can now be consistently
identified as a category of wetland.

107

A LAKES

0

A6

B SUMPLANDS

C DAMPLANDS

OC3

D CHANNELS: RIVERS and CREEKS

C8

0

\ ”

Q C7

oC?

0

0C4

o

C5

0

FLOODPLAINS and PALUSPLAINS

E1

E4

£V.° J Floodplain
l~ i Palutplain

A1

Loch McNess

C1

A2

Nowergup Lake

C2

A3

Lake Joondalup

C3

A4

Jandabup Lake

C4

AS

Lake Monger

C5

A6

Lake Richmond

C6

A7

Lake Cooioongup

C7

A8

Bibra, North Lakes

C8

C9

B1

Lake Karrinyup

D1

B2

Wallabuenup Swamp

D2

B3

Balcatta

D3

B4

Lake Claremont

D4

B5

Lake Pinjar

B6

Bollard Bullrush Swamp

B7

Lake Banganup

E1

B8

The Spectacles

E2

B9

Wright Road Swamp

E3|

E4I

0

5 km

E5

L

1

E6

Lake Adams
Sydney Rd, Gnangara
Piney Lake East
Nicholson Rd, Forrestdale
West of Yangebup
Canning Vale Area
Johnson Rd, Casuarina
Bindiar Lake
Yeal Swamp group

Canning River
Ellen Brook
Serpentine River
Gin Gin Brook

West Rd, Keysbrook
Muchea Area
" Timaru ” near Brand
Hwy and Chandalla Bk
Chandalla Brook
Canning River

Figure 6. — Typical range of wetland categories in the Darling System.

108

Figure 7. — Maps showing types and distribution of wetland categories for four selected areas.

HYDROGRAPHS OF TYPICAL BASIN WETLANDS

LAKE

SUMPLAND

DAMPLAND

0 Bibra Lake
• Jandabup Lake
A Lake Joondalup
A Lake Cooloongup
■ Lake Coogee

o Henderson System

• Stakehill Swamp

^ North Lake Swamp

* Wright Road Swamp
■ Mungala Swamp

o Gnangara Pine Forest 1
• Gnangara Pine Forest 7
^ Stoney Rd, Gnangara
A Lake Adams
■ Ballajura area

Figure 8. — Hydrographs of some typical basin wetlands on the Swan Coastal Plain.

Journal of the Royal Society of Western Australia. Vol. 69. Part 3. 1987.

Various wetland categories have a range of
multifarious ecological functions and the proposed
classification may parallel these functional delineations.
For instance lakes, damplands and creeks are utilised by
fauna in different ways because each of these wetlands
arc essentially different habitats as determined by the
longevity and type of water input. For example some
avifauna and reptiles use open water lakes for a specific
range of purposes, whereas the vegetated sanctuary of
many typical damplands may be utilised by mammals
and other species of avifauna in a different capacity.
Nutrient pathways and trophic inter-relationships also
may be fundamentally distinct bctw'ccn these various
primary wetland types. Botanists. zoologists,
educationalist, recreational and land use planners may
be able to make preliminary' assessments of the di^ ersity.
dependence, complexity etc. of wetlands from the class
to which it belongs. The advantage of the proposed
classification also is that it can be used as a basis for any
wetland study regardless of the ultimate discipline of the
study (e.g. hydrology, stratigraphy, botany, zoology), and
so circumvents the problem of a proliferation of
nomenclature arising from specific applications/studies.
Thus it provides a non-genelic framew-ork upon which to
base further detailed work.

Finally it is apparent that with the proposed
classification a wetland can still be placed in its
appropriate category^ even if it has been substantially
altered by clearing of vegetation and disturbance of soils.
As long as the mechanisms of water maintenance and
basic geometry have not been destroyed then the
inherent geomorphic wetland entity remains and can be
identified and named.

Acknowlcdgemem.s — The manuscnpi was critically reviewed by D.
Glassford. I. LeProvost. V. Semeniuk and P. A. S, Wurni. Their help is
gratefully acknowledged. Some of the research for this study was funded by
veSRG. Research and Educational Consultants and LeProvost. Semeniuk
and Chalmcr. Environmental Consultants.

References

Adam. P.. Unwin, N.. W'eincr. P.. Sim. 1. and W'iner, P. (1985). — Coastal
^Vctlamis of New South ^^'a^es. Report for Department of
Environment and Planning.

Allen, T. (1976). — Outline of the hydrogeology of the superficial formations
on the Swan Coastal Plain H ..1. Gcoi. Surv. Ann. Kept. 1975: 31-
42.

Allen. T. ( 1981 ). — fJroundwalcr Resources of the Svvan Coastal Plain, near
Perth. Western Australia. In: Groundwater Resources of the Swan
Coastal Plain. CSIRO. Proceedings of Symposium by CSIRO and
Western Australian Slate Committee of Water Research
Foundation.

Anderson, J A. R. (1983). — The Tropical Peat Swamps of Wesleni
Malesia. In: Gore. A. J. P. (Ed.) Ecosystems of the World. 4H.
.Xfires; Swamp. Bog. Fen and Moor. Regional Studies. Elsevier
Scicntilic Publishing Co., pl8]-l99,

Baird. A. M. (1984). — Observations on regeneration after fire in the Yule
Brook Reserve near Perth. Western Australia. J. Roe. Soc.
11.4.67:1-13.

Bayly, J. A. E. and W'lUiams, W. D. (1973). — Inland X\'utcrs and Their
Ecology^ Longman Cheshire.

Bates. R. L. and Jackson. J. A. (1980). — Glo.'isuiy of Geology. .American
Geological Institute.

Bell. D. T.. Loneragan. W. .A. and Dodd. J. (1979), — Preliminary
vegetation survey of Star Swamp and vicinity. Western Australia.
Herbarium Res. Notes Number 2: 1 -2 1 .

Bloom. A. L- (1978). — Geomorphology. 1 Systematic .inanlysis of Late
Cenozotc Landforms. Preniice-IIall Inc., pi 97-278.

Briggs. S. V. (1981). — Freshwater Wetlands. In: R. H Groves (Ed.).
-iusirahan I'egelaiion. Cambridge University Press. London,
p335-.^60.

Campbell. E. O. (1983). — Mires of Australasia. In; Gore .A. J. P. (Ed.)
Ecosysiemi of 'he li orld. 4B. Mires; Swamp. Bog. Fen and .Moor.
Regional .*>tudk’s. Elsevier Scientific Publishing Co.. pl53-180.

Cant. D. J. ( 1 982)-Fluvial Facies Models and their .Application. In: Scholle.

P. A. and Spearing. D. (Eds.) Sandstone Dcposifional
Environments, .-im. .-l.s5oc. Petrol. Geoi. Mem 31: 115-121.

Chorlcy. R. J. (1969). — The Drainage Basin as the Fundamental
Geomorphic Llnit. In; Chorley. R. J.. (Ed.) Introduction to Fluvial
Processes. Methuen & Co. Ltd. p30-5 1 .

Clarke. E. de C.. Pridcr. R. T., Tcichert. C. (1971). — Elements of Geology
for A usiralian Students, University of W..A . Press.

Congdon. R. A. and McComb. A. J (1976). — The nutrients and plants of
Lake Joondalup, a mildly enirophic lake c.xperiencing large
seasonal changes m volume. Roy. Soc. W.A.. 59: 14-23.

Cowardin. L. M.. Carter. V., Golet. F. C. and LaRoe. E. T.
(1979). — Classification oj wetlands and deepwater habitats of the
United Staled. U.S. Dept, of the Inlerior; Fish & Wildlife Service
Dec.

Cowling, S. L. ( 1 977). — Classification of the wetland habitats of waterbirds.
Aust. Mar. Sci. Bull.. 58.

Dahl. E. (1956). — Ecological salinity boundaries in poikilohaline waters.
Oikos. 7: 1-21.

Davis. S. N. and DeWiesi. R. J. (1966). — Hydrogeology. John Wiley &
Sons Inc., pi 1 8.

Denny. P. (Ed.) (1985). — The Ecolo.gy and .Management of .African wetland
vegetation. W'. Junk Publishers.

Department of Conservation and Environment (1980). — Wetlands
Guidelines for protection and management. Dept. Conservation &
Environment Bulletin No. 79.

Drever. J. 1. (1982). — The Geochemistry of Natural H ater. Prentice-Hall.
Inc.

Evans. G. A. and Sherlock. N. V. (1950). — Butler's Swamp Claremont.
H 'cstern .Australian Nalurali.st.,2: 1 52.

Fairbridgc, R. W, (Ed.) ( 1 968). — The Encyclopedia of Oeomorphology. I 'ol.
Ill Dow’den. Hutchinson Sc Ross Inc.

Lceder. M. R. ( 1 982). — Sedimentology Process Product. .Allen & Linwin.

Gilluly. J.. Walers. .A. C. and Woodford. .A. O. (1951). — Principles of
Geology. Freeman & Co.. p224-231 .

Gold. F. C and Larson. J. S. (1974). — Classification of Freshwater
Wetlands in the Glaciated Northeast. Resource Publication 1 16.
US. Fish and Wildlife Seivice Washington. D.C.

Goodrick. G. M. (1970). — .A Surv’cy of Wetlands of coastal New South
Wales. CSIRO Division of Wildlife Research Technical
Memorandum No. 5.

Gore. A. J, P. ( 1 983). — Introduction. In: Gore A. J. P. (Ed.) Ecosystems of
the World. 4.A. .Mires.- Swamp. Bog. Fen. and Moor. General
Studies. Elsevier Scientific Publishing Co., pi -34.

Gorham, E. ( 1953). — Some Early Ideas Concerning the Nature. Origin and
Development of Peat Lands. Jour. Ecology.. 41: 257-273.

Gorham. E. (1957), — The Development of Peatlands. Quart. Review
Biology.. 32: 147-164,

Hakanson, L, and Jansson, M. ( 1983). — Principles of Lake Sedimentology.
Springer-Verlag, p5-l7.

Hammer. I'. T.. (1986). — Saline Lake Lcosysrems of the H'orld. Dr. W'.
Junk Publishers.

Hammer, IL T. Shamess. J.. and Haynes. R. C. (1983). — The distribution
and abundance of algae in saline lakes of Saskatchewan. Canada.
In: U. T. Hammer (cd.) Saline Lakes Proc. 2nd Int. Symp
Aihalassic Inland Saline Lakes. Dev. Hydrobiol 16 Junk. The
Hague. p)-26.

Hill, J. R. Jr. (1978). — Corps of Engineers efforts related to Wetland
Protection. In: Montanan, J. H. and Kusler. J. A. (Ed.) Proceedings
of the National Wetlands Protection Sympo.sium. 1977. Reston. Va.
O.S. Dept of the Interior Fish and Wildlife Serv ice FWS/OBS-
78/79. Washington D.C.

Hofsletter. R. H. (1983). — Wetlands in the United Slates. In; Gore. A. J. P.
(Ed.) Ecosystems oJ the H 'orld. 4B. Mires; Swamp. Bog, Fen and
.Moor. Regional Studies. Elsevier Scientific Publishing Co.. p201-
244.

Holmes, A. (1944). — Principles of Physical Geology. Nelson. p437-440 and
P67I-675.

Howard Williams. C. and Gaudet. J. J. (1985).— The Structure and
functioning of African Swamps. In: Denny. P. (Ed.) The Ecology (S
Management of .African Wetland Vegetation. VV. Junk Publishers,
p 153-1 76.

Hutchinson. G. E. ^1957). — A Treatise on Limnology. 1 V>/. I. Wiley & Sons.

Ivanov. K, E. ( 1 98 1 ). — H atcr .Movement in .Mirelands. .Academy Press.

Jacobs. S. W. L. (1983). — Vegetation. In: Haigh. C (Ed.). H'etlands in
N.S.W. Parks and Wildlife, pi 4-1 9.

Junk. W. J. (1983). — Ecology ofSwamps on the Middle Amazon. In; Gore,
,A. J. P. (Ed.) Fcosystetm of the H'orld. 4B. Mires; Swamp. Bog. Fen
and Moor. Regional Studies. Elsevier Scientific Publishing Co..
p269-294.

Kulczynski. F. (1949). — Peal bogs of Polesic, Mem .icad. Sei. Cracovic..
Ser. B: 12356.

Leopold. L, B.. Wolman, M. G. and Miller. J. P. (1964). — Fluvial Processes
in Geomorphology. W. H. Freeman & Co.. San Francisco.

110

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

Logan. B. W.. Read. J. F-, Hagan. G. M., Hoffman, P.. Brown. R. G..
Woods, P. J. and Gcbclcin. C. D. (1974). — Evolution and
Diagencsis of Quaternary Carbonate Sequences, Shark Bay,
Western Australia. .-Iw. /I.v5w. Petrol, (ieoi, Mem. 22.

McArthur, W. M, and Beiienav, E. (I960). — The Development and
Disinbulion of the Soils of the Swan Coastal Plain Western
Australia, CSfROSoU Publicaiton'^io. 16.

Martin. A. C.. Hotkiss. N.. Uhler, F. M. and Bourn, W. S
(1953). — ClassiHcation of Wetlands of the United States. V.S. Fbh
Wild!. Servian. Spec. Sa Rep. H ildl. No. 20.

McComb, J- .A. and McComb, A. J. (1967). — A Preliminary .Account of the
Vegetation of Loch McNcss, a swamp and fen formation in
Wc.slem Australia. Roy. Soc. H'.A . 50: 105-112.

Miall. A. D. (1977). — A Review of the braided river depositional
Environment: i'tirr/j 617. Rev., 13: 1*62.

Mill. H. R, (J900-19I0). — In: Sir Dudley Stamp (Ed.) 1966. A Glossary of
Geographical Terms.

Monkhouse. E. J. (1965). — A Dictionary' of Geography. E. Arnold and
Chicago Aldinc.

Moore. P. D. and Bellamy. D. J (1974). — Peatlands. London. Elek Science,
p9-183.

Moore. W. G. (1949). — A Dictionary of Geography. Penguin Books.
Harmondsworlh. Middlesex.

Morisawa. M, (1968). — Streams.' Their dynamics and morphology. New
York. McGraw-Hill Book Co.

Muir, B. G. (1983). — Drainage. Swamp Structure and Vegetation
Succession at Melaleuca Pk. Northern Swan Coastal Plain.
Herbarium Research Notes. No. 9: 27-39.

Paijmans, K.. Galloway. R. W., Faith. D. P., Fleming. P. M.. Haantjens, H.
A.. Hcyligers, P. C.. Kalma, J. D. and Loflier. E. (1985). — Aspects
of Australian Wetlands. CSIRO Aust. Div. Hater Land Resources
Techntcal Paper No. 44: 1-71.

Pen. L. J (1983). — Pcnpheral Vegetation of the Swan and Canning
Estuaries 1981. Dept, of Conservation and Environment. Bull. 113.

Pisano, E. (1983). — The Magellanic Tundra Complex. In; Gore. ,A. J. P.
(Ed.) Ecosystems of the World. 4B. .\fires. Syy-amp. Bog. Fen and
Moor. Regional Studies. Elsevier Scientific Publishing Co.. p295-
330.

Reeves, C. C. Jr. (1968). — Introduction to Palaeoltmnology. Elsevier.

Reineck, H. E. and Singh, I, B. (1973). — Depositional Sedimentary
Environments. Springer- Verlag.

Riggert, T. L. (1964-1966). — Wetlands of Western Australia Department of
Fisheries and Fauna West. Aust.

Ruttner. F. (1953). — Fundamentals of Limnology University of Toronto
Press.

Ruuhijarvi, R. (1983). — The Finnish Mire Types and Their Regional
Distribution. In: Gore. A. J. P. (Ed.) Ecosystems of the World. 4B.
Mires; Swamp. Bog. Fen and Moor. Regional Studies. Elsevier
Scientific Publishing Co.. p47-68.

Schumm, S. A. { 1977). — The Fluvial System. John Wiley & Sons.

Seddon. G. { 1 972). — A Seme of Place. University of W,A Press.

Scmcniuk, V. (1986). — Terminology for geomorphic units and habitats
along the tropica] coast of Western Australia. J Roy. Soc. W.A. 68:

Shipley. J. T. (1982). — Dictionary of Word Origins. Littlefield. Adams &
Co.

Sjors, H. (1983). — Mires of Sweden. In: Gore. A. J. P. (Ed.) Ecosystems of
the World. 4B. .Mires: Swamp, Bog. Fen and .Moor. Regional
Studies. Elsevier Scientific Publishing Co.. p69 94.

Smith, D. G. (1983). — Anastomosed fluvial deposits: modern examples
from Western Canada. In; J. D. CoHinson and J. Lewin (Ed.)
Modern and Ancient Fluvial Syslenu. Special Publication Number
6 of the Inlcmalionql Association of Sedimcnlologisis. Blackwell
Scientific Publications. pl55-!68.

Speck. N. H. and Baird, A. M. (1984). — Vegetation of Yule Brook Reserve
near Perth, Western Australia, ./, Roy. Soc. H 66: 147-162.
Slrahler, A. N. and Strahlcr. A. H. (1978). — .Modern Physical Geography.
John Wiley Sops.

Swayne. B. (1956). — A Concise Glossary of Geographic Terms. In: Sir D.
Stamp (Ed.) ( 1 966) .4 G/os.sary of Geographic Term.s.

Tansley. A. G. (1939). — The British Islands and Their Vegetation.

Cambridge University Press. Cambridge.

Tingay. A. and Tingay, S, R. (1976). — ^The Wetland of System Six. Dept.
Conservation & Environ. Bull. 28.

Trowbridge, A. C. (Ed.) ( 1 962). — Dictionary of Geological Terms. American
Geological Institute, Washington D.C
Twidale, C. R. (1968). — Geomorphology. Nelson, pl48-203.

UNESCO (1971), — Article I. part 1, Convention on wetlands of
international significance especially as waterfowl hibitats.
Published in Australia, 1976, for Department of Foreign Affairs by
Australian Government Publishing Service, Treatv Series 1975
No. 48.

Watson, L. E. and Bell. D. T. (1981). — The Ecology of Star Swamp and
surrounding bushlands. North Beach. Western Australia. J. Roy.
Soc. W.A..63: 103-117.

Wetlands Advisory Committee ( 1 977) — The Status of Welland Reserves in
System 6. Report of the Wetlands Advisory (Tommittee to the
Environmental Protection Authority.

Whiitow. J. B. ( 1984). — Pengidn Dictionary of Physical Geography. Penguin
Books Hammondswonh. Middlesex.

Wetzel. R. G. ( 1 983). — Limnology. Saunders College Publishing.

Zoltai, S. C. and Poiletl, F. C. (1983). — Wetlands in Canada: Their
Classification, Distribution and Use. In: Gore, A. J. P. (Ed.)
Ecosystems of the World. 4B. Mires: Swamp. Bog, Fen and Moor.
Regional Studies. Elsevier Scientific Publishing Co., p245-268.

Ill

W'

fX p.

m

n

i?;

X

\4^“

5^

>h ^

* ;

#

►•<^K

rH-#

IV

«i

a

£?

I*-

*» -*

i

•- *1‘ . }*!

?>

A»

V«?

V ^

/ *

Kii

't

1^

iiA-

^ isi- -V W

s.*£

r

? -f- <

«•■!''. '?'.■' ■■ ‘ i.'-.

Kj

^ * fn _<r

* >*

'P' #

». ... 'r i' ./ a'* X* .* .

^ tbP ^ vA

V;> ..

''n. V**"

t

^ '-I. P r<:

- *' r'^ vi« A

^ i .-P y€ > X J*X

.. , .,....;.-3ii^-'y :'-.^'^i,^-. ?ys.:

-» V',-'

^ ^». ^ -f- ^ ^ $!d-^r.

c4i^ h

, ,.«^'*‘‘V A iVf -r ^\ivT> ^ ^ ^

" ’* 't'

pfKPti'^T'' %

* -f .. 1^ yAiT> 'v

V' ^ V *- >lf ^ ^ .

•m

%.W85S«>--,SS4i-f>'.V' 5 *'>i!t' ,*4?.. ' < A.-.'

V >>

^#,^7, %V*A - ^ .c t.

.. ■ ^‘4S:^iv- yiiv*

. ,- P-. '

■e*® s-i

-»-.<'#>i- 5 ff;

1 '4r • "^j ' 'AA'* "^V

.■>sr

■ /%..■ ■'. ■•'’ '':;:'/ 7 /--' v-.'-.' . •■",

^ ^ 1 - 2 '' £i*

i' < 'S 'V>3.i ", f‘fC-^■'^■^•‘f^ 'a ''’^ r>-f

’* ^ ' **; .< ^ k'^ ’ t- ;v£^

A*

*SyX-'

;.7^v^'^cv -»?.

f i. "i >i^\-

K^xf.

r<»v

,< -, ; ., ^ ‘^,*t;3. }► Ij -JA- ^ / 'J4? •’

> ;

f4?

fii .’f' ^..v ' . I: A' ‘ 'C

k‘^'€'-m

Journal of ihc Royal Society of Western Australia. Vol. 69. Part 3. 1 987. p. 113-116.

Rb-Sr Geochronology of granitoids from Mount Mulgine, Western Australia

by J. R. De Laeter and J. L. Baxter

School ofPhysics and Geosciences. Curtin University of Technology. Bentley. Western Australia. 6102.

Manuscript received 15 July 1986; accepted 16 December 1986

Abstract

The Mulgine Granite is highly fractionated with greisen sheets containing tungsten-molybdenum
mineralisation. The Mulgine Granite forms a dome-shaped body which has been intruded by an
unmineralised discordant porphyritic-biotite adamellite.

The best estimate of the Rb-Sr whole rock age and initial ^^Sr/^^Sr ratio of the Mulgine Granite is
2684 * 79 Ma and 0.701 ± 0.005 respectively. It is unlikely that it had an extended crustal pre-
history. The cross-cutting porphyritic-biotite adamellite has a Rb-Sr whole rock age of 2596 ± 35
Ma with an initial ^^Sr/®%r ratio of 0.704 i 0.004. Although the experimental errors do not show a
significant difference in the ages, the isotopic data indicate that the porphyritic-biotite adamellite is
younger than the Mulgine Granite: a result consistent with the field relationships. A subset of the
samples of the porphyritic-biotite adamellite gives an approximate Rb-Sr whole rock age of 2330 Ma
and an initial ^^Sr/^^Sr ratio of 0.713. Two Rb-rich samples of granitoid from the Mulgine Granite
give similiar model Rb-Sr ages which indicate that a later thermal event has affected both granitoid
suites.

Introduction

The Yilgarn Block in Western Australia is one of the
largest segments of .Archaean crust in the world. Gee et
al. (1981) subdivided the Yilgarn Block into three
granitoid-greenstone provinces (the Murchison,
Southern Cross and Eastern Goldfields provinces), and a
predominantly gneissic terrain (the Western Gneiss
Terrain), which occupies an area around the western
periphery' of the granitoid-greenstone province.

Granitoids account for about 70 per cent of the crustal
exposure of the Yilgarn Block, which has an area of
approximately 650,000 km- (Gee et al. 1981). Older
synkinemalic granitoids have been described by
■Archibald and Bettenas (1977) and Watkins and Tyler
{ 1985) in the Kalgoorlie and Cue districts of the Yilgarn
Block respectively. Younger post-tectonic granitoid
batholiths intrude these foliated rocks.

The most common compositional types in both
synkinematic and post-tectonic plutons arc adamellite
and granodiorite in contrast to many other granitoid-
greenstone terrains where more tonal ite and

trondjcmilic rocks predominate (Libby 1979).

In the Murchison Province the greenstone belts arc
intruded by a number of discrete or coalescing ovoid
plutons (Muhling 1969). The synkinemalic granitoids
are emplaced prior to deformation and metamorphism
within the greenstone belts (Watkins and Tyler 1985),
each commonly containing co-planar ' foliation.
Deformation and metamorphism in the synkinematic
granitoids and greenstone belts are consequently often
co-eval.

From an extensive Rb-Sr whole-rock study of
granitoids and gneisses across most of the Yilgarn Block,
Arriens (197!) found three distinct episodes between
3100-2900 Ma, 2700-2550 Ma and 2300-2200 Ma.
.Arriens pointed out that many of the granitoids which
had Rb-Sr ages between 2700 and 2600 Ma have an
initial ^^Sr/^^r ratio of approximately 0.704 suggesting
a crustal pre-history for these samples. .Arriens (1971)
estimated that these granitoids could not have existed
for longer than 120 Ma as crustal materials, assuming an
average Rb/Sr ratio for the granitoids of 0.85. However.
Gee et aL (1981) argue that this period of crustal
prehistory may be 200 to 400 Ma in duration. Oversby
(1975) considers that the high ^ values recorded in some
Eastern Goldfields granitoids indicates that these rocks
had existed as crustal material for at least 300 Ma before
the - 2600 Ma lectono-ihermal event. On the other hand
Bicklc c/ al. (1983) argue that Pb-Pb whole rock isotopic
data from synkinemalic plutons in the Diemals area
indicate that the crustal history of the precursor material
is less than - 200 Ma.

De Laeter et al. (198 la) have reviewed the
geochronological data in the Yilgarn Block. Few results
from the Murchison Province have been reported.
Granitoids from the Cue-Mount Magnel-Paynes Find
region give a Rb-Sr whole-rock age of 2706 ± 264 Ma
(.Arriens 1971). Two muscovites from Mount Mulgine
give model Rb-Sr ages of 2632 and 2614 Ma (Arriens,
1971 ). Muhling and De Laeter (1971) also reported Rb-
Sr whole-rock ages for granite-adamellite and
granodiorite from the Poona baiholith (a complex
granitoid body cast of Cue), of 2535 ±. 23 Ma and 2550
± 51 Ma respectively. Fletcher (Pers. Comm.) has
obtained Sm-Nd model ages of 2820 Ma and 2730 Ma
on two of these samples.

Figure I. Regional map of Mount Mulginc in the Murchison Province of Western Australia.

1 14

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

Geology of the area

The Mt. Mulgine district contains a mafic volcanic
supracrustal succession intruded by the synkinematic
Mulgine Granite and a discordant porphyritic (lath
granite) adamellite (Fig. 1). Both granitoid suites are
unusually highly fractionated with Rb-Sr ratios ranging
from 0.88 to 49.5.

The Mulgine Granite is a 1 km diameter ovoid stock
near the axis of a regional anticline. The northern,
eastern and western margins of the stock are parallel to
layering in the adjacent volcanic sequence (Baxter ! 979).
Foliation and lineation in the volcanic rocks and the
granite are parallel. Greisen sheets intrude the volcanic
sequence along the northern perimeter of the stock. The
greisen has induced polymetallic scheelile-molybdeniie-
fluorite mineralisation in potassium metasomatized
ultramafic rocks in the adjacent volcanic sequence.
Pegmatites and greisen sheets associated with the
Mulgine Granite occur up to 2 km from the stock
contact. The volatile rich nature of the Granite is
indicated by the wide halo of hydrothermal alteration
around the stock.

Porphyritic-biotite adamellite intrudes the southern
margin of the Mulgine Granite (Fig. 1). This is an
apophysis of a batholith of coarse-grained porphyritic
and even-grained adamellite lying to the south and east
of Mulgine Hill.

give model ages of 2362 Ma and 2282 Ma respectively.
The choice of initial ratio does not affect the model ages
to any significant extent.

Subsequently, another five drill-core samples of the
Mulgine Granite were obtained, and if these are
combined with the original seven samples, a Rb-Sr
whole rock age of 2644 ± 17 Ma and Rj = 0.7026 ±
0.0012 is obtained with a MSWD of 28. Thus the effect
of the additional samples is to increase the scatter in the
data, and to lower the age slightly. A Model 2 assessment
of the age, initial ratio and errors is 2643 ± 87 Ma and
^ Rj = 0.7026 ± 0.0061 respectively.

Experimental procedures

Samples were reduced to -200 mesh using a jaw
crusher and an agate Tema-type mill. After chemical
extraction, the samples were analysed in a 30.5 cm
radius of curvature, 90* magnetic sector field, solid
source mass spectrometer. Techniques are essentially
those reported by De Laeter ei al. (1981b).

The value of *^Sr/^^Sr for the NBS 987 standard was
0.7102 ± 0.0001 normalised to a *^Sr/^^Sr value of
8.3752. A value of 1.42 x 10'”yr’* was used for the decay
V constant of ®^Rb. All the Rb-Sr ages quoted in this paper
have been corrected where necessary, to this decay
constant. The data have been regressed using the least
squares programme of McIntyre et al. (1966). All errors
are given at the 95% confidence level.

Results and discussion

Two suites of samples were available for analysis — the
Mulgine Granite and the postkinematic discordant
porphyritic-adamellite. Isotopic data are listed in Tables
1 and 2 for these two granitoids.

Mulgine Granite

Initially nine samples of the Mulgine Granite were
collected and subsequently analysed. Seven of the nine
samples form an isochron with an age of 2694 ± 30 Ma
and initial ^"^Sr/^^Sr ratio (R,) of 0.7000 ± 0.0018 (Fig.
2). The mean square of weighted deviates (MSWD) is
6.3, indicating that there is a real dispersion in the data
greater than the experimental errors associated with the
measurements. A better estimate of the age, initial ratio
and associated errors is 2684 ± 79 Ma and R, = 0.7007

± 0.0053. This Model 4 age implies a real scatter in age
and initial ratio. Taking into account the average Rb/Sr
ratios of these samples, it is unlikely that there was an
extended period of crustal pre-history for these samples.
The remaining two samples (numbers 255 and 247) have
a much higher Rb/Sr ratio with respect to the group, and

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75

"Rb/”Sr

R7 R7 R/x

Figure 2. — Sr/®°Sr vs ® Rb/^°Sr isochron diagram for the Mount Mulgine
Granite.

It is of interest to note that the average age for the two
muscovite samples reported by Arriens (1971) is 2623
Ma. These two samples, presumably from pegmatites
associated with the Mulgine Granite, were obtained
from drill-core by Newmont Pty Ltd, who held leases for
molybdenum prospecting at Mount Mulgine. The Rb/Sr
values for these muscovites are exceptionally high.

Porphyritic Adamellite

Ten samples of the porphyritic-biotite adamellite were
originally analysed and the results are displayed in
Figure 3. Six of the ten samples fall on an isochron of age
2570 ± 65 Ma and Rj = 0.7055 ± 0.0071 with a MSWD

115

Journal of the Royal Society of Western Australia, Vol. 69, Part 3, 1987.

of 2 . 4 . Subsequently three additional samples were
analysed, and if the resulting nine samples are fitted to
an isochron, an age of 2596 ± 35 Ma and = 0.7039 ±
0.0043, with an improved MSWD of 1.8 is obtained.

The remaining four samples (numbers 270, 266, 267
and 271) give a poorly fitted isochron with a Model 3
age and initial ratio of 2332 ± 296 Ma and 0.7133 ±
0.062 respectively. This age, despite its large error, is in
good agreement with the average model age 2322 Ma, of
samples 255 and 247 from the Mulgine Granite.

Conclusion

The Rb-Sr whole rock analyses of two suites of
granitoids from the Mount Mulgine district indicate that
the Mulgine Granite is older than the discordant
porphyritic-biotite adamellite, thus supporting the field
relationships. The low initial ratio of the Mulgine
Granite suggests that the rocks were derived from the
mantle, with a relatively short crustal history prior to the
Rb-Sr isochron age. The best estimate of this whole-rock
age is 2684 ± 79 Ma with an initial ratio of 0.701 ±
0.005. However if one takes into account all the isotopic
evidence available, it could be argued that an age of
approximately 2650 Ma and an initial ratio of 0.702 is
more appropriate for the Granite.

The discordant porphyritic biotite adamellite gives a
younger age of 2596 ± 35 Ma with an initial ratio of
0.704 i 0.004. In terms of the errors associated with the
two sets of samples, it is not possible to state that the
ages and initial ratios are significantly different,
although the isotopic data are consistent with the field
evidence.

A subset of the porphyritic adamellite suite gave an
approximate age of 2330 Ma with a high initial ratio of
0.713. This younger age is in good agreement with model
ages from two Rb-rich samples from the Mulgine
Granite, and reflects a younger Proterozoic
metamorphic event which has reset some of the samples
from each granitoid suite. It is possible that this event
could be associated with the basic magmatic
emplacement of Proterozoic dykes in the Yilgam Block.

Acknon/ec/gefnenfs —

This project was proposed by Associate Professor D, I. Groves of the
Geology Department of the University of Western Australia, who also
commenled on an earlier draft of this manuscript. Initial field work and
sample collection was earned out by Mr F. D. Turner, under the
supervision 9 f Associate Professor Groves. The authors would like to thank
Dr K. Watkins. Mrs P. R. Harris. Mr E. Schmidt and Mr D. J. Hosie for
their assistance! This project was supported by the Australian Research
Grants Committee.

References

Archibald, N. J. and Beitenay. L. F. ( 1 977). — Indirect evidence for tectonic
reactivation of a pre-greenstone sialic basement in Western
Australia £arlh Planet. Set. Lett., 33: 370-378.

Arriens, P. A. (1971). — The Archaean Geochronology of Australia. Spec.
Publ. Geol. Soc. Aust., 3 : 1 1-23.

Baxter, J. L. (1979). — Molybdenum, tungsten, vanadium and chromium in
Western Australia, Aust. Geol. Survey. Min. Res. Bull.

Bickle, M. J., Chapman. H. J.. Bcticnay, L. F.. Groves, D. I. and De Laeier.
J. R, (1983). — Lead ages, reset Rb-Sr ages and implications for the
Archaean Crustal Evolution of the Diemals area. Central Yilgam
Block, Western Australia. Geochim. Cosmochim. .Acta, 47: 907-
914.

De Laeter. J. R., Libby, W. G. and Trendall, A. F. (1981a). — The older
Precambrian Geochronology of Western Australia. Spec. Publ.
Geol. Soc. /lujl.. 7:145-157.

De Laeier, J. R.. Williams. I. R.. Rosman. K. J. R. and Libby, W. G.
(1981b). — A definitive 3350 Ma age from banded gneiss. Mount
Narryer area. Western Gneiss Terrain. West Aust. Geol. Surv. Ann.
Rept. 1980:94-98.

Gee, R. D„ Baxter, J. L., Wilde, S. A. and Williams, I. R. (1981). — Crustal
development in the Archaean Yilgam Block, Western Australia.
Spec. Publ. Geo!. Soc. .Aust., 43-56.

Libby, W. G. (1979). — Regional variation in granitic rock. Report Geol.
Surv. 9 : 53-109.

McIntyre. G. A., Brooks. C., Compslon, W. and Turek, A. (1966). — The
statistical assessment of Rb-Sr isochrons. J. Geophvs. Res., 71:
5459-5468.

Muhling, P. C. (1969). — The geology of the granitic rocks of the Poona-
Dalgaranga area, Murchison and Yalgoo Goldfields West. Aust.
Geol. Surv. Ann. Rept l968:5C)-52.

Muhling, P. C. and De I.aeter. J. R. (1971). — .Ages of granitic rocks in the
Poona-Dalgaranga area of the Yilgam Block, Western Australia.
Spec. Pub! Geol. Soc. Au.'tt., 3 : 25-31.

Oversby. (1975). — Lead Isotope systemalics and ages of Archaean acid
intrusives in the Kalgoorlie-Norseman area. Western Australia.
Geochim. Cosmochim. Acta, 39 : 1 107-1 125.

Watkins, K. P. and Tyler. I. M. (1985). — Structural and stratigraphic
relationships in the Archaean granite-greenstone terrain around
Cue, W.A- West. Aust. Geo! Sun.. Rept. 14: 46-56.

116

INSTRUCTIONS TO AUTHORS

The Journal publishes (after refereeing)

^papers dealing with original research done in Western Australia into any branch of the natural sciences;

^papers concerning some biological or geological aspect of Western Australia:

^authoritative overviews of any subject in the natural sciences, integrating research already largely published in
the more specialized national or international journals, and interpreting such studies with the general
membership of the Society in mind:

^analyses of controversial issues ofgreat scientific moment in Western Australia.

Prospective authors of papers in the last two categories should consult the Hon. Editor for further advice.

C ontributions should be sent to I'he Honorary Editor. Royal Society of Western Australia. Western Australian
Museum, Francis Street, Perth, Western Australia, 6000. Publication in the Society’s Journal is available to all
categories of members and to non-members residing outside Western Australia. Where all authors of a paper live in
Western Australia at least one author must be a member of the society. Papers by non-members living outside the
State must be communicated through an Ordinary or an Honorar>’ Member. Submission of a paper is taken to mean
that the results have not been published or are not being considered for publication elsewhere. Free reprints are not
provided. Reprints may be ordered at cost, provided that orders arc submitted with the return galley proofs. Authors
arc solely responsible for the accuracy of all information in their papers, and for any opinion they express.

Manuscripts. The original and two copies must be submitted. They should be typed on opaque while paper with
double-spacing throughout and a 3 cm margin on the left-hand side. All pages should be numbered consecutively,
including those carrying tables and captions to illustrations, which appear after the text. Illustrations, both line
drawings and photographs, arc to be numbered as figures in a common sequence, and each must be referred to in the
text. In composite figures, made up of several photographs or diagrams, each of these should be designated by a letter
(c.g. Figure 2B).

Authors are advised to use the most recent issue of the Journal as a guide to the general format of their papers.
Words to be placed in Italics should be underlined. To facilitate editing, papers must be accompanied by a table of
contents, on a separate sheet, showing the status of all headings.

References must be set out as follow's:

Paper Jackson A 1931 The Oligochacta of South-Western Australia. J R Soc West Aust 17:17-136.

l^ook Jacobs M R 1955 Growth Habits of the Eucalypls. ForTimb Bur, Canberra.

C liaptcr in book Dell J 1 983. The Importance of the Darling Scarp to Fauna. In: .Scarp Symposium (ed J D Majer).
W A Inst Technol, Bentley. 17-27.

The Title should begin with a keyword. The Abstract should not be an expanded title, but should include the main
substance of the paper in a condensed form. The metric system (SI units) must be used. Taxonomic papers must
follow the appropriate International Code of Nomenclature, and geological papers must adhere to the International
Stratigraphic Guide. Spelling should follow the Concise Oxford Dictionary.

Authors should maintain a proper balance between length and substance, and papers longer than 1 0 000 words would
need to be of exceptional importance to be considered for publication. Authors will normally be charged page costs
where papers exceed 8 printed pages.

Illustrations. These should be prepared to fit single or double column wudihs. Illustrations must include all necessary
lettering, and be suitable for direct photographic reduction. No lettering should be smaller than I mm on reduction.
To avoid unnecessary handling of the original illustrations, which arc best prepared between 1.5 and 2 times the re-
quired size, authors are advised to supply extra prims already reduced. .Additional printing costs, such as those for fold-
ing maps or colour blocks, will be charged to authors.

Suppleinenlary Publications. Extensive sets of data, such as large tables or long appendices, may be classed as
Supplementary Publications and not printed with the paper. Supplcmentarv Publications will be lodged with the
Society’s Library (C7- Western Australian Museum. Perth. WA 6000) and with the National Library of Australia
(Manuscript Section. Parkes Place. Barton, ACT 2600) and photocopies mav be obtained from either institution
upon payment of a fee.

JOURNAL OF THE

ROYAL SOCIETY OF WESTERN AUSTRALIA
CONTENTS VOLUME 69 PART 3 1987

Page

Grazing pressure by the tammar {Macropus eugenii Desm.) on the vegetation of Garden Island, Western
Australia, and the potential impact on food reserves of a controlled burning regime. D. T. Bell, J. C.
Moredoundt and W. A. Loneragan 89

Wetlands of the Darling System — A geomorphic approach to habitat classification. C. A. Semeniuk 95

Rb-Sr Geochronology of granitoids from Mount Mulgine, Western Australia. J. R. de Laeter and J. L.
Baxter

Edited by B. Dell and I. Abbott

Registered by Australia Post — Publication No. WBG 0351 .

No claim for non-receipt of the Journal will be entertained unless it is received within 12 months after publication of Part 4 of each Volume.
The Royal Society of Western Australia. Western Australian Museum, Perth.

53510/3/87-000-L/4223

By Authority WILLIAM BEN60W, Acting Government Printer