PROCEEDINGS

OF THE

ROYAL SOCIETY OF VICTORIA

VOLUME 93

ROYAL SOCIETY’S HALL
9 VICTORIA STREET, MELBOURNE 3000

I

*

' •

Contents of Volume 93 Number 1

Article Page

1 Coastal tree fern communities at Western Port, Victoria.By Fiona Ferwerda,

R. J. Williams and D. H. Ashton 1

2 Coastal archaeology in Victoria Part 2: Adaptation technology and volcanism

By P. J. F. Coutts 15

3 Past and present distributions and translocations of Macquarie perch Macquaria
australasica (pisces: Percichthyidae), with particular reference to Victoria

By P. L. Cadwallader 23

4 Three corophioids (Crustacea: Amphipoda) from Western Port, Victoria

By J. L. Barnard and Margaret M. Drummond 31

5 Ostracoda from Australian inland waters - notes on taxonomy and ecology

By P. De Deckker 43

Contents of Volume 93 Number 2

Article Pa S e

6 The Bruun Rule-the relationship of sea-level change to coastal erosion and

deposition.By H. Allison and Maurice L. Schwartz 87

7 A study of Mt Kororoit, Victoria-a small volcanic vent

By W. S. Fischer and L. Thomas 99

8 Chapman’s “Mallee Bores” and “Sorrento Bore” Ostracoda in the National
Museum of Victoria, with the description of Macidocksella new genus

By K. G. KcKenzie 105

9 Studies on Western Australian Permian brachiopods 2. The family

Rugosochonetidae Muir-Wood 1962.By N. W. Archbold 109

10 Stratigraphy, sedimentology and hydrocarbon prospects of the Dilwyn Formation

in the central Otway Basin of southeastern Australia By G. R. Holdgate 129

11 Tertiary fluvial sediments at Morrison, Victoria.By Paul Bolger 149

Short communication: %

Taxonomic status of the Victorian fossil whale, Ziphius ( Dolichodon )
geelongensis McCoy 1882 . By R- Ewan Fordyce 157

INSTRUCTIONS TO AUTHORS

Detailed lists of Instructions are available from the Executive Officer. The following

is an abbreviated version only.

Papers considered for publication may be Reviews,
Reports of experimental or descriptive research, or Short
Communications. Length of Papers may vary; Short
Communications should not be longer than 1,500 words.

Two copies of the manuscript, with any accompanying
Plates and Figures, should be submitted initially to the
Executive Officer at the Society’s Hall, 9 Victoria Street,
Melbourne, 3000.

Manuscripts must be typed on quarto or A4 paper,
double-spaced on one side of paper only, and with ample
margins. Captions to Text Figures and Explanations of
Plates must be attached to the Ms as final pages. Underlin¬
ing should be restricted to generic and specific names of
biological taxa. Measurements must be expressed in the
metric system (SI units).

References should be listed alphabetically at the end of
the paper. Abbreviations of titles of periodicals must
conform with those in A World List of Scientific Periodi¬
cals (1963-4, 4th ed., London, Butterworth). References
to books should give the year of publication, number of
edition, city of publication, name of publisher. Titles of
books and (abbreviated) names of periodicals should be
underlined, in the typed list of references.

Maximum size for Plates is 15*5 cm x 21 cm. Do
not mount photographs. These must have clear
definition, may be submitted as either glossy or flat
prints, at actual size. The Plate number should be
clearly labelled on the back of the photograph.

Line drawing for Text Figures should be made in black
ink on white card or drawing linen. Maximum size (full
page) is 15*5 cm x 21 0 cm: single column width

Honorary Editor:

Dr. L. A. Frakes,

Department of Earth Sciences,
Monash University,

Clayton, Victoria 3168.

Assistant Editor;

Dr. P. A. Jell,

National Museum of Victoria,
285-321 Russell Street,
Melbourne, Victoria 3000.

7*5 cm. Figures are preferably submitted at actual size:
they may well be drawn larger and photographed by the
author to be submitted as glossy prints of required size.
Graphic scales must be included with the drawings, and
on maps and geographical plan views, compass directions
should be indicated. Lettering on Figures must be inserted
by the author, and special care is needed to ensure that all
letters and numerals arc still readable when the Figure is
reduced.

Oversized illustrations, tables or maps are accepted for
publication as Foldouts only with the understanding that
the author meet any additional costs involved in their
production. Maximum size for Foldouts is 21 cm x
31 *5 cm.

Short Tables will be typeset within the text. Extensive
Tables which arc likely to cover one page of print should
be typed, on an electric typewriter to ensure clarity and
evenness, within the dimensions 27*9 cm x 22*5 cm.
They should then be photographed to reduce them to full
page size, 15*5 cm x 21 cm, and submitted as glossy
prints. They will be reproduced photographically, not
type-set, and hence must be finally correct when submit¬
ted, since they cannot be corrected at the proof stage.

The Society supplies authors with 25 Reprints free of
charge; joint authors receive a total of 30. Additional
reprints may be ordered by returning the order blank
which accompanies the galley proofs.

The author of a paper or the Institution from
which a paper is received may be requested to con¬
tribute towards the cost of publication 15% of the
cost of printing that paper.

Honorary Research Secretary:
Dr. I. A. Staff,

Department of Botany,

La Trobe University,

Bundoora, Victoria 3083.

Executive Officer:

Dr. R. R. Garran,

Royal Society’s Hall,

9 Victoria Street,

Melbourne, Victoria 3000.

COASTAL TREE FERN COMMUNITIES AT
WESTERN PORT, VICTORIA

By Fiona Ferwerda, R. J. Williams, and D. H. Ashton

2 ? dcc mi

School of Botany, University of Melbourne, Parkville, Australia 3052

Abstract: An occurrence of the rough tree fern, Cyathea australis , on coastal bluffs near Point Leo,
Western Port, is described. The presence of this species, normally found as an understorey component in
cool, humid forests in Victoria, is unusual. It is concluded that the establishment and survival of Cyathea
australis in a coastal environment is dependent upon the existence of suitable micro-habitats, offering
sufficient soil moisture and protection from salt laden S and SE winds. Such micro-habitats are created by
small-scale rotational slumping, which occurs frequently in the study area. The slumping is in turn related
to a probable rise in water table levels, exacerbated by extensive clearing of native forest in the immediate
hinterland.

Tree ferns are characteristic of cool humid gullies in
Victoria, so it is surprising to find them on coastal bluffs
within 50 m of the high tide mark. The rough tree fern,
Cyathea australis , occurs on basalt bluffs at Shoreham,
Coles’ Beach, Merricks and Somers (Figs 1 & 2) on
Western Port. An isolated occurrence is found at the
base of cliffs composed of lateritized Tertiary clays at
Freeman’s Point on the NE coast of French Island. In
all coastal sites it is confined to wet or unstable soils.

The tree fern sites at Somers and Shoreham are at pre¬
sent being overwhelmed by blackberry thickets 2 to 3 m
deep. At the main Coles’ Beach study site, blackberries
had increased upslope alarmingly between 1974 and
1977, from well entrenched colonies at the bluff-dune
junction. The need to describe this unusual plant com¬
munity was therefore urgent and such a project was in¬
itiated in June 1977 by a group of second year ecology
students —Fiona Ferwerda, Jane Mallen, Wayne An-
trobus and Jan Brown.

DISTRIBUTION OF Cyathea australis

This tree fern ranges from Queensland to the NW tip
of Tasmania. In Victoria its distribution is generally
restricted to areas receiving an annual rainfall of over
750 mm. It is therefore most commonly found in hilly
highland regions up to altitudes of approximately 900 m
(Fig. 3). Where annual rainfall exceeds 1000 mm, it is
found as an understorey component in forests on valley
sides or even on ridge tops. In areas of lower rainfall,
however, it is confined to sheltered, moist habitats. An
interesting example of this in the Creswick area is its oc¬
currence along the moist sides of shallow mine shafts
abandoned since the late 1800’s (R. Hatelv pers. comm.
1980).

The occurrence of C. australis along the coastal slopes
of Western Port lies close to its general annual rainfall
limit in Victoria, and as a consequence it is found only
on the cooler east to south aspects and where there is
moderate protection from salty onshore winds (Fig. 4a,
b). It is absent from the taller bluffs on the Bass Strait
coast at Flinders, where there are similar aspects, rain¬
fall and microtopography, but where there are probably

rougher seas and stronger and more salt-laden S and SE
winds. The tree fern communities at Shoreham and
Somers are protected by the canopies of Eucalyptus and
Banksia forests and woodlands, or by thickets of coast
tea-tree (Leptospermum laevigatum) nearer the shore.

At Western Port, Dicksonia antarctica is rare, and
single specimens only have been found in two localities
at Coles’ Beach. One occurs in the incised course of
Short Creek and the other, although salt burnt, grew in
a seepage area of Melaleuca ericifolia , 10 m from the
beach sand. In 1976 it was cut down and removed,
presumably by a gardener.

ENVIRONMENT OF THE WESTERN PORT SITES
Climate

The climate of the Western Port coastal region is
typically temperate and maritime. The well distributed
annual rainfall ranges from 750-900 mm and is relatively
* reliable (Fig. 5a, b). Frosts are neither common nor
severe. Very wet years (1 standard deviation above the
mean) have occurred this century nine times.

Geology and Soils

Accounts of the geology of the Mornington Peninsula
and Western Port region by numerous authors were
summarized by Spencer-Jones et al. (1975). The
Shoreham, Coles’ Beach, Point Leo area is dominated
by the Thorpdale Volcanics of Eocene age which are
locally overlain by veneers of Tertiary sands, sandy clays
and silts. Shore platforms may be variably covered by
calcareous dune and beach sand.

Grey to yellow podzolic soils are formed on Tertiary
sediments, whereas brown podzolic and brown earth
soils have developed from basalts. The degree of
leaching and texture differentiation in the profile is
dependent on topography and proximity to the sea.

METHODS

At Coles’ Beach, three transects were located at right
angles to the shore line, on a 300 m sector of the coast.
Transect I was established on a steep, exposed cliff;
Transect II on the unstable bluff supporting Cyathea

1

2

FIONA FERWERDA, R. J. WILLIAMS, AND D. H. ASHTON

Fig. 1 - Map of Western Port showing tree fern localities (+) on the coast. Coles’ Beach is between Mer-
ricks and Point Leo; the. French Island occurrence is at Freeman’s Point. Rainfall isohyets (mm) are
shown by the dashed lines. Rainfall data from Shapiro (1975).

Fig. 2-General map of the Coles’ Beach area showing the land
use of the hinterland since 1863.

australis; and Transect III on a stable bluff (Fig. 2).
Along each transect, 4 x 1 m quadrats were established
at 5 m intervals. In the tree fern community, a further 10
random quadrats were sampled. Within each quadrat,
Braun-Blanquet cover values were recorded for each
species. Soil holes were augered every 5 m, and soil tex¬
ture and colour were recorded. Slope profiles at each site
were surveyed with tape and clinometer or dumpy level.

TOPOGRAPHY OF THE COLES’ BEACH SITE

This area, up-thrown by the nearby Tyabb Fault,
slopes gently seaward, the rolling terrain terminating
abruptly at a height of 8-10 m above mean high tide
level. The coastal features are the product of marine at¬
tack and sub-aerial denudation from surface and sub¬
surface water. Gravitational displacement is clearly im¬
portant where internal resistance has been overcome by
water lubrication.

On the more exposed headlands (e.g. Transect I),
where sand sheets are absent, marine erosion has pro¬
duced wide shore platforms and reefs (Hills 1976, Bird
1976). The bases of cliff slopes are subject to marine at¬
tack by high tides and storm waves. Due to the local
resistance of less decomposed basalt, almost vertical
cliffs are produced which suffer frequent rock-falls and
slumps after heavy rains (Fig. 6). In this paper, ‘cliffs’ are
defined as those slopes greater than 35° on which there is

COASTAL TREE FERN COMMUNITIES

3

substantial exposure of underlying rock. The vegetal
cover of such cliffs decreases dramatically with increase
of slope (Fig. 7). ‘Bluffs’ are regarded as those vegetated
slopes less than 35°, which rise sigmoidally from the
beach. They may be further stabilized and degraded to
gentler slopes by the combined action of soil creep and
hill-wash (Bird 1977).

Stabilized bluffs are common between Merricks and
Point Leo where they reach slopes of up to 21° (Fig. 8).
At Coles’ Beach there is little evidence of slumping at
this relatively low angle of slope, and soil profiles at the
base reflect a general mixture of wind-blown dune sand
with clays, silts and loams washed down from above.
The original beach at this point occurs at a depth of 100
cm, where sand is mixed with flat, waterworn basalt peb¬
bles 2 to 15 cm in diameter.

Steeper bluffs of 30° to 33° are frequently unstable,
and complex slumping is a conspicuous feature which
has reduced the local slope angle to 10° to 20°. The ma¬
jor tree fern occurrences are associated with this diverse
microtopography (Fig. 9). At Coles’ Beach, judging
from prolific seepage and 0.5 m deep cracks in sloppy
clay, some areas are still moving. At many sites in this
area, mixtures of subsoil, basalt fragments and topsoil
have collected against the upper sides of tree trunks and

logs to depths of 30 to 40 cm (Fig. 10). Headward ero¬
sion from some slumps has now reached the top-soils of
the plateau edge, although others have ceased move¬
ment and the arcuate clifflets at their upper limits have
grassed over.

Due to the rotational nature of the slumps (Selby
1970), back-walls (1 to 3 m high) of decomposed basalt
have remained after the tongues of mud and debris have
extended out over areas 20 to 30 m long and 5 to 10 m
wide. The protected cirque-like basins which are so-
formed, are wet and hummocky and some incipient
drainage lines have been deflected by later flows. In
some places muds have extended out over the dune for
up to 4 m. The original slope surfaces, which are often
reduced to narrow interfluves by lateral erosion from ad¬
jacent slumps, frequently display incipient instability by
the presence of terracettes 1 to 10 m long and 10 to 30
cm high.

At the foot of the slope, erratic variations in clay and
sand horizons indicate a fluctuating history of earth
flows and dune encroachment. The original beach of
pebbles and yellow sand occurs at a depth of about 1.3
m. The nature of the slump deposits can be seen in Fig.
4c where truncation has been affected by high tide ero¬
sion between Transects I and II.

Fig. 3 —Victorian distribution of Cyathea australis per 10 minute grid. Shaded areas represent those
regions receiving greater than 1000 mm annual rainfall. Much of the map was based on data provided by

Mrs Betty Duncan, Monash University.

mmi

4

FIONA FERWERDA, R.

J.

WILLIAMS, AND D.

H. ASHTON

I-ig. 4—Interior of a woodland of (a) Cyathea australis—Acacia melanoxylon and Clematis aristata at Coles’ Beach, (b) A relatively
large tree fern of C. australis 120 cm tall, on a grassy slope facing SE at Coles’ Beach. This fern is often ‘salt burnt’ following on-shore
winds from this quarter. Eucalyptus viminalis woodland is in the background, (c) Alternating top soil bands and clay rubble in an
earth flow truncated by high tides at Coles’ Beach north.

COASTAL TREE FERN COMMUNITIES

5

Fig. 5a—Variation of the annual rainfall at Cowes, Phillip Island, 10 km east of Coles’ Beach.

Fig. 5b —Seasonal distribution of rainfall (solid line) and air temperature (broken line) at CovVes.
Cowes has the most complete set of rainfall statistics of any recording station near the study area. Annual
rainfall figures for the years 1954-1957 were missing from the record. These figures were interpolated
from a regression line correlating annual rainfall data from Cowes and Frankston over the period

1919-1975.

The occurrence of landslides in Victoria appears to be
primarily controlled by the rock or soil type. The
seasonal distribution of rainfall may be broadly related
to landslide activity, but heavy rains are usually only a
‘catalyst’ (Evans & Joyce 1974). Another factor causing
instability on the Coles’ Beach bluffs is likely to be the
behaviour of ground water, especially its depth fluctua¬
tions. A dumpy level survey of Short Creek indicated
that the water table spring at its source was 21 m above
high tide mark on 12/12/1979 at a distance of 210 m
from the beach (Fig. 11). This intersection of the water
table was about 2 m above the junction of the basalt

flows and the Tertiary sands and silty clays. The creek
was brackish near its source with a salinity of
5.1 %t.d.s. —a value only l/7th of that of sea water. The
water tables in the Coles’ Beach bluffs are likely to be of
shallow origin and intake recharge may be derived from
areas such as the broad sand-capped hill, 60 m above sea
level and 1 km inland, which was cleared for the original
‘Larnoo’ homestead in 1863. A very broad divide ex¬
tends directly from this hill along a 2° slope to the small
forested area on the ‘hinterland plateau’ above the
unstable bluff site (Fig. 2). It seems probable that
ground-water discharge occurs down this slope and is in-

6

FIONA FERWERDA, R. J. WILLIAMS, AND D. H. ASHTON

Table 1

The Percentage frequency of Species Occurrence in 10 Random Quadrats (1+4 m) in the Unusual
Woodland of Tree Fern (Cyathea australis) — Blackwood (Acacia melanoxylon) on slumped clays on the Coles'
Beach Site, 1978. Figures in brackets indicate the frequency where cover exceeds 5%.

LIFE FORM AND STRATA

SPECIES

FREQUENCY

Tree and shrub strata

Trees

Eucalyptus viminalis

20

Acacia melanoxylon

40 (30)

Banksia integrifolia

20

Shrubs

Helichrysum dendroideum

10

Bursaria spinosa

10

Tree fern

Cyathea australis

30 (20)

Lianes

Clematis aristata

80 (50)

Field and ground strata

Ground ferns

Pteridium esculentum

90 (40)

A diant urn aethiopicum

90 (10)

Graminoids

Poa labillardieri

100 (70)

Lepidosperma laterale

40

Holcus lanatus

30

Microlaena stipoides

20

Echinopogon ovatus

10

Dicotyledon forbs and scramblers

Geranium potentilloides

100

Galium parisiense

70

Acaena anserinifolia

40

Senecia jacobaea

40

Epilobium glabellum

30

Rubus ulmifolius

20

Stellaria pun gens

20

Sonchus asper

20

Dichondra repens

20

Lobelia alata

20

Oxalis corniculata

10

Glycine clandestina

10

G. latrobeana

10

Hypochoeris radicata

10

Rubus parvifolius

10

Cirsium vulgare

10

Gnaphalium japonicum

10

Bryophytes

Fissidens asplenifolius

50

Ptychomnion aciculare

10

Sematophyllum amoenum

10

Lophocolea semiteres

10

E. viminalis

form at Transect 1.

Fig. 7 — The relationship between the amount of bare ground
and the angle of slope in the Coles’ Beach area. Bare ground
cover was subjectively assessed and the curvilinear relationship
estimated by eye.

COASTAL TREE FERN COMMUNITIES

7

E.vimmalia

Fig. 8 —Vegetation and land profiles of the stable bluff along Transect III, Coles’ Beach. B = Banksiu;

E = Eucalyptus.

tersected by the slumping bluff surface about 20 m above
high tide approximately at the junction of the decom¬
posed basalt and the Tertiary sediments. The relative
stability of the bluffs to the north and south of the cen¬
tral area may be preserved by a deflection of ground
water discharge along the lateral drainage lines which
run parallel to the coast along slopes of 3 to 5° in this
area.

Sodium predominance in the clay complexes in this
maritime site is certain to render the soils more fluid and
dispersable in wet conditions (Leeper 1957). The soils of
the basalt areas are brown, friable loams over
red/yellow mottled sticky clays, whilst those on the Ter¬
tiary deposits are grey fine sandy loams over heavy
yellow clay. In both soil types the presence of magnetic
buckshot gravel in the lower A horizons suggests

periodic waterlogging —a prerequisite of mass soil move¬
ment.

It is well known that clearing of eucalypt forest for
pasture development can result in reduced water use and
a rise in the level of the water table. It seems highly likely
therefore that this has been the case in the Coles’ Beach
area following general clearing from 1910 to 1938. A
greater flow of ground water would not only increase the
likelihood of slumping but also tend to cause instability
higher up the slope of the bluff.

VEGETATION AT THE COLES’ BEACH SITE

At this site the largely intact vegetation appears as a
moderately low open-forest or woodland which
descends from an almost level plateau to a set of 2 or 3

■Trae lain Woodland —

Pln “ f k * T1 Aamzuti

.

] Sand
kXXS Sandy Clay
CsS\S3 Clay

EEEZ3 sin

I* Basalt Rock Ru
li*lag) Baach Rabttas

E3

§8

BS

i

1 f

f

i |

i

11

3 r

%

Fig. 9 — Vegetation and land profiles of the unstable bluff along Transect II, Coles’ Beach. B = Banksia;

E = Eucalyptus ; A = Acacia.

FIONA FERWERDA, R. J. WILLIAMS, AND D. H. ASHTON

Fig. 10-Build-up of hill wash against a young tree fern (12 to
13 years old), on the lower slopes of an earth flow, Coles’
Beach. The rate of accumulation against the 6 year old trunk
could be about 1 cm per year.

grassy dunes 1 to 2 m high and 30 to 40 m broad. It is
likely that these deposits with scattered, resistant
Banksia have protected some of the bluff vegetation
from the worst excesses of salt spray. The vegetation has
been mapped on a structural basis, and for convenience
named according to the dominant plant form. A purely
floristic analysis may have grouped the communities
differently but it is not likely to have modified the con¬
clusions. The major communities described below are
closely correlated with local topography and soil as
shown in maps (Figs 12, 13) and profiles (Figs 6, 8, 9).
Ground vegetation is often most luxuriant at the
dune/slope junction where moisture supplies are reliable
but not excessive. The tallest trees usually occur on the
mid to lower slopes in response to less exposure and bet¬
ter moisture status. An analogous situation occurs on
French Island near Freeman’s Point, where a narrow
zone of wet scierophyll shrubs (Olearia argophylla,
Rapanea howittiana) appear beneath taller E. obliqua
trees.

The communities are:

1. Sand Dune Complex consisting of;

(a) Foredune grassland of Spinifex hirsutus, Agropyron
junceum, Ammophila arenaria and Cakile maritima.

(b) Hind dune complex of eraminoids ( Scirpus nodosus,
Lagurus ovatus, Stipa cornpactus) together with Acaena
anserinifolia and scattered bracken. Small bushes of
Banksia integrifolia are establishing in areas of this
zone, either by seed or by root suckers. This situation
appears to be successional and is a similar pattern to
previous descriptions (Bird 1976).

2. Fernland of bracken — ( Pteridium exculentuni) at
the base of the main slope. This is often tangled with
Clematis aristata, associated with or without an
understorey of maidenhair fern (Adiantum aethi-
opicum). In many sites, vigorous patches of blackberry
(Rubus ulmifolia) have overcome the bracken stands.

3. Woodland of Banksia integrifolia on two types of

site; on the sands and mixed clay soils at the bluff/dune
junction and on the hind dune associated with dune
plants such as Rhagodia baccata . On some of the basalt
slopes, especially the lower slopes, it is associated with
an understorey of tussock grass ( Poa labilliardieri ) and
various forbs. It seems that these latter sites are similar
to those occupied by eucalypts. Some Banksia com¬
munities bear evidence of old eucalypt stumps and fallen
trees. Possibly, the eucafypt component requires fire for
adequate perpetuation whereas Banksia integrifolia may
continue to regenerate vegetatively.

4. Woodland of Acacia melanoxylon in association
with Cyathea australis and Clematis aristata on slumped
sites. This community has a low, scattered tree canopy
but the ferny, herbaceous stratum —consisting o [Adian-
turn aethiopicum, Pteridium esculentum, Poa labillar-
dieriy Geranium potentilloides and Galium paris -
iense— is luxuriant (Fig. 4a). Areas of damp clay-rubble
associated with the tree ferns are almost invariably col¬
onized by the moss, Fissidens asplenifolius. The percen¬
tage frequency of species in this community is illustrated
in Table 1.

5. Open Forest of Eucalyptus viminalis—E. obliqua on
the slopes. This is a predominantly grassy community
with an understorey of Poa labillardieri and forbs. On
the gentler slopes E. radiata and E. ovata also occur.
Banksia integrifolia is a common understorey or

(a)

200 160 120 60 40 Om

(c)

200 160 120 80 40 Om

Fig. 11 — Land profiles in the Coles’ Beach area in relation to
Tertiary sand and sandy clay cappings (stippled), the limit of
which is shown by a dashed line. The depth of the ground water
table along Short Creek was determined by survey offsets from
the transect lines.

a. Section 100 m west of Transect I. (Source: Contour Map,
10,000/4, 6, 1973). Vertical exaggeration x 5.

b. Survey of Transect II.

c. Survey of Short Creek.

Slumping areas occurred on the bluff slopes of each of
the three transects. The extensive slumping on (b) is
associated with a definite perched water table.

COASTAL TREE FERN COMMUNITIES

9

Fig. 12 — Landform map of the Coles’ Beach site showing the distribution of tree ferns.

° minant species and is associated with Helichrysum
oust r °!^ eum anc * Bursaria spinosa. Large-crowned C.
this > 6 sorn . et * mes occur on terraced grassy slopes in
Estahr^^ 0 ^ ancl are ^ rom l ^ e beach (Fig. 4b).
slope -! Shment 9^ Bittosporum undulatum into these
Mor ?° mniunitie s Is very common, as it is over much of
bird nin ^ ton Peninsula, where both seed sources and
speci VeCt0rS are present - The mature development of this
mu 9 s w bl undoubtedly greatly modify these com-
6 Low r~ aS the ‘ nvas ^ on °* blackberry thickets,
tered i ^ RASSY Woodland complex of shrubs and scat-
tepfif /° W trees °1 Casuarina stricta and Banksia in -
t 0 t |^ la ' In Edition, the highly disturbed cliffed site,
bunted n ° rt ^ ma ‘ n stu dy area, also supports

°f ft. ’ Prostrate Acacia melanoxylon , scattered bushes
lonotf 1 ^- 0 toccata and tussocks of both Lomandra
? a and Poa labi ll<trdteri.
radiatn^ Plateau Open-forest of E. obliqua, E.
ihese ^ • Vlm ^ na ^ s an d E. ovata. The frequencies of

Se ason ^ eCles c b an ge locally, probably in response to
as soci ^ at 9 r ^°Sgiug conditions. These dominants are
Xantho^ h Panthonia racemosa, Poa sieberiana ,
rrnoea minor and Leptospermum juniperinum.

The understorey has been modified to some extent by
cattle grazing.

STATUS OF DEVELOPMENT OF Cyathea australis

The majority of tree ferns on the west coast of
Western Port have trunk heights less than 1.5 m and
crown diameters of about 4 m. Flowever, at Shoreham
and Somers, occasional large specimens reach heights of
2.9 and 3.2 m respectively. The frequency histograms of
tree fern heights at Coles’ Beach, Shoreham and Somers
are compared in Fig. 14. It is clear that the Shoreham
stand is bimodally distributed whilst that at Coles’ Beach
has only one major peak of trunk heights. The coin¬
cidence of population frequency peaks at the two sites
may mean that establishment has been synchronised by
similar events. The reduced frequency peaks of taller
classes at Shoreham (and to some extent, Somers), sug¬
gest that there have been additional earlier phases of
establishment.

Cyathea australis at Coles’ Beach is actively
regenerating, since 41 % of plants are sessile, with crown
0.3 to 3.0 m in diameter. Young plants are found where

10

FIONA FERWERDA, R. J. WILLIAMS, AND D. H. ASHTON

A-B

Transect II

-Dune ridge |,'

Limits of bluff

C—0

Transect ID

u

*

Tree ferns (according to >iza>

1 Foredune grassland E

o

Pinetree

• ’ • Hind dune complex of graminoids L

Drainage line

Fernland of Pteridium escutentum

' v v I

VAX I Thickets of biackberries iBufcua sp)
Woodland Banfraia intnorilolia

Woodland Ama^anox vlon c. mmrain

Open Forest - E. vimnalis E otoliqua

Mixed plateau Open Forest

Fig. 13 —Map of the main structural vegetation units at Coles’ Beach.

Height (cm)

Fig. 14 — Frequency histogram of tree fern heights at the three
major tree fern localities — Shoreham, Coles’ Beach and
Somers.

the ground stratum of bracken, maidenhair fern,
grasses, forbs and Clematis vines is incomplete. At
Shoreham, regeneration is absent beneath dense
blackberries —the few small plants occurring only where
this weed is still sparse. At Somers, the situation is
similar, and regeneration occurs only where the dense
wreaths of the introduced Asparagus asparagoides are
patchy. It is expected that regeneration of the tree fern
at Coles’ Beach will eventually be curtailed by
blackberry growth. However, the regeneration of the
tree fern on moist clay surfaces is rapid, since at
Shoreham, fernlings appeared in late 1978 —within six
months of a track being graded through a tree fern site
that had been cleared of dense blackberries.

From general ecological observations, the mature tree
fern appears to be intolerant of prolonged drought and
excessive insolation. The development and survival of
the prothallus is likely to be dependent on the
maintenance of moist and locally humid microclimates.
These conditions are frequently satisfied by moist clay
soil on slopes where obliterating litter cannot ac¬
cumulate. At Coles’ Beach, 50% of all tree ferns occur
in seepage areas or erosion furrow's in and between clay
slumps; the remainder occur on open slopes where bare
soil has been exposed by terracettes, hillwash or wombat

COASTAL TREE FERN COMMUNITIES

11

5 6

Fig. 15 —Block diagrams depicting the possible sequence of development of the tree fern habitat at Coles’
Beach:

1. ClifF kept active by marine erosion.

2-3. With the build up of protective dunes, sub-areal processes denude cliffs to bluffs.

4. Hinterland cleared of forest causing a rise in water table levels.

5-6. Slumping activity increases, creating suitably moist habitats for the establishment and survival of
Cyathea australis.

12

FIONA FERWERDA, R. J. WILLIAMS, AND D. H. ASHTON

burrowing. A few in the latter sites have succumbed to
drought. The ages of slumps providing conditions for
tree fern establishment are not generally known. Since
the biology of this species is incompletely known, the
age structure of the stands can only be surmised.

Examination of five of the larger tree ferns at the
Coles’ Beach site in November 1979 revealed that an
average of 14 fronds had been produced in the current
spring season. The number of older, green and decadent
fronds average 17. The length of stem subtending the
older fronds averaged 11 cm. If allowance is made for a
few fronds being older than one year, the annual stem
growth could be about 10 cm, or the equivalent of two
spirals of frond insertions.

Some corroboration of these results was obtained by
measuring the growth of tree ferns beyond the char-line
of the 1968 fires in Ferntree Gully National Park. The
average increment of six trunks of C. australis , 2.5 to 4.9
m tall, was 11.3 cm per year. Thus for enriched post-fire
conditions in a wet sclerophyll environment at an
altitude of 200 m the annual growth rate was 10.5 cm.
On the basis of these estimates the minimum age of the
trunks of the Coles’ Beach ferns may be only 5 to 13
years.

The period from spore germination to trunk develop¬
ment is more difficult to estimate. Observations are
limited to the development of a transplanted rosette of
C. australis 30 cm in diameter, in a Surrey Hills garden
(Melbourne) over 20 years. A trunk, 33 cm high,
developed after seven years and an extrapolation of
subsequent height growth measurements (5.5 cm/yr. to
zero) indicated that the juvenile rosette period was likely
to be about four years. A similar time to develop to the
rosette stage seems reasonable in view r of glasshouse
observations of prothallial and fernling stages. The age
of the larger, mature tree ferns at Coles’ Beach could
therefore be between 12 and 20 years old. The largest
specimen at Shoreham may be 35 to 40 years old, whilst
that at Somers may be at least 50 to 55 years old, since
growth rate is likely to diminish with increasing age.

Tree ferns may have arrived on the coastal bluffs by
chance spore dispersal from the fern gullies of the Red
Hill district, 10 to 15 km up-wind to the west and north
west of this coastline. At the Shoreham bluffs, they are
known to have been present for the last 62 years
(1918-1980) (M. Wainwright pers. comm., 1980),
although on the lower, weedy stretches of nearby perma¬
nent creeks (Stony Ck. and East Ck.) they appear to be
absent.

At the French Island site, a cliff slump was estimated
to have occurred 14 years ago, judging from the maxi¬
mum number of whorls of branches on vigorously in¬
vading Pinus pinaster saplings. The oldest tree fern on
this slump is 50 cm high and is estimated to be 9 years
old. It is growing in a moist habitat, and although salt-
burnt, is protected locally from the prevailing SW winds
by a projecting clifflet. It would appear that Cyathea
australis is readily able to colonise such habitats when
they become available, although at the French Island
site, the persistence of the ferns is in jeopardy due to im¬

pending marine erosion of both the slump and the pro
jecting clifflet.

The impression gained from this initial study is that
the colonization of tree ferns on most bluff sites has been
a relatively recent phenomenon. The steles of tree ferns
are resistant to decay for many years. No remanants of
large ferns have been found. The mature tree fern is
resistant to all but the most intense fires (Jarrett & Petrie
1929). Since none of the tree ferns at Coles’ Beach show
evidence of charring on their lower trunks, they are cer¬
tainly younger than the last fire in 1941 (Lorraine Coles
pers. comm.).

CONCLUSIONS

An hypothesis which may account for the habitat
complex and the general youth of the tree fern popula¬
tion at the Coles’ Beach site concerns the rise of water
table following general land clearance on the gentle
hinterland slopes. If this has happened, then clays of the
mid-slopes of the vegetated bluff, which may have been
saturated by ground water only in very exceptional
decades or centuries, have now been much more fre¬
quently under its influence. Instability may have been
triggered by flood rains, such as occurred in 1952 and
1956, and have continued with variable activity until the
present day. A summary of the possible generalised se¬
quence of development of the unusual tree fern habit on
the west coast of Western Port is shown in Fig. 15.

ACKNOWLEDGEMENTS

It is a pleasure to acknowledge the help offered by the
following people: Miss Lorraine Coles and Mrs M,
Wainwright, who provided valuable information on
local history; Mrs Betty Duncan for use of Cyathea
distribution maps of Victoria; Drs E. C. F. Bird,
P. G. Ladd and Pauline Ladiges and Messrs R. Lakey
and P. McCumber for discussions and critical
assessments of the manuscript; Messrs D. G. Moore and
R. Eager for help with the field work.

REFERENCES

Bird, E. C. F., 1976. Coasts, 2nd Edition. A.N.U. Press,
Canberra.

Bird, E. C. F., 1977. Cliffs and bluffs and the Victorian coast.
Victorian Nat. 94: 4-9.

Bird, E. C. F., 1979. Victorian coastal geomorphology. Proc.
R. Soc. Viet. 92: 19-35.

Evans, R. S. & Joyce, E. B.,1974. Landslides in Victoria,
Australia. Victorian Nat. 91: 240-245.

Hills, E. S., 1976. Physiography of Victoria. Whitcombe
& Tombs, London.

Jarrett, E. S. & Petrie, A. H. K., 1929. The vegetation of
the Blacks’ Spur Region. A study in the ecology of some
Australian mountain Eucalyptus forests. II. Pyric succes¬
sion. J. Ecol. 17: 249-281.

Leeper, G. W., 1957. Introduction to soil science. 3rd Edition.

Melbourne University Press, Melbourne.

Selby, M. J., 1970. Slopes and slope processes. Waikato
Branch of the New Zealand Geographical Society (Inc.)
Publication No. 1. 59 pp.

COASTAL TREE FERN COMMUNITIES

13

Shapiro, M. A., 1975. Westernport Bay Environmental Carrillo-Rivera, J. J., 1975. Geology of the Westernport

Study 1973/74. Ministry for Conservation, Victoria. Sunkland. Proc. R. Soc. Viet., 87: 43-68.

Spencer-Jones, D., Marsden, M. A. H., Barton, G. M. &

COASTAL ARCHAEOLOGY IN VICTORIA
PART 2: ADAPTATION TECHNOLOGY & VOLCANISM

By P. J. F. Coutts

Victoria Archaeological Survey
Victoria Avenue, Albert Park 3206

INTRODUCTION

In Part 1 of this paper the morphology of coastal ar¬
chaeological sites in Victoria was described. Much as it is
desirable to place coastal sites in a rigorous cultural
framework, the present state of knowledge of coastal ar¬
chaeology in Victoria does not permit this. Instead we
can do little more than review some of the more in¬
teresting aspects of coastal archaeology. Topical issues
that could be discussed are: the relationship through
time between resources orientated and generalised
economic strategies; the relationship between the level
of stone technology, coastal economies and produc¬
tivity; the importance of fishing in local economies; the
relative importance of sandy beach and estuarine
habitats and their degree of exposure to ocean swell;
changes in coastline and marine habitats induced by
fluctuations in sea level; and changes in sea water
temperatures and their effect on abundance and variety
of marine fauna.

In this paper problems relating to the technology of
flint, interpreting various archaeological sites, and man’s
need to adapt to a changing coastal environment are
discussed. Adaptations of particular interest are those
arising from natural processes (e.g. rises in sea level and
volcanism) or from the actions of man himself (e.g.
fire).

CHANGES IN COASTAL VEGETATION
In evaluating the economic potential of coastal areas
at any prehistoric time the resources of the sea and the
immediate hinterland must be assessed in concert.
Vegetation is a key factor as it largely determines the
range of animal and vegetable foods available. Changes
in vegetation patterns almost certainly precipitated
adaptation. It is likely that coastal vegetation has chan¬
ged during the past 30 000 years, as sea-level rose and
fell. Analyses of pollen from south-eastern South
Australia, from Wilsons Promontory and from Lake
Keilambete suggest that fluctuations in climate during
the past 10 000 years have not been severe enough to
modify the main floral associations to any great extent
(Dodson 1974a, 1974b, Dodson & Wilson 1975, Hope
1974, Yczdani 1970).

However, in Gippsland, using pollen data that span
the last 7000 years Hooley, Southern and Kershaw
(1980) demonstrated a dramatic shift from Casuarina to
Eucalyptus woodland about 3000 years before the pre¬
sent (BP) and a further decline in Casuarina about 200
BP. Both periods of change are associated with peaks in
charcoal debris in soil profiles, which suggests that the
changes were fire initiated, and probably by man. The

B

major vegetation change 3000 BP must have affected
coastal Aboriginal economy in that area.

Aboriginals periodically set fire to coastal vegetation
along other parts of the Victorian coastline during the
Late Prehistoric period, but the antiquity of this practice
is not known (e.g. Port Phillip —observations by Knop-
wood (Nicholls 1977: 35, 37, 39)). Thus patterns of
coastal vegetation have varied from time to time and
resulting changes in Aboriginal economy can be ex¬
pected in archaeological records.

COASTLINE CHANGES

Changes in the coastline induced changes in coastal
ecology, which in turn affected Aboriginal settlement
patterns (Lampert & Hughes 1974), and possibly social
and economic patterns, and material culture. During the
past 20 000 years the Victorian coastline has receded
(Fig. 1) as the sea-level rose after the last glacial period.
The coastline stabilised about 6000 BP, which is the
oldest dating for most archaeological sites along the pre¬
sent coastline. However, in several coastal areas, Cape
Nelson, Cape Woolamai, the southern tip of Wilsons
Promontory, Point Hicks and Gabo Island, where the
inshore submarine profiles are steep, sites having ar¬
chaeological records older than 6000 BP may be found.
Indeed if the interstadial high of circa 35 000 BP
reached minus 10 m then these areas may still have the
only examples of littoral exploitation dating from those
periods.

Port Phillip probably began to fill about 9000 BP and
its comparatively flat floor would have precluded exten¬
sive occupation anywhere as sea-level rose. Thus most
middens around Port Phillip are likely to post-date 6000
BP. Western Port began to fill around 8000 BP when the
western margins were inundated. The islands in Western
Port would not have been formed until the sea-level was
very close to its present height about 6000 BP. Evidence
of slightly older littoral exploitation may also be found
on the western fringes of the two largest islands in
Western Port and on the mainland between Flinders and
Hastings. The only other area that may have ar¬
chaeological littoral sites pre-dating 6000 BP is the
coastal strip along the eastern edge of Cape Otway,
where the underwater terrain is steeper than in most
other coastal areas. Even there the sites are unlikely to
be much more than 500-1000 years older than those in
other coastal areas in Victoria.

If the sea-level circa 6000 BP (see Gill & Hopley 1972,
Thom, Hails & Martin 1969, Thom, Hails, Martin &
Phipps 1972) was 2-3 m higher than at present, the man-

15

16

P. J. F. COUTTS

made littoral deposits in the relatively flat zones of
Western Port, Corner Inlet and the Gippsland Lakes
would have been inundated and covered by sediments.
Aboriginal sites formed then would be further inland
than the present shoreline.

As the sea rose to its present level, areas rich in food
resources were created (Bowdler 1977: 213). Such areas
include Western Port, Port Phillip, Mallacoota Inlet
and the Gippsland Lakes. However, it is not clear yet
whether the prehistoric inhabitants living in these areas
had the required technologies to exploit the resources
effectively (Lampert & Hughes 1974).

One of the most important events affecting coastal
ecology and therefore human settlement was the inunda¬
tion of the land bridge joining Tasmania and mainland
Australia. Unfortunately the impact of these changes on
Aboriginal communities, occupying the Victorian
coastline has not yet been documented archaeologically.

Much of the present Victorian coastline is eroding to¬
day and there is increasing evidence that many sites less
than 500 years old that are or were situated in the vicin¬
ity of the foredunes or on low cliff tops, are being eroded
away or have disappeared already. Other periods of ac¬
tive erosion may have occurred over the past few thou¬
sand years, initiated by minor fluctuations in sea-level.
So far there is only one published instance of a changing
coastal geomorphology affecting prehistoric human set¬
tlement in Victoria (Coutts 1967, Hope & Coutts 1971).
On the west coast of Yanakie Isthmus, Wilsons Pro¬
montory, there are two dune systems associated with
two series of sites ‘A' and 4 B’ (Coutts 1970, 1981).
Because ‘A’ series middens are exposed in the eroded
cliffs along the beach they must have once extended fur¬
ther out onto the beach and consequently, when these
sites were occupied the coastline must have been further
seaward than it is today.

‘A’ series sites are found on the summits of
Pleistocene dunes which run almost at right angles to the
present coastline. Peat deposits which outcrop on the
beach between these dunes can be linked stratigraphic-
ally with soils containing the ‘A’ series middens.
Radiocarbon dating of the latter indicate that they were
formed about 3000-6500 years ago and a single date
(6010 110 BP) from the peat deposit confirms the

stratigraphic relationships (Coutts 1967). The peat con¬
tains freshwater gastropods and was probably landlock¬
ed at one time, presumably by a foredune further to the
west. Results of analyses of the midden materials in¬
dicate that nearly all the shellfish in the ‘A’ series sites
were collected from rock platforms; others were from
estuaries and sandy beaches. The present coastline fron¬
ting the dunes is sandy beach, almost totally devoid of
intertidal rock platforms.

Some evidence from the later 4 B’ series dunes (dating
from 3000 BP) suggests that Aboriginal economy may
have diversified towards the end of the prehistoric
period and that the preferences also changed for the
stone materials used. Whilst the change in the methods
of collecting shellfish may be seen as an adaptive
response to a changing coastal environment, the changes

in material culture seem to have been widespread
throughout Victoria (Coutts & Witter 1977, Hope &
Coutts 1971) and cannot be explained in this way.

FLINT AND COASTAL RESOURCES USAGE
Large quantities of flint are found in many Victorian
coastal archaeological sites. It derives from Gambier
limestone (Gill 1957) and has been washed in from
offshore, often buoyed by kelp (Fig. 2, Boutakoff 1963,
Hossfeld 1966, Mitchell 1949). Gill (1957) suggested that
Aboriginals obtained flint from caves, sink holes and re-
emerged Pleistocene strand lines along the coast near the
South Australian/Victorian border. Although similar
flint can be obtained from inland areas of south and
north-eastern South Australia (Wright 1971) and from
isolated localities in western Victoria (Hossfeld 1966, P.
Kenley pers. comm.) it is considered unlikely that the
flint nodules found along the Victorian coast came from
any of these sources.

The following observations are pertinent to the
distribution of flint on the Victorian coast and its use by
Aboriginals:

(1) Flint occurs in coastal archaeological sites and
on beaches between Wilsons Promontory and
the South Australian border, increasing in fre¬
quency from east to west.

(2) Flint implements are rarely found in ar¬
chaeological sites east of Wilsons Promontory.

(3) Flint was used by Victorian Aboriginals at
Wilsons Promontory around 5000 BP (Coutts
1970) and at Thunder Point near Warrnambool,
around 4300 BP (Coutts 1978). In western Vic¬
toria, so called Gambieran implements made
from flint have been found in unstratified con¬
text (Mitchell 1949). The flint is thought to
derive from local outcrops rather than from
coastal beaches (Clark 1979). In South Australia
none of the implements from Devon Downs or
Fromm’s Landing (sites less than 5000 years old
and little more than 200 km from the coast) were
made from flint (Hale & Tindale 1930, Mulvaney
1960, Mulvaney, Lawson & Twidale 1964), but a
few fragments of dark blue-grey flint have been
excavated at Roonka, a site which dates from
about 7000 BP (Pretty 1977). In the central
Western District of Victoria, flint has been
found in sites which date to less than 3000 BP.
No flint artefacts have been found at Keilor,
Green Gully or Cloggs Cave (Flood 1974,
Mulvaney 1970, Wright 1970), sites which have a
much greater antiquity than those mentioned
above, and which are no more than 30 km from
the present coastline.

However, it should be remembered that
15 000 years ago, the coastline was some hun¬
dreds of kilometres further south, and the pros¬
pects for collecting flint from beach sources and
conveying it to these sites are unknown.

(4) Flint and quartz are the two most common types
of stone materials found in recent coastal ar-

MaMacoota

17

<u

t/5

Fig. 1 Approximate shoreline positions in Bass Strait during the post-glacial sea-level ri:

18

P. J. F. COUTTS

chaeological sites and in general, retouch is rare,
there is little deliberate shaping of the flakes, and
blade manufacture is minimal (Coutts 1970). At
earlier sites a wide variety of implements were
made from flint including most types of backed
blade, bifacial choppers, and scrapers (Camp¬
bell & Noone 1941-43, Mitchell 1949).

(5) Inland, flint artefacts are found at Penola in
South Australia; Willaura, Glen Thompson,
Murtoa, Condah Swamp, Dooen Swamp, In-
verleigh, Lake Bolac and at sites in the Gram¬
pians in western Victoria (Campbell & Walsh
1952, Coutts & Lorblanchet 1981, Mitchell
1949).

(6) On the shingle beaches at Port Macdonald in
South Australia are large quantities of black
rather brittle flint. Grey flints are often the most
common components (being used extensively for
backed blades) of flint assemblages on the Vic¬
torian coastline. Mulvaney (1962), in discussing
his basically unretouched flint and bone
assemblages from Glen Aire has noted that
finely produced stone implements appear to be
absent from late Victorian assemblages and he
suggests that ‘the basic industrial materials of re¬
cent Victorian prehistory were of organic
origin’, as opposed to stone.

Finely produced implements, such as backed blades,
were no longer being made in any quantity at the end of
the prehistoric period (Coutts 1981). However this does
not necessarily mean that Aboriginals at any other time
in prehistory were any less dependent on organic
materials. As Mulvaney (1962) observed, ‘these
materials could not be expected to survive for ar¬
chaeological discovery’; and it is this fact that makes it
difficult to argue either way. The presence of certain
classes of stone tools in more ancient Victorian sites,
suggests that organic materials were being worked with
them. There has also been a long tradition of working
with bone in Victoria. At least one bone implement
similar to those which were made several thousand years
later at Glen Aire, has been found at Cloggs Cave in
eastern Victoria (Flood 1974). It is not certain that a lack
of retouch on tools—such as at Glen Aire—or the
absence of finely worked implements, means that there
was a reduction in the degree of stone tool use. The
crux of the problem is an inability to recognise when a
flake (retouched or otherwise) has been used.

No attempt will be made to resolve this dilemma here.
However, a working hypothesis is proposed to explain
why finely worked tools were abandoned in some parts
of Victoria. It is based on premises of increasing popula¬
tion pressure and changing availability of raw materials,
in particular, flint. Rising sea-level following the last
glaciation, caused erosion of the Tertiary limestone out¬
crops, now below water, carrying quantities of flint
nodules onto the shorelines of south-eastern South
Australia and western Victoria where it was collected
and used by Aboriginals.

It is not known how r long it might take for the flint to

find its way onto the beaches of Victoria in sufficient
quantities to be regularly exploited. By 14 000 BP (Jen¬
nings 1971) the Bassian depression had been inundated,
though Tasmania and mainland Australia were still con¬
nected by a land bridge through Wilsons Promontory
and Flinders Island and the shoreline was within several
kilometres of its present position.

Between 14 000 and 6000 BP the coastline stabilised.
By at least 5000 BP flint had been washed up onto the
beaches of Wilsons Promontory and was used by
Aboriginals. If the frequency with which it was used at
the Promontory is any indication of its local availability,
it was not common 5000 years ago. However, 4000 years
later it was a dominant material at coastal arch¬
aeological sites. During the late Pleistocene period it
might be assumed that population levels in Victoria, as
elsewhere in Australia, were increasing (Birdsell 1977)
and that tribal and linguistic boundaries w r ere continual¬
ly adapting to changing environments and in response to
changing population pressures. This period might be
regarded as one of experimentation when Aboriginals
were still discovering suitable outcrops for stone tool
manufacture. It appears that they concentrated on ex¬
ploiting sources of quartz, silcretes and quartzites, and
to a much lesser degree, various other types of chert.
After the close of the Pleistocene period, the Aboriginal
population tended to increase or possibly stabilise, and
pressure on stone resources would have increased.
Moreover it is likely that the most visible and desirable
sources of stone gradually became the property of small
groups of Aboriginals.

At the end of the prehistoric period for example
(about 1840 in Victoria) seven out of 35 Victorian tribes
(Tindale 1974) had control of major ‘flint catchment
areas’ along the coast. Five out of the seven tribes shared
the Kulinic group of languages and distribution of flint
implements tends to fall within the boundaries of the
Drual sub-group of languages (Fig. 2, Oates & Oates
1970). Trade and exchange networks existed in the cen¬
tral Western District at the end of the prehistoric period
and these could have sponsored the trading of flint be¬
tween coastal and inland areas (Lourandos 1977,
McBryde 1977, Mulvaney 1976).

’ Following Hayden’s thesis (1977), there could have
been a period of rationalisation as the value of some
types of stone material increased. This period peaked
after the introduction of a blade technology and the pro¬
duction of finely made implements such as backed
blades. Since blade production can be inherently
wasteful of materials, economy was achieved by utilising
waste products to make utilitarian implements (Watson
1968). In South Australia and south-west Victoria the
blade technology was applied to flint which became
available about the time of the peak of the period.
However sources of chert, jasper and other fine grained
materials became less accessible to many Aboriginals
and available sources of these materials probably dimin¬
ished. As a result a much more utilitarian approach to
the use of stone was adopted inland, leading to the
demise of finely made tools, essentially because their

19

20

P. J. F. COUTTS

manufacture was uneconomic and wasteful of stone
resources. This trend spread to the coast where abun¬
dant flint materials were available and a ‘throw away*
technology developed.

These changes would not have happened everywhere
simultaneously, but would have depended upon such
factors as the availability of raw materials and the
strength of local traditions pertaining to the manufac¬
ture of stone tools.

To test this theory and its variations more data are re¬
quired on flint implements from coastal and inland sites,
on the sources of flint, and on ways of assessing ‘stone
economy’.

VOLCAN1SM AND COASTAL ADAPTATION

Tower Hill, situated near the coast at Koroit in
western Victoria, erupted during the Holocene period
and had a pronounced effect on local physiography. The
explosive eruption covered the landscape with ash
which varied in thickness (Gill 1947, 1967). Moreover,
the concommitant intense heat and bushfires would have
killed off and burnt back the vegetation in the surroun¬
ding areas, and would have had dramatic effects on local
wildlife. Consequently the eruption of this volcano pro¬
bably precipitated a disjunction in social and economic
patterns of Aboriginal groups. Ideally a range of ar¬
chaeological materials need to be located above, below
and within the ejectamenta so that cultural adaptations
can be monitored.

Date of the Tower Hill eruption

The Tower Hill eruption and deposition of tuff is
presumed to have been a rapid event on the basis of the
type of volcanism involved (Gill 1967) and the whole
event may have taken place in something less than 50
years. Such a time span is not normally measureable in
the archaeological record, though the effects of the erup¬
tion are manifested in all subsequent local records in the
guise of tuff enriched sediments. If the date of the erup¬
tion can be established, the Tower Hill tuffs can be used
as a chronological marker for the Warnambool district.

Gill (1953, 1955, 1967, 1972) has several dates from
archaeological and geological deposits above, below and
within the ejectamenta and the author has several from
archaeological deposits above them (Coutts 1981).
These dates need to be scrutinised carefully.

The date from Bushfield (Gill 1967), said to come
from an archaeological horizon below the tuff (Coutts
1981: Table 7), can be excluded because it was obtained
‘from the C0 2 fraction, and the provenance of the
mineralisation is uncertain’ (Gill 1967: 358). However,
Gill (1978) considers that the dates of 6500 ±200 BP
(Gill 1967) obtained from marine shells from the Merri
canal, and 6570 ± 115 BP (E. D. Gill pers. comm.) and
5850 ± 320 BP (Gill 1967) from marine shells of the Per-
tobe Coquina, which overlies the tuff, are reliable. A
date of 8700 BP (E. D. Gill pers. comm.) has been ob¬
tained from a laminated mammillary calcrete at Den-
nington which is presumed to be the same age as the

calcrete underlying the tuff at Thunder and Pickering
Points.

Two suites of dates are available for midden deposits
which clearly postdate the eruption: one from middens
found within soil horizons associated with mobile dune
systems at Armstrong Bay and the other from middens
overlying the tuffs at Thunder Point.

The earliest C' 4 date for Armstrong Bay is 5680 BP
(Campbell 1967) and is similar to Gill’s earliest date of
5120 BP (Gill 1967). However our investigations (Coutts
1977, 1978) and those of Campbell (1967) failed to
locate tuffs (which outcrop on the beach) underneath the
dunes and the middens at the eastern end of Armstrong
Bay. Gill (1953, 1955, 1967, 1971) on the other hand has
found the tuff under the dunes on the western side of the
Bay, near Gormans Lane, but this is well away from
where the archaeological work has been conducted.
Thus the relationships between the dunes, the middens
and the tuff remain unclear in the eastern area, although
Gill is certainly correct in concluding that the tuff is
older than the dunes. A radiocarbon date from ar¬
chaeological deposits a little above stratified tuff
deposits at Thunder Point is 4130 ±200 BP (Coutts
1977), and it can safely be assumed that the Tower Hill
volcano erupted some time before this.

Some disagreement between Gill and the author per¬
sists over the interpretation of a date obtained from
shells from Pickering Point. In 1972, Gill claimed to
have found shells ‘within the Tower Hill tuff which dated
to 7300 ± 150 BP’ (Gill 1972). The photographic record
(Gill 1972), Gill’s brief description of the site from which
the C 14 sample was taken, the results of the later textural
analysis of the sediments plus archaeological excava¬
tions at Pickering Point to determine the stratigraphic
context of the archaeological debris, suggest that what
was dated were shells from dispersed archaeological
debris recemented into tuff rich sediments sometime
after the eruption. Hence the significance of the
7300 ± 150 BP date in the context of dating the Tower
Hill eruption is unclear.

More recently Mortlock (1977) has attempted to date
the Tower Hill ejectamenta directly by thermolu¬
minescence dating; but as this results in the scoria
predating the tuff, which is inconsistent with the
geological evidence, these dates must be regarded as
unreliable.

On evidence available at present, the Tower Hill
volcano erupted between about 6600 and 8700 BP.

EVIDENCE FOR CULTURAL CHANGES

There are no reliable archaeological data from the
period before the eruption of Tower Hill (Coutts 1976).
For the post-eruption period, there are copious data
from surface collections and from excavations at
Thunder Point and Armstrong Bay.

Unfortunately stratigraphic and cultural evidence
from Armstrong Bay is confused. Investigations by Gill
have suggested that there are at least two soil horizons
associated with the post-eruption coastal dune system
and which are connected with Aboriginal occupation.

COASTAL ARCHAEOLOGY IN VICTORIA

21

Gill (1955) dated occupational debris, which included
large numbers of bone points from the latest of the two
soils as 1750±20 BP. More recently he reported that the
bone industry was associated with the older soil horizon
(Gill 1967). In contrast Kenyon (1912) and Mahony
(1912) claimed that the bone points at Armstrong Bay
came from the surface (confirmed later by Mitchell
1958), and Mahony went so far as to say that no
Aboriginal material could be found in the oldest
horizons of the deeper blowouts. Campbell (1967) con¬
cluded that there has been one period of soil develop¬
ment at Armstrong Bay. However his evidence is con¬
fusing because he obtained very different radiocarbon
dates, 5680 BP and 1280 BP for two different exposures
of this soil.

More recent archaeological work in this area (author),
has revealed the presence of one soil horizon at the
western end of the beach. A single radiocarbon date of
2450 BP for the upper part of the soil places it in the
middle of the present sequence of radiocarbon dates for
this area (Coutts 1981: Table 1). At the eastern end of
the beach there is evidence for at least two closely spaced
soils containing archaeological materials, within the
beach face of the leading dune system, and dating from
570 ± 80 BP to 610 ± 110 BP and 2685 ± 110 to 2925 ± 95
BP respectively.

Judged on the dates alone, Aboriginals were ex¬
ploiting the coastal resources at Armstrong Bay some
2000 years after the eruption. They camped intermittent¬
ly in the vegetated areas of the dunes for the next 5000
years, and it is probable that throughout the history of
the dune system at Armstrong Bay there may have been
continuous soil development, punctuated by periods of
dune stability and destabilisation during which parts of
the vegetated landscape and the archaeological deposits
were buried under loose sand.

A 3000 year old midden deposit in the leading edge of
the present foredune suggests the coastline at that time
was further seaward. Today these middens are approx¬
imately 3 m above high water mark and are covered by
more than 6 m of sand.

A key feature of the Armstrong Bay site is the variety
of faunal remains found in them. Gill (1967) and Camp¬
bell (1967) have suggested that the middens in the older
soil horizons contain shellfish from both rock platforms
and sandy beaches. Archaeological work in this area
suggests that for a period of 2500 years shellfish col¬
lecting was confined to rock platforms with almost no
evidence of exploitation of sandy beach animals. Today,
the Armstrong Bay coastline is a sandy beach. Thus, if
the ecology of predominant shellfish taxa found in mid¬
dens reflects the type of coastline in the vicinity, it seems
likely that the Armstrong Bay coastline has undergone a
number of changes in physiography since the volcano
erupted.

The same cannot be said of the middens at Thunder
Point, which are on the tail of the volcanic ash shower
and in a different physiographic situation. From 4300
BP until 800 BP there is no evidence of any significant
shift in shellfish collecting strategies, although there is

evidence that the coastline itself in the vicinity of Warr-
nambool has undergone significant changes over this
period. Clearly one or more of the occupation phases at
Armstrong Bay was associated with the manufacture of
bone points, spatulates etc., (Coutts et al. 1976), and
from at least 3000 BP Aboriginals utilised coastal flint,
making a variety of tools including backed blades. They
hunted a w'ide range of terrestrial animals but evidence
of fishing is minimal. At Thunder Point, there is barely
any material culture evidence as the site was associated
almost exclusively with shellfish gathering.

CONCLUDING REMARKS

In these two papers discussion has been limited to
aspects of coastal archaeology in Victoria. It is apparent
that there are many varieties of coastal middens and that
their interpretation has and will prove to be difficult. In¬
deed coastal archaeology in Victoria is still in its infancy.
The need for intensive, long term regional and local
studies cannot be over-stressed. Some of the most in¬
teresting areas will be adjacent to Port Phillip, Western
Port, Tower Hill, the Gippsland Lakes, and Mallacoota
Inlet.

ACKNOWLEDGEMENTS

My sincere thanks to Edmund Gill for his advice and
for reading and commenting on this manuscript during
its preparation. I would also like to thank Mr P. Kenny
(Mines Department) and Ms A. McConnell (VAS) for
comments on the geological aspects of this manuscript.

REFERENCES

The references listed below include only those which
have not already been cited in Coutts 1981.

Birdsell, J. B., 1977. The recalibration of a paradigm for the
first peopling of greater Australia. In Sunda and Sahul:
Prehistoric studies in south-east Asia, Melanesia and
Australia, J. Allen, J. Golson & R. Jones, eds, Academic
Press, London, 113-167.

Boutakoff, N., 1963. The geology and geomorphology of the
Portland area, Mem. geol. surv. Vic. 22.

Bowdlf.r, S., 1977. The coastal colonisation of Australia. In
Sunda and Sahul: Prehistoric studies in south-east Asia ,
Melanesia and Australia , J. Allen, J. Golson & R. Jones,
eds, Academic Press, London, 205-246.

Campbell, T. D. & Noone, H. V. V., 1941-43. South Australia
microlithic stone implements. Rec. S. Aust. Mus. 7:
281-307.

Campbell, T. D. & Walsh, G. C., 1952. Aboriginal imple¬
ments from campsites in the south of South Australia and
Victoria. Mankind 4: 339-342.

Clark, D. J., 1979. The Gambieran stone tool industry. B.A.

Hons Thesis, Latrobe University (unpubl.).

Coutts, P. J. F.. 1967. Coastal dunes and field archaeology in
south-east Australia. APA0 2: 28-34.

Coutts, P. J. F., 1976. The prehistory of Victoria: a review.
Rec. Viet. Archaeol. Surv. 2.

Coutts, P. J. F., 1981. Coastal Archaeology in Victoria, Part
1: The morphology of coastal archaeological sites. Proc. R.
Soc. Viet. 92: 67-80.

Coutts, P. J. F., & Lorblanchet, M., 1981. Archaeology
and rock art in the Grampians, Victoria, Australia. V.A.S.
Rept. (unpubl.)

22

P. J. F. COUTTS

Dodson, J. R., 1974a. Vegetation and climatic history near
Lake Keilambete, western Victoria. Aust. J. Bot. 22:
709-717.

Dodson, J. R., 1974b. Vegetation history and water fluctua¬
tions at Lake Leake, south-eastern South Australia 1,
10 000 BP to present. A us/. J. Bot. 22: 719-741.

Dodson, J. R. & Wilson, 1. B., 1975. Past and present vegeta¬
tion of Marshes Swamp in south-eastern South Australia.
Aust. J. Bot. 23: 123-150.

Gill, E. D., 1947. An hypothesis relative to the age of some
western district volcanoes. Proc. R. Soc. Viet. 60: 189-194.

Gill, E. D., 1953. Geological evidence in western Victoria
relative to the antiquity of the Australian Aborigines. Mem.
nat. mus. Viet. 18: 25-92.

Gill, E. D., 1957. The Australian Aborigines and fossils.
Victorian Nat. 74: 93-97.

Gill, E. D., 1971. Applications of radiocarbon dating in Vic¬
toria, Australia. Proc. R. Soc. Viet. 84: 71-88.

Gill, E. D., 1975. Calcrete hardpans and rhizomorphs in
western Victoria, Australia. Pacific Geol. 9: 1-16.

Gill, E. D., 1978. Radiocarbon dating of the volcanoes of
western Victoria, Australia. Victorian Nat. 95: 152-158.

Gill, E. D. & Hopley, D., 1972. Holocene sea-levels in eastern
Australia. Mar. Geol. 12: 223-242.

Hale, H. M. & Tindale , N. B., 1930. Notes on some human
remains in the lower Murray Valley, South Australia. Rec.
S. Aust. Mus. 4: 145-218.

Hayden, B., 1977. Sticks and stones and ground-edge axes: the
upper Palaeolithic in south-east Asia. In Sunda and Sahul:
Prehistoric studies in south-east Asia , Melanesia and
Australia. J. Allen, J. Golson & R. Jones, eds. Academic
Press, London, 73-112.

Hooley, A. D., Southern, W. & Kershaw, A. P., 1980.
Holocene vegetation and environments of Sperm Whale
Head, Victoria, Australia, J. Biogeog. 1: 349-362.

Hope, G. S. & Coutts, P. J. F., 1971. Past and present
Aboriginal food resources at Wilsons Promontory, Vic¬
toria. Mankind?,: 104-114.

Hope, G. S. & Coutts, P. J. F., 1971. Past and present
Aboriginal food resources at Wilsons Promontory, Vic¬
toria. Mankinds: 104-114.

Hossfeld, P. S., 1966. Materials used in Australian Aboriginal
stone implements. In Aboriginal man in south and central
Australia , B. C. Cotton, ed.. Government Printer,
Adelaide, 163-168.

Jennings, J. N., 1971. Sea-level changes and land links. In
Aboriginal man and environment in Australia , D. J.
Mulvaney & Golson, eds, A.N.U. Press, Canberra, 1-13.

Kenyon, A. S., 1912. Camping places of the Aborigines of
south-east Australia. Viet. Hist. Mag. 2: 97-110.

Lourandos, H., 1977. Aboriginal spatial organisation and
population: south-western Victoria reconsidered. APAO

12: 202-225.

McBryde, I., 1977. Wil-im-ee Moor-ring: or, where do axes
come from? Mankind 11: 354-382.

Mitchell, S. R., 1949. Stone-age craftsmen. Tait Book Co.,
Melbourne.

Mortlock, A. J., 1977. Progress in thermoluminescence
dating. Mem. Viet. Archaeol. Surv. 1: 10-17.

Mulvaney, D. J., 1960. Archaeological excavations at
Fromms Landing on the lower Murray River, S. A. Proc. R.
Soc. Viet. 72: 53-85.

Mulvaney, D. J., 1970. Green Gully revisited: the later ex¬
cavations. Mem. natn. Mus. Viet. 30: 15-78.

Mulvaney, D. J., 1975. The Prehistory of Australia. Penguin
Books, Melbourne.

Mulvaney, D. J., 1976. The chain of connection:'the material
evidence. In Tribes and boundaries in Australia, N. Peter¬
son cd., AIAS, Canberra, 72-94.

Mulvaney, D. J., Lawson, G. H. & Twidale, C. R., 1964.
Archaeological excavations of rock shelter 6, Fromms
Landing, S.A. Proc. R. Soc. Viet. 77: 479-516

Nicholls, M., ed., 1977. The diary of the Reverend
Robert Knopwood 1803-1838. Tasmanian Historical
Research Association, Launceston.

Oates, W. J. & Oates, L. F., 1970. A revised linguistic survey
of Australia. Australian Aboriginal Studies No. 33, AIAS,
Canberra.

Pretty, G. L., 1977. The cultural chronology of the Roonka
Flat: a preliminary consideration. In Stone tools as
cultural markers , R. V. S. Wright, ed.. Prehistory and
material culture series No. 12, AIAS, Canberra, 288-331.

Thom, B. Ci., Hails, J. R. & Martin, A. R. H., 1969.
Radiocarbon evidence against higher post-glacial sea-levels
in eastern Australia. Mar. Geol. 7: 161-168.

Thom, B. G., Hails, J. R., Martin, A. R. H. & Phipps, C. V.
G., 1972. Post-glacial sea-levels in eastern Australia —a rep¬
ly. Mar. Geol. 12: 223-242.

Tindale, N. B., 1957. Culture succession in south-eastern
Australia from Late Pleistocene to the present. Rec. S.
Aust. Mus. 13: 1-49.

Tindale, N. B., 1974. Aboriginal tribes of Australia. ANU
Press, Canberra.

Watson, W., 1968. Flint implements. British Museum,
London.

Wright, R. V. S., 1970. Flaked stone material from GGW-1.
Mem. natn. Mus. Viet. 30: 79-92.

Wright, R. V. S., cd., 1971. Archaeology of the Gallus Site,
Koonalda Cave. Australian Aboriginal Studies No. 26,
AIAS, Canberra.

Yezdani, G. H., 1970. A study of the Quaternary vegetation
history in the volcanic lakes region of western Victoria.
Ph.D. Thesis, Monash University.

PAST AND PRESENT DISTRIBUTIONS AND TRANSLOCA¬
TIONS OF MACQUARIE PERCH Macquaria australasica (PISCES:
PERCICHTHYIDAE),WITH PARTICULAR REFERENCE TO

VICTORIA

By P. L. Cadwallader

Fisheries and Wildlife Division, Ministry for Conservation, Snobs Creek Freshwater Fisheries
Research Station and Hatchery, Private Bag 20, Alexandra, Victoria 3714

Abstract: Details of past and present distributions and translocations of Macquarie perch in Victoria
are presented. Prior to 1970 the species was recorded at 52 localities within its natural geographical range,
in the Murray-Darling River system. Since 1970 it has been recorded at only 20 of these localities. Of the
waters which arc outside its natural range and into which it has been released, Macquarie perch has been
taken in only three since 1970 and in only one with any regularity. In most waters, both within and out¬
side its natural range, only relict populations remain. Siltation has probably been the main cause of the
decline in range and abundance of the species in Victoria, but several other factors, including the con¬
struction of dams and weirs, interaction with introduced fish, overfishing, “river improvement” and local
pollution, have probably all contributed to its demise. Data on the distribution of Macquarie perch out¬
side Victoria are summarised.

INTRODUCTION

Past and present distribution patterns of Australian
native freshwater fish are poorly documented. In addi¬
tion, transfers of fish from one catchment to another
often went unrecorded, thereby causing confusion for
taxonomists who now try to elucidate the relationships
bet ween the various stocks of a species. However, there
are more records for distributions of commercial or
sporting fish than for many of the smaller species. In this
paper, 1 have summarised data on past and present (post
1970) distributions and translocations of Macquarie
perch, Macquaria australasica Cuvier and Valenciennes
(Pisces: Percichthyidae), a native freshwater sport fish
whose survival is now seriously threatened (Llewellyn &
MacDonald 1980). Although I include some data on the
distribution of the species outside Victoria, most of the
information presented relates to the status of Macquarie
perch in Victoria. Such information is a necessary prere¬
quisite in any attempt to determine the factors responsi¬
ble for the decline or elimination of Macquarie perch
from specific areas and also as a basis for planning the
release of hatchery-produced fish in the future.

METHODS

Information on the distribution of Macquarie perch
was obtained from the following published works: Stead
(1913), McKeown (1934), Butcher (1946), Whitley
(1964), Roughley (1966), Lake (1967, 1971, 1978),
Wharton (1968, 1973), Anon. (1973, 1974), Scott et al.
(1974), Tunbridge & Rogan (1976), Cadwallader (1977,
1979), Cadwallader & Rogan (1977), Hungerford (1977),
Wed lick (1977), Bishop & Bell (1978), Bishop & Tilzey
(1978), Tunbridge (1978), Cadwallader & Eden (1979),
Pratt (1979), and Llewellyn & MacDonald (1980), and
from the records of the Victorian Piscatorial Council
(V.P.C.), the Ballarat Fish Acclimatisation Society and
the Fisheries and Wildlife Division of the Ministry for
Conservation. The last source included the unpublished

results of several fish surveys and an old manuscript en¬
titled “Particulars of stocking streams with Murray
perch, English perch and English tench, etc.” covering
the period 1909-1955. Information was also received in
response to a circular distributed to angling clubs and to
Fisheries and Wildlife officers throughout Victoria, and
from general contact and correspondence with anglers in
particular areas.

In addition, many Victorian waters known to have
contained Macquarie perch in the past were surveyed by
the author between 1975 and 1980, and by other
Fisheries and Wildlife Division staff during routine
survey work, particularly in relation to management of
the trout fishery.

RESULTS

Distribution of Macquarie Perch in Victoria

The past and present distributions of Macquarie perch
in Victoria are shown in Fig. 1. The natural geographical
range of the species is confined to the Murray-Darling
River system, north of the Great Divide, and this range
is now much less than in the past (cf. open circles and
solid circles in Fig. 1). There have been many recorded
translocations of Macquarie perch both within and out¬
side this natural range (Table 1) and the present status
(post 1970) of Macquarie perch in the waters into which
they have been introduced is indicated in Fig. 1 (cf. open
squares and solid squares). In addition, there have been
several transfers of Macquarie perch from natural
waters to farm dams within the same catchment; details
of these transfers are not included in this paper.

Several of the earliest records of translocations refer
to “Murray perch”, which probably included Macquarie
perch as well as golden perch Macquaria ambigua
(Richardson) and perhaps also silver perch Bidyanus bi-
dyanus (Mitchell), although the last species was usually
distinguished by the term “grunter”. The terms “Murray
bream” and “Goulburn bream” also occur in the early

23

24

P. L. CADWALLADER

100 km

DISTRIBUTION OF MACQUARIE PERCH

25

records; “bream” is the common name for Macquarie
perch in some areas, but in other areas may also have
applied to silver perch. Some of the early records of
translocations of “Murray perch” are qualified by the
term “of the Macquarie variety” (e.g. Table 1: the 1910
releases into the Yarra River) and several early records
(from 1913 onwards) for the Yarra River refer to “Mac¬
quarie perch” being caught by anglers after earlier
releases of “Murray perch”. In the records of a series of
translocations of “Macquarie perch” taken from the
Goulburn Weir in 1922 (see Table 1), the consignments
were reported to include small proportions of golden
perch (about 5%) and silver perch (about 2.5%). Thus,
although the species composition of the earliest
translocations of “Murray perch” is in doubt, Macquarie
perch probably formed the bulk of the fish, particularly
of those fish taken in the Goulburn River catchment
(from the Goulburn River itself, the Goulburn Weir or
Broken River) from where many later batches of “Mac¬
quarie perch” were taken.

Fig. l^Past and present distributions of Macquarie perch in
Victoria. The grey shaded area indicates the presumed past
distribution of Macquarie perch; •, natural population still
(post 1970) present; O, natural population no longer present;
■ , population derived from introduced fish, still present; □,
introduced but no longer present. Key to locality numbers: 1,
River Murray (Tom Groggin); 2, Berringamah Creek; 3, Dart¬
mouth Dam and inflowing waters; 4, Mitta Mitta River; 5,
Lake Hume; 6, Kiewa River; 7, Lake Sambell; 8, Buckland
River; 9, Lake Buffalo; 10, Buffalo River; 11, King River; 12,
Meadow Creek; 13, Ovens River; 14, River Murray (Bur-
ramine); 15, River Murray (Tocumwal); 16, River Murray
(Barmah); 17, Boosey Creek; 18, Broken Creek; 19, Lake
Nillahcootie; 20, Broken River; 21, Stony Creek; 22,
Honeysuckle Creek; 23, Seven Creeks (upper reaches); 24,
Faithfuls Creek; 25, Seven Creeks (lower reaches); 26, Hughes
Creek (upper reaches); 27, Hughes Creek (lower reaches); 28,
Sunday Creek; 29, Strath Creek; 30, King Parrot Creek; 31,
Yea River; 32, Murrindindi River; 33, Home Creek; 34, Lake
Eildon and inflowing rivers; 35, Goulburn River (Thornton);
36, Goulburn River (Cathkin); 37, Goulburn River (Seymour-
Kerrisdale); 38, Goulburn River (Tabilk-Goulburn Weir); 39,
Goulburn River (Shepparton); 40, Lake Victoria; 41, River
Murray (Echuca); 42, Campaspe River; 43, Axe Creek; 44,
Coliban River; 45, River Murray (Torrumbarry); 46, Loddon
River; 47, Laanecoorie Reservoir; 48, Bet Bet Creek; 49, Deep
Creek ( = Tullaroop Creek); 50, Lake Daylesford; 51, River
Murray (Swan Hill); 52, River Murray (Euston); 53, Avoca
River; 54, Wimmera River; 55, Burnt Creek; 56, Marma Lake;
57, Taylors Lake; 58, Richardson River; 59, Mokepilly Creek;
60, Glenelg River; 61, Wannon River; 62, Grange Burn; 63,
Hopkins River; 64, Mt. Emu Creek; 65, dam at Mortlake; 66,
Lake Wendouree; 67, Moorabool River; 68, Barwon River; 69,
Werribee River; 70, Kororoit Creek; 71, Middle Gully Creek;
72, Deep Creek; 73, Plenty River; 74, Yarra River; 75, La
Trobe River; 76, dam at Cheltenham; 77, dam at Bayswater;
78, dam at Ringwood; 79, Edwardes Lake, Preston; 80, dam at
Greenvale; 81, Coburg Lake; 82, Fish Hatchery at Studley
Park.

Distribution of Macquarie Perch Outside Victoria
Whitley (1964) mentioned that Macquarie perch oc¬
curred in the upper reaches of the Murray-Darling River
system in southern Queensland, New South Wales, Vic¬
toria and South Australia and some eastern streams of
New South Wales. However, there are no corroborative
reports of Macquarie perch occurring in southern
Queensland, and the occurrence of Macquarie perch in
South Australia is based on one doubtful record from
the Murray River (Scott et al. 1974). Reports of their oc¬
currence in the River Murray upstream of the South
Australian border have been incorporated into the
distribution data for Victoria. Macquarie perch have
also been recorded from the Edwards (or Kyalite) River;
Wakool River; Wyangala Reservoir on the Lachlan
River and in the Lachlan and Abercrombie Rivers
upstream of the Reservoir. In the Murrumbidgee River
catchment, they have been taken at Narrandera, Wagga
Wagga and just below Burrinjuck Dam on the Murrum¬
bidgee River itself; in Burrinjuck Dam and the Yass and
Goodradigbec Rivers that enter it; near Tantangara
Dam in the headwaters of the Murrumbidgee River; in
the Cotter River system and the Cotter Reservoir; in
Googong Reservoir and the Queanbeyan River that
flows into it, and in the lower reaches of the Umeralla
River. Macquarie perch have also been recorded in
several coastal rivers and their tributaries in New South
Wales, including the Hunter River; Gross River and Ne¬
pean River (Hawkesbury River catchment); Kangaroo
River, Mongarlowe River, Northangara Creek and
Tallowa Dam Impoundment (Shoalhaven River catch¬
ment); Clyde River; and Warrnambool River (Tuross
River catchment); and in several Sydney water supply
dams such as the Avon Reservoir.

Stead (1913) mentioned that the Macquarie perch oc¬
curring in the Shoalhaven, Hawkesbury and Hunter
Rivers and “probably a number of others farther north”
were of a smaller variety than those occurring in the
Murray-Darling system. At present, the Macquarie
perch in these coastal rivers are generally thought to
have been introduced (by either Aboriginal or European
man) from stocks occurring west of the Great Dividing
Range, in the Murray-Darling River system, but there
are no records of Macquarie perch having been transfer¬
red to these rivers (Bishop & Tilzey 1978). This is in con¬
trast to the detailed description given by Stead (1913) of
the transfer in 1912 of 414 Macquarie perch, ranging in
length from 76 mm to 229 mm, from the upper reaches
of the Murrumbidgee River near Cooma to the Snowy
River near Dalgety.

According to Bishop & Tilzey (1978) the numbers of
Macquarie perch in the Murray-Darling system in New
South Wales appear to have been drastically reduced
during the last 20 years, as also have the numbers of
Macquarie perch in the Mongarlowe River. Pratt (1979)
reports a similar decline in the numbers of Macquarie
perch in the Canberra region, where the Macquarie
perch populations are now small and localised.

26

P. L. CADWALLADER

Name of water

Table 1

Known Translocations of Macquarie Perch in Victoria
See text for further details and discussion of the terms “Murray perch” and “bream”, etc.

Release locality

Source locality

Ref. No. Catchment

(see Fig. 1)

Date Name of water

Ref. No.

(see Fig. 1)

A. Translocations within the natural geographical range of Macquarie perch

Bet Bet Creek

48

Loddon River

1930

Goulburn Weir

38

Boosey Creek

17

Broken Creek

1917

Not recorded

-

1962

Broken River

20

Coliban River

44

Campaspe River

1936

Broken River

20

1962

Broken River

20

Lake Daylesford

50

Loddon River

1931

Goulburn Weir

38

Deep Creek

49

Loddon River

1930

Goulburn Weir

38

(= Tullaroop Creek)
Faithfuls Creek

24

Goulburn River

1922

Stony Creek

21

Honeysuckle Creek

22

Goulburn River

1937

Broken River

20

Hughes Creek

26

Goulburn River

1922

Goulburn Weir

38

(upper reaches)

1980

Dartmouth Dam

3

Meadow Creek

12

Ovens River

1973

Seven Creeks

23

Lake Sambell

7

Ovens River

1928

Ovens River

13

Seven Creeks

23

Goulburn River

1921

Goulburn River

36

(upper reaches)

1921

Seven Creeks
(lower reaches)

25

1922

Goulburn River

36

Sunday Creek

28

Goulburn River

1917

not recorded

—

Lake Victoria

40

Goulburn River

1937

not recorded

—

1937

Broken River

20

B. Translocations to

outside the natural geographical range of Macquarie perch

Avoca River

53

Avoca River

1927

Goulburn Weir

38

Barwon River

68

Barwon River

1931

Lake Eildon

34

1935

Broken River

20

1938

not recorded

_

Burnt Creek

55

Wimmera River

1920

not recorded

_

Deep Creek

72

Maribyrnong River

1917

not recorded

_

Glenelg River

60

Glenelg River

1930

Goulburn Weir

38

Grange Burn

62

Glenelg River

1922

Goulburn Weir

38

1926

Goulburn Weir

38

Hopkins River

63

Hopkins River

1913

not recorded

Kororoit Creek

70

Kororoit Creek

1912

not recorded

La Trobe River

75

La Trobe River

1927

Goulburn Weir

38

Marina Lake

56

Wimmera River

1922

Goulburn Weir

38

Middle Gully Creek

71

Maribyrnong River

1926

Goulburn Weir

38

(= Riddells Creek)

Mokepilly Creek

59

Wimmera River

1913

not recorded

Moorabool River

67

Barwon River

1935

Broken River

20

Mortlake, dam

65

Hopkins River

1920

not recorded

Mt. Emu Creek

64

Hopkins River

1920

not recorded

1922

Goulburn Weir

38

1926

Goulburn Weir

38

Plenty River

73

Yarra River

1912

not recorded

Richardson River

58

Wimmera River

1910

Goulburn Weir

38

Taylors Lake

57

Wimmera River

1933

Wakool River

NSW

1935

not recorded

Wannon River

61

Glenelg River

1917

not recorded

1920

not recorded

_

1922

Goulburn Weir

38

Catchment

Goulburn River

Goulburn River
Goulburn River
Goulburn River

Goulburn River
Goulburn River

Goulburn River
Goulburn River
Goulburn River

Mitta Mitta Rive
Goulburn River
Ovens River
Goulburn River

Goulburn River

Goulburn River

Goulburn River

Goulburn River
Goulburn River
Goulburn River

Goulburn River
Goulburn River
Goulburn River

Goulburn River
Goulburn River
Goulburn River

Goulburn River

Goulburn River
Goulburn River

Goulburn River
Murray River

Goulburn River

DISTRIBUTION OF MACQUARIE PERCH

Tabu; 1 ( Continued)

27

Re marks

2 oq^° v * ; 200 Macquarie perch released in Deep and Bet Bet Creeks at Maryborough.

5 0-in Urray perch re,easecl in Boosey Creek at Tungamah.

40o ^ small Macquarie perch released.

Mac aC( ? Uarie perch nelted near Nalinga on the Broken River released in the Coliban River downstream of Malmsbury Reservoir,
the S ne perch taken in order t0 stock the Coliban and Campaspe arms of Lake Eppalock. Considered to be insufficient fish for a release
29 i 1 d * so the were stocked in farm dams in the Coliban-Campaspe catchment.

19 \? n ' ; Macquarie perch released.

0V -; 200 Macquarie perch released in Deep and Bet Bet Creeks at Maryborough.

300

F e b Sr ? a ^ Macc I uar i e P erch released in the upper reaches of Faithfuls Creek.

“18 r ,? Macquarie perch, average length 150 mm, released.

lns °f cod, bream and golden perch taken to the Ruffy area and released in local waters.

15 No**'’ ll Macquarie perch > range 215-390 mm and weight range 100-600 g.

lNQ - ~ - w, released in upper reaches of Hughes Creek.

50 28 Macquarie perch, 115-230 mm long, from the upper reaches of the Seven Creeks system released.

45 ivi acc * Uar * e perch, 150-400 mm long, released.

_ WJaCnilQrJrt _ I._. .1 ^

Creeif CqUarie perch taken at Ca thkin on the Goulburn River released near Strathbogie above the Gooram Falls in the upper reaches of the S

Ks system.

Ab,

Str athbo°i SmaU COd and Macquarie perch were taken in the Seven Creeks system below the Gooram Falls and released upstream

Ja n .^? Uarie perch taken at Cathkin on the Goulburn River released near Strathbogie in the upper reaches of the Seven Creeks systen
Oct! 2n ^ urray perc h released in Sunday Creek at Broadford.

Nov’.Macquarie perch released in Lake Victoria, Shepparton.

•’ 33 Macquarie

arie perch released.

14 Ma *. 128 Macc l uarie perch released. 16 Nov.; 250 Macquarie perch released.

April- 97 ^ Macquarie perch released in the Barwon River at Princes Bridge near Geelong.

Aprij! R ® perc h fry, lengths 60-215 mm, released in the Barwon and Moorabool Rivers.

De c «* 2 n ^ ac 9 uarie perch released.

Jan.*’ 2 nn ! vIurray P e rch» average length 150 mm, released in Burnt Creek near Horsham.

20 isj 0v U Murray perch released in Deep Creek at Romsey.

Feb.- 2 nn^^ Macquarie perch released in Glenelg River at Casterton.

N°v’, 2 on^ aCquarie perc,1 » wei S ht ran ge 60-680 g, released in Wannon River and Grange Burn near Hamilton.

Dec.-’ , J; Murray perch released in Wannon River and Grange Burn near Hamilton.

50 Mur ° Murray P erch released in Hopkins River at Ararat.

9-12 N ray pcrc h released in Kororoit Creek, Melbourne.

Feb. ; 230 \ 1072 Macquarie perch released in La Trobe River at Traralgon.

No v \ 3 0n \ Iacquarie perc h. weight range 60-680 g, released in Marma Lake, Murtoa.

Murray perch released in Middle Gully Creek and Reservoir, Macedon.

De c . _

April; 93 n Murray perch released in Mokepilly Creek at Stawell.

Dee^’jQo w GrcR ,ry > lengths 60-215 mm, released in the Barwon and Moorabool Rivers.

Dee. ; 200 m rray perck ’ avera ge length 150 mm, released in dam at Mortlake.
p eb.; 3 -),. Murray perc h. average length 150 mm, released in Emu Creek (sic), Skipton.

^ov.- 200 \, acquarie P erc h> from 60 to 680 g, released in Emu Creek (sic), Skipton.
lOQ Mur ^ Urray perc h released in Emu Creek (sic), Skipton.

^ 0v • -Dee^ic 4 FCdl re l easec l ,n Plenty River, south Yan Yean.

Ap ri|; go’’ ; fyIurray perc h released in Richardson River at Donald; length and weight ranges of 18 tagged fish; 115-200 mm, 30-131
Apri]. 5 Q , and pcrch released.

500 M d anC * perc h* lengths 250-460 mm, released.

^ec.; 4 Qo urra y per ch released in Wannon River near Hamilton.

Fe b. ; 375 m 11 ^ perc h* average length 150 mm, released in Wannon River, 200 at Coleraine and 200 at Hamilton.

Wa Pnon p . acquane P e rch, from 60 to 680 g, released in Wannon River at Coleraine, and a further 200 of the same size range release<

K,ver and Grange Burn, Hamilton.

28

P. L. CADWALLADER

Table 1 ( Continued)

Release locality

Name of water

Ref. No.
(see Fig. 1)

Catchment

Lake Wendouree

66

Barwon River

Werribee River

69

Werribee River

Wimmera River

(?)

54

Wimmera River

Yarra River 74 Yarra River

Miscellaneous lakes and ponds in

or near

Melbourne

dam, Cheltenham

76

sea

dam, Greenvale

80

Yarra River

dam, Ringwood

78

Yarra R. or
Dandenong Ck.

Edwardes Lake, Preston

79

Yarra River

dam, Bayswater

77

Dandenong Creek

Coburg Lake

81

Yarra River

fish hatchery, Studley Park

82

Yarra River

DISCUSSION

In their natural geographical range in Victoria, Mac¬
quarie perch have, during the last decade, been recorded
at only 20 of the 52 localities where they had previously
been recorded. Of these 20 localities, three (Meadow
Creek and the upper reaches of both Hughes and Seven
Creeks) are waters into which they were introduced from
elsewhere within their natural range. Of the waters out¬
side their natural range which have been stocked, only in
three (Wannon, Barwon and Yarra Rivers) have Mac¬

Source locality

Date

Name of water

Ref. No.

(See Fig. 1)

Catchment

1926

Goulburn Weir

38

Goulburn River

1927

Goulburn Weir

38

Goulburn River

1910

Goulburn Weir

38

Goulburn River

1920

not recorded

—

-

1922

Goulburn Weir

38

Goulburn River

1926

Goulburn Weir

38

Goulburn River

1930

Goulburn River

38

Goulburn River

1910

Goulburn Weir

38

Goulburn River

1920

not recorded

-

—

1922

Goulburn Weir

38

Goulburn River

1926

Goulburn Weir

38

Goulburn River

1938

Murray River

?

Murray River

1948

Kyalite River

NSW

Murray River

1949

Broken River

20

Goulburn River

1950

Broken River

20

Goulburn River

1907/8

Goulburn River

38

Goulburn River

1909

Goulburn Weir

38

Goulburn River

1910

Goulburn Weir

38

Goulburn River

1911

not recorded

_

_

1912

not recorded

-

—

1914

not recorded

—

—

1915

not recorded

—

—

1917

not recorded

—

—

1920

not recorded

—

-

1926

Goulburn Weir

38

Goulburn River

1927

Goulburn Weir

38

Goulburn River

1930

Goulburn River

38

Goulburn River

1931

Goulburn River

38

Goulburn River

1932

Goulburn River

38

Goulburn River

1933

Goulburn River

37

Goulburn River

1934

Goulburn River

?

Goulburn River

1938

Broken River

20

Goulburn River

1943

Broken River

20

Goulburn River

1915

not recorded

_

_

1915

not recorded

-

-

1920

not recorded

_

_

1922

Goulburn Weir

38

Goulburn River

1922

Goulburn Weir

38

Goulburn River

1922

Goulburn Weir

38

Goulburn River

1922

Goulburn Weir

38

Goulburn River

1927

Goulburn Weir

38

Goulburn River

quarie perch been taken since 1970, and only in the
Yarra with any regularity. In most waters, both within
and outside the natural range, only relict populations re¬
main, with recent surveys recording only a few fish, in
most cases often only single individuals, and anglers’
reports similarly referring to the capture of only a few or
single individuals. At present the largest population of
Macquarie perch in Victoria is in the newly-formed Lake
Dartmouth on the Mitta Mitta River. In the Goulburn
River catchment there are several isolated populations,

DISTRIBUTION OF MACQUARIE PERCH

Table 1 ( Continued)

29

Re marks

2l No 2 °°') MUrray PCrCh released in Gran S e Burn and Wannon River at Hamilton.

Nov .n ’ 264 Nlacquarie P erch released in Wannon River at Hamilton.

400 Murrain 4 -a 1Urray re !?¥cn m LakC WendoUree ’ Ballarat; length and weight ranges of 26 tagged fish: 100-220 mm, 14-140 e
Feb • tS 7 V\ PCrC1 ’ avera ^ c ^ en Sth 150 mm, released in Melton Weir on Werribee River.

" 110 M aCqUariC PCrCh ’ fr0m 60 10 680 8 ’ re,eased in Melton Weir on Werribee River.

Nov

lln -'-’ ^ vwv 6, iwvajLU III IVICIIUIJ weir c

19 Anr I U 1 l,rray perc . h released m Melton Reservoir on Werribee River.

Nov.-De 2 >_ Macquarie perch taken at Tabilk released in Werribee River below the Exford Weir.

inov -n ptiv.ii laivcn <xi faun* ieieasea in werrioee Kr

De c.’ 3nn’i 250 Murray perch reIeascd Wimmera River at Horsham
Feb • 4 «:n \ urray P crc h, average length 150 mm, released in Wimmera River at Horsham
Nov’- 3sn\7 CqUarie perch ’ from 60 t0 680 released near Horsham (no waters mentioned).
April’. Murray perch released in Wimmera River at Horsham.

An w’ 8 cod and Macquarie perch released in Wimmera River.

60 Ma,

July- a i Cquarie per ch and Murray cod released in Wimmera River.
Anri’i. Macquarie perch released in Wimmera River at Eversley.

pr, h ISA m - 1 .w,t«owv» ui rv iiiiiiici a ixivci cii cversiey.

500 Ma cquarie perch released in Wimmera River at Eversley.

■ *urrav hroom — _ 1 . r_ i_;n. i i • .,

60-iV; uv ' : 975 perch and bream released between Studley Park and Heidelberg; length and weight ranges of 48 tagged fish: 135-215

O en *" n 1 roducccf Uiere 1 ^ fr ° m Tabilk re ‘ eaSed ” Yarra RivCT betWeen Abbotsford and Heidelberg to supplement fish which had previo
GCI -'N0V.;

6 °H85o
Nov.-d^ *

o nd 31 Fairfield ( "? f 'the Macquarie variety”-V P.C. records) released above Dights Falls (3100 fish), at Swan Street Bridge (

9-13 Nov m 575 ^’ gth and Welghl ranges of 97 ta 8S ed fish: 100-255 mm, 14-215 g S '

m 00 Murrav „ N , 1Un ', ay ? 1 5® ed t' 6 DightS Falls (4 °° fish) ’ at Swan Street Brid 8 e ( 400 > and « Johnston Street (200)

N 0v • inn perch reIeased at Johnston Street. y '

Nov.’Id U Murray pcrch released at Fairfield.

Jan *; 4mn\ 4325 Murray perch released at Studley Park.

Dec • 22 on M Urray perch released at Richmond.

Si »&* average length 150 mm, released at Swan Street Bridge (500 fish) and Studley Park (1700)

Ma C q Ua Perch released at Fairfield.

A PriI; 34S C M rch released above falls (Dights ?): 500 on 14-15 Nov., 150 on 17 Nov. and 382 on 24 Nov.

!s M arc h; iSr PerCh ' kT!' 150 r 2. m u-n n8 ’, lakCn at TabHk released in Yarra River between Dights Falls and Alphington.

8 M^i* ^ IVlaCQllflnp n^rrh Inlrpri ot TiLill/ ralanmrl Vn— n n:_ ^ ®

uSh 1 -

16 Macquarie perch taken at Tabilk released in Yarra River.

Macquarie pcrch taken at Tabilk released at Heidelberg.

4 Macquarie perch, 100-240 mm long, taken at Kerrisdale released at Fairfield.

^acoti • pcicn, iuu-z^u mm long, taK<

208 M a ^ ane perch released near Burke Road Bridge.

40 CQd^nd 3 - 16 PCrCh re,eased at w arrandyte.

N ° v -D ec .«

Perch released at Warrandyte.

N° v .~Dee .’ 3 Murray perch and 2 grunters released.
*' * Murray perch and 2 grunters released.

Dec

Fe b.; 8() Murray perch, average length 150 mm, released.

Fe b.; 5 () M acquar | e P e rch, from 60 to 680 g, released.

Fc b.; 50 L aCquade P erch > from 60 to 680 g, released.

Feb.; 2o a |V? Uar * e percF ’ f rom 60 to 680 g, released.

^ c ans” n f acc l ua rie perch, from 450 to 910 g, placed in ponds at the hatchery.
Macquarie pcrch sent to hatchery.

u PDer y at ,° ne or two localities, Hughes Creek and the
Pon u ,nH eaCh f S °* the Seven Cr eeks system, can the
u Pper r^u be considered viable. The population in the
from fi if ru S the S even Creeks system was derived

%ed I?™ in the area in I921 “ 22 and per-
D e rch* i rS | aS lbe natura l population of Macquarie
probahi the ° Wer reaches has been almost eliminated
Si^ . as a result of siltation (Cadwallader 1979).
in the j° n bas Probably been the most important factor
aecline of Macquarie perch in Victoria (Cad¬

wallader 1978). Although they may thrive in artificial
impoundments, Macquarie perch naturally are riverine
fish and are normally found in deep holes, but require
shallow water flowing over a gravel-pebble-boulder
substrate for spawning (Wharton 1968, Cadwallader &
Rogan 1977). Silt, by filling the deep holes, destroys
Macquarie perch habitat and, by settling on the river
bed, provides conditions unfavourable for the demersal
eggs of Macquarie perch. It also affects the composition
of the benthic fauna which forms the main component

30

P. L. CADWALLADER

of the diet of Macquarie perch (Cadwallader & Eden
1979). Significantly, most of today’s riverine popula¬
tions of Macquarie perch are in the upper reaches of
catchments where siltation loads are not heavy and
where there are still deep holes interspersed with shallow
riffles.

The construction of dams and weirs for hydro¬
electric, irrigation and water conservation schemes has
also played a role in the decline of Macquarie perch.
Dams prevent Macquarie perch from moving upstream
(see Appendix 3 in Cadwallader 1977). Moreover, since
no dam in Victoria (including those on the River Murray
adjoining Victoria) is fitted with multi-level water
offtakes, water released downstream is taken from near
the base of the water column and is too cold (e.g. 9°C
from Lake Eildon) to induce spawning of Macquarie
perch, which require a temperature threshold of 16.5°C
for spawning during late spring-early summer (Wharton
1968, Cadwallader & Rogan 1977).

Interaction (including competition and predation)
with introduced fish is another factor to be considered in
the decline of Macquarie perch. For example, although
the evidence is circumstantial, the European perch Perea
fluviatilis (Linnaeus) has been implicated in the decline
of Macquarie perch in Lake Eildon (Cadwallader &
Rogan 1977). Various other factors, such as overfishing,
“river improvement” and pollution (particularly that
arising from spraying crops along river banks), have also
probably contributed to the decline of Macquarie perch
in some areas, but, again, no studies have been under¬
taken to examine the role of these factors in the decline
of Macquarie perch.

ACKNOWLEDGEMENTS

I am grateful to the many people who provided data on
the distribution of Macquarie perch. In particular, I
thank Fisheries and Wildlife officers Pat Sheridan, Jack
Rhodes and Jim Crozier for comments on the distribu¬
tion of Macquarie perch in their areas of operation, and
Charles Barnham (Fisheries and Wildlife Division) who
unearthed the archival material on early translocations
of Macquarie perch. 1 also thank Ray Donald who took
part in most of the Macquarie perch surveys, Darwin
Evans for his comments on the manuscript, and Keith
Bishop (N.S.W. State Fisheries) for providing un¬
published data on the distribution of Macquarie perch in
New South Wales and for his comments on the
manuscript.

BIBLIOGRAPHY

Anon., 1973. Native fish in the Campaspe and Coliban Rivers.

Fish Wild!. Div., Vic .. Freshwat. Fish. News/. 5: 18-19.
Anon., 1974. Endangered native fish. Fish. Wild/. Div., Vic.,

Fresh wat. Fish. News/. 7: 9-15.

Bishop, K. A. & Bell, J. D., 1978. Observations on the fish

fauna below Tallowa Dam (Shoalhaven River, New South

Wales) during river flow stoppages. Aust. J. mar. Freshwat .
Res. 29: 543-549.

Bishop, K. A. & Tilzey, R. D. J., 1978. We/come ReeJ Pro-
ject Environmental Study—Aquatic Life. Metropol. Wat.
Sewerage and Drainage Bd., Sydney, 110 pp.

Butcher, A. D., 1946. The Freshwater Fish of Victoria and
Their Food. Fish. Game Dept., Vie.. 64 pp.

Cadwallader, P. L., 1977. J. O. Langtry’s 1949-50 Murray
River investigations. Fish. Wild/. Pap., Vic. 13, 70 pp.

Cadwallader, P. L., 1978. Some causes of the decline in
range and abundance of native fish in the Murray-Darling
River system. Proc. R. Sac. Viet. 90: 211-224.

Cadwallader, P. L., 1979. Distribution of native and in¬
troduced fish in the Seven Creeks River system, Victoria.
Aust. ./. Ecol. 4: 361-385.

Cadwallader, P. L., & Eden, A. K., 1979. Observations
on the food of Macquarie perch, Macquaria austra/asica
(Pisces: Percichthyidae), in Victoria. Aust. J. mar.
Freshwat. Res. 30: 401-409.

Cadwallader, P. L., & Rogan, P. L., 1977. The Macquarie
perch, Macquaria austra/asica (Pisces: Percichthyidae), of
Lake Eildon, Victoria. Aust. J. Eco/. 2: 409-418.

Hungerford, R. (Ed.), 1977. Complete Rook of Australian
Fishing. Rigby, Adelaide. 1043 pp.

Lake, J. S., 1967. Freshwater Fish of the Murray-Darling
River System. N.S.W. State Fish. Res. Bull. 7: 48 pp.

Lake, J. S., 1971. Freshwater Fishes and Rivers of Aus¬
tralia. Nelson, Melbourne. 61 pp.

Lake, J. S., 1978, Australian Freshwater Fishes. Nelson,
Melbourne. 160 pp.

Llewellyn, L. C. & Macdonald, M. C., 1980. Australian
freshwater basses and cods. In Freshwater Fishes of South¬
eastern Australia , R. M. McDowall, ed., Reed. Sydney,
142-149.

McKeown, K. C., 1934. Notes on the food of trout and Mac¬
quarie perch in Australia. Rec. Aust. Mus. 19: 141-152.

Pratt, B., 1979. The Canberra Fisherman. ANU Press,
Canberra. 120 pp.

Roughley, T. C., 1966. Fish and Fisheries of Australia. Angus
and Robertson, Sydney. 328 pp.

Scott, T. D., Glover, C. J. M. & Southcott, R. V., 1974. The
Marine and Freshwater Fishes of South Australia (2nd Edi¬
tion). Government Printer, South Australia. 392 pp.

Stead, D. G., 1913. An account of some experiments in the ac¬
climatisation of two species of Australian freshwater perch.
Rept. Aust. Assoc, Adv. Sci. 14(D): 279-288.

Tunbridge, B. R., 1978. A survey of the fish populations in the
Mitta Mitta River and tributaries before the construction of
Dartmouth Dam. In Dartmouth Dam Project: Report on
Environmental Studies. S.R.W.S.C., Victoria.

Tunbridge, B. R. & Rogan, P. L., 1976. A guide to the
inland angling waters of Victoria. Government Printer, Vic¬
toria. 149 pp.

Wedlick, L., 1977. How to catch Redfin and other

Australian Perch. Wcdneil, Melbourne. 80 pp.

Wharton, J. C. F., 1968. Spawning areas of the Macquarie
perch Macquaria austra/asica above the Eildon Lake (Vic¬
toria). Aust. Soc. LimnoL Newsl. 6(1): 11-13.

Wharton, J. C. F., 1973. Spawning induction, artificial
fertilization and pond culture of the Macquarie perch (Mac-
quaria austra/asica (Cuvier 1830)). Aust. Soc. LimnoL Bull.
5: 43-65.

Whitley, G. P., 1964. Native Freshwater Fishes of Australia.
Jacaranda, Brisbane. 127 pp.

THREE COROPHIOIDS (CRUSTACEA: AMPHIPODA) FROM
WESTERN PORT, VICTORIA

By J. L. Barnard* and Margaret M. Drummond!

♦Smithsonian Institution, Washington, D.C. 20560, USA

tNational Museum of Victoria, 285-321 Russell Street, Melbourne, Victoria 3000

Abstract: Species in three genera of the superfamily Corophioidea are described for the first time in
modern context. Baracuma alquirta gen. et $>p. nov. (Ischyroceridae) lies close to the traditional Cerapus
of the literature, a cosmopolitan genus of diverse form. A new species of the cosmopolitan Laet-
matophilus , L. dabberi, (Podoceridae) is fitted into a world key. The unusual Leipsuropus parasiticus
(Podoceridae) is redescribed.

New material of three corophioid Amphipoda from
Victoria permits detailed descriptions of a new genus
Baracuma (Ischyroceridae), a new species of Laet-
matophilus (Podoceridae) and the unusual Leipsuropus
parasiticus (Podoceridae). Baracuma is close to Cerapus
and Rurtanga Barnard 1961. A key to the species of
Laetmatophilus is provided to distinguish this new
species within the genus. Leipsuropus is one of the most
unusual gammaridean amphipods known because of the
selective loss of uropod 2; generally, in an evolutionary
cycle, gammaridean amphipods lose uropod 3 first and
then uropod 2 but in Leipsuropus a small uropod 3 is re¬
tained.

The study was initiated and the major part of it car¬
ried out at the Marine Studies Group of the Ministry for
Conservation in Victoria during J. L. Barnard’s visit to
Australia in 1976. Materials came from the two major
benthic Surveys of Western Port: Crib Point Benthic
Survey, 1964-5 (CPBS) and the Westernport Bay En¬
vironmental Study, 1973-4 (WPBES). The Australian
Museum, by courtesy of Dr J. K. Lowry, Curator of
Crustacea, loaned specimens from the Old Collection
for use in the Leipsuropus parasiticus study, and made
available type material for examination. Details of the
Western Port Surveys have been published previously by
Barnard and Drummond (1978) and much of the
literature on Australian amphipods can be traced by
consulting the same reference.

LEGENDS

Capital letters and numbers on the figures denote
parts, as follows:

A, antenna; B, body or carcass; C, coxa; D, dactyl; E,
epistome, left view; F, accessory flagellum; G,
gnathopod; H, head; I, inner plate of maxilliped; J,
ramus; K, variable, see legend; L, lower lip = labium; M,
mandible; N, molar; O, palp; P, pereopod; Q, pleopod;
R, uropod; S, maxilliped; T, telson; U, upper
lip = labrum; V, brood plate; W, pleon; X, maxilla; Y,
gill; Z, gland.

The figures each contain illustrations from a master
specimen listed first in the caption of each figure and no
lower case letters are placed on these figures; subsidiary
specimens on each figure are denoted by lower case let¬
ters as specified in the caption for each figure,
ci

SYSTEMATICS

Superfamily COROPHIOIDEA
Family ISCHYROCERIDAE

Three kinds of Cerapus— like genera are now
recognizable. The first of these to be described as
distinct from Cerapus was Rurtanga Barnard (1961)
which differed markedly in many characters. Later, Mc¬
Cain (1969) described a second species, Rurtanga waiora
which lessened the distance between the old concept of
Cerapus and the more distant Rurtanga, To some extent
our new genus also lessens this distance, but examina¬
tion of species of so-called Cerapus , of which C.
tubularis Say is the type-species, reveals generic distinc¬
tions.

Genus Baracuma gen. nov.

Diagnosis: Thorn-like appearance of rostrum weak. Ar¬
ticle 1 of antenna 1 untoothed, articles 2 and 3 elongate,
longer than article 1, primary flagellum much longer
than any peduncular article, accessory flagellum absent.
Antenna 2 slender, slightly longer than antenna 1,
flagellum of male as long as article 5 of peduncle. Man¬
dibular palp with 3 normal articles. Inner plate of max¬
illa 1 with one large seta. First 3 and last 2 coxae short,
not touching serially; pereonite 2 in neither sex differen¬
tially elongate in comparison to pereonite 1; male coxa 1
much shorter than segment 1; coxa 5 of both sexes
longer than pereonite 5, coxa 5 in female almost asetose
and folding to meet partner ventrally; only female
pereonite 5 much longer than pereonite 1. Gnathopod 1
normally subchelate. Gnathopod 2 in female simple
though article 6 slightly inflated, article 2 not heavily
setose anteriorly; gnathopod 2 in male very large, essen¬
tially carpochelate, these two teeth of carpus gaping. Ar¬
ticle 3 of pereopod 4 elongate. Pleopod 3 with 2 rami.
Uropod 3 with one small ramus. Uropod 2 with one
vestigial, mostly fused ramus bearing 2 hooks. Telson
narrow, cleft halfway, apices armed with rows of studs.
Male with ventral keel.

Type Species: Baracuma alquirta gen. et sp. nov.
Relationship: In the absence of the type specimen of
Cerapus tubularis Say 1817, and of an unequivocal
diagnosis of the genus, comparison of Baracuma with

31

32

J. L. BARNARD AND MARGARET M. DRUMMOND

Fig. 1 —Baracuma alquirta new species, holotype male “b” 3.92 mm; a = female “a” 3.56 mm; f = female “f” 3.40 mm.

COROPHIOIDS FROM WESTERN PORT 33

Fig. 2 — Baracuma alquirta new species, holotype male “b” 3.92 mm; a = female “a” 3.56 mm.

34

J. L. BARNARD AND MARGARET M. DRUMMOND

Fig. 3 — Baracuma alquirta new species, holotype male “b” 3.92 mm; a = female “a”

thoracic sternites.

C. tubularis must be made from re-description of the
species by S. I. Smith (1880), Stebbing (1906) and the il¬
lustrations of the female by Bousfield (1973). Whether
Smith and Bousfield were even dealing with the same
species seems uncertain in view of, for example, the
difference in the telson, as illustrated, and in the relative
lengths of the pereonites as described or figured by the
two.

Smith found only 3 pairs of gills and 3 pairs of brood
plates, but classic Cerapus tubularis specimens in
Smithsonian collections have formulas identical with
Baracuma. In a very detailed description, Smith did
not mention either a ventral keel on the male nor an ex¬
ceptionally elongate female coxa 5 with folding capabili¬
ty. Both are very noticeable features.

The genus differs from C. tubularis as conceived of
from Bousfield’s illustrations of the female, and from
examination of specimens at the Smithsonian Institu¬
tion, in the irregular form of coxae 1-4, the elongate

coxa 5 of both sexes, the apically narrowed poorly cleft
telson, and the male ventral keel. Baracuma differs from
the classic C. crassicornis (Bate), as described and
figured by Sars (1895), additionally, in the absence of a
dorsal tooth on article 1 of antenna 1, and in the shorter,
thicker second and third peduncular articles.

Baracuma differs from Runanga in the absence of an
accessory flagellum, the longer coxa 5 and its folding
capability in the female, the poor development of setae
on female coxa 5, the lack of long dense setae on the
anterior margin of article 2 of female gnathopod 2, and
the narrow telson.

Runanga waiora McCain 1969 bears a scale-like ac¬
cessory flagellum, has a broad telson, and does not have
the special characteristics of female coxa 5 found in
Baracuma.

Baracuma appears to be fairly close to the Australian
species of Cerapus , but again, the narrow telson, the
shape of female coxa 5, and the male ventral keel are

COROPHIOIDS FROM WESTERN PORT

35

strong distinctions. Australian species of Cerapus differ
from the Northern Hemisphere concept of the genus in
the lack of dorsal teeth on article 1 of antenna 1, the
elongate articles 2 and 3 of antenna 1, the lack of sexual
distinction on pereonite 1 and its coxa, and in the
elongate article 3 of pereopod 4.

The narrow telson distinguishes Baracuma from all
other Cerapus -like taxa. This telsonic form has the ap¬
pearance of being most primitive in the group; but other
adaptations, such as elongation of body segments and
coxa 5, appear to be advanced characters.

Baracuma alquirta gen. et sp. nov.

Figs 1-3

Description of the Male: Head as long as pereonite 5,
the longest pereonite; pereonites 3 and 4 shortest.
Rostrum short, blunt, broadly tapered. Eyes darkly
pigmented. Peduncular articles of antenna 1 subequal in
length, first article broad, dorsoventrally expanded,
lacking distodorsal cusp; articles 1 and 2 long and rec¬
tangular; flagellum much longer than peduncular article
3, 7-articulate, first article nearly as long as combined
length of next two; locus of accessory flagellum (near 2
setae) marked only by invagination attached to tangent
reflecting internal ridge probably representing vestigial
accessory flagellum. Peduncle of antenna 2 rather
stouter than that of antenna 1, article 5 longer than arti¬
cle 4; flagellum subequal to or longer than article 5 of
peduncle; first article longer than combined lengths of
second and third.

Mandibular incisors with 6 teeth; left lacinia mobilis
broad, right rather narrower, both toothed; left mandi¬
ble with 3 rakers, right with 2; each molar with scaled
flake, articles 2 and 3 of palp slender, subequal in
length, only article 3 with setae.

Lower lip with inner lobes and short, splayed man¬
dibular processes.

Inner plate of maxilla 1 small, with one long stout
apical seta; outer plate with 11 spines; right and left
Palps alike with about 6 apical spine-setae and 3
subapical setae. Both plates of maxilla 2 apically setose,
inner with a few widely-spread apico-medial and medial
setae. Inner plate of maxilliped truncate, with 3 stout
apical teeth and a crescent of apical setae; outer lobe
spined, and with a few medial setae; article 2 of palp
medially setose; article 3 with subapical spray of long
setae; article 4 stout, blunt, tipped with 3 setae.

Coxae 1-4 very shallow, different from each other and
irregular in shape. Coxa 1 very small, rounded, set on
extreme anteroventral angle of pereonite 1; coxa 2
shorter than pereonite 2, attached well towards posterior
margin of its pereonite, resultant gap between coxae 1
and 2 revealing distinctively enlarged part of ventral
keel. Coxa 5 longer than pereonite 5, bulged anterodor-
sally to overlap coxa 4, and with vestigial posterodorsal
lobe extending beyond posterior margin of pereonite.
Coxa 6 larger than coxa 7, both irregular in shape. Coxal
gills present on pereonites 3-6, largest and inflated on
pereonite 4; strap-shaped on pereonites 3 and 5; very
small on pereonite 6.

C2

Gnathopod 1 subchelate, articles 5 and 6 subequal.
Gnathopod 2 carpochelate, outer tooth on article 5
straight in line with posterior margin, deep sinus be¬
tween outer and inner teeth; article 6 narrow, less than
half as wide as article 5.

Article 2 of pereopod 3 broader than that of pereopod
4, bulging anteroproximally; article 3 of pereopod 4
longer than articles 4, 5 or 6. Pereopod 5 shorter and
stouter than pereopods 6 or 7, articles 4, 5 and 6
inflated, articles 4 and 6 almost surrounding article 5;
dactyl stout, recurved. Pereopods 6 and 7 similar,
pereopod 7 slightly the longer, dactyls of both with
broad, bulging base and two-toothed apex.

Epimera posteroventrally rounded; epimeron 3 with
several short ventral setae.

Pleopods decreasing in size; pleopod 1 with 2 long
setose rami, inner longer, outer broader; outer rami of
pleopods 2 and 3 leaf-shaped, setose; inner rami of both
very small, ovate.

Peduncle of uropod 1 reaching beyond the apex of
urosomite 3; outer ramus longer than inner, with short
lateral setae and apical spine. Uropod 2 with one very
short ramus tipped with a seta. Peduncle of uropod 3
inflated; ramus vestigial, fused, with 2 apical and 2 ac¬
cessory hooks.

Telson nearly 80°/o as long as wide, cleft about 50%,
narrowing to rounded apices each with 2 sets of apical
studs.

Description of female: Antennae shorter than those
of male, peduncle of antenna 2 relatively more slender;
flagella of both antennae 5-articuIate. Pereonite 5 much
longer than in the male, longer than any other pereonite,
though pereonite 4 also elongate; pereonites 1 and 2
shortest. Coxae 1-4 shallow, of different shapes; coxa 4
longer than 1, 2 or 3; coxa 5 enormously elongate,
longer than pereonite 5, much longer than deep, almost
asetose and folding to meet partner of opposite side in
midventral line.

Hand of gnathopod 1 better developed than in male.

Brood plates present on pereonites 2-5; that on
pereonite 5 large and distally expanded; others short and
slender.

Pleopod 2 and epimera like male “b”, but epimeron 3
with only 2 ventral setae. Pleopod 3 with only one seta
on inner ramus, 10 on outer.

Ventral keel absent except for protruding support for
maxillipede on sternite 1. Female otherwise resembling
male.

Only domiciliary tube found in collection about 8 mm
long, very dark, blackish-green in colour, round in sec¬
tion, broader at one end, smoothly and evenly con¬
structed, containing female.

Illustrations: Maxilla 1 enlarged more than maxilla 2.
Unfolded coxa 5 of female drawn to same enlargement
as coxa 5 of male by overlaying respective fifth
pereopods; brood plates of coxae 3 and 4 enlarged to
this same degree. Gnathopod 1 and pereopods 3-7 at
same magnification. Apex of pereopod 6 like that shown
for pereopod 7. Male telson unflattened.

36

J. L. BARNARD AND MARGARET M. DRUMMOND

Holotype: NMV J1256, male “b” 3.92 mm.

Type Locality: CPBS 41N/1, Australia, Victoria,
Western Port. 24 Steptember 1973, 12.8m, fine gravel
and sand with mud.

Paratypes: Type-locality, female “A” 3.56mm, J1266,
CPBS 32N/367, female “f” 3.40 mm, J1267; CPBS 12N,
1975, male, female in domiciliary tube, J1287.
Material: CPBS, 40 samples from 15 stations (92
specimens); WPBES, 7 samples from 5 stations (15
specimens).

Distribution: Australia, Victoria, Western Port, 5-18.3
m, sand, gravelly sand, muddy sand, coarse sand and
shell, coarse sand and mud.

Family Podoceridae
Genus Laetmatophilus Bruzelius
Key to the species of Laetmatophilus

1. Head with one or more distinct dorsal teeth.2

Head lacking distinct teeth.5

2. Pereonites transversely rugose but lacking

multiple transverse teeth dorsally.

. L. tuberculatus

Pereonites with multiple transverse teeth dorsally.3

3. All pereonites with teeth on midline. L. hala

Some pereonites with teeth set sagitally in pairs on

either side of midline.4

4. Anteroventral corner of head with sharp tooth, at

least 5 pereonites and pleonites with 4 or more

teeth each. L. hystrix

Anteroventral corner of head quadrate, only one
pereonite or pleonite with more than 2 dorsal

teeth...

.L. armata

5. Outer ramus of uropod 1 much less than half as long

as inner ramus (dactyl of male gnathopod 2 not

overlapping palm). L. dabberi

Outer ramus of uropod 1 more than half as long as
inner ramus.6

6. Hand of gnathopod 1 very slender (inch L. sp.

Sivaprakasam 1970).7

Hand of gnathopod 1 stout.8

7. Dactyl of male gnathopod 1 fitting palm.

... w . L. durbanensis

Dactyl of male gnathopod 1 overlapping palm.

. L. leptocheir

8. Palm of male gnathopod 2 with 2 teeth. L. purus

Palm of male gnathopod 2 with 3 teeth. L. tridens

Laetmatophilus dabberi sp. nov.

Figs 4, 5

Diagnosis: Head lacking dorsal teeth, anteroventral cor¬
ner with small cusp, cephalic lobe with cusp, anteroven¬
tral corner of first antennal podium with small cusp.
Pereonite 1 with 2 low transverse humps, pereonite 2
with one similar hump, pereonite 3 to pleonite 2 with
dorsal hump or carina on midline, pleonite 3 weakly
humped; pleonites laterally with plaques, especially in
female; pereonites 5-7 coalesced. Coxae 2-4 with ventral

points. Articles 2 and 3 of gnathopod 2 with sharp
anteroventral cusp(s), articles 3-4 with sharp pos-
terodistal cusp; dactyls of gnathopods 1-2 fitting palm;
male gnathopod 2 palm with 3 teeth, 2 of these near
defining corner sharp, third tooth flat and representing
most of palm; female gnathopod 2 with strongly convex
simple palm. Outer ramus of uropod 1 much less than
half as long as inner ramus.

Description: See illustrations. Specimens mostly with
missing appendages, only pereopods 4 and 6 of adult
pereopods 3-7 recovered and illustrated; right and left
gnathopod 2 of male “b” of different sizes (illustrated);
young male gnathopod 2 like female. Gills thin, borne
on coxae 2-7 in female, 2-6 on male; brood plates very
broad and setose, borne on coxae 2-4. Peduncles of
pleopods longer than wide, each with 2 coupling hooks,
no other major setae, rami subcqual, article 1 elongate,
counts of articles on pleopods 1-3, outer and inner, of
male “h”=3-4, 5-5 and 5-5.

Illustrations: Detached antenna 2 of male “m” drawn
to same enlargement as body of male “h”. Medial tex¬
ture on dactyl of gnathopod 1 with basal limit marked
on Figure 5 Glh by dashed line, then enlarged in Figure
5 DGlh.

Holotype: NMV J1279 male “b” 3.64 mm.

Type Locality: WPBES 1746/2, Western Port, 25
November 1974, 24 m, gravelly coarse sand, Victoria,
Australia.

Paratypes: Type locality, female “a’’ 3.14 mm J1280,
juvenile “c” 1.91 mm J1282, male “m” 2.21 mm J1281,
male “p” 2.60 mm J1283; WPBES 1747/3, female “f”
3.20 mm J1284, male “g” 3.34 mm J1285, male “h” 2.80
mm J1286.

Relationship: The short outer ramus of uropod 1
coupled with lack of multiple teeth on body segments
and unextended dactyl of gnathopod 1 distinguish this
species from any others known. Laetmatophilus hystrix
and L. hala from Australia and Hawaii respectively,
have multiple teeth on body segments. The species in
couplets 7 and 8, from Africa or the Indian Ocean, all
have a much longer outer ramus of uropod 1 than does
L. dabberi , those of couplet 7 also having a thin hand on
gnathopod 1. Both L. purus and L. tridens have a male
gnathopod 2 like Podoeerus with the dactyl and palm
occupying the full posterior margin of the hand.
Material: WPBES, 2 samples from 2 stations (9).
Distribution: Australia, Victoria, Western Port, sand.

Genus Leipsuropus Stebbing 1899

Leipsuropus Stebbing 1899, p. 241; 1906, p. 698.

Type Species: Cyrtophium parasiticum Haswell (mono-
typy). Unique.

Diagnosis: Accessory flagellum vestigial; antenna 1
shorter than antenna 2; some coxae touching each other
or weakly overlapping; pereonites 6-7 amalgamated;
urosome with 3 segments, uropod 1 well developed,
uropod 2 absent, uropod 3 forming small setose leaf
lacking rami.

Remarks: Barnard (1969) did not accept HaswelPs and
Stebbing’s observations on the absence of uropod 2 but

COROPHIOIDS FROM WESTERN PORT

37

Fig. 4- Laetmatophilus ciabberi new species, holotype male “b” 3.64 mm; a = female “a” 3.14 mm; h = male “h” 2.80 mm; m = male

“m” 2.21 mm.

38

J. L. BARNARD AND MARGARET M. DRUMMOND

Fig. 5 — Laetrnatophilus dabberi new species, holotype male “b” 3.64 mm; a = female “a” 3.14 mm; h = male “h” 2.80 mm.

they were indeed correct as shown in the illustrations for
the type-species presented below. The diagnosis is re¬
vised to show the presence of a vestigial accessory
flagellum and the fusion of pereonites 6-7.

Leipsuropus parasiticus (Haswell 1879)

Figs 6, 7

Cyrtophium parasiticum Haswell 1879, p. 274; 1882, p.
271; 1885, p. 108, pi. 17 figs 1-7.

Leipsuropus parasiticus Stebbing 1906, p. 699; 1910, p.
650.

Description: Rostrum small, thin, blunt; ocular lobes
forming lateral nacelles, apex of head broad in lateral
view, forming weak cavity for reception of antenna 1.
Article 2 of antenna 1 about twice as long as article 1,
scarcely longer than article 3, primary flagellum com¬
posed of 4 articles together shorter than 3, first article
elongate, accessory flagellum vestigial, antenna 1 strong¬

ly setose ventrally. Antenna 2 much larger than antenna
1, article 5 of peduncle almost 1.4 times as long as article
4, these two articles moderately setose ventrally,
flagellum about as long as article 4 of peduncle, compos¬
ed of one elongate article tipped with 2 vestigial articles.

Epistome with large anterior tooth, upper lip incised
ventrally. Right mandibular incisor with 5 teeth, left
with 4, right lacinia mobilis with 3 or 4 teeth, left with 4,
right rakers 2, left 3, right molarial seta elongate, left
short; palp article 3 short, clavate, heavily setose. Man¬
dibular lobes of lower lip thin. Inner plate of maxilla 1
obsolescent, asetose, outer plate with 8 spines, right and
left palps similar. Inner plate of maxilla 2 lacking sub¬
marginal setae. Palp article 4 of maxilliped short, stub¬
by, with 2 apical rows of setae.

Coxae poorly setose; coxa 1 broadly produced
anterodistally, boot-shaped; ventral margins of coxae
2-4 sinuate, coxae 2 and 3 rounded-quadrate, coxa 4

COROPHIOIDS FROM WESTERN PORT

39

Fig. 6- Leipsuropus parasiticus (Haswell), male “d” 2.95 mm; g= female “g” 3.32 mm; k = female “k” 2.85 mm; m = male “m”

4.41 mm.

40

J. L. BARNARD AND MARGARET M. DRUMMOND

broader than tall; female coxae 2-4 anteroposteriorly
elongate, much broader than tall, sinuate, coxa 3 larger
than coxa 2, coxa 4 very large.

Gnathopod 1 of both sexes similar, article 2 not
cuspidate. Gnathopod 2 of male enlarged, article 2 with
sharp apicolateral cusp, apicomedial lobe blunt and
enlarged, articles 4 and 5 with posterior conical projec¬
tion, article 5 almost obsolescent; hand large, longer
than broad, palm carved into giant proximal defining
tooth separated by large sinus from sinuate, irregularly
scalloped distal marginal blade, entire palm and
posterior margin of hand moderately setose, dactyl
long, and in terminal males, strongly overreaching palm;
female gnathopod 2 very small, appearing almost as it
regenerant or stunted, articles 2, 4 and 5 unornamented,
hand of simple gammarid type with almost transverse
and simple palm and weak defining tooth, dactyl scar¬
cely overreaching palm.

Articles 5-7 poorly setose on pereopods 3 and 4, more
strongly on pereopods 5, 6 and 7.

Pereonites with transverse dorsal rugae, weaker in
female than in male, pereonites 6 and 7 fused together;
pereonites 2-3 and 4-5 with especially well developed
anterior or posterior projections above coxae. Pleonites
dorsally untoothed; epimera naked.

Uropod 1 well developed and spinose, outer ramus
shorter than inner; uropod 2 absent; uropod 3 forming
small setose leaf, lacking rami and hidden beneath
telson. Telson broad, short, linguiform, each side with
triad of dorsal penicillate setules.

Male “t’\ Port Jackson: Hand of gnathopod 2 five
percent narrower than in figured male (measuring
anterior margin to base of proximal tooth), main palmar
projection and proximal narrow tooth much shorter
than in figured male; hand thus appearing to be much
narrower than in fully developed Victorian male but ac¬
tual width scarcely distinct and most of narrowness ow¬
ing to poorly developed sculpturing.

Observations: Most specimens lacking all or parts of
antenna 2 and pereopods 5-7, often lacking pereopods
3-4, illustrated pereopods picked from different
specimens, best specimen juvenile male “p” with
pereopods 6 and 7 observed, article 5 of pereopod 5
stouter than that illustrated for male “n”; antenna 1 of
male “m” as showm for female “k”.

Intersexes: Intersexes were found at five Western Port
stations:

(1) CPBS 32S/1, J1257: One individual (out of 13 in
sample) with very large, asetose brood plates and
typically female coxae 2-4 in addition to well developed
penes. Gnathopod 2, though smaller than normal for a
male of the size, larger than the normal female hand,
with 2 proximal teeth and 2 sinuses, with the finger
fitting into the most proximal sinus.

(2) CPBS 32S/367: One specimen (out of two) closely
resembling the above. (In the Ministry for Conservation
Marine Studies Division Collection.)

(3) CPBS 33N/3, J1276: One specimen, resembling the
two above, but with only one palmar sinus.

(4) CPBS 35N/2, J1277: One specimen with half-

grown, asetose brood plates and typical female coxa 4,
well developed penes and broad, inflated gnathopod 2,
bearing one proximal tooth and with one sinus.

(5) CPBS 61N/1, J1278: One specimen with large,
asetose brood plates, female coxae 2-4 and well
developed penes; but gnathopod 2 hand closer to male
type, with fairly oblique palm.

Illustrations: Antenna 2 drawn to same magnification
as antenna 1; spination obscured by agglutinates, pro¬
bably not complete; female gnathopod 2 drawn as right
sided attached to left coxa; female gnathopods 1-2,
pereopod 3 and coxa 4 equally magnified.

Types: AM “G. 5388, Cyrtophium parasiticum Hasw.,
Port Jackson, Type , Old Coll 3 sp.”, later added to card
“3-4 fms 1879 P.T.O. 320.1”; 3 specimens mounted on
white plate with gum in alcohol; one female and one
male removed Nov 4 1976; female corresponds with il¬
lustrations herein; male gnathopod 2 satisfactory but
urosome too occluded with deposit to determine iden¬
tity; both specimens replaced in 2 tiny vials with white
plate carrying third specimen. Special techniques will be
required to remove occlusions from male to prepare as
lectotype.

Other Material: Male “t” 3.40 mm, AM P3426, Cyr¬
tophium parasiticum and 13 specs. Port Jackson Old
Coll ( = narrow-handed form).

Voucher Material: CPBS 32S/1, male “d” 2.95 mm J
1270, female “g” 3.32 mm J1269, female “k” 2.85 mm,

J1268; WPBES 1746/2, young male “p” 2.55 mm,
J1272, female “q”, J1271; WPBES 1747/2, male “m”
4.41 mm, J1273, male “n” 4.35 mm, J1274.

Material: CPBS, 11 samples from 6 stations (26
specimens); WPBES, 3 samples from 3 stations (17
specimens); AM P3426, 1 sample (14 specimens).
Distribution: Australia, Western Port, Victoria, and
Port Jackson, New South Wales, 12-24 m, sand, muddy
sand, and muddy sand and gravel.

ACKNOWLEDGEMENTS

In an earlier paper (Barnard & Drummond 1978)
acknowledgement of the help from many individuals
and organizations during the Westernport Bay Survey
was expressed and is reiterated here. In connection with
the present study special acknowledgement is made to
Dr Alistair Gilmour, Officer-in-Charge of the Marine
Studies Group in 1976, and to Danuta Karpow, for
technical assistance.

At the Smithsonian Institution thanks are due to
Carolyn Cox and Irene F. Jewett who prepared the sket¬
ches for press, and to Elizabeth B. Harrison and Janice
Clarke for technical assistance. The authors are deeply
indebted to Dr Dillon S. Ripley, Secretary of the
Smithsonian Institution, who made it possible for J. L.
Barnard to come to Australia; and that author extends
his personal thanks to members of the Marine Studies
Group, and in particular to Dr Alistair Gilmour and Dr
Jerry D. Kudenov, for many kindnesses during his visit.

This publication is No. 324 in the Ministry for Con¬
servation, Victoria, Environmental Studies Series.

COROPHIOIDS FROM WESTERN PORT

41

Fig. 7 — Leipsuropus parasiticus (Haswell) male “d” 2.95 mm; g = female “g” 3.32 mm; k = female “k” 2.85 mm; n = male “n”

4.35 mm.

REFERENCES

Barnard, J. L., 1961. Gammaridcan Amphipoda from
depths of 400 to 6000 metres. Galathea Rept. 5: 23-128.

Barnard, J. L. & Drummond, M. M., 1978. Gammari-
dean Amphipoda of Australia, Part III: The Phox-
ocephalidae, Smithsonian Contr. Zool. 245: 1-551.

Bousfield, E. L., 1973. Shallow-water Gammaridean
Amphipoda of New England. Cornell University Press,
Ithaca.

Haswell, W. A., 1879. On Australian Amphipoda.
Proc. Linn. Soc. N.S. W. 4: 245-279.

Haswell, W. A., 1882. Catalogue of the Australian

stalk-and-sessile-eyed Crustacea. Aust. Mus. Sydney (plus
addenda et corrigenda).

Haswell, W. A., 1885. Notes on the Australian Am¬
phipoda. Proc. Linn. Soc. N.S. W. 10: 95-114.

McCain, J. C., 1969. A new species of deep sea am-
phipod (Gammaridea) belonging to the genus Runanga.
N. Z. J. mar. freshw. Res. 3: 17-19.

Sars, G. O., 1895. An account of the Crustacea of Nor¬
way with short descriptions and figures of all the species. 1,
Alb. Cammermeyers Forlag, Christiania.

Say, T., 1817. On a new genus of Crustacea and the
species on which it was established. J. Acad. nat. Sci.
Phi lad. 1: 49-52.

Sivaprakasam, T. E., 1970. A new species and a new
record of Amphipoda from the Madras coast../. mar. biol.
Ass. India 10: 274-282.

Smith, S. I. 1880. On the Amphipodus genera, Cerapus ,
Unciola , and Lepidactylis, described by Thomas Say.
Trans. Conn. Acad. Arts Sci. 4: 268-277.

Stebbing, T. R. R., 1899. On the true Podocerus and
some new genera of amphipods. Ann. Mag. nat. Hist, ser 7,
3: 237-241.

Stebbing, T. R. R., 1906. Amphipoda 1. Gammaridea.
Das Tierreich 21.

Stebbing, T. R. R., 1910. Crustacea. Part 5. Am¬

phipoda. Sci. Res. Trawling Exped. H.M.C.S. “Thetis”.
Mem. Aust. Mus. 4: 565-658.

OSTRACODA FROM AUSTRALIAN INLAND WATERS-
NOTES ON TAXONOMY AND ECOLOGY

By P. De Deckker

Zoology Department, University of Adelaide, Adelaide, S.A.

(Present address: Department of Biogeography and Geomorphology, Research School of Pacific Studies,
Australian National University, Canberra, A.C.T. 2601)

Abstract: One new ostracod genus, Ampullacypris , and ten new species of ostracods are described:
Candonocypris incosta, Cypricercus salinus, C. unicornis , Heterocypris vatia, Ilypdromus amplicolis, l.
candonites, /. dikrus, Kapcypridopsis asynunetra , Limnocy there dorsosicula, L. milt a. Three other
species are re-described, namely Candonocypris novaezelandiae (Baird 1843), Cypretta baylyi McKenzie
1966 and Ampullacypris oblongata (Sars 1896); 2 cosmopolitan species Eucypris virens (Jurine 1820) and
Sarscypridopsis acu/eata (Costa 1847) are recorded for the first time in Australia.

Ecological notes for these species as well as for Limnocy there mowbrayensis Chapman 1914 are
presented.

INTRODUCTION

Knowledge of the taxonomy and ecology of ostracods
from Australian inland waters is poor compared to that
of other microcrustaceans, although ostracods are quite
common in a variety of habitats. This paper presents
new data on non-marine ostracods for use in future
ecological studies and for studies of Quaternary
material. Since ostracod shells are readily fossilized
these data may be useful in palaeolimnological studies
(see De Deckker 1981b).

Material for this study is deposited in the Department
of Crustacea, National Museum of Victoria under the
registry numbers: J1134-J1162. The following abbrevia¬
tions are used in the text: C = carapace, H = height,
L = length, LV = left valve, RV = right valve.

SYSTEMATICS

Subclass Ostracoda Latreille 1806

Order Podocopida Muller 1894
Superfamily Cytheracea Baird 1850
Family Limnocytheridae Klie 1938
Subfamily Limnocytherinae Klie 1938

Genus Limnocythere Brady 1867
Type Species: Limnocythere inopinata (Baird 1843).

Limnocythere dorsosicula n. sp.

Figs 1, 2a-j

Diagnosis: Member of Limnocythere with three to six
small posterodorsal spines on the right valve; two small
dorsal bosses separated by a main depression in the mid¬
dle and never higher than the hinge in lateral view.
Outline of hemipenis as in Fig. 1H.

Description: carapace (External) —Rectangular,
faintly reticulated, and pitted to smooth; three main
depressions on each valve: one in the centre where a
vertical column of four muscle scars is often visible,
another just above and a third in front just below the
hinge line; greatest height at about one quarter to one

fifth of length from anterior; greatest width at about 0.6
of length from anterior; right valve with three to six
small spines along its edge posterodorsally; in dorsal
view, anterior narrow and pointed; two small dorsal
bosses, separated by the main depression in the middle,
smooth, never higher than the hinge line in lateral
profile. Sexual dimorphism pronounced —length:height
ratio of valves greater in males.

(Internal)—Hinge with a broad tooth in right valve
and a matching depression in the left one at both ends;
inner lamella broadest anteriorly and peripheral selvage
faint; radial pore canals numerous and straight from
which many hairs protrude at a distance from the outer
lamella anteriorly.

anatomy (Antennula) — (Fig. 1A) Six-segmented;
length: width ratio of the last five segments: 2:1, 1:1,
1:1.3, 2:1, 4.2:1, longest distal seta bifid at about mid¬
length.

(Antenna) —(Fig. IB) Two pectinate distal claws and
another thinner and barren; distal segment small and
squa’rish.

(Mandible) —(Fig. ID) Mandibular coxale with seven
teeth; palp with distal segment very small and squarish
and with three thin setae; distal seta on penultimate seg¬
ment thicker than the other three and pectinate.

(Maxillula) — (Fig. 1C) Distal palp elongate with three
setae; 3rd lobe with three others, two of which are
biramous.

(Maxilla) —(Fig. IF) Short and stocky; no setae on 1st
segment.

(Thoracopoda I) —(Fig. 1G) Longer than maxilla;
distal end of 1st segment with two unequal setae; one at
proximal end; another at mid-length.

(Thoracopoda II) —(Fig. IE, J) Longer than thora¬
copoda I with distal claw almost twice its length and
three times that of the maxilla claw; in female, setae
pectinate but barren in male where the distal seta on the
2nd segment has a bifid tip.

(Hemipenis) —For outline see Fig. 1H.

(Genitalia) —For outline see Fig. II.

(Furca) —(Fig. 1H, I) One small and barren setae near
the reproductive organs.

43

44

P. DE DECKKER

Fig. 1 — Limnocythere dorsosicula n. sp. Lake Tcrangpom, Vic. A, B, E-H are drawn from the holotype
adult male and C, D, I, J from the paratype female. Scale = 100/*. A, antennula. B, antennna. C, max-
illula —palp and lobes. D, mandible—palp. E, thoracopoda II. F, maxilla. G, thoracopoda I. H,

hemipenis. I, genitalia. J, thoracopoda II.

OSTRACODA FROM INLAND AUSTRALIA

45

Colour of Shell Light brown to transparent.

Size: L H L H

holotype adult

male LV 410/* 230/* RV 410/* 230/*

paratype adult

female LV 450/* 230/* RV 450/* 230/*

Type Locality: Lake Terangpom, west of Lake Cor-
angamite, western Victoria.

Derivation of Name: From Latin dorsum ( = black)
and sicula (= small spine) for the diagnostic posterodor-
sal spines on the right valve.

Ecology and Distribution: Only two collections of
this species are known to me, one from Lake
Terangpom in water of 2.03°/ oo salinity and the other
from South Nerrin Nerrin Lagoon in water of 1.96°/ 0 o
salinity, both in western Victoria. At first glance, it ap¬
pears that this species is indicative of freshwater
(<3%o) despite the fact that some species of Lim¬
nocythere live in saline waters (see L. milta n. sp. below
and De Deckker 1981c). However, L. dorsosicula which
has been recovered in many samples of a core from Lake
George (see De Deckker 1981b), is found in some of
these samples co-occurring with other ostracod species
indicative of either fresh water or water of salinity

<10°/oo.

Remarks: L. dorsosicula is easily distinguishable from
L. notodonta Vavra 1906 from west Java since the latter
species has a maximum of four posterodorsal spines on
the right valve. The anterior of the shell of the former
species is narrow and pointed whereas in the latter
species, the shell is much broader and rounded at both
extremities. L. dorsosicula differs from L. mowbrayensis
Chapman 1914 as the latter has broad alae, which are
rounder or pointed and curved backward at about mid¬
height near the centre of the shell. Dorsal spines have
also been noticed on one fossil juvenile specimen of L.
mowbrayensis from Pillie Lake, S. A. (De Deckker
1981b), whereas this feature appears common on
specimens from Pulbeena Swamp illustrated by Brehm
(1939) for L. percivali (later synonymized to L.
mowbrayensis by Hornibrook (1955) and Deevey
(1955)). Hornibrook (1955), however, did not mention
any spines on his specimens from the same locality. In
addition, L. mowbrayensis is characterized by a dorsal
boss at mid-length which extends above the hinge line
when seen in lateral view. L. stationis Vavra 1891, which
inhabits European waters, also possesses posterodorsal
spines but only on the left valve.

Limnocythere milta n. sp.

Figs 2k-r, 3

Diagnosis: Member of Limnocythere with faintly
reticulated valve; with a vertical depression above the
central muscle scars separating a small smooth boss
anteriorly from the broad posterior; depression above
and in front of the boss; row of fine denticles along the
posteroventral margin of left valve. Maxilla and
thoracopodae I and II with three long, pectinate setae
on the inside of the 1 st segment.

Description: carapace (External) —Subrectangular
and finely reticulated all over except for the anterior
boss above and in front of the central muscle field and
along the anterodorsal margin; this boss is separated
from the posterior of the shell by a vertical groove just
above the central muscle field; there is a depression adja¬
cent to the boss dorsally which gives it a bilobate ap¬
pearance in dorsal view; greatest height at about one
quarter to one third of length from the anterior; mouth
region concave and at about mid-length; dorsum gently
curved; in dorsal view shell narrow, anterior pinched
and pointed; greatest width at about 0.66 of length from
the anterior; left valve slightly longer than right one
posteriorly; shell compressed posteroventrally where the
inner lamella is broad.

(Internal)-Inner lamella broad anteriorly in both
valves and of almost similar width posteroventrally:
posteriorly at mid-height and above, selvage absent;
numerous straight radial pore canals from which many
hairs protrude anteriorly at a distance from the outer
margin; central muscle field with a vertical row of four
scars; two narrow horizontal ones in the middle
separated by two circular to oval ones; one antennal scar
in front of the row at the level of the top scar and an ad¬
ditional scar above the vertical row of four; all these
scars are met by depressions on the outside of the shell.
Four to six minute spines along the margin of the left
valve posteroventrally; hinge with broad tooth at both
ends in the right valve with matching socket in the left
valve; the posterior tooth is the largest.
anatomy (Antennula) —(Fig. 3A) Six-segmented; length
width ratio of last five segments: 2:1, 1.2:1, 1:2, 1:1, 5:1;
longest seta bifid at Vs from its base.

(Antenna)-(Fig. 3B) Three barren distal claws; distal
segment almost rectangular.

(Mandible)-(Fig. 3C, D) Mandibular coxale with
seven teeth, the inner two acicular; palp with distal seg¬
ment almost squarish; at the distal end of the 1 st seg¬
ment, thick seta (a bristle?) stout, pointed and pilose;
on 2 nd segment there are four setae, two long ones and
two pectinate and shorter (one is a p bristle?); distal end
of third segment with one long and barren seta and
another half its length and pectinate (7 bristle?); three
unequal setae on distal end of last segment.

(Maxillula) —(Fig. 3E) Epipod with 14 long and one
small plumose Strahlen plus a shorter barren one; palp
two-segmented with distal segment rectangular; for
chaetotaxy see Fig. 3E.

(Maxilla) —(Fig. 3F) Distal claw stout and curved;
three thick and pectinate setae on inner side of 1 st seg¬
ment and a longer pectinate one near its base outside.

(Thoracopoda I)-(Fig. 3G) Similar to maxilla but
larger.

(Thoracopoda II) —(Fig. 3H) Similar to thoracopoda
I but larger and with no basal seta on the outside of the
1 st segment.

(Genitalia) —Weakly chitinized (see Fig. 31).

(Furca) —(Fig. 31) Single barren seta.

(End of body) (Fig. 3J) With tuft of hairs and one
biramous short seta.

46

P. DE DECKKER

Fig. 2—a-j, Limnocythere dorsosicu/a n.sp. Lake Terangpom, Vic. a, c* g, i, j female paraiypes; b-d, f
male paratypes; h male holotype. a, C showing RV. b, RV external, c, LV external, d, C showing RV. e,
C dorsal, f, C dorsal, g, RV internal, h, LV internal, i, LV internal, j, anterior detail of i. k-r, Lim¬
nocythere milta n. sp. Small lake N.W. of Lake Werowrap, Vic. k. 1, o-r female paratypes; m, n female
holotype. k, C showing RV. 1, LV external, m, LV internal, n, RV internal, o, RV external, p, C showing
LV. q, C dorsal, r, C ventral. (Scales: 1 =200/x for a-i, =35/zfor j; 2 = 200/* for k-r.)

OSTRACODA FROM INLAND AUSTRALIA

Fig. 3 — Limnocy there milta n. sp. Small lake NW of Lake Werowrap, Vic. Drawn from holotype adult
female. Scale = 100 fi. A, antennula. B, antenna. C, mandible—palp. D, mandible —coxale. E, maxillula.
F, maxilla. G, thoracopoda I. H, thoracopoda II. I, genitalia. J, end of body.

48

P. DE DECKKER

Colour of Shell: Yellow to light brown.

Size: L H L H

Holotype adult

female LV 545 g. 310 fi RV 540/i 31 Op

Type Locality: Small lake north-west of Lake
Werowrap, Red Rock area, near Colac, western Victoria
(38° 15'23"S, 143°29'35”E).

Derivation of Name: From Greek miltos meaning red
earth for the Red Rock area.

Ecology and Distribution: L. milta is known only
from the type locality where salinity was 15.42°/oo and
pH 9.5. This lake is known to dry up occasionally. No
males have yet been found.

Remarks: L. milta differs from L. aspera Henry 1923, as
the latter does not possess the typical posteroventral
spines along the margin of the left valve.

Limnocythere mowbrayensis Chapman 1914

1914 Limnicythere mowbrayensis Chapman p. 60.

1955 Limnicythere sicula; Hornibrook p. 268.

1955 Limnicythere mowbrayensis; Hornibrook, p. 268.
1978 Limnocythere mowbrayensis; McKenzie, p. 181.

1980 Limnocythere sp., De Deckker & Geddes, p. 691.

1981 Limnocythere mowbrayensis; De Deckker, p. 37.
Diagnosis: Member of Limnocythere with almost
straight dorsum and deeply concave ventrum; two large
dorsal bosses, which, in lateral view, extend above the
hinge line, are separated by a vertical groove which is
situated above a vertical row of four muscle scars; in
front of the row, there is a broad lateral process which,
on most occasions, is pointed and curved backwards.
Discussion: L. mowbrayensis has recently been re-
described by De Deckker (1981a). Limnocythere sp.,
briefly described by De Deckker and Geddes (1980) from
an ephemeral salt lake near the Coorong Lagoon, is here
considered to be L. mowbrayensis as it is almost iden¬
tical to the specimens of L. sicula described by Chapman
(1919), later synonymized by Hornibrook (1955) to L.
mowbrayensis , as it has poorly developed lateral pro¬
cesses.

Ecology and Distribution: L. mowbrayensis cannot
swim: it is usually found crawling among filamentous
algae. It is a fresh water species which can tolerate
slightly saline waters up to 6%>o. This upper record
refers to the Limnocythere sp. of De Deckker and Ged¬
des (1980) mentioned above, and is not surprising as
some other Limnocythere species can inhabit saline
waters (e.g. L. milta ; see De Deckker 1981c).

L. mowbrayensis has also been recorded at 2.8°/oo in
Fresh Dip Lake, near Robe, S. A. Apart from the
ephemeral locality near the Coorong Lagoon where L.
mowbrayensis was collected only once, all other
localities are permanent; this species has never been
found in temporary pools.

L. mowbrayensis has been recorded from southern
Australia (even Kangaroo Island) and as fossil from
north-western Tasmania (from where it was originally
described) and New Zealand.

Superfamily Cypridacea Baird 1845
Family Cyprididae Baird 1845

Subfamily Herpetocypridinae Kaufmann 1900

Genus Candonocypris Sars 1896

Type Species: Cypris candonioides King 1855 ( = Can¬
donocypris novaezelandiae (Baird 1843).

Diagnosis: Adult with smooth elongated shell and with
broad inner lamellae anteriorly; selvage prominent and
raised posteroventrally in the right valve. Two jointed
sensory seta on the 2nd segment of the antennula.
Thoracopoda II with two setae at mid-length on the last
segment.

Remarks: The opinion held by Sars (1894) that the well
defined selvage placed far away from the edge of the
right valve anteriorly was a diagnostic feature of Can¬
donocypris species is no longer valid as this feature is
not present in C. incosta which, on other features of the
shell and anatomy, is considered here to be a true Can¬
donocypris.

Two Australian species are included in Can¬
donocypris namely, C. incosta n. sp. and C. novaezelan¬
diae (Baird 1843). Herpetocypris caledonica Mehes
1939, from New Caledonia, definitely represents a Can¬
donocypris species since he illustrated the distal segment
of thoracopoda II with two setae at mid-length. No type
material could be examined, as it has not been deposited
in the Natural History Museum in Basle, Switzerland
contrary to Mehes’ (1939) indication. (C. Stocker pers.
comm. 26 Jan. 1981).

Candonocypris incosta n. sp.

Figs 4, 5

1914 Candona lutea ; Chapman, p. 59, fig. 6.

1971b Ilyodromas cf. smaragdinus ; McKenzie, p. 396.
1977 Ilyodromus cf. smaragdinus; Danielopol &
McKenzie, p. 309.

Diagnosis: Member of Candonocypris with peripheral
selvage anteriorly in both valves and broad and near the
outer margin in the posterior of the right valve.
Description: carapace (External) —Smooth and
elongated, ellipsoid shell with dorsum arched and with
ventrum flat except in front of the middle where it is
slightly concave. Both ends of the valves tapering but
posterior more pointed. Greatest height at about mid-
length. Shell narrow in dorsal view. Obvious overlap of
the left valve antero- and posterodorsally.

(Internal) —Inner lamellae similar in both valves
anteriorly and approximately twice as broad anteriorly
compared to the posterior area. Selvage peripheral
anteriorly and only prominent posteroventrally in the
right valve. This selvage is met by a depression in the left
valve where the selvage is peripheral.
anatomy: The species fully described by McKenzie
(1971b) as Ilyodromus cf. smaragdinus from New
Guinea is here synonymized to C. incosta. Its diagnostic
anatomical features are the short third segment of the
antennula with a length width ratio of about 1.6:1 (Fig.
5A), strongly arched palps on the male maxilla (Fig. 51,

OSTRACODA FROM INLAND AUSTRALIA

49

Fig. 4 — Candoncypris incosta n. sp. Spring at base of limework quarry at Pulbeena Swamp, Tas.
Scales I 000 /x. a, b, i, k, m female paratypes; c, d male holotype; e-h, j-1 male paratypes. a, LV internal,
b, RV internal, c, LV internal, d, RV internal, e, RV external, f, RV external, g, LV internal, h, RV inter¬
nal. i, C dorsal, j, C dorsal, k, C showing LV. 1, C showing RV. m, C showing RV.

50

P. DE DECKKER

Fig. 5—Candonocypris incosta n. sp. Spring at base of limework quarry at Pulbeena Swamp, Tas. C, F,
K are drawn from a paratype adult male, the remainder from the holotype adult male. Scales: 1 = 100/* for
A-F, H-O; 2 = 50/* for G. A, antennula. B, antenna. C, maxilla. D, mandible—palp. E, mandible-cox-
ale. F, maxillula-palp and lobes. G, rake-like organ. H, thoracopoda I. I, maxilla-endopodite. J, max¬
illa. K, thoracopoda II. L, Zenker organ. M, hemipenis. N, furcal attachment. O, furca.

OSTRACODA FROM INLAND AUSTRALIA

51

Fig. 6 — Candonocypris novaezelandiae (Baird 1843). a-d, Kangaroo Creek Reservoir, Adelaide, S.A. c-k,
t, Milbrook Reservoir, Adelaide, S.A. 1-s, Small farm dam near Gilmandyke Creek, S. of Bathurst,
N.S.W. Scales: 1 = 5(XV for a-1, s, t; 2 = 50/x for r. a, b, f-h, k: adult male; e-e, i, j, 1, m, s, t adult female;
n-p, r juvenile female; q juvenile, a, RV internal, b, LV internal, c, RV internal, d, LV internal, e, C dor¬
sal. f, RV external, g, LV external, h, C showing RV and hemipenis. i, LV external, k, C showing LV and
hemipenis. 1, RV internal, m, LV internal, n, RV internal, o, LV internal, p, C showing LV. q, C show¬
ing LV. r, RV internal, anterior detail of n. s, C showing LV. t, C showing LV.

P. DE DECKKER

Fig. 7 — Candonocypris novaezelandiae (Baird 1843). Kangaroo Creek Reservoir, Adelaide, S.A. I, J,
drawn from adult female remainder from adult male. Scales: 1 = 100/* for A-D, F-M; 2 = 50/* for E. A,
antennula. B, antenna. C, mandible-palp. D, maxillula-palp and lobes. E, rakc-likc organ. F,
thoracopoda I. G, maxilla — endopodite. H, maxilla —endopodite. I, maxilla. J, thoracopoda II. K,

hemipenis. L, furcal attachment. M, furca.

OSTRACODA FROM INLAND AUSTRALIA

53

J), outer extremity of copulatory sheath at mid-length
forming a broad hump with right angle (Fig. 5M), and
furca with almost equal claws (Fig. 50).

Colour of Shell: White to transparent.

Size:

L

H

L

H

holotype
adult male

LV

1 220/4

600/4 RV

1 160/4

560/4

paratype

adult

female

LV

1 410/4

700/4 RV

1 390/4

660/4

Type Locality: Spring at base of limework quarry at
Pulbeena Swamp in north-western Tasmania.
Derivation of Name: From Latin in ( = without) and
costa (= ridge) for the absence of prominent selvage in
the right valve anteriorly in comparison with C.
no vaezelandiae.

Ecology and Distribution: This species has been col¬
lected from the type locality only once; it was found
crawling on and within the topmost centimetre of sedi¬
ment in freshwater. It is also recorded from Lake
Peunde, near Mt. Wilhelm (about 3 750 m), Bismarck
Range in New Guinea (McKenzie 1971b). A few
specimens of C. incosta (labelled Candona luted) are
present in Chapman’s (1914) slide of fossil specimens
from Mowbray Swamp in north-western Tasmania. This
species has since been re-collected at that site (De Dcck-
ker 1981b).

Remarks: This species differs from C. novaezelandiae
because of the absence of a prominent selvage at a
distance from the edge of the shell in the right valve.
This prominent selvage was originally thought to be
characteristic of Candonocypris species but this is not
the case. C. incosta has to be included in Candonocypris
as it shares many other shell and anatomical features
with the type species C. novaezelandiae . Both species,
for example, possess the prominent selvage posteroven-
trally in the right valve. In the type lot thoracopoda II
possesses the two typical setae at mid-length on the last
segment (Fig. 5K). This feature is also found on
specimens of llyodromus cf. smaragdinus described by
McKenzie (1971b). As this feature is not found in ll¬
yodromus (verified from type material of llyodromus
smaragdinus , held in the Oslo Museum), this alone
justifies the transfer to Candonocypris of McKenzie’s
specimens. These are now synonymized to C. incosta as
they have identical shell features and anatomies. Addi¬
tionally, the outline of the hemipenis and the very short
two-jointed sensory seta on the second segment of the
antennula (in llyodromus species it is very long and
three-segmented) in C. incosta and C. novaezelandiae ,
further justify the grouping of these two species under
the same genus.

The only feature different in McKenzie’s specimens is
the broader extension of the copulatory sheath over the
lateral lobe. This difference is not considered to be im¬
portant. Comparison with the New Guinea material
kept in the British Museum and the species described
here indicates that all other details of the hemipenis are
the same.

Candonocypris novaezelandiae (Baird 1843)

Figs 6, 7

1843 Cypris Novae Ze land iae Baird, p. 268.

1855 Cypris candonioides King, p. 66.

1855 Cypris sydneia King, p. 65

1889 Herpetocypris Stanley ana\ Sars, p. 35.

1894 Candonocypris assimilis Sars, p. 36.

1894 Candonocypris candonoides; Sars, p. 35.

1919 Candonocypris assimilis; Chapman, p. 28.

1955 Candonocypris assimilis; Hornibrook, p. 271.

1956 Candonocypris candonoides; Hornibrook in Gill &
Banks, p. 19.

1969 Candonocypris assimilis; Hussainy, p. 305.

1971 Candonocypris novaezelandiae; Eagar, p. 55.

1975 Candonocypris assimilis; Okubo, p. 157.

1976 Candonocypris novaezealandiae; Chapman &
Lewis, p. 95.

1976 Candonocypris assimilis; Chapman & Lewis, p. 95.
Diagnosis: Member of Candonocypris with prominent
selvage in right valve usually half way between the outer
and inner margins and following the curvature of the
shell; posteroventrally in the right valve and near the in¬
ner margin, selvage is prominent.

Description: carapace (External) —Smooth shell like a
flattened ellipsoid with dorsal area slightly arched;
overlap of left valve over right one vent rally and to a
lesser extent dorsally at both extremities of the hinge
area; right valve larger and like a flatter ellipsoid com¬
pared to left one.

(Internal) —Broad selvage all around and placed at a
distance from the anterior edge of the right valve; in the
left valve, it is faint and peripheral; in both valves, inner
lamella twice as broad anteriorly; posteriorly in the right
valve, the selvage is prominent, especially posteroven¬
trally where it is near the inner margin: lhis area is met
by a depression in the left valve where the selvage is
faint.

anatomy: Characterized by a long third segment of the
antennula with a length to width ratio of 2.2:1 (Fig. 7A);
male maxillar palps differently arched (Fig. 7G, H);
outer extremity of copulatory sheath at mid-length
forming a narrow but prominent hump (Fig. 7K); furca
with unequal claws (Fig. 7M).

Colour of Shell: Green to beige brown.

Size Range: L H

adult male 1 400-1 500/4 700-800/4

adult female 1 650-1 800/x 750-850/4

Note: LV narrower but taller than RV in both sexes.
Ecology and Distribution: This freshwater species is
usually found in farm dams and eutrophic waters, even
sewerage lagoons. It is commonly found in high
numbers crawling in among decaying vegetal matter and
black organic muds, especially near lake shores. Hus¬
sainy (1969) was the first to describe the male of C.
assimilis (synonymized here to C. novaezelandiae) from
Lake Purrumbete. Males have since been found in a
number of permanent waterbodies (e.g. Milbrook and
Kangaroo Creek Reservoirs) but not in ephemeral
waters or small waterbodies such as farm dams.

Adults of C. novaezelandiae are a benthic species and

54

P. DE DECKKER

have never been seen to swim. Juveniles, on the other
hand, are good swimmers, having natatory setae of their
antennae much longer than in adults.

Remarks: Examination of the type material of C.
novaezelandiae (empty valves only) kept in the British
Museum confirmed the suggestion ot Eagar (1971) that
this species is synonymous with C. candonioides. Addi¬
tionally, since in many collections taken in ephemeral
waters, morphs representing both C. candonioides and
C. assimilis , as illustrated by Sars (1894, Plate V.l and
2 ), are found together, it is suggested here that they
represent the same species: C. candonioiodes synoni¬
mized to C. novaezelandiae. For example, forms oi C.
assimilis as illustrated in Fig. 6 n-p from a small farm
dam near Gilmandyke Creek, south ot Bathurst in New
South Wales are considered to be young specimens of C.
novaezelandiae found in the same collection and il¬
lustrated in Fig. 6 1, m, s. There are no morphological
differences in anatomy except that appendages of C.
novaezelandiae are bigger. In the latter, the colour of
the shell is green with yellow to brown diagonal bands
caused by the ovaries, whereas shells ot C. assimilis
morphs are beige brown in colour with similar bands for
the ovaries (Sars 1894).

The latter morph is smaller and has a slightly arched
dorsum (the greatest height is at the middle) whereas the
C. novaezelandiae morph is larger, more arched dorsally
(greatest height at about 0.66 of length from the
anterior) and with the right valve much larger than the
left anteriorly and posteriorly. This synonymy is further
confirmed by the fact that the anatomy of male
specimens described by Hussainy (1969) from Lake Pur-
rumbete in Victoria for C. assimilis is identical to that of
male specimens of C. novaezelandiae found in Milbrook
Reservoir in South Australia. The presence of well
formed ovaries in juveniles in some ostracod species is
not uncommon in the family Cyprididae and this would
therefore explain why previous authors have considered
C. assimilis morphs to represent the last molt stage of
the species.

From the original illustration and short description of
Cypris sydneia King 1855, it appears that King’s species
represents the C. assimilis morph because of the outline
and colour of the shell, limited ability to swim, and the
habit of crawling on mud.

C. novaezelandiae is found in New Zealand (Sars
1894, Chapman 1963, Chapman & Lewis 1976),
Australia (Sars 1894, 1896a, Henry 1923) and Japan
(Okubo 1975). Originally Sars (1894) stated that this
species was also present in South Africa as he had raised
it in his aquaria from samples of dried mud collected at
Knysna, Cape of Good Hope, but later (Sars 1924) sug¬
gested that this had been caused by contamination by
Australian material in his aquaria.

Genus Ilyodromus Sars 1894

Type Species: Candona stanleyana King 1855.
Diagnosis: See Danielopol & McKenzie (1977, p. 305).
Remarks: The genus Ilyodromus has recently been
redescribed by Danielopol and McKenzie (1977) who

provided a diagnosis for /. stanleyanus and redescrip¬
tion of I. varrovillius (King 1855) from New Zealand
specimens. Both species were originally described from
Australia. These authors also discussed all the other Il¬
yodromus species and their geographical distribution.

Ilyodromus amplicolis n. sp.

Figs 8 , 9 1-r

Diagnosis: Member of Ilyodromus with striated shell;
anterior and posterior ends broadly rounded; a slight
concavity in front of the hinge anterodorsally; inner
lamellae broad anteriorly and posteriorly; lateral lobe of
hemipenis broad and rectangular in shape; maxilla palps
of male similar to each other.

Description: carapace (External) — Weakly calcified;
elongated ellipsoid with joint striations all over; dorsum
straight along the hinge line and slightly concave
anterior to it; dorsally behind the hinge line it is flat and
inclined; anterior and posterior broadly rounded but the
latter is narrower; ventrum almost flat except in the mid¬
dle of the mouth region where it is concave; greatest
height at 0.33 of length from the anterior, left valve
slightly larger.

(Internal) —Inner lamella very broad and similar in
both valves; anteriorly, the width of the inner lamella is
one-third of the length of the shell; there it is slightly
broader and it extends all around the shell except in the
hinge area dorsally; selvage faint and peripheral in both
valves.

anatomy (Antennula) —(Fig. 8 A) Seven-segmented;
length width ratio of the last six segments: 1 . 2 : 1 , 1 . 6 : 1 ,
1 . 2 : 1 , 1 . 6 : 1 , 2 : 1 , 2 : 1 ; natatory setae as long as last five
segments; 3-segmented sensory organ on second seg¬
ment with distal end pointed.

(Antenna) — (Fig. 8 B) Natatory setae short: two
longest ones shorter than the penultimate segment and
two minute ones reaching the proximal end of the same
segment.

(Mandible) — (Fig. 81) Mandibular coxale with seven
teeth; palps with a bristle short, barren and slim, /3 bris¬
tle stout, pointed and densely pilose, 7 bristle thick,
slightly longer than the last segment and pilose in the
distal half.

(Rake-like organ) — (Fig. 8 C, D) Seven to nine teeth
with an additional bifid one on the inner side.

(Maxillula) —(Fig. 8 E) Distal palp trapezoid and two
smooth Zahnborsten on third lobe.

(Maxilla) —Sexually dimorphic; in male (Fig. 8 G, H)
palps strongly and similarly arched and one slightly nar-
rer in the proximal 0.33 of its length; in female (Fig.
8 F) three plumose setae, the middle one being twice the
length of the other two which are equal; in both sexes,
epipod with five long and a shorter plumose Strahlen;
for chaetotaxy of protopod, see Fig. 8 L.

(Thoracopoda I) —(Fig. 8 J) Third segment well
divided; inner distal seta of second segment shorter than
half the length of the 3rd segment and outer seta on
distal segment 0.2 of the length of the claw.

(Thoracopoda II) —(Fig. 8 N) Three-segmented with
large distal pincers; distal setae unequal: shorter one

OSTRACODA FROM INLAND AUSTRALIA

55

Fig. 8 — Uyodromus amp/icolis n. sp. Granite rock pool on top of Boyagin Rock, between Brookton and
Pingelly, W.A. A, B, D, E, G-J, L-N drawn from paratype adult male; remainder from holotype adult
female. Scales: 1 = 100/* for A, B, E-N; 2 = 50/* for C, D. A, anlennula. B, antenna. C, rake-like organ. D,
rake-like organ. E, maxillula —palp and lobes. F, maxilla —endopodite. G, maxilla —endopodite. H,
maxilla—endopodite. I, mandible —palp. J. thoracopoda I. K, furca. L, maxilla-protopodite. M,

hemipenis. N, thoracopoda II.

56

P. DE DECKKER

hook-shaped and about 0.33 of the length of the other.

(Hemipenis) —(Fig. 8M) Lateral lobe broad and rec¬
tangular; inner lobe broad, subrectangular but about 0.8
of the width of the lateral lobe and almost same length;
near the base of the lateral lobe on the inside, small
knob-like protuberance.

(Zenker organ) —More than 30 rosettes.

(Furca) —(Fig. 8K) Claws almost equal with pectinate
and thick posterior seta half the length of the posterior
claw and 0.66 longer than the pectinate and narrow
anterior seta.

(Furcal attachment) —Median branch long, divided
distally and with a broad, but short, spike at right angle
near its proximal end ventrally.

Colour of Shell: White to transparent ventrally and
bluish green dorsally.

Size:

L

H

L

H

holotype

adult

female

LV

2 (XXV

920/a

RV

2 020/a

960/a

paratype
adult male

LV

1 540 /a

720/a

RV

1 550/a

_

Type Locality: Granite rock pool on top of Boyagin
Rock, between Brookton and Pingelly, W.A.

Derivation of Name: From Latin amplus ( = large) and
colis (= penis) for the unusually large penis.

Ecology and Distribution: This species has been col¬
lected in the following localities: granite rock pools in
Sullivan Rocks, 11 km south of Gleneagle, W.A. (or 63
km south of Perth on Albany Highway); roadside ditch
north of Scadden, W.A. (56 km north of Esperance on
road to Norseman). /. amplicolis appears to be restricted
to fresh, temporary pools.

Remarks: I amplicolis differs from /. varrovillius (King
1855) and /. stanleyanus (King 1855), which have similar
shell outlines, by the absence of long natatory setae on
its antennae (in specimens of these two species examined
in Sars’ collection, the natatory setae extend past the tip
of the antennal claws). No males have been found in the
latter two species.

Uyodromus candonites n. sp.

Figs 9a-k, 10

Diagnosis: Ilyociromus w-ith subrectangular shell in
lateral view, with posterior broadly rounded and
anterior tapering; valves faintly striated; inner lamella
anteriorly almost three times the width of the posterior
in both valves; faint selvage peripheral in the right valve
and broader, 0.33 of width from the outer margin on the
in ner lamella posteriorly and ventrally; natatory setae
of antenna atrophied; maxilla palps in male similar,
hook-shaped and angular; lateral lobe of hemipenis
digitate and broadest distally.

Description: carapace (External) — Subrectangular in
lateral outline with posterior broadly rounded and
almost forming a right angle with the dorsum which is
almost flat; anterior tapering but rounded and
anterodosal area inclined; ventrum almost flat except in
the mouth region which is slighly concave 0.4 of length

from the anterior; surface of shell faintly striated with
two generations of striae (Fig. 9k; in dorsal view, like a
flattened ellipsoid with both ends pointed; simple nor¬
mal pore canals scattered with broad rim.

(Internal) —Inner lamella anteriorly almost three
times the width of the posterior in both valves; selvage
peripheral and faint in the right valve and broader and
0.33 from the outer margin on the inner lamella
posteriorly and ventrally; anteriorly the inner lamella is
faintly reticulated like all Ilyodromus species.
anatomy (Antennula) —(Fig. 10A) Seven-segmented:
length width ratio of last six segments: 2:3, 1.8:1, 1.2:1,
1.4:1, 1.8:1, 1.3:1; sensory organ on second segment
3-segmented and short; natatory setae as long as all
segments together.

(Antenna)—(Fig. 10B) Three claws on penultimate
segment and a fourth one on the distal one; natatory
setae extremely short except for the outer one which is as
long as half the length of the penultimate segment.

(Mandible) —(Fig. 10G) Mandibular coxale with seven
teeth; palp 3-segmented and with a bristle stylet-like, 0
bristle stout, pointed and densely pilose, y bristle
slightly longer than distal segment, stout and densely
pilose in the distal two thirds; epipod with five long
plumose Strahlen plus a shorter one at mid-length and a
smaller barren seta near its base.

(Rake-like organ) —Seven to nine teeth plus a bifid
one on the inner side of each rake.

(Maxillula) —(Fig. 10C) Distal part short and
trapezoidal; third lobe with two smooth Zahnborsten;
epipod plate with 22 Strahlen.

(Maxilla) —Sexually dimorphic: in male (Fig. 10D, E)
palps similar, narrow-, angular and hook-shaped; in
female (Fig. 10F) three plumose setae w r ith middle one
twice the length of the other two which are of almost
equal length; for chaetotaxy of protopod, see Fig. 10F,
(Thoracopoda I) —(Fig. 101) 3rd segment divided;
distal seta of 2nd segment as long as half of the length of
the 3rd segment; outer seta on 4th segment 0.25 of the
length of the distal claw.

(Thoracopoda II) —(Fig. 10K) Three-segmented;
distal setae unequal with shorter one curved and about
0.33 of the length of the other; distal pincers small.

(Hemipenis) —(Fig. 10H) Lateral lobe digitate with
distal end broadest; inner lobe bilobate distally and
curved inward.

(Zenker organ) —With about 27 rosettes.

(Furca) —(Fig. 10J) Both claws of almost equal
length; posterior seta thick; pectinate and half the length
of the posterior claw; slim anterior seta barren and
about half length of the other seta.

(Furcal attachment) —median branch long, divided
distally and w r ith a broad, short and curved spike at right
angle near its proximal end ventrally.

Colour of Shell: Green.

Size: L H L H

holotype

adult male LV 1 140/x 600 /a RV 1 140 /a 600/a
Type Locality: Small granite rock pool at summit of
Mt. Chudalup, near Northclifle, W.A.

OSTRACODA FROM INLAND AUSTRALIA

57

Fig. 9-a-k Uyodromus candonites n. sp. a-e, k Small granite rock pool on Muirillup Rock, near Nor-
thcliffe, W. A. f-j Small granite rock pool at summit of Mt Chudalup, near Northcliffe, VV.A. Scales:
1 = 500/* for a-j; 2- 10/* for k. a-e, k females; f, i-j male paratypes; g, h male holotype. a, LV internal, b,
RV internal, c, C dorsal, d, C showing LV. e, C showing RV. f, C showing RV. g, LV internal, h, RV in¬
ternal. i, RV external, j, LV external, k, C showing LV, detail of d. 1-r, Uyodromus amp/icolis n. sp.
Granite rock pool on top of Boyagin Rock, between Brookton and Pingelly, W.A. Scale: 1 = 500/* I, m
holotype female; n, o paratype male; p, r paratype female. 1, RV internal, m, LV internal, n, LV internal,
o, RV internal, specimen distorted, p, C dorsal, q, C showing RV, same specimen as p. r, C showing LV,

specimen distorted.

P. DE DECKKER

Fig. 10 — Ilyodromus candonites n. sp. Granite rock pool on top of Boyagin Rock, between Brookton and
Pingelly, W.A. A-E, H-K drawn from holotype adult male, F, G from paratype adult female. Scale:
= 100/x. A, antennula. B, antenna. C, maxillula—palp and lobes. D, maxilla —endopodite. E, max¬
illa— endopodite. F, maxilla. G, mandible—palp. H, hemipenis. I, thoracopoda I. J, furca. K,

thoracopoda II.

OSTRACODA FROM INLAND AUSTRALIA

59

Fig. 11 —Uyodromus dikrus n. sp. Dam at Wasley Well, near Nallan, W.A. Scales: 1 =500/* for a-l;
2 = 100/* for m, = 20/* for n, = 200/* for o. a-d, j, 1, o female paratypes; e, f, i-k male paratypes; g, h, m, n
male holotype. a, LV internal, b, RV internal, c, LV external, d, RV external, e, RV internal, f, LV inter¬
nal. g, LV external, h, RV external, i, C dorsal, j, C ventral, k, C showing LV, penis and some appen¬
dages. I, C showing RV. in, LV external, posterior detail of g. n, LV external, detail of g. o, C showing

RV, anterior detail of 1.

P. DE DECKKER

Fig. 12 — Ilyodromus dikrus n. sp. Dam at Wasley Well, near Nallan, W.A. A, B, D, E, G, H, K, L, N
drawn from holotype adult male, the remainder from paratype adult female. Scale: 1 = 100/x. A, anten-
nula. B, antenna. C, maxillula-palp and lobes. D, thoracopoda 1. E, hemipenis. F, maxilla. G, max¬
illa-right endopodile. H, maxilla—left endopodite. I, mandible-palp. J, mandible—coxale. K, Zenker
organ. L, thoracopoda II. M, furcal attachment. N, furca.

OSTRACODA FROM INLAND AUSTRALIA

61

Derivation of Name: From the genus Candona plus the
Greek suffix —ties (= like) as the lateral profile of this
species is reminiscent of many species of Candona.
Ecology and Distribution: This species has only been
collected in Western Australia. It occurs in many tem¬
porary granite pools near Northcliffe —at and near sum¬
mit of Mt. Chudalup, and on and near Muirillup Rock.
The size of /. candonites is variable: the length of adult
specimens can vary between 1 100/* and 1 400/*.
Remarks: Ilyodromus candonites differs from /.
midulus specimens examined in Sars’ collection on the
following important details: the natatory setae of the
antenna almost reach the tip of the claws in the latter
species, and its shell is more elongated: it is faintly con¬
cave dorsally in front of the hinge (/. candonites is flat)
and the selvage is near the inner margin posteriorly in
the left valve and is broader posteriorly in the right
valve. The greatest extension of the shell posteriorly in /.
midulus is at mid-height whereas it is near the ventrum
in /. candonites. The latter species differs from type
specimens of I. substriatus Sars 1894 and I. obtusus Sars
1894 from Sars’ collection (which have short natatory
setae on the antenna extending to the middle of the
penultimate segment), on the following features of the
shell: I substriatus has a broad selvage posteriorly in the
right valve which is met by a depression in the left valve
where the selvage is faint and along the periphery of the
inner margin; in I. obtusus the selvage is faint and along
the outer margin in both valves. No males are known for
Sars’ species.

Ilyodromus dikrus n. sp.

Figs II, 12

Diagnosis: Member of Ilyodromus like an inclined
parallelogram with rounded ends in lateral view; ob¬
vious depression anterior to the hinge dorsally; with the
greatest extension of the shell anteriorly 0.4 from the
dorsum plane; inner lamella broad throughout in both
valves; male maxilla palps asymmetrical, the narrower
one being more arched; outer seta of 4th segment
thoracopoda more than half the length of the distal
claw; hemipenis with digitate lateral lobe and inner lobe
like an elongated rectangle reaching almost the tip of the
lateral lobe.

Description: carapace (External) —Inclined parallelo¬
gram with rounded ends in lateral view, with obvious
depression anterior to the hinge dorsally; shell with
longitudinal striations of two generations (Fig. 11 n) all
over except in the anterior area near the margin; simple
type normal pores; greatest extension of the shell
anteriorly at 0.4 from dorsum plane and 0.6 posteriorly;
ventrum concave just before mid-length. In dorsal view
extremely narrow' and with both ends pointed.

(Internal) —Inner lamella similar in both valves and of
similar width anteriorly and posteriorly: it is broadest
anteriorly where the valve extends the furthest, and nar¬
rowest in the mouth region above the concavity.
anatomy. (Antennula) —(Fig. 12A) Seven-segmented:
length width ratio of last six segments: 1:1, 1.8:1, 1:1,

1.3:1, 1.7:1, 2.5:1; natatory setae as long as last six
segments, sensory organ on second segment elongate.

(Antenna) —(Fig. 12B) Three distal claws on the
penultimate segment with a shorter one on the distal seg¬
ment; natatory setae extending much further than the
tip of the claws.

(Mandible) —(Fig. 121, J) Mandibular coxale with
seven teeth; palp with a bristle stylet-like, (3 bristle stout
and densely pilose, y bristle broad, almost twice the
length of the distal segment and pilose in the distal half;
epipod plate with four pilose Strahlen.

(Rake-like organ) —Seven to nine teeth, plus one bifid
tooth on inner side of each rake.

(Maxillula) —(Fig. 12C) Distal segment of palp
trapezoidal and third lobe with two smooth
Zahnborsten; epipod with about 18 plumose Strahlen.

(Maxilla)—Sexually dimorphic: male (Fig. 12G, H)
palps asymmetrical with the narrower more strongly ar¬
ched; the other is broadest at mid-length; female (Fig.
12F) palp W'ith three short plumose setae, the middle one
almost twice the length of the other two which are of
similar length; for chaetotaxy of protopod see Fig. 12F.

(Thoracopoda I) —(Fig 12D) Seta at mid-length on
outer side of fourth segment thick and more than half
the length of the distal claw; proximal seta on first seg¬
ment 0.33 of the length of the distal one.

(Thoracopoda II) —(Fig. 12L) Three-segmented;
distal pincers small and distal setae unequal: longest seta
1.6 times the length of the shorter and slightly curved
one.

(Hemipenis) —(Fig. 12E) Lateral lobe digitate and in¬
ner lobe like an elongated rectangle reaching almost the
tip of the lateral lobe; the broad tip of the inner lobe is
covered with small hooks.

(Zenker organ) —(Fig. 12K) Elongate, with 25 roset¬
tes.

(Furca) —(Fig. 12N) Claws almost equal; posterior
seta slim, pectinate, twice the length of the other barren
seta and 0.66 of the length of the posterior claw.

(Furcal attachment) —(Fig. 12M) Median branch
thick, bifurcate distally and with broad spike at right
angle near its base.

Colour of Shell:

White.

Size:

L

H

L

H

holotype
adult male

LV

1 270/i

560/*

RV

1 270/*

560/*

paratype

adult

female

LV

1 470/*

660/*

RV

1 470/*

660/*

Type Locality: Dam at Wasley Well, near Nallan, 21
km NNE of Cue, W.A. (27°16'54 ,; S, 118°09'06"E).
Derivation of Name: From Creek dikros ( = forked)
for the forked appearance of the distal end of the
thoracopoda I which has a long outer seta on the last
segment.

Ecology: this species has only been collected once from
the type locality: water was fresh and turbid.

Remarks: Although this species appears at first glance
to resemble the elongated I. varrovillius (King 1855), it is
easily separated from the latter by its long seta on the

62

P. DE DECKKER

Fig. \3 — Ampullacypris oblongata (Sars 1896). Roadside pool, on Gibb River Road, 58 km E. of Derby,
W. A. Scale: = 500/z. a, b, h, j-m females, c-g, i, n males, a, C showing LV. b, C showing RV. c, C show¬
ing LV. d, RV internal, e, LV internal, f, LV external, g, RV external, h, C dorsal, i, C dorsal, j, RV in¬
ternal. k, LV internal. 1, LV external, m, RV external, n, C ventral.

OSTRACODA FROM INLAND AUSTRALIA

63

last segment of the thoracopoda I and by its inclined
parallelogram outline in lateral view. A varrovi/lius in
Sars’ collection has short natatory setae on the antenna.

Ilyodromus williamsi (McKenzie 1966)

1966 Isocypris williamsi McKenzie, p. 266.

Remarks: The transfer of this species to Ilyodromus is
suggested here because this species possesses many
typical anatomical features of that genus. These are:
3-segmented sensory organ on the 2nd segment of the
antennula; slim stylet-like a bristle; stout, pointed and
densely pilose 0 bristle; thick, stout y bristle which is
pilose in its distal half; trapezoid palp and smooth
Zahnborsten on maxillula; presence of two setae on 1st
segment of thoracopoda I; thick and pectinate posterior
seta on furca and furcal attachment with stout spike
forming a right angle with the median branch near its
base. All these were seen on the holotype.

Although A williamsi has a smooth shell (when ex¬
amined under a binocular microscope), contrary to most
Ilyodromus species, it is still included in that genus for
the reasons given above. It is worth noting, however,
that striations on the shell of many Ilyodromus
specimens, all belonging to the one species and collected
together, can vary; on some specimens of A viridulus ,
for example, striations are only visible anteriorly and
posteriorly, on others the shell is smooth, and others the
shell is finely striated.

McKenzie (1971a) has already pointed out that A
williamsi was not an Isocypris sensu stricto on shell
characters alone. This species in fact is closely related to
I. dikrus as they both have a similar shell outline but A
williamsi has a faint selvage at a distance from the outer
lamella anteriorly in the left valve and has a very short
outer seta on the distal segment of the thoracopoda I.

A williamsi is only known from the type locality,
about 16 km west of Inverway, N.T.

Genus Ampullacypris n. gen.

Type Species: Ampullacypris oblongata (Sars 1896).
Diagnosis: Smooth ellipsoidal shell with normal pore
canals and flattened when viewed dorsally; inner lamella
broad anteriorly and posteriorly in both valves; central
muscle field consisting of a row of three in front and two
behind plus a hollow inclusion above and in front of the
upper adductor scar; two mandibular scars in front and
below the adductor scars; two toothed Zahnborsten and
rectangular palp on maxillula; mandibular palp with a
bristle smooth and slim, 0 bristle longer (but not stout)
and densely pilose, y bristle longer than last segment,
stout and pilose on its distal half; maxilla palps on male
asymmetrical; 1 st segment of thoracopoda I with one
long seta, furcal shaft smooth and posterior seta on fur¬
ca thick and pectinate; furcal attachment with median
branch long and no distal ornament plus dorsal branch
forming narrow elongated loop.

Derivation of Name: from Latin ampulla meaning
flask (as the type species has been described from
specimens originally grown in an aquarium by G. O.
Sars) and the generic name Cypris.

Remarks: Ampullacypris n. gen. is closely related to
Psych rodrom us Danielopol & McKenzie 1977 and Il¬
yodromus Sars 1894. It differs from these two genera on
the following important anatomical feature: the distal
end of the furcal attachment does not have a wedge
shaped spike and the dorsal branch forms a loop. Am¬
pullacypris differs from Psychrodromus by possessing a
smooth furcal shaft and from Ilyodromus by its two¬
toothed Zahnborsten and a rectangular palp on the
maxillula. The a, (3, and y bristles on the maxilla of Am¬
pullacypris are like those of Psychrodromus as is the
two-segmented, short sensory organ on the 2 nd segment
of the antennula.

Ampullacypris oblongata (Sars 1896)

Figs 13, 14

1896 Cypris oblongata Sars, p. 29.

1901 ? Amphicypris oblongata ; Sars, p. 18.

1923 Amphicypris oblongata; Henry, p. 268.

Diagnosis: Smooth ellipsoidal shell with posterior nar¬
rower than anterior and ventral area almost flat; in dor¬
sal view, shell narrow and greatest width at about a third
from the anterior. Inner lamella broad anteriorly and
posteriorly in both valves. Lateral lobe of hemipenis
crescent-shaped; Zenker organ with 42 rosettes.
Description: carapace (External)-smooth ellipsoidal
shell with posterior narrower than anterior and ventral
area almost flat except in the mouth area in the middle
where it is faintly concave. Valves similar with left one
slightly longer and overlapping the other slightly ven-
trally. In dorsal view, shell narrow and greatest height at
about a third from the anterior. Shell hirsute pos¬
teriorly.

(Internal) —Inner lamella broad anteriorly and
posteriorly in both valves and selvage faint and
peripheral except in the right valve ventrally; thin flange
along the periphery of the right valve. Marginal pore
canals common, short and straight. Central muscle field
consisting of a row of 3 scars in front with the central
one the smallest; two scars are situated behind the front
row and are at the level of the two lower scars; two man¬
dibular scars in front and below the adductor scars and a
hollow inclusion above and in front of the upper adduc¬
tor scar.

anatomy: (Antennula) —(Fig. 14B) Seven-segmented,
length width ratio of the last six segments: 1 : 1 , 2 . 2 : 1 ,
5:3, 8.5:5, 7:4, 3:1; 2nd segment with two segmented,
short sensory organ.

(Antenna) —(Fig. 14C) Natatory setae reaching the tip
of the claw's; 3 claws of equal length on penultimate seg¬
ment and reaching the tip of the other claw on the last
segment.

(Mandible) —(Fig. 14A, G) Palp with a bristle smooth
and slim, /3 bristle longer (but not stout) and densely
pilose, 7 bristle longer than last segment, stout and
pilose on its distal half.

(Rake-like organ) —(Fig. 14D) Seven teeth plus one
inner bifid tooth.

Maxillula) —(Fig. 14F) Distal palp rectangular and
two toothed Zahnborsten on the third lobe.

64

P. DE DECKKER

Fig. 14 _ a mpullacypris oblongata (Sars 1896). Roadside pool, Gibb River Road, 58 km E. of Derby,
W.A. A-D, G-N drawn from adult male, E, F, O from adult female. Scales: 1 = 100/x for A-C, E-O;
2 = 5Qn for D. A, mandible — eoxale. B, antennula. C, antenna. D, rake-like organ. E, maxilla -
endopodite. F, maxillula— palp and lobes. G, mandible-palp. H, thoracopoda I. I, maxilla-left
endopodite. K, right maxilla. L, thoracopoda II. M, Zenker organ. N, hemipenis. O, furca. P, furcal

attachment.

OSTRACODA FROM INLAND AUSTRALIA

65

Fig. 15 — Heterocvpris vatic/ n. sp. Hexham Swamp, Newcastle, N.S.VV. Scales: 1 = 500/t for a-d, f-m,
= 200/i for e, o, =40/i for n; 2= 100/t for p-r. a-e, j, 1, o, p female paratypes; f, g, k, m, n, q, r male
paratypes; h-i male holotype. a, LV external, b, RV external, c, LV internal, d, RV internal, e, RV inter¬
nal, anterior detail of d. f, LV external, g, RV external, h, LV internal, i, RV internal, j, C dorsal,
k, C ventral. 1, C dorsal, m, C ventral, n, RV external, anterior detail of g. o, RV internal, posterior
detail of d. p, C dorsal, anterior detail of j. q, C ventral, posterior detail of k. r, C ventral, anterior detail

of k.

66

P. DE DECKKER

(Maxilla) —Sexually dimorphic: male (Fig. 141, J)
palps asymmetrical: right one broad and triangular with
outer side forming a rounded right angle and with two
long bristles near the base of the palp; other smaller,
narrower and more arched plus, at the base of the palp,
with two shorter and also smooth bristles. Female (Fig.
14E) palp with three unequal plumose setae, the middle
one being longer than the other two together.

(Thoracopoda I) —(Fig. 14H) Proximal end of 1st seg¬
ment with a long seta only; penultimate segment
divided; seta at the inner distal end of the last 3 segments
and at the division of the penultimate segment; last seg¬
ment with an outer seta, as long as inner one, near the
distal end.

(Thoracopoda II) —(Fig. 14K) 3-segmented; distal
seta of second segment longer than half the length of the
third segment; distal setae unequal with shorter one
curved and about one-third the length of the other
straight one; distal pincers small.

(Hemipenis) — (Fig. 14M) Broadly ellipsoidal in shape
with lateral lobe crescent-shaped.

(Zenker organ) —(Fig. 14L) Narrow long and with 42
rosettes.

(Furca) —(Fig. 14N) Distal claws unequal; posterior
seta pectinate, much thicker and longer than anterior
one and half the length of the posterior claw.

(Furcal attachment) —(Fig. 14 O) Median branch long
and slightly curved with dorsal branch like a narrow,
elongate loop normal to the median branch and forming
an obtuse angle with the straight ventral branch.
Colour of Shell: Beige brown.

Size: L H

From Sars* (1896) female: carapace 1 900/* 800/t

male : carapace 1 600/* —

Specimens examined here:

adult female : carapace 1 840/* 880/*

adult male : carapace 1 520/* 760/*

Ecology and Distribution: This species was raised by
Sars (1869b) from a dry sample of sand collected 64 km
east of Roebuck Bay in W.A. In Sars’ collection, there
are a number of samples of A. oblongata for which the
given locality is central Australia and others for which
females and one male, all undissected and preserved in a
hardened polyvinyl alcohol slide (Oslo Museum —Sars’
collection No. 11 600) arc syntypes. The male is
designated here as lectotype. In addition, the vial con¬
taining about 10 specimens of A. oblongata in the “old
spirit collection of Sars” held in the Oslo Museum,
under the number of 53.3/2 are also syntypes. The
specimens described here have been collected in a road¬
side pool on the Gibb River road, 58 km east of Derby,
W.A. The specimens labelled “ Eucypris ” cf. oblongata
(Sars, 1896) by McKenzie (1966) do not belong to the
species described here as one of the specimens studies by
him has peripheral tubercles on the right valve.

Subfamily Cyprinotinae Bronstein 1947
Genus Heterocypris Claus 1892

Type Species: Heterocypris incongruens (Ramdohr
1808).

Heterocypris vatia n. sp.

Figs 15, 16

Diagnosis: Member of Heterocypris with anterior edge
of right valve bent outward; lateral lobe of hemipenis
boot-shaped with “sole” of the boot convex; inner lobe
of hemipenis with scattered minute hooks.

Description: carapace (External) —Bean-shaped in
lateral view with dorsum curved; greatest height at 0.4 to
0.5 from the anterior; posterior slightly more broadly
arched than anterior; ventral area nearly flat except in
the mouth region which is faintly concave; in dorsal
view like a flattened ellipsoid with both ends pointed;
anteriorly the extremity is bent slightly clockwise;
anteriorly and posteroventrally the right valve bends
outward along the edge; the left valve bends inward to
meet the right valve all along its periphery except ven-
trally where it overlaps the other; shell pseudopunctate
with numerous rounded wart-like tubercles on the
anterior of the shell; a hair protrudes from each
tubercle.

(Internal) — Right valve faintly tuberculate all along its
periphery except dorsally; inner lamella broadest
anteriorly in both valves; in right valve, selvage broad
following the curvature of the shell halfway between the
outer and inner margins anteriorly, whereas it is near the
inner margin posteroventrally; the inner lamella between
the outer margin and the selvage is convex anteriorly
and posteroventrally; in left valve, selvage faint and
peripheral and presence of narrow flange all along;
radial pore canals numerous and straight.
anatomy (Antennula) —(Fig. 16A) Seven-segmented;
length width ratio of the last six segments: 1:1, 2.25:1,
1.5:1, 1.5:1, 1.7:1, 2.5:1; small, rod-shaped, sensory
organ at mid-length on the second segment; natatory
setae slightly longer than all segments together.

(Antenna) —(Fig. 16C) Sexually dimorphic: in female
the claw attached to the small 3rd segment is narrower
and smaller.

(Mandible) —(Fig. 16E, F) Mandibular coxale with
seven teeth (Fig. 16F); inner tooth longer than the
previous two and pointed and near its base two setae,
one of which is pilose; endopod (Fig. 16E) with a bristle
short and narrow, (3 bristle of same length, wrinkled and
covered with a few short hairs, y bristle longer than last
segment, stout and thickly pilose externally in its distal
half; epipod with five plumose Strahlen plus a shorter
one half way and a short, stout and pilose seta at its
base.

(Rake-like organ) —With seven teeth and inner one
bifid.

(Maxillula) —(Fig. 16D) Endopod with about 17
plumose Strahlen; length width ratio of palps: 3:1, 2:1;
third lobe with two toothed Zahnborsten and near their
base presence of a short and thick tufted bristle.

(Maxilla) —Sexually dimorphic: male palps strongly
asymmetrical (Fig. 161, J) with left one narrower and
strongly arched; female palp with three plumose setae,

OSTRACODA FROM INLAND AUSTRALIA

67

Fig. 16 -Heterocypris vatia n. sp. Hexham Swamp, Newcastle, N.S.W. A-E, G, I-M drawn from
holotype adult male, remainder from paratypc adult female. Scale: =200/*. A, antennula. B, rake-like
organ. C, antenna. D, maxillula-palp and lobes. E, mandible-palp. F, mandible-coxale. G,
thoracopoda I. H, maxilla-endopodite. I, left maxilla. J, maxilla-right endopodite. K, thoracopoda
ILL, hemipenis. M, furca. N, furcal attachment.

68

P. DE DECKKER

Fig. 17 —Eucypris virens (Jurine 1820). Pond close to Reel Inlet (coastside), 19 km S. of Mandurah, W.A.
Scales: 1 =500/* for a-g, i-j; 2 = 40/* for h, =20/* for k, =30/* for 1. All females, a, LV internal, b, RV in¬
ternal. c, LV internal, d, RV internal, e, C dorsal, f, C showing RV. g, C showing LV. h, C dorsal,
anterior LV detail of j. i, C ventral, j, C dorsal, k, C showing RV, posterior detail off. 1, C showing RV,

anterior detail of f.

OSTRACODA FROM INLAND AUSTRALIA

69

the two outside ones being of similar length; for
chaetotaxy, see Fig. 161.

(Thoracopoda I) —(Fig. 16G) Penultimate segment
weakly divided and distal claw 1.2 times the length of
the last two segments together.

(Thoracopoda 11) — (Fig. 16K) Distal setae on last seg¬
ment unequal: the longer one four times the length of
the other which is hook-shaped.

(Hemipenis) — (Fig. 16L) Outer lateral lobe boot¬
shaped with “sole” part of the boot convex and “heel”
part slightly angular and forming an obtuse angle; inner
lobe broadly rectangular and covered with numerous
short hooks.

(Zenker organ) —Both ends rounded with 48 rosettes.

(Furca) — (Fig. 16M) Setae almost equal with posterior
one finely pectinate; claws unequal; anterior one 1.6
times the length of the other.

(Furcal attachment) —(Fig. 16N) Median branch
straight with short dorsal branch normal to it; ventral
branch curved and 2.5 times the length of the dorsal
one.

(Eye) —Cups of nauplius eye fused; brown in colour.
Colour of Shell: Translucent pale brown.

Size: L H L H

holotype

adult male LV 1 700/* 1 000/* RV 1 710/4 1 000/*

paratype

adult

female LV 2 400/* 1 400/* RV 2 340/* 1 360/*

Type Locality: Hexham Swamp, behind the University
campus at Newcastle, New South Wales.

Derivation of Name: From Latin vatius meaning bent
outward for the diagnostic feature of the right valve.
Ecology and Distribution: This species has only been
collected once; water at the type locality is known to be
fresh.

Remarks: It was thought that this species belonged to
H. leana (Sars 1896) because of its large size. The female
specimens described by Sars (1896a) were 2.70 mm long
and came from Hay, N.S.W. However, after examina¬
tion of Sars’ collections in the Oslo Museum, it became
obvious that none of the male specimens labelled H.
leana by Sars have the same outline of the lateral lobe as
the specimens from Hexham Swamp; all Sars’ specimens
have a small and pointed protuberance in the “heel”
Part, the boot-shaped lateral lobe of the hemipenis. This
feature is not seen in H. vatia. However, no specimen
from Hay was found in Sars’ collection; only specimens
which are labelled as “Victoria” A or C are found.
Therefore, designation of a lectotype will prove to be
difficult. However, a 2.4 mm long male specimen col¬
lected from Goulburn Billabong, Alexandra, Vic. by R.
Shiel corresponds to Sars’ description of H. leana and
possesses the pointed “heel” on the lateral lobe of the
hemipenis. This substantiates the separation of the two
taxa into different species which have a large shell but
different anatomy. H. vatia differs from all other
Heterocypris species recently reviewed in Victor and
Fernando (1980).

Subfamily Eucypridinae Bronstein 1947
Genus Eucypris Vavra 1891
Type Species: Eucypris virens (Jurine 1820)

Eucypris virens (Jurine 1820)

Figs 17, 18

1820 Monoculus virens Jurine, p. 171.

1900 Eucypris virens ; Daday, p. 143.

Diagnosis: Subrectangular shell with dorsum arched
and greatest height in the middle; length height ratio of
carapace: 1.45 to 1.65; shell convex ventrally just in
front of the slightly concave mouth region; in dorsal
view oval in shape with anterior more pointed than
posterior; wart-like protuberances (Fig. 17h, k, 1) near
the outer margin anteriorly best seen in dorsal view. Col¬
our of shell: pale green.

Remarks: E. virens is a cosmopolitan species well
known outside Australia; description of the shell and
anatomy is therefore unnecessary but illustrations are
provided in Figs 17, 18. This species has already been
recorded from New Zealand (Barclay 1968, Chapman &
Lewis 1976). In Australia, it is a common inhabitant of
temporary pools and is usually found in fresh waters but
has been recorded in slightly saline water; the highest
salinity record for E. virens is 4.4°/ 00 in a Western
Australian locality (Geddes et al. 1981). So far E. virens
has been collected in southern Australia (W.A., S.A.,
Vic.).

Variations in the outline of E. virens have been com¬
monly noted, even on specimens collected in the same
locality. These variations are illustrated in Fig. 17. They
are best seen in lateral view and correspond to variations
in shell outline already noted by Muller (1900) who
designated the following variations: E. virens var.
acuminata which has a more elongated shell (see Fig.
17a, b, 0 and E. virens var. obtusa which has a more
compressed shell and more broadly curved outline
posteriorly (see Fig. 17 c, d, g). These variations
may be ecologically significant but remain as yet
unexplained.

It is likely that Eucypris pratensis Eagar 1970, record¬
ed only from three localities near Wellington in New
Zealand (Eagar 1970), is also a variant of Eucypris
virens .

Eucypris virens in Australia is a parthenogenic species
although both sexes have been recorded in other parts of
the world (North Africa (Gauthier 1928); pond in the
delta of the River Don, USSR —material received from
Dr. E. I. Shornikov).

Subfamily Cypricercinae McKenzie 1971
Genus Cypricercus Sars 1895
Type Species: Cypricercus cuneatus Sars 1895.

Cypricercus salinus n. sp.

Figs 19 a-1, 20

Diagnosis: Smooth, triangular shell, elongated ellipsoid

70

P. DE DECKKER

Fig. 18 — Eucypris virens (Jurine 1820). Pond very close to Reel Inlet (coastside), 19 km S. of Mandurah,
W.A. Scale: 1 = 100jt. Drawn from adult female. A, antennula. B, antenna. C, thoracopoda I. D, max¬
illa— protopodite. E, maxillula — palp and lobes. F, maxilla —cndopodite. G, mandible — palp. H, furca.

1, furcal attachment. J, thoracopoda II.

OSTRACODA FROM INLAND AUSTRALIA

71

Fig. 19 — a-1, Cypricercus salinus n. sp. Small lake N. of Lake Terangpom, Vic. Scale: =200/*. a, b male
holotype; c-e, i, j male paratypes; f-h, k, 1 female paratypcs. a, RV internal, b, LV internal, c, LV exter¬
nal. d, RV external, c, C showing LV. f, LV internal, g, RV external, h, LV external, i, C dorsal, j, C ven¬
tral. k, C dorsal. 1, C anterior, m-q, Cypretta baylyi McKenzie 1966. Granite rock pool on top of Boyagin
Rock, between Brookton and Pingclly, W.A. Scale: = 200/* for m-p, =40/* for q. m, n male; o-q
unknown sex. m, RV internal, n, LV internal, o, LV external, p, C dorsal, q, LV external, detail of o
e (central muscle scar area).

72

P. DE DECKKER

in dorsal view; distal end of lateral lobe of hemipenis
broader than its base.

Description: carapace (External)-Triangular in
lateral view with greatest height at about middle;
anterior and posterior ends broadly rounded; anterodor-
sal area almost straight whereas posterodorsally it is
slightly arched; ventral area almost flat except in the
mouth region in the middle where it is concave. Left
valve slightly larger anteriorly and overlapping ven-
trally, especially in the mouth region. In dorsal view,
like a flattened ellipsoid.

(Internal)-Inner lamella twice as broad anteriorly in
both valves compared to posteroventral area; selvage
near the rounded outer margin in right valve anteriorly,
and at about 0.33 of the width of the inner lamella from
the outer margin posteroventrally; in left valve, selvage
bordering the rounded outer margin antero- and
posteroventrally and separated from it by a broad
groove all along; ventrallv it is distant from the outer
margin.

anatomy (Antennula) — (Fig. 20A) Length width ratio
of the last six segments: 1:1.4, 1.7:1, 1.2:1, 1.2:1, 1:6.5,
2.7:1; natatory setae slightly longer than all segments
together.

(Antenna) —(Fig. 20B) Natatory setae much longer
than the last three segments and claws together; four ter¬
minal claws in both sexes.

(Mandible) —(Fig. 20C) Endopod with a and 7
bristles long, narrow and smooth; (3 bristle short, stout
and pilose; mandibular coxale with seven teeth: inner
one slightly longer than the two adjacent.

(Rake-like organ) —Seven teeth plus one bifid on the
inner side of each rake.

(Maxillula) —(Fig. 20E) third palp with two smooth
Zahnborsten; length width ratio of palps 4.5:1, 2.5:1.

(Maxilla) —Sexually dimorphic: in male (Fig. 20 H, I)
distal palps unequal with the narrow one strongly arched
and forming a right angle; female (Fig. 20 D) endopod
with three unequal setae. For chaetotaxy of epipod, see
Fig. 20 D, H.

(Thoracopoda I) —(Fig. 20F) Third segment divided
at mid-length where an inner seta as long as the distal
half of the third segment and the fourth one together oc¬
curs.

(Thoracopoda II) — (Fig. 20L) Distal setae unequal
with small one hook-shaped; seta at mid-length of the
last segment shorter than hook-shaped distal seta; small
distal pincers present.

(Hemipenis) —(Fig. 20G, K) Outline of copulatory
sheath triangular with greatest -length on inner side;
lateral lobe arched inward and distal end broader than
its base.

(Zenker organ) —(Fig. 20M) Elongate with 13 broad
rosettes.

(Furca) —(Fig. 20J) Anterior claw 1.6 times longer
than posterior one; anterior seta slightly longer than
posterior one.

(Furcal attachment)-(Fig. 20N) Ventral branch
almost as long as median one and hook-shaped distally;
dorsal branch short with a spike at each end on the distal

side to form an almost closed eyelet which is typical of
the genus as illustated for the type species C. cuneatus
by McKenzie (1977). (In the latter species, the eyelet is
closed).

(Eye) —Cups of nauplius eye fused; dark brown in
colour.

Colour of Shell: Pale Green.

Size: L H L H

holotype adult

male LV 600/a 380 /a RV 580/a 360 /a

paratype adult

female LV 640/a 390/a RV 620/a 370/a

Type Locality: Small lake (38°06'06"S, 143°18'47"E)
north of Lake Terangpom, west of Lake Corangamite,
western Victoria.

Derivation of Name: From Latin salinus meaning
saline as this species has been collected in many slightly
saline waters.

Ecology and Distribution: C. salinus has been col¬
lected from lakes in the vicinity of Colac and Camper-
down, Victoria (e.g. Lakes Martin, Koreetnung, Kariah,
Wingeel, Terangpom). The salinity range of the species
is: 0. 34-12.3%o. Salinity of the water at the type locali¬
ty was 4.9°/oo.

Cypricercus unicornis n. sp.

Figs 21, 22

Diagnosis: Pseudopunctate, ellipsoid shell with asym¬
metrical valves: left valve longer especially posteriorly
where it often tapers to a broad and rounded spike; no
spike in the right valve; lateral lobe of hemipenis digitate
and straight.

Description: carapace (External) —Ellipsoid in lateral
view with greatest height at about middle in the right
valve; surface of shell deeply pseudopunctate except
dorsally and ventrally. Anterior broadly rounded, ven-
trum almost flat and posterior tapered. Valves asym¬
metrical: left valve, which overlaps the right one all
around, has protuberance posteriorly which extends
slightly outward. The extension of this protuberance is
variable: in some specimens, it is almost non-existent. In
dorsal view, like a flattened ellipsoid except in the
posterior area of the left valve where the protuberance
occurs. Normal pore canals of two types; some simple
and others simple with a broad rim.

(Internal) —Inner lamella twice as broad anteriorly
compared to posterior; selvage faint and peripheral in
the right valve whereas it is broad and follows the
periphery of the left valve where it is separated from the
rounded outer margin only by a narrow but deep
groove. The posterior protuberance in the left valve is
hollow. Radial pore canals numerous, narrow and
straight.

Anatomy (Antennula) —(Fig. 22A) Length width ratio
of the last six segments: 1:1, 2.6:1, 1.6:1, 2:1, 2.3:1,4:1;
natatory setae slightly longer than all segments together.

(Antenna) —(Fig. 22B) Four terminal claws in both
sexes: the claw attached to the small distal segment is

OSTRACODA FROM INLAND AUSTRALIA

73

Fig. 20 — Cypricercus salinus n. sp. Small lake N. of Lake Terangpom, Vic. A-C, F-N drawn from
holotype adult male, D, E from paratype adult female. Scale: = 100/x. A, antennula. B, antenna. C, man¬
dible. D, maxilla. E, maxillula — palp and lobes. F, thoracopoda I. G, hemipcnis. H, left maxilla. I, max¬
illa-right endopodite. J, furca. K, hemipenis. L, thoracopoda II. M, Zenker organ. N, furcal attach¬
ment.

74

P. DE DECKKER

Fig. 21 — Cypricercus unicornis n. sp. Granite pool, Newmann’s Rocks, 140 km E. of Norseman, W. A.
Scales: 1 = 500/i for a-n; 2 = 40^ for o, =20 n for p, =60/i for r; 3= 10ji for q. a, b, e, f, j-1, p female
paratypes; c, d, r male holotype; g-i, m-o, q male paratypes. a, RV internal, b, LV internal, c, RV inter¬
nal. d, LV internal, e, RV external, f, LV external, g, RV external, h, LV external, i, C showing LV. j, C
showing RV. k, C dorsal. 1, C ventral, m, C dorsal, n, C ventral, o, C ventral, posterior detail of n. p, C
ventral, anterior detail of 1. q, RV external, detail of g. r, LV internal, posterior detail of d.

OSTRACODA FROM INLAND AUSTRALIA

75

Fig. 22 — Cypricercus unicornis n. sp. Granite pool, Newmann’s Rocks, 140 km E. of Norseman, W.A.
A-G, I-M drawn from holotype adult male, H, N from paratype adult female. Scale: = 100/*. A, anten-
nula. B, antenna. C, mandible-palp. D, maxillula— palp and lobes. E, thoracopoda I. F, maxilla —left
endopodite. G, right maxilla. H, maxilla —endopodite. I, hemipenis. J, hemipenis. K, thoracopoda II. L,
furca. M, Zenker organ. N, furcal attachment.

76

P. DE DECKKER

strongly pectinate in male; natatory setae extend further
than the tip of the claws.

(Mandible) —(Fig. 22C) As for C. salinus .

(Rake-like organ) As for C. salinus.

(Maxillula) —(Fig. 22D) Third palp with two toothed
Zahnborsten; length width ratio of palps 5:1, 3:1.

(Maxilla)-Sexually dimorphic: in male (Figs. 22F, G)
distal palps unequal with the narrow one strongly arched
whereas the other is narrow at mid-length and only
slightly curved; female (Fig. 22H) endopod palp with
three unequal setae. For chaetotaxy see Fig. 22G.

(Thoracopoda I)-(Fig. 22E) Third segment divided
at mid-length with an inner seta slightly longer than half
of the third segment.

(Thoracopoda II) —(Fig. 22K) Distal setae unequal
with longest seta twice the length of the other; seta at
mid-length on last segment almost reaching the tip of the
terminal pincers and the distal seta on the penultimate
segment slightly longer than half of the last segment.

(Hemipenis) — (Fig. 221, J) Outline of copulatory
sheath triangular with greatest length on inner side;
lateral lobe digital and straight with distal end rounded.

(Zenker organ) —(Fig. 22M) Elongated with 17 broad
rosettes.

(Furca) — (Fig. 22L) Length of anterior claw over
posterior one 1.7; setae small and almost equal; furcal
shaft extremely long: 2.5 times the length of the anterior
claw.

(Furcal attachment) —(Fig. 22N) Median branch
broad and about three times the length of the ventral
branch; dorsal branch curved inward and with a basal
eyelet.

(Eye) —Cups of nauplius eye fused; brown in colour.
Colour of Shell: Green to pale green.

Size: L H L H

holotype adult

male LV 880/* 400/* RV 780/* 380/*

paratype adult

female LV 870/* 400/* RV 790/* 370/*

Type Locality: Granite pool, Newmann’s Rocks, 140
km east of Norseman, W.A.

Derivation of Name: From Latin unus ( = one) and
cornu (= horn) for the protuberance on the posterior of
the left valve.

Ecology and Distribution: This freshwater species has
been found in three widely separated localities in
Australia: Newmann’s Rocks in W.A. ( = type locality),
swamp at Booligal, N.S.W., and roadside pool, 13 km
east of Rocky River, Kangaroo Island, S.A.

Remarks: The extension of the posterior protuberance
on the left valve is variable: in some valves from the type
locality, the protuberance was restricted to a slight
lengthening of the left valve over the right one.

Subfamily Cyprettinae Hartmann 1963
Genus Cypretta Vavra 1895

Type Species: Cypridopsis (Cypretta) tenuicaudata

Vavra 1895.

Cypretta baylyi McKenzie 1966
Figs 19 m-q, 23, 24

1966 Cypretta baylyi McKenzie, p. 273.

Diagnosis: Cypretta with pitted to reticulated shell and
ventral margin in front of the concave mouth region,
0.33 from the anterior, forming a strongly convex lump
in both valves. Anterior to this lump, the shell margin is
flattened.

Description: The original description of C. baylyi by
McKenzie (1966) is sufficient, and does not warrant ad¬
ditional description here except for the diagnostic
features of the male anatomy since McKenzie (1966)
only dealt with female specimens.

(Maxilla) —(Fig. 24F, G) Male palps broadly arched
but of different width.

(Hemipenis) —For outline see Fig. 24J; lateral lobe
broad and tongue-like.

(Zenker organ)-(Fig. 24K) One end funnel-shaped,
with 11 rosettes.

Colour of Shell: Green to dark green.

Remarks: In one male specimen, two inner distal setae,
instead of one, were seen on the second segment of the
thoracopoda I. The surface of the shell of C. baylyi is
known to vary extensively from finely punctate, as in
Fig. 19q, to regularly reticulate, as in Fig. 23L. The
finely punctated specimens are usually smaller, narrower
in dorsal view and the convex lump anterior to the
mouth region is less pronounced.

Ecology and Distribution: C. baylyi is a freshwater
species and a very common inhabitant of temporary
pools in Western Australia (collections from Dr. I. A. E.
Bayly and personal ones). It was originally described
from near Inverway in the Northern Territory, and has
not been recorded in eastern Australia. A brief review of
all Cypretta species is provided by Sohn and Kornickcr
(1973).

On one occasion numerous specimens of C. baylyi
were found crawling on soft mud in a granite pool below
2 cm of water. This is presumably the typical mode of
locomotion for this species which is devoid of long
natatory setae on the antennae, an unusual feature for
Cypretta species which are commonly good swimmers.

Subfamily Cypridopsinae Kaufmann 1900
Kapcypridopsis McKenzie 1977
Type Species: Kapcypridopsis barnardi McKenzie 1977

Kapcypridopsis asymnietra n. sp.

Figs 25, 26

Diagnosis: Kapcypridopsis with valves asymmetrical
posteriorly: right valve with posterodorsal hump extend¬
ing well beyond the smoothly curved left valve; lateral
lobe of hemipenis digitate and with a concave and blunt
distal end.

Description: carapace (External) —Pseudopunctate,
subrectangular in lateral view with dorsum gently arched
and ventrum almost flat except in the mouth region
where it is slightly concave; right valve larger than left

OSTRACODA FROM INLAND AUSTRALIA

77

Fig. 23 — Cypretta baylyi McKenzie 1966. Granite pool on top of Boyagin Rock, between Brookton and
Pingelly, W.A. Scales: 1 = 100/x for a-k, = 10/z for 1; 2 = 50 fi for m. a-f, 1, m females; g, h males, a, RV in¬
ternal. b, LV internal, c, C dorsal, d, LV external, e, RV external, f, C ventral, g, RV internal, h, LV
internal, i, LV external, j, RV external, k, C dorsal. 1, C dorsal, detail of c. m, RV internal, posterior

detail of a.

78

P. DE DECKKER

Fig. 24 — Cypretta baylyi McKenzie 1966. Granite rock pool on top of Boyagin Rock, between Brookton
and Pingelly, W.A. A, B, D-K drawn from adult male; L, M from adult female. Scale: = 100/4. A, anten-
nula. B, antenna. C, maxillula —palp and lobes. D, mandible—palp. E, thoracopoda II. F, maxilla—left
cndopodite. G, right maxilla. H, maxilla-endopodite. I, thoracopoda I. J, hemipenis. K, Zenker organ.

L, furca. M, furcal attachment.

OSTRACODA FROM INLAND AUSTRALIA

79

Fig. 25 — Kapcypridopsis asynmietra n. sp. Natural granite rock pool near Frenchman’s Bay Road,
Albany, at turn off to the Blow Holes, W.A. Scales: 1 =200for a-d, f-k; 2 = 50/x for e, 1, =20 n for m.
a-d, h, 1 male paratypes; e-g, i-k, m female paratypes. a, RV internal, b, LV internal, c, C showing LV. d,
RV external, e, RV internal, posterior detail of f. f, RV internal, g, LV internal, h, C dorsal, i, C showing
LV. j, C showing RV. k, C ventral. 1, C showing LV, posterior detail of c. m, C ventral, posterior detail

of k.

P. DE DECKKER

Fig. 26 — Kapcypridoposis asymmetra n. sp. Natural granite rock pool near Frenchman’s Bay Road,
Albany, at turn off to the Blow Holes, W.A. A-F, H, J, K drawn from holotype adult male, remainder
from paratype adult male. Scale: = 100/*. A, antennula. B, antenna. C, mandible—palp. D, thoracopoda
1. E, right maxilla. F, maxilla-left endopodite. G, maxillula-palp and lobes. H, thoracopoda II. I, fur-
ca. J, hemipenis. K, Zenker organ. L, maxilla-endopodite.

OSTRACODA FROM INLAND AUSTRALIA

81

one and overlapping it slightly all around; postero¬
dorsally, however, the right valve possesses a hump
which extends well beyond the edge of the left valve in
that area, which is also the furthest point of extension of
the valve. Edge of both valves rounded all around; nor¬
mal pore canals rare and scattered.

(Internal) —Inner lamella broad anteriorly and nar¬
row posteriorly; selvage peripheral in both valves except
posteroventrally in the right valve, where it borders the
inner lamella which does not follow the curvature oflhe
shell in that area; narrow groove outside and all along
the selvage in both valves.

Anatomy: (Antennula) —(Fig. 26A) Seven-segmented;
length width ratio of last six segments: 1 : 1 , 1 : 1 , 1:1.3,
LI, 1 . 3 : 1 , 4:1; natatory setae as long as all segments
together.

(Antenna) —(Fig. 26B) Four claws: three almost equal
on penultimate segment and a shorter pectinate one on
distal segment nearly reaching the tip of the other claws
in male; in female it is barren and shorter; unequal
natatory setae reduced to three and not reaching the
distal end of the penultimate segment.

(Mandible) —(Fig. 26C) Mandibular palp with a bris¬
tle slim, long and barren, (3 bristle thick, short, pointed
and densely pilose, 7 bristle slim, twice the length of the
distal segment and pilose in its distal third.

(Rake-like organ) —Six to seven teeth plus one inner
bifid on each rake.

(Maxillula) —(Fig. 26G) Epipod with about 17
Strahlcn; distal palp rectangular and third lobe with
toothed Zahnborsten.

(Maxilla) —Sexually dimorphic: in male (Fig. 26 E, F)
palps asymmetrical but both strongly and similarly
arched; in female (Fig. 26L) middle seta faintly plumose
and twice the length of the other equal setae; for
chaetotaxy of protopod, see Fig. 26E.

(Thoracopoda I) — (Fig. 26D) Third segment divided;
distal seta on segment and inner seta, at mid-length of
the third segment where it is divided, of equal length and
longer than half the third segment; inner distal seta of
third segment slightly shorter than the other two men¬
tioned above.

(Thoracopoda II) —(Fig. 26H) Distal pincers narrow
but long; distal setae unequal with shorter curved one
0.4 of the length of the other straight one.

(Hemipenis) — (Fig. 26J) Lateral lobe digitate with
blunt and concave end; inner lobe like a broad and
pointed hump reaching half the length of the outer lobe;
copulatory sheath like a narrow tongue near the outer
lobe and almost completely covered by it.

(Zenker organ) —(Fig. 26K) Both ends rounded and
with 12 rosettes.

(Furca) —(Fig. 261) Weakly chitinized, whip-like and
with a short seta near its base.

(Furcal attachment) —Thin and bifurcate distally.
Colour of Shell: Dark green to almost black, except in
the eye region where it is reddish brown.

Size: L H L H

holotype adult

male LV 580/x 340/i RV 590 n 360 n

paratype adult

female LV 640/* 400/* RV 680/x 410/x

Type Locality: Natural granite rock pool near Fren¬
chman’s Bay Road, Albany, at turn off to The Blow
Holes, W.A.

Derivation of Name: From Greek a - ( = not) and sym-
metros (- symmetrical) for the asymmetrical valves
posterodorsally.

Ecology and Distribution: This species has only been
collected twice, in the same year, at the type locality in
2.5 cm of water; it is not found in nearby artificial holes
dug in granite.

Remarks: Differences between the type species from the
Cape Province in South Africa and S. asymmetra^xc the
posterodorsally asymmetrical valves (in the latter), the
broad inner lamella (in the former), the number of
reduced natatory setae on the antenna, the number of
teeth on the rake-like organ, and the outline of the
hemipenis. These are considered to be specific
differences only.

Genus Sarscypridopsis McKenzie 1977
Type Species: Sarscypridopsis gregaria (Sars 1896).

Sarscypridopsis aculeata (Costa 1847)

Figs 27, 28

1847 Cypris aculeata Costa, p. 11.

1867 Cypridopsis aculeata; Brady, p. 117
1900 Cypridopsis aculeata; Muller, p. 85.

1968 Cypridopsis obstinata Barclay, p. 75.

1977 Sarscypridopsis aculeata; McKenzie, p. 49.
Diagnosis: Member of Sarscypridopsis with sub-
triangular shell in lateral profile; with dorsum straight
along the hinge line and forming an obtuse angle with
the almost straight posterodorsal area; surface of shell
pitted and with or without spines; shallow depression
above the hingeline as both valves are higher than the
hinge line; ventral overlap of right valve over left.
Description: This cosmopolitan species has been
described fully on numerous occasions (Muller 1900,
Sars 1928, illustrated Swain 1976). It is not further
described here but it is illustrated as the first Australian
record.

Ecology and Distribution: Sarscypridopsis aculeata is
a cosmopolitan species, commonly found in temporary
pools. It has been collected in pools in WA, SA, and
Victoria. The salinity range for the species is freshwater
to 11.2°/oo with one additional record at 21.3°/oo. This
record is much higher than in European waters where
the upper salinity is 1.95 °/ 00 (see De Deckker 1981c).
This might be the result of an acclimatization to
Australian conditions where water in temporary pools is
commonly saline.

Remarks: S. aculeata is definitely not a Cypridopsis
since its right valve overlaps the left one ventrally — this
is the opposite in Cypridopsis. The difference is also
reflected in the male anatomy where the right prehensile
palp of the maxilla is larger than the left in Cypridopsis ,
the opposite of Sarcypridopsis and Plesiocypridopsis

82

P. DE DECKKER

Fig. 27 — Sarscypridopsis aculeala (Costa 1847). a-d, h, k-n, o, pond very close to Reel Inlet (coastside),
19 km S. of Mandurah, W.A. e, f, shallow lake near south-western Highway, 15 km N. of Horney or im¬
mediately S. of Yarloop, W.A. g, i, j, q Lake Sadie, E. end of Wilson Inlet, near Denmark, W. A. Scales:
1 = 250/* for a-n, q; 2 = 100/* for o, p. a-h, k-p females; i-j, q juveniles, a, RV internal, b, LV internal, c, C
showing LV. d, C showing RV. e, RV internal, f, LV internal, g, C showing LV. h, C showing RV. i, RV
external, j, LV internal, k, C ventral. 1, C dorsal, m, LV dorsal, n, RV dorsal, o, C showing RV, detail of
h. p, C showing LV, dorsal detail of g. q, C showing RV.

OSTRACODA FROM INLAND AUSTRALIA

83

Fig. 28 — Sarscypridopsis aculeata (Costa 1847). Pond very close to Reel Inlet (coastside), 19 km S. of
Mandurah, W.A. Drawn from an adult female. Scale: = 100/z. A, antennula. B, antenna. C, thoracopoda
I. D, maxilla — endopodite. E, maxillula — palp and lobes. F, mandible—palp. G, thoracopoda I. H, fur-
ca.

84

P. DE DECKKER

Rome 1965. Unfortunately, no males of S. aculeata have
yet been found to confirm the transfer of this species to
Sarscypridopsis.

S. aculeata has a green to dark green shell and often
many large pseudopores. In addition, between these
pseudospores, either hairs or small spines, are present
(see Fig. 27o, p). These differences are presumably con¬
trolled ecologically. It appears therefore that species
which are almost identical with S. aculeata but which are
extremely spinose, as described by Sars (1924) for
Cvpridopsis spinifera Sars 1924 from South Africa, are
to be synonymized with S. aculeata. The non-spinose
species Cvpridopsis obstinata Barclay 1968, described
from New Zealand by Barclay (1968) is considered here
to be synonymous with S. aculeata as the anatomy of the
type specimen of the former has been examined and is
identical with the specimen of S. aculeata illustrated in
Fig. 28.

ACKNOWLEDGEMENTS
This paper was written during the tenure of a Com¬
monwealth Postgraduate Research Award under the
supervision of Professor W. D. Williams in the Zoology
Department, University of Adelaide.

Many of the collections described here have been
received from Dr I. A. E. Bayly, Mr R. Croome, Ms M.
Davies, Dr B. V. Timms and Mr M. J. Tyler. I wish to
thank these people as well as Dr M. Christiansen (Oslo
Museum) and Dr G. Poore (National Museum of Vic¬
toria) for the loan of type specimens.

I also wish to acknowledge Mr P. G. Kempster for his
prompt help in photographical matters and Miss A.
Sebastyan who typed the manuscript.

REFERENCES

Barclay, M. H., 1968. Additions to the freshwater ostracod
fauna of New Zealand. N.Z. J. Mar. Freshwat. Res. 2:
67-80.

Brehm, V., 1939. Eine neue, subfossile Limnocythere von
Neuseeland. Zool. Anz. 127: 191-193.

Chapman, F., 1914. Notes on Testacea from the Pleistocene
marl of Mowbray Swamp, North West Tasmania. Mem.
twin. Mus . Melb. 5: 55-61.

Chapman, F., 1919. On an ostracod and shell marl of Pleis¬
tocene age from Boneo Swamp, West of Cape Schanek, Vic¬
toria. Proc. R. Soc. Viet. 32: 24-32.

Chapman, M. A., 1963. A review of the freshwater osiracods
of New Zealand. Hydrobiologia 22: 1-40.

Chapman, M. A. & Lewis, M. H., 1978. An introduction
to the freshwater Crustacea of New Zealand. Collins,
Auckland.

Danielopol, D. L. & McKenzie, K. G., 1977. Psychrodromus
gen. n. (Crustacea, Ostracoda), with redescription of the
cypridid genera Prionoeypris and llyodromus. Zool. Scr. 6:
301-322.

Df. Deckker, P., 1981a. Taxonomic notes on some
Australian ostracods with description of new species Zool.
Scr. 10: 37-55.

De Deckker, P., 1981b. Taxonomy, ecology and palaeo-
ecology of ostracods from Australian inland waters. Un¬
published Ph. D. Thesis, University of Adelaide.

De Deckker, P., 1981c. Ostracods of athalassic salt

lakes: a review. In Salt Lakes: Proceedings of an Interna¬
tional Symposium , W. D. Williams, ed.. Junk, The Hague.

De Deckker, P. & Geodes, M. C., 1980. Seasonal fauna
of ephemeral saline lakes near the Coorong Lagoon, South
Australia. Aust. J. Mar. Freshwat. Res. 31: 677-699.

Deevf.y, E. S., 1955. Paleotimnology of the upper

swamp deposit, Pyramid Valley. Rec. Cant. Mus. 6:
291-344.

Eagar, S. H., 1970. A new species of Eucypris (Ostracoda)
from Wellington. N.Z. ./. Mar. Freshwat. Res. 4: 195-202.

Eagar, S. H., 1971. A checklist of the Ostracoda of New
Zealand. J. R. Soc. N.Z. I: 53-64.

Gauthier, H., 1928. Ostracodes et cladoceres de PAfrique du
Nord (2 C note). Bull. Soc. Hist. Nat. Afr. Nord. 19: 69-79.

Geddes, M. C., Df Deckker, P., Williams, W. D.,
Morton, D. & Topping, M., (1981). On the chemistry and
biota of some saline lakes in Western Australia. In Salt
Lakes: Proceedings of an International Symposium, W. D.
Williams, ed., Junk, The Hague.

Gill, E. D. & Banks, M. R., 1956. Cainozoic history of the
Mowbray Swamp and other areas of northwestern
Tasmania. Rec. Q. Viet. Mus. 6: 1-41.

Henry, M., 1923. A monograph of the freshwater En¬
tomostraca of New South Wales, Part HI. Ostracoda. Proc.
Limn. Soc. N.S.W. 48: 267-286.

Hornibrook, N. de B., 1955. Ostracoda in the deposits of
Pyramid Valley Swamp. Rec. Cant. Mus. 6: 267-278.

Hussainy, S. U., 1969. Description of the male of Can-
donocypris assimilis G. O. Sars 1894 (Cyprididae,
Ostracoda). Pap. Proc. R. Soc. Viet. 82: 305-307.

King, R. L., 1855. On Australian Entomostracans. Pap. Proc.
Roy. Soc. Tas. 3: 56-75.

McKenzie, K. G. f 1966. Mytilocypris, a new ostracode genus
from Tasmania. Pap. Proc. R. Soc. Tasm. 100: 27-30.

McKenzie, K. G., 1971a. Distribution of freshwater

Ostracoda. Bull. Centre Rech. Pau-SNPA 5 suppl.: 179-190.

McKenzie, K. G., 1971b. Ostracoda from Lake Pcunde,
near Mt. Wilhelm, New Guinea. Zool. Anz. 186: 391-403.

McKenzie, K. G., 1977. An illustrated Key to South African
continental Ostracoda. Ann. S. Afr. Mus. 74: 45-103.

Mehes, G., 1939. Ostracodes de la Nouvclle Caledonic.
Rev. Suisse Zool. 46: 549-565.

Muller, G. W., 1900. Deutschlands Siisswasser-ostracoden.
Zoologica 12: 1-112.

Okubo, I., 1975. Studies on Ostracoda in fish ponds —I —Two
species in fish ponds on the Chiba Prefectural Freshwater
Fisheries Experimental Station. Bull. Jap. Soc. scient. Fish.
41: 155-165.

Rome, D. R., 1969. Morphologic de 1’attache de la furca
chez les Cyprididae et son utilisation en systematique. In The
Taxonomy, Morphology and Ecology of Recent Ostracoda ,
J. W. Neale, ed., Oliver & Boyd, Edinburgh.

Sars, G. O., 1894. Contributions to the knowledge of the
freshwater Entomostraca of New Zealand as shown by ar¬
tificial hatching from dried mud. Torch. Vidensk. Selsk,
Krist. (1894) 5: 1-62.

Sars, G. O., 1896a. On some freshwater Entomostraca from
the neighbourhood of Sydney, partly raised from dried mud.
Arch. Math. Naturv. 18: 1-81.

Sars, G. O., 1896b. On some West Australian Entomo¬
straca raised from dried sand. Arch. Math. Naturv. 19: 1-35.

Sars, G. O., 1924. The freshwater Entomostraca of the Cape
Province (Union of South Africa) Part II. Ostracoda. Ann.
South Afr. Mus. 20: 105-193.

Sars, G. O., 1928. An account of the Crustacea of
Norway with Short descriptions and figures of all the species.
9: Ostracoda. Bergen Museum, Bergen.

OSTRACODA FROM INLAND AUSTRALIA

85

Sohn, I. G. & Kornickkr, L. S., 1973. Morphology of
Cypretta kawatai Sohn & Kornicker, 1972 (Crustacea,
Ostracoda), with a discussion of the genus. Smith. Contrib.
Zoo/. 141: 1-28.

Swain, F. M., 1976. Evolutionary development of cypridopsid
Ostracoda. Abh. Verb, naturwiss. Ver. Hamburg. (NF)

18/19 (Suppl.): 103-118.

Victor, R. & Ff.rnando, C. H., 1980. On Herpetocypris
makua (Tressler) 1937, a freshwater ostracod (Crustacea:
Ostracoda) from the Hawaiian Islands, with notes on the
other species of the genus. Can. J. Zoo/. 58: 1288-1297.

/

PROCEEDINGS

OF THE

ROYAL SOCIETY OF VICTORIA

Volume 93

NUMBER 2

ROYAL SOCIETY'S HALL
9 VICTORIA STREET, MELBOURNE 3000

Contents of Volume 93 Number 2

Article Page

6 The Bruun Rule —the relationship of sea-level change to coastal erosion and

deposition .By H. Allison and Maurice L. Schwartz 87

7 A study of Mt Kororoit, Victoria —a small volcanic vent

By W. S. Fischer and L. Thomas 99

8 Chapman’s “Mallee Bores” and “Sorrento Bore” Ostracoda in the National
Museum of Victoria, with the description of Maddocksella new genus

By K. G. KcKenzie 105

9 Studies on Western Australian Permian brachiopods 2. The family

Rugosochonetidae Muir-Wood 1962.By N. W. Archbold 109

10 Stratigraphy, sedimentology and hydrocarbon prospects of the Dilwyn Formation

in the central Otway Basin of southeastern Australia. By G. R. Holdgate 129

1 1 Tertiary fluvial sediments at Morrison, Victoria.By Paul Bolger 149

Short communication:

Taxonomic status of the Victorian fossil whale, Ziphius ( Do/ichodon)
geelongensis McCoy 1882 .By R. Ewan Fordyce 157

THE BRUUN RULE-THE RELATIONSHIP OF SEA-LEVEL
CHANGE TO COASTAL EROSION AND DEPOSITION

By H. Allison* and Maurice L. Schwartz!

♦Institute of Earth Resources, Division of Land Resources Management, CSIRO, Private Bag,
P.O., Wembley, 6014, Western Australia.

tWcstern Washington University, Bellingham, Washington, 98225, U.S.A.

Abstract: The Bruun Rule in shore erosion is presented systematically, starting from initial assump¬
tions and axioms towards a rigorous mathematical treatment. Mathematical treatment is given with detail
to assist geologists in full understanding of the development. Cases of beach excavation, littoral drift, ar¬
tificial beach nourishment and formation of a cuspate shoreline are treated by the theory, based on the
Bruun Rule.

INTRODUCTION

The concept of shore erosion due to sea-level rise ex¬
pressed by Bruun (1962), later becoming widely known
as the “Bruun Rule” (Schwartz 1967), belongs to the
class of concepts, which are enthusiastically supported
by some scientists, while immediately triggering a
negative response from others.

The chronology of Bruun Rule studies (Fisher 1980a)
includes laboratory and/or field observations, by
Schwartz (1965, 1967), Dubois (1975, 1976, 1977a,
1977b), Hands (1976, 1977, 1980), Fisher (1977a, 1977b,
1977c, 1980b), and Rosen (1977, 1978a, 1978b, 1980); all
of them strongly supporting Bruun’s concepts. On the
other hand, the concept was criticized by Swift (1976) as
being of limited applicability. Another criticism is that
by Kaplin (1973), who is skeptical about the validity of
Bruun’s concepts; although supportively stating
simultaneously that at least some of Bruun’s concepts
were known by the Soviet school as early as 1946.
Valuable comments on applicability of the concept are
given by Gill (1979).

It appears that most of the controversy exists solely
due to differences in interpretation of the Bruun Rule.
The differences arise, in their turn, due to lack of
rigorous mathematical formulation as well as lack of
clarity in the statements of the initial assumptions of the
Buie. Hence, our task in this paper is to present the
Bruun Rule as a rigorous theory based on clearly stated
assumptions. It is shown that the Bruun Rule has, in
fact, a much wider field of application, than was
previously thought. The Bruun theory can be applied
for both “closed” and “open” beach systems. We start
from formulation of the Bruun Rule as in Schwartz
(1965, 1967).

THE BRUUN RULE AND ITS INITIAL assump¬
tions

The Bruun Rule states the following (Fig. 1):

1. A rise in sea level causes erosion of the upper beach
and shoreward displacement of the shore-water
boundary.

2. The change in sea level corresponds to translation of
the transverse beach profile while retaining its
original shape.

3. The material eroded from the upper beach is equal in
volume to the material deposited on the nearshore
bottom.

4. The rise of the nearshore bottom is equal to the rise
in sea level.

5. The relationship between sea level rise a and
shoreward displacement s of the beach profile is given
by the formula (Bruun 1962):

where: / is the length of the transverse profile,

h is the profile height, being the sum of sea
depth at the distance / from the shore and
the shore elevation above the sea level.

LIMITS OF THE BRUUN RULE

As stated by Gill (1979) the rule only applies 1, where
there is sufficient energy; 2, where equilibrium has been
attained; 3, where there is sufficient space in the subtidal
area; and 4, if there is sufficient sediment.

EXPERIMENTAL SUPPORT OF THE BRUUN
RULE

The two classes of experiments, supporting the Bruun
Rule, are laboratory (Schwartz 1965, 1967) and field ex¬
periments. The advantage of laboratory experiments is
that the “equilibrium profile” which is practically
unobservable on a beach, due to continuous variations
in wave climate, can be maintained in a laboratory
where the wave climate can be set constant.

Laboratory Experiments

In the first experiment (Schwartz 1965), the wave
basin was 81.25 cm wide and 115 cm long with variable
gradient of the bottom adjustable at 0°, 2.5°, and 4.5°.
The waves generated had a period of 0.33 sec ±5%, an
amplitude of 8 ±2 mm and a wave length of 15 ±1 cm.
Water depth ranged from 5 to 10 cm, the sand used was
a natural Ottawa sand, washed and sorted.

It was found that 30 mins of wave attack produced a
beach profile which did not change with subsequent
wave action, and this profile was assumed to be the
“original profile” figuring in the Bruun statements. The
water level was then raised by 10 mm, the wave

87

88

H. ALLISON AND M. L. SCHWARTZ

Y

Fig. 1—Definition sketch of the variables in the Bruun Rule.
The proto-profile AOB is translated towards the shore (positive
^-direction) to its new position AOB. However, no deposition
occurs at the values of x<0. The segment DO is assumed to be
the sharp boundary between the regions. In reality there will
always be a smooth transition from the point D to the curve
BO. However, this is neglected in the Bruun theory.

generator was set up again and as soon as the
equilibrium profile was reached water depth and depth
of sediment at the outer edge of the shelf were
measured.

The profile retained its original shape after water level
rise and was translated towards the shore, causing the
shore recession. During the experiment the slope of the
bottom was varied and it is remarkable that this change
did not have any significant effect on water depth or
sand heights accumulated at the lower portions of the
beach.

In the second laboratory experiment (Schwartz 1967)
the wave tank was larger in size (100 x 232.5 cm) and the
wave generator operated with variable periods. The sand
was the same as in the previous experiment and water
level was raised in different increments and at various
rates.

Profile shapes were observed before and after water
level rise and the results indicated support for statements
1, 2 and 3 of Bruun’s Rule. The most important state¬
ment (4), that the increase in sand height deposition on
the outer shelf equals the rise of water level (hence,
maintaining the constant water depth there), was sup¬
ported by the experimental results.

These experimental results support the first 4
statements of the Bruun Rule. However, formula (1)
was not checked during the laboratory experiments.
This was done in some of the following field ex¬
periments.

Field experiments

In 1964 an experimental program was established to
study the variation of beach profiles under the variation
of sea level due to neap and spring tides (Schwartz 1967,
1979). At Herring Cove Beach in the Cape Cod National
Seashore (USA) during a slight breeze, small waves ap¬

proaching the breaker zone were observed with 15-20 ctyi
heights and periods of 3 sec. The direction of wave ap¬
proach during the summer is predominantly from ths
north-west. Shore drift at this time, is consequently*
toward the south. Sediment supply for this part of ths
Cape was provided by shore drift from the cliffs Ctf
glacial drift that form the outer coast. The nearshoKj
bottom is characterized by a steep slope prior to grading
off to a gentler slope.

At nearby Nauset Light Beach which bears 12° we$t
of north and is exposed to the fetch of the Atlantic
Ocean, small waves from the southeast approach the
breaker zone on a calm day with a period of 8 secs and a
height of 30 cm. Erosion of the back beach cliff of
glacial drift supplies sediment that drifts northward.
The nearshore bottom slopes gently to a bar 500 m
offshore at low tide.

The results of profiling during low and high water
levels indicate parallel translation without significant
change in shape during sea-level rise. The more dynamic
regime at Nauset Light Beach means greater translation.

Another field experiment, strongly supporting state¬
ment 2 of the Bruun Rule, was conducted by Dubois
(1976) who measured nearshore profiles of Lake
Michigan as water level seasonally rose from April to
July, 1971. There were a total of 17 measurements of
beach profiles during this period. The profiles presented
by Dubois clearly indicate the parallel translation
without significant changes in shape of profiles as water
level rose.

Using the Bruun formula (1), in which the value / was
taken as a distance from foreshore base to the position
of breaking waves, Dubois (1977a) calculated shore
recessions and compared them with the observed values
for various wave conditions recorded. The agreement
between the values of shore recession observed in the.
field and those calculated on the basis of Bruun’s for¬
mula (1) was remarkable.

The above field observations were limited to only two
sites and a few months in both Schwartz’s and Dubois’
experiments. The following field experiment by Hands
(1976, 1980) was devoted to more long-term shore ero¬
sion.

Hands (1976, 1980) described the response of Lake
Michigan shores to increased water level over a 9 year
period at 34 sites. He used formula (1) and demon¬
strated good agreement with observed erosion. His most
interesting result was the clear transition zone between
the area of bottom erosion and no erosion. Hands was
the first to point out that actual length of the bottom
profile (value / in eq. (1)) is not important, because it is
the value /i//, which is critical in the formula (1), and
this value does not change much whether the length of a
transition zone is considered or not.

Observations by Fisher (1977a, 1977b, 1980b) of
Rhode Island shoreline retreat over 35 years with a scope
of 113 sites take into account an advance of sea-shore
boundary due to a submergence (drowning) of the shore
without erosion, separating this from shore retreat due
to shore erosion, to which he applies Bruun’s formula

BRUNN RULE

89

(1). The greatest contribution of Fisher is to consider the
“open” system, in which not all the sediment volume
eroded is deposited on the nearshore bottom. Sediment
budget calculations by Fisher require knowledge of the
position of the point between the beach erosion zone
and the offshore deposition zone (point Z on Fig. 1) to
which various names had been prescribed (inflection,
fulcrum, null point).

It will be demonstrated below, however, that
knowledge of position of such a point is not needed for
either “closed” or “open” systems.

The longest project undertaken up to date was that of
Rosen (1978b, 1980), who described the shore erosion
on 146 beach units, due to sea level rise, over 100 years.

The study area consisted of 350 km of estuarine
shoreline in the southern half of Chesapeake Bay.
Bruun’s formula ( 1 ) was applied to calculate an average
over 100 years’ shore recession rate and the results were
compared to the measured values.

On the basis of the above laboratory and field ex¬
periments it is clear that experimental support exists for
the Bruun Rule. We now have to analyze the Bruun Rule
statements themselves.

ANALYTICAL DESCRIPTION OF THE BRUUN

effect

Statement 2 of the Bruun Rule, termed the “Bruun
Effect” (Schwartz 1967), must be accompanied by state¬
ment l, which indicates in what direction (onshore or
offshore) the profile is translated during sea level rise.
Considering these two statements together as separated
for the time being from the other Bruun statements, we
must introduce a notation a for the rise of the profile,
which is not required at this stage to be equal to the sea
level rise a.

The bottom profile is considered as a function f(x,s)
of the variable * and the parameter s.

Assume that the initial shape of the profile, before
erosion occurred (5 = 0), was given by a certain arbitrary
function f 0 (x ). Then the initial condition can be written
as

Ax, 5)1 =fo(x) (2)

|5=0

The translated profile, shifted by s and lifted by a(s) is
given by

Ax, 5 ) =f 0 (x - 5) + a(s) (3)

Let us assume now, that we accept Bruun formula (1),
equating a-a, which, when substituted into (3) gives the
result:

Ax,a)=Mx-f> + a (4)

This is the analytical form of the Bruun effect,
because, when a = 0 , the function Ax,a) reduces to the
initial profile shape f () (x); for any given positive sea level
rise a , the profile is shifted shorewards in the positive
^-direction by the value al/h and raised by the value a ,
retaining its original shape f Q (x ), hence statements 1 and
2 of Bruun’s theory follow from (4). (For the negative a

(sea level fall), the profile moves offshore and
downwards).

Let us consider now the case in a certain sense op¬
posite to rise in sea level, namely, elevation of the beach
profile by the value 6 ( 5 ) due, for example, to artificial
nourishment of a beach. Elevation of the beach profile
while the absolute sea level remains still is equivalent to a
fall of the relative sea level with respect to the nearshore
bottom. Hence, according to Bruun’s statements 1 and
2 , the beach should retain its original shape, but must be
translated seawards.

If, once again, the Bruun formula (1) is utilized, with
the value of profile rise b substituted instead of a , a for¬
mula analogous to (4) emerges:

Ax,b)=f 0 (x+^) + b (5)

When there is no beach nourishment (b = 0), the
beach profile reduces to the original profile shape f 0 (x).

For any positive b >0 (beach nourishment) the profile
is shifted seawards by the value bl/h , while retaining its
original shape. (For b negative due, for example, to
sinking of sediments offshore, littoral drift or bottom
excavation, the profile is moving shorewards, i.e. ero¬
sion occurs.)

If it happens that sea level rise a is accompanied by
beach nourishment, resulting in the bottom rise b , then,
combining equations (4) and (5), the resultant beach
transformation can be described as:

Ax,a,b) =/o[x+ (b - a)] + (b +a) (6)

where f 0 (x) is the original shape of the beach.

Note that the basic equations (4), (5) and ( 6 ) were ob¬
tained solely on the basis of the first two statements of
Bruun Rule.

BRUUN RULE FOR CLOSED BEACH SYSTEMS

Statement 3 of the Bruun Rule is the definition of the
closed beach system, in which the volume of erosion is
balanced by the volume of deposition and no exchange
of beach material with the outer world exists.

It may seem obvious that for the calculation of
volumes of erosion and deposition one needs to know
the position of the fulcral (null) point (see Fig. 1) and
this need was, in fact, expressed by Dubois (1977a) and
Fisher (1980). The difficulty here is that there could be,
in principle, several such fulcral points, as shown in Fig.
2. Determination of their positions requires knowledge
of beach profile shape and foreknowledge of the same
values which one is going to calculate on the basis of
equating the volumes of erosion and accretion.

This difficulty, however, can be by-passed if one
prescribes opposite signs to the volumes of erosion and
accretion according to the signs of difference in or¬
dinates between the protoprofile f 0 (x) and the translated
profile J[x,s). (Sign plus corresponds to accretion; sign
minus to erosion.) Then the definite integral taken along
the A'-axis will be equal to zero, when volumes (area be¬
tween the curves in Fig. 2) of erosion and accretion
balance each other.

90

H. ALLISON AND M. L. SCHWARTZ

Y

Fig. 2 —Multiple fulcrum points, corresponding to multiple
areas of erosion (horizontal dashing) and deposition (vertical
dashing).

The important issue here is the limits of integration.
As already mentioned, the value / (offshore beach
length) is supposed to have been chosen in such a way
that no erosion or deposition occurs offshore beyond
this length, i.e. to the left of the point x=0 in Fig. 2.
Therefore, it is reasonable to accept point a*=0 as the
lower limit of integration. On the other hand, if the
beach profile retreat is equal to value s during sea level
rise, it is reasonable to take the upper limit of integra¬
tion as (l + s ). Consequently, we conclude that Bruun’s
statement 3 can be equivalently written analytically (in
the co-ordinate system of Figs 1, 2) as

1= T [f (x,s)-f (x)) dx = 0 (7)

0 o

The first two statements of the Bruun Rule were
shown in the previous paragraph to lead to the analytical
expression (3) of the functionals), in which the rise of
bottom profile as a whole was denoted a(s). According
to statement 4 we now equate the rise of profile a(s) to
the sea level rise a(s). Hence, the function^x.s), figuring
in (3) can be written

J[x,s) =/o (x-s) + a(s) (8)

Let us emphasize that equations (7) and (8) comprise
statements 1-4 of the Bruun Rule, but the Bruun for¬
mula (1) (statement 5) is not used. On the contrary, we
are going to demonstrate, that these four statements
form the complete set of axioms, from which formula
(1) can be derived. We start from several examples of
profiles, having simple analytical expressions.

Analytical Profiles
Parabolic profile

Consider the profile, described by a parabola

Mx)=x* (9)

Let us note, that when x = l, the right end value of
function (9) equals the profile height h (see Fig. 3), hence
h = l 2 . Substituting (9) into (8) and then into (7), one ob¬
tains

l+s

| [(x +s) 2 + a(s) - x 2 )dx=

l+s

| [ - 2xs + s 1 + a(s)]dx =

- x 2 s + s 2 x +xa(s) | = - Ps - Is 2 + (/+s)a(s) = 0

o

Hence

a(s) =

l 2 s + Is 2
l+s

/(l + y) '

( 10 )

(ID

where the value l 2 was equated to h. Bruun’s formula (1)
emerges rigorously from equations (7) and (8) for the
parabolic profile.

Now the position of the fulcrum point can be found
as the point of intersection of the functions f 0 (x) and
J(x,s) by solving the equation
(notice that a(s) = hs/l = l 2 s/l = Is)

x 2 = (x-s) 2 + a(s) = (x-s) 2 + Is

0 = - 2 xs + s 2 + Is

$(/+$)__ l+s (12)

2 s 2

The fulcrum point appears to be at the distance
(l + s)/2 from the coordinate system origin.

Let us find the volume (area, as in Fig. 1) of accretion.
This is the integral between the points ^=0 and the
fulcrum point x= Vi(l+s)

Vi (l+s) Vl(l*s)

V a = \[(x - s) 2 + Is - x 2 ]dx = s\[(l + s) - 2 x]dx

0 o

V4(i+»)

= s[(l + s)x-x 2 ] | =14s(/ + 5) 2 (13)

0

t

h

1

LINEAR
f (x) = kx

u

/ \ hs
a(s)=-j-

V

h

1

PARABOLIC

/ 0 (x)=x 2

a(s)=T

*T

h

1

CUBIC

/ 0 (x)=x 3

a(s)=-!}p- • G(-|-)

r r

h

1

MONOMIAL

/ 0 (x)=x"

a(s)= +■ • G n (-|-)

Gn=1 + ^ + n^*

J

r

h

|

EXPONENT

f 0 (x)=e ax -i

, , hs +oc(e*+t)"i"

T

ARBITRARY WITH

FLAT BOUNDARIES

°< s >= A,
*0+^)

fr

ARBITRARY with
SEAWARD SLOPE CC
AND SHOREWARD
SLOPE A

o(s )= hiiia+ell:

Fig. 3 —Analytical and non-analytical profiles with correspon¬
ding formula, relating sea level rise a to shore retreat s.

BRUNN RULE

91

Analogously, the volume of erosion, i.e. between the
fulcrum point jc= Vi(l+s) and the shoreward limit
x= /+$, is equal to

Preserving only the terms with the value s/l of power
two and less, one obtains in the same manner as for the
parabola and cubic function

J+, / +5

Vg= j [(x-s) 2 + /s-x 2 ]cfr = .s[(/ + $)*-A' 2 ]

Zi (i+s) Zi U*s)

= — l As(l+s) 2 (14)

From this example it is seen that volumes of erosion
and accretion are equal in absolute values, as it should
be according to the Bruun statement 3.

However, the actual volumes can be found only when
the position of the fulcral point has been calculated. The
impressive result obtained was that the Bruun formula is
exact for the parabolic profile. We shall see now, that
this is not necessarily correct for other profile shapes.

/s[l+4.

o(s) =-

s , n(n- 1)

7 + ~~5~~

l(\ + s/t)

hs//.G n (s/l) (20)

It is seen, that for large n the value of the correction
factor can also be large and departure from the Bruun
formula can be significant. For example, for n= 10 and
s/l=0A the value G„= 1.5. We shall discuss these results
later and consider now another class of profiles, for
which departure from the Bruun formula can also be
large.

Exponential profile

Cubic profile

Let/<>(*) =** 05)

then at the right end of the profile, where x= /, the value
y’o(jc)| v _ /. = l 3 = h, the value h being the profile height in
the cooTdinate system of Fig. 3. Substituting (15) into (8)
and then into (7), we arrive at

/♦a ^ /♦*

I = j [(.v-s) J + a{s) - x’l dx = - x*s + -2 .vV - xs* + xa(s) \

0 “ o

= l 3 s+ y/ 2 5 2 + /5 J + Ls 4 -(/ + s) a(s) = 0 (16)

Using the identity / J = /?, it follows:

Letf 0 (x) = e ax -1 (21)

The profile has a zero ordinate when x=0 and the
height of the profile at x=l is equal to h = e al - 1. By
varying the (positive) parameter a various shapes can be
obtained (Fig. 3). Substituting the translated function

fix,s) = e*'™' - 1 + a(s) = e*e‘ a * - 1 + a(s) (22)

together with (21) into the integral (7) we obtain, with
some simple manipulations

I = l ] s [e ax (e raa - 1) + a(s) dx (23)

a ,s) = hs(1 + II + F + 1/2 ? } = hs/, - G Z /r ) ( ]7 )

/(l+f)

One can recognize here the right-hand side of the
Bruun formula, being multiplied by the correction fac¬
tor G(s//). When the ratio s/l is small in comparison to
unity, the value of the correction factor is also small.
For example, when 5 //= 0.1, the value G($//)= 1.05,
meaning that use of Bruun formula (1) for a cubic
profile results in an error of 5%. For the more realistic
value s/l- 0.01, the factor G (s/l) equals 1.005, meaning
an error of 0.5%. We show below, that for other
analytical profiles the errors could be much larger.
Monomial profile

Let/o(.v) = *'’ (18)

where n> 1. At the point x=l the height of the profile
is h = l n . Substituting (18) into (8) and then (7), we can
expand the integrand:

(x - s) n + tf(s) — x n = — nx s + /7 ( ^[ 0 x 1 ' s 2

__ n(n — OU? — iijr -3 s 3 + ... (- 1 ) B y + 0 ( 5 ). (19)

Upon integration, the powers of (l + s) appear, which
are expanded in the series such as

(/+5)" = /"(l + A)" = /"|l +n. j +

n{n -

1) s 2

— T*

(I + s)"-' = ( \ = r-'[ \ + (n - \ ) ^ +

(n-\)(n-2) s 1 ,

2 • p ■ ■ ■ i

= ± (e-“ - 1) (e"e“ - 1) + (/+ s)a(s)

= i 1 - e") + (1 - e‘“)] + (/ + s)a(s) = 0

As the value s is considered to be small, we employ
now the familiar Taylor expansions, retaining only the
terms with s in the power two and less

1 -er= 1 -(1 +as + ° 2 ^+ .. .)c= ^(as+ a j~)

1 -e-° s = 1 -(1 (24)

Substitution of (24) into (23) leads to

-s(e“'- l)-°^(e“'+ l) + (l + s)a(s) = 0

wherefrom, if one recalls that (e^-l) = /?, the result
emerges

a(s) =

hs-\-a(e al 4-1)5 2 /2
/(!+/-)

(25)

It is seen that for a small in comparison with unity the
Bruun formula follows once again. However, for large a
(see Fig. 3) the second term in the numerator may
become predominant and it grows alarmingly fast with
the increase of a. We again temporarily postpone the
discussion of the reasons for such large errors and in¬
stead return back to the case of the linear profile, which
we avoided discussing before, quite deliberately.

92

H. ALLISON AND M. L. SCHWARTZ

Linear profile

Let/o(x) = kx;fix,s) = k(x-s) + a(s) (26)

where the slope k- j-
The integral (7) then reduces to:

I = j [k(x-s) + a(5) - kx\clx= j [ - ks + a(s)]dx

0

= - ks + a(s) = 0

0

(27)

Hence,

a (s) = ks = hs/l

(28)

and the Bruun formula emerges, exactly. Notice, that in
this particular case of linear profile (and in this case
only) the integrand in (27) equals zero identically, in¬
dependently of the value of x, as soon as (28) is used. If
we recall the case of the parabolic profile, where to find
the fulcrum point we need to equalize the integrand to
zero, it becomes obvious that for a linear profile every
point x is the fulcrum point; hence no erosion or accre¬
tion really occurs.

As any analytical function fix) can be expanded as a
power series, it follows from equations (11), (17), (20)
and (28), that Bruun formula (1) is rigorous only for
generalised parabolic profile

fix) = c l x+c 2 x 2 (29)

where the constants Ci and c 2 are arbitrary. For higher
order of profiles the Bruun formula (1) holds only as an
approximation, as summarised in Fig. 3.

Non-Analytical Profiles and Boundary Conditions

One may notice that when the profile f 0 (x) is shifted
shorewards by the value 5, the portion of the profile
which originally corresponded to the values .v<0, ap¬
pears at the positive part of the .v-axis (see Figs 1, 2).

This means, that it is insufficient to know the profile
shape Jo(x) strictly within the interval 0 <*</; to
calculate the integral (7) one needs also to know the
behavior of the function f 0 (x) beyond the points 0; /,
namely, in the wider interval [-s;l + s].

Hence, the outer boundaries of the segment [0;/] are
important, and the shape of the profile at these boun¬
daries -s<.v<0 and /<£</ +5 forms some kind of
boundary condition (not to be confused with the boun¬
dary conditions for differential equations).

We are going to demonstrate now that these are
precisely the boundary conditions, which determine
validity or invalidity of the Bruun formula (1).

Let us consider an arbitrary function fix) with its do¬
main of definition in the interval [-$<.*</ + $]. (The
subscript zero is dropped out here for simplicity of nota¬
tion only.)

Using the general formulae (7) and (8), the integral I is
as follows:

I = Tl fix-s) + a(s)-fix)\dx= I, + I 2 - I 3 (29)

where

I. = T/l* - s)dx; 1 2 = ! a(s)dx; I, = j .Rx)dx (30)
0 0 0

By a change of variables the integrals may be rewritten

as:

I, = l \fix ~ s)dx = \fix)dx = [fix)dx+ \fix)dx= (31a)

O -J 0 -J

= \j\x)dx+ \f( - x)dx.

0 0
(+J

1 2 = j a(s)dx = (/+ s)a(s) (31 b)

0

Ij = T Ax)dx= \Jlx)dx+'\J{x)dx= j f(x)dx +

oo/o

\f(x + l)dx (31c)

0

Combining the expressions (31a, b and c) according to

i

(29) it is seen, that the integrals \f[x)dx cancel each other
and: 0

1 = i[/(~A-)-y(x+/)]^ + (/ + 5)a(5) (32)

0

According to equation (7), the value I should equal
zero and the rigorous expression for the relationship 0 ( 5 )
follows:

0 ( 5 ) =

\lfix+l)-fi-x)]dx

0 _ _

/+S

(33)

Returning back to the temporarily dropped notation
/ 0 (jc), it is convenient to rewrite (33) in the following
form:

a(s) = h'!; F (s) = i l f»(x + f) -/»( - x)]dx (34)

I T J 0

The main advantage of (34) is that it makes clear the
role of boundary conditions. It is seen from (34) that
behavior of the profile f Q [x) outside the interval [0; f]
contributes to the value F(s), while shape of the profile
within this interval is of no importance at all.

To clarify this statement, consider the two classes of
non-analytical profiles.

(1) Consider one class of profiles, which are arbitrary
between the points [0, /] and horizontal beyond these
points (Fig. 3). The profile is assumed to have the height
/?, then

fi(x+ 1) =/<,(/) = h

M-x)=M0) = 0 (35)

Calculation of the function F (s) in (34) gives, after
substitution of (35) in it

F(s)= ! hdx=hs (36)

0

Hence, for such a profile

a(s) =

hs

l + 5

hs

K\+ S f)

(37)

and Bruun formula (1) follows, if the value s/l is small in
comparison to unity.

(2) Consider another class of profiles, which again are
arbitrary between the points [0,1], but have extensions

BRUNN RULE

93

beyond the terminal points as linear functions (Fig. 3)

fo(x + f) = h + 0x— shoreward

f 0 ( -x) = - ax- seaward (38)

The parameter a is, in fact, the seaward slope of the
profile, while (3 is the shoreward slope.

Substitution of (38) into (34) gives

F {s) =](/? + /3x+ ax ] dx=[hs + (0 + a) ] (39)

0 *•

It is seen, that the seaward slope a is accompanied in
formula (39) by the corresponding shoreward slope (3
and both slopes play equal roles.

Particularly, when the slopes are horizontal, (i.e.
xx = l3 = 0) formula (39) reduces to (36).

Another case of reduction to (36) occurs when
a= -/3, i.e. seaward and shoreward slopes are of op¬
posite signs.

An interesting case emerges when slopes a = (3 = k,
where k = h/l. Then it follows from (39)

F(5) = /js + A-5 2 =/75(1+^-) (40)

and the Bruun formula appears rigorously once again

, v hs{ 14" "T) hs( 1T -y) » /.

a(s) = _ r = I = hs/l ( 4 j)

/+s /(1+^-)

Hence, when both seaward and shoreward slopes of
otherwise arbitrary profiles are equal to the value h/l of
the profile, the Bruun formula (1) is exact.

The results obtained so far for the profiles of various
shapes are depicted on Fig. 3 with the corresponding ex¬
act formula a(s).

One can sec from Fig. 3 that steepness of monomial
and exponential profiles grows with the increase of their
order and so does the error in the use of the Bruun for¬
mula (I). Hence, the boundary conditions, rather than
the shape of the profiles , are responsible for accuracy of
the Bruun formula. It is seen from (39), that the value
F(s) can depart from the Bruun value hs if the seaward
and/or shoreward slopes are large. Reversely, one can
always estimate the error, given by the Bruun formula,
by means of exact formula (34).

Consequently, we may conclude that Bruun’s formula
(1) follows from the exact formula (34) as a very ac¬
curate approximation, providing the seaward and
shoreward slopes of the profile are not too steep. In
turn, formula (34) is based rigorously on equations (7)
and (8), which are the mathematical formulation of the
first four statements of the Bruun rule.

We show that the same Bruun statements form the
basis for the theory of erosion/accretion in open beach
systems.

bruun rule for open beach systems

It is convenient to define an open beach system as one
in which an additional volume ( + V) of sediment is sup¬
plied from outside the system (for example, beach
nourishment and/or littoral drift) or, alternatively,
some volume of sediment (-V) is removed from the

system (for example, due to excavation of the nearshore
bottom and/or to littoral drift).

Beach Nourishment

Consider the case of a positive volume (+ V) of sedi¬
ment being supplied to the system, while the sea level a is
still. This case was partly touched in section 4 where
equations (3) and (4) were derived solely on the basis of
the first two of Bruun’s statements.

In this section we denote the horizontal profile
displacement due to supply or removal of sediment as r ,
leaving the notation s for horizontal profile leireat solely
due to sea-level rise. Correspondingly, equation (3) in
the new notation is

Ax, r) =f 0 (x + r) + b(r) (42)

where b(r) is positive, when rise of the profile occurs due
to income of positive volume ( + V) of sediment into a
beach system and horizontal displacement /• due to the
same factor is taken as positive for seaward displace¬
ment of the profile.

The increment in volume of sediment in a beach
system due to translation of the profile given by (42) can
be calculated as an integral I, analogous to (29).

However, the limits of integration must be different
now. As the profile is shifting seawards (positive r), part
of the shifted profile will appear left of point a' = 0 (Fig.
4), beyond the limiting length / offshore. But our initial
assumption was that a profile is undisturbed to the left
of point .v=0. Hence, the lower limit of integration
should be taken as ,v=0.

On the other hand, as the profile is shifted seaward,
no disturbance of the profile occurs beyond point x *=/,
which should be taken now as the upper limit of integra¬
tion. Consequently, the additional volume ( + V) of sedi¬
ment supplied to the beach should be equal to the in¬
tegral (we again temporarily drop the subscript zero in
Mx))

I=J \A* + r) + b(r)-J{x)]dx= I, +1 2 - Ij = V (43)

0

where

Ii= J Ax+r)dx= j )\x)dx + J j{x)dx + \f(x)dx (44a)

0 r r l

I 2 = j b(r)dx=lb(r) (44b)

0

Ia= j J(x)dx. (44c)

0

Adding to I, eq. (44a) the self-cancelling pair of in¬
tegrals

'\f(x)dx- ]/(x)dx (45)

O 0

we can rewrite I, as

I.= 1 j\x)dx+ \f(x)dx + "\f(x)dx- \ J\x)dx (46)

0 r l 0

where the sum of the first two integrals exactly equals I 3 ,
thus cancelling it.

94

H. ALLISON AND M. L. SCHWARTZ

Fig. 4—Deposition of sediments, resulting in rise of profile by
the value b(r) and seaward shift by the value r (r is positive
seaward).

Consequently

1= i f[x)dx- j J[x)dx + lb(r) = V (47)

/ 0

and the formula for b(r) emerges (returning back to
subscript zero)

b(r) = X(V- ]f 0 (x)dx+ ) fo(x)dx) (48)

/ / 0

One has to notice that the profile shape between the
points x= r and x= l does not figure in (48) and value of
the integrals is determined by the behavior of the profile
shape in the neighborhood of its terminal points .v = 0
and x=L Let us assume that the profile is nearly fiat
around ,v=0 and x '=/, i.e. at the interval [0,/*] the func¬
tion f 0 (x) = 0 and at the shoreward end f 0 (x) = h at the in¬
terval [/, /4-r], then the first integral in (48) reduces to
the value hr and the second to zero. Formula (49) then
emerges as:

b(r) = (V- hr)/l

(49)

It can be shown, by analogy with the previous section,
that (49) is a very good approximation of the exact result
of (48) when the boundary conditions are such that the
seaward and shoreward slopes are not too steep and the

difference in their slope angles is small.

The expression for r follows from (49)

r=(V-bl)/h

(50)

On the other hand, we did have another expression

for the same value r in equation (5)

-i

ll

=*is:

(51)

Combining (50) and (51), we obtain

r=I b = J

2h ' 21

(52)

Consequently, equation (42), becomes dependent on
the volume of sediment supplied to the beach

f(x, V) =f 0 (x + 2 ^-) + 2 / (53)

We want to emphasize, that the values r and b in (52)
and formula (53) are independent of the profile shape.

If the beach loses sediment (due to excavation of the
bottom, littoral drift, etc.), then the value Lin (52) and
(53) should be taken as negative and the beach profile
shifts shorcwards and downwards, because the values r
and b become negative.

When beach nourishment (excavation) is accom¬
panied by sea level rise, it follows from equations (6)
and (53), that the new beach profile is described by the
equation

f(x,a, V) =/ 0 [x+ j t ( ^ -al)]+ j- ( j +af), (54)

where positive values V and a correspond to beach
nourishment and rise in sea level.

Equation (54) was obtained, after all, solely on the
basis of the first four Bruun Rule statements.

The previous treatment was performed for a two-
dimensional cross-section of a beach, which may also be
considered a slice (of unit thickness) of an actually three-
dimensional beach.

Bruun Rule for Three-Dimensional Beach System
with Littoral Drift

Consider the three-dimensional portion of shoreline
in the coordinate system of Fig. 5 (note that the y coor¬
dinate is now directed alongshore). Any vertical plane,
parallel to the plane ZOX, dissecting the shore, forms in
its cross-section a two-dimensional profile, such as
described in the previous sections. However, there is no
need to assume that all the cross-sections have the same
or similar shape. For each cross-section with a longshore
coordinate y equation (54) holds; therefore independent
of the shape of the particular cross-section, its seaward
motion (denoted R(y) from now on) and vertical uplift

(denoted a{y)) is equal to

(55a)

a{y) = a+[V(y)/21(y)]

(55b)

where a is the rise in sea level as before and the functions
h(y)\ l(y); V(y) are dependent on the longshore coor¬
dinate y. Consider the equation of continuity or the
volume conservation equation for longshore sediment
drift (see, for example, Komar, 1976)

8V/dt= -8Q/dy

where Q is the longshore volumetric rate of sediment
transport (measured, for example, in m 3 /day).

The physical sense of equation (56) is quite obvious: it
states that the time-rate (speed) of volume of sediment
removal (sign negative in 56) from a beach equals the
longshore rate in change of volumetric transport along
the beach. This is illustrated in Fig. 5, where both func¬
tion Q{y,t) and dQ/dy are plotted along the longshore
coordinate y for one specific type of dependancy Q(y).

It is reasonable to assume further that the basic
parameters h(y) and /O'), figuring in (55), do not change
much in time due to shore erosion (accretion), and only
the values, R, U, a and V in (55) are variable in time.

BRUNN RULE

95

Fig. 5 — Definition sketch for longshore littoral drift. Axis y
directed alongshore. Profile shapes could be different for
various y, such as y x and y 2 .

Consequently, equations (55) can be rewritten to incor¬
porate the above assumption as

R(y ’ t) = h( Lj [V iM- a(t) / (y)] (57a)

a(y,t) = a(t) + ^2 (57b)

Notice that sea level rise a in (57) is assumed to be
dependent on time, but independent of longshore coor¬
dinate y , meaning that extremely long coastlines, where
sea-level varies along the coastline, are not considered.
Equations (57) give the complete steady-state solution of
the problem of beach nourishment (excavation) in the
presence of variations in the time sea level rise a{t ) for
any distribution of V(yJ) along the shore and in time,
providing the values h(y) and l(y) are known.

Differentiating (57) with respect to time t one obtains

dR/dt = Vid V/dt - l(y)da/dt]/h(y) (58a)

da/dt = da/dt+ Vil(y\dV/dt (58b)

Substituting instead of dv/dt in (58) the equal value
- dQ/dy from (56), it follows

dR/dt = — [ Vi • dQ/dy + l{y) • da/dt\/h(y) (59a)

da/dt = da/dt — ViKyY dQ/dy (59b)

Integrating equations (59) with respect to time, one
obtains the final formulae for calculation of shoreward
displacement R and profile rise a as functions of
longshore sediment transport Q and variations in the
time of sea level rise a(t)

R(y, 0 = - h ^[ { j d - Q ^t + /O). a(l)] (60a)

( 60 b)

We illustrate the use of (60) by two examples.

Example 1

Let <7(/) = 0 and Q(y,t)=y sin < ot> meaning that the
sediment transport rate is linearly distributed along the
shoreline and varies periodically in time.

Then

dQ(y,t)/dy = sin ut (61)

and from (60) one obtains by integration:

R(y> t)= ~ 2 Jbf os ■ + c (62a)

<Xy,t)= - cos ut + D (62b)

where C and D are the constants to be determined from
initial conditions. Assuming them as R(y,t) = 0 and
fly,0 = 0 (no initial erosion-accretion) when f = 0, it is
easy to find from (62)

C= 2ji(y)’ D= 2zky) (63)

hence, the seaward shore displacement and rise of the
bottom are:

m 0 = 2 Jjfyf 1 - cos (64a)

«0’.0 = 2 J(^( 1 - COS “0, (64b)

meaning periodic (for example; seasonal) variation in
shore erosion, dependent on the original longshore
shape (functions h(y) and l{y)) of the coastline.

Example 2

Let <7(/) = 0 and a tropical cyclone (hurricane) moves
alongshore, causing a propagating wave-like longshore
transport Q(y,t) with magnitude Q 0

Q(y, 0 = Qo cos (Ky + (at) (65)

Then

- KQo sin (Ky + ut) (66)

and upon substitution of (66) into (60a, b) and integra¬
tion

R(y. 0=- 5 J^yf os {Ky + ut) + C(y) (67a)

a{y, 0 = - (Ky + + D(y) (67b)

Here instead of constants C and D, as in the previous
example, the functions C(y) and D(y) should be found
from initial conditions. One obtains from (67) for / = 0

f osKy (68a >

D(y)= i^/f osKy (68b)

and the final result emerges from (67) upon substitution
of (68)

96

H. ALLISON AND M. L. SCHWARTZ

^Cv,0=^-S- 2 r[ cos AT-cos (Ky + oj/)] (69a)

2u)n(y)

a[yyt) = ^-Q-2\cos Kv-cos^Ky^- ut)] (69b)

2wl{y)

Note the appearance of the time independent term cos
Ky in (69), meaning that after passage of a cyclone (hur¬
ricane) an irreversible cuspate change of a coastline oc¬
curs, if the rate of sediment transport could be approx¬
imated as a propagating wave (65).

DISCUSSION

The first four statements of the Bruun Rule form a
non-contradictory set of axioms, which need no other
axioms for development of the theory of shore erosion;
incorporating not only sea level rise, for which the
statements were originally intended, but also the prob¬
lems of beach nourishment, excavation and littoral
drift.

The most important of all assumptions of the theory,
its cornerstone, is the statement, that beach profile
essentially preserves its shape during process of shore
evolution, be it due to sea-level rise, wave attack, beach
nourishment etc. Of course, it is recognised, that some
temporal variations, perhaps caused by a sudden storm,
can be superimposed on the conservative shape of beach
profile, but these variations are usually short-lived and
can be discarded in the first approximation.

The experimental data available support this assump¬
tion. No doubt, further experimental work is necessary.

The most important limitation of the above theory is
its static nature.

This means, that the beach system is considered only
in two states: initial state, before evolution started and
final state, after evolution was completed. The transition
process between these two states is beyond the scope of
the present theory. Clearly, this shortcoming must be
rectified. This is our direction for future research.

ACKNOWLEDGEMENTS

Discussions with Mr R. A. Perry and Mr E. D. Gill
(both CSIRO) were most stimulating. The authors thank
reviewers for their valuable comments.

REFERENCES

Allison, H., 1980. Enigma of the Bruun’s formula in shore
erosion. In Proc. of the Per Bruun Symposium , M. L.
Schwartz & J. J. Fisher, eds., IGU & Western Washington
University, Bellingham, 67-78.

Bruun, P. ? 1962. Sea level rise as a cause of shore erosion:
Am. Soc. Civil Engineers Proc., Jour. Waterways and Har¬
bors Div. 88: 117-130.

Bruun P., 1980. The “Bruun Rule”, Discussion on boundary
conditions. In Proc. of the Per Bruun Symposium , M. L.
Schwartz & J. J. Fisher, eds., IGU & Western Washington
University, Bellingham, 79-83.

Gill, E. D., 1979. Change from quartz arenite to calcarenite at
Warrnambool, Victoria, Australia. Viet, Naturalist 96:
151-153.

Dubois, R. N., 1975. Support and refinement of the Bruun
Rule on beach erosion. J. Geol. 83: 651-657.

Dubois, R. N., 1976. Nearshore evidence in support of the
Bruun Rule on shore erosion. ./. Geol. 84: 485-491.

Dubois, R. N., 1977a. Predicting beach erosion as a function
of rising water level. ./. Geol. 85: 470-476.

Dubois, R. N., 1977b. Nearshore evidence in support of the
Bruun Rule on shore erosion, Reply. J. Geol. 85: 492-494.

Fisher, J. J., 1977a. Coastal erosion revealed in the U.S.S.R.:
Geotimes 22 (9): 24-25.

Fisher, J. J., 1977b. Rate of erosion relative to sea level rise
and profile of equilibrium, Rhode Island and North
Carolina. Astrets 10th In VI Quaternary Congress , Birm-
ingham, 140.

Fisher, J. J., 1977c. Relationship of shoreline erosion rates to
eustatic sea level rise, Rhode Island coast. Geol. Soc. A met'.
Abstract 9 (3): 265.

Fisher, J. J., 1980a. Holocene sea level rise, shoreline erosion
and the Bruun Rule-overview. In Proc. of the Per Bruun
Symposium , M. L. Schwartz & .1. J. Fisher, eds. IGU &
Western Washington University, Bellingham, 1-5.

Fisher, J. J. 1980b, Shoreline erosion, Rhode Island and
North Carolina coasts —test of Bruun Rule. In Proc. of the
Per Bruun Symposium, M. L. Schwartz & .1. J. Fisher, eds.,
IGU & Western Washington University, Bellingham, 32-54.

Hands, E. B., 1976. Some data points on shoreline retreat at¬
tributable to coastal subsidence. Proceedings of the
Anaheim Symposium , International Association of
Hydrological Sciences, p. 629-645.

Hands, E. B., 1977. Implications of submergence for coastal
engineers. In Coastal Sediment ' 77 Proceedings, Fifth Sym¬
posium of the Waterways , Port, Coastal and Ocean Divi¬
sion of the A.S.C.E., p. 149-166.

Hands, E. B., 1980. Bruun’s concept applied to the Great
Lakes. In Proc. of the Per Bruun Symposium , M. L.
Schwartz & J. J. Fisher, eds., IGU & Western Washington
University, Bellingham, 63-66.

Kaplin, P. A., 1973. Recent history of the coasts of the world
oceans. University of Moscow-, Moscow, 265 p. (in Rus¬
sian).

Komar, P. D., 1976. Beach processes and sedimentation.
Prentice-Hall, Englewood Cliffs, N. J., 429 p.

Rosen, P. S., 1977. Nearshore evidence in support of the
Bruun Rule on shore erosion: Discussion. J. Geol. 85:
491-492.

Rosen, P. S., 1978a. Predicting beach erosion as a func¬
tion of rising water level: Discussion. ./. Geol. 86: 763.

Rosen, P. S., 1978b. A regional test of the Bruun Rule on
shoreline erosion. Mar. Geol. 26: m7-ml6.

Rosen, P. S. t 1980. An application of the Bruun Rule in the
Chesapeake Bay. Proc. of the Per Bruun Symposium, In
M. L. Schwartz & J. J. Fisher, eds., IGU & Western
Washington University, Bellingham, 55-62.

Schwartz, M. L., 1965. Laboratory study of sea level rise as a
cause of shore erosion. J. Geol. 73: 528-534.

Schwartz, M. L., 1967. The Bruun theory of sea level rise as a
cause of shore erosion. J. Geol. 75: 76-92.

Schw artz, M. L., 1979. Case history of a coastal investigation
at the Cape Cod National Seashore. In Proceedings of the
First Conference on Scientific Research in the National
Parks; R. M. Linn, ed., National Park Service Transactions
and Proceedings ser. 5, 2: 757-759.

Schwartz, M. L. & Milicic, V., 1978. The Bruun Rule con¬
troversy. Coastal Research 5: 13-14.

BRUNN RULE

97

Schwartz, M. L. & Milicic, V., 1980. The Bruun Rule: A
historical perspective. In Proc. of the Per Bruun Sym¬
posium , M. L. Schwartz & J. J. Fisher, eds., IGU &
Western Washington University, Bellingham, 6-12.

Swift, D. J., 1976. Coastal sedimentation. In Marine sediment
transport and environmental management, Stanley, D. J. &
Swift, D. J., eds., John Wiley & Sons, New York, 255-310.

-

A STUDY OF MT KOROROIT, VICTORIA-A SMALL VOLCANIC VENT

By W. S. Fischer* & L. Thomas

Department of Geology, University of Melbourne, Parkville, Vic. 3052
*Now at Esso Exploration Inc.

Abstract: A geological reconnaissance and gravity survey indicate that Mt Kororoit is composed
mainly of scoriaceous material and began to grow in the early stage of formation of the surrounding lava
sheet.

regional geology

Mt Kororoit (also known as Mt Misery, at 37°39'S,
144°40'E) is a small conical hill about 1500 m in
diameter at the base, rising about 100 m above the sur¬
rounding countryside. It is one of many flat-topped
volcanic cones on the Werribee Plains, which lie on the
northern and western outskirts of Melbourne. The
plains consist mainly of sheets of basaltic lavas, dotted
with vents, of the Pliocene to Pleistocene Newer
Volcanics. The lavas lie unconformably on folded
Palaeozoic (mainly geosynclinal) sediments, which pro¬
trude through the basalt in a number of places, especial¬
ly to the northwest of the plains. Deposition of these
sediments ceased in the Middle Devonian, and was
followed by a period of granite intrusion during the Up¬
per Devonian. Sedimentation resumed in the Cainozoic;
in adjacent areas it continued during the Lower Car¬
boniferous, Permian and Mesozoic (Vandenberg et al.
1973).

GEOLOGY OF MT KOROROIT

Although occasionally noted in the literature on Vic¬
torian volcanoes, we believe that Mt Kororoit has not
previously been studied in any detail. A geological map
of the volcano based on aerial photograph interpreta¬
tion with some field work is presented in Figure 1.
Basically, the hill appears to be a scoria pile elongated
somewhat to the west, with a minor change of slope pro¬
bably marking the edge of the scoria. Cultivation of the
surrounding area has made the background difficult to
define.

Shallow pits have been dug in the scoria on the north¬
west flank of the mountain. These expose some vesicular
lava containing small feldspar phenocrysts and pieces of
fresh scoria. Outcrops of unweathered rock show the
scoria to consist of small irregular lapilli welded together
with spatter material. Small scale flow features can often
be recognised in the spatter.

Four features are prominent on the aerial photo¬
graphs:

(a) The resistant capping on the summit of the moun¬
tain.

(b) Two thin ridges which trend northeast and
southeast from the cap.

(c) A series of concentric arcuate structures on the
northwestern side of the northeast ridge.

(d) A series of arcuate structures about 1 km northwest
of the cap.

Close investigation in the field shows that the cap rock
consists largely of agglomerate material, although
basalt is found in some places. Three almost concentric
tiers form the cap, the upper showing consistent dips of
about 40° towards the centre of the cap, while the third
and lowest has a much gentler dip. Some thin layering is
also present in the cap outcrop, being aligned parallel to
the dip.

A large number of volcanoes in the Gisborne district
show the distinct flat-topped basalt capping. It has been
suggested by Edwards and Crawford (1940) that the cap¬
pings represent basaltic crater infillings later exposed by
the erosion of the scoria crater walls which once con¬
fined the lava. Some of these volcanoes also show evi¬
dence of the crater having been breached, resulting in a
lava flow forming what is now the gentlest slope. This is
not as strongly evident at Mt Kororoit.

Another possibility is that the basalt cap is more than
just a crater infilling, and is in fact the top of the former
lava conduit.

Originating at the cap rock and striking radially, the
highly vesicular basalt ridges have been variously
described as radial dykes (Singleton 1973) and as a fine
example of radial squeeze-ups (Vandenberg et al. 1973).
These seem similar in composition to the summit cap¬
ping.

Within the outcrop, one can recognise flow patterns
lying sub-parallel and parallel to the dip of the outcrop.
It is most noticeable where one finds thin elongate
vesicles aligned in curved patterns. Often, however, the
vesicles are very fine or there are no vesicles present at
all.

Along the northeastern ridge, one encounters dips to
the southeast of between 20° and 70°. Dips along the
southeastern ridge are generally about 30° to the north¬
east and east. The forks of the southeastern “dyke” ex¬
tend in discontinuous outcrop further than was noted
from aerial photographs. Between these forks lie a
number of small outcrops of basalt, many of which are
linear, sub-parallel to and dipping in the same direction
as the nearby ridges.

The concentric arcs of rock to the northwest of the
northeast ridge are very prominent on the aerial
photogaphs, but are much more difficult to recognise on
the ground. At this level they are not readily

99

100

W. S. FISCHER AND L. THOMAS

Holden
No. 8

o 500 M.

\

I'v; ~,| Scoria

☆

Sample locality

Basalt ridge

A

Trig point

Rocky ground

Road

v v Lava flow

Inferred geological

r;

boundary

Obscured by
agriculture

•

Bore hole

Fig. 1—Geological Sketch Map, Mt Kororoit, Victoria.

MT. KOROROIT—A VOLCANIC VENT

101

distinguishable from other areas of rocky ground
presumably weathered from the basaltic material.

The arcuate features northwest of the cap have flat
tops and relatively steep fronts, and are probably either
tumuli in, or lobes of, a lava flow extending in that
direction.

Samples were taken of the two major rock types for
density determinations, and a brief description of these
is given in Table 1.

Table I

Brief Description of Rocks used in Density Measurement

. Rock Locality

Type Description of Samples

Scoria Red-brown colour, fairly well con¬
solidated, fine vesicles showing
some flattening; in part made of
fine Iapilli aggregates; some well
preserved phenocrysts.

Basalt Medium to dark brown colour;
vesicular basalt; very hard; fine to
coarse gradation of vesicles-
elongate, aligned and compressed;
occasional vesicles filled with mud¬
dy material.

Kl, road cutting
about 1 km east
of the cap.

K2; edge of
basalt cap.

K 3; northeast
ridge.

gravity survey

The gravity survey at Mt Kororoit formed part of an
undergraduate field exercise. Four major traverses, run¬
ning approximately north, south, east and west were laid
out from the trig marker on the summit. These lines ex¬
tended at least 1 km from the summit. Additional lines
were also laid out on the flanks of the hill to provide ad¬
ditional control on the anomaly shapes on the hill
slopes. The average station spacing was 60 m; the lines
are shown on Fig. 2.

Two gravity meters were used; Worden No. 169, loan¬
ed by the Bureau of Mineral Resources, and Scintrex
No. 255-G. Both meters were calibrated immediately
before the survey. All station elevations were deter¬
mined by spirit levelling. Loop closure errors in both the
gravity and levelling data were adjusted numerically
after gross errors had been detected and removed.

Densities

Density determinations were carried out on 45 rock
samples collected from three localities in the field (Fig.
1). The results arc given in Table 2. Although the two
classes of sample were quite distinctive in appearance,
their densities were similar; in fact they appear
statistically to be drawn from the same population (Stu¬
dent’s test; p>0.95). Accordingly, a weighted mean
density of 1.66 g.cm -3 w r as taken as the density of the
hill material for Bouguer reduction purposes.

The indirect ‘density profile’ method of Nettleton
(1976) was also used. The density indicated by this
method was about 1.6 g.cm -3 in good agreement with
the sample results.

G

Table 2

Summary of Wet Bulk Densities for the Rock Types
Sampled

Rock Type: (Locality) Mean Density

S.D.

No. of

(g

.cm 3 )

Samples

Scoria: (Kl)

1.63

0.09

20

Basalt: (K2) & (K3)

1.68

0.06

25

Weighted mean density of

scoria and

basalt

= 1.66 g.cm -3

Standard Deviation (S.D.)

= 0.06

Data Reduction

The gravity data were corrected for latitude, elevation
and Bouguer plate effects, using the scoria density value
(1.66 g.cm -3 ) and a datum plane at the approximate
level of the surrounding land (150 m above sea level).
Terrain corrections were determined for isolated sta¬
tions, but found to be negligible.

No regional gravity gradient was observed on the ma¬
jor traverses, so the Bouguer gravity values were ar¬
bitrarily adjusted before display by subtracting 45.62
mGal from each value. The resulting values were then
three-point smoothed before plotting (Fig. 2).

The residual gravity contour map (Fig. 2) clearly
show's a negative anomaly, with a maximum amplitude
of 1.8 mGal. The anomaly is hour-glass shaped, with the
long axis approximately north-south and the narrowest
point just where the present volcanic cap appears.

INTERPRETATION
Gravity Data

The form of the gravity anomaly, together with the
surface features, suggests that a shallow subsurface
depression filled w'ith less dense material exists. The
shape of the anomaly may reflect the shape of the
depression, or variations of the density of the fill.

The near-surface rocks near Mt Kororoit are the
Newer Volcanic lava flows, and in a nearby bore
(Holden No. 8, see Fig. 1) a total of 74 m of flows were
encountered overlying Silurian mudstones.

The form of the gravity anomaly around the two
minima can be satisfied by assuming that the source is
approximately 75 m in thickness, with its surface at the
level of the surrounding terrain, and with a density con¬
trast of approximately -0.9 g.cm If the material fill¬
ing the depression is scoria similar to that sampled, then
the bulk density of the basalt flows should be about 2.56
g.cnr \ which is a reasonable value for such rocks. We
therefore propose that the fundamental basin shape of
the anomaly is due to the presence of an island of scoria
in the basalt sheet.

The ‘neck’ of the anomaly coincides approximately
(but not exactly) with the position of the prominent sur¬
face ridges to the northeast and southeast. The
amplitude of this part of the anomaly (positive, with
respect to the larger anomaly) is about 0.5 mGal, which
could arise from a shallow sheet of more dense material
(density equal to that of the surrounding lavas) of ap¬
proximately 10 m thickness buried within the cone.

102

W. S. FISCHER AND L. THOMAS

MT. KOROROIT — A VOLCANIC VENT

103

A more detailed interpretation of the gravity anomaly
has been attempted, but is not very illuminating, as the
details depend to a great extent on the interpreter’s star¬
ting assumptions.

We modelled, for instance, a hypothetical feeder pipe
below the cone as a vertical circular cylinder of radius 25
m with its top at 75 m below the datum plane. With a
density contrast of +0.4 g.cm 1 such a column would
give a circular gravity anomaly with a maximum value of
0.25 mGal and a half width of the order of 200 m. While
such a component could be added to any model, we see
no compelling evidence for a significant column of lava
at the centre of the cone.

Combined Data

We propose the following sequence of events to ex¬
plain the field and gravity data which we have reported.
Eruption at Mt Kororoit began at an early stage in the
development of the surrounding lava sheet, with effu¬
sion of mainly scoriaceous material forming an island in
the lava. Periods during which more fluid material was
erupted were interspersed, probably adding to the flows
forming the sheet around the cone, and accounting for
the lobes to the northwest of the cone. Discontinuities of
eruption allowed the magma to withdraw in the narrow
central vent, and so produce the inward dips in the tiers
of lava remaining at the top of the cone.

Near the end of activity, lava flowed from the cone
towards the east over a bed of scoria. The ridges north¬
east and southeast of the cap represent the edges of this
flow, and dip inwards toward the flow as a result of sub¬
sidence while the flow was still plastic. The dips could
also result from confinement in a channel of scoria now
eroded. The rocks sampled at the ridge may be more
vesicular than the body of the flow and so give the lower
densities observed.

The final phase of eruption began with a burst of spat¬
ter (fire fountaining) which agglomerated to form the
present cap rock, and ended with an eruption of scoria
which blanketed the flanks of the cone.

Gases derived from beneath the flow may have
formed the arcuate structures to the north of the north¬
east ridge, by a blistering process. Subsidence of the flow
may have been aided by the weight of the scoria erupted
in the final phase and which now covers the flow.

The topography suggests that the final flow must have
occurred after most of the surrounding lavas were in
place; however, subsequent erosion has most probably
severely modified the external shape of the cone.

SUMMARY

Observations at Mt Kororoit indicate that the cone is
composed of a pile of scoria, with basalt layers in¬
terspersed, which began to form during the early stages
of the formation of the surrounding lava flows. Erup¬
tion must have continued, intermittently, for much of
the period during which the flows were emplaced.

ACKNOWLEDGEMENTS
We thank our colleagues (staff and students) in the
Department of Geology for their assistance during
various stages of this study, and the Bureau of Mineral
Resources for making a gravimeter available.

REFERENCES

Edwards, A. B. & Crawford, W., 1940. The Cainozoic
volcanic rocks of the Gisborne district, Victoria. Proc. R.
Soc. Via. 52: 281-311.

Nettleton, L. L., 1976. Gravity and Magnetics in Oil Pros¬
pecting. McGraw Hill, New York.

Singleton, O. P., 1973. Geology and petrology of the
Macedon district. In Regional Guide to Victorian Geology ,
L McAndrew & M. A. H. Marsden, eds., Department of
Geology, University of Melbourne, 2nd Edition.
Vandenberg, A. H. M., Marsden, M. A. H. & McAndrew,
J., 1973. Geology of the Melbourne district. In Regional
Guide to Victorian Geology , J. McAndrew & M. A. H.
Marsden, eds., Department of Geology, University of
Melbourne, 2nd Edition.

■

r .

CHAPMAN’S “MALLEE BORES” AND “SORRENTO BORE”
OSTRACODA IN THE NATIONAL MUSEUM OF VICTORIA,
WITH THE DESCRIPTION OF MADDOCKSELLA NEW GENUS

By K. G. McKenzie

Riverina College of Advanced Education, Wagga Wagga, N.S.W.

Abstract: Fifty-four Ostracoda in the collections of the National Museum of Victoria, which were
named or described either by Chapman or by Chapman and Crespin in two early papers on Victorian
Cainozoic Ostracoda, are reassessed and reassigned as to genus and species. The new pontocypridid genus
Macldocksella , type species Maddocksella lumefacta (Chapman 1914), is described and illustrated from
among these taxa.

INTRODUCTION

No Australian palaeontologist can afford to ignore the
voluminous writings of Frederick Chapman on the con¬
tinent’s fossil faunas. From his arrival in Victoria during
1902 to his death in 1943 Chapman’s output was so pro¬
digious that it became synonymous with the develop¬
ment of Australian palaeontology. He undertook, vir¬
tually singlehanded, the description of all groups of
fossils found in this country; his bibliography includes
over a dozen papers on Ostracoda (McKenzie 1974).

Chapman had a decade of ostracode studies behind
him, in the impressive company of men such as Jones
and Sherborn, when he arrived in Australia. For this
reason his ostracode taxonomy was never questioned.
Unfortunately, with the passage of time, Chapman’s
generic level assignments need revision and many of his
species names, especially those referring to taxa des¬
cribed by G. S. Brady, are now known to be incorrect.
Most of the errors made by Chapman can be accounted
for by the unremitting pace at which he must have work¬
ed. Sometimes, sex dimorphs or juveniles and adults of
the same species are given different names. When he
referred species to taxa described by G. S. Brady, which
he did regularly, his only references were Brady’s
Papers, notably Brady’s “Challenger” Report (Brady
1880); to my knowledge he never rechecked Brady’s type
materials, most of which were then deposited at the
Hancock Museum, Newcastle-upon-Tyne.

SYSTEMATIC PALAEONTOLOGY

Family Pontocyprididae Muller 1894
Genus Maddocksella gen. nov.

Genus A; McKenzie 1964, pp. 448-453.

Australoecia ; McKenzie 1969, p. 11.

Australoecia; Maddocks 1969, pp. 49-50.

Australoecia ; McKenzie 1974, pp. 158 (Text-

fig. 3g), 166.

Australoecia n. subgen.; McKenzie 1979, pp.

90-94.

Etymology: For Dr R. F. Maddocks, who has several
Papers, including an important monograph, on pon-
tocypridids.

Type Species: Maddocksella tumefacta (Chapman
1914) (Fig. 1).

Diagnosis: Argilloeciine pontocypridids characterised

by an inflated and robust shell, strong left valve overlap
and an adductor rosette of 5 large wedge-shaped scars.
Geological Age: Eocene to Recent.

Discussion: As pointed out by Maddocks (1969, p. 49),
the right valve overlap displayed by the type species of
Australoecia McKenzie (1967, pp. 67-8) is not matched
by some different species otherwise referrable to it
because these have a marked left valve overlap. Further,
the new genus is characterized by an inflated and robust
shell whereas the shell in Australoecia victoriae McKen¬
zie 1967 is less well calcified and cigar shaped rather than
inflated. These three characters sufficiently differentiate
between Maddocksella and Australoecia and also bet¬
ween Madocksella and Argilloecia Sars, the only other
genus which bears any resemblance to the new taxon.
The adductor scar pattern of the new genus resembles
that of Australoecia but is clearly different from that of
Argilloecia.

The strong left valve overlap which characterises Mad¬
docksella vis a vis Australoecia is not fortuitous nor is it
confined to only one or a few species. Indeed, all the
many Tertiary records belong in Maddocksella . On this
evidence, Maddocksella is the ancestral taxon and
Australoecia is a radiation from the ancestral stock
which is represented in the known living fauna of
southern Australia only by the type species Australoecia
victoriae.

Other Species: These include Maddocksella mackenziei
(Maddocks 1969). Several as yet undescribed species of
Maddocksella are known to occur in the Tertiary of Vic¬
toria and South Australia (McKenzie 1974, 1979). Genus
A sp. A of McKenzie (1964) represents yet another
species which lives in Oyster Harbour, near Albany,
Western Australia. Other living species are known from
Sahul Shelf, oft' northwestern Australia (McKenzie 1974,

p. 166).

Note that McKenzie (1979, p. 90) refers the deep sea tax¬
on Australoecia abyssophila Maddocks 1969 to the
genus Abyssocypris van den Bold 1974.

Ecology: Maddocksella appears to be restricted to
coastal waters, including protected bays and estuarine
harbours where it is usually found living in sublittoral,
muddy silts and fine sand facies. Since empty shells are
often washed onto beaches, it is unlikely that the prefer¬
red depth for this genus is much greater than 15-30 m. It

105

106

k. g. McKenzie

Table 1

Generic and Specific reassignments of Ostracoda described in Chapman’s (1914) “Mallee Bores”

Paper

Chapman’s Name

Reassignment

NMV Reg No

1.

Cytherella pulchra G. S. Brady

-Cytherella [non pulchra]

PI 2539

2.

Cytherella polita G. S. Brady

= Cytherella [non polita ]

P12538

3.

Cytherella muricultus Chapman

= Cytherelloidea Alexander 1929

PI 2536-7

4.

Cytherella lata G. S. Brady

-Cytherella [non lata]

P12535

5.

Cytherella subtruncata Chapman

- no change

P12541

6.

Cytherella punctata G. S. Brady

= Platella Coryell & Fields 1937 [non punctata]

PI2540

7.

Cytherura ouyenensis Chapman

= Loxocythere Hornibrook 1952

PI2529

8.

Cytheropteron postumbonatum Chapman

= Bythoceratina Hornibrook 1952

P12532

9.

Cytheropteron reticosum Chapman

= no change

PI2534

10.

Cytheropteron batesfordiense Chapman

= n. gen.

PI 2531

11.

Cytheropteron batesfordiense aculeata Chapman

= n. gen. [same sp. as 10]

PI 2530

12.

Cytheropteron praeantarcticum Chapman

= Oculocytheropteron Bate 1972

PI 2533

13.

Cytheropteron rostratum Chapman

= non Cytheropteron

P12553

14.

Cythere rastromarginata G. S. Brady

= Cletocvthereis Swain 1963

PI2518

15.

Cythere scabrocuneuta G. S. Brady

= Trachvleberis Brady 1898 [male, non

scabrocuneataj

PI 2520

16.

Cvthere sdniil/u/atu G. S. Brady

= Parakrithe van den Bold 1946 [non scintillulata]

P12519

17.

Cy there scutigera G. S. Brady

= Trachvleberis [non scutigera]

PI 2521

18.

Cy there wyvillethomsoni G. S. Brady

= n. gen. [non wyvillethomsoni]

PI 2522

19.

Krithe eggeri Chapman

= Parakrithe [same sp. as 16]

PI 2523

20.

Loxoconcha australis G. S. Brady

= no change

PI 2524

21.

Xestoleberis curt a G. S. Brady

= Xestoleberis [non curia]

PI 2525

22.

Xestoleberis marginata G. S. Brady

= Xestoleberis [non marginata]

PI 2526

23.

Xestoleberis varieguta G. S. Brady

- [slide empty]

PI 2527

24.

Cytherura capillifera Chapman

= n. gen.

PI 2528

25.

Cy there dictyon G. S. Brady

= Bradleya Hornibrook 1952 [non dictyon]

PI 2507

26.

Cythere flexicostata Chapman

= n. gen.

PI 2508

27.

Cvthere lactea G. S. Brady

= Tenedocythere Sissingh 1972 [non lacteal]

PI 2509

28.

Cythere lepralioides G. S. Brady

= Cytheralison Hornibrook 1952 [non lepralioides]

[juv.]

PI 2510

29.

Cythere lubbockiana G. S. Brady

= ?Keijia Teeter 1975 [non lubbockiana]

P12511

30.

Cythere mi li tar is G. S. Brady

= Ponticocytherds McKenzie 1967 [ah', clavigera

G. S. B.]

PI 2512

31.

Cvthere normani G. S. Brady

= Quasibradleya Benson 1972 [non normani]

PI 2513

32.

Cvthere obtusalata G. S. Brady

= Loxoconcha Sars 1866 [non obtusalata]

PI 2514

33.

Cythere ova Us G. S. Brady

- Cytheralison [non ovalis] [adult male of 28]

PI 2515

34.

Cythere parallelogramma G. S. Brady

= n. gen. [same sp. as 26] [non parallelogramma]

PI 2516

35.

Cvthere postdeclivis Chapman

= Cytheralison [adult male, same sp. as 28, 33]

PI 2517

36.

Macrocvpris decora G. S. Brady

= Tasmanocypris McKenzie 1979 [non decora]

PI 2496

37.

Macrocvpris tumida G. S. Brady

= Maddocksella [non tumida]

PI 2497

38.

Bvthocvpris tumefacta Chapman

= Maddocksella [same sp. as 37]

PI 2498

39.

Bvthocvpris tumefacta Chapman

= Maddocksella [same sp. as 37, 38]

PI2499

40.

Bairdia amvgdaloides G. S. Brady

- Neonesidea Maddocks 1969 [non amygdaloides]

PI 2500

41.

Bairdia australis Chapman

= Neonesidea

PI 2501

42.

Cvthere canaliculata Reuss

= Callistocythere Ruggieri 1953 [non canaliculata]

PI 2502

43.

Cythere crispata G. S. Brady

= pectocytherid n. gen. [non crispata]

PI 2503

44.

Cythere dasyderma G. S. Brady

= Cytheralison [female, same sp. as 28, 33, 35]

PI 2504

45.

Cythere demissa G. S Brady

= Keijia [non demissa]

PI 2505

46.

Cythere dictyon G. S. Brady

= Trachvleberis [non dictyon] [female, same sp.

as 15]

PI 2506

Notes: 1. For description of Maddocksella see text.
2. juv. = juvenile.

CHAPMAN’S OSTRACODA

107

Table 2

Generic and Specific reassignments of Ostracoda described in Chapman, Crespin and Keble (1928) —the “Sorrento Bore”

Paper.

Chapman’s and Crespin’s Name

Reassignment

NMV Reg No

1.

Cythere sorrentae Chapman & Crespin

= Tenedocythere (?] [juv.]

PI4431

2.

Cvthere caudispinosa Chapman & Crespin

= Oertliella Pokorny 1964 [?]

PI4432

3.

Cvthere baragwanathi Chapman & Crespin

= Osticythere Hartman 1980

P14433

4.

Bythocythere keblei Chapman & Crespin

= n. gen. [A-l juv.]

PI4434

5.

Cytherura praemucronata Chapman & Crespin

= Pokornyella Oertli 1956 s. 1.

PI 4435

6.

Cytherella sulcosa Chapman & Crespin

= no change

PI4436

7.

Cytherella intermedia Chapman & Crespin

= Cytherelloidea

PI 4437

8.

Cytherella araneosa Chapman & Crespin

= same sp. as 6

PI4438

Note: juv. = juvenile.

Fig. 1 — Maddocksella tumefacta (Chapman 1914).
a, A-l juvenile left valve, internal view (outline) NMV 12499
lectoparatype; b, adult carapace, external view NMV 12498 lec-
totype; c, adult right valve, internal view NMV 12497 lec¬
toparatype. Note: normal pore canals not illustrated.

is thus a useful shallow water marine and inshore index
in the Cainozoic of Australia.

ACKNOWLEDGEMENTS

I am grateful to Dr Peter Jell, Curator of Invertebrate
Fossils, for access to the collections and use of facilities
at the National Museum of Victoria; and to Mr A.
Tynan, Hancock Museum, Newcastle-upon-Tyne, for
access to G. S. Brady’s types. Mrs Margaret Nichol
typed the manuscript. The research is supported by
ARGC Grant No. E7615127.

REFERENCES

Jrady, G. S., 1880. Report on the Ostracoda dredged by
H.M.S. Challenger during the years 1873-76. Challenger
Reps, Zool. 1 (3): 1-184. pi. 1-44.

Chapman, F., 1914. Description of new and rare fossils ob¬
tained by deep boring in the Mallee. Pt. 3, Ostracoda to
fishes. Proc. R. Soc. Via. 27: 28-71, pi. 6-10.

Chapman, F., Crespin, I., & Keble, R. A., 1928. The Sor¬
rento Bore, Mornington Peninsula, with a description of
new or little-known fossils. Rec. geol. Surv. Viet. 5 (1):

I- 195, pi. 1-12.

Maddocks, R. F., 1969. Recent ostracodes of the family Pon-
tocyprididae chiefly from the Indian Ocean. Smithson. Con-
tribs Zool. 1: 1-56, figs. 1-35.

McKenzie, K. G., 1964. The ecologic associations of an
ostracode fauna from Oyster Harbour, a marginal marine
environment near Albany, Western Australia. Pubbl. staz.
zool. Napoli 33 (suppl.): 421-461.

McKenzie, K. G., 1967. Recent Ostracoda from Port Phillip
Bay, Victoria. Proc. R. Soc. Viet. 80: 61-106, figs 1-10, pi.

II- 13.

McKenzie, K. G., 1969. Discussion. In The Taxonomy, Mor¬
phology and Ecology of Recent Ostracoda, J. W. Neale, ed.,
Oliver and Boyd. Edinburgh, 11-13.

McKenzie, K. G., 1974. Cenozoic Ostracoda of southeastern
Australia with the description of Hanaiceratina new genus.
Geoscience and Man 6: 153-182.

McKenzie, K. G., 1979. Appendix 2. Notes on Ostracoda
from Willunga Embayment Boreholes WLG 38, WLG 40
and WLG 42. In: B. J. Cooper “Eocene to Miocene
Stratigraphy of the Willunga Embayment”. S. Aust. geol.
Surv., Rep. Investigations 50: 90-101.

STUDIES ON WESTERN AUSTRALIAN PERMIAN

BRACHIOPODS

2. THE FAMILY RUGOSOCHONETIDAE MUIR-WOOD 1962

By N. W. Archbold

Department of Geology, University of Melbourne, Parkville, Victoria 3052

Abstract: Representatives of the family Rugosochonetidae (Chonetidina, Brachiopoda) from the
Western Australian Permian sequences are documented. The genus Neochonetes Muir-Wood 1962 is dis¬
cussed with the establishment of the new subgenus Neochonetes (Somrneria ). The following species are
revised or described: Neochonetes ( Somrneria) pratti (Davidson), Neochonetes (Somrneria) robustus sp.
nov., Neochonetes ( Somrneria) tenuicapillatus sp. nov., Neochonetes ( Somrneria) afanasyevae sp. nov.,
Svalbardia narelliensis sp. nov. Additional material indicating the presence of further species of
Neochonetes (Somrneria) and a species of Chonetinella is described.

INTRODUCTION

Brachiopods of the suborder Chonetidina are abun¬
dant in the diverse faunas of the west Australian Per¬
mian sedimentary sequences. Representatives of the
Anopliidae have been documented elsewhere (Archbold
1980a). This paper concludes the investigation of
members of the Rugosochonetidae, some results of
which have already appeared (Archbold 1981a, 1981b).
Neochonetes Muir-Wood, in the form of the new
subgenus Neochonetes (Somrneria), is particularly abun¬
dant at certain stratigraphical levels and species of the
genus are useful for intrabasinal correlations.

Neochonetes (Somrneria) has a disjunct or bipolar
distribution. Such a distribution for Permian chonetids
was anticipated by Afanas’yeva (1978) and has been
demonstrated for Tornquistia , Svalbardia and Quin-
quenella (Archbold 1980a, 1981a, 1981b).

Collections

All figured and measured specimens are housed in the
following institutions as indicated by the prefix to
registered numbers. CPC —Commonwealth Palaeon¬
tological Collections of the Bureau of Mineral
Resources, Geology and Geophysics, Canberra, A.C.T.
GSWA —Geological Survey of Western Australia,
Perth, Western Australia. NMVP —National Museum
of Victoria, Melbourne, Victoria. UWA —Geology
Department, University of Western Australia,
Nedlands, W.A. SM —South Australian Museum,
Adelaide, S.A.

Stratigraphy

Marine Permian sedimentary sequences occur in the
Perth, Carnarvon, Canning and Bonaparte Gulf Basins
(Fig. 1) with representatives of the Rugosochonetidae in
each. The sequence of the Perth Basin was revised by
Playford et al. (1976), those of the Carnarvon and Can¬
ning Basins were reviewed by Playford et al. (1975) with
the latter revised by Yeates et al. (1975) and Crowe and
Towner (1976), and the Permian succession of the
Bonaparte Gulf Basin was revised by Dickins et al.
(1972).

Age

The biostratigraphic scheme of Glenister & Furnish
(1961) based on ammonoids is still largely followed for
Western Australia with one important exception. Those
authors were unable to recognise the distinction of the
Kungurian stage. Dickins (1976) attempted correlation
of the Western Australian Permian sequences with the
International Time Scale and he indicated the
Baigendzinian-Kungurian boundary at about the level of
the Baker Formation of the Carnarvon Basin. Dickins
(1956) and Waterhouse (1976) favoured a slightly lower
position for the boundary at the base of the Baker For¬
mation and Archbold (1981b) suggested that the boun¬
dary may be within the Nalbia Greywacke below the
Baker Formation. This latter suggestion is compatible
with information on ammonoids provided by Bogoslov¬
skaya (1976) and Cockbain (1980).

Terminology

The terminology applied to the Chonetidina has been
clearly defined by Muir-Wood (1962, 1965) and
Sarycheva (1970).

Palaeoecology of Massed Chonetid Occurrences

Western Australian chonetids occur in two types of
coquinites, those where the individuals are disarticulated
and invariably worn and those where the individuals are
conjoined with excellent preservation of fine external,
surface ornament.

Specimens of Svalbardia narelliensis sp. nov. occur by
the thousands in large slabs of rock from high in the
Noonkanbah Formation, Canning Basin. All specimens
are disarticulated, many are worn, and other fossils are
restricted to one or two isolated individuals. With few'
exceptions all ventral valves are convex up; dorsal valves
show no preferred orientation. Elias (1962, 1966) sug¬
gested: 1, that chonetids may be gregarious and when
they occur in great numbers, to the virtual exclusion of
other invertebrates, they indicate waters shallower than
normal for articulate brachiopods; 2, that nested ventral
valves possibly indicate deposition in gently agitated
water, above wave base. One such valve in valve ar-

109

110

N. W. ARCHBOLD

rangement of ventral valves concave up for Svalbardia
narelliensis sp. nov. is figured herein (Fig. 2Y). Ac¬
cumulations of Neochonetes ( Sommeria ) afanasyevae
sp. nov. are also of disarticulated valves in the three
basins in which the species occurs. The faunas of the
three formations containing the species are molluscan
dominant, which, in the scheme of Thomas (1958) are
shallower water assemblages than brachiopod dominant
assemblages.

On the other hand, accumulations of Neochonetes
(Sommeria) in the Callytharra, Madeline and Wandagee
Formations are often of articulated shells suggesting
deposition in water below wave base. The same is true of
chonetid conquinites of Tornquistia magna from the

Bulgadoo Shale. These formations all possess
brachiopod dominated assemblages inferred to reflect
deeper water (Thomas 1958).

SYSTEMATIC PALAEONTOLOGY

Suborder Chonetidina Muir-Wood 1955
Superfamily Chonetacea Bronn 1862
Family Rugosochonetidae Muir-Wood 1962
Subfamily Svalbardiinae Archbold 1981

Genus Svalbardia Barkhatova 1970
Type Species: Chonetes capitolinus Toula 1815.
Diagnosis: See Archbold (1981b, p. 3).

Discussion: Svalbardia has been discussed by Archbold
(1981b).

Svalbardia narelliensis sp. nov.

Fig. 2

Holotype: CPC19167U, a dorsal valve.

Etymology: From Narelli Rockhole, Canning Basin,
Western Australia.

Location and Occurrence: Upper Noonkanbah For¬
mation, Canning Basin.

The following B.M.R. localities yielded several hum
dred specimens of Svalbardia narelliensis which were ex¬
amined for the present investigation. Additional
localities in the Canning Basin that have yielded
Svalbardia are documented in Dickins and Jell (1974)
who identified the species as a representative of
Neochonetes . L.3., Lai. 20°08'S, Long. 127°58'E, 4.8
km west of Balgo Mission; L.101., Lat. 20°08'S, Long.
127°58'E, 4.8 km west of Balgo Mission, L.610., Lat.
20°09'S, Long. 127°56'E, 1.2 km south of Narelli
Rockhole.

Measurements: The specimens measured for Table 1
are paratypes from L610 and have the registered
numbers CPC 19167A to 19167Z; CPC19168A to
19168E.

Description: external. Shell length is about three-
quarters maximum width which occurs about mid¬
length of the shell. The ventral valve is convex, usually
without a sulcus, although a slight flattening or a very
shallow sulcus is visible on rare specimens. The anterior
commissure is not flexed. The dorsal valve is normally
flat or rarely gently concave. The interareas are narrow,
that of the ventral valve being striate parallel to its
width. The exterior of the shell is smooth except for con¬
centric growth lines. Worn ventral valves exhibit a sur¬
face feature of pits arranged in concentric rows formed
by exposure of the taleolae. Worn dorsal valves, in
which growth lines may still be visible, exhibit a pseu-
docapillate exterior. Hinge spines are worn on all ex¬
amined specimens. However, spine base canals 1 to 1.5

Table 1

Size Ranges of Populations of Svalbardia narelliensis sp. nov. (mm)

Maximum Width
12.4-20.5

Hinge Width
8.6-19.0

Ventral Length
10.9-15.7

Dorsal Length
8.9-11.6

Thickness
2.7- 4.3

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 111

mm apart are clearly visible in the ventral interarea of
several specimens. Where traces of spines remain they
indicate the spines emerged at a low angle to the hinge.
internal, the ventral median septum is over three-
quarters valve length and arises under the delthyrium as
a high pronounced structure. One ventral valve
(CPC19167H) possesses two parallel, weakly impressed
striations adjacent to the septum. Vascular trunks and
muscle scars are poorly impressed. Specimen
CPC19167G exhibits a pair of small adductor scars adja¬
cent to the posterior third of the median septum. The
teeth are small and sharp.

The dorsal median septum is just over two-thirds
valve length and is higher and broader posteriorly. The
alveolus is small but distinct and, at times, deep. The
lateral septa are stout and arise anterior to the alveolus,
where they are fused with the median septum to form a
low platform. The sockets are deep with pronounced in¬
ner socket ridges. The brachial ridges are indistinct and
their anterior recurved portions are only slightly raised
in mature individuals. The cardinal process is low but
distinct, externally quadrilobed, internally bilobed. The
interior of the dorsal valve is finely papillose with the
anterior papillae being arranged in radiating rows.

Delthyrial structures are poorly known, however, a
minute pseudodeltidium is present in the apex of the
delthyrium of specimen CPC19167G. None of the
specimens exhibit a chilidium.

Discussion: Svalbardia narelliensis is morphologically
similar to Svalbardia thomasi Archbold (1981b) from
the Baker Formation and Nalbia Greywacke of the Car¬
narvon Basin. S. narelliensis possesses, in mature in¬
dividuals, only weakly raised anterior recurved portions
of the brachial ridges. In other respects S. narelliensis is
similar to S. thomasi and they are of approximately the
same age. S. narelliensis attains a larger size than S.
thomasi. S. thomasi was compared with many boreal oc¬
currences of the genus by Archbold (1981b); the boreal
species usually being larger and possessing prominently
raised anterior recurved portions of the brachial ridges.
Age: Svalbardia in the Nalbia Greywacke and Baker
Formation (Archbold 1981b) indicates an earliest
Kungurian age. Svalbardia occurs high in the Noonkan-
bah Formation, beneath the Middle Kungurian Light-
jack Formation (sec discussion under Neochonetes
(Sommeria) afanasyevae sp.nov.). It therefore appears
that the Noonkanbah Formation contains Baigendzi-
nian and earliest Kungurian faunas.

Table 2

Representative Reports of Species of Neochonetes Muir-Wood Indicating Time Range
and Geographical Distribution of Stocks within the Genus.

1. Stock of Neochonetes (Neochonetes) carboniferus
(Keyserling 1846)

Chonetes sarcinulata var. carbonifera Keyserling 1846;
North Urals; Late Carboniferous. Neochonetes car¬
bon iferus Afanas’yeva 1975 b; Russian Platform;
Kasimovian-Gzhelian. Neochonetes? donetzianus
Afanas’yeva 1975b; Donetz Basin; Middle Car¬
boniferous. Chonetes carboniferus Semenova 1972;
Kuibyshev Region; Bashkirian. Chonetes ex gr. car¬
boniferus Lapina 1958; Kharaulakh Mts, NE USSR;
Late Carboniferous. Neochonetes acanthophorous
Winkler-Prins 1968; Cantabrian Mts Spain; Bashkirian.
Chonetes pseudovariolatus Loczy 1897; Kansu Pro¬
vince, China; Late Carboniferous. Chonetes
pseudovariolatus Schellwien 1911; Nth Nan-Shan,
China; Late Carboniferous. Chonetes carbonifera Chao
1928; China, widespread; Middle & Late Carboniferous.
Neochonetes puanensis Liao 1979; Western Guizhou
Province, China; Gzhelian. Chonetes cf. carboniferus
Ozaki 1934; Korea; Late Carboniferous. Neochonetes
permicus Grushenko 1975; Donetz Basin; Asselian.

2. Stock of Neochonetes (Neochonetes) granulifer
(Owen 1852)

Chonetes granulifer Owen 1852; Iowa, U.S.A.; Penn¬
sylvanian. Chonetesdominus King 1938; Texas, U.S.A.;
Pennsylvanian. Chonetes granulifer. var. Dunbar &
Condra 1932; Nebraska, U.S.A.; Pennsylvanian.
Neochonetes granulifer Spencer 1970; Kansas, Missouri,

U.S.A.; Pennsylvanian-Sakmarian. Neochonetes gra¬
nulifer Brand 1970; England; Late Carboniferous.
Chonetes sp. C Brand 1970; Scotland; Late Car¬
boniferous. Chonetes (Chonetinella) granulifer Boger &
Fiebig 1963; Germany; Stephanian. Leptaena variolata
d’Orbigny 1842; Bolivia; Asselian-Sakmarian. Chonetes
variolatus Thomas 1930; Peru; Asselian-Sakmarian.

3. Stock of Neochonetes (Sommeria) pratti (Davidson
1859)

Chonetes pratti Davidson 1859; Western Australia;
Sakmarian. Chonetes arabicus Hudson & Sudbury 1959;
Oman Peninsula; Sterlitamakian. Neochonetes vario¬
latus Termicr et al. 1974; Afghanistan; Sterlitamakian.
Neochonetes variolatus Fantini Sestini 1965b;
Karakorum; Sakmarian? Neochonetes forbesi Czar-
niecki 1969; Spitzbergen; Asselian-Sakmarian. Chonetes
variolatus Zavodovsky & Stepanov 1971; Kolyma,
USSR; Asselian. Neochonetes tschernyschewi
Barkhatova 1970; Urals; Sakmarian. Neochonetes
fredericksi Archbold 1979; Pechora Basin, USSR; Ar-
tinskian. Chonetinasuperba Gobbett 1964; Spitzbergen;
Late Kungurian. Chonetina? cf. C. superba Brabb &
Grant 1971; Alaska; Late Kungurian. Neochonetes sp.
Bamber & Waterhouse 1971; Yukon; Late Kungurian.
Neochonetes asseretoi Fantini Sestini 1964; Iran; Late
Kungurian-Kazanian. Chonetes wageri Muir-Wood
1941; Himalaya; Chhidruan. Neochonetes (Sommeria)
sp. herein; Western Australia; Chhidruan.

112

N. W. ARCHBOLD

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 113

Subfamily Rugosochonetinae Muir-Wood 1962
Genus Neochonetes Muir-Wood 1962
(= Quaciranetes Sadlick 1963)

Type Species: Chonetes dominus King 1938
Diagnosis: Neochonetes includes medium to large sized
weakly to moderately concavo-convex, rugosochone-
tinids with finely capillate external ornament and a
feebly to distinctly developed sulcus; dorsal interior with
distinct alveolus, lateral septa, median septum and
brachial ridges; inner socket ridges well developed and
outer socket ridges may be present. Ventral valve with
prominent septum, two parallel vascular trunks forming
prominent ridges, and hinge spines at a low to moderate
angle to hinge line.

Discussion: The long time range and large number of
species undoubtedly account for the rather broad
generic diagnosis. Within Neochonetes several groups or
stocks may be recognised and generalised evolutionary
trends defined.

One of the early stocks of Neochonetes , arising in the
Bashkirian, characterised by hinge spines at a low angle
(may curve to higher angle distally), a low convexity of
the shell, a moderately transverse outline often with
small ears, the hinge line usually being the position of
maximum width, and an obsolescent sulcus, can be
referred to as the group of Neochonetes carboniferus
(Table 2).

Closely allied to the Neochonetes carboniferus stock
•s the group of Neochonetes granulifer (Table 2). This
stock also possesses hinge spines at a low angle (one ex¬
ception being the Late Mississippian Neochonetes
oklahomensis (Snider) as redescribcd by Branson (1964)
which has hinge spines at 45°) a moderate to low con¬
vexity and a broad shallow sulcus, the latter usually
more strongly developed than in the Neochonetes car¬
boniferus stock. Younger variants of the Neochonetes
granulifer stock are often, but not invariably, more
transverse than older variants and possess hinge spines
at a greater angle to the hinge. South American occur¬
rences of the stock remain poorly understood despite at¬
tempts at the redescription of Neochonetes variolatus
(d’Orbigny) by Koninck (1847) and Kozlowski (1914).
The binomen Chonetes variolatus has at times been ap¬
plied to members of the Neochonetes carboniferus slock
(Gorskii & Timofeeva 1950) indicating a degree of con¬
fusion as to the nature of Neochonetes variolatus. The
°nly representative of Neochonetes from the Late Per¬
mian of New Zealand (Waterhouse 1964, 1976), a large
species with hinge spines at a low angle and a gentle

broad sulcus may be a migratory descendent of the N.
granulifer stock. Waterhouse (1964) compared the New
Zealand N. beatusi with several species of the N.
granulifer stock. A species in the Bowen Basin (Dear
1971) possesses a maximum width at midlength of the
shell and a greater convexity of the ventral valve.

A third group within Neochonetes is the Neochonetes
pratti stock here named Neochonetes (Sonuneria). This
stock is characterised by large size, ventral hinge spines
at an angle of about 40° to 45°, often a prominent
sulcus and a maximum width invariably anterior of the
hinge.

Boreal species of the stock replace the older N. car¬
boniferus stock in the northern Urals during the
Sakmarian (Barkhatova 1964). Few boreal species are
well documented, exceptions being the early Artinskian
Pechora Basin species N. fredericksi Archbold 1979 and
the Sakmarian N. tschernyschewi Barkhatova 1970 from
the Urals.

During the Kungurian and subsequent stages,
Neochonetes again underwent subtle changes in mor¬
phology, but stocks are not clear because of the paucity
of documented species. Small species of Neochonetes
from the Zechstein of Germany and England
(Schauroth 1856, Trechmann 1944) and the Kazanian of
Armenia (Sokolskaya 1965) exhibit a trend to weakly
developed ornament and an obsolescent sulcus. Discus¬
sions and illustrations of the Zechstein species (David¬
son 1880, Malzahn 1957 and Muir-Wood 1962) indicate
that they belong to a separate stock within Neochonetes.

Other representatives of Neochonetes of Kungurian
and younger age do not exhibit obsolescent ornament or
an obsolescent ventral sulcus, although the species may
be small in size. Chonetes pinegensis Kulikov (1974)
from the Kazanian of northern European USSR,
Neochonetes cf. pinegensis of Stepanov et al. from the
Kazanian of the Kanin Peninsula and Chonetes sp. of
Licharew (1913) from the late Kungurian of Kirillov are
all small species with distinct ornament and ventral
sulcus. The wide variety of morphologies of Kungurian
and younger neochonetids probably reflects greater
isolation of specific gene pools resulting from the
elimination of seaways for migratory exchanges and
from other environmental factors such as increase in
salinity (the latter undoubtedly affecting the Zechstein
stock).

Subgenus Neochonetes (Sommeria) subgen. nov.
Type Species: Chonetes prattii Davidson 1859.

Fig. 2 — Svalbardia narelliensis sp. nov.

All specimens from the Noonkanbah Formation, Canning Basin; A-C, Holotype CPC 19167U, Dorsal valve in ventral, dorsal and
posterior views. x3; D, CPC 19168B, Dorsal valve in ventral view. x3; E, CPC 19168C, Dorsal valve in ventral view. x3; F-H, CPC
19167Z, Dorsal valve in ventral, dorsal and posterior views. x3; I, CPC 19167R, Ventral valve in dorsal view. x3; J, CPC 19i68 E,
Dorsal valve in dorsal view. x3; K-M, CPC f 19167V, Dorsal valve in ventral, dorsal and posterior views. x3; N, CPC 19168D, Dor¬
sal valve in dorsal view. x3; O, CPC 19167B, Ventral valve in dorsal view. x2; P-Q, CPC 19167X, Dorsal valve in ventral and dorsal
views. x3; R, CPC 19167H, Ventral valve in dorsal view. x2.5; S, CPC 19167G, Ventral valve in ventral view. x3; T-U, CPC
19168F, Dorsal valve in ventral and dorsal views. x3; V, CPC 19168A, Dorsal valve in ventral view. x3.25; W, CPC 19167F, Ventral
valve in ventral view. x3; X, CPC 19167K, Ventral valve in dorsal view. x3.25; Y, CPC 19168G, Valve in valve arrangement, dorsal

view. x3.5.

114

N. W. ARCHBOLD

Fig. 3 — The internal morphology of Neochonetes (Sommeria)
pratti (Davidson); after Muir-Wood (1962, fig. 3, p. 13).

Etymology: Named in honour of Ferdinand von Som¬
mer, pioneer geologist in Western Australia.

Diagnosis: Similar to Neochonetes (Neochonetes) but
sulcus usually conspicuously developed, gentle fold
often developed in the dorsal valve, a greater convexity
of the ventral valve and hinge spines at about 40° to 45°.
Maximum width of mature shells usually anterior of the
hinge. Interior as for Neochonetes (Neochonetes).
Discussion. Although, as discussed above, several
stocks can be identified in the genus Neochonetes , to
give these full generic status is considered premature
because the stocks do exhibit a degree of morphological
overlap. The stock of Neochonetes pratti , outlined
above, from Western Australia and the Boreal Realm is
considered to belong to the new subgenus Sommeria.
Many of the poorly known boreal species are compared
with Neochonetes (Sommeria) pratti below. Fig. 3 shows
the internal morphology of Neochonetes (Sommeria)
pratti.

Neochonetes (Sommeria) pratti (Davidson) 1859
Figs 4, 5, 6.

1859 Chonetes prattii Davidson, The Geologist 2: 116,
pi. 4, figs 9-12.

1890 large Chonetes Etheridge, Annu. Rep. Dept. Mines
N.S.W. 1889: 239.

1892 Chonetes pratti; Newton, Geol. Mag. (3) 9: 468,
542, pi. 14, figs 1-12.

1893 Chonetes pratti; Newton, Rep. Br. Assoc. Advmt.
Set. 1892: 725.

1903 Chonetes pratti; Etheridge, Bull. geol. Surv. W.
Aust. 10: 23.

1907 Chonetes pratti; Etheridge, Bull. geol. Surv. W.
Aust. 27: 31, pi. 8, fig. 2; pi. 9, fig. 7; pi. 10, fig. 2.
1910 Chonetes pratti; Glauert partim , Bull. geol. Surv.
W. Aust. 36: 86.

1931 Chonetes pratti; Hosking, J. Proc. Roy. Soc. W.
Aust. 17: 19.

1952 Chonetes pratti; Guppy et al., 19th Int. Geol.
Cong., Alger., Sym. Gond ., 110.

1962 Neochonetes pratti; Muir-Wood, Monograph Br.
Mus. (Nat. Hist.) p. 13, pi. 11, figs 7-8.

Holotype: (by monotypy) BM(NH), BB 41082. A com-
plete shell (valves separate) Figured by Davidson (1859),
Newton (1892), and Muir-Wood (1962).

History of Discovery: The locality of the type speci¬
men, from the collection of Mr. Pratt was unknown to
Davidson (1859). Newton (1892, 1893) considered that
the type specimen came from the Irwin River, Western
Australia, after examination of specimens of a chonetid
brachiopod from that locality, particularly in view of
the nature of preservation of the specimens available to
him. He further considered the holotype part of the
Strzelecki collection but this is unlikely as Strzelecki’s in¬
vestigations and collections were made in eastern
Australia at an earlier date.

Permian sediments were discovered by Commander J.
L. Stokes between 1837 and 1843 near Port Keats
(Bonaparte Gulf, Northern Territory). Neochonetes
(Sommeria) afanasyevae sp. nov. from that locality is
not close to Chonetes pratti and is preserved as moulds
and casts, whereas the type specimen of Chonetes pratti
is a well preserved shell. Samuel Peace Pratt (1789-1863)
was an active member of the Geological Society of Lon¬
don and spent much time arranging the collections of
the Society (Woodward 1907). As a result Pratt had ac¬
cess to collections from the Irwin River sent to the Socie¬
ty by the Gregory Brothers and von Sommer (Archbold
1981c) and hence the specimen in his collection probably
comes from one of these sources.

It appears that Davidson’s original specimen probably

Fig. 4 — Neochonetes (Sommeria) pratti (Davidson).

All specimens from the Fossil ClilT Member of the Holmwood Shale, Fossil Clilf, Irwin River; A-B, GSWA F 11038, Dorsal valve in
dorsal and ventral views, xl.75; C, GSWA F 11022, Ventral valve in dorsal view x2; D-E, GSWA F 11039, Dorsal valve in dorsal
and ventral views, x1.75; F-G, GSWA F 11023, Ventral valve in dorsal and ventral views, xl.75; H-I, GSWA F 11027, Ventral valve
views, x 1.5; J-K, GSWA F 11041, Dorsal valve in ventral and dorsal views, xl.75; L-M, GSWA F 11044, Dorsal valve in dorsal and
ventral views xl.75 and 1.5; N-O, GSWA F 11018, Shell in ventral and dorsal views xl.75; P, GSWA F 11031, Ventral valve in ven¬
tral view x2; Q, GSWA F 11046, Dorsal valve in ventral view x2.5; R-S, GSWA F 11043, Dorsal valve in dorsal and ventral views x
1.75; T, GSWA F 11045, Dorsal valve in ventral view x2; U, GSWA F 11030, Ventral valve in ventral view x2.

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 115

116

N. W. ARCHBOLD

Table 3

Size ranges of Populations of Neochonetes ( Sommeria ) pratti (mm)

Stratigraphic

Maximum

Hinge

Ventral

Dorsal

Thickness

Horizon

Width

Width

Length

Length

Fossil Clift'

18.0-29.0

13.8-24.0

11.5-21.0

11.7-19.0

4.0- 8.8

Callytharra

12.0-28.0

12.0-28.0

11.5-21.2

8.3-19.0

4.0-10.0

Jimba Jimba

30.0

24.7

20.0

17.3

7.7

Nura Nura

14.4-31.4

14.4-31.4

7.0-14.0

7.8-14.0

NJ

oo

•fe.

came from Fossil Cliff on the Irwin River where the
Fossil Cliff Member of the Holmwood Shale outcrops.
Material: Perth Basin: GSWA F 11017-11046, 4 con¬
joined shells, 14 ventral valves (VV) and 12 dorsal valves
(DV) from Fossil Cliff, Irwin River, Fossil Cliff Member
of the Holmwood Shale. Carnarvon Basin: CPC
19863-19865, 19867-19868, 3VV and 2DV from BMR
locality GW74, about 1.2 km west of Callytharra Spr¬
ings, near base of Callytharra Formation; CPC 19866,
DV from BMR locality GW87, Lat. 25°32'S, Long.
115°30'E about 40 to 45 m above base of Callytharra
Formation; GSWA F 11047A-110471, one conjoined
shell, 5VV, 4DV, from GSWA locality 30137A, Glen-
burgh (1920) run 7/079, pt. 1312, RMH BK6, Yard Grid
387-803, Callytharra Formation; CPC 19869, a conjoin¬
ed shell, from BMR locality W03, Lat. 25 D 02'75"S,
Long. 114°58'E, type section of Jimba Jimba
Calcarenite. Canning Basin: CPC 19880A-19880D,
three conjoined shells and 1VV from BMR locality
KNuA, 2.4 km south-west of Paradise Homestead (Lat.
18°02'50"S, Long. 124°31'00"E), Nura Nura Member of
Poole Sandstone; CPC 19881A-19881V, three internal
moulds of conjoined shells, 5 internal moulds of VV, 2
external moulds of VV, 7 external moulds of DV and 6
internal moulds of DV, from BMR locality KPA54, 23.5
km at 120° from Mt. Tuckfield in the southern part of
the St. George Range, near base of Pool Sandstone.
Diagnosis: Large Neochonetes (Sommeria). Ventral
sulcus well developed; dorsal fold low but distinct. Ex¬
terior ornament of fine capillae, on average 4 per mm at
1 and 3 cm from umbones.

Description: External. Length of the shell is just over
two-thirds the maximum width. Maximum width about
mid-length of the shell, hinge width being less than max¬
imum width. The ventral valve is strongly convex,
especially in gerontic individuals. The sulcus is distinct,
sometimes deep and produces a flexure on the anterior
commissure of the valve. The dorsal valve is gently con¬
cave with a low median fold or flexure corresponding to
the sulcus of the ventral valve. The interareas are nar¬
row. The chilidium is small and seldom preserved; the

pseudodeltidium appears to be absent. Ornament is of
fine capillae which increase by either bifurcation or in¬
tercalation. Growth lines are distinct, variable and may
be lamellose. Spinule bases are randomly scattered on
well preserved shells. Spines along the ventral interareas
are distinct with at least six on each side of the umbo of
large individuals. The angle of emergence is about 40°.
Spines up to 1.5 mm in length have been observed but
are broken.

Internal. The ventral median septum is up to half valve
length and is high posteriorly, arising 1 to 2 mm in front
of the umbo. Anteriorly two parallel vascular trunks
form prominent ridges adjacent to the septum. The ad¬
ductor muscle scars the indistinct while the diductors are
strongly impressed. The teeth are small and sharp. With
the exception of the region of the muscle scars, which is
smooth or striate, the interior of the ventral valve is
papillose. The cardinal process is externally
quadrilobate and internally bilobed. The dorsal median
septum is two-thirds valve length and arises in front of
the deep alveolus as do the short lateral septa. The
sockets are deep with pronounced inner socket ridges
and feeble outer socket ridges. The brachial ridges are
distinct. The interior of the dorsal valve is papillose, the
papillae being arranged in radiating rows.

Discussion: A report of the species from the Mingenew.
Formation, Perth Basin, by Etheridge (1907b) has been
shown to refer to the strophalosiid Mingenewia (Ar¬
chbold 1980b). The occurrence of the species in the
Nura Nura Member, previously reported by Guppy et
al. (1952), is confirmed but several individuals from that
horizon are more transverse than specimens from the
Fossil Cliff Member or the Callytharra Formation.
Ecological factors are probably the cause of the varia¬
tion. The Nura Nura specimens were found in a coarse
sandstone, quite unlike the fine grained muddy marls
and siltstones of the more southerly occurrences.

Zavadovsky & Stepanov (1971, pi. 23, figs 8a-8b)
figured a large specimen of a Neochonetes from the Ear¬
ly Permian (Asselian) Paren horizon of the Kolyma
region, far north-east USSR, as Chonetes variolatus

Fig. 5 — Neochonetes (Sommeria) pratti (Davidson)

A-U from Callytharra Formation, Carnarvon Basin; V-W from Jimba Jimba Calcarenite, Carnarvon Basin; A-B, GSWA F
11047B, Dorsal Valve in dorsal and ventral view, x 1.75 and xl.5 respectively; C-D, CPC 19866, Dorsal valve in ventral and dorsal
views. x2; E-G, GSWA F 11047E, Shell in ventral, dorsal and anterior views, x 1.75; H-I, GSWA F 11047G, Dorsal valve in dorsal
and ventral views, xl.75; J, CPC 19864, Ventral valve in dorsal view, xl.5; K, GSWA F 11047D, Ventral valve in dorsal view.
xl.75; L-N, GSWA F 11047F, Dorsal valve in dorsal, ventral and posterior views, xl.75; O-P, GSWA F 110471, Dorsal valve in dor¬
sal and ventral views, xl.75; Q-R, GSWA F 11047C, Ventral valve in ventral and dorsal views, xl.75; S, CPC 19868, Dorsal valve in
ventral view. x2.5; T-U, CPC 19867, Dorsal valve in dorsal and ventral views. x2.5 and 2.25 respectively; V-W, CPC 19869, Shell in

ventral and dorsal views, xl.5.

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 117

118

N. W. ARCHBOLD

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 119

(d’Orbigny). It is slightly more transverse than
Neochonetes (Sommeria) pratti and possesses similar in¬
terareas. The external ornament is poorly known
because of decortication of the shell.

Czarniecki (1969) compared Neochonetes forbesi
(Gobbett), from the Asselian-Sakmarian Treskelodden
Beds and the Upper and Lower Wordiekammen
Limestone of Spitzbergen, with N. ( S.) pratti. Gobbett’s
species is, however, much smaller (one-third the size)
than the Western Australian species. Nevertheless, com¬
parison is warranted as Neochonetes forbesi possesses a
Well rounded outline (Czarniecki 1969, pi. 4, fig. 7a).
Neochonetes (Sommeria) 1 . superbus (Gobbett) from the
Middle Brachiopod Chert, Bunsow Land, Spitzbergen
of Late Kungurian age, is a large species comparable in
size with the largest specimens of Neochonetes (Som¬
meria) pratti. Neochonetes (Sommeria) 1 . superbus is,
however, a distinct species, by being more transverse.
Possessing weaker ornament and possessing a much
weaker convexity in profile.

Other reports of neochonetids from the Permian of
the USSR and Asia that indicate species related to N.
(5.) pratti are mentioned briefly. Detailed comparisons
are impossible since the species mentioned are usually
based solely on ventral valves or on poorly preserved
material. Chonetes lobata Griinewaldt (1860, pi. 3, fig.
6) from the Urals is a finely capillate species, widest
anterior of the hinge with a shallow sulcus. Chonetes
dalmanoides of Fredericks (1915, pi. 9, fig. 5) from beds
near Krasnoufimsk, European USSR, underlying those
from which in 1912 he recorded Chonetes variolatus
d’Orbigny, is a smaller species with a rounded outline
and shallow sulcus and is probably closer to
Neochonetes (Neochonetes) than N. (Sommeria).
Chonetes dereimsi Douglas 1936 of probable Kungurian
age from south western Persia possesses a similar
swollen ventral valve to that of N. (S.) pratti and is finely
capillate, but, it is half the size of adult specimens of N.
(S.) pratti. Neochonetes variolatus of Fantini Sestini
(1965b) from the Karakorum is represented by ventral
v alves with a distinct sulcus and a maximum width
anterior of the hinge line in some specimens. The early
Permian Neochonetes sp. of Acharyya et al. (1975, pi.
2, figs H, 1, J) is also similar to Neochonetes (Sommeria)
pratti in outline but appears to have less well developed
dorsal septa, and the specimens are smaller than mature
individuals of the Western Australian species and hence
may be juveniles. Similar comments can be made for the
record of two varieties of Chonetes carboniferus from
the Sakmarian of Sikkim recorded by Sahni &

Srivastava (1956, pi. 36, figs 12-16). This is a species
that possesses a sulcus invariably more weakly
developed than the Western Australian species. The
small Sterlitamakian species, Chonetes arabicus Hudson
& Sudbury (1959, pi. 3, figs 6-16 and ? pi. 4, figs 14-18)
also possesses a hinge line shorter than the maximum
width of the shell as well as a distinct sulcus, but the
Arabian species, while the same age as the Western
Australian species, is much smaller in size.

Age: Latest Tastubian to Aktastinian.

Neochonetes (Sommeria) robustus sp. nov.

Fig. 7

71912 Chonetes pratti; Glauert, Rec. W. Aust. Mus.
1:75.

1965 Chonetes pratti ; Edgell, Ann. Rept. geol. Surv.
W. Aust . 1964: 65, pi. 34, fig. 2.

Holotype: CPC 19886M from BMR Locality WB182.
Material: Paratypes: CPC 19885A-19885C, 2 external
moulds of DV and 1 internal mould of a VV, from BMR
locality WB51, 3.6 km on a bearing 225° from Keogh
Hill, Madeline Formation, Carnarvon Basin; CPC
19886A-P, 2 external moulds of DV, 1 internal mould of
a DV, 4 external moulds of VV, 3 internal moulds of VV
and 6 internal moulds of conjoined shells from BMR
locality WB182, 3.5 km bearing 358° from Monument
Bore, Madeline Formation; GSWA F 5285, 5287, 5288,

3 internal moulds of VV and 1 external mould of a VV
from hillside 4.8 km west of Arrino, Mingenew Forma¬
tion, Perth Basin.

Table 4

Size Ranges of N. (5.) robustus sp. nov. (mm)

Maximum Hinge Ventral Dorsal Thickness

Width Width Length Length

10.2-20.0 10.5-15.0 6.8-13.5 7.8-12.8 4.0-5.6

Diagnosis: Small for genus, strongly convex ventral
valve, weakly concave dorsal valve. Sulcus weak to ab¬
sent. Shell outline subquadrate, globular, length of shell
between 0.67 and 0.75 the maximum width.
Description: External. Maximum width is at about
mid length of the shell. The sulcus is either absent or
present only in the form of a gentle median flattening of
the ventral valve. The interareas are narrow. Ornament
consists of fine capillae, 4 per 1 mm on specimen CPC
19886A 9 mm from the umbo, and delicate growth lines.
Hinge spines apparently emerge at about 40° to the
hinge.

Internal. The ventral septum arises close to the umbo,

Fig. 6 —Neochonetes ( Sommeria ) pratti (Davidson)

All specimens from the Nura Nura member, Poole Sandstone, Canning Basin; A-C, CPC 19881 A, Internal mould of shell in dorsal,
ventral and anterior views. x2; D, CPC 19881R, latex replica of dorsal valve internal mould. x2; E-F, CPC 19881B, Internal mould
of shell in ventral and dorsal views. x2.25; G-I, CPC 19881C, Internal mould of shell in dorsal, ventral and postero-dorsal views.
x2, x2 and x3 respectively; J-K, CPC 1988ID and E, Latex replica of internal mould of ventral valve. x2.5, latex replica of external
mould of ventral valve. x2; L, CPC 19881U, Latex replica of dorsal valve internal mould, xl .75; M, CPC 19881H, Internal mould
of ventral valve. x3; N, CPC 19881G, Internal mould of ventral valve. x2; O, CPC 19881 I, Latex replica of internal mould of ven¬
tral valve. x2.75; P, CPC 19880B, Shell in dorsal view. x2; Q, CPC 198800, Latex replica of external mould of dorsal valve. x2; R,
CPC 19880A, Shell in ventral view. x2.5; S, CPC 19880C, Shell in dorsal view, xl.75; T, CPC 19880W, External mould of dorsal

valve. x2.5.

120

N. W. ARCHBOLD

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 121

Fig. % — Neochonetes ( Sommeria) robustus sp. nov.
Ventral spine arrangement of specimen CPC 19886 C, x8.

is high posteriorly and may extend anteriorly for three
quarters of the valve length. The vascular trunk scars
and muscle scars are usually weakly impressed but ap¬
pear normal for the genus. The delthyriutn is distinct for
the size of the specimens; teeth are small and sharp.
Margins of the interior of the ventral valve are only
weakly papillose.

The dorsal interior possesses a strong median septum

(about two-thirds valve length) a short, strong lateral
septa. All three septa arise from a low platform around
the deep, circular alveolus. The sockets are distinct with
strong inner socket ridges which coalesce with the car¬
dinal process. The cardinal process is normal for the
genus. The brachial ridges are usually ill-defined and the
anterior margin of the dorsal valve is papillose. The
papillae are randomly arranged.

Discussion: The specimens from the Mingenew Forma¬
tion are small, distinctly convex and without a sulcus.
Traces of radial capillae are present on GSWA F 5285.

One crushed ventral valve, Fig. 7T, (maximum width
12.1 mm, length 7.9 mm), which is strongly convex and
possesses a weakly developed sulcus from Locality
627/1, Lat. 14°26', Long. 129°43\ 7.2 miles (11.5 km)
north-north west of Table Hill, Port Keats district, Nor¬
thern Territory is provisionally referred to N. ( S .)
robustus sp. nov. Dickins et al. (1972) considered the
locality to be equivalent with the Noonkanbah Forma¬
tion of the Canning Basin. The associated fauna in¬
cludes Wyndhamia sp. nov. and Neospirifer sp. nov.
that are otherwise restricted to the Madeline Formation

Table 5

Measurements of Neochonetes (Sommeria) tenuicapillatus sp. nov. (mm)
* = holotype; e = estimate

Specimen

Number

Hinge

Width

Maximum

Width

Length

Ventral

Length

Dorsal

Thickness

Formation

CPC 19870

18.9

21.0

14.5

13.2

_

Bulgadoo

CPC 19871

17.2

18.6

—

11.8

—

Bulgadoo

CPC 19872

20.Oe

22. Oe

—

12.0 +

-

Bulgadoo

CPC 19873

—

19.0

—

—

5.5e

Bulgadoo

CPC 19874

22. Oe

25.Oe

13.5

12.5

4.3

Quinnanie

CPC 19875

23.3

24.6

16.9

15.0

5.3

Quinnanie

CPC 19876

25.1

30.0

19.6

—

8.5e

Cundlego

CPC 19877

25.0

27.0

17.6

—

5.8

Cundlego

NMVP 60726

26.5e

28.Oe

—

16.5

—

Wandagee

NMVP 60727

—

30.Oe

—

-

—

Wandagce

NMVP 60728

23.2

25.5

16.8

15.6

4.2

Wandagee

NMVP 60729

24.0

25.0

12.5

-

-

Wandagee

NMVP 60730

18.5

24.0

14.0

12.6

—

Wandagee

NMVP 60731*

18.9

22.5

15.7

14.0

6.3

Wandagee

NMVP 60732

17.8

20.0

13.5

11.3

4.5

Wandagee

NMVP 60733

14.0

16.4

11.0

9.5

—

Wandagee

NMVP 60734

12.7

15.3

9.8

8.7

—

Wandagee

CPC 19879

17.7

19.7

14.0

12.5

5.0

Wandagee

NMVP 60735

22.0 +

27.8

18.5

16.5

8.0

Noonkanbah

NMVP 60736

24. Oe

27.Oe

16.0

13.5

—

Noonkanbah

CPC 19878

15.Oe

18.0e

12.5

11.5

4.0

Nalbia

Fig. 1 — Neochonetes ( Sommeria) robustus sp. nov.

A-M, O-P, from the Madeline Formation, Carnarvon Basin; N, Q-S from equivalents of the Mingenew Formation, Perth Basin;
A-B, CPC 19886E, Internal mould of shell in dorsal and ventral views. x3 and x2 respectively; C-E, Holotype, CPC 19886M, Inter¬
nal mould of shell in lateral profile, dorsal and ventral views. x2.5, x3.25 and x2.5 respectively; F-G, CPC 198860, Internal mould
of shell in ventral and dorsal views. x2.5 and x3.5 respectively; H-I, CPC 19886K, Internal mould of shell in dorsal and ventral
views. x3.5 and x2.5 respectively; J, CPC 19886J, Latex replica of dorsal valve internal mould. x3.25; K, CPC 19886A, Latex
replica of ventral valve external mould. x2.5; L-M, CPC 19885 B, External mould of dorsal valve and latex replica. x2.5; N, GSWA
F 5287, Internal mould of ventral valve. x3; O, CPC 19885E, Latex replica of dorsal valve external mould. x4; P, CPC I9886H,
Latex replica of dorsal valve external mould. x3; Q-R, GSWA F 5285, External mould and internal mould of ventral valve. x4; S,
GSWA F5288, Internal mould of ventral valve. x3.5; T, CPC 19169, Decorticated ventral valve in ventral view. x3; Specimen from

Locality 627/1, Port Keats Group, Bonaparte Gulf Basin.

122

N. W. ARCHBOLD

Table 6. Size Ranges of Populations of Neochonetes (Sommeria) afanasyevae sp. nov. (mm)

Maximum

Hinge

Ventral

Dorsal

Formation

Width

Width

Length

Length

Thickness

Port Keats (L.M)

15.0-22.4

12.0-21.0

10.0-14.5

10.2-13.5

—

Lightjack

8.2-28.2

10.2-27.8

6.2-12.5

10.2-17.5

1.3-1.6

Coolkilya

15.8-22.2

15.4-20.2

12.5-15.2

11.2-15.0

2.1-2.5

Total Range

9.2-28.2

10.2-27.8

6.2-15.2

10.2-17.5

1.3-2.5

indicating an Early Baigendzinian (Early Late Artin-
skian) age for the locality.

N. (S.) robustus sp. nov. is retained in N. (Sommeria)
on the basis of the angle of the ventral hinge spines (Fig.
8) despite the distinctive shape, convexity and general
lack of sulcus.

An undescribed species from the Sirius Formation of
the Bowen Basin, Queensland, (Dear 1972) appears
closest to the Western Australian species.

Age: Early Baigendzinian.

Neochonetes (Sommeria) tenuicapillatus sp. nov.
Fig. 9

1915 Chonetes pratti ; Etheridge, Bull. geol. Surv. W.
Aust. 58: 36.

1918 Chonetes prattei ; Etheridge, Proc. R. geog. Soc.
Aust. Sth. Aust. Brch. 18: 25.

Holotype: NMVP 60731 from locality WC(32)1.
Material: Paratypes: (all from Carnarvon Basin) CPC
19870-19873, 2 external moulds of DV, 1 internal mould
of a DV and l internal mould of a VV from BMR
Locality MG236 (Reg. No. F20836), 3.2 km S.E. of
Donnelly’s Well about 33 m above base of the Bulgadoo
Shale; CPC 19874-19875, 2 conjoined shells from BMR
Locality ML55, N of Minilya River, 53 m above base of
Quinnanie Shale; CPC 19876-19787, 2 conjoined shells,
from BMR Locality F20875, approximately 14.6 km
south-east of Middalya Homestead, 0.2 km north-west
of well, Cundlego Formation; CPC 19879, a conjoined
shell, from BMR Locality ML58, north bank of Minilya
River, 40-44 m above base of Wandagee Formation,
Lat. 23°44'S, Long. 114°25'E; NMVP 60726-60734, five
conjoined shells, 1 VV, 1 external mould of a DV and 1
internal mould of a DV (collected by Dr. C. Teichert
(1938 to 1940) on Wandagee Station on the Minilya
River. (WC = Wandagee Series, Calceolispongia Stage)
followed by the zone number (in brackets) and then the
locality number. NMVP 60726-60727 from WC(36);
NMVP 60728-60729 from WC(22-25)5, NMVP 60730
from WC(22-25)30; NMVP 60731 from WC(32)1;

NMVP 60732 from WC(27-32)3; NMVP 60733-60734
from WC(22-25)12. CPC 19878, a conjoined shell, from
BMR Locality MG247, small syncline about 4 km NNW
of Wandagee Hill Trig, Nalbia Greywacke. Canning
Basin: NMVP 60735-60736, 2 conjoined shells, from
Ml. Marmion, Noonkanbah Formation, Canning Basin.
Description: External. The length of the shell is be¬
tween two-thirds and three quarters the maximum
width. The maximum width is about mid-length of the
shell and is consistently greater than the hinge width.
The sulcus is distinct, broadens anteriorly and is match¬
ed by a low fold on the dorsal valve. The ventral valve is
strongly convex and the dorsal valve concave. Interareas
are low. The chilidium, seldom preserved, is small and
curved around the base of the cardinal process. The
pseudodeltidium, also seldom preserved, is low and seals
the apex and sides of the delthyrium. The external orna¬
ment consists of concentric growth lines, occasionally
lamellose, and very fine capillae. Very slight erosion of
the shell (though growth lines may still be visible)
obscures the capillae. Spinule bases are minute and ran¬
domly distributed when preserved. Hinge spines emerge
at about 40° to the hinge and are short.

Internal. The ventral septum arises close to the um¬
bo and is up to half the valve length. Parallel vascular
trunks and muscle scars are usually weakly impressed.
The teeth are small and sh£rp; the delthyrium small. The
margins of the interior of the ventral valve are papillose.

The dorsal interior possesses a median septum which
is between half and two-thirds the valve length. The
lateral septa are short and strong and all three septa arise
anteriorly of a distinct, circular alveolus. The sockets
are deep with pronounced inner socket ridges and feeble
outer socket ridges. The inner socket ridges coalesce
with the cardinal process which is normal for the genus.
The brachial ridges are not distinct, even in mature
specimens. The anterior of the interior of the dorsal
valve can be thickly papillose; the papillae are randomly
arranged.

Discussion: Neochonetes (Sommeria) tenuicapillatus is

Fig. 9 —Neochonetes (Sommeria) tenuicapillatus sp. nov.

A-D from the Bulgadoo Shale, Carnarvon Basin; E-G from the Quinnanie Shale, Carnarvon Basin; H-J from the Cundlego Forma¬
tion, Carnarvon Basin; K-Q, T-V from the Wandagee Formation, Carnarvon Basin; specimens R, S, W, from the Noonkanbafe
Formation, Canning Basin; specimen X-Y from the Nalbia Greywacke, Carnarvon Basin; A, CPC 19872, Latex replica of dorsal
valve internal mould. x2; B, CPC 19870, Latex replica of dorsal valve external mould, x2; C, CPC 19873, Internal mould of ventral
valve. x2; D, CPC 19871, Latex replica of dorsal valve external mould, x 1.75; E-F, CPC 19874, Shell in dorsal and ventral views,
xl.75; G, CPC 19875, Shell in dorsal view, xl.75; H, CPC 19876, Shell in ventral view, xl.25; 1-J, CPC 19877, Shell in ventral and
posterior views, xl.5; K-M, Holotype NMV P60731, Shell in dorsal, ventral and posterior views. x2, x2 and x 1.5 respectively; N.
NMV P 60732, Shell in ventral view. x2; O, NMV P 60727, Latex replica of dorsal valve internal mould, xl.5; P, NMV P 60726.
Dorsal valve external mould, xl.5; Q, NMV P 60730, Shell in dorsal view, xl.75; R-S, NMV P 60735, Shell in ventral and dorsal
views, xl.5; T, NMV P 60734, Shell in ventral view. x2.5; U-V, NMV P 60733, Shell in ventral and dorsal views. x2.25; W, NMV P
60736, Incomplete shell in dorsal view, x 1.75; X-Y, CPC 19878, Shell in dorsal and ventral views. x2 and x2.5 respectively.

'• 35v/«r-

;;*#• • ^*A£ r .X,

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 123

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 125

Fig. 11 — Neochonetes ( So/nmeria) afanasyevae sp. nov.
Ventral spine arrangement of specimen CPC 19887A, x8.

chiefly distinguished from N. (S.) pratti by the very fine,
delicate external ornament. Brachial ridges and ventral
muscle scars do not appear to be as strongly developed
as in N. (S.) pratti, although these features are probably
not of specific importance.

N. (S.) tenuicapillatus is closest to Neochonetes (Som-
tneria) fredericksi Archbold (1979) from the Artinskian
(probably Aktastinian) of the Pechora Basin. The latter
species has a length about two-thirds the maximum
width of the shell and about 5 capillae per 1 mm.

Chonetes dubia Hamlet (1928) from Bitauni and Noil
Toko, Timor, of Late Baigendzinian age is similar in
outline, size and convexity to Neochonetes ( Sommeria)
tenuicapillatus , but differs by possessing higher in¬
terareas and a more prominent ventral umbo. Chonetes
dubia exhibits weakly developed radial ornament
(Hamlet 1928, pi. 1, figs 12, 14b).

Range: Late Baigendzinian to Early Kungurian.

Neochonetes (Sommeria) afanasyevae sp. nov.

Fig. 10

1906 Chonetes sp. (? C.pratti); Etheridge, S. Aust.
Pari. Paper 55: 41.

1907 Chonetes sp. (? C.pratti); Etheridge, S. Aust.
Pari. Paper 54: 7, pi. 6, fig. 9.

1909 Chonetes pratt /(?); Basedow, Z. deutschen Geol.
Gesell., 61(3): 325.

1952 Chonetes pratti; Guppy et al., 19th Int. Geol.
Gong., Alger., Syrn. Gond., III.

Fig. 12 —A-C, Neochonetes (Sommeria) sp. A
All specimens from the Hardman Formation, Canning Basin.
A, CPC 19883. Ventral valve in ventral view. x4; B, CPC
19882, Ventral valve in ventral view. x3.75; C, CPC 19884,
Latex replica of ventral valve external mould, xl.25.

D, Chonetinella sp.

Coolkilya Greywacke, Carnarvon Basin. UWA 88108, Decor¬
ticated ventral valve in ventral view. x4.25.

1957 Chonetes sp. Thomas, J. Pal. Soc. India, 2: 180.
1958a Chonetes sp. Thomas, Abstracts ANZAAS 1958,
Sec. C., p. 2.

Holotype: CPC 19902A from locality 1027A or 1095.
These are similar sections some 2 to 4 km apart on the
same ridge. The collections were inadvertently mixed.

Fig. 10 — Neochonetes ( Sommeria ) afanasyevae sp. nov.

A-G from the Lightjack Formation, Canning Basin; H-M from the Lower Marine Beds, Port Keats Group, Bonaparte Gulf Basin;
N-V from the Coolkilya Greywacke, Carnarvon Basin. A-B, Holotype CPC 19902A, Dorsal valve internal mould and latex replica,
*2.25; C, CPC 19901 A, Latex replica of ventral valve internal mould, x2; D, CPC 19900, External mould of dorsal valve, x2; E,
CPC 19889B, Natural cast of dorsal valve, x2; F, CPC 19902C, Latex replica of ventral valve internal mould, x3.5; G, CPC 19902E,
Latex replica of ventral valve external mould, x3; H, CPC 19887C, Natural cast of dorsal valve, x2.5; 1, CPC 19888C, Latex replica
of dorsal valve internal mould, \2; J, CPC 19888D, Latex replica of dorsal valve internal mould, x2.5; K, CPC 19887B, Natural
cast of ventral valve, x2.5; L, SMP 2131, Natural cast of ventral valve, originally figured by Etheridge (1907a, pi. 6 fig. 9), x2; M,
CPC 19887A, Latex replica of ventral valve external mould, x2.5; N, CPC 19904D, Latex replica of dorsal valve internal mould,
X2.25; O, CPC 19908B, Latex replica of dorsal valve internal mould, x2.25; P, CPC 19907A, Latex replica of dorsal valve internal
mould, x2; Q, CPC 19909C, Latex replica of dorsal valve internal mould, x2.5; R, CPC 19905B, Latex replica of dorsal valve inter¬
nal mould, x2; S, CPC 19907D, Latex replica of dorsal valve external mould, x2.5; T, CPC 19905 A, Latex replica of ventral valve
external mould, x2.5; U, 19909A, Latex replica of ventral valve external mould, x2; V, CPC 19907B, Latex replica of ventral valve

internal mould, x2.75.

126

N. W. ARCHBOLD

Etymology: For Dr. G. A. Afanas’yeva for her substan¬
tial contributions to the study of chonetacean
brachiopods.

Material: Paratypes. CPC 19887 A-19887D,
19888A-19888D 1 natural ferruginous cast of a DV and
2 natural casts of VV, 1 external mould of a DV, 2 inter¬
nal moulds of DV and 1 internal mould of a VV from
Fossil Head, south of Port Keats Mission, Bonaparte
Gulf, Northern Territory, Lower Marine Beds, Port
Keats Group; CPC 19889A-19889B, 1 natural cast of a
DV and 1 natural cast of a VV, from BMR Locality
1027A, Lat. 17°54'15"S, Long. 123°52'15"E, Lightjack
Formation, Liveringa Ridge, Canning Basin; CPC
19901 A-19901C, 19902B-19902E, 3 internal moulds of
VV, 3 external moulds of VV and 1 internal mould of a
DV, from BMR Locality 1027A or Locality 1095, Lat.
18°04'40"S, Long. 124°03'00"E, Lightjack Formation,
Liveringa Ridge; CPC 19900, 1 external mould of a dor¬
sal valve, from BMR Locality KLA3, Bore near Liver¬
inga Homestead, Lightjack Formation; CPC 19903,
19904A-D, 19905A, B, 19906, 19907A-D, 3 internal
moulds of VV, 1 external mould of a VV, 4 internal
moulds of DV and 4 external moulds of DV from BMR
Locality F 17060, ridge and outliers of Kennedy Range,
traverse NW of Paddy’s Outcamp by Teichert, Thomas
and Johnstone 1948, Coolkilya Greywacke, Carnarvon
Basin; CPC 19908A, B, 19909A-D 3 internal moulds of
VV, 2 internal moulds of DV and 1 external mould of a
DV, from BMR Locality F 17082, 1.2 km SE of
Southern Cross Bore, Middalya Station, Coolkilya
Greywacke.

Diagnosis: Planar-convex; ventral valve weakly convex;
sulcus obsolescent; exterior ornament of fine capillae;
ventral and dorsal septa distinct but thin and sharp.
Description: External. Length of the shell is between
0.67 and 0.75 of maximum width. Maximum width is
about mid-length of the shell and is greater than hinge
width. Interareas are low. Pseudodeltidial and chilidial
structures have not been preserved on any specimens.
The external ornament consists of concentric growth
lines, occasionally lamellose, and fine capillae which are
spaced, on average, 5 per mm at 1 cm from the umbo
and between 4 and 5 per mm at 2 cm from the umbo;
they increase in number by both bifurcation and inter¬
calation. Weathering may obscure the capillae and yet
leave prominent growth lamellae. Hinge spines are poor¬
ly known, and are short and emerge at 40°-45° to the
hinge.

Internal. The ventral septum arises close to the um¬
bo, under the delthyrium, and is up to three-quarters of
the valve length. The septum is thin and sharp. Parallel
vascular trunks are usually absent and smooth muscle
scars arc weakly impressed. The teeth are small and
sharp and the delthyrium is small. Interior margins of
the ventral valve are weakly papillose.

The dorsal interior possesses a median septum which
is about two-thirds the valve length; it is thin and sharp.
The lateral septa are short, at a small angle to the me¬
dian septum, and are thin and strong. All three septa
arise anteriorly of a distinct, circular alveolus. The
sockets are small, yet distinct, with small, sharp inner

socket ridges and feeble outer socket ridges. The inner
socket ridges coalesce with the cardinal process which is
small, externally quadrilobate and internally bilobate.
The brachial ridges are never strongly developed and the
anterior of the interior of the valve is finely papillose.
Discussion: Neochonetes ( Sommeria ) afanasyevae , is a
distinctive species of the genus, with few recorded
species possessing comparable morphological char¬
acters. Neochonetes sp. of Bamber & Waterhouse (1971,
pi. 19, figs. 13-18) from the Lower Ufimian of the North
Richardson Mountains, Yukon Territory is similar in
size, shell outline, dorsal interior and development of
the brachial ridges, weakly developed sulcus, and weak
external ornament.

The near absence of the sulcus makes N. (5.)
afanasyevae atypical for the subgenus Sommeria.
However, the species is retained in the subgenus because
of the angle of the hinge spines (Fig. 11).

Range. Middle Kungurian.

Neochonetes (Sommeria) sp. A.

Fig. 12 A-C.

Material: CPC 19882-19883, 2 small VV, from BMR
Locality KLB II, Mount Hardman, from beds 1.25 m
thick at about 40 m below top of hill, Hardman Forma¬
tion, Canning Basin; CPC 19884, external mould of
large VV, from BMR Locality CR 1188, Lat.
19°30'00"S, Long. 125°32'15"E, Hardman Formation,
Canning Basin.

Table 7

Measurements of Neochonetes (Sommeria) sp. A (mm)

Specimen

Number

Hinge

Width

Mid

Width

Ventral

Length

Locality

CPC 19882

8.6

9.8

6.8

KLB 11

CPC 19883

10.9

10.3

7.3

KLB 11

CPC 19884

31.0e

29.Oe

17.6

CR 1188

Observations: This distinct species is large, the max¬
imum width is not always coincident with the hinge
width. The sulcus is broad and deep in its centre. Ex¬
terior ornament of capillae is relatively coarse with be¬
tween 2 and 3 per mm. Convexity of the ventral valve is
not pronounced.

Comparisons: Despite the inadequate material these
specimens appear closely comparable with Neochonetes
(Sommeria) wageri (Muir-Wood, in Muir-Wood &
Oakley 1941) in outline and thickness. However, Muir-
Wood’s species possesses finer ornament and a shallower
sulcus.

Age: Chhidruan.

Subfamily Chonetinellinae Muir-Wood 1962
Genus Chonelinella Ramsbottom 1952
Type Species: Chonetes fiemingii Norwood & Prattcn
1855.

Diagnosis: Strongly concavo-convex, capillate
rugosochonetids with a prominent, narrow ventral
sulcus and distinct dorsal fold.

Discussion: Chonetinella has been broadly interpreted
by Grant (1976) to include species which approach
representatives of Neochonetes (Sommeria) subgen.

WESTERN AUSTRALIAN PERMIAN BRACHIOPODS 2. FAMILY RUGOSOCHONETIDAE 127

nov. However, as no material from Western Australia
adds to the understanding of the genus, Chonetinella is
not discussed further here.

Chonetinella sp.

Fig. 12 D.

Material: UWA 88108, one decorticated VV, from
UWA Locality WF 8.5 (registered no. 28011); Tham-
nopora horizon, 1300 links south of gate in Shed Pad-
dock, Wandagee Station, (collector Dr. C. Teichert),
Coolkilya Greywackc, Carnarvon Basin.
Measurements: Maximum width 11 mm, hinge width
10 mm, length of valve 7 mm, thickness estimated at 3.5
mm.

Observations: Its small size, distinct convexity and nar¬
row deep sulcus refer it to Chonetinella. Traces of
capillae are visible on the anterior portion of the valve.
The specimen indicates the presence of a distinct, highly
sulcate chonetid in the higher portion of the Permian se¬
quence of the Carnarvon Basin.

Age: Middle Kungurian.

ACKNOWLEDGEMENTS
I thank Dr. J. M. Dickins, Bureau of Mineral
Resources, Geology and Geophysics, Dr. P. Jell, Na¬
tional Museum of Victoria and Dr. A. E. Cockbain,
Geological Survey of Western Australia, for providing
specimens from collections in their care. Dr. G. A.
Thomas, University of Melbourne, and two reviewers
provided fruitful comments. The assistance of the staff
of the Baillieu Library, University of Melbourne is
acknowledged. Val Le Maitre and Isabel McDonald
typed the manuscript. The work was carried out while
the author was in receipt of a University of Melbourne
Postgraduate Award.

REFERENCES

References supplementary to those in Archbold 1980a.
Acharyya, S. K., Ghosh, S. C., Ghosh, R. N. & Shah,
S. C., 1975. The Continental Gondwana Group and
associated marine sequences of Arunuchal Pradesh (NEFA)
Eastern Himalaya. Himalayan Geology 5: 60-81.
Afanas’yeva, G. A., 1975b. Chonetacea (Brachiopoda)
srednego i pozdnego karbona Russkoy platformy. Paleont.
Zhur. 1975(2): 96-113.

Archbold, N. W., 1979. Revision of two Permian
brachiopod species names. J. Paleont. 53: 1260.
Archbold, N. W., 1980a. Studies on Western

Australian Permian brachiopods. 1. The family Anopliidae
(Chonetidina). Proc. R. Soe. Viet. 91: 181-192.
Archbold. N. W., 1980b. Mingenewia n. gen.

(Strophalosiidina, Brachiopoda) from the Western
Australian Permian. J. Paleont. 54: 253-258.

Archbold, N. W., 1981a. Quinquenella (Chonetidina,

Brachiopoda) from the Permian of Western Australia. J.
Paleont. 55: 204-210.

Archbold, N. W., 1981b. Svalbardia (Chonetidina,
Brachiopoda) from the Kungurian (Permian) of Western
Australia. Alcheringa 5: 1-8.

Archbold, N. W., 1981c. Western Australian Geology:
An historical review to the year 1870. J. Roy. Soe. W. Aust.
63: 119-128.

Bogoslovskaya, M. F., 1976. Kungurskiya ammonoidei
srednego PreduraPya. Paleont. Zhur. 1976(4): 43-50.

Branson, C. C., 1964. Neochonetes oklahomensis

(Snider). Okla. Geol. Notes. 24: 95-97.

Cockbain, A. E., 1980. Permian ammonoids from the
Carnarvon Basin —A review. Anna. Rep. Geol. Surv. W.
Aust., 1979: 144-149.

Crowe, R. W. A. & Towner, R. R.,1976. Permian
stratigraphic nomenclature Noonkanbah 1:250,000 sheet.
Annu. Rep . Geol. Surv. W. Aust. 1975: 56-58.

Czarniecki . S., 1969. Sedimentary environment and
stratigraphical position of the Treskelodden Beds
(Vestspitsbergen). Praee Museum Ziemi, Warsaw. 16:
201-336.

Davidson, T., 1859. Palaeontological notes on the

Brachiopoda. No. 2. On the families Strophomenidae and
Productidae. Geologist 2: 97-117.

Davidson, T., 1880. A monograph of the British fossil
Brachiopoda. Volume 4, Part 3. Supplement to the British
Permian Brachiopoda. Palaeontograph. Soe. Mon. 34:
243-248.

Dear, J. F., 1971. Strophomcnoid brachiopods from

the higher Permian faunas of the Back Creek Group in the
Bowen Basin. Pubis, geol. Surv. Qld., 347, Palaeont. Pap.
21: 1-39.

Dickins, J. M., 1956. Permian pclecypods from the

Carnarvon Basin, Western Australia. Bull. Bur. Miner.
Resour. Geol. Geophys. Aust. 29: 1-42.

Dickins, J. M., 1976. Correlation chart for the Permian
System of Australia. Bull. Bur. Miner. Resour. Geol.
Geophys. Aust. 156B: 1-26.

Dickins, J. M. & Jell, P. A., 1974. Permian fossils
from the Canning Basin, 1971 and 1972. Bur. Miner.
Resour. Geol. Geophys. Aust., Record 1974/77: 1-14. (Un-
publ.)

Dickins, J. M., Roberts, J. & Veevers, J. J., 1972.
Permian and Mesozoic geology of the northeastern part of
the Bonaparte Gulf Basin. Bull. Bur. Miner. Resour. Geol.
Geophys. Aust. 125: 75-102.

d’Orbigny, A., 1842. Voyages dans L'Amerique meri-

dionale execute pendant les annees 1826, 1827, 1828, 1829,
1830, 1831, 1832 et 1833. Tome 3; 4/ Partic: Paleontologie.
Bertrand, Paris. Levrault, Strasbourg.

Douglas, J. A., 1936. A Permo-Carboniferous fauna

from Southwestern Persia (Iran). Mem. geol. Surv. India
Palaeont. indica. N.S. 22, 6: 1-59.

Dunbar, C. O., & Condra, G. E., 1932. Brachiopoda
of the Pennsylvanian System of Nebraska. Bull. geol. Surv.
Neb. Ser. 2, 5: 1-377.

Elias, M. K., 1962. Comments on recent paleo-

ecological studies of Late Paleozoic rocks in Kansas. Kan¬
sas geol. Soe., 27th Field ConJ., Guidebook 106-115.

Elias, M. K., 1966. Depth of Late Paleozoic Sea in Kan¬
sas and its megacyclic sedimentation. Bull. geol. Surv. Kan¬
sas 169(1): 87-106.

Fantini Sestini, N., 1964. Diagnosi di forme nuove,
Neochonetes asseretoi sp. n. Riv. Ital. Pal. Strut. 70: 899.

Fantini Sestini, N., 1965a. The geology of the Upper
Djadjerud and Lar Valleys (North Iran). II Paleontology.
Bryozoan, brachiopods and molluscs from Ruteh
Limestone (Permian). Riv. Ital. Pal. Strut. 71: 13-110.

Fantini Sestini, N., 1965b. Permian fossils of the

Shaksgam Valley. In, Italian Expedition to the Karakorum
( K2) and Hindukush. Sei. Rep. IV, Paleont. Zool. Bot
149-215. E. J. Brill, Leiden.

Fredericks, G. N., 1915. Fauna verkhne paleozoiskoi
Tolshchi/okrestnostei goroda Krasnoufimska Permskoi

128

N. W. ARCHBOLD

Gub. Trudy geo/. Kom. N.S. 109: 1-117.

Glenister, B. F. & Furnish, W. M., 1961. The Permian
ammonoids of Australia. J. Paleont. 35: 673-736.

‘Gorskii, 1. I. & Timofeeva, 1. L., 1950. Verk-

hnepaleozoiskaya fauna iz Dzhon-garskogo Alatav.
Petrozavodsk, Karelo Finskii Gos. Univ. pp. 1-82.

Grunewaldi, M. Von., 1860. Beitrage zur kenntniss
der Sedimentaren Gebirgsformaiionen in den Berghaupt-
mannschafien Jckatherinburg, Slatousi and Kuschwa,
sowie den angrenzenden Gegendcn des Ural. Mem. Acad,
imp. Sci. St. Petersbourg, Ser. 7, 2, 7: 1-144.

Grushenko, N. V., 1975. Brakhiopody nizhnei permi i
ix znachenie dlya stratigrafii vostochno-Ukrainskogo
nefterazonosnogo basseina. In, Lapkin, Yu, (ed.) t
Stratigrafiya verkhnego paleozoya i nizhnego mezozoya
dneprovsko-donetskoi bpadniy. Moskva, “Nedra”, pp.
83-118.

Guppy, D. J., Lindner, A. W„ Rattigan, J. H. &
Casey, J. N., 1952. The stratigraphy of the Mezozoic and
Permian sediments of the Desert Basin, Western Australia.
19th Irtt. Geol. Cong. Alger., Sym. Gond., 107-114.

Hamlet, B., 1928. Permische Brachiopoden, La-

mellibranchiaten und Gaslropoden von Timor. Jaarboek
van het Mijnwezen in Nederlands Indie. 56(2): 1-115.

Hudson, R. G. S. & Sudbury, M., 1959. Permian
Brachiopoda from south-east Arabia. Notes, Mem. Moyen-
Orient , Mus. nat. Hist, nat., 7: 19-55.

Keyserling, A. F. M. L. A. Von, 1846. Wis-
senschaftlich Beobachtung auf einer Reise in das Petschora-
Land im Jahre 1843. Carl Kray, St. Petersburg, pp. 1-336.

King, R. H., 1938. New Chonctidae and Productidae
from the Pennsylvanian and Permian strata of north-
central Texas. J. Paleonl. 12: 257-279.

Koninck, L. G. de, 1847. Recherches sur les animaux
fossiles. 1. Monographie des genres Productus et Chonetes.
H. Dessain. Liege. 246 pp.

Kozlowski, R., 1914. Les Brachiopodes due Car-

bonifere superieur de Bolivie. Ann. paleont. 9, 1-100.

Kulikov, M. V., 1974. O rasselcnii i usloviyakh
obitaniya fauny v severnoi chasti Kazanskogo morya.
Trudy Vses. ordena Lenina nauehno-issled. geol. Inst.
(VSEGEI) n.s. 182: 138-153.

Lapina, N. N., 1958. Nekotorye dannye o brak-
hiopodovoi fauna verkhnego paleozoya severnoi chasti
Kharaulakhskikh gor. Nauehno-issled. Inst. Geol. Arkt.,
Sb. Strat. Paleont. Biostratigr., 8: 21-30.

Liao Zhuo-Ting, 1979. Uppermost Carboniferous
brachiopods from Western Guizhou. Acta palaeont. sin.,
18, 527-546.

Licharew, B. K., (Likharev, B. K.), 1913. Fauna perm-
skikh’ otlozhenii okrestnostei goroda Kirillova Novgorod-
skoi Gubernii. Trudy Geol. Kom. N.S. 85: 1-99.

Malzahn, E., 1957. Neue fossil funde und vertikale ver-
breitung der niederrheinischen Zechsteinfauna in den
Bohrungen Kamp 4 und Friedrich Heinrich 57 bei Kamp-
Lintformt. Geol. Jb. 73: 91-12(j.

Muir-Wood, H. M. & Oakley, K. P., 1941. Upper
Palaeozoic faunas of north Sikkim. Mem. geol. Surv. In¬
dia. Palaeont. indica N.S. 31, 1-91.

Newton, R. B., 1892. On the occurrence of Chonetes

pralti, Davidson, in the Carboniferous rocks of Western
Australia. Geol. Mag., (3) 9: 468-469, 542-544.

Newton, R. B., 1893. On the occurrence of Chonetes

pratti, Davidson in the Carboniferous rocks of Western

Australia. Rep. Br. Ass. Advmt. Sci. for 1892, 725-726.

Owen, D. D., 1852. Report of a Geological Survey of
Wisconsin, Iowa and Minnesota and incidentally of a por¬
tion of Nebraska Territory. Philadelphia, Lippincott,
Grambo and Co. pp. 1-638.

Ozaki, K., 1934. On some brachiopods from the reddish
purple shale of the Koten series exposed in the Heizyo coal
field. J. Shanghai Sci. Inst. Sec. 2, 1: 89-98.

Playford, P. E., Cockbain, A. E. & Low, G. H., 1976.
Geology of the Perth Basin, Western Australia. Bull. geol.
Surv. West. Aust. 124: 1-311.

Sahni, M. R. & Srivastava, J. P., 1956. Discovery of
Eurydesma and Conularia in the eastern Himalaya and
description of associated faunas. J. Pal. Soc. India 1(1):
202-214.

Sarycheva, T. G. 1970. Stovar’ terminov po morfologii
produktid (Brachiopoda). Moskva, Nauka. 84 pp.

Schauroth, K. F., 1856. Ein neuer beitrag zur Palaon-
tologie des deutschen Zechsteingebirges. Zeit. deutsch.
geol. Ges. 8: 211-245.

Schellwein, E., 1911. Palaozoische und Triadische

fossilien in Ostasian. In, Futterer, K. (ed.)., Durch Asien,
Vol. 3, pp. 125-174. Dietrich Reimer. Berlin.

Semenova, E. G., 1972. Brakhiopody bashkirskogo

yarusa i vereiskogo gorizonta Kuibyshevskoi oblasti. In,
Ifanova, V. V. & Semenova, E. G.
SrednekamennougoTnye i permskie brakhiopody vostoka i
severa evropeiskoi chasti SSSR pp. 7-68. Akad. Nauk SSSR
“Nauka”. Moskva.

Sokolskaya, A. N., 1965. Podotryad Chonetoidea. In,
Ruzhentsev, V. E. & Sarycheva, T. G., Razvitie i smena
morskikh organizmov na rubezhe paleozoya i mesozoya.
Akad. Nauk SSSR, Paleont. Inst., Trudy 108: 209-211.

Spencer, R. S., 1970. Evolution and geographic varia¬
tion of Neochonetes granulifer (Owen) using multivariate
analysis of variance. J. Paleont. 44: 1009-1028.

Termier, G., Termier, H., Lapparent, A. F. De &
Marin, P., 1974. Monographie due Permo-Carbonifere de
Wardak. Doc. Lab. Geol. Fac. Sci. Lyon., H.S. 2: 1-167.

Thomas, G. A., 1958. Thcr Permian Orlhotetacea of
Western Australia. Bull. Bur. Miner. Resour. Geol.
Geophys. Aust. 39: 1-158.

Thomas, H. D., 1930. An Upper Carboniferous fauna
from the Amotape Mountains, North-Western Peru. Geol.
Mag. 67: 394-408.

Trechmann, C. T., 1944. On some new Permian fossils
from the Magnesian Limestone near Sunderland. Q. Jl.
geol. Soc . Lond. 100: 333-354.

Waterhouse, J. B., 1964. Permian brachiopods of New
Zealand. N.Z. geol. Surv. paleont. Bull. 35: 1-289.

Winkler-Prins, C. F., 1968. Carboniferous Produc-

tidina and Chonetidina of the Cantabrian Mountains
(N.W. Spain): Systemalics, stratigraphy and palaco-
ecology. Leid. geol. Meded. 43: 41-126.

Woodward, H. B., 1907. The History of the Geological
Society of London.Geological Society. London. 336 pp.

Yeates, A. N., Crowe, R. W. A., Passmore, V. L.
Towner, R. R. & Wyborn, L. I. A., 1975. New and revised
stratigraphic nomenclature, northeast Canning Basin. An-
nu. Rep. geol. Surv. W.Aust. 1974: 49-51.

Zavodovsky, V. M. & Stepanov, D. L., 1971. Typ
Brachiopoda. In, Kulikov, M. V. (ed.). Polevoi Atlas
permskoi fauny i flory Severo-Vostoka SSSR. Magadan-
skoe Knizhnoe Izd-vo. Magadan, pp. 70-182.

STRATIGRAPHY, SEDIMENTOLOGY
AND HYDROCARBON PROSPECTS OF THE
DILWYN FORMATION IN THE CENTRAL
OTWAY BASIN OF SOUTH EASTERN AUSTRALIA

By G. R. Holdgate

State Electricity Commission of Victoria

Abstract: A sedimentary and facies analysis from deep bore data in the central Otway Basin in¬
dicates the Early Eocene Dilwyn Formation comprises up to seven stacked deltaic cycles. The oldest,
Delta Cycle A, derived much of its sediment from two main fluvial fed sources coming from north of the
basin, and possibly north of the present divide. The fluvial sequences coalesced south of a hinge line at the
Tartwaup Fault, and filled the deep northwest-southeast trending Portland Trough. This delta cycle com¬
prises the prodclta sequences of the Pember Mudstone Member, grading up into two main high construc¬
tive elongate delta channels similar to those of the present day Mississippi birdfoot. Subsequent Deltas B
to G are high sand-shale ratio cycles elongate parallel to the Portland Trough margins. Each cycle begins
with marine shales. Size analysis and isopachs of sand suggest the constituent sand bodies of each cycle
include barrier bars and delta front channels. Fluvial channel sands dominate in the upper parts of the
cycles and towards the north of the basin. Each cycle was terminated abruptly by renewed marine trans¬
gression. The cycling is best explained by channel switching and abandonment of sediment supply, than
by the custacy model. Extrapolation into offshore areas of the basin indicates some hydrocarbon poten¬
tial exists in the more deeply buried sands of the prodelta sequences, and an explanation for the sub¬
marine canyons on the continental slope is proposed.

INTRODUCTION AND METHODS

In the Otway Basin of south-eastern Australia
Palaeocene to Middle Eocene sediments are referred to
as the Wangerrip Group, including a basal Pebble Point
Formation conformably overlain by the Dilwyn Forma¬
tion which comprises the major portion of the Wanger¬
rip Group. This study concentrates on the subsurface
central part of the Basin between the Victoria-South
Australia border and the Warrnambool Ridge (Fig. 1)
which encompasses the two major onshore Gambier and
Tyrendarra Embayments. Most sedimentological work
was carried out on the former embayment, using 16 deep
groundwater and stratigraphic exploration bores drilled
by the Department of Minerals and Energy, Victoria
(DMEV). Cuttings samples were examined every 3 m, as
well as core samples cut approximately every 100 m, and
wire line geophysical logs including gamma ray, spon¬
taneous potential (SP) and resistivity. Similar data from
other bores drilled for oil and water in the area and in¬
terpretation based on geophysical logs for most deep
DMEV bores in the Tyrendarra Embayment have com¬
pleted the programme. Examination of the extensive oil
company seismic data coverage of this area has not been
attempted in this study.

The Dilwyn Formation was first described by Baker
(1943, 1950) from coastal cliffs near Princetown and
later named the Dilwyn Clay (Baker 1953). Bock and
Glenie (1965) introduced the name Dilwyn Formation
with a lower Pember Mudstone Member and an upper
Dartmoor Sand Member, to describe the subsurface
sediments. In the western Otway Basin near the Glenelg
River Boutakoff and Sprigg (1953) named similar strata
as the Dartmoor Formation, now regarded as a junior
synonym of the Dilwyn Formation (Abele et al. 1976).
In South Australia this formation is known as the

Knight Group (Ludbrook 1971). For convenience both
the outcrop and subsurface occurrences are referred to
as the Dilwyn Formation.

The Dilwyn Formation in the subsurface has been
referred to only in general terms in well completion
reports and basin summaries, and the more detailed
descriptions of Abele et al. (1976), apply mainly to the
sparse outcrops. Its depositional environment has been
considered to be paralic (Bock & Glenie, 1965). This
study examines the information available from bores,
and makes an assessment of the depositional en¬
vironments from these data using techniques developed
by oil companies for similar paralic environments. The
conclusions reached probably apply to the Dilwyn For¬
mation throughout the Otway Basin.

RESULTS

Subdivisions

In the study area the Dilwyn Formation can be sub¬
divided into three parts described below, the
characteristic wire line logs of which are shown on Figs 5
& 6 .

1—The Pember Mudstone Member is the oldest unit
and consists of tan to grey siltstones, mudstones and
shales, usually pyritic, carbonaceous, micaceous and
locally glauconitic. Carbonate cemented sands are com¬
mon particularly in the upper parts. Owing to later
uplift and erosion many of the basin margin outcrops
occur within this unit.

2 —The undifferentiated Dilwyn Formation is the
thickest unit generally comprising two-thirds of the total
formation. It is characterised by sands predominating
over shales, and cyclic repetitions of sands-silts-clays.
This unit includes the Dartmoor Sand Member of Bock
and Glenie (1965) which cannot be distinguished in this

129

130

G. R. HOLDGATE

0»Ui

"%ntr

FIGURE 1

p ia* so'

Fig. 1 — Central Otway Basin general geology. (From Portland 1:250 000 Geological Sheet)

DILWYN FORMATION, OTWAY BASIN

131

central area of the Otway Basin. It is transitional with
the Pember Mudstone Member below. When the overly¬
ing Burrungule Member is absent the Dilwyn Formation
appears on bore hole sections (Fig. 6) to be unconform-
able with overlying carbonate units such as the Gambier
Limestone.

3 —The Burrungule Member is mainly recognized south
of the Tartwaup Fault. However, thin non-marine coaly
equivalents may be present for a short distance to the
north of the fault. It includes all the well burrowed grey
muddy siltstones with lesser sands which conformably
overlie the undifferentiated Dilwyn Formation and
underlie the Mepunga Formation and other marine car¬
bonate deposits. Not uncommonly it includes a
calcareous marine fauna of foraminifera and shelly
fossils. It has not been described from the Otway Basin
Victoria before, but is in the South Australian portion
(Harris 1966).

Recognition in the Subsurface

The Dilwyn Formation occurs in all bores throughout
the study area except where removed by subsequent ero¬
sion. It underlies younger marine carbonate units such
as the Mepunga Formation, Gambier Limestone and
Heytesbury Group, or the Whaler’s Bluff and
Bridgewater Formations. North of the Tartwaup Fault
these overlying carbonate sequences truncate the beds of
the Dilwyn Formation. It overlies conformably the Peb¬
ble Point Formation or unconformably the Otway
Group in the far north.

The Dilwyn Formation sands are readily dis¬
tinguishable from those of the Mepunga Formation and
Pebble Point Formation by their lack of brown oxida¬
tion and better sorting and roundness. Upper
Cretaceous Paaratte Formation sands are similar to
Dilwyn sands but, in all cases, the Pebble Point Forma¬
tion intervenes (Holdgate 1977c). Bores in the far NW
lack the Pebble Point Formation but here the Dilwyn
rests unconformably on Early Cretaceous Otway
Group. Tertiary uplift on basin margins and structural
highs has caused truncation of the Dilwyn Formation.
In some bores complete removal has occurred, e.g.
Myaring 2 where Pleistocene Whaler’s Bluff Formation
rests unconformably on Pebble Point Formation.

Description of Lithologies

Sands —typically well sorted and clean, with clear
frosted well rounded quartz grains of mean size 1.0 phi
in the south to 0.4 phi in the north. More detailed sand
size analyses are made in subsequent sections. The sands
occur in sand bodies which range from 15 m to 30 m in
thickness but can exceed 80 m locally. Some sand bodies
coarsen upwards but others remain constant in grain size
throughout. The lower contacts can be gradational with
underlying muddy sands and silts, or abrupt with clays
and shales. The upper contacts are generally abrupt and
overlain by ligneous clays and silts. The sand bodies
comprise over half of the formation thickness but
average less than one quarter of the thickness in the
Pember Mudstone and Burrungule Members. Core

recoveries are low in the sands but improve with depth.
They often fall below 20% due to the overall lack of
consolidation. On wire line logs they commonly show
short normal resistivities of 20 ohms/m 2 . The cemented
sands can be up to 30 or 40 ohms/m 2 . Most sands are
fresh water flushed and on pumping can produce water
flows up to 125 litres/second (Lawrence 1976). Water
quality is good with low salinities. Calculations using
neutron and density logs in Warrain 7 bore show
porosities to decrease from 50% at 665 m to 28% at
1350 m (Laing in Holdgate 1977a).

Shales —usually tan to grey, and laminated by fine
white silty interbeds. In the lower part of the Pember
Mudstone the shales become massive without obvious
bedding, and here can be over 100 m thick. Dark grey
and black shales occur near the base of the formation in
the deep parts of the basin. Gamma ray logs show the
shales range between 0.015 and 0.02 mr°/hr. Maximum
intensities mainly occur in shales of the Pember
Mudstone, and the lower shales in the cyclic sequences.

Silts are most common in the Burrungule Member.
The mottling and burrowed appearance of these silts is
due to infillings of animal burrows by cleaner white silts.
This lithology is also present in the rest of the formation
particularly in the lower shales of the cyclic sequences.

Cemented sandstones are common in the upper parts
of the Pember Mudstone and as occasional interbeds in
the undifferentiated Dilwyn Formation near the base of
the cyclic sequences. On the resistivity logs they have a
spike-like appearance. From their stratigraphic position
it can be inferred that the carbonate cementing media
may be of a primary origin. Chemically, they range be¬
tween dolomite, siderite, silica and pyrite.

Calcareous fossils are rare except in the Burrungule
Member. Planktonic foraminifera have been obtained
from this and the Pember Mudstone and more rarely
from the lower shales in the undifferentiated part.
Siliceous foraminifera including species of Cyclammina
are more common throughout the formation occurring
mainly in the shales. Sharks teeth can also be found.

Carbonaceous material is present in both sands and
shales as disseminations and discrete beds (stringers).
Thin coal seams up to a few metres thick occur in the
upper parts of the cyclic sequences and in the Bur¬
rungule Member. Their poor definition from the gamma
ray logs suggests they contain high ash contents.

Mica is common throughout the formation with large
flakes lying parallel to bedding. In the Nelson Bore the
heavy mineral fractions seldom exceed 0.5% by weight
(Baker 1961), and most of this is authigenic pyrite and
heavy carbonates. From Baker’s table of percentage
distribution (Baker 1961, Table 13), the non-authigenic
minerals in the Dilwyn formation comprise about
50% zircon with the rest evenly separated between tour¬
maline, rutile, garnet, cassiterite and minor metamor-
phic minerals. This mature population distribution in¬
dicates extensive or long transport histories from a
granite terrain. There appears to be no major change in
the heavy mineral population throughout the Dilwyn
Formation.

132

G. R. HOLDGATE

Distribution and Thickness
The formation ranges in thickness between 6.1 m in
the north-west to greater than 1247 m in the south-east,
and occurs in the subsurface throughout the study area.
The isopach map (Fig. 3) shows the general distribution
of sediments. Thicknesses over 600 m are limited to that

part of the study area south of the Tartwaup Fault
where a deep trough is developed. The depocentre for
this trough occurs in theTyrendarra Embayment. North
of the Tartwaup Fault the isopach lines strike north-west
parallel to and deepening south-west away from the
Kanawinka Fault.

DILWYN FORMATION, OTWAY BASIN

133

Fig. 3 —Central Otway Basin isopach of Dilwyn Formation (metres).

Ji

Fig. 4 — Central Otway Basin structure contours to top of Dilwyn Formation (metres).

134

G. R. HOLDGATE

The structure contour lines on top of the Dilwyn For¬
mation (Fig. 4) indicate a structural high around Dart¬
moor plunging south-west towards Nelson township,
part of which has been referred to as the Stokes River
Anticline (Kenley 1971). To the north of this anticline
the structure lines indicate progressive deepening away
from the Kanawinka Fault towards the State border. To
the south there is rapid deepening into the Tyrendarra
Embayment. The Dartmoor Ridge of Boutakoff (1952),
which as a gravity defined structure was postulated to
divide the Gambier and Tyrendarra Embayments, is not
in evidence in the subsurface and has little effect on the
early Tertiary formation. Portland Trough is herein pro¬
posed to describe the thick east-west Tertiary sequence
which straddles this ‘ridge’.

The Portland Trough commences shortly east of
Nelson township as indicated by the isopach lines
around Wanwin 1 and Warrain 7 bores, and deepens
and broadens to the south-east in the general direction
of Portland. To the north it is limited by the rapid
shallowing of section along the line of the Tartwaup
Fault, to the west by structural shallowing in the Nelson-
Caroline area, and opens to the south and south-east in¬
to Portland Bay. All the DMEV bores in the centre of
the trough failed to penetrate the full sequence, which
probably exceeds 2000 m. The only offshore well south
of Portland (Shell Dev. Voluta No 1) drilled a thinner
sequence of Dilwyn Formation, which is taken here as
indicating that a southern margin to this trough occurs
offshore in Discovery Bay.

The trough has two north trending re-entrants which
have important palaeoenvironment implications discus¬
sed later. One occurs on the north-western edge of the
trough around the Wanwin 1 and Mumbannar 6 bore
sites, and a larger re-entrant occurs in the Tyrendarra
Embayment trending north towards the Hamilton area.

Age

In the subsurface, the Dilwyn Formation generally
ranges from Middle Palaeocene to Middle Eocene in age
(Abele et al. 1976) although McGowran (1978) considers
the time interval between Late Early Eocene and Early
Middle Eocene may be absent. In the study area it in¬
cludes the L. balmei and M. diversus spore-pollen zones
of Stover and Partridge (1973) (see Ripper 1976). The T.
collactea (Late Middle Eocene) foraminiferal zone of
McGowran (1973) occurs in the Burrungule Member.

The Pember Mudstone contains planktonic
foraminifera of Palaeocene to Early Eocene age which
have been identified in core material from Malanganee 4
(835.5 m) by Abele in Holdgate (1977b) and Wanwin 1
(1215 m) by Abele in Holdgate (1975). Further iden¬
tifications by Dr C. Abele (pers. comm.) include Ar-
donachie 2 (680 m), Gorae 2 (1369 m), Heywood 13
(1595.5 m), Mumbannar 1 (456 m) and Narrawong 15
(1552.8 m and 1675 m).

The undifferentiated Dilwyn Formation contains
sparser foraminifera of Palaeocene to Early Eocene age
including Cobboboonee 2 (1368.5 m) and Heywood 13
(900 m) (Dr C. Abele pers. comm.).

The Burrungule Member contains a planktonic
foraminiferal fauna of late Middle Eocene age, in War-
rain 7 (256 m and 272 m), Abele in Holdgate (1977a) and
possibly Wanwin 1 (170 m), (Abele in Holdgate 1975).

Cyclic Sedimentation in the Undifferentiated
Dilwyn Formation

From an examination of the wire line logs there occur
characteristic gamma ray and resistivity log traces
through the undifferentiated Dilwyn Formation which
show cyclic repetition. These can be compared with
similar cyclic gamma and electric log profiles considered
to represent subsurface deltaic sequences (Galloway
1968, Fisher 1969, Weber 1971, Selley 1976). The princi¬
ple used to determine the depositional environments
from logs is related to the fact that gamma ray and elec¬
tric log profiles reflect clay content in clastic sequences.
From this the vertical changes in environment can be in¬
terpreted as well as lateral facies changes.

The correlation between vertical log profiles in a
deltaic sequence and grain size obtained from side wall
coring has been demonstrated by Weber (1971). To in¬
terpret Dilwyn Formation environments without the aid
of sidewall cores, grain size analyses of cuttings samples
were made on three bores. Cuttings samples every 3 m
were washed free of drilling mud and sieved. The verti-
cle distribution of grain sizes plotting relative distribu¬
tion of the whole phi intervals in a cumulative plot
coarsening from right to left is illustrated on Fig. 7 for
the three bores. Cutting samples are not ideal for deter¬
mining size distribution, but the correspondence bet¬
ween the wire line logs and size distributions is close,
confirming that the cyclic sequences on the logs relate to
similar cycles in the grain size. Each cycle commences
with a sudden change from a sand in the preceding cycle
to a shale or silty shale immediately above. In some
bores this contact is associated with carbonate cemented
sands and these along with calcareous planktonic
foraminifera indicate the shales are of marine origin.
The shales gradually decrease upwards to be replaced by
two thick massive unfossiliferous sand beds at the top of
the cycle, separated by a second thinner shaly interval.
Grain sizes increase upwards through the transitional
shaly silt and silty sand intervals but mean sizes stabilise
through the massive sands.

Up to seven cycles can be recognized in each bore
which can be readily correlated on logs throughout the
study area and into the adjacent Tyrendarra Embay-
ment (Figs 5 & 6). From the cuttings size analyses each
cycle has a repeatable grain size distribution to the one
below which is characteristic to each bore. Adjacent
bores have similar vertical repeatability but different size
distributions. Mean grain size appears to decrease away
from the basin margins. North of the Tartwaup Fault
the upper cycles are eroded from the structural highs
and basin margins, so that progressively older cycles
subcrop below the younger carbonate sequences. In¬
dividual cycles average 80 m in thickness but decrease
towards the basin margins where there was a continuous
positive influence. Basinwards there is an increase in

DILWYN FORMATION, OTWAY BASIN

135

COBBOBOONEE 2

Fig. 5 —Central Otway Basin Dilwyn Formation. East west log cross-section and reduced wireline logs.

J2

136

G. R. HOLDGATE

shale as the proportion of sand decreases, and the major
contributing factor to the thickening in the Portland
Trough is the large increase in thickness of the Pember
Mudstone and Burrungule Members.

The seven cycles are listed A to G in stratigraphic
order. Cycles B to G characteristically have two sands
and two shales. The lowest shale in each cycle has the
higher gamma ray count, and can be fossiliferous and

DILWYN FORMATION, OTWAY BASIN

137

bioturbated. The lower sand in each cycle often shows a
coarsening upwards sequence as indicated by grain sizes,
and a gradually decreasing gamma log. In some bores
the lower sand has a box-like log character produced by
sharp contacts. These sand bodies are generally thicker
than adjacent sands with transitional contacts. The up¬
per sand bodies are usually box-like with sharp contacts.
The lateral correlations between sand bodies for cycles
A to C are shown on the bore log in Figs 5 and 6.

Cycle A differs from the other cycles whereby it can
grade laterally from a single 90 m thick box-like sand
body (Mumbannar 6) to a 15 m thick coarsening-up car¬
bonate cemented sand body (Malanganee 4) (Fig. 8).
Sizing the cuttings from the Mumbannar 6 bore in¬
dicates coarser sands occur in the upper and lower
regions of the sand body, with little variation in grain
sizes dominated by the 1 to 2 phi class range through the
middle. On the bore log cross section through the Tyren-
darra Embayment (Fig. 9), Cycle A consists of a series
of three or more stacked box-like sand bodies, each up
to 100 m thick. The relationships between vertical grain
size profiles and wire-line log profiles conform to those
deduced by Weber (1971) for fluvial distributary chan¬
nels and barrier sands in the subsurface Niger Delta,
although it should be noted that the Dilwyn sand bodies
can have up to twice the thickness of the Niger Delta
sand bodies.

Geometry of the Cyclic Units

Subsurface geometry of the individual cycles is
depicted by a series of isopachs of each cycle (Figs 10 to
13). The Tyrendarra Embayment contains similar cor¬
rectable cycles which are included on the isopachs.
Each cycle comprises a number of facies and en¬
vironments, which in the sense of Busch (1971) make up
an increment of sedimentation. The sum of all in¬
crements constitutes a genetic sequence of strata—in this
case the undifferentiated Dilwyn Formation. It is en¬
visaged that time-stratigraphic lines would transgress
across the cycles but retain some conformity to the cycle
boundaries.

Figures 11 to 13 show that for the most widespread
Cycles B to D the major depocentres are coincident with
the Portland Trough. The sand percentages for each cy¬
cle as derived from the gamma ray log indicate the basin
margin areas are generally sandier. When considering
the isopach for Cycle A (Fig. 10) some differences are
apparent. This cycle has two depocentres —one in the
Mumbannar/Warrain area, and one in the Gorae/
Portland area. These trend at right angles across the
main trough axis and contain the highest sand percen¬
tage. They are coincident with and extend southward
from the re-entrants described previously at each end of
the Portland Trough. Figures 8 and 9 show two bore
hole log cross sections at right angles to these trends with
the datum horizon placed on top of Cycle A. The
channel-like nature of the sands is clearly visible, as are
the lateral facies changes in these channels to the shales
and carbonate cemented sands of the upper part of the
Pember Mudstone.

Geometry of the Constituent Sand Bodies

Isopachs of each sand in Cycles A, B and C are shown
on Figs 15 to 19, which depicts subsurface geometry of
these sand bodies. Sand bodies for Cycles D to G are not
shown due to partial erosion and correlation difficulties.
The following points are noted:

1. The isopach of Cycle A sand body is the same as the
isopach for the whole of Cycle A comprising two main
depocentres opening to the south in the Mumbannar
and Gorae areas. These have thicknesses over 60 m and
are separated by an intervening low sand area in the
Glenaulin and Kentbruck area (Fig. 15).

2. The isopach of the lower sand in Cycle B (Fig. 16) in¬
cludes two parts —a SW trending depocentre with
thicknesses over 60 m in the Mumbannar/Caroline area,
connecting to an eastward trending depocentre through
the Glenaulin, Heywood and Narrawong bores. Areas
of minor sand thickness surround this sand body to the
north, south and west.

3. The upper sand in Cycle B (Fig. 17) comprises a
V-shaped depocentre through the middle of the study
area with thicknesses exceeding 30 m. This forms a
seaward perimeter to a low sand area in the north central
area. A separate more localised trough occurs in the
Tyrendarra bores.

4. The lower sand in Cycle C (Fig. 18) comprises a long
narrow linear depocentre parallel to the Portland
Trough axis, with thicknesses increasing south-
eastwards to 60 m in the Gorae area. It is flanked to the
north and south by low sand areas.

5. The upper sand in Cycle C (Fig. 19) comprises two
depocentres; one in the Wanwin/Ardno area which
opens into South Australia, and the other in the Tyren¬
darra area. These are separated by a low to absent sand
zone through the Heywood/Cobboboonee areas.

6. An isopach of the main correctable sand body
which occurs in the middle of the Pember Mudstone is
included for comparison (Fig. 14). It is a narrow linear
shaped body with somewhat sinuous outline approx¬
imating in position to the Portland Trough. At its centre
around Kentbruck it exceeds 60 m in thickness.

Dipmeter Analysis

As an additional aid to subsurface facies analysis
dipmeter logs were available to derive sedimentary dips
such as crossbedding, the directions of which may be
used to indicate direction of source and directions of
flow in channel sands. Outlines of the techniques for ex¬
amining dipmeter logs for such features are described by
Schlumberger (1970). Using the methods described,
sedimentary dips are divided into two main cate¬
gories-those where two or more sequential dips of
similar directions increase in dip downhole (red pat¬
terns), and those which decrease in dip downhole (blue
patterns). Ideally for sedimentary dips in channels red
patterns indicate downdip thickening towards the chan¬
nel axis, the blue patterns indicate direction of fill. For
beach barriers and offshore bars dip direction may occur
in either direction normal to the shoreline depending on
whether the sediment was derived from the landward or

138

G. R. HOLDGATE

WANWIN N° 1

GAMMA mr°/h ,
0 0 005 0 Qt 0 015

OftA'N SHI OiiTBiBilTIOtl
IN C\M>'NGS

IN CUTTINGS

not sampled

not sompied

MEMBER

no* sampled

PEMBER

MUDSTONE

not sampled

not sampled

GAMMA mr # /h

Fig. 7 —Reduced wireline logs and grain size distribution of Dilwyn Formation sands in 3 Bores —Gam-

bier Embayment (Victoria) Otway Basin.

DILWYN FORMATION, OTWAY BASIN

139

DRIK DRIK N° 1 GLENAULIN N° 2

cumulative percentage of
phi classes

sand silt shale basalt marl

lithology log

A to G : cyclic units

LOCALITY MAP

Fig. 7 (Continued)

MALANGANEE N°4 MUMBANNAR

140

G. R. HOLDGATE

cr

o

cr

Q

lO

o

Z

Fig. 8 — Stratigraphic cross-section and grain size distribution of the Mumbannar Channel.

DILWYN FORMATION, OTWAY BASIN

141

Table 1

Dipmeter Results

Bore Name and

Sand Unit

Red Patterns

Blue Patterns

Relationship between Dip Direc¬
tion and Main Trends of the Sand
Isopachs (Figs 14-19)

Wire-line Log Profile
of Sand Body

Voluta No 1 Pember
Mudstone Sand

4°-15°@350°N
3°-13°@ 40°N

9°-13°@40°N

At right angles

Coarsening upward

Voluta No 1 Cycle

A — Sand Body

2°-10°@220°S

3°- 8°@165°S

4°-17°@165°S

Parallel

Box-like

Caroline No 1 Cycle

_

5°- 7°@105°E

Parallel? (axis not clear in this

Box-like

B — Lower Sand

—

2°- 5°@140°E

area)

Caroline No 1 Cycle

B —Upper Sand

-

8°@235°

Parallel?

Box-like

Caroline No 1 Cycle

C Lower Sand

9°@05°N

8°- 9°@ 05°N

Diagonal to right angles

Coarsening Upward

Caroline No 1 Cycle

C —Upper Sand

2°- 6°@320°N

2°-15°@180°S

Right angles for red
patterns, parallel? for blue pat¬
terns

Box-like

seaward side. In the study area only two bores have
three-arm dipmeter surveys run through parts of the
Dilwyn Formation. These are the oil wells of Caroline
No 1 and Voluta No 1 (Fig. 2). All dips are recalculated
after subtracting the structural dip as determined by the
methods given by Schlumberger (1970) and the results
are tabulated in Table 1. The results generally indicate
that dip directions parallel the depositional axis for box¬
shaped sand bodies, whereas for coarsening upward
sand bodies major dip directions occur at right angles
and dip towards the depocentres.

DISCUSSION AND CONCLUSIONS
Depositional Environments in the Dilwyn
Formation

The Pember Mudstone represents a prodelta sequence
of marine shales and silty shales which grades upwards
into delta front facies. These sediments partially filled a
deep cast-west trending trough in the central part of the
Otway Basin between the Gambier and Tyrendarra Em-
bayments. Sands within the Pember Mudstone are
characterised by coarsening upwards log profiles, ex¬
cellent sorting coefficients, and linear geometries parallel
to the main Portland Trough margins, and hence pro¬
bably represent palaeoshorelines which formed
periodically with marine regressions during this deposi¬
tional phase. Cross bedding dips to the north, normal to
lhe shoreline trends may indicate the sediments were
derived from the seaward side.

The cyclic sequences which typify the undifferentiated
Dilwyn Formation consist of two genetically different
units, subdivision of which is based on geometry and

facies equivalents. The first genetic unit, Cycle A,
represents the delta front and delta plain equivalents to
the upper part of the Pember Mudstone —the two thus
forming a single large deltaic sequence. This cycle com¬
prises thick channel and stacked channel sands which
form two large lobate deltas similar to the Mississippi
birdsfoot deltas (Fisk et al. 1954). These are named the
Mumbannar Channel and the Portland Channel for
convenience. The channels prograded southwards from
two distinct fluvial sources which followed down the
north-south trending re-entrants at each end of the
Portland Trough, and cut at right angles across the main
basin trends. The channel sands eroded into and pro-
graded over the delta front silts and shales of the
Pember Mudstone. Their lateral facies equivalents bet¬
ween the delta lobes comprise sparsely fossiliferous
laminated and cross bedded silty shales and silts in¬
terspersed by carbonate cemented sands, and represent
deposition under more brackish water conditions in in¬
terdistributary bays.

A marine transgression over these birdsfoot deltas
represented by the fossiliferous shales at the base of Cy¬
cle B marks the end of this Mississippi-type high con¬
structive delta phase, and the beginning of a series of
stacked tidal dominated delta cycles B to G. This second
genetic sequence of strata resembles more closely the
Niger Delta examples (Weber 1971) and consists of up to
six high sand to shale ratio cycles, without large seaward
prodelta components. The single major depocentre for
each cycle is broadly coincident with the boundaries of
the Portland Trough, and with each successive cycle this
preferred lineation becomes more apparent (Figs 10-13).

142

G. R. HOLDGATE

Fig. 9 —Wireline log cross-section across the Tyrendarra Embayment of the Otway Basin, below the top

of Cycle A.

However, when the individual sand bodies which make
up each delta cycle are examined, it is apparent they
have multi-environmental origins.

The lower sand in Cycle B shows box-like log profiles
occurring in a south-west trending depocentre through
the Mumbannar-Caroline areas, indicating a westerly
shift occurred to the Mumbannar channel axis of Cycle
A. This large distributary channel prograded seawards
through a barrier bar sequence typified by the coarsen¬
ing upwards log profiles in bores which lay along the
northern edge of the Portland Trough in the Glenaulin,
Heywood and Narrawong areas. The lack of any similar
large channel-like developments in the Gorae area in¬
dicates a diversion of the previous Cycle A Portland
Channels, possibly into the Mumbannar area. Dipmeter
data in Caroline 1 indicate some contradicting evidence
for channel flow directions, with blue pattern dips trend¬
ing south-east normal to the channel sides. This may be
explained by the multi-environmental origin for this
sand body, which includes some parts derived as channel
deposits and others from barrier bar deposits. On the
seaward side of the combined channel-barrier system
there are up to 30 m thick shales which probably thicken
offshore as the sands thin.

After a thin shale interval the re-establishment of the
Cycle A channel directions took place. The V-shaped
depocentre of the upper Cycle B sand comprises box¬
shaped log profiles in the western arm, suggesting
mainly a fluvio-deltaic environment. Blue pattern dip

directions indicate channel flows to the south in the
Caroline 1 area approximately parallel to the depocentre
axis. Mixed box-like and coarsening upwards profiles in
the eastern arm suggest some barrier sand intervals. The
sharp upper contact with the overlying marine shales is
associated with carbonate cemented sands in inter-
distributary areas, and marks the end of the Cycle B
delta phase.

The lower shales of Cycle C progressively coarsen up¬
wards into a long linear sand body typical of delta front
barrier sands. Parallel red and blue pattern dip direc¬
tions to the north normal to the depositional axis suggest
this barrier complex was similar to barrier sands in the
Pember Mudstone, with sediments being derived
primarily from ofTshore. The sands thin to the north
probably grading into back barrier lagoonal shales, and
offshore to the south into marine shales. This barrier
complex is overlain by shales of probable delta front
facies.

The upper sands of Cycle C show maximum develop¬
ment accompanied by box-like profiles at the western
end of the study area. The lack of bores in the South
Australian section means further environmental inter¬
pretation is difficult, although all indications suggest a
re-established Mumbannar type fluvio-deltaic channel
sequence flowing SSW. Blue pattern dips in Caroline 1
located near the depocentre of the channel also trend
SSW. An area of no sand development in the Heywood
area may represent a local unconformity or non-

DILWYN FORMATION, OTWAY BASIN

143

Figs 10-13-Central Otway Basin isopachs (metres) of Cycles A (10), B (11), C (12), D (13). Encircled
numbers are sand percentages. Numbered left to right on top then beneath.

deposition area separating the westerly channels from
further channels developing in the south-east.

Cycles D to G follow similar patterns of sedimenta¬
tion. Some basalts are intercalated at various levels in
these cycles such as Drik Drik 1, Homerton 3, and Cod-
rington 1. These are probably sills as the basalt date of
37 million years on the Codrington 1 basalt (Bowen
1974) is considerably younger than the palynological age
of equivalent sequences elsewhere. Thicker basalts at the
top of the Dilwyn Formation in the Cobboboonce 2,
Hey wood 10 and 13 bores arc also probably sills (Abele
etal. 1976).

The deltaic sequences ended in the Middle Eocene
when a marine transgression deposited silts and shales of
the Burrungule Member across the Portland Trough.
Marine transgression over the area was complete by the

Late Eocene. Fluvio-deltaic sequences were not re¬
established during or subsequent to this transgression
phase.

Rates and Sources of Sedimentation

The large volumes of sediment comprising the Dilwyn
Formation must have been deposited rapidly from at
least two northerly sources in the vicinity of the present
Western Highlands. In the centre of the Portland
Trough minimum sedimentation rates for the Pember
Mudstone and undifferentiated Dilwyn Formation are
approximately 0.3 m/1000 yrs over a period of 4.5
million years, assuming an upper age limit at the Middle-
Late Eocene boundary. Actual sedimentation rates were
probably much higher considering that diastems occur

144

G. R. HOLDGATE

at the end of each cycle, and subsequent post deposi-
tional consolidation has occurred.

The major depocentre of the Dilvvyn Formation oc¬
curs in the Portland Trough which contains about 1000
km 3 of sediment covering some 2400km 2 . By com¬
parison to present day deltaic depocentres, this is similar
in area to deltas of the Brazos (Colorado), Danube or
Rhone Rivers, which discharge up to 100 million tonnes
of sediment per year (Smith 1966). The sizes and
geometry of the sand bodies in the Dilwyn Formation
are equal to and in many cases exceed the dimensions of
those described for the Niger Delta (Weber 1971) which
has an annual sediment discharge of 25 million tonnes.
This may be owing to the Dilwyn sands being derived
from only two major point sources, whereas the Niger
Delta sands occur at the mouths of a multitude of
smaller delta distributaries.

The lack of any major rivers of comparative size and
discharge rates in this part of Victoria today presents
some difficulties for the reconstruction of palaeo-
drainage patterns. Either the dividing range in Eocene
times was higher and experienced greater precipitation
and erosion to sustain large rivers with short lengths; or
else the deltas of the Portland Trough were supplied
from rivers with larger watersheds coming through from
the Murray Basin north of the divide.

The existence of a substantially higher dividing range
in the Eocene has been refuted in the Eastern Highlands
by the valley filling Tertiary basalts and their ages ob¬
tained by Wellman (1974) who infers that the dividing
range has maintained its heights and extent to the pre¬
sent. Climatically the Lower Eocene in southern
Australia is considered to have experienced greater
precipitation (e.g. Harris 1965, Gill 1975, Martin 1977,
Kemp 1978), but for the lower lying Western Highlands
is unlikely to have produced substantial river deltas
unless accompanied by topographically higher and/or
greater areas of watershed.

Major rivers may have come from north of the divide
into this part of the Otway Basin during Eocene and pre-
Eocene times (Denham & Brown 1976, Gostin & Jenkins
1980, Harris et al. 1980). Supportive evidence from this
study includes:

1—The major re-entrant at the eastern end of the
Portland Trough trends northwards towards the present
low saddle in the dividing range between the Grampians
to the east and the Dundas Tablelands to the west. The
re-entrant at the western end of the Portland Trough
trends northwesterly towards the low basement saddle
of the Padthaway Ridge which divides the Otway from
the Murray Basins in South Australia.

2. The cessation of deltaic sedimentation near the end
of the Early Eocene coincides with folding and uplift
north of the Tartwaup Fault, as indicated by the un¬
conformity on top of the Dilwyn Formation and the sug¬
gested time break (McGowran 1978). Marine transgres¬
sions subsequently occurred from the south. This uplift
could have blocked the Murray Basin river sources along
the divide. The Murray rivers then became confined to
the Murray Basin, and without this major sediment

source open marine transgressions in the Middle Eocene
could begin from the south (the Burrungule Member).
It, therefore, seems likely that the postulated Early Ter¬
tiary outfall for the Murray River in the Spencer Gulf
(Williams & Goode 1978) could have occurred subse¬
quent to this Middle Eocene rise of the Eastern
Highlands. This date would accord with the oldest (Late
Eocene) dated channel sediments near the Mt Lofty-
Flinders Block west of Morgan (Goode & Williams
1980). By Oligocene times there was negligible clastic in¬
put into the Otway Basin and marine carbonate shelf
sediments (Heytcsbury Group) spread right across the
basin reaching practically to the edge of the divide.
Failure of the river systems to re-establish their large
deltas subsequent to the Late Miocene marine regression
indicate that the Eocene sediment regime was reliant
upon a large sediment supply such as from north of the
divide and was not repeated.

Eustacy and Cyclic Sedimentation

The cyclical sedimentation may be attributed to two
types of related causes:

i, those from outside the local area, e.g. tectonic
changes, or world wide eustatic sea level changes;

ii, those entirely within the local environment, e.g.
delta cycles produced by the shifting and aban¬
donment of successive delta lobes.

The relative contribution each one makes to any given
sequence of strata is subject to individual interpretation.
In the Otway Basin cyclic sedimentation is considered
in terms of time, stratigraphic thickness and area
distribution.

The primary cycles consist of four main transgressive-
regressive events (Bock & Glenie 1965), which relate to
tectonics and eustacy and their interactions with basin
development (Glenie el al. 1968). The Dilwyn Formation
is described by them as one major coarsening upward-
decreasing marine influenced regressive cycle, commenc¬
ing with a rapid transgressive phase at the base (Pebble
Point Formation, and Pember Mudstone Member),
grading up into a slow regressive phase (undifferentiated
Dilwyn Formation and Dartmoor Sand Member). The
Burrungule Member was not recognised at this time and
the Mepunga Formation became the base of a new cycle
of marine transgression.

Second order cycles were referred to by Bock & Glenie
(1965) as subcycles, but were not elaborated upon. The
time span and stratigraphic thickness of second order
cycles are probably more closely analogous to the
eustatic cycles identified in the Gippsland Basin (Par¬
tridge 1976). As these eustatic cycles are best defined
from good quality seismic and palynological data they
are not so easily identified in the study area. By inference
some of the major intra-Dilwyn Formation events may
relate to contemporaneous eustatic events in the Gipp¬
sland Basin as suggested by Partridge (1976), but remain
to be better defined.

Third order cycles are those created by local changes
in depositional environment. These are the fundamental

DILWYN FORMATION, OTWAY BASIN

145

Figures 14-19 —Central Otway Basin isopachs (metres) of Pember sand (14), Cycle A sand (15), Cycle B
lower sand (16), Cycle B upper sand (17), Cycle C lower sand (18), Cycle C upper sand (19).

146

G. R. HOLDGATE

factors which contribute to the cyclic sedimentary se¬
quences in the Dilwyn Formation. Each cycle com¬
mences with a rapid marine transgression across the area
followed by slower regression which accompanied the
differential filling by clastic sediments of the newly
created basin. These transgressive events are interpreted
to have been similar to those occurring in modern lobate
deltas such as the Mississippi Delta (Fisk el al. 1954). In
the Mississippi case a series of overlapping delta lobes
were created by the shifting of distribution channels into
new areas during Pleistocene to Recent times. Once
abandoned, each delta lobe continued to subside and
hence underwent subsequent partial destruction during
renewed marine transgression.

Reworking of the topmost sand in each Dilwyn cycle
accompanies the seven transgressive phases, as identified
by the sharp upper contacts associated with carbonate
cemented sands. This contact signals the start of a new
cycle by abandonment of sediment supply. The earliest
transgressive phase was the deepest and most stable and
was accompanied by the deposition of more than 100 m
of prodelta shales (the Pember Mudstone) in inter¬
distributary areas, and more than 600 m of sands and
shales at the lobate delta mouths of the main
distributaries (Cycle A and Pember Mudstone). In this
cycle the rate of deposition would necessarily exceed the
rate of subsidence (Curtis 1970).

A major intra-formational event at the end of Cycle A
occurred and altered the depositional pattern from one
of thick lobate high constructive deltas to one of thin¬
ner, individually more linear and parallel to the palaeo-
shorelines, stacked delta cycles with characteristics
similar to tidal dominated deltas. It is tempting to
associate this change with one of the Gippsland Basin’s
eustatic events within the M. diversus Zone, but equally
its causes may be related to fundamental changes in the
local distribution patterns. Each cycle averages about 80
m in thickness (which may tentatively indicate approx¬
imate water depths by the methods outlined by Klien
1974), even though the component lithologic units show
marked lateral thickness variations. By comparison with
examples given by Curtis (1970) this is more likely to in¬
dicate a widespread and uniform subsidence rate with an
equal rate of sediment supply. The relative influence of
the marine environment does not appear to decrease
from one cycle to the next, rather grain size analyses
show that each cycle is a replica of the one below. This
appears to contradict the overall first order cycle scheme
of Bock & Glenie (1965) which predicts a decrease up¬
wards in marine influence concurrent with an increase in
grain size. In fact, at no time was the Portland Trough
remote from marine influence. Areas peripheral to the
trough became ones of non-deposition and erosion as
subsidence continued without an adequate sediment
supply, so that when marine transgression did occur
throughout the basin in the Late Eocene a major uncon¬
formity had been formed.

Hence in the study area, and also probably
throughout the Otway Basin, the depositional cycles of
the Dilwyn Formation can be considered more in terms

of the depositional processes acting within, rather than
the vertical and lateral facies changes being a conse¬
quence of any basin wide tectonic or eustatic cycle. By
analogy the underlying Late Cretaceous Sherbrook
Group which comprises a similar clastic sequence could
also be considered in these terms.

Offshore Trends

Despite the lack of adequate well control offshore (1
only), the following trends are considered likely to oc¬
cur:

The Portland Trough is open to the southeast and
probably extends some distance offshore into Portland
Bay. In doing so it may close against the offshore exten¬
sion of the Warrnambool Ridge, or it may cross the
front of this ridge to connect with a secondary Tertiary
depocentre off the Port Campbell Embayment (Fig. 1).

North to south bore hole cross sections (Wopfner el
al. 1971, Abele el al. 1976) and seismic interpretations
(e.g. Robertson et al. 1978) indicate that the Wangerrip
Group thins out beneath the middle and outer con¬
tinental shelf areas. Offshore wells in the Port Campbell
Embayment and South Australian part of the Gambier
Embayment indicate this to be the case. In most in¬
stances deltas are observed to thin offshore in a similar
manner.

It can be anticipated from the delta model that in the
offshore areas the Dilwyn Formation would become in¬
creasingly shaly to the detriment of sands. The only
offshore well in the study area (Voluta 1) was sited along
the line of the major distributary channel of the Pember
Mudstone and encountered a sand section over 250 m
thick, indicating this lobate delta extended seawards of
the present shoreline by up to 7 km. Similar thick sand
sequences occur in the closest DMEV bores at Portland
and near Cape Bridgewater. From palynological results
(Ripper 1976) and lithostratigraphic correlation (this
paper) none of these DMEV bores reached the base of
the Tertiary as previously suggested by Glenie and Reed
(1961) and Leslie (1966). This is due to the substantial
thickening in the overlying Heytesbury Group, and the
depth limits of the DMEV drilling rigs. The full Dilwyn
sequence has not been penetrated in this area. Criteria
for recognition of the Pebble Point Formation have
been discussed by Holdgate (1977c). Similar lithologies
have not been observed in any of the Portland bores,
but are present in the sidewall cores in Voluta 1 at 1300
m and 1320 m.

The suspected lack of delta cycles B through G in
Voluta 1 indicates either they did not prograde this far
offshore, or else were removed by subsequent erosion
prior to deposition of the Late Eocene Mepunga Form¬
ation. The closing of the isopach contours on Fig. 3
suggests considerable thinning of the total Dilwyn For¬
mation offshore in Discovery Bay. No Middle Eocene
(Burrungule Member) sediments are present in Voluta 1
suggesting that it is on a structurally high position
relative to the Portland Trough similar to areas north of
the Tartwaup Fault. This fact would suggest some ero-

DILWYN FORMATION, OTWAY BASIN

147

sion of the topmost Dilwyn beds has occurred between
the Lower and Upper Eocene.

Deep water canyons (Bridgewater canyons), and
buried canyons on the continental slope 36 km
southwest of Cape Bridgewater (Fig. 1) have been
discussed by Hopkins (1966) and Von der Borch (1968).
Both authors assign a later Tertiary age for the canyons.
It appears more than coincidental that as river cut
features in an area where substantial rivers onshore are
lacking, they should occur on line to the major Early
Eocene channels of the Dilwyn Formation. This age is
similar to that assigned to other canyon cutting episodes
in southern Australia (Von der Borch 1968). It,
therefore, could be argued that while the present canyon
sediment fill has a Late Tertiary age they reflect deeper
subsurface early Tertiary cutting events when, larger
rivers occurred in this area, and when a marked shelf
break may not have existed. In this way they are similar
to the association of past and present canyons on the
continental slopes off the Gippsland Basin.

Hydrocarbon Prospects

In Otway Basin summaries (Wopfner et al. 1971,
Robertson et al. 1978) the Dilwyn Formation is con¬
sidered to be less attractive as a hydrocarbon prospect
than the underlying Cretaceous sediments, due mainly
to fresh water flushing and the lack of sufficient over¬
burden.

The Portland Trough has not been drilled for oil, and
its definition has only recently been established by
DMEV drilling. For the following reasons this trough
may hold more promise:

1. The Dilw r yn Formation is thicker here, and buried
deeper than other parts of the basin. If trends continue,
the offshore extensions to the trough should see the
lower part of the Pember Mudstone reach burial depths
over 2000 m, giving a better chance to find these
sediments at sufficient maturity.

2. The Tertiary sediment pile in the trough is thick, and
was deposited more rapidly than in surrounding areas of
the basin. In the Dilwyn Formation the sediments were
deposited in a deltaic environment. These factors are
considered to have an important bearing on the
hydrocarbon producing areas in the offshore Gippsland
Basin (Kantsler et al. 1978).

3. As a source rock the thick shales of the Pember
Mudstone contain abundant coaly organic matter. Some
vitrinite reflectance values (Ro max %) in the centre of
the onshore trough are up to 0.56% at 1560.9 m (A. J.
Kantsler pers. comm.). Deeper burial offshore could see
these values increased to bring the sediments into the oil
window at 0.60%.

4. The sands of the Pember Mudstone have porosities
of 28% or above (Laing in Holdgate 1977a), and include
beach and offshore bar environments. They occur as
lenses and linear bodies which in the more distal regions
of the trough away from the main distributary centres
are likely to be isolated from major fresh water flushing
paths. They are also likely candidates for porosity
pinchouts.

5. None of the deeper Tertiary sands in the Portland
Trough have been tested, although shows of hydrocar¬
bons on gas detectors and encouraging wire line log in¬
terpretations have been made (Laing in Holdgate
1977a). The upper beds of the Sherbrook Group and the
Pebble Point Formation also contain some potential as
provided by the shale seals of the Pember Mudstone and
the unconformable relationship between the Upper
Cretaceous and Tertiary sediments.

ACKNOWLEDGEMENTS

This paper is published with permission of the Direc¬
tor of Geological Survey Division, Department of
Minerals and Energy, Victoria. The writer also wishes to
thank colleagues in the Survey who participated in
discussions on this work, and the draughting group of
E. & G. Division, State Electricity Commission of Vic¬
toria.

REFERENCES

Abele, C., Kenley, P. R., Holdgate, G. R. &
Ripper, D., 1976. Otway Basin. In Geology of Victoria. J.
G. Douglas & J. A. Ferguson, eds, Geot. Soc. Aust. Spec.
Publ. 5: 198-229.

Baker, G., 1943. Features of a Victorian limestone

coastline. J. Geol ., 51: 359-386.

Baker, G., 1950. Geology and physiography of the

Moonlight Head district. Proc. R. Soc. Viet. 60: 17-44.
Baker, G., 1953. The Relationship of Cyctammina

bearing sediments to the older Tertiary deposits south-east
of Princetown, Victoria. Mem. natn. Mus. Vic. 18:
125-134.

Baker, G., 1961. Studies of Nelson Bore Sediments,
Western Victoria. Bull. geol. Surv. Viet. 58.

BorK, P. E. & Glenie, R. C., 1965. Late Cretaceous and
Tertiary depositional cycles in south-western Victoria.
Proc. R. Soc. Viet. 79: 153-163.

Boutakoff, N., 1952. The structural pattern of south¬
west Victoria. Min. Geol. J. Viet. 4(6): 21-29.

Boutakoff, N. *& Sprigg, R. C., 1953. Summary

report on the petroleum possibilities of the Mount Gambier
Sunk lands. Min. Geot. J. Via. 5(2): 28-42.

Bow'en, K. G., 1974. Potassium —argon dates —deter¬
minations carried out for the Geological Survey of Victoria.
Rep. geol. Surv. Viet. 1974/79 (Unpubl).

Busch, D. A., 1971. Genetic Units in delta prospecting.

Bull. Am. Assoc. Petrol. Geol. 55: 1137-1154.

Curtis, D. M., 1970. Miocene deltaic sedimentation,
Louisiana Gulf Coast. In Deltaic sedimentation modern
and ancient. J. P. Morgan & R. H. Shaver, eds, Soc. Eco.
Pal. and Miner. Spec. Publ. 15: 293-308.

Denham, .1. 1. & Brown, B. R., 1976. A new look at the
Otway Basin. A PEA J. 16: 91-98.

Fisher, W. L., 1969. Facies characterisation of Gulf
Coast Basin delta systems, with some Holocene analogues.
Trans. Gulf Cst ,455. geol. Socs 19: 105-125.

Fisk, H. N., McFarlan, E., Jr, Kolb, C. R., &
Wilbert, L. J., Jr, 1954. Sedimentary framework of the
modern Mississippi delta. ./. sedim. Petrol. 24: 76-99.
Galloway, W. E., 1968. Depositional systems of the
Lower Wilcox Group, North Central Gulf Coast Basin.
Trans. Gulf Cst Ass. geol. Socs 18: 275-289.

Gill, E. D., 1975. Evolution of Australia’s unique flora
and fauna in relation to the Plate Tectonics Theory. Proc.
R. Soc. Via. 87: 215-234.

148

G. R. HOLDGATE

Glenie, R. C. & Reed, K. J., 1961. Bores 2 & 3

Portland, Victoria —subsurface geology and engineering

data. Min. Geol. J. Viet. 6: 37-46.

Glenie, R. C., Schofield, J. C. & Ward, T. W. t 1968.
Tertiary sea levels in Australia and New Zealand.
Palaeogeogr. Palaeoclimat. Palaeoecol. 5: 141-163.

Goode, A. D. T. & Williams, G. E., 1980. Possible

western outlet for an ancient Murray River in South
Australia 3. Reply. Search 11: 227-230.

Gostin, V. A. & Jenkins, R. J. F., 1980. Possible

western outlet for an ancient Murray River in South
Australia 1. An alternative viewpoint. Search 11: 225-226.

Harris, W. K., 1965. Basal Tertiary microfloras from
the Princetown area, Victoria, Australia. Palaeon-
tographica B 115: 75-106.

Harris, W. K., 1966. New and redefined names in South
Australian Lower Tertiary stratigraphy. Quart, geol.
Notes, geol. Surv. S. Aust. 20: 1-3.

Harris, W. K., Lindsay, J. M., & Twidale, C. R.,
1980. Possible western outlet for an ancient Murray River
in South Australia 2. A discusion. Search 11: 226-227.

Holdgate, G. R., 1975. Wanwin No 1. Water bore
completion report. Rept. geol. Surv. Viet. 1975/25 (un-
publ.).

Holdgate, G. R., 1977a. Warrain No 7. Well comple¬
tion report. Rept . geol. Surv. Viet. 1977/54 (unpubl.).

Holdgate, G. R., 1977b. Malanganee No 4. Well com¬
pletion report. Rept. geol. Surv. Viet. 1977/66 (unpubl.).

Holdgate, G. R., 1977c. Subsurface stratigraphy of the
Victorian section, Gambier Embayment —Otway Basin.
Part 1 The Pebble Point Formation. Rept. geol. Surv. Viet.
1977/10 (unpubl.).

Hopkins, B. M., 1966. Submarine canyons. BHP Tech.
Bull. 26: 39-43.

Kantsler, A. J., Cook, A. C., & Smith, G. C., 1978.
Rank variation, calculated palaeotemps. The Oil & Gas
Jour. 20 Nov.: 196-205.

Kemp, E. M., 1978. Tertiary climatic evolution and

vegetation history in the south-east Indian Ocean region.
Palaeogeogr. Palaeoclimat. Palaeoecol. 24: 169-208.

Kenley, P. R., 1971. Cainozoic geology of the eastern
part of the Gambier Embayment, south-western Victoria.
In The Otway Basin of south-eastern Australia. H. Wopf-
ner & J. G. Douglas, eds, Spec. Bull. geol. Survs. SA &
Viet., 89-153.

Klien, G. de V., 1974. Estimating water depths from
analysis of barrier island and deltaic sedimentary .se¬
quences. Geology 2: 409-412.

Lawrence, C. R., 1975. Geology, hydrodynamics and
hydrochemistry of the southern Murray Basin. Mem. geol.
Surv. Viet. 30.

Lawrence, C. R., 1976. Groundwater. In Geology of
Victoria. J. G. Douglas & J. A. Ferguson, eds, Geol. Soc.
Aust. Spec. Puhl. 5: 411-417.

Leslie, R. B., 1966. Petroleum exploration in the Otway
Basin. Proc. 8th Comm. Min. Met all, Congr ., Aust & NZ.
5: 203-216.

Ludbrook, N. H., 1971. Stratigraphy and correlation

of marine sediments in the western part of the Gambier Em-
bayment. In The Otway Basin of south-eastern Australia.
H. Wopfner & J. G. Douglas, eds. Spec. Bull. geol. Survs.
SA & Viet. 47-66.

Martin, H. A., 1977. The Tertiary stratigraphic

palynology of the Murray Basin in New South Wales. 1.
The Hay-Balranald-Wakool Districts. J. Proc. R. Soc .
N.S.W. 110: 41-47.

McGowran, B., 1973. Observation bore No 2, Gambier
Embayment of the Otway Basin. Tertiary micropaleon¬
tology and stratigraphy. SA. Min. Res. Rev. 135: 43-55.

McGowran, B., 1978. Early Tertiary foraminifera

biostratigraphy in Southern Australia. A progress report.
Bull. Bur. Miner. Resour. Geol. Geophys. Aust. 192:
83-95.

Partridge, A. D., 1976. The geological expression of
eustacy in the Early Tertiary of the Gippsland Basin. A PEA
J. 1976: 73-79.

Ripper, D. T., 1976. Otway Basin, Victoria/South

Australia —spore pollen, microplankton bar-chart compila¬
tion for Otway Basin bores. Rept. geol. Surv. Viet. 1976/90
(unpubl.).

Robertson, C. S., Cronk, D. K., Mayne, S. J., &
Townsend, D. G., 1978. A review of petroleum explora¬
tion and prospects in the Otway Basin region. Bur. Min.
Resour. Geol. Geophys. Record 1978/91 (unpubl.).

Schlumberger, 1970. Fundamentals of dipmeter inter¬
pretation. Course notes to dipmeter interpretation con¬
ference, Melbourne July 1972. Schlumberger Ltd.

Selley, R. C., 1976. Subsurface environmental analysis
of North Sea sediments. Bull. Am. Assoc, petrol. Geol. 60:
184-195.

Smith, A. E., Jr, 1966. Appendix. Modern Deltas:
Comparison Maps. In Deltas in their geological
framework , M. L. Shirley, cd., Houston geol. Soc.,
233-251.

Stover, L. E., & Partridge, A. D., 1973. Tertiary and
Later Cretaceous spores and pollen from the Gippsland
Basin, south-eastern Australia. Proc. R. Soc. Viet. 85:
237-286.

Von Der Borch, C. C., 1968. Southern Australian sub¬
marine canyons: their distribution and ages. Marine Geol.
6: 267-279.

Weber, K. J., 1971. Scdimcntological aspects of oil fields in
the Niger Delta. Geologie Mijnb. 50: 559-576.

Wellman, P., 1974. Potassium-argon ages on the Caino¬
zoic volcanic rocks of eastern Victoria, Australia. J. geol.
Soc. Aust. 21: 359-368.

Williams, G. E., & Goode, A. D. T., 1978. Possible
western outlet for an ancient Murray River in South
Australia. Search 9: 442-447.

Wopfner, H., Kenley, P. R., & Thornton, R. C. N.,
1971. Hydrocarbon occurrences and potential of the Otway
Basin. In The Otway Basin of south-eastern Australia. H.
Wopfner & J. G. Douglas, eds, Spec. Bull. Geol. Survs. S/4
& Viet. 385-435.

TERTIARY FLUVIAL SEDIMENTS AT MORRISON, VICTORIA

By Paul Bolger

Geological Survey of Victoria, 107 Russell Street, Melbourne, Victoria, 3000

{Present Address : Herman Research Laboratory, State Electricity Commission of Victoria, Howard

Street, Richmond, Victoria 3121)

Abstract: Tertiary fluvial sediments underlying the Newer Volcanics in the Morrison area, Victoria,
are subdivided into two units. The basal Ballark Conglomerate is very coarse grained, consisting of
locally derived Ordovician bedrock clasts. Originally considered to be of Permian age its conformable
relationship to the overlying ‘Sub-basaltic sediments’ suggests a Tertiary age. The‘Sub-basaltic sediments’
grade vertically from the Ballark Conglomerate, are finer grained and contain abundant vein quartz and
subordinate Ordovician sandstone clasts. Extensive areas of the sediments have been silicified. Clasts of
the silcrete occur in the upper beds of the ‘Sub-basaltic sediments’, enabling subdivision into an upper and
lower unit. Palaeocurrcnt analyses in the Ballark Conglomerate and the basal beds of the ‘Sub-basaltic
sediments’ and examination of the distribution of Newer Volcanic flows in this area and further west, sug¬
gest a major drainage reversal from northward to southward prior to basalt extrusion.

INTRODUCTION

Tightly folded Ordovician bedrock is overlain by flat-
lying Tertiary conglomerate, gravel, sand, silt and clay
and basalts of the Newer Volcanics in the Meredith-
Morrison-Elaine area. The area is located within an
elevated block of Lower Palaeozoic bedrock that is
marginal to the Tertiary Otway and Port Phillip Basins
and the ‘Ballan Graben’ (Fig. 1).

The Tertiary sequence is subdivided into a thick,
massive basal conglomerate, named the Ballark Con¬
glomerate (Bolger 1977), and an overlying unit of in-
terbedded gravel, sand, silt, clay and rare tuff referred to
informally in this paper as ‘the Sub-basaltic sediments’.
These units are exposed along the valleys of the
Moorabool River and Tea Tree Creek at Morrison (Fig.
1). Selwyn and Ulrich (1866), Brough-Smyth (1869),
Dunn (1907), Hunter (1909), Harris and Thomas (1949),
Makram and Neilson (1970) and Jahnke (1973) all refer¬
red to the massive conglomerate at Morrison, but it was
never described in detail. The purpose of this paper is to
describe the Ballark Conglomerate and the overlying
sediments and to discuss their age, and depositional en¬
vironment.

BALLARK CONGLOMERATE

The Ballark Conglomerate (Bolger 1977) is a coarse
grained deposit, cropping out along the valleys of the
Moorabool River and Tea Tree Creek and exposed in a
water race on the slopes leading down to Dolly’s Creek
at Morrison (Fig. 1). It is also known from bores and
mine workings in the area. The type locality of the
Ballark Conglomerate is the exposure on the south side
of the valley of Tea Tree Creek (Grid reference 166.393,
Meredith 1:63 360 map).

The Ballark Conglomerate (Fig. 3) is a poorly sorted
orthoconglomerate with a maximum observed clast
diameter of 1 m and a median maximum diameter bet¬
ween 10 and 15 cm. The framework consists of 70%
arenaceous clasts (quartzitic sandstone and grey-

wacke), 15% slate and siltstone and 15% quartz. Sand¬
stone and quartz cobbles and boulders are rounded to
well rounded and tend to be ellipsoidal. Slate casts are
platey, often have rounded edges and are strongly im¬
bricated. Quartz pebbles are usually equant and sub¬
rounded. The matrix consists of angular to subangular,
coarse to very coarse sand and granule gravel consisting
of quartz, muscovite and deeply weathered, platey, slate
clasts weakly bound by interstitial clay and silt. The
Ballark Conglomerate is usually friable, although 3.5
km north of Morrison along the east bank of the
Moorabool River, the clasts are closely compacted and
cemented by limonite. It is massive and homogeneous
and apart from trough-crossbedded coarse-sand lenses
up to 0.5 m thick, bedding is not recognisable. Along
the Moorabool River, exposures of the conglomerate are
up to 30 m high and at the type locality are 8 m high.
Selwyn and Ulrich (1866) recorded about 120 m of
‘Miocene gravel’ from Tea Tree Creek. Brough-Smyth
(1896) recorded a thin basal unit comprising gravel,
black clay containing fossil trees and grey sandy clay
containing gold, overlain by more than 100 m of con¬
glomerate in the Golden Rivers mine along Tea Tree
Creek. This basal unit is not exposed but is here included
in the Ballark Conglomerate.

The base of the Ballark Conglomerate is not exposed
but it overlies Ordovician bedrock along Tea Tree Creek
(Brough-Smyth 1869). In the Borhoneyghurk 1 bore
near Morrison 135 m of conglomerate rests on Ordovi¬
cian bedrock. The basal 35 m contains granite boulders,
up to 1.5 m in diameter, which may be glacial erratics.
The basal 35 m is thus inferred to be of glacigene origin,
and excluded from the Ballark Conglomerate. Although
the contact between the Ballark Conglomerate and the
Ordovician bedrock is not exposed, outcrops along the
Moorabool River suggest a sharp, steep boundary and a
high angular unconformity is inferred. The Con¬
glomerate grades into the overlying sub-basaltic
sediments with a decrease in the percentage of slatey

149

150

P. BOLGER

clasts. The top of the Ballark Conglomerate is taken to
be the uppermost bed of coarse conglomerate contain¬
ing sandstone, quartz and more than 8% slatey clasts.

SUEi-BASALTIC SEDIMENTS

Up to 40 m of terrigeneous sediments comprising
poorly consolidated interbedded gravel, sand, silt, clay
and rare tuff underlie the Newer Volcanics along the
Moorabool River and Tea Tree Creek and are known
from the bores between Morrison and Elaine. To the
east and west of Morrison, silicified quartz gravel and
sand forms part of an extensive plateau (Fig. 1). Fer¬
ruginous sands and gravels occur north of Morrison.

The gravels consist mostly of quartz clasts with
smaller amounts of sandstone, slate and siliceous sand¬
stone and conglomerate. Quartz pebbles are usually less
than 10 cm in diameter and occur either as equant, sub-
angular to rounded, or as well rounded, platey and
rodlike grains. Platey clasts are imbricated. Sandstone
and silcrete clasts are usually less than 10 cm in length
and are well rounded and ellipsoidal. The coarsest beds
are at the base of the unit, where there are well rounded
cobbles and pebbles of sandstone and quartz up to 30
cm in length (Fig. 4). Fine to medium grained cross-
bedded and horizontally-bedded sand, pebbly sand,
clayey sand, silt and clay are interbedded with the
gravel. Sand beds are composed almost exclusively of
common (‘plutonic T ) quartz. Massive silt and clay units
are up to 15 m thick. A 1 m thick planar-laminated tuff
band exposed on the Moorabool River about 5 km north
of Morrison consists largely of subangular quartz and
plagioclase sand grains and pebbles of basalt in a clayey
matrix.

The sub-basaltic sediments, at least in part, are con¬
sidered to conformably overlie the Ballark Con¬
glomerate. Locally the base of the sub-basaltic beds is
transitional with the underlying Ballark Conglomerate,
and comprises massive homogeneous pebble to cobble
conglomerate, containing coarse sandstone and quartz
but rare to absent slate or siltstone clasts. The basal
coarse units grade vertically into interbedded gravel,
sand, silt and clay (Fig. 4). The decrease in grain size is
accompanied by a decrease in the number of lithic clasts.

Silcrete clasts do not occur in the Ballark Con¬
glomerate, the transitional gravels at the base of the sub-
basaltic sediments, nor in the silcretes on either side of
the Moorabool River, but they are present in the upper
beds of the sub-basaltic sediments. This suggests the ex¬
istence of two units within the sub-basaltic sediments
(Fig. 2), although the sparsity of silcrete clasts in the
sediments makes detailed mapping difficult. It is sug¬
gested that a period of silcrete formation took place bet¬
ween deposition of the lower and upper units of the sub-
basaltic sediments and that the upper unit disconfor-
mably overlies the lower. The duration of this break in
deposition is not known.

PALAEONTOLOGY AND AGE

Ballark Conglomerate

The age of the Ballark Conglomerate has been

disputed for many years. Dunn (1907), Kenny (1937),
Harris and Thomas (1949) and Makram and Neilson
(1970) considered the Ballark Conglomerate to be of
Permian age. Harris and Thomas (1949) inferred a Per¬
mian age largely on the basis of large granite boulders in
the basal 35 m of the Borhoneyghurk 1 bore. The abun¬
dance of slatey clasts in the Ballark Conglomerate gives
it a similar appearance to Permian glacigene con¬
glomerates in the Ballan and Bacchus Marsh areas to the
north of Morrison. However, the wide variety of erratics
recorded in the Permian beds (Bowen & Thomas, 1976)
is not found in the Ballark Conglomerate, nor are
faceted or striated pebbles. There is therefore no direct
evidence in the exposures at Morrison to suggest either a
glacial origin or a Permian age for the Ballark Con¬
glomerate.

Selwyn and Ulrich (1866) and Brough-Smyth (1869)
regarded the Ballark Conglomerate as Miocene,
underlying Pliocene auriferous gravels and Newer
Basalt. Hunter (1909, p. 116) concluded that ‘the
presence of large boulders and fossil logs tend to show
that this material is of early Tertiary age’. The confor¬
mable relationship with the Tertiary sub-basaltic
sediments establishes a Tertiary age for the Ballark Con¬
glomerate.

No megafossils have been collected from the Ballark
Conglomerate, but plant spores have been recovered
from one sample of the conglomerate matrix in the top 5
m outcropping on the north side along Tea Tree Creek.
The samples contained very few species but included a
single specimen of Tubuliforidites antipodica which sug¬
gests a late Miocene age (Tripompollenites bellus Zone)
(D. T. Ripper, V. Archer pers. comm. 1978). However,
the micro-flora is not diagnostic and the precise age of
the Ballark Conglomerate is still uncertain.

A basaltic tuff overlying the conglomerate contains
clasts of basalt which resemble the early Miocene Maude
Basalt, exposed 17 km south of Morrison (Bowler 1963),
rather than the Older Volcanics of the Ballan Graben or
the Newer Volcanics within the Morrisons area (R. A.
Day pers. comm. 1981). The presence of these clasts
supports the notion that the Ballark Conglomerate and
the basal beds of the sub-basaltic sediments are of Early
to Middle Tertiary age.

Sub-Basaltic Sediments

The sub-basaltic sediments contain rare indeterminate
compressions of fossil wood, but there are no palaeon¬
tological data to determine their precise age. However,
similar sediments which are intimately associated with
the Plio-Pleistocene Newer Volcanics throughout Vic¬
toria are customarily regarded as Pliocene (Abele et al.
1976) and are probably older in some areas. Thomas and
Baragwanath (1950) partly equated gravel, sand and clay
underlying the Newer Volcanics with the Rowsley For¬
mation, in which they include sandy sediments overlying
the brown coal in the Maddingley No. 1 Open Cut at
Bacchus Marsh. The upper part of the sub-basaltic
sediments at Morrison is considered to be partly
equivalent to the Rowsley Formation. The basal beds

TERTIARY SEDIMENTS AT MORRISON

151

Fig. 1 — Geological map of the Morrison area. A marks site of section A of Fig. 2 and section B is at and
in the vicinity of Borhoneyghurk No. 1 Bore.

152

P. BOLGER

may be considerably older, perhaps Early to Middle
Tertiary.

LITHOFACIES AND DEPOSITIONAL ENVIRON¬
MENT

Poor exposure along the Moorabool River and Tea

Tree Creek does not allow detailed facies analysis of the
Tertiary sediments. However, 8 lithofacies are recog'
nised in the sediments at Morrison and are briefly
described below. Lithofacies nomenclature is after Miall
(1977).

TERTIARY SEDIMENTS AT MORRISON

153

Fig. 3-Poorly sorted
Ballark Conglomerate, Tea
Tree Creek, Morrison.

Fig. 4 — Basal units of Sub-basaltic sediments showing fining
upward into interbedded gravel and sand.

Lithofacies Types

Lithofacies Gm(bc) comprises the coarse, poorly bed¬
ded and sorted, clast supported conglomerate of the
Ballark Conglomerate. Platey clasts are strongly im¬
bricated. Individual depositional units are generally
unrecognisable although there are very rare trough
cross-bedded coarse-sand lenses up to 1 m thick (Fig. 5).

Lithofacies Gm(sb) comprises the coarse cobble to
pebble, upward fining conglomerates in the sub-basaltic
sediments. Imbricated pebbles are present but less abun¬
dant in this lithofacies than in lithofacies Grn(bc) due to
the smaller number of slate clasts. Individual units are
1-2 m thick (Fig. 4).

Trough cross-bedded gravels ranging from granule to
pebble size comprise lithofacies Gt. Individual troughs
with theta cross-bedding are up to 1 m deep. Gravel
bands have erosional bases and thin Ienticles of clay and
silt are common.

Lithofacies St comprises trough cross-bedded very
coarse sand with intercalated lenses of granule gravel up
to 0.5 m thick. Troughs are generally less than 1 m deep.
The sands are often clayey.

Units of horizontally bedded medium sand less than 1
m thick comprise lithofacies Sh, and planar cross-
bedded coarse sands with individual sets up to 0.7 m,
comprise lithofacies Sp. These occur locally in the sub-
basaltic sediments.

Lithofacies Fm occurs in both the Ballark Con¬
glomerate and the sub-basaltic sediments and comprises
structureless clay, silt, silty and sandy clay, car¬
bonaceous clay containing some plant remains and very
thin gravelly bands. It occurs at the base of the Ballark
Conglomerate (Brough-Smyth 1869) but is not exposed
and is known only from old mining records. In the sub-
basaltic sediments it occurs in thin units interbedded

K

154

P. BOLGER

with gravel and sand and also in thick units up to 15 m
thick on the west bank of the Moorabool River.

Lithofacies Distribution and Interpretation of
Depositional Environment

Three lithofacies assemblages are recognised, one in
the Ballark Conglomerate and two in the sub-basaltic
sediments.

Assemblage 1 comprises lithofacies Gm(bc), rarely St
and Fm at the base and is restricted to the Ballark Con¬
glomerate. This assemblage is characteristic of deposits
accumulated by superposition of longitudinal bars in
gravel rivers (Miall 1977).

Assemblage 2 comprises the lithofacies Grti(sb), Gt ,
St, Sh , Sp and Fm and is restricted to the sub-basaltic
sediments. Lithofacies Gm(sb) appears to be common at
the base of the sub-basaltic sediments and comprises
much of the transition zone overlying the Ballark Con¬
glomerate. Assemblage 2 contains graded cycles with
basal Gm(sb) and Gt units overlain by St or less com¬
monly Sp or Sh and sometimes Fm units. These cycles
are considered to represent channel filling events in
gravel streams.

The thick lithofacies Fm and locally St and Gm(sb)
comprise Assemblage 3, which occurs within the upper¬
most beds of the sub-basaltic sediments. This
assemblage represents deposition largely under condi¬
tions of low flow, with some point bar deposition. Its
thickness in this area suggests prolonged deposition of
fine sediment from suspension in standing water. The
lithofacies represent abandoned channel and overbank
deposits.

PALAEOCURRENT ANALYSIS

Pebble imbrication, determined by measuring the dip
direction of the AB plane of platey clasts, was used to
infer transport directions for both the Ballark Con¬

glomerate and the basal gravel units of the sub-basaltic
sediments. The data are summarised in Fig. 6.

The results suggest that currents depositing both the
Ballark Conglomerate and the overlying sub-basaltic

Fig. 6 —Summary of palaeocurrent data. Filled roses refer to
Ballark Conglomerate. Open roses refer to lower unit, Sub-
basaltic sediments.

TERTIARY SEDIMENTS AT MORRISON

155

sediments moved from south to north, implying a
drainage divide south of Morrison. This would suggest
an uplifted zone to the south of Meredith, with streams
transporting detritus into the ‘Ballan Graben\ The
orientation of poorly exposed trough cross-beds is con¬
sistent with transport directions from the south and
south-west.

However, this interpretation of the palaeogeography
is not consistent with the interpretation of the drainage
system based on the distribution of Newer Volcanics in
this area, which suggests a southerly palaeoslope. The
basalt flows and associated deep leads in the Ballarat-
Rokewood area, west of Morrison have southerly trends
and flowed between elevated areas of Ordovician
bedrock often capped with ferruginised Tertiary
sediments. The youngest basalt flow at Morrison was
also a valley flow restricted by cappings of silcrete, fer¬
ruginous Tertiary sediments and Ordovician bedrock.
The modern Moorabool River system north of Mor¬
rison, comprising two lateral streams around this basalt
flow-, is southward flowing and follows the pre-basaltic
drainage system.

It is suggested that the drainage system changed from
northward flowing streams which deposited the Ballark
Conglomerate and lower beds of the sub-basaltic
sediments, to a southward flowing system reflected by
the distribution of Newer Volcanic flows. It is con¬
sidered to be largely the result of uplift on the Rowsley
and Spring Creek Faults.

A drainage reversal of this type is problematical.
There is no topographic evidence to support widespread
river capture in the areas to the north of Morrison.
Other than the appearance of silcrete clasts in the upper
unit of the sub-basaltic sediments, no change in pro¬
venance reflected by the petrography of the sediments is
recognised above or below the disconformity. However,
this is not surprising as the likely source rocks both to
north and south of Morrison are Ordovician sandstones,
slates and associated quartz veins.

The timing of the inferred drainage reversal is not
known, although it may coincide with final marine
regression from the Durdidwarrah-Anakie area to the
south in the late Miocene-Pliocene (Bolger & Russell in
Prep.).

The widespread occurrence of silcretes and their
possible time relationship to the drainage reversal raises
the possibility of silcrete formation resulting from severe
alteration of the hydrological system imposed by the
drainage reversal. Although spatially related to basalts
of the Newer Volcanics, the genetic affinities of the
silcretes and basalts are not clearly established (Bolger
1977).

CONCLUSION

Up to 120 m of coarse grained lithic detritus derived
largely from the local Ordovician bedrock was deposited
at Morrison to form the Ballark Conglomerate. The
high lithic content in the Ballark Conglomerate is
atypical of most Tertiary clastic deposits in Victoria and
has led to some confusion about its age. However, its

conformable relationship with more typical quartzose
Tertiary fluvial deposits suggests that the Ballark Con¬
glomerate is Tertiary.

The linear distribution of the Ballark Conglomerate
and the nature of the basement topography suggest that
it was deposited in a steep-sided narrow channel. The
Conglomerate was deposited by a high energy gravel
river flowing to the north or north-east. The single nar¬
row channel eventually filled and a more widespread
stream system was established. Extensive sheets of
gravelly quartzose sediments comprising the basal beds
of the sub-basaltic sediments were deposited above the
Ballark Conglomerate and the Ordovician bedrock.
Current directions remained northward. These deposits
and the surrounding Ordovician bedrock were subse¬
quently extensively silicificd to form silcrete.

A southward flowing drainage system was initiated,
probably in response to movement of the Spring Creek
and Rowsley Faults. Incision into the lower units of the
sub-basaltic sediments and the Ballark Conglomerate
produced terraces. Detritus eroded from the silcrete cap¬
pings was deposited by southward flowing streams to
form the upper parts of the sub-basaltic sediments at
Morrison. Sedimentation at Morrison was terminated
by the extrusion of the youngest basalt flow over the
sub-basaltic sediments. Lateral streams with deep valleys
have subsequently been entrenched.

ACKNOWLEDGEMENTS

The author thanks Dr J. G. Douglas for encourage¬
ment during the course of this investigation and T. G.
Russell for constructive criticism of the manuscript.
Figures were drafted by Mr G. Held of the S.E.C. V. The
work was undertaken and the paper published with the
permission of the Director of the Geological Survey of
Victoria.

REFERENCES

Abele, C., Kenley, P. R., Holdgate, G. & Ripper, D.,
1976. Otway Basin. Spec. Publ. geo/. Soc. Aust. 5: 198-229.
Bolger, P. F., 1977. Explanatory notes on the Meredith
and You Yangs 1:50 000 geological maps. Kept. geol. Surv.
Viet. 1977/14.

Bolger, P. F. & Russell, T. G. (in prep). Late Tertiary
marine transgression in the Brisbane Ranges, Victoria.
Bowen, R. L. & Thomas, G. A., 1976. Permian. Spec.

Publ. geol. Soc. Aust. 5: 123-142.

Bowler, J. M., 1963. Tertiary stratigraphy and

sedimentation in the Geelong-Maude area, Victoria. Proc.
R. Soc. Via. 76: 69-137.

Brough-Smyth, R., 1869. The goldfields and mineral
districts of Victoria. John Ferres, Govt Primer, Melbourne.
Dunn, E. J., 1907. Geological parish plan, Ballark.

Mines Dept. Melb.

Harris, W. J. & Thomas, D. E., 1949. Geology of the
Meredith area. Min. geol. J. Viet. 3: 43-51.

Hunter, S., 1909. Deep leads of Victoria. Mem. geol.
Surv. Viet. 7.

Jahnke, F. M., 1973. Structural geology of the Mor¬
risons area. BSc (Hons) Univ. Melb. (unpubl).

Makram, A. E. & Neilson, J. L., 1970. Notes on the

156

P. BOLGER

geology of the Meredith 1:63 360 military sheet. Rept. geol.
Surv. Viet. 1970/84 (unpubl.).

Miall, A. D., 1977. A review of the braided river

depositional environment. Earth-Sci. Rev. 13: 1-62.
Selwyn, A. R. C. & Ulrich, G. H. F., 1866. Notes on

the physical geography, geology and mineralogy of Vi<;s
toria. Intercolonial Exhibition Essays 1866-67, 83.

Thomas, D. E. & Baragwanath, W., 1950. Tb

geology of the brown coals of Victoria—Part 3. Min. geoL
J. Viet. 4: 41-63.

SHORT COMMUNICATIONS

TAXONOMIC STATUS OF THE VICTORIAN FOSSIL WHALE, ZIPHIUS
(DOL1CHODON) GEELONCENS1S McCOY 1882

Frederick McCoy (1882) based Ziphius ( Doliehodon)
geelongensis on a supposed mandibular tooth of a beaked
whale (Family Ziphiidae, Suborder Odontoceti). The holotype
and only described specimen is now curated as NMV P7464,
National Museum of Victoria, Melbourne. It was collected by
either Mr Legge, Mr Price, or Mr Nelson, from ‘Waurn Ponds
Quarries”, near Geelong, Victoria, Australia (grid reference
about BT6I1682 (1:100 000 map, Series R 652, Sheet 7721,
Geelong) or about 38°16'S, 144°16'E). The holotype almost
certainly came from the Waurn Ponds Member of the Jan Juc
Formation (Torquay Group), which Abele et al. (1976, fig. 13)
indicated is of Late Oligocene to earliest Miocene age. The
description of the specimen given by McCoy (1882) obviates
the need for discussion of all but a few morphological details,
below. Collection details of the holotype were outlined by
Mahoney and Ride (1975).

DISCUSSION

McCoy believed the holotype of Z. geelongensis to be a man¬
dibular tooth similar to those of the extant strap-toothed
whale, Mesoplodon layardi (Gray 1865) (Ziphiidae) as shown,
for example, by Gray (1865, fig. b) and McCann (1962, fig.
5a).Presumably, for this reason. Chapman (1917, p. 32)
employed the combination Mesoplodon geelongensis. Unlike
these authors, however, I believe the holotype to be a fragment
of bone, probably rib. Haversian canals are seen in a thin sec¬
tion of a fragment of the specimen, and there is no trace of
enamel or dentine prisms. While most of the element is dense
and compact, macroscopic vacuities like those of cancellous
bone are present near the axis at both ends of the element.
These are not evident in McCoy’s figures (1882, pi. 69).
Although teeth of adult ziphiids often lack an enamel
croum and do not retain an open pulp cavity at the apical end
of the tooth when the more basal part of the cavity is occluded
(e.g. Flower 1872, Christensen 1973, fig. 2), McCoy (1882, p.
25) assumed the “crown" (enamel?) to have “surmounted the
large pulp cavity on the outer face of the distal end”. McCoy
stated that the surface consists of a very thin layer of cement,
but none is identifiable, and the surface layer appears to be
weathered bone. Finally, there is no evidence, in section, of
occlusion of the axial “pulp cavity” by layers of cementum
which, in other odontocetcs, form axially concentric rings.

It is impossible to ascertain to which taxon the undiagnostic
holotype should be referred, although its large size,
anteroposterior flattening, and the occurrence of other ceta¬
cean bones in the Waurn Ponds Limestone Member indicate
that it is probably a cetacean. Thus, I suggest that the name
Ziphius ( Doliehodon ) geelongensis McCoy 1882 is a nomen
dubium and that it should be allowed to lapse. This suggestion
is unlikely to affect even the more obscure aspects of cetacean
systematics for the species has received little mention in
literature. It has been mentioned only occasionally in local
literature (e.g. Tate 1888, p. 247, Hall & Pritchard 1894, p.
185, Dennant & Kitson 1903, Gregory 1914, Mulder 1914,
Richards 1922, Mahoney & Ride 1975, p. 163), and this sup¬
posed record of Ziphius and Mesoplodon was not included in
standard lists of fossil and recent cetacean taxa (e.g. Kellogg
1928, Simpson 1945, Romer 1966).

REFERENCES

Abele, C., Gloe, C. S., Hocking, J. B., Holdgate, G.,
Kenley, P. R., Lawrence, C. R., Ripper, D. &
Threlfall, W. F., 1976. Tertiary. 177-274. In DOUGLAS,
J. G. & FERGUSON, J. A. (Eds.). Geology of Victoria.
Geol. Soc. AustSpec. Publ. 5. 528 p.

Chapman, F., 1917. New or little-known Victorian fossils in
the National Museum. Part XXL Some Tertiary cetacean
remains. Proc. R. Soc. Viet. 30: 32-43.

Christensen, I., 1973. Age determination, age distribution
and growth of bottlcnose whales, Hyperoodon ampullatus
(Forster), in the Labrador Sea. Norw. J. Zool. 21: 331-340.

Dennant, J., & Kitson, A. E., 1903. Catalogue of the des¬
cribed species of fossils (except Bryozoa and Foraminifcra)
in the Cainozoic fauna of Victoria, South Australia and
Tasmania. Rec. Geol. Surv. Viet. 1: 87-147.

Flower, W. H., 1872. On the recent ziphioid whales, with a
description of the skeleton of Berardius arnouxi. Trans
Zool. Soc. Lond. 8: 203-234.

Gray, J. E., 1865. Notes on the whales of the Cape; By E. L.
Layard, Esq., of Cape-Town, Corr. Memb. With descrip¬
tions of two new species. Proc. Zool. Soc. Lond. 1865:
357-359.

Gregory, J. W., 1914. The correlation of the Australian
marine Kainozoic deposits —evidence of the echinoids,
bryozoa, and some vertebrates. Rep. Brit. Assoc. Adv.
Sci., Australia 84: 376.

Hall, T. S. & Pritchard, G. B., 1894. The older Tertiaries of
Maude, with an indication of the sequence of the Eocene
rocks of Victoria. Proc. R. Soc. Viet. 7: 180-196.

Kellogg, A. R., 1928. The history of whales —their adaptation
to life in the water. Quart. Rev. Biol. 3: 29-76, 174-208.

McCann, C., 1962. Key to the Family Ziphiidae. Beaked
whales. Tuatara 10: 13-18.

McCoy, F., 1882. Ziphius ( Doliehodon ) geelongensis
(McCoy). Prodromus of the palaeontology of Victoria; or,
figures and descriptions of Victorian organic remains ,
Decade 7: 23-26.

Mahoney, J. A., & Ride, W. D. L., 1975. Index to the genera
and species of fossil Mammalia described from Australia
and New Guinea between 1838 and 1968 (including cita¬
tions of type species and primary type specimens). W. Aust.
Mus. Spec. Publ. 6. 250 p.

Mulder, J. F., 1914. Notes on the Waurn Ponds Limestone
fossil beds. Geelong Nat. (ser. 2) 6: 23-26.

Richards, H. C., 1922. Post-Cretaceous rocks of Australia.
Proc. Pan-Pacific Sci. Congr., Bernice P. Bishop Mus.,
Honolulu, Spec. Publ. 1: 744-753.

Romer, A. S., 1966. Vertebrate Paleontology (3rd Edn.) Univ.
of Chicago Press, Chicago. 468 p.

Simpson, G. G., 1945. The principles of classification and a
classification of mammals. Bull. Arner. Mus. Nat. Hist 85-
1-350.

Tate, R., 1888. Census of the fauna of the older Tertiary of
Australia. J. Proc. R. Soc. N.S.W. 22: 240-253.

R. Ewan Fordyce
Dept, of Earth Sciences
Monash University,
Clayton , Victoria 3168

157

ROYAL SOCIETY OF VICTORIA

1981

Patron:

His Excellency Sir Henry Winneke, KCMG, OBE, QC,
KStJ, Governor of Victoria.

President:

G. D. Aitchison, ME, PhD.

Vice- Presiden ts:

D. M. Churchill, BSc, BA, PhD.

Professor J. W. Warren, MA, PhD.

Immediate Past President:

Professor L. L. Stubbs, DAgrSc.

Hon. Secretary:

T. A. Darragh, MSc, DipEd.

Hon. Treasurer:

H. G. Stevens, FCA.

Hon. Librarian:

J. G. Douglas, MSc, PhD.

Hon. Editor:

L. A. Frakes, MA, PhD.

Hon. Research Secretary:

1. A. Staff, MSc, PhD, DipEd.

Hon. Development
Manager:

Professor G. Seddon, BA, MSc, PhD.

Hon. Assistant Secretary:

P. A. Jell, BSc, PhD.

The above Office-bearers are ex officio members of the Council of the Society.
Other members of the Council are:

J. K. Aitken, LLM.

Sir Robert Blackwood, MCE, BEE.
Professor M. J. Canny, MA, PhD.

M. G. Lay, BCE, MEngSc, PhD.

M. J. Littlejohn, PhD.

Professor J. F. Lovering, MSc, PhD.
Professor J. D. Morrjson, PhD, DSc, FAA.

A. E. Perriman, BSc.

Mrs. Delys B. Sargeant, BSc, MEd.

G. A. Sklovsky, IngChem, ESC1L, LiciSci, IngDoc.
J. H. Thompson, BCom.

Executive Officer:

R. R. Garran, MSc, PhD.

Honorary Architect: F. Suendermann, FRAIA, MRAPI.

Honorary Financial Adviser: D. S. Clarebrough.

Honorary Solicitors: Phillips, Fox and Masel.

Honorary Auditor: I. J. Curry, AASA.

Trustees: D. M. Churchill, BSc, MA, PhD.

Professor J. S. Turner, OBE, MA, PhD, MSc, FAA.
Professor J. W. Warren, MA, PhD.

The Hon. Mr. Justice A. E. Woodward, LLM, QC, OBE.

INDEX TO VOLUME 93

A K

Amphipoda

31

Kapcypridopsis asymmetra

76

Ampullacypris oblongata

63

L

B

Laetmatophilus dabberi

36

Ballark Conglomerate

149,150

Leipsuropus

36

Baracuma

31

Leipsuropus parasiticus

38

Baracuma alquirta

35

Limnocythere dorsosicula

43

Beaches—open systems

89, 93

Limnocy there milt a

45

— littoral drift

94

Limnocythere mowbrayensis

48

— nourishment

93

— profiles

90

M

Brachiopoda

Bruun effect

Bruun Rule

Burrungule Member

109

89

87

131

Macquaria australasica

Maddocksella

Maddocksella tumefacta
“Mallee Bores”

23

105
107

106

c

Morrison

149

Mt Kororoit

99

Candonocypris incosta

48

Murray River — Early Tertiary outfall

144

Candonocypris novaezelandiae

53

Chonetinella sp.

127

N

Coasts-archaeological sites

16

Neochonetes

111

— vegetation

Coles Beach

Cyathea australis

Cyclic sedimentation

Cyprice reus sal in us

Cypricercus unicornis

15
2,7
1,9
134, 144

69

72

Neochonetes ( Sommeria) Pratti

114

Neochonetes (Sommeria) robust us

Neochonetes (Sommeria) tenuicapillatus
Neochonetes (Sommeria) afanasyevae

Neochonetes (Sommeria) sp. A

119

121

124

126

Cypretta baylyi

76

O

D

Ostracoda

43, 105

Dilwyn Formation

129

P

E

Palaeocurrent analysis

154

Eucypris virens

69

Palaeodrainage —Murray River outfall

144

— Moorabool River at Morrison

155

F

Pember Mudstone Member

129

Flint

16

S

Fossils— Brachiopoda
-Whale

109

157

Sarscypridopsis aculeata

81

Sea-level changes

15

G

“Sorrento Bore”

107

Gravity survey

101

Submarine canyons

146

Svarlbardia narelliensis

110

Heterocypris vatia

H

66

T

Tower Hill

20

Hydrocarbon prospects

147

V

I

Volcanic vent

99

llyodromus amplicolis

54

Volcanism

20

Ilyodromus candonites

56

Z

llyodromus dikrus

61

llyodromus wi/liamsi

63

Zip hi us (Dolichodon) geelongensis

157

■

•i

*

X

T J

%

Sy-V* * -•

■*#l „ p •»*- ••

u*«

VjLinlffij

Viv?

.;. * .’v.: : >j. • >v ^

■■'. ':-■ .,i-- ' r' • ''■ .; j r » * * ' > 1 ' • •

' i - 1 * *'.. ' '"-••• ' ‘ L‘„‘ z *- ..

I

*•)

W

V .

-J •

- +~z

jE? *

.*

i

JC*‘

' ’ • - T *i,y^SS'

PROCEEDINGS

OF THE

ROYAL SOCIETY OF VICTORIA

Volume 94

ROYAL SOCIETY'S HALL
9 VICTORIA STREET, MELBOURNE 3000

Contents of Volume 94 Number i

Article Page

1 Classification and evolution of the brachiopod Family Rugosochonetidae Muir-

Wood 1962 .By N. W. Archbold 1

2 Channel incision at Eaglehawk Creek, Gippsland, Victoria, Australia

...By Juliet F. Bird 11

3 The peak of the Flandrian Transgression in Victoria, S.E. Australia —faunas and

sea level changes.By E. D. Gill& J. G. Lang with appendix by S. E. Boyd 24

4 Ordovician and Silurian stratigraphy and structure in the Wombat Creek — Benam-

bra area, northeastern Victoria.By Paul F. Bolger 35

5 Occurrence of the Tasmanian mudfish, Ga/axias cleaveri Scott, on Wilsons Pro¬
montory—first record from mainland Australia By P. D. Jackson & J. N. Davies 49
Nomenclatural Note:

Sommenella , a new name for the Permian chonetacean brachiopod subgenus
Sommeria Archbold 1981.By N. W. Archbold 10

Contents of Volume 94 Number 2

Article

6 Seals in Tasmanian Prehistory.By Jim Stockton 53

7 Wind-induced movements of beach sand at Portsea, Victoria By C. L. So 61

8 Wetlands of Victoria III. Wetlands and waterbirds of the Western Districts from

Port Phillip Bay to Mount Emu Creek.By A. H. Corrick 69

9 Studies on Australian Mangrove Algae II. Composition and geographic distribu¬
tion of communities in Spencer Gulf, South Australia.By W. R. Beanland &

Wm. J. Woelkerling 89

Contents of Volume 94 Number 3

Article Page

10 Discovery of Cheirocratus (Crustacea:Amphipoda) on Australian shores By J.

Laurens Barnard & Margaret M. Drummond 107

11 A preliminary study of movement of fishes through a Victorian (Lerderderg River)

fish-ladder. .By J. P. Beumer & D. J. Harrington 121

12 Trilobites from the Mount Ida Formation (Late Silurian-Early Devonian),

Victoria.By David J. Holloway & John V. Neil 133

13 Morphological and geographical disjunctions in forms of Eucalyptus nitida Hook,

f. (Myrtaceae): with special reference to the evolutionary significance of Bass
Strait, southeastern Australia.By Julie C. Marginson & Pauline Y. Ladiges 155

Contents of Volume 94 Number 4

Article Page

14 Upogebiu niugini (Crustacea:Decapoda) a new shrimp from Papua New Guinea

.By Gary C. B. Poore 169

15 The geology of Cape Everard, Victoria — By Margaret C. Fry &C. J. L. Wilson 173

16 Holocene ostracods, other invertebrates and fish remains from cores of four maar

lakes in southeastern Australia.By P. DeDeckker 183

17 Mammals of southwestern Victoria from the Little Desert to the coast By P. W.

Menkhorst & C. M. Beardsell 221

INSTRUCTIONS TO AUTHORS

Papers considered for publication may be Reviews,
Reports of experimental or descriptive research, or Short
Communications. Length of papers may vary; Short Com¬
munications should not be longer than 1500 words.
SUBMISSION

The original and two duplicates of the typescript and all
Tables and Figures should be submitted initially to the Ex¬
ecutive Officer at the Society’s Hall, 9 Victoria Street,
Melbourne 3000. Use International A4 bond paper for the
first copy; machine-copied duplicates are preferred to car¬
bons. Use double space throughout; leave wide margins;
number all pages. Underline only those words to be italicis¬
ed in running text.

FORMAT

Papers are to be written in concise, simple English and
should be organised as follows: 1, a brief title, capitalised,
and if possible beginning with a key word. 2, author’s name
and address, using superscripts to distinguish addresses of
multiple authors. 3, short abstract (200 words or less)
describing the results of the work. 4, the main text; if
necessary use four or fewer grades of headings as follows:
GRADE ONE HEADING

Grade Two Heading: May be followed by running text on
the same line or used as a superior heading.

Grade three heading

Grade four heading: Followed by running text on the same
line. In taxonomic works headings in an hierachical
classification should be centred but not underlined; at
Genus level and above the taxonomic category should
precede the name; refer to other references in the text as
Gill (1968), (Gill 1968), or (Gill 1968, p. 162, fig. 1), and
refer to illustrations of the paper in the text as “Fig. 1”,
“Figs 1, 2” or “Figs 1-4”; all measurements are to be ex¬
pressed in the Metric System (SI units); footnotes are not
allowed; group all acknowledgments at the end and keep
them brief. 5, references should conform to the examples
below; abbreviations of titles of periodicals must conform
with those in A World List of Scientific Periodicals ( 1963-4,
4th ed., London, Butterworths). 6, captions for Tables and
Figures typed on a separate sheet or sheets.
ILLUSTRATIONS

Maximum size is 15.5x21.0 cm and single column

width is 7.5 cm; all Tables and Figures to be reproduced
directly from copy must be designed for these sizes leaving
room for captions below if possible and must be supplied at
reproduction size; photographs should be supplied as glossy
prints, unmounted, labelled on back; Figures that are com¬
posites of several individual photographs should be
mounted on thin flexible cardboard or supplied as glossy
prints of the w'hole figure; line drawings should be made in
black ink and supplied as glossy or flat prints at reproduc¬
tion size; scales must be included and compass directions
must be indicated where necessary; illustrations will be
designated Figures or Tables; on composite Figures items
should be labelled A, B, C, etc. (not a, b, c,).

Oversized illustrations, tables or maps may be accepted
for publication as Foldouts only with the understanding
that the author must meet all costs involved in their produc¬
tion. Maximum size for Foldouts is 21 x31.5 cm.

Short tables will be typeset within the text. Extensive
Tables which arc likely to cover one page of print should be
typed on an electric typewriter to ensure clarity and even¬
ness, within the dimensions 28 x 22.5 cm. They should then
be photographed to reduce them to printed page size
(15,5 x 21 cm) and submitted as glossy prints. They must be
finally correct when submitted as they cannot be corrected
in proof.

REFERENCES

Gill, E. D., 1968. Radiocarbon dating. Victorian Nat. 85:
161-164.

MacPherson, J. H. & Gabriel, C. J., 1962. Marine molluscs
of Victoria. Melbourne University Press, Melbourne,
xv+ 475 p.

Parsons, W. T., 1982. Weeds. In Atlas of Victoria, J. S.
Duncan, cd., Victorian Government Printing Office,
Melbourne, 122-125.

REPRINTS

The Society supplies authors with 25 Reprints free of
charge; joint authors receive a total of 30. Additional
reprints may be ordered by returning the order blank that
accompanies the galley proofs. The author of a paper or the
Institution from which a paper is received may be requested
to contribute towards the cost of publication of the paper.

PROC. R. SOC. VICT. vol. 94, no. 1, 1-9, March 1982

CLASSIFICATION AND EVOLUTION OF THE BRACHIOPOD
FAMILY RUGOSOCHONETIDAE MUIR-WOOD 1962

By N. W. Archbold

Department of Geology, University of Melbourne, Parkville, Victoria 3052

Abstract: The Rugosochonetidae are reclassified into six subfamilies namely: Rugosochonetinae
Muir-Wood 1962, Plicochonetinae Sokolskaya 1960, Undulellinac Cooper & Grant 1975, Lamellosiinae
Cooper & Grant 1975, Quinquenellinae Archbold 1981 and Svalbardiinae subtam. nov. The
Chonetinellinae Muir-Wood 1962 should be allowed to lapse. Phylogeny ot the Rugosochonetidae is
discussed and its geographic distribution and possible migration routes are also documented.

The brachiopod family Rugosochonetidae, with
a history spanning some 160 million years from the Mid¬
dle Devonian until the end of the Permian, became the
most diverse (generic level) family of the Chonetidina
during the Carboniferous and maintained this
dominance in Permian brachiopod faunas (Afanas’yeva
1975a, 1978a). At present, the family includes 29 genera
with 19 having been identified since the major studies of
Sokolskaya (1960) and Muir-Wood (1962).

Elevation of the Rugosochonetinae Muir-Wood
(1962) to family status (Cooper & Grant 1975) is justified
because of the change in scope and content of the fami¬
ly. Nevertheless, considerations of the content and
phylogeny of the family indicate the necessity of modi¬
fying the existing sub-familial classification. The scheme
discussed below differs substantially from that given by
Cooper & Grant (1975) whose review was restricted to
North American Permian genera.

For this review the summary papers on Car¬
boniferous and Permian chonetaceans by Afanas’yeva
(1975a, 1978a) have been an invaluable compilation of
data on the ranges and distributions of rugosochonetid
genera. As a result the discussions herein on generic
distributions and migrations are supplementary to those
papers, and supply necessary corrections and more re¬
cent information. The terminology applied to the
Rugosochonetidae is that used by Archbold (1981e).

Taxonomic Criteria

Cooper (1970) demonstrated the value of the dor¬
sal internal structures of articulate brachiopods for
generic and higher levels of classification of the phylum
and hence it is not surprising that the Rugosochonetidae
was defined by Cooper & Grant (1975) on those struc¬
tures. However, dorsal internal structures of the
Chonetidina change through ontogeny (Greene 1908,
Sokolskaya 1949) and therefore analysis of genera
should be based on large collections.

Sokolskaya (1946) stressed external ornament in
distinguishing stocks of chonetids, a character also used
by Cooper & Grant (1975) in their classification of sub¬
families within the Rugosochonetidae. Many members
of the Rugosochonetidae lack true radial ornament but
the dorsal valve of most smooth genera is pseudo-
capillate when worn which reflects the distribution of

fine radiating taleolae, the long axis of which is parallel
to, rather than normal to, the exterior surface of the
valve. These fine taleolae, which occur just below the
thin primary layer ot the dorsal valve, are found in all
members of the Undulellinae, Lamellosiinae and the
majority, if not all, of the Svalbardiinae subfam. nov. 1
consider pseudocapillate ornament to be a significant
taxonomic criterion within the Rugosochonetidae.

PROPOSED CLASSIFICATION

Suborder Chonetidina Muir-Wood 1955
Superfamily Chonetacea Bronn 1862
Family Rugosochonetidae Muir-Wood 1962
Subfamily Rugosochonetinae Muir-Wood 1962
Genera Included: Rugosochonetes Sokolskaya 1950
( = Nix Easton 1962); Waagenites Paeckelmann 1930
( = Dienerella Reed 1931); Mesolobus Dunbar & Condra
1932; Chonetinella Ramsbottom 1952; Neochonetes
Muir-Wood 1962; Arctochonetes Ifanova 1968;
Schistochonetes Roberts 1971; Parameso/obus
Afanas’yeva 1975; Jakutochonetes Afanas’yeva 1977;
Dagnachonetes Afanas’yeva 1978; Tenuichonetes Jing &
Hu 1978.

Subfamily Plicochonetinae Sokolskaya 1960
Genera Included: Plicochonetes Paeckelmann 1930;
Striatochonetes Mikryukov 1968; Rugaria Cooper &
Grant 1969.

Subfamily Svalbardiinae subfam. nov.

Genera Included: Svalbardia Barkhatova 1970;
Lissochonetes Dunbar & Condra 1932; Dyoios Stehli
1954; Quaclrochoneies Stehli '954; Eolissochonetes
Hoare 1960; Sulcataria Cooper & Grant 1969;
C/ionetineies Cooper & Grant 1969; Komiella
Barkhatova 1970; Capillonia Waterhouse 1973;
Leurosina Cooper & Grant 1975; Leiochonetes Roberts
1976.

Subfamily Undulellinae Cooper & Grant 1975
Genera Included: Undulella Cooper & Grant 1969;
Micraphelia Cooper & Grant 1969.

Subfamily Lamellosiinae Cooper & Grant 1975
Genera Included: Lamellosia Cooper & Grant 1975.

Subfamily Quinquenellinae Archbold 1981
Genera Included: Quinquenella Waterhouse 1975.

2

N. W. ARCHBOLD

DISCUSSION AND DIAGNOSIS FOR PROPOSED
CLASSIFICATION

Family Rugosochonetidae Muir-Wood 1962

(nom. trans. Cooper & Gram 1975, p. 1212 ex.
Rugosochonetinae Muir-Wood 1962, pp. 32, 64.)
Amended Diagnosis: Small to medium sized, costate,
capillate, smooth or lamellose chonetaceans. Dorsal in¬
terior with pronounced lateral septa, long median sep¬
tum and deep alveolus. Cardinal process externally
quadrilobed and internally bilobed; it may be externally
bilobed in early members of the family. Ventral sulcus
absent to strongly developed; median septum of variable
length, high posteriorly; hinge spines oblique to nearly
vertical. Pseudodeltidium and chilidium may be present.

Discussion: Cooper & Grant (1975) raised the taxon
from sub-family to family status but did not provide a
diagnosis although it is clear from their comments
(Cooper & Grant 1975, p. 1212) that the family was
recognised on the basis of internal structures, especially
the nature of the cardinal process.

The family Rugosochonetidae contains a diverse
group of genera united by common internal dorsal
features. Six subfamilies can usefully be recognised at
present.

Subfamily Rugosochonetinae Muir-Wood 1962

Amended Diagnosis: Small to large sized rugoso-
chonetids with radially capillate or costate external or¬
nament. Ventral sulcus feebly to strongly developed.
Hinge spines at low to moderate angle. Dorsal fold pre¬
sent in several genera; brachial ridges often well
developed.

Discussion: The Rugosochonetinae is restricted to
include only those genera with an external ornament
varying from capillate to costate. The Chonetinellinae
Muir-Wood (1962) is permitted to lapse. Muir-Wood
(1962) assigned three genera to that subfamily:
Neochonetes, Chonetinella and Waagenites. The first
was reassigned to the Rugosochonetinae by Cooper &
Grant (1975). Chonetinella is a broadly interpreted
genus (Grant 1976) and includes species which approach
Western Australian Permian species of Neochonetes
(Archbold 1981 e). Waagenites is still poorly known, its
dorsal interior never having been adequately illustrated,
and is provisionally assigned to the Rugosochonetinae.
At present the genus is broadly interpreted (Waterhouse
& Piyasin 1970, Grant 1976) and includes species with
poorly developed sulci. The Chonetinellinae as
characterised by Cooper & Grant (1975), grouped
together rugosochonetids with a distinct sulcus. They
also included Chonetinetes Cooper & Grant (1969)
within the subfamily because of that genus being similar
in gross morphology to Chonetinella. To group chonetid
genera together on the basis of the presence of a promi¬
nent sulcus is a dubious criterion; the development of
heterochronous, homeomorphic ventral sulci in different
stocks of chonetaceans has previously been discussed by
Archbold (1980a). Chonetinetes is provisionally assign¬
ed herein to the Svalbardiinae subfam. nov.

Discussion of remaining rugosochonetids is
restricted to poorly understood genera. The variable ex¬
ternal ornament of Mesolobus has been extensively
discussed by many authors. Hoare (1960) demonstrated
that North American, early Pennsylvanian species are
capillate while younger species are smooth. The type
species, from high in the Pennsylvanian, was considered
by Weller & McGehee (1933) and King (1965) to be
smooth, whereas Girty (1915) and Dunbar & Condra
(1932) agreed with Norwood & Pratten (1855) in con¬
sidering the species to be capillate. The query remains;
how smooth are the smooth species? Girty (1915, p. 63)
noted that “when large series of specimens from
different horizons are examined, individuals more or less
intermediate in character are found. That is, associated
with the smooth variety are a few shells which show
faint, yet unmistakable traces of radial sculpture”.
Sutherland & Harlow (1973) showed that even smooth
species of Mesolobus occasionally show faint capillae
commonly near the anterior margin. Following Dunbar
& Condra (1932) the ornament of Mesolobus is con¬
sidered to be finely capillate, at times “obsolescent”.

Arctochoneies Ifanova (1968) is assigned to the
Rugosochonetinae. The bifurcating ventral median sep¬
tum of Arctochonetes appears to be-a stronger develop¬
ment of the short median septum and pair of ridges on
either side of the adductor muscle field of Neochonetes.

A new genus belonging to the Rugosochonetinae
(Fig. 1) is typified by Neochonetes unbonoplicatus, from
the Sakmarian Nenets Beds, Sula River, Northern
Timan Mountains, as figured by Barkhatova (1964). The
ventral valve is capillate, possesses a distinct, posteriorly
developed sulcus which changes anteriorly to a swollen
fold, separated from the lateral flanks of the valve by a
valley on either side. The species has not been formally
described and hence is a nomen nudum , but rather than
describe the new species and genus here, on the basis of
the only figured specimen, description is left to those
with access to a collection of specimens so that the on¬
togeny of the ventral valve can be fully assessed.

Subfamily Plic.ochonetinae Sokolskaya 1960
Amended Diagnosis: Small, strongly convex capillate to
costate rugosochonetids. Hinge spines oblique to high
angle. Fold and sulcus absent. Interior generalised,
often poorly known.

Discussion: Although not adopted by Muir-Wood
(1962), this subfamily was redefined and reconstituted
by Cooper & Grant (1975). The subfamily still appears
to include a heterogeneous group of genera and further
work is required to define the scope of the subfamily.
Plicochonetes appears morphologically far removed
from Dagnachonetes, regarded herein as the ancestral
rugosochonctid, and this suggests that the Rugoso-
chonetidae may be polyphyletic. Muir-Wood (1962)
showed that the hinge spines of Plicochonetes are slight¬
ly curved and extended at a high angle to the hinge.
Striatochonetes Mikryukov (1968), a finely costellate
genus with high angle hinge spines, is inadequately
known internally and hence is provisionally included in

CLASSIFICATION RUGOSOCHONETIDAE

3

Fig. 1—The inferred phylogeny of the Rugosochonetidae Muir-Wood showing the relationship of the
constituent genera. Subfamilies are separated by curved, dotted lines.

4

N. W. ARCHBOLD

the Plicochonetinae following Mikryukov (1968). The
type species, Strophomena setigera Hall 1843, well
figured externally by Hall (1843, 1862), possesses hinge
spines at a high angle which may indicate a relationship
for the genus with either the Strophochonetinae or the
Retichonetinae, both of the Chonetidae, although it
does not preclude the genus from the Rugoso-
chonetinae. Rugaria Cooper & Grant (1969), with obli¬
que hinge spines, was well described by its authors and is
interpreted herein as a direct descendant of
Plicochonetes ,

Subfamily Svalbardiinae subfam. nov.
Diagnosis: Small to medium sized, externally smooth
rugosochonetids. Dorsal exterior pseudocapillate when
worn. Hinge spines at low to moderate angle.

Discussion: Members of this subfamily are distinguish¬
ed from the Quinquenellinae by the absence of accessory
septa and the presence of dorsal pesudocapillae when
worn, and the Undulellinae by the possession of hinge
spines at a much lower angle to the hinge. The Un¬
dulellinae are also characterised by pronounced develop¬
ment of the brachial ridges and subtle variations in the
arrangement of the dorsal septa. Probably all the consti¬
tuent genera of the Svalbardiinae possess the distinctive
taleolate shell structure which results in a
pseudocapillate shell ornament, especially of the dorsal
valve when the shell is worn. This is considered to be a
unifying feature of this stock of the Rugosochonetidae
and is shared with the Undulellinae and the l.amcllo-
siinae. Confirmation of the pseudocapillate ornament is
required for the dorsal valves of Quadrochonetes and
Leiochonetes. Pseudocapillate shell structure may be
noted for the following genera as figured and/or discuss¬
ed by the various authors:

Lissochonetes Dunbar & Condra (1932, p. 171, pi. 20,
fig. 48)

Sulcataria Cooper & Grant (1975, pi. 478, fig. 62)
Dyoros Stehli 1954, Cooper & Grant (1975, pi. 481, fig.
10, pi. 485, figs 12, 13; subgenus Ltssosia Cooper &
Grant 1975, pi. 487, fig. 18; subgenus Tetragonetes
Cooper & Grant 1975, pi. 489, figs 26, 32 and pi. 490,
fig. 72)

Chonetinetes Cooper & Grant (1975, pi. 477, figs 2, 49)
Leurosina Cooper &. Grant (1975, pi. 495, fig. 3)
Komiella Barkhatova 1970, Licharew (1934, pp. 12, 100)
Capillonia Waterhouse 1973 (sec Waterhouse 1964, pi.
3, figs 1, 10 as discussed by Archbold 1981b)
Svalhardia Barkhatova, as discussed by Archbold
(1981b)

Chonetinetes and Dyoros both possess a pseudocapillate
ventral valve as well as dorsal valve suggesting a close
relationship for the two genera.

Eo/issochonetes Hoare 1960 is stated to possess “no
trace of true radial striation” (my italics). A hint of a
pseudocapillate ornament is shown for Quadrochonetes
Stehli by Cooper & Grant (1975, pi. 502, fig. 17).

The shell structure of Leiochonetes Roberts (1976) is not
known but Roberts’ description indicates a comparable
feature may be present.

Following Brunton (1972) I consider that the
microstructure of the brachiopod shell is important to
systematics. The distinctive shell structure in the
Svalbardiinae, Undulellinae and Lamellosiinac, possibly
reflecting the positions of setae along the mantle edge,
just below the primary layer of the shell, which were
differentially filled with shelly material, structurally
different to that of the remainder of the secondary layer,
appears to unite the three subfamilies closely. Similar
shell structure has not been noted for true capillate
rugosochonetids which, when worn, are smooth.

Lissochonetes , despite good illustrations of the
type specimens by Geinitz (1867) and Mudge &
Yochelson (1962) and discussions by Dunbar & Condra
(1932), Muir-Wood (1962) and Cooper & Grant (1975),
remains poorly known. It should be restricted to weakly
sulcate species with delicate dorsal internal structures
until large collections are available. Komiella , with stout
lateral septa and a long median septum fused anteriorly
of a deep alveolus (Archbold 1981b), appears useful for
separating species from the ill-defined Lissochonetes.
The type species of Komiella , of Kazanian age, has been
recorded from as early as the Middle Carboniferous
(Afanas’yeva 1977) indicating that the genus spans a
considerable time interval.

Subfamily Undulellinae Cooper & Grant 1975

Amended Diagnosis: Small smooth rugosochonetids.
Exterior of dorsal valve pseudocapillate when worn.
Hinge spines at high angle (nearly 90° to the hinge).
Brachial ridges and dorsal median septum prominent;
cardinal process small.

Discussion: The Undulellinae are morphologically close
to the Svalbardiinae, being distinguished from that sub¬
family by details of the hinge spines and dorsal interior.
The pseudocapillate appearance of the dorsal valve
when worn is added to the subfamilial diagnosis, in
order to emphasise the relationship of the Undulellinae
to the Svalbardiinae.

Subfamily Lamellosiinae Cooper & Grant 1975
Amended Diagnosis: Concentrically lamellose rugoso¬
chonetids with no radial ornament. Pcsudocapillate
shell when worn.

Discussion: I follow Cooper & Grant (1975) and include
this subfamily in the Rugosochonetidae, but as the dor¬
sal interior is unknown, the familial assignment is only
tentative. Support for the rugosochonetid affinities of
Lamellosia comes from the pseudocapillate nature of
the shell when worn.

Subfamily Quinquenellinae Archbold 1981

Amended Diagnosis: Small, smooth rugosochonetids,
not pseudocapillate when worn. Dorsal interior with
short lateral septa, long accessory septa and a variably
developed median septum.

Discussion: The development of long accessory septa
and the lack of a pesudocapillate dorsal valve when
worn sets this subfamily apart from other smooth

CLASSIFICATION RUGOSOCHONETIDAE

5

rugosochonetids. Phylogenetic implications of the lack
of pseudocapillae are discussed below.

PHYLOGENY OF THE RUGOSOCHONETIDAE
Introduction

The inferred phylogeny of the Rugosochonetidae
is shown in Fig. 1.

While it is beyond the scope of the present review
to discuss fully the applicability of particular stage
names in subdividing the Carboniferous and Permian
Periods, it should be noted that several of the subdivi¬
sions in Fig. 1 are provisional and serve merely as a
guide to time control for the development of the family.

The problem of the relationship of the Gzhelian
and Asselian Stages, especially with respect to the
"Orenburgian” stage, has been reviewed by Waterhouse
(1978a). The view that much of the Orenburgian is basal
Asselian is strongly indicated by the re-examination of
the classic brachiopod faunas, and a review of other
fossil groups, of the Samara Bend by Prokofev (1975),
who maintained that the Gzhelian is the youngest stage
of the Carboniferous, a view consistent with the most re¬
cent monographic study of the Carboniferous of
Fergana (Sikstell et al. 1975).

The Chhidruan as used herein equals the Punja-
bian of Stepanov (1973) and Waterhouse (1976). Use of
the name Punjabian is avoided because of the earlier in¬
formal use of the same name by Reed (1936, 1939) for
early Permian faunas of the Salt Range.

The phylogeny diagram is dendritic in style.
Diverging branchlets do not necessarily imply that two
genera or subfamilies become less similar to each other
through time. Several genera appear to result from small
changes to the ancestral genus; e.g. Neoehonetes from
Rugosochonetes (Archbold 198 le). Other genera appear
abruptly with no obvious antecedent' (e.g. QuacJ-
rochonetes) as do the two subfamilies, Lamellosiinae
and Quinquencllinae. No scale of variation is intended
by the curved branchlets although genera do not exhibit
a constant morphology, but rather show changes in
morphology from species to species. These trends will
not be in a constant direction—as might be inferred
from straight lines. A species may show a constant mor¬
phology (reflecting a constant gene pool); a genus, of
more than one species, never will. Diverging branchlets
may be interpreted from Fig. 1 to imply divergent evolu¬
tion of two genera although, as stated above this is not
necessarily the case.

Variations in the external gross morphology of
the shell from genus to genus within chonetacean
families such as the strength of the sulcus is probably
related to environmental factors, such as the nature of
the substrate. The development of heterochronous
homeomorphs occurred in different families and sub¬
families (Archbold 1980a).

The ancestry of the Rugosochonetidae is likely to
lie within the Chonetidae. The Parachonetinae of
Johnson (1970) appear an ideal group to be the ancestor
of the Rugosochonetidae because of the similarity of the

dorsal internal structures of the two groups. Para-
ehonetes, common in Emsian age rocks (Johnson 1970),
is a suitable ancestor for the Rugosochonetinae. It seems
likely that the Rugosochonetidae is polyphyletic in
origin, but it appears possible, judging from illustrations
of Parachonetes by Johnson (1966, 1970) that the
Plicochonetinae and the Rugosochonetinae may have
both arisen independently from the Parachonetinae by
modification of the external ornament and convexity of
the ventral valve. The earliest member of the
Rugosochonetidae appears to be the Eastern European,
Eifelian genus Dagnachonetes which possesses a
simplified, bilobed cardinal process (Afanas’yeva 1978b)
but the derivation of the Plicochonetinae from
Dagnachonetes appears unlikely. Striatochonetes , ap¬
pearing in the Givetian, may not belong to the
Plicochonetinae and is certainly far removed mor¬
phologically from Dagnachonetes and yet both genera
are close in time. Similarly the origin of the small, highly
convex Frasnian genus Plicochonetes is obscure and ap¬
pears distinct from Dagnachonetes. Plicochonetes has a
subtantial time range if the referral of the Artinskian
Plicochonetes minor to the genus is correct (Ting 1965).
Rugaria was probably derived from Plicochonetes by
modification of the hinge spines and cardinal process.

Evolution of tub Rugosochonetidae

The Rugosochonetidae first appear in the
Eifelian, and then reappear in the earliest Carboniferous
with the genus Rugosochonetes. Rugosochonetes has a
substantial time range and early species of Neoehonetes
are similar to species of Rugosochonetes. The local
development of Schistochonetes in northwestern
Australia, from Rugosochonetes occurred in the Visean
by modification of the external ornament (Roberts
1971). Neoehonetes arose from Rugosochonetes in the
Bashkirian, or a little earlier, and various stocks subse¬
quently developed within the genus (Archbold 1981 e).
Jakutoehonetes appears to have been a local develop¬
ment, in northeastern USSR, from Neoehonetes during
the Late Carboniferous by slight modification of the
sulcus and fold (Afanas’yeva 1977). Arctochonetes, by
modification of the ventral median septum, developed
from a Neoehonetes ancestor in the Artinskian.

Mesolobus (Sutherland & Harlow 1973) is most
closely related to Neoehonetes and probably evolved
from that genus in the early Moscovian. Hoare (1960)
considered that Eolissochonetes evolved from Meso¬
lobus but Eolissochonetes has been shown to be older
than Mesolobus (Afanas’yeva 1975a, Hoare et al. 1979).
In the late Pennsylvanian, members of Mesolobus with
an obsolescent ornament died out in North America but
in the Kasimovian of the Moscow Basin Para mesolobus,
with a stronger radial ornament flourished (Ivanov &
Ivanova 1936, Afanas’yeva 1975b) and has been widely
recorded from the Late Carboniferous of southern
Europe including Spain (Winkler-Prins 1968, 1970), and
the Karnic Alps (Schellwein 1892, Heritsch 1931,
Vinassa de Regny & Gortani 1905). Species of
Paramesolobus are usually poorly known but an

6

N. W. ARCHBOLD

approximate assessment of the Permian range of the
genus can be made from illustrated accounts of
chonetids usually referred to either of Schellwein’s
species Chonetes sinuosa or Chonetes latesinuata.
Puramesolobus is known from the Asselian-Sakmarian
of Thailand (Yanagida 1967), Japan (Nakamura 1959,
Tazawa 1976), Spitzbergen (Gobbelt 1964, pi. 15, fig.
10) and the Karnic Alps (in the form of Chonetes sp.
nov. Heritsch 1938). It has been reported from the
Artinskian of the Karakorum (Renz 1940) and the
Kungurian of China, in the form of Chonetes plicati-
forniis Chan & Lee (1962). Younger Permian forms
have been described by Coogan (1960) from California
(a form with weaker capillae), by Cooper & Grant (1975)
from Texas, in the form of Mesolobusl permianus and
a Chhidruan form is known from Japan (Hayasaka
1925, pi. 5, figs 5, 6). Cooper & Grant (1975) considered
that Mesolobusl permianus represented a convergence
towards Mesolobus, but the Texan occurrence can be
explained as a descendant species of Paramesolobus.
The new genus, discussed above within the Rugoso-
chonetinae, was apparently derived from the
Paramesolobus stock during the Sakmarian by
modification of the fold and sulcus. Tenuichonetes may
have evolved from either Mesolobus or Paramesolobus
during the Artinskian.

Chonetinella evolved during the Bashkirian,
possibly from the same stock of Rugosochonetes that
gave rise to Neochonetes, by the development of a
distinct sulcus and fold. The origin of Waagenites is
obscure. The earliest species, from the Sakmarian of the
Urals, is Chonetes (Dienerelta) J'asciger Mirskaya et al.
(1956) which possesses the characteristic deep sulcus and
very coarse costae of the genus. Waagenites faseiger has
invariably been overlooked by subsequent authors who
have assumed that Waagenites speciosus from the late
Artinskian or Kungurian of Thailand was the earliest
species of the genus. Waterhouse & Piyasin (1970) and
Grant (1976) noted that Waagenites speciosus was very
different from Waagenites grandicosta (Waagen) the
type species of the genusi Muir-Wood (1962) was not
able to elucidate all the dorsal internal structures of
Waagenites but she did indicate that the dorsal interior
was unlike that of Neochonetes. Grant (1976) stated that
the dorsal interior possessed short lateral septa, and a
short median septum, low and near the valve centre, but
unfortunately he did not figure any of his topotypes and
from his diagnosis one cannot determine the precise
relationship of the three dorsal septa. It appears that W.
speciosus does not belong in Waagenites s.s. Illustra¬
tions of the species by Grant (1976) and Yanagida (1971)
revealed that the ventral sulcus (strongly developed in
the ancestral and type species of the genus) is weakly
developed or even absent in the Thai species. Two com¬
ments can be ottered regarding the taxonomic position
of Waagenites speciosus. Firstly, the dorsal interior
structures of W. grandicosta (Waagen) require full
description and need to be figured. Secondly, the present
author agrees with Yanagida (1971) who considered that
the species is close to Neochonetes in details of mor¬

phology of the shell and dorsal internal structures.
Nevertheless the relatively coarse costcllae of the Thai
species would result in a modification of the generic
diagnosis of Neochonetes in order to accommodate the
species in that genus. A new generic name is probably re¬
quired for the species, the new genus being a develop¬
ment from a neochonetid slock. Huang (1932) described
and Liao (1980) recorded several species of “ Chonetes ”
or “Waagenites” from the Late Permian of Kweichow,
China, some specimens of which are large, coarsely
costate and possess a weak sulcus and hence they are
possibly descendants from the Thai species. Waagenites
is a generic name that should be applied with caution
until the type species is well understood. The evidence of
Mirskaya et al. (1956) strongly suggests that the genus
already possessed a well developed, deep sulcus in the
Sakmarian and hence reports of the genus from the
Chhidruan of Primorya (Licharew & Kotlyar 1978) and
Japan (Tazawa 1976) also require re-examination.

The Svalbardiinac first appeared with Leio -
chonetes during the middle Visean in New South Wales,
although Roberts (1976) discussed two other poorly
known Early Carboniferous occurrences of smooth
chonetaceans that may be allied. Leiochonetes is a
small, unspecialised smooth rugosochonetid and hence
is an ideal ancestor for the group. Leiochonetes which
possibly does not possess a pseudocapillate dorsal valve
or a similar genus, may also have independently given
rise to the Quinquenellinae by modification of the dorsal
septa (Archbold 1981a). Two principal lineages may be
delimited within the Svalbardiinae.

The Dyoros lineage, appearing in the Kasimovian
with QuadrochoneteSy is characterised by the develop¬
ment of a pronounced ventral sulcus. Arising from
Quadrochonetes in the late Sakmarian, Dyoros became
a major element of Texan Permian chonetacean faunas
(Cooper & Grant 1975). Dyoros possesses prominent
dorsal septa and this trend appears to have been accen¬
tuated by the development of Chonetinetes with a
modified elevated cardinal process. The dorsal septal ar¬
rangement of Chonetinetes is consistent with an origin
for the genus from Dyoros. The ancestry of
Quadrochonetes is not well understood as the genus
possibly lacks pseudocapillate shell structure which sug¬
gests an origin from Leiochonetes or an unknown
ancestor. Dyoros and Chonetinetes are unusual for the
Svalbardiinae in that they both possess pseudocapillate
shell structure of both valves.

The Svalbardia lineage represents a broader,
more varied development with subgroupings, although
relationships are obscured because of uncertainty over
the internal morphology of Lissochonetes. Accepting
the diagnosis of Lissochonetes given by Muir-Wood
(1962, p. 77), it appears possible to derive Sulcataria
from Lissochonetes as Sulcataria also possesses poorly
developed lateral septa with a more prominent,
posteriorly placed, dorsal median septum. Species of
Lissochonetes and Sulcataria are generally small.

Eolissochonetes possibly evolved from Lisso¬
chonetes by modification of the lateral septa and pro-

CLASSIFICATION RUGOSOCHONETIDAE

7

duction of a long thin median .septum. However, early
species of Eolissochonetes are dose to representatives of
Komiella and both genera appeared at about the same
time. The earliest species of Lissoehonetes is Lisso-
chonetes montinis (McKellar 1965) occurring as early as
the late Visean (Roberts 1975) and probably the species
evolved directly from Leiochonetes . Lissoehonetes mon¬
tinis has a dorsal septal arrangement similar to that in
Eolissochonetes morsei (Hoare et at. 1979) and is also
similar to species assigned to Lissoehonetes from the
Late Carboniferous of Kazakhstan by Sokolskaya
(1968) and the Late Carboniferous Magarsk Horizon of
the Gizhiga River Basin by Afanas’yeva (1977) that
would now be assigned to Komiella. Younger species of
Eolissochonetes (Hoare 1960, 1961) exhibit the distinc¬
tive internal morphology of the genus and are larger.
Komiella may be derived from early species of
Eolissochonetes or both genera may be derived directly
from early species of Lissoehonetes.

Leurosina is internally similar to Komiella sug¬
gesting derivation from that genus; it differs in anterior
curvature of the ventral valve and the prominent raised
anterior recurved portions of the brachial ridges
(Cooper & Grant 1975, pi. 479, fig. 78). The latter
feature also occurs in Svalbardia and CapilIonia.
Svalbardia possesses short hinge spines while the
younger and larger Capillonia possesses long delicate
hinge spines.

The two representatives of the Undulellinae are
morphologically close to the Svalbardiinae, differing
only in details of the hinge spines and the dorsal interior,
and hence the Undulellinae can be derived from the
Svalbardiinae, probably from an unspecialised “Lisso-
chonetes” or perhaps Leurosina. The origin of the
Lamellosiinae is obscure but the pseudocapillate shell
structure of Lamellosia suggests an origin within the
Svalbardiinae.

MIGRATIONS AND ENDEMISM

With the origin of the family in eastern Europe,
the descendant genus, Rugosochonetes, attained a
cosmopolitan distribution during the Carboniferous.
One descendent of Rugosochonetes namely Sehisto-
chonetes remained a localised endemic development in
northwestern Australia while another, Neoehonetes, at¬
tained a wide distribution in the Permian. Mesolobus
reveals endemic development and then extinction in the
North American Pennsylvanian, with a re-introduction
of the descendent, essentially European genus
Paramesolobus into Japan and North America in the
Late Permian. Chonetinefla has been reported widely
from the Late Carboniferous and Early Permian but it
appears premature to determine stocks within the genus
and possible migration effects. Waagertites appeared in
the Early Permian as a rare element in the Ural seas,
later spreading its range to the Tethys, including the
Caucasus (Licharew 1936), the Salt Range, Pakistan
(Waagen 1894, Reed 1944), the Himalaya (Waterhouse
& Gupta 1979) and Burma (Diener 1911). Jakuto-
chonetes is an endemic development in the Late Car¬

boniferous of the Kolyma-Omolon region of the USSR,
while Arctochonetes is restricted to the Artinskian of the
Boreal region of the USSR.

The Svalbardiinae appeared in eastern Australia
in the Visean and became widely distributed by the Late
Carboniferous. Lissoehonetes and/or Komiella had
penetrated North America, Kazakhstan, northeastern
USSR, European USSR (Ivanov& Ivanova 1936), Spain
(Winkler-Prins 1968) and South America (Amos 1960,
Mendes 1959) by the Late Carboniferous. Endemic
development took place in North America in the Late
Carboniferous with the development of Quadro¬
chonetes. During the early Permian, endemic develop¬
ment of the Quadrochonetes stock continued in North
America with the development of the genus Dyoros
which subsequently penetrated the Boreal sea in the Late
Artinskian and Kungurian (Ifanova 1968, 1972).
Solomina (1978, p. 106, pi. 9, fig. 3) recorded, with
some question, the possible occurrence of Dyoros in the
Late Carboniferous Khaldan Suite from the Southern
Orulgan region of northeastern USSR. Her specimens
are inadequate for precise determination, nevertheless,
the illustration indicates a strongly sulcate form that
may be a species of Quadrochonetes or Dyoros. Dyoros
entered northern Gondwana waters during the
Chhiduran in the form of ISulcataria pentagonal
(Waterhouse 1978b, Waterhouse & Gupta 1979).
Sulcataria and Leurosina appear to be endemic
developments of the Svalbardiinae in North America.
Svalbardia exhibits a bipolar or disjunct distribution in
the Kungurian (Archbold 1981b), while Capillonia is
restricted to the Kazanian and younger Permian of New
Zealand (Waterhouse 1973) and eastern Australia in the
form of Lissoehonetessemicircularissolida (Dear 1971).

The Undulellinae are restricted to the Permian of
North America as are the Lamellosiinae. The Quin-
qucncllinae exhibit a bipolar or disjunct distribution in
the Permian, being found in Western Australia, the
Himalaya, possibly south-east Asia and northeastern
USSR (Archbold 1981a). The occurrences of Para¬
mesolobus , Dyoros and Rugaria on both sides of the
Pacific at times during the Permian indicate some poten¬
tial for migration between the two regions by several
rugosochonctid genera.

ACKNOWLEDGEMENTS

I thank Dr. G. A. Thomas, University of
Melbourne, for critically reading an earlier draft of the
manuscript. The assistance of the staft' of the Baillieu
Library is acknowledged. Isabel McDonald typed the
manuscript.

REFERENCES

References not listed here will be found in Archbold
(i 9 8ie).

Afanas’yeva, G. A., 1978b. Novyye Khonetatsei is devona
Nakhichcvanskoy ASSR. Paleont. Zh. 1978(3): 64-71.
Amos, A. J., 1960. Algunos Chonetacea y Productacea del
Carbonifero Inferior y Superior del Sistema de Tepuel,

8

N. W. ARCHBOLD

Provincia de Chubut. Revta. Assoc, geol. argent. 15:
81-107.

Archbold, N. W., 1981 e. Studies on Western Australian Per¬
mian brachiopods. 2. The family Rugosochonetidae.
Proc. R. Soc. Viet. 93: 109-128.

Brunton, C. H. C., 1972. The shell structure of Chonetacean
brachiopods and their ancestors. Brit. Mus. (Nat.
Hist.) Bull. Geol. 21(1): 1-26.

Chan, Li Pei & Li, Li, 1962. Early Permian brachiopods from
the Maokou Suite of east Uchatka Chin Lin. Acta.
Palaeont. Sin. 10(4): 472-493.

Coogan, A. H., 1960. Stratigraphy and paleontology of the
Permian Nosoni and Dekkas formations. Geol. Dept.
Univ. Calif. Pub/. 36(5): 243-316.

Cooper, G. A., 1970. Generic characters of brachiopods.

Proc. Nth. Atner. Paleont. Conv. C.: 194-263.
Cooper, G. A. & Grant, R. E., 1969. New Permian
brachiopods from West Texas. Smithson. Contrib.
Paleobiol. 1: 1-20.

Easton, W. H., 1962. Carboniferous formations and faunas
of central Montana. Prof Pap. U.S. geol. Surv. 348:
1-126.

Geinitz, H. B., 1867. Carbon formation und Dyas in
Nebraska. Nova. Acta Leopold., deutsche Akad.
Natur. Leopold., Halle 33: 1-91.

Girty, G. H., 1915. Fauna of the Wewoka formation of
Oklahoma. Bull. U.S. geol. Surv. 544: 1-353.

Hall, J., 1843. Geology of New York. Part IV. Comprising
the Survey of the Fourth Geological District. Carroll &
Cook, Albany. 783p.

Hall, J., 1867. Natural History of New York. Palaeontology.
Containing descriptions and figures of the fossil
Brachiopoda of the Upper Helderberg, Hamilton, Por¬
tage and Chemung Groups. Nat. Hist. N. Y. pi. VI,
palaeont. 4(1): 1-428.

Heritsch, F., 1931. Verstcinerungen aus dem Karbon der
Karawanken and Karnischen Alpen. Abhand. Geol.
Bundesanst. Wien 23(3): 1-56.

Heritsch, F., 1938. Die stratigraphische stellung des

Trogkofelkalkes. N. Jahr. Min. Geol. Palaeont., B 79:
63-186.

Hoare, R. D.. 1960. New Pennsylvanian Brachiopods from
south west Missouri. J. Paleont. 34: 217-232.

Hoare, R. D., 1961. Desmoinesian Brachiopoda and Mollusca
from south west Missouri. Univ. Missouri Studies 36:
1-263.

Hoare, R. D., Anderson, J. R. & Sturgeon, M. T., 1979.
Eolissochonetes morsei n. sp. (Brachiopoda) from the
Pennsylvanian of Kentucky. J. Paleont. 53: 1179-1181.
Huang, T. K., 1932. Late Permian brachiopoda of south west
China. Palaeont. Sinica. Ser. B. 9(1): 1-138.

Ifanova, V. V., 1968. Nekotoryye Rannepermskiye

Chonctidac Pechorskogo Basseyna. Paleont. Zh. 1968
(3): 29-33.

Ifanova, V. V., 1972. Permskie brakhiopody Pechorskogo
Basseina. In, Ifanova, V. V. & Semenova, E. G.
SrednekamennougoTnye i Permskie brakhiopody
vostoka i severa evropeiskoi chasti SSSR. Akad. Nauk.
SSSR. Izd-vo “Nauka”, Moskva, pp. 72-161.

Jing, Y. & Hu, S., 1978. Brachiopods of the Kuhfeng Forma¬
tion in South Anhui and Nanking Hills. Acta Palaeont.
Sinica 17: 101-127.

Johnson, J. G., 1966. Parachonetes, a new lower and middle
Devonian brachiopod genus. Palaeontology 9:
365-370.

Johnson, J. G., 1970. Great Basin lower Devonian

Brachiopoda. Mem. geol. Soc. Amer. 121: 1-421.

King, R. H., 1965. The chonetid brachiopod Mesolobus and
some of its species. J. Paleont. 39: 293-295.

Liao Zhuo-Ting, 1980. Brachiopod assemblages from the
Upper Permian and Permo-Triassic boundary beds,
South China. Can. .J. Earth Sci. 17: 289-295.

Licharew, B. K., (Likharev, B. K.), 1934. Fauna permskikh
otlozhenii Kolymskogo Kraya. Akad. Nuak. SSSR,
Trudy Sow izuchen. proizvod. sil. ser. Yakutskava
14(1)’ 1-148.

Licharew, B. K., (Likharev, B. K.), 1936. Permskie
Brachiopoda severnogo Kavkaza. Semeistva:
chonetidae Hall et Clarke i Produetidae Gray. Mono,
po Paleont. SSSR 39(1): 1-151.

Licharew, B. K. & Kotlyar, G. V., 1978. Permskie
brakhiopody Yuzhnogo Primor’ya. In, Verkhnii
Paleozoi Severn Vostochnoi Azii. Vladivostok, pp
63-75, 96-99.

Mikryukov, M. F., 1969. Novye rod i vid khonetid Russkoi
Platformy. In, Markovsky, B. P., (ed.). Novye Vidy
Drevnikh Fastenni i Beznozvoiochnykh SSSR.
“Nedra” Moskva, Vyp. 2: 90-92.

McKellar, R. G., 1965. An upper Carboniferous brachiopod
fauna from the Monto District, Queensland. Publ.
geol. Surv. Qld. 328: 1-15.

Mendes, J. C., 1959. Chonetacea e Productacca Car¬
boniferous de Amazonia. Univ. Sao Paulo, Fac. Filos.,
Cienc. Letras, Bol. 236, Geol. 17: 1-83.

Mirskaya, M. F., Shestakova, M. F. & Chudinova, I. L,
1956. O nekotorykh novykh okamenelostyakh is
nizhnepermskikh otlozhenii Kamskogo Priurafya.
Uchenye Zapiski Molotovskogo Gos. Un-ta. 7(4):
27-43.*

Mudge, M. R. & Yochelson, E. L., 1962. Stratigraphy and
paleontology of the uppermost Pennsylvanian and
lowermost Permian rocks in Kansas. Prof Pap. U.S.
geol. Surv. 323: 1-213.

Nakamura, K., 1959. Some Lower Permian Sakamotozawa
brachiopods. J. Fac. Sci. Hokkaido Univ. Ser. 4 10:
199-207.

Norwood, J. G. & P ratten, H. 1855. Notice of the genus
Chonetes as found in the western states and territories
with descriptions of eleven new species. J. Acad. nat.
Sci. Philadelphia, Ser. 2 3: 23-32.

Paeckelmann, W., 1930. Die Brachiopoden des deutschen
Unterkarbons. 1 Teil: Die Orthiden, Strophomeniden
und Choncten des mittleren und oberen Unterkarbons.
Abh. Preuss. Geol. Landesanst. N.F. 122: 143-326.

Prokof’ev, V. A., 1975. Brakhiopody verkhnego Karbona
Samarskoi Luki. Trudy Kses. Nauchno-issled.
Geol.-razved. Neft. Inst. (VNIGNJ) 162: 1-144.

Ramsbottom, W. H. C., 1952. The fauna of the Cefn Coed
Marine band in the coal measures at Abcrbaiden, near
Tondu, Glamorgan. Bull. geol. Surv. Gt. Br. 4: 8-32.

Reed, F. R. C., 1931. New fossils from the Productus
Limestone of the Salt Range with notes on other
species. Mem. geol. Surv. India, Palaeont. Indica,
N.S. 17: 1-56.

Reed, F. R. C., 1936. Some fossils from the Eurydesma and
Conu/aria beds (Punjabian) of the Salt Range. Mem.
geol. Surv. India, Palaeont. Indica, N.S. 23(1): 1-36.

Reed, F. R. C., 1939. Non-marine lamellibranchs etc., from
the “Speckled Sandstone” Formation (Punjabian) of
the Salt Range. Rec. geol. Surv. Ind. 74: 474-491.

Reed, F. R. C., 1944. Brachiopods and Mutlosca of the Pro¬
ductus Limestone of the Salt Range. Mem. geol. Surv.
India, Palaeont. Indica , N.S. 23(2): 1-678.

Renz, H., 1940. Die Palaozoischen faunen von 1935. Meta-

CLASSIFICATION RUGOSOCHONETIDAE

9

zoen. Wiss. Ergeb. Niederland. Exped, Karakorum.
3(1), 2: 118-247. E. J. Brill, Leiden.

Roberts, J., 1971. Devonian and Carboniferous brachiopods
from the Bonaparte Gulf Basin, North western
Australia. Bull. Bur. Miner. Resour. Geol. Geophys.
Aust. 122, vol. I: 1-319, vol. 2, 59 plates.

Roberts, J., 1975. Early Carboniferous brachiopod zones of
eastern Australia. ./. geol. Soc. Aust. 22: 1-31.
Roberts, J., 1976. Carboniferous chonctacean and producta-
coan brachiopods from eastern Australia. Palaeon¬
tology 19: 17-77.

Sciiellwien , E., 1892. Die Fauna des Karnischen Fusu-
Iinenkalks. Palaeontographica 39: 1-56.

Sikstell, T. A., (ed.), 1975. Biostratigrafiya verkhnego
paleozoya gornogo obramfeniya yuznoi Fergany .
Minist. Geol. Uzbek. SSR, Sred. nauchno-issled. inst.
geol. i mineral. Syr’ya, (SAIGIMS). Tashkent.
Sokolskaya, A. N., 1946. Osnovnye puti evolvutsii semeistva
Chonetidae. A kad. Nauk SSSR, Otdel biol Nauk, Izv.
6: 731-740.

Sokolskaya, A. N., 1949. Vozrastiyc izmeneniya khonetid i
ikh taksonomicheskoe znachenie. A kad. Nauk SSSR,
Paleont. Inst., Trudy 20: 268-278.

Sokolskaya, A. N., 1950. Chonetidae russkoi platformy.

A kad, Nauk SSSR, Paleont. inst., Trudy. 27: 1-108.
Sokolskaya, A. N., 1960. Nadsemeistva Chonctacea. In,
Orlov, Yu. A. (ed.). Osnovy Paleontologii 15: 221-223.
Moskva.

Sokolskaya, A. N., 1968. Podotryad Chonetidina. In,
Sarycheva, T. G. (ed.) Brakhiopody verkhnego
paleozoya vostochnogo Kazakhstana. A kad. Nauk
SSSR, Paleont. Inst., Trudy 121: 66-73.

Solomina, R. V., 1978. Nekotorye sredneverkhnekamen-
nougol’nye brakhiopody severnogo verkhoyan’y'a.
A kad. Nauk SSSR, Sibirsk. Otdel., Inst. Geol. Geofiz.,
Trudy 386: 99-123.

Stehli, F. G., 1954. Lower Leonardian Brachiopoda of the
Sierra Diablo. Bull. Amer. Mus. Nat. Hist. 105:
261-358.

Stepanov, D. L., 1973. The Permian System in the USSR.

Mem. Can. Soc. Petrol. Geol. 2: 120-136.

Sturgeon, M. J. & Hoare, R. D., 1968. Pennsylvanian
brachiopods of Ohio. Bull. geol. Surv. Ohio 63: 1-95.

Sutherland, P. K. & Harlow', F. H., 1973. Pennsylvanian
brachiopods and biostratigraphy in southern Sangre dc
Cristo Mountains, New Mexico. Mem. New Mexico
Bur. Mines Miner. Resour. 27: 1-173.

Tazawa, J., 1976. The Permian of Kesennuma, Kitakami
Mountains; a preliminary report. Earth Science
(Chikyu Kagaku) 30(3): 175-185.

Ting, P., 1965. The Permian and Triassic brachiopods from
Yangkang Valley, Tienching District, Tsinghai Pro¬
vince, Acta Palaeont. Sinica 13: 260-290.

Vinassa De Regny, P. & Gortani, M., 1905. Fossili car-
boniferi del M. Pizzul e del piano di Lanza nellc Alpi
Carnichc. Bull. Soc. Geol. Ital. 24: 521-597.

Waterhouse, J. B. v 1973. New brachiopod genera from the
New Zealand Permian. J. Roy. Soc. N.Z. 3: 35-42.

Waterhouse, .1. B., 1975. New Permian and Triassic
brachiopod taxa. Pap. Dep. Geol. Univ. Qld. 7: 1-23.

Waterhouse, J. B., 1978a. Chronostratigraphy for the world
Permian. Amer. Avsoo. Petrol. Geol., Studies in
Geology 6: 299-322.

Waterhouse, J. B., 1978b. Permian Brachiopoda and
Mollusca from north west Nepal. Palaeontographica
Abt. A. 160: 1-175.

Waterhouse, J. B. & Gupta, V. .1., 1979. Late Middle Per¬
mian brachiopods from Marbal Pass, Kashmir, India.
Bull. Ind. Geol. AstfQCl 12(1): 1-42.

Waterhouse, J. B. & Piyasin, S., 1970. Mid-Permian
brachiopods from Khao-Phrik, Thailand. Palaeon¬
tographica Abt. A. 135: 83-197.

Weller, J. M. & McGehee, R., 1933. Typical form and range
of Mesolobus mesolobus. J. Paleont. 7: 109-110.

Winkler-Prins, C. F., 1970. Brachiopod descriptions. In,
Wagner, R. H. & Winkler-prins, C. F. The
stratigraphic succession, flora and fauna of Cantabrian
and Stcphanian A rocks at Barruelo (prov. Palencia)
N.W. Spain. Univ. Liege, Colloquie St rat. Car¬
bon if ere, Cong. Colloques 55: 531-542.

Yanagida, J., 1967. Early Permian brachiopods from north-
central Thailand. Geol. Palaeont. S.E. Asia 3: 46-97.

Yanagida, J., 1971. Permian brachiopods from Khao-Phrik
near Rat Buri, Thailand. Geol. Palaeont. S.E. Asia 8:
69-96.

NOMENCLATURAL NOTE

Sommeriella , a new name for the Permian chonetacean brachiopod subgenus

Sommeria Archbold 1981

It has been pointed out by Dr. R. W., Hud¬
dleston (Chevron Oil Field Research Company, La
Habra, California) that the subgeneric name Sommeria
Archbold 1981 is a junior homonym of Sommeria
Huebner 1825, a genus of African Lepidoptera. I
therefore propose the new name Sommeriella to replace
Sommeria Archbold 1981 for the Permian brachiopod
subgenus.

REFERENCES

Archbold, N. W. 1981. Studies on Western Australian Per¬
mian brachiopods. The Family Rugosochonetidae
Muir-Wood 1962. Proc. R. Soc. Viet. 93: 109-128.

Huebner, J., 1818-1825. Zutrage zur Sammlurig exotischer
Schmetterlinge bestehead in Bekundigung einzelner
Fliegmuster neuer Oder rarer Nichteuropdisher Gat -
tungen . Drittes Hundert. 40 p. 172 pis. Augsburg.

N. W. Arch bold
Department of Geology
University of Melbourne
Parkville , Victoria 3052

PROC. R. SOC. VICT. vol. 94, no. 1, 11-22, March 1982

CHANNEL INCISION AT EAGLEHAWK CREEK,
GIPPSLAND, VICTORIA, AUSTRALIA

By Juliet F. Bird

Department of Geography, Melbourne State College, 757 Swanston Street, Carlton, Victoria 3053

Abstract: A hundred years ago, when the land around Glengarry was subdivided into small
farms, Eaglehawk Creek was a shallow shifting watercourse interspersed with patches of swamplands.
Efforts to drain the land and contain the creek have led to incision, creating a channel up to 15 m deep.
The progress of incision can be documented from historical evidence including the personal recollection
of residents of the area. Phases of rapid deepening associated with major floods have alternated with
periods of stability recorded as terraces in the channel sides. The present phase <5f relatively slow erosion
probably reflects the absence of high flows in the last three years rather than the attainment of a stable
grade.

Many of the river channels in Ciippsland, in
eastern Victoria have undergone extensive geomor-
phological change since the area was settled by Euro¬
peans 150 years ago. Among the most striking examples
of change is the channel incision which has occurred
along the northern tributaries of the Latrobe River. This
paper describes the history of that erosion with par¬
ticular reference to Eaglehawk and Stoney Creeks (Fig.
1), and discusses the factors that have led to the conver¬
sion of the shifting surface streams of the pre-
agricultural landscape into the gullies up to 15 m deep
which are seen in the area today (Fig. 2).

In the United States there have been many studies
of channel changes in the post settlement period: while
most researchers have concluded that human in¬
terference has been the main cause, a few, such as Bull
(1964) have suggested that in some cases climatic events,
such as periods of exceptionally heavy rainfall, may
have been significant. Cooke and Reeve (1976) reviewed
the extensive literature on arroyos, deep gullies which
are widespread in the western USA. They concluded
from their own studies in Arizona and California that
arroyos resulted from floristiC change, particularly that
associated with livestock grazing, and destruction of
valley floor vegetation with a consequent increase in the
susceptibility of channels to erosion. Malde and Scott
(1977), in a study of contemporary arroyo development
near Santa Fe in New Mexico, analysed the processes of
gullying but emphasised the problem of isolating any
single cause, because the onset of erosion in the 1880s
coincided with several other changes, including a transi¬
tion from grassland to scrub vegetation, and a general
lowering of water tables in the area. In the mid-west of
the United States channel erosion is also prevalent.
Daniels (1960) presented a case study of Willow County
Ditch in Iowa, which has entrenched several metres into
its former floodplain since the beginning of the century;
he considered that while regional modification of run off
may have been a contributory factor, the most impor¬
tant cause was an increase in stream gradient resulting
from channel straightening.

In Europe eroding channels are less common, but
an example is the Devon gully described by Gregory and
Park (1976) which was believed to be due to modifica¬
tion of run off caused by urbanisation. Incision of chan¬
nels in a forested area in Luxembourg was also ascribed
to change in surface run off, in this case the result of
concentration along culvert underpasses and drainage
ditches (Imeson & Jungerius 1977).

Recognition of the impact of settlement on river
channels in Australia came early, and the remarks of
John Robertson (1853, quoted in Bride 1898) concerning
the development of gullying in the Western District of
Victoria are often quoted. Abbott (1884, p. 105) was
more specific when he described changes in run-ofi in in¬
land New South Wales: “The difference between stocked
country and that which has never been stocked is ap¬
parent even after a few years. The surface becomes
firmer and water runs where it never ran before . . it

does not now take half the amount of rain to put water
in the rivers that it did thirty years ago, just after it was
first settled.”

More recently Woodyer (1968) emphasised the
dominance of incised channel forms in New South
Wales; he quoted Dury as saying that these had
developed in the 200 years since European settlement.
Pickup (1975) analysed the process of incision along
Crawfords Creek, near Picton, where instability was in¬
itiated by a severe flood in 1949. In a later paper Pickup
(1976) described a regional prevalence of incision with
particular reference to the Cumberland Basin, but here
analysis of precipitation records, which extend back as
far as 1880, suggested a recent increase in rainfall with
consequent growth in the magnitude and frequency of
flooding as the most likely cause.

Goedc (1972), in a study of Tea Tree Rivulet in
northeastern Tasmania, emphasised the role of vegeta¬
tion clearing, particularly alongside the gullied sector, in
promoting channel instability, which was probably also
influenced by a trend towards an increased number of
small rain falls since about 1917. In South Australia
channel deepening over the past century was described

12

J. F. BIRD

Fig. 1 — Study area on Eaglehawk and Stoney Creeks.

CHANNEL INCISION GIPPSLAND

13

Fig. 2 —Eaglehawk Creek 200 m downstream from Christensons, looking downstream. Figure arrowed for scale.

by Bourmann (1976) who attributed it to increased
runoff following deforestation and ploughing. Bour¬
mann also cited specific cases of erosion developing
along channels dug to drain swamplands.

In Victoria Bird (1980) concluded that incision
along the Lang Lang River had several causes, which she
was unable to rank in order of importance. In the lower
part of the valley, swamps had been drained and the
river course had been straightened, which resulted in
lowered base levels and increased channel gradient.
Upstream sectors had also been straightened, while con¬
struction of a levee bank blocked a depression which
formerly permitted egress of flood waters from the Lang
Lang channel onto swamplands to the north.

Eaglehawk and Stoney Creeks are particularly
favourable for a study of geomorphological change. Air
photographs arc available for 1935, 1964, 1970 and
1976. A railway line, built in 1883, crosses Eaglehawk
Creek just below the main incised sector and
maintenance surveys made by the Victorian Railway
Authority record erosion and sediment deposition at the
bridge site at frequent intervals. Most of the changes
have occurred within the last seventy years, so that
farmers in the area today recall them either personally,
or through stories related by their parents, many of
whom were the original settlers who opened up the land
for agriculture. While personal memory is often an
unreliable source of evidence, it is likely to be accurate
where events, such as destruction of important access
tracks, seriously affect farm management.

THE LOCAL SETTING

Eaglehawk and Stoney Creeks rise in Palaeozoic
hill country, at an altitude of about 380 m and in the up¬
per part of their courses they flow through dissected ter¬
rain covered with open eucalypt forest which has been
extensively logged. Part of the area has been cleared and
replanted with conifers for a nearby paper mill. On leav¬
ing the forest the streams cross a foothill zone formed by
coalescing fans with surface gradients of 0.010. Under
natural conditions the fans enclosed swampy tracts (Fig.
1), where stream courses became indistinct before re-
emerging as defined channels to cross the terraces of the
Latrobe. Eaglehawk Creek flows directly into the
Latrobe River, unlike the other watercourses in this
region, which terminate in the back-swamps at the edge
of the flood plain.

The foothill zone was originally occupied by
squatters in the 1840s but little vegetation clearing oc¬
curred until after the land was resumed for release under
the Selection Acts of the 1860s. Small farmers were then
encouraged to settle in the area to grow food for the
miners who passed through on their way to the Walhalla
goldfields, but the swampy terrain proved unsuitable for
cropping, and there was little progress until the con¬
struction of the railway in 1883. After that improved
transport ensured rapid development, mainly of dairy
farming to supply local butter factories, and by the
1930s most of the forest had been cleared and replaced
with improved pasture.

14

J. F. BIRD

THE DEVELOPMENT OF EROSION

The shifting stream courses (Fig. 1) (The infor¬
mation on flow routes and vegetation derives from the
land selection files of the 1860s, many of which include
plans of the individual subdivisions.) were a nuisance to
the local farmers, and in 1894 they tried to control
Eaglehawk Creek by constructing a dam to block the
flood overflow which formerly travelled across a low' col
into the catchment of Four Mile Creek. About the same
time, the farmer who owned the land west of the
Glengarry North Road enlarged one of the channels tak¬
ing Eaglehawk Creek across his property with the dual
aim of eliminating the changes of course and draining
the swamps. The local Shire Council (Rosedale) then ex¬
tended this channel east of the road, directing the water
into the depression leading to the northernmost of the
three bridges which carried the railway across branches
of Eaglehawk Creek. Present stream alignment still
follow's this northern route, though subsequent erosion
and meandering have eliminated the signs of its man¬
made origin.

The first record that Eaglehawk Creek was
changing dates from 1910, when engineers noted in¬
stability beneath the railway bridge; nine years later the
bridge had to be partially rebuilt following a flood which
widened the underlying waterway. Thereafter it was
sediment deposition rather than erosion which caused
bridge maintenance problems. Incision becomes evident
about 1 km above the railway, though in this lower part
the incised channel is partially infilled with sediment.
Evidence for the early development of erosion is sparse,
but in 1914 Eaglehawk Creek was still small enough to
be crossed by a minor footbridge at Christensons, while
each winter floods caused problems when they over¬
flowed down the access track to the local school. By
1920 there was a large waterfall 300 m above Langs, and
in 1929 the channel just above the Glengarry North
Road was said to be very deep. (Mrs Timmins pers.
comm.). The 1935 air photographs show' active incision
extending to about 500 m below Wellingtons, and
ground photographs taken at this time indicate that the
channel was already 10 m deep above the junction with
Stoney Creek. By 1939 the Forest (now Frasers) Road
crossing at Wellingtons was threatened, and ten years
later erosion at this point was described as severe. Above
the bridge, bedrock outcropping in the channel slowed
the progress of erosion, but where the valley is infilled
incision is still evident 1 km upstream.

Under natural conditions Stoney Creek, like
Eaglehawk, lacked a defined channel across the fan zone
and was prone to change its course. At the time of settle¬
ment it flowed on an alignment similar to the present one
as far as the site of Bermingham’s Farm, then became
dispersed in a swamp before reforming into a channel
just above the railway. According to a local farmer
(Jack Lang pers. comm.) his father, in 1932 or 1933, had
been impressed with the effectiveness of the eroding
Eaglehawk channel for swamp drainage and the rapid
removal of floodwater from the hills so he decided to
direct Stoney Creek into it by way of a ditch around the

base of the spur dividing the two catchments. Incisiop
spread rapidly back up the ditch into the old channel
with the head of the erosion already 500 m above th c
confluence by 1935. By the early 1940s incision extended
back to Berminghams, where it repeatedly cut the access
track to the hills: attempts to maintain a bridge at this
point (Fig. 3) were finally abandoned about 1952. (Dat£
from Raymond Smith, son of the owner of a property
upstream from this point, who depended on this bridge
for access.) Figure 4 includes a cross section of the chan-
nel at the bridge site today. Like Eaglehaw'k, Stoney
Creek has now cut down to bedrock in several places,
and although further upstream extension of incision is
likely, it will be a much slower process than in the lower
course.

An average figure for the rate of upstream
transmission of the head of gullying can be deduced
from the historical evidence. Over the period 1920-1935
it moved from the vicinity of the Glengarry North road
bridge to above Christensons, a distance of 2 km, a rate
of about 133 m a year. Between 1935 and 1939, wheji
erosion is recorded at Wellingtons, the head of the inci¬
sion migrated a further 500 m, indicating a similar an¬
nual rate. In Stoney Creek movement in the first 3 years
was very rapid —500 m — but it subsequently slowed to a
rate comparable to that in Eaglehawk Creek.

Average figures for headward extension though
useful for comparison with gully development else¬
where, are somewhat misleading, because it is clear that
incision has been an episodic process, with periods of
rapid development in high flows interspersed with years
of near stability w'hen no major floods occurred. Valley
side benches in the channel of Eaglehawk Creek between
Stoney Creek and Wellingtons are relics of former chan¬
nel floor levels temporarily stabilised behind weirs. In
the early years incision appears to have been ac¬
complished through the headward migration of promi¬
nent knickpoints, which may have originated on harder
bands within the fan sediments, or behind weirs which
became breached. About 1952 the head wall which had
marked the upstream limit of incision became degraded
into a series of rapids and though a few steps persist
behind intact drop structures, or associated with rocky
sectors or blocking logs, none are more than 40-60 cm
high. Even the upper limit of incision is ill defined, as
above Frasers Road stable rocky sectors alternate with
patches of infill where the former valley floor is pre¬
served as a terrace 1-2 m above the present channel
floor. The contrast between the present smooth profile,
and the irregular one described in early records may be
due to diminution of downcutting due to the attain¬
ment of a stable grade, or may reflect a change in the
nature of the sediment load. In the early years incision in
the lower part of the fan released a load dominated by
silt, sand and gravel; as erosion has extended back into
the hills it has cut into increasingly coarse material so
that it is likely that a greater proportion of the sediment
now moves as bcdload. Goede (1972) suggested that the
absence of a prominent hcadwall in an eroding gully
may be attributed to a preponderance of coarse material

CHANNEL INCISION GIPPSLAND

15

\ 1 * v '

,.,r •• W‘ ■>

B

Fig. 3-A, Bridge across Stoney Creek at Berminghams, looking upstream. About 1936. (Source: R. Smith, Glengarry). B. Same
site in August 1944. Remains of the old bridge in the foreground. (Source: State Rivers and Water Supply Commission photograph

collection).

16

J. F. BIRD

Fig. 4 —Cross-sections of Eaglehawk and Stoney Creeks at the positions indicated.

CHANNEL INCISION GIPPSLAND

17

in the sediment load. The change in sediment size may
also be reflected in the the changing shape of the cross
profile. Historical data suggest that deepening was
dominant in the early years, so that sections attained a
depth of 10 m or more by 1935; since that time the max¬
imum depth has only increased to around 15 m, but the
rate of sediment delivery downstream, together with the
visible signs of bank undercutting, show that widening,
particularly on bends, is still active. Schumm (1977) has
demonstrated that channel width-depth ratios are
related to sediment size, and the transition from domi¬
nant down-cutting to dominant widening may be
associated with the increasing proportion of bed load in
transit. This may also explain the lower width-depth
ratios noted in Stoney Creek, which is deeper and nar¬
rower (Fig. 4) and carries finer debris.

SEDIMENT DEPOSITION

Erosion along Eaglehawk and Stoney Creeks has
mobilised an estimated 750 000 m 3 of sediment ranging
in size from cobbles 6 cm or more in diameter down to
fine sand and silt. While the former remain on the bed of
the higher reaches, and much of the silt has been carried
as suspended load through to the Latrobe River,
redeposition of the sand and gravel components of the
load has created many problems. Where the creek bed
has deepened most of this material has been retained on
the channel floor, forming a deep, porous infill, so that
surface water flow is much less common than it used to
be, thereby reducing the value of the stream as a source
of stock water. The area within which this has occurred
is shown as a slight convexity on the long profile. The
most damaging effect of sediment deposition has been
the blocking of the waterway beneath the rail and main-
road bridges. The original 1881 railway survey and a
profile of the bridge site drawn in 1919 both show a
clearance of nearly 3 m between river bed and bridge
deck. 13y 1933 this waterway had been almost obliterated
by deposition within the channel and early in 1934 floods
banked up behind the bridge to flood 10 hectares of
land, leaving 1.5 m of sediment in places. Later that year
a flood of even greater magnitude swept the channel
clear, transporting sediment further downstream and
restoring the waterway profile of 15 years earlier. The
improvement was shortlived, and by 1935 railway
authorities were forced to excavate material from
beneath the bridge to re-establish the channel and pre¬
vent the creek from flooding over the line and disrupting
train services. For a time it looked as if Eaglehawk
Creek might abandon this course altogether and revert
to one of its old routes to the south. The railway
authorities were unconcerned; they already had bridges
that would cope with such a change because when the
line was constructed the creek still sometimes flowed in
this direction. It was seen as more of a threat by the
Shire Council and the Country Roads Board, which
maintained the road. The road, built after the railway,
had made no provision for an alternative waterway, so
that a course diversion would necessitate the construc¬
tion of a new bridge.

The deposition of sediment across farmland in

the region known locally as the “delta” has been
detrimental to those farmers directly affected, as the
value of pasturelands is reduced when they are
blanketed by several centimetres of coarse sand and
gravel. The cumulative effect of sediment deposition has
been to raise land levels by 2 or 3 m in places, burying
fence posts and tree trunks.

ATTEMPTS TO CONTROL EROSION AND
SEDIMENTATION

In the 1930s attempts to deal with problems
resulting from the erosion of Eaglehawk and Stoney
Creeks were confined to periodic excavation of the
waterway beneath the main-road and rail bridges, but it
became increasingly evident that the problem of infilling
could only be solved by the insertion of sediment traps
and erosion controls upstream. At first attempts to do
this were on a very small scale consisting mainly of
brushwood weirs across the channel floor. None of the
structures lasted more than a few months before they
were washed out by floods, and though the Rosedale
Shire Council had consultations with the Soil Conserva¬
tion Board and the State Rivers and Water Supply Com¬
mission in 1942 over a proposal for construction of a
major dam across Eaglehawk Creek, no further action
was taken for some years. In 1949 the Country Roads
Board replaced the old main-road bridge with a new one
having a 2 m clearance above the river bed; the follow¬
ing year the Railways, too, replaced their bridge with a
new, higher level structure on what was considered an
improved alignment. Floods in 1952 washed out the ap¬
proach to the new road bridge (Fig. 5) and went over the
raised rail bridge, but problems at the bridge sites have
subequently diminished.

In 1950, following the failure of the early at¬
tempts to check erosion and sedimentation, local
farmers met to form the Eaglehawk Creek Control
League; in 1953 this was taken over by the Latrobe River
Improvement Trust when its area of jurisdiction was ex¬
tended to include Eaglehawk Creek. In 1956 the Trust,
in conjunction with the State Rivers and Water Supply
Commission, constructed extensive new' erosion control
works, consisting of about 45 stone and wire-mesh
weirs, placed at intervals along Eaglehaw'k and Stoney
Creeks, but within a few months of completion more
than two-thirds had been rendered useless because the
river outflanked them by undercutting the soft bank
sediments. The broken weirs then made the situation
worse, because they trapped tree trunks and branches
which fell into the river as the forested bank eroded,
forcing the creek to migrate around them. By 1957
Eaglehawk Creek just upstream from the confluence
with Stoney Creek occupied a vertical sided slot 13 m
deep and 30 m wide. For a decade attempts to control
erosion were abandoned because it was clear that only a
large dam and sediment trap w r ould be effective, and its
cost was more than was justified now that the new road
and rail bridges had been constructed overcoming the
problem of severing the major transport link through
the area.

In 1968 a further attempt to control erosion

18

J. F. BIRD

Fig. 5 — Aerial photograph of the main road and rail bridges taken just after the 1952 floods, showing sand splays. (Source: Vic¬
torian Railways).

through insertion of concrete drop structures was made
using Drought Relief Funds to provide local employ¬
ment, but a flash flood of January 1971 caused severe
damage to the new weirs (Fig. 6), and today there are
none on Stoney Creek, and only two intact on the upper
part of Eaglchawk. A concrete block structure protects
the Glengarry North Road Bridge, while a low concrete
wall across the creek bed just above the main bridge
traps some of the sediment arriving from upstream.
Current erosion “control” is limited to annual bulldoz¬
ing by the Latrobe River Improvement Trust designed to
remove obstacles from the channel floor, and redirect
the current away from the eastern bank where under¬
mining results in the loss of valuable farmland. The
enlarged waterway beneath the bridges and periodic
straightening of the channel through the sediment splays
has resulted in progressive downstream migration of the
toe of the “delta” or sand splay area. If the process con¬
tinues coarse sediment may eventually be delivered onto
the flood plain of the Latrobe River or even, like the silt,
into the Latrobe River itself.

CAUSES OF INSTABILITY AT EAGLEHAWK AND
STONEY CREEKS

The historical and field evidence so far reviewed
indicates that erosion began along Eaglehawk Creek
about the turn of the century, with a phase of rapid
deepening lasting until about 1940 since when channel

widening has been dominant except in peak flood years
such as 1952, 1971 and 1978, when a new channel has
been incised into the pre-existing creek floor. Incision of
Stoney Creek started in the early 1930s, but like
Eaglehawk Creek it is now tending to widen, rather than
deepen its channel.

It is unlikely that precipitation in the area has
changed sufficiently to trigger channel instability. Local
rainfall recording began in the 1880s (Rainfall figures
analysed include Toongabbie (daily and monthly)
1889-1920, 1929-1930; Morwcll (monthly) 1888-1891,
1899-1915, 1934-80; Traralgon (monthly) 1903-1962.)
and analysis of five year running means shows no signifi¬
cant increase since then. Retention of a forest cover over
the upper part of the catchments makes it unlikely that
the flow regime within the creek has changed greatly,
though logging may have had a temporary effect on the
magnitude of flood peaks. Bank erosion is widespread
along the Latrobe River, but has not been transmitted
up Eaglehawk Creek. Two weirs on the lower end of the
Creek stop any headward migration of instability and
serve as partial silt traps.

Early maps show that the fans once carried open
forest vegetation interspersed with denser areas des¬
cribed as “scrub” or “fern” most of which had been
cleared by 1935. The impact of this clearing on river flow
is unlikely to have been great; the fans occupy only a
small and relatively dry part of the catchment and peak

CHANNEL INCISION GIPPSLAND

19

flows in the streams, both of which are prone to flash
flooding, coincide with periods of heavy rain in the up¬
per catchment associated with prolonged spells of
easterly weather and the development of cut off low
pressure cells. While it is clear that flash floods may
form the main erosive force, the climatic events which
lead to flash flooding are not a post-settlement
phenomenon, and the fact that the floods have become
such an effective agent of erosion must be due to other
change.

Destruction of bank vegetation is one con¬
tributory factor. In the past many local farmers con-
siderd scrub growth along creeks was undesirable
because it limited access by stock and harboured vermin,
but today its role in bank stabilisation is acknowledged.
Trees are only removed where they have been under¬
mined and are liable to fall into the river, because when
this happens they are expensive to remove and there is a
risk that they will promote channel widening by diver¬
ting the water flow against the bank. In many places to¬
day the sides of Eaglehawk and Stoney Creeks are
unvegetated because they are too steep for plants to
become established, but the absence of any severe flood
in the last three years has enabled a number of riparian
species to colonise higher sections of the channel floor.
While they persist these will assist bed stabilisation, but
if they are damaged either by flooding or bulldozing
there is a risk that increased sediment mobilisation will
follow.

The excavation of a continuous channel through
the swamp which once formed a local base level for the
upper part of Eaglehawk Creek has also caused erosion.
Water now flows to a new base level at the Latrobe
River. Figure 7 shows evolution of the new long profile
in diagrammatic form. This erosion began in the upper
part of the natural channel across the terrace as the
relatively steep gradients of the head of a water course
became adjusted to the lower gradients of a middle
reach. The deepening was then transmitted upstream
through headward migration of knickpoints. The chan¬
nel is deepest where it passes through higher parts of the
undulating surface of the fan. This adjustment of long
profile is complicated by the redeposition of eroded
material as regional gradients diminish from fan to ter¬
race. The redeposition creates a new and steeper gra¬
dient at the lower end of the sand splays. As deepening
upstream diminishes and the rate of arrival of new sedi¬
ment declines, the channel starts to erode back through
this oversteepened sector creating a new, lower, sand
splay as this material in turn is redeposited downstream.
Erosion of the channel back up into the fan is com¬
parable to the sequence described by Schumm and
Hadley in Wyoming and New Mexico (1957): trenching
appears to be initialed at the toe of fan once surface gra¬
dients exceed a threshold value, which in the case of
Eaglehawk Creek lies at about 0.012. Below the lowest
splay the water emerges minus its coarser sediment load,
and possessed of superfluous energy. Several authors,

Fig. 6 —Wood and concrete drop structure across Eaglehawk Creek outflanked by bank erosion. Looking upstream.

20

J. F. BIRD

including Komura and Simmins (1967) and Gregory and
Park (1974) have shown that this leads to local channel
degradation, and this is evident in Eaglehawk Creek.
Basalt boulders have been dumped at the head of the
channel below the splays in an attempt to prevent the
development of deep water holes in the channel floor,
eroded by clear water emerging from the depositional
zone.

Stoney Creek underwent changes in long profile
similar to those of Eaglehawk Creek. It, too, once
flowed into a swamp which provided a local base level,
but when it was diverted into Eaglehawk Creek in 1933 it
adjusted to the changes within the main stream. A

photograph taken in 1935 shows the confluence at a
stage when they were still at different levels, with the
mouth of Stoney Creek about a metre above the channel
of Eaglehawk. This disjunction was the origin of a
knickpoint which was rapidly transmitted upstream.
Stoney Creek began to deepen later than Eaglehawk
Creek, and the cross profile (Fig. 4) together with the
comparative lack of channel floor vegetation suggests
that the transition from a dominance of deepening to
one of widening occurred more recently. Much of the re¬
cent sediment deposited in the splays downstream pro¬
bably originated form Stoney Creek rather than
Eaglehawk Creek.

Forested hills with patches of alluvium valley Pre-1900

Piedmont fan zone -- Terraces

A

c. 1900

Fig. 7 —A, Under natural conditions Eaglehawk Creek had two sectors with a clearly defined channel separated by a swamp which
served as base level for the upper channel. B, About 1895 a drain was cut through the swamp, linking the upper and lower channels.
C, The steepened reach of the man-made channel through the swamp began to incise back into the upper part of the fan zone. At the
same time redeposition of eroded materials built up a new prograding fan across the terrace below the swamp. D, As the incised
channel extended back into the hills the rate of erosion slowed, and supply of sediment to the downstream depositional area
diminished. The channel at the lower end of the post-settlement fan then began to cut back, creating a secondary sediment splay at
the downstream end. Water minus its sand and gravel load emerging from the toe of this secondary splay caused degradation of the

channel immediately downstream.

CHANNEL INCISION GIPPSLAND

21

Changes in long profile have not been the only
cause of channel instability in Eaglehawk Creek.
Although, as stated earlier, (here is no evidence for
changes in the stream hydrograph due to rainfall trends
or catchment modification, interference within the fan
zone itself has accentuated variations in run ofY. One
factor in this has been the elimination of the local
swamps which formerly served as natural flood retarding
basins. Even more important is the retention of run off
within the one main channel. Earlier in this paper the
natural pattern of run off was described: as is common
in fan terrain, watercourses were subject to frequent
lateral shifts, so that when flood waters overtopped one
channel they created a new one off to the side. When
measures were taken to prevent this all the flood flows
were retained within the one alignment. The dam built in
1894 to block the overflow into Four Mile Creek ensured
that an increased proportion of the floodwater was re¬
tained within Eaglehawk Creek; a corollary of this is
that flood peaks in Four Mile Creek have diminished
and this is one of the few streams in the area today
which shows no signs of incision, but rather of having
an over-fit channel. A further major increment to the
flow of Eaglehawk Creek occurred with the diversion in¬
to it of Stoney Creek, effectively increasing the catch¬
ment of the middle reaches from 39 km 2 to 53km 2 . This
diversion did not affect lower Eaglehawk Creek,
because, as the map (Fig. 1) shows, it already carried the
Stoney Creek flow under natural conditions.

Eaglehawk Creek between the railway bridge and
the Stoney Creek confluence is therefore receiving the
runoff from an additional 14 km 2 of catchment, as well
as containing the flood peaks which formerly passed out
of the channel. There is no evidence from which to
estimate the magnitude of the flood formerly required to
overtop the banks and flow into alternative channels,
but it is known to have occurred several times each year.
This implies that even moderate floods with a return
period of less than a year used to be dispersed, but were
thereafter retained within the main creek, and as Dury
(1977) has suggested, these are particularly important in
determining channel capacity.

CONCLUSION

In suggesting that river channels in Victoria were
stable over the few hundred years prior to European set¬
tlement it is necessary to clarify the definition of stabil¬
ity. Course changes took place, and fan construction
was actively occurring in many areas. A corollary of this
is that the rivers were carrying sediment, so that catch¬
ment erosion must have been quite widespread. The
post-settlement onset of vertical instability has often
been initiated through attempts to control pre¬
settlement lateral instability. Channelling and swamp
drainage have created stream-bed gradients considerably
steeper than those occurring naturally and the response
of many rivers has been to incise their former channel
floors.

The rate of incision diminishes as streams again
approach a stable grade, or when they encounter
bedrock. Minor control structures are rarely effective on

channels prone to flash flooding, particularly if they are
located in poorly consolidated sediments, and stabilisa¬
tion can only be achieved with massive dams, the cost of
construction of which usually far outweighs the benefits
obtained. Many problems result from the less visible end
of the process, the redeposition of eroded material,
which masks farmland, devalues creek water supplies,
obliterates fence lines, and threatens road and rail links.
The insertion of small sediment traps along the channel
may help by retaining some of the sediment within the
creek bed at the cost of transforming surface into sub¬
surface flow.

Eaglehawk and Stoney Creeks have passed
through the phase of maximum instability as they ad¬
justed to human modification, but their present near¬
stable condition is a precarious one. Another severe
flood is likely to wash out the two remaining intact drop
structures which together hold a head of just over a
metre, and initiate a knickpoint which, as it migrates
headwards, will threaten the causeway crossing at
Frasers Road. A flood of this magnitude could be caused
by a recurrence of the meteorological conditions ex¬
perienced in 1952, or by a less rare climatic event exacer¬
bated by human interference in the catchment such as
renewed logging or a re-opening and enlargement of the
old quarry just above Frasers Road. The answer to the
difficulties posed by eroding channels appears to be to
adapt to changes that have occurred rather than attempt
to reverse them, but also to guard against any en¬
vironmental modifications which might result in the in¬
itiation of a further phase of marked instability.

ACKNOWLEDGEMENTS

The author would like to thank the following
Victorian organisations for permission to consult rele¬
vant files: State Rivers and Water Supply Commission;
Latrobe River Improvement Trust; Victorian Railways;
Education Department; Department of Crown Lands;
Public Record Office. Jennie Wragge and Neville Green
assisted with the surveying. Tony Boyd printed the
photographs and Rob Bartlet prepared the diagrams.
Many local landowners also assisted with historical in¬
formation.

REFERENCES

Abbott, W. E., 1884. Water supply in the interior of New
South Wales J. Proc. Roy. Soc. N.S. W. 18: 85-111.
Bird, J. F., 1980. Some gcomorphological implications of
flood control measures on the Lang Lang River, Vic¬
toria. Aust . Geogr. Stud. 18: 169-183.

Bourman, R., 1976. Environmental geomorphology: examples
from the area south of Adelaide Roy. Geogr. Soc.
Austr., Proc. S. Aust. Branch 75: 1-23.

Bride, T. F., 1898. Letters from Victorian Pioneers.
Heinemann, Melbourne.

Bull, W. B., 1964. History and causes of channel trenching in
western Fresno County, California. Am. J. Sci. 262:
249-258.

Cooke, R. U. & Reeve, R. W., 1976. Arroyos and environ¬
mental change in the American Southwest. Clarendon
Press, Oxford.

22

J. F. BIRD

Daniels, R. B., 1960. Entrenchment of the willow drainage
ditch, Harrison County, Iowa. Am. J. Sci. 258:
161-176.

Dury, G., 1977. Peak flows, low flows and aspects of geomor-
phic dominance. In River channel changes, K. J.
Gregory, ed., Wiley, Chichester, 61-74.

Goede, A., 1972. Discontinuous gullying of the Tea-tree
Rivulet, Buckland, Eastern Tasmania. Pap. Proc. R.
Soc. Tasm. 106: 5-15.

Gregory, K. J. & Park, C. C., 1974. Adjustment of river
channel capacity downstream from a reservoir. Water
Resources Res. 10: 870-873.

Gregory, K. J. & Park, C. C., 1976. The development of a
Devon gully and Man. Geography 61: 77-82.

Imeson, A. C. & Jungerius, P. D., 1977. The widening of
valley incisions by soil fall in a forested Keuper area,
Luxembourg. Earth Surf. Proc. 2: 141-152.

Komura, S. & Simmons, D. B., 1967. River bed degradation
below dams. J. Hyclrol. Div. Am. Soc. Civil Eng.
93-HY4: 1-14.

Malde, H. E. & Scott, A. G., 1977. Observations of contem¬
porary arroyo cutting near Santa Fe, New Mexico,
U.S.A. Earth Surf. Proc. 2: 39-54.

Pickup, G., 1975. Downstream variations in morphology, flow
conditions and sediment transport in an eroding chan¬
nel. Z. Geomorph. N.E. 19: 443-459.

Pickup, G., 1976. Geomorphic effects of changes in river run¬
off, Cumberland Basin, New South Wales. Aust.
Geogr. 13: 188-193.

Schumm, S. A. & Hadley, R. F., 1957. Arroyos and the semi-
arid cycle of erosion. Am. J. Sci. 255: 161-174.

Schumm, S. A., 1977. The fluvial system. Wiley, New York.

Woodyer, K. D., 1968. Bankfull frequency in rivers. J.
Hydro/. 6: 114-142.

PROC. R. SOC. VICT. vol. 94, no. 1, 23-34, March 1982

THE PEAK OF THE FLANDRIAN TRANSGRESSION IN
VICTORIA, S.E. AUSTRALIA-FAUNAS AND SEA LEVEL

CHANGES

By E. D. Gill* and J. G. Lang,* with an appendix by S. E. Boyd|

* CSIRO Division of Applied Geomechanics, Institute of Earth Resources, P.O. Box 54, Mount

Waverley, Victoria 3149

t National Museum of Victoria, 285-321 Russell Street, Melbourne, Victoria 3000

Abstract: The peak of the Flandnian Transgression (ca. 6000 years ago in Australia) has been
studied at two sites on the Victorian coast 440 km apart in a direct line. The Warrnambool site in the west
is on a stable high where marine Upper Miocene strata are still horizontal, while the Seaspray site in the
east is in an area of known earth movements over the same period of time. The sediments at Warrnam¬
bool are characterized by calcarcnite, and at Seaspray by quartz arenite.

When the sea reached its peak, Warrnambool Bay extended over the Lake Pertobe flats where
open bay fossils establish the facies. Two spits of different coloured sands (separate sand systems) enclos¬
ed the western end of Warrnambool Bay, and so established a lagoonal system where shell beds were
deposited at the peak of the transgression. At Seaspray a similar lagoonal shell bed was emplaced behind
a coastal dune.

At Warrnambool the present emergence of the top of the shell bed is 1,76 m. Taking into account
compaction and other factors, an emergence of 2 m at the peak of the Flandrian Transgression is
estimated. A similar figure was obtained at Seaspray, indicating that for the past 6000 years measurable
earth movement has not occurred.

Following a line of latitude the coast of Victoria
covers a little over 900 km, but following the coast it is
about twice the distance. The peak of the Flandrian
Transgression in Australia was about 6000 years B.P.
(Gill & Hopley 1971, Thom & Chappell 1975). Around
the coast of Victoria the commonest evidence is shell
beds of quiet water facies behind coastal dunes and in
inlets of various kinds. These shell beds are one of the
best categories of evidence for sea level change because
(1) they are low energy deposits, (2) the shells were laid
below water level, and (3) except near the entrances,
lagoons have a water level at or near mean sea level ex¬
cept in flood time. As such shell beds were laid down
below water level, and upon draining the fine-grained
sediments compacted appreciably (Gill & Lang 1977),
the level of the top of the shell bed was usually below
mean sea level at the time of deposition.

These shell beds have been studied along the
whole coast of Victoria and beyond, but two sites have
been chosen for special treatment. At both sites un¬
disturbed cores of the shell beds have been taken, the
faunas have been studied, and precise levelling has been
carried out. The recently established Australian Height
Datum (A.H.D.) (Roelse et al. 1971) makes it possible to
relate accurately the surveys of the two sites, although
440 km apart in a straight line. One site is at Warrnam¬
bool in western Victoria and the other at Seaspray in
eastern Victoria (Fig. 1).

WARRNAMBOOL SITE

The remarkable sequence of onlapping
calcarenite formations at Warrnambool (Gill 1977 and
Fig. 2) in an embayment eroded by a Pliocene river may
be a suitable world Quaternary standard section. They

onlap a basalt dated at 1.95 million years old. For sec¬
tion with dates see Reeckmann and Gill (1981). The for¬
mations of the past 400 000 years have been shown to be
couplets of shallow marine to beach deposits overlain by
aeolian sediments. Because of the rapid cementing of
these highly calcareous formations, they retain their
original morphology to a high degree, so that their facies
can be demonstrated. Their superposition, plus relative
dating with C 13 and O' 8 isotopes, and chronometric
dating with C 14 and U/Th provide a time framework.
Each of these couplets marks the peak of an interglacial
transgression, of which the Flandrian Transgression is
the latest.

The bedrock is a shelf calcisiltite, the Port Camp¬
bell Limestone, which is still horizontal for about 100
km. Also no seaward warping is discernible because the
declivity from the Miocene shore near Hamilton to the
present coast is of the order of 1.5 m/km. This is a
remarkable stability even for mid-plate Australia. The
Warrnambool sequence overlies the Warrnambool
High, a platform of ancient rocks (Kenley 1976), and so
is an ideal place to study the peak of the Flandrian
Transgression and consider geoidal changes. By contrast
the Seaspray site is in an area of continuing tectonic
movement.

Other evidence of a higher level for the peak of
the Flandrian Transgression are relict cliffs, emerged
shore platforms, channels in Holocene platforms ex¬
tending up into the coastal scrub, overgrown
honeycomb weathering in the lichen and dicotyledon
zones, dune building seaward of Holocene cliffs, and
penetration by the sea of river tracts not now intruded
by it.

23

24

E. D. GILL AND J. G. LANG

Fig. 1 — Map showing localities of Warrnambool and Seaspray, Victoria, Australia.

Lake Pertobe

Three years before the settlement of Warrnam¬
bool began, William Pickering reported on the site to
Governor La Trobe on 7 December 1844. He described
the ample supplies of fresh water, and referred to the
“lake which is at present salt, but by throwing a dam
across its mouth ... it could be kept fresh all the year.’'
This is apparently the earliest reference concerning Lake
Pertobe, which lies on flats 1-3 m above the sea at the
western end of Warrnambool Bay. Following variations
in rainfall, it ranges from a body of fresh water to (rare¬
ly) a dry lake bed. Under natural conditions it emptied
via Pertobe Creek into the Merri River, which
debouches at the S.W. corner of Warrnambool Bay
(also called Lady Bay). In the Latrobe Library,
Melbourne, an unfinished water colour (or draft) made
in 1879 is preserved, and Fig. 3 is a pen and ink copy. It
shows Lake Pertobe, Pertobe Creek and the Merri River
in their natural condition (cf. Fig. 2). The spring tidal
range in Warrnambool Bay is 0.9 m.

The Lake Pertobe sediments are sandier to the
east, nearer the sand spits that formed the lagoon. The
sea entered through a gap between the two termini of the
spits 6000 years ago (Fig. 7), and since the sea departed
from the lagoon, waves have broken through from time
to time. A photograph by Thomas Washbourne taken
between 1872 and 1879 (date limits determined by
Thomas Wicking from structures in the photo) shows a
wide gap with a sandy floor. Such intrusions could not
have been frequent or sustained for long, otherwise
there would be more sand in Lake Pertobe.

Auger holes have been sunk at a number of
places in this area (c.g. see Gill 1953, figs 13, 14), and in
1979 CSIRO drilled a number of bores in the Lake Per¬
tobe flats and the present estuary of the Merri River.
Years ago bores were sunk in the Lake Pertobe flats to
explore the possibility of building a harbour there, but
none of the organizations concerned have been able to

produce the logs. In recent years the City of Warrnam¬
bool has built an Adventure Park in the area. This in¬
volved extensive testing and excavations which have
shown the wide extent of the shell bed and the various
facies.

Caravan Park Section

Sewerage trenches dug at the Ocean Beach
Caravan Park in Pertobe Road provided an opportunity
to study a section in detail and relate it to A.H.D. (Fig.
5). For the general stratigraphy of the area see Gill
(1976). The shell bed and its equivalents elsewhere were
given the stratigraphic name Pertobe Coquina, follow¬
ing Pettijohn, but now Schreiber (1978) defines a co-
quinoid as “an autochthonous (in situ) deposit of shelly
material, usually having a fine-grained matrix, which
may build up, under certain conditions, to biostromes.”
This description fits the Lake Pertobe deposit very well,
so it is re-named Pertobe Coquinoid.

Figure 5 shows the stratigraphic section with a
disconformity between the lagoonal coquinoid and the
freshwater peat. During the interval, the shell bed was
drained (or partly so), and compacted, with some
development of secondary carbonate, i.e. an incipient
soil formed. The time gap is given approximately by the
radiocarbon dates (6570 years for the coquinoid, and
1875 years for the middle and lowest third of the peat).
The shells for the C' 4 assay were from 0.27 to 0.37 m
above A.H.D. As the area was tectonically stable for the
period concerned, the fall in water table to allow the in¬
cipient soil to develop infers a fall in sea level. Evidence
for this fall is widespread.

Pertobe Coquinoid

Below the Lake Pertobe flats, including the
estuary of the Merri River, there is a bed of mostly mud¬
dy lagoonal sediments rich in shells, 1-5 m thick, dated
7040 to 5850 years B.P. by radiocarbon (Gill 1971a,

Fig. 2 — Locality plan of Warrnambool and Dennington. At Excavation 2 (E.2) is a fossil shore platform with estuarine shells dated
3750 years B.P. The Lake Pertobe shell bed is 6000 years B.P. For description of Dennington section line see Gill 1967a.

FLANDRIAN TRANSGRESSION

25

26

E. D. GILL AND J. G. LANG

J o

J3 X>
O -o

_j a

r3 O
? ;=
•o ^
o c
J= c
,S2 r ~»

tH "«

§ 5
o §
53 o

as

c

23

1973). The base of the shell bed is characterized by open
bay shells such as Katelysia, Ostrea and Mytilus,
whereas the major part of the bed is characterized by the
lagoonal genera such as Homcilina, Notospisula ,
Velacumantus, Bembicium , and Salinator , succeeded by
fine sandy, muddy and peaty brackish to freshwater
sediments as indicated by Hydrococcus tasmanicus
(tidal salt marsh), Coxiella australis (brackish lake) and
Potomopyrgus niger (freshwater, usually just above
tidal influence). The usual ecological range of these
species is given in the Appendix.

Below the shell bed is a layer of fine sand.
Although most of the sands in this area are calcarenites,
this deposit is entirely siliceous and very well sorted. Its
origin is probably in the dry period of about 21 000 to
8500 years ago when a terra rossa with calcrete subsoil
formed over the calcarenites (Gill 1975). At present, car¬
bonates are not held in the subsoil, but in that dry
period a calcrete 0.5-1 m thick accumulated. At the peak
of the dry period the A horizon of the terra rossa was
winnowed away except in the valley floors. Leaching
during pedogenesis removed most of the carbonate from
the A horizon, leaving essentially quartz and clay. As
the bedrock is a consolidated dune sediment, this quartz
sand is fine to medium and very well sorted, and the win¬
nowed sand was washed into Warrnambool Bay area
while sea level was lower. The sand is grey, except for
that on a fossil shore platform under the Pertobe Co-
quinoid at the foot of Cannon Hill (Gill 1953, fig. 13),
where it is red. This red sand was probably washed from
Cannon Hill and collected on the platform following a
small drop in sea level, and/or because the growth of the
spits that created the lagoon lowered the tidal range,
although the distance from the entrance of the lagoon is
so short that the latter is unlikely.

Tower Hill Tuff

Under the Pertobe Coquinoid at the foot of Can¬
non Hill is a deposit of Tower Hill Tuff (Gill 1950,
1967a, 1972, 1976, 1978, 1979). Pieces of stratified tuff
were noted in the spoil from the dragline excavation of a
channel for the Johnson Adventure Park below Cannon
Hill. Three characteristics of the tufT are significant:

1. It is stratified with different grain sizes in the various
layers, as seen in the airlaid tuff in many places on the
Warrnambool terrain.

2. 11 has the open texture of airlaid tuff, and not the
compact texture of waterlaid tuff, as revealed by
sew-erage trenches in the housing area north of the
Princes Highway and east of McGregors Road, East
Warrnambool, where fossil water weeds, ripple marks
and trails were found.

3. The top of a large piece of tuff brought up by the
dragline was oxidized to a yellow colour, suggesting
short-term pedological activity. The difference between
the C 14 date for the tuffand that for the oyster bed is 720
years.

In a bore at the foot of Cannon Hill on the Per¬
tobe flats, where the surface is 1.88 m above A.H.D.
(virtually mean sea level), the shell bed occupied the top

FLANDRIAN TRANSGRESSION

27

Fig. 4 —Oblique aerial photograph showing the gap through which, under natural conditions, the sea occasionally broke through.
On the right of the gap is the spit of light grey sand, and on the left is the spit of light brown sand (Fig. 7). Photo by courtesy “The

Standard”, Warrnambool.

2.26 m, including some fill which formed the path where
the boring was done. AC 14 shell sample from 0.3 m
below A.H.D. was dated 6250± 160 years B.P.
Stratified tuff was cored from 2.26-2.87 m, the tuff being
mottled red (2.5 YR 8/4) at the base. Below the tuff was
red (10 YR 4/6) calcarenite grading to pink (7.5 YR 8/4)
and then into normal very pale brown (10 YR 8/3) at
3.77 m, typical of a terra rossa soil on calcarenite. If the
sea had been present at that time, the soil would have
been washed away, and the tuff also, but 7300 years B.P.
sea level was lower and the tuIT fell on a land surface.
Secondary tuff was described over the Pertobe Co-
quinoid and piled against the Cannon Hill cliff (Gill
1950); it is better to refer to this as lulfaceous colluvium.

On the top of the cliff immediately east of the
Thunder Point trigonometrical station is a hollow con¬
taining well stratified tuff. East of this is higher ground
with traces of tuff on a calcrete surface. Further east is a
slope in the lee of the prevailing winds where a layer of
unstratified tuff lightly cemented by calcarenite overlies
finely laminated mammillary calcite as described from
Dennington, the western suburb of Warrnambool and
dated 8700± 150 years B.P. (GaK-3920) (Gill 1975). At
various levels in this unstratified tuff are edible marine

shells of moderate size accompanied by charcoal, which
are therefore Aboriginal middens. The shells are from
two contrasting facies. Subninel/a undulata came from
the open ocean rock platforms to the south, while
MytHus eclu/is planulatus , a bay facies shell, must have
come from Warrnambool Bay to the north. Sea level
was appreciably lower when the Tower Hill Tuff' was
emplaced, as is shown by the bore at the foot of Cannon
Hill cited above, by airlaid stratified tuff at Tower Hill
beach extending down below L.W.L., and by similar
tuff below L.W.L. under the Kelly Swamp deposits (Fig.
2 ).

The dune ridge at Thunder Point, on which the
tuff' deposits lie, is Last Interglacial in age, left stranded
as the sea retreated. As the dune bedding extends to
about 3 m below present sea level, its age is estimated to
be about 95 000 years. Following the return of the sea to
its present level, this dune has been cut back just past its
centre-line, so that all the dips are landward. The island
platforms of this rock left in the sea, and the dips of the
dune rock make it possible to reconstruct the dune.
Thus, 7300 years ago, the south edge of the dune was
about 0.25 km further seaward and, instead of the pre¬
sent islands off Point Pickering, the dune ridge extended

28

E. D. GILL AND J. G. LANG

FORMATION &

COLOUR &

LITHOLOGY

A.H.D.

AGE IN Cl4

COMPACTION

THICKNESS

& FACIES

LEVEL

YEARS B.P.

0.28 m

FILL

Surface

Modern

+■ 1.85 m

Moyne

0.26 m

PEAT

V 1 57 m

Alluvium

Black

Freshwater

1875 yr

4.8-8.4 kg/cm 2

swamp

1.46 m

At top muddy

V

Light gray

lagoon with

6570 yr on

Pertobe

2.5 yr 7/2

Homalina

shell sample

with irregular

from + 0.27

Coquinoid

mottles of

to + 0.37 m

strong brown

7.5 yr 5/G

7.2-10 kg/cm z

Juvenile soil

SHELL

some leaching

formed on

BED

and deposition

chemically

-

t- H . W. L .

of secondary

reduced

springs

carbonate;

lagoonal

in open

elongate con-

deposit before

sea

cretions up to

peat formed

At base open

8 cm in long

bay facies with

. <4

- A.H.D.

diameter.

Ostrea and

Bottom of

T

Other dates

excavation

- 0.15 m

in same

2 m from

shell bed

surface. A «

f- L.W.L.

6500 yr in

nearby excava-

springs

Merri Canal

At the base of

tion penetrated

in open

near Woollen

Cannon Hill

to quartz silt.

sea

Mill. Shells

the Pertobe

about HWM

Coquinoid over-

lies subaerially

5850 yr.

deposited, fine

Shells over

bedded, sur-

fossil shore

ficially oxidized

platform at

TOWER HILL TUFF

foot of

(7300 yr)

Cannon Hill

•Penetrometer Readings

Fig. 5 —Geological section in Lake Pertobe area at the Ocean Beach Caravan Park, Warrnambool, Western Victoria.

over 1 km east of the present tip of Point Pickering
(Figs 6, 7). The middens show that the Aborigines chose
a high point for their eating place, yet protected from
the cool prevailing S.W, winds. They gathered food
from the open coast to the south and from the bay to the
north. As the water deepened in the bay, beds of large
oysters flourished there, but as yet their shells have not
been found in the middens.

Palaeogeography

When the sea first returned from the Last Glacial
low level, Warrnambool Bay was more extensive as it
occupied also what are now the Lake Pertobe flats, and
a shore platform was formed on the north side (Gill
1953, figs 13, 14). It now remains to explain how the
Pertobe flats came into existence. Two processes are
possible: 1, the prograding of the shore, whereby the
land is progressively built seawards; 2, the growth of a
sand barrier or spits which cut oft’ the area, after which it
gradually infilled with Stillwater sediments. A small fall
in sea level aids both processes (Gill 1981a), and there is
evidence that this occurred. As the muddy Pertobe Co-
quinoid covers the whole of the area, and is bounded on
the seaward side by two opposing spits of different col¬
oured sands, the second process is the one that occurred.
A spit of light grey calcarenite grew north from the
Pickering promontory (the Merri River was not there

then) to near where the Surf Life Saving Club is now
(west shore), while from the Last Interglacial calcarenite
on the west side of the Hopkins River mouth a spit of
light brown calcarenite grew west to near the S.L.S.C.
site (north shore). Behind these spits was a lagoon in
which the Pertobe Coquinoid was deposited; the sea
entered through the gap between them (Fig. 7). When
the sea retreated the gap filled, but has always been a
weak point. Under natural conditions the sea used to
break through there and cross Pertobe Road into the
lake. Captain Barrow’s original survey of the harbour
(1854) shows a submarine sand spit in this area. The
aerial photograph (Fig. 4) shows that in spite of
foreshore works the gap between the two spit ends can
still be readily recognized. From a high point it is easy to
see the fine light grey sand of the west shore as distinct
from the coarser light brown of the north shore (Fig. 7).
The presence of two sand systems has not been prev¬
iously recognized. At one time it was thought that the
source of the sand filling the harbour was to the east, so
a seawall was built opposite the Pertobe Cutting in 1919.
This was not successful because it was at the end of the
light brown sand system; the movement of brown sand
finished there and did not continue on into the harbour.

FLANDRIAN TRANSGRESSION

29

Fig. 6-Superimposed on map of Lake Pertobe area is the

shore line of about 6000 years B.P. (dashed line).
Chronology

Seven radiocarbon dates (uncorrected for
seawater age —about 450 years) provide time control for
the geological history of the Lake Pertobe flats:

5850±320 (PIC-9) Homalina , foot of Cannon Hill,
over fossil shore platform.

6250 ± 160 (GX-6788) Ostrea, Mytilus and Notospisula,
bore, foot of Cannon Hill.

6460±110 (SUA-985) Ostrea, N. end Lake Pertobe
(date recalculated).

6500 ±200 (PIC-10) Homalina, Merri Canal near Warr-
nambool Woollen Mill.

6570 ± 200 (SUA-780) Fauna with Mytilus, Ocean Beach
Caravan Park excavation.

7040 ± 205 (GX-6789) Peaty layer (weed bed) with pieces
of large shells (including Katelysia, open bay genus) plus
whole weed shells (cerithiids) under Merri River estuary.
7300 ± 150 (GaK-2856) Mytilus and Subninel/a midden
in Tower Hill TufL with calcarenite, E. of Thunder
Point.

During the rise of the sea to the peak of the Flan-
drian Transgression, the two spits that border the Lake
Pertobe flats were initiated. As they grew, Stillwater
areas were created behind them with consequent
changes in flora and fauna. The peaty layer dated 7040
years B.P. is the first evidence of this in the excavations.
As the spits extended, so the area of lagoon behind them
increased. Practically no waves reached this zone, and
the tidal currents were weak because of a spring range of
only 0.9 m. Shells were moved, but not far, as most are
complete and many paired. The shells are so packed in
some places that gentle winnowing to remove the mud is
suspected. But the large heavy oyster shells provide a
biocoenosis, as there was certainly not enough energy to
move them. When the buildings at the corner of Pertobe
Road and Price Street were constructed, an extensive
oyster bed was encountered (“Warrnambool Standard”
17 April 1944). Another oyster bed was recently revealed
in excavations at the north end of Lake Pertobe near the
railway gates (C 14 date 6460 years).

Thus in various parts of the Lake Pertobe flatt¬
en

Fig. 7 —The palaeogeography of the Lake Pertobe area, Warr¬
nambool, showing the more extensive Warrnambool Bay at the
peak of the Flandrian Transgression about 6000 years B.P.,
and the two spits that created the lagoon (the area now oc¬
cupied by the Pertobe flats).

the Katelysia-Ostrea-Mytilus open bay facies is dated
7040-6250 years, while the Homa/ina-Notospisula-
Velacumantus lagoonal facies is dated 6500-5850 years.
This time overlap is to be expected because of the
gradual growth of the spits, but the dates do suggest that
open bay conditions did not last very long. The spits
must soon have extended far enough to establish a
Stillwater regime with mud deposition (Fig. 7).

The dates so far obtained for the top and bottom
of the shell bed give a formation time for the thickest
part of 1190 years, but for the greater part of it only 730
years.

Role Of The Merri River

The Merri River flows across the south end of the
Pertobe flats, and debouches into Warrnambool Bay.
While the Pertobe Coquinoid was being deposited the
river did not enter this area, but debouched near Den-
nington 4 km N.W. of the present mouth. Three bores
sunk through the present bed of the river (which is now
diverted through the Merri Canal) on the east side of the
road to Thunder Point penetrated only Pertobe Co¬
quinoid. There is no proper channel and practically no
riverine sediments; the river just floods across the
lagoonal shell bed.

During the Last Glacial period the Merri River
flowed through the water gap at Dennington and direc¬
tly south to the sea. The contours on the continental
shelf define the valley through which it flowed (Gill
1967a, fig. 15.1). When the sea returned to its present
level, the Dennington Spit (Gill 1981b) began to grow to
the N.W., forcing the river to flow around it. About
4000 years ago the sea was still reaching the back of what
is now Kelly Swamp (Fig. 2), because sea shells overlying
an aeolianite shore platform south of Dennington (Gill
1967a, fig. 15.7) were dated at 3980± 150 years B.P.
(GX-58). That the genus Homalina was common in¬
dicates that the Dennington Spit had grown far enough
to create a sheltered area at the mouth of the river. The
Merri River continued to arch round the growing spit

30

E. D. GILL AND J. G. LANG

until this course became too difficult, and it diverted
S.E. along an old interdune swale to Warrnambool Bay.
The dating of sediments along this course could deter¬
mine approximately when the diversion occurred.

Interpretation of Sea Level Stand

Although the top of the shell bed is now
emerged, it was below low water when deposited
because the shells are mostly low water to subtidal
species. The top of the shell bed at the Caravan Park is
1.31 m above A.H.D., which is 1.78 m above low water
level. The emergence is thus 1.76 m plus (1) the depth of
water above low tide, (2) the compaction due to draining
on emergence (probably not very much here), (3) any
reduction of the top of the shell bed by erosion as the sea
retreated, (4) the lowering of the top of the bed by
subaerial weathering following exposure (as shown by
secondary carbonate deposition) and (5) leaching by
acids following deposition of the peat. To allow 0.24 m
for these five factors to round off the emergence at 2 m is
a conservative figure. The figure most difficult to quan¬
tify is the depth of water over the shell bed. As the

distance from the sea into the Lake Pertobe lagoon was
so very short, no difference in tidal range from that in
Warrnambool Bay is to be expected.

The geology cannot be explained as a function of
progradation because of (1) the emergence of the shell
bed, and because (2) the western part of the post-
Flandrian embayment was cut off by a bay bar, then the
enclosed area infilled with muddy sediments, i.e. there
was not a gradual migration seaward of the beach ridge
due to progradation.

SEASPRAY SITE

The Gippsland Lakes form an extensive body of
water behind a siliceous sandy beach and barrier about 7
m high (Bird 1978, Jenkin 1968, Ward 1977). It is a
quartz sand system and not a calcarenite system as at
Warrnambool. The beach is of open ocean type with
high energy waves due to the Southern Ocean swell. A
high contrast thus exists between the quiet waters of the
lakes system and the surf of the open beach; the
differences are clear in both the sediments and the biota.
It is thus easy to distinguish which Quaternary beds were

SI6000 mE

Fig. 8 —Map of drain and bore sites between Seaspray and Lake Reeve.

FLANDRIAN TRANSGRESSION

31

Bore 2

Fig. 9 —Longitudinal section along drain from

deposited inside the barrier and which outside (Figs.
8 - 10 ).

At the peak of the Flandrian Transgression about
6000 years B.P. the lakes system was far more extensive
than now, reaching some 30 km further southwest. The
village of Seaspray stands on the lagoonal deposits of
this more extensive system, and on a riverine terrace of
Merriman’s Creek. In 1962 a drain some 3 km long was
excavated from Seaspray to Lake Reeve, which is the
S. W. end of the present lake system. Under natural con¬
ditions the delta of Scott Creek prevented this drainage.
This drain provided an excellent section through the
emerged shell beds, which were studied and dated by
radiocarbon (Gill 1966, 1970, 1971a, b). The CSIRO
Division of Applied Geomechanics drilled a series of
bores (with undisturbed cores at critical places) normal
to the coast across this shell bed in the vicinity of Scott
Creek, where the mid-Holocene shore is 1.7 km inland
from the present shore (Fig. 8). The delta of Scott Creek
provided the only access across the swamp for the drill
rig.

Since this work was done, A.H.D. permanent
bench marks have been established, and these have been
used to determine the height of the shell bed surface in
the area. All the pegs put in during our original survey
of points of geological interest have now gone. For¬
tunately our survey was tied into the engineering survey
made by the Shire of Rosedale for the engineering
works, and it is now possible to relate the local arbitrary

Seaspray to Lake Reeve as shown in Fig. 8.

engineering datum to A.H.D., and hence also the
geological sites (Figs 8, 9).

On the transect the original pegs have also gone,
but it has been possible to relate the bore sites to A.H.D.
through the level of the road. From photographs of the
drilling rig when in position at the site of Bore 1 it has
been possible to fix within 1 m the point where the
transect crossed the centre of the bitumen road. Another
layer of bitumen (without screenings) has been added
since the survey, but considering the wear of the road
and the thickness of the bitumen, the pavement level can
be only 1 or 2 mm different, and this can be ignored. The
survey from a permanent survey mark to the transect
(0.6 km) and back closed within 1mm, and the level of
the centre of the road was determined as 1.773 m
A.H.D.

Figure 10 shows a cross-section along the
transect, and indicates the relationship of the shell bed
to A.H.D. The shell bed is lower along the transect
across the flats than it is in the drain behind the dune
(Fig. 9). Scott Creek crosses this area and the bed has
evidently suffered erosion following fall in sea level.
Note that bore sites 3, 5 and 7 of the eight pegged out
were not drilled. Fig. 9 shows the levels obtained for the
top of the shell bed along the drain parallel to the coast,
and along the transect normal to the coast.

Interpretation of Sea Level Stand

It should be noted that the top of the shell bed

32

E. D. GILL AND J. G. LANG

was surveyed and not the top of the marine bed. At the
Lake Pertobe Caravan Park site there is a sharp break
between the shell bed and the overlying freshwater peat,
but at Seaspray the shell bed is followed by a gradual
transition from lagoonal to terrestrial deposits, as also
occurs over most of the Lake Pertobe flats. To define the
top of the marine bed would require a great deal of
micro-palaeontological work. Mr. A. C. Collins deter¬
mined Ammonia and Trochammina inflat a from this
zone; they arc estuarine foraminifera. So it is important
to observe that the figures for the top of the shell bed are
all minimal figures for the height of the marine deposits.

Lake Reeve has a tidal range of only about 15 cm
and is virtually at mean sea level. As the water in the
drain flows freely down to A.H.D. (M.S.L.) at Lake
Reeve, and the shell bed was visible in the side of the
drain, and the spoil was crowded with shells, the top of
the shell bed there was well above A.H.D. (Fig. 9). This
has now been confirmed and quantified by the detailed
survey. As most of the species of molluscs lived at low
water or below, the shell bed is emerged. As it is a quiet
water regime, waves could not lift a whole shell bed
above the level at which it was deposited. The fauna in¬
dicates that the sea had free access to the area, the spring
tidal range is 2.4 m, and the top of the shell bed in the
drain is 0.58 m A.H.D. The amount of emergence must
therefore be of similar order to that at Warrnambool,
where it is given as 2 m. The shell bed was not a result of

progradation. During the Holocene the barrier has
migrated to and fro a short distance (Gill 1967b), but
otherwise there is a bed over 1 km wide of homogeneous
coquinoid built up in quiet water behind a coastal bar¬
rier.

COMPARISON OF SHELL BEDS 440 km APART
Around the Victorian coast there are numerous
similar unoxidized mid-Holocene shell beds which arc
slightly emerged. Their correlation is clear. However,
the precise surveys of two distant sites in contrasting tec-
tonic environments provided an opportunity for com-
parison. For a long distance through the Port Fairy and
Warrnambool districts of S.W. Victoria the Last In¬
terglacial 7 m shoreline can be shown to be without
displacement, and the mid-Holocene beds are un¬
disturbed. Although Seaspray in Eastern Victoria is in
an area of tectonic movements, no displacement of the
mid-Holocene shell bed could be detected, which sug¬
gests that the coast there has been stable for the past
6000 years.

ACKNOWLEDGEMENTS

We are indebted to Mrs. S. E. Boyd of the Na¬
tional Museum of Victoria for the Appendix on Lake
Pertobe fossils, to Miss Rhyllis Plant of the Museum for
drafting Figures 6 and 7, and to Mr. T. A. Wicking of
Warrnambool for historical information on that area.

Fig. 10 — Cross section from relict coast to present shore, N.E. of Seaspray as shown in Fig. 8.

FLANDRIAN TRANSGRESSION

33

REFERENCES

Barrow, J., 1854. Survey of Lady Bay, Warrnambool. Votes
Proc. Leg. Council 1853-4, vol. 3.

Bird, E. C. F., 1978. The geomorphology of the Gippsland
Lakes region. Ministry Conservation Viet. Publ. 186.
Gill, E. D., 1950. An hypothesis relative to the age of some
Western District volcanoes. Proc. R. Soc. Viet. 60:
189-194.

Gill, E. D., 1953. Geological evidence in Western Victoria
relative lo the antiquity of the Australian Aborigines.
Mem. Nat. Mas. Melb. 18: 25-92.

Gill, E. D., 1966. Australian research in Quaternary
shorelines. Aust. J. Sci . 28: 407-411.

Gill, E. D., 1967a. Evolution of the Warrnambool-Port Fairy
coast and the Tower Hill eruption. Western Victoria.
Landform Studies from Australia and New Zealand.
A.N.U. Press. 340-364.

Gill, E. D., 1967b. Evolution of shoreline barriers. Viet.
Naturalist 84: 282-283.

Gill, E. D., 1970. Current Quaternary shoreline research in
Australasia. Aust. J. Sci. 32: 426-430.

Gill, E. D., 1971a. Application of radiocarbon dating in
Victoria, Australia. Proc. R. Soc. Via. 84: 71-85.
Gill, E. D., 1971b. The far-reaching effects of Quaternary sea
level changes on the flat continent of Australia. Proc.
R. Soc. Viet. 84: 188-205.

Gill, E. D., 1972. Eruption date of Tower Hill volcano,
Western Victoria, Australia. Viet. Naturalist 89:
188-192.

Gill, E. D., 1973. Second list of radiocarbon dates on samples
from Victoria, Australia. Proc. R. Soc. Viet. 86:
133-136.

Gill, E. D., 1975. Calcrete, hardpans and rhizomorphs in
Western Victoria. Pacific Geol. 9: 1-16.

Gill, E. D., 1976. Quaternary: Warrnambool-Port Fairy
District. Geol. Soc. Aust. Spec. Publ. 5: 299-304.
Gill, E. D., 1977. Quaternary shorelines report —Victoria.
Aust. Quat. Newsl. 10: 20-23.

Gill, E. D., 1978. Radiocarbon dating of the volcanoes of
Western Victoria, Australia. Viet. Naturalist 95:
152-158.

Gill, E. D., 1979. 1978-9 research on Quaternary shorelines
in Australia and New Zealand —Victoria. Aust. Quat.
Newsl. 13: 53-57.

Gill, E. D., 1981a. The separation of progradation due to fall
of sea-level from progradation due to sand over¬
supply. Oceanisl: 367-371.

Gill, E. D., 1981b. The contribution of geology to the search
for the ‘Mahogany Ship’. Warrnambool Inst. Adv.
Educ. J. Social Issues 1: 13-15.

Gill, E. D. & Hopley, D., 1971. Holocene sea levels in Eastern
Australia —a discussion. Mar. Geol. 12: 223-242.

Gill, E. D. & Lang, J. G., 1977. Simple measurement of com¬
paction in marine geological formations from engineer¬
ing data commonly available. Mar. Geol. 25: M1-M4.

Jenkin, J. J., 1968. The geomorphology and Upper Cainozoic
geology of Southeast Gippsland, Victoria. Mem. Geol.
Surv. Viet. 27.

Kenley, P. R., 1976. Otway Basin, western part. Geology of
Victoria. Geol. Soc. Aust. Spec. Publ. 5: 147-152.

Reeckmann, S. A. & Gill, E. D., 1981. Rates of vadose
diagenesis in Quaternary dune and shallow marine
calcarenites* Warrnambool, Victoria, Australia.
Sedim. Geol. 30: 157-172.

Roelse, A., Granger, H. W. & Graham, J. W., 1971. The
adjustment of the Australian Levelling Survey
1970-1971. Div. Nat. Mapping Tech. Rep. 12.

Schreiber, B. C., 1978. Encyclopedia of Sedimentology (Ed.
R. W. Fairbridge & J. Bourgeois). Stroudsburg, Pa.

Thom, B. G. & Chappell, J., 1975. Holocene sea levels relative
to Australia. Search 6: 90-93.

Ward, W. T., 1977. Geomorphology and soils of the
Stratford-Bairnsdale area, East Gippsland, Victoria.
CSIRO Soils & Land Use Ser. 57.

C2

34

E. D. GILL AND J. G. LANG

APPENDIX
The Fossil Mollusca
By S. E. Boyd

Austrocochlea constricta (Lamarck 1822)-usually found on intertidal rock platforms, but the species has a wide tolerance of
habitat and salinity.

Bembicium auratum (Quoy & Gaimard 1834) —quieter waters of sheltered muddy inlets and estuaries, in the upper littoral zone*
Velacumantus australis (Quoy & Gaimard 1834)—found in the shallow waters of estuaries, mangrove swamps and mud flats.
Coxiel/a striata (Reeve 1842) —brackish water lakes usually away from direct marine influence.

Potomopyrgus niger (Quoy & Gaimard 1838)-freshwater streams, usually just above the region of tidal influence in coastal
streams.

Hydrococcus tasmanicus (Tenison-Woods 1876) —on mud between plant roots in tidal saltmarsh areas in sheltered bays and inlets*
Parcanassa burchardi (Philippi 1851)-found on intertidal mud flats.

Salinator fragilis (Lamarck 1822) —mid to upper littoral zone in sheltered marine inlets, on mud and scagrass flats.

Mytilus edulis planulatus (Lamarck 1819) —in sheltered waters attached to rocks and jetty piles. Able to take advantage of very
small areas of shelter.

Ostrea angasi Sowerby 1871 - O. sinuata Cotton & Godfrey 1938 (non Lamarck) —usually found in estuaries, but marine beds ar£
also lound. In Port Phillip Bay it occurs in areas of silty sand and silty clay from low water to approximately 11 fathoms*
Notospisula trigonel/a (Lamarck 1818) —a very tolerant species occurring in the sands and sandy muds of the shallows of sheltered
bays and estuaries.

Homalina deltoidalis (Lamarck 1818) —estuaries, in the mud and muddy sand of the middle to lower littoral zone.

Eumarcia fumigata (Sowerby 1853) —found in shallow water in sand and sandy mud.

Venerupis crenata Lamarck 1818 —in holes and crevices of rocks.

Katelysia rhytiphora Lamy 1937 —intertidal and subtidal in sand and sandy mud, in areas of Zostera, etc.

Soletellina biradiata (Wood 1815) —found buried in silty mud in shallow water.

PROC. R. SOC. VICT. vol. 94, no. 1, 35-47, March 1982
ORDOVICIAN AND SILURIAN STRATIGRAPHY AND
STRUCTURE IN THE WOMBAT CREEK — BENAMBRA AREA,

NORTHEAST VICTORIA

By Paul F. Bolger

Geological Survey of Victoria, 107 Russell Street, Melbourne, Victoria 3000
Present address: Herman Research Laboratory, State Electricity Commission, Howard Street,

Richmond, Victoria 3121

Abstract: Undifferentiated Ordovician low grade metasediments are unconformably overlain by
a Silurian succession consisting of the Middle (?) Silurian Mitta Mitta Volcanics and the Late Silurian
Wombat Creek Group. The Ordovician beds have undergone several periods of folding, regional
metamorphism and granitic intrusion in the Benambran Deformation. Silurian rocks arc preserved in a
graben which is considered to have been active during deposition. The Wombat Creek Group consists of
three newly described formations, in ascending order, the Toaks Creek Conglomerate, Gibbo River
Siltstone and Tongaro Sandstone. Facies relationships of these units suggest it is a transgressive sequence
with the tluvial(?) to shallow marine Toaks Creek Conglomerate overlain by the shallow marine Gibbo
River Siltstone and “deeper” marine Tongaro Sandstone. The Silurian sequence in the Graben correlates
with a similar sequence at Limestone Creek (VandenBerg et al. in press) and other Silurian sequences in
southeastern Australia.

A thick Silurian sequence exposed in the Wom¬
bat Crcek-Gibbo River area about 20 km north of
Benambra is preserved in a structure here named the
Wombat Creek Graben. The graben is bounded by the
Wombat Creek and Morass Creek Faults and trends
northwest-southeast, extending from Soldier Creek to
near Taylors Crossing (Fig. 1). There is a small outlier
along Morass Creek near Benambra (Fig. 1). The
Silurian sequence unconformably overlies multiply
deformed Ordovician sandstone, slate and siltstone and
consists of the Middle(?) Silurian Mitta Mitta Volcanics
and the Upper Silurian Wombat Creek Group.

The Ordovician beds were deformed during the
Early(?) Silurian Banambran orogeny of Browne (1947).
To the west they grade into high grade regional
metamorphic and granitic rocks of the Omeo Metamor-
phic Complex. The Silurian sequence in the Wombat
Creek Graben was tightly folded during the Early Devo¬
nian Bindian Deformation (VandenBerg et al. in press)
and was also intruded by acid to intermediate plutonic
rocks. Syenite, trachyte and granite porphyry were
emplaced east of Benambra during the Triassic.
Cainozoic faulting in the area has resulted in rejuvena¬
tion of the mature topography, accompanied by stream
diversion and river capture, and in the late Pliocene, the
Morass Creek Basalt was extruded north of Benambra.

The purpose of this paper is to define and
describe the poorly known Silurian stratigraphy within
the Wombat Creek Graben, and to discuss the deposi-
tional environments of the sediments and the structural
relationships of the Silurian sequence to older rocks.

Map grid references are quoted in the text with
the prefix (GR...) and refer to the Benambra 1:100 000
topographic sheet, No 8424, Scries R652, Division of
National Mapping, Department of National Develop¬
ment.

PREVIOUS WORK

The earliest investigations in the area were made
by Stirling (1887, 1888, 1889) and Ferguson (1899) who
compiled the first stratigraphic column of the Wombat
Creek Group. Dunn (1907a, b) made a brief study of the
distribution of ore minerals in limestones of the Wom¬
bat Creek Group. Whitelaw mapped the isolated
limestone outcrops in detail in 1913, but his work was
not published until 1954. Chapman (1906, 1912, 1917,
1920) identified fossils collected from the Wombat
Creek Group by earlier workers. While examining possi¬
ble sites for the location of a dam on the Mitta Mitta
River, Kenny (1937) examined the contact between the
volcanics and the Wombat Creek Group at the junction
of the Gibbo and Mitta Mitta Rivers, and suggested that
the volcanics unconformably overlie the sediments.

Andrews (1938) described the stratigraphy and
structural relationships of the Silurian and Ordovician
and suggested that an intense deformation event, which
he named the Mitta Mitta Movement, affected the Or¬
dovician prior to the accumulation of the Silurian se¬
quence. This event was later named the Benambran
Orogeny by Browne (1947).

Crohn (1950) gave the name Wombat Creek For¬
mation to the sequence of conglomerate, limestone,
sandstone and shale exposed along the Mitta Mitta
River. Talent (1959a) upgraded this to group status
without defining constituent formations, and suggested
a Middle to Late Silurian age, thus inferring an Early
Silurian age for the Benambran Orogeny. Bcavis (1962)
discussed the structure of the Wombat Creek Group in
his regional study of the Omeo metamorphic Complex,
and Singleton (1965), and Talent (1965, 1969) briefly
described the stratigraphy of the Wombat Creek Group
and its relationship to the Mitta Mitta Volcanics. Talent
et al. (1975) discussed the stratigraphy and correlation
of the Silurian rocks in the Wombat Creek Graben.

35

P. F. BOLGER

localitymap

• Yass

Wombat Creek 1
Indi River — . \

Limestone Creek .

Yalmy

Melbourne Nowa Nowa

Quidong

CREEK FAULTS

Quaternary

Alluvium, Colluvium

Pliocene

Morass Creek Basalt

Taylors

Cross ing

Triassic

Ml Leinster Complex

Tongaro Sandstone

- limestone

- conglomerate

Gibbo River Si/tstone

- limestone

Pridolian

Ludlovian

Toaks Creek Conglomerate

Mitta Mitta Volcanics

Wenlockian

Omeo Metamorphic Complex

Early Silurian

Undifferentiated Ordovician
low grade metasediments

Ordovician

Palaeozoic granitic rocks

. I Pyles *
| \ Limestone’
V N Deposit

The Brothers

_ bedding

_ cleavage
- lineation

trend of small scale
fold axes

n fault

Fossil locality

grapto/ite

shelly

conodont

ABENAMBRA

Lake"
Orneo

Indi River
Limestone

geological boundary

river, creek
*1 type section

Mt Pleasant

Fig. 1—Geological map of the Wombat Creek —Mitta Mitta River area, Benambra, Victoria.

PALAEOZOIC GEOLOGY BENAMBRA

37

STRATIGRAPHY

Ordovician

Distribution: Ordovician sedimentary and low grade
metasedimentary rocks surround the Wombat Creek
Graben and outcrop extensively in eastern Victoria
beyond the study area. The Ordovician rocks have been
regionally metamorphosed to biotite grade, and pass
westwards into higher grade regional metamorphic
rocks of the Omeo Metamorphic Complex.

Lithology: The Ordovician sequence in northeastern
Victoria comprises an undetermined thickness of well
bedded sandstone and mudstone. The beds are steeply
dipping and tightly, often isoelinally folded. Slaty
cleavage is well developed in many of the finer beds. No
attempt has been made to subdivide the Ordovician
lithologically.

Grey to yellow coarse to fine sandstone beds up
to 1 m thick, are usually graded and sometimes contain
internal lamination, convolute lamination and ripple
marks. The upper parts of some massive beds display
gently undulating bedding. Laminated and thinly
bedded grey-green argillites form the upper parts of
graded beds. Thin, medium to fine grained sandstones
are sometimes graded and often display ripple-drift
cross-lamination, planar lamination and ripple marks.
The sandstones generally have planar bases with occa¬
sional load casts. Sole marks are not commonly observ¬
ed but this may be due to lack of exposure of the under¬
sides of beds. There are shaly intraclasts in some sand¬
stones. The sandstones are quartz rich (quartzose
wackes in the classification of Okada, 1971), with quartz
comprising up to 70% of the rock and 95% of the
framework. Quartz grains are mostly equant to
elongate, angular to subangular, monocrystalline grains
with undulose extinction. Weathered feldspars comprise
up to 15% of some rocks, but are generally less than
5%. In thin section, alkali feldspar appears to be subor¬
dinate to plagioclase (andesine (?) to labradorite). There
are occasional biotite plates and muscovite is common.
Equigranular and foliated quartzite occurs as uncom¬
mon sand sized grains. The matrix, comprising up to
40% of some rocks, consists of clays, muscovite,
chlorite, rare biotite and very fine quartz. Platy minerals
and some bladed quartz grains are stongly aligned.
Massive grey to yellow quartzites up to 2 m thick are
usually structureless but occasionally contain internal
lamination. They are coarse to fine sandstones compos¬
ed almost entirely of quartz with rare feldspar,
muscovite and heavy minerals. The quartzites have
usually been recrystallized with framework grains within
a fine grained polycrystalline quartz aggregate. They are
often intersected by quartz veins. Finer beds may be
cherty. The Ordovician sediments exhibit abundant AE,
subordinate AB and ABC, rare BC and inttfrturbidite E
divisions of Bouma's (1962) turbidite cycle. No complete
A —E Bouma sequences have been recorded.

Discrete units of laminated grey to black slate,
more than 20 m thick on the Eustace Gap track, contain
poorly preserved graptolites. These units have not been

mapped in detail, but they may represent marker
horizons within the Ordovician sequence. The slates are
composed of varying proportions of silt-sized angular
quartz and dctrital muscovite, set in a very fine matrix of
clays, muscovite, biotite and chlorite. A strongly
developed slaty cleavage is due to alignment of platy
minerals and some quartz grains. Some very dark grey
beds contain abundant carbon and often contain oxi¬
dized pyrite cubes.

Age: Graptolites have been found at a number of
localities in the Mitta Mitta River area. They arc
generally poorly preserved and strongly distorted by sla¬
ty cleavage. Black slate units within the area contain
species including Climacograptus affinis, C. cf.
tubtdiferus, C. spiniferns, C. bicornis, C. caudatus , Or-
t hog rapt us cf. quadrimucronatus , O. cf. amp/exicaulis,
Dicranograptus rani os us, D. Ilians , Dicellograptus sp.,
and numerous unidentifiable diplograptids (Bolger
1978). These suggest a broad age range of Gisbornian to
Eastonian. One rather dubious occurrence of
Phyllograptus nobilis associated with species of
“Diplograptus” and “Climacograptus” has been record¬
ed from the Gibbo River (Harris & Keble 1932) and sug¬
gests a Darriwilian (Da3) age for the Ordovician in this
area. More recently Kilpatrick and Fleming (1980) have
found early Bendigonian graptolites including
Tctragraptus fruticosus in knotted schist in the Eskdale
area west of the Wombat Creek Graben, indicating that
a large part of the Ordovician sequence (so far un-
fossiliferous) may be of Early Ordovician age.

Silurian

Silurian rocks in the area are confined to the
Wombat Creek Graben and comprise a thick sequence
of acid volcanics (Mitta Mitta Volcanics) and ter-
rigeneous and carbonate sediments (Wombat Creek
Group).

Mitta Mitta Volcanics (Singleton 1965)

Distribution, Thickness and Lithology: The Mitta
Mitta Volcanics comprise a suite of dacite and
rhyodacite outcropping from the junction of the Mitta
Mitta and Gibbo Rivers to near Yankee Point. They
form steep bluffs and resistant ridges on both sides of
the Mitta Mitta River. The rocks are massive and apart
from an outcrop of columnar jointed rhyodacite on the
Mitta Mitta River (GR 595327) the volcanics are struc¬
tureless. They are fine grained, with small phenocrysts
of quartz, plagioclase and rare K-feldspar and biotite in
a chloritic devitrified groundmass. The sparsity of struc¬
tural data and lack of younging criteria in the Mitta Mit¬
ta Volcanics do not allow an accurate estimation of
thickness although the sequence is considered to be at
least 500 m thick on the Mitta Mitta River.

Type Section: The best exposures of the Mitta Mitta
Volcanics were along the Mitta Mitta River prior to in¬
undation by the Dartmouth Dam, and this would have
been the logical type section had flooding not occurred.
The only other accessible area of exposure is along (he

38

P. F. BOLGER

main ridge east of Limestone Gap between the Mitta
Mitta River and Toak’s Creek Track and this is propos¬
ed as the type area for the Mitta Mitta Volcanics (bet¬
ween GR 555388 and 588327).

Relationships: The Mitta Mitta Volcanics are faulted
against Ordovician metasediments in complex fault
zones near Eustace Gap (GR 585428, 589425) and on the
Mitta Mitta River near Yankee Point (GR 538414) where
there is a fault zone 100 m wide. Contacts with the
Wombat Creek Group are also faulted, with much
deformation of the latter at the contact along the Mitta
Mitta River (GR 592322). The presence of pebbles of
Mitta Mitta Volcanics within the basal conglomerate of
the Wombat Creek Group suggest that the Wombat
Creek Group post-dates the volcanics.

Age: The Mitta Mitta Volcanics post-date the Ordovi¬
cian beds and pre-date the Upper Silurian Wombat
Creek Group, but a more accurate age determination is
not possible. The volcanics are considered on regional
grounds to be correlatives of the Thorkidaan Volcanic
Group (VandenBerg et ai in press) outcropping east of
Benambra. Tentative correlation with the Douro Group
(Pogson & Baker, 1974) in the Yass area of New South
Wales implies a Middle Silurian (Wenlockian) age for
the Mitta Mitta Volcanics.

Wombat Creek Group (Crohn 1950)

Distribution, Type Section, Thickness: The Wombat
Creek Group comprises three formations: the Toaks
Creek Conglomerate, the Gibbo River Siltstone and the
Tongaro Sandstone, which all outcrop within the Wom¬
bat Creek Graben from the Toaks Creek track
southwards to near Mt Frazer. A small outlier along
Morass Creek, Benambra, near “The Brothers” contains
sediments referred to the Tongaro Sandstone (Fig. 1).

The proposed type section is a composite section
comprising three segments (Fig. 1): (i) a section along
the Toaks Creek walking track from Toaks Creek (GR
553380) to near Limestone Gap (GR 537363); (ii) the ex¬
posures along the Limestone Gap Track from Limestone
Gap (GR 542355) to the Wombat Creek Fault (GR
541359); (iii) exposures along Wombat Creek from GR
547342 to GR 541336. The thickness in this composite
section is probably in excess of 3800 m (Fig. 2) although
this may be an overestimate due to repetition by tight
folding.

Relationships and Boundary Criteria: The presence
of pebbles of Mitta Mitta Volcanics within con¬
glomerates of the Wombat Creek Group indicates that
the sediments post-date the Volcanics although the con¬
tact is now a low angle thrust fault. At the contact the
Mitta Mitta Volcanics have behaved competently and
were little deformed by the faulting, whereas the Wom¬
bat Creek Group sediments are faulted, folded and in¬
tensely fractured. Along the Gibbo River, the Mitta Mit¬
ta Volcanics have been thrust over the Wombat Creek
Group to give the false appearance of an unconformity
and reversed superposition. The contact with the Mitta
Mitta Volcanics along the Toak’s Creek track is com¬
plexly faulted.

The top of the Wombat Creek Group is faulted
against graptolitic Ordovician sediments along the
Wombat Creek Fault, which has a crush zone 50 m wide
well exposed along the Limestone Gap track (GR
541349). Further south, the boundary with the Ordovi¬
cian rocks is difficult to locate accurately as the Ordovi¬
cian beds adjacent to the western edge of the Wombat
Creek Graben are often quartz-rich and difficult to
distinguish from the Tongaro Sandstone. It is possible
that the graptolitic Ordovician shales exposed along the
Limestone Gap Track and Wombat Creek may be
preserved in narrow fault slivers, with the Tongaro
Sandstone having a more widespread distribution west
of the Wombat Creek Fault, than is depicted in Fig. 1.
Age: Etheridge (in Ferguson 1899) examined fossils
from the Wombat Creek Group and assigned them a
Late Silurian age. Chapman (1920) suggested that some
beds were Middle Devonian, while the remainder were
“Yeringian”, which was then considered to be Late
Silurian. Talent (1960) re-examined Chapman’s faunas
and concluded that there was no evidence for a Devo¬
nian age. The age of the Group is presently considered
to be Late Silurian (Ludlovian-Pridolian), although the
Toaks Creek Conglomerate may even be as old as Llan-
doverian (Talent 1959a, 1965, Talent et a/. 1975).

Toaks Creek Conglomerate

(New name, named after Toaks Creek, Benambra
district)

Distribution, Type Section, Thickness: The Toaks
Creek Conglomerate is a wedge shaped unit of massive
conglomerate with subordinate sandstone, siltstone and
pebbly mudstone which outcrops in the area from the
Toaks Creek Track to the junction of the Mitta Mitta
and Gibbo Rivers. The thickest section and best ex¬
posures of the Toaks Creek Conglomerate are along the
Toaks Creek walking track between Limestone Gap and
Toaks Creek in the northern part of the graben, and this
is the proposed type section (GR 537363 to GR 553380).
Its thickness is probably up to 2300 m. Good exposures
along the Mitta Mitta River are now inundated by the
waters of the Dartmouth Dam.

Lithology: The Toaks Creek Conglomerate is
characterized by massive, clast-supported conglomerate
consisting of well rounded pebbles, cobbles and occa¬
sional boulders of quartzite, vein quartz, black and
green chert, rhyodacite, rare andesite, argillaceous and
rare granitic clasts in a very coarse to fine sandy matrix.
The clast composition becomes less variable towards the
top of the unit where the conglomerate consists almost
entirely of quartzite pebbles. Thin sandstone units con¬
sisting of common quartz as well as some vein quartz,
chert and siltstone grains, are interbedded with con¬
glomerates and have the same composition as the con¬
glomerate matrix.

The conglomerates are massive and usually
poorly bedded. Some thin beds grade upwards from len¬
ticular pebble conglomerate to coarse and medium
grained sandstone. Bedding thickness varies from 5 cm
to several metres. Large conglomerate-filled channels

PALAEOZOIC GEOLOGY BENAMBRA

39

with erosional bases into siltstone are recognizable along
the spur between the Mitta Mitta and Gibbo Rivers.
Elsewhere scour and fill structures, pebble imbrication
and rare cross stratification are observed.

> Typ«

Section

A thin unit of siltstone and feldsparthic sand¬
stone occurs at the base of the Toaks Creek Con¬
glomerate near Toaks Creek. This unit is continuous for
several kilometres southwards to the Mitta Mitta-Gibbo
River junction where it passes into interbedded con¬
glomerate, sandstone, siltstone and pebbly mudstone up
to 30 m thick. The pebbly mudstone contains sparsely
scattered clasts, up to 0.5 m, of limestone, rhyodacite,
vein quartz, chert and quartzite in a green grey struc¬
tureless mudstone matrix. This unit wedges out and is
faulted beneath the Mitta Mitta Volcanics along the
Gibbo River. Thin discontinuous siltstone units are in¬
terbedded with conglomerate. They are structureless to
laminated, grey-green in colour, locally partly silicified
and consist of angular to subangular quartz with in¬
terstitial clays and chlorite. At the contact with the Mitta
Mitta Volcanics the siltstones are strongly fractured.
Palaeontology and Age: Trimerellid brachiopods in¬
dicating a late Llandovery or younger Silurian age have
been found in siltstones intercalated in the Toaks Creek
Conglomerate (Talent 1959a, 1965, Talent et al. 1975).
Rare reworked favositid corals have also been found.
Chapman (1906) reported the long ranging Atrypa
reticularis in the Toaks Creek Conglomerate. A lime-

Q shelly fossil beds
F-. - H bedded, grey crystaine Imestone

Kg interbedded grey quartzitic sandstone and dark grey siltstone

wol bedded, grey, fcw to coarse sandstone, quartzitic. with ttwi
r.» *1 ritorbodded dark grey siltstone. shale

jjgjgj me<*um to coruse sancMarvMjjartzose, sometimes feldspathic

V “ coarse pebble and cobble grain supported conglomerate
lie "J comprising mostly quartzite, some granitic and chert clast

pebbly mudstone, mudstone, minor conglomerate

!•••! coarso pebble and cobble gram supported conglomerate containing
»• abundnnt acid volcanic and chert pebbles

rhyodacite, dacite

Fig. 2 —Columnar sections through the Wombat Creek Group.

40

P. F. BOLGER

stone clast in the pebbly mudstone unit at the Mitta
Mitta-Gibbo River junction yielded a single simplexi-
form element of Panderodus unicostatus but this did
not enable determination of an older age limit for the
Toaks Creek Conglomerate (Cooper 1977).

Gibbo River Siltstone

(New name, named after Gibbo River, Benambra
district)

Distribution, Type Section, Thickness: The Gibbo
River Siltstone is a thin unit of siltstone, calcareous
siltstone, conglomerate, sandstone and, at the base, a
number of limestone lenses. It is best exposed in its type
section along the Limestone Gap Track, where it is 350
m thick (GR 542355 to GR 541352). In the Gibbo River
area and the eastern flank of the Lower Tableland, its
outcrop width is exaggerated by folding, but it is pro¬
bably more than 700 m thick. Individual limestone
lenses are up to 100 m thick (Whitelaw 1954).
Lithology: The dominant lithology is grey-green to buff
coloured siltstone and calcareous siltstone, consisting of
silt-sized angular-subangular quartz grains and some
shell fragments in a clay-chlorite-mica matrix. Beds are
generally massive, often bioturbated, with some planar
lamination and rare small scale cross lamination and
scour-and-fill structures. Near the Wombat Creek Fault,
the siltstones possess a well developed slaty cleavage.
Lenses of fossiliferous grey and white crystalline
limestone occur at the base of the Gibbo River Siltstone

and directly overlie the Toaks Creek Conglomerate in
the western part of the Graben. A thin fossiliferous
sandstone underlies the limestone, along the Mitta Mitta
River. Eastwards along the Gibbo River, limestone
lenses are enveloped by conglomerate and fossiliferous
siltstone. The limestone lenses outcrop as broad mounds
in the Limestone Gap-Quart Pot Flat area, but form ver¬
tical bluffs in the Mitta Mitta and Gibbo River ex¬
posures. They are planar bedded with beds up to 30 cm
thick. The limestones are usually bioclastic (mostly
corals, crinoids and brachiopods), and range from
grainstones to packstones. Pelletal grains are abundant
in limestone near Quart Pot Flat (GR 563332). The lens
at Limestone Gap (GR 542355) contains angular ir¬
regularly shaped clasts of calcareous mudstone, and is
strongly silicified and dolomitized. Recrystallized
dolomitic limestone occurs at the hairpin bend of the
Gibbo River (GR 608314). Coarse-grained marble with
its original depositional texture completely obliterated
outcrops near the mouth of Morass Creek.

Numerous thin conglomerates up to 10 m thick
are interbeddcd with siltstones in the Morass Creek and
Gibbo River areas. They contain well rounded quartzite
and rare chert clasts. Along the Mitta Mitta River, the
Gibbo River Siltstone becomes coarser grained near the
top and comprises siltstone interbedded with graded and
cross laminated fine to coarse quartz sandstone, a thin
intraformational conglomerate, and a massive struc¬
tureless cobble conglomerate, up to 25 m thick, contain-

TOAKS CREEK MITTA MITTA RIVER GIBBO RIVER
WOMBAT CREEK

Tongaro Sandstone

°o5>

° ^ O ° g-

o o o o O O O

o o

o ° o °— 1°.£

° o Toaks Creek 0 o Zo ooTiD EjGTbbo River Siltstone;

o Conglomerate 0 ° ° °

0 o O o-

o o ° ° -• —

O ° ° o ° „ o

°ooOo ° 0 o o ooo

T-

Mitta Mrt t a' V ole a ri i cs: ■ *

Fig. 3 —Facies relationships of Silurian units within the Wombat Creek Graben.

PALAEOZOIC GEOLOGY BENAMBRA

41

ing mostly quartzite and some granitic clasts.
Relationships: The Gibbo River Siltstone conformably
overlies the Toaks Creek Conglomerate along the Mitta
Mitta River and in the Limestone Gap-Quart Pot Flat
area, although in the Gibbo River area, the basal beds
are laterally equivalent to parts of the Toaks Creek Con¬
glomerate (Fig. 3). VandenBerg (1976) suggested that
the limestones are allochthonous blocks contained in the
upper units of the Toaks Creek Conglomerate.
However, although the contacts with the surrounding
rocks are not well exposed, the limestone lenses appear
to be enclosed in fossiliferous, often calcareous siltstone
of the Gibbo River Siltstone and are considered to be
autochthonous.

The boundary with the overlying Tongaro Sand¬
stone is sharp along Limestone Gap track, but along the
Mitta Mitta River, the Gibbo River Siltstone is sandy
towards the top and appears to grade into the Tongaro
Sandstone.

Palaeontology and Age: The Gibbo River Siltstone
contains a diverse fauna including brachiopods,
crinoids, trilobites and corals in the siltstones, and cor¬
als, stromatoporoids, crinoids, brachiopods and rare
conodonts in the limestone lenses. The fauna has not
been examined in detail and awaits future study.

The siltstones contain species of I sort his,
Platystrophia, Hesperorthis, Atrypoidea, (?) Pen-
la merus, HowelleHa and Nucleospira (M. J. Garratt
pers. comm. 1974) which suggest a Late Silurian
(Ludlow) age. Mucophyllum liliiforme, Propora confer -
ta and Favosites allani have been identified from
limestones and indicate a Wenlock to Ludlow age
(Talent 1959a, 1960, Talent et al. 1975). Cooper (1977)
extracted elements of the conodonts Ozarkodina ex -
cavata, Panderodus unicostatus , and Walliserodus sp.
but only a broad Early Silurian to Early Devonian age
could be inferred from this collection.

Chapman (1906) described a species of a
ccphalaspid fish from siltstones in the formation, but
this was later reidentified as a cast of a compound coral
(Hills 1958).

Tongaro Sandstone

(New name, named after Parish of Tongaro)

Distribution, Type Section, Thickness: The Tongaro
Sandstone, comprising sandstone, siltstone, con¬
glomerate and limestone is the uppermost unit of the
Wombat Creek Group and outcrops between the
Limestone Gap Track and the Mitta Mitta River. The
proposed type section on the Mitta Mitta River between
GR 586312 and GR 588260 contains good exposures on
the spurs overlooking the river. The thickness in this sec¬
tion is approximately 2000m and a sequence 1200 m
thick is exposed in a reference section along Wombat
Creek (between GR 547342 and C.R 541336).

Lithology: The Tongaro Sandstone is characterized by
well bedded quartzitic sandstone and interbedded grey
to black siltstone. The sandstones are grey coloured, fine
to medium grained quartz wackes ( sensu Okada 1971)

consisting of common quartz sand, rare muscovite,
altered feldspar, chert and mudstone in a muddy matrix.
Some sandstones have a bimodal framework consisting
of well rounded coarse sand and angular fine sand
grains. Beds are usually less than 30 cm thick with planar
boundaries. Sedimentary structures include planar
lamination, cross lamination and ripple marks. Graded
units arc rare. Lenses of well-bedded dark grey
limestone up to 30 m thick (Whitelaw 1954) outcrop
sporadically along the Mitta Mitta River and northwest
towards Wombat Creek. They are planar bedded, with
beds up to 50 cm thick. A fossiliferous packstone-
wackestone exposed along the Mitta Mitta River is
strongly foliated.

The Tongaro Sandstone varies laterally. In the
northwest along Wombat Creek,, quartz sandstone is less
abundant than along the Mitta Mitta River, while thinly
bedded dark grey to black siltstones increase in
thickness. Massive, poorly bedded lenticular con¬
glomerates, which wedge out southwards, form steep
bluffs and strike ridges near Wombat Creek southwest
of Limestone Gap. The conglomerates consist of ran¬
domly orientated, well rounded cobbles and pebbles of
quartzite. Individual conglomerate lenses are up to 200
m thick.

A small outlier of the Wombat Creek Group is
exposed along Morass Creek near Benambra township
and is tentatively referred to the Tongaro Sandstone. It
consists of a thin, poorly exposed grey limestone, known
as Pyle’s limestone deposit, slaty siltstone and quartzite
pebble conglomerate. The limestone has been com¬
pletely recrystallized and contact metamorphosed to
skarn by the intrusion of the Triassic Mount Leinster
Complex. Wollastonite is recognizable in thin section
and Whitelaw (1954) reported the occurrence of garnet.
Relationships: The Tongaro Sandstone conformably
overlies the Gibbo River Siltstone along the Mitta Mitta
River. The base of the Tongaro is marked by the first
plane bedded structureless quartzitic sandstone. The
relationships are complicated by faulting and overturn¬
ing along the Limestone Gap Track, although here the
Tongaro Sandstone also appears to overlie the Gibbo
River Siltstone. Along strike, the Tongaro grades
laterally into the upper parts of the Gibbo River
Siltstone. The top of the Tongaro Sandstone is faulted
by the Wombat Creek Fault against graptolitic Upper
Ordovician quartzite, sandstone and slate along Wom¬
bat Creek and along the Limestone Gap Track where
there is a 50 m wide crush zone. Due to the lithological
similarity with the Ordovician the faulted boundary is
more difficult to locate between Wombat Creek and the
Mitta Mitta River.

Palaeontology and Age: Apart from unidentifiable
fragmentary shelly fossils in sandstones along the
Limestone Gap Track, the non-calcareous part of the
Tongaro Sandstone is unfossiliferous. Limestones out¬
cropping along the Mitta Mitta River contain halysitid
corals (Talent 1959a, 1965, Talent et al. 1975) and pen-
tamerid brachiopods.

42

P. F. BOLGER

Conodonts recovered from limestone along the
Mitta Mitta River include a single costate element of
Panderodus unicostatus, which has ornament of a type
only found in post-Wenlockian collections (Cooper
1977). The presence of halysitid corals indicates a
Silurian age, thus confining the Tongaro sandstone in
the Mitta Mitta River to the Late Silurian. Pyles
limestone deposit near Benambra contains poorly
preserved spiriferid brachiopods and conodonts in¬
cluding Spathognathodus remscheidensisl , S. inc/inatus
inclinatus, Hmdeodella prisci/la and Neoprionodus sp.,
which suggest a Pridolian or younger age (Bischoff, in
Talent et al. 1975) for this deposit.

CORRELATION

The Mitta Mitta Volcanics and the Wombat
Creek Group can be broadly correlated with other
Silurian successions in southeastern Australia. The most
similar sequence is exposed in the Indi River-Limestone
Creek area east of Benambra, where the Silurian se¬
quence comprises an acid volcanic unit, the Thorkidaan
Volcanic Group, conformably overlain by the Enano
Group, which consists of terrigenous and carbonate
sediments with thin intercalated volcanics (A. H. M.
VandenBerg et aL in press). Tentative correlation
between units in the Wombat Creek and Indi River suc¬
cessions (Table 1) is based more on lithological
similarities than on palaeontological evidence, although
the Gibbo River Siltstone at Wombat Creek and the
Cowombat Siltstone along the Indi River contain similar
faunas and are probably contemporaneous units (Talent
1959a).

Further east in the Yalmy area, possible
equivalents to the Wombat Creek Group are the Sardine
Beds (Talent et al. 1975) while Silurian sediments are
known in bores near Nowa Nowa (Talent 1959a). A
poorly known succession at Quidong north of Yalmy
contains beds of similar age to the Wombat Creek
Group (Talent et al. 1975).

Precise correlation with the better known succes¬
sion at Yass in New South Wales is not possible
although the regional stratigraphy suggests possible cor¬
relation of the Mitta Mitta Volcanics with the Hawkins
Volcanics of the Douro Group, and the Gibbo River
Siltstone is probably equivalent to parts of the Silverdale
Formation and Black Bog Shale.

DISCUSSION OF THE WOMBAT CREEK GROUP

Depositional Environments

The Toaks Creek Conglomerate forms a wedge-
shaped body in the northwest of the Wombat Creek
Graben. Sedimentary structures include graded bedding,
channels, rare cross-stratification and imbrication. The
conglomerate has probably been deposited largely by
traction currents although one pebbly mudstone unit
near the base suggests mass movement. The Toaks
Creek Conglomerate could be a type of “delta-fan”
deposit accumulated at the margins of the marine basin.
The occurrence of shelly fossils in interbedded

mudstones indicates a marine origin for at least part of
the Toaks Creek Conglomerate, but the possibility that
part of the unit is of fluvial origin cannot be overlooked.
The Gibbo River Siltstone comprises siltstone, limestone
and thin conglomerate and sandstone. The lenticular
limestones are mostly grainstones, and some pack-
stones, cemented by ferroan calcite. They are inter¬
preted to be carbonate buildups deposited under open
shelf conditions. Siltstones for the most part are con¬
sidered to be deposits settled from suspension onto a
shelf inhabited by a diverse benthonic biota which has
bioturbated the substrate. However they do contain rare
sedimentary structures such as scour and fill structures,
which suggest some current activity. The conglomerates
in the Gibbo River Siltstone are massive, structureless
closed framework deposits. Sandstones exposed near the
top are graded, cross-laminated and associated with thin
intra-formational conglomerate and show evidence of
deposition by turbidity currents.

The Tongaro Sandstone in the Mitta Mitta River
area consists of thin bedded quartz sandstone and in¬
terbedded dark siltstone enclosing small randomly
distributed carbonate bodies. The sandstones contain
few sedimentary structures, although small scale cross-
stratification and planar lamination are observed. In the
Wombat Creek area to the northwest the Tongaro Sand¬
stone contains lenticular closed framework con¬
glomerates. The conglomerates are interbedded with
thinly bedded dark mudstone and thin, sometimes ripple
marked, laminated and small scale cross-stratified
quartz sandstone containing fragmented shelly fossils.
Insufficient data are available to conclusively determine
the deposition^ environment of the Tongaro Sand¬
stone. The sandstones do not display well developed
Bouma sequences. However, they are thin, laterally con¬
tinuous and internally planar and cross-laminated, and
resemble turbidites of Facies D and E described by Mutti
and Ricci Lucchi (1978).

It is therefore suggested that the sandstones are
mass-flow deposits. The association of turbiditic sand¬
stones with conglomerates near Wombat Creek suggests
that the conglomerates were deposited below wave base.
It is not clear whether the limestone bodies along the
Mitta Mitta River are allochthonous or were deposited
in situ.

Nature and Origin of Terrigenous Detritus

The basal Toaks Creek Conglomerate contains
abundant pebbles of acid volcanic rocks derived from
the underlying Mitta Mitta Volcanics. Towards the top
of the unit the conglomerates become compositionally
more mature and contain mostly quartzite and some
chert pebbles.

The source of the quartzite clasts is uncertain,
although clearly they are less muddy and less micaceous
than the local Ordovician sandstones. Similarly chert,
which is a major component of the Toaks Creek Con¬
glomerate, is rare to absent in the Ordovician nearby.
Argillaceous clasts in the Toaks Creek Conglomerate are
not strongly foliated, although some have a weak fissil-

PALAEOZOIC GEOLOGY BENAMBRA

43

Table 1

Correlation of Wombat Creek Graben Sequence with other Silurian Sequences in Southeastern

Australia.

1, Nomenclature this paper. 2, Nomenclature, VandenBerg et al. 1981. 3, Nomenclature, VandenBerg
(unpubl.). 4, Nomenclature, Campbell in Talent et al. 1975. 5, Nomenclature, Pogson & Baker 1974.

LOCHKOVI AN

WOMBAT

CREEK

L

U

D

L

0

V

I

A

N

INDI RIVER

LIMESTONE

CREEK

YALMY

QUIDONG

LATE

ORDOVICIAN

EARLY-MIDDLE
ORDOVICIAN

Early-Late

Ordovician

Ordovician

SARDINE 0

o

° 0 0
0 0 0
0 n °
BEDS

0

0 0 0

• o° .

0 ?

YALMY

GROUP

Warbisco

Shale

Pinnak

Sandstone

Unnamed

Ordovician

YASS

ELMSIDE

FORMATION

COWRIDGE

SILTSTONE

BARAMBOGIE

GROUP

ROSEBANK BOOROO

SHALE PONDS

BLACK DOG GROUP

MERRIANGAAH

SILTSTONE

TOMBONG
: BEDS x

1 TONGARO SANDSTONE

2 GIBBO RIVER SILTSTONE

3 TOAKS CRE.SK CONGLOMERATE

4 MITTA MITTA VOLCANICS

5 TOWANGA SANDSTONE

6 COWOMBAT SILTSTONE

7GIBSONS FOLLY FORMATION
8 THORKIDAAN VOLCANIC GROUP

Jerrawa

Beds

acid to intermediate voIconics
conglomerate, minor sandstone
[ | sandstone, minor conglomerate, siltstone

siltstone, minor limestone, conglomerate

| ; ! j limestone, minor mudstone

| siltstone, volcanics (undifferentiated)
[ | shale, siltstone

ity, and no clasts of higher grade regional metamorphic
rocks have been observed.

The Gibbo River Siltstone is considered to be a
transgressive facies which has covered the coarse clastic
wedge of the Toaks Creek Conglomerate with a thin
cover of siltstone and limestone. Further offshore there
was continued settling of silt and clay from suspension
with intermittent influx of coarse detritus transported in¬
to the basin by mass flows. It is in part laterally
equivalent to the Toaks Creek Conglomerate, but
represents a more offshore marine environment.

Towards the top of the Wombat Creek Group
the coarse sediments become compositionally mature,
consisting almost entirely of quartzite and rare granitic
clasts. Again the quartzite clasts are less muddy than the
Ordovician sandstones.

The quartzite clasts are petrographically similar

to quartzitic sandstones of the Lower Silurian Yalmy
Group (A. H. M. VandenBerg pers. comm.) which con¬
formably overlies Upper Ordovician graptolitic shales in
the Yalmy area to the east. It is suggested that deposi¬
tion in the Early Silurian may have continued as far west
as the Wombat Creek Graben, covering the Ordovician
sediments. Following extrusion of the Mitta Mitta
Volcanics, erosion of the volcanics occurring outside the
graben, as well as the Low'er Silurian sediments and
minor high level intrusions provided detritus to the sub¬
siding basin. However, erosion did not proceed to suffi¬
cient depth to expose Ordovician beds or high grade
regional metamorphic rocks.

44

P. F. BOLGER

STRUCTURAL RELATIONSHIPS BETWEEN
ORDOVICIAN AND SILURIAN

Ordovician Structures

The most prominent structural element in the Or¬
dovician is bedding which is folded into tight to isoclinal
folds with steeply dipping to vertical axial surfaces.
Isoclinal fold hinges have not been observed, but are in¬
ferred from changes of facing of beds. Major folds in
the Ordovician trend between 120° and 200°, with the
dominant direction NW-SE (Fig. 4a, b, c). Small scale,
variably plunging folds, which may be minor folds on
the Hanks of major structures, are common in some
areas and have an east-west trend. The folds are
characterized by an axial plane slaty cleavage, defined by
a preferred orientation of platy minerals and elongate
quartz grains. The cleavage is strongly developed and
penetrative in the slates, but is only weakly developed in
the sandstones.

In thin section, a fissility defined by platy
minerals aligned parallel to bedding can be seen to have
been crenulated by the slaty cleavage. This earlier sur¬
face may be a folded slaty cleavage which is axial planar
to an earlier set of folds, although no refolded folds
have been observed in this area. However further north
near Lockhart Gap, McKay (1969) has recorded
mesoscopic folds refolded by northwest-trending
regional structures. East-west trending folds which have
been refolded by the northwest-southwest trending
regional structures of the Onieo Metamorphic Complex
have been recognized at Albury (Heilman 1976) and
Tallandoon (Rogerson 1976). It thus appears that the
Ordovician beds in northeastern Victoria have under¬
gone at least two deformations.

Silurian Structures

The only prominent structural element observed
throughout the Wombat Creek Group is Bedding. Beds

4a ORDOVICIAN BEDS

Junction of Razorback Spur Track and Merrimac
Spur Track Junction to Limestone Gap
Poles to bedding S° • (36 points)

Poles to cleavage S° (16 points)

4b ORDOVICIAN BEDS
Razorback Spur Track West of
Merrimac Spur Track Junction
Poles to bedding S° • (20 points)

N

4c ORDOVICIAN BEDS
Benambra-Corryong Road
Poles to bedding S° • (27 points)
Poles to cleavage S-j * (2 points)

Poles to bedding S s • (43 points)

Fig. 4 — Stereographic projections of bedding and cleavage surfaces.

PALAEOZOIC GEOLOGY BENAMBRA

45

are usually steeply dipping and sometimes overturned.
No fold hinges have been observed, and the average fold
trend of 150° is inferred from the direction and attitude
of dips (Fig. 4d).

In contrast to the underlying Ordovician slates,
strong foliation is only developed near major faults.
There is a poorly defined reticulate cleavage in siltstones
near the junction of the Mitta Mitta and Gibbo Rivers.
Adjacent to the Wombat Creek Fault there is slaty
cleavage in parts of the Gibbo River Siltstone and a pro¬
minent foliation, defined by stretched brachiopods, is
parallel to bedding in a limestone lens of the Tongaro
Sandstone on the east bank of the Mitta Mitta River.

Because of the near absence of bedding and the
paucity of dipsin the Mitta Mitta Volcanics, the struc¬
ture of this unit is not well known, but bedding trends
are parallel with those of the Wombat Creek Group.

Discussion of the Benambran Deformation (or
Mitta Mitta Movement)

The Bcnambra-Wombat Creek area is structur¬
ally significant in that it is the type area of the Benam¬
bran Orogeny (Browne 1947) —a tectonic event recorded
from many localities through the Lachlan Fold Belt (e.g.
Packham 1969, Crook et al. 1973). The deforma¬
tion in the type area has been inferred from the struc¬
tural differences between the Ordovician and Silurian
beds (Talent 1959a). The Benambran Deformation was
originally called the Mitta Mitta Movement by Andrews
(1938, p. 151) but this name has failed to gain accep¬
tance.

The contacts between the Ordovician and
Silurian are always faulted and no angular unconformity
has been observed. However the fold shapes and degree
of deformation appear to differ between the Ordovician
and Silurian in the study area. The variably plunging,
tight, small scale folds in the Ordovician are not ob¬
served in the Silurian and the Ordovician probably has a
refolded cleavage. In contrast, the continuity of marker
beds in the Wombat Creek Group suggests tight but
relatively simple folding in the Silurian rocks.

The metamorphic grade of the Ordovician in the
Wombat Creek area is much higher than the Silurian,
with the Iow r est grade (biotite zone) rocks passing
westwards into the high temperature-low pressure Omeo
Metamorphic Complex. Heilman (1976) has shown that
the regional metamorphism at Albury post-dates early
east-west trending folds but pre-dates northwest-
southeast trending folds which are the prominent fold
trends in the Omeo Metamorphic Complex. However,
McKay (1969) and Rogerson (1976) suggested that
metamorphic crystallization accompanied and post¬
dated the northwest-southeast folding.

The Omeo Metamorphic Complex is intruded by
Early to Middle Silurian S-type granitic rocks with
Rb/Sr ages of 430±8 m.y. (Brooks & Leggo 1972,
recalculated by J. Richards 1980 pers. comm.) and K/Ar
ages from isolated granitic intrusions of 440±9 m.y. (J.
Richards pers. comm. 1980).

The Ordovician in northeastern Victoria has

undergone at least two periods of folding with concomi¬
tant metamorphic and intrusive events. The folding and
thermal events are here referred to as the Benambran
Deformation, which has traditionally been considered to
be Early Silurian in northeastern Victoria (Talent 1969,
VandenBerg 1978).

In New South Wales, the Benambran Deforma¬
tion is considered to range from the Early Silurian and
comprise three separate tectonic events inferred from
unconformities within the Silurian sequence at Orange
(Packham 1969). The oldest of these, referred to as the
Cobblers Creek Orogenic Phase, is of Late Ordovician to
Llandovery age, while the Panuara Phase and the
Quarry Creek Phase range from middle to late Llan¬
dovery and late Llandovery to Wenlock respectively.
The tendency has been to correlate these events
throughout the Lachlan Fold Belt (Scheibner 1973,
Talent et al.A915), thereby assigning minor breaks in
deposition and slight angular discordances to intense
deformation events.

There is good evidence to infer that deformations
took place between the Late Ordovician and Llandovery
and the late Llandovery and the late Wenlock-Ludlow' at
a number of localities in southeastern New South Wales
(Crook et al. 1973). The former event is referred to the
Benambran Deformation and the latter to the
Quidongan Deformation (Crook et al. 1973).

The age of the Benambran Deformation in its
type area in northeastern Victoria is not closely con¬
strained. The oldest possible age of the Toaks Creek
Conglomerate, based on the occurrence of poorly
preserved trimerellid brachiopods, is late Llandovery
(Talent 1959a). However, tentative correlations with
other areas in southeastern Australia (Table 1) suggest
that the Mitta Mitta Volcanics and the Wombat Creek
Group may be Middle and Late Silurian (Wenlock and
Ludlow' to Pridoli) respectively. If these “preferred” ages
are accepted, there is a wide gap between deposition of
graptolitic shales in the Late Ordovician and the extru¬
sion of the Mitta Mitta Volcanics.

Evidence from east of the Wombat Creek
Graben at Yalmy suggests that deposition was con¬
tinuous during the l ate Ordovician and Early Silurian
(A. H. M. VandenBerg pers. comm.). It may have also
extended westwards into the Wombat Creek Graben and
the Indi River-Limestone Creek Graben. If so, the
possibility that the Benambran Deformation is actually
a late Early Silurian to Middle Silurian event cannot be
precluded. The real Benambran Deformation in eastern
Victoria may, in fact, be equivalent to the Wenlockian
Quidongan Deformation, and events referred to the
Benambran Deformation in southeastern New South
Wales may be a pre-Benambran event not recognised in
eastern Victoria.

SUMMARY

Turbiditic Early to Late Ordovician sandstone
and shale suffered multiple deformation, high tempera¬
ture-low pressure regional metamorphism and granitic

46

P. F. BOLGER

intrusion during the Benambran Deformation. The
resulting rocks comprise the Omeo Metamorphie com¬
plex.

The Wombat Creek Graben was initiated during
or after the Benambran Deformation. In the Middle (?)
to Late Silurian it was a locus for accumulation of a
thick acid volcanic pile (the Mitta Mitta Volcanics) and
subsequent fluvial (?) to shallow marine clastic and
calcareous sediments (Toaks Creek Conglomerate and
Gibbo River Siltstone) and turbiditic marine clastic
sediments (Tongaro Sandstone). Clastic detritus was
derived from the Mitta Mitta Volcanics extruded outside
the graben, possibly from Lower Silurian clastic
deposits and from the highest levels of the Omeo
Metamorphie Complex. High grade rocks of the
Metamorphie Complex were not exposed during the
deposition of the Wombat Creek Group.

ACKNOWLEDGEMENTS

The author wishes to thank O. P. Singleton for
supervision of the project during 1974 and R. A. Cas,
M. J. Garratt, T. G. Russell and A. H. M. VandenBerg
for critical reading of the manuscript. M. J. Garratt and
A. H. M. VandenBerg assisted in identification of
fossils. Limestone samples were processed through the
courtesy of the Geological Survey of South Australia
and conodonts recovered were examined by B. J.
Cooper. H. R. Thorne and P. D. Wood of State Rivers
and Water Supply Commission provided some field data
and helpful discussion. The paper is published with per¬
mission of the Director of the Geological Survey Divi¬
sion, Department of Minerals and Energy, Victoria.

REFERENCES

Andrews, E. C., 1938. The structural history of Australia
during the Palaeozoic: The stabilization of a continent.
J. Proc. R. Soc. N.S.W. 71: 118-187.

Beavjs, F. C., 1962. The geology of the Kiewa area. Proc. R.
Soc. Viet. 75: 349-410.

Bolger, P. F., 1978. New graptolite localities from

northeastern Victoria. Rep. geo/. Surv. Viet. 1978/44.
(Unpubl.).

Bouma, A. H., 1962. Sedimentology of some flysch deposits. A
graphic approach to facies interpretation. Elsevier,
Amsterdam, 168 pp.

Brooks, C. & Leggo, M. D. 1972. The local chronology and
regional implications of a Rb-Sr investigation of
granitic rocks from the Corryong District, southeastern
Australia. J. geol. Soc. Aust. 19: 1-19.

Browne, W. R., 1947. A short history of the Tasman
Geosyncline of Eastern Australia. Science Progress 35:
623-637.

Chapman, F. f 1906. New or little-known Victorian fossils in
the National Museum, Melbourne. Part VII —A new
cephalaspid, from the Silurian of Wombat Creek.
Proc. R. Soc. Viet. 18: 93-100.

Chapman, F., 1912. Reports on fossils. Silurian and Devonian
fossils from the Mitta Mitta District, northeast Vic¬
toria. Rec. geol. Surv. Viet. 3: 215-217.

Chapman, F., 1917. Preliminary notes on new species of
Silurian and Devonian fossils from northeast Gipps-
land. Rec. geol. Surv. Viet. 4: 103-104.

Chapman, F., 1920. Palaeozoic fossils of eastern Victoria.

Part IV. Rec. geol. Surv. Viet. 4: 175-194.

Cooper, B. J., 1977. Preliminary report on conodonts from
the Wombat Creek Group, north eastern Victoria. S.
Aust. Dept. Mines Rept. 77/139 (Unpubl.).

Crohn, P. W., 1950. The geology, petrology and

physiography of the Omeo District, north-eastern Vic¬
toria. Proc. R. Soc. Viet. 62: 1-70.

Crook, K. A. W., Bein, J., Hughes, R. J. & Scott, P. A.,
1973. Ordovician and Silurian history of the south¬
eastern part of the Lachlan Geosyncline. J. geol. Soc.
Aust. 20: 113-138.

Dunn, E. J., 1907a. Geological notes on the Mitta Mitta River,
north-east District. Rec. geol. Surv. Viet. 2: 122-123.
Dunn, E. J., 1907b. Limestone deposits at Wombat Creek and
the Mitta Mitta River. Rec. geol. Surv. Viet. 2:
123-124.

Ferguson, W. H., 1899. Report on collection of fossils, etc.,
from Wombat Creek. Mon. Progr. Rept. geol. Surv.
Viet. 3: 1-17.

Harris, W. J. & Keble, R. A., 1932. Victorian graptolite
zones, with correlations and description of species.
Proc. R. Soc. Viet. 44: 25-48.

Hellman, P. L., 1976. Structural analysis of the Albury
District, N.S.W. ./. Proc. R. Soc. N.S.W. 109:
103-113.

Hills, E. S., 1958. A brief review of Australian fossil
vertebrates. In Studies on fossil vertebrates , T. S.
Westoll, ed., Athlonc Press, London, 86-107.

Kenny, J. P. L., 1937. Dam site, Mitta Mitta River, below
Gibbo Junction. Rec. geol. Surv. Viet. 5: 469-470.
Kilpatrick, D. J. & Fleming, P. D., 1980. Lower Ordovician
sediments in the Wagga Trough: discovery of early
Bendigonian graptolites near Eskdale, north-east Vic¬
toria. J. geol. Soc. Aust. 27: 69-73.

McKay, W. J., 1969. Metamorphie and igneous rocks in the
Tallangatta District, Northeast Victoria. B.Sc. (Hons)
Thesis, Aust. Nat. Univ. (Unpubl.).

Mum, E. & Ricci Lucchi, F., 1978. Turbidites of the
Northern Apennines: introduction to facies analysis.
Int. Geol. Review 20: 125-166.

Ok ada, H., 1971. Classification of sandstone: analysis and
proposal. J. Geol. 79: 509-525.

Packham, G. H., 1969. Southern and Central Highlands Fold
Belt. Tectonics and sedimentation. J. geol. Soc. Aust.
16: 216-226.

Pogson, D. J. & Baker, C. J., 1974. Revised stratigraphic
nomenclature for the Yass 1:100 000 sheet. Quart.
Notes geol. Surv. N.S.W. 16: 7-9.

Rogerson, R. J., 1976. Metamorphism, folding and plutonism
in the Wagga Metamorphie Belt of N.E. Victoria. Bull.
Aust. Soc. Explor. Geophys. 1: 41-43.

Scheibner, E., 1973. A plate tectonic model of the Palaeozoic
tectonic history of New South Wales. J. geol. Soc.
Aust. 20: 405-426.

Singleton, O. P., 1965. Geology and mineralization of Vic¬
toria. In Geology of Australian Ore Deposits, J.
McAndrew, ed., AIMM, Melbourne, 1: 440-449.
Stirling, J., 1887. Second progress report on preliminary
geological traverse of the western boundary of County
of Benambra. Quart. Rep. Mining & Surveyors
Registrars Dec. 1887, Appendix D: 75.

Stirling, J., 1888. Preliminary notes on the geology of the
Wombat Creek Valley, its caves and silver lodes.
Quart. Rep. Mining & Surveyors Registrars Sept. 1888:
78-80.

Stirling, J., 1889. Report on the tin lodes at Wombat Creek.

PALAEOZOIC GEOLOGY BENAMBRA

47

Quart. Rep. Mining & Surveyors Registrars March
1889, Appendix A: 65-67.

Talent, J. A., 1959a. Notes on Middle Palaeozoic

stratigraphy and diastrophism in eastern Victoria. Min.
geol. J. Viet . 6: 57-58.

Talent, J. A., 1959b. Subsurface Silurian sediments, Parish
of Novva Nowa South, Victoria. Bull. geol. Surv. Viet.
57: 45-48.

Talent, J. A., 1960. Contributions to the stratigraphy and
palaeontology of the Silurian and Devonian of Gipps-
iand. Ph.D. Thesis, Melbourne University. (Unpubl.).

Talent, J. A., 1965. The stratigraphic and diastrophic evolu¬
tion of central and eastern Victoria in Middle
Palaeozoic times. Proe. R. Soe, Viet. 79: 179-195.

Talent, J. A., 1969. The Geology of East Gippsland. Proe. R.
Soe. Viet. 82: 37-60.

Talent, J. A., Berry, W. B. N. & Boucot, A. J., 1975. Cor¬
relation of the Silurian rocks of Australia, New
Zealand and New Guinea. Spec. Pap. geol. Soe. Amer.
150: 1-108.

VandenBerg, A. H. M., 1976. Silurian-Middle Devonian of
Eastern Victoria. In Geology of Victoria, J. G.
Douglas & J. A. Ferguson, eds, Spec. Publ. geol. Soe.
Aust. 5: 62-70.

VandenBerg, A. H. M., 1978. The Tasman Fold Belt System
in Victoria. Tectonophysics 48: 267-297.

VandenBerg, A. H. M, Bolger, P. F., Carey, S. P., O’Shea,
P. J. & Nott, R. J., 1979. Geology of the Limestone
Creek area, northeast Victoria. In Victoria Exploration
Potential, Seminar. Aust. Min. Found, and Dept, of
Mins. & Energy, Melbourne.

VandenBerg, A. H. M., Bolger, P. F. & O’Shea, P. J. in
press. Geology and mineral exploration of the
Limestone Creek area of northeast Victoria. Rept.
geol. Surv. Viet. 72.

Whitelaw, H. S., 1954. Some limestone and marble deposits
in east Gippsland. Min. geol. J. Viet. 5(3): 23-33.

*

PROC. R. SOC. VICT. vol. 94, no. 1, 49-52, March 1982
OCCURRENCE OF THE TASMANIAN MUDFISH, GALAXIAS
CLEAVERI SCOTT, ON WILSONS PROMONTORY-FIRST
RECORD FROM MAINLAND AUSTRALIA

By P. D. Jackson and J. N. Davies

Fisheries and Wildlife Division, Arthur Rylah Institute for Environmental Research, P.O. Box 137,

Heidelberg, Victoria 3084

Abstract: The Tasmanian mudfish, Galaxias cleaveri Scott, is recorded from Wilsons Promon-
tory, Victoria which is the first report from outside Tasmania. This occurrence is considered in the con¬
text of recent land bridges between the mainland and Tasmania.

Scott (1934) described the Tasmanian mudfish,
Galaxias cleaveri , and later (Scott 1936) created the
genus Saxilaga for G. cleaveri and another species, S.
anguilliformis. Subsequently he (Scott 1942) described a
third species of mudfish, Galaxias upcheri. Andrews
(1976) considered the two Saxilaga species not gen-
erically separate from Galaxias and judged S.
anguilliformis and G. upcheri to be synonyms of G.
cleaveri. Recent observations by McDowall and
Frankenberg (1981) support these views.

Galaxias cleaveri is one of the most specialised of
the galaxiids, having adopted a benthic mode of life in
swamps and drains, where it is apparently able to
aestivate during periods of drought (Scott 1934). Ii has
been recorded from low-lying coastal areas in north,
west and south-western Tasmania (Fig. 1). This paper
documents the occurrence of G. cleaveri on mainland
Australia.

METHODS

Specimens of G. cleaveri were obtained during a
survey of the fish fauna of Wilsons Promontory between
29 April and 16 December 1980. A number of sampling
techniques were employed during the survey but all
specimens of G. cleaveri were taken by electrofishing.

Fish were anaesthetised with quinaldine before
being fixed in 10% neutral formalin. After about two
weeks, they were transferred to 70% alcohol. Body
measurements, according to Andrews (1976), were taken
with vernier calipers, read to the nearest 0.1 mm. The
four largest specimens were X-rayed and all vertebrae
having unmodified centra at both ends counted.

On 16 December 1980, the ph, conductivity,
salinity and dissolved oxygen were measured at one
point in the sampling site. Depth, to the nearest 5 cm,
was recorded every 5 m along a single transect across the
site.

RESULTS

Locality

Specimens of G. cleaveri were taken from a single
locality on a small, swampy tributary of Freshwater
Creek, on the south-eastern side of the Promontory
(Fig. 1; Lat. 39°4'S Long. 146°26'E). There was no

discernible flow, mean depth was 16 cm and the
substrate was *mud. Conductivity was 1364fimhos at
25°C and pH was 5.3. There was no measurable salinity
or dissolved oxygen and a ‘metallic’ film was present
over much of the water surface, possibly due to the
presence of metallic sulphides formed under anaerobic
conditions (T. Pearce, pers. comm. 1981). This film was
not present when the site was first visited on 29 October
1980.

Dense stands of tea-tree, Melaleuca sp., were
present together with areas of eel grass, Triglochan sp.
The sampling area was partially cleared to construct a
walkway for hikers. Outside the area the tea-trees were
only centimetres apart, making electrofishing impos¬
sible.

Specimens

Five specimens of G. cleaveri were captured on
29 October 1980 and ten on 16 December 1980. Ten were
preserved for identification and have been deposited in
the National Museum of Victoria (NMVA 2037). Other
species captured were the common jollytail, Galaxias
maculatus (Jenyns) (82 specimens, 44-95 mm total
length), and the short-finned eel, Anguilla australis
Richardson, (2 specimens, 231-235 mm total length).

The standard lengths of preserved specimens of
G. cleaveri ranged from 37.4 to 75.4 mm with a mean of
52.5 mm. Morphometric and meristic data recorded
from the 10 specimens are shown in Table 1. Also shown
are the ranges of data recorded by Andrews (1976) from
three separate collections, each of 10 fish, from
Tasmania.

DISCUSSION

Table 1 shows that, except for the ratios of in¬
terorbital width to head length and head length to stan¬
dard length, the mean body ratios for Wilsons Prom¬
ontory specimens all fall within the range of data record¬
ed by Andrews (1976). The deviant proportions are only
marginally outside that range. Similarly, meristic varia¬
tions agree closely.

The occurrence of G. cleaveri on Wilsons Pro¬
montory undoubtedly reflects the recent land connec¬
tions between Victoria and Tasmania. Lying on the con-

49

50

P. D. JACKSON AND J. N. DAVIES

■7

Melbourne

King rv
Island ( J

Wilsons ( see enlargement

Promontory

<T\ Flinders
VJ Island

100

I

0

1

k m

100

Fig. 1—Current known distribution of Galaxias cleaveri with Tasmanian localities taken from McDowall and Frankenbera

(1981). Insert shows locality on Wilsons Promontory.

tinental shelf, Tasmania has been part of the Australian
land mass throughout its history with the last severing of
the land bridge apparently taking place about 12 000 to
13 500 years ago (Galloway & Kemp 1981). Oppor¬
tunities have existed several limes in the past for disper¬
sal of fauna via confluent streams (Walker 1981) and
McDowall (1981) has pointed out that the freshwater
fishes of southern Victoria arc nearly all represented by
conspecific populations in Tasmania. Andrews (1976)
believed that the present distribution of the Australian
Galaxiidae could be explained by the hypothesis that
species groups were once widely distributed over
southeastern Australia and became fragmented by the
formation of Bass Strait. He expressed surprise at the
apparent absence of G. cleaveri from the Australian
mainland and from the Bass Strait islands, particularly
since the species appears to be tolerant of brackish
water. The lack of any records of G. cleaveri from Vic¬

toria may, at least in part, be explained by the fact that
low-lying coastal swamps have seldom been adequately
sampled, often because they are relatively inaccessible.
Additional sampling may extend the range of G. cleaveri
in Victoria.

Galaxias cleaveri has undoubtedly been
fragmented in its range in Tasmania by the draining and
clearing of swamps (Frankenberg 1974). In New
Zealand, a similarly adapted galaxiid, Neochanna bur -
rowsius (Phillipps), is seriously threatened with extinc¬
tion for the same reason (Cadwallader 1975). Indeed,
the Nature Conservation Council of New Zealand has
decided to create reserves for the preservation of N. bur -
rowsius (Eldon 1979). The occurrence of a population of
G. cleaveri on Wilsons Promontory may thus be for¬
tuitous as the site is within the bounds of a National
Park, where the habitat can be protected.

TASMANIAN MUDFISH ON WILSONS PROMONTORY

51

ACKNOWLEDGEMENTS

Wc thank the Victorian National Parks Associa¬
tion for permission to sample in Wilsons Promontory
National Park, and the staff of the Park, particularly
Peter Thomas, for their assistance with the sampling.
Thanks to Martin Gomon (National Museum of Vic¬
toria) who prepared the X-rays of the specimens and
reviewed the manuscript. Thanks also to Phillip An¬
drews (Tasmanian Museum) and Bob McDowall
(Fisheries Research Division, New Zealand) for critically
reading the manuscript.

REFERENCES

Andrews, A. P., 1976. A revision of the family Galaxiidae
(Pisces) in Tasmania. Aust. J. mar. Freshwater Res.
27: 297-349.

Cadwallader, P. L., 1975. Distribution and ecology of the
Canterbury mudfish, Neochan na burro wsius
(Phillipps) (Salmoniformes: Galaxiidae), J. Roy. Soc.
N.Z. 5: 21-30.

Eldon, G. A., 1979. Habitat and interspecific relationships of
the Canterbury mudfish, Neoehanna burrowsius
(Salmoniformes: Galaxiidae), N.Z. .//. mar. freshw
Res. 13: 111-119.

I'rankenberg, R., 1974. Native freshwater fish. In Bio¬
geography and Ecology in Tasmania , W. D. Williams,
ed., W. Junk, The Hague, 113-140.

Galloway, R. W. & Kemp, E. M., 1981. Late Cainozoic en¬
vironments in Australia. In Ecological Biogeography
of Australia, A. Keast, ed., W. Junk, The Hague,
53-80.

McDowall, R. M., 1981. The relationships of Australian
freshwater fishes. In Ecological Biogeography of
Australia , A. Keast, ed., W. Junk, The Hague,
1253-1273.

McDowall, R. M. & Frankenberg, R. S., 1981. The galaxiid
fishes of Australia. Rec. Aust. Mus . 33: 443-605.

Scott, E. O. G., 1934. Observations of some Tasmanian
fishes, with descriptions of new species. Pap. Proc. R.
Soc. Tasm. 1933: 31-53.

Scott, E. O. G., 1936. Observations on fishes of the family
Galaxiidae. Part 1. Pap. Proc. R. Soc. Tasm. 1935:
113-129.

Table 1

Biometric Data from Ten Specimens of Galaxias cleaveri from Wilsons Promontory and
Thirty Specimens from Tasmania (Tasmanian Data from Andrews 1976)

A Morphometric data

Body proportions Wilsons Promontory Tasmania

Mean

s.d.

Min.

Max

As a percentage of standard length:

Head length

21.7

1.3

16.7

20.8

Snout lip to dorsal fin origin

71.7

1.0

65.1

77.0

Snout tip to ventral fin origin

53.1

1.8

49.7

54.7

Dorsal length of caudal peduncle

14.0

1.7

12.1

19.3

As percentage of head length:

Eye diameter

15.2

2.2

10.4

17.7

Upper jaw length

28.1

1.6

20.8

31.4

Lower jaw length

26.5

2.1

22.7

32.0

Gape width

34.6

4.0

30.3

44.6

Interorbital width

43.3

1.2

33.9

43.0

Pectoral fin length/pectoral fin base to ventral
origin

39.5

5.1

24.6

42.9

Ventral fin length/ventral fin base to anal origin

43.3

2.4

26.9

46.2

Minimum length of caudal peduncle/dorsal
length of caudal peduncle

57.5

9.8

41.8

70.1

B Meristic data

Variation (Number of fish in parenthesis)

Wilsons Promontory

Tasmania

Dorsal rays

12(7), 13(3)

11(10), 12(15), 13 (5)

Anal rays

13(7), 14(3)

12 (3), 13(13), 14(12), 15(2)

Pectoral rays

13(6), 14(4)

12 (1), 13(19), 14 (9), 15(1)

Ventral rays

6(4), 7(6)

5 (1), 6 (5), 7(21)

Vertebrae

56(2), 57(2)

56 (2), 57 (4), 58(4), 59(1)

52

P. D. JACKSON AND J. N. DAVIES

Scott, E. O. G., 1942. Description of Tasmanian mud trout,
Galaxias ( Galaxias ) upcheri sp. nov.: with a note on the
genus Brachygalaxias Eigenmann, 1924, and its ocur-
rence in Australia. Rec. Queen Via. Mus. 1: 51-57.

Walker, K. F., 1981. The distribution of freshwater mussels
(Mollusca: Pelecypoda) in the Australian zoo
geographic region. In Ecological Biogeography of
Australia , A. Keast, ed., W. Junk, The Hague,
1231-1249.

PROCEEDINGS

OF THE

ROYAL SOCIETY OF VICTORIA

Volume 94

NUMBER 2

ROYAL SOCIETY’S HALL
9 VICTORIA STREET, MELBOURNE 3000

Contents of Volume 94 Number 2

Article Page

6 Seals in Tasmanian Prehistory.By Jim Stockton 53

7 Wind-induced movements of beach sand at Portsea, Victoria.By C. L. So 61

8 Wetlands of Victoria III. Wetlands and waterbirds of the Western Districts from

Port Phillip Bay to Mount Emu Creek.By A. H. Corrick 69

9 Studies on Australian Mangrove Algae II. Composition and geographic distribu¬
tion of communities in Spencer Gulf, South Australia.By W. R. Beanland &

Wm. J. Woelkerling 89

PROC. R. SOC. VICT. vol. 94, no. 2, 53-60, June 1982

SEALS IN TASMANIAN PREHISTORY

By Jim Stockton

Department of Prehistory, Research School of Pacific Studies, Australian National University, P.O.

Box 4, Canberra, ACT 2600

Abstract: Aboriginal sealing began over 8000 years ago and continued up to the early
part of the last century. Species exploited included Arctocephalus pusillus doriferus, A.
forsteri, Mirounga leonina and Hydruga lepionyx. Historical information, derived from
Warneke (1976), on the locations of seal colonies at the time of the arrival of European ex¬
plorers is plotted as a series of maps. Archaeological data on the antiquity and distribution of
seal remains are presented. Aboriginal methods of sealing are reviewed, and their possible
effects on the prey species considered. It is concluded that there are insufficient data available
at present to suggest a change in seal numbers through time, in spite of the long duration of
the predalor/prey relationship.

In prehistoric midden sites in coastal north¬
western Tasmania the numerous remains of seals attest
to the importance of seals as a meat source to the
Aborigines. In this paper I have attempted to synthesize
all the available ethnographic information on
Aboriginal sealing, and to look at the antiquity of seal
remains in Tasmania.

The data for this paper fall into three broad
classes, biological, historical, and archaeological.

SYNOPSIS OF RELEVANT BIOLOGICAL DATA

Biological data relevant to the prehistoric use of
seals was taken from a variety of published and un¬
published sources (Gaskin 1972, Horton 1978, Hyett &
Shaw 1980, Thoughton 1965, Warneke 1975, 1976 and
pers. comm., Wood Jones 1925). In the period when the
colonists first entered Tasmanian waters, there appear to
have been four common species of seal around
Tasmania and in Bass Strait, consisting of Southern
Elephant Seal, Australian Sea Lion, Australian Fur
Seal, and New Zealand Fur Seal. The Leopard Seal is
and was an occasional visitor, but was never abundant.
Although Wood Jones (1925) referred to Australian Sea
Lion remains having been found in Aboriginal middens
in Tasmania, subsequently they have not in fact been
identified in excavations and will not be dealt with in this
section. It is quite possible that Australian Sea Lion re¬
mains will be found in future excavations.

Southern Elephant Seal, Mirounga leonina

At the start of the Nineteenth Century, large col¬
onies of the Southern Elephant Seal occurred on the
Hunter Islands, New Year Island, and on King Island
(Micco 1971) (Fig. 1), but today the nearest breeding
grounds are on Macquarie Island (Green 1973).
Reproductive behaviour of this species has been sum¬
marised by Warneke (1976). Breeding sites are usually
sandy beaches. Mature males are ashore from early
August to early December while mature and pregnant
cows start arriving in September. Groups of females
form harems with large male ‘beachmasters’. On average

a female is on the breeding site for 28 days during which
she does not eat. Lactation lasts about 23 days (Hyett &
Shaw 1980). During this time pups gain about 5 kg in
weight per day (Warneke 1976). The male Southern
Elephant Seal is the largest seal occuring in Tasmanian
waters. At maturity it may attain lengths of 6 m and
weights of 3 600 kg. Females weigh about 900 kg at
maturity and attain lengths of 3-4 m.

The Southern Elephant Seal also spends a period
ashore during December, January and February when
adults of both sexes haul out on suitable beaches to
moult. This process take about 20 to 40 days, during
which time the animals cease feeding and rarely go near
the sea unless disturbed. When the moult is complete the
animals return to the sea to feed, and large concentra¬
tions of Elephant Seals are not seen ashore until the
following breeding season. However breeding sites may
not be completely abandoned as there are usually some
resting young present all year round (Hyett & Shaw
1980).

Australian Fur Seal, Arctocephalus pusillus doriferus
and New Zealand Fur Seal, Arctocephalus forsteri

As the Arctocephalus remains in middens cannot
usually be identified to species, the life history of both
species will be considered together.

Fur seals were recorded in early accounts at
numerous locations throughout Bass Strait. The highest
concentrations were around King Island and the
Furneaux Group (Fig. 2). The Australian Fur Seal is the
only seal which is still a regular breeding species in Bass
Strait (Green 1973). The distribution map of modern
seals is based on Pearse (1979) (Fig. 3). The figure is
somewhat misleading as the number of sites cannot be
directly equated with abundance; some of the locations
had very few individuals. Although there is little detail
of the historic colonies of seals in the southeast and
south, they were present.

Australian Fur Seal males grow to about 2.25 m
long with weights of up to 360 kg, while females grow to
around 1.5 m and 90 kg (Hyett & Shaw 1980). New

53

54

J. STOCKTON

New
Year I.

Fig. 1 — Historic locations of Mirounga leonina colonies (source: Warneke 1976). Scale and north marker

as for Figure 2.

Zealand Fur Seal males grow up to 2 m and 110 kg,
while females are up to 1.6 m and 80 kg (Hyett & Shaw
1980). Breeding locations, usually exposed rocky
beaches, are occupied by males from October. Most
females give birth in November (Australian Fur Seal)
and January (New Zealand Fur Seal). Pups weigh
around 4-5 kg. Australian Fur Seal pups congregate in
groups of up to 50 while their mothers are away feeding.

Observations on the vulnerability to capture of
Australian Fur Seal have been made (R. M. Warneke
pers. comm.). Young pups less than eight weeks of age
are almost helpless on land. If harassed they quickly
take refuge in the sea, but soon tire and return to shore.
If the hunter lay down and remained still the pups would
approach quite closely. By three months of age they are
strong active swimmers and are much more difficult to
catch by pursuit; they are quite curious however, and
will approach a stationary crouching person. Around
20% of pups die of natural causes in the first three
months (Horton 1978).

The reconstructions of the distribution of
Southern Elephant Seal and of fur seals (Figs 1, 2) are
based on historic records (Warneke 1976). In spite of the
limitations of the historical records, the maps show the
highest density for both groups is on the inaccessible
islands of Bass Strait.

Leopard Seal, Hydruga leptonyx

The Leopard Seal ranges widely through the
Southern Ocean but is a rare visitor to Tasmania. The
species is migratory, with both immature and adult seals
moving northwards in the autumn and winter. The
younger animals tend to wander further afield than
reproductively mature adults. Fully grown a female may
weigh 450 kg and attain a length of 4 m. Males weigh
about 270 kg and attain lengths of about 3 m at
maturity. On land Leopard Seals are quite agile (Gaskin
1972).

HISTORIC ACCOUNTS OF ABORIGINAL
SEALING

Information on the hunting and use of seals is
contained in the journals of Robinson (Plomley 1966)
and Kelly (1920). There are two types of Aboriginal
weapon for hunting seals described in the literature, a
club and a spear. In his account of a voyage around
Tasmania in 1815-1816, Kelly (1920) describes
Aboriginal women stalking seals at George Rocks. This
fascinating account describes Aboriginal women sneak¬
ing up on seals by imitating seal behaviour. After first
washing their bodies to lessen the chance of the seal
catching their scent, they approached slowly from the
downwind side to the rocks where the seals lay sleeping.
When observed by a seal the women would imitate the
seals’ movements by raising their heads, looking around,
and scratching. The six women stayed on the rocks with
the seals (where they were washed by the occasional
wave) for nearly an hour before making their attack.
Each then clubbed two seals. By clubbing a seal on the
nose it can be stunned into unconsciousness, and then
dispatched. Over the next few days the women con¬
tinued to catch seals from the rocks, even though the
animals were becoming increasingly shy. When Kelly’s
party and Aboriginal helpers returned to the mainland,
the men danced around a pile of seal carcasses on the
beach, sticking their spears into the dead seals as if kill¬
ing them (Kelly 1920). There is an interesting problem in
this northeast region in that the people did not have
watercraft (Jones 1976). All the seal colonics on the
islands of the Furneaux Group and George Rocks were
inaccessible to them until the arrival of the colonists and
their boats. Some of these islands are as little as 3 km
offshore.

The reference to killing seals by spearing is con¬
tained in a hearsay account in Robinson’s journal where
Robinson’s native companion WOORRADY describes
sealing at Cox Bight by the Bruny Island and southwest

SEALS IN TASMANIAN PREHISTORY

55

A Colony of seals which were probably Arctocephalus spp.

Fig. 2 —Historic locations of Arctocephalus spp. colonies (source: Warneke 1976)

tribes who “went off in catamarans to the De Witt
Islands and to the difTerent rocks, and speared seal and
brought them back to the mainland. Also went to the
Eddystone and speared seal.” (Plomley 1966). On one
occasion Robinson observed a large seal at Cox Bight
which the natives killed, but he does not say how it was
dispatched (Plomley 1966). However, he does say that it
was cut into Hitches which were then carried to the
camp. If butchering at the kill site was a common prac¬

tice for large seals, then their bone remains will be under
represented in middens, having been left behind at the
site of the kill. Vanderwal (1978) has excavated seal re¬
mains in a midden on Maatsuyker Island and interprets
the bulk of archaeological debris to suggest local con¬
sumption rather than a return to the mainland with the
meat, but the reasons for this statement are not given.
The seal bones recovered by Jones (1966) at West Point
are all from young animals and bones from the flippers

56

J. STOCKTON

Fig. 3 —Modern observations of fur seals (source: Pearse 1979)

and skull over represented (R. Jones pers. comm.).

Also of interest is Robinson’s mention of
Aborigines catching a Leopard Seal at Sandy Cape
(Plomley 1966) and an unsuccessful attempt to catch a
seal at Hunter Island (Plomley 1966). The Hunter Island
seal he calls a “Leopard Oil Seal from its being spotted
^ like a leopard”, which Plomley (1966) interprets as a
Leopard Seal.

Seal meat was a food which the Tasmanians were
very fond of (Plomley 1966) but hunting seals was a high
risk activity. Robinson records in his diary that WOOR-
RADY had told him that many hundred natives had
been lost on sealing trips to the De Witt group and
Eddystone Rock (Plomley 1966). Although this is pro¬
bably exaggerated, access to these islands does involve
water crossings of 5 km and 27 km respectively. Further
confounding this claim, a voyage of 27 km should have
been beyond the capacity of Tasmanian watercraft
(Jones 1976). Steep Island and Albatross Island in the
northwest involve water crossings of 3.2 km and 11.3
km, and neither appear to have been visited by
Aborigines in the recent prehistoric period (Bowdler
1974, R. Jones per. comm.). The reference to Ed¬
dystone Rock may be a mistake in Robinson’s transla¬
tion of the location. On another occasion TRUCANINI
told Robinson that the seals fought with the men at Cox
Bight, and that once “a seal caught a black man by the
thigh and they both fell over the rocks into the water
and the seal carried the man down. He however came up
but was ever after lame” (Plomley 1966). WOORRADY
relates at least four attacks by seals in the Cox Bight area
(Plomley 1966). The risks of seal hunting fall into three
categories. Seals occasionally attacked the hunters.

site name
Little Duck Bay
Louisa Creek 1
Louisa River 1
Louisa River 3
Louisa River Cave
Maatsuyker Island
Rocky Cape, North Cave
Rocky Cape, South Cave
Sisters Creek
Stockyard site
Sundown Point
Thornton River
West Point (118)

West Point (77)
Venables Corner

modern 2000 4000 6000 8000

years before present

Fig. 4- Dates of seal remains in Tasmanian archaeological sites. The dotted line for Sisters Creek is based
on stratigraphic correlation with Rocky Cape.

SEALS IN TASMANIAN PREHISTORY

57

Long distance water crossings were dangerous given the
inherent structural limitations of Tasmanian watercraft
(Jones 1976). There was a risk of injury while landing on
wave swept rocky seal islands.

ARCHAEOLOGY OF SEAL REMAINS

How long have seals formed part of the diet of
the coastal people of Tasmania? 1 have plotted the C14
dates ± two standard deviations for all the sites where
seal remains occur throughout the deposit (Fig. 4). The
Rocky Cape sequence demonstrates some 8 000 years of
continuous exploitation of seal. It was about this time
that the rising post-glacial sea level reached the vicinity
of the Cape and shows that seals (and seal hunters) came
with the encroaching sea. The next dated appearance of
seal bone is at Sisters Creek at around 6 000 BP. No top
date is available for the Sisters Creek site, but on
stratigraphic correlation with Rocky Cape its upper
layers (which do not contain fish bones) are taken to be
less than 3 500 years old (R. Jones pers. comm.). The re¬
maining dates show the occurrence of seal remains from
around 3 000 BP to the present for west and southwest
sites. The figure shows a pattern of long-term exploita¬
tion of seals in coastal sites up to the contact period, but
the pattern is not quite that simple. In the northwest,
seal bones were not recorded in the excavations of the
midden layers of Cave Bay Cave on Hunter Island
(Bowdler 1979). Similarly in the east and southeast at
Little Swanport (Lourandos 1970), Alum Cliffs

1980) seal bones were not found. That so many sites do
not contain seal bones cannot be dismissed as a problem
of excavation sample size. This is especially true of the
large excavations of the midden layers in Cave Bay
Cave, which is geographically close to the northwest and
covers a similar time span to sites which contain seal
bones. Why do seals appear to be abundant in some
locations, but absent from others?

DISCUSSION

The distribution of archaeological seal remains
was compiled from published and unpublished sources.
The majority of the observations are of fragmentary re¬
mains noted during field reconnaissance. Under these
conditions it is easy to distinguish seal bone from other
mammals, but difficult to identify species. A number of
people contributed records of their observations of seal
remains in middens in response to a letter circulated in
1980. Subsequently, it was possible to check many of the
observations from northwest Tasmania, but records
from other areas were not checked. Identifications to
generic level for Arctocephalus spp. and specific level
for other seals were available for remains from ex¬
cavated sites (Jones 1966, 1971, Bowdler 1979, Vander-
wal 1978). The distribution of archaeological sites with
prehistoric seal remains shows a marked concentration
on the northwest coast (Figs 5-8), from Sandy Cape to
Cape Grim, with a lower frequency along the remaining

58

J. STOCKTON

coast. Of lowest frequency are the Bass Strait coast east
of Port Sorell, and Tasman Peninsula and Bruny Island.
The patterns for Arctocephalus spp. (Fig. 6) and
Mirounga (Fig. 7) remains are similar. The remains are
found on the northwest, west and southwest coast.
These patterns reflect the locations of analysed excava¬
tions where accurate generic identifications are
available, and so the absence of records from other
areas may bias the picture. This could be tested by
sampling sites in other areas for seal remains. This could
be done quite easily by collecting material from deflated
sites around the coast. Occurrence of the Leopard Seal
(Fig. 8) is less common than the other species. The
species was recorded ethnographically, and an
Aboriginal word, "TOPER”, has been recorded for it
(Plomley 1976). The present day distribution of fur seal
is in contrast to the archaeological and contact period
distributions. The modern pattern is strongly centered
on the south and southeast coast. Several hypotheses
can be advanced in an attempt to explain this.

Did Aborigines exterminate or drive away the
seals of the coastal breeding populations before Euro¬
pean arrival, as Jones (1966) speculated for the Southern
Elephant Seal at West Point. There is no evidence that
seals breed on “mainland” coasts, so it is extremely
unlikely that Mirounga was breeding on the Tasmanian
coast at any time. Secondly modern Cl4 dates have been
obtained for sites containing Southern Elephant Seal at
Sundown Point (Ranson pers. comm.) and at Venables
Corner. Both of these sites also contain fur seal.

Did Aboriginal activity reach just the fringe of
the range of seals? Were the main seal areas on inac¬
cessible islands in Bass Strait or the far south? Was it
just the stray wandering seal which came ashore on the
mainland and fell prey to the Aborigines? Because of the
extremely high density of seal remains along the north¬
west coast, I have plotted the distribution for the region
at a larger scale (Fig. 9). The high density of site infor¬
mation here is partly due to the fact that the coast from
Cape Grim to Macquarie Harbour has been surveyed
more frequently and more intensively than any other
part of Tasmania. Also, shell middens in the northwest
are more numerous and larger than elsewhere in
Tasmania. Consequently, in this well studied area we
find an almost continuous occurrence of seal remains.
This suggests a resident population of seals, or regular
visitors, all along the coast. At West Point, Jones’(1966)
tentative minimum numbers estimates for seals in the
whole site is several thousand individuals. Supplemen¬
ting this, analysis of the Southern Elephant Seal canine
teeth has shown that all specimens were young animals,
with some teeth belonging to seals less than three mon¬
ths old. All this evidence suggests a large breeding col¬
ony of Southern Elephant Seals in the immediate
vicinity —but where was it located? There are no
offshore islands or rocks any distance from the coast,
but there are a few little islands less than 100 m away.
Some of these have shell midden on their surface, and in
1980 two were rookeries for the Little Penguin, Eudyp -

SEALS IN TASMANIAN PREHISTORY

59

tula minor. The simplest explanation of the evidence is
to put breeding Southern Elephant Seal colonies on the
shore of the Tasmanian mainland, or on these islands
just off shore, where they are quite accessible to
Aborigines. In fact, given the age structure of the seals
found at West Point, is it possible the Aborigines arc
just scavenging dead juveniles? (Horton 1978). On the
other hand, there is historical evidence for a concentra¬
tion of Mirounga to the northwest of Tasmania on King
Island, New Year Islands and Hunter Island (Peron in
Micco 1971). The prevailing winds are southwesterly i.e.
onshore. With significant seal activity in the offshore
waters it is inevitable that some freshly dead, sick or
tired individuals are going to end up on the coast. The
presence of the remains of juveniles of less than three
months of age at West Point is not surprising as the pups
are weaned at about one month (Warneke 1975). This
evidence does not imply that there was a breeding colony
at West Point. The same argument also applies to Arc -
tocephalus. During a seven year period Warneke

from the Seal Rocks colony of A. pusillus doriferus.
This is just 2% of the total sample tagged in the same
period. The sealer Kelly estimated that the Albatross
Island colony off Hunter Island had a population in the
order of 12 000 (Plomley 1966). Reid Rocks further to
the northwest probably had more than 2 000 animals
and the many colonies around King Island possibly ex¬
ceeded 50 000 (R. M. Warneke pers. comm.). With
these population numbers, beach scavenging could ac¬
count for all the fur seal remains in northwest Tasma¬
nian coastal middens (R. M. Warneke pers. comm.). No
reference to historic sealing on the northwest coast of
Tasmania has been found (Warneke 1976).
CONCLUSIONS

Quantitative studies of seal remains in large sites
may give clues to the times of the year when seals were
killed. With an idea of what ages and sexes were killed it
should be possible to describe the population structure
of the seal groups that the Aborigines preferred to prey
upon. Alternatively, it may suggest that the Aborigines
were collecting dead animals. Work along these lines us¬
ing growth rings on teeth is presently being conducted by
Horton (1980) on material from Maatsuyker Island and
O’Connor (1980) for West Point. Since the various
species vary so widely in size and habits, any reconstruc¬
tion concerned with seasonal availability or significance
of seals in the diet will need to be species specific.

The distribution maps for seal remains indicate
few records of seal bone in the northeast and southeast.
Whether this is a reflection of the true distribution or an
artefact of a lack of archaeological w'ork in the region
could be tested by systematic site survey and sampling.

Fig. 9 —Distribution of seal remains in northwest Tasmania.

Location and observer of numbered sites:

I, Curvier Point —pers. obs. 2, Little Duck Bay — Bowdler
(1979). 3, Suicide Bay-pers. obs. 4, Valley Bay —Jackson
(pers. comm.), pers. obs. 5, Studland Bay — Lourandos
(1970).6, Calm Bay —Jackson (pers. comm.). 7, Maxies
Point —Lourandos (1970), Jackson (pers. comm.), 8, Green
Point-Lourandos (1970), Jackson (pers. comm.), pers. obs.
9, West Point (Nettley Bay —Jones (1966), pers. obs. 10, West
Point (south) —Jones (1966), Jackson (pers. comm,), pers. obs.

II, Bluff Hill Point —Jackson (pers. comm.), pers. obs. 12,
Sundown Point-Ransom (pers. comm.), Jackson (pers.
comm.), pers. obs. 13, Nelson Bay —pers. obs. 14, Rebecca —
Jackson (pers. comm.). 15, Gannct Point Hazard Bay —
Jackson (pers. comm.). 16, Ordinance Point —Pullcine (1929),
Jackson (pers. comm.), pers obs. 17, Thornton River —pers.
obs. 18, Daisy River—pers. obs. 19, Wild Wave River —pers.
obs. 20, Sandy Cape Beach —pers. obs. 21, Sandy
Cape/Venables Corner—Pullcine (1929), Legge (1929),
Jackson (pers. comm.), Ranson (pers. comm.), pers. obs. 22,
Johnson River —Jackson (pers. comm.). 23, Rupe ?oint —
Jackson (pers. comm.). 24, Conical Rocks —Jackson (pers.
comm.). 25, Ahrberg Bay-Jackson (pers. comm.). 26, Trial
Harbour —Cane (pers. comm.). 27, Sloop Rocks —Jackson
(pers. comm.). 28, Ranger Point —pers. obs. 29, Burgess
Point —pers. obs. 30, Homestead Lagoon (Stockyard
site) —Bowdler (1979). 31, North Point —pers. obs. 32, Rocky
Cape—Jones (1966), pers. obs. 33, Lee Archer —Jones (pers.
comm.). 34, Sisters Creek-Jones (J966).

60

J. STOCKTON

The large numbers of seals in the Furneaux Group and
Maria Island at contact suggests that seal remains should
be common in middens on the adjacent coasts.

The overall impression of the combined ar¬
chaeological and ethnographic data is a picture of long
term exploitation of seals by Aborigines. An early
speculation by Jones (1966) that the Aborigines had in
prehistoric times caused the local abandonment of the
Tasmanian mainland coast by the Southern Elephant
Seal has not been resolved. Jones proposed the idea of
C14 dating the latest phase of prehistoric Aboriginal
sealing to see if the species continued to be found up to
the contact period. This has now been done by Bowdler,
Ranson, Vanderwal and the author, who have shown
that the species is present up to the contact period.
However, we still do not know if the seals were living on
the coast in colonies, or only coming ashore occasion¬
ally. The presence of juvenile Mirounga teeth (less than
three months old) at West Point does not resolve the
problem as weaning occurs at about one month of age,
after which the juveniles will be travelling widely in
order to feed. There is insufficient evidence to suggest a
decline or increase in seal numbers, in spite of at least
8 000 years of continuing predator/prey relationship.
This relationship was to alter drastically when the arrival
of European technology made it possible to exploit
previously inaccessible colonies in Bass Strait.

ACKNOWLEDGEMENTS

The distribution maps which forced me to think
about the problems outlined here were compiled from
contributions by Scott Cane, David Horton, W. D. and
L. Jackson, Rhys Jones, Don Ranson, Ron Vanderwal
and the Tasmanian Aborignal Sites Index. I am deeply
indebted to their willingness to share information. For
comments on early drafts of this paper I am indebted to
R. H. Green, John Luly, R. M. Warneke, Jeanette
Hope, Rhys Jones and Phil Hughes. R. M. Warneke
generously gave access to unpublished material, and
discussed many of the difficult questions raised in this
paper.

REFERENCES

Bowdler, S., 1974. An account of an archaeological recon¬
naissance of Hunter’s Isles, north-west Tasmania,
1973/4. Rec. Queen Via. Mus . 54: 1-22.

Bowdler, S., 1979. Hunter Hill. Hunter Island. Unpublished
PhD thesis, Australian National University.

Green, R. I I., 1973. The Mammals of Tasmania. The Author,
Launceston.

Gaffney, L. & Stockton, J,, 1980. Results of the Jordan
River Midden Excavation. Australian Archaeology 10:
68-78.

Gaskin, D. E., 1972. Wales, Dolphins and Seals. Heinemann.

Healey, L. & Stockton, J., 1980. Problems and potentials of
archaeological evidence for prehistoric biophysical

description in the Derwent estuary. The Artefact 5:
145-154.

Horton, D. R., 1978. Preliminary notes on the analysis of
Australian coastal middens. Australian Institute of
Aboriginal Studies Newsletter 10: 30-33.

Horton, D. R., 1980. Age determination in Australian Fur
Seals based on canine teeth. Bull. Aust. Mammal Soc.
6: 41.

Hyett, J. & Shaw. N., 1980. Australian Mammals. Nelson,
Australia.

Jones, R., 1966. A speculative archaeological sequence for
northwest Tasmania. Rec. Queen Viet. Mus. 25: 12.

Jones, R., 1971. Rocky Cape and the problem of the Tas¬
manians. Unpublished PhD thesis, ANU.

Jones, R., 1976. Tasmanian aquatic machines and off-shore
islands. In Problems in Economic and Social Ar¬
chaeology , G. de Ci. Sicveking. I. H. Longworth & K.
E. Wilson, eds, Duckworth, London, 235-263.

Kelly, J. 1920. First discovery of Port Davey and Macquarie
Harbour, by James Kelly Pap. Proc. R. Soc. Tasm.
1920: 160-181.

Legge, R. W., 1929. Tasmanian Aboriginal Middens of the
West Coast. Australasian Association for the Advance¬
ment of Science 19: 323-328.

Mtcco, H. M., 1971. King Island and the Sealing Trade , 1802.
Roebuck, Canberra.

Lourandos, H., 1970. Coast and hinterland: the ar¬
chaeological sites of eastern Tasmania. Unpublished
MA thesis, ANU.

O’Connor, S., 1980. Bringing it all back home: an analysis of
the vertebrate faunal remains from the Stockyard Site,
Hunter Island, north-west Tasmania. Unpublished
B.A. Honours thesis, University of New England.

Pearse, R. J., 1979. Distribution and conservation of the
Australian fur seal in Tasmania. Viet. Naturalist 96:
48-53.

Plomley, N. J. B. (ed.), 1966. Friendly Mission. The Tasma¬
nian Journals and Papers of George Augustus Robin¬
son 1829-1834. Tasmanian Historical Research
Association, Hobart.

Plomley, N. J. B., 1976. A Word-list of the Tasmanian
Aboriginal Languages. The author in association with
the Government of Tasmania.

Pulleine, R., 1929. The Tasmanians and their stone culture.
Australasian Association for the Advancement of
Science 19: 296-314.

Stockton, J., 1978. Archaeological investigation of the Der¬
went River Estuary, south-east Tasmania: Alum Cliffs
test excavation interim report. The Tasmanian
Naturalist 53: 8-9.

Thoughton, E.. 1965. Furred Animals of Australia. Angus
and Robertson, Sydney.

Vanderwal, R. L., 1977. The Shag Bay Rockshelter,
Tasmania. The Artefact 2: 161-170.

Vanderwal, R. L., 1978. Adaptive technology in south west
Tasmania. Australian Archaeology 8: 107-127.

Warneke, R. M., 1975. Dispersal and mortality of juvenile fur
seals Arctocephalus pusillus doriferus in Bass Strait,
Southeast Australia. Rapp. R.-v. Reun. Cons int. Ex-
plor. Mer 169: 296-302.

Warneke, R. M. 1976. Preliminary report on the distribution
and abundance of seals in the Australian region. Scien¬
tific Consultation on Marine Mammals, Food and
Agriculture Organisation of the United Nations,
Bergen, Norway.

Wood Jones, F., 1925. Mammals of South Australia Part 3,
The Monodelphia. Government Printer, Adelaide.

PROC. R. SOC. VICT. vol. 94, no. 2, 61-68, June 1982

WIND-INDUCED MOVEMENTS OF BEACH SAND AT
PORTSEA, VICTORIA

By C. L. So.

Department of Geography and Geology, University of Hong Kong, Hong Kong

Abstract: An attempt is made to monitor beach changes associated with wind-induced
movements of beach sand at Portsea, Victoria. This reveals that given adequate wind strength and at¬
mospheric aridity, a process of sand-layer levelling, arising from alternate drying of sand and activation
of sand movement by wind, takes place within a favourable time span of the tidal cycle. The moving sand
ending up in sinks may constitute a permanent loss unless intra-compartmcntal and inter-compartmental
transfers of the beach sand are made good by feedbacks from the dune-dominated coastal hinterland and
the swell-dominated offshore subsystem. Where recovery is not complete, the cumulative, though
diminished, net loss ol beach sand plays a significant role in both the spatial and the temporal distribu¬
tions of sand on the beach.

Beach changes associated with wind-induced
movements of beach sand have not been given the atten¬
tion they deserve. Well-based observations are confined
to coastal dunes and dune topography. Mobility of
beach levels of aeolian origin is considered to lag
behind. However, recognition of transport surfaces on
the foreshore as sites of sand mobilization without
significant change of sand level renders this assertion
untenable. Moreover, relevance of sand movements by
wind to beach sediment budget further calls for their in¬
vestigation. The beach at Portsea, with ample catchment
for sand and a multi-directional wind system, provides
an adequate base for such study.

THE COASTAL ENVIRONMENT

The ocean coast of Portsea between Mt. Levy
and Sphinx Rock (Fig. 1, Inset A), formed since the
Post-glacial rise ol sea level (Keble 1950), encompasses a
sandy beach abutting a series of dune-capped cliffs and
shore platforms cut in Pleistocene dune calcarenite (Bird
1975, 1977). Forming the coastal hinterland is an un¬
dulating terrain 30-40 m high (Fig. 1, Inset B). Its overly¬
ing Holocene dunes (Bird 1972) rest on a basement of
calcarenite formed by the consolidation and cementa¬
tion of older dunes (Bowler 1966). Deflation, periodical¬
ly affecting the dune surface, which is now partly under
scrub or grass, initiates sand migration and blowout
development. The lack of spatial pattern in the asym¬
metry of some dune profiles points to less well-defined
sand drift. Absence of surface drainage testifies to
permeability of the dune sand and percolation of rain¬
water into the ground. This enhances backshore sand
movement.

Part of the beach compartment (Fig. 1) has been
chosen for monitoring wind action on sand. The beach
material is calcareous sand derived from shelly organ¬
isms thriving in coastal waters to the south, and from
eroded calcarenite cliffs and overlying dunes. Except
where a berm develops, beach profiles rise landward at
angles of 10-12° but occasionally flatten off to 6-8°.
Where a hardened calcarenite layer persists seaward

beyond the beach sand, remnants of dissected intertidal
platforms and reefs are exposed at low water.

The coast is characterised by a microtidal swell
environment (Davies 1972). Northerly and north¬
easterly winds are prevalent in autumn, winter and
spring, while southerly and south-westerly winds are
prevalent in summer. Westerly winds may blow
throughout the year but easterly winds are weak and in¬
frequent. Winds strengthen with passage of cyclonic
depressions south of the Victorian coastline. Despite a
high frequency of gales in spring and summer (Bowler
1966), the mean monthly velocities are the highest in
winter (Maher &. McRae 1964), being largely those of
northerly winds. Running from west-northwest to east-
southeast, the coast is affected by wave action associated
with onshore southerly winds, a favourable fetch and
offshore depth in excess of 18 m. The more significant
impact of the wind system, in the present context, is its
role in the spatial tranfser of sand particles. Thus, given
sufficiently dry conditions and a widening sand surface
associated with an ebb tide, the southerly winds may
move sand from the foreshore to the backshore which, if
not acting as a sink, may subsequently allow northerly
winds to return particles to the foreshore. Some or all of
these may end up in the offshore zone. The decrease in
rainfall in summer favours such inducement of sand
movement by the southerly winds.

MONITORING SAND MOVEMENTS

Sand-level changes were recorded simultaneously
with wind data. A sector of the beach (Fig. 1), 120 m
long and 40-50 m wide, was chosen that had relatively
uniform internal and external parameters including sand
source, beach face slope, beach materials, offshore
profile, and backshore and dune hinterland configura¬
tion. Fifty-six w'ooden stakes, each 30 cm long and 1.25
cm in diameter, were placed in a network designed by
random-stratified sampling. This design was modified to
account for different zones such as lower intertidal, mid-
intertidal, upper-intertidal, storm-wave and backshore,
and also for variation in beach topography. The posi-

61

62

C. L. SO

0

1

10 20 30

1 1 1

40

1

METRES

—3—

BEACH CONTOUR IN METRES
(DATUM: L. W. M. WILLIAMSTOWN )

TTTf

miiiiii

CLIFF IN CALCARENI

PLATFORM

FENCE

TE

Fig. 1 — Location of study area and shore-zone physiography at Portsea.

tion of the stakes was fixed by compass and tape and
plotted on a base map, produced by levelling and
tacheometry. Readings of sand levels were taken hourly
over a six-hour period spanning where possible the se¬
cond half of an ebb tide and the first half of the follow¬
ing flood. To collect sand moving in saltation or in
creep, plastic sand traps 3.5 cm in diameter and 3.4 cm
deep were used.

INTRA-COMPAR rMENTAL AND INTER¬
COM PARTMENTAL TRANSFERS

Sand movement was found to be associated with
winds above a critical force. A macroclimatic wind (Jen¬
nings 1957) of 46 km/hr with shear velocities up to 19
km/hr failed to initiate sand movement on 15 November
(Table 1), but a macroclimatic wind of 44 km/hr with
shear velocities of 19-22 km/hr started the process on 18

BEACH SAND MOVEMENT AT PORTSEA

63

November despite the drizzle, pointing to threshold
velocities of 19-22 km/hr required to move sand. With
sand transport varying as the cube of excess of velocity
over this threshold (Bagnold 1954), strengthening of
wind offset the effect of rain as seen on 11 November.
Raising the threshold velocity were salt crusts formed on
the sand surface, reported on 25 November, and air
dampness (Belly 1964), locally enhanced by splash and
spray associated with plunging breakers.

Sand movement on 4 November 1977

Sand movement was localised in place and in
time. The greater sand loss on 4 November than on 14
and 18 November (Fig. 2), taking place at low tide, was
associated with strengthening of wind brought about by
a passing, complex, low pressure system accompanied
by a cold front. It took a subsequent drop in ground-
level wind speed to below 19 km/hr, the onset of rain
and increasing splash and spray with a rising tide to ter¬
minate the process. In the course of sand movement,
phases of differentia! cut and fill were involved (Fig. 3)
such that along rows EF and GH sand loss dominating
the first two hours was then increasingly held up or
replaced by gain. Hourly changes of beach levels (Fig. 4)
revealed a change from dominantly falling sand levels,
especially along profiles AB, EF and GH, to rising ones
towards the backshore late in the monitoring period,
pointing to intra-compartmental and inter¬
com part mental transfers of sand. Where the hourly
sand levels merged together, transport surfaces oc¬
curred.

Nine of the ten sand traps operating at noon were
filled by 1.00 p.m. More rapid infilling of those placed
near the high water mark than those located down the
beach pointed to sand migration from the lower-
foreshore source region to the upper-foreshore receptor,
overspill of which then efleeted inter-compartmental
transfer. In size and rounding, the migrating sand in
traps was comparable to those on the foreshore and in
the blowouts, the only variation being its higher percen¬
tage of medium grains indicative of sorting effect. In
respect of pivotability (Shepard & Young 1961) deter¬
mined with a ‘rock-and-roll* shape-sorter (Kuenen 1964),
the sand displayed a better spread of values, and higher
percentages of the more pivotable fractions, than the
foreshore or the blowout sand.

Sand movement on 14 November 1977

Sand movement spanned half a spring flood tide
and part of the following ebb. The reducing wind force,
moist air persisting after rain, and abundant splash and
spray generated by a rough sea cut down sand-level
changes. Part of the upper intertidal zone and adjacent
backshore became a receptor of the migrating sand (Fig.
2B). Elsewhere alternate gain and loss dominated (Fig.
3). The upper part of profile EF changed function from
a transport surface to a receptor with overspill (Fig. 4)
while gain in the upper part of the profile was concurrent
with loss from the lower part, pointing again to intra-
compartmental and inter-compartmental transfers of
sand.

Sand movement on 18 November 1977

Covering the late stage of an ebb tide and almost
entirely its succeeding Hood counterpart, subdued move¬
ment of slightly wet sand took place under a narrow
range in wind speeds and directions (Fig. 2C). This
resulted in small cumulative changes (Fig. 3) and limited
depth of disturbance (Fig. 4) of the foreshore. Profile
EF showed limited change at the seaward end from 1.00
p.m. onwards (Fig. 4), a node of no change of the mid-
tide level, and sand transfer from the lower to the upper
foreshore.

Fig. 2—Isolines of sand loss from Portsea Beach on 4,
14 and 18 November 1977.

64

C. L. SO

Fig. 3-Cumulative graphs showing cut and fill
at selected points on Portsea Beach on 4, 14
and 18 November, 1977.

Sand movement on ii November 1977

To depict patterns of cut and fill, more sand traps
were laid out to form a close network with the stakes us¬
ed. Some cut was experienced (Fig. 5A) despite the fall
of 2-3 mm of rain at Point Lonsdale (Table 1). Data on
fill indicated the amount of sand passing over a

transport surface. Its decrease in value landward from a
peak (Fig. 5B) revealed substantial sand migration from
source regions. Towards the back of the beach where the
fill thinned out within the zero isoline for cut, there was
a gain of sand.

Fig. 4—Hourly sand-level changes on Portsea
Beach on 4, 14 and 18 November, 1977.

BEACH SAND MOVEMENT AT PORTSEA

65

Table 1

Wind and Rain Conditions in the Period of Sand-Movement Monitoring

-r o
£ B

o >
S= o

•5 e

Ground-level wind

Wind and rain observations at

Point Lonsdale

observations at site

(9 km

from site)

Wind

Rain (mm)

Date

Speed

0900

1500

(1977)

Time

Direction

(km/h)

0900

1500

Dircc-

Speed Direc-

Speed

tion

(km/h)

tion

(km/h)

Nov. 4

1200-1500

NW, ESE

13-30

WNW

72

N

44

Nil

0.60

Nov. 11

1100-1330

SSW to sw

20-30

wsw

90

wsw

44

2.00

3.00

Nov. 14

1100-1500

s to SW

22-30

sw

76

wsw

80

Nil

Nil

Nov. 18

1200-1700

S to SSW

19-22

SSW

50

wsw

44

3.00

Nil

Oct. 31

1245-1500

SSW to SW

12-19

WNW

11

SE

30

Nov. 8

1000-1500

S, SE

5-14

ESE

12

W

14

Nil

Nil

Nov. 15

1000-1500

S to SW

15-19

sw

46

WNW

32

Nov. 17

1000-1500

SE to S

6-13

SE

30

E

16

Nov. 22

1200-1400

W to SW

6-9

S

12

WNW

20

Nov. 25

1000-1500

ESE to SE

7-11

E

36

ENE

34

Fig. 5 — Isolines of cut and fill of Portsea Beach
on 11 November 1977.

Sand-layer levelling

Such intra-compartmental and inter-
compartmental transfers of sand reflect deflation by
wind action. This is tantamount to a process of sand-
layer levelling in which removal of dry foreshore sand by
wind, halted by the increasingly moist sand at depth,
renews activity once the sand exposed at that level
becomes sufficiently dry to be entrained. Alternate dry¬
ing of sand and activation of sand movement by wind,
preferably over a wide foreshore by spanning the latter
part of an ebb tide and the first part of its succeeding
flood, are held responsible for some loss of foreshore
sand. Despite atmospheric drying of the upper foreshore
for a relatively long period, the lower foreshore on the
upwind side of an onshore wind witnesses maximum
sand loss. This may be made good during the next flood
tide or sustained during the following ebb. Thus, mainly
by saltation and surface creep, sand moves inland and
onto the coastal hinterland. Unless return of sand is
effected by reversals of wind directions, the resultant
sand deposition, often by accretion and to some extent
by encroachment, is such that to the lower foreshore
sand, the upper foreshore, the backshore and to some
extent the coastal hinterland act as interceptors, recep¬
tors and sinks.

A wind simulation experiment carried out on
samples of foreshore and blowout sand in the laboratory
reveals a threshold in the percentage of sand moved
when wind speed increases from 15 km/h to 19 km/h
with the blowing time kept constant at eight minutes
(Table 2). Further increase in wind speed with substan¬
tial reduction in blowing time increases the percentage
of moving sand.

66

C. L. SO

Table 2

Results of Laboratory Simulation of Wind Action on Sand

Simulated wind action

% of simulated moving fraction

Wind speed

Foreshore

(km/h)

Exposure time

sand

Blowout sand

15

8 min.

2.88

2.59

19

8 min.

44.30

64.48

22

30 sec.

52.28

52.00

26

10 sec.

71.59

65.98

30

5 sec.

79.06

75.13

DISCUSSION

Wind-induced movements of beach sand at Port-
sea have to be viewed in perspective if their impact on
the spatial distribution and redistribution of sand and
on the long-term sand budget of the beach is to be ap¬
praised. The primary movement, intra-compartmental
transfer, is influenced by numerous local factors. Lenses
of moistened or compressed grains resist wind action.
The size of sand particles determines the ease with which
the grains can be entrained. Shell or pebble layers pre¬
sent impose frictional effect on creep and inhibit sand
movement where they survive instabilities and tur¬
bulence they help to create in the wind. Subsurface ir¬
regularities exhumed in the course of sand movement
provide local base levels for deflation. The same is true
of water tables of the beach although Portsea is relative¬
ly free from the impact of a raised water table associated
with excessive seepage or inability of streams to main¬
tain open channels to the sea. Periodic splash and spray
accompanying plunging breakers or strong onshore
winds, and obstacles, hollows or wet surfaces created by
man on the foreshore, curtail sand migration.

While ridges and swales temporarily halt sand
movement or channel it into pathways through wind¬
ward scour and leeside accumulation giving patterns of
cut and fill not readily explained in simple spatial terms,
sand transfer from the lower to the upper foreshore is
largely independent of beach gradient, profile or cur¬
vature. The extent of intra-compartmental transfer of
sand is determined by the net catchment made available
by an expanding or contracting foreshore associated
with ebb or Hood. It also varies with the nature and
direction of winds. Locally, conditions range between
increase in catchment associated with offshore, relatively
dry northerlies in an ebb tide and decrease in catchment
associated with onshore, relatively moist southerlies in a
flood tide. In the former case, transfer of sand from the
upper to the lower foreshore covers a catchment encom¬
passing the backshore. The resultant inter-
compartmental transfer not only returns sand from
backshore receptors to the foreshore compartment, but
also causes much sand to be lost to the offshore zone,
especially with a spring ebb tide. In the latter case,

transfer of sand from the lower to the upper foreshore
by onshore, moist wind lags behind, especially with ^
contracted catchment associated with a flood tide. Be¬
tween these extremes occur many catchments of in¬
tervening sizes representing different combinations of
wind directions besides north and south, varied moisture
content of the wind, and tidal states other than spring
tides.

In such circumstances, sand movement operates
along a number of pathways. With onshore wind of ade-
quate strength in an ebb tide and dry conditions, defla¬
tion brings about surface erosion of the beach but may
be held up by a subsurface shell or pebble layer (Fig. 6).
Whether beach erosion undergoes or by-passes this
negative feedback loop, it reduces the amount of beach
material on the foreshore and the beach gradient. This
permits deposition in the following flood tide should
material be made available for deposition, helping to
replenish the beach materials previously lost. Surface
erosion of the beach through deflation, bringing about
intra-compartmental transfer of sand in the foreshore
subsystem, provides a link to the backshore subsystem
through its supply of sand from an enlarging catchment
to the backshore, thus initiating inter-compartmental
transfer of sand (Fig. 6). Increasing the sand cover of
backshore dunes thickens the loose sand mantle which
in turn encourages sand removal by an onshore wind.
Where a sizable transgressive sand sheet under a strong
onshore wind makes available to the backshore sand in
quantities too large to be trapped and fixed by the
limited vegetation present, conditions are especially com
ducive to spilling of sand inland. Removal of sand by an
onshore wind thus deprives the dune cover of a sand
supply and impoverishes sand deposition on the
backshore, imposing on the backshore in the long run a
limit to sustained deposit ion (Fig. 6). Sand lost by the
backshore, under unidirectional transport characteristic
of wind action, is then either stored up temporarily in a
catchment of the coastal hinterland or permanently lost
to a sink.

The development of events as shown in Fig. 6
depicts the tendency for intra-compartmental transfer of
sand, and backshore deposition arising from inter-

BEACH SAND MOVEMENT AT PORTSEA

67

Fig. 6-Model showing the effect o!' onshore
winds on the foreshore and the baekshore in an
ebb tide.

compartmental transfer of sand, to be self-arrested to
some extent should conditions required for the feedback
mechanism to work be made available so that the
foreshore’s loss is not always the backshore’s gain. This
confines the function of the baekshore to that of a
transport surface serving the coastal hinterland. The
model presents the operation of processes associated
with an onshore wind during an ebb tide. A flood tide
imposes on the functioning of this system a contracted
foreshore relative to the baekshore, cutting down the
catchment for sand supply and eventually reducing
inter-compartmental transfer.

A change in wind direction gives the situation a
El

face-lift. Combination of an offshore wind with an ebb
tide turns the baekshore, now on the upwind side, into a
catchment and feeder of sand. Substantial movement of
sand on the baekshore, unless self-arrested to an ap¬
preciable extent through exposure of the indurated dune
surface, sustains inter-compartmental transfer of sand
to an expanding foreshore and receptor provided by a
retreating tide, and the backshore’s loss is the
foreshore’s gain except where the moving sand ends up
in the sea. It takes the next flood tide to turn the bulk of
this receptor into a sink, to cut short transfer of sand
from the baekshore by diminishing the foreshore, and to
cause much sand gained through inter-compartmental
transfer from the baekshore to be moved to the offshore
zone, imparting to the foreshore the mere role of a
transport surface.

Broadly along these lines of action, the beach at
Portsea, dominated by onshore southerlies in summer
and offshore northerlies in winter, experiences differen¬
tial transfer of sand in and across its compartments with
every fluctuation of wind components and variation of
the tide. Alternations of onshore and offshore winds
over a relatively large time span are superimposed on the
more regular cycles of ebb and flood of the tide occupy¬
ing relatively small intervals on the time scale. Response
of the beach to such external adjustment of changing
process parameters is likely to find expression in a range
of changes between two poles. On one hand, intra-
compartmental and inter-compartmental transfers of
sand, operating in association with onshore southerlies
and under conditional self-regulation, make themselves
felt, on a gradually diminishing scale with a change from
ebb to flood tides, in the provision of fill to the back-
shore at the expense of cut on the foreshore. On the
other hand, in another changing phase from ebb to
flood, intra-compartmental transfer of sand imposing a
cut on the baekshore, and inter-compartmental transfer
effecting a fill to the foreshore in association with the
offshore northerlies hold sway. Whatever combination
of process parameters prevails in association with one of
these situations, the general tendency is for much mov¬
ing sand entrained by .wind to end up in sinks outside the
foreshore subsystem and to some extent the baekshore
subsystem. This constitutes a permanent loss unless and
until such intra-compartmental and intercom-
partmental transfers of the beach sand are made
good. Feedbacks induced by wind action from the
coastal hinterland dominated by dunes, and their
counterparts induced by wave action from the offshore
zone dominated by swell, may contribute towards this.
Where the recovery process is not complete, the net loss
of beach sand, although it is relatively diminished in
amount, is cumulative. It is likely to play a significant
role in the spatial distribution and redistribution of sand
on the beach in general, and in the long-term sand
budget of the beach in particular.

CONCLUSION

This study reveals that given the optimal wind
strength and a combination of favourable cir-

68

C. L. SO

cumstances, marked changes of beach levels result from
wind-induced intra-compartmental and inter-compart-
mental transfers of sand in a shore system. Such mobili¬
ty of beach levels recorded over short-term observations
points to the ‘catastrophic’ loss of sand within short
periods. Part of the significance of this lies in its bearing
on how far the magnitude of a process may over-ride its
frequency as so far indications are such that any inter¬
pretation of the role of aeolian agents in a beach en¬
vironment solely on the principle of uniformitarianism
may be taken too far. In the functioning of a process-
response system, the periodic encroachment of drifting
sand on the backshore, or on the offshore zone as the
case may be, reinforces the assertion that the
catastrophic loss of sand is spasmodic in action but
cumulative in effect, and that there is the tendency for it
to play such a significant role in the spatial distribution
and redistribution of sand in a beach environment as not
to be readily ignored.

ACKNOWLEDGEMENTS

I thank Dr E. C. E. Bird of the Department of
Geography, University of Melbourne, for advice and
help with the research. 1 am grateful to the Regional
Director (Victoria) of the Bureau of Meteorology for
permission to have access to meteorological records for
reference. The research was carried out with the award
of a Leverhulme Fellowship to the author.

REFERENCES

Bagnold, R. A., 1954. The physics of blown sand and desert
dunes. Methuen, London.

Belly, P., 1964. Sand movement by wind. Tech. Memo.
Coastal Eng. Res. Center U.S. No. 1.

Bird, E. C. F., 1972. Ancient soils at Diamond Bay, Victoria.
Viet. Nat. 89: 349-353.

Bird, E. C. F., 1975. The shaping of the Nepean Peninsula,
Victoria, Australia. Viet. Nat. 92: 132-141.

Bird, E. C. F., 1977, Sites of special interest in (he Victorian
coastal region: a report on geological and geomor -
phological aspects. Town and Country Planning
Board, Melbourne.

Bowler, J. M., 1966. Geology and geomorphology—Port
Phillip Bay. A lent. Natn. Mus. Viet. 27: 19-67.

Davies, J. L., 1972. Geographical variations in coastql
development. Oliver and Boyd, Edinburgh.

Jennings, J. N., 1957. On the orientation of parabolic or
U-dunes. Geog. Journ. 123: 474-480.

Keble, R. A., 1950. Mornington Peninsula. Mem. Geo/. Sur\>.
Viet. 17.

Kuenen, Ph. H., 1964. Pivotabilily studies of sand by a shape-
sorter. 207-15. In Deltaic and Shallow Marine
Deposits , L.M.J.U. van Straaten, ed., Elsevier, Lon¬
don.

Maher, J. V., & McRae, J. N., 1964: Upper wind statistics,
Australia. Surface to 55,000 feet. Bureau of
Meteorology, Melbourne.

Shepard, F. P., & Young, R., 1961. Distinguishing between
beach and dune sands. J. Sediment. Petrol. 31:
196-214.

PROC. R. SOC. VICT. vol. 94, no. 2, 69-87, June 1982

WETLANDS OF VICTORIA III. WETLANDS AND
WATERBIRDS BETWEEN PORT PHILLIP BAY AND MOUNT

EMU CREEK

By A. H. Corrick

Fisheries and Wildlife Division, Arthur Rylah Institute for Environmental Research, 123 Brown

Street, Heidelberg, Victoria 3084

Abstract: Wetlands in the Western District were categorized by water regime and salinity, and
subcategories were based on vegetation. Waterbird distribution and numbers were recorded during
ground inspection and by counts at 135 sites in March, July and October 1980.

In all 1437 wetlands totalling 72 500 ha were located. They were of 6 categories, 23 subcategories
and in sewage and salt evaporation systems. Permanent saline wetland (37 200 ha) and permanent open
freshwater (16 100 ha) were the most extensive categories and permanent open freshwater (397) and
freshwater meadows (371) the most numerous. Since European settlement 34 a /o of the original area of
freshwater wetland has been lost; most reduced have been shallow and deep freshwater marshes, with
79°/o and 66% lost respectively. Little saline wetland has been lost. Three hundred and seventy four im¬
poundments (3580 ha), 4 salt evaporation systems (2180 ha) and 11 sewage oxidation systems (1680 ha)
have been created since settlement.

Eighty six of the 110 waterbird species recorded were seen during the study period. On most
wetland subcategories 5 to 10 species comprised more than 90% of the waterbirds using that subcategory.
Many of these most abundant species occurred on most wetland subcategories. Wetlands with shallow
permanent water support the highest densities of non-breeding birds (8 to 24 birds/ha) and average the
most species (4 to 7) per visit. Assessment was made of the most important species in the various
categories and whether they were breeding there or not. Some comments are offered on duck distribution
based on duck band returns. Factors influencing the distribution of birds are discussed and the lack of
freshwater meadows and shallow freshwater marshes in reserves for conservation is highlighted.

Earlier reports in this series (Corrick & Norman
1980, Corrick 1981) document the number, area and
types of wetlands lost as well as the extent and waterbird
use of remaining wetlands in south-eastern Victoria; in
this report 1 present similar information for wetlands in
western Victoria.

Study Area

The boundaries of the study area (Fig. 1) are
Mount Emu Creek in the west, the main divide of the
western highlands in the north, the eastern watershed of
the Werribee River in the east and the coastline of Port
Phillip Bay and Bass Strait in the south. It is approx¬
imately 19 000 km 2 or 8% of the area of the state.

The population of the area is approximately
350 000 (1978 estimate from Cowie 1980). Of these 45%
live in Geelong and nearby settlements on the Beliarine
Peninsula, 21% live in Ballarat, 15% in Bacchus Marsh
and outer suburban centres of Melton and Wet ribee and
the remainder in the few large country towns (Colac
10 500 and Camperdown 3600), numerous smaller
towns and on rural properties. During summer the
population of all coastal towns is increased sharply by
holiday makers.

Physical Divisions

The study area may be divided into western
highlands, volcanic plains, coastal plains and Otway
Ranges (Hills 1964, Cowie 1980).

The western highlands occupy the northern
quarter of the area. Elevation in this division is about
400 m and relief generally low near Ballarat where areas
of basalt have filled river valleys; both the ancient in¬
terfluves and eruption points extend above this general
level. To the south of Ballarat the Yarrowee and
Moorabool Rivers are well embedded and about Bac¬
chus Marsh relief is more pronounced where the Wer¬
ribee River and its tributaries have formed steep gorges.
The Otway Ranges, which occupy the southern tenth of
the area, have been created by upwarping and faulting
during the Tertiary to form a broad dome which grades
gradually northward to the surrounding plain but dips
sharply to the coast along the southern side. Although
dissection is mature the topography is rounded due to
land slip. The drainage patterns mirror the basic struc¬
ture with the Gellibrand and Barwon Rivers draining the
northern flanks and many small streams diaining the
southern slopes directly to the sea. On the northern edge
of the Otway Ranges the coastal plains form a narrow
strip between the ranges and the volcanic plain. They ex¬
tend eastward to include the Beliarine Peninsula and
westward to Peterborough and Warrnambool on the
coast and are a partially dissected sedimentary surface
which is flat or undulating with a veneer ol Quaternary
dune limestones and sands in places.

The final physical division, the basalt plains,
covers the remaining half of the study area. The plains
have a slight southward slope and prominent eruption

69

70

A. H. CORRICK

cent flows south and east of Lake Corangamite and
south of Skipton. Lakes and swamps occur in volcanic
craters, in depresssions caused by lava collapse (both in
stony rises and elsewhere) and where drainage patterns
have been interrupted particularly along the edges of
lava flows (Currey 1970). Lunettes, parna dunes and
lacustrine deposits formed during the retreat of a much
larger Lake Corangamite (Currey 1963) and by present
lakes (Gill 1963) are common. Over much of the plains
drainage is internal, either to lakes, swamps or to
ground water and has been greatly modified by drainage
works.

Climate

The region’s climate as it affects this study in¬
cluding wetland maintenance is taken from Central
Planning Authority (1956), Bureau of Meteorology
(1959), Department of National Development (1966),
Bureau of Meteorology (1968), Hounam & Powell
(1964) and Cameron (1979).

Rainfall distribution is mainly influenced by
orographic features, namely the western highlands and
Otway Ranges where rainfall is increased; the latter cast
a rain shadow over the lowlands to the west of Port
Phillip Bay. Median (10 and 90 percentile) annual rain¬
fall is 900 to 1800 mm (800 to 1800 mm and 1200 to
2400 mm) in the Otway Ranges, 700 to 900 mm (600 to
800 mm and 800 to 1200 mm) on the coastal plains, 700
to 1000 mm (400 to 600 mm and 800 to 1200 mm) in the
western highlands, 500 to 700 mm (400 to 500 mm and
600 to 800 mm) throughout the western plains and is 400
to 500 mm (300 to 400 mm and 600 to 800 mm) in the
rain shadow to the west of Port Phillip Bay. Annual
variation in rainfall is amongst the lowest in Australia
(Cameron 1979). January is the driest month
throughout the area; falls in the wettest months on the
eastern lowlands (September and October) and
elsewhere (July, August and September) exceed January
falls by 1 to 2 and 2 to 3 limes respectively.

Temperatures range from January average max-

WETLANDS OF VICTORIA III

71

ima of 21 to 27°C (minima 10 to 16°C) to July average
maxima of 5 to 10°C (minima 1 to 4°C inland and 4 to
10°C near the coast). Frost can be expected on fewer
than 5 days per year close to the coast and on up to 15
days per year inland and on higher parts.

Evaporation decreases towards the sea and
towards increasing altitude; it ranges from less than
150 mm per year in the Otway Ranges to just under
1000 mm per year on the plains where it is highest in
January (110 to 150 mm) and lowest in July (25 to
40 mm). Evaporation exceeds rainfall from September
until May and this is reflected in both the patterns of
run-off which result in stream flows and in the occur¬
rence of periods of non-effective rainfall (i.e. falls in¬
sufficient to start germination and maintain growth).
Over most of the plains, where Berrybank (30 years of
records) can be used as an example, non-effective falls
have been recorded in most months in the six month
period April to September; however, these events have
not been consecutive. From the spring months onward
periods of non-effective rainfall become both more fre¬
quent and protracted. In the six month period October
to March, 2, 3 and 4 consecutive months of non-
effective falls have occurred in 93%, 66% and 33% of
years respectively. Close to the coast and in the ranges
the frequency of non-effective falls during summer
decreases e.g. at Beech Forest (51 years of records) 2 and
3 months of non-effective falls have occurred in only
15% and 2% of years respectively.

Hydrology

The Werribee (gauged mean annual discharge of
78 x JO 3 ML from a catchment of 1100 km 2 ),
Moorabool (70 x 10 3 ML from 1100 km 2 ) and Yarrowee
(92 x 10 3 ML from 870 km 2 ) Rivers and Mount Emu
Creek (67 x 10 3 ML from 1240 km 2 ) rise in the western
highlands and the Barwon (139xl0 3 ML from
1040 km 2 ), Curdies (130x 10 3 ML from 780 km 2 ) and
Gellibrand (135 x 10 3 ML from 560 km 2 ) Rivers rise in
the Otway Ranges. Smaller streams which drain the
southern slopes of the Otways include the Aire (27 x 10 3
ML from 25 km 2 ) and the Carlisle Rivers (37x 10 3 ML
from 77 km 2 ) (Bibra & Riggs 1971).

Stream flows are highest in August or September
and are lowest, often reaching zero, in January,
February or March. Floods can occur throughout the
year but are least frequent during summer. Flood fre¬
quencies and summer flows have been reduced in most
rivers by the construction of dams and the subsequent
diversion of irrigation and urban water.

On the plains countless minor drains have either
eliminated or limited the extent of swamps and lakes.
Major drains constructed in the 1950s control the levels
of Lakes Colac and Corangamite, diverting excess water
to the Barwon River and preventing a recurrence of the
extensive expansion these lakes underwent following
successive years of above average rainfall during the
1950s.

Seasonal variation (Bibra & Riggs 1971), type of
basalt substrate (Maddocks 1967) and various catch-

E 2

ment to surface area ratios and throughflows combine to
produce lakes and swamps of a very wide range of
salinities (e.g. Williams 1964, Bayly & Williams 1966).
The geochemistry of waters in the region is discussed by
Maddocks (1967) and detailed studies which include
seasonal changes of individual basins have been made
(e.g. Pollard 1971a, Timms & Brand 1973, Walker 1973,
Geddes 1976, Timms 1976, Williams & Buckney 1976).

METHODS
Wetland Distribution

All wetlands and drained areas larger than 1.0 ha
(excluding river flats inundated during floods) were
located from aerial photographs (Division of National
Mapping and Department of Crown Land and Survey,
1:85000 enlarged x2, flown 1961 Ballarat and 1976
Melbourne map sheets and 1:33000 flown 1975-77 for
the remainder), from topographic maps and during
ground surveys between July 1978 and November 1980.
Water source and regime were determined, plant com¬
munities identified and areas calculated by planimeter
measurement from aerial photographs.

Wetland Classification

Water regime and salinity were used for defini¬
tion of wetland categories while subcategories were
based on vegetation important in determining waterbird
usage. Wetlands were classified as fresh if salinity re¬
mains below' 3000 parts per million (ppm) (Williams

1964) for the greater part of the period of inundation.
Changes in depth and area (e.g. Lake Corangamite,
Currey 1963, Parliamentary Public Works Committee

1965) resulting from long term trends in the amount of
rainfall have been ignored and the areas calculated and
categories assigned using conditions prevailing during
this study.

Sewage oxidation basins have not been included
in previous surveys (Corrick & Norman 1980, Corrick
1981); however, in the present study area, along with
salt evaporation basins, their area is substantial. In¬
dividual basins in these systems are small (often < 1 ha)
and collectively show a wide range of chemical composi¬
tions and limnological development and could be as¬
signed individually to one of several wetland sub-
categories (e.g. salt evaporation basins range in salinity
from sea water to saturation and thus cover sub¬
categories 6.1, 6.5 and 7.1). The condition of the in¬
dividual basins may also change at anytime with changes
in management practices. Classification of individual
basins was considered to be inappropriate; the systems
were considered as single wetlands outside the category
system (Tables 2 and 3) but their distribution and water-
bird use have not been shown.

The diagnostic salinities, w r atcr regimes and
vegetation of all wetland categories and subcategories
found are summarized in Table 1. Subcategories not
described by Corrick and Norman (1980) or Corrick
(1981) are described below.

Red Gum-dominated (Subcategories 2.3 and 3.5)

Red Gum (Eucalyptus camaldulensis Dehnh.)

72

A. H. CORRICK

Table 1

Characteristic Depth, Duration of Inundation of the Wetland Categories and Typical Vegeta¬
tion of Wetland Subcategories in the Study Area

Duration of

Category Depth (m) inundation Subcategories Typical vegetation

freshwater

2, Meadows

<0.3

<4 months

.1

Herb-dominated

Annual moist soil species

.3

Red Gum-dominated

Eucalyptus camaldulensis

.4

Lignum-dominated

Muehlenbeckia cunn¬
inghamii

3, Shallow marshes

<0.5

<6 months

.1

Herb-dominated

Annual moist soil and
aquatics

.3

Cane Grass-
dominated

Eragrostis australasica

.4

Lignum-dominated

M. cunninghamii

.5

Red Gum-dominated

Eucalyptus camaldu/ensis

4, Deep marshes

<2

12 months

.2

Reed-dominated

Phragmites australis, Typha
sp. Scirpus valid us

.3

Sedge-dominated

Lepidosperma longitudinale

.4

Rush-dominated

Eleocharis sphacelata

.5

Open water

Submerged aquatics with
moist soil annuals in the lit¬

toral zone.

.6

Cane Grass-
dominated

Eragrostis australasica

.7

Lignum-dominated

M. cunninghamii

5, Permanent open
water

>0

permanent

.1

Shallow (<3m)

Submerged aquatic species,
emergent species in the lit¬
toral zone.

.2

Deep (>3 m)

Submerged aquatics'

.3

Impoundment

Submerged aquatics with
emergent species in the lit¬
toral zone 2

SALTWATER

6, Semipermanent

<2

<8 months

.1

Salt pan

Lepi/aena spp. Ruppia sp 3 .

.2

Salt meadow

Halophytes with Ruppia sp.
and Lepilaena spp. in
shallows

.3

Salt flats

Dense ground cover of
halophytes

.5

Hypersaline lakes

nonQ

7, Permanent

>0

permanent

.1

Shallow (<3 m)

Ruppia spp. Lepilaena spp.

.2

Deep (> 3 m)

Ruppia spp. Lepilaena spp.

.3

Intertidal flats

Zostera spp. various alga,
none in places 4

' Reduced by turbidity and depth. 2 Depends on grazing. 3 No vegetation when dry. 4 Sand flats are
usually devoid of vegetation.

woodland up to 20 m tall occurs on both freshwater
meadows and shallow freshwater marshes. Associated
vegetation is similar to herb-dominated subcategories
but Cane Grass ( Eragrostis australasica (Steud.) Hub¬
bard) and Lignum (Muehlenbeckia cunninghamii
(Meissn.) Mueller) may also occur. Mature trees, both in
wetlands and on higher ground, provide nest hollows for
waterfowl; however, grazing and cultivation prevent
regeneration so that mature trees are not being replaced.

Cane Grass-dominated (Subcategories 3.3 and 4.6)

Cane Grass has upright stems (<5 mm in

diameter) up to 1.4 m high which range from scattered
tussocks to continuous extensive stands in which the
stems become progressively more tangled if flooding
persists for several seasons. Grazing reduces stems to
ground or water level and prevents regrowth until after
flooding. In highly turbid water associated species may
be absent.

Lignum-dominated (Subcategorics 2.4, 3.4 and 4.7)

Lignum forms dense tangled bushes (up to 2.5 m
high and 3 m in diameter) in water up to 1 m deep.
Although the stems are woody, persistent grazing will

WETLANDS OF VICTORIA III

73

eventually break these down and along with cultivation
will remove bushes entirely. Both Red Gum and Cane
Grass are often associated with Lignum which also oc¬
curs as scattered bushes above the shoreline of perma¬
nent open waters and persists in drained areas which
have not been cultivated.

Hypersaline lakes (Subcategory 6.5)

In lakes of this category salinity only drops below
50 000 ppm in exceptionally wet seasons. The water
usually reaches saturation each year as the water level
recedes. Smaller basins may be 3 m deep but large areas
Ure shallow (<0.4 m) so that summer drying occurs.
Rooted aquatic plants may be present for the short

period when salinity is low but growth is reduced as
salinity rises. Zooplankton, particularly Parartemia ziet-
ziana, which will hatch in salinities to 202 000 ppm and
survive almost to saturation (Geddes 1976), are abun¬
dant.

Waterbird Distribution and Abundance

When wetlands were visited throughout the study
period waterbirds were either counted on the whole, or
part, of each wetland; alternatively only the species pre¬
sent were recorded. Large wetlands were visited several
times although usually only one complete count of
waterbirds present was made on each. The counts, 780
in all, were used to indicate habitat preferences, provide

Table 2

Number of Wetlands of Each Category and Number of Areas of Each Wetland Subcategory in

Each Wetland Size Range

Category/subcategory

Number of wetlands in the following size
(ha) ranges

Total number

1-5

6-10

11-25

26-100

>100

Subcategory

Category

2 Freshwater meadow

.1 Herb-dominated

237

80

44

8

1

370

.4 Lignum-dominated

1

1

Number of wetlands

237

80

45

8

1

371

3 Shallow freshwater marsh

.1 Herb-dominated

154

34

33

25

1

247

.3 Cane Grass-dominated

1

3

4

.4 Lignum-dominated

1

1

1

3

.5 Red Gum-dominated

1

1

Number of wetlands

156

34

37

26

1

254

4 Deep freshwater marsh

.1 Shrub-dominated

1

1

.2 Reed-dominated

1

1

2

4

.3 Sedge-dominated

1

1

2

1

1

6

.4 Rush-dominated

4

2

1

1

8

.5 Open-water

47

2

14

13

5

81

.6 Cane Grass-dominated

1

3

2

6

.7 Lignum-dominated

1

1

Number of wetlands

52

5

16

13

6

92

5 Permanent open freshwater

.1 Shallow

2

2

1

8

9

22

.2 Deep

1

1

2

4

.3 Impoundment

318

25

13

8

10

374

Number of wetlands

321

27

14

16

19

397

6 Semipermanent saline

.1 Salt pan

101

41

36

32

10

220

.2 Salt meadow

19

16

11

16

5

67

.3 Salt flats

3

3

6

13

4

29

.4 Hypersaline lake

1

2

6

3

3

15

Number of wetlands

119

50

46

38

14

267

7 Permanent saline

.1 Shallow

2

4

5

10

16

37

.2 Deep

1

5

6

.3 Intertidal flats

1

1

2

4

Number of wetlands

2

4

6

11

18

41

Salt evaporation basin

4

4

Sewerage oxidation basin

6

3

1

1

11

Totals

893

203

165

112

64

1437

74

A. H. CORRICK

a species list (common names according to RAOU 1978)
for each category and subcategory and enable both the
frequency of occurrence of species to be compared and
the regional significance of concentrations to be gauged.

During August, September and October 1978
and September 1979 records were kept of sightings of
waterbirds not on wetlands included in the study. Obser¬
vations aided by 10x50 binoculars were made from a
vehicle over 1100 km of roads through open farmland
between Geelong and Colac and near Ballarat. Approx¬
imately 0.3 km on each side of the road was covered ex¬
cept where hedges, trees and buildings obscured views.

So that seasonal changes in wetland area and
waterbird use and populations could be monitored on a
quantitative basis 135 wetlands (or parts of large
wetlands) of known area between 38°00' and 38°30'S
and 143° to 144°E were selected and counts made in
March, July and October 1980. Buoys, stakes and
natural features were used to mark site boundaries and

areas were calculated from aerial photographs, bV
triangulation and by rangefinder measurements. Obser¬
vations were aided by 10x50 binoculars and 25x60
telescope. The area of water was estimated each month.
Analysis of Waterfowl Banding Data

Ducks have been banded in Victoria since 1950,
81% at the Wildlife Research Station at Lara (38°0l'S
144°25'E). The distribution of the returns of bands from
ducks shot during open seasons between 1951 and 1980
has been plotted on a 10' grid of the area and the number
compared with recoveries from elsewhere in the state.

RESULTS

Wetland Distribution

During this study 1437 wetlands (Table 2) total'
ling 72 500 ha (Table 3) were located, including 1680 h4
of sewage oxidation and 2180 ha of salt evaporation
systems (which were not categorized). Impoundments

Table 3

Area of Wetland Categories and Subcategories in Wetlands of Various Size Ranges

Category/subcategory

Area (ha) of wetlands in the following
size (ha) ranges

Total area

1-5

6-10

11-25

26-100

>100

Subcategory

Category

2 Freshwater meadow

2410

.1 Herb-dominated

630

620

700

323

123

2400

.2 Lignum-dominated

16

16

3 Shallow freshwater marsh

2540

.1 Herb-dominated

368

261

557

1060

190

2440

.3 Cane Grass-dominated

5

30

35

.4 Lignum-dominated

3

20

26

49

.5 Red Gum-dominated

16

16

4 Deep freshwater marsh

2320

.1 Shrub-dominated

9

9

.2 Reed-dominated

1

2

365

368

.3 Sedge-dominated

3

7

30

43

69

152

.4 Rush-dominated

11

12

30

66

119

.5 Open-water

75

15

213

580

540

1420

.6 Cane Grass-dominated

8

36

148

192

.7 Lignum-dominated

53

53

5 Permanent open freshwater

16100

.1 Shallow

9

16

19

376

11400

11800

.2 Deep

4

30

680

710

.3 Impoundment

580

187

212

416

2180

3580

6 Semipermanent saline

8080

.1 Salt pan

239

261

460

1260

1540

3760

.2 Salt meadow

43

111

86

279

385

904

.3 Salt flats

7

12

82

500

1220

1820

.5 Hypersaline lake

3

15

98

117

1360

1590

7 Permanent saline

37200

.1 Shallow

3

23

85

660

14400

15200

.2 Deep

5

20900

20900

.3 Intertidal flats

15

57

1020

1090

Salt evaporation basin

2180

2180

Sewerage oxidation basin

18

27

19

1620

1680

Total

2000

1580

2670

5850

60400

72500

WETLANDS OF VICTORIA III

75

Table 4

Numbers and Area (ha) of Wetlands Destroyed by Drainage Works and the Area Lost and the Area of Other Wetland

Categories Created by Partial Drainage

Category

Destroyed

Area (No)

Reduced or altered to:
area

lost FM SFM

DFM

POFW

SPS

PS

Total

affected

net

lost *

freshwater meadow

1660 (162)

67

31 (6)

1760

390

^hallow freshwater

marsh

5090 (228)

3860

1020 (137) 133 (16)

4(1)

10110

9340

Peep freshwater marsh

2280 (16)

1690

166 (9) 408 (13)

16(1)

4(1)

4560

4540

Permanent open

freshwater

590 (2)

1100

36 (2)

13 (1)

670 (7)

2410

2410

Semipermanent saline

53 (5)

278

1100 (10)

1430

-1390

Permanent saline

0

940

153(1)

1050 (4)

560 (1)

2890

2330

Total area (number)

9670 (413) ’

7940

1370 (153) 767 (31)

29 (2)

4(1)

2820 (22)

560 (1)

23160

17610

'total affected less area of category formed by partial drainage of other categories.

(374) and herb-dominated freshwater meadows (370) are
the most numerous subcategories but are on average
small in area (9.6 ha and 6.5 ha respectively). Other sub-
categories which are numerous include herb-dominated
shallow freshwater marshes (247 average 9.9 ha) and salt
pans (220 average 17.1 ha). These four subcategories ac¬
count for 85% of the wetlands in the region studied but
only 18% of the area of wetland in it. Permanent saline
wetland (37 200 ha) and shallow and deep permanent
open freshwater (12 500 ha), account for 54% and 24%
of the area and 3% and 2% of the number of wetlands
respectively. In all categories except permanent saline
the number of wetlands in the various size ranges
decreases with increasing size. For freshwater meadows
and shallow freshwater marshes the largest area is in the
11-25 and 26-100 ha size classes; for all other categories
the > 100 ha class contains the greatest area even though
it contains the fewest wetlands. Overall, 62% of
wetlands and 3% of the area is in wetlands 1-5 ha in area
while 4% of wetlands and 82% of the area is contained
in wetlands >100 ha in area.

Six wetland subcategories each occur at 4 or
fewer sites and total less than 100 ha. A further 3 sub¬
categories occur at 4 or fewer sites but their area is
greater than 100 ha. These nine subcategories, together
with sedge and Cane-Grass-dominated deep freshwater
marshes and hypersaline lakes are restricted to 5 or
fewer 10' grids (Fig. 2). Natural freshwater wetlands
(52%) of the total wetland area) are found throughout
the lowlands and are absent from the ranges; saline
wetlands are found on the basalt plains and along the
coast. The most widespread subcategories are also the
most numerous (i.e. freshwater meadows, shallow
freshwater marshes and impoundments).

Drainage works of various forms have destroyed
or altered 623 wetlands which covered 23 160 ha (Table
4); of these 413 (9670 ha) have been lost completely with
freshwater meadows 162 (1660 ha), shallow freshwater
marshes 228 (5090 ha) and deep freshwater marshes 16

(2280 ha) having decreased most. The net result, allow¬
ing for the area of some categories created by partial
drainage of others, is that 390 ha of freshwater meadow,
9340 ha of shallow freshwater marsh, 4540 ha of deep
freshwater marsh and 2410 ha of permanent open
freshwater have been lost. Only the area of semiper¬
manent saline wetland has increased since settlement (by
1390 ha). These changes, when compared to the area
present before European settlement, represent a loss of
14% of the area of freshwater meadow, 79% of shallow
freshwater marsh, 66% of deep freshwater marsh, 16%
of permanent open freshwater, 6% of permanent saline
wetland and an increase in the area of semipermanent
saline wetland of 17%.

Seasonal Changes in Wetland Area

Changes in area of the count sites visited in
March, July and October 1980 (Table 5) show that sub¬
categories with permanent water show little seasonal
change in area while salt and freshwater meadows,
shallow freshwater marshes and salt pans will dry during
summer and fill during winter. Drying rate depends on
rainfall, temperature, catchment area, bottom type and
contour. Wetland categories and subcategories not in¬
cluded in the count sites, particularly various sub¬
categories of deep freshwater marshes, would show
similar variation to open water areas of this category. In
impoundments the seasonal fluctuations are increased
by diversions of water and tidal flats vary in extent ac¬
cording to the lunar cycle.

Waterbird Distribution and Abundance

During the study 86 species of waterbird were
recorded (Table 6, Appendices 1 & 2) and a further 24
have been recorded by other authors. The five most
abundant species on each of the wetland subcategories
(28 species in all) surveyed during counts in March, July
and October 1980 are listed in Table 6. Wetland sub¬
categories used by less abundant but regularly seen

76

A. H. CORRICK

Freshwater meadows 2.1 Herb-dominated

i l i

143' . 144 s _ 145°

38° -

i*+o . _ »***♦

^J»p A ♦

5^3

.£1

Freshwater meadows 2.4 Lignum-dominated
I i i

Shallow freshwater marshes 3.1 Herb-dominated Shallow freshwater marshes 3.3 Cane Grass-dominated

ill ill

Deep freshwater marshes 4.1 Shrub-dominated

HECTARES A 1-25 ■

Fig. 2 — The distribution (plotted on a 10' grid) of

Deep freshwater marshes 4.2 Reed-dominated

area (ha) of each wetland subcategory of the study

WETLANDS OF VICTORIA III

77

Deep freshwater marshes 4.3 Sedge-dominated
i i i

Deep freshwater marshes 4.4 Rush-dominated
i i I

Deep freshwater marshes 4.5 Open water

Deep freshwater marshes 4.6 Cane Grass-dominated
i I i

Deep freshwater marshes 4.7 Lignum-dominated
I I l

Permanent open freshwater 5.1 Shallow

Permanent open freshwater 5.2 Deep

HECTARES

▲ 1-25

Permanent open freshwater 5.3 Impoundments

144’ 145°

38° -

M

▲

'A

A]

n

Cr

A

■

B

ft

m

♦

♦

\

%

A

a

2

A

A

L

□

:

£

m

▲

A

A

■

A

A

A

a

▲

A

A

■

A

A

■

a

B

A

A

♦

m

A

▲

1

c

C

□

□

7

Ksnciisr;

26-150

+ 150-500

• >500

Fig. 2(continued)

78

A. H. CORRICK

Semipermanent saline wetlands 6.1 Salt pans Semipermanent saline wetlands 6.2 6alt meadows

ill ill

Semipermanent saline wetlands 6.3 Salt flats
i l i

Semipermanent saline wetlands 6.5 Hypersaline lakes
i l l

Permanent saline wetlands 7.1 Shallow

i I i

Permanent saline wetlands 7.2 Deep

Fig. 2(continued)

WETLANDS OF VICTORIA III

79

species are presented in Appendix 1 and Appendix 2 lists
the remaining rare species, which either occur regularly
in very low numbers* are misplaced nomadic or
migratory species (e.g. various sandpipers, Garganey
Teal) or are more common on other habitats of the
region (e.g. Hooded Plover— ocean beaches, Black-
faced Shag-offshore waters).

The 780 counts made of watcrbirds during
ground surveys have been used to indicate subcategories
utilized by species listed in Appendix 1. The more abun¬
dant species (Hoary-headed Grebe, Black Swan,
Masked Lapwing, Australian Shelduck, Silver Gull,
Table 6) were also the most frequently recorded occurr¬
ing on a wide variety of wetlands. The counts also in¬
dicate species with particular habitat preferences, thus
more than 50% of the records of some species came
from a single wetland category, e.g. Latham’s Snipe —
freshwater meadow; Brolga —shallow freshwater marsh;
Marsh Harrier, Dusky Moorhen and Purple Swam-
p| ien _-dcep freshwater marsh; Banded Stilt —hyper¬
saline lakes and Great Crested Grebe, Pelican, Pied Cor¬
morant, Great Egret, Chestnut Teal, Blue-billed Duck
and Freckled Duck —permanent saline wetlands.

Data from counts conducted in March, July and
October 1980 (Table 6) show the distribution of species
and numbers amongst the subcategories studied. Hoary-
headed Grebe, Black Swan, Shelduck, Grey Teal, Coot,
Masked Lapwing and Silver Gull are abundant on most
categories while other species, e.g. Great Crested Grebe,
Blue-billed Duck, Dusky Moorhen and Purple Swamp-
hen, are restricted to only a few; shorebirds arc obvious¬
ly absent from deep permenant open freshwater and cor¬
morants from hypersaline lakes. Count sites on wetlands
with shallow permanent water have the highest density
of birds (means of 8-24 birds/ha) and the most species
(means 4-7 species/visit) while sites on seasonal wetlands
have lower population densities (means from 1.5-13.2
birds/ha) and fewer species (1.5-4 species/visit). On all
subcategories the 5 most abundant species accounted for
more than 70% of all birds/ha and in many cases for
more than 95% of this total. Observations suggest that
intertidal flats, sewage oxidation and salt evaporation
systems, a subcategory and habitats not included in the 3
counts during 1980, also carry comparatively high den¬
sities of birds. The results of detailed surveys of these
habitats are provided by Tarr (1952), Morgan (1954),
Watson (1955), Wheeler, W. R. (1955), Barkla (1978)
and personal communication from members of the
Australasian Wader Studies Group in 1981.

The numerical distribution of all species across
the wetlands of the area cannot be considered in detail
here; however, it is clear that the shallow lakes (salinities
<50 ppt), which comprise a large proportion of the
total wetlands area (Table 3) support a large proportion
of the populations of some non-breeding waterbirds.
The seasonal and annual changes which occur in these
lakes can provide conditions particularly favourable to a
species and large concentrations which are not neces¬
sarily annual events can occur. Wetlands at which
regionally significant concentrations of non-breeding

birds were recorded include:

Lake Corangamite (southern half)—7500 Coot,
1400 Great Crested Grebe, 8200 Black Swan, 100
Chestnut Teal on 30 November 1979.

Lake Corangamite (northern half)—14 000
Hoary-headed Grebe, 1900 Musk Duck on 10 April 1979
and 650 Freckled Duck on 26 August 1980 (Corrick
1980).

Lake Rosine —6000 Pink-eared Duck, 3000 Blue¬
billed Duck 25 July 1979.

Lake Murdeduke— 19 000 Coot on 9 April 1979.

Lake Gnarpurt —900 Musk Duck 31 April 1979.

Lake Beeac-11 000 Banded Stilt 23 February
1979, 1000 Whiskered Tern on 17 October 1979.

Lake Bookar —7000 Hoary-headed Grebe on 26
June 1979.

An unnamed shallow freshwater marsh at
38°08'S 143 °45'E — 30(F Avocet on 18 November 1979.

Seasonal changes apparent in Table 6 may be due
to changes in the area of available habitat (see Table 5),
or to arrival and departure of migratory species (e.g.
Sharp-tailed Sandpiper and Whiskered Tern). Nomadic
species (Pink-eared Duck, Hardhead, Grey Teal and
Black Swan) may also show numerical changes related
to climate and wetland conditions outside the study
area, which are not regular seasonal events, and finally
the effect of the duck shooting season may also modify
populations. The long term variability of numbers of a
nomadic species, Grey Teal, is shown by Figure 3 (see
also Morgan 1954). The population is generally low
from July to December inclusive and high with greater
variation from year to year for the remainder of the
year.

During roadside counts of waterbirds (not on
wetlands included in the study), 494 sightings of 31
species were made, 222 sightings of 25 species were on
farm dams and ponds < 1.0 ha in area. In all 953 dams
were checked and of these 158 (17%) had waterbirds on
or close by; Masked Lapwing (on 44 dams), Pacific
Heron (on 30), White-faced Heron (on 27), Hoary-
headed Grebe (on 26) and Black Duck (on 22) were seen
most frequently. Of the 158 dams with waterbirds 73%
had only one species present, 18% had two, 5% three,
3% four and 1% five species. Away from dams the most
frequently seen species were White-faced Heron (73
sightings), Masked Lapwing (41), Straw-necked Ibis and
Pacific Heron (36), Silver Gull (35) and Australian
Shelduck (22). Of the common species on dams only
two, Black Duck and Hoary-headed Grebe, were not
seen more frequently away from dams.

The habitats used by breeding waterbirds are
often entirely different and often distant from the non-
breeding records given previously. Time did not permit
detailed searches of each category of wetland for nests
so except for colonial nesting species relatively few
breeding records were obtained. However, the de¬
pendence of some species on particular conditions or
vegetation for successful breeding are apparent. Some
species breed throughout the area, e.g. White-faced
Heron which next in trees away from water, and Masked

80

A. H. CORRICK

long grass and dense vegetation and Pied Oystercatcher
and Red-Capped Plover use beaches and adjacent
dunes). Finally several species nest colonially (e.g.
Straw-necked and Sacred Ibis use extensive reed beds in
Reedy Lake near Geelong; these species of ibis with
Pelican, Glossy Ibis, Black Swan, Pied and Great Cor¬
morant, Silver Gull and Gull-billed Tern use islands in
wetlands; Pied Cormorant use trees on an island in
Swan Bay and on the Melbourne Sewage Farm at Wer-
ribee and Fairy Terns use large areas of exposed sand
above high tide line). The continued success of these col¬
onies is dependent on the continued seclusion of the sites
both from predators and disturbance by man and the
maintenance of appropriate nesting substrates.

Analysis of Wathrfowi. Banding Data

Bands have been returned from all but one (in the
Otway Ranges) of the 10' grid squares entirely within the
area (Fig. 4) and all but seven partly in the area, 5 of
which are mainly ocean, I outer suburban, and the last
forested ranges along the northern boundary.
Recoveries of all species are concentrated in squares
about the main banding location at Serendip (Table 7),
which include tidal Hat, semipermanent saline wetlands,
salt evaporation and sewage oxidation systems along the
western shores of Port Phillip Bay and semipermanent
and permanent saline wetlands and deep freshwater
marshes at the mouth of the Barwon River and in the
four 10' grid squares which include Lakes Corangamite,
Martin and Colac. A higher proportion of bands from
Black Duck, than from other species, have come from
10' squares without large wetlands, particularly in the
northeastern part of the area. Grey Teal have come from
more squares (73) and a higher proportion have come
from the grid squares containing Lake Murdeduke and
Modewarre than have returns of other species. Chestnut
Teal have come from only 32 squares, more along the
southern coast and few in the ranges, while most returns

Table 5

The Mean and Standard Error (SE) of the Area (Expressed as a °7o of the Total Possible Area)
and the Number of Sites at Which Waterbird Counts were made in March, July and October

1980

March July October

Category

Mean

SE

Mean

SE

Mean

SE

number

Freshwater meadow

0

44

8

63

8

27

Shallow freshwater marsh

0

58

11

69

10

19

Deep freshwater march
Permanent open freshwater

65

14

96

2

98

2

9

Shallow

91

5

95

2

96

3

9

Deep

Semipermanent saline

100

0

100

2

100

0

2

Salt pan

12

6

92

3

96

3

29

Salt meadow

0

71

12

71

12

9

Hypersaline lakes

Permanent saline wetland

43

16

99

6

100

0

8

Shallow

72

7

86

5

97

2

24

Deep

0

100

100

300—1

o>

I 200-

Q.

O

o

>.

o

o

JFMAMJJASOND

Month

Fig. 3—The range (vertical line), mean (horizontal line) and
95% confidence limits of the number of Grey Teal caught per
day of trapping each month between 1951 and 1977 at Serendip
Wildlife Research Station 38°0rS and 144°25'E.

Plover which nest on the ground around wetlands and in
pasture. Some species are restricted to either particular
wetland categories (e.g. Brolga to freshwater meadows
and shallow freshwater marshes, Black-winged Stilt to
freshwater and saltwater meadows. Blue-billed Duck to
deep freshwater marshes (Wheeler, J. R. 1953) and Bit¬
tern to deep freshwater marshes); or utilize non-wetland
habitats (e.g. Shelduck use tree hollows, Black Duck use

WETLANDS OF VICTORIA III

81

CHESTNUT TEAL
»

143'

144

145

PACIFIC BLACK DUCK

I i

143' . r> 144 x

145

snnncnnn

38 -

I

||\

cor

■gg rarrna i

" CT a_aia^l

GREY TEAL

38 -

143 144 145

4 >t>i*T'A

^BaaIa

pQoon

IBannnninnl
cinnnnnnncD
pnmnnsi
Emmacsl

3 E*£:T

PROPORTION
OF RECOVERIES

A < 1%

AUSTRALIAN SHELDUCK
i

143°

38“

■i^

145

^ 1 % TO< 2%

4^2%TO<10% # $510%

Fig. 4 —The distribution of band recoveries from Australian Shelduck, Pacific Black Duck, Grey Teal
and Chestnut Teal by shooting in 10' grid squares covering the study area. Recoveries between 1951 and
1980 are included. The number recovered in each square is displayed as a proportion of the total number
recovered from each species in the study area.

of Australian Shelduck, few of which were banded at
Serendip, are from the western plains.

DISCUSSION

The total area of wetland including sewage oxida¬
tion and salt evaporation systems located in this study, is
72 500 ha, which compares with 72 200 ha and 75 900 ha
located in similar surveys in Gippsland and South Gipp-
sland (Corrick & Norman 1980, Corrick 1981); however,
there are many more wetlands (1437 cf. 320 and 202)
and the proportions of various categories are different.
In South Gippsland, for example, 91% of the wetland
area is intertidal flats while here only 1.6% is of that
subcategory. Around the Gippsland Lakes shallow and
deep permanent saline wetlands comprise respectively,
13% and 21% of the total while in this study they com¬
prise 22% and 31%. The proportion of freshwater
wetland is higher (35% of total cf. 33% and 3%) and the
area of freshwater meadow and shallow freshwater
marsh is some four times higher than in the previous
studies.

As in the other areas studied there have been
widespread changes in wetland area since European set¬
tlement. These changes include the large number of
wetlands which have been drained, the 374 impound¬
ments spread almost throughout the region and the ex¬

tensive sewage (1680 ha) and salt evaporation (2180 ha)
systems on Port Phillip Bay. In all 34% of the original
area of freshwater wetland has been lost, 14% of
freshwater meadow, 79% of shallow marshes, 66% of
deep marshes and 16% ol permanent open water. These
changes in area include 37% of the number of wetlands.
These losses are not as severe as those in South Gipp¬
sland where 95% of the area of freshwater wetlands has
been lost but worse than around the Gippsland Lakes
where 22% has been lost. As in other parts of the state
little saline wetland has been lost.

The impact of these changes on waterbird
populations is difficult to assess. Certainly the large
number of impoundments and increased area of cleared
land have benefited a few species (c.g. Hoary-headed
Grebe, Ibis and White-faced Heron) enabling them to
occur more widely. The large populations of migratory
waders along the western shores of Port Phillip Bay are
probably due to a combination of factors including the
increased productivity of intertidal flats due to outfalls
of treated sewage and the presence of adjacent sewage
and salt works ponds which enable migratory waders to
feed for longer periods each day than is possible on in¬
tertidal flats alone. Although the sewage ponds and im¬
poundments are used by flocks of waterfowl it is unlike¬
ly that the populations of these species have benefited

82

A. H. CORRICK

Table 6

Abundance of Birds on Wetland Categories Included in counts in March (M), July (J) and October(O) 1980.

Birds are ranked in order of abundance (birds/ha) from I (highest) 10 10 (lowest). The table includes the 5 most abundant species
from each category in each month and indicates the rank (or its presence ( + ) at ranks over 10) on the other categories and in other
months. The number of sites, the number of species and the total number of birds/ha is also given. The category and subcategories

arc listed in full in Table 1.

Wetland category and subcategory

Species

2.1

3.1

4.5

5.1

5.2

6.1

6.2

6.5

7.1

M J

O

M J

O

M

J

O

M

J

O

M

J

O

M J

O

M J

O

M

J

O

M

J

O

Great Crested Grebe

4

2

1

4

+

4

4

7

Hoary-headed Grebe
Little Black Cor-

9

4

4

7

3

2

3

3

2

2

1

1

2

4 7

3

4

3

2

3

4

morant

Little Pied Cor-

8

10

10

4

4

4

8

4

morant

4

4-

3

4-

4

4

White-faced Heron

3

10

9

3

9

4

5 4-

+

4

5

4

Straw-necked Ibis

4

1

6

7

+

4

Yellow-billed S’bill

7

2 4

Black Swan

1

2

7

4

4

7

4

2

4-

5

2

6

6 3

1

1

5

6

1

10

7

3

Australian Shelduck

8

8

4

4

7

'4-

4

5

4

8

2

1

2

5

7

4

Pacific Black Duck

6

4

6

6

4-

4-

6

6

Grey Teal
Australasian

5

3

10

4

9

9

5

4

9 1

2

7

3

5

8

Shoveler

4

10

4

10

5

7

6

9

4-

4

4

4

4

Pink-eared Duck

5

4

3

4

4

2

2

Hardhead

4

8

8

2

4-

4-

3

9

4

5

Blue-billed Duck

6

4

8

4

4

10

Musk Duck

4

4-

9

6

3

4

4

6

Dusky Moorhen

5

5

+

Purple Swamphen

4

+

1

2

2

3

6

+

Eurasian Coot

4

6

+

1

1

1

1

1

1

3

5

4- 2

4

6

6

1

1

1

Masked Lapwing

4

9

5

9

8

9

4-

4-

1 5

7

4

8

4

5

6

4

4

Red-capped Plover

8

4

4-

-1-

4-

4

4

4 8

3

6

4

4

4

Black-winged Stilt

4

5

4-

4

10

4

4

Red-necked Avocet
Sharp-tailed Sand-

4

8

10 9

6

9

1

4

piper

+

10

4

4-

4-

7

5

2

4

4

4

Red-necked Stint

7

4

4-

4-

6

8

2

3

4

6

9

Curlew Sandpiper

5

4

4

4

Silver Gull

2

6

2

4

4-

4

3

4

5

10

4

4 7

4

3

3

4

5

8

10

Whiskered Tern

3

1

10

4-

4

4

4

1

2

4

4

Number of Sites

0 19

23

0 13

15

7

9

9

9

9

9

2

2

2

3 28

29

0 8

7

3

8

8

20

28

24

Total number of

species

1 13

23

0 14

22

15

17

21

24

21

22

6

6

6

13 21

26

0 7

17

6

7

7

31

19

21

Mean total

M 0

0

11.9

22.6

0.62

2.9

0

0

9.5

number/ha

2.9-49

7-65

0-4.8

5.3-17

with 95%

J 4.8

1.5

13.9

14.2

0.5

0.9

3.8

0.65

7.6

limits

1.6-12

0.5-3.

1

4.7-40

4.5-41

0.4-1.7

1.2-9.5

0-2.5

4.6-12

O 4.7

3.2

12.7

21.2

0.74

3.5

13.2

1.1

11.4

2.4-8.7

1.5-6.1

1

.1-34

7.3-58

2.1-5.6

3.7-42

0.02-3.3

6.7-19

because these wetlands provide habitat similar to
shallow and deep permanent open fresh and salt water
wetlands and represent only 8% of the area of these
categories already present in the study area. The large
number of rare species recorded along the western
shores of Port Phillip Bay reflect the very prolonged and
intensive observations made in the area (Morgan 1954,
Watson 1955, Smith 1962 to 1978, Barkla 1978) com¬
pared to the few studies in other parts (Binns 1953,
Hirth 1976, Missen & Timms 1976), rather than some

unique feature of the habitats present.

The detrimental effect of the loss of wetland on
waterbird populations is more certain. Sites at which
counts were made on freshwater meadows, and shallow
and deep water marshes had up to 45, 21 and 117
birds/ha respectively during 1980. The 14 300 ha of
these categories which have been lost would have sup¬
ported substantial populations and similar habitats have
not been provided by wetlands created since settlement.
Species most affected would presumably be those

83

WETLANDS OF VICTORIA III

restricted to freshwater (e.g. Latham’s Snipe, Purple
Swamphen, Crakes, Dusky Moorhen and Whiskered
Tern) and particularly those which rely on such
categories for breeding habitat (e.g. Brolga, Australa¬
sian Bittern and Black-winged Stilt).

The factors which control the number of species
of waterbirds in the area at any one time are complex
and depend not only on the condition and area of the
wetland categories available but also on rainfall patterns
in other parts of Australia which control the movement
of many Australian waterfowl (Frith 1967, Cowan 1973,
Braithwaite 1975) and day length which regulates the
movement of migratory species. Within the area the
distribution of the numbers and species present also
depends on a number of interacting factors, e.g. the
amount and distribution of rainfall, seasonal changes in
wetland area, the potential food available and habitat
condition of individual basins and the normal range of
the species involved. Counts of waterbirds show the
effects of these factors and also suggest the important
factors acting on the distribution of some species within
the area. Species such as Pied Oystercatcher, Black¬
faced Cormorant, Crested Tern, Fairy Tern, Pacific
Gull and many migratory waders occur on marine and
estuarine wetlands and do not extend to inland saline
areas; Australasian Bittern, crakes and rails occur on
deep freshwater marshes with Cane Grass, Lignum or
reeds and Banded Stilts are found most often on hyper-
saline lakes. The distribution of many other species can
be linked more precisely to the distribition of potential
food particularly as it is influenced by salinity. The up¬
per salinity tolerances of various species of potential
food include fish (Galaxias maculatus ) 30 parts per thou¬
sand (ppt) (Chessman & Williams 1974); Gastropoda
(Coxiella) 100 ppt, Insecta 90 to 120 ppt and Isopoda
(Haloniscus) 159 ppt (Bayly & Williams 1966);
Ostracoda (Platycypris) 176 ppt and Anostraca
( Parartemia ) 298 ppt (Geddes 1976) and plants Ruppia
60 ppt (Yezdani in Aston 1973) and Lepilaena 65 ppt
(this study). Thus fish eating species (grebe, cormorant,
Pelican) will be confined to salinites of less than 30 ppt;
grazing species (Swan and Coot) will not occur at
salinities greater than 60 ppt unless they are feeding on
grassland above the waterline; species (e.g. Flardhead)
feeding on benthic organisms such as Coxiella and
chironomid larvae will be excluded from waters greater

than 100 ppt and the prey of birds (Banded Stilt and
Avocet) feeding on highly saline waters must be
restricted to only a very few species of microcrustaceans.

The few studies of primary and secondary pro¬
duction (Hammer 1970, Walker 1973, Paterson &
Walker 1974, Marchant & Williams 1977) show that pro¬
duction will be highest during summer when incident
radiation and water temperature are highest, provided
that the water regime and salinity are suitable. Thus
Marchant and Williams (1977) showed that the popula¬
tion of Parartemia in hypersaline lakes comprises
cohorts which originate from hatchings which are most
likely to occur following increases in water-level during
winter and spring but are unlikely to occur during sum¬
mer when levels are declining and Pollard (1971b) show¬
ed that the population size of Galaxias maculatus in
Lake Modewarre depends on flood flows in the spawn¬
ing areas during spring. Production of aquatic plants
such as Ruppia and Lepilaena which comprise the major
food of Coot and Swan has not been studied but clearly
turbidity and bottom type as well as season and salinity
influence their distribution and growth (e.g. Mayer &
Low 1970, Congdon & McComb 1979, Vcrhoeven
1979). It follows that variations in the usage by birds of
wetlands of the same category, or of the same wetland
from year to year may result from minor seasonal
differences in rainfall, turbidity and water chemistry and
my observations show that concentrations of birds do
occur on particular wetlands presumably when ap¬
propriate food is most abundant.

Although there are limited waterbird population
data available for other areas of Victoria (e.g. Loyn
1978, Corrick & Norman 1980, Corrick 1981) and
regional comparisons are complicated by the annual and
seasonal variations which occur in the numbers of many
species, counts made during the study indicate that the
area supports important numbers of many species.
Species for which the area is important include all col-
onially nesting species (e.g. Straw-necked Ibis, Sacred
Ibis, Pied Cormorant, Fairy Tern, but particularly
Pelican, the only known recently active colony in Vic¬
toria; Gull-billed Tern, the only breeding since 1971-72
(cf. Bourke el al. 1973) and Glossy Ibis, the first
breeding attempt since 1973 (cf. Cowling & Lowe 1981)
and for the non-colonially nesting species Brolga,
Australasian Bittern and Chestnut Teal all of which

Table 7

Number of Ducks Banded in Victoria and Recoveries by Shooting during Open Seasons from 1951
to 1980 . The percentage of bandings within the area made at Screndip Wildlife Research Station (38°01'S

144°25'E) is shown.

Banded

Total

in area

Australian Shelduck

3832

3625

Pacific Black Duck

7516

5848

Grey Teal

67736

60965

Chestnut Teal

6045

5766

Recovered

(°/o at Seren-
dip)

Total

in area

(%)

(14.2)

982

338

(34)

(74.4)

2209

517

(23.4)

(97.7)

13508

4054

(30)

(87.8)

899

281

(31)

84

A. H. CORRICK

have restricted distributions in Victoria. The area is also
important for non-breeding populations of many
nomadic or migratory species (e.g. Great Crested Grebe,
Hoary-headed Grebe, Black Swan, Grey Teal, Pink¬
eared Duck, Hardhead, Freckled Duck, Blue-billed
Duck, Musk Duck, Banded Stilt, Avocet, Coot,
Whiskered Tern and several migratory waders in par¬
ticular Curlew Sandpiper, Red-necked Stint and Sharp¬
tailed Sandpiper.

Wetlands reserved specifically for conservation in
National Parks and Wildlife Reserves include 868 ha of
deep freshwater marsh (37% of the area in the region),
1000 ha of permanent open freshwater (6%), 2870 ha of
semipermanent saline wetland (36%) including 960 ha or
60% of hypersaline lakes, 4650 ha of shallow and deep
permanent saline wetland (13%) and 190 ha of intertidal
flats (17%). Only a very small proportion of freshwater
meadows (1%) and shallow freshwater marshes (5%)
are included in these reserves. In addition large areas
(e.g. 11 300 ha of permanent open freshwater, 2700 ha
of semipermanent saline wetland and 30 100 ha of per¬
manent saline wetland) of lakes which are on public land
are Lake Reserves which recongize their value for
recreation, water supply and drainage as well as for con¬
servation. Although their future as wetland is assured,
the value of some areas to waterbirds will be lowered
unless the requirements of waterbirds are considered.
High speed or intensive boating on shallow lakes and in¬
creased activity near breeding colonies pose particular
threats. At present the large non-breeding populations
of waterbirds which use the open lakes and many of the
breeding colonies are well protected by reserves;
however, some categories (freshwater meadow and
shallow freshwater marshes) and the species which
utilize them are not. Significant areas of these latter
categories should be reserved to ensure that adequate
and representative areas of each wetland type and its
associated communities are conserved.

ACKNOWLEDGMENTS

It is a pleasure to thank Mr K. G. Bode for
technical assistance throughout the project, Mr M.
O’Sullivan for assistance during bird counts and Mrs L.
Keetch for processing data.

REFERENCES

Abbott B. & Hughes, P., 1972. Sonic birds of the Bellarine
Peninsula. Geelong Nat. 9: 47-50.

Aston, H. L, 1973. Aquatic plants of Australia. Melbourne
University Press, Melbourne.

Barkla, J. G., 1978. Winter birding on ‘‘the farm”—1978.
Bird Observer 563: 57-58.

Bayly, I. A. E. & Williams, W, D., 1966. Chemical and
biological studies on some saline Lakes in south-east
Australia. Aust. J. mar. Freshwater Res. 17: 177-288.
Bibra, E. E. & Riggs, H. C. W., 1971. Victoria River Gougings
to 1969. State Rivers and Water Supply Commission,
Melbourne.

Binns, G., 1953. Birds of Terang, South-western Victoria.
Emu 53: 211-221.

Bourke, P. R., Lowe, V. T. & Lowe, T. G., 1973. Notes on
the Gull-billed Tern. Aust. Bird Watcher 5: 69-79.

Braithwaite, L. W. 1975. Managing waterfowl in Australia.

Proc. Ecol. Soc. Aust. 8: 107-128.

Bureau of Meteorology, 1959. Climatological Survey
Region 4 -Corangamite, Victoria. Bureau of Meteor¬
ology, Melbourne.

Bureau of Meteorology, 1968. Review of Australia’s Water
Resources. Monthly rainfall and evaporation. Bureau
of Meteorology, Melbourne.

Cameron, R. J., 1979. Year Book Australia, No. 63.

Australian Bureau of Statistics, Canberra.

Campbell, A. J., 1924. The Garganey Teal Querquedula quer-
quedula. Emu 24: 146.

Carter, M. .1., 1968. A Victorian sight record of the Ruff
Rhilomachus pygnax. Aust. Bird Watcher 3: 100-102.
Central Planning Authority, 1956. Resources Survey: Cen¬
tral Highland Region. Central Planning Authority,
Melbourne.

Chessman, B. C. & Williams, W. D., 1974. The distribution of
fish in inland saline waters in Victoria, Australia. Aust.
J. mar. Freshwat. Res. 25: 167-172.

Congdon, R. A. & McComb, A. J., 1979. Productivity of Rup-
pia: Seasonal changes and dependence of light in a
Australian estuary. Aquat. Bot. 6: 121-132.

Corrick, A. H., 1980. Freckled Duck on lakes in the Western
District, Victoria. Aust. Bird Watcher 8: 254-255.
Corrick, A. H., 1981. Wetlands of Victoria II. Wetlands and
waterbirds of South Gippsland. Proc. R. Soc. Viet. 92:
187-198.

Corrick, A. H. & Norman, F. I., 1980. Wetlands of Victoria
I. Wetlands and waterbirds of the Snowy River and
Gippsland Lakes catchment. Proc. R. Soc. Viet. 91:
1-15.

Cowan, I. M., 1973. The conservation of Australian water¬
fowl. A.E.A.C. Special publication No. 2. Australian
Government Publishing Service Canberra.

Cowie, I. M., 1980. Victorian Year Book 1980. Number 100.

Aust. Bureau of Statistics Melbourne.

Cowling, S. J. & Lowe, K. W., 1981. Studies of ibises in Vic¬
toria, 1: records of breeding since 1955. Emu 81: 33-39.
Currey, D. T., 1963. The former extent of Lake Corangamite.

Proc. R. Soc. Viet. 77: 377-386.

Currey, D. T., 1970. Lake systems. Western Victoria. Aust.
Soc. Limnol. Bull. 3: 1-13.

Department of National Development 1966. Atlas of
Australian Resources. Department of National
Development, Canberra.

Frith, H. J., 1967. Waterfowl in Australia. Angus and
Robertson, Sydney.

Geddhs, M. C., 1976. Seasonal fauna of some ephemeral
saline waters in western Victoria with particular
reference to Parartemia zietziana Sayce (Crustacea:
Anostraca). Aust. J. mar. Freshwat. Res. 27: 1-22.
Gill, E. D., 1963. Rocks contiguous with the basaltic cuirass
of western Victoria. Proc. R. Soc. Viet. 77: 331-355.
Gilmore, A. M., Emison, W. B. &. Wheelf.r, J. R. 1979.
Vertebrate fauna of the Ballarat area, Victoria. Mem.
Natn. Mus. Viet. 40: 51-103.

Goodrick, G. N., 1970. A survey of the wetlands of coastal
New South Wales. CSIRO Div. Wildl. Res. Tech.
Mem. 5.

Hammer, U. T., 1970. Primary production in saline waters.

Aust. Soc. LimnoL Bull. 3: 20.

Hills, E. S. 1964. The physiography of Victoria. Whitcomb
and Tombs, Melbourne. 4th ed.

Hirtii, G. J., 1976. The birds of Trialla and Lake Coradgil,
Lesilie Manor. Victoria. Aust. Bird Watcher 6:
310-317.

WETLANDS OF VICTORIA III

85

Hounam, C. E. & Powell, P. A., 1964. Climate of the basalt
plains of Western Victoria. Bureau of Meteorology,
Melbourne.

Land Conservation Council, 1973. Report on the Melbourne
study area. Land Conservation Council, Melbourne.

Loyn, R. H., 1978. A survey of birds in Westernport Bay. Vic¬
toria 1973-74. Emu 78: 11-19.

Maddocks, Cj. E., 1967. The geochemistry of surface waters in
the western district of Victoria. Aust. .1. mar.
Freshwqt, Res. 18: 35-52.

Marchant, R. & Williams, W. D., 1977. Population dynamics
and production of a brine shrimp Parartemia zietziana
Sayce (Crustacea: Anostraca), in two salt lakes in
western Victoria, Australia. Aust. J. mar. Freshwat.
Res. 28: 417-438.

Mayer, F. L. & Low, J. B., 1970. The efleet of salinity on
widgeon-grass. J. Wildf. Mgmt. 34: 658-661.

Missen, R. & Timms, B., 1974. Seasonal fluctuations in water-
bird populations on three lakes near Camperdown,
Victoria. Aust. Bird Watcher 5: 128-135.

Morgan, D. Cj., 1954. Seasonal changes in populations of
Anatidac at the Laverton Saltworks, Victoria, 1950-53.
Emu 54: 263-278.

Parliamentary Public Works Committee, 1965. Final
Report. The Flooding of Lakes Corangamite. Gnar-
purt and Murdeduke Inquiry. Parliamentary Public
Works Committee, Victoria.

Paterson, C. G. & Walker, K. F., 1974. Seasonal dynamics
and productivity of Tarty tarsus barbitarsis Freeman
(Diptera: Chironomidae) in the benthos of a shallow,
saline lake. Aust. J. mar. Freshwat. Res. 25: 151-165.

Pollard, D. A., 1971a. Faunistic and environmental studies
on Lake Modcwarrec, a slightly saline athalassic lake in
south-western Victoria. Aust. Soc . Limnol. Bull. 4:
25-42.

Pollard, D. A., 1971b. The biology of landlocked form of the
normally catadromous salmoniform fish Galaxids
maculatus (Jcnyns). 1. Life cycle and origin. Aust. J.
mar. Freshwat. Res. 22: 91-123.

Raou, 1978. Recommended English names for Australian
Birds. Emu 77: 245-313.

Riggert, T. L., 1966. Wetlands of Western Australia ,
1964-1966. Department of Fisheries and Fauna,
Western Australia, Perth.

Smith, F. T. FL, 1962a. An Australian sight record of the Buff-
breasted Sandpiper. Aust. Bird Watcher 1: 185-192.

Smith, F. T. FL, 1962b. Some recent wader records from the
vicinity of Melbourne, Victoria. Aust. Bird Watcher 1:
211-214.

Smith, F. H. T., 1963. An Australian sight record of the Red¬
necked Phalarope {Phalaropus lobatus). Aust. Bird
Watcher 2: 1-4.

Smith, F. H. T., 1964. Wader observations in southern
Victoria, 1962-1963. Aust. Bird Watcher 2: 70-84.

Smith, F. H. T., 1965. The White-winged Black Tern in
Southern Victoria. Aust. Bird Watcher 2: 128-134.

Smith, F. H. T., 1966. Wader records and observations in
mid-southern Victoria, 1963-1965. Part 1. Aust. Bird
Watcher 2: 246-266.

Smith, F. H. T., 1967. Wader records and observations in mid¬
southern Victoria, 1963-1965. Part 2. Aust. Bird Wat¬
cher 3: 19-29.

Smith, F. H. T., 1968a. An Australian sight record of Wilson’s
Phalarope. Aust. Bird Watcher 3: 91-99.

Smith, F. FI. T., 1968b. The pectoral Sandpiper in mid¬
southern Victoria. Aust. Bird Watcher 3: 122-128.

Smith, F. H. T., 1968c. The Long-toed Stint in southern Vic¬

toria. Aust. Bird Watcher 3: 132-140.

Smiih, F. H. T. t 1969a. Waders of the Geelong District.
Geelong Nat. 6: 66-67.

Smith, F. H. T., 1969b. Field notes —a season’s waders. Bird
Observer 453: 5-7.

Smith, F. FL T., 1969c. Additional records of the Long-toed
Stint. Aust. Bird Watcher 3: 167-168.

Smith, F, H. T., 1969d. The Dunlin near Melbourne. Aust.
Bird Watcher 3: 193-195.

Smith, F. H. T., 1969c. Mallard records from Werribee
Sewerage Farm. Aust. Bird Watcher 3: 195.

Smith, F. H. T., 1970. The Dunlin —a new wader for the
Geelong district. Geelong Nat. 7:6.

Smiih, F. H. T., 1974a. A Victorian record of the Asiatic
Dowitchcr. Aust. Bird Watcher 5: 111-118.

Smith, F. H. T., 1974b. A second Victorian record of the
Asiatic Dowitchcr. Aust. Bird Watcher 5: 199-200.

Smith, F. H. T., 1976. An Australian sight record of the
White-rumped Sandpiper. Aust. Bird Watcher 6:
317-320.

Smith, F. H. T., 1977. Another Victorian record of the Buff-
breasted Sandpiper. Aust. Bird Watcher 1: 59-60.

Smith, F. H. T., 1981. A retraction of Victorian Dunlin
records. Aust. Bird Watcher 9: 43.

Smith, F. FI. T. & Swindley, R. J., 1975. A Victorian record
of Baird’s Sandpiper. Aust. Bird Watcher 6: 35-40.

Smith, F. H. T., Swindley R. J. & Barkla, J. G., 1978. A
second Australian record of the White-rumped Sand¬
piper. A ust. Bird Watcher 1: 194-197.

Sympson, R. T.. 1968. The Dunlin at Altona, Victoria. Aust.
Bird Watcher 3: 141.

Tarr, H. E., 1952. Birds of Melbourne and Metropolitan
Board of Works Farm, Werribee, and Little River.
Bird Observer Club, Monthly notes July 1952.

Timms, B. V., 1976. A comparative study of the limnology of
Three Maar Lakes in Western Victoria. I. Physio¬
graphy and Physioehemical features. Aust. ./. mar.
Freshwat. Res. 21: 35-60.

Timms, B. V. & Brand, Cj. W., 1973. A limnological survey of
the Basin Lakes, Nalangil, Western Victoria, Aus¬
tralia. Aust. Soc. Limnol. Bull. 5: 32-42.

Verhoeven, J. T. A., 1979. The ecology of A!//pp/V/-dominated
communities in Western Europe. 1. Distribution of
Ruppia representatives in relation to their autecology.
Aquat. Bot. 6: 197-268.

Walker, K. F., 1973. Studies on a saline lake ecosystem. Aust.
J. mar. Freshwat. Res. 24: 21-27.

Watson, I. M., 1955. Some species seen at the Laverton
Saltworks, Victoria, 1950-53, with notes on seasonal
changes. Emu 55: 224-248.

Wheeler, .1. R., 1953. Notes on the Blue-billed Duck at Lake
Wendouree, Ballarat. Emu 53: 280-282.

Wheeler, J. R., 1959. Little Bitterns move south. Aust.
Bird Watcher 1: 53.

Wheeler, J. R., 1976. A new wader for Australia. Geelong
Nat. 13: 52-53.

Wheeler, W. R., 1955. Char jiarii formes at the Laverton
Saltworks, Victoria, 1950-1953. Emu 55: 279-295.

Wheeler, W. R., 1964. Black-faced Cormorant in Victoria.
Aust. Bird Watcher 2: 89-90.

Wheeler, W. R., 1967. A handlist of the birds of Vic¬
toria. Victorian Ornithological Research Group,
Melbourne.

Wheeler, W R., 1975a. The Darter in southern Victoria.
Geelong Nat. 12: 57-62.

Wheeler, W. R., 1975b. Altona Survey Group 25th Anniver¬
sary. Aust. Bird Watcher 6: 134-135.

86

A. H. CORRICK

Williams, W. D., 1964. A contribution to lake typology in Vic¬
toria, Australia. Verh. int. Ver. Limnol. 15: 158-168.

Williams, W. D. & Buckney, R. T., 1976. Stability of ionic
proportions in five salt lakes in Victoria, Australia.
A ust. J. mar. Freshwat. Res. 27: 367-377.

Appendix 1

Wetland Subcategories used by less common, but regularly present Waterbirds. Most common
species have been listed in Table 6. Only those subcategories with areas greater than 100 ha arc included, x
and S, frequently recorded; (x) and (S), infrequently recorded; x and (x), throughout subcategory; S and
(S), shores and shallows only; B breeding. Names of categories and subcategories appear in Table 1.

Category

Subcategory

2

.1

3

.1

.4

4

.5

.6

.1

5

.2

.3

.1

.2

6

.3

.5

.1

7

.2.

3

Australasian Grebe

X

X

X

X

X

Australian Pelican

X

X

X

X

X

B

X

X

Great Cormorant

X

X

X

X

X

B

X

X

Pied Cormorant

X

B

X

X

Pacific Heron

X

X

(X)

X

s

S

(S)

Catte Egret

X

X

Great Egret

X

X

X

X

X

s

S

X

s

X

Little Egret

(S)

X

s

X

Rufous Night Heron

X

Australasian Bittern

X

Glossy Ibis

X

X

X

B

Sacred Ibis

X

X

X

X

s

S

B

X

Royal Spoonbill

X

X

X

X

X

S

S

X

X

S

X

Freckled Duck

(X)

X

X

X

X

Cape Barren Goose

X

X

Chestnut Teal

X

X

X

X

Maned Duck

S

Marsh Harrier

X

X

B

X

X

X

Australian Crake

X

X

X

S

X

Brolga

B

B

X

B

S

Pied Oystercatcher

X

Grey Plover

X

Lesser Golden Plover

X

X

S

X

Red-kneed Dotterel

X

X

X

X

S

S

Mongolian Plover

X

Double-banded Plover

S

X

X

X

S

X

Red-capped Plover

X

X

X

X

S

S

B

B

X

X

s

X

Black-fronted Plover

X

X

X

X

S

S

Banded Stilt

X

(S)

X

X

(S)

Avocet

X

X

X

(S)

X

X

X

(X)

Ruddy Turnstone

(X)

X

Eastern Curlew

X

Grey-tailed Tattler

X

Common Sandpiper

(X)

Grcenshank

X

s

X

Marsh Sandpiper

X

s

X

Latham’s Snipe

X

X

X

X

X

Black-tailed Godwit

(X)

X

Bar-tailed Godwit

X

Red Knot

(S)

(S)

X

Great Knot

X

Pacific Gull

X

Gull-billed Tern

X

X

X

X

X

X

X

X

X

B

(X)

Caspian Tern

(X)

(X)

(x)

X

Fairy Tern

X

Crested Tern

X

WETLANDS OF VICTORIA III

87

APPENDIX 2

Vagrant or Rarely Reported Species which Utilize
Wetlands of the Area

In some cases (e.g. Black-faced Shag, Hooded
Plover) species are more common on oiher habitats in
the area (e.g. rocky shores, ocean beaches or ocean
waters). The references cited usually assess specific
status or review past records. (* indicates the species was
seen during the study).

Darter* (Wheeler, W. R. 1975a, Gilmore et al.
1979), Black-faced Shag* (Wheeler, W. R. 1964), In¬
termediate Egret* (Watson 1955, Abbot & Hughes 1972,
Gilmore et al. 1979), Little Bittern (Wheeler, .1. R. 1959,
Gilmore et al. 1979), Wandering Whistling-Duck
(Wheeler, W. R. 1967), Plumed Whist ling-Duck*
(Wheeler, W. R. 1967, Gilmore et al. 1979), Mallard*
(Smith 1969e, Gilmore et al. 1979), Garganey leal
(Campbell 1924), Bull-banded Rail, Lewin’s Rail* and
Spotless Crake (Watson 1955, Wheeler, W. R. 1967,
Gilmore et al. 1979), Black-tailed Native-hen (Watson
1955, Gilmore et al. 1979, LCC 1973), Painted Snipe*

(Wheeler, W. R. 1967, Smith 1969a, Gilmore et al.
1979), Sooty Oystercatcher (Wheeler, W. R. 1967,
Smith 1969a), Hooded Plover *(Wheelcr, W. R. 1967),
Large Sand Plover (Smith 1964, 1969a), Oriental Plover
(Smith 1964, 1969a, Wheeler, W. R. 1967), Whimbrel
(Smith 1969a,b), Little Curlew (Smith 1964, 1969a),
Wood Sandpiper* (Smith 1964, 1969b) Terek Sand¬
piper* (Smith 1967, 1969a), Asian Dowitcher (Smith
1974a,b), Pectoral Sandpiper (Smith 1968a,b), Baird’s
Sandpiper (Smith & Swindley 1975), Long-toed Stint
(Smith 1968b, 1969c), White-rumped Sandpiper (Smith

1976, Smith et al. 1978), Dunlin (Sympson 1968, Smith
1969d, 1970, 1981), Sanderling* (Wheeler, W. R. 1955,
Smith 1969a), Buff-breasted Sandpiper (Smith 1962a,

1977, Wheeler, W. R. 1975b), Broad-billed Sandpiper
(Smith 1962b, 1969b, Wheeler, W. R. 1955), Ruff
(Carter & Smith 1968, Smith 1969b), Red-necked
Phalarope (Smith 1963, Wheeler, J. R. 1976), Wilson’s
Phalarope (Wheeler, J. R. 1976, Smith 1968a, 1969b),
White-winged Tern (Smith 1965, LCC 1973), Common
Tern* (LCC 1973) and Little Tern (LCC 1973).

PROC. R. SOC. VICT. vol. 94, no. 2, 89-106, June 1982

STUDIES ON AUSTRALIAN MANGROVE ALGAE:

II. COMPOSITION AND GEOGRAPHIC DISTRIBUTION OF
COMMUNITIES IN SPENCER GULF, SOUTH AUSTRALIA

By Warwick R. Beanland and Wm. J. Woelkerling
Department of Botany, La Trobc University, Bundoora, Victoria, Australia 3083

Abstract: This study of algal communities associated with the temperate mangrove ecosystems of
Spencer Gulf, South Australia documents the occurrence of 49 species including 10 Chlorophyta, 2
Cyanophyta, 9 Phacophyta, and 28 Rhodophyta. Pertinent morphosystematic and distributional data are
presented for each species. The Spencer Gulf mangrove algal flora is far more diverse than previously
thought but is pedestrian and depauperate compared with the southern Australian marine algal flora as a
whole. Most species are widespread on a global basis, although several typically tropical taxa also occur
on Spencer Gulf mangroves and possible explanations for their occurrence are provided. Frequency data
indicate that Calog/ossa leprieurii occurs most commonly but that most species found occur only rarely or
sporadically. Comparisons of the Spencer Gulf mangrove algal flora with those of mangrove ecosystems
elsewhere in Australia suggest that the Spencer Gulf flora is comparatively species rich and shows distinct
similarities to and diflcrences from the mangrove algal flora in Victoria.

Data on Australian tropical and warm temperate
mangrove algal communities are scant and pertain main¬
ly to scattered floristic records (and one biomass produc¬
tion estimate) in Queensland and New South Wales
(Cribb 1979, King 1981a, 1981b, 1981c, Saenger et al.
1977). Along the cool temperate southern Australian
coast, mangrove ecosystems occur as geographically dis¬
junct stands within Victoria, South Australia, and
Western Australia (Fig. 1), but the only detailed account
(Davey & Woelkerling 1980) of associated mangrove
algae deals with community composition and
geographic distribution in Victoria. Data for Western
Australia are lacking entirely.

Within South Australian mangrove ecosystems,
Cladophora sp., Enteromorpha compressa (Linnaeus)
Grcville, Hormosira banksii (Turner) Decaisne, Ulva
lactuca Linnaeus and various diatoms have been re¬
garded as common (Butler el al. 1977a, 1977b, Specht
1972, Womersley & Edmonds 1958, Wood 1937),
whereas Bostryehia and Calog/ossa , two of the most
characteristic and cosmopolitan genera of mangrove
algae (Post 1963), have been reported as apparently ab¬
sent (Womersley & Thomas 1976, Womersley 1981a).
Based on these isolated records, mangrove algal com¬
munities in South Australia would appear to differ
markedly from those in Victoria where species of
Bostryehia and Calog/ossa are the most frequently oc¬
curring algae, species of Enteromorpha and Ulva tend to
occur only sporadically, Cladophora occurs only rarely
and Hormosira appears to be absent (Davey &
Woelkerling 1980). The dearth of detailed data from
South Australia has precluded a more thorough com¬
parative assessment of these apparent differences.

Among those regions in South Australia where
mangrove ecosystems occur (Fig. 1), Spencer Gulf has
been regarded as noteworthy (Womersley 1981b) in the
sense that at least three typically tropical benthic algae
(A cel a bid aria caly cuius Quoy et Gaimard, Hormophysa

triquetra (Lamouroux) Kuetzing, Sargassum decurrens
J. Agardh) are present, presumably because summer
water temperatures are high enough for a sufficient
period to allow survival (Womersley 1981c). Whether
Spencer Gulf mangrove ecosystems also harbour
typically tropical macroscopic algae has remained
unknown as has the extent to which the algal com¬
munities in Spencer Gulf mangroves contrast with those
in more tropical regions of Australia.

This account presents results of detailed studies
on the floristic composition, frequency of species occur¬
rence and geographic distribution of mangrove algal
communities of Spencer Gulf, South Australia. It also
examines the extent to which Spencer Gulf mangrove
algal communities differ from those in Victoria and
those in more tropical regions of Australia, and it in¬
cludes comparisons of mangrove and open coast algal
communities in terms of composition, diversity, and oc¬
currences of endemic taxa.

Study Sites

The 10 mangrove algal communities (Fig. 2)
selected for detailed study included the southern-most
stands on both the eastern (Wallaroo) and western
(Tumby Bay) shores, three stands in the far north (Blan¬
che Harbour, Port Augusta, Red Cliff), two stands
along tidal creeks (Arno Bay, Port Davis) and three
other larger stands (Cowleds Landing, Franklin Har¬
bour, Port Broughton)*

In the two tidal creeks, pneumatophores occur¬
red in permanently submerged areas only at Arno Bay
(Fig. 4). Spencer Gulf mangrove ecosystems, like others
in southern Australia (Macnae 1966), are based pri¬
marily in the mid to upper eulittoral zone, are
dominated solely by Avicennia marina (Forster)
Vierhapper, and usually are associated with a salt-marsh
in the littoral-fringe (Butler*e/ al. 1977a, 1977b). Data
relating to tree height and stand size at the study sites are

89

90

W. R. BEANLAND AND Wm. J. WOELKERLING

Macnac (1966) and Saengcr ei ol. (1977).

summarized in Tabic 1. Abundant pneumatophores
(Figs 3, 4) beneath and beyond the edge of the A vicennia
canopy are the main substrate for macroscopic algae.

Study site selection also took account of certain
physical parameters which divide Spencer Gulf into two
distinct regions. The tides have a phase difference be¬
tween the far north (i.e. north of Port Davis) and the
rest of the Gulf and change with greater amplitude in the
north (Easton 1974). As a result, mean lowest tides dur¬
ing summer months tend to occur during the daytime in
the south and at night in the north (Australian National
Tide Tables 1981). Thus the mangrove algal com¬
munities south of Port Davis are emergent longer during
periods of extreme salinities, air temperatures, and

Table 1

Measurements of Avicennia Height and Fringe
Width, and Approximations of Fringe Length, at
Ten Mangrove Communities in Spencer Gulf, South
Australia

Locality

Height

m

Approximate

Fringe

Length

km

Maximum
Fringe Width
m

Arno Bay

5.0

2

25

Blanche Harbour

3.5

5

110

Cowleds Landing

4.0

30

100

Franklin Harbour

4.5

18

75

Port Augusta

3.0

2

200

Port Broughton

6.0

4

80

Port Davis

4.0

4

50

Red Cliff

5.0

5

320

Tumby Bay

1.5

3

75

Wallaroo

3.0

2

90

desiccating conditions that commonly occur at neap tide
in summer.

Water circulation in Spencer Gulf inhibits ex¬
change between the northern (i.e. north of Franklin
Harbour) and southern parts, and Tronson (1974) ex-
timates that at least two years are required for a total ex¬
change to occur. A salinity gradient ranging from 36
g/kg in the far south to 45 g/kg in the far north also oc¬
curs (Bullock 1975, Womersley 1981a: 217). While mean
monthly surface water temperatures along the central
South Australian open coast range from 14-19°C, the
range part way up Spencer Gulf is about 12-25°C and at
the head it is 13-28°C (Womersley 1981a: 216). Thus
summer surface water temperatures in the central and
northern parts of the gulf are 6-9°C warmer than those
of the adjacent open coast. The extent to which these
physical factors affect mangrove algal community com¬
position and species distribution within the Gulf has not
been examined previously, and sites for the present
study were chosen partly to obtain data relevant to such
an examination.

MATERIALS AND METHODS

Two procedures were employed to obtain entire
Avicennia pneumatophores for algal community com¬
position and frequency data from each study site during
the period March to July 1981. One involved using a
restricted random sampling regime (Goldsmith & Har¬
rison 1976) to collect 50 pneumatophores from each of
three 10 m wide, 50 m long belt transects. Each belt
transect was bisected lengthwise into two laterally con¬
tiguous 5 m wide strips contoured so that one strip was
situated just within the shaded margin of the Avicennia
canopy while the other strip lay in the adjacent sun-
exposed seaward fringe of pneumatophores beyond the

SOUTH AUSTRALIAN MANGROVE ALGAE

91

canopy (Fig. 3). In the tidal creek at Arno Bay, however,
one strip of each transect was situated along the creek
bank at low tide while the adjacent strip lay within the
permanent creek water. From each transect, 25 pairs of
random co-ordinate pneumatophores were collected;
one of each pair came from the strip under the shaded
canopy (or in the creek) while the other came from the
strip in the sun-exposed region (or along the creek
bank).

The second collecting procedure involved gather¬
ing additional pneumatophores from throughout the
mangrove fringe if they appeared to harbour algal
species not obtained during the transect sampling. This
was the only procedure used at Port Broughton, where
pneumatophores bearing algae occurred so intermit¬
tently that transect sampling was not undertaken.

All pneumatophores were preserved in 1:10 com¬
mercial formalin-seawater solution and returned to the
laboratory for analyses. Algal community composition
data were compiled from assays of all pneumatophores
collected at each locality. These data did not take ac¬
count of any diatoms (Bacillariophyta) present and in¬
cluded only those species of blue green algae
(Cyanophyta) which formed macroscopic colonies. Per¬
manent slides, liquid preserved material, and/or dried
voucher specimens have been deposited in the La Trobe
University Botany Department Herbarium (LTB —see
Holmgren & Keuken 1977).

Species frequency data for each locality except
Port Broughton were calculated from results of a species
presence/absence survey of the 150 pneumatophores
collected in the three transects by using the formula:

F

Fig. 2—Location of study sites in Spencer Gulf, S.A.

92

W. R. BEANLAND AND Wm. J. WOELKERLING

F = EN/N, where
F = the absolute frequency;

EN = the number of pneuniatophores on
which a particular alga occurred;

N = the total number of pneumatophores
surveyed.

Following Davey & Woclkerling (1980), the relative pro¬
fusion of taxa has been determined from absolute fre¬
quency data, and species have been assigned to one of
five categories: Rare (F<0.05); Sporadic (F = 0.05 to
0.24); Occasional (F = 0.25 to 0.49); Common (F -0.50
to 0.75); Abundant (F>0.75). In the text, the word
‘prevalent’ is used to include both common and abun¬
dant frequency classes.

COMMUNITY COMPOSITION AND SPECIES
DISTRIBUTION

Forty-nine species of algae, including 10
Chlorophyta, 2 Cyanophyta, 9 Phaeophyta and 28
Rhodophyta, were recorded. At any particular locality
5-26 species were detected (Table 2), and with the excep¬
tions of Port Augusta, Port Broughton and Port Davis,
red algae predominated. Taxa from all four algal Divi¬
sions were encountered in 7 of the 10 mangrove stands;
macroscopic colonies of blue green algae were not
detected at Arno Bay, Port Broughton or Port Davis
and brown algae were not observed at Port Broughton
or Port Davis. Sixteen species were found at 5 or more
localities while 15 other species were recorded from only

Table 2

Summary of the Mangrove Algal Community Com-'
position at Spencer Gulf Study Sites

Locality

chlorophyta

<

>*

O

z

<

>*

u

<

H

>*

I

CL.

c

UJ

<

X

RHODOPHYTA

) TOTAL

Arno Bay

8

_

3

13

24

Blanche Harbour

5

2

I

5

13

Cowleds Landing

5

2

4

15

26

Franklin Harbour

8

2

4

9

23

Port Augusta

5

1

4

3

13

Port Broughton

3

—

—

2

5

Port Davis

4

—

—

3

7

Red ClifT

4

2

3

12

21

Tumby Bay

7

1

1

7

16

Wallaroo

7

1

1

16

25

one locality each. Relationships between distribution
and frequency data are considered later.

COMMUNITY COMPOSITION LIST

Details of locality occurrences of species in the following
list may be obtained from Table 4, which also sum¬
marizes frequency data.

Pig. 3 _ Portion of the mangrove ecosystem at Covvleds Landing showing sun exposed and canopy shaded
regions along the seaward margin at low tide.

SOUTH AUSTRALIAN MANGROVE ALGAE

93

Fig. 4 —Portion of the mangrove ecosystem at Arno Bay showing the eulittoral (E) and sublittoral (S)
creek environment at low tide. Note emergent pneumatophores in foreground; permanently submerged
pneumatophores (not visible) also occur within the creek, which is up to 1 m deep at low tide.

Division Chlorophyta
Order Ulvales
Family Ulvaceae
Genus Enteromorpha Link 1820
Enleromorpha sp.

Specimens Examined: LTB 12190, 12233, 12243, 12259,
12281, 12292, 12316, 12327, 12348, 12359.

Remarks: Enteromopha plants occurred on
pneumatophores at all sites except Red Clift' and also
grew loose on the sun-exposed mud Hat at Tumby Bay.
Most plants were less than 2 cm tall and none could be
identified to species with confidence.

Genus Pereursaria Bory 1828

P. percursa (C. Agardh) Rosevinge 1893: 963. Abbott &
Hollenberg 1976: 70, fig. 23. Bliding 1963: 20, figs 5, 6.
Type Locality: Denmark.

Reported Distribution: Widespread.

Specimens Examined: LTB 12172, 12197, 12223, 12251,
12279, 12300, 12320.

Remarks: P. percursa occurred attached to pneu¬
matophores under the A. marina canopy and in the sun-
explosed seaward margin, and was encountered as an
epiphyte on CystophyUum onustum at Red Cliff. The
only previous record of this taxon in southern Australia
was on mangroves at Hovell’s Creek, Victoria (Davey &
Woelkerling 1980).

Genus Ulva Linnaeus 1753

U. lactuea (Linnaeus) C. Agardh, 1821: 409. Bliding
1968: 540, figs 3-5. Womerlsey 1956: 353.

Type Locality: Sweden.

Reported Distribution: Widespread.

Specimens Examined: LTB 12167, 12231, 12244, 12262,
12311.

Remarks: Plants up to 20 cm tall occurred on sun-
exposed pneumatophores at Arno Bay and Wallaroo.
Smaller specimens, up to 3 cm tall, were encountered at
Port Davis, Port Augusta and Franklin Harbour. U.
laciuca has been found on Avicennia pneumatophores
in Queensland (Cribb 1979, Saenger el al. 1977) and Vic¬
toria (Davey & Woelkerling 1980).

Order Cladophorales
Family Cladophoraceae
Genus Chaetomorpha Kuetzing 1845

C. aerea (Dillwyn) Kuetzing 1849: 379. Womersley 1956:
355.

Type Locality: Cromer, Britain.

Reported Distribution: Widespread.

Specimens Examined: LTB 12181, 12360.

Remarks: Filaments 10-20 cells long, 250-350 /*m broad
and attached by a subclavate, lobed, basal cell occurred
on pneumatophores at Arno Bay and Wallaroo.

C. capillaris (Kuetzing) Boergesen 1925: 45, fig. 13.

94

W. R. BEANLAND AND Wm. J. WOELKERLING

Feldmann 1937: 207, fig. 17. Womersley 1956: 356.
Type Locality: Nice, France.

Reported Distribution: Mediterranean and Atlantic
Ocean. In Australia from Kangaroo Island, Spencer
Gulf and Western Port.

Specimens Examined: LTB 12293.

Remarks: Entangled mats of C. capillaris occurred in
patches on pneumatophores. Davey & Woelkerling
(1980) found this species on mangroves throughout
Western Port.

Genus Cladophora Kuetzing 1843
Cladophora sp.

Specimens Examined: LTB 12176, 12189, 12214, 12246,
12260, 12278, 12301, 12312, 12344, 12356, 12361.
Remarks: Plants of Cladophora were found attached to
pneumatophores at all localities except Port Davis, and
at Red Cliff and Cowleds Landing plants also grew on
Cyst ophy Hum onus turn. Cells were mostly 40-70 /tm
broad and 150-230 /im long, but specimens were too
small (<2 cm tall) or too young to identify to species
reliably.

Genus Cladophorella Fritsch 1944

C. marina Chapman 1956: 442, fig. 92b, c.

Type Locality: Orongo Bay, New Zealand.

Reported Distribution: From Orongo Bay and Uruiti
Bay, New Zealand; South Australia.

Specimens Examined: LTB 12178, 12210, 12261, 12275,
12295, 12313, 12343.

Remarks: The Spencer Gulf collections have been refer¬
red to Cladophorella marina because their morphology
agreed more closely with that species than with other
species assigned to the genus (Table 3). Chapman (1956)
did not supply data on wall diameter or cell dimensions,
but the latter have been calculated from his figure 92b.
Plants commonly formed dense veneer-like mats on
lower portions of pneumatophores both in the sun ex¬
posed and shaded regions of the communities.
Womersley (1956, 1971) did not record this species from
southern Australia but Cribb (1965, 1979) recorded the
related C. calcicola from Queensland. The relationships
of Cladophorella Fritsch 1944 to the genus Wittrockiella
Wille 1909 require clarification and are under current
review (Van den Hoek, H.B.S. Womersley, pers.
comm.).

Genus Rhizoclonium Kuetzing 1843

R. implexum (Dillwyn) Kuetzing 1845: 206. Abbott &
Hollenberg 1976: 92, fig 45. Newton 1931: 93. Rueness
1977: 243.

Type Locality: Ireland.

Reported Distribution: Widespread.

Specimens Examined: LTB 12185, 12234, 12252, 12294,
12310.

Remarks: Plants occurred on Avicennia pneu¬
matophores at Arno Bay, Franklin Harbour, Port Davis
and Tumby Bay and also occurred entangled with
Bostrychia and Percursaria at Wallaroo. Most filaments
were 17-22 ym broad, unbranched, and composed of

cells 2-6(-9) diameters long, each containing a reticulate
chloroplast mostly with 5-8 pyrenoids. The Spencer GU|f
plants agreed with the concept of R. implexihn
presented by Abbott & Hollenberg (1976) and Rucn£$s
(1977). This species apparently has not been recorded
previously from southern Australia (Womersley 1956, b.
361, 362).

R. riparium (Roth) Harvey 1849: pi. 239. Abbott &
Hollenberg 1976: 92, fig. 46, pi. 1, fig. 9. Rueness 1977:
243. Womersley 1956: 361.

Type Locality: Northern Europe.

Reported Distribution: Widespread.

Specimens Examined: LTB 12182, 12196, 12212, 1222$,
12241, 12270, 12282, 12308, 12338.

Remarks: Rhizoclonium riparium occurred On
pneumatophores amongst other algae at all sites excebt
Blanche Harbour and Port Broughton and was abun¬
dant at Arno Bay and Port Davis. Most filaments wefe
20-50 yin broad, rarely possessed rhizoidal branches,
and were composed of cells 1-2.5 (-4) diameters lon^;
each contained a reticulate chloroplast with 10 or more
pyrenoids. The Spencer Gulf plants agreed most closely
with the concept of R. riparium presented by Rueness
(1977). Californian plants (Abbott & Hollenberg 197$)
appear to be more robust and possess more rhizoidal
branches. Davey & Woelkerling (1980) reported this
species from four mangrove communities in Victoria.

Order Chaetophorales
Family Ckaetophoraceae
Genus Sporocladopsis Nasr 1947

S. novaezelandiae Chapman 1949: 496, fig. 4; 1956: 433,
fig. 85a-c.

Type Locality: Bay of Islands, New Zealand.
Reported Distribution: North Island of New Zealand;
Queensland, southern Australia.

Specimens Examined: LTB 12288, 12319, 12336.
Remarks: Sterile plants occurred on pheumatophores as
inconspicuous epiphytes. This species was not recorded
from southern Australia by Womersley (1956, 1971).

Division Cyanophyta
Order Oscillatoriales
Family Rivulariaceae
Genus Rivularia Roth 1797

R. atra Roth 1806: 340. Drouet 1973: 164; 1978: 237.
Umezaki 1961: 105, pi.21, fig. 2a-c. Womersley 1946:
132.

Type Locality: West Germany.

Reported Distribution: Widespread.

Specimens Examined: LTB 12187, 12202, 12269, 12276,
12296, 12314, 12333.

Remarks: R. atra formed solid colonies, up to 1 cm
diameter on pneumatophores in the sun-exposed and
shaded regions of the communities. Trichomes were
3-12 ym in diameter, sheathed in a thick hyaline coat,
usually bore heterocysts of 10-12/xm diameter, and
tapered to a thin hair. Drouet (1973) placed this species
in synonomy with Calothrix Crustacea Schousboe &
Thuret.

SOUTH AUSTRALIAN MANGROVE ALGAE

95

R. polyotis (J. Agardh) Bornet & Flahault 1886: 360.
Umezaki 1961: 103. Womerslcy 1946: 134.

Type Locality: Mediterranean coast of France.
Reported Distribution: Widespread.

Specimens Examined: LTB 12203, 12277, 12315, 12334.
Remarks: R. polyotis occurred on pneumatophores;
trichomes, 2-8 /*m in diameter, were sheathed in a thick
hyaline coat, bore basal heterocysts 6-10 /un in diameter,
and usually tapered to a thin hair. Plants formed soft,
hollow colonies up to 2 cm in diameter.

Division Phaeophyta
Order Ectocarpales
Family Ectocarpaceae
Genus Ectocarpus Lyngbye 1819

E. siliculosus (Dillwyn) Lyngbye 1819: 131. Russell
1966: 275 et seq ., figs. 3, 4. Womerslcy 1967: 190.

Type Locality: Europe.

Reported Distribution: Widespread in temperate and
boreal seas.

Specimen Examined: LTB 12253.

Remarks: E. siliculosus occurred sporadically on con¬
tinually submerged pneumatophores, grew up to 2 cm
tall, and bore plurilocular sporangia. E. siliculosus also
has been recorded from Victorian mangrove com¬
munities (Davey & Woelkerling 1980).

Ectocarpus sp.

Specimen Examined: LTB 12267.

Remarks: Two small plants with plurilocular sporangia
were found in the sun-exposed seaward margin at Port
Augusta. Cells were 16-20 /im broad, 20-35 /im long and
plurilocular sporangia, borne terminally, were 50-80 /on
long and 25-30/mi broad. The plants were insufficiently
developed for reliable species identification.

Genus Giffordia (Batters) Hamel 1939

G. sordida (Harvey) Clayton 1974: 785, fig. 12G1, G2,
15A-F, 26A-K.

Type Locality: Georgetown, Tasmania.

Reported Distribution: Southern Australia, Queens¬
land.

Specimen Examined: LTB 12362.

Remarks: G. sordida was abundant at Wallaroo in
June, occurring attached to pneumatophores and/or in¬
termixed with other algae in all regions of the com¬
munity. Plants up to 15 cm tall, with filaments 20-50 /im
in diameter were collected. Unilocular sporangia occur¬
red rarely.

Genus Kuetzingiella Kuckuck 1956
Kuetzingiella sp.

Specimens Examined: LTB 12245, 12266, 12280, 12289,
12317, 12337.

Remarks: Filaments consisted of 20-30 cells approx¬
imately 10 /cm broad and 5 /an long. Plurilocular
sporangia 100-200 /an long and 18-22 /an broad usually
were present. According to Clayton (1974), the tax¬
onomy of southern Australian Kuetzingiella species is
uncertain and in need of revision.

Order Sphacelariales
Family Sphacelariaceae
Genus Spliacelaria Lyngbye 1819
S. furcigera Kuetzing 1855: 27, pi.90. Womersley 1967:
199.

Type Locality: Karak Is., Persian Gulf.

Reported Distribution: Widespread.

Specimens Examined: LTB 12200, 12254, 12263, 12303.
Remarks: Plants up to 1 cm tall were found on
pneumatophores at Red Cliff, Port Augusta, Franklin
Harbour and Arno Bay. They also grew as epiphytes on
Cystophyllum onustum at Red Cliff and Cowleds Land¬
ing. Some plants bore plurilocular or unilocular
sporangia and most produced long-armed, biradiate
propagules.

S. tribuloides (Meneghini) De Toni 1895: 502. Taylor
1960: 211, pi.29 fig 6. Womersley 1967: 201.

Type Locality: Italy.

Reported Distribution: Widespread in tropical and
temperate seas.

Specimens Examined: LTB 12201, 12264, 12304.
Remarks: Several plants up to 1 cm tall were found on
pneumatophores at Port Augusta, Red Cliff' and
Franklin Harbour. Those at Red Cliff were intermixed
with S. furcigera . Short-armed, triradiate propagules
occurred on most plants.

Order Dictyotales
Family Dictyotaceae
Genus Dictyota Lamouroux 1809

Dictyota sp.

Specimens Examined: LTB 12214, 12351, 12363.
Remarks: Sterile plants, up to 3 cm tall, were epiphytic
on Cystophyllum onustum at Red Cliff and Cowleds
Landing, and were attached to pneumatophores in the
sun-exposed seaward margin at Wallaroo in June.
Plants were insufficiently developed for reliable species
identification.

Order Fucales
Family Cystoseiraceae
Genus Cystophyllum J. Agardh 1848.

C. onustum (Mertens) J. Agardh 1848: 230. Womersley
1967: 254.

Type Locality: Western Australia.

Reported Distribution: All around Australia; Indian
Ocean.

Specimens Examined: LTB 12214, 12349.

Remarks: A large, solitary plant was attached to the
base of a mangrove tree at Red Clift'; a second plant was
found attached to shells under the canopy at Cowleds
Landing. Both plants harboured a number of epiphytes.

Family Hormosiraceae
Genus Hormosira (Endlicher) Meneghini 1838

H. banksii (Turner) Decaisne 1842: 331. Clarke &
Womersley 1981: 497 eqseq. King 1981a: 325, figs 9, 12;
1981b: 107; 1981c: 569 et seq. Womersley 1967: 249-250.
Type Locality: ‘Novae Hollandiae’.

96

W. R. BEANLAND AND Wm. J. WOELKERLING

Reported Distribution: From King George Sound,
Western Australia, around southern Australia; New
South Wales; Lord Howe Island; Norfolk Island; New
Zealand.

Specimens Examined: LTB 12166, 12199, 12224, 12225,
12326, 12329, 12347.

Remarks: Plants up to 25 cm tall with swollen vesicles,
were attached to the lower half of pneumatophores and
loose lying on mud surfaces at Wallaroo, Red Cliff,
Franklin Harbour, and Cowlcds Landing. The loose ly¬
ing form was also encountered at Tumby Bay. King
(1981a, 1981b, 1981c) provided data on an extensive
loose lying community of H. banksii within the
mangroves of southern Botany Bay in New South
Wales; he estimated that mean dry weight biomass
ranged from 280 g m~ 2 in late winter (August) to 638 g
nr 2 in mid-summer and suggested an annual biomass
production rate of approximatley 400 g nr 2 . Clark &
Womersley (1981, p. 500) reported an unattached
population of plants occurring among mangroves at
Port Arthur, South Australia at the north end of St.
Vincent Gulf. Individuals up to 50 cm long were en¬
countered.

Division Rhodophyta
Order Bangiales
Family Erythropeltidaceae
Genus Aslerocylis (Hansgirg) Schmitz 1896.

A. ornata (C. Agardh) Hamel 1925: 40.

A. ramosa (Thwaites) Schmitz 1896: 314. Abbott
& Hollenberg 1976: 283. Taylor 1960: 287.

Type Locality: British Isles.

Recorded Distribution: Widespread in temperate
seas.

Specimens Examined: LTB 12211, 12257, 12302, 12340,
12358, 12364.

Remarks: A. ornata was attached to pneumatophores in
the sun-exposed seaward margin at Port Broughton and
Cowlcds Landing and occurred at Wallaroo in June but
not in March. Plants were also encountered as epiphytes
on Chonclria sp. at Arno Bay and Wallaroo, on
Polysiphoma leges and Sphacelaria furcigera at Red
Cliff, and on Cystophyllum onustum at Cowleds Land¬
ing.

Genus Erythrolrichia Areschoug 1850

E. carnia (Dillwyn) J. Agardh 1883: 15. Abbott &
Hollenberg 1976: 286, fig. 228. Newton 1931: 242.
Taylor 1960: 291.

Type Locality: Wales, Gt. Britain.

Reported Distribution: Widespread.

Specimen Examined: LTB 12247.

Remarks: E. cornea grew on pneumatophores, on Ulva
lactuca and on Chondria sp.

Order Nemalionales
Family Acrochaetiaceae
Genus Audouinella Bory 1823

A. botryocarpa (Harvey) Woelkerling 1971: 37. Searles
& Schneider 1978: 1 (X).

Type Locality: King George Sound, W. Australia.
Reported Distribution: Bunbury, Western Australia
to Point Lonsdale, Victoria, and Tasmania; New
Zealand; North Carolina.

Specimen Examined: LTB 12248.

Remarks: Plants were attached both to continually
submerged and mud-flat pneumatophores at Arno Bay,
and usually bore monospores. Woelkerling (1970) pro¬
vided a detailed account of this alga.

A. daviesii (Dillwyn) Woelkerling 1971: 28, figs 7A-J,
22A-B; 1973: 550, fig. 32-43.

Type Locality: Bantry Bay, Ireland.

Reported Distribution: Widespread.

Specimens Examined: LTB 12249, 12365.

Remarks: Plants of A. daviesii grew attached to con¬
tinually submerged and mud-flat pneumatophores and
also on Ulva lactuca at Arno Bay. Plants were en¬
countered at Wallaroo in June but not in March.

A. savianna (Meneghini) Woelkerling 1973: 560-565, figs
56-60.

Type Locality: Genoa, Italy.

Reported Distribution: Widespread.

Specimen Examined: LTB 12250.

Remarks: Monosporangial plants were found attached
to continually submerged and mud-flat pneumatophores
at Arno Bay. A detailed account of this taxon in
southern Australia was provided by Woelkerling (1971)
using the name A. thuretii (Bornet) Woelk. Subsequent
comparisons of the type collections of A. savianna and
A . thuretii (Woelkerling 1973) indicated that the two
taxa were conspecific, with A. savianna having priority.

Order Gelidiales
Family Gelidiaceae

Genus Gelidiella Feldmann & Hamel 1934
G. nigrescens (Feldmann) Feldmann & Hamel 1934:
533. Feldmann & Hamel 1937: 222, fig. 7.

Type Locality: Algeria.

Reported Distribution: Uncertain.

Specimens Examined: LTB 12175, 12323.

Remarks: Dense stands of tetrasporangial plants col¬
onized the lower half of pneumatophores at Wallaroo.
Plants were more common under the canopy than in
sun-exposed regions. A single plant also was collected at
Franklin Harbour. The two species of Gelidiella record¬
ed in this study closely lit the reproductive and mor¬
phological descriptions in Feldmann and Hamel (1934,
1937), and apparently have not been recorded previously
from southern Australia.

G. lemiissima Feldmann & Hamel 1937: 226, figs 11,
12A-E.

= G. panosa (Bornet) Feldmann & Hamel 1934:
534.

Type Locality: Biarritz, France.

Reported Distribution: Uncertain.

Specimens Examined: LTB 12173, 12207, 12238, 12335,
12366.

Remarks: G. tenuissima was common under iheAvicen-
nia canopy at Wallaroo and Arno Bay and occurred in¬
frequently at Cowleds Landing, Port Augusta and Red

SOUTH AUSTRALIAN MANGROVE ALGAE

97

ClifL Plants were usually fertile and bore either
cystocarps or tetrasporangia.

Genus Gelidium Lamouroux 1813

G. pusillum (Stackhouse) Le Jolis 1863: 139. Chapman
1969: 89. Dixon & Irvine 1977: 129, fig. 48A-J. May
1965: 371.

Type Locality: England.

Reported Distribution: Widely distributed in tropical
and temperate waters.

Specimen Examined: LTB 12367.

Remarks: Sterile plants occurred infrequently on
pneumatophores under the Avicennia canopy.

Order Cryptonemiales
Family Corallinaceae

Remarks: With few exceptions, southern Australian
representatives of this family are poorly known and
never have been the subject of monographic studies.
Moreover, species concepts among nongeniculate taxa
generally are rather confused and many questions con¬
cerning generic concepts also remain unanswered. Con¬
sequently the corallines found during the study have not
been identified to species, and placement into genera is
based on concepts presented by Johansen (1981).

Genus Heteroderma Foslic 1909
Heteroderma sp.

Specimens Examined: LTB 12368, 12369.

Remarks: Heteroderma sp. was common on pneu¬
matophores at Wallaroo, and occurred infrequently at
Cowleds Landing. Many plants possessed either female
or tetrasporangial conceptacles.

Genus Jama Lamouroux 1812

Jania sp.

Specimen Examined: LTB 12372.

remarks: One small sterile plant, 5 mm tall, was found

at Wallaroo in July.

Genus Lithothamnium Philippi 1837

Lithothainnium sp.

Specimen Examined: LTB 12318.

Remarks: Specimens occurred on the lower portion of
pneumatophores under the canopy at Franklin Har¬
bour. Most plants bore either female or tetrasporangial
conceptacles. The concept of Lithothamnium as a genus
is under review (Woclkerling 1981).

Genus Neogoniolithon Setchell & Mason 1943
Neogoniolithon sp.

Specimens Examined: LTB 12370, 12371.

Remarks: Crusts up to 4 mm thick occurred on
pneumatophores in sun-exposed and shaded regions.
Most plants had male or female or tetrasporangial con¬
ceptacles.

Genus Phymatolithon Foslie 1898

Phymatolithon sp.

Specimen Examined: LTB 12373.

Remarks: Tetrasporangial plants encrusted the lower
portions of pneumatophores.

Order Ceramiales
Family Ceramiaceae
Genus Centroeeras Kuetzing 1841
C. clavulatum (C. Agardh) Montagne 1846: 140. Abbott
& Hollenberg 1976: 604, fig. 547. May 1965: 371. Taylor
1960: 537.

Type Locality: Caloa, Peru.

Reported Distribution: Widely distributed in tropical
and temperate seas.

Specimens Examined: LTB 12188, 12209, 12217, 12256,
12286, 12321, 12357.

Remarks: Sterile plants up to 3 cm tall were found en¬
tangled with other algae at Wallaroo, and attached to
pneumatophores at Red Cliff, Franklin Harbour, Arno
Bay and Tumby Bay. Plants were epiphytic on
Cyst Ophy Hum onustum at Red Cliff and Cowleds Land¬
ing. No sun or shade preference was evident. Cribb
(1979) recorded this species from Queensland
mangroves.

Genus Spyridia Harvey 1833
S. filamentosa (Wulfen) Harvey 1833: 336. Womersley
& Cartledge 1975: 222, fig. 1A-D.

Type Locality: Adriatic Sea.

Reported Distribution: Widespread in tropical and
temperate seas.

Specimens Examined: LTB 12170, 12198, 12221, 12283,
12285, 12309, 12325, 12354.

Remarks: Plants usually grew on the lower halt of
pneumatophores in both the sun-exposed and shaded
regions of the community. S. filamentosa was en¬
countered growing on mud surfaces and attached to
shells at Tumby Bay and Red Cliff, and as an epiphyte
on Cystophyllum onustum at Red Cliff and Cowleds
Landing. All specimens were sterile. This species occurs
on mangroves in Queensland (Cribb 1979, Sacnger et al.
1977).

Family Rhodomelaceae
Genus Bostrychia Montagne 1842
B. moritziana (Sonder in Keutzing) J. Agardh 1863: 862.
Post 1963: 57; 1964: 244.

Type Locality: French Guiana.

Reported Distribution: Widespread in tropical and
temperate seas.

Specimens Examined: LTB 12180, 12195, 12230, 12273,
12305, 12330.

Remarks: B. moritziana was locally abundant at
Wallaroo, Port Davis, Red Cliff, Blanche Harbour,
Cowleds Landing and Franklin Harbour. Plants were
most common under the Avicennia canopy and often
were intermixed with B . radicans. Tetrasporangial
plants were rare, and cystocarpic or male plants were
not found. Davey and Woelkerling (1980) found this to
be the most widely distributed species of Bostrychia in
Victorian mangrove ecosystems, and Cribb (1979) and
Saenger et al. (1977) recorded the species from
Queensland.

98

W. R. BEANLAND AND Wm. J. WOELKERLING

B. radicans (Montagne) Montagne 1850: 286.

Type Locality: Sinnamary, French Guiana.

Reported Distribution: Widely distributed in tropical
and temperate seas.

Specimens Examined: LTB 12177, 12194, 12229, 12236,
12272, 12287, 12306, 12334.

Remarks: B. radicans was the most abundant and
widespread species of Bostrychia in Spencer Gulf, oc¬
curring at all sites except Port Broughton and Port
Augusta. Plants often formed mats on pncumatophores
under the Avicennia canopy and in some cases bore
spermatangia, cystocarps or tetrasporangia. The only
previous record of this taxon in southern Australia was
from the mangrove environment in Victoria (Davey &
Woelkerling 1980); Saenger et al. (1977) and Cribb
(1979) recorded this species from Queensland.

Genus Caloglossa J. Agardh 1876

C. leprieurii (Montagne) J. Agardh 1876: 499.

Type Locality: Cayenne, French Guiana.

Reported Distribution: Widespread in tropical and
temperate seas.

Specimens Examined: LTB 12179, 12193, 12232, 12235,
12258, 12274, 12307, 12331.

Remarks: C. leprieurii was often in association with
species of Bostrychia and sometimes forming dense,
pure stands on pneumatophores in both sun-exposed
and shaded regions. Tetrasporangial and cystocarpic
plants were found on occasions. This species occurs
commonly on mangrove pneumatophores elsewhere in
Australia (see Cribb 1979, Davey & Woelkerling 1980).

Genus Chondria C. Agardh 1817
Chondria sp.

Specimens Examined: LTB, 12204, 12240, 12291,
12374.

Remarks: Chondria sp. occurred infrequently at
Wallaroo, Tumby Bay and Red Cliff. Plants were at¬
tached to pneumatophores in the sun-exposed seaward
margin, and were epiphytic on Cystophyllum onustum
at Cowleds Landing and on Ulva lacluca at Arno Bay.
Plants were insufficiently developed for reliable species
identification.

Genus Diplocladia Kylin 1956

D. patersonis (Sonder) Kylin 1956: 504. May 1965: 383.
Type Locality: Cape Paterson, Victoria.

Reported Distribution: South Australia, Tasmania,
Victoria.

Specimens Examined: LTB 12171, 12290, 12324.
Remarks: D. patersonis plants up to 10 cm tall were col¬
lected both in sun-exposed and shaded areas at
Wallaroo, Franklin Harbour and Tumby Bay. The
Wallaroo samples collected during winter (June) bore
tetrasporangia. Davey & Woelkerling (1980) reported
this species on mangroves at two localities in Victoria.

Genus Herposiphonia Nageli 1846

Herposiphonia sp.

Specimens Examined: LTB 12339, 12375.

Remarks: The two collections of Herposiphonia obtain¬
ed during this study probably are referrable to different
species but both were sterile, and as noted by Abbott &
Hollenberg (1976: 720) such specimens often are difficult
to identify to species level with confidence. Plants (sp.
“A”) in the Cowleds Landing collection (LTB 12339)
had determinate branches 40-60 /*m in diameter, which
were 7-14 segments long and had 6-7 pericentral cells.
These specimens appeared to be most similar to //.
delicatula Hollenberg (1968: 540). Plants (sp. “B”) in the
Wallaroo collection (LTB 12375) had determinate bran¬
ches 70-80 /im in diameter which were mostly 12-18
segments long and had 8-9 pericentral cells. These
specimens appeared to be most similar to H. tenella f.
secunda (C. Agardh) Hollenberg (1968).

Genus Laurencia Lamouroux 1813
Laurcncia sp.

Specimens Examined: LTB 12168, 12208, 12216, 12355.
Remarks: Sterile plants up to 3 cm tall were found on
pneumatophores at Wallaroo and Red Cliff, and other
plants occurred epiphytically on Cystophyllum onustum
at Red Cliff and Cowleds Landing. Species identification
could not be made with certainty but these specimens
appeared to be most similar to Laurencia shepherd'd
Saito & Womersley (1974).

Genus Lophosiphonia Falkenberg in Schmitz &
Falkenberg 1897

L. subadunea (Kuetzing) Falkenberg 1901: 496, pi.9,
figs. 21-24. Cribb 1956: 139. May 1965: 380. Taylor
1960: 605.

Type Locality: Corsica.

Reported Distribution: Arabia; southern Australia;
Bahamas; Mediterranean; Queensland; Texas.
Specimen Examined: LTB 12376.

Remarks: L. subadunea was common at Wallaroo in
June, but absent in March. Plants were sterile, up to 2
cm tall, and occupied the lower portion of A. marina
pneumatophores under the canopy. According to Cribb
(1956) L. subadunea has not been recorded from
southern Australia.

Genus Polysiphonia Greville 1824

P. infestans Harvey 1855: 539. Womersley 1979: 481,
fig. 6A-E.

Type Locality: Princess Royal Harbour, King George
Sound, W. Australia.

Reported Distribution: From North Beach Reef,
Perth, southwards and along the southern Australia
coast. Botany Bay, New South Wales.

Specimens Examined: LTB 12232, 12299.

Remarks: Sterile plants up to 3 cm tall, were en¬
countered on the lower portion of pneumatophores.

P. scopulorum Harvey 1855: 540. Womersley 1979: 467,
fig. 2A-E.

Type Locality: Rottnest Island, W. Australia.
Reported Distribution: From Rottnest Island, W.
Australia to Lawrence Rock, Victoria.

Specimens Examined: LTB 12205, 12218.

SOUTH AUSTRALIAN MANGROVE ALGAE

99

Table 3

Morphologic Affinities of the Spencer Gulf Cladophorella With Other Described Species

Species of
Cladophorella

Cell

dia.

range

(pm)

Cell

length:

width

ratio

Cell
wal 1
struct¬
ure

Cell

wall

dia.

(pm)

Reported habitat and
references

C. oalcicola

23-39

3-10

Lamellate

8

Hot house walls, Cambridge Botanical
Gardens (Fritsch 1944). A . marina
mangrove environment in Queensland
(Cribb 1979). Semi-marine cavern in
Queensland (Cribb 1965).

C. frit8ohii

55-88

1 *3-3*5

Lamell ate

3-15

Freshwater environment in East

Pakistan (Islam 1964).

Co sundarbanensis

15-55

2*5-6

Not

Lamellate

2-4

Mangrove environment in Bangladesh
(Islam 1973). Brackish water in

East Pakistan (Islam 1964).

C . marina

80-240

1-5

Lamel late

Not

Supplied

Marine environment, New Zealand
(Chapman 1956).

Spencer Gulf
Cladophorella

140-250

1-4

Lamell ate

10-50

Avicennia marina mangrove environment
in Spencer Gulf.

Remarks: Sterile plants, 2 cm tall, were epiphytic on C.
onustum under the canopy, and attached to a single
pneumatophorc in a tidal creek, at Red Cliff. Cribb
(1979) recorded this species from Queensland
mangroves.

P. subtilissima Montagne 1840: 199. Womersley 1979:
469, fig. 2F-I.

Type Locality: Cayenne, French Guiana.

Reported Distribution: Tropical and sub-tropical
Eastern America, French Guiana, Hawaiian Islands; in
southern Australia from Coffin Bay, South Aust., to
Port Phillip Bay, Victoria; Tasmania; Botany Bay, New
South Wales.

Specimen Examined: LTB 12255.

Remarks: Tetrasporangial plants, up to 3 cm tall, were
attached to continually submerged and mud-flat
pneumatophores at Arno Bay.

P. teges Womersley 1979: 494, fig. 10A-C.

Type Locality: Frenchmans Bay, Albany, W.
Australia.

Reported Distribution: Type locality and Spencer
Gulf, S.A.

Specimens Examined: LTB 12174, 12265, 12284, 12352.

Remarks: Specimens occurred epiphytically on
Cystophyllum onustum at Cowleds Landing and on the
lower portions of pneumatophores at Wallaroo, Port
Augusta, and Blanche Harbour. Tetrasporangial plants
were collected at the latter two locations. The only
previously known specimens from Spencer Gulf were
epilithic (Womersley 1979, H. B. S. Womersley pers.
comm.).

FREQUENCY DATA

Frequency data for the 42 species collected from
the belt transects are summarized in Table 4. Seven taxa
{Bos try chi a rad icons, Caloglossa leprieurii , Clad¬
ophorella marina , Enteromorpha sp., Gelidiella
tenuissima , Rhizoclonium riparium , Rivularia atra) oc¬
curred commonly (F = 0.50 to 0.75) or abundantly
(F>0.75) at one or several localities, and based on mean
frequency values [i.e. LF/N, where LF is the sum of all
recorded frequencies >0 and N is the total number of
localities at which the alga occurred; see Table 4],
Caloglossa leprieurii is the most conspicuous alga in
Spencer Gulf mangrove ecosystems. Although all seven
taxa were recorded from a majority of study sites, none

100

W. R. BEANLAND AND Wm. J. WOELKERLING

Table 4

Frequency Data for Algae Associated with Spencer Gulf Mangroves.
Taxa found outside sampling transects are recorded as present (P).

Taxon

ARNO BAY

BLANCHE

HARBOUR

COWLEDS

LANDING

FRANKLIN

HARBOUR

PORT

AUGUSTA

PORT

BROUGHTON

PORT DAVIS

RED CLIFF

TUMBY BAY

WALLAROO

MEAN

FREQUENCY

(F)

CHL0R0PHYTA

Chaetanorpha aerea

0.11

0.09

0.10

C. capillarie

0.05

0.05

Cladophora Sp.

0.17

0.27

0.02

0.20

0.21

P

0.03

0.04

0.26

0.16

Cladophorella marina

0.31

0.02

0.75

0.17

0.06

0.01

0.18

0.48

0.25

Enteranorpha sp.

0.27

0.26

0.02

0.22

0.57

P

0.55

0.30

0.31

Percurearia percursa

0.33

0.30

0.02

0.14

0.15

0.33

0.21

Rhizocloniun implexum

0.06

0.05

0.06

0.34

0.06

0.11

R. riparian

0.52

0.06

0.44

0.37

P

0.84

0.26

0.04

0.13

0.33

Sporocladopsis novaezelandiae

0.05

0.02

0.04

0.06

Ulva lactuca

0.24

0.24

0.02

0.02

0.13

0.13

CY-AN0PHYTA

Rivularia atra

0.09

0.54

0.34

0.49

0.20

0.05

0.50

0.31

R. polyotiz

0.11

0.26

0.02

0.22

0.15

PHAE0PHYTA

Cystophyllum onus turn

P

P

Dictyota sp.

P

P

P

Ectocarpus eiliculosue

0.02

0.02

Ectocarpus sp.

0.02

0.02

Giffordia sordida

P

Rormosiva bankeii

0.01

0.01

0.09

0.04

Kuetzingiella sp.

0.07

0.02

0.05

0.05

0.03

0.18

0.07

Sphacelaria furcigera

0.09

0.11

0.02

0.01

0.12

S. tribuloidee

0.04

0.02

0.01

0.04

was encountered at all nine transect localities.
Moreover, all seven taxa also occurred only rarely
(F<0.05) or sporadically (F = 0.05 to 0.24) at one or
more localities and except at Wallaroo, frequency values
for Gelid iella tenuissima never exceeded 0.13.

Nine additional taxa (Bostrychia moritziana ,
Centroceras clavulatum , Chondria sp., Cladophora sp.,
Kuetzingiella sp., Percursaria percursa , Rhizoclonium
implexum , Spyridia filamentosa , Viva lactuca) also oc¬
curred at a majority of study sites, but with relatively
few exceptions, these species occurred rarely (F<0.05)
or sporadically (F —0.05 to 0.24) and none was ever
common or abundant.

In addition to the above 16 taxa, 4 ( Diplocladia
pa t arson is y Gel id iella nigrescens , Heteroderma sp.,
Neogoniolilhon sp.) occurred with occasional frequen¬
cies (F = 0.25 to 0.49) at Wallaroo. Overall, however,
these algae occurred at only 1-3 study sites and were rare
or sporadic except at Wallaroo. The remaining 21 algae
were detecied only at a minority of study sites, always
were rare or sporadic, and thus appear to be of relatively
little significance in Spencer Gulf mangrove ecosystems.

At a particular locality, 3 to 12 species occurred
with frequencies of 0.25 or more, and except for Blan¬
che Harbour and Tumby Bay, at least one species was
common (F = 0.50 to 0.75) or abundant (F>0.75). Four

of the 7 species found at Port Davis were common or
abundant, while other localities only 1 or 2 common or
abundant taxa were present. At all study sites except
Port Broughton, either Caloglossa leprieurii or
Bostrychia radicans or both occurred with frequencies
of 0.25 or more. Frequency data were not obtained for
Port Broughton, but neither Bostrychia nor Caloglossa
plants were found there. At Wallaroo, eight red algae
with frequencies of 0.25 or more were present; no more
than 3 such red algae occurred at any other site.
Similarly 5 green algae with frequencies of 0.25 or more
grew at Arno Bay while at all other localities no more
than 3 such species occurred. Blue green algae became
occasional to abundant (F>0.25) at Cowleds Landing,
Franklin Harbour, Port Augusta, and Wallaroo but
brown algae occurred only rarely or sporadically at
localities where they were present.

DISCUSSION

The marine algal flora of cool temperate
Australia is rich and diverse in species, often with a
biomass predominance of larger brown algae (especially
Fucales) and a high percentage (71%) of endemic genera
(Womersley 1981b). This contrasts with the com¬
paratively depauperate algal flora in Spencer Gulf
mangrove ecosystems, mostly consisting of red algae

SOUTH AUSTRALIAN MANGROVE ALGAE

101

Table 4 (Continued)

Taxon

UJ cc

58

LU >—i
—I Q

Si

►- 8

88
CL <

8S

O. CO

52-b

Audouinella botryoaarpa

0.09

0.09

A. dainesii

0.06

0.06

A. savicama

0.03

0.03

Asterocytis omata

0.01

0.02

P

0.02

0.01

0.02

Boetrychia moritsiana

0.10

0.16

0.26

0.46

0.20

0.09

0.21

B. radicans

0.44

0.15

0.12

0.50

0.30

0.43

0.36

0.33

CalogloBsa leprieurii

0.72

0.25

0.45

0.56

0.60

0.77

0.58

0.08

0.50

Centrocerae clavulatwt

0.01

0.04

0.06

0.22

0.06

0.08

Chondria sp.

0.12

0.01

0.03

0.04

0.03

0.05

Diplocladia patersonis

0.01

0.02

0.28

0.10

Erythrotrichia cornea

0.18

0.18

Gelidiella nigrescens

0.32

0.32

G. tenuissima

0.13

0.07

0.01

0.03

0.51

0.15

Gelidiitn pusillum

P

Berposiphonia sp. “A"

0.05

0.07

0.06

Berposiphonia sp. “B“

0.07

0.10

Beteroderma sp.

0.07

0.32

0.17

Jania sp.

P

Laurenoia sp.

0.04

0.08

0.06

Lithotharnniien sp.

0.12

0.12

Lophosiphonia subadunca

P

Beogoniolithon sp.

0.16

0.30

0.23

Phymatolithon sp.

0.03

0.03

Polysiphonia infestans

0.02

0.04

0.03

P. saopulonen

P

P. subtilissima

0.05

0.05

P. tegee

0.04

P

0.03

0.24

0.10

Spyridia filamentosa

0.09

0.02

0.01

0.06

P

0.17

0.23

0.25

0.12

which have wide distributions. Based on published
species estimates (Womersley 1981b: 301) only 10
(9.8%) of the 97 Chlorophyta, nine (4.4%) of the 203
Phaeophyta and 28 (3.5%) of the 800 Rhodophyta
species known to occur in southern Australia seas were
encountered in Spencer Gulf mangrove ecosystems.
Only one alga ( Diploclaaia patersonis ), of the 49
reported is endemic to southern Australia; the remaining
48 taxa are more widespread and nearly all have been
reported outside Australian waters. All species recorded
in Spencer Gulf are known to occur on open coastlines.

Four species (Cladophorella marina , Gelidiella
nigrescens, Gelidiella tenuissima , Lophosiphonia
subadunca) found during this study, are generally con¬
sidered tropical, and have not been reported previously
from the southern coast of Australia.

The genus Cladophorella has not been reported
from southern Australia (see Womersley 1956, 1971)
and the only record of Gelidiella from the southern
region of Australia is G. rame/losa which was originally
described as Acrocarpus ramellosus by Sonder (1848)
from an unspecified locality in Western Australia (see
also Kuetzing 1868, Tab. 34 d-f). Three other tropical

algae ( Acetabularia calycuius, Hormophysa triquelra,
Sargassum decurrens) have been recorded outside the
mangrove environment in Spencer Gulf (Womersley
1981b).

Numerous hypotheses have been proposed to ex¬
plain disjunct distributions of marine benthic algae.
These include:

1. Spread by shipping (Farnham et al. 1973, Lewis &
Kraft 1979, Womersley 1981b);

2. Ocean currents (Womersley, 1981b);

3. Continental drift (Chapman 1953); and

4. The possibility that they are relict populations
(Womersley 1981b).

The establishment of algal species distant from their
modern recorded natural distributions was tentatively
attributed to transport on shipping for Schottera
nicaeensis (Lamouroux ex Duby) Guiry and Hollenberg
in Port Phillip Bay (Lewis & Kraft 1979) and for
Sargassum mulicum and other algae in Britain (Farn¬
ham 1980, Farnham et al. 1973). This hypothesis is bas¬
ed on the apparent absence of these species from
previous floristic records and their recent appearance
near busy shipping ports. Spread by ocean currents over

102

W. R. BEANLAND AND Wm. J. WOELKERL1NG

long distances seems to be of rare occurrence, due to the
short life of algal spores and drifting fragments
(Womersley 1981b).

An alternative hypothesis is that the isolated
populations in Spencer Gulf are relics of warmer
climatic periods when their natural distributions may
have been quite different, and that they are able to sur¬
vive in Spencer Gulf because the summer water
temperatures are sufficiently high for reproduction to
occur (Womersley 1981b). These populations may be
relict from the Cretaceous era (70-130 million years ago)
when an inland sea (Brown et al. 1968) joined southern
Australia to the tropical waters of the Arafura sea.

Continental drift has been another hypothesis
used to explain the disjunct distribution of algae (Chap¬
man 1953). Chapman (1953) used an example of three
genera (Eck Ionia, Macrocyst is , Splachnidium) which oc¬
cur in both South Africa and South America. This
hypothesis fails to explain the occurrence of tropical
species in Spencer Gulf because, prior to continental
drift (Coulomb 1969), southern Australia was connected
to the Antarctic.

On the available data, it seems equally possible,
that the species with tropical affinities are remnants of
an earlier climatic period or, since Spencer Gulf includes
a number of ports, that the algae may also have been
transported by shipping. The least likely hypothesis to
explain the occurrence of tropical algae in Spencer Gulf
is continental drift.

Of the 10 localities examined, Blanche Harbour,
Port Augusta and Red Cliff were subject to the higher
salinities and surface water temperatures of the far
north of Spencer Gulf while the remaining seven sites
were situated in areas of more moderate salinity and
temperature further south. The total flora in the far nor¬
thern region (30 taxa) was less diverse than in more
southern sites (47 taxa), and only one taxon ( Ectocarpus
sp.) appeared to be confined to the far north. Hence,
high salinity or surface water temperature may restrict
mangrove algal community composition. Davey and
Woelkerling (1980) also found that the estuarine
(brackish water) influence adversely affects algal diversi¬
ty on Victorian mangroves. Species of Bostrychia and
Caloglossa, however, occur in the estuarine environ¬
ment in Victoria, the high salinity region of northern
Spencer Gulf, and are widespread in tropical
Queensland (Cribb 1979) and outside of Australian
waters (Post 1963). Their widespread distributions may
be attributed to their high resistance to osmotic shock,
desiccation or temperature extremes and/or salinity ex¬
tremes (e.g. Biebl 1962, Davis & Dawes 1981, Yarish el
al. 1979, Yarish el al. 1981). Further studies would be re¬
quired to determine whether the more characteristic
mangrove algae have a relatively higher tolerance to
stressed conditions than other species with more limited
distributions.

The total number of species (Chlorophyta,
Phaeophyta, Rhodophyta) in Spencer Gulf mangrove
algal communities is far more diverse than heretofore
realised; 43 (88%) of the species found during this study

were not recorded previously from South Australian
mangrove ecosystems. The total number of mangrove
algae in Spencer Gulf (Table 5) is comparable to that
reported (Cribb 1979, Saenger el al. 1977) from Avicen¬
na pneumatophores in tropical Queensland (47 taxa),
but more diverse than reported (Davey & Woelkerling
1980) from Victorian mangroves (23 taxa). It was
thought previously (Davey & Woelkerling 1980) that the
species richness of temperate Victorian mangrove com¬
munities was comparable to that reported by Saenger et
al. (1977) from tropical Queensland (23 taxa). The re¬
cent floristic contribution of Cribb (1979), however, has
indicated that the Queensland communities are more
species rich (Table 5).

Species composition differs substantially between
these three regions of Australia. Of the 10 recorded
species of Chlorophyta in Spencer Gulf, two occurred in
Queensland and three in Victoria; of the 28 recorded
Rhodophyta, six occurred in Queensland and four in
Victoria; of the nine Phaeophyta, none have been
reported in Queensland or in Victoria. Thus, only 20%
of the total eucaryotic macroscopic algal flora in
Spencer Gulf has been reported from Queensland and
only 17% from Victoria. Percursaria percursa and
Diplocladiapatersonis are common to Spencer Gulf and
Victorian mangrove communities and do not occur on
tropical Queensland mangroves. Because no distinction
has been made in most cases between those taxa present
in mangrove ecosystems and those taxa present in salt
marsh ecosystems in New South Wales (King 1981a:
323-324, Saenger et al. 1977), only a few algal taxa
(Table 5) are recorded unequivocally from New South
Wales mangroves and meaningful comparisons with the
Spencer Gulf flora are not possible. Absence of detailed
data prevents biogeographic comparisons of the Spencer
Gulf algal flora with those of Northern Territory,
Western Australia, and other parts of South Australia
(Gulf St. Vincent, Streaky Bay —Ceduna region).

Bostrychia and Caloglossa were thought not to
occur on South Australian mangroves (Womersley
1981a), but this study has revealed that they are two of
the most widespread and common algae in Spencer Gulf
communities. All species of these two genera, also have
been found on mangroves in Queensland and Victoria.
Eight additional species of Bostrychia and two addi¬
tional species of Caloglossa occur on Queensland
mangroves, and two additional species of Bostrychia oc¬
cur on Victorian mangroves. Species of Catenella ,
widespread in other mangrove ecosystems, were not
found in Spencer Gulf.

Frequency data from Spencer Gulf (Table 4) and
Victoria (Davey & Woelkerling 1980) suggest that (he
mangrove algal communities of the two regions are
similar in some respects but differ in others. In both
regions, Caloglossa leprieurii is the most conspicuous
taxon. Bostrychia radicans was occasional, common, or
abundant at many localities in both areas, while species
of Enteromorpha and Diva appear to occur with a
similar range of frequencies in Spencer Gulf and
Victoria.

SOUTH AUSTRALIAN MANGROVE ALGAE

103

Table 5

Biogeographic Comparisons of the Algal Flora Recorded from Mangrove Ecosystems
in South Australia, New South Wales, Queensland and Victoria, with Those of
Spencer Gulf (Reference sources are numerically coded)

Spencer
Gulf, S.A.
(Present
Study)

South
Australia
(L7,8,9)

N.S.W.

(4,5,6)

QLD.

(2,6)

VIC.

(3)

Reported
Australian
Mangrove Algae
(Present Study
+ 1-9)

Chlorophyta

10

3

_

14

6

22

Chrysophyta

—

—

—

—

1

1

Phaeophyta

9

1

1

3

3

13

Rhodophyta

28

-

6

30

13

53

Total

47

4

7

47

23

89

Bostrychia

3

_

3

10

4

12

Caloglossa

1

—

1

3

1

3

Catenella

-

-

1

2

1

2

Total

4

_

5

15

6

17

References

L Butler el al. (1977b); 2, Cribb (1979); 3, Davey & Woelkerling (1980); 4, King (1981a); 5, King (1981b);
6, Saenger (1977); 7, Specht (1972); 8, Womersley & Edmonds (1958); 9, Wood (1937).

In contrast, Cladophorella marina and Rivularia
atra , fairly conspicuous components of Spencer Gulf
mangrove ecosystems, were not found in Victoria, and
conversely Bostrychia inlricata and Catenella nipae were
widespread and often prevalent in Victoria but were not
found in Spencer Gulf mangroves. Hormosira banksii ,
not recorded by Davey & Woelkerling (1980) from Vic¬
torian mangroves, occurs only rarely or sporadically in
Spencer Gulf mangroves studied. This contrasts with the
reports of Butler el al. (1977a, 1977b) who regard H.
banksii as common in South Australian mangroves and
with the record of Clarke and Womersley (1981: 500)
from St. Vincent Gulf.

In Victoria, 3-5 prevalent (F>0.50) species oc¬
curred at half of the 16 study sites and at least 1
prevalent species was present at every locality. In
Spencer Gulf, however, only one site (Port Davis, 4
prevalent species) harboured more than 2 prevalent
species and at two localities (Blanche Harbour, Tumby
Bay) no prevalent species occurred al all. Reasons for
the generally lower number of prevalent species at
Spencer Gulf study sites remain unclear, and whether a
reduction in the number of prevalent species occurs as
the overall diversity increases is also uncertain.

Both in Spencer Gulf and in Victoria, a majority
of algal taxa found in mangrove ecosystems always ap¬
pear to be rare (F<0.05) or sporadic (F = 0.05 to 0.24).
In Victoria, only 10 of the 23 (43%) taxa found
registered frequencies greater than 0.24. In Spencer Gulf
16 of 41 (39%) similar taxa occurred. Thus in both
regions, only a minority of mangrove algae appear to be
potentially well adapted to the environmental stresses
present.

ACKNOWLEDGEMENTS

Sincere thanks are due to Dr. Ian Price for identi¬
fying taxa of the Gelideaceae, Professor H. B. S.
Womersley for confirming identifications of taxa of
Rhizoclonium , Dr. Michael Wynne for examining some
collections of Rhodomelaceae and Mr. Doug Reilley of
the Red Cliff environmental research station for
hospitality during field trips.

REFERENCES

Abbott, I. A. & Hollenberg, G. J., 1976. Marine Algae of
California. Stanford University Press, Stanford,
California.

Agardh, C. A., 1817. Synopsis Alga rum Scandinaviae, Ad-
jecla Dispositione Universal! Alga mm. Berling, Lund.
Agardh, C. A., 1821. Species Alga rum. Vol 1. Berling, Lund.
Agardh, J. G., 1848. Species, Genera el Ordines Alga rum.
Vol. /. Species genera et ordines fucoidearum. T. O.
Weigel, Leipzig.

Agardh, J. G., 1863. Species, Genera el Ordines Alga rum.
Vol. 2, Pi. 3, Fasc. 3. Species genera et ordines
floridearum. T. O. Weigel, Leipzig.

Agardh, J. G. 1876. Species, Genera et Ordines Alga rum.
Vol. 3, Pt. I. Epicrisis systematis floridearum. T. O.
Weigel, Leipzig.

Agardh, J. G., 1883. Till algernes systematic Nya bidrag.

Lunds Univ. Arsskr. 19(2): 1-182.

Areschoug, J. E., 1850. Enumeratio phycearum in maribus
Scandinaviae cresentium. Sectio posterior, Ulvaceas
continens. Nova Acta R. Soc. Scient. upsal. Ser. 2, 14:
385-454.

Australian National Tide Tables, 1981. Australian Govern¬
ment Publishing Service, Canberra.

104

W. R. BEANLAND AND Wm. J. WOELKERLING

Biebl, R., 1962. Protoplasmatische-okologische Unter-

suchungen an Mangrovalgen von Puerto Rico. Pro¬
toplasma 55: 572-606.

Bliding, C., 1963. A critical survey of european taxa in
Ulvalcs. Pt. 1. Capsosiphon, Percursaria, Blidingia,
Enteromorpha. Op. hot. Sue. hot. Lund. 8(3): 1-160.

Bliding, C., 1968. A critical survey of european taxa in
U1 vales. Pi. II. Ulva, Ulvaria, Monosfro/na, Kornman-
nia. But. Noliser 121: 535-629.

Bohrgesen, F., 1925. Marine algae from the Canary Islands. I.
Chlorophyccae. Kgl. Danske Vidensk. Selsk. Biol.
Medd. 5(3): 1-123.

Bornet, E. & Flahault, C., 1886. Revision des Notocacees
Heterocystees. Ann/s. Sci. not. Ser. 7, 4: 343-373.

Bory de Saint-Vinc ent, J. B., 1823. Auduinel/a. In: Diction-
naire Classtque d'Histoire Nature lie, Vol. 3. CAD-
CHI, pp. 340-341, Ray ei Gravier, Paris.

Bory De Sajnt-Vincknt, .). B., 1828. Percursaria. In: Diction-
naire Classtque d’Histoire Nature/le , Vol. 13. PAN-
PI V, p. 206, Ray et Gravier, Paris.

Brown, D. A., Campbell, K. S. W. & Crook, K. A. W., 1968.
The Geological Evolution of Australia and New
Zealand. Pergamon Press, London.

Bullock, D. A., 1975. The general water circulation of
Spencer Gulf, South Australia, in the period February
to May. Trans. P. Soc. S. Aust. 99: 43-53.

Butler, A. J., Depers, A. M., McKillup, S. C. & Thomas, D.
P., 1977a. Distribution and sediments of mangrove
forests in South Australia. Trans. R. Soc. S. Aust. 101:
35-44.

Butler, A. J., Depers, A. M., McKillup, S. C.&Thomas, D.
P., 1977b. A survey of mangrove forests in South
Australia. S. Aust. Nat. 51: 39-49.

Chapman, V. J., 1949. Some new species and forms of marine
algae from New Zealand. Farlowia 3: 495-498.

Chapman, V. J., 1953. Distribution of marine algae in relation
to continental drift. Proc. VII. Pacif. Sci. Congr. 5:
40-43.

Chapman, V. .1., 1956. The marine algae of New Zealand. Pt.
1. Myxophyceac and Chlorophyceac. ./. Linn. Soc.
Bol. Loud. 55: 333-501.

Chapman, V. J., 1969. The Marine Algae of New Zealand. Pt.
HI. Rhodophyceae, issue I: Bangiophycidae and
Florideophycidae ( Nernanionafes, Bonnentaisoniales,
Ge/idiales). J. Cramer, Lehre.

Clarke, S. M. & Womersley, H. B. S., 1981. Cross fertiliza¬
tion and hybrid development of forms of the brown
alga Hormosira banksii (Turner) Decaisne. Aust. J.
Bol. 29: 497-505.

Clayton, M. N., 1974. Studies on the development, life
history and taxonomy of the Ectocarpales (Phaeo-
phyta) in southern Australia. Aust. ./. Bot. 22: 743-813.

Coulomb, J., 1969. Sea floor spreading and continental drift.
D. Riedel, Holland.

Cribb, A. B., 1956. Records of marine algae from south¬
eastern Queensland. II. Polysiphonia and
Lophosip/wnia. Pap. Dep, Bot. Univ. Qd 3: 131-147.

Cribb, A. B., 1965. An ecological and taxonomic account of
the algae of a semi-marine cavern Paradise Cave,
Queensland. Pap. Dep. Bot. Univ. Qd 4: 259-282.

Cribb, A. B., 1979. Algae associated with mangroves in
Moreton Bay, Queensland. In Northern Moreton Bay
Symposium , A. Bailey & N. C. Stevens, cds. Royal
Society of Queensland, Brisbane, 63-69.

Davey, A. & Woelkerling, W. J., 1980. Studies on
Australian mangrove algae. I. Victorian communities:
Composition and geographic distribution. Proc. R.

Soc. Viet. 91: 53-66.

Davis, M. A. & Dawes, C. J., 1981. Seasonal photosynthetic
and respiratory responses of the intertidal red algae,
Bostrychia hinderi Harvey (Rhodophyta, Ceramiales)
from a mangrove swamp and a salt marsh. Phyco/ogi a
20: 165-173.

Decaisne, J., 1842. Essais sur une classification ses algues 0[
des poly piers calcifcrs de Lamouroux. Annls. Sci. nat .
Bol. Ser. 2, 17: 297-380.

De Toni, J. B., 1895. Syl/oge Alga rum Vol. 3. De Toni,
Padova.

Dixon, P. S. & Irvine, L. M., 1977. Seaweeds of the British
Isles. Vol. I. Rhodophyta. Pt. I. Introduction ,
Nema/fales, Gigartinales. British Museum (Natural
History), London.

Dkouet, F., 1973. Revision of the Nostocaceae with Cylitu
drical Trichomas. Hafner Press, New York.

Dkouet, F., 1978. Revision of the Nostocaceae with coru
striated trichomas. J. Cramer, Lehre.

Easton, A. K., 1974. A reappraisal of the tides in Spencer
Gulf, South Australia. Aust. J. mar. Freshwat. Res ,
29: 247-277.

Falkenberg. P., 1901, Die Rhodomclaceen des Golfes von
Neapel und dcr angrenzenden Meers-abschnitte. Fan no
Flora Golf. Neapel 26: i-xvi, 1-754.

Farnham, W. F., 1980. Studies on aliens in the marine flora of
southern England. In The Shore Environment, Vol. 2:
Ecosystems , .1. H. Price, D. E. G. Irvine, & W. F.
Farnham, eds. Academic Press, London, 875-914.
Farnham, W. F., Fletcher, R. L. & Irvine, L. M., 1973. AL
tached Sargassum found in Britain. Nature, Lond. 243:
231-232.

Feldmann, J., 1937. Rechcrches sur la vegetation marine de la
Mediterranee. La cote des Alberes. Revue algol. 10:
1-339.

Feldmann, J. & Hamel, G., 1934. Observations sur quelques
Geldiacees. Revue gen. Bot. 46: 529-549.

Feldmann, J. & Hamel, G., 1937. Les algues marines de la cdte
des Alberes. II. Chlorophycees. Revue algol. 9:
141-329.

Foslie, M. H., 1898. List of species of the Lithothamnia.

Kgl. Norske Vidensk. Selsk. Skr. 1898(3): 1-11.

Foslie, M. H., 1909. Algologiske notiscr. VI. Kgl. Norske
Vidensk. Selsk. Skr. 1909(2): 1-63.

Fritsch, F. E., 1944. Cladophorella calcicola nov. gen. et sp.,
a terrestrial member of the Cladophorales. Annls. Bot.
N.S. 8: 157-171.

Goldsmith, F. B. & Harrison, C. M., 1976. Description and
analysis of vegetation. In Methods in Plant Ecology , S.
B. Chapman, cd., Blackwell Scientific, Oxford, 85-155.
Greville, R. K., 1824. Flora Edinensis . William Blackwood,
Edinburgh.

Hamel, G., 1925. Floriddes de France III. Revue algol. 2:
39-67.

Hamel, G., 1939. Phaeophycees de France. Fasc 5. Paris.
Harvey, W. H., 1833. Algae. In The English Flora of Sir
James Edward Smith. Class XXIV. Cryptogamia. Vol.

5, Pt. I, W. J. Hooker, ed., Longman, Rees, Orme,
Brown, and Green, London, 248-415.

Harvey, W. FL. 1849. Phycologia Britanica Fasc. 40, pis.
235-240, L. Reeve, London.

Harvey, W. FL, 1855. Some account of the marine botany of
the colony of Western Australia. Trans. R. Ir. Acad.
22: 525-566.

Holmgren, P. K. & Kueken, W., 1977. Additions to ‘The Her¬
baria of the World’, cd. 6(11). Taxon 26: 483-491.
Hollenberg, Ci. J., 1968. An account of the species of the red

SOUTH AUSTRALIAN MANGROVE ALGAE

105

alga Herposiphonia occurring in the central and
western tropical Pacific Ocean. Pacif Sci. 22: 536-559.

Islam, A. K. \L, 1964. The genus C/adophorella newly found
in East Pakistan. Revue algol, 7: 276-289.

Islam, A. K. M., 1973. The flora of Sundarbans mangrove
forest, Bangladesh. Bangladesh J. Bot. 2: 11-36.

Johansen, H. W., 1981, Coralline Algae, a first synthesis.
CRC Press: Boca Raton, Florida.

King, R. J., 1981a. Mangrove and saltmarsh plants. In Marine
Botany: An Australasian Perspective, M. N. Clayton &
R. J. King, eds., Longman Cheshire, Melbourne,
309-328.

King, R. J., 1981b. The macroalgae of mangrove communities
in eastern Australia. Phycologia 20: 107-108.

King, R. J., 1981c. The free-living Hormosira banksii (Turner)
Decaisnc associated with mangroves in temperate
eastern Australia. Bot. mar. 24: 569-576.

Kuckuck, P., 1956. Ectocarpacecn-studien. IV. Herponetna,
Kuetzingiella nov. gen., Far/owiella nov. gen.
Helgolander wiss. meeresunters 5: 361-382.

Kuetzing, F. T., 1841. Ueber Ceramium Ag. Linnaea 16:
727-746.

Kuetzing, F. T., 1843. Phycologia Genera/is. F. A.
Brockhaus, Leipzig.

Kuetzing, F. T., 1845. Phycologia Germanica. W. Kohne,
Nordhausen.

Kuetzing, F. T., 1849. Species A/garutn. F. A. Brockhaus,
Leipzig.

Kuetzing, F. T., 1855. Tabulae Phyco/ogicae. Vol. 5. Kuetz¬
ing, Nordhausen.

Kuetzing, F. T., 1868. Tabulae Phycofogicae. Vol. 18. Kuetz¬
ing, Nordhausen.

Kyi.in, 1956. Die Gattungen der Rhodophyceen. Glecrups,
Lund.

Lamouroux, J. V. F., 1809. Exposition des characteres du
genre Dictyota , et tableau des espcces qu’il renferme../.
Bot., Paris 2: 38-44.

Lamouroux, J. V. F., 1812. Extrait d’un memoire sur la
classification des polypiers coralligenes non entiere-
ment pierreux. Nouv. Bull. Sc. Soc. Philomat. Ser. 2,
3: 181-188.

Lamouroux, J. V. F., 1813. Essai sur les genres de la famille
dcs Thalissiophytes non articulees. Ann. Mus. Hist,
nat ., Paris 20: 21-47, 115-139, 267-293.

Le Jolis, A., 1863. Listc des algues marines dc Cherbourg.
Mem. Itnp. Soc. Sci. nat. Cherbourg 10: 1-168.

Lewis, J. A. & Kraft, G. T., 1979. Occurrence of a European
red alga (Schottera nicaeensis) in southern Australian
waters. J. PhycoU 15: 226-230.

Link, H. F., 1820. Epistola dealgis aquaticus in genera dispon-
endis. In Novae Physicae Berolineses , C. G. D. Nees,
cd., Marcus, Bon, 1-8.

Linnaeus, C., 1753. Species Plantation. 1 Ed., L. Salvii:
Stockholm.

Lyngbye, H. C., 1819. Tentamen Hydrophytologiae Danicae.
Schuttzianis, Copenhagen.

Macnae, W., 1966. Mangroves in eastern and southern
Australia. Aust. ./. Bot. 14: 67-103.

May, V., 1965. A census and key to the species of
Rhodophyceae (red algae) recorded from Australia.
Contr. N.S.W. natn. Herb. 3: 349-429.

Meneghini, G., 1838, Cenni sulla organografia e fisiologia
delie alghe. Accad. Sci. Padova 4: 325-88.

Montagne, J. F. C., 1840. Seconde centurie de plantes
cellularies exotiques nouvelles, decades I et II. Annls,
Sci. nat. Bot. Ser. 2, 13, 193-207, pi. 5, 6.

Montagne, J. F. C., 1842. Algae. In R. de la Sagra, Histoire

Physique Politique et Naturelle de L’/le de Cuba.
Botanique-Plantes Ce/lulares, 1-104, A. Bertrand,
Paris.

Montagne, .1. F. C., 1846. Phyccae. In Exploration Scientifi-
que de l'Algeria Flore d l Algeria Cryptogamie Paris, M.
C. Durieu de Maisonncuve, cd., Impreimerie Royalc,
Paris, 1-197.

Montagne, J. F. C., 1850. Cryptogamia guaynensis, sue plan-
tarum ccllularium in Guayana Gallica annls. 1835-1849
cl. Leprieur collectarum enumeratio universalis. Annls
Sci. nat. Bot. Ser. 3: 14, 283-309.

Nageli, C., 1846. Herposiphonia Z. wiss.Bot. 1:238-256.

Nasr, A. H., 1947. Synopsis of the marine algae of the Egyp¬
tian Red Sea coast. Bull. Fac. Sci. Fouad I Univ. 26:
1-155.

Newton, L., 1931. A Handbook of the British Seaweeds.
British Museum (Natural History), London.

Philippi, P., 1837. Beweis dass die Nulliporen Pfianzcn sind.
Arch. Naturgesch. 3: 387-393, pi. 9, fig. 2-6.

Post, E., 1963. Zur Verbrietung und Oekologie dcr

Bostrychia-Caloglossa- Assoziation. hit. Revue ges.
Hydrobiol. Hydrogr. 48: 47-152.

Post, E., 1964. Bostrychietum aus dem National park von
Melbourne. Revue algol N.S. 7(3): 242-255.

Rosenvinge, L. K, 1893. GrAilands halvager. Meddr.
Groenland 3, 795-981, 2 pi.

Roth, A. W., 1797. Catalecta Botanica. Vol. 1. J. G. Mueller,
Leipzig.

Roth, A. G., 1806. Catalecta Botanica. Vol. 3. J. G. Mueller,
Leipzig.

Russell, G., 1966. The genus Ectocarpus in Britain I. The at¬
tached forms. J. mar. biol. Ass. U.K. 46: 267-294.

Rueness, J., 1977. Norsk Algenfiora. Universitctsforlaget,
Oslo.

Saenger, P., Specht, M. M. Specht, R. L. & Chapman, V.
J., 1977. Mangal and coastal salt-marsh communities
in Australasia. In Ecosystems of the World I. Wet
Coastal Ecosystems, V. J. Chapman, ed., Elsevier
Scientific Publishing Company, New York, 293-345.

Saito, Y. & Womersley, H. B. S., 1974. The southern
Australian species of Laurencia (Ceramiales:
Rhodophyta). Aust. J. Bot. 22: 815-874.

Schmitz, F., 1896. Bangiaceae. In Die Natur/ichen Pfianzen-
familien Vol. L, Part 2, A. Breler Si K. Prantl, eds.,
Wilhelm Engelmann, Leipzig, 307-316.

Schmitz, F. Sc Falkenberg, P., 1897. Rhodomelaceae. In Die
Naturlichen Pfianzenfatnilien, Vol. 1, Part 2, A.
Engler & K. Prantl, eds., Wilhelm Engelmann, Leip¬
zig^ 421-480.

Searles, R. B. Si Schneider, C. W., 1978. A checklist and
bibliography of North Carolina seaweeds. Botanica
marina 21: 99-108.

Setchell, W. A. & Mason, L. R., 1943. Goniolithon and
Neogonio/ithon: Two genera of crustaceous coralline
algae. Proc. Nat. Acad. Sci. 29: 87-92.

Son der, O. Cj., 1848. Algae. In Plantae Preissianae sive
Enumeratio Plan (arum quas in Australasia Occidentali
et meridionali Occidentals anis, Vol. 2, C. Lehmann,
cd., Sumptibus, Hamburg, 148-195.

Specht, R. I.., 1972. TheVegetation of South Australia. 2nd
Ed., Government Printer, Adelaide.

Taylor, W. R., 1960. Marine algae of the Eastern Tropical
and Subtropical Coasts of the Americas. University of
Michigan Press, Ann Arbor.

Tronson, K., 1974. The hydraulics of the South Australian
gulf system I. Circulation. Aust. J. mar. Freshwat.
Res. 25: 413-26.

106

W. R. BEANLAND AND Wm. J. WOELKERLING

Umezaki, I., 1961. The marine blue-green algae of Japan.
Mem. Coll. Agr. Kyoto Univ. 83: 1-148, pi. 1-21.

Wille N., 1909. Algologische Noiizen. XV. Ueber Wit-
trockieUa nov. gen Nytt. Mag. Naturvid. 47: 209-225.

Woelkerling, W. J., 1970. Aerochaetium botryocarpum
(Harv.) J. Ag. ( Rhodophvta ) in southern Australia. Br.
phycol. J. 5: 159-171.

Woelkerling, W. J., 1971. Morphology and taxonomy of the
AudouineHa complex (Rhodophyta) in southern
Australia. Aust. J. But., Suppl. Ser. 1: 1-91.

Woelkerling, W. J., 1973. The morphology and systematic
of the AudouineHa complex (Acrochaetiaceae,
Rhodophyta) in northeastern United States. Rhodora
75: 529-621.

Woelkerling, W. J., 1981. New insights into the systematics
of the Corallinaceac (Rhodophyta). In Xlll Interna¬
tional Botanical Congress. Abstracts, Australian
Academy of Science, Canberra, 161.

Womersley, H. B. S., 1946. Studies on the marine algae of
southern Australia. Introduction and No. I. The
genera Isactis and Rivularia (Myxophyccae). Trans. R.
Soc. S. Aust. 70: 127-136.

Womersley, H. B. S., 1956. A critical survey of the marine
algae of southern Australia 1. Chlorophyta. Aust. J.
mar. Freshwat. Res. 7: 343-383.

Womersley, H. B. S., 1967. A critical survey of the marine
algae of southern Australia II. Phacophyta. Aust. J.
Bot. 15: 189-270.

Womersley, H. B. S., 1971. New records of taxa of marine
Chlorophyta in southern Australia. Trans. R. Soc. S.
Aust. 95: 113-120.

Womersley, H. B. S., 1979. Southern Austalian species of
Po/ysiphonia Grcville (Rhodophyta). Aust. J. Bot 27:
459-528.

Womersley, H. B. S. 1981a. Marine ecology and zonation of
temperate coasts. In Marine Botany: An Australasian
Perspective , M. N. Clayton & R. J. King, eds.,
Longman Cheshire, Melbourne, 211-240.

Womersley, H. B. S., 1981b. Biogeography of Australasian
marine macroalgae. In Marine Botany: An Australa¬
sian Perspective , M. N. Clayton & R. J. King, eds.,
Longman Cheshire, Melbourne, 292-307.

Womersley, H. B. S., 1981c. Aspects of the distribu¬
tion and biology of Australian marine macro-algae. In
The Biology of Australian Plants , J. S. Pate & A. J.
McComb, eds., Univ. W. Australia Press, Nedlands,
W.A., 294-306.

Womersley, H. B. S. & Cartledge, S. A., 1975. The
southern Australian species of Spyridia (Ceramiaceae:
Rhodophyta). Trans. R. Soc. S. Aust. 99: 221-234.

Womersley, H. B. S. & Edmonds, S. J., 1958. A
general account of the intertidal ecology of South
Australian coasts. Aust. J. mar. Freshwat. Res. 9:
217-260.

Womersley, H. B. S. & Thomas, I. M., 1976. Intertidal
ecology. In Natural History of the Adelaide Region , C.
R. Twidale, M. J. Tyler & B. P. Webb, eds., Royal
Society of South Australia, Adelaide, 175-185.

Wood, J. Ci. 1937. The Vegetation of South Australia.
1st ed. Government Printer, Adelaide.

Yarish, C., Casey, S. & Edwards, P., 1981. The effects
of salinity, and calcium and potassium variations on
the growth of two estuarine red algae. Phycologia 20:
118-119.

Yarish, C., Edwards, P. & Casey S., 1979. A culture
study of salinity responses in ecotypes of two estuarine
red algae. J. Phycol. 15: 341-346.

PROC. R. SOC. VICT. vol. 94, no. 3, 107-120, September 1982

DISCOVERY OF CHEIROCRATUS (CRUSTACEA:
AMPHIPODA) ON AUSTRALIAN SHORES

By J. Laurens Barnard* and Margaret M. Drummond!

* Smithsonian Institution, Washington, D.C. 20560, USA
t National Museum of Victoria, 285-321 Russell Street, Melbourne, Victoria 3000

Abstract: Two new species of the gammaroid amphipod genus Cheirocratus, C. bassi and C.
praedens, and a new genus and species, Prosocratus buicheri with close affinities to Cheirocratus , all from
Australian waters, are described. A new genus, Incratella , is erected to accommodate C. inermis Ledoyer,
1967. A revised diagnosis of Cheirocratus and a key to all known species in the genera Cheirocratus, In¬
cratella and Prosocratus are presented.

For more than fourteen decades the marine am¬
phipod genus Cheirocratus has been well known in Arc¬
tic, North Atlantic and Mediterranean seas. Recently a
species of the genus was described from Madagascar
(Ledoyer 1967) which we propose should be transferred
to a new genus Incratella.

We have discovered in southern Australia, and here
report on two new species of Cheirocratus both of which
show close ties to Mediterranean species; we report a
third new taxon which we consider to be generically
distinct from, though very closely allied to Cheirocratus.
We assume the flow of evolutionary deployment pro¬
ceeds from Australia towards the North Atlantic as it
does in so many other groups of Amphipoda (see Bar¬
nard 1972, 1974, Barnard & Drummond 1978).

Most of the materials examined came from the two
benthic surveys conducted in Western Port, Victoria,
sponsored by the Victorian Government and supported
by industrial organizations: Crib Point Benthic Survey,
1964-5 (CPBS) and Westernport Bay Environmental
Study 1973-4 (WPBES). Acknowledgements to those
concerned in these undertakings, as well as station data,
have been detailed in a previous publication (Barnard &
Drummond 1978). Additional material came from
plankton samples collected in Western Port by R. H.
Miller, University of Melbourne (RHM); from
Tasmania (collected by T. Walker); from the Bass Strait
survey at present being conducted by the Victorian In¬
stitute of Marine Science (VIMS); from the Queensland
University Survey of Middle Banks, Moreton Bay,
Queensland, from Port Phillip (PPBES) and from the
N.S.W. Fisheries Estuarine Benthic Survey (EBS).

LEGENDS

Capital letters and numbers on the figures denote
parts, as follows: A, antenna; B, body or carcass; C,
coxa; D, dactyl; E, epistome, left view; F. accessory
flagellum; G, gnathopod; H, head; 1, inner plate; J,
ramus; K, variable, see legend; L, lower lip = labium; M,
mandible; N, molar; O, palp; P, pereopod; Q, pleopod;
R, uropod; S, maxilliped; T, telson; U, upper
lip = labium; V, brood plate; W, pleon; X, maxilla; Y,
gill; Z, gland.

The figures each contain illustrations from a master
specimen listed first in the caption of each figure and no

lower case letters are placed on these figures; subsidiary
specimens on each figure are denoted by lower case let¬
ters to left of capitals as specified in the caption for each
figure. Lower case to the right of capitals indicate: m,
medial; r, right; s, setae removed.

SYSTEMAT1CS

Genus Cheirocratus Norman 1867
Type Species (by monotypy): Cheirocratus mantis Nor¬
man 1867, (= Gammarus assimilis Liljeborg 1852 accor¬
ding to Stebbing 1906: 417).

Diagnosis: Body ordinary, urosomites free, dorsally
denticulate and spinosetose transversely. Rostrum ob¬
solescent, lateral cephalic lobes mamiliform, sinus
present. Eyes present. Antenna 1 much shorter than
antenna 2, ratio of peduncular articles = 16:16:5 (in type
species), primary flagellum as long as peduncle, ac¬
cessory flagellum 2+ articulate. Antenna 2 large and
elongate, flagellum scarcely shorter than peduncle.
Labrum as broad as long, weakly notched apically.
Mandibular incisor toothed, molar triturative, ratio of
palp articles = 10:18:11 (in type species), article 3 weakly
falcate or strongly sickle-shaped, setae = (A)DE. Inner
lobes of labium well developed. Maxillae medially
setose, inner plate of maxilla 1 ovatotriangular, fully
setose medially, outer plate with 11 (rarely 9) spines,
palps ^symmetrical]. Inner plate of maxilla 2 with ob¬
lique facial row of setae or strongly setose medially.
Outer plate of maxilliped medially spinose, palp article 3
unlobed, dactyl shorter than article 3, unguiform,
[?w r ithout nail). Coxae of ordinary length, poorly setose,
coxa 1 slightly to strongly expanded apically, coxa 4
scarcely or not, Iobate. Gnathopods diverse; female
gnathopods simple, feeble, wrists elongate, not Iobate,
hands thin, lacking palms; male gnathopod 1 like
female, gnathopod 2 greatly enlarged, wrist short, not
Iobate, hand large, elongate, ovate, rectangular or
trapezoidal, palm oblique, elongate, smooth or
sculptured. Pereopods 3-4 ordinary. Article 2 of
pereopods 5-7 scarcely expanded or not, almost linear,
scarcely Iobate or not, posterior margins weakly ser-
ratosetulate. Pleopods ordinary. Rami of uropods 1-2
marginally spinose, evenly extended, peduncle or
uropod 1 [?without basofacial armaments]. Uropod 3

107

J. L. BARNARD AND M. M. DRUMMOND

Fig. 1 — Cheirocratus bassi sp. nov. Unattributed figures — holotype male “h 5.80 mm; p —female p

4.41 mm.

CHEIROCRATUS IN AUSTRALIA

109

Cheirocratus bassi sp. nov. Unattributed figures = holotype male “h” 5.80 mm:

4.41 mm.

female

110

J. L. BARNARD AND M. M. DRUMMOND

extended, magniramous, almost aequiramous, peduncle
elongate, rami 1-articulate, lanceolate. Telson short,
deeply cleft, gaping, lobes tapering, well spinose apic-
ally. Coxal gills [?2-6], ovate, occasionally pediculate.
Oostegites narrow.

Variants : Telson poorly armed (C. bassi, C. praedens );
male gnathopod 2 like zeylanca melitas , hand ovate,
palm undefined and heavily setose, dactyl riding onto
medial face of hand (C. sundevalli ); mouthparts in
diagnosis based on C. sundevalli of Sars (1895).

Relationship: Like Cheirocratella but female
gnathopod 2 simple.

Species: See Chevreux & Fage (1925); armatus G. S.
Karaman 1977a; assirnilis (Liljeborg 1852) (Sars 1895)
(Chevreux & Page 1925); bassi Barnard & Drummond
herein; intermedius Sars 1895; monodontus G. S.
Karaman 1977b; praedens Barnard & Drummond
herein; robustus Sars 1895 (Stephensen 1928, 1929,
1940); sundevalli (Rathke 1843) (Sars 1895).

Key to Cheirocratus, Incratella and Prosocratus
(modified after Karaman 1977)

(Note: C. bassi is placed twice in key, second time ignor¬
ing telson to show position near species from Europe)

1. Male gnathopod 2 simple.

. Prosocratus butcheri gen. et sp. nov.

Male gnathopod 2 subchelate.2

2. Telson with thin setae but no spines.3

Telson with both setae and spines.5

3. Urosomites untoothed, article 1 of mandibular palp

as long as article 3, article 2 of pereopod 7

lobate ..... ... Incratella inermis

Some urosomites toothed, article 1 of mandibular
palp longer than article 3, article 2 of pereopod 7
not lobate. .4

4. Urosomite 2 untoothed.C. armatus

Urosomite 2 with 2 dorsal teeth ... C. bassi sp. nov.

5. Dactyl of male gnathopod 2 closing on toothed

posterior palm, hinge part of palm with

crenellate tooth or teeth.6

Dactyl of male gnathopod 2 overriding palm onto
face of hand, hinge part of palm smooth or with
smooth hump.7

6. Palm of male gnathopod 2 with 4 major teeth

spread throughout palm, palm poorly setose ...

.....C. assirnilis

Palm of male gnathopod 2 with 2 major teeth or
tooth groups, one at hinge, one at defining cor¬
ner, middle of palm lacking teeth but heavily set¬
ose .......... ...C. intermedius

7. Dorsomedial peduncular spination of uropod 3

concentrated into group, adult male without
cephalic notch, telson with only small apical

flexible setules.C. bassi sp. nov.

Dorsomedial peduncular spination of uropod 3
widespread, adult male retaining cephalic notch,
telson with one or more stout and stiff apical
spines.8

8. Articles 4-6 of pereopod 7 stout, medial face of

hand on male gnathopod 2 with humped setae,

no spines or ridges.C. robustus

Articles 4-6 of pereopod 7 slender, medial face of
hand on gnathopod 2 with setae, ridge and
spines..9

9. Urosomite 2 with 4 dorsal teeth, epimeron 3 with

straight posterior margin.10

Urosomite 2 with 2 dorsal teeth, epimeron 3 with
convex posterior margin ... C. praedens sp. nov.

10. Urosomite 1 with one large dorsal tooth.

.....C. monodontus

Urosomite 1 with several small dorsal teeth (at
least 3).C. sundevalli

Cheirocratus bassi sp. nov.

Figs 1-3

Diagnosis: Cephalic notch lost in adult male; male
gnathopod 2 without major palmar teeth, palm formed
by inner facial ridge near posterior margin, this ridge
with teeth and spines near apex, medial face with setae
only near posterior margin, dactyl overriding posterior
margin; articles 4-6 of pereopod 7 slender; urosomite 1
with 3 closely contiguous dorsal teeth; urosomite 2 with
2 widely spaced dorsal teeth; spines on peduncle of
uropod 3 forming one group only, not queued; apices of
telsonic lobes with minor armaments only.

Description of Male Holotype “h”: Rostrum minute,
lateral cephalic lobe weakly excavate anteriorly but
notch found in juvenile obliterated in adult, antero-
ventral corner weakly produced; eyes round, pigment
widely spread but ommatidia sparse.

Antenna 1 of medium extension, ratio of pedun¬
cular articles = 16:12:5, article 1 with 2 apical sharp teeth
near spines, one above and one below, primary
flagellum longer than peduncle, with thick but sparse
aesthetascs, accessory flagellum 2-articulate. Antenna 2
elongate, much longer than antenna 1, gland cone large,
extending 50% or more along article 3, article 5 of
peduncle longer than 4, flagellum elongate.

Upper lip broad, entire, rounded.

Mandibular incisors weakly toothed, right lacinia
mobilis thin, narrow, left broad, flat, each 4-toothed;
raker spines main row with 7, accompanied by 7 vestigial
fern-like setae near molar, left rakers like right but sub¬
sidiary row with only 2 rakers; molars large, poorly
triturative; tiny setule on the left, longer weak seta on
the right; mandibular palp large, ratio of
articles = 11:14:9, articles 1 and 2 with inner setae, ar¬
ticle 3 weakly falcate, setae of article 3 = DE.

Inner lobes of lower lip large and fleshy, outer lobes
with medial gape. Inner plate of maxilla 1 broadly
pyriform, leaf-like, fully setose medially, also with facial
setae, outer plate slender, with 11 spines; palps
2-articulate, symmetrical, apices truncate, with simple,
bifid pectinate spines and facial setae, and outer
marginal row of 3 facial setal-spines. Inner plate of max¬
illa 2 broader and shorter than outer, somewhat
geniculate, with fully developed oblique facial row of
setae, some outer marginal setal-spines, sparse medial

CHEIROCRA TUS IN AUSTRALIA

Fig. 3 — Cheirocratus bassi sp. nov. Unattributed figures = holotype male “h” 5.80 mm; j = male “j” 5.34
mm; p= female “p” 4.41 mm; e marks figure drawn to same size as uropod 3 on Fig. 2;

Kl, dorsal urosome; K2, cuticle of pereopod 5.

112

J. L. BARNARD AND M. M. DRUMMOND

and dense apical setae; outer plate almost evenly rec¬
tangular, with 3-4 apicolateral setae and dense apical
setae. Inner plate of maxilliped broad, truncate, with 2
inner apical spines, medially and apicolaterally setose;
outer plate ovate, medially and apically spinose; palp ar¬
ticle 2 elongate, article 3 shorter, unlobate, dactyl
shorter, uniform, with partly immersed nail and ac¬
cessory setule; palp poorly setose laterally.

Coxae 1-5 of medium length, extending almost
equally, only coxa 2 with long setal-spines; coxa 1 ex¬
panded apically and with short armaments, other coxae
poorly armed; coxa 2 constricted midposteriorly; coxae
3-4 subrectangular, 4 not significantly excavate, 3
slightly longer than 4; coxa 5 bilobed; coxae 6-7 shorter
than 5.

Gnathopod 1 slender, small, simple; wrist elongate,
weakly but broadly lobate posteriorly; hand very
slender, weakly bent, irregularly tapering, posterior
margin at base sinuous, palm absent, dactyl short, curv¬
ed, serrate. Wrist of gnathopod 2 short, not lobate,
posterior margin with moderately developed setal
clumps; hand very large, elongate, ovate, anterior
margin setose, posterior margin smooth, convex, with
dense inner facial setae near base, medial face with
heavy ridge near posterior margin, broken into 2 pro¬
cesses near hinge, distal process with long spine, long
seta, 4-5 small jewel-like spines, proximal process with
one medium spine; dactyl fitting into slot formed by
ridge and palmar margin, strongly bent, short; dactyl in¬
terpreted to be overriding margin onto face of palm;
longer setae on wrist of gnathopod 1 and hands of both
gnathopods extremely thick and with basal bulbs.

Pereopod 4 smaller than pereopod 3.

Article 2 of pereopods 5-7 scarcely expanded, not
lobate, posterior margins weakly serratosetulate; legs
slender; pereopods 6 and 7 subequal in length.

Pleopods ordinary. Pleonites 1-5 each with dorsal
posterolateral setule in weak notch; epimera 1-3 with
sinuous posterior margin and sharp medium posteroven-
tral tooth marked with weak setule; ventral margin of
epimeron 2 with one weak spinule, of epimeron 3 with 2
weak anteroventral spinules.

Urosomites 1-2 dorsally toothed, urosomite 1 with 3
dorsal teeth in tight group, this complex armed with 2
setal spines; urosomite 2 with 2 widely spread dorsal
teeth each armed with setal-spine.

Uropods 1-2 extending equally, outer rami scarcely
shortened, all rami with marginal spines and weakly
spinose apices; peduncle of uropod 1 with both margins
spinose, no basofacial spine, apicolateral margin with
small sharp cusp, large spine and small spine. Uropod 2
peduncle with 2 lateral spines and one medial. Uropod 3
strongly extended, peduncle slightly elongate, apicodor-
sal margin with 2 spines in tandem, midmedial margin
with cluster of 3 spines, apicomedial margin with cusp;
rami elongate, lanceolate, weakly spinose along
margins. Telson very short, cleft to base, lobes slightly
separate at base, each lobe tapering to sharp point and
bearing weak subapical setule, one dorsal setule, pair of
dorsolateral setules greatly distad.

Coxal gills on pereonites 2-6, sausage-shaped.

Cuticle under medium power appearing to be peb¬
bled, under oil-immersion pebbles seen to be poorly
developed (only well refractive on low power), com¬
posed of complex concentric interrupted line-pairs, in
places yielding to or basally underfounded by sharp
scales; cuticle also with occasional spikes or studs and
fine bulbar setules.

Description of Female “p”: Lateral cephalic lobe with
deep notch like juvenile. Antenna 1 like male; flagellum
of antenna 2 only as long as articles 4-5 of peduncle
combined.

Gnathopod 1 generally like that of male but hand
less grotesque, tapering evenly, wrist more softly
rounder posteriorly. Gnathopod 2 feeble, simple, wrist
and hand both elongate, dactyl with 2 inner spinules
each with partner setule, palm and posterior margin
confluent and bearing several groups of apically curled
setae.

Both epimera 2-3 with 2 strong anteroventral spine-
setae, much stronger than in male holotype. Brood
plates very slender.

Description of Male “g”: Right maxilla 1 with 9 spines
on outer plate.

Illustrations: Mandibular incisors and laciniae
mobiles when shown together spread apart; main spines
of mandibular raker row with bases heavily broadened
and from side view appearing to form secondary spine
row; maxillae 1-2 equally magnified. Small version of
one telsonic lobe magnified equally to uropod 3. Larger
telsonic versions greatly enlarged.

Holotype: NMV J1631 male “h” 5.80 mm.

Type Locality: WPBES 1749/1, Australia, Victoria,
Western Port, 25 November 1974, 14 m, sand with mud,
silt, clay.

Paratypes: NMV J1632-1636. The type-locality, 6
other specimens, including male “j” 5.34 mm; CPBS
Al/1, female “p” 4.41 mm; RHM, male “g’\ 5.86 mm; 21
(2) dredged WP, 20/2/78, female “q”.

Relationship: Cheirocratus bassi resembles C. armatus
particularly in the lack of thick spines on the telson and
the lack of significant incision in the cephalic lobe. The
other known species of Cheirocratus have apically
spined telsons and markedly incised cephalic lobes in
addition to individual gnathopodal distinctions. C. bassi
differs from C. armatus in the presence of teeth on the
second urosomite. C. monodontus and C. praedens sp.
nov. each have only one tooth on urosomite 1 instead of
the 3 in C. bassi; C. sundevalli has 4 (instead of 2) on
urosomite 2; and in both C. robustus and C. assimilis
pereopod 7 is unequal in length to pereopod 6 and
stouter, in contrast to C. bassi in which pereopods 6 and
7 are slender and subequal.

Material: CPBS, 77 samples from 33 stations (258
specimens); WPBES, 8 samples from 5 stations (23
specimens); RHM 27/10/71 (1 specimen); PPBES, 4
samples from 2 stations (4 specimens).

Distribution: Australia, Victoria, Western Port, Port
Phillip and Bass Strait, 0-36 m, fine to coarse sand,
muddy sand, gravelly sand, sand and shell.

CHEIROCRATUS IN AUSTRALIA

113

Etymology: The species is named for George Bass,
physician, who discovered and named Western Port in
1797.

Cheirocratus praedens sp. nov.

Figs 4, 5 (upper)

Diagnosis: Cephalic notch persistent in adult male; male
gnathopod 2 without major palmar teeth, palm formed
by inner facial ridge near posterior margin, this ridge
with teeth and spines near apex, apical process very
broad with about 6 spines, medial face with setae in mid¬
dle stripe and near posterior margin, dactyl overriding
posterior margin; articles 4-6 of pereopod 7 slender;
urosomite 1 with one medium sized dorsal tooth,
urosomite 2 with 2 widely spaced dorsal teech; spines on
peduncle or uropod 3 widely spread; apices of telsonic
lobes with at least one thick spine each.

Description of Male Holotype “k”: Rostrum minute,
lateral cephalic lobe strongly notched anteriorly,
anteroventral corner rounded; eyes round, pigment fully
covered by morula of ommatidia.

Antenna 1 short, reaching about half way along pe¬
duncular article 5 of antenna 2, ratio of peduncular ar¬
ticles is 16:14:5, article 1 with 2 apical sharp teeth near
apex, one medial and one lateral; primary flagellum
longer than peduncle, with thick but sparse aesthetascs;
accessory flagellum 2 articulate. Antenna 2 elongate,
much longer than antenna 1, gland cone large, article 5
of peduncle longer than 4, flagellum elongate.

Upper lip broad, entire, almost truncate below.

Mandibular incisors weakly toothed, right lacinia
mobilis thin, left flat broad each 4 toothed; raker spines
main row with 9, accompanied by 3-4 vestigial fern-like
setae near molar, left and right sides similar; molars
large, poorly triturative, with conspicuous medium¬
sized seta each; mandibular palp large, ratio of articles is
9:11:8, article 3 weakly falcate, articles 1 and 2 with in¬
ner setae (article 1= medial, 3 large; apicomedial, 1
medium 1 setule; apicolateral, 1 setule; article
2 = medial, 15 mixed setae), setae of article 3 = DE
(medial = 17, apical = 3 large and 1 small). Inner lobes of
lower lip large and fleshy, outer lobes with medial gape.
Inner plate of maxilla 1 broadly pyriform, leaf-like,
fully setose medially, also with facial setae; outer plate
slender, with 11 spines, palps 2-articulate, symmetrical,
apices truncate, with simple, bifid and pectinate spines
and facial setae, also with outer marginal row of 3
setulcs (widespread). Inner plate of maxilla 2 broader
but scarcely shorter than outer, somewhat geniculate,
with fully developed oblique facial row of setae, some
outer marginal setal spines, sparse medial and dense
apical setae, outer plate almost evenly rectangular, with
2 apicolateral setae and dense apical setae.

Inner plate of maxilliped broad, truncate, with 2
inner apical spines, medially and apicolaterally setose,
outer plate ovate, medially and apically spinose, palp
article 2 elongate, article 3 shorter, unlobate, dactyl
shorter, unguiform, with partly immersed nail and
accessory setule; palp poorly setose laterally.

Coxae 1-5 of medium length, extending equally, only
coxa 2 with long setal spines but some on coxa 1 short
and thick, other coxae poorly armed; coxa 1 expanded
apically, coxa 2 weakly constricted midposteriorly,
coxae 3-4 subrectangular, 4 not significantly excavate, 3
and 4 equally extended, coxa 5 bilobed, coxae 6-7
shorter than 5.

Gnathopod 1 slender, small, simple, wrist elongate,
broadly rounded but not lobate posteriorly, hand very
slender, almost straight, tapering, sinuosity very minor,
palm absent, dactyl short, curved, serrate. Wrist of
gnathopod 2 short, not lobate, posterior margin with
moderately developed setal clump, hand very large,
elongate, ovate, posterior margin smooth, convex, with
dense inner facial setae on posterior margin and stripe
down middle of face, medial face with heavy ridge near
posterior margin. Ridge broken into broad process near
hinge, armed with 6 small, jewel-like spines, plus proxi¬
mal process with one spine; dactyl wiping ridge between
both processes, strongly bent, short, dactyl interpreted
to be overriding palm onto face of hand; anterior
margin of hand setose, longer setae on wrist of
gnathopod 1 and hands of both gnathopods extremely
thick and with basal bulbs.

Pereopod 4 scarcely smaller then pereopod 3.

Article 2 of pereopods 5-7 scarcely expanded, not
lobate, posterior margins weakly serratosetulate, legs
slender.

Pleopods ordinary. Pleonites 1-2 each with dor¬
solateral setule, epimera 1-3 with sinuous posterior
margin and sharp posteroventral tooth marked with
weak setule. Ventral margin of epimeron 2 with 4 strong
setal spines, of epimeron 3 with 5 similar armaments.

Urosomite 1 with strong dorsomedial tooth,
urosomite 2 with pair of small dorsal teeth. Uropods 1-2
extending equally, outer rami scarcely shortened, all
rami marginally and apically weakly spinose; peduncle
of uropod I with both margins spinose, without
basofacial spine, apicolateral margin with small sharp
cusp, large spine and small spine; peduncle of uropod 2
with 3 lateral and 2 medial spines. Uropod 3 strongly ex¬
tended, peduncle scarcely elongate, apicodorsal margin
with 2 spines in tandem, medial margin with 2 spines in
group and 2 spines in tandem, apicomedial margin with
cusp; rami elongate, lanceolate, weakly spinose along
margins.

Telson very short, cleft to base, lobes slightly
separate at base, each tapering to sharp point with weak
subapical inner notch armed with small stout spine (one
side with second spinule) and one setule; dorsolateral
surface with setule and partner (variable, either spine or
long plusetule).

Coxal gills on pereonites 2-6, sausage-shaped.

Cuticle under medium power appearing to be pebbl¬
ed, under oil-immersion pebbles seen to be poorly
developed (only well refractive on low power), compos¬
ed of complex concentric interrupted line-pairs, in
places yielding to or basally underfounded by sharp
scales; cuticle also with occasional spikes or studs and
fine bulbous setules.

114

J. L. BARNARD AND M. M. DRUMMOND

Fig. 4 — Cheirocratus praedens sp. nov. Holotype male “k” 12.10 mm.

CHEIROCRA TUS IN AUSTRALIA

115

Illustrations: Many parts similar to those of C. bassi
and not repeated for this species: upper lip, epistome,
mandibles, lower lip, cuticle; maxilla 2 figured in outline
only. Minor distinctions from illustrations of C. bassi ,
noted but not illustrated separately: inner plate of max¬
illa 1 with 26 medial, 11 facial and 3 apical setae, palp
apex with 10 spines (2 more thin medials and one extra
bifurcate) in pattern similar to C. bassi ; inner plate of
maxilliped with 15 medial setae, outer with 2 thin apical
spines and 11 thick medials, article 3 of palp with 4
lateral setae in groups of 3 and 1; pereopods 3 and 4
slightly better armed; pereopod 5 with 3 posteroventral
cusps and 2 setules on article 2 and pair of small
posteroventral spines on article 5; pereopods 6 and 7
with 2 posteroventral cusps and one setule on article 2
and article 4 more elongate (see formulae below);
peduncle of uropod 1 with 5 lateral and 6 medial spines,
of uropod 2 with 3 lateral and 2 medial; outer ramus of
uropod 1 with 4 small lateral spines and 3 small medials,
apex with 2 spines, one large, one small, and 2 cusps;
outer ramus of uropod 2 with 4 large lateral and 4-5
small medial spines, apex with one large and one small
spine, and 2 cusps; inner ramus of uropod 1 with 3 small
lateral and 5 large medial spines, apex with one medium
spine and 2 cusps; inner ramus of uropod 2 with 3 small
lateral and 6 large medial spines, apex with one large
spine, one small spine and 2 cusps. Length ratio of ar¬
ticles 2, 4, 5 of pereopod 6 = 65:53:48 and of pereopod
7 = 61:50:51. (For Cheirocratus bassi these ratios are
65:47:50 and 62:42:51.)

Holotype: NMV J1637 male “k” 12.10 mm.

Type Locality: Tasmania, 400-600 m east of Mid¬
dleton, BBN/T.L., weed bottom, 5 October 1973, col¬
lected by Terry Walker.

Relationship: Cheirocratus praedens differs from C.
monodontuSy from the Mediterranean Sea, in the
presence of only 2 (not 4) dorsal teeth on urosomite 2,
the bulging posterior margin of epimeron 3, the lack of
ventral armament on epimeron 1, the very sparse arma¬
ment of the telson and the produced midfacial ridge on
the hand of male gnathopod 2. It differs from C. bassi,
its Australian compatriot, in the presence of only one
tooth (not 3) on urosomite 1, the absence of spines on
urosomites 1 and 2, the presence of a cephalic notch in
fully adult male, the stronger medial cavity and more
heavily armed hand of male gnathopod 2, the weaker
development of gnathopod 1 in the adult male, in which
article 5 is narrower and article 6 less sculptured, and the
narrower hand of male gnathopod 2 relative to coxa 2,
with larger dactyl.

This species shows many similarities to C. sundevalli
(Rathke) from Europe, but differs from that species in
having fewer dorsal teeth on the urosomites (1 and 2
contrasted with 3 and 4), in the convex posterior margin
of epimeron 3, the more pointed telsonic lobes with
fewer spines (1-2 as against 3), the larger apical process
on the inner palmar face of gnathopod 2 with more
spines (6 as contrasted with 2), and the shorter peduncle
of uropod 3.

The holotype of this species also differs from C. bassi

in various characters assumed to reflect its large adult
size: the ventral spines on urosomites 1 and 2 (absent in
bassi), thicker armaments on the telsonic apices,
stronger or denser spination on peduncle of uropod 3
and epimera 2-3, broader, coxa 3 and the longer reach
on the inner plate of maxilla 2.

Material: A second specimen of Cheirocratus praedens
was taken in a lately-sorted sample from Bass Strait
(VIMS station 110, 16m, fine, shelly sand, 3/11/80).
This female, 11.2 mm in length with setose brood plates,
has simple gnathopods closely resembling those of
females figured for C. bassi and Prosocratus butcheri.
From these two species, however, it is readily
distinguished by the single, very large tooth on the first
urosomite. In other respects the female conforms closely
to the description of the male holotype.

Distribution: Tasmania, intertidal.

Etymology: From the Latin prae meaning “in front”
and dens meaning “a tooth” —refers to the single tooth
on the first urosome.

Genus Incralella nov.

Diagnosis: body unornamented, urosome not dorsally
setose. Rostrum [?small, lateral cephalic lobes deeply
notched, ? sinus present]. Antenna 2 elongate, antenna
1 much shorter than 2, ratio of peduncular articles
= 35:21:10, primary flagellum as long as peduncle, ac¬
cessory flagellum 2-articulate. Articles 4-5 of peduncle
on antenna 2 thin, elongate, flagellum longer than article
5.

Labrum about as long as broad. Mandibular molar
large but poorly triturative, ratio of palp
articles = 11:15:13, article 3 slightly curved, blunt,
setae =DE. Inner lobes of labium large. Maxillae
medially setose; inner plate of maxilla 1 broadly ovate,
fully setose medially, outer plate with [?5] spines, palps
^symmetrical with elongate article 1]. Inner plate of
maxilla 2 with oblique facial row of setae. Coxae of or¬
dinary length to slightly shortened, almost glabrous,
coxa 1 strongly expanded, coxa 4 quadrate, unexcavate.
Female gnathopods simple, feeble, wrists elongate,
unlobate, hands thin, lacking palms; male gnathopods
unknown.

Article 2 of pereopods 5-7 moderately expanded,
weakly (5) to moderately (7) lobate, posterior margins
lacking large setae.

Rami of uropods 1-2 marginally but poorly spinose,
evenly extended, peduncle of uropod 1 [?without
basofacial armaments.] Uropod 3 extended, magnira-
mous, almost aequiramous, peduncle slightly elongate,
rami 1-articulate, lanceolate. Telson short, cleft, gaping,
lobes weakly tapering, not spinose apically.

Coxal gills [?2-6] ovate. Oostegites [?narrow].

Type Species: Cheirocratus inermis Ledoyer 1967.
Relationship: Like Cheirocratus but urosomites un¬
toothed. Male unknown.

Species: Incratella inermis (Ledoyer 1967) Griffiths,
1975; Madagascar and southern Africa, sublittoral, 1.
Etymology: The generic name Incratella is contrived
from word fragments.

116

J. L. BARNARD AND M. M. DRUMMOND

Fig. 5-Upper, Cheirocratus praedens sp. nov. Holotype male “k” 12.10 mm.
Lower, Prosocratus butcheri gen. nov. et sp. nov. Holotype male “n” 4.34 mm.

CHEIROCRATUS IN AUSTRALIA

117

Genus Prosocratus nov.

Diagnosis: Body ordinary, urosomites free, dorsally
denticulate and spinosetose transversely. Rostrum ob¬
solescent, lateral cephalic lobes subtruncate, sinus ab¬
sent. Eyes present. Antenna 2 elongate, antenna 1 much
shorter than antenna 2, ratio of peduncular articles is
16:11:5 (in type-species), primary flagellum as long as
peduncle, accessory flagellum 2-articulate. Antenna 2
large and elongate, scarcely shorter than peduncle.
Labrum as broad as long, weakly notched apically.
Mandibular incisors scarcely toothed, molar triturative,
ratio of palp articles = 11:13:9 (in type-species), article 3
weakly sickle-shaped, setae = DE. Inner lobes of labium
well developed. Maxillae medially setose, inner plate of
rpaxilla 1 ovatotriangular, fully setose medially outer
plate with 11 spines, palps symmetrical. Inner plate of
piaxilla 2 with oblique facial row of setae, strongly
setose medially. Outer plate of maxilliped medially
spinose, palp article 3 unlobed, dactyl shorter than 3,
unguiform, nail almost fused, with secondary scale.
Coxae of ordinary length, poorly setose, coxa 1 weakly
expanded, coxa 4 scarcely lobate. Gnathopods diverse;
female gnathopods simple, feeble, wrists elongate, not
lobate, hands thin, lacking palms; male gnathopod 2
like female but article 2 very broad, gnathopod 1 en¬
larged, strongly subchelate, wrist elongate, hand broad,
dactyl very long and strongly overlapping transverse
sculptured palm. Pereopods 3-4 ordinary. Article 2 of
pereopods 5-7 w'eakly expanded, not lobate, posterior
margins weakly setulate. Pleopods ordinary. Rami of
uropods 1-2 marginally spinose, evenly extended or
outer ramus of uropod 2 shortened, peduncle of uropod
1 without basofacial armament. Uropod 3 extended,
magniramous, almost aequiramous, peduncle elongate,
rami 1-articulate, lanceolate. Telson short, fully cleft,
gaping, lobes tapering, poorly spinose apically. Coxal
gills 2-6 ovate, occasionally pediculate. Oostegites
narrow.

Type Species: Prosocratus butcheri sp. nov.
Relationship: Differing from Cheirocratus in the axial
reversal of male gnathopodal dominance, gnathopod 1
dominating instead of gnathopod 2.

Etymology: From the Greek proso , “in advance of’ and
kratos “power”, refers to the dominance of the first
gnathopod.

Prosocratus butcheri sp. nov.

Figs 5 (lower), 6, 7

Diagnosis: With the generic characters, thus male
gnathopod 1 dominant; female with articles 4-6 of
pereopod 7 slender; urosomite 1 with 3 closely con¬
tiguous dorsal teeth; spines on peduncle of uropod 2 in 2
groups of 2 each; apices of telsonic lobes with minor ar¬
maments only.

Description of Holotype Male “n”: Rostrum minute,
lateral cephalic lobe scarcely emarginate anteriorly,
anteroventral corner rounded; eyes round, pigment not
fully covered by morula of ommatidia.

Antenna 1 of medium extension, ratio of peduncular

articles = 16:12:5 article 1 without teeth, primary
flagellum scarcely longer than peduncle, with thick but
sparse aesthetascs, accessory flagellum 2-articulate.
Antenna 2 elongate, longer than antenna 1, gland cone
large, article 5 of peduncle longer than 4, flagellum
elongate.

Upper lip broad, entire, almost truncate below.

Mandibular incisors scarcely toothed; right lacinia
mobilis thin, left flat, right 4-toothed, left with 2 pairs of
teeth weakly divided at extremities; raker spines main
row r with 5, accompanied by 4 vestigial fern-like setae
near molar, left and right sides similar; molars large,
poorly triturative, each with weak seta; mandibular palp
large, ratio of articles = 11:13:9, article 3 weakly falcate,
articles 1 and 2 with inner setae (article 1 with 1 long, 1
small; article 2 with 7 irregular weakly hooked, of
various sizes), setae of article 3 = DE (medial with 11,
apical with 2). Inner lobes of low r er lip large and fleshy,
outer lobes with medial gape. Inner plate of maxilla 1
broadly pyriform, leaf-like, fully setose medially, and
with facial setae; outer plate slender, with 11 spines;
palps 2-articulate, symmetrical, apices truncate, with
simple, bifid and pectinate spines, facial setae, and an
outer marginal setule. Inner plate of maxilla 2 slightly
broader but scarcely shorter than outer, somewhat
geniculate, with fully developed oblique facial row of
setae, some outer marginal setal spines, sparse medial
and dense apical setae; outer plate almost evenly rec¬
tangular, with 2 apicolateral setae, and dense apical
setae. Inner plate of maxilliped broad, truncate, with 2
inner apical spines, medially and apicolaterally setose;
outer plate ovate, medially and apically spinose; palp
poorly setose laterally, article 2 elongate, article 3
shorter, not lobate, dactyl shorter than article 3,
unguiform, with partly immersed nail, accessory setule
and scale-flake.

Coxae 1-5 of medium length, extending equally, only
coxa 2 with long setae, one on coxa 1 short and thick,
other coxae poorly armed; coxa 1 expanded apically,
coxa 2 weakly constricted midposteriorly, coxae 3-4
subrectangular, 4 not significantly excavate, 3 and 4
equally extended, coxa 5 bilobed, coxae 6-7 shorter than
5.

Gnathopod 1 stout, wrist elongate, with moderately
dense setal clumps posteriorly; hand short, broad,
medial face weakly setose, posterior margin with notch
bearing 2 spines (as if situated on false palm), palm
almost transverse, weakly sculptured, dactyl very long
and overlapping palm. Gnathopod 2 with strongly
broadened article 2 and sharp anterior tooth plus ac¬
cessory tooth; remainder of appendage slender and like
that of female, wrist of medium length, hand much
longer, thin, tapering, without palm, dactyl short and
not folding back far, posterior margins of wrist and
hand with groups of apically bent setae.

Pereopods 3 and 4 subequal.

Article 2 of pereopods 5-7 weakly expanded, not
lobate, posterior margins weakly setulate, legs slender.

Pleopods ordinary. Pleonites 1-2 each with dor¬
solateral setule; epimera 1-3 with sinuous posterior

118

J. L. BARNARD AND M. M. DRUMMOND

Fig. 6 — Prosocratus butcheri gen. nov. et sp. nov. Unattributed figures = holotype male “n” 4.34 mm;
c = female “c” 3.41 mm; p = male “p” 4.18 mm.

margin and sharp posteroventral tooth marked with
weak setule, ventral margin of epimeron 3 with weak
spinule. Urosomitc 1 with pair of medium teeth embrac¬
ing third small tooth enfolding 2 spine setae; urosomite
2 with pair of more widely spaced small teeth each em¬
bracing spineseta. Uropods 1-2 extending equally, outer
ramus of uropod 2 shortened, all rami weakly spinose

marginally and apically; peduncle of uropod 1 with both
margins spinose, without basofacial spine, apicolateral
margin with small sharp cusp, large spine and small
spine; uropod 2 peduncle with 1 lateral and 1 medial
spine at apex. Uropod 3 strongly extended, peduncle
scarcely elongate, apicodorsal margin with 2 spines in
tandem, medial margin with 2 spines in basal group;

CHEIROCRATUS IN AUSTRALIA

119

rami elongate, lanceolate, weakly spinose along
margins. Telson short, cleft to base, lobes slightly
separate at base, lobes tapering to sharp point, each
apex normally with setule (on holotype missing on one
side), then another setule more basally, then pair of long
plumose setules about M60.

Coxal gills on pereonites 2-6, ovate, sausage-shaped
or adz-shaped.

Cuticle under medium power not textured, under oil
immersion showing very faint tiny surficial thorn-scales
and possible granules.

Female: Like male but gnathopods 1-2 simple like male

120

J. L. BARNARD AND M. M. DRUMMOND

gnathopod 2, and article 2 slender; gnathopod 1 actually
shorter than 2, with wrist longer than hand; brood plates
slender, with few setae.

Young Male “p”: Hand narrower than in adult
(Fig. 6).

Holotype: NMV J1639 male “n” 4.34 mm.

Type Locality: WPBES 1707, Australia, Victoria.
Western Port, 7 January 1974, intertidal, sand.
Paratypes: NMV J1640-1642.

Paratype Locality: Female “c”, 3.41 mm; QUBS
Moreton Bay Q., male “p” 4.18 mm; 1 male and 1
female from type locality.

Relationship: Although we have placed this species in a
genus of its own because of the axial reversal of
gnathopodal dominance in the male and the quite
distinctive structure of that dominant gnathopod com¬
pared to the dominant gnathopod of other Australian
taxa in this family group, we must note the very strong
resemblance between this species and Cheirocratus bassi
in dozens of small characters. Most of the morphology
of this species could have been illustrated simply by
reference to the drawings for C. bassi. This is somewhat
unnerving as it means there may be very few increments
of evolution between this taxon and its ancestors which
may lie near C. bassi and that the attribution of generic
importance we give it may be exaggerated.

Smaller differences between P. butcheri and C. bassi
are to be found in (1) shape of the lateral cephalic lobes
of which the anteroventral corner is rounded in P. but¬
cherly weakly produced in C. bassi (2) peduncle of
antenna 2 which is rather stouter in P. butcheri , par¬
ticularly articles 3 and 4; (3) the shorter and stouter
gland cone in P. butcheri which reaches barely to M50
on articles 3 whereas in C. bassi it extends more than
halfway; (4) the left lacinia mobilis, which has really
only 2 branches in P. butcherly 4 in C. bassi ; (5) the
stronger molar seta and (6) the broader plates of maxilla
2 in P. butcheri ; (7) subequal and stouter pereopods 3
and 4 in P. butcheri ; (8) longer dorsal teeth on urosome
1 and shorter urosome 2 in P. butcheri.

The females of these two species are not easy to
distinguish at a glance, but may be separated by the first
gnathopod which in P. butcheri has article 6 distinctly
shorter than article 5, whereas in C. bassi the articles are
subequal.

Placed side by side ovigerous females of these species
of similar length reveal comparative differences;
urosomite 2 is longer in C. bassi and the teeth on
urosomite 1 are shorter. The gland cone in P. butcheri is
stouter and shorter. Prosocratus appears heavier, due to
slightly stouter second antennae, second articles of
pereopods, subequal and stouter pereopods 3 and 4.
Material: CPBS, 1 sample (1 specimen); WPBES 9
samples from 6 stations (32 specimens); WP dredged 1
sample (1 specimen); EBS 1 sample (3 specimens);
QUBS 3 samples from 3 stations (20 specimens).
Distribution: Western Port, Victoria, to Moreton Bay,
Queensland, intertidal, sand, muddy sand, weed
( Posidonia ).

Etymology: This species is named for A. Dunbavin

Butcher, former Deputy Director of Conservation in
Victoria, in recognition of his major role in the benthic
survey programmes in Victoria 1964-74 which have
proved so vastly productive of new amphipod taxa.

ACKNOWLEDGEMENTS

We are grateful to Dr Alistair Gilmour formerly
Officcr-in-Charge, and Dr Leon Collett, of the Marine
Studies Group, Ministry for Conservation, Victoria,
and to Dr Barry Wilson, Director National Museum of
Victoria, for their support of this study. At Smithsonian
Institution we are indebted to Roland M. Brown, Janice
Clark and Deborah Feher for their assistance. The
Secretary of Smithsonian Institution Dr S. Dillon Ripley
made it possible for J. L. Barnard to visit Australia in
pursuit of this study. This publication is No. 380 in the
Ministry for Conservation Environmental Study Series.

REFERENCES

Barnard, J. L., 1972. Gammaridean Amphipoda of

Australia, Part 1. Smithson. Contr. Zool. 103: 1-333.
Barnard, J. L., 1974. Gammaridean Amphipoda of

Australia, Part II. Smithson. Contr. Zool. 139: 1-148.
Barnard, J. L. & Drummond, M. M., 1978. Gammaridean
Amphipoda of Australia, pari III: The Phoxoccphalidae.
Smithson Contr. Zool. 245: 1-551.

Chevreux, E. & Fage, L., 1925. Amphipodes. Faune de
France. 9: 1-448.

Griffiths, C. L., 1975. The Amphipoda of South Africa. Part
5, The Gammaridae and Caprellidae of the Cape Province
west of Cape Agulhas. Ann. S. Afr. Mus. 67: 91-181.
Karaman, G. S., 1977a. Cheirocratus armatus n. sp. from
Suez Region with some remarks to some other members of
this genus (Fain. Gammaridae). Poljopr. Sum. 23: 43-52.
Karaman, G. S., 1977b. New member of the genus
Cheirocratus Nor. from Mediterranean Sea, C. monodon-
tus n. sp. (Fam. Gammaridae), Glasn. Rep. Zav. Zast. Prir.
Zbirke Titogradu. 10: 59-68.

Ledoyer, M., 1967. Amphipodes gammariens de quelques
biotopes de substrat meuble de la region de Tulcar
(Republique Malgache [sic]). Etude systemalique et ecolo-
gique. Ann. Univ. Madagascar 6: 17-62.

Liljeborg, W., 1852. Norges Crustacder. Ofversigt

Konglelige Vetenskaps-Akademiens Forhandlinger, At-
tonde A rgangen 8: 19-25.

Norman, A. M,, 1867. Report on the Crustacea. Nat. Hist.
Trans. Northumb. 1: 12-29.

Ratiike, H., 1843. Beitrage zur Fauna Norwegcns.
Verhandl. Kaiserl. Leopoldinisch-Carolinischen Akad.
Naturforscher. Breslau 20 (1): pp. 1-264.

Sars, G. O., 1895. Amphipoda. An account of the Crustacea
of Norway with short descriptions and figures of all the
species. Christiana, Alb. Cammermeyers Forlag.

Stebbing, T. R. R., 1906. Amphipoda I. Gammaridea. Das
Tierreich 21: 806.

Stepiiensen, K., 1928. Storkrebs II. Ringkrebs 1. Tanglopper
(Amfipoder). Damn. Fauna 399.

Stephensen, K., 1929. Amphipoda. Tierwelt N.-u Ostsee,
Leipzig.

Stephensen, K., 1940. The Amphipoda of northern Norway
and Spitsbergen with adjacent waters. Fasc 3. Troms<\> Mus.
Skr. 3: 279-362.

PROC. R. SOC. VICT. vol. 94, no. 3, 121-132, September 1982

A PRELIMINARY STUDY OF MOVEMENT OF FISHES
THROUGH A VICTORIAN (LERDERDERG RIVER)

FISH-LADDER

By J. P. Beumer and D. J. Harrington

Arthur Rylah Institute for Environmental Research, Fisheries and Wildlife Division, 123 Brown

Street, Heidelberg, Victoria 3084

Abstract: The fish community below the spillway of a diversion weir on the Lerderderg River, Vic¬
toria and the upstream and downstream movements of fishes through a fish-ladder on the weir were
studied over a seven-month period from June to December 1980. Representatives of all five species of fish
recorded below the weir spillway were taken in a two-way trap installed in the fish-ladder. River blackfish
(Gaciopsis marmoratus Richardson) and brown trout ( Salmo trutta Linnaeus) were taken most fre¬
quently. The other species taken were short-finned eel {Anguilla australis Richardson), Australian smelt
(Retropinna semoni (Weber)) and roach (Rutilus rutilus (Linnaeus)). Mean velocity at the centre of the
outflowing weir orifice was 103.9 ±30.9 cm/sec. Although mean water velocities within steps, 18.9-49.3
cm/sec, in the fish-ladder were not considered high, larger river blackfish and brown trout, longer than
180 mm in length, were the major users of the fish-ladder. Two alterations to existing tunnel off-take and
downstream water diversion operations are suggested to increase access-time for fishes to the entrances of
the fish-ladder.

Although dams are continually being constructed in
Australia to provide water for domestic, industrial or ir¬
rigation purposes, relatively few of the existing barriers
have fish-ways or fish-ladders incorporated (Harris
1980). Even for those barriers with some form of fish¬
way, little attention has been paid to monitoring the
effectiveness of their design (Beumer 1980, Harris 1981).
The study of Kowarsky & Ross (1981) on the Fitzroy
River (Queensland) fish-ladder is the most recent and
comprehensive in Australia.

The first fish-ladder to be incorporated on a barrier
in Victoria forms part of the Lerderderg River Diversion
Weir, approximately 45 km north-west of Melbourne.
Pre-impoundment distribution studies on the fishes in
the Lerderderg River suggested that at least two native
species, the short-finned eel, Anguilla australis Richard¬
son 1841 and the river blackfish, Gadopsis marmoratus
Richardson 1848 and one exotic species, brown trout,
Salmo trutta Linnaeus 1758 would be affected by con¬
struction of the weir (Beumer & Harrington 1980a).

In this paper, we report on a seven-month (June to
December 1980) monitoring of the fish movement
through the Lerderderg River fish-ladder and the fish
community immediately downstream of the spillway
and entrance of the fish-ladder.

STUDY AREA

The Lerderderg River is a 60 km long tributary of the
Werribee River which flows into Port Phillip Bay
(Maver & Farmar-Bowers 1979). The diversion weir is
located at the lower end of a gorge section on the river.
The storage area is approximately 2 ha with a capacity of
about 65 ML. A series of tunnels of more than 6 km in
length and a further weir on Goodmans Creek carry
water from the weir into Coimadai Creek where it is
stored in Merrimu Reservoir (Fig. 1). The catchment lies
within a forestry reserve, the only other activity in the
area being gold-prospecting.

The fish-ladder is a pool-type with full weirs and
submerged alternate orifices (Fig. 2a). This design was
chosen to provide a choice of passage for the different
species of fish, either by migrating through an orifice or
over a weir (Clay 1961). A total of 36 vertical and 6
horizontal steps form the ladder, each step being 90 cm
long. A two-way trap to monitor fish movement was
located at Step 36. Steps above the trap-bay are 107 cm
wide while those below the bay are 102 cm wide. The
weirs are 10 cm thick and each step drops 10 cm to the
next. The orifices are 20 x 20 cm 2 in area and located 15
cm above the level of each pool base. Resting pools with
a minimum depth of 20 cm are located at the right-angle
bends at steps 21, 30 and 31 (Fig. 2b). All pools were
normally operated at full depth with 2-5 cm of water
spilling over the weirs. Discharge from the fish-ladder is
at right angles to the direction of normal flow at a point
level with the edge of the spillway. A compensation flow
pipe, carrying water for agricultural requirements,
discharges at a point level with but to the east of the fish-
ladder entrance. The fish-ladder has a 1:10 slope up to
and including the trap-bay and is then at a constant level
to the reservoir proper.

METHODS

Weekly maximum and minimum water temperatures
(°C) were recorded at three locations: (1) in the trap-bay
for the entire monitoring period (24 June to 2
December); (2) above the main body of the reservoir for
inflowing water and (3) 100 m downstream of the
spillway for outflowing water from 7 October onwards.
The height (cm) of water in the trap-bay was read from a
metal rule attached to the upstream wall of the bay.
Discharge figures were recorded by the State Rivers and
Water Supply Commission of Victoria (S.R.W.S.C.) at
O’Brien’s Crossing, approximately 9 km upstream of the
diversion weir.

The fish-ladder trap and the area immediately

B

121

122

J. P. BEUMER AND D. J. HARRINGTON

Fig. 1-Catchment area (-) showing diversion weirs and

tunnels on Lerderderg River and Goodmans Creek.

downstream of the spillway were monitored every 7 days
for the occurrence and movement of fishes from 24 June
to 2 December 1980, inclusive. After this date, the reser¬
voir was drained to allow maintenance of the radial
gates’ control mechanism.

Two dead brown trout on the bank near the fish-
ladder soon after the commencement of the monitoring
led to covering the fish-ladder with a “Nylex” plastic
screen of 12 mm diagonal mesh to prevent further
losses. This screen was attached at a number of points in
such a manner that access to steps 14, 28 and 33 could be
gained readily to allow water velocities to be measured
with a Teledyne Gurley flow-meter at 9 standard loca¬
tions (Fig. 2a): eight within each of these steps and one
in the centre of the outflowing orifice. The velocities
recorded within the steps at the 8 positions were sub¬
jected to a two-way analysis of variance, using a ran¬
domised complete block design (Sokal & Rohlf 1969)
where dates are blocks and step and position are factors,
to test for the effects of date, step and position on the
velocity. The position, open or closed, of the middle
radial gate and side valves (Fig. 2b) was recorded and
periods of tunnel operation were supplied by
S.R.W.S.C.

The area immediately downstream of the fish-ladder
entrance and below the weir spillway was electrofished

with a 240 v. d.c. unit (Moore 1968) at each visit. The
area, approximately 240 m 2 , consisted of large boulde rs
and a gravel substrate and had a maximum depth of %
cm. All fishes were anaesthetised with Quinaldine, iden¬
tified and measured (total length (TL) to the nearest
mm). Fishes longer than 100 mm were marked below the
1st dorsal fin with small blue fingerling tags (Floy
FTF-69), serially numbered with a black legend, to
determine movement of individual fishes. Eels larger
than 340 mm were tagged with anchor tags (Floy Fb-
68A). Fishes were released in the area of capture.

Movement of fishes through the ladder w^s
monitored by a trap covered with galvanised mesh (4
mm diagonal) and installed in the trap-bay. Funnels ex¬
tended from the trap into the upstream and downstream
orifices of the trap-bay (Figs. 3, 4) so that all fishes
would be captured. A section, 9 cm wide, was removed
from the bottom of the cone of each funnel to facilitate
fish entry. Guide rails were added later to facilitate
movement of the funnels. Upstream and downstream
catches were separated by a median division in the trap.
Lids with an overhanging lip of 25 mm covered each
trap section. Weights were placed on each lid to ensure a
complete seal. The trap was checked every 7 days. The
trap-bay was covered by a hinged steel grate and locked.
All fishes were removed from the trap, handled as
above, then released into the reservoir or below the
spillway depending on whether these were taken in the
lower or upper section of the trap respectively. For
individual species, the Student’s ‘t’ test was used to
determine the significance of differences between the re¬
spective mean TL of catches from below the spillway
and lower trap section.

RESULTS

Both maximum and minimum water temperatures in
the trap-bay were lowest during the first half of the
monitoring period after which they rose to the highest
values in late November (Fig. 5). Maximum and
minimum temperatures in the trap-bay were significantly
correlated with those recorded upstream (P<0.01) and
downstream (P<0.05). Water height in the trap-bay was
relatively constant except on 7 October and 2 December
(Fig. 5) when the side-valves and the middle radial gate
respectively were open. Discharge figures at O’Brien’s
crossing show peaks during early July, late August and
mid-September. The middle radial gate and the side-
valves were open on 13 and 3 sampling occasions respec¬
tively, while the diversion tunnel operated for a total
period of approximately 20 days during late June to late
September. The monthly minimum flow regime below
the diversion weir instituted by S.R.W.S.C. is: 120
ML/day (May-October); 24 ML/day (November) and
60 ML/day (December-April). Of this flow, the fish-
ladder carries 4 ML/day, the compensation flow pipe
may release a maximum of 24 ML/day with any
necessary additional flow from the side-valves and/or
the middle radial gate.

Significant differences exist between velocities
recorded at different positions within the steps

FISH-LADDER LERDERDERG RIVER

123

Top View

Fig. 2 — a. Top and side view of four steps in fish-ladder showing positions 1-9 at which velocities
measured, direction of water over weirs and through alternate orifices.

Fig. 2 —b. Diagram of fish-ladder relative to diversion weir and tunnel.

124

J. P. BEUMER AND D. J. HARRINGTON

Fig. 3 —Section through two-way fish-trap
used for monitoring movement in fish'
ladder. Arrows show direction of flow.

Lower section (-) of cone of each fun'

nel removed.

Fig. 4 —Two-way fish-trap, with upstream funnel on right-hand side, in situ in step 36 of the fish-ladder.

FISH-LADDER LERDERDERG RIVER

125

Fig. 5 — Physical parameters and abundance of fishes of the weir. Discharge recorded at O’Briens Crossing.

126

J. P. BEUMER AND D. J. HARRINGTON

Table 1

Mean Velocity± s.d. for each Position and Analysis of Variance for Effects of Date, Step and

Position on Velocity

Position

1

2

3 4

5 6

7 8

Mean

49.3

20.9

18.9 39.1

43.5 29.4

28.2 25.2

Velocity (cm/sec ± s.d.)

24.1

10.4

9.9 19.5

35.3 15.3

12.4 6.4

Source of variation

d.f.

SS

MS

F

Dale

16

18279.3

1142.46

3.63*

Step

2

2367.8

1183.89

3.77|

Position

7

54913.2

7844.74

24.96*

Step x Position

14

5891.0

420.79

1.34

Residual

368

115670.7

314.32

Total

407

197122.0

484.33

*P<0.01

tP<0.05

(Pc0.01), on different dates (PcO.Ol) and to a lesser
extent at which step (Pc 0.05) (Table 1). The two
highest mean velocities within the steps were at the sur¬
face (position 1) and bottom (position 5) proximal to the
inflowing orifice (Table 1) while the mean velocity
(103.9±30.9 cm/sec) at the centre of the outflowing
orifice (position 9) was more than twice any recorded
within the steps.

Almost equal numbers of fishes were taken in the up¬
per (34 specimens) and lower (35) sections of the trap
during the monitoring period. Of the fishes in the trap
not one was damaged. Brown trout eggs were found in
the lower trap section on several occasions. Brown trout
and river blackfish were the major species taken in each
section of the trap, with the former species being the
only one taken in the trap from June until early October
(Fig. 5). In the upper section of the trap the majority of
the catches was taken from mid-October onwards, with
numbers of river blackfish and eels increasing as maxi¬
mum temperatures rose above 15°C. In the lower
section, brown trout were taken only during June to
mid-August, with blackfish, roach and an eel taken from

late October to early December. Below the spillway,
roach, smelt and brown trout were the major species
present from June to the end of September, after which
river blackfish and eels became more abundant in the
catches. No fishes were found in the trap-bay outside the
trap.

All the brown trout taken in the upper section of the
trap were longer than 210 mm TL (Fig. 6). One spent
female was recorded at the end of July. None of the
river blackfish, all greater than 180 mm TL, was ripe
(eggs or milt freely expressed by slight pressure on the
abdomen) but all were larger than the minimum spawn¬
ing size of 120 mm TL (Jackson 1978). The short-finned
eels, from 330 to 570 mm TL, were all immature adults
with the yellow feeding body colouration. The roach
taken was a juvenile.

Brown trout captured in the lower trap section (Fig.
7) were larger than those taken below the spillway on all
corresponding sampling occasions. Comparison of the
mean total length between below spillway and lower trap
section catches for this species showed significant
differences (P<0.01) (Table 2) for particular dates and

Table 2

Comparison of Size between below-Spillway and lower Trap Section Catches

Date

Below Spillway

No. TL±s.d.

Lower Trap Section

No. TL±s.d.

Salmo trutta

1 July

6

129.5 ±27.4

11

293.0± 31.0*

15 July

5

151.0± 70.2

3

369.0± 53.4*

1 July-12 Aug.

26

148.0±46.3

18

313.3 ± 42.6*

Gadopsis marmoratus

18 Nov.

5

210.6±47.6

5

212.2± 26.1

25 Nov.

4

157.5 ± 64.4

7

215.7 ±46.3

28 Oct.-25 Nov.

16

185.1 ±48.3

14

222.4 ±42.9f

* P<0.01
t P<0.05

FISH-LADDER LERDERDERG RIVER

127

for the period 1 July to 12 August. In the lower trap sec¬
tion, ripe males and females were present on 1 July with
ripe females present during mid-July to mid-August.
Brown trout taken below the spillway were juveniles or
immature adults except for one ripe female taken on 15
July.

River blackfish taken in the lower trap section were
greater than or equal in total length to those captured
below the spillway (Fig. 8). Statistical comparison of the
mean total lengths showed no significant differences for
particular dates (Table 2) between the lower trap section
and below-spillway samples although a significant
difference (P<0.05) existed when the period 28 October
to 25 November was considered. In the lower trap sec¬
tion not one of the river blackfish was ripe although all
were of spawning size. Those taken below the spillway
were either immature and mature adults or juveniles.

Roach larger than 90 mm were not taken in the lower
section of the trap (Fig. 8) although these occurred fre¬
quently in samples from below the spillway. These
samples included juveniles, immature and mature
adults, and one ripe male taken on 7 October. Only one
short-finned eel, a small brown elver, was taken in the

lower trap section (Fig. 9). Eels taken below the spillway
were brown elvers (mean TL= l^3.0± 11.6 mm; range
129-175 mm) and immature adults, again with the yellow
feeding body colouration.

A total of 78 fishes was marked with fingerling tags:
39 river blackfish, 33 brown trout, 4 roach and 2 short-
finned eels. A further six eels were marked with anchor
tags. Of the 84 tagged fishes, 21 were recaptured once
(Table 3), and three, all brown trout, were multiple
recaptures. Five of the nine river blackfish recaptured
were taken in the lower trap section, indicating upstream
movement between 7 and 28 days after release. Four of
these five fish originated from the reservoir, as these
were captured initially in the upper section of the trap.

DISCUSSION

While both the number of species and specimens
captured increased with elevated water temperatures
during the monitoring period, the correlation between
inflowing water temperatures and fish-ladder tem¬
peratures and also between fish-ladder temperatures and
outflowing water temperatures suggest that the shallow

128

J. P. BEUMER AND D. J. HARRINGTON

Fig. 7 Length-frequency distribution of brown trout, Salmo trutta , in the lower trap section and below the spillway.

reservoir does not alter the normal temperature regime
of the Lerderderg River. Furthermore, fish behaviour
within the river is not disrupted by fluctuations in
temperature which are often associated with deeper and
larger reservoirs, e.g. Dartmouth Dam (Blyth 1980) or
with diversion operations. Fishes were affected by fluc¬
tuations in discharge with reduced catches related to
peaks of discharge. This is due to a combination of
decreased density of fishes and their displacement at
times of increased water volume and to a lesser extent to
the reduced visibility and consequent lower efficiency of
electrofishing operations in rapidly flowing turbid
waters.

The fish-ladder design permits larger brown trout
and river blackfish to migrate upstream and down¬

stream. Movement through the fish-ladder may be ran¬
dom, as suggested by some of the recapture data for
river blackfish, or be specific as recorded for brown
trout captured on upstream spawning migrations. Short-
finned eels, roach and smelt did not utilise the fish ladder
to the same extent. This may reflect the unsuitable
velocity or the time of sampling. However, Davidson
(1949) found an average swimming speed of 60.8 cm/sec
for specimens of Anguilfa rostrata Lesueur of a size
similar to that of the brown elvers recorded here. This
speed is higher than that recorded in any of the eight
positions within a step. Sorensen (1951) found eels, A.
anguilla Linneaus, 100-150 mm TL, capable of
migrating upstream at water temperatures of 20°C
against currents of between 90 and 130 cm/sec, speeds

Frequency (%)

FISH-LADDER LERDERDERG RIVER

129

Fig. 8 —Length-frequency distribution of river blackfish, Gadopsis mar mo rat us, and roach, Rutilus
rutilus, in the lower trap section and below the spillway.

130

J. P. BEUMER AND D. J. HARRINGTON

90 180 270 360 450 540 630

Total length (mm)

Fig. 9 —Length-frequency distribu¬
tion of short-finned eel, Anguilla
australis , in the lower trap section
and below the spillway.

within the range recorded in this study. The migration of
short-finned eels, A. australis , is temperature dependent
and follows a period of dormancy in winter (Beumer
1979). The upstream movement of the brown elvers
follows a spring-summer pattern (Beumer & Harrington
1980b) and the results suggest this movement was just
commencing when the monitoring period ended.

Although apparently suitable for larger brown trout
and river blackfish, the fish ladder was not shown to be
utilised by small individuals of either species. Studies of
other fish species have shown that larger individuals are
capable of swimming against velocities equivalent to
those recorded here, especially when in peak condition
(Farlinger & Beamish 1978). Even though these
velocities may be high in certain sections of the fish-
ladder, e.g. greater than 100 cm/sec through the
outflowing orifice of a step, several factors, including
the size of the brown trout and river blackfish, the
relatively short distance over which these velocities ex¬
tend and the presence of significantly lower velocity
backwater regions in the steps, reduce the energy expen¬
diture of fishes migrating upstream. The mean velocities
recorded at all eight positions within the steps are below
the threshold value of 60 cm/sec above which rheotactic
coastal off-shore fishes (mean length range 30-60 mm)
are displaced downstream (Schuler & Larson 1975).
Furthermore, studies on marine fishes have shown burst
swimming speeds of more than 100 cm/sec for one sec¬
ond (Dorn et al. 1979) and, assuming that the fishes in
this study have similar swimming capabilities, the

velocities recorded within the study may be within their
capacity and allow upstream migration if made over an
extended period. Successful swimming performances of
fishes require high ambient oxygen levels (Kutty &
Saunders 1973) and are temperature-dependent (Otto &
Rice 1974). The oxygen levels, although not measured in
this study, are assumed to be non-limiting as a result of
high mixing and turbulence of the water while passing
down through the orifices and over the weirs of the fish-
ladder. An increase in swimming performance with in¬
creasing temperature has been recorded for certain
species of freshwater fish (Otto & Rice 1974). The in¬
crease in numbers of fishes, both below the spillway and
in the trap in the latter half of the monitoring period, as
a result of increasing activity may in part reflect a similar
performance for fish species in the Lerderderg River.

Reasons for the limited use of the fish-ladder by
smaller fishes are uncertain. While the possibility of
escape by returning upstream or downstream from the
trap or by escape through the mesh (4 mm diagonal,
2.83 mm square) should not be discounted, several fac¬
tors suggest that this would be minimal. Fishes actively
migrating upstream do so in response to and by orien¬
tating into the downstream-flowing current. The
absence of fishes in the trap-bay outside the trap and the
body-depth (>5 mm) of brown elvers (>100 mm TL)
suggested that there was no escape from the trap
through the mesh. While the extent of escape was not
evaluated, it is assumed that the opportunity to escape
through the funnels would be equal for all species and

FISH-LADDER LERDERDERG RIVER

131

Table 3

Recapture Data for each Species of Fish in 1980
Unless indicated, release and recapture sites are below diversion weir spillway.

Species

TL (mm)

Release Date

Recapture Date(s)

Gadopsis marmoratus

185“

3 Nov.

25 Nov/

202“

28 Oct.

25 Nov/

239“

11 Nov.

25 Nov.

250“

3 Nov.

25 Nov/

262“

28 Oct.

18 Nov/

165

28 Oct.

18 Nov.

252

7 Oct.

21 Oct.

254

3 Nov.

18 Nov.

266

18 Nov.

25 Nov/

Anguilla australis

585

21 Oct.

28 Oct.

Radius rutilus

165

30 Sept.

28 Oct.

196

30 Sept.

21 Oct.

Salmo trutta

103

5 Aug.

19 Aug.

118

26 Aug.

3 Nov./18 Nov./25 Nov./2 Dec.

125

9 Sept.

30 Sept.

134

2 Sept.

9 Sept./21 Oct./28 Oct.

145

3 Nov.

18 Nov.

157

19 Aug.

9 Sept.

158

3 Nov.

11 Nov.

247

30 Sept.

7 Oct.

275

26 Aug.

30 Sept./7 Oct.

“ initially captured in upper trap section.
b recaptured in lower trap section.

the recorded catches in the trap still reflect the migrating
trends and fish-ladder usage. Some fishes 100 mm TL or
smaller are able to attain speeds of 25 body lengths/sec
(Wardle 1975) and the recorded water velocities of this
study are within this limit. Smaller brown trout, river
blackfish, roach and smelt may not make upstream
migrations similar to those of larger, more mature in¬
dividuals but rather utilise the area below the spillway as
a suitable habitat. Alternatively, smaller individuals of
these four species may be incapable of attaining the re¬
quired speeds or be inhibited from entering the fish-way
by the presence of larger individuals.

As with other fish-ladders (Sakowicz & Zarnecki
1962, Dominy 1973, Kowarsky & Ross 1981) and their
capability to allow fishes to pass, a number of problem
areas exist with the Lerderderg River ladder. Possible
modifications to the design of the fish-ladder to improve
its effectiveness by reducing velocities and turbulence in¬
clude bevelling each orifice on both sides (top and
bottom only) or, at least, on the entry side, and the
inclusion of a slot in each weir in such a manner that the
flow passing through the fish-ladder forms two intersect¬
ing sinusoidal lines (Sakowicz & Zarnecki 1962). The
capacity of the fish-ladder and compensation pipe com¬
bined provide the required downstream flow for only
one month (November) each year. For the rest of the
year, the middle radial gate and/or the side-valves must
be opened and this may divert the upstream migrating
fishes away from the fish-ladder entrance. A possible
solution is the controlled release from the three radial

gates in such a manner as to extend the area and in¬
tensity of the velocity of release water from the fish-
ladder entrance outwards. However, the current
operating mechanism for these gates allows only for
releases from the middle gate (to an initial maximum
opening of 45 cm) before the two outer gates may be
opened. It is unlikely that species other than short-
finned eels would make their way upstream over the
spillway. A screen placed along the lower spillway
margin to guide fishes to the fish-ladder entrance may
partly resolve this situation. Fishes, originating from the
reservoir, have also been caught between the bottom of
the middle gate and the spillway (K. Long, S.R.W.S.C.
pers. comm.) when water was released from the reser¬
voir. No further fish were found outside the fish-ladder
after its covering with the Nylex screen. The screen also
prevented leaf-litter and other extraneous material from
entering the fish-ladder, thereby reducing maintenance.

Operational procedures for the diversion of water
through the tunnel may also result in physical displace¬
ment of reservoir and migrating fishes to Goodmans
Creek partly because of the proximity of the fish-ladder
intake to the tunnel entrance. The present procedure of
diverting the entire volume of the reservior above the
fish-ladder intake level when full results in extreme fluc¬
tuations of reservoir level over a relatively brief period.
A gradual continual removal of excess water above the
intake level of the fish-ladder would allow more stable
reservoir water levels and a standard head of water
acting on the upstream entrance of the fish-ladder. In

132

J. P. BEUMER AND D. J. HARRINGTON

addition this would allow the maintenance of flows
through the fish-ladder in summer and autumn when
fishes with temperature-dependent swimming perfor¬
mances and behaviour patterns may by-pass the diver¬
sion weir. Further monitoring with the trap at these
times would provide evidence of seasonal movements by
fishes and may also be of use in controlling the invasion
of the reservoir and upper reaches of the catchment by
undesirable species of fish, e.g. common carp, Cyprinus
carpio Linnaeus 1758, already established in the Melton
reservoir.

ACKNOWLEDGEMENTS

We thank Mark Nelson and other staff from the
Arthur Rylah Institute for assistance with field work; the
State Rivers and Water Supply Commission for
discharge figures; Kevin Long, S.R.W.S.C., for
cooperation and assistance with access to the diversion
weir; Brian Barry of Bacchus March for construction of
the fish-ladder trap; Alicia McShane for drafting sup¬
port and David Powell for biometrical support and ad¬
vice. Drs. Peter Jackson and Phillip Cadwallader and
Mr. Jim Pribble read the manuscript and made many
constructive comments.

REFERENCES

Beumer, J. P., 1979. Feeding and movement of Anguilla
australis & A. reinhardtii in Macleods Morass, Victoria,
Australia. J. Fish Biol. 14: 573-592.

Beumer, J. P., 1980. Fish ladders —steps in the right
direction? Wildl. Aust. 17: 38-39.

Beumer, J. P. & Harrington, D. J., 1980a. Pre¬
impoundment distribution of fishes in the Lerdcrdcrg
River, Victoria. Victorian Nat. 97: 68-72.

Beumer, J. P. & Harrington, D. J., 1980b. Techniques for
collecting glass-eels and brown elvers. Aust. Fish. 39:
16-22.

Blyth, J. D., 1980. Environmental impact of reservoir con¬
struction: the Dartmouth Dam invertebrate survey: a case
history. In An Ecological Basis for Water Resources
Management. W. D. Williams, ed., Aust. Nat. Univ. Press.
Canberra, 174-181.

Clay, C. H., 1961. Design of Fishways and Other Fish
Facilities. Dept, of Fisheries of Canada, Ottawa. 301 pp.
Davidson, V. M., 1949. Salmon and eel movement in con¬
stant circular current. J. Fish. Res. Bd. Can. 7: 432-448.
Dominy, C. L., 1973. Effect of entrance-pool weir elevation
and fish density on passage of alewives (Alosa

pseudoharengus ) in a pool and weir fishway. Trans. Aty
Fish. Soc. 102: 398-404.

Dorn, P., Johnson, L. & Darby, C., 1979. The swimming
performance of nine species of common California insho re
fishes. Trans. Am. Fish. Soc. 108: 366-372.

Farlinger, S. & Beamish, F. W. H., 1978. Changes in blo^d
chemistry and critical swimming speed of largemouth ba$ Sj
Micropterus salmaides, with physical conditioning. Tran s
Am. Fish. Soc. 107: 523-527.

Harris, J. H., 1980. Structures affecting fish migration i n
streams draining the south eastern coastal region of
Australia. Ecol. Select. Est. Organ. Proj. (12-045-16), Ubi.
of N.S. W. Data List. 4: 1-42.

Harris, J. H., 1981. Fishways and fish passage problems | n
Australia. Proc. 16th Assembly Aust. Fresh Wat. Fis} t
9-20.

Jackson, P. D., 1978. Spawning and early development of t| le
river blackfish, Gadopsis marmoratus Richardson (Gado^.
siformes: Gadopsidae), in the Mackenzie River, Victoria.
Aust. J. mar. Freshwat. Res. 29: 293-298.

Kowarsky, J. & Ross, A. H., 1981. Fish movement
upstream through a central Queensland (Fitzroy Rive r )
coastal fishway. Aust. J. mar. Freshwat. Res. 32: 93-109.

Kutty, M. N. & Saunders, R. L., 1973. Swimming
performance of young Atlantic salmon ( Salmo solar) as
affected by reduced ambient oxygen concentration. J. Fisb.
Res. Bd. Can . 30: 223-227.

Maver, J. L. & Farmar-Bowers, Q. G., 1979. En¬
vironmental studies and effects of Dartmouth and
Lerderdcrg Dam projects, Victoria. Gen. Eng. Trans. 3 ;
41-48.

Moore, W. H., 1968. A light-weight pulsed D.C. fish shocker.
J. Anim. Ecol. 5: 205-208.

Otto, R. G. & Rice, J. O., 1974. Swimming speeds of yellow
perch ( Perea flavescens) following an abrupt change in en¬
vironmental temperature. J. Fish. Res. Bd. Can. 3l;
1731-1734.

Sakowicz, S. & Zarnecki, S., 1962. Pool passes—biological
aspects in their construction (Engl, trans.). Rocz. Nauk
Roln. Ser. D. 66: 5-171.

Schuler, V. J. & Larson, L. E., 1975. Improved fish protec¬
tion at intake systems. Proc. Am. Soc. Civil Eng. 101EE6:
897-910.

Sokal, R. R. & Rohlf, F. J., 1969. Biometry-the principles
and practice of statistics in biological research. W. H.
Freeman & Co., San Francisco. 776 pp.

Sorensen, I., 1951. An investigation of some factors affecting
the upstream migration of the eel. Rep. Inst. Freshw. Res.,
Drottningholm. 32: 126-132.

Wardle, C. S., 1975. Limit of fish swimming speed. Nature
(Lond.). 255: 725-727.

PROC. R. SOC. VICT. vol. 94, no. 3, 133-154, September 1982

TRILOBITES FROM THE MOUNT IDA FORMATION
(LATE SILURIAN-EARLY DEVONIAN), VICTORIA

By David J. Holloway* and John V. Neil!

* National Museum of Victoria, 285-321 Russell Street, Melbourne, Victoria 3000
t 23 Michael Street, Bendigo, Victoria 3550

Abstract: A poorly preserved fauna of sixteen trilobite species is described from coarse quartzose
sandstones of the Mount Ida Formation in the Heathcote district of central Victoria. Most of the taxa are
described under open nomenclature but two new species are named, Scutellum droseron and Cheirurus
(Crotalocephalina) oxina. The trilobites support the view of Philip (1967), based on brachiopods, that
Unit 3 of the Mount Ida Formation is of early Gedinnian age, whereas stratigraphic relations indicate that
the lower part of the formation is of Pridolian age.

The stratigraphy of the Silurian-Lower Devonian
sedimentary sequence in the Heathcote district of central
Victoria was originally described by Thomas (1937) and
has been summarised recently by VandenBerg & Garratt
(1976) (Fig. 1). The Mount Ida Formation, the upper¬
most unit in the sequence, consists of approximately
2100 m of sandstones with interbedded mudstones,
shales and conglomerates that were interpreted by
Talent (1965a) as having been deposited in a shallow-
water environment close to the western margin of the
Melbourne Trough. The formation was subdivided by
Thomas (1937) into four units, apparently largely on the
basis of the poorly preserved shelly fauna which is
dominated by brachiopods but also includes molluscs,
trilobites, corals, bryozoans, and rare sponges,
ostracodes and echinoderms. The faunas of Units 1 and
2 are small and characterized by the abundance of the
brachiopods Molongia cf. elegans Mitchell and
Notoconchidium thomasi Gill respectively; Unit 3 con¬
tains the most abundant and diverse faunas in the
Heathcote sequence. No fossils have been found in Unit
4, which contains conglomerates.

With the exception of a short paper on two species of
brachiopods (Gill, 1951), the only descriptions of the
fauna of the Mount Ida Formation are in a monograph
by Talent (1965b) on the faunas of the whole sequence.
Since that work was published large collections of fossils
have been made from the Mount Ida and these contain
numerous trilobites, including several forms not
previously recorded. Most of the trilobites described in
the present study were obtained from Unit 3 of the for¬
mation; Units 1 and 2 have yielded only a few species
(Table 1). The trilobites occur as internal and external
moulds in hard, grey to brown or red quartzose sand¬
stones that are so coarse-grained that fine morphological
features of the exoskeleton have not been preserved. In
addition, the surfaces of the moulds are commonly en¬
crusted with a layer of iron minerals that further
obscures detail. Partially or completely articulated
specimens are very rare, and furthermore the isolated
parts of the exoskeleton have commonly been broken
prior to preservation. This, together with the fact that
the associated brachiopods are all preserved with their
valves separated, is indicative of transport and deposi¬
tion in a turbulent environment.

AGE AND CORRELATION

Unit 3 of the Mount Ida Formation has always been
considered to be of Early Devonian age but there has
been disagreement about its precise correlation with
standard European sequences. Owing to the scarcity of
stratigraphically useful fossils in Units 1 and 2 and in the
underlying Mclvor Sandstone, there has also been con¬
siderable uncertainty as to whether the Silurian-
Devonian boundary should be placed within the lower
part of the formation, at its base, or even below it.
Talent (1965a, b) considered the fossil assemblage of
Unit 3 to be indicative of a broad Gedinnian-Siegenian
age and placed the Silurian-Devonian boundary
(Skalian-Gedinnian boundary in his 1965a correlation
chart, although at that time he included the Skalian
within the Devonian) somewhere within the Mclvor
Sandstone. Philip (1967) placed the boundary much
higher, at the base of Unit 3 ol the Mount Ida, which he
considered to be of early Gedinnian age on the basis ot
the brachiopod fauna. A similar brachiopod fauna oc¬
curs in the upper part of the Boola Beds in the Tyers
area (see Philip 1962, pp. 244-6) and also in the
Maradana Shale in the Manildra district of New South
Wales; the latter was considered to be of early Gedin¬
nian age by Savage (1974). Unit 2 of the Mount Ida For¬
mation was correlated by Philip (1967) with the upper
part of the Florence Quartzite of Tasmania, which he
regarded as pre-Gedinnian because ol the presence ot
encrinurid trilobites (but see discussion below on the up¬
per range of encrinurids). Strusz et al. (1972) equated
the upper part of the Mount Ida Formation with strata
in the Yea and Seymour districts containing Monograp-
tus thomasi and M . aequabiiis notoaequabilis , and from
this concluded that Unit 3 is late Lochkovian (approxi¬
mately late early to middle Siegenian). They placed the
Silurian-Devonian boundary in the Heathcote sequence
at the top of the Mclvor Sandstone which was con¬
sidered to be of Pridolian age because of the entry in
that unit of the brachiopod Notoconchidium .

VandenBerg & Garratt (1976) stated that the upper
part of the Mclvor Sandstone laterally becomes the
Clonbinane Sandstone Member of the Humevale For¬
mation and that this unit contains M. thomasi and M.
aequabiiis aequabiiis at the top. Consequently they
regarded the upper part of the Mclvor as Early Devo-

133

134

D. J. HOLLOWAY AND J. V. NEIL

faulted

MOUNT IDA
FORMATION

MclVOR

SANDSTONE

Unit 4
Unit 3

Unit 2

i

Unit 1

Unit 3

Unit 2

DARGILE

FORMATION

Unit 1

WAPENTAKE

FORMATION

COSTERFIELD

SILTSTONE

base not exposed

Fig. 1 - Silurian-Early Devonian stratigraphic succession in the
Heathcote district.

nian. Talent et al. (1975) also reported the occurrence of
graptolites of the M. hercynicus type in the Clonbinane
Sandstone Member, which they considered to be a
lateral equivalent of the basal part of the Mclvor
Sandstone. However the identification of Devonian
graptolites from this horizon cannot be confirmed and
appears to be incorrect as the Ludlovian species
Bohemograptus bohemicus has been recorded well
above the Clonbinane Sandstone Member at Strath
Creek (Garratt 1978) and has also been identified by Dr.
R. B. Rickards (pers. comm.) from an horizon in the
Yea district just below the Rice’s Hill Sandstone
Member, which Garratt (1978) equated with the Clon¬
binane Sandstone Member. Assuming that the reported
equivalence of all these units is correct, the Mclvor
Sandstone must be of Ludlovian age and is probably
middle to late Ludlovian in view of the occurrence of
early Ludlovian graptolites in Unit 2 of the Dargile For¬
mation, much lower in the Heathcote sequence (Jaeger
1966).

The few species of trilobites found in Units 1 and 2
of the Mount Ida Formation are of no value for correla¬
tion, but it is of interest to note the presence of en-
crinurids near the base of Unit 2. The highest occurrence
of encrinurids previously recorded in the Heathcote se¬
quence was in Unit 4 of the Dargile Formation (the
report by Strusz (1980) of encrinurids occurring in Unit
3 of the Mount Ida Formation was based on incorrect
locality information provided by us). Encrinurids were

long thought to have died out in the Pridolian and their
highest occurrences in certain sequences have been used
to suggest the position of the Silurian-Dcvonian boun¬
dary. Recently, however, they have been recorded from
strata of Lochkovian age in the Tajna and Mitkov Beds
in Podolia (Mozdalevskaya et al. 1968, Nikiforova
1977), the Kokbailal horizon in central Kazakhstan
(Maksimova 1975), the Kunjak horizon in Tien-Shan
(Biske et al. 1977), and the West Point Formation in
Quebec (Bourque & Lesperance 1977). It is not known
whether encrinurids persisted into the Devonian in
eastern Australia. Excluding the Mount Ida Formation,
the youngest occurrences are in the Florence Quartzite
of Tasmania (Gill 1948a), the Derriwong Beds, Trundle
Beds and Wallace Shale of New South Wales (Landrum
& Sherwin 1976, Talent et al. 1975, Strusz 1980), and in
limestone lenses outcropping in the Rockhampton
district of Queensland (McKellar 1969). The ages of
these units lie within the range Pridolian-Lochkovian
but are not known with greater certainty.

Several trilobites from Unit 3 of the Mount Ida For¬
mation provide evidence on the age of the fauna.
Ananaspis serrata (Foerste 1888) was originally des¬
cribed from the Yass Basin of New South Wales, from
beds now considered to lie within the upper part of the
Elmside Formation, which contains conodonts in¬
dicative of the earliest Gedinnian woschrnidti Zone
(Link & Druce 1972). Scutellum droseron sp. nov. and
Cheirurus (Crotalocephalina) oxina sp. nov. are very
close morphologically to species occurring in the early
Gedinnian Kokbaital horizon of central Kazakhstan.
Proetus (Coniproetus) sp. nov. resembles P. (C.) affinis
Boucek 1933 from the Lochkovian of Czechoslovakia.
A slightly younger age than that suggested by the
preceding species is suggested by Sthenarocalymene sp.
A (Chatterton, Johnson & Campbell 1979). This species
occurs also in the Garra Formation of New South

Table 1

Stratigraphic Distribution of Trilobites within the
Mount Ida Formation

Unit 1 Unit 2 Unit 3

Scutellum droseron sp. nov. X

Proetus (Coniproetus) sp. X

Tropidocoryphinae gen. et sp. indet. X X

Harpidella sp. 1 X

Harpidelta sp. 2 X

Cheirurus (Crotalocephalina) oxina
sp. nov. X

Encrinurinae gen. et sp. indet. X

Sthenarocalymene sp. A. (Chatterton,

Johnson & Campbell 1979) X

Homalonotinae gen. et sp. indet. 1 X X

Homalonotinae gen. et sp. indet. 2 X

Homalonotinae gen. et sp. indet. 3 X

A nanaspis serrata (Foerste 1888) X

Odontochile cf. formosa Gill 1948 X X

Acastella sp. X

Acanthopyge (Lobopyge) sp. X

Leonaspis sp. X

EARLY DEVONIAN TRILOBITES

135

Wales, which has yielded late Lochkovian-Pragian con-
odonts (Chatterton, Johnson & Campbell 1979).

In summary, the trilobite fauna of Unit 3 of the
Mount Ida Formation tends to support the early Gedin-
nian age suggested by Philip (1967) on the basis of the
brachiopods, although as only a few trilobite species
have been found to be useful for correlation the
evidence cannot be considered to be sufficient by itself to
give such a precise age. If the middle to late Ludlovian
age suggested above for the Mclvor Sandstone is cor¬
rect, Units 1 and 2 of the Mount Ida Formation must be
at least in part of Pridolian age.

LOCALITIES AND REPOSITORY

The localities referred to in the text are those shown
on the locality map given by Talent (1965b, fig. 1) and
also on the 1:31,680 geological parish maps published by
the Geological Survey of Victoria (Thomas 1940 a, b,
1941). On these maps there is considerable duplication
of locality numbers in different parishes and in order to
avoid ambiguity the numbers are prefixed here by the
initial letter of the parish, as follows: D, Parish of
Dargile; H, Parish of Heathcote; R, Parish of Red-
castle. Some of these localities do not represent actual
outcrops but only loose boulders scattered over the
ground; it is unlikely, however, that these boulders were
moved an appreciable distance from where they were
originally exposed. Locality R25 proved to be much
more productive than any of the others and the greater
part of the material collected came from here. The des¬
cribed specimens are housed in the palaeontological col¬
lections of the National Museum of Victoria (catalogue
numbers prefixed NMVP).

SYSTEMATIC PALAEONTOLOGY

Family Scutelluidae R. & E. Richter 1955

Genus Scutellum Pusch 1833

Type Species (by original designation): Scutellum
costatum Pusch 1833 from the Frasnian of Poland (not
the Givetian of Germany as stated by Chatterton,
Johnson & Campbell 1979; see R. & E. Richter 1926).

Scutellum droseron sp. nov.

Fig. 2P-V

Name: Greek droseros meaning dewy, referring to the
tuberculate surface sculpture.

Type Material: Holotype, NMV P75100, internal
mould of cranidium and counterpart external mould,
from locality R25; Fig. 2P,R,U. Paratypes, NMV
P75101-4 (cranidia), NMV P75105-6 (librigenae), from
locality R25; NMV P75107 (incomplete thorax and
pygidium), from 100 m south of locality R31.
Diagnosis: Dorsal surface of exoskeleton covered with
coarse tubercles that are contiguous on glabella and
become scale-like on outer part of librigena. Glabella in¬
creases in width only slightly between occipital furrow
and front of lp impression but anterior two-thirds is

strongly expanded so that width at frontal lobe is 2.5
times that at occipital furrow. Occipital ring short for
genus, with large median tubercle or small spine on its
posterior margin and large tubercle situated at outer ex¬
tremity of occipital muscle impression. Glabellar muscle
impressions deep but partly obscured by surface tuber-
culation; anterior limb of lp deeper than posterior limb;
2p gently convex forwards and slightly oblique; 3p
poorly defined, transverse. Outline of palpebral lobe
more than a semi-circle; anterior branch of facial suture
almost straight, diverging at 25° to sagittal axis, lying
just outside axial furrow at frontal glabellar lobe. Genal
angle elongate, approximately 35°.

Description: Palpebral width of cranidium almost
equal to width at a-a and approximately 1.4 times the
length (sag.). Occipital ring convex backwards in dorsal
view (with palpebral lobe horizontal), flattened (sag., ex¬
sag.) and merging anteriorly with occipital furrow,
which is bounded in front by a short, abrupt slope. Oc¬
cipital muscle impression subrectangular, extending in¬
wards about two-thirds the distance to the sagittal axis,
lp impression parabolic, somewhat expanded (tr.) prox-
imally, enclosing a small inflated lobe; inner end of 2p
impression level with glabellar midlength; 3p impression
wider and shallower than 2p, lying opposite widest part
of frontal lobe. Axial furrow narrow, shallowest adja¬
cent to occipital ring and deepest at lateral muscle im¬
pression, diverging at 55°-60° to sagittal axis in front of
lp impression. Preglabellar furrow is gently impressed
abaxially and anterior border is short (exsag.) with a
rounded crest; furrow and border both fade adaxially
and become almost obsolete medially; outer face of
anterior border slightly convex (sag., exsag.) and
subvertical, decreasing in height abaxially. Anterior
border and tubercles on front of glabella bear terrace
lines; tubercles immediately behind preglabellar furrow
merge into one or more low, transverse ridges.

Palpebral lobe almost equal in height to glabella;
steeply inclined portion of fixigena medial to palpebral
lobe separated from subhorizontal posterior portion by
a furrow directed slightly obliquely forwards from e
towards lateral muscle impression; in front of palpebral
lobe fixigena slopes anterolaterally. Lateral muscle im¬
pression depressed. Lateral border forms a narrow,
rather sharp, upturned rim inside which the librigena is
flattened or slightly concave, except towards base of eye
(not preserved) where is rises slightly. Posterior border
furrow on librigena shallow and poorly defined proxi-
mally and fading abaxially; posterior border weakly
convex, increasing in length (exsag.) abaxially.

Thorax and pygidium poorly preserved, covered
with coarse tubercles. Pygidial pleural ribs decreasing in
convexity abaxially; median rib (assuming there are
seven pleural ribs) not appreciably wider than pleural
ribs and non-bifurcate distally.

Remarks: S. droseron is closest morphologically to S.
micfwevitchi Maksimova 1968 from the Kokbaital
horizon (Early Devonian) of central Kazakhstan. Both
species differ from characteristic members of the genus
in having a sculpture of coarser tubercles that are con-

136

D. J. HOLLOWAY AND J. V. NEIL

EARLY DEVONIAN TRILOBITES

137

tiguous on the glabella; a very strongly expanded
anterior part of the glabella; a relatively short (sag., ex¬
sag.) occipital ring that lacks a depressed anterior band
medial to the occipital muscle impression; deep glabellar
impressions; and a tubercle situated at the distal edge of
the occipital muscle impression (present in at least some
specimens of S. rnichnevitchi ; see Maksimova 1968, pi.
4, fig. 4). As compared with S. rnichnevitchi , S.
droseron has somewhat smaller tubercles on the
glabella; an axial furrow that diverges more gradually in
front of the lp impression; and a less divergent anterior
branch of the facial suture.

The only Australian species with which worthwhile
comparisons may be made is S. calvum Chatterton 1971
from the Receptaculites Limestone (late Emsian-early
Eifelian) near Yass, New South Wales. S. calvum is a
characteristic representative of Scutellum and so S.
droseron differs from it in the features listed above. In
addition, S. droseron has a deeper occipital furrow, a
more strongly curved outline of the palpebral lobe, a
more divergent anterior branch of the facial suture, a
relatively narrower anterior part of the fixigena and
shorter (exsag.) abaxial part of the anterior border, and
a longer, more acute genal angle. S. calvum has gently
arcuate, transverse ridges on the front of the glabella as
does S. droseron but they were described by Chatterton
(1971) as terrace lines. The ridges in S. droseron are
relatively longer (sag., exsag.) and lower than terrace
lines, and have terrace lines developed on them.

Family Proetidae Salter 1864
Subfamily Proetinae Salter 1864

Genus Proetus Steininger 1831
Subgenus Proetus (Coniproetus) Alberti 1966

Type Species (by original designation): Proetus conden-
sus Pribyl 1965 from the Koneprusy Limestone
(Pragian), Menany, Prague district, Czechoslovakia.

Proetus (Coniproetus) sp. nov.

Fig. 2A-M

Material: At least 20 cranidia, 6 librigenae, 1
hypostome and 40 pygidia from localities R25, R31, ap¬

proximately 100 m south of R31, and various places be¬
tween R25 and R30.

Description: Glabella almost as wide as long (sag.), ex¬
panding forwards across occipital ring, subparallel-sided
between occipital furrow and 6 - 6 , and thereafter
decreasing in width (weakly constricted opposite 7 in
some specimens) and curving steeply downwards. Oc¬
cipital ring with median tubercle and prominent lateral
lobes; occipital furrow deep, medial portion convex for¬
wards and outer portion deflected anterolaterally.
Glabellar furrows lp and 2p faintly visible abaxially but
3p furrow indistinguishable. Preglabellar field absent,
glabella slightly overhanging anterior border furrow.
Margin of palpebral lobe strongly flexed at 6 but 7-6 and
6 -e only weakly curved. Between 7 and /3 anterior branch
of suture diverges at about 13° to sagittal axis; posterior
branch of suture parallel to axial furrow from e to f (the
latter lying opposite midlength of occipital lobe) and
thereafter turning sharply outwards.

Librigena gently convex inside border furrows, rising
beneath eye to become vertical; faint ridge subparallel to
outer margin runs midway between lateral border fur¬
row and base of eye, curving backwards and inwards
posteriorly. Lateral border flattened; posterior border
narrower than lateral border and convex (exsag.);
posterior border furrow sharply impressed. Genal angle
pointed or extended into short spine. Doublure on
librigena convex (tr.) anteriorly, becoming flattened op¬
posite front of eye and narrowing towards posterior ex¬
tremity; inner edge broadly rounded beneath genal
angle, with no sign of panderian notch. Traces of terrace
lines preserved towards outside of lateral border and
doublure.

Pygidium 1.75 times as wide as long, weakly
segmented on axis and pleurae. Axis strongly arched
(tr.), 0.3 times as wide anteriorly as maximum width of
pygidium, narrowing strongly to bluntly rounded ter¬
minus. First ring furrow weaker than articulating furrow
and remainder successively fainter, defining 3-5 or
possibly 6 axial rings. Pleurae curve gently downwards
abaxially and bear 4-5 pleural furrows, of which only the
first is sharp; first interpleural furrow faintly
distinguishable; broad, weakly convex border defined by
slight change of slope.

Remarks: Specimens assigned to this species show varia-

Fig. 2-A-M, Proetus (Coniproetus) sp. nov.; A, C, D, H, I, K, L from locality R31; B, E, F, G, M from
locality R25; J from 100 m south of locality R31. A, D, cranidium NMV P75109; lateral and dorsal views,
x5. B, E, cranidium NMV P75108; lateral and dorsal views, x4. C, latex cast of pygidial mould NMV
P75117; ventral view, x3.5. F, latex cast of librigenal mould NMV P75113; ventral view, x3.5. G, latex
cast of external mould of pygidium NMV P75115; dorsal view, x3.5. H, librigena NMV P78290; dorsal
view, x3.5. I, latex cast of incomplete external mould of cranidium NMV P75111; dorsal view, x4.5. J,
hypostome NMV P75112; ventral view, x5.5. K, latex cast of external mould of librigena NMV P75114;
dorsal view, x4. L, latex cast of external mould of cranidium NMV P75110; dorsal view, x3.5. M. latex
cast of external mould of pygidium NMV' P75116; dorsal view, x 3.5. N, O, Harpidelta sp. 1, from locali¬
ty R25. N, cranidium NMV P78291; dorsal view, x6.5. O, cranidium NMV P48746; dorsal view, x8.
P-V, Scutellum droseron sp. nov.; T from 100 m south of locality R31, remainder from locality R25. P,
R, U, holotype cranidium NMV P75100; x2; P, R, latex cast of external mould in anterior and dorsal
views; U, counterpart internal mould in dorsal view. Q, librigena NMV P75105; dorsal view, x2. S, V,
cranidium NMV P75101; lateral and dorsal views, xl.5. T, incomplete thorax and pygidium NMV
P75107; dorsal view, x0.9. Except where otherwise stated, specimens are internal moulds.

138

D. J. HOLLOWAY AND J. V. NEIL

tion in the convexity (sag.) of the glabella, the weak con¬
striction of the glabella opposite 7 in some individuals,
the degree of taper of the pygidlal axis, and possibly the
length of the genal spine (compare Fig. 2A, D with 2B,
E). Some of the observed differences may be due to tec¬
tonic distortion.

P. (C.) affinis Boucek 1933 from the Lochkovian of
Czechoslovakia is similar to this species but in the
former the eye is situated further from the lateral
border, the hypostome is wider, the pygidial axis is more
strongly segmented, and the pygidial border is better
defined. Only a cranidium of the type species, P. (C.)
condensus , has been illustrated (Pribyl 1965, pi. 1, fig-
1), but Owens (1973, text-fig. 3C, D) figured a cephalon
and a pygidium of P. (C.) glandiferus Novak 1890,
which he considered to be a synonym of P. (C.) conden¬
sus. These specimens differ from the present species in
that the occipital ring is wider (tr.) than the basal part of
the glabella; the palpebral lobes are situated slightly
further forward; the pygidium is more strongly
segmented and more transverse; there is a greater
number ( 7 - 8 ) of axial rings and they are more convex
(sag., exsag.) and have muscle scars impressed abaxially;
and the pygidial border is better defined.

The only previously described species of the
subgenus from Australia is P. (C.) irroratus Chatterton,
Johnson & Campbell 1979 from the Garra Formation
(Early Devonian) of central New South Wales. It is not
very close to the present species, which has a relatively
shorter glabella, more posteriorly placed palpebral
lobes, more weakly divergent section 7-/3 of the facial
suture, no eye socle, and a more weakly segmented
pygidium. There are similarities with the fragmentary
cranidium figured by Chatterton, Johnson & Campbell
(1979, pi. 104, fig. 23) as P. (Coniproelus)? sp., but
cranidia from the Mount Ida Formation have a more
conical glabella, and no preglabellar field.

The present species appears to occur also in the
Humevale Formation near Lilydale, Victoria. Formal
naming of the species will be deferred until work in pro¬
gress on the Lilydale faunas is complete.

Subfamily Tropidocoryphinae Pribyl, 1946
Tropidocoryphinae gen. et sp. indet.

Fig. 4S-V

1965b Proetidae indet. gen. and sp. B. Talent, p. 48, pi.
25, figs. 4, 5.

Material: A cephalon NMV P59652, and pygidium
NMV P59653 (formerly 46739 and 46740 respectively in
the Geological Survey of Victoria collections), from
locality R26; also a cranidium and 4 pygidia from ap¬
proximately 300 m east of locality R25.

Remarks: The cephalon and pygidium from locality
R26 were figured by Talent (1965b). The cephalon is a
badly crushed internal mould with parts of the exo¬
skeleton adhering. The glabella is gently convex, sub¬
parabolic in outline, and lacks lateral furrows. The
occipital furrow is firmly impressed, transverse or
slightly convex forwards medially and deflected for¬

wards distally. The axial furrow is shallow but distinct
and merges with the anterior border furrow in front of
the glabella. The anterior and lateral borders are flat¬
tened and bear fine terrace lines around their outer
margins; the border furrows are broad, shallow and
poorly defined. Talent (1965b) could find no trace of
eyes and concluded that this species was blind but it
seems more likely that the eyes were obliterated during
crushing of the specimen because the palpebral lobe,
outlined by the facial suture, is visible on the left cheek.
The pygidium is wider than long, with a relatively nar¬
row, strongly arched axis and weakly convex pleurae
having a slightly raised rim around the outside. The axial
rings and pleural ribs are flattened (sag., exsag.). The
pleural furrows are very shallow and the interpleural
furrows even fainter. There are traces of very fine ter¬
race lines on the pleurae, expecially around the margins.

This species appears to be closely related to ‘ Proteus*
bowningensis Mitchell 1887 from the upper Ludlovian
to lower Gedinnian of the Yass Basin, New South
Wales. ' P . * bowningensis was assigned by Owens (1973)
to Latiproetus Lu 1962 but Holloway (1980) considered
that until the type species was better known, that genus
should not be used. Talent (1965b) considered the pre¬
sent material to bear some resemblance to Lepp
doproetus Erben 1951 but members of that genus differ
in the structure of the anterior and lateral cephalic
borders (see Alberti 1969, Holloway 1980) and in ad¬
dition have a glabella that is subquadrate instead of
narrowing anteriorly, a more transverse pygidium with a
broader, more conical axis, and pygidial axial rings that
stand higher at the posterior edge than the ring behind.

The specimens from 300 m east of locality R25 are
much smaller and more poorly preserved than those
from R26 but appear to belong to the same species.

Family Aulacopleuridae Angelin 1854
Subfamily Aulacopleurinae Angelin 1854

Genus Harpidella McCoy 1849

Type Species (by monotypy): Harpes? megalops McCoy
1846 from the upper Llandoverian at Boocaun, Cong,
County Galway, Ireland.

Harpidella sp. 1

Fig. 2N, O

Material: Five cranidia from locality R25.

Remarks: These cranidia are assigned to Harpidella
because of the very large, posteriorly placed palpebral
lobe; the presence of an eye ridge; and the distinct
glabellar furrow' 2p (opposite the front of the palpebral
lobe in Fig. 20). In having an elongated glabella that
does not narrow strongly in front of the Ip lobe, very
small lp lobes, and weakly curved lp furrows,
they resemble most closely the species H. distincta,
H. tantula and H. kobayashii described by Pribyl &
Vanek (1981) from the Pragian to Zlichovian of
Czechoslovakia.

A species of Harpidella from the underlying Mclvor

EARLY DEVONIAN TRILOBITES

139

Sandstone was described and figured by Talent (1965b,
pi. 24, figs. 7, 8) as Proetidae indet. gen. and sp. A. It is
nOt very similar to the present material, having a
relatively shorter, anteriorly narrowing glabella and
larger lp lobes.

Harpidella sp. 2

1965b Otarion? n. sp. Talent, p. 48, pi. 25, fig. 6.
Material: A cranidium NMV P59654 (formerly 37979
in the Geological Survey of Victoria collections), from
locality R9.

remarks: This cranidium which was figured by Talent
(1965b) is very similar to Harpidella sp. 1 but the lp
lobes are slightly larger and more triangular in shape,
the preglabcllar field is longer and more convex (sag.),
afid the anterior border is not evenly curved in dorsal
view but is strongly flexed medially.

Family Cheiruridae Salter 1864
Subfamily Cheirurinae Salter 1864

Remarks: The presence of continuous glabellar furrows
2p and 3p is a feature of most Devonian cheirurinids.
All such species were formerly assigned to Crotalo-
cephalus Salter 1853, which was generally considered to
be a subgenus of Cheirurus Beyrich 1845, but in recent
years some of them have been included in four new taxa:
Crotalocephalina Pribyl & Vanek 1964; Crotalocepha-
lides Alberti 1967; Pilletopeltis Pribyl in Pillet 1973; and
Geracephalina Kobayashi & Hamada 1977. This has led
to confusion about the identity of Crotalocephalus itself
because there seems to be no valid designation of a type
species, a problem that has been discussed by Lane
(1971). For present purposes it will be assumed, in ac¬
cordance with traditional usage, that Crotalocephalus
includes species such as Cheirurus (Crotalocephalus)
sternbergii (Boeck 1827) and C. (C.) pengellii (Whid-
borne 1889). Thus Pilletopeltis Pribyl in Pillet 1973
(nomen novum for Boeckia Pillet 1965 non Malm 1870,
and a senior synonym of Pilletopeltis Pribyl & Vanek
1973), which has C. (C.) sternbergii as type species, is
here considered to be a synonym of Crotalocephalus.
Geracephalina is also considered to be a synonym of
Crotalocephalus because the feature on which it is
based-the presence of a sagittal furrow on the 3p and
occasionally the 2p lobe —seems to be of no more than
specific importance, and in at least some cases is
teratological (Prantl 1947).

Genus Cheirurus Beyrich 1845
Subgenus Cheirurus (Crotalocephalina) Pribyl & Vanek
1964

Type Species (by original designation): Cheirurus gib-
bus Beyrich 1845 from the Dvorce-Prokop Limestone
(Pragian), Podoli (Dvorce), Prague, Czechoslovakia.
Diagnosis: Subgenus of Cheirurus having continuous
glabellar furrows 2p and 3p, and a pygidium with short,
broad marginal spines.

Remarks: In addition to the shape of the marginal

spines on the pygidium Pribyl & Vanek (1964, 1973)
considered C. (Crotalocephalina) to be characterised
especially by a “transversely arched and longitudinally
elliptical, somewhat laterally compressed exoskeleton”
(our translation). This is certainly true of species such as
C. (Crotalocephalina) gibbus and C. (Crotalocephalina)
globifrons Hawle & Corda 1847, in which the pleural
regions on the cephalon, thorax and pygidium are
markedly reduced in width (tr.) relative to the axis. Thus
in C. (Crotalocephalina) gibbus the cheeks and the
thoracic pleurae are only about one half the width (tr.)
of the occipital ring and the thoracic axial rings respec¬
tively, whereas in C. (Crotalocephalus) sternbergii and
related species they are almost equal in width to the axis,
or even wider than it. There are, however, a number of
species that resemble C. (Crotalocephalina) gibbus in
the shape of the pygidial spines and yet their pleural
regions are no more reduced in width than those of C.
(Crotalocephalus) sternbergii. These species, which in¬
clude C. (Crotalocephalina) chlupaci Pribyl & Vanek
1962, C. (C.) expansus Balashova 1965, C. (C.) hex-
aspinus (Maksimova 1960), and C. (C.) oxina sp. nov.,
also appear to us to show no significant difference from
C. (Crotalocephalus) sternbergii in the degree of
transverse arching of the exoskeleton.

In our opinion, most of the other characters listed by
Pribyl & Vanek in their diagnosis of C. (Crotalocepha¬
lina) cannot be used to distinguish this taxon from C.
(Crotalocephalus) either. The rate of expansion of the
glabella is variable in both subgenera but the amount of
variation is only slight. Glabellae with inflated frontal
lobes and 2p and 3p furrows that are almost straight are
found also in some species of C. (Crotalocephalus), such
as C. (C.) pengellii (see Lutke 1965, pi. 20, figs. 13, 14)
and C. (C.) copiosus Haas 1968, but these features are
not present in C. (Crotalocephalina) oxina sp. nov. The
genal spines in C. (Crotalocephalina) hexaspinus
(Maksimova 1960) are no shorter than those in most
species of C. (Crotalocephalus). From published il¬
lustrations we are unable to confirm the lack of a prean¬
nulus on the thoracic segments of C. (Crotalocephalina)
gibbus or C. (C.) globifrons , but one is present on the
thoracic segment of C. (C.) oxina described below. At
any rate, a preannulus is present in some species of C.
(Cheirurus) but is poorly developed or absent in others.
Finally, although we can find no published photographs
of C. (Crotalocephalus) showing the ventral surface of
the pygidium, the reconstruction given by Pribyl &
Vanek (1973, text-fig. 3) shows no apparent differences
from C. (Crotalocephalina) in the width of the pygidial
doublure.

There is even some difficulty in separating C.
(Crotalocephalus) and C. (Crotalocephalina) on the
basis of the length and thickness of the pygidial spines
because there is almost continuous variation between the
extremes. In this respect, species such as C. (Crotalo¬
cephalus) pauper Barrande 1852, C. (Crotalocephalus)
maurus Alberti 1966, and C. (Crotalocephalus)
africanus Alberti 1967 lie almost midway between C.
(Crotalocephalus) sternbergii and C. (Crotalocephalina)

140

D. J. HOLLOWAY AND J. V. NEIL

gibbus and in fact bear a closer resemblance to species of
C. (Cheirurus). It seems, therefore, that C.
(Crotalocephalus) and C. (Crotalocephalina) intcrgrade
morphologically and we consider it inappropriate to
recognise them as independent genera, as advocated by
Lane (1971). He proposed that they belong to separate
evolutionary lineages but we can see no evidence to sup¬
port this.

Cheirurus (Crotalocephalina) oxina sp. nov.

Fig. 3M-U

Name: Greek oxina meaning a harrow or rake, referring
to the appearance of the pygidium.

Type Material: Holotype, NMV P75137, internal
mould of pygidium and counterpart external mould,
Fig. 3P, R, S; paratypes, NMV P75138-9, P75143,
P75149 (cranidia), NMV P75140 (hypostome), NMV
P75141-2 (thoracic segments), NMV P75144, P75147-8
(pygidia); all from locality R25.

Diagnosis: Glabella subparallel-sided; furrows 2p and
3 p almost as oblique as lp, so that 2 p lobe is only
slightly longer sagittally than distally. Fixigena broad,
palpebral lobe lying distant from lateral border furrow;
genal spine very short and strongly divergent. Thoracic
pleurae slightly wider (tr.) than axial rings. Marginal
spines on pygidium subtriangular, flattened in cross-
section; first and second spines deflected backwards half
way along their length so that tips of first pair are level
with posterior end of axis; third pair much shorter than
others and directed straight backwards. Posteromedian
projection absent, notch belween third pair of spines
narrowing to an acute angle.

Description: Occipital ring twice as long sagittally as
distally and gently convex (sag.); outermost two-fifths of
occipital furrow occupied by slit-like apodemal pit,
medial portion of furrow clearly distinguishable at
posterior of depressed area between Ip lobes (Fig. 30).
lp furrow' directed obliquely backwards from axial
furrow at approximately 65° to sagittal line, lp apodeme
1.5 times as wide as occipital apodeme. 2p and 3p
furrows shallow slightly adaxially, outer end of 2 p op¬
posite glabellar midlength (sag.). Frontal glabella lobe
0.4 times length (sag.) of glabella. Posterior border
strongly rounded (exsag.) adaxially but decreasing in

convexity beyond fulcrum; lateral border with slightly
angular outward projection in line with posterior border
furrow, and a similar but smaller projection imme¬
diately behind u). Palpebral lobe situated with anterior
edge opposite front of 3p lobe and posterior edge op¬
posite outer end of 2p furrow. Anterior branch of facial
suture falls steeply in front of 7 , becomes horizontal op¬
posite anterior pit in axial furrow, and curves inwards
opposite widest part of frontal lobe; posterior branch of
suture runs parallel to posterior border to meet lateral
margin of cheek at 90°. Regularly spaced tubercles pre¬
sent on glabella, excluding occipital ring; inner part of
fixigena coarsely pitted.

Middle body of hypostome divided into subcircular
anterior lobe and subcrescentic posterior lobe by middle
furrow that is well defined laterally but obsolete me¬
dially. Lateral margins of hypostome converge behind
shoulder at 50°; posterior margin transverse.

Axial ring of thoracic segment strongly arched (tr.)*
approximately equal in length sagittally and distally but
contracting to two-thirds this length half way between.
Medial portion of ring divided by weak transverse
change of slope into small, flattened (sag.), lenticular
anterior band and convex posterior band bearing a pair
of large tubercles. Pleura of usual form for subfamily*
articulated portion comprising half the width; pleural
band increases in height near fulcrum where it merges
with a flattened, lanceolate spine that is sharply
downturned at approximately 45° to the horizontal and
curves backwards slightly distally.

Ratio of pygidial widths across tips of marginal
spines approximately 1.5:1.0:0.3. First axial ring more
prominent than remainder, inclined slightly forwards,
and contracted medially; second and third rings (ex¬
cluding well developed pseudo-articulating half ring on
second) of constant length (sag., exsag.); posterior ter¬
minus a small subcircular swelling. Axial furrow distinct
at first axial ring, poorly defined adjacent to rings 2 and
3 and axial terminus, except at the distal ends of each or
the ring furrows where it contains a deep pit; on internal
moulds axial furrow deeper than on external surface.
First pleural furrow' deep, curving outwards abaxially;
first interpleural furrow sharply impressed proximally
but fading towards the margin and gently curving
backwards. Second pleural furrow also deep, narrower

Fig. 3 — A-L, Sthenarocalymene sp. A (Chatterton, Johnson & Campbell 1979); D from locality R54, re¬
mainder from locality R25. A, B, E, cranidium NMV P47977; lateral, dorsal and oblique views, x2.5. C,
latex cast of external mould of cranidium NMV P75130; lateral view, x2. D, librigena NMV P75136;
dorsal view, x2.75. F, latex cast of external mould of cranidium NMV P75133; dorsolateral view', x2.
G, latex cast of external mould of cranidium NMV P75131; dorsal view, x2.5. H, J, pydigium NMV
P47980; lateral and dorsal views, x2. I, latex cast of external mould of pygidium NMV P47961; dorsal
view, x2. K, latex cast of external mould of pygidium NMV P75134; dorsal view, x2. L, latex cast of
pygidial mould NMV P75135; ventral view, x2.5. M-U, Cheirurus (Crotalocephalina) oxina sp. nov.,
from locality R25. M, incomplete cranidium NMV P75138; dorsal view, x 1.5. N, thoracic segment NMV
P75141; dorsolateral view, x2.75. O, latex cast of external mould of cranidium NMV P75139; dorsal
view, x 1.5. P, R, S, holotype pygidium NMV P75137, x2.5: P. R, latex cast of external mould in dorsal
and lateral views; S, latex cast of counterpart in ventral view. Q, hypostome NMV P75140; ventral view,
x 3.75. T, cranidium NMV P75143; dorsal view, x 1.25. U, incomplete pygidium NMV P75144; dorsal
view, x 1.75. V, W, Encrinurinae gen. et sp. indet., from locality R52. V, pygidium NMV P75146; dorsal
view, x 3.5. W, librigena NMV P75145; oblique view', x 3. Except where stated otherwise, specimens are

internal moulds.

EARLY DEVONIAN TRILOBITES

141

142

D. J. HOLLOWAY AND J. V. NEIL

(tr.) than first and directed more strongly backwards;
anterior pleural band on second segment much lower
than posterior band; second interpleural furrow faint.
Third pleural furrow absent on smaller pygidium but
deep on larger one, which is an internal mould (Fig. 3P,
U). Pygidial doublure deeply notched posteromedially
and narrowing gradually towards anterolateral ex¬
tremities; it slopes gently inwards over most of its ex¬
tent, but around posteromedian notch is flexed sharply
upwards to stand almost vertically. Ventral surface of
marginal spines separated from doublure by short,
abrupt slope.

Remarks: In the width of the fixigenae, the very short
and subtriangular pygidial spines, and the lack of a
posteromedian projection on the pygidium, C. (C.) ox¬
ina resembles C. (Crotalocephalina) expansus Balashova
1965 from the Early Devonian Kokbaital horizon of
Kazakhstan (see Maksimova 1968, pi. 33, figs. 1-4). C.
(C.) expansus differs from C. (C.) oxina in that the
glabella expands more strongly forwards; there is a weak
sagittal furrow on the 3p lobe; the frontal glabellar lobe
is relatively larger and more inflated; the first pygidial
axial ring does not appear to be contracted medially and
a pseudo-articulating half ring is not developed on the
second axial ring; the axial terminus is better defined and
more transverse; the anteriormost pygidial spines are
not directed as strongly backwards; and the notch be¬
tween the third pair of spines is broader and more
rounded in outline. C. (Crotalocephalina) hexaspinus
(Maksimova 1960) also has a pygidium with very short
marginal spines and no posteromedian projection, but
the spines differ in shape from those of C. (C.) oxina ,
and the cranidium bears little resemblance to that of the
Australian species (see Maksimova 1968, pi. 34).

Two cheirurinid species have previously been named
from the Early Devonian of southeastern Australia and
each is known only from a single incomplete cranidium,
so that there is some doubt about their subgeneric
assignment. C. (Crotalocephalus) regius Foldvary 1970
from the Trundle district of New South Wales differs
from C. (C.) oxina in having a relatively longer frontal
glabellar lobe, more oblique 3p furrow, relatively nar¬
rower cheeks, and more prominent tuberculation. C.
(Crotalocephalus) packhami Strusz 1964 from the Garra
Beds near Wellington, New South Wales, has a frontal
glabellar lobe that is more subquadrate in dorsal view
and a 2p lobe that is significantly longer sagittally than
distally.

Etheridge & Mitchell (1917) described two Silurian
cheirurinid species from the Yass Basin of southern New
South Wales as Crotalocephalus silverdalensis and C.
sculptus. Both of these are from beds now assigned to
the Yarwood Siltstone Member of the Black Bog Shale,
which is of late Ludlovian age (Link 1970). C. sculptus
was said to differ from C. silverdalensis in having a pro¬
portionally longer and narrower glabella, a larger fron¬
tal lobe, furrows 2p and 3p that are more acutely flexed
medially, and thoracic segments with a more convex (tr.)
axis and relatively longer pleural spines. On examining
casts of the type specimens of both species, we can see

no significant differences in the proportions of the
glabella or the length of the pleural spines, and we at¬
tribute the other differences to tectonic distortion, c.
silverdalensis and C. sculptus are alike in all other ob¬
servable characters and we consider them to be
synonymous. They are not very similar to C. (C.) oxina ,
and in fact their generic or subgeneric assignment is
uncertain. In the length and thickness of the pygidial
spines they are closer to C. (Crotalocephalina) than to
C. (Crotalocephalus), but they differ from species nor¬
mally assigned to either of these taxa in the form of the
occipital and lp furrows. These furrows do not merge
medially but are separated by a very short (sag.), slightly
depressed portion of the lp lobe bearing subdued
tubercles. The occipital and Ip furrows do merge
medially, however, in the specimens from the Boola
Beds (Early Devonian) of eastern Victoria that Philip
(1962) referred to as C. (Crotalocephalus) silverdalensis.
These specimens also differ from the Yass species in the
shape of the glabella, the inflation of the frontal
glabellar lobe, the form of the 2p and 3p furrows, the
thickness and disposition of the pygidial spines, and the
lack of a posteromedian projection on the pygidium.

Family Encrinuridae Angelin, 1854
Subfamily Encrinurinae Angelin, 1854

Encrinurinae gen. et sp. indet.

Fig. 3V, W

Material: A librigena and 7 pygidia from locality R 52 .
Description: Lateral border on librigena relatively nar¬
row, covered with closely spaced tubercles; lateral
border furrow deep and U-shaped in cross section.
Shallow vincular furrow present on lower edge of border
posteriorly but dying out before reaching pseudo-
glabellar region.

Pygidium approximately 1.3 times as wide as long,
lateral margins converging at about 100° towards
rounded posterior extremity. Axis strongly arched (tr.)
anteriorly but decreasing in height posteriorly and merg¬
ing with postaxial region; 10-15 axial rings are
distinguishable but there is no evidence of median
tubercles on any of them. Pleurae curve steeply
downwards abaxially, composed of 77-10 pleural ribs
which near back of pygidium are deflected slightly in¬
wards abaxially, posterior pair tending to fuse distally
behind axis. On external moulds pleural ribs are flat-
topped and pleural furrows are short (exsag.) and
sharply incised.

Remarks: In the number of axial rings and pleural ribs
and the apparent lack of median tubercles on the axis,
these pygidia resemble the species from the Dargile For¬
mation (Ludlovian) described by Talent (1965b) as En-
crinurus simpliciculus , which we consider to be a junior
synonym of Cromus spryi (Chapman 1912). The type
locality of C. spryi is in the South Yarra area of
Melbourne and was said by Strusz (1980) to lie within
the Anderson Creek Formation of Wenlockian age, but
the strata in this area were assigned to the Dargile For-

EARLY DEVONIAN TRILOBITES

143

mation by VandenBerg (1974). Strusz (1980) noted that
E. simpliciculus resembles species of Cromus in the
glabellar tuberculation but stated that it differs from
them in the more posterior position of the eye, the
preglabellar furrow that is incomplete medially, the lack
of a median longitudinal furrow on the cranidial portion
of the pseudoglabellar region, and the form of the
pygidiunj. We can see no significant difference between.
E. simpliciculus and C. spryi in any of these features.
Although the specimens of C. spryi figured by Strusz
(1980, pi. 1, figs. 1-3) appear to have wider pygidial
pleurae than the types of E. simpliciculus , we attribute
this to tectonic flattening.

Family Calymenidae Milne Edwards 1840

Subfamily Flexicalymeninae Siveter 1976

Genus Sthenarocalymene Siveter 1976

Type Species (by original designation): Sthenarocaly-
mene lirella Siveter 1976 from the Am pyx Limestone
(Llandeilo to lowermost Caradoc), Oslo-Asker district,
Norway.

Remarks: This genus has recently been discussed by
Holloway (1980) who gave reasons for considering it to
be a senior synonym of Apocalymene Chatterton &
Campbell 1980.

Sthenarocalymene sp. A
Fig. 3A-L

1965b Gravicalymene cf. angustior (Chapman); Talent,
p. 49, pi. 26, figs. 3, 4, ?5.

1979 Apocalymene sp. A. Chatterton, Johnson &
Campbell, p. 813, pi. 104, figs. 28, 29; pi. 106, figs 10,
16-27.

Material: At least 15 cranidia, 4 librigenae and 20
pygidia from localities R25 and R54.

Remarks: Our material differs only in a few details from
the description and photographs given by Chatterton,
Johnson & Campbell (1979). They stated that the
palpebral lobes are situated so that d-d passes through
the 2p furrow, but in our material they are slightly
further back. In addition, the fifth pygidial axial ring is
possibly not as well defined in the Mount Ida specimens,
the axis is more distinctly separated from the postaxial
ridge, and the exoskeletal granules tend to increase in
size and become stronger towards the margin of the
pygidium. Some of these differences may be due to the
larger size of the Mount Ida specimens.

The approximately contemporaneous species S.
angustior (Chapman 1915) from the Humevale Forma¬
tion at Lilydale, Victoria has not been well illustrated,
but examination of type and topotype specimens in the
National Museum of Victoria shows that they differ
from Sthenarocalymene sp. A in having a glabella that
narrows forwards more strongly and is more rounded
anteriorly; a palpebral lobe that is situated slightly
further forward (with 5 opposite the 2p furrow); a
higher anterior border; a narrower pygidial axis that is
indistinctly separated from the postaxial ridge; no cinc¬

ture on the pygidium; and pleural and interpleural
furrows that are deeper abaxially. S. hetera (Gill 1945)
and 5. kilmorensis (Gill 1945), which were stated to be
of Gedinnian-Siegenian age by Chatterton, Johnson &
Campbell (1979), are in fact from the Dargile Formation
(Ludlovian) at Kilmore East, Victoria. They are prob¬
ably synonymous, the differences in the shape of the
glabellae evident in the figures given by Gill (1945, pi. 7,
figs. 9, 12) being due almost certainly to tectonic distor¬
tion. These species differ from Sthenarocalymene sp. A
in having much higher anterior borders, glabellae that
narrow more strongly anteriorly and do not project as
far forward with respect to the central part of the cheek,
and relatively longer lp lobes.

Family Homalonotidae Chapman 1890
Subfamily Homalonotinae Chapman 1890

Remarks: The classification of Silurian and Devonian
homalonotinids is in a confused state because the ma¬
jority of described species are in need of revision or are
known only from fragmentary material. Tomczykowa
(1975) recently reviewed all of the established genera and
subgenera (elevating the latter to generic rank) and
reassigned some of the species, but did not undertake a
revision of any of the type species. She suggested that
less weight in classification should be given to the form
of the rostral plate, as the presence or absence of a
rostral process was in her view largely of adaptive
significance. In diagnosing the genera she therefore
employed only features of the dorsal exoskeleton, some
of which appear to us to be of doubtful taxonomic value
at this level. There are no distinctive differences between
any of the genera in the shape of the glabella and the
strength of the thoracic segmentation, and the degree of
glabellar lobation is variable even within individual
species (see Clarke 1913, pi. 2). Of the other features
listed by Tomczykowa in her Table 5, the shape of the
pygidium and its degree of trilobation and segmentation
are extremely variable within the generic groupings she
has made. It is difficult to believe that there are signifi¬
cant differences in any of these features between her
species Trimerus novus and the specimens she has il¬
lustrated as Digonus vialai (Gosselet) and Parahoma-
lonotus forbesi (Rouault) (see Tomczykowa 1975, pi. 3,
figs 5-7, pi. 4, figs 10, 11, pi. 6, figs 1, 3), or between her
species Trimerus lobatus and Dipleura praecox (Tomc¬
zykowa 1975, pi. 2, figs 7, 11). Furthermore the species
Homalonotus clarkei Kozlowski which has a pygidium
with an acutely pointed posterior termination is assigned
by her to Dipleura , the type species of which has a
pygidium that is broadly rounded posteriorly.

The ventral morphology of the exoskeleton is known
in only a few homalonotinid species at present but we
believe that consideration of the structures on the
cephalic and pygidial doublurcs may help in the elucida¬
tion of relationships within the subfamily. For example,
it is apparent that some species possess different styles of
vincular furrows on their pygidial doublures (compare
Salter 1865, pi. 12, fig. 9a with Fig. 4Q herein), whereas
others have none at all (Wolfart 1968, pi. 3, figs lb, 2d)

144

D. J. HOLLOWAY AND J. V. NEIL

EARLY DEVONIAN TRILOBITES

145

and our investigations have shown that a vincular
furrow is present on the cephalic doublure in some
species. These differences presumably reflect differences
in enrollment patterns. We also believe that insufficient
evidence exists at present to dismiss the form of the
rostral plate as a taxonomically useful character.

Homalonotinae gen. et sp. indet. 1
Fig. 4A-H

1965b Trimerus (Dipleura?) sp. Talent, p. 49-50 (in
part), pi. 26, figs 1, 2 (not text-fig 6).

Material: A cephalon, a cranidium, 3 thoracic
segments and 7 pygidia from localities D2, D3 and D4.
Description: Cephalon subtriangular, posterior margin
deflected forwards distally to well-rounded genal angle,
anterior outline weakly curved laterally and interrupted
medially by forwardly projecting anterior border (edge
of border broken off on specimen). Glabella wider
posteriorly than long, narrowing uniformly forwards;
anterior outline transverse; no trace of glabellar
furrows. Axial and preglabellar furrows broad, shallow
and poorly defined; occipital furrow more sharply im¬
pressed than posterior border furrow. Cheek gently con¬
vex (exsag., tr.), sloping anteriorly and laterally from
distal end of occipital ring. Palpebral lobe situated op¬
posite glabellar midlength; anterior branch of facial
suture converges parallel to axial furrow as far as front
of glabella, thereafter becoming indistinct; posterior
branch directed laterally from e but not distinguishable
distally.

Pleurae of thoracic segments have large articulating
facets and broadly rounded tips. Medial to fulcrum,
pleural furrow deep with steep anterior slope and more
gently inclined posterior slope; beyond fulcrum it
shallows and gradually curves forwards in a broad arc.

Pygidium approximately as wide as long, postero¬
lateral margins gently convex. In transverse profile
pleurae slope steeply downwards from axial furrow in a
gentle curve; axis lacks independent convexity anteriorly
but becomes slightly inflated posteriorly. In lateral
profile crest of axis inclined at 50° to plane of lateral
margin. Axis composed of strongly tapering segmented
portion of 10 axial rings and a subparallel-sided ter¬
minus extending to posterior extremity of pygidium;

ring furrows distinctly flexed medially in dorsal view.
Anterior width of axis almost two-thirds maximum
pygidial width. Axial furrow weaker than ring furrows
anteriorly but increasing in depth slightly posteriorly.
Pleura on anteriormost segment divided into short (ex¬
sag.) anterior band and gently convex (exsag.) posterior
band by deep, sharp pleural furrow that is continuous
with articulating furrow proximally and runs onto ar¬
ticulating facet distally; anterior band expands abaxially
as far as inner edge of facet, posterior band expands
slightly beyond fulcrum. Remainder of pleurae com¬
posed of 7 ribs defined by shallow rib furrows that fade
towards lateral margin; between posteriormost rib and
terminal piece of axis is a gently concave, subtriangular
region formed in part of axial furrow and eighth rib
furrow. Doublure narrow and semitubular; flattened
and steeply inclined portion of border roll immediately
behind articulating facet bounded below by slight vin¬
cular furrow (Fig. 4F).

Remarks: This material is very close to the species from
the Humevale Formation described by Gill (1949) as
Trimerus lilydalensis , and may actually belong to it. The
collections of the National Museum of Victoria contain
several topotype and other specimens of T. lilydalensis
that are better preserved than the types, but in the pre¬
sent state of knowledge of Silurian and Devonian
homalonotinids they cannot be assigned with certainty
to any established genus. The closest similarity seems to
be with certain species presently referred to Burmeisteria
(Digonus), such as B. (D.) noticus (Clarke 1913) and B.
(D.) aecraensis Saul 1967. T. lilydalensis differs from
these species, however, in that the rostral suture does
not run along the margin of the anterior border, so that
the rostral plate has a small, subtriangular dorsal por¬
tion; the rostral plate lacks a medial process; and there is
a well-developed vincular furrow on the cephalic
doublure, directed anteromedially from the outer end of
the hypostomal suture. There are also similarities with
Trimerus (Dipleura), especially in the form of the rostral
plate and in the fact that the anterior branch of the
facial suture joins in a smooth curve in front of the
glabella (Wolfart 1968, pis 1-3; note that the appearance
of a transverse rostral suture in some of the illustrations
by Hall & Clarke 1888, pis 2-5, is due to breakage of the
specimens, and the diagnosis and reconstruction given

Fig. 4 —A-H, Homalonotinae gen. et sp. indet. 1; A, B, from locality D4; C, D from locality D2; E-H
from locality D3. A, B, cephalon NMV P78292; lateral and dorsal views, x 1.5. C, fragmentary thoracic
segment NMV P78293; dorsolateral view (anterior at top of photograph), x2. D, fragmentary thoracic
segment NMV P78294; dorsolateral view (anterior at top of photograph), x2. E, H, pygidium NMV
P78295; dorsal and lateral views, x 1.75. F, G, pygidium NMV P59656, figured by Talent (1965b, pi. 26,
fig. 2); F, anterior part of border roll in right ventrolateral view, showing slight vincular furrow, x 2.5; G,
dorsal view, x 1.5. I-L, Homalonotinae gen. et sp. indet. 2, from locality H25. I, J, incomplete cephalon
NMV P78296; dorsal and lateral views, x 1.5. K, L, pygidium NMV P78297; lateral and dorsal views,
x2. M-R, Homalonotinae gen. et sp. indet. 3; O, P, from locality R25, remainder from locality R55. M,
N, Q, pygidium NMV P78298; lateral, dorsal and ventral views, x 1.25. O, P, librigena NMV P78299;
ventral and dorsal views (anterior at top of photographs), x 1.5. R, fixigena NMV P78300; dorsal view,
x 1.25. S-V, Tropidocoryphinac gen. et sp. indet.; U from locality R26, remainder from approximately
300 m east of locality R25. S, cranidium NMV P78301; dorsal view, x6. T, latex cast of external mould
of pygidium NMV P78302; dorsal view, x5. U, partly exfoliated cephalon NMV P59652, figured by
Talent (1965b, pi. 25, fig. 4); dorsal view, x3. V, pygidium NMV P78303; dorsal view, x5.25. Except
where otherwise stated, specimens are internal moulds.

146

D. J. HOLLOWAY AND J. V. NEIL

by Sdzuy, in Moore 1959 are incorrect in this respect).
Species of T. (Dipleura) differ from T. Ulydalensis in
lacking a vincular furrow on the cephalic doublure, and
in having a more weakly segmented pygidium that is
rounded posteriorly instead of pointed.

The cranidium from the underlying Mclvor Sand¬
stone described and figured by Talent (1965b, text-fig. 6)
as belonging to the same species as pygidia from the
Mount Ida Formation differs from the cephalon describ¬
ed above in the shape and convexity of the glabella and
in the presence of indistinct lp lobes.

Homalonotinae gen. et sp. indet. 2
Fig. 4I-L

Material: A cephalon and a pygidium from locality
H25.

Description: Glabella narrows forwards at a constant
rate and bears faint traces of lp and 2p furrows; at the
front it curves abruptly downwards into what appears to
be a deep and sharply recessed preglabellar furrow. Ax¬
ial furrow broad and shallow, except near from of
glabella; paraglabellar area elliptical and extending for¬
wards almost to point opposite middle of palpebral
lobe. Anterior branch of facial suture converges gently
and passes close to anterolateral extremity of glabella;
lateral to eye, librigena descends vertically to lateral
border, or is even inclined slightly inwards.

Pygidium longer than wide. Axis approximately one-
half width of pygidium anteriorly; posteriorly it nar¬
rows, increases in convexity, and merges with strongly
projecting post-axial region. There are at least 10 axial
rings. Pleurae composed of 7 pleural ribs, anteriormost
one divided into anterior and posterior bands by deep
pleural furrow.

Remarks: Amongst established Silurian and Devonian
homalonotinid genera, the only one with a deep, sharply
recessed preglabellar furrow seems to be Homalonotus
itself (see Salter 1865, pi. 12, fig. 2; McLearn 1924, pi.
27, fig. 14). The present species is also not unlike
Homalonotus in the shape of the glabella, the indistinct
glabellar Iobation, and the relatively well-defined
paraglabellar areas. The pygidium is not as strongly
segmented as in described species of Homalonotus , but
is more like that of species such as Burmeisteria
(Digonus) clarkei (Kozlowski 1923, pi. 1, figs. 14, 15).

Homalonotinae gen. et sp. indet. 3
Fig. 4M-R

Material: A librigena, a fixigena, a fragmentary
thoracic segment, and 2 pygidia from localities R25 and
R55.

Remarks: The librigena has a gently convex central
region separated from the less steeply inclined lateral
border by a broad flexure that dies out posteriorly. The
anterior branch of the facial suture curves in a broad arc
to converge gradually with the outer margin of the
cheek, so as to isolate an acute anterior projection on
the librigena. The posterior portion on the librigenal
doublure is narrow and steeply inclined; the anterior

portion expands strongly forwards and near its outer
edge bears a vincular furrow that dies out anteriorly.
The pygidia differ from those of Homalonotinae gen. et
sp. indet. 1 in that the posterior extremity is more acute
and is slightly upturned; the axis is more convex (tr.) and
does not appear to taper as strongly; the pleural rib
furrows are weaker; and the vincular furrow is much
stronger.

Family Phacopidae Hawle & Corda 1847
Subfamily Phacopinae Hawle & Corda 1847

Genus Ananaspis Campbell 1967
Type Species (by original designation): Phacops fecun-
dus Barrande 1846 from the Kopanina Formation
(Ludlovian), Kolednik near Beroun, Czechoslovakia.

Ananaspis serrata (Foerste 1888)

Fig. 5A-P

1965b Phacops sp. cf. P. serra/ws Foerste; Talent, p. 50
pi. 26, figs 6-8.

1971 "Phacops”serratus Foerste; Sherwin, p. 83, pi. \ y
figs 1-10, pi. 2, figs 1-5 (with full synonymy).

1977 Paciphacops (Paciphacops) serratus (Foerste)-
Campbell, p. 32.

Type Material: The neotype, selected by Sherwin
(1971) is Australian Museum F27132, a partly exfoliated
thorax and pygidium with disarticulated and inverted
cephalon, figured by Sherwin (1971, pi. 1, figs 1, 5-7,
10), from the Upper Trilobite Bed ( = Elmside Forma¬
tion, early Gedinnian), near Bowning, New South
Wales. The specimen on which Foerste based his
species-an internal mould of the dorsal exoskeleton
with disarticulated cephalon —is apparently lost.

Other Material: At least 50 cephala, 20 pygidia and an
incomplete dorsal exoskeleton from localities R9, R24,
R25, R30, R31, R54, approximately 300 m east of R25*
and approximately 100 m south of R31.

Remarks: The taxonomic importance of Paciphacops
(Paciphacops) Maksimova 1972 to which the present
species was assigned by Campbell (1977), has been ques¬
tioned by Holloway (1980) who considered it to be at
most a subgenus of Ananaspis. Paciphacops is ap¬
parently distinguished from Ananaspis (s.s.) only in the
presence of perforate glabellar tubercles in large-eyed
morphs and thickened sclera between the lenses in all
small-eyed and most large-eyed morphs. Perforations in
the glabellar tubercles have not been observed in the pre¬
sent material or in the topotypes, although the quality of
the preservation is such that they would probably have
been obliterated even if they were originally present. Eye
dimorphism does not appear to be present in A. serrata
but none of the specimens has thickened sclera. Thus we
can see no morphological grounds for separating A. ser¬
rata at the subgeneric level from other members of
Ananaspis (s.s.)

The most distinctive feature of A. serrata is the
presence of axial spines on the occipital ring, thorax and

EARLY DEVONIAN TRILOBITES

147

pygidium. Similar spines (but not on the pygidium) are
also developed in A. claviger (Haas 1969) from the
Siegcnian of Nevada and in some members of
Viaphacops Maksimova 1972, although in the latter
their presence on the thorax is extremely variable even
amongst individuals from the same population, and this
has led to the suggestion that they are a sexually dimor¬
phic character (Eldredge 1973, Campbell 1977). There is
some evidence that the axial spines in A. serrata are also
dimorphic as they are present in only one of the pygidia
from the Mount Ida Formation (Fig. 5N), and their
absence in the other pygidia cannot in all cases be ex¬
plained by breakage. The only thorax of A. serrata
available to us from the Mount Ida is too poorly
preserved to retain traces of axial spines (Fig. 5K). It is
not possible to argue a strong case for dimorphism on
the basis of the present material, but it is of interest to
note that Etheridge & Mitchell (1895) suggested that A.
serrata may be a sexual dimorph of their species A.
crossleii which occurs in the same beds in the Yass Basin
and lacks axial spines.

Our specimens of A. serrata are mostly larger than
those figured by Sherwin (1971) but they are alike in the
majority of characters. The main points of difference are
as follows.

1 . The only eye in which Sherwin was able to count the
lenses had 14 dorsoventral lens files of up to five lenses
each. The most complete lens count obtained for our
material is from the specimen in Fig. 5F in which the eye
contains 17 files with a maximum of five lenses per file,
although the posteriormost file seems to contain only a
single lens. In another specimen, however, the eyes con¬
tain at least six lenses in the longest file. There is also
evidence of some variation in the overall size of the eye
but the information is inadequate to determine if dimor¬
phism is present.

2. The genal angle in at least some of the Mount Ida
specimens is more pointed than it is in the topotypes (see
Fig. 5H).

3. Most of the pygidia in our collection contain 7 axial
rings and 5-6 pleural furrows, the same numbers given
by Sherwin, but one of the largest (the only one with
medial spines on the axial rings) has 9 axial rings and 7
or possibly 8 pleural furrows.

Sherwin (1971) stated that the cephalon of A. serrata
from the Mount Ida Formation figured by Talent
(1965b, pi. 26, figs 6, 8) differs from topotype cephala in
that the glabella more strongly overhangs the anterior
border and the medial portion of the vincular furrow is
deeper, but we are unable to confirm this. Although
Sherwin describes the medial portion of the vincular
furrow as “very shallow’’, it is in fact moderately deep in
the neotype, which retains the exoskeleton on the
doublure (Sherw'in 1971, pi. 1, fig. 10). Any slight
difference in the depth of the furrow in Talent’s
specimen can be adequately accounted for by the fact
that it is an internal mould.

The species A. claviger (Haas) referred to above has
been compared with A. serrata by Sherwin (1971). In ad¬
dition to the differences noted by him, A. claviger has

eyes that are relatively larger than those of A. serrata
and extend closer posteriorly to the axial furrow and the
posterior border furrow; the sclera is thickened between
the lenses; the intercalating ring is more distinctly
separated medially from the remainder of the glabella;
and the pygidium has a coarser tuberculate ornament
and better defined pseudo-articulatmg half rings on
segments 2 to 5 at least.

Several specimens of aphacopid have been obtained
from mudstones within the Mount Ida Formation at
locality R56. The cephalon bears genal spines and there
are very long axial spines on the occipital ring and
thorax but apparently not on the pygidium. The
material is inadequate for meaningful comparison with
A. serrata.

Family Dalmanitidae Vogdes 1890
Subfamily Dalmanitinae Vogdes 1890

Genus Odontochile Hawle & Corda 1847

Type Species (1CZN Opinion 537 (1959)): Asaphus
hausmanni Brongniart 1822 from the Dvorce-Prokop
Limestone (Pragian), Prague district, Czechoslovakia.

Odontochile cf. formosa Gill 1948
Fig. 6M-U

cf. 1948b Odontochile formosa Gill; p. 20, figs 1, 2.
1965b Dalmanitidae gen. indet. B Talent, p. 51, pi. 27,
fig. 1.

1965b Dalmanitidae gen. indet. C Talent, p. 51, pi. 27,
figs 6, 8 (not pi. 27, fig. 7 =Acastella? sp.)

Material: At least 18 cephala, 35 pygidia, several
fragmentary fixigenae and thoracic segments, and a
detached visual surface from localities R20, R25, R31,
between localities R25 and R30, approximately 300 m
east of locality R25, and approximately 100 m south of
locality R31.

Remarks: Odontochile formosa was erected by Gill
(1948b) for specimens from the Humevale Formation
(Early Devonian) at Kinglake West, Victoria, and has
been redescribed and discussed by Jell & Holloway (in
press) on the basis of material from a different
stratigraphic horizon and locality. The present material
is too fragmentary for a confident assignment to O. for¬
mosa but closely resembles that species in features such
as the shape of the glabella and the proportions of the
frontal lobe; the anterior border that is well-rounded in
outline and lacks a medial process (see Fig. 6S); the
number of lenses (14) in the longest dorsoventral file of
the eye; the convex (tr.) eye platform that closely abuts
the lateral border furrow (Fig. 6S); and the lack of a
distinct inner flange on the pygidial doublure. The
pygidia show considerable variation in the number of
segments, smaller specimens having 10-12 axial rings
and 11-12 or possibly 13 pleural furrows, whereas the
largest has 17 axial rings and 15 pleural furrows. We
have therefore been in some doubt as to whether more
than one species is represented, although the difference
in the size of the pygidia suggests that the variation is

148

D. J. HOLLOWAY AND J. V. NEIL

EARLY DEVONIAN TR1LOBITES

149

due to ontogenetic changes. Specimens of O. formosa
from the same area as the types show much less varia¬
tion in the pygidial segmentation, even small specimens
having 16-17 axial rings and 12-13 pleural furrows,
whereas the largest have one or two more of each.

The presumed ontogenetic changes would account
for the differences between the pygidia described by
Talent (1965b) as Dalmanitidae gen. indet. B and
Dalmanitidae gen. indet. C. It has already been pointed
out by Shergold (1968) that the cranidium figured by
Talent (1965b, pi. 27, fig. 7) as Dalmanitidae gen. indet.
C. in fact belongs to an acastid.

Family Calmoniidae Delo 1935
Subfamily Acastinae Delo 1935
Remarks: We agree with Eldredge (1971) that
Acastavinae Struve 1958 is a synonym of Acastinae.

Genus Acastella Reed 1925

Type Species (by original designation): Phacops
(Acaste) Downingiae var. 5, spinosus Salter 1864 from
the Upper Whitcliffe Beds (latest Ludlovian), Ludlow,
Shropshire, England.

Acastella sp.

Fig. 5Q-Z

1965b Dalmanitidae gen. indet. C Talent, p. 51, pi. 27,
fig. 7 (not pi. 27, figs. 6, % = Odontochile cf. formosa
Gill 1948).

Material: At least 3 cephala, 15 cranidiaand 13 pygidia
from localities R25, R31, and approximately 100 m
south of R31.

Description: Anterior margin of cephalon deflected
gently forwards medially around front of glabella; maxi¬
mum curvature at mid line. Glabella subquadrate, in¬
creasing in width anteriorly as far as 2p furrow,
narrowing slightly to 3p furrow and thereafter expand¬
ing so that width at frontal lobe is almost equal to length
(sag.). Occipital ring more strongly arched transversely
than remainder of glabella, well rounded sagittally and
exsagittally but with highest point close to posterior

margin; occipital furrow deflected slightly backwards
abaxially towards apodemal pit. Ip furrow directed ob¬
liquely backwards from axial furrow, becoming
transverse adaxially and deflected slightly forwards
proximally; lp apodemal pit isolated from axial furrow
and distinctly expanded at its inner end. 2p furrow
shallower than lp, transverse or slightly oblique, ter¬
minating relatively far from axial furrow distally. 3p
furrow lying at about 60° to sagittal axis, expanding
abaxially and abruptly shallowing just medial to axial
furrow. Axial furrow rises gently between occipital ring
and a point level with 2p furrow, thereafter falls steeply
to outer end of 3p furrow and is very deep, but flattens
out adjacent to frontal lobe and becomes indistinct (Fig.
5U). Palpebral lobe situated with 5 opposite 2p furrow
and e close to posterior border furrow; anterior branch
of facial suture converges slightly between y and 0. In¬
complete eye preserved on specimen in Fig. 5R has at
least 6 lenses in the longest dorsoventral file. Lateral
border indistinctly separated from central part of cheek,
which rises steeply adaxially so that recessed region
below eye is vertical. Posterior border short (exsag.) but
prominent, expanding slightly abaxially; posterior
border furrow deep proximally but abruptly dying out
distally. Posterior edge of cephalon curves backwards
distally, indicating that a genal spine was probably pre¬
sent although it is not preserved (Fig. 5R, U). Medial
part of cephalic doublure convex (sag.) anteriorly and
concave behind; hypostomal suture transverse or gently
convex forwards. Lateral part of doublure narrows
towards genal angle and curves sharply upwards and
outwards towards inner margin which is almost in con¬
tact with inner surface of cheek.

Posterior margin of pygidium of almost uniform
curvature in dorsal view, deflected slightly upwards
medially in posterior view. Axis composed of 6-7 axial
rings; posterior to third ring, rate of taper of axis
decreases slightly, apodemes abruptly decrease in size,
and axial rings are more flattened (sag., exsag.). Pleurae
curve steeply downwards abaxially to a gently inclined
border. There are 3-4 successively shallower pleural
furrows and 2-3 weak interpleural furrows. Doublure
with a narrow, subhorizontal outer part and an almost

Fig. 5 — A-P Ananaspis serrata (Foerste 1888); B, C, Irom locality R24; K Irom 100 m south ol locality
R31; M from locality R25; remainder from 300 m east of locality R25. A, E, latex cast of external mould
of cephalon NMV P78304; lateral and dorsal views, x2. B, C, cephalon NMV P78305; lateral and
antcrovenlral views, x2.5. D, G, cephalon NMV P78306; lateral and dorsal views, x2. F, latex cast of
external mould of cheek NMV P78307; dorsolateral view, x2.5. H, latex cast of external mould of
cephalon NMV P78308; oblique view, x 2.25. L latex cast of cephalic mould NMV P78309; ventral view,
x 2. J, N, O, pygidium NMV P78310, X 2; .1, N, latex cast of external mould in lateral and dorsal views;
O, latex cast of counterpart mould in ventral view. K, incomplete cephalon and thorax NMV P7831 1;
dorsolateral view, x 1.75. L, latex cast of external mould of pygidium NMV P783I2; dorsal view, x3.
M, pygidium NMV P78313; dorsal view, x 1.75. P, latex cast of external mould of cephalon NMV
P78314; anterior border and doublure in anterolateral view showing vincular furrow, X 3. Q-Z, Acastella
sp; Q from locality R31; T from 100 m south of locality R31; remainder from locality R25. Q, pygidium
NMV P78315; dorsal view, x3.75. R, U, cephalon NMV P78316; lateral and dorsal views, x2.25. S,
cephalic doublure NMV P78317; ventral view, x2.75. T, cranidium NMV P78318; dorsal view, x3. V,
cephalic doublure NMV P78319; ventral view, x2.25. W, latex cast of pygidial mould NMV P78320;
ventral view, x2.75. X, latex cast of external mould of pygidium NMV P78321; dorsal view, x3.5. Y,
latex cast of external mould of cranidium NMV P78322; dorsal view, x3.5. Z, cranidium NMV P78323,
dorsal view, x3.75. Unless otherwise staled, specimens are internal moulds.

150

D. J. HOLLOWAY AND J. V. NEIL

vertical inner flange, the junction between the two being
subangular.

Remarks: There is some variation between specimens in
the width of the glabella at the frontal lobe relative to
that at the 2p furrow, the width (tr.) of the glabellar fur¬
rows, and the width of the pygidial pleurae as compared
with that of the axis. However, in the lack of convincing
evidence to the contrary, we assume that only a single
species is represented.

Shergold (1968, p. 21) suggested that the incomplete
cranidium from the Mount Ida Formation figured by
Talent (1965b, pi. 27, fig. 7) belongs to Acaste. We
prefer to assign the present material (as well as that of
Talent) to Acastella because of the rather pointed
anterior cephalic outline (as shown by the shape of the
front of the doublure; Fig. 5S, V); because genal spines
were probably present; and because in some of the
cranidia (Fig. 5T, Z) it can be seen that the anterior
branch of the facial suture diverges from the
preglabellar furrow medially and joins at an obtuse
angle, isolating a small, triangular portion of the
cranidium in front of the glabella. This last feature is
present in most previously described species of Acastella
and is particularly well develolped in A. tiro R. & E.
Richter (1954, pi. 5, fig. 73d).

The only other acastids recorded from Australia are
the species Acastella frontosa and Acaste longisulcata
described by Shergold (1968) from the Early Devonian
Humevale Formation near Lilydale, Victoria. A. fron-
tosa differs from the present material in having a larger
subtriangular portion of the cranidium enclosed by the
facial suture in front of the glabella, a relatively longer
glabella, and a more strongly segmented pygidium with
relatively broader pleural regions. The holotype of A.
longisulcata is an internal mould of a pygidium figured
by Chapman (1915, pi. 15, fig. 15) as Phacops crossleii
Etheridge & Mitchell 1896. This pygidium does in fact
belong to a phacopid, as indicated by the form of the ar¬
ticulating facets and the slight forward deflection of the
axial rings towards their distal ends. This deflection of
the axial rings is caused by the pattern of insertion of the
appendage muscles (see Campbell 1975, fig. 1, pi. A, fig.
6). The other specimens figured by Shergold as A.
longisulcata are acastid.

Family Lichidae Hawle & Corda 1847
Subfamily Ceratarginae Tripp 1957

Genus Acanthopyge Hawle & Corda 1847
Subgenus Acanthopyge (Lobopyge) Pribyl & Erben
1952

Type Species (by original designation): Lichas
Branikensis Barrande 1872 from the Dvorce-Prokop
Limestone (Pragian) at Prague, Czechoslovakia.
Remarks: This taxon was considered by Tripp (in
Moore 1959) to be a synonym of Acanthopyge (s.s.) but
we accept the arguments of Chatterton, Johnson &
Campbell (1979) for regarding it as a separate subgenus.

Acanthopyge (Lobopyge) sp.

Fig. 6A-K

Material: At least 20 cranidia, a hypostome, 3 pygidia
and an almost complete dorsal exoskeleton from locality
R25.

Description: A pair of large tubercles is present on the
depressed median part of the Ip lobe but there appears
to be no definite arrangement of tubercles on other parts
of the cranidium. Posterolateral cranidial lobe separated
from median part of lp lobe by broad longitudinal fur¬
row, not separated from remainder of fixigena abaxially,
or else bounded only by a faint vestige of a furrow. Me¬
dian glabellar lobe rises steeply from median part of lp
lobe to reach a maximum height opposite front of
palpebral lobe, subparallel-sided or even narrowing
slightly forwards before expanding around front of
anterolateral cranidial lobe; anterolateral extremities of
median lobe acute and depressed. Pit-like depressions
present where longitudinal furrow meets axial and lp
furrows. Palpebral lobe rising steeply from cheek and
curving outwards distally, much lower than
posterolateral cranidial lobe. Gentle eye ridge runs from
front of palpebral lobe towards junction of axial and
longitudinal furrows.

Anterior margin of hypostome gently convex. Mid¬
dle body 1.6 times as wide as long, lateral margin
notched at outer end of middle furrow; distinctly
swollen maculae situated at lateral extremities of
posterior lobe.

Fig. 6-A-K, Acanthopyge (Lobopyge) sp., from locality R25. A, cranidium NMV P78324; lateral view,
x 5.5. B, E. cranidium NMV P78325; lateral and oblique views, x 7. C, cranidium NMV P78326; dorsal
view, x 5. D, latex cast of external mould of cranidium NMV P78327; dorsal view, x4.5. F, cranidium
NMV P78328; dorsal view, x6. G, pygidium NMV P78329; dorsal view, x3.75. H, hypostome NMV
P78330; ventral view, x6.25. I, cranidium NMV P78331; dorsal view, x5. J, almost complete dorsal
exoskeleton NMV P78332; dorsal view, x3.25. K, pygidium NMV P78333; dorsal view, x4.5. L,
Leonaspis sp., from 300 m east of locality R25. Cranidium NMV P78334; dorsal view, x 3. M-U, Odon-
tochile cf. formosa Gill 1948; M, N, S, from locality R25; O, Q, R, U, from locality R31; P from between
localities R25 and R30; T from 100 m south of locality R31. M, pygidium NMV P78335; dorsal view,
x2.25. N, latex cast of external mould of pygidium NMV P78336; dorsal view, x 3.5. O, latex cast of ex¬
ternal mould of glabella NMV P78337; dorsal view, x2.5. P, incomplete cranidium NMV P78338;
dorsolateral view, x 1.75. Q, U, pygidium NMV P78339; dorsal and lateral views, x0.9. R, cranidium
NMV P78340; dorsal view, x2. S, cephalon with glabella and anterior border broken off to reveal exter¬
nal mould of doublure NMV P78341; dorsal view, x 1.5. T, latex cast of pygidial mould NMV P78342;
ventral view, x 1. Unless otherwise stated, specimens are internal moulds.

EARLY DEVONIAN TRILOBITES

151

152

D. J. HOLLOWAY AND J. V. NEIL

Thorax composed of 11 segments. Axis strongly
rounded (tr.), subparallel sided as far as sixth or seventh
segment and thereafter narrowing gently. Pleurae with a
gently convex (tr.) articulated portion approximately
one-half the width of the axis and with downturned
distal spines curving backwards on more posterior
segments.

Pygidial axis as wide anteriorly as long (sag.), nar¬
rowing only slightly backwards and broadly rounded
posteriorly. Anteriormost axial ring short (sag., exsag.)
and prominent, turned slightly backwards abaxially.
Second ring bears a large tubercle distally and another
closer to midline; between these tubercles the ring is
similar in form to first ring except that it is lower and
turned more strongly backwards; medial part of ring is
very faint. Second ring furrow curves downwards abax¬
ially into a weak apodeme. Third ring vaguely defined,
apparently with a row of tubercles. Posterior pleural
bands on first and second segments more convex than
anterior bands, anterior band on second segment wider
distally than posterior band. First marginal spine diverg¬
ing at about 10° to sagittal axis, second spine directed
straight backwards, third spine poorly preserved but
situated close to sagittal axis. Pleurae steeply inclined
behind axis, with a gently convex postaxial ridge
bounded laterally by a shallow furrow. Doublure nar¬
row, flattened, and bearing concentric terrace lines.

Remarks: Despite the poor preservation of the material
it is possible to distinguish this species from the other
species of A. (Lobopyge) that have been described from
the Early Devonian of southeastern Australia. In A. (L.)
australiformis Chatterton, Johnson & Campbell 1979
and A. (L.) sinuata (Ratte 1886), both from the Garra
Formation near Wellington, New South Wales, the me¬
dian glabellar lobe rises more gently from the lp lobe,
the palpebral lobes are relatively higher, and the pygidial
axis is longer and narrower. In addition, the anterior
margin of the hypostome in A. (L.) australiformis is
more convex and the second pair of marginal spines on
the pygidium curve slightly inwards distally, while A.
(L.) sinuata has a definite pattern of large tubercles on
the cranidium, the anterior pleural bands on the
pygidium are relatively shorter (exsag.) and the anterior-
most marginal spines on the pygidium are more
divergent. A. (L.) australis (McCoy 1876) from the
Humevale Formation near Lilydale, Victoria is in need
of revision, but it has a relatively longer pygidial axis
with a posterior lobe that is distinctly inflated and bears
a number of large tubercles, and the pleural region
behind the second segment is larger (Gill 1939, pi. 5, fig.
1). The fragmentary pygidia of A. (Lobopyge) described
by Chatterton, Johnson & Campbell (1979, pi. 109, figs
22, 23) from the W'arroo Limestone near Yass, New
South Wales are more weakly tuberculate than those
from the Mount Ida Formation and the anterior
marginal spines are more divergent, at least proximally.

Amongst overseas species, the closest resemblance is
with A. (L.) consanguinea (Clarke 1894) from the Early
Devonian of New York, A. (L.) richteri (Vanek 1959)
from the Lochkovian of Czechoslovakia, and A. (L.)

pragensis (Boucek 1933) from the Pridolian or Lochko¬
vian of Czechoslovakia. It is not possible definitely to
distinguish the Mount Ida species from A. (L.) richteri
or A. (L.) pragensis on the basis of published illustra¬
tions, but A. (L.) consanguinea has a distinctive ar¬
rangement of larger tubercles on the cranidium, and the
anterior margin of the hypostome is more arcuate
(Whittington 1956, pi. 131). Chatterton, Johnson &
Campbell (1979) included consanguinea in Acanthopyge
(Acanthopyge), but on the basis of pygidial characters it
clearly belongs to Lobopyge.

Family Odontopleuridae Burmeister 1843
Subfamily Odontopleurinae Burmeister 1843

Genus Leonaspis R. & E. Richter 1917

Type Species (by original designation): Odontopleura
Leonhardi Barrande 1846 from the Kopanina Forma¬
tion (Ludlovian) at Kolednik near Beroun, Czecho¬
slovakia.

Leonaspis sp.

Fig. 6L

Material: A cranidium from approximately 300 m east
of locality R25.

Remarks: The most distinctive features of this
cranidium are a relatively short (sag.) occipital ring with
a large median tubercle or spine situated close to the
posterior margin; a strongly inflated median glabellar
lobe; relatively wide fixigenae; a facial suture that runs
directly forwards in front of the eye before curving
gently inwards; and an ornament of moderately coarse
granules. These features are also characteristic of L. rat -
tei (Etheridge & Mitchell 1896) from the Ludlovian-
Gedinnian of the Yass Basin, New South Wales (see
Chatterton 1971, pi. 22, figs. 8-14) but the present
material is inadequate for meaningful comparison with
that species.

ACKNOWLEDGEMENTS

We thank K. S. W. Campbell (Australian National
University) for encouraging us to work on the fauna and
for assistance in the initial stages; A. Ritchie and R. K.
Jones (Australian Museum) for providing casts of com¬
parative species; and K. N. Bell, A. Jenkin and H. E.
Wilkinson for assistance in collecting the specimens.

REFERENCES

Alberti, G. K. B., 1966. Uber einige neue Trilobiten aus dem
Silurium und Devon, besonders von Marokko. Senck -
enberg. leth. 47: 111-121.

Alberti, G. K. B., 1967. Neue obersilurische sowie unter- und
mitteldevonischer Trilobiten aus Marokko, Deutschland
und einigen anderen europaischen Gebieten. I. Senck -
enberg. leth. 48: 463-478.

Alberti, G. K. B., 1969. Trilobiten des jiingeren Siluriums
sowie des Unter- und Mitteldevons. I. Mit Beitragen zur
Silur-Devon Stratigraphie einiger Gebiete Marokkos und
Oberfrankens. Abh. Senck. Nat. Gesell. 520: 1-692.

EARLY DEVONIAN TRILOBITES

153

Biske, J. S., Gorianov, V. B. & Rzonsnickaja, M. A.,
1977. Tien-shan. In The Silurian-Devonian boundary, Mar-
tinsson, A., Ed., IUGS Series A , 5: 227-237.

Boucek, B., 1933. O novych trilobitech ceskcho gotlandienu,

I. Vest. st. geol. Ust. csl. Repub. 9: 171-179.

Bourque, P-A. & Lesperance, P. .1., 1977. Gaspe Penin¬
sula, Quebec. In The Silurian-Devonian boundary, Mar-
tinsson, A., Ed., IUGS Series A. 5: 245-255.

Campbell, K. S. W., 1975. The functional anatomy of
phacopid trilobites: musculature and eyes. J. Proc. R. Soc.
N.S. W. 108: 168-188.

Campbell, K. S. W., 1977. Trilobites of the Haragan, Bois
d’Arc and Frisco Formations (Early Devonian) Arbucklc
Mountains region, Oklahoma. Bull. Okla. geol. Surv. 123:
1-227.

Chapman, F., 1915. New or little-known fossils in the National
Museum. Part XVIII. Some Yeringian trilobites. Proc. R.
Soc. Viet. 28: 157-171.

Chatterton, B. D. E., 1971. Taxonomy and ontogeny of
Siluro-Devonian trilobites from near Yass, New South
Wales. Palaeontographica (A) 137: 1-108.

Chatterton, B. D. E., Johnson, B. D. & Campbell, K. S.
W., 1979. Silicified Lower Devonian trilobites from New
South Wales. Palaeontology 22: 799-837.

Clarke, J. M., 1913. Fosseis Devonianos do Parana.

Monogrqfias Div. geol. miner. Bras. 1: 1-353 (not seen).
Eldredge, N., 1971. Patterns of cephalic musculature in the
Phacopina (Trilobila) and their phylogenetic significance.

J. Paleont. 45: 52-67.

Eldredge, N., 1973. Systcmatics of Lower and lower Middle
Devonian species of the trilobite Phacops Emmrich in
North America. Bull. Am. Mus. nat. Hist. 151: 285-338.
Etheridge, R. & Mitchell, J., 1895. The Silurian trilobites
of New South Wales, with references to those of other parts
of Australia. Part III. The Phacopidac. Proc. Linn. Soc.
N.S. W. 10: 486-51L

Etheridge, R. J. & Mitchell, J., 1917. The Silurian
trilobites of New South Wales, with references to those of
other parts of Australia. Part IV. The Calymeneidae,
Cheiruridae, Harpeidae, Bronteidae, etc., with an appen¬
dix. Proc. Linn. Soc. N.S. W. 42:480-510.

Foldvary, Ci., 1970. A new species of trilobite, Cheirurus
(Crotalocephalus) regius n. sp. from the Early Devonian of
Trundle district, central N.S.W. J. Proc. R. Soc. N.S. W.
103: 85-86.

Garratt, M. J., 1978. New evidence for a Silurian (Ludlow)
age for the earliest Baragwanathia flora. Alcheringa 2:
217-224.

Gill, E. D., 1939. The Silurian trilobite Lichas australis.

Mem. natn. Mus. Viet. 11: 140-142.

Gill, E. D., 1945. Trilobita of the family Calymenidae from
the Palaeozoic rocks of Victoria. Proc. R. Soc. Viet. 56:
171-186.

Gill, E. D., 1948a. Eldon Group fossils from the Lyell
Highway, Western Tasmania. Rec. Q. Viet. Mus. 2: 57-74.
Gill, E. D., 1948b. A gens of dalmanitid trilobites. J. Proc. R.
Soc. N.S. W. 82: 16-24.

Gill, E. D., 1949. Palaeozoology and taxonomy of some
Australian homalonotid trilobites. Proc. R. Soc. Viet. 61:
61-73.

Gill, E. D., 1951. Two new brachiopod genera from Devonian
rocks in Victoria. Mem. natn. Mus. Viet. 17: 187-205.
Haas, W., 1968. Trilobiten aus dem Silur und Devon von
Bithynien (NW-Turkei). Palaeontographica (A) 130:

60-207.

Haas, W., 1969. Lower Devonian trilobites from central
Nevada and northern Mexico. J. Paleont. 43: 641-659.

Hall, J. & Clarke, J. M., 1888. Natural History of New
York. Palaeontology: Volume VII. Text and Plates. Con¬
taining descriptions of the trilobites and other Crustacea of
the Oriskany, Upper Helderberg, Hamilton, Portage,
Chemung and Catskill Groups. 236 pp., 46 pis. Geological
Survey of New York, Albany.

Holloway, D. J., 1980. Middle Silurian trilobites from
Arkansas and Oklahoma, U.S.A. Part 1. Palaeon¬
tographica (A) 170: 1-85.

Jaeger, H., 1966. Two late Monograptus species from Vic¬
toria, Australia, and their significance for dating the
Baragwanathia flora. Proc. R. Soc. Viet. 79: 393-413.

Kobayashi, T. & Hamada, T., 1977. Devonian trilobites of
Japan in comparison with Asian, Pacific and other faunas.
Spec. Pap. palaeont. Soc. Japan 20: i-vii, 1-202.

Kozlowski, R., 1923. Faune devonienne de Bolivie. Annls
Paltont. 12: 1-112.

Landrum, R. E. & Sherwin, L., 1976. Warburgella from
central New South Wales. Rec. geol. Surv. N.S. W. 17:
135-146.

Lane, P. D., 1971. British Cheiruridae (Trilobita). Palaeon-
togr. Soc. [Monogr.J: 1-95.

Link, A. G., 1970. Age and correlations of the Siluro-
Devonian strata in the Yass Basin, New South Wales. J.
geol. Soc. Aust. 16: 711-722.

Link, A. G. & Druce, E. C., 1972. Ludlovian and Gedin-
nian conodont stratigraphy of the Yass Basin, New South
Wales. Bull. Bur. Miner. Resour. Geol. Geophys. Aust.
134: 1-136.

Lutke, F., 1965. Zur Kenntnis herzynischer Trilobiten aus
dem Unter- und Mitteldevon des Harzes. Palaeon¬
tographica {A) 124: 151-236.

Maksimova, Z. A., 1968. Srednepaleozojskie trilobity tsentral
’nogo Kazakhstana. Trudy vses. nauchno-isled. geol. Inst.
165: 1-208.

Maksimova, Z. A., 1975. Trilobity. In Kharaktcristika fauny
pogranichnykh sloev Silura i Dcvona tsentral ’nogo
Kazakhstana, Menner, V. V., Ed., Material'y po geologii
tsentral 'nogo Kazakhstana 12: 119-133.

McKellar, R. G., 1969. Brachiopods and trilobites from
Siluro-Devonian strata in the Rockhampton district. Pubis
geol. Surv. Qd 337: 1-13.

McLearn, F. H., 1924. Palaeontology of the Silurian rocks
of Arisaig, Nova Scotia. Mem. geol. Surv. Can. 137: 1-179.

Moore, R. C. 1959. (Ed.) Treatise on Invertebrate Paleonto¬
logy, Part O, Arthropoda I. xix + 560 pp. Geological So¬
ciety of America and University of Kansas Press,
Lawrence.

Mozdalevskaya, T. L., 1968. Atlas of Silurian and Early
Devonian fauna of Podolia. (Appendix to the guide.)
VSEGEI, Leningrad.

Nikiforova, O. I., 1977. Podolia. In The Silurian-Devonian
boundary, Martinsson, A., Ed., IUGS Series A, 5: 52-64.

Owens, R. M., 1973. British Ordovician and Silurian Proe-
tidae (Trilobita). Palaeontogr. Soc. / Monogr.J : 1-98.

Philip, G. M., 1962. The palaeontology and stratigraphy of
the Siluro-Devonian sediments of the Tyers area, Gipps-
land, Victoria. Proc. R. Soc. Viet. 75: 123-246.

Philip, G. M., 1967. Late Silurian-early Devonian relation¬
ships in the central Victorian and western Tasmanian clastic
sequences, Australia. In International Symposium on the
Devonian System, Calgary, Oswald, D. H. Ed., 1967.
Volume 2: 913-920. Alberta Society of Petroleum
Geologists, Calgary.

Pillet, J., 1973. Les trilobites du Devonien inferieur et du
Devonien moyen du Sud-Est du Massif armoricain. M£m.
Soc. scient. Anjou 1: 1-307.

D

154

D. J. HOLLOWAY AND J. V. NEIL

Prantl, F., 1947. Some abnormalities in Crotalocephalus
Salter (Trilobitae). Bull. int. Acad, tcheque Sci. 48: 1-16.

Pribyl, A., 1965. Proetidm trilobiti z novych sberu v ceskem
siluru a devonu. (Proetidae aus neueren Aufsammlungen
im bohmischen Silur und Devon (Trilobitae) I). Cas. ndrod.
Muz. 134: 91-98.

Pribyl, A. & Vanek, J., 1964. Nekolik pozndmek ke
klasifikaci rodu Cheirurus Beyrich (Trilobita). Cas. ndrod.
Muz . 133: 93-95.

Pribyl, A. & Vanek, J., 1981. Studie zur Morphologic und
Phylogenie der Familie Otarionidae R. and E. Richter, 1926
(Trilobita). Palaeontographica (A). 173: 160-208.

Richter, R. & E., 1926. Die Trilobiten des Oberdevons.
Beitrage zur Kenntnis devonischer Trilobiten. IV. Abh.
preuss. geol. Landesanst. 99: 1-301.

Richter, R. & E., 1954. Die Trilobiten dcs Ebbe-Sattels und
zu vergleichende Arten. (Ordovizium,
Gotlandium/Devon). Abh. Senck . Nat. Gesell. 488: 1-76.

Salter, J. W., 1865. A monograph of the British trilobites
from the Cambrian, Silurian and Devonian formations.
Palaeontogr. Soc. fMonogr.J (2): 81-128.

Saul, J. M., 1967. Burmeisteria (Digonus) accraensis , a new
homalonotid trilobite from the Devonian of Ghana. J.
Paleont. 41: 1126-1136.

Savage, N. W., 1974. The brachiopods of the Lower Devonian
Maradana Shale, New South Wales. Palaeontographica (A)
146: 1-51.

Shergold, J. H., 1968. On the occurrence of the trilobite
genera A caste & Acastella in Victoria. Proc. R. Soc. Viet.
81: 19-30.

Siierwin, L., 1971. Trilobites of the subfamily Phacopinae
from New- South Wales. Rec. geol. Surv. N.S. W. 13: 83-99.

Strusz, D. L., 1964. Devonian trilobites from the Wellington-
Molong district of New' South Wales. J. Proc. R. Soc.
N.S.W.. 97: 91-97.

Strusz, D. L., 1980. The Encrinuridae and related trilobite
families, with a description of Silurian species from
southeastern Australia. Palaeontographica (A) 168: 1-68.

Strusz, D. L., Bischoff, G., Cooper, B. J., Gill, E. D. &
Talent, J. A., 1972. Central and eastern Victoria. In Cor¬
relation of the Lower Devonian rocks of Australasia,
Strusz, D. L., J. geol. Soc. Aust. 18: 427-455.

Talent, J. A., 1965a. The stratigraphic and diastrophic evolu¬
tion of central and eastern Victoria in middle Palaeozoic
times. Proc. R. Soc. Viet. 79: 179-195.

Talent, J. A., 1965b. The Silurian and Early Devonian faunas
of the Heathcote district, Victoria. Mem. geol. Surv. Via.
26: 1-55.

Talent, J. A., Berry, W. B. N. & Boucot, A. J., 1975.
Correlation of the Silurian rocks of Australia, New Zealand
and New' Guinea. Spec. Pap. geol. Soc. Anter. 150: 1-108.

Thomas, D. E., 1937. Some notes on the Silurian rocks of the
Heathcote district. Min. geol . J. 1: 64-67.

Thomas, D. E., 1940a. Heathcote sheet, 1:31,680 geological
parish plans. Department of Mines, Victoria.

Thomas, D. E., 1940b. Redcastle sheet, 1:31,680 geological
parish plans. Department of Mines, Victoria.

Thomas, D. E., 1941. Dargile sheet, 1:31,680 geological parish
plans. Department of Mines, Victoria.

Tomczykowa, E., 1975. The trilobite subfamily Homa-
lonotinae from the Upper Silurian and Lower Devonian of
Poland. Acta palaeont. pol. 20: 3-46.

VandenBerg, A. H. M., 1974 (compiler). Melbourne Sheet SJ
55-1, Australian 1:63,360 Geological Series. Geological
Survey of Victoria.

VandenBerg, A. H. M. & Garratt, M. J., 1976.
Melbourne Trough. Spec. Pub!, geol. Soc. Aust. 5: 45-62.

Vanek, J., 1959. Celed Lichaidae Hawle & Corda, 1847 ze
stredoceskeho slarsiho paleozoika. Bohan, cent. 1: 81-168.

Whittington, H. B., 1956. Beecher’s lichid protaspis and
Acanthopvge consanguinea (Trilobita). J. Paleont. 30:
1200-1204.

VVolfart, R., 1968. Die Trilobiten aus dem Devon Boliviens
und ihre Bedeutung fur Stratigraphie und Tiergeographie.
In Beitrage zur Kenntnis des Devons von Bolivien, Wolfart,
R. & Voges, A., Beih. geol. Jb. 74: 5-201.

PROC. R. SOC. VICT. vol. 94, no. 3, 155-167, September 1982

MORPHOLOGICAL AND GEOGRAPHICAL DISJUNCTIONS
IN FORMS OF EUCALYPTUS NITIDA Hook. f.
(MYRTACEAE): WITH SPECIAL REFERENCE TO THE
EVOLUTIONARY SIGNIFICANCE OF BASS STRAIT,
SOUTHEASTERN AUSTRALIA

By Julie C. Marginson and Pauline Y. Ladiges
School of Botany, University of Melbourne, Parkville, Victoria 3052

Abstract: Variation in seedling characters of Eucalyptus nitida (shining peppermint) and related
species was analysed phenetically using multivariate classification and ordination techniques. The prin¬
cipal aims were to clarify the identity of the peppermint species occurring on some Bass Strait islands and
to examine patterns of variation in seedling morphology across the geographic range of populations cur¬
rently referred to E. nitida. Related species were included as aids to interpreting patterns. Analyses were
based on seedling and juvenile foliage, the latter resolving the taxa more completely. On the basis of
seedling morphology, Bass Strait island peppermint appears to be conspccific with Tasmanian E. nitida ,
and these populations appear to be separate from Victorian and South Australian populations currently
referred to E. nitida. It is suggested that the degree of distinctness is sufficient to warrant separate tax¬
onomic recognition of the Australian mainland populations.

Systematic affinities of Tasmanian peppermints are discussed, and an explanatory hypothesis is pro¬
posed to account for the present distribution of peppermint species and the high degree of endemism
among Tasmanian peppermints.

Current usage recognizes six species of Eucalyptus as
occurring variously on the larger islands of Bass Strait
(Table 1). This contrasts with about 75 species in Vic¬
toria and 25 on the Tasmanian mainland. Five of the six
species belong to the informal subgenus Symphyomyr-
tus as recognized by, Pryor and Johnson (1971) and are
relatively widespread on both sides of Bass Strait. They
are E. viminalis Labill., E. dairympleana Maiden, E.
globulus Labill., E. ovata Labill. and E. brookerana A.
M. Gray. Apart from the occurrence of E. globulus on
Rodondo Island just south of Wilsons Promontory
(Kirkpatrick et al. 1974), these species are confined to
the seven largest islands.

The sixth species, E. nitida Hook, f., is a member of
the peppermint group of eucalypts and is placed in the
informal subgenus Monocalyptus (Pryor & Johnson
1971). It is considered to occur not only on the Tas¬
manian mainland but also in far southeastern South
Australia, in scattered localities close to the coast of Vic¬
toria from the South Australian border to Sperm Whale
Head, and in the Grampians (Willis 1970) (Fig. 1). In re¬
cent years, the name has been applied to populations of
Bass Strait island peppermint (e.g. Curtis & Morris
1975). These populations occur on all the larger islands,
on some smaller islands of the Furneaux Group and on
two islands of the Kent Group, Deal and Dover Islands.

Willis (1972, p. 416) described all the above popula¬
tions as “undoubtedly conspecific”. Previously, Vic¬
torian and South Australian populations were known
variously as E. dives {e.g. Ewart 1930, Hall et al. 1963),
E. vitrea (e.g. Black 1964, Boomsma 1972, Parsons et
al. 1972) and E. radiata (e.g. Parsons 1966). Similarly,
island populations had been given names of Tasmanian
mainland species such as E. tenuiramis (e.g. Brett 1938),
E. simmondsii (e.g. Jackson 1965, Hall et al. 1970,
Hope 1973) and E. amygdalina (Hope 1973).

Significantly, Blakely (1934, 1965) applied the name E.
tenuiramis Miq. to specimens collected from Deal and
Flinders Islands which he had sighted at the National
Herbarium of New South Wales. Willis (1967) strongly
disagreed with the application of this name, mainly on
the ground that the glaucousness of leaves and buds
described by Miquel has not been found to occur in any
Bass Strait island specimens. The name E. tenuiramis
has since been applied strictly to lowland populations
from southeastern Tasmania having connate juvenile
leaves and a dense waxy bloom on shoots, buds and
fruits. Subsequently, Willis (1970) reduced E. simmond¬
sii of northwestern Tasmania to synonymy with E.
nitida and made his claim for conspecificity, but without
offering any reasons for his decisions.

With the other five island species occurring on both
sides of Bass Strait, the notion of a comparable distribu¬
tion for E. nitida is plausible. However, there are two
prima facie reasons why Willis’s claim should not be ac¬
cepted uncritically: firstly, of the four systematic groups
of eucalypts that occur in Tasmania, the peppermints
show' the highest degree of endemism (Table 2); second¬
ly, variation within E. nitida sensu Willis is con¬
siderable, particularly with respect to habit and bark
type, fruit size and shape, and juvenile leaf and stem
morphology.

Bass Strait island peppermint is generally a small
tree, dominant in low open-forest communities; it is 2 to
10 m tall, of poor form, sometimes multi-stemmed, and
has smooth bark with a short, scaly to sub-fibrous butt;
the more or less flat-topped turbinate to pyriform fruits
are relatively large (to 9 mm wide); coppice leaves are
ovate, sessile and opposite for many pairs. Herbarium
specimens conform reasonably well with the type
specimens of E. nitida which were collected from several
Tasmanian mainland localities. The nearest Victorian

155

156

J. C. MARGINSON AND P. Y. LADIGES

Fig. 1—Generalised distribution limits of £. nitida sensu Willis and localities of seed collections:
E. nitida (•), other species (*). Abbreviations: am = E. amygdalina , co = E. coccifera , di = E. dives , el = E.
elata , n = E. nitida , pu = £. pulchella , ra = £. radiata, ri = E. risdonii , te = £. tenuiramis.

populations of peppermint currently referred to £.
nitida occur on Wilsons Promontory, but they dilTer
from island peppermint in several respects: bark is
rougher, buds and fruits are smaller, and coppice leaves
are lanceolate.

In some parts of its range, Tasmanian E. nitida is not
separable from E. amygdalina (Hall el al. 1963, A. Gray
pers. comm.). Some Victorian and South Australian
populations include trees of intermediate status and
bearing some resemblance to £. radiata (Parsons et al.
1977) or E. pauciftora (Willis 1970, Boomsma & Lewis
1980). Throughout the geographic range, there appear
to be few readily interpretable disjunctions in adult form
that might clarify the taxonomy.

The diagnostic importance of juvenile leaves has
been recognized for almost a century (Maiden 1922).
More recently, Brooker (1979) observed that seedling
morphology can be an extremely valuable aid in assign¬
ing species to series within the genus Eucalyptus.
Ladiges et al. (1981) found that differences in seedling
morphology between E. ovata and E. brookerana were
sufficient to justify recognition of these taxa at the
specific level. It was therefore considered desirable to

use seedling characters to test the adequacy of E. nitida
sensu Willis, principally with respect to Bass Strait
island populations. A seedling trial was established in a
uniform environment to examine the patterns of varia¬
tion in seedling morphology within £. nitida sensu
Willis, and to assess the distinctness of E. nitida seed¬
lings with respect to those of related species.

Related species were considered to be all the
members of subseries Amygdalininae of series Piperitae
(Pryor & Johnson 1971). Details of their geographic and
altitudinal distributions, growth forms and bark types
are summarized in Table 3. Typically, they occur on in¬
fertile, acidic soils derived from sedimentary, metamor-
phic or granitic parent materials; average annual rainfall
ranges from 600 to 1300 mm. The member species are
distinguished from other members of the subgenus
Monocalyptus on the basis of possessing some combina¬
tion of the following characters (modified from Wilcox
1979):

1. “Peppermint” bark —finely textured, shortly
fibrous, loosely coherent rough bark which per¬
sists on larger branches.

EUCALYPTUS NITIDA AND BASS STRAIT

157

Table 1

Occurrence of Eucalyptus Species on the Islands of
Bass Strait

Island

Area

Species

(km 2 3 4 5 6 )

vim dal

glo ova bro

nit

King Island

1,100

+ -

+ - +

__ *

Hunter Group

Hunter Island
Three Hummock

70

+

+

Island

68

+

+ ?

Robbins Island

96

+

+

Furneaux Group

Flinders Island
Cape Barren

1,333

+ +

+ + -

+

Island

450

+

+ +

+

Clarke Island

75

+

+ +

+

Prime Seal Island

10

+

Babel Island

4

+

Great Dog Island

3

+

Kent Group

Deal Island

15

+

Dover Island

2.5

+

Erith Island

3.0

-

Rodondo Island

0.8

+

vim = £. viminalis , dal = £. dairympleana, glo = £. globulus ,
ova = £. ovata , bro = £. brookerana , nil = £. nitida.

+ = present, - = notably absent.

* The map provided by Jackson (1965) showing £. nitida (syn.
£. simmondsii ) as occurring on King Island is considered to
be incorrect (P. Barnett, King Island, pers. comm.)

2. Seedling leaves opposite and sessile for first 4 to 6
nodes; juvenile leaves opposite and sessile for an
indefinite number of following nodes.

3. Verrucae (protruding oil glands) on margins and
midribs of seedling and juvenile leaves, and on
seedling stems.

4. Characteristic essential oils in juvenile and mature
foliage, particularly piperitonc, phellandrene and
cineolc.

5. Mature leaves symmetrical, with more or less
longitudinal secondary venation, and with the
apex acuminate or hooked.

6. Buds numerous, clavate and pedicellate; capsules
pyriform-turbinate with a flattened disc.

The following taxa were omitted from the study: (1) E.
radiata ssp. robertsonii sensu Johnson & Blaxell (1973)
occurring in upland areas of New South Wales and Vic¬
toria; (2) E. robertsonii sensu Jackson (1965) and Hall et
al. (1970) considered by Johnson & Blaxell (1973) to be
an undescribed subspecies of E. amygdalina ; and (3) the
two species that form subseries Piperitinae of series
Piperitae (Pryor & Johnson 1971) considered by
Brooker (1977) to be better classified with the ash group
of eucalypts (series Obliquae subgenus Monocalyptus).

METHODS

Shed Collection and Seedling Growth

Of the nine species included in the seedling trial,
eight were represented by seed collections from in¬
dividual mother trees. E. nitida sensu Willis was
represented by seed collected from five trees on Dover
Island, six on Flinders Island, four in southern
Tasmania, two each on Wilsons Promontory and in the
Grampians, and one each in South Gippsland at Bellbird
Swamp near Longford and in southeastern South
Australia near Wandillo (Fig. 1). £. tenuiramis and E.
radiata were represented by seed from two trees each; E.
dives , E. pulchella, E. amygdalina and £. risdonii by
seed from one tree each; all these trees were considered
to be typical members of each taxon and accordingly
were located in southeastern Tasmania or central Vic¬
toria. The subalpine Tasmanian species E. coccifera was
represented by seed from one tree considered to be
typical and from one tree representing a population con¬
sidered to be anomalous on the basis of both adult mor¬
phology and its lowland locality (Devil’s Kitchen,
Tasman Peninsula). The ninth species, E. elata , was
represented by seed obtained through the Forests Com¬
mission of Victoria from the vicinity of Nerriga, N.S.W.
It is not known whether the seed was from a single tree
or was pooled from several trees. Voucher specimens
from the parent trees of all species except £. elata are
held in herbaria: MELU or FR1.

Seed was germinated on moist filter paper in petri
dishes. Young seedlings were pricked out into plastic
buckets containing a general purpose potting mix and
were allowed to grow in a heated glasshouse for seven
months. Initially, each bucket contained six seedlings
from one seed collection; after three months, these were
thinned out to the most robust three per bucket. This
growth system was chosen to suit the scale of the trial
while providing sufficient replication and a reasonable
volume of soil for individual seedlings. The trial was ter¬
minated when competitive effects on growth rate (ex¬
pressed as shoot height) became apparent in some
buckets, causing a reduction in the number of assessable
seedlings.

Phenetic Analysis of Seedling Characters

Individual seedlings were subject to multivariate
analysis of morphological characters. Following the
distinction made between seedling and juvenile leaves by
Blake (1953) and Chippendale (1973), two data matrices
were prepared. The first was based on seedlings after
four months’ growth; seedling leaf characters were
assessed on one of the paired leaves at the third node
above the cotyledonary node and the matrix incor¬
porated data from 96 seedlings using 35 characters
(Table 4). The second matrix was based on seedlings
after seven months’ growth. Juvenile leaf characters
were assessed on one of the paired leaves al the seventh
node, or at the next highest node at which a clear change
to juvenile foliage had occurred. The matrix incor¬
porated data from 85 seedlings using 36 characters.
Changes in the character set used for the second matrix
were due to changes in seedling morphology with age.

158

J. C. MARGINSON AND P. Y. LADIGES

Table 2

Tasmanian Species of Eucalyptus. Geographic and Systematic Distribution
(compiled primarily from Pryor and Johnson 1971)

Species

Tas.

S.A.

E. delegatensis

+

E. obliqua

+

+

E. paucijlora

+

+

E. regnans

+

E. sieberi

+

E. amygdalina

+

E. coccifera

+

E. nitida

+

+ ?

E. pulchella

+

E. risdonii

+

E. tenuiramis

+

E. barberi *

+

E. brookerana *

+

E. ovata

+

+

E. rod way i

+'

E. cordata

+

E. dalrympleana

+

E. globulus

+

E. gunnii

+

E. morrisbyi

+

E. perriniana

+

E. rubida

+

+

E. urnigera

+

E. vernicosa

+

E. viminalis

+

4-

Vic.

N.S.W.

Qld.

Systematic Group

+

+

Subg. Monocalyptus

+

+

+

Sect. Renantheria

+

+

+

Series Obliquae

+

+

Subg. Monocalyptus

Sect. Renantheria

+ ?

Series Piperitae

Subseries Amygdalininae

Subg. Symphyomyrtus

+

Sect. Maidenaria

+

+

+

Series Ovatae

Subseries Ovatinae

Subg. Symphyomyrtus

+

+

Sect. Maidenaria

+

+

Series Viminales

+

+

+

+

+

+

+

* Tentative placement of E. barberi Johnson and Blaxell and E. brookerana A. M. Gray after Ladiges et al. (1981).

Each raw data matrix was analysed separately using
classification and ordination programs from CSIRO’s
TAXON P2 package as follows: data were input to the
polythetic program MSEUC which produces an inter-
element dissimilarity matrix from mixed data using a
squared Euclidean metric with Burr’s standardization
(see Williams 1976). The dissimilarity matrix was then
input to SAHN which generates a hierarchy based on
“distances” between individuals (seedlings). Output
from this agglomerative procedure was restricted to the
last 14 fusions (15 groups). Group data were then input
to the diagnostic programs GPCOM and CRAMER to
assist in interpreting relationships between groups and
the characters that contributed to their formation. An
inter-group dissimilarity matrix was generated using
GPDIS and input to the ordination program PCOA
which finds principal coordinates for group centroids.
The inter-group dissimilarity matrix was also input to
MST which calculates a minimum spanning tree linking
“nearest” group centroids (see Gillison 1978). Groups
were also redefined using GPRES and input to BACRIV
which calculates correlation coefficients between
characters and the vectors (axes) generated by PCOA.
Output from both MST and BACRIV assists in inter¬
preting the ordinations. Further details of these methods
can be obtained from CSIRO Division of Computing
Research.

RESULTS

The limited size of the seedling trial, especially with
respect to species represented by progeny from only one
or two trees, necessitates caution in interpreting the
analyses. Only the broader elements of the patterns
found will be discussed, and only then in relation to in¬
terpreting variation within E. nitida sensu Willis.

Seedling Leaf Analysis

The classification based on seedling leaves produced
a major dichotomy separating Victorian and South
Australian seedlings of E. nitida from Bass Strait island
seedlings (Fig. 2). Tasmanian mainland seedlings oc¬
curred equally on both sides of the dichotomy, reflecting
the extent of their variability. Moreover, classification to
the 15-group level did not produce a clear separation of
some E. nitida seedlings from those of three other
species: Group 2 incorporated seedlings of E. dives with
E. nitida from the Grampians and South Gippsland;
Groups 3 and 4 incorporated E . radiata with E. nitida
from Wilsons Promontory and South Gippsland; and
Groups 6 and 7 incorporated E . coccifera with E. nitida
from Flinders Island and the Tasmanian mainland. Ten
of the 15 groups included some representatives of E.
nitida.

The characters contributing most to the major
dichotomy in the classification were those related to leaf

EUCALYPTUS NITIDA AND BASS STRAIT

159

Table 3

Tasmanian Peppermints and their Allies (Subseries Amygdalininae)
(Compiled primarily from Hall et at. (1970), Willis (1972), Curtis and Morris (1975))

Species

Geographic distribution

Altitudinal range

Typical habit

Bark type

E. coccifera

Tas. Central Highlands.
Sub-alpine.

600-1500 m

Shrub to small tree

1 -6(-12) m

Smooth

E. risdonii

Tas. Restricted to a small
area near Hobart.

Near sea level to

150 m

Shrub to small tree
3-8(-20) m

Smooth with circular scars

E. tenuiramis

Tas. South-eastern quarter.

Near sea level to

450 m

Tree 10-30 m

Smooth with circular scars

E. pulchella

Tas. South-eastern quarter.

150-550 m

Tree 10-30 m

Smooth with ±subfibrous
butt

E. amygdalina

Tas. Eastern half.

Near sea level to

850 m

Tree 15-35 m

Peppermint — variable

E. nitida

Tas. Western half; Bass
Strait islands.

Vic. Patchy, southern
coast, Grampians S.A.
Extreme south-east.

Near sea level to

850 m

Small to medium
tree

(5-)10-20(-35) m

Variable. Typically
sub-fibrous on trunk, but
may be smooth-barked

E. radiata

Vic. Widespread in

Central Highlands and
foothills.

N.S.W. Eastern third.

150-1200 m

Tree 15-35 m

Peppermint

E. dives

As for E. radiata , but
on drier sites.

150-120 m

Tree 10-30 m

Peppermint to larger
branches

E. elata

Vic. Far east coast,
Macalister R. valley.

N.S.W. South east coast.

All alluvial sites.

Near sea level to
200(-750) m

Tree

20-35(-50) m

Smooth with sub-fibrous
butt

shape (principally the ratio of length to maximum
width) plus the occurrence of verrucae on leaf margins
(Table 5). Another leaf shape ratio (LM:LL), internode
length, stem glaucousness and leaf colour were import¬
ant in producing the next two dichotomies. Taking only
those ten groups that incorporated E. nitida , the four
mainland groups (1-4) included seedlings that had green,
ovate to lanceolate leaves with acute apices; internodes
were relatively long; verrucae were absent from the leaf
margins. The six groups that incorporated island and
Tasmanian E. nitida (5-10) included seedlings with green
to blue-green, broadly ovate leaves; apices were more or
less obtuse; internodes were relatively short; and ver¬
rucae were conspicuous on the margins of all seedlings
in these groups. The last-mentioned character clearly
separated all Tasmanian and island seedlings of E. nitida
from all Australian mainland seedlings, including those
from Wilsons Promontory. Together with the presence
of more or less undulate margins, these glands were
most prominent in progeny of trees from Dover Island
in the Kent Group.

Ordination of the group centroids together with links
between nearest neighbours are shown in Fig. 3. The
first three axes (vectors) accounted for 64.4% of the
total variation in the data matrix. The relative positions
of group centroids confirmed the pattern of clear separa¬
tion between Victorian and South Australian seedlings
of E. nitida (Groups 1-4), and Bass Strait island and

some Tasmanian mainland seedlings (Groups 5-9). The
remaining Tasmanian mainland seedlings (Group 10)
occupied an intermediate position with respect to the
characters separating the groups on axis 1. These in¬
cluded the same characters that produced the major
dichotomy in the classification. Axis 2 was most strongly
correlated with internode length and total height and the
extreme position of Group 10 suggests that some Tas¬
manian mainland populations may have an inherently
slow growth rate relative to other populations of E.
nitida , at least during early growth.

Juvenile Leaf Analysis

The classification based on juvenile leaves again pro¬
duced a major dichotomy separating Victorian and
South Australian samples of E. nitida from both Bass
Strait island and Tasmanian mainland populations (Fig.
4). All seedlings of E. nitida were contained within seven
groups and were clearly differentiated from all other
species included in the trial. To that extent, the analysis
based on juvenile leaf characters produced a more
readily interpretable result.

The characters contributing most to the major
dichotomy in the classification were again principally
related to leaf shape; in addition, secondary vein angle
and oil gland density were important (Table 6). Leaf col¬
our and the ratio of length from leaf base to widest point
over total length were important in producing the next
two dichotomies respectively. Taking only those seven

160

J. C. MARGINSON AND P. Y. LADIGES

Table 4

Seedling Characters Scored for Data Matrices

Binary

1. Leaf margin (5).revolute/not

2. Leaf margin.undulate/not

3. Verrucae on stem.present/absent

4. Verrucae on abaxial midrib.present/absent

5. Verrucae on leaf margin.present/absent

6. Stem.glaucous/not

7. Lignotuber.present/absent

8. Petiole (J) .present/absent

9. Leaf lamina (J) .concolorous/discolorous

Multistate

10. Leaf base shape.acutc/obtuse/cordate/connate

11. Leaf angle to stem.<80°/80-100°/> 100°

12. Leaf margin .. .entire/repand/sparsely serrulate/scrrulate

13. Leaf colour .... yellow-green/green/blue-grecn/glaucous

14. Midrib colour.yellow/pale red/intense red

15. Margin colour.yellow/pale red/intense red

16. Stem colour.yellow/pale red/intense red

17. Anthocyanin on abaxial leaf surface (S).absent/pale

/intense

18. Leaf orientation (J) .horizontal/twisted/vertical

Numerical

19. Leaf length (- LL) (mm)

20. Maximum leaf width ( = MW) (mm)

21. Width 5 mm from leaf base ( = WB) (mm)

22. Width 5 mm from leaf apex ( = WAS) (mm)

23. Width 10 mm from leaf apex (- WA10) (mm)

24. Leaf length from base to widest point ( = LM) (mm)

25. Total height ( = TH) (mm)

26. Length 3rd or 7th internode (= IL) (mm)

27. Mean internode length ( = ML) (mm)

28. Leaf apex angle (°)

29. Secondary vein angle (°)

30. Total number of leaf pairs (J)

31. Number of leaf pairs opposite

32. Leaf thickness (mm x 10 2 )

33. Oil gland density (per cm 2 )

34. Ratio LL:MW

35. Ratio LM:LL

36. Ratio WB:MW

37. Ratio WA5:WA10

38. Ratio height to 3rd node:TH (5)

39. Ratio IL:ML

S = scored only for the matrix based on seedling leaves at
third node.

J= scored only for the matrix based on juvenile leaves at
seventh or higher node.

groups that incorporated E. nitida, seedlings in the two
Australian mainland groups had predominantly green
linear to lanceolate leaves with extremely acute apices;
secondary vein angles approached semi-longitudinal; oil
gland density was high, especially in Group 2 from
Wilsons Promontory and South Gippsland. Seedlings in
the five island and Tasmanian mainland groups had
more or less elliptical, green to blue-green leaves with
correspondingly higher apex and secondary vein angles;
oil gland density was relatively low to zero. Interest¬
ingly, the trend in oil gland density appears to be the

reverse of that reported by Ladiges et al. (1981) for E.
ovata and E. brookerana ; that is, within E. nitida sensu
Willis, values were higher at lower latitudes and low at
higher latitudes.

Ordination of the group centroids together with links
between nearest neighbours are shown in Fig. 5. The
first three axes (vectors) accounted for 66.6% of the
total variation. The relative positions of the group cen¬
troids again confirms the pattern of clear separation be¬
tween Groups 1 and 2 incorporating seedlings from the
Australian mainland, and Groups 3 to 7 incorporating
all island and Tasmanian seedlings. As indicated by both
the classification and the nearest neighbour linkages,
Group 2, incorporating E. nitida from Wilsons Prom¬
ontory and South Gippsland, is more similar to Group
14 incorporating E. radiafa than it is to Group 1 incor¬
porating E. nitida from the Grampians and South
Australia. This is principally due to differences ex¬
pressed on axis 3, notably in oil gland density. Axis 1
was correlated most strongly with a number of
characters relating to leaf shape; axis 2 was correlated
positively with internode length and negatively with
total number of leaf pairs.

Comparison of the two analyses indicates that, at
least on the basis of seedling morphology, Bass Strait
island peppermint populations are phenetically most
closely related to the Tasmanian species E. nitida sensu
stricto. The patterns of variation in seedling mor¬
phology suggest that island and Tasmanian mainland
populations are distinguishable from Australian
mainland populations currently referred to E. nitida ,
principally on the basis of (1) leaf shape and presence or
absence of verrucae on seedling leaf margins apparent in
seedling 3 to 4 months old and (2) leaf shape and oil
gland density apparent in juvenile leaves of seedlings 6
to 7 months old.

DISCUSSION

Taxonomic Implications

Since E. nitida is the only peppermint species con¬
sidered to occur on both sides of Bass Strait at present,
the disjunctive variation pattern within that species
means that all Australian mainland peppermint popula¬
tions can be distinguished from all Tasmanian and
island populations, regardless of any taxonomic
difficulties that may occur within those regions. The in¬
termediate geographic position of Wilsons Promontory
might reasonably lead to the expectation that its pepper¬
mint populations have some intermediate taxonomic
status. However, results of the present study plus
preliminary results of a study of adult morphology sug¬
gest that the resemblance with E. radial a is paramount.

The discontinuous pattern of variation within £.
nitida sensu Willis suggests that consideration should be
given to recognizing new taxa. In particular, it would
appear that there are grounds for recognizing popula¬
tions of peppermint occurring in western Victoria in¬
cluding the Grampians and in southeastern South
Australia at the species level. As indicated by Pryor and

EUCALYPTUS NITIDA AND BASS STRAIT

161

Table 5

Seedling Leaf Analysis

Means and frequencies (%) for the Seven Characters that Contributed Most to the First Three Dichotomies in the
Hierarchical Classification (a) and in the Ten Groups Incorporating Seedlings of E. Nitida sensu Willis (b)
Group numbers and combinations are those indicated in Fig. 2. Numbers of seedlings are shown in brackets.

(a) (b)

Characters Group Combinations Groups incorporating E. nitida sensu Willis

1+2+3+

5+6+7+

Vic., S.A

. groups

Bass Strait island, Tas. groups

4+10+11

8 + 9+13

1st dichotomy

+ 12

+ 14+15

1

2

3

4

5

6

7

8

9

10

(39)

(57)

(6)

(7)

(6)

(8)

(21)

(5)

(5)

(8)

(6)

(6)

Seedling leaf apex angle (°)

60

85

65

73

52

53

83

84

96

84

77

67

Ratio seedling leaf length:

maximum width

2.9

1.8

2.9

2.0

2.7

2.9

1.9

1.6

1.5

1.7

1.9

2.7

Verrucae on seedling leaf

margin

28%

95%

0

0

17%*

12%*

o

o

#

100%

100%

100%

100%

100%

2nd dichotomy

1+2 + 3 + 4 10+11 + 12

(27)

(12)

Ratio seedling leaf length

from base to widest:total

length

0.25

0.48

0.37

0.26

0.18

0.18

0.34

0.29

0.28

0.30

0.25

0.45

Length 3rd internode (mm)

39.7

18.6

47.3

46.6

35.7

31.0

33.5

26.2

32.0

27.6

20.7

13.0

3rd dichotomy 5 + 6 + 7+ 14+15

8 + 9+ 13

(48) (9)

Stem glaucousness

Seedling leaf colour —

0

100%

0

0

0

0

0

0

0

0

0

0

green

•65%

0

100%

100%

100%

100%

71%

20%

80%

100%

0

0

blue-green

35%

33%

0

0

0

0

29%

80%

20%

0

100%

100%

glaucous

0

67%

0

0

0

0

0

0

0

0

0

0

* contributed by E. radiata only.

Johnson (1971), the name “E. vitrea ” may not be
available for these populations because the type
specimen is believed to be of hybrid origin, the putative
parental species being E. radiata and E. pauciflora
(series Obliquae). It is nonetheless possible that E.
pauciflora may have been involved in the evolution of
what appears now to be a stable taxon (Boomsma 1972).
Seedlings and coppice foliage from present-day popula¬
tions in western Victoria and southeastern South
Australia have certain ash-like characters, notably the
tendency of juvenile leaves to be falcate and to be
oriented in the vertical plane due to twisting of petioles.
Systematic Affinities

Blakely (1934) placed the nine species represented in
this study in three subseries of peppermints based prin¬
cipally on juvenile leaf shape. Angustifoliae included E.
pulchella, E. elata, E. radiata , E. amygdalina and E.
nitida; Latifoliae included E. nitida (as E. simmondsii ),
E. dives and E. coccifera ; and Connatae included E.
tenuiramis and E. risdonii. The first two of these
subseries incorporated species from both sides of Bass
Strait. Pryor and Johnson (1971) fused Blakely’s three
subseries into one ( Amygdalininae ), but they recognized

two superspecies. The first included E. risdonii and E.
tenuiramis and corresponds to Blakely’s subseries Con¬
natae. These two species are morphologically similar in
several important characters (such as having connate
juvenile foliage), and moreover, in terms of geographic
range, the distribution of the former is contained within
the distribution of the latter. Their relative positions in
the analyses reported here lend weight to the view that
they are sufficiently closely related to justify grouping in
a superspecies. The second superspecies recognized by
Pryor and Johnson (1971) included E. radiata, E.
amygdalina and E. nitida. None of the analyses reported
here linked any of the three possible species pairs, let
alone all three together in one group; that is,
resemblances based on seedling morphology appear to
be relatively weak or lacking. The justification for
grouping these three species into a superspecies appears
therefore to warrant further examination.

If Australian mainland populations currently re¬
ferred to E. nitida are to be treated as taxonomically
separate at the species level from Tasmanian and island
populations, then all Tasmanian peppermints would be
endemic (Table 2). This degree of endemism may prove

162

J. C. MARGINSON AND P. Y. LADIGES

10 —

5 —

c

o

CO

E

o

0

11 12 10 1 2 3 4

amyg pul ✓ | I

/ GRAMPIANSX
/ S.AUST. J
TASMANIA

W. PROM.
S.G’LAND
E.radiata

GRAMPIANS
E.dives

13 5 6 7 8 9 14

E. elata

FLINDERS
DOVER
TASMANIA

T l I tenu

FLINDERS 1
DOVER 1

DOVER

FLINDERS

TASMANIA

15

risd

E. coccifera

Fig. 2 —Seedling leaf analysis. Hierarchical classification of 96 seedlings based on leaf morphology at the
third node and truncated at the 15-group level. Groups 1-10 incorporate seedlings of E. nitida sensu
Willis. Abbreviations: amyg = E. amygdatina , pul = E. pulchella , risd = E. risdonii , tenu = E. tenuiramis.

to be of systematic and phylogenetic significance. The
present study placed considerable emphasis on con¬
tinuously varying morphological characters relating par¬
ticularly to leaf shape and such manifestations as apex
angle and base shape. It is therefore not surprising that
various combinations of these characters were impor¬
tant in the analyses and that the major dichotomies did
not correspond to the major geographic disjunction that
is Bass Strait. Juvenile leaf shape appears to be a plastic
character likely to vary over a wide range of more or less
related systematic groups within the genus. Its
systematic and evolutionary significance is difficult to
elucidate; there are, for instance, no broad-scale pat¬
terns sifting broad- and narrow-leaved species of
Eucalyptus into particular climatic or edaphic zones
(contrast leaf orientation). An alternative approach em¬
phasizing discretely varying characters may indicate
grounds for treating the Tasmanian peppermints as a
distinct systematic group, possibly a subseries within a
peppermint series.

Evolutionary Significance

More than 50% of Tasmanian eucalypts are con¬
sidered to be endemic (Table 2); this contrasts with only
8% of Victorian species. An explanatory hypothesis is
required that will account for the present distribution of
E. nitida and phenetically related species whilst
recognizing the high degree of endemism among Tas¬
manian peppermints.

The islands of Bass Strait are considered to be rem¬
nants of a land bridge connecting the Australian
mainland to Tasmania (Jennings 1971). During the
Pleistocene, the land bridge formed with each period of
glaciation and associated lowering of sea level. It is
estimated that Tasmanian temperatures would have
been about 5°C lower than at present, and the tree-line
would have been as low as near present sea level on the
west coast to 4-500 m above sea level on the east coast
during periods of glaciation (Galloway 1965, Macphail
1979). Forests, woodlands, and scattered trees would

EUCALYPTUS NITIDA AND BASS STRAIT

163

Fig. 3 —Seedling leaf analysis. Ordination of group centroids and links between nearest neighbours.
Groups as indicated in Fig. 2. Axis 3 is represented as the diameter of spheres (following Gillison 1978).

have been restricted to a coastal fringe, to the central
midlands of Tasmania and to the land bridge (J. Hope
1973, G. Hope 1978, Macphail 1979). At the time of
maximum glaciation during the late Pleistocene, open
grassland communities with scattered eucalypts would
have been widespread in southeastern Australia in¬
cluding the low-lying central part of the land bridge
(Hope 1978). Outcropping granite hills that form the
present islands of Bass Strait may have provided refuges
for Eucalyptus forest communities as suggested by Ladd
(1979) for the granite highlands of Wilsons Promontory.
The most recent interruption of the land bridge to form
Bass Strait probably occurred between 12 000 and
13 500 years ago (Jennings 1971, Hope 1973), initially

separating Tasmania from the Australian mainland to
the north of King and Flinders Islands and probably the
Kent Group.

During interglacial periods when Bass Strait re¬
formed, selection pressures may have resulted in
divergence and speciation. During subsequent periods of
glaciation, boundaries of Tasmanian eucalypt species
probably shifted northward, mainly in response to
climatic constraints. New habitats would have become
available to lowland species as the islands increased in
size and coalesced in sequence from south to north to
form land corridors through both eastern and western
ends of Bass Strait. As the land bridge formed, probably
by connecting the eastern land corridor to Wilsons Pro-

164

J. C. MARGINSON AND P. Y. LADIGES

TASMANIA

Fig. 4 —Juvenile leaf analysis. Hierarchical classification of 85 seedlings based on leaf morphology at the
seventh or higher node and truncated at the 15-group level. Groups 1-7 incorporate seedlings of £. nitida
sensu Willis. Abbreviations: amyg = £. amygdalina , cocc = £. coccifera , pul = £. pulchella , rad = £.
radiata , risd = £. risdonii , tenu = £. tenuiramis.

montory, northbound species occupying the land cor¬
ridor may have met more widely tolerant southbound
species capable of migrating against the climatic gra¬
dients into new habitats. Genetic isolation may or may
not have been complete, depending on the degree of
divergence that had occurred among related species (or
populations) during the previous interglacial period of
geographic isolation. The subsequent rise in sea level
and loss of suitable habitats would, perforce, lead to the
contraction of species boundaries both north and south,
and may have left relatively few species able to survive
on the islands of the Strait. But it may have lead to new
opportunities for divergence and speciation, and
geographic and phylogenetic origins may become
obscured.

The distribution of E. nitida sensu stricto in
Tasmania and on some islands of Bass Strait parallels
that reported by Hope (1973) for many members of the
vertebrate fauna. Following her interpretation, it is sug¬
gested that the extent of phenetic similarity between
island and Tasmanian mainland populations of the

species may be taken to indicate the degree of genetic
isolation of that species at any time when it may have
been in contact with related species migrating southward
on the land bridge. If present island populations of E.
nitida were to be intermediate between Tasmanian and
Australian mainland species or populations, this would
suggest that, during the last period of glaciation,
populations originating from either side of the Strait
and meeting on the land bridge were not reproductively
isolated. We report here evidence that, at least with
respect to seedling morphology, island populations are
not only more similar to Tasmanian mainland popula¬
tions than they are to any other population or species
represented in this study, but also are not intermediate
between Tasmanian and any Australian mainland
populations currently referred to E. nitida sensu Willis.
The extent of the disjunction implies at least a degree of
genetic isolation, if not complete isolation. In so far as
island populations differ from Tasmanian mainland
populations, some divergence may have occurred since
the sea-level rose isolating the present islands; alter-

EUCALYPTUS NITIDA AND BASS STRAIT

165

Fig. 5 —Juvenile leaf analysis. Ordination of group centroids, and links between nearest neighbours.
Groups as indicated in Fig. 4. Axis 3 is represented as the diameter of spheres (following Gillison 1978).

natively, the rising sea may have disrupted a clinal se¬
quence of variation, the island populations being at the
northern end of a gradient. This study provides no way
of choosing between these two options.

Australian mainland peppermint species and popula¬
tions may have had various origins. Some may have
originated from one or more ancestral species in
response to directional selection pressures on the
mainland. Others may have resulted from back-
migration following radiation in Tasmania.

Further studies are in progress testing the hypotheses
outlined above. These will involve sampling on a
population basis, taking account of adult morphology
and applying the methods of phylogenetic systematics

and variance biogeography as described by Wiley
(1980). In the lattermost context, it is of interest that E.
nitida appears to be absent from the present native flora
of King Island while the tall, well formed species, E.
brookerana , has not been recorded on Flinders Island.
This pattern parallels that described for some members
of various vertebrate groups (summarized by Hope
1973) and has been attributed principally to higher rain¬
fall on the western side of land bridges formed during
the Pleistocene. In contrast, Tasmanian eucalypt species
that occur on both islands are geographically
widespread on the Australian mainland (E. viminalis, E.
globulus and E. ovata ), again resembling some
vertebrate distribution patterns.

166

J. C. MARGINSON AND P. Y. LADIGES

Table 6

Juvenile Leaf Analysis

Means and frequencies (%) for the Seven Characters Contributing Most to the First Three Dichotomies in the Hierar¬
chical Classification (a) and in the Seven Groups that Incorporated Seedlings of E. nitidasensu Willis (b)

Group members and combinations are those indicated in Fig. 4. Numbers of seedlings arc shown in brackets.

(a) (b)

Characters Group Combinations Groups incorporating E. nitida sensu Willis

3 + 4 + 5 + 6 +

Vic.,

S.A.

Bass Strait island.

Tas. groups

1+2+11 +

7 + 8 + 9+10

groups

1st dichotomy

12+13+14

+ 15

1

2

3

4

5

6

7

(31)

(54)

(8)

(8)

(14)

(9)

(7)

(2)

(4)

Secondary vein angle (°)

28

46

28

30

43

47

50

49

35

Apex angle (°)

26

64

28

25

53

58

72

54

42

Ratio juvenile leaf length:

maximum width

7.3

2.6

6.0

6.0

3.0

3.0

2.2

3.4

5.0

Maximum juvenile leaf width (mm)

11.3

20.9

14.5

14.0

21.1

21.1

24.7

15.6

10.0

Oil gland density (per cm 2 )

505

88

279

1012

88

171

92

32

0

3+4+5+6+

2nd dichotomy

7 + 8 + 15

9+10

(41)

(13)

Juvenile leaf colour — yellow-green

15%

0

0

0

0

0

0

0

0

green

63%

8%

63%

100%

86%

89%

14%

0

100%

blue-green

22%

23%

37%

0

7%

11%

86%

100%

0

glaucous

0

69%

0

0

7%

0

0

0

0

3rd dichotomy

1+2+13+14 11 + 12

Ratio juvenile leaf length from base

to widest point over total length

0.22

0.55

0.36

0.17

0.26

0.33

0.38

0.44

0.26

ADCKNOWLEDGEMENTS

Most of the seed collections were obtained from
CS1RO Division of Forest Research (courtesy M. I. H.
Brooker), the Forests Commission of Victoria, M. A.
Marginson, Melbourne, and A. M. Gray, Kingston,
Tasmania. The interest of A. M. Gray in this project
was a source of considerable stimulation. Advice,
assistance and information from the following people is
gratefully acknowledged: Drs D. H. Ashton, B. Gott,
M. Harding, S. Murray-Smith, D. Ross; Messrs P.
Barnett (King Island), G. M. Chippendale and J.
Whinray (Flinders Island).

REFERENCES

Black, J. M., 1964. Flora of South Australia. Part 3. Govt.
Printer, Adelaide.

Blake, S. T., 1953. Botanical contributions of the Northern
Australia Regional Survey. 1. Studies on northern
Australian species of Eucalyptus. Aust. J. Bot. 1: 185-352.
Blakely, W. F., 1934. A Key to the Eucalypts. 1st edition. The
Worker Trustees, Sydney.

Blakely, W. F. t 1965. A Key to the Eucalypts. 3rd edition.

Forestry and Timber Bureau, Canberra.

Boomsma, C. D., 1972. Native Trees of South Australia.

Woods and Forests Dept., Adelaide.

Boomsma, C. D. & Lewis, N. B., 1980. The Native Forest
and Woodland Vegetation of South Australia. Woods and
Forests Dept., Adelaide.

Brett, R. G., 1938. A survey of Eucalyptus species in
Tasmania. Pap. Proc. R. Soc. Tasm. 1937: 75-109.

Brooker, M. I. H., 1977. Internal bud morphology, seedling
characters and classification in the ash group of eucalypts.
Aust . For. Res. 7: 197-207.

Brooker, M. I. H., 1979. A revision of the informal series
Foecundae Pryor and Johnson of the genus Eucalyptus
L’Herit. and notes on variation in the genus. Brunonia 2:
125-170.

Chippendale, G. M., 1973. Eucalypts of the Western
Australian Goldfields. Aust. Govt. Publishing Service,
Canberra.

Curtis, W. M. & Morris, D. I., 1975. The Student's Flora
of Tasmania. Part 1. 2nd edition. Govt. Printer, Hobart.

Ewart, A. J., 1930. Flora of Victoria. Govt. Printer,
Melbourne.

Galloway, R. W., 1965. Late Quaternary climates in
Australia. J. Geol. 73: 603-618.

Gillison, A. N., 1978. Minimum spanning ordination —a
graphic-analytic technique for three-dimensional ordina¬
tion display. Aust. J. Ecol. 3: 233-238.

Hall, N., Johnston, R. D. & Chippendale, G. W., 1970.
Forest Trees of Australia. Aust. Govt. Publishing Service,
Canberra.

Hall, N., Johnston, R. D. & Marryatt, R., 1963. The
Natural Occurrence of the Eucalypts. Forestry and Timber
Bureau Leaflet No. 65, Canberra.

Hope, G. S., 1978. The Late Pleistocene and Holocene vegeta-
tional history of Hunter Island, north-western Tasmania.
Aust. J. Bot. 26: 493-514.

EUCALYPTUS NITIDA AND BASS STRAIT

167

Hope, J. H., 1973. Mammals of the Bass Strait islands. Proc.
R. Soc. Viet. 85: 163-195.

Jackson, W. D., 1965. Vegetation. In A tlas of Tasmania, J. L.

Davies, ed.. Lands and Surveys Dept., Hobart, 30-35.
Jennings, J. N., 1971. Sea-level changes and land links. In
Aboriginal Man and Environment in Australia , D. J.
Mulvaney and J. Golson, eds, A.N.U. Press, Canberra,
1-13.

Johnson, L. A. S. & Blaxell, D. F., 1973. New taxa and
combinations in Eucalyptus- II. Contrib. N.S.W. Nat.
Herb. 4: 379-383.

Kirkpatrick, J. B., Massey, J. S. & Parsons, R. F., 1974.
Natural history of Curtis Island, Bass Strait. 2. Soils and
vegetation. With notes on Rodondo Island. Pap. Proc. R.
Soc. Tasm. 107: 131-144.

Ladd, P. G., 1979. A Holocene vegetation record from the
eastern side of Wilsons Promontory, Victoria. New Phytol.
82: 265-276.

Ladiges, P. Y., Gray, A. M. & Brooker, M. I. H., 1981.
Patterns of geographic variation, based on seedling mor¬
phology, in Eucalyptus ovata Labill. and E. brookerana A.
M. Gray and comparisons with some other Eucalyptus
species. Aust. J. Bot. 29: 593-603.

Macphail, M. K., 1978. Vegetation and climate in southern
Tasmania since the last glaciation. Quat. Res. 11: 306-341.

Maiden, J. H., 1922. A Critical Revision of the Genus
Eucalyptus. Vol. 6, Part 6. Govt. Printer, Sydney.
Parsons, R. F., 1966. The soils and vegetation at Tidal River,
Wilson’s Promontory. Proc. R. Soc. Viet. 79: 319-354.
Parsons, R. F., Kirkpatrick, J. B. & Carr, G. W., 1977.
Native vegetation of the Otway Region, Victoria. Proc. R.
Soc. Viet. 89: 77-88.

Parsons, R. F., Scarlett, N. H. & Rosengren, N. J., 1972.
Ecology of some Eucalyptus woodlands near Hall’s Gap,
Victoria. Victorian Nat. 89: 41-49.

Pryor, L. D. & Johnson, L. A. S., 1971. A Classification of
the Eucalypts. A.N.U. Press, Canberra.

Wilcox, M. D., 1979. The peppermint group of eucalypts.

N.Z.J. Forestry Sci. 9: 262-266.

Wiley, E. O., 1980. Phylogenetic systematics and vicariance
biogeography. Syst. Bot. 5: 194-220.

Williams, W. T. (ed.), 1976. Pattern Analysis in Agricultural
Science. C.S.I.R.O., Melbourne.

Willis, J. H., 1967. Systematic notes on the indigenous
Australian flora. Muelleria. 1: 117-163.

Willis, J. H., 1970. The shining peppermint ( Eucalyptus
nitida). Victorian Foresters* Newsletter. 26: 4-5.

Willis, J. H., 1972. A Handbook to Plants in Victoria. Vol. 2.
Melbourne University Press, Melbourne.

PROCEEDINGS

OF THE

ROYAL SOCIETY OF VICTORIA

Volume 94

NUMBER 4

ROYAL SOCIETY’S HALL
9 VICTORIA STREET, MELBOURNE 3000

Contents of Volume 94 Number 4

Article b Page

14 Upogebia niugini (Crustacea:Decapoda) a new shrimp from Papua New Guinea

.By Gary C. B. Poore 169

15 The geology of Cape Everard, Victoria — By Margaret C. Fry & C. J. L. Wilson 173

16 Holocene ostracods, other invertebrates and fish remains from cores of four maar

lakes in southeastern Australia.By P. DeDeckker 183

17 Mammals of southwestern Victoria from the Little Desert to the coast .... By P. W.

Menkhorst & C. M. Beardsell 221

PROC. R. SOC. VICT. vol. 94, no. 4, 169-172, September 1982

UPOGEBIA NIUGINI (CRUSTACEA) A NEW SHRIMP FROM

PAPUA NEW GUINEA

By Gary C. B. Poore

National Museum of Victoria, 285-321 Russell Street, Melbourne, Victoria 3000

Abstract: Upogebia niugini sp. nov. is the fourth species of Upogebia from the seas north of
Australia possessing ventral rostral spines. Its affinities with other species are discussed.

A small collection of Crustacea from Port Moresby,
housed in the National Museum of Victoria (NMVJ),
contains specimens of the thalassinidean shrimp genus
Upogebia. Although closely related to known species
these distinctive specimens demand description as a new
species. The higher syslematics and diagnosis of the
genus, with particular reference to Australia, have been
given by Poore & Griffin (1979).

Tribe Thalassinidea

Family Upogebiidae

Upogebia niugini sp. nov.

Figs 1, 2

Description of Holotype: Carapace

Anterior region of carapace about one-third longer
than posterior region. Cervical groove well defined dor-
sally, not visible laterally; an alternative suture running
diagonally anierior to groove. Linea thalassinica exten¬
ding whole length of carapace.

Rostrum widest about one-third way along, 1.3 times
as long as wide. Ventral surface with three spiniform
teeth, all reaching as far forward as tip of rostrum. Dor¬
sal surface with four irregular longitudinal rows of den¬
ticles extending onto anterior half of gastric region.
Lateral margin of rostrum with eight teeth.

Gastric region 1.3 times as wide as rostrum and
separated from it by broad lateral grooves; lateral crest
with 11 (right) and 12 (left) tubercles. Majority of gastric
region and rostrum with dense cover of plumose setae
dorsally.

Anterolateral margin with five denticles; lateral
region posterior to this with four minute denticles. Line
of cervical groove defined laterally only by seven den¬
ticles.

Antennule and antenna

Antennule: first article with one or two distoventral
spines, flagellum shorter than peduncle. Antenna: first
article with one distoventral spine; second article with
three small proximal dorsal spines and three distal ven¬
tral spines; third article with three ventral spines;
scaphocerite a small bifid scale.

Mouthparts

Mandible with a prominent proximal tooth on den¬

ticulate mesial margin. Maxillae, first and second max-
illipeds typical of genus. Third maxilliped with 2-articled
exopod; coxa with an epipod and two mesial hooked
spines; ischium with only feeble mesial spine row and
one spine laterally.

Pereopods

Pereopod 1 subchelate. Coxa with one mesial den¬
ticle. Ischium with one minute ventral denticle. Merus
2.5 times as long as w'ide, with 12 ventral short spines
and with one dorsal spine distally. Carpus with about 13
curved spines scattered over dorsal edge, one on ventral
margin; distal mesial margin bearing two spines, one
prominent. Propodus 2.3 times as long as wide, with ir¬
regular longitudinal row's of strong curved spines: 13
along anterior margin, 7-8 in each of three rows along
mesial surface and three along the posterior margin
leading to the fixed finger; its lateral surface with few
denticles near posterior margin. Fixed finger a short,
broadly based, spur. Dactyl 0.75 times as long as pro¬
podus, anterior margin denticulate.

Pereopod 2 coxa with mesial denticles; ischium with
minute denticles posteriorly, one distal spine anteriorly;
carpus with one posterior and four anterior spines.

Pereopod 3 coxa with mesial spines; ischium with
seven denticles posteriorly; carpus with one posterior
and one anterior spine.

Pereopod 4 unarmed, similar in form to pereopod 3.

Pereopod 5 subchelate, dactyl about 3 times as long
as fixed finger.

Tailfan

Uropods longer than telson; exopod oval, 1.7 times
as long as wide; endopod triangular, widest in proximal
half.

Telson 1.2 times as wide as long, posterior margin
concave; dorsally two broad transverse carinae in
proximal half.

Holotype: Female (with left pereopods 1-4, right
pereopods 3, 5), c.l. 9.1 mm (NMV J1653).

Type Locality: Papua New Guinea, Port Moresby, 80
m off eastern side of Esade Reef, shelly-muddy sedi¬
ment, 15 m, coll. J. E. Watson and J. Carey using
SCUBA, 28 July 1981.

Paratypes: Male (without pereopods), c.l. 6.9 mm;
female (with left pereopod 5 only), c.l. 8.1 mm, both
from type locality (NMV J1654).

169

170

G. C. B. POORE

a-d, j

UPOGEBIA FROM NEW GUINEA

171

Fig. 1 Upogebia niugini sp. nov., holotype. a , b, lateral and dorsal views of anterior region of carapace;
c, telson and uropod; d> antennules; e, mandible;/, g, maxillae 1, 2; h-j, maxillipeds 1-3; k, lateral view
of base of ischium of maxilliped 3. (Figures are without setae, mouthparts and antennules are from left

side.) Scale = 1 mm.

172

G. C. B. POORE

Variation: The ventral rostral spines of all three
specimens are uniform. The dorsal spines of the female
paratype are in only two, longitudinal rows, that are,
especially on the rostrum, more regular than in the
holotype.

Etymology: The specific epithet niugini is from the
Pidgin language and is sometimes used to refer to the
nation of Papua New' Guinea.

Remarks: Upogebia niugini belongs to the group of In-
dopacifk species possessing ventral rostral spines: U.
acanthochela Sakai, U. acutispina de Saint Laurent &
Ngoc-Ho, U. ceratophora De Man, and U. monoceros
De Man. Upogebia talismani Bouvier, from the warm
temperate and tropical Atlantic, also belongs to this
group, first separated from other species of Upogebia by
De Man (1928) in his key. De Saint Laurent & Ngoc-Ho
(1979) defined the group more closely with additional
characters:

— rostrum with one or several ventral spines;

— anterolateral margin of the carapace armed with a
series of spinules;

— posterior margin of the telson more or less concave;

— mandible without a sharp anterior tooth;

— fixed finger of pereopod 1 reduced to a strong
spiniform projection; mesial face of the propodus
armed with one or several row's of spines;

— coxae of the pereopods with fine mesial spinules;

— branchial filaments simple.

Upogebia niugini is most closely related to U .
acutispina from northwestern Australia, many features
of the spination of the pereopods being virtually in¬
distinguishable. However, the three ventral rostral
spines are shorter in U. niugini , the telson relatively
broader, and the rostrum narrower than in U.
acutispina. Upogebia monoceros from Java and U.
ceratophora from eastern Indonesia (De Man 1928, de
Saint Laurent & Ngoc-Ho 1979) possess only one ventral
rostral spine; U. acanthochela from the Yellow Sea has
two (Sakai 1967).

The number of morphologically similar species in
this group, revealed by a relatively small amount of
sampling, suggests a high rate of species radiation in
Upogebia in the seas around the Indonesian archi¬
pelago.

Another Indopacific species, Upogebia spinifrons
(Haswcll) from northeastern Australia also possesses
ventral rostral spines but differs from the group defined
by de Saint Laurent & Ngoc-Ho (1979) in several
respects. Dorsal spines on the rostrum and gastric region
are absent or obsolete, the posterior margin of the telson
is not concave, the fixed finger of pereopod 1 is substan¬
tial and toothed, and the mesial face of the propodus of
pereopod 1 is without spine rows (Poore & Griffin 1979).

ACKNOWLEDGEMENTS

I am indebted to Jan Watson, Melbourne, for mak¬
ing this material available for study, to Michele de Saint
Laurent, Paris, for helpful comments on the
manuscript, and to Ron Vanderval, Melbourne, for ad¬
vice on the etymology of the specific name.

REFERENCES

Man, J. G. De, 1928. The decapoda of the Siboga-Expedition.
Part VII. The Thalassinidae and Callianassidae collected by
the Siboga-Expedition, with .some remarks on the
Laomediidae. Siboga-Exped. 39a (6): 1-187.

Poore, G. C. B., & Griffin, D. J. G., 1979. The
Thalassinidea (Crustacea: Decapoda) of Australia. Rec.
Aust. Mus. 32: 217-321.

Saint Laurent, M. de, & Ngoc-Ho, N., 1979. Description
de deux especes nouvclles du genre Upogebia Leach, 1814
(Decapoda, Upogebiidae). Crustaceana 37: 57-70.

Sakai, K., 1967. Three new species of Thalassinidea
(Decapoda, Crustacea) from Japan. Rec. Crust. Japan 3:
39-51.

PROC. R. SOC. VICT. vol. 94, no. 4, 173-181, December 1982

THE GEOLOGY OF CAPE EVERARD, VICTORIA

By Margaret C. Fry and C. J. L. Wilson
Department of Geology, University of Melbourne, Parkville, Victoria 3052

Abstract: The Ordovician sediments of Cape Everard constitute a sequence of sandstones,
grcywackes and siltstones interbedded with minor black shales. Four lithofacies are recognized and are in¬
terpreted as proximal turbidiles, thought to be deposited by turbidity currents. The association of
lithofacies suggests that sedimentation occurred on and around the upper mid-fan region of a submarine
fan system. Palaeocurrent indicators suggest turbidity current transport in a northerly direction. The
strata are folded into tight folds and have undergone low grade regional metamorphism, with little
disruption of the original sedimentary texture. These sediments are intruded by granite with insignificant
contact metamorphic effects and are overlain by a possibly Miocene calcarenite.

Cape Everard, which includes Point Hicks, is a
granite headland 40 km southeast of Cann River
township in East Gippsland (Fig. 1). Exposure is limited
to shore platforms and cliffs as most of the Cape and ad¬
jacent areas arc covered by a complex of Holocene
transgressive dunes superimposed on older (Pleistocene)
sand ridges (Rosengren 1978). The granite is considered
by Douglas (1974) to be a small outlier of the Early
Devonian Bega Batholith (396 ± 16 Ma, K/Ar; Richards
& Singleton 1981) and intrudes folded Ordovician
sedimentary rocks. Unconformably overlying the
granite are restricted calcarenite deposits and the granite
is intruded by three olivine bearing basaltic dykes (1 m
wide).

The best exposures of the Ordovician are found on
the shore between th£ mouths of the Thurra and Mueller
Rivers, in areas 1 and 2 (Fig. 1). Here a 200 m thick suc¬
cession of turbidites crops out and is comprised of three
main lithologies namely: coarse sandstone, siltstone and
shale. Minor chert (<2 m thick) has been found with
comparable rocks in the Thurra River. All these rocks
bear the weak imprint of lower greenschist facies
regional metamorphism and have near vertical dips (Fig.
2) and one generation of upright isoclinal folds which
plunge 15° towards 356°. Faults with unknown
displacements parallel the N-S trending bedding and are
only recognised by comparison of sedimentary cross-
sections; these are considered to be bedding plane
thrusts and are similar to those described by Wilson et
al. (1982). Other vertical faults, often accompanied by
quartz veins, trend to the northwest (Fig. 2). They ap¬
pear to be later features and arc accompanied by minor
crenulation cleavage.

These Ordovician sediments are similar to the
Mallacoota Beds described by Fenton et al. (1982),
which are of Late Ordovician age on the basis of grap-
tolitcs found at Seal Creek (P. de Hedouville, un¬
published data, April 1982). A conodont fauna (I.
Stewart in Webby et al . 1981, p. 30) obtained from a
silcified shale unit in a road cutting on the north side of
the Thurra River bridge at Cape Everard, also suggests a
Late Ordovician age. The sediments at Cape Everard are
coarser grained and not as well-laminated as the Malla¬
coota succession. It is therefore the primary purpose of
this paper to describe the sediments and determine the

environment of deposition and compare it with the
Mallacoota sequence. At the same time, the other rock
types recognised in the area will be briefly described.
Data were collected following the methods used in Fen¬
ton et al. (1982) and Wilson et al. (1982) and further
details may be found in Fry (1981).

ORDOVICIAN SEDIMENT PETROGRAPHY

The sequence is characterised by thick-bedded and
coarse-grained deposits that are comparable to those
described by Walker and Mutti (1973). The turbidites
show numerous sedimentary structures and different
combinations of the divisions of Bouma (1962).
Measured sections illustrating the sedimentary features
in areas I and 2 are shown in Fig. 3. Sections 9A and 9B
are situated on adjacent limbs of a fold. Younging
evidence in Sections 14 and 15 suggests that a fold lies
between them, but the lithology, thickness and sedimen¬
tary structures do not correlate. This is interpreted as be¬
ing due to a bedding-plane thrust sub-parallel to the
axial-surface of a tight-fold.

Three main lithologies arc distinguished in the rocks
at Cape Everard.

Sandstone and greywackes are mainly clast-supported
with clay matrix contents of <5 to 20%. Greywackes
(with more than 10% matrix) only comprise about 25%
of this lithology. Quartz with undulose extinction com¬
prises 95-100% of the framework grains, whereas meta¬
morphic quartz displaying re-crystallization is restricted
to coarse sandstone and “coarse clasts”. Grains are
subangular to rounded (Fig. 4) and size (<1 mm-4.0
mm) varies within individual beds and between beds.
Plagioclasc is present as rare grains. The matrix is com¬
posed of muscovite (<2 mm), partly recrystallized and
deformed during the regional metamorphism. The ac¬
cessory minerals are hornblende, zircon, tourmaline and
iron oxide. Intraclasts of shale and mud are common.

Bodies interpreted as concretionary features, post¬
date sedimentation but predate deformation and are
comparable to similar features that are folded and
metamorphosed in other areas in East Gippsland (e.g.
Eaton 1980). They occur at different intervals within
some coarse sandstone beds as zones up to 8 cm thick
that show slight colour variation and sometimes have
the appearance of distinct units. These coarse layers lack

173

174

M. C. FRY AND C. J. L. WILSON

Fig. 1 — Map of Cape Everard area showing location of major rock types and areas 1,2, and 3 detailed in
Fig. 2. The contoured data are poles to joints and quartz veins in the granitoid rocks, and contours are
>1% per 1% area, >5% per 1% area, > 10% per 1% area and >20% per 1% area.

clear fining-upwards grading and are discontinuous
along their length (1-4 m) with rounded terminations.
There are no obvious compositional differences in the
clastic sand-grain content between these layers and the
adjacent sandstones, but the matrix of these bodies may
contain a trace of calcite.

Siltstones are matrix supported (>35%) with quartz
(<1 mm) and rare altered plagioclase. The matrix is
composed of recrystallized muscovite and chlorite. Ac¬
cessory minerals are the same as in the sandstones.
Shales are a minor component of the sediments at Cape
Everard. They are grey to black and are similar to those
described by Fenton et al. (1982). They are composed of
quartz, muscovite and chlorite.

ORDOVICIAN SEDIMENT LITHOFACIES

Four major lithofacies crop out along the coast and
are summarized below.

Massive sandstones

These are composed of coarse sands that fall within
the A division of Bouma (1962) and comprise 13% of all
beds. Individual sheets range from 1-100 cm thick with a
median thickness of 40 cm, and are generally not lateral¬
ly persistent for more than a few tens of metres of out¬
crop, with marked pinch and swell. Grading, wher.
present, is usually confined to the top of the A division
and may be the only internal sedimentary structure. The
base may be either erosional or sharp (Fig. 5A) with rare
flame structures (Fig. 5B). Amalgamation of the graded
beds has resulted in apparently massive sand sequences
more than 10 m thick, such as Sections 5, 6, and 7 (Fig.
3). Other sands occur as coarse-grained lens-like bodies

(<15x60 cm) that may show discontinuous coarse
sandy layers up to 10 cm thick at the base.

Proximal turbidites

These comprise 57% of all beds and are composed of
the AB, ABE and AE divisions of Bouma (1962). They
consist predominantly of graded sandstone and
siltstone, 5-35 cm thick (e.g. Sections 8, 9 and 10). Many
fine sandstones and silt units have ripple cross-bedding
or are slumped in the middle parts of thicker sand bodies
(Fig. 5C and Sections 9B, 12, 13 in Fig. 3). There is also
gradation between the two sets of structures. Coarse
clasts (Fig. 5D) with lengths up to 12 cm, distributed in
irregular fashion (in A division) or aligned with bedding
(in B division) are particularly noticeable in Sections 4, 5
and 6 (Fig. 3).

Clasts of shale that range in length from 0.5-30 cm
also occur in the A and B divisions of the more sandy
units (see Sections 1 and 2 in Fig. 3). They appear to be
rip-up clasts and occur in beds with an abundance of
coarse sandstone clasts, suggesting an erosional history.

Distal turbidites

Where the lowest division of a bed is B, combina¬
tions of the Bouma sequences such as BCDE, BCE, CE
and DE comprise 18% of all beds. They range in bed
thickness from 1 to 100 cm with a median of 16 cm and
have most of the features described by Fenton et ai
(1982). Also common in the C division of these
sediments is the occurrence of both coarse-grained clasts
and shale clasts (Fig. 5E). The clasts are finer and
smaller than those observed in the proximal turbidites
and reveal internal folding and are enclosed in a bed that
displays complete irregularity, except where micaceous

GEOLOGY OF CAPE EVERARD

175

Fig. 2 —Geological maps of areas 1, 2 and 3 (of Fig. 1) showing trends in bedding and location of
sedimentological sections. Stercographic projections, on the lower hemisphere of an equal area net show
the orientation of poles to bedding planes, contours are the same as Fig. 1.

176

M. C. FRY AND C. J. L. WILSON

Fig. 3 -Measured sections of the Cape Everard Ordovician sequence in areas 1 and 2. Locations of the

sections are shown in Fig. 2.

1

W c*. 1 ^

W PiSfr

GEOLOGY OF CAPE EVERARD

LEGEND

LITHOLOGY

SANOSTONE

8CT8TONE

SHALE

M MTERBEOOED FINE SANDSTONE ANO SHALE

SEDIMENTARY STRUCTURES

GRADED BED
PLANE LAMMATION
□ COARSE CLASTS
|'-j SHALE CLASTS
V SLUMP WOUCED DEFORMATION
i* TECTONKJALLY INOUCED DEFORMATION

r FAULT

E EAST END OF SECTION

W WEST ENO OF SECTION
1 4 CROSS SECTION NUMBER

c CLAYSTONE
*c SILTY CLAYSTONE
i *®rn» c * * 9A.TSTONE

I* FINE SANOSTONE
MEDIUM SANOSTONE
c» COARSE SANOSTONE

e

TO W E E E E c ltTT6^ E

Fig. 3 (continued)

w

178

M. C. FRY AND C. J. L. WILSON

Fig. 4 — Photomicrograph illustrating coarse sandstone texture
in Melbourne University Geology Department sample R27859.
A recrystallised quartz grain is seen at A and a rare feldspar
grain at B.

minerals and shale clasts are folded around the coarse
sand clasts (Fig. 6 ). It is suggested that some C division
clasts are remnant bedding clasts disrupted by slumping
whereas the others, particularly the smaller ones, may
represent shale introduced with the sediment load.

Thin sandstone beds interbedded with shale

These units make up 11% of all beds and occur as
sequences between 0.8-2 m thick. The presence of cross
bedding and ripple cross bedding in the fine sandstones
indicates a Bouma C division with the seemingly massive
overlying shale characteristic of E division. The abrupt
change (Fig. 5F) between the two divisions could be in¬
terpreted as showing a change to pelagic deposition and
the E division would be expected to show no internal
grading or sedimentary features. It is on this basis that
Fenton et al. (1982) suggested the fine sandstones were
reworked by intermittent bottom currents, with subse¬
quent slow deposition of a pelagite layer. However,
close examination revealed overall grading and ripple
cross bedding within each unit, suggesting that the beds
are fine-grained tail deposits of individual turbidite
events.

ENVIRONMENT OF DEPOSITION

In the Ordovician sediments of Cape Everard there is
a preponderance of coarse sandstones which represent
beds deposited in a high flow' regime. Sand deposition
was interrupted now and then by accumulations of silt
and mud (under normal deep marine conditions) which
form intercalations of siltstone and shale in a sequence
composed predominantly of sand. These changes could
be accounted for in terms of modern facies interpreta¬
tion (Walker 1978) as deposits near or within a rapidly
changing suprafan region.

Repetition of massive and graded sandstones (Sec¬
tions 4, 5 and 6 ) with large lenses of coarse material (e.g.
Section 12) suggests channel fill or overbank deposits in
the upper or mid-fan region (Walker 1978). Coarse sand
and shale clasts in the Bouma A and B divisions also
suggest rapid erosion with short transport distance.

Such an origin also explains some of the clasts recognis¬
ed in C division whereas others may be remnants j n
slump beds. Fining upward sequences are restricted but
are seen in Sections 12 (between X and Y), 18 and 19 .
These are repeating beds of diminishing size which
Walker and Mutti (1973) interpret as a sequence formed
in a prograding sumbarine fan. However, we favour the
interpretation of Fenton et al. (1982) that at Cape
Everard and Mallacoota these sequences are part of a
channelled suprafan deposit and indicate proximity t 0
the channel. The interbedded siltstones and shale with
graded CE Bouma divisions suggest that these are fine
tail deposits of individual turbidite events rather than
regular reworking of the sands by intermittent bottom
currents as proposed by Fenton et al. (1982) for similar
rocks at Mallacoota. Similarly as these interbedded
siltstones and shales are generally grouped together (e.g.
Sections 3, 11, 12 and 14 to 18) as packets of distal tu r -
bidites, they may be interpreted as deposits derived from
more than one channel (M. W. Fenton pers. comm.).

The initial dips of C-division cross-bed lamellae
range from 15° to 25° and have been used as indicators
of directions of sediment transport. They show a nor¬
therly current direction (Fig. 7) which is in agreement
with measurements on similar rocks by Fenton et al.
(1982) and Cas et al. (1980). The use of underlying scour
or tool-marked surfaces was not possible as all exposed
surfaces were essentially two-dimensional.

There is good correlation of individual graded beds
near the fold hinge in Sections 9A and 9B, but as
distance from the hinge increases the correlation
decreases. This lateral change would be a response to the
changing environment induced by the flow r and position
within the channelled suprafan region.

Two isolated contact metamorphosed cherts crop
out in the Thurra River but their relation to the sedimen¬
tary sequence described here is unclear. Such cherts may
be deposited in regions free of turbidity llow due either
to topographic highs (channel levees) or to channel
movement away from the area (Fenton et al. 1982). This
lends support to the idea that the Cape Everard se¬
quence is part of an extensive deposit formed on a rapid¬
ly changing channelled suprafan.

A lower mid-fan depositional environment has been
favoured by Fenton et al. (1982) for the Mallacoota
Beds at Mallacoota. However at Cape Everard there is
strong evidence for channelled lithofacies that suggests
an upper mid-fan. Distal outer fan and basin plain
associations and proximal inner fan, slope and shelf
associations arc absent. A possible explanation for this
is that the Ordovician in this part of Victoria was not a
sedimentary wedge built off a prograding continental
margin, instead it was a deeper water, channelled
suprafan deposit (Walker & Mutti 1973). The detritus
would be in part supplied from older fan sediments that
were made available for redeposition by contem¬
poraneous tectonism or sedimentary subsidence. Such
an interpretation would explain the occurrence of sand¬
stone and shale clasts within the coarse sandstone units.
The intralayer slumps in the finer sandy units apparently

GEOLOGY OF CAPE EVERARD

179

Fig. 5 —A, coarse sandstone lens with sharp base and without grading, resting on a thin E division shale
and C division in section 5. B, flame structure in a coarse A division in a massive sandstone in section 13.
C, slumped sandstone-siltstone unit in section 13. D, irregular coarse clasts in an A division in Section 5.
The clasts are grain supported with grain sizes greater than the surrounding sand. E, shale clasts compos¬
ed of clay and phyllosilicate plates in a C division in section 18. F, intercalation of shale and siltstone in

section 16.

indicate minor slope failure or may have resulted from
impact of large gravity flows that crevassed the graded
sandstone-siltstone facies. The source-rocks for the se¬
quence are probably, as previously proposed (Fenton et
al 1982), the Cambrian and Precambrian sediments of
the Ross Orogen of eastern Antarctica (Tcssensohn et al.
1981).

METAMORPHOSED ORDOVICIAN

Limited outcrops of contact metamorphosed
sediments can be seen in the Thurra River and as pods
isolated by the granites, for example in area 3 (Fig. 2).
They are sandy units that have been transformed to
quartzites with large recrystallized muscovite. Minor
shale intercalations contain spots of chlorite, muscovite,

180

M. C. FRY AND C. J. L. WILSON

Fig. 6 —Photomicrographs illustrating structures associated
with sandstone and shale clasts (a) slump between siltstone and
shale unit in Melbourne University Geology Department sam¬
ple R27873; (b) bedding being folded adjacent to a sandstone
clast in Melbourne University Geology Department sample
R27863. Width of both micrographs are 4 mm.

quartz and opaques. The sediments have retained their
folded pattern but display some localised refolding (area
3, Fig. 2).

GRANITOID ROCKS

Two distinct granites, crop out around the Point
Hicks headland (Fig. 1). One is a muscovite bearing
granite (average grain size 1 mm), found close to the Or¬
dovician sediments; the other is a coarse-grained granite
(up to 10 mm) with euhedral orthoclase phenocrysts (up
to 20 mm). Both granites exhibit a hypidiomorphic-
granular texture and are granites according to the
I.U.G.S. classification with the modal compositions
shown in Table 1. The boundary between the two types
appears to be gradational. The more leucocratic, coarse¬
grained granite contains patches of idiomorphic tour¬
maline (1 to 10 cm) intergrown with the groundmass
phase, especially near the boundary with the muscovite
granite. A late-stage magmatic origin is favoured for
much of the tourmaline as it rarely exhibits replacement
textures.

Jointing is intense within all the granitoids with an
average spacing of 1 m (Fig. 8). Their orientation (Fig.
1 ) corresponds to the regional trend of S 2 structures

Table 1

Modai. and Textural Variation of Point Hicks Granitoid
Rocks

Mineral

coarse-grained

granite

muscovite-bearing

granite

quartz

20

25

K-feldspar

50

50

plagioclase

10

5

biotite

15

5

muscovite

5

15

hornblende

Tr

Tr

accessories: zircon, tourmaline, sericite and apatite

recognised in the Mallacoota region (Wilson et al. 1982)
and this may reflect the stress field and deformation
events occurring during the cooling of the granite. The
joints occur in zones up to 30 m wide, particularly in the
muscovite bearing granite, with more closely spaced
joints (1 to 8 cm) that are often associated with quartz
veins (2 mm to 3 cm). Thicker quartz veins are less com¬
mon, but are found in the muscovite granite and contain
pyrite, sphalerite and chalcopyrite.

MIOCENE CALCAREN1TE

East of the Point Hicks lighthouse the granite is
overlain by a 3 m thick, cross-bedded calcarenite in
which the foresets dip 33° towards 036°. The calcarenite
contains calcareous clasts in a calcite matrix with less
abundant well sorted and subrounded clasts of fine
quartz. The biogenic fraction containing fragments of
brachiopods, molluscs, forams, and bryozoans and, is
unsorted. An isolated exposure of a similar rock also oc¬
curs 3 km to the north in the Pleistocene dunes (Fig. 1).
This calcarenite is comparable to the Late Miocene
“Bairnsdale Limestone” described in other parts of East
Gippsland by Mallett (1977) and Parker (1979) and may
represent a shallow water intertidal or beach deposit.

CONCLUSIONS

Deformation has not modified the sedimentary
features in the Late Ordovician turbidite sequence,
recognized at Cape Everard. Sedimentary structures and
lithofacies suggest that:

1 , the sequence is dominated by interbedded packets of

sandy sediments (channels) with minor finer-grained,

N

-10

- 4

s

Fig. 7 — Palaeocurrent directions using 10° intervals. Measure¬
ment taken from C-division ripple cross-bedding; 38
measurements taken from different beds occurring in all sec¬
tions, except in sections 2 and 9.

GEOLOGY OF CAPE EVERARD

181

Fig. 8 —Aerial view illustrating N-S and E-W joint sets in the coarse-grained granite adjacent to the Point

Hicks lighthouse.

thin-bedded muddy sediments (thinner bedded chan¬
nel levee or fan fringe deposits).

2 , individual sand units were deposited by north flowing
currents in a rapidly changing channelled suprafan.

3, coarse sandstone and shale clasts are directly related
to the flow regime of the turbidity current, being
deposited in or near channels on the channelled
suprafan. Irregular clasts occur in the Bouma A divi¬
sions and a laminated distribution is observed in the
B divisions. Clasts in the C division are either in¬
troduced or more commonly remnant bedding
resulting from slumping of the deposited sediments.
A coarse-grained granite with a less orthoclase- and

more muscovite-rich western margin intrudes the Or¬
dovician sediments at Point Hicks. Minimal contact
metamorphic effects are observed because of paucity of
outcrop and the quartz-rich nature of the adjacent
sediments. Overlying the granite is a calcarenite
deposited during a Miocene marine transgression.

ACKNOWLEDGEMENTS

We wish to thank the National Parks Service for giv¬
ing us permission to work at Cape Everard and
Transport Australia (Victoria region) for admittance to
the lighthouse reserve at Point Hicks. Dr. J. B. Keene is
thanked for his discussions about the sedimentology of
the region and M. W. Fenton is thanked for his com¬
ments on the manuscript. D. D. Campbell is thanked for
assistance with the drafting.

REFERENCES

Bouma, A. H.. 1962. Sedimentology of some flysch deposits. A
graphic approach to facies interpretation. Elsevier,
Amsterdam, I68p.

Douglas, J. G., 1974. Explanatory notes on the Mallacoota
1:250 000 geological map. Kept. geol. Surv. Viet.
1974/6; 1-48.

Eaton, P. C., 1980. The geology of Cape Conran, East Gipps-
land. BSc (Hons) Thesis, Univ. Melb. (unpubl.).

Fenton, M. W., Keene, J. B. & Wilson, C. J. L., 1982.
The sedimentology and environment of the Mallacoota
Beds, Eastern Victoria. J. geol. Soc. Aust. 29: 107-114.

Fry, M. C., 1981. The geology of Point Hicks, Croajingolong,
Victoria. BSc (Hons) Thesis, Univ. Melb. (unpubl.).

Mallett, C. W., 1977. Tertiary foraminifera. PhD Thesis,
Univ. Melb. (unpubl.).

Parker, W. A., 1979. The Bairnsdale Limestone. BSc (Hons),
Univ. Melb. (unpubl.).

Richards, J. R. & Singleton, O. P., 1981. Palaeozoic
Victoria, Australia: Igneous rocks, ages and their inter¬
pretation. ./. geol. Soc. Aust. 28: 395-421.

Rosengren, N., 1978. The physiography of coastal dunes,
East Gippsland, Victoria. MA Thesis, Univ. Melb. (un¬
publ.).

Tessensohn, F., Duphorn, K., Jordan, H., Kleinschmidt,
G., Skinner, D. N. B., Vetter, U., Wright, T. O. &
Wyborn, D., 1981. Geological comparison of base¬
ment units in Northern Victoria Land, Antarctica.
Geol. Jh. B41: 31-88.

Walker, R. G., 1978. Deep water sandstone facies and ancient
submarine fans: models for exploration for
stratigraphic traps. Bull. Am. Assoc. Petrol. Geol. 62:
932-966.

Walker, R. G. & Mum, E., 1973. Turbidiate facies and
facies associations. In Turbidites and deep water
sedimentation, G. V. Middleton & A. H. Bouma, eds.
Soc. Econ. Paleontol. Mineral. (Pacific Section) Short
Course, Anaheim, California, 119-157.

Webby, B. D., VandenBerg, A. H. M., Cooper, R. A.,
Banks, M. R., Burrett, C. F., Henderson, R. A.,
Clarkson, P. D., Hughes, C. P., Laurie, J., Stait,
B., Thomson, M. R. A. & Webers, G. F., 1981. The
Ordovician System in Australia, New Zealand and
Antarctica. Correlation Chart and Explanatory Notes.
Inter. Union Geol. Sci. Publ. 6: 1-64.

Wilson, C. J. L., Harris, L. B. & Richards, A. L.,
1982. Structure of the Mallacoota area, Victoria. J.
geol. Soc. Aust. 29: 91-105.

PROC. R. SOC. VICT. vol. 94, no. 4, 183-220, December 1982

HOLOCENE OSTRACODS, OTHER INVERTEBRATES AND
FISH REMAINS FROM CORES OF FOUR MAAR LAKES IN
SOUTHEASTERN AUSTRALIA

By P. De Deckker

Department of Biogeography and Geomorphology, Australian National University, P.O. Box 4,

Canberra, A.C.T. 2600

Abstract: Cores from 4 maar lakes in Western Victoria yielded ostracods (most abundant),
foraminifers, gastropods, cladocerans, isopods, sponges, chironomids, trichopterans, and some fish re¬
mains; these faunas are described herein. From the present day ecology of most of these organisms,
especially the ostracods, changes in lake levels and salinities are inferred. During the last 9 000 years, fluc¬
tuation in water level and consequently salinity is nearly always synchronous in 3 of the lakes:
Bullenmerri (salinity today 4.5-8.5°/oo), Gnotuk (today 55-63°/oo) and Keilambete (today 55-62°/oo).
Lake Purrumbete (salinity -0.4% o today) has remained fresh during the last 7 000 years probably
resulting from continuous connection to a river.

The following events are inferred as having occurred synchronously in the 3 Lakes Bullenmerri,
Gnotuk and Keilambete and are thought to result from climatic changes since these lakes arc small,
enclosed basins. 1, during the last 100 years, lake levels have fallen significantly; 2, during the 2 000-3 000
yBP period lake levels were low; 3, between 3 800 and 6 400-6 500 yBP water levels were high and the
highest lake levels occurred between 5 700 and approximately 6 400 yBP; 4, before 8 300 yBP water levels
were at their lowest (i.e. highest salinities) in Lakes Keilambete and Gnotuk (no record for Lake
Bullenmerri) for the Holocene.

Core from Lakes Bullenmerri, Gnotuk, Keilambete
and Purrumbete exhibit changes in the fossil lacustrine
fauna. These changes are attributed to changes of lake
salinity which in turn arc attributed to variation in water
levels in the lakes; the water levels indicate fluctuations
in climate during the past 10 000 years. The lakes cover
a broad spectrum of salinities (see Table 1) and cor¬
respondingly have different faunas. Thus the recovery of
different ostracods (the commonest fossil invertebrates)
from the cores, and knowledge of salinity tolerance of
the species today provide palaeosalinity data. With in¬
formation on the present hydrology of each lake, a cor¬
relation of the salinity curve with that of a climatic one
may be attempted as radiocarbon dates (Barton &
Polach 1980, Bowler & Hamada 1971, Dodson 1974),
are available for correlation between the cores.

Lakes Bullenmerri, Gnotuk, Keilambete and Pur¬
rumbete are located near Camperdown, approximately
170 km west southwest of Melbourne and about 30 km
from the sea (Fig. 1). They are situated within the
Western Victorian Newer Volcanic Province which is of
Pliocene to Recent age (Ollier & Joyce 1964) and overlies
Miocene limestone (Joyce 1975).

Each lake occurs inside a shallow volcanic crater
described as a maar. According to Ollier (1968), a maar
is a land form caused by volcanic explosion consisting of
a crater which reaches or extends below, general ground
level; it is considerably wider than it is deep and has a
surrounding rim constructed of material ejected from
the crater.

Physical data on the four lakes are provided in Table
1. Lakes Bullenmerri and Gnotuk are adjacent craters
formed by distinct volcanic explosions (Ollier 1970) and
only once in human memory is Lake Bullenmerri known
to have overflowed into Lake Gnotuk (Currey 1970).

Lake Bullenmerri is clover-leaf shaped, and has steep
sides (Fig. 2). Present water level is 21 m below the point
of overflow into Lake Gnotuk.

Lake Gnotuk is much smaller, oval shaped, and flat-
bottomed (Fig. 2). The water level is about 40 m below
that of Lake Bullenmerri. Circular Lake Keilambete is
flat-bottomed. Salinity varied between 55.3 and
62.4°/oo in the last 15 years (Maddocks 1967, Hussainy
1969a, Bowler 1970).

Lake Purrumbete has steep flanks with a gentle slope
towards its centre (Fig. 2).

Lakes Bullenmerri, Gnotuk and Keilambete are in¬
ternal drainage basins: water in the lakes results from
precipitation and crater slope run-off and in each case
there does not seem to be much interference with
groundwater (Bowler 1970, Currey 1970). An exception
occurs when Lake Bullenmerri overflows into Lake
Gnotuk. Lake Purrumbete can also overflow into Cur-
dies River which is at about the same altitude as the lake
at its present level. Water chemistry of the four lakes has
been studied by Maddocks (1967). The fauna of Lakes
Bullenmerri, Gnotuk and Purrumbete was studied dur¬
ing the 1969-72 period (Timms 1973, 1980, 1981) and the
1967-68 period (Hussainy 1969a). The flora of the four
maars was examined by Yezdani (1970) and Tudor
(1973). The latter concentrated on the diatoms.

Palaeolimnological work has already been carried
out on these four maar lakes. Yezdani (1970) described
changes in the aquatic flora (using pollen and diatoms)
of Lakes Gnotuk and Bullenmerri. Tudor (1973), using
diatoms only, described changes in water quality for
some periods of the history of Lakes Keilambete and
Gnotuk. Bowler (1970, 1981) examined the sediments in
cores from Lake Keilambete and its margins; he
established a water level curve for the last 30 000 years

183

184

P. DE DECKKER

Lake Gnarpurt

Lake

Kariah

Terangpom's^)

Lake Bookar

Lake

Corangamite

Lake Colongulac

Lake Keilambete

iCAMPERDOWN

Lake Werowrap

Lake Gnotuk

TERANG Lake Bullenmerri

Lake Purrumbete

? The
^ Basins

Lake Elingamite

Fig. 1-Map showing the location of the four maar lakes in Victoria; Camperdown is situated 190 km

west of Melbourne.

(Bowler & Hamada 1971, Bowler 1981). Dodson (1974)
presented a palynological curve for the same lake for the
last 10 000 years. His data reveal changes in vegetation
surrounding the lake, accompanied at times, by
modifications in the aquatic vegetation because of
changes in water salinity. Churchill et al. (1978) publish¬
ed a water level curve derived from salinities indicated
by diatom communities for Lakes Bullenmerri and
Gnotuk extending back to 5 500 and 7 500 years respec¬
tively. Barton (1978), Barton & Polach (1980) and Bar¬
ton & McElhinny (1981) have collated a 10 000 year
geomagnetic secular variation record from many cores
of Lakes Bullenmerri, Gnotuk and Keilambete. Finally,
Dodson (1979) presented a pollen record from a core
taken from the deepest part of Lake Bullenmerri and
covering the 8 000 to 16 000 yBP period.

METHODS

A 6 m long pneumatic corer, similar to that designed
by Mackereth (1958) fitted with an orientating device
(Barton & Burden 1979) was used to core each lake. The
54 mm diameter cores were originally taken for
palaeomagnetic investigation of the sediments (Barton
1978). Cores were cut into 1 to 2 m sections to facilitate
transport. Rubber bungs inserted at the ends of most
sections compressed sediment by about 2 cm. The cores
were later split open lengthwise and sedimentological

decription was completed, often under a binocular
microscope. Sampling was carried out by extracting 3 g
of sediments each time. The numbers of each sample for
all cores refer to their appropriate levels in cm below the
top of the core. Each sample was kept in a sealed 200 ml
jar in a 10% hydrogen peroxide solution for one to two
weeks depending on the separation rate of clays and
dissolution of organic matter. The contents of the jar
were then gently washed with a water jet over a 200 /nn
sieve (a finer sieve would have retained valves of uniden¬
tifiable juvenile ostracods). The residue was dried in a
low temperature oven and picked under a binocular

Table 1

Morphometric and Salinity Data on the Four Maars

Bullenmerri

Gnotuk

Keilambete

Purrumbete

Surface area (ha)

448'

208'

21V

522'

Volume (10 6 m 3 )

192 1

32'

13.3 3

157'

Maximum depth
(m)

66', 67 6

18.5', 20 6

11 2 , 10 6

45', 42 5

Mean depth (m)

39.3 1

15.3'

9.54

28.5'

Salinity (TDS Voo)
1979-80

4.49-8.57

55-63

62.4

0.37-0.44

Other periods

7.8-8.5'

56-62'

(Jan. 1980)
55.3\ 61.3 6

0.42-0.50'

1 Timms, 1976; 2 Bowler, 1970; 3 calculated from Bowler, 1970;
Maddocks, 1967; 5 Barton, 1978; 6 Hussainy, 1969a.

HOLOCENE OSTRACODS FROM MAAR LAKES

185

Fig. 2-Bathymetry and location of core sites for each of the four maar lakes.

microscope. Every ostracod was examined and counted.
When specimens were particularly numerous (> 1 500
individuals), an estimate of their number was made. The
presence of other fossil remains was also noted. Conven¬
tional l4 C dates are used throughout this paper.

DESCRIPTION OF THE CORES
Lake Bullenmerri

A core 533.5 cm long was taken on 1 March, 1977 at
a depth of 55.5 m in the northwestern part of the lake

(Fig. 2, location K). The coring site could differ from
this location by about 100 m (C. E. Barton pers.
comm.). The core is labelled BK. Details of the core are
given in Fig. 6. The core consists mainly of fairly
homogeneous organic mud. Two colorations occur:
a, brown to dark brown to grey brown organic mud
with abundant fine (100-200 /*m thick) or occasion¬
ally coarse (ca. 1 mm) laminations. The fine lamina¬
tions are usually black whereas some of the coarse
ones vary from beige brown to orange brown to
white in colour;

186

P. DE DECKKER

LAKE GNOTUK - CORE GH

Fig. 3-Detailed lithological description of core GH from Lake Gontuk. For legend see Fig. 4. Black dots
indicate the position of the samples in the core.

b, dark grey to black unlaminated organic mud below
level 400 cm.

After treatment of some of the samples with H 2 0 2 ,
small grains (>200 /*m) consisting mainly of scoria
fragments were found. Their presence in the core is
referred to in Fig. 9 and their significance will be discuss¬
ed later. The “sulfureous orange muds” described by
Barton (1978) at the bottom of his much longer cores
(ca. 10 m long) are not encountered in core BK.

Lake Gnotuk

A core 362.5 cm long was taken on 7 March, 1977 at
a water depth of 19 m near the centre of the lake. Its
exact location is queried by Barton (1978). The core is
labelled GH. Details of the core are given in Fig. 3.

A variety of sediment types was encountered and

these are described in descending order:

— brown to dark brown organic mud with numerous
white to beige carbonate bands (ca. 1 mm) down to
32 cm. A small hiatus with contorted bedding was
noticeable at level 21 cm.

— dark brown to black organic mud alternating with
light and dark thick layers (ca. 1 cm) with many
ostracod shells (Diacypris compacta) down to 88 cm.
The shells are sometimes so abundant that the layers
have a sandy appearance.

— brown to dark brown to black organic mud with
some pale brown to olive green layers and many very
fine black laminations (ca. 100 /tm) down to 175 cm.

HOLOCENE OSTRACODS FROM MAAR LAKES

187

Fig. 3 — ( continued ).

The average distance between these thin black
laminations ranged between 450 /*m and 600 /an.
dark brown to black organic mud with thin (< 1 mm)
grey or black or brown layers at irregular intervals
down to 231 cm.

between 231 cm and 266 cm, dark brown to black
organic mud with abundant white to grey layers,
especially between level 231 and 250 cm (large
crystals of aragonite occur in some of these light col¬
oured layers)

grey-green organic mud down to 285 cm grading into
grey to dark grey organic mud down to 325 cm. Pale
coloured thin bands more common in the upper
part. A small truncation of bedding was noticeable
around level 295 cm.

— grey to dark grey organic mud down to 333.5 cm. No
sediments recovered between that layer and level 346
cm.

— grey clay from 346 cm to 362 cm. The upper 6 cm
appear to have been mixed. This entire layer is
probably displaced as suggested by the gap above it.

Lake Keilambete

Two cores were taken: a short one, 127 cm long and
labelled KIC was collected on 5 December, 1976, slightly
east of centre of the lake where it is about 10 m deep.
This core was taken to obtain the uppermost layers of
sediments not recovered in the longer core. The latter,
labelled KG and collected on 29 April, 1975, is 419 cm
long and was taken at the centre of the lake at a depth of
10.5 m. For location of both cores, see Fig. 2, and their
lithological description, see Figs 4, 5. The top of core
KG is at about 40 cm below the water sediment interface
(see correlation between the two cores —Table 3 and Figs
4, 5, 8 and 9). Forty-nine samples were examined from
core KIC and 172 from core KG. Bowler (1970, 1981)
provided a detailed stratigraphic log of a 440 cm long
core (labelled K4) from Lake Keilambete. This will not
be repeated here as the core was taken from another part
of the lake with distinct facies differences. Notably, the
grey mud recovered at the bottom of core KG (Fig. 5)
was not found by Bowler in core K4. Additionally, the
sandy layers mentioned by Bowler ( op . c/7.) in his cores
K4 and K5 are not found in core KG. The broad
sedimentary divisions given by Bowler (1981) are used
here:

— top of core to 200 cm (core K4) = Upper Keilambete
Muds consisting of fine grained dark calcareous
muds with paler carbonate rich bands and occasional
sandy horizons (150 cm, 100 cm). This unit ap¬
parently terminates at level 153 cm in core KG and
comprises the whole of core KIC.

— 200 cm to 375 cm (core K4 ) = Lower Keilambete
Muds consisting of fine grained dark muds which are
weakly calcareous. The basal part of this unit cannot
be defined in core KG as the other two units describ¬
ed by Bowler (1981) as the basal saline sands (zone
375-440 cm) and the basal soil (400-440 cm) are miss¬
ing in core KG. It is thought that the basal soil pro¬
bably is facies equivalent to the grey to brown mud
found in core KG below level 350.5 cm and finally
grading into the grey mud below level 385.5 cm.
Throughout the entire core, there are many bands,

sometimes more than 1 cm thick, which consist mainly
of ostracod shells (Fig. 4). Below levels 355 cm and 387
cm in core KG, bedding is disturbed: these are probable
signs of aerial exposure of the lake floor.

Lake Purrumbete

A 581 cm long core, labelled PC, was taken on 2
June, 1975, at a depth of 40.8 cm near the centre of the
lake (Fig. 2). Note the bathymetric map provided by
Barton (1978) differs from Timms’ map (1976). The core
is entirely homogeneous dark brown organic mud. Small
gas vesicles are common in the more fluid upper 113 cm
of the core; below that level, the organic mud becomes

188

P. DE DECKKER

(°>

TOP

@T

7

0 *

OB

t

in
m

c
o

t

CO

(§>

1E-GR

(™>

t®i

LAKE KEILAMBETE - CORE KIC

|@h

**v \

jjGR

BE-GR «

t

BE-GR

I -

BE-GR

f

scale

Icm

BOTTOM

BE Beige

BL Black

BR Brown

GR Grey

0 Olive green

OR Orange

P Purple

Pi Pink

d dark

p pale

A A Coxiella

es Ostracoda

Sandy appearance
due to high numbers
of ostracods

^ Thin laminations

___ White calcareous
layer

___ Calcareous nodules
or crystals

111 No sediment
Rubber bung
• Sample

Fig. 4-Detailed lithological description of core KIC from Lake Keilambete.

HOLOCENE OSTRACODS FROM MAAR LAKES

189

more compact with increasing depth. 195 samples,
usually taken at 3 cm intervals were analyzed from core
PC (Fig. 10).

SYSTEMATICS

Distribution of fossils in the cores is given in Figs
6 - 10 .

OSIRACODA

Ostracods, which have a calcareous shell, are readily
preserved in lake sediments. Their fossils are infor¬
mative on the environmental conditions in the lakes at
the time these ostracods lived. The study of fossil
ostracods from Australian salt lakes is of importance
since the halobiont ostracod fauna is unusually diverse
(De Deckker 198Id) as many species have distinct ranges
of salinity tolerance. In addition, as the ostracod fauna
in salt lakes is represented mostly by planktonic species,
shells of the same species should be fairly evenly
distributed on lake floors. This should permit easy cor¬
relation between cores taken from any part of a lake.

Australocypris robust a De Deckker 1974
Fig. 13M-Y

1974 Australocypris robusta De Deckker, p. 65.
Description: Adult: valves large (ca. 3 mm), oblong,
extremely thin, and smooth to pseudopunctate; greatest
height at about 0.4 from anterior which is broadly
rounded compared to the narrower posterior; ventral
margin almost straight except for the slightly concave
mouth region; inner margin narrow anteriorly and
posteriorly. Juvenile: valves subtriangular to almost
round; greatest height between 0.3 and 0.5 from
anterior.

Ecology: A. robusta is a good swimmer but is also
found on lake floors. Its salinity range in Victoria is
7-145°/oo (De Deckker 1981a) but in South Australia, in
lakes near the Coorong Lagoon, it is 15-38°/oo (De
Deckker & Geddcs 1980). The broader salinity range of
the Victoria specimens probably relates to the fact there
is no other Australocypris species in salt lakes in that
area (i.e. interspecific competition is lacking) whereas in
the lakes near the Coorong Lagoon, 3 additional species
(A. insularis (Chapman 1966), A. rectangularis De
Deckker 1978 and A. clispar De Deckker 1981) probably
have to compete for similar niches. The salinity range of
the present day Victorian specimens is attributed to the
fossil material as, so far, no fossil remains of A. in¬
su laris have yet been found in Victorian lacustrine
sediments. In Victorian lakes, A. robusta is encountered
in high numbers between 45 and 77.5°/oo salinities and
nearly always it co-occurs with large numbers of D.
compacta. At higher salinities, these 2 species are found
with P. baueri.

Remarks: A. robusta specimens are rarely found intact,
especially at the adult stage, because they are very brit¬
tle-fracture of the shell often results from slight com¬
paction even during a very careful extraction of a sample
from a core. Juveniles of A. robusta are distinguished
from both Diacypris species by: the greatest height of A.

robusta being very close to the anterior margin (ca. 1/4),
its shell more rounded and its valves much broader in
dorsal view. No carapaces of A. robusta were ever
found nor large quantities of large specimens typical of
layers occasionally found on the shore of lakes which
dry up. This suggests that the edge of the lake, where
such a phenomenon would occur, has never been near
the coring sites (in Lakes Gnotuk and Keilambete).

Candonocypris novaezelandiae (Baird 1843)

Fig. 12W

Description and Synonymy: See De Deckker (1981b).
Ecology: This freshwater species is common in farm
dams and eutrophic waterbodies. For more details see
De Deckker (1981b). C. novaezelandiae lives today in
Lake Purrumbete (Hussainy 1969, Timms 1981, called
by them C.assimilis) down to 33 m (Timms 1973).
Remarks: Very few specimens were found in the cores.
The adult shells were either partly decalcified or partly
perforated. This w r as probably caused by waters of low
pH. This phenomenon can be expected as C.
novaezelandiae is often crawling in among decaying
vegetal debris.

Diacypris compacta (Herbst 1958)

Fig. 15A-P

1958 Pseudocypris compacta Herbst, p. 181.

1960 Diacypris compacta; Herbst, p. 143.

1981 e Diacypris occidentalis McKenzie; De Deckker p.
54.

1981 e Diacyprisparacompacta McKenzie; De Deckker,
p. 54.

1981 e Diacyprisparva Hartmann; De Deckker, p. 54.

Description: Adult: valves pesudopunctate, almost cir¬
cular in shape and narrow in dorsal view; greatest height
at about 0.3 to 0.5 from anterior margin; posterior area
broadly rounded. Inner lamellae broader anteriorly.
Left valve overlaps right one all along its periphery but
overlap is more obvious in dorsal area where a hump is
present in left valve only. This hump is variable; occa¬
sionally it is pointed. Juvenile: valves more triangular in
lateral view and dorsal hump usually absent.

Ecology: This is an ubiquitous halobiont species; its
salinity range is very broad and it is an excellent swim¬
mer. It occurs in large numbers over its entire salinity
range. In the lakes near the Coorong Lagoon in South
Australia, its range is 8-132°/oo (De Deckker & Geddes
1980) and in western Victorian lakes it is 14-181 °/oo
(one specimen was even collected at 0.34°/oo). In West
Australian lakes, it has been collected in ten lakes in the
range of 2.9-87.9°/oo (Geddes et al. 1981).

As for A. robusta , the salinity range of D. compacta
in Victoria appears to be wider; this probably results
from the absence there of the Diacypris species which
have a higher salinity tolerance as those from the lakes
near the Coorong Lagoon, e.g. D. dictyote , D. fodiens
and D. xvhitei. D. compacta has been found on many
occasions in extremely large numbers (e.g. ca. 20-40 ml
of settled ostracods filtered from 1 m of lake water)
resulting from “blooms” of that species. This phen¬
omenon usually occurred at salinities around 45-77°/oo

190

P. DE DECKKER

LAKE KEILAMBETE - CORE KG

Fig. 5 —Detailed lithological description of core KG from Lake Keilambete. For legend see Fig. 4.

in Victorian lakes and in lakes near the Coorong
Lagoon. In the cores from Lakes Gnotuk and
Keilambete, there are layers (up to 2 cm thick) which
have a sandy texture and which consist mainly of D.
compacta shells. This feature is thought to derive from
such “blooms”. Once D. compacta was found in large
numbers at 124°/oo in Pink Lake in western Victoria; in
the Coorong area, the “bloom” condition extends down
to lower salinities: 21-69°/oo with also two high records
at 96 and 123°/oo (see De Deckker & Geddes 1980).

It is not known whether the wide variations of the
dorsum of D. compacta , which is a diagnostic feature
for the species, is of any ecological sigpificance.

Diacypris dictyote De Deckker 1981
Fig. 14EE-1I, KK

1980 Diacypris n.sp.l De Deckker & Geddes, p. 692.

1981 Diacypris dictyote De Deckker, p. 49
Description: see De Deckker (1981 e).

Ecology: So far this species has only been found living

in South Australia, in many ephemeral lakes near the
Coorong Lagoon, (De Deckker & Geddes 1980), some
on Kangaroo Island and on the Yorke Peninsula. In the
Coorong area, the salinity range of this species over a
year was 12-143°/oo with one record at 195°/oo.
Remarks: D. compacta, which also has a sharp dorsal
“keel” is easily distinguished from this species by its
smooth and smaller shell, the absence of spines and its
narrow shape in dorsal view.

Diacypris dietzi (Herbst 1958)

Fig. 14U-Z, AA, JJ

1958 Pseudocypris dietzi Herbst, p. 177.

1960 Diacypris dietzi', Herbst, p. 143.

Description: Triangular shell in lateral view with
greatest height at about middle; dorsum steeply inclined
and straight behind the highest point of the shell; no
strong overlap of left valve over right one in dorsal area;
valves narrow in dorsal view.

HOLOCENE OSTRACODS FROM MAAR LAKES

191

Ecology: This species is not common in salt lakes: in a
survey of 79 lakes in western Victoria it was collected
only six times whereas D. compacta was collected twenty
times and Australocypris robusta nineteen times. In the
Victorian lakes, its salinity range was 35-I27°/oo with
one specimen collected at 0.34°/oo. In the lakes near the
Coorong Lagoon, its salinity range is broader:
4-141 °/oo with a few specimens found once at 216°/oo
(De Deckker & Geddes 1980). Presence of valves of
fossil D. dietzi in low numbers in the cores cannot sug¬
gest more than a broad range of salinity.

Diacypris whitei (Herbst 1958)

Fig. 14BB-DD.

1958 Pseudocypris whitei Herbst, p. 185.

1968 Diacypris whitei ; Herbst, p. 143.

Description: Smooth, rectangular shell with arched
dorsum and broadly rounded anterior. Shell depressed
dorsally in front of hinge area and in dorsal view, oval in
shape. Greatest height just before mid-length; convex
area of the ventrum behind mid-length. Left valve
slightly larger than right valve all along. Selvage thin and

peripheral on both valves except in the posteroventral
area of right valve.

Ecology: This rare species inhabits highly saline lakes in
which halophytes cannot grow. Its salinity range for this
species in the lakes adjacent to the Coorong Lagoon in
South Australia is 14-195°/oo (De Deckker & Geddes
1980). The low salinity records were taken during winter
when water was plentiful for a short period of time and
consequently water salinity temporarily reduced. This
species has not yet been recorded living in Victoria.
Remarks: Only a few valves of this species have been
found in one sample each from Lakes Keilambete (KG
410) and Gnotuk (GH 315). It is distinguished from the
other Diacypris species by its more rectangular outline
and oval shape in dorsal view.

Ilyocypris australiensis Sars 1889
Fig. 12Z, AA-II

1889 Ilyocypris australiensis Sars, p. 46.

Description: Adult: rectangular and pitted shell occa¬
sionally covered with fine denticles especially along
periphery; 3 main depressions on shell; a central round
one, another round one above it and below the hinge
line and a third one vertically enlogated in front, starting
below the hinge line and ending at mid-height between
the other 2. Greatest height at about 1/4 from anterior.
Hinge adont. Inner lamella broadest anteriorly and
selvage broad all along in both valves. Juvenile: com¬
pared to adults, length height ratio of valves greater and
height of shell of the hinge much greater than at
posterior.

Ecology: /. australiensis occurs in temporary
freshwater pools but it has been found in slightly saline
lakes in Victoria. In the latter, it is usually found at
salinities ranging between 4 and 7°/oo. The uppermost
salinity record of 10.37°/oo is from Lake Kariah.
Remarks: The ornamentation of the shell of /.
australiensis is very variable: the shell can be nearly
smooth, faintly spinose or reticulated all over. For fur¬
ther details, see De Deckker (1981a).

Leptocythere lacustris De Deckker 1981
Fig. 12A-0

1981a Leptocythere lacustris De Deckker, p. 129.
Description: See De Deckker (1981a).

Ecology: L. lacustris is a benthic species which requires
permanent water conditions. Its salinity range is
19-28°/oo with one collection at 2.8%o. Being of
marine ancestry its salinity range probably extends up to
35°/oo; for more detail see De Deckker (1981a).
Remarks: The shell ornamentation of L. lacustris varies
from almost smooth to coarsely reticulated.

Limnocythere dorsosicula De Deckker 1981
Fig. 12Q-X

1981b Limnocythere dorsosicula De Deckker, p. 43.
Description: See De Deckker (1981b).

Ecology: This species is known from four localities:
two in Victoria (Lake Terangpom and South Nerrin
Nerrin Lagoon) and two in New South Wales (Lake
Bathurst and The Morass). Its salinity range is
0.42-3.3°/oo and therefore indicates fresh or slightly

192

P. DE DECKKER

LAKE BULLENMERRI - CORE BK

Fig. 6—Distribution of fossil remains and grains in the upper part of core BK from Lake Bullcnmerri.
Numbers in the ostracod columns are the number of ostracod valves recovered per 3 gm of sediment.
Triangles indicate the position of samples taken from the core. Underlined dates are those which were ob¬
tained from core BK, others were obtained by correlation with other dated cores.

HOLOCENE OSTRACODS FROM MAAR LAKES

193

/

&

^ if

*

■* / 1 #
f / /

•$r i

£ 4

J 4

/ J

0 \

330 -

390 -

395 -

'm.

400 -

490 -

BOTTOM

LAKE BULLENMERRI - CORE BK

Fig. 6 — (continued) Distribution of fossil remains and grains in the lower part of core BK from Lake

Bullenmerri.

Austratocypns

Diacypris compact a Reticypris robust a

194

P. DE DECKKER

* <0% %

•c °

MS. %%%
'Z. V o O

O _ 4>

o o

^.■c> °-
<» 6,0

*M2- 'S'fc

-o

¥

■ *

o o

o CM

CM CM

CM CM

O

UJ < (—
«r rOCM

o

o

Fig. 7 —Distribution of fossil ostracods and other remains in core GH from Lake Gnotuk. Numbers in
the ostracod columns are the number of ostracod valves recovered per 3 gm of sediment. Dots in the other
columns refer to the presence of remains recovered only in low numbers. In Foraminifera column:
E = Elphidium sp., A = Ammonia beccarii, T = Triloculina rotunda. For l4 C dates, refer to the text and
caption of Fig. 6. The underlined dates are those obtained from core GH.

LAKE GNOTUK - CORE GH

HOLOCENE OSTRACODS FROM MAAR LAKES

195

LAKE KEILAMBETE - CORE KIC

20 -

40 -

60 -

*2E

80

100

120 -

130 -

FlG. 8-Distribution of fossil ostracods and other remains in core KIC from Lake Keilambete. Some
remarks as for Fig. 7 except for Foraminifera column.

saline water. Very likely it requires permanent water
conditions.

Mytilocypris praenuncia (Chapman 1936)

Fig. 13A-L

1936 Cypris praenuncia Chapman, p. 298.

1978 Mytilocypris praenuncia; De Deckker, p. 24.
Description: Adult: valves large (ca. 3 mm), smooth,
fairly thin, subtriangular in shape with a steeply inclined
dorsum. Greatest height at about 0.3 from anterior;
posteroventral area of shell broadly curved in lateral
view. Inner lamellae broad anteriorly and posteriorly.
Juvenile: more triangular in shape with posteroventral
area more pointed; distance of greatest height from
anterior between 0.3 and 0.5 of length.

Ecology: M. praenuncia can swim easily but is often
seen on the lake floor or in among beds of halophytic
plants such as Ruppia sp., Lepilaena sp. and the
charophyte Lamprothamniutn papulosum. Its salinity
range varies between 5 and 42°/oo in Victorian waters.
The same species has previously been recorded in the
Coorong area (De Deckker & Geddes 1980) at salinities
between 12 and 35°/oo with one additional record at
43°/oo. It is usually found in samples with P. baueri and
can occur with Diacypris spinosa (salinity range:
5-16°/oo) at lower salinities. As the latter species has
never been found in the cores studied here the records of
the fossil M. praenuncia probably represent the upper
part of the salinity range of the species. The presence of
M. praenuncia indicates waters of much lower salinities
than for A. robusla.

Remarks: M. praenuncia may be distinguished from A.
robusta by its very obvious triangular shape, its broader
inner lamella and its more compressed outline in dorsal
view. Remarks made for large concentrations of shells

of A. robusta on shore lines also apply for this species.

Platycypris baueri Herbst 1957
Fig. 13Z, AA-LL

1957 Platycypris baueri Herbst, p. 217.

Description: Adult: valves smooth, very thin and oval
to rectangular in shape; greatest height at about 0.6
from anterior. Slightly concave ventrum about 0.3 from
anterior. In dorsal view, valves very narrow. Inner
lamella almost non-existent posteriorly and broad only
in the anterodorsal margin. Muscle scar area minute.
Juvenile: oval in largest to almost circular in smallest
specimens. Faint concavity in ventral area also present
0.3 from anterior.

Ecology: P. baueri is a good swimmer but it also bur¬
rows into soft lake sediments. It has the broadest salinity
range of any ostracod found in Australia; in Victorian
lakes it is 9.3-176°/oo (Geddes 1976) and 5-182°/oo (De
Deckker unpubl.) and for Ihc lakes near the Coorong
Lagoon it is 5-195°/oo (De Deckker & Geddes 1980).
This species occurs in low numbers at low salinities and
usually is much more abundant at salinities above ap¬
proximately 70°/oo (Geddes 1976, De Deckker & Ged¬
des 1980). High water salinity of the order of
100±50°/oo is inferred when fossil P . baueri is found
with no other ostracod species or only with A. robusta.
On the other hand when fossil P. baueri is found with
M. praenuncia, water salinity is thought to be of the
range of the latter species, viz. 20-43°/oo.

Remarks: Some specimens of P. baueri were recovered
from the cores with both valves still attached in the
hinge area (Fig. 13FF). This is surprising as valves,
especially for this species, become separated fairly rapid¬
ly after death of the animal. It is suggested that these
fossil carapaces belong to animals which were burrowing

196

P. DE DECKKER

Fig. 9- Distribution of fossil ostracods and other remains in core KG from Lake Keilambete. Some
remarks as for Fig. 7 except that for l4 C dates, refer to Table 3. The underlined date is the only one obtain¬
ed from core KG.

LAKE KEILAMBETE - CORE KG

HOLOCENE OSTRACODS FROM MAAR LAKES

197

X#

# # /

g A.S'

0 ^ g

ff

Xj y. >Jo

4?

V

c?

v <f*

'#

w

/

Jp

0/

TOP

0-_

r snnp

/T'-' i Vr' l
vi\2r
G

A

150-P

250-

.9^

290-

300-

580-

ASE_

BOTTOM

LAKE PURRUMBETE - CORE PC

Fig. 10— Distribution of fossil remains and grains in core PC from Lake Purrumbete. Dots indicate the
position of samples taken from the core. Triangles indicate only the presence of remains and grains in the

core.

198

P. DE DECKKER

Fig. 11-Lake-level curves for Lakes Bullenmerri and Purrumbete, and salinity-level curves for Lakes
Gnotuk and Keilambete postulated from the data obtained from fossil ostracods and other remains.
These curves are compared with the water-level curve for Lake Keilambete of Bowler (1981).

in sediments and remained there until death. These
ostracods arc likely to be of a younger age than the other
organisms found with them in the samples. There is
however no way of controlling this possible discrepancy,
unless signs of bioturbation are observed, and it will be
ignored here as only few carapaces of P. baueri have
been recovered in the cores.

Genus Reticypris McKenzie 1978
The species R. herbsti McKenzie 1978 and R. clava De
Deckker 1981 are easily recognized on anatomical
features but may not be distinguished on shell character
alone. However, their fossils recovered from Lakes
Gnotuk and Keilambete cores have been identified at the
specific level because of their association with other

HOLOCENE OSTRACODS FROM MAAR LAKES

199

ostracod species (see discussion on ecology below). The
descriptive notes refer to both species.

Reticypris spp.

Fig. 14A-T, LL-MM

Description: Adult: shell subrectangular with dorsum
slightly arched; valve reticulated all over except along
periphery; inner lamellae broader anteriorly; greatest
height of shell between 0.3 and 0.5 from anterior. Left
valve larger than right one with often obvious overlap of
left valve in the dorsal area. Juvenile: often shell more
circular to subtriangular in shape; reticulation on shell
more patchy (Fig. 140-Q). In some specimens, ventral
ridge visible on both valves.

Ecology: In collections from Victorian lakes, R. clava
always accompanied low salinity tolerant ostracods such
as Mytilocypris splendida or M. praenuncia and occa¬
sionally Diacvpris spinosa. Salinity for these collections
ranged between 4 and 42°/oo. R. herbsti , on the other
hand, was collected in other lakes in Victoria with D.
compacta at salinities between 99 and 172°/oo.
Therefore, the absence of D. compacta and presence of
either a Mytilocypris species or D. spinosa as fossil with
Reticypris valves should help in identifying R. clava.
The opposite association would point out to the
presence of R. herbsti .

In the lakes near the Coorong Lagoon, data for both
Reticypris species are less clear: the salinity range for R.
clava was 5-131°/oo but it was never found in high
numbers above 68°/oo; for R. herbsti , the salinity range
was 12-141 °/oo with 3 additional records at 195, 216 and
218°/oo (De Deckker & Geddes 1980). The latter species
occurred in high numbers between 104 and 124°/oo.
Apart from the broader ranges for each species, the
salinity values at which each species was found in high
numbers is similar to that for the Victorian specimens.
The latter values from Victoria are used here for the in¬
terpretation of the cores.

Foraminifera

Foraminifers have recently been recognized as com¬
mon benthic inhabitants of salt lakes in Australia (De
Deckker & Geddes 1980, Cann & De Deckker 1981).
Although they may prefer salinities close to that of sea
water they “survive” fluctuating salinities and some even
withstand periods of lake desiccation (Cann & De Deck¬
ker 1981). Their transport into lakes like that for
ostracods is probably by birds (De Deckker 1977).

Some of the foraminifer species found in the cores
can survive periods of desiccation, e.g. Elphidium sp.
sensu Cann & De Deckker (1981) whereas others ap¬
parently cannot as they are only found in permanent
water e.g. Ammonia beccarii and Triloculina rotunda.

Ammonia beccarii (Linne 1758)

Fig. 15W-Z, AA-FF
1758 Nautilus beccarii Linne, p. 710.

1949 Ammonia beccarii ; Frizzell & Keen, p. 106.
Description: Finely perforated, trochospiral text more
convex dorsally; sutures usually thick and smooth; ven-
trally gradation of umbilicus from empty to a plug and
often finely to coarsely spinose. Some aberrant growth

forms (Fig. 15BB, FF) have been found in some
samples.

Ecology: This cosmopolitan species indicates a salinity
close to that of sea water, although it is known to sur¬
vive a broad range of salinities, from 7-67°/oo (Brad¬
shaw 1957). It only reproduces and grows best at
salinities between 20-40°/oo (Bradshaw 1957). At level
362-367 cm in core KG from Lake Keilambete, the large
number of specimens of all sizes of A. beccarii represent
a series of thriving populations and indicates a salinity
similar to that of sea water.

Discorbis sp.

Fig. 15GG

Remarks: One specimen was found at level GH135 in
the Lake Gnotuk core. This species has not been found
in present day salt lakes.

Elphidium sp. sensu Cann & De Deckker 1981
Fig. 15Q-R, U-V

Description: See Cann & De Decker (1981).

Ecology: The salinity range of this species is not yet
known although it is known to “survive” high salinities:
at 88°/oo no pseudopodia were seen protruding from
the test but the same specimens, put in sea water,
became active (Cann & De Deckker 1981).

Triloculina rotunda d’Orbigny 1893
Fig. 15S-T

1893 Triloculina rotunda d’Orbigny, p. 20.
Description: Test triloculine, oval in shape, flattened at
aperture and ellipsoidal in section. Aperture with nar¬
row bifid tooth.

Ecology: This species is rare in the core samples and has
only been collected in a lake with a salinity range of
17-24°/oo and in permanent water. T. rotunda was
never found in the ephemeral Coorong lakes samples,
collected over a year by De Deckker & Geddes (1980).

Mollusca

Only shells (no opercula) of the halobiont gastropod
Coxiella sp. have been recovered in the cores. The
gastropod Potamopyrgus niger (Quoy & Gaimard 1835)
and the bivalve Sphaerium sp., are both found today in
Lake Purrumbete (Timms 1981) and their absence in the
cores has some relevance to the lake histories.

Coxiella sp.

Fig. 17P-Z, DD-EE

Description: Conical to elongate shell with up to 7
whorls; round to oval aperture with broad lip in adults;
extent of umbilicus variable; shell finely ribbed and
sutures deep. Up to 10 mm in length.

Ecology: Species of Coxiella can withstand lake desic¬
cation phases and also survive high salinity ranges by
sealing their aperture with the operculum. No salinity in¬
formation can be obtained from their fossils except that
they exclude fresh water as Coxiella is a halobiont
genus. In lakes below 100°/oo salinity Coxiella spp.
often graze on algal mats or crawl in among halophytic
grasses such as Ruppia sp., Lepilaena sp. and the
charophyte Lamprothamnium papulosum. Live speci¬
mens of Coxiella have never been found at great water

G

200

P. DE DECKKER

depths: in Lake Bullenmerri, Timms (1973, 1981) col¬
lected juveniles (length <5 mm) of Coxie/la striata
(Reeve 1842) down to 25 m and larger specimens of the
same species rarely below 6 m. He also recorded large
numbers of emptied juvenile shells between depths of 12
and 25 m. Consequently, a large concentration of adult
Coxiella shells in cores (e.g. 201.5-202.5 cm in core KG
from Lake Keilambete) very likely indicates that lake
depth was probably less than 25 m and most probably
less than 6 m.

After death of the animals, shells become filled with
gas resulting from body decay and float and are often
blown by the wind onto the lake shore. As a result exten¬
sive layers of Coxiella shells are common on many lake
shores. Such layers should easily be recognized in cores.
Juveniles of Coxiella, on the other hand, have been seen
to float upside down at the surface tension in some
lakes. Their failure to remain near the surface and
subsequent death by sinking to the bottom of the lake,
would explain the presence of few shells of juveniles
found in a number of samples: their occurrence in
sediments in this case adds no information on the depth
of the lake at the time of their death.

Remarks: As pointed out by Mellor (1979) and De
Deckker & Geddes (1980), the taxonomy of all Coxiella
species is in a confused state. Doubt is now placed on
the value of previously considered diagnostic features of
the shell and this is the reason for which no specific iden¬
tification is attempted here. See Fig. 17P-Z for an il¬
lustration of variations in shell morphology.

Timms’ data (1973, 1981) on C. striata found living
in Lake Bullenmerri today are used to interpret the fossil
material. Ecological requirements are likely to be similar
for both fossil and living material.

Chapman (1919) identified the fossil Coxiella
striatula (Menke 1842) from the Pleistocene ( sic Chap¬
man 1919) deposit at Bonco Swamp in Victoria but gave
no consideration to the fact that it is found with other
molluscs, apparently freshwater inhabitants.

The freshwater gastropod Potamopyrgus niger
(Quoy & Gaimard 1835) occurs in Lake Purrumbete
(Timms 1973, 1981) over a wide depth range (0.5-35 m)
but is most common between 1 and 6 m (Timms 1973).
As this gastropod was never recorded in any of the
cores, it appears that especially in the case of Lake Pur¬

rumbete, which is thought to have remained fresh for
the period represented in the entire core, the shore of the
lake has never been close to the coring site, otherwise
shells of P. niger would have been found. This remark
also applies to the freshwater bivalve Sphaerium sp.,
found today in Lake Purrumbete between 0.5 and 22.5
m in depth (Timms 1981) but never recovered in any of
the cores.

Cladocera

Only ephippial sac remains have been recorded from
the cores —cuticular fragments of cladocerans were
noticed in some samples but were not studied further.
Two main types of ephippium were found: one belong¬
ing to the halobiont Daphniopsis pusilla and the other to
the mainly freshwater inhabitant Daphnia spp.

Daphniopsis pusilla Serventy 1929
Fig. 16A-D

1929 Daphniopsis pusilla Serventy, p. 65.

Description of ephippium: Ephippial sac almost rec¬
tangular and asymmetrical: posterior side forming
almost a right angle with small extension of dorsal
chitinous rod (which is occasionally bifid) whereas
anterior side forming an acute angle with the longer and
often bifid chitinous rod. Greatest length of ephippial
sac at about 0.65 from its dorsal side. In dorsal view,
very compressed.

Ecology: The ecology of D. pusilla, which is endemic to
Australia, has been recently reviewed by De Deckker &
Geddes (1980). The salinity range of this cladoceran in
the lakes near the Coorong is 5.8-68.l°/oo, although
few specimens were recorded at the 68.1°/oo. Geddes
(1976) noted that hatching of the species occurred bet¬
ween 4.4 and 33.4°/oo. The presence of ephippial sacs at
a particular level in a core should imply that the lake
water had been at some stage between 4.4 and 33.4°/oo
for that level. On lake floors today, occasional bundles
of 20 or more ephippial sacs of D. pusilla entangled
together by the bifid chitinous rods are found. This
phenomenon was not observed in fossil material.
Remarks: The most diagnostic feature of the ephippium
for this species is the acute angle formed by the anterior
dorsal chitinous rod and the anterior side of the ephip¬
pial sac.

Fig. 12-A-O, Leptocythere lacustris De Deckker 1981. A, LV external, BK 159. B, LV external,
BK 210. C, RV external, KIC 27. D, RV external, BK 210. E, LV external, GH 135. F, RV external,
BK 177. G, RV external, GH 135. H, LV internal, KIC 5. I, LV internal, BK 187. J, LV internal,
BK 210. K, LV internal, BK 210. L, RV internal, BK 187. \1, C dorsal, GH 135. N, C dorsal, GH 135.
O, LV internal, hinge posterior of I. P, LV internal, hinge anterior of I. Q-X, Limnocy theredorsosicula
De Deckker 1981. Q, LV external, male, BK 187. R, LV external, male, BK 171. S, LV external, female,
BK 210. T, LV external, female, BK 206. U, RV external, juvenile (female?), BK 210. V, LV external,
juvenile (male?), BK 210. W, LV external, female, BK 175. X, C dorsal, BK 182.5. Y, Candonocypris
novaezelancliae (Baird 1843), RV internal, juv., BK 17. Z, AA-II, Ilyocypris australiensis Sars 1889, Z,
RV external, BK 187. AA, RV external, BK 187. BB, LV internal, BK 47ft CC, RV external, BK 210.
DD, LV external, BK 210. EE, LV external, BK 472. FF, RV internal, juv., BK 193. GG, LV external,
juv., BK 158. HH, RV internal, juvenile male, BK 206. II, RV internal, juv., BK 189. JJ, RV internal,
juv., BK 193.

Scales: 1-200 /tm for A-N; 2-50 /mi for O-P; 3-200 /on for Q-Y; 4-250 /mi for Z-JJ. Note: BK, GH, KIC,
GH, PC = Core label followed by depth in cm from top of core. C = carapace; LV, RV = left and right

valves; juv. = juvenile.

HOLOCENE OSTRACODS FROM MAAR LAKES

201

202

P. DE DECKKER

HOLOCENE OSTRACODS FROM MAAR LAKES

203

Daphnia spp.

Fig. 16E-H

Description of ephippium: Epphipial sac ellipsoid in
shape and at least twice as long as wide; in dorsal view
narrow to bulbous; external surface sometimes faintly
reticulated. Dorsal chitinous rod longer anteriorly and
forming a right angle with posterior side of the ephippial
sac; posterior angle obtuse. Internal capsule often with a
ridge along its periphery (broadest anteriorly and
posteriorly) and with a vertical groove in the middle
separating the two egg spaces; external surface of cap¬
sule reticulated.

Ecology: No identification at the species level of ephip¬
pial sacs of daphniid species is yet possible. Their
presence, however, indicates a water salinity between
fresh and 5.8°/oo. The upper record refers to a collec¬
tion made in January 1980 from a small lake near Lake
Coragulac in Victoria. Five records of Daphnia spp.
observed in a survey of 79 lakes during the same period,
ranged between 1.90 and 4.91 °/oo. Daphnia carinata
King 1853 was recorded by Geddes et al. (1981) from
three localities in Western Australia where salinity
values ranged between 3.57 and 4.76°/oo. The value of
5.8°/oo is the upper value recorded so far in Australia
for a Daphnia species and it will be regarded as the max¬
imum value for the fossil material studied here.
Remarks: Sars (1885) accurately illustrated the mor¬
phology of the ephippium of D. lutnholtzii Sars 1885 by
providing adequate illustrations of ephippial sacs and
internal egg capsules. These resemble the material
recovered from the Lakes Bullenmerri and Purrumbete
cores, but the latter deflated or partly shrunk after the
drying process prior to picking and preparation for SEM
photography. However, Sars* illustrations show a dou¬
ble row of tiny spines along the dorsal chitinous rod at¬
tached to the ephippial sac. These were rarely seen on
the specimens recovered from the cores (e.g. Fig. 16F).
No further identification has been carried out.

Isopoda

Remains of the aquatic halobiont isopod Haloniscus
searlei have been recovered from a number of core
samples and are described below.

Haloniscus searlei Chilton 1920
Fig. 15HH-PP

1920 Haloniscus searlei Chilton, p. 724.

Description of remains found in the cores:

Cones: Slightly curved and hollow, partly calcareous?
and brittle; external surface consisting of parallel rows
of disconnected and alternating faint and arched
grooves; some with occasional rimmed triangular pores
with two small pores inside and one at base of triangle.
The cones with pores correspond to distal segments of
the posterior appendages of the animal, and those
without pores belong to spines attached to the telson.
Others : These are of various shapes and are illustrated in
Fig. 15KK-PP. The fragment illustrated in Fig. 15MM is
a proximal segment of one of the appendages whereas
fragments illustrated in Fig. 15NN-PP are thought to be
part of ventral parts of the animal’s head. In most cases,
the external surface of these fragments is characterized
by faint grooves similar to those found on the cones
(Fig. 15HH-II).

Ecology: The biology of Haloniscus searlei has been
thoroughly reviewed recently by Williams (in prep.). It is
an Australian endemic oniscoid isopod which is aquatic
and tolerates a high range of water salinities:
3.6-191.7°/oo. It is also known to survive periods of
lake desiccation (De Deckker & Geddes 1980, Williams
in prep.).

Remarks: Similar fragments have been found in other
lacustrine deposits (Pillie Lake in South Australia) (De
Deckker et al. 1982). Their presence cannot provide
much ecological information as H. searlei is found in
both ephemeral and permanent saline lakes. Although
H. searlei occurs in ephemeral lakes, annual rainfall is
necessary each year for the animal to survive as it cannot
survive complete desiccation (Ellis & Williams 1970).
This explains its absence in Central Australian lakes
which are dry for long periods.

Porifera

Only three specimens of the asexual reproductive
bodies of spongillid sponges have been recovered from
the Lake Purrumbete core. These gemmules are all
distinct and will be briefly described below. They all

Fig. I3-A-L, Mytilocypris praenuncia (Chapman 1936). A, LV internal, KG 135.5. B, RV internal,
KG 135.5. C, RV internal, KG 135.5. D, RV external, juv., KG 201.5. E, LV internal, KG 164.5. F, LV
internal, KG 181.5. G, RV internal, KG 181.5. H, RV internal, juv., KG 201.5. I, LV internal, juv.,
KG 135.5. J, LV internal, juv., KG 135.5. K, LV internal, juv., KG 135.5. L, RV external, juv.,
KG 135.5. M-Y, Australocypris robusta De Deckker 1974. M, LV external, juv., KG 380. N, LV inter¬
nal, fragment, GH 62.5. O, LV internal, juv., KG 362. P, RV internal, GH 300.5 (note juvenile
Diacypris sp. inside). Q, LV dorsal, juv., KG 377. R, RV internal, KG 362. S, LV internal, fragment,
KIC 95. T, RV external, juv., KIC 95. U, LV internal, juv., KIC 95. V, LV dorsal, juv., KG 360. W, LV
external, partly broken, juv., KG 360. X, RV external, juv., KG 360. Y, RV internal, partly broken,
KG 360. Z-LL, Platycypris baueri Herbst 1957. Z, LV internal, partly broken, juv., KC» 274. A A, RV in¬
ternal, partly broken, KG 404. BB, LV internal, partly broken, KG 404. CC, RV internal, KG 404. DD,
RV internal, juv., KG 393. EE, LV internal, juv., KG 393. FF, C dorsal, KIC 26. GG, LV external, juv.,
KIC 26. HH, LV external, juv., KIC 26. II, RV external, juv., KIC 26. J J, LV external, ? juv., KIC 26.
KK, LV external, KIC 26. LL, RV internal, anterior detail of CC.

Scales: 1-1 000 /im for A-L, N, P-T, Y; 2-500 urn for M, U, AA-KK; 3-500 for O and 250 / 4 m for V;

4-100 / 4 m for LL.

204

P. DE DECKKER

HOLOCENE OSTRACODS FROM MAAR LAKES

205

belong to the Australasian genus Heterorotula Penney &
Racek 1968. All three indicate freshwater conditions as
none respond to the description of H. capewelli (Bower-
bank 1863) which is known to tolerate athalassic saline
environments from Central Australia to the Dividing
Range (Racek 1969).

Heterorotula nigra (Lendenfeld 1887) (Fig. 16P-R) is
distinguished by the usually granulated gemmoscleral
shafts and the absence of megascleres from the
pneumatic layer. The foramen has not been examined.

Heterorotula multidentata (Weltner 1895) (Fig.
160-T) is recognized here by its smooth to microspined
megascleres and the occasional reinforcement of the
outer coat of the pneumatic layer by megascleres. The
foramen is simple and bears no collar.

The gemmulc of Heterorotula sp. (Fig. 16N,S)
resembles H . capewelli as many megascleres are present
around its periphery. It is distinguished from the latter
species on the following details: the gemoscleres are pre¬
sent only on the outside of the pneumatic layer in H.
capewelli (for comparison see Penney & Racek 1968,
Plate 8, Fig. 5); the megascleres in H. sp. are more pec¬
tinate and the edge of the gemmosclere rotules is spinose
rather than crenulate as in H. capewelli.

Remarks: Few megascleres, cemented on trichopteran
cases were noticed in samples from the Lake Pur-
rumbete core (Fig. 17FF-HH). These belong to
Heterorotula spp., and although megascleres recovered
from lacustrine sediments can provide some palaeo-
ecological information as already demonstrated by
Racek (1966) for Guatemalan material, no similar at¬
tempt was made to systematically recover spongillid
remains.

No remains of Radiospongilla sceptioides (Haswell
1882) have been found in the cores even though the
species has been recorded twice from Lake Purrumbete
(Nov. 1969, May 1970) by Timms (1973, 1981).
Vertebrata

Four types of fish remains were recovered from the
cores: jaw bones, scales, otoliths and other bone
fragments. Only some items of the first type could be
identified and subsequently provide palaeoccological in¬
formation. The other remains in lakes Bullenmerri,
Gnotuk and Keilambete, which are not connected to
rivers indicate permanent water when fish were present.

Two categories of jaws are recognized: those with a
single or a double row of teeth (Fig. 17A-C, F-G) and
those with more than two rows along most of their
length (Fig. 17E, I-J). For their classification see Table
2. Also a few remains bearing teeth (Fig. 17D, H, L) are
identified as being part of the mouth of fishes —they
usually consist of straight rods with fewer teeth and oc¬
casionally have a fiat base (Fig. 17D); others are large
vomer plates covered with many teeth (Fig. 17H). Only
the latter could not be identified at the generic level.

Table 2 lists also the species which are found today in
Lakes Bullenmerri and Purrumbete. In addition, the
hardyhead Craterocehalus stercusmuscarum (Guenther
1867) is also listed as it is a common inhabitant of slight¬
ly saline lakes in Victoria. The following palaeo-
ecological information can be obtained from the fossil
material: jaws with more than two rows of teeth belong
to fish which live in water of salinities between <3 and
13.4°/oo; for the other jaws, the salinity range referred
to lies between <3 and 30.3°/oo.

Although Lake Gnotuk is devoid of fish today very
likely fish would more easily be introduced in it, com¬
pared to Lake Keilambete, as the former lake can receive
overflowing waters from the adjacent less saline Lake
Bullenmerri. The record of two fish species (see Table 2)
from Lake Gnotuk in 1916 probably resulted from the
last overflow recorded in 1841. As the lake level dropped
continuously since 1841 (Currey 1970), salinity would
have progressively increased and eventually fish would
not have been able to survive the highly saline waters
such as those found today. Salinity of Lake Gnotuk was
between 3-13.4°/oo at the time of deposition of level 6.5
cm (see Fig. 7), a period probably following an overflow
from Lake Bullenmerri as jaws with more than two rows
of teeth are found in the samples. Such an overflow
must have occurred on a number of occasions since fish
remains are sparsely distributed in the upper 200 cm of
the Lake Gnotuk core.

Diptera-Chironomidae

Head capsules, mandibles and labia of chironomid
larvae can be useful in tracing past changes of lake pro¬
ductivity as a number of chironomid species are very
sensitive to changes of sediment types and oxygen con¬
centration. The only work dealing with the recovery of
chironomids in Australia is that of Paterson & Walker

Fig. 14 —A-T, LL-MM, Rcticypris. A, RV external, KG 372. B, LV external, KG 372. C, C showing RV,
juv., KG 372. D, C showing RV, KG 372. E, C dorsal, KG 372. F, RV external, KG 372. G, LV external,
KG 351.5. H, LV internal, KG 372. 1, LV internal, KG 351.5. J, RV external, KG 372. K, LV external,
KG 372. L, RV internal, KG 351.5. M, RV internal, KG 351.5. N, C ventral, KG 351.5. O, LV external,
juv., KG 372. P, LV external, juv., KG 372. Q, LV external, juv., KG 372. R, RV internal, juv.,
KG 351.5. S, LV dorsal, KG 351.5. T, RV dorsal, KG 351.5. LL, RV external, anterior detail of J. MM,
RV external, anterior of F. U-Z, AA, JJ, Diacyprisdietzi (Herbst 1958). U, LV internal, KG 393. V, LV
internal, KG 393. W, LV external, KG 398.5. X, LV external, KG 398.5. Y, LV external, KG 398.5. Z,
RV external, KG 398.5. AA, LV dorsal, KG 398.5. JJ, LV external, anterior detail of U. BB-DD,
Diacypris whitei (Herbst 1958). BB, L.V external, KG 410. CC, C dorsal (note valves dislocated), KG 410.
DD, RV external, KG 410. EE-II, KK, Diacypris dictyote De Deckker 1981. EE, RV external, KG 410.
FF, LV external, KG 410. GG, C dorsal, KG 410. HH, LV internal, KG 410. 11, RV internal, KG 410.
KK, LV external, anterior detail of FF.

Scales: 1-20 /xm for A-T; 2-250 for U-Z, AA-1I and 20 /xm for JJ and 10 /xm for KK; 3-50 /xm for LL-

MM.

206

P. DE DECKKER

HOLOCENE OSTRACODS FROM MAAR LAKES

207

Table 2

Native Fishes in the Maar Lakes with Notes on Salinity Tolerance of each Species and on Jaw Morphology

Present day occurrence 1 2

Salinity

tolerance

Type of Jaws

more than

double row

along most vomer

Lake

Lake

Lake Lake

°/oo 2

single or of length plate with

Purrumbete Bullenmerri

Gnotuk Keilambete

double row of jaw many teeth

philypn odon grandiceps

+

+

+ 3

<3-7.3

+

Pseudophrites urvil/i

+

+

+ 3

<3-3.3

+ +

Nannoperca australis

+

<3-3.3

+

Anguilla australis

■o

occidental is

+

3

3

< 3-13.4 s

+ +

Ga/axias maculatus

+

2

<3-30.3

Retropinna victoriae

+

C

c

<3-8.8

Craterocephalus

c

c

stercusmuscarum

< 3-8.8 4

' Timms (1973); 2 Chessman & Williams (1974); 3 Timms (1973) said that these two species were collected once from L. Gnotuk on
8.12.1916-record from National Museum of Victoria; 4 Craterocephalus eyeresii (salinity 3.8-30.9 D / co in Chessman & Williams
(1974) and up to 110°/™ in Glover & Sim (1978) has been recorded in the Murray Darling drainage system in northern Victoria.
This species will be ignored here as it has not been recorded in any of the salt lakes in central Victoria; 5 Record (or A. cf. australis
in Chessman & Williams (1974).

(1974) from Lake Werowrap in Western Victoria. They
recovered head capsules of Procladius paludicola,
Chironoinus duplex and Tanytarsus barbitarsis from
two one metre long cores. The latter two species were
mutually exclusive. T. barbitarsis appears to be
restricted to highly saline waters (in Victoria up to
82°/oo— Paterson & Walker 1974) whereas C. duplex is
an indicator of a freshwater condition, although it has
been found by these authors in Lake Coragulac between
ca. 5.1-5.8°/oo,

Iii the present study, only three head capsules were
found; they could not be identified. They are illustrated
in Fig. 16 I-M. This paucity of material might result
from the treatment of the samples with dilute hydrogen
peroxide which is inappropriate for the recovery of
chironomid remains.

COLEOPTERA

Many fragments of beetles have been recovered but
none have been identified. It is of interest to note that

their occurrence, in the cores of Lakes Gnotuk and
Keilambete, corresponds to the less saline water phases
as extrapolated from other fossil remains such as
ostracods and pollen. No systematic search for col-
eopteran remains was conducted; only large fragments
such as elitra, thoraxes, and more .rarely cephalon
fragments and some appendages were noted.
Trichoptera

Only cases of trichopterans were found in samples
from Lakes Bullenmerri and Purrumbete (Fig. 17 AA-
CC, FF-HH). These most likely belong to the family
Leptoceridae (A. Neboiss pers. comm.). Some lep-
tocerids are good swimmers and therefore their presence
in the samples is not surprising even for those which are
thought to have been deposited in deep water (> 30 m)
far away from the shore. The leptocerid cases found
here cannot be indicative of water quality as some
species also occur in saline waters. It is worth noting,
however, that no leptocerid cases have been found in

Fig. 15-A-P, Diacypris compact^ (Herbst 1958). A, LV internal, GH 62.5. B, LV internal, KG 346. C,
LV internal, GH 361. D, RV internal, KG 338. E, RV internal, juv., GH 62.5. F, LV external, GH 350.
G, LV external, KG 346. H, LV external, KG 338. I, LV external, juv., GH 350. .1, C dorsal, KG 338. K,
C ventral, KG 338. I , RV external, KG 47.5. M, LV external, KG 39.5 (note aragonite crystals). N, RV
external, KG 39.5. O, LV internal, juv., GH 62.5. P, RV external, detail of L. Q-R. U-V, Elphidium sp.
Q, side view, GH 135. R, side view, GH 135. U, side view, K1C 104. V, side view, KIC 104. S-T
Trilocu/ina rotunda d’Orbigny 1893 . S, side view, GH 135. T, apertural view, GH 135. W-Z, AA-FI-’
Ammonia beccarii (Linn* 1758). W, spiral view, KG 362. X, spiral view, KG 362. Y, spiral view, KG 362.
Z, umbilical view, KG 362. AA, umbilical view, KG 362. BB, spiral view, KG 362. CC, spiral view,
KG 362. DD, spiral view, KG 362. EE, spiral view, KG 362. FF, apertural view, KG 362. GG, Discorbis
sp. spiral view, GH 135. HH-PP, Haloniscus searlei Chilton 1920. HH, detail of KK. II, detail of JJ. JJ,
tragment ot spine attached to telson, KIC. KK, fragment of distal segment of a posterior appendage,
GH 29. LL, fragment of distal segment of a posterior appendage, GH 29. MM, fragment of proximal
segment ot an appendage, KIC 90. NN, fragment of ventral portion of cephalon, GH 29. OO, fragment
ot ventral portion of cephalon, GH 29. PP, fragment of ventral portion of cephalon, GH 29.

Scales: 1-200/un for A-O, MM-PP; 2-20/un for Pand40/mi for HH-II; 3-100/mi forQ-Z, AA; 4-200/mi

for KK-LL.

208

P. DE DECKKER

HOLOCENE OSTRACODS FROM MAAR LAKES

209

Lakes Gnotuk and Keilambete even in the samples
representing the less saline phases.

“Unidentified Cones”

Small calcareous cones (average diameter: 50-100
fim) (Fig. 16 U-Z) have been found attached to vegetal
fragments in samples from Lakes Bullenmerri, Gnotuk
and Purrumbete. They have not been identified. They
are not part of fern sporocarps and it is suggested that
they could be of fungal origin (H. Aston pers. com.). It
is interesting to note that they occur in the samples from
Lake Gnotuk where a freshwater phase is postulated.
The presence of these cones in the samples studied here
is recorded in Figs. 6, 7, 10 but will not be discussed fur¬
ther until they are identified.

Diatomophyceae

A few specimens of the large mesohalobic diatom
Campilodiscus sp. were found. A systematic search for
diatoms was not undertaken as they have been dealt with
by Yezdani (1970) and Tudor (1973) for various portions
of cores from the lakes studied here except Purrumbete.

SEQUENCE OF FOSSIL REMAINS FROM THE
CORES AND THEIR PALAEOECOLOGICAL
SIGNIFICANCE

Note that salinity values estimated in this section
represent annual averages, bearing in mind that salinities
for Lakes Gnotuk and Keilambete fluctuate by ca.
8-10°/oo each year, whereas for the less saline Lake
Bullenmerri, the fluctuation is by about 3°/oo.

Lake Bullenmerri

The main components of the various faunas in core
BK are shown on Fig. 6. In addition a few large diatoms
(Campilodiscus sp.), one valve of Diacypris compacta
and one fragment of a mytilocypridinid ostracod were
recovered. Fish remains include jaws, scales and bones;
insect fragments consist mainly of coleopterans, with a
few trichopteran cases.

Major zones are defined in the core on the basis of
faunal assemblages and also on the presence or absence
of some species (Fig. 6). These are examined in descen¬
ding order and water depth refers here to the height of
the water column above the core site.

0-112 cm: Fossil remains are extremely rare indicating
great depth with lake level (and therefore water salinity)
similar to that of today (50-60 m) or even higher. No
ostracod shells are to be found at such depth as there are
no truly planktonic ostracods in Australia which live at a

low salinity with little fluctuation, nor would any ben¬
thic ones be found living at such depth because the lake
would be anoxic. (Although some ostracod species, such
as Diacyprisspinosa and Mylilocypris splendida, inhabit
lakes of low salinity, it appears that they only live in
lakes with broader fluctuating salinities.)

112-144 cm: Fish bones and scales are found in most
samples but with few other fossils. The fish jaw at level
128 cm has more than two rows of teeth so the salinity of
the water was apparently less than 13.4°/oo at the time.
The presence of few Coxiella sp. could indicate a lower
water level as today Coxiella striata is restricted to dep¬
ths less than 25 m in Lake Bullenmerri (Timms 1973,
1981). Their small numbers might only indicate that the
lake level was in the vicinity of 25 m and that this se¬
quence is a transitory one between the deep water one
above and the shallow one below.

144-238 cm: This sequence yields the most diverse
fauna: all the ostracod species recorded throughout the
core are present in it; fish bones and scales are more
abundant than above; Coxiella is present in substantial
numbers in nearly all samples, and a few insect
fragments and the body of a water mite are encountered.
In some samples, fine scoria material and other ter¬
rigenous grains (>200 /tin) are found. These indicate
that the core position was a short distance from the lake
shore and consequently testify to a major drop in lake
level: the water level was less than 25 m because Coxiella
sp. abound. This is confirmed by the diversified ostracod
fauna which consist mainly of benthic animals requiring
oxygenated sediments to live on, except for C.
novaeze/andiae (level 154 cm). Only two specimens of
the latter ostracod were found. Lake salinity during this
period probably fluctuated more, a change supported by
the presence of some species with different salinity
tolerance in the same samples. Salinity fluctuated most
often between 2 and 7°/oo as L. dorsosicula and /.
austra/iensis are the most common species. Salinity may
have gone higher at times (either for some years or dur¬
ing parts of some years) as indicated by the presence of
more salt tolerant species such as P. baueri and L.
lacustris. The presence of the latter species points to per¬
manent water conditions. On two occasions (198-204
cm, 214-226 cm), water levels must have increased as
ostracods and other fossils are either absent or rare.
238-290 cm: Fish and a few insects are present; Coxiella
is rare which in turn indicates a high water level as for se¬
quence 114-145 cm. At level 285 cm, the vomer plate of
an eel ( Anguilla sp.) suggests water salinity of less than

Fig. 16-A-D, Daphniopsispusilla Serventy 1929. A, ephippial sac, GH 87.5. B, ephippial sac, GH 87.5.
C, ephippial sac, KG 92.5. D, ephippial sac, KG 92.5. E-H. Daphnia sp. E, ephippial sac, PC 579. F,
ephippial sac, PC 167. G, internal capsule of ephippial sac, BK 416. H, ephippial sac, PC 137. I-M:
Chironomidae. I, head case, PC 417. J, head case, GH 129. K, head case, GH I. L, head case, detail of
J. M, head case, detail of K. N-R: Porifera. N, Heterotula sp., gemmulc, PC 229. O, Heterotula
multidentata (Weltner 1895), gemmule, PC 235. P, Heterotula nigra (Lendenfeld 1887), gemmulc,
PC 232. Q-R, Heterotula nigra, detail of P. S, Heterotula sp., detail of N. T, Heterotula multidentata ,
detail of N. U-Z, “Unidentified cones”. U, BK 177. V, BK 356. W, BK 194.5. X, BK 356. Y, BK 362. Z,
BK 362.

Scales: 1-200 /tm for A-D; 100 /tm for 1-K, W-X; 40 /mi for Q, S; 20 /tm for R; 400 /mi for Y-Z; 2-200 /tin
for E-H, N-P; 3-40 /tm for L-M, T; 100 /mi for U-V.

210

P. DE DECKKER

HOLOCENE OSTRACODS FROM MAAR LAKES

211

13.4°/oo. Coxiella occurs in large numbers at that
level — water level could have dropped for a short period
of time.

290-346 cm: Fish bones are rare and insect fragments
present in most samples.

346-413 cm: Few fish bones are found, insect fragments
rare. Between 348 and 370 cm “unidentified cones” are
abundant in every sample. The absence of many fossils,
especially at levels 370-413 cm may indicate a general in¬
crease in water level. The lake would then stratify with
the bottom becoming anoxic, stagnant and inhospitable
to organisms such as ostracods. At level 410 cm large
Campilodiscus sp, diatoms were recovered.

413-474 cm: Ostracods are present in some samples;
ephippia of Daphnia spp. occur in nearly all samples
treated (sometimes up to 50 per 3 gm of sediment).
Salinity probably ranged from fresh to 5.8°/oo, remain¬
ing lower than 5.8°/oo for a number of seasons. Some
Daphnia, which are truly planktonic species, at times
tolerate slightly saline waters. In central Victoria, Mar¬
tin Lake was sampled bimonthly for one year and yield¬
ed Daphnia sp., /. australiensis and P. baueri . Water
salinity fluctuated there between 3.25 and 7.92°/oo
although adult Daphnia sp. were only found at
4.83°/oo. Additionally, in the same sample, halobiont
ostracods {sensu De Deckker 1981b) were recorded:
Reticypris clava, M. praenuncia, M. splendida and D.
spinosa. Thus, the lower diversity of ostracod species in
the Lake Bullemerri core samples and the absence of
halobiont species point to a water salinity probably less
than 3°/oo at most times. This would explain the
absence of the halobiont Coxiella in the samples.
Ostracods with daphniid ephippia between 460 and 474
cm suggest a salinity range as in Martin Lake
(3.26-7.92°/oo). Ostracods in many samples indicate a
decrease in water level and this is substantiated by the
presence of scoria fragments (found especially at levels
between 430 and 438 cm where, surprisingly, ostracods
are absent) and other terrigenous grains in some of the
samples. The shore of the lake was, at times, in the
vicinity of the coring site, but it is not possible to be
more precise.

474-535 cm: Fish bones and scales are common in most
samples. The absence of ostracods probably indicates a
high water level associated with lake stratification.
Water remained near fresh at least for level 474-508 cm.

14 C dates from core BK are shown on Fig. 6. Good
correlation exists with other cores 14 C dated by Barton &
Polach (1980) especially between cores BK and BB of
Barton (1978) (with corresponding levels 480 cm of
given age 7 510±490 yBP (ANU-1659) and 525
cm = 8 140± 110 vBP (ANU-1657)). It is assumed that
cores BK and BB cover similar time sequences as they
are located close together (Fig. 2), and that the rate of
sedimentation was fairly uniform through time as shown
for other cores taken in the lake by Barton (1978). This
is further demonstrated as levels BK 475-485 and BB 480
have statistically the same age. It becomes therefore,
possible to estimate the timing of events described for
Lake Bullenmerri. These are summarized below.
(Note: —Comparisons of lake levels here refer to the
height of the water column above the deepest part of the
lake for today and the past. The additional 5 m of
sediments deposited on the lake floor in approximately
the last 8 000 years is not considered in the calculations.)
0.112 cm (0-1 700 yBP): Lake stratification and high
water level similar to that of today or even higher.
112-144 cm (1 700-2 000 yBP): Water level lower than
that of today but more than 35 m and lake stratification.
144-238 cm (2 000-3 600 yBP): Water level less than 25
m and water salinity most often between 2 and 7°/oo
with possible periodic fluctuations to fresh (level 154
cm = 2 250 yBP) and more than 8°/oo (2 800 yBP).
Water level higher at least on two occasions: 3 000 and
3 200-3 400 yBP.

238-290 cm (3 600-4 400 yBP): Water level lower than
that of today with stratification of water. At level 285
cm (4 300 yBP) salinity was below 13.4°/oo and water
level could have been below' 35 m.

290-413 cm (4 400-6 400 yBP): Water level probably
equivalent to that of today and lake stratified-level
probably shallower at level 290-346 cm (4 400-5 250
yBP) and the highest at level 370-413 cm (5 700-6 400
yBP).

413-474 cm (6 400-7 400 yBP): Drop in lake level and at
times, especially between 430-438 cm (6 700-6 800 yBP)
and 460-474 cm (7 100-7 400 yBP), probably of the
order of 35 m or less. Surprisingly, salinity less than
3°/oo at most times.

474-508 cm (7 400-8 000 yBP): Water level probably
similar to that of today or above it and salinity less than
3°/oo; lake stratification present.

Fig. 17 — A-N Fish, A, jaw, BK 189. B, jaw, GH 166.5. C, jaw, fragment of B. D, jaw?, BK 358. E, jaw,
GH 6.5. F, jaw, GH 166.5. G, jaw, BK 114. H, vomer plate Anguilla sp., BK 285. I, jaw, GH 6.5. .1,
jaw, BK 128. K, jaw?. BK 398.5. L, scale BK 117. M, scale BK 126. N, scale GH 117. O, scale, GH 117.
P-Z. DD-EE. Coxiella sp. P, apertural view, BK 169.5. Q, dorsal view, KG 201.5. R, apertural view,
KG 201.5. S, apertural view, BK 189. T, apertural view, KG 201.5. U, apertural view, K 201.5. V, aper¬
tural view, GH 170. W, apertural view, KG 201.5. X, apertural view, BK 126. Y, dorsal view, GH 187.
Z, ventral view, BK 158. DD, detail of X showing gnawing or leaching marks on shell surface. EE, detail
of W showing aragonite crystals on shell surface. AA-CC, FF-HH. Trichoptera. AA, case of leptocerid,
PC 17. BB, case of leptocerid PC 185. CC, case of leptocerid PC 179. FF, case detail of BB (note sponge
megasclere). GG case detail of AA (note sponge megasclere). HH, case detail of AA (note sponge
megasclere).

Scale: 1-400/tm for A-C, F, H; 200 fx m for D; 300 /an for E, I, K; 600 /tm for G, P, W, Z, BB-CC; 30 /an
for FF; 2-200/an for J, L-O, Z-AA; 100 /an for Q-R, T-V, W-X; 50 /an for DD-EE; 20 /an for HH; 3-20
/an for GG.

212

P. DE DECKKER

508-535 cm (8 000-8 700 yBP): Water level probably
similar to that of today with salinity more than 3°/oo
and lake stratification. Therefore the lake level was like¬
ly to have been lower than for period 7 400-8 000 yBP.

Churchill el al. (1978) curve for surface water level at
Lake Bullenmerri between 2 500 and 5 500 yBP cor¬
responds in broad terms with the data presented here
although there is some disagreement regarding the
amplitude of lake level fluctuation. Both works agree
with the lake level having been the highest before 5 500
yBP. My data do not identify the extremely low lake
level around 5 000 yBP shown by Churchill et al. (1978)
but fossil invertebrate data for that interval is poor.

Lake Gnotuk

The lithological units described previously for the
362.5 cm long core correspond to most zones based on
ostracod assemblages apart from one section of the core
(92-230 cm) which yielded very few ostracods (Fig. 7).
Ostracod assemblages are described in descending order.
0-22 cm: A few D. compacta as well as rare Diacypris
juveniles. Insect fragments and Coxiella sp. are common
at some levels. At level 6.5 cm, two jaws with more than
a double row of teeth were found among abundant fish
remains (Fig. 17 A, I). These two indicate a salinity less
than 13.4°/oo. Note that fishes with similar types of
jaws (see Table 2) were recorded from the lake in 1916; it
is likely that these would result from Lake Bullenmerri
overflowing into Lake Gnotuk and consequently would
allow fish to populate both lakes.

22-92 cm: The great abundance of D. compacta results
from “blooms” of that species usually recorded at
salinities between 45 and 77.5°/oo today. (Samples
registering such phenomenon also contained A. robusta
but in lower numbers.) Salinity values for corresponding
fossil material are therefore in the vicinity of
45-77.5°/oo whereas when numbers of D. compacta are
lower, the salinity range should be broadened to that of
when the two species have been found together in some
Victorian lakes at 98-l()0°/oo, and up to 144°/oo.) At
level 82.5 cm D. pusilla ephippia indicate a salinity of
4.4-68°/oo, and the water would have had to go at least
below 33.4°/oo for the cladoceran to hatch.

92-233 cm: This large portion of the core is depauperate
in ostracods. Its upper part, however, is fossiliferous
down to 202 cm whereas it is barren below it. In most
upper samples Coxiella shells are present and even
numerous at times (ca. 10 specimens per 3 gm sediment)
and fish bones are occasionally found. At level 166.5
cm, two jaws are recovered but little information on
salinity can be drawn from them as they only possess
one to two rows of teeth inferring a salinity range of
3-30.3°/oo. The fauna at level 135 cm indicates perma¬
nent saline water conditions in the vicinity of 35°/oo. At
levels 171, 175.5, 182, 190 cm are “unidentified cones”
which are common in samples from Lake Bullenmerri
for which salinity must have been in the vicinity of
2-7°/oo.

233-270 cm: This zone was probably deposited under
similar conditions to those for levels 22-92 cm but A.

robusta valves are rare with only fragments recovered.
For section 245-260 cm, where Coxiella juveniles are
also found, salinity was in the vicinity of 45-77.5°/ 00 .
For the other levels, where D. compacta are found, the
salinity range is <3-182°/oo.

270-317 cm: Reticypris vaives are common in most
samples and are present in great numbers at times. As
explained before, the absence in these samples of the low
salinity ostracods Mytilocypris spp. and D. spinosa War¬
rants the specific identification of R. herbsti for the
specimens found in this core. Salinity of the lake in the
presence of R. herbsti in high numbers (levels 272,
284-300, 313-316 cm) was of the order of 99-172°/oo. At
level 315 cm, the presence of two valves of the highly
saline D. whitei supports the values suggested above.
Salinity was probably lower when R . herbsti numbers
were lower and with A. robusta co-occurring. When the
latter species was common (>200 valves per 3 gm sedi¬
ment) salinity was between 45 and 77.5°/oo. Near level
294 cm, disruption in the bedding resulted from the lake
having dried.

317-333.5 cm: Salinity of the lake must have varied
because of the different associations and variations in
abundance of ostracods. The salinity range was pro¬
bably similar to that of level 270-317 cm. This is sup¬
ported by a collection made once in an unnamed lake
near Lake Bolac where the three species were collected
together at 99.4°/oo.

333.5-346 cm: No sediments.

346-349 cm: Mixed sediments.

350-363.5 cm: R. herbsti , D. compacta and P. baueri are
found together in most samples. These three species
have been found together in various lakes in Victoria at
salinities between 99-172°/oo. A. robusta and D. dietzi
arc poorly represented and occur only in a few samples.
Their presence does not contradict the salinity range
postulated for this zone.

14 C dates from core GH are shown on Fig. 7. The
alternation of light and dark coloured bands with diffus¬
ed carbonate rich layers between 84 and 109 cm in core
GB ends at level 79 cm in core GH. Dodson (1974) iden¬
tified this band in his core from Lake Gnotuk and dated
it between 3 790± 100 yBP and 3 530± 100 yBP. He
suggested that this layer, which he described as being
dolomite-rich, represented a period of low water level
and hypersalinity. The ostracods suggest that salinity
should have between 45 and 77.5°/oo.

The carbonate layer in core GB at level 130 cm could
not be correlated with any layer in core GH but the 10
cm thick layer below (dated at 4 140 ±70 yBP
(ANU-1987)) probably corresponds to level 115-125 cm
in core GH.

The well defined zone characterized in core GB by
black to dark grey mud at 188-210 cm (dated as
5 750 ±70 yBP (ANU-1988)) corresponds to layer
166-191 cm in core GH. Also, the base of the approx¬
imately 17 cm thick layer consisting of white laminae in
dark grey to black organic mud at level 263 cm in core
GB, correlates with level 249 cm in core GH. 10 cm of
this layer above level 260 cm in core GB was dated as

HOLOCENE OSTRACODS FROM MAAR LAKES

213

7 290±100 yBP (ANU-1989). It appears that layer
300.5-301.5 cm (rich in R. herbsti) in core GH cor¬
responds to the one labelled “ostracod layer” in core GB
(at about 316 cm) by Barton (1978). If this is correct, the
date of 9 240±120 yBP (ANU-1990) relates to level
295-305 cm in core GH. This correlation remains uncer¬
tain though as the description by Barton (1978) of that
section of the core does noi mention the conspicuous
alternation of dark and pale layers seen in GH. Finally,
the pale grey layer in core GB below 323 cm is not
recorded in core GH until below level 346 cm (note that
there is a gap of 12 cm above that layer in core GH). The
,4 C date of 7 780±330 yBP (ANU-2487) for level
352-361 cm in core GH suggests that the sediments
recovered in that core below level 334 cm arc either con¬
taminated or displaced.

As it appears that the levels in core GB which are
equivalent to those in core GH, are always on the
average 15 cm above the latter ones in respect to the top
of each core, the top of core GH should be 25 cm below
the water sediment interface, as the top of core GB is
said to be 10 cm below the same interface by Barton
(1978). The results are summarized below:

0-22cm (700-1 200 yBP): Little information available
but probably low salinity (around 10°/oo) most of the
time as halobiont ostracods are rare and insect
fragments are present. At level 6.5 cm (850 yBP) salinity
was below 13.4°/oo.

22-92 cm (1 200-3 000 yBP): Increase in salinity which is
maintained between 45 and 77.5°/oo except on one oc¬
casion at level 82.5 cm (2 900 yBP) when water salinity
had to go below 33.4Voo for a short period of time.
Note that there is some disagreement between Dodson’s
dating for the dolomite-rich layers (between 3 790 yBP
(1-4611) and 3 530 yBP (1-4612) and the date of 3 000
yBP extrapolated from the dates given by Barton &
Polacli (1980) for level 92 cm in GH.

92-233 cm (3 000-7 250 yBP): Return to less saline con¬
ditions and water salinity was probably around 10°/oo.
233-270 cm (7 250-8 250 yBP): Salinity of the water
ranging definitely between 45 and 77.5°/oo for levels
246-250 cm (-1 700 yBP) and probably in the same
range for the remainder.

270-317cm (8 250-9 500 yBP): Salinity of the water fluc¬
tuated; it was often between 99-172°/oo when R. herbsti
was present alone and sometimes between 45-77.5°/oo
when A. robusta was present in high numbers. There is
evidence at level 294 cm (8 900 yBP) of a dry phase
shown by disrupted bedding.

317-335.5 cm (9 500-10 000 yBP): Salinity of the water
fluctuated; it remained constantly high in the vicinity of

100°/oo.

350-363.5 cm No date is available because there is a gap
in the core above this level, and it is thought that this
material could have been reworked, although some of
the fauna ( P. baueri, D. dietzi) is not found elsewhere
in the core. The ,4 C date of 7 780±330 yBP for level
352-361 cm remains unexplained in comparison with
core GB which is presumed to be much older (> 10 000
yBP) by correlation. It will not be considered further.

The plotted curve for corrected annual salt ac¬
cumulation of Churchill et al. (1978) ought to be revised
in the light of Timms’s (1975) remarks on Currey’s
original data (1970) and, since no such correction was
considered in the present work, it is not further discuss¬
ed. The uncorrected water level curve of Churchill et al.
(1978) indicates a number of water level fluctuations not
evidenced by the invertebrate remains. These are: a ma¬
jor drop in water level at 5 000 yBP and around 3 600
yBP, increase in level around 4 500 and 1 700 yBP and
also a fairly high level for the period of 4 000 to 3 000
yBP. For other periods, data from both works are com¬
patible.

Lake Keilambete

The overlapping parts of the upper core KIC with the
top of core KG are described together.

Core KIC (Fig. 8)

0-38 cm: Water salinity lower than that of today.

0-10 cm: The range of R. c/ava in Victoria today is
12-42°/oo, with one rare record at 5°/oo; at salinities
below 17.5°/oo, the species is not found with M.
praenuncia but always accompanied by D. spinosa.
(This species was never recorded in the core.) I infer that
salinity was between 17.5-42°/oo.

10-38 cm: The range of M. praenuncia in Victoria today
(8-43°/oo, with an additional collection with very few
specimens at 5°/oo) is postulated for the water during
this period of sedimentation. It is likely that salinities
below 10°/o o were rarely reached as no low salinity
water inhabitants are present. The absence of P. baueri
in some samples cannot be explained since that species is
tolerant to a broad range of salinities. In addition, the
presence of this species in high numbers at other times
indicates temporary fluctuations to higher salinities
(70°/od) tor levels 10-12 and 25-30 cm. The presence of
M. praenuncia in this level indicates that salinity must
have also gone below 43°/oo at times. At level 27 cm,
one L. lacustris was found indicating permanent water
of salinity range 19-35%o for it. No explanation can be
provided for the poor representation of D. compacta.
38-72 cm: Note that a few quartz grains are found at
level 69 cm in KIC —this level is probably facies
equivalent to the sand lens occurring at level 100 cm in
Bowler’s (1970) core K4. The water level must have been
low at that particular time. The absence of ostracods
suggests the presence of a stratified layer and very
diluted (fresh?) water otherwise saline ostracods would
have been found in the core since there are a number of
planktonic species.

Cores KIC and KG (Figs. 8, 9)

32-140 cm in KG (72-127 cm at least in KIC): This zone,
characterized by the high numbers of D. compacta
(1 500 valves per 3 gm sediment) in nearly all samples,
can be subdivided into a few distinct events as registered
by changes of ostracod species.

32-62 cm in KG (72-101 cm in KIC): D. compacta
“bloom” with salinity of the lake water probably
between 45-77.5°/oo because of the presence of A.
robusta. The salinity range could have fluctuated up to

214

P. DE DECKKER

144°/oo as A. robusta are few in number. At level 49-53
cm in KG (88-91 cm in KIC) D. compacta is less abun¬
dant: the salinity range for the lake water at the time has
to be broadened to 42-145°/oo (it is likely that salinity
did not drop below 42°/oo as M. praenuncia is absent).
Note that fragments of H. searlei are recorded in core
KIC during this short event.

62-140 cm in KG (102-127 cm in KIC—no record
below): Water salinity below that of today for most
times. The salinity range was approximately 19-43°/oo
as D. spinosa and M. splendida (both with a range of
5-18°/oo) are absent. The recorded high numbers of D.
compacta representing “blooms” of that species at
various levels in KG can be explained by temporary ex¬
cursions to high salinities ranging between 45 and
77.5°/oo. At level 104 cm, two Elphidium sp. (Fig. 15
U-V) are found: salinity was probably similar to that of
sea water. Only once, at level 99-102 cm, were high
numbers of M. praenuncia found associated with a D.
compacta “bloom”. As such a phenomenon has never
been recorded in the Victorian lakes today, it is thought
that this level represents two distinct events. The occur¬
rence of the fragile shells of P. baueri in low numbers is
consistent with the salinity values given above since this
animal can be found over a broad range of salinities and
is usually recorded in small numbers below 70°/oo
salinity. At level 81-84 cm in KG (= 114-123 cm in KIC),
few A. robusta valves are found. This event, recorded in
both cores, represents fluctuations to higher salinities
(up to a maximum possible value of 145°/oo for A.
robusta during short periods of time for a phase which
saw salinities remaining generally between 19-43 c /oo
(for M. praenuncia). Some insect fragments are found in
a few samples from both cores. The presence of D.
pusilla ephippia at level 92.5 cm in core KG, indicates a
salinity range of 4.4-68°/oo, with values having to drop
below 33.4°/oo, at least temporarily, for the animal to
hatch. This is consistent with other data as for this level
D. compacta numbers are very low.

Core KG (Fig. 9)

141-280 cm: This zone covers two distinct events:
141-210 cm: Numbers of M. praenuncia fluctuate often
and D. compacta valves are present in most samples but
their numbers are very low (< 10 valves per 3 gm sedi¬
ment). Valves of P. baueri are also found in some
samples. Salinity postulated for this event is of the order
of 19-43°/oo. The low species numbers cannot be ex¬
plained when compared to the zone above, except by
suggesting that salinities were low (20°/oo) and as a con¬
sequence there would be very few D. compacta. Insect
fragments are present in a number of samples. At level

201.5-202.5 cm the conspicuous layer with many Cox-
iella shells also recorded in Bowler’s core K4 (Bow'ler
1970) is considered to represent a phenomenon
registered over most of the lake floor: water depth was
probably less than 6 m because shells of adults are
found. A few quartz grains (> 250 /im) also found in this
layer in core KG confirm this assumption.

210-280 cm: The low numbers of M. praenuncia and P.
baueri probably indicate unfavourable conditions for

both species. The absence of P. baueri between 210 and
245 cm is considered to represent the less saline portion
of this event.

280-323 cm: No data available as no octracods are
recovered except for level 302 cm where one D. compac¬
ta is found.

323-348.5 cm: Fairly high numbers of D. compacta and
subdivided into a series of events:

323-325.5 cm: D. compacta present alone in fairly large
numbers —salinity ranged between 43-182°/oo (the
range of this species is 3-182°/oo in Victorian lakes, but
M. praenuncia is absent here).

325.5- 332 cm: D. compacta and A. robusta co-occur
and both are abundant at times. Salinity range:
28-145°/oo (this corresponds to the present day range of
A. robusta in Victorian lakes when it is found only in
large numbers).

332-336 cm: D. compacta “bloom” (at level 332: 6 000
valves per gm of sediment!!) accompanied by many A.
robusta. Salinity range: 45-77.5°/oo.

336-348.5 cm: D. compacta in fair numbers and A.
robusta, at limes, in high numbers for the species (at
level 346 cm: 720 valves!). Presence of R. herbsti in
small numbers (species identification extrapolated
because of the core and occurrence of A. robusta and D.
compacta as explained before and this remark refers to
all Reticypris specimens found in the samples below
level 342 cm). Salinity range broadened to a maximum
value of 145°/oo ( = upper limit of A. robusta) when A.
robusta is found in high numbers (341.5 cm, 346 cm)
and between 45-77.5°/oo for other times.

348.5- 393 cm: Period of high salinity at most times with
extensive fluctuations: when R. herbsti is the most abun¬
dant species, salinity was about 99-172°/oo (level 351
cm, 377 cm). High numbers of A. robusta (357-362 cm,
375.5 cm) represent a salinity range of 45-77.5°/oo. The
presence of few D. dietzi is consistent with the given
salinity values. When it is found in high numbers (level
390 cm) with R. herbsti and quite a few' A. robusta,
salinity was around 75°/oo. At levels 362-367 cm, a large
quantity of all sizes of A. beccarii indicates permanent
water around 35°/oo. On two occasions (levels 355-357
cm and 387.5-392 cm) the lake dried as shown by the
disturbed bedding.

393-419 cm: Salinity fluctuations and values often very
high. At level 401 cm, P. baueri is numerous and accom¬
panied by many R. herbsti and a few valves of three
other species (D. compacta, D. dietzi and A. robusta).
This association indicates a salinity range of 99-172°/oo.
This is confirmed by the presence of D. dictyote at levels
404 and 410 cm and D. whitei at 410 cm. Level 404 cm
probably experienced a higher salinity (as level 401 cm)
as P. baueri and R. herbsti are numerous. The same
range is extrapolated for level 413 cm when A. robusta
and D. compacta are absent and R. herbsti in smaller
numbers and P. baueri more common than usual. Bet¬
ween levels 401 and 413 cm, salinity probably remained
high as R. herbsti and P. baueri are either abundant or
common in the samples. Salinity was probably lower at
level 417 cm, as A. robusta is recorded with few D. com-

HOLOCENE OSTRACODS FROM MAAR LAKES

215

Table 3

Correlations for Levels of Cores KIC and KG with corresponding ones already l4 C Dated in Cores Studied by Bowler

(1970), Bowler & Hamada (1971),

Dodson (1974) and Barton (1978)

KIC

KG

K4 KF

KJ

Dodson

,4 C date

Lab.

Bowler Barton

(1978)

Dodson Justification of correlation

(1970)

(1974)

610 ± 110

N517

19-30.5

10-20

Marl band at 15-20 cm in K4 = 24-30.5 cm in

KIC.

765 ±135

15245

30.5-35.5

105-110 5 cm below marl band in Dodson = 30.5-35.5 in

KIC.

935 ±110

N518

30.5-42

21-33

Start at 1 cm below marl band —see sample

N517.

1 970±110

N519

63.5-73.5

24.5-34.5

55-65

4 cm below 2 ihin carbonate layers (52 cm in

K4 = 20.5 cm in KG and 60 cm in KIC).

2 410 ±120

N520

90-101

53-63

79-90

4 cm below start of ostracod rich mud and 8 cm

below beige layer in both cores.

2 600 ±110

N521

110-120

72-82

102-112

10 cm band with strong lamination; this level is

compressed in KIC as many distinctive layers
are much thinner than in KG.

2 610 ± 90 ANU2035

72-92

90-110

50-70

2 970±120

N522

85-97

130-140

132-141 in K4 with dark brown to black weakly

calcareous mud = 87-98.5 in KG.

3 500 ± 100 ANU2054

?

141-161

110-120

Cannot be correlated as

3 580 ±125

N523

?

165-175

no diagnostic layer present.

4 200 ±125

N524

151-161

190-202

End of lamination and start of weakly

calcareous mud at 202 in K4= 161 in KG.

4 630 ± 80 ANU2055

-167-187

202-222 160-180

5 cm band of thin lamination at around 210 cm

in KF is probably similar layer at 173-176 cm
in KG.

4 930 ±200

15244

275-280Cannot be correlated as no diagnostic laver.

5 250±135

N525

193.5-204.5 235-245

Shell layer of Coxiella at 242 in K4 corresponds

to 158 in KG.

5 980± 110 ANU2056

?

267-287 225-245

Cannot be correlated as no diagnostic laver in

the middle of black mud of KF.

6 440 ± 145

N526

?

290-300

Cannot be correlated as

6 470 ± 110

16225

7

370-375 no diagnostic layer.

7 850± 165

N527

283-303

335-345*

The 2 carbonate bands ending at 355 cm in

K4 = those ending at 303 in KG.

8 640 ± 80 ANU1807

360-375

390-405

The “striated” layer of Barton (1978) at 389 cm

in KF = the layer with disrupted layering at
355-357 cm in KG.

9 670 ± 135

16226

480-485 Cannot be correlated as no diagnostic laver.

10 190± 90 ANU1808

390-410

420-440

The “striated” layer of Barton (1978) at 420 cm

in KF=the band with disrupted layering at
388.5-392 cm in KG.

14 300 ± 300

N528

395-412

This level (swamp plant debris) is not present in

-

KG.

* Erroneously labelled as 325-345 by Bowler & Hamada (1971).
Layers which are underlined are those which were originally dated.

pacta and R. herbsti, but it cannot be adequately defin¬
ed. Between levels 407 and 419 cm, the lake was pro¬
bably subject to drying up at times, as no lamination is
visible in the grey clay.

Correlation with other cores, which is possible on
lithological grounds, is necessary for the dating of
events in cores KIC and KG here as a number of l4 C
dates associated with cores from Lake Keilambete are
already available (Table 3, Fig. 9).

The results are summarized below:

0-10 cm in KIC (0-300 yBP): Water salinity:
17.5-42°/oo. At level 5 cm, permanent water conditions
and salinity: 19-35°/oo.

10-38 cm in KIC (300-900 yBP): Water salinity:
10-42°/o O ; at level 27 cm, same conditions as for level 6
cm and temporary fluctuations to higher salinities at
10-12 cm (300 yBP) and 24-30 cm (600^-750 yBP).

38-72 cm in KIC (900-2 000 yBP): Little data available.
At level 69 cm in KIC ( = 2 000 yBP) water level must
have been low.

72-101 cm in KIC ( = 32-62 cm in KG) (2 000-2 500
yBP): Water salinity 45-77.5°/oo. For levels 80-101 cm
in KIC ( = 2 250-2 500 yBP) the salinity range has to be
broadened to 42-145°/oo.

62-140 cm in KG (2 500-[3 800-4 000] yBP): Salinity
below that of today and of the order of 19-43°/oo;

HI

216

P. DE DECKKER

possibly with records of higher salinities up to
45-77.5°/oo for levels 99-102 cm (2 900 yBP), 108-112
cm (3 100 yBP), 131-134 cm (3 600 yBP). Also, at levels
81-84 cm (2 600-2 800 yBP), the presence of a few A.
robusta suggests higher salinity: up to a maximum of
145°/oo. At level 92.5 cm (2 800 or 3 000 yBP), salinity
was below 33.4°/oo at least once.

141-210 cm in KG (4 000-5 500 yBP): Salinity between
19.43°/oo but probably around 20°/oo. At level

201.5- 202.5 cm (5 300 yBP) a Coxiella- rich layer
signifies a water level below 6 m.

210-280 cm in KG (5 500-7 200 yBP): Conditions of
slightly saline to near fresh water, at times. Between
210-245 cm (5 500-6 400 yBP), the absence of P. baueri
reflects the less saline portion of this phase. The absence
of low salinity ostracods could be explained by the lake
being stratified.

280-323 cm (7 200-8 300 yBP): The suggestion of the
presence of stratified layer, as for level 210-280 cm, also
applies here.

323-348.5 cm (8 300-9 000 yBP): Water salinity fluctua¬
tions around today’s value.

323-325.5 cm (8 300 yBP): Salinity 43-182°/oo.

325.5.5- 332 cm (8 300-8 500 yBP): Salinity 28-145°/oo.
332-348.5 cm (8 500-9 000 yBP): Salinity 45-77.5°/oo.

348.5- 393 cm (9 000 +vBP): [Note: available l4 C dates
are conflicting and therefore no timing for the various
events is presented here.] Salinity values high and exten¬
sive fluctuations of water level.

351 cm: Salinity 99-172°/oo.

355-357 cm: The lake dried temporarily.

357-362 cm: Salinity 45-77.5°/oo.

362-367 cm: Permanent water — 35°/oo salinity.

375 cm: Salinity 45-77. 5°/oo.

377 cm: Salinity 99-172°/oo.

387.5- 392 cm: The lake dried temporarily sometimes
during that period.

Around 390 cm: Salinity around 75°/oo.

393-419 cm (9 700 yBP): Wide fluctuation of salinity
which was often very high similar to the 348.5-393 cm
section.

401 cm: Salinity 99-172°/oo.

404 cm: Salinity 99-172°/oo.

407-419 cm: Lake subject to drying up.

413 cm: Salinity 99-172°/oo.

The w'ater level curve for Lake Keilambete,
calibrated by l4 C dates, w r as first proposed by Bowler
and Hamada (1971) and later a more detailed version
was produced by Bowler (1981).

It should be noted that Bowler’s (1970) core K4 was
not recovered from the deepest part of the lake w r hereas
Barton’s (1978) cores (those studied here) were taken
near the centre, and therefore yielded different sed¬
iments (than core K4) resulting from changes in lake
levels.

Around 1 300 yBP, an increase in salinity detected in
core K4 is not found in core KG. Between about 900 and
2 000 yBP, core K4 suggests that water level was high.

From 2 000-4 000 yBP, water level fluctuated but re¬
mained generally low. There is an exception around

3 000 yBP. The two peaks of high salinity and low water
level between 2 000 and 3 000 yBP registered in core K4
are also recorded in core KG and the two opposite peaks
(lower salinity and higher water levels correspondingly)
are detected in core KG.

Between 4 000-8 300 yBP core K4 suggests less saline
conditions. Dodson’s (1974) record of Pediastrum
(salinity less than 3.5°/oo) for the period 5 000-6 500
yBP is slightly inconsistent with the ostracod data ob¬
tained in core KG as, for the 4 000-5 500 yBP period,
the extrapolated salinity range is 10-43°/oo with a pro¬
bable lowering of the lake level down to 6 m or less at
about 5 300 yBP (Coxiella- rich layer). It appears there¬
fore, that salinity must have fluctuated at times between
less than 3.5°/oo and more than 19°/oo.

Between 5 500-6 500 yBP freshwater conditions
prevailed most of the time as Pediastrum is abundant
but occasional returns to slightly saline conditions are
necessary to justify the presence of M. praenuncia in
some samples. During the 6 500-7 200 yBP period,
slightly saline conditions must have prevailed at times as
Pediastrum is absent w'hilc P. baueri and M. praenuncia
co-occurred. Botryococcus in these samples indicates
oligotrophic conditions.

Dodson (1974) recorded Ruppia between approx¬
imately 6 900-8 200 yBP (calculated from his diagram)
with the highest value around 7 800 yBP. It is likely that
this phenomenon corresponds to the lowering of the
lake level registered at around 7 900-8 000 yBP by
Bowler (1981).

The highest lake level of Bow'ler (1981) could be ex¬
plained by a period of lake stratification which would
exclude benthic ostracods.

Between 8 300 and approximately 10 000 yBP, the
generally low and fluctuating water level drawn by
Bowler (1981) is in agreement with the ostracod data
especially during the period older than 9 500 yBP which
experienced the highest salinities.

Dodson (1974) discussed the formation of the
creamy yellow band of dolomite at depth 96-103 cm in
his core ( = marl band of Bowler 1970), and concluded
that it represented a dry period in the lake history. In the
corresponding band in core KIC at level 24-30.5 cm,
valves of P. baueri abound and those of M. praenuncia ,
common on either side of this band, are numerically
low. Salinity of the lake must have therefore been high
(above 70°/oo) at times with fluctuations down below
43°/oo. At level 27 cm, water must have been permanent
and of lower salinity (17-35°/oo) as indicated by the
presence of L. lacustris. It is most unlikely then that the
lake dried during the period of the dolomite formation
but salinity could have been high at times.

Lake Purrumbete

No ostracods have been recovered in the 195 samples
taken throughout the core which consists of homog¬
eneous dark brown organic mud. At first glance there is
no indication of the lake having been saline. On the
other hand, most samples yielded daphniid ephippia and
egg capsules (Fig. 10). Their state of preservation was

HOLOCENE OSTRACODS FROM MAAR LAKES

217

often very good as most sacs or capsules were still
swollen after treatment in H 2 0 2 and prior to the drying
of the residues in an oven. It appears, therefore, that
during approximately the last 7 000 years water of Lake
Purrumbete remained either fresh or as the upper salini¬
ty range recorded for Daphnia spp. viz. 5.8°/oo. (Bar¬
ton (pers. comm.) suggested that by using intensity of
magnetization correlation with one of his ,4 C dated core
PD, level around 5 m in core PC studied here is approx¬
imately equivalent to 6 140± 110 yBP). As mentioned
before, the absence of shells of the freshwater gastropod
P. niger and the bivalve Sphaerium sp. in the samples
suggest that the shore of the lake was never close to the
coring site and that the height of the water column
above this site remained higher than 35 m at all times.
This would also explain the absence of the benthic
ostracods Gomphodella australica (Hussainy 1969)
found today in the lake by Timms (1973) in collections
between 0.5-1 m, Candonocypris novaeze/andiae ( = C.
assimilis in Hussainy 1969b, Timms 1973) recorded
down to 33 m by Timms (1973), and of the free swim¬
ming ostracod Newnhamia fenestrata King 1855
inhabiting waters near the shore of the lake at present.

No fluctuation of water level, resulting from changes
of climate, has been registered during the last 7 000
years of the lake history, probably due to a connection
of Lake Purrumbete to the Curdies River which would
have permitted exchange of water and salts.

CONCLUSIONS

The four maar lakes are situated in a subhumid area
close to a semi-arid area today. Any change in evapora¬
tion and/or precipitation in the area is likely to affect
levels of lakes, especially those which have closed
basins, such as maars. Unfortunately, at present a
change of this ratio cannot be properly assessed for a
number of reasons. First of all, it is not possible to plot
an accurate water level curve from known past salinities
for various phases of the lakes as it appears that the
amount of total dissolved salts (TDS) did not remain
constant in all the lakes; the waters of Lakes Bullenmerri
and Gnotuk which now have a similar volume of TDS
(Currey 1970) must have mixed at some stage. Prior to
mixing the TDS volume in Lake Bullenmerri must have
been different as water salinity is thought to have been
between 3.26-7.92°/oo but then water depth is con¬
sidered to have been less than half of today’s between
7 100 and 7 400 yBP. Another example applies to water
depth of Lake Keilambete thought to have been below 6
m at about 5 300 yBP when salinity was between 19 and
45°/oo. It appears that TDS are either lost or introduced
into the lakes by percolation and via the water table.
Finally, it is not possible to assess how much of the TDS
volume is lost periodically by precipitation of salts
especially in Lakes Keilambete and Gnotuk.

Although at present no hydrological budget can be
calculated for the four maar lakes, the synchronous fluc¬
tuations of water levels and salinities, recognized mainly
from fossil ostracod data, in Lakes Bullenmerri, Gnotuk

and Keilambete should inform on climate in central Vic¬
toria during the last 10 000 years. Data for the lakes is
schematized in Fig. 11 and comments are given below. It
is interesting to note, however, that Lakes Gnotuk and
Keilambete, which have similar salinities and faunas to¬
day, registered almost identical changes of ostracod
faunas at most times. Water levels and salinities, are in¬
ferred to have responded to changes of climate. Unfor¬
tunately, it is not yet possible to state how these changes
related to either evaporation or precipitation.

The following sequence of events is deduced from the
foregoing analyses:

(a) During about the last 100 years, lake levels for the
three maars (Lake Purrumbete is not discussed here)
have decreased drastically (Currey 1970; Bowler 1970,
1981).

(b) At 300 yBP and around 600-750 yBP, there were
fluctuations of salinity to higher values in Lake
Keilambete.

(c) At 1 300-1 800 yBP there is a discrepancy for Lake
Gnotuk with high salinity values whereas the other lakes
have a high water level (Keilambete with a suspected
very low salinity).

(d) At around 2 000 yBP a change in water level in
Lakes Keilambete and Bullenmerri is supported by 14 C
dated trees which were drowned (Yezdani 1970, Bowler
1970). [Fig. 11. indicates that water level rose before the
particular tree existed at Lake Keilambete. This
discrepancy is caused by the approximation in dating
events but, after consideration of the limits of error for
,4 C dates, the above statement is still considered valid.]
(c) During the 2 000-3 000 yBP period lake levels were
low in all three lakes.

(0 At about 3 000 yBP there was a change in level in
Lakes Bullenmerri and Gnotuk; it is noticeable a bit
later in Keilambete.

(g) Between 3 000 and 3 600-3 800 yBP lake levels fluc¬
tuated in Bullenmerri and Keilambete.

(h) Between 3 800 and 6 400-6 500 yBP water levels
were high in all three lakes. The highest lake level occur¬
red between 5 700 and approximately 6 400 yBP.

(i) The changes in water levels recorded at about the
same time in Lakes Gnotuk and Bullenmerri before
6 400 yBP are not detected in Lake Keilambete.

(j) Between 7 400 and 8 000 yBP water was high in Lake
Bullenmerri also presumably in Lake Keilambete. It ap¬
pears not to be the case at Lake Gnotuk.

(k) There was a drastic change of water level for Lakes
Keilambete and Gnotuk at 8 300 yBP. This corresponds
to a probable change in level seen by a change of fauna
in Lake Bullenmerri at the same time.

(l) Before 8 300 yBP salinities in Lakes Keilambete and
Gnotuk were the highest ever recorded in the lakes for
the last 10 000 years. Lake Keilambete water level and
salinity seem to have fluctuated more.

Lakes Gnotuk and Keilambete appear to be more
sensitive recorders of “climatic change” since salinity
fluctuated more drastically and frequently there. This is
a direct result of their smaller volume of water and
shallower water depth compared to Lake Bullenmerri.

218

P. DE DECKKER

These 2 lakes (Gnotuk and Keilambete), dried up during
the very arid phase prior to the last 10 000 years and this
would explain the flat bottom topography of each lake
as pedogenesis must have prevailed during that period.
(This phase is already documented for Lake Keilambete
in Bowler and Hamada (1971)). Lake Bullenmerri did
not dry up during that period (Dodson 1979).

It is interesting to note that at times the similar Lakes
Gnotuk and Keilambete did not register identical and
synchronous salinity changes. The total dissolved solids
content of the water of Lake Gnotuk must have changed
fairly drastically after each flooding from Lake
Bullenmerri. Lake Keilambete therefore should prove to
be the most reliable and accurate recorder. However
there are also difficulties in interpreting changes of
salinity for Lake Keilambete since some salts must have
been lost during the high water levels with lake overflow.

ACKNOWLEDGEMENTS

I am grateful to Dr C. E. Barton who provided me
with cores from the 4 maar lakes which he used for his
study of Holocene magnetic stratigraphy. He also made
available a vast amount of unpublished data.

The paper was written during the tenure of a Com¬
monwealth Postgraduate Research Award at the Univer¬
sity of Adelaide under the supervision of Professor W.
D. Williams. Being able to accompany him on many
field trips in western Victoria enabled me to collect
limnological data, especially on the ecology of
ostracods, pertinent to the interpretation of the fossil re¬
mains recovered in the cores.

I also wish to thank Dr J. R. Dodson for offering
comments on a draft of the manuscript and Mr J. Head
who provided useful comments regarding the radio¬
carbon dating.

REFERENCES

Barton, C. E., 1978. Magnetic studies of some Australian lake
sediments. PhD Thesis, Australian National University
(unpubl).

Barton, C. E. & Burden, F., 1979. Modifications to the
Mackereth corer. LimnoL Oceanogr. 24: 977-983.
Barton, C. E. & McElhinny, M. W., 1981. A 10000 yr
geomagnetic secular variation record from three
Australian maars. Geophys. J.R.astr. Soc. 67:
465-485.

Barton, C. E. & Polach, H. A., 1980. ,4 C ages and mag¬
netic stratigraphy in three Australian maars. Radio car¬
bon 22: 728-739.

Bayly, I. A. E. & Williams, W. D., 1966. Chemical and
biological studies on some saline lakes of south-east
Australia. Aust . J. Mar. Freshwat. Res. 17: 177-228.
Bowler, J. M., 1970. Late Quaternary environments: a study
of lakes and associated sediments in south-eastern
Australia. PhD Thesis, Australian National University
(unpubl).

Bowler, J. M,, 1981. Australian salt lakes: a palaeo-
hydrological approach. Hydrobiologia 82: 431-444.
Bowler, J. M. & Hamada, T., 1971. Late Quaternary
stratigraphy and radiocarbon chronology of water level
fluctuation in Lake Keilambete, Victoria. Nature,
Load. 237: 330-332.

Bradshaw, J. S., 1957. Laboratory studies on the rate 0 f
growth of the foraminifer “ Streblus beccarii (Linne)
var. tepida (Cushman)”. ./. Pat eon t. 31: 1138-1147.

Cann, J. H. & De Deckker, P., 1981. Fossil Quaternary
and living Foraminifera from athalassic (non-marine)
saline lakes, southern Australia. J. Paleont. 55 :
660-670.

Chapman, F., 1919. On an ostracod and shell marl 0 f
Pleiostocene age from Boneo Swamp, West of Cape
Schanck, Victoria. Proc. R. Soc. Viet. 32: 24-32.

Chessman, B. C. & Williams, W. D., 1974. Distribution 0 f
fish in inland saline waters in Victoria, Australia. Aust.
J. Mar. Freshwat. Res. 25: 167-172.

Churchill, D. M., Galloway, R. W. & Singh, q.,
1978. Closed lakes and the palaeoclimatic record.
Climatic change and variability , A, B. P. Pittock, l.
A. Frakes, D. Jensen, J. A. Peterson and J. w.
Zilman, eds, Cambridge University Press, Cambridge,
97-108.

Currey, D. T., 1970. Lake systems. Western Victoria. Bull.
Aust. Soc. Limn. 3: 1-13.

De Deckker, P., 1974. Australocypris a new ostracod genus
from Australia. Aust. J. Zool. 22: 91-104.

De Deckker, P., 1977. The distribution of the “giant”
ostracods (family: Cyprididae Baird, 1845) endemic to
Australia. In Aspects of Ecology and Zoogeography of
Recent and Fossil Ostracoda, H. Lofller & D. L.
Danielopol, eds, Junk, The Hague, 285-294.

De Deckker, P., 1978. Comparative morphology and review
of Mytilocypridinid ostracods (Family Cyprididae).
Aust. J. Zool. ( suppl. ser .) 58: 1-62.

De Deckker, P., 1981a. Taxonomy and ecological notes for
some ostracods from Australian inland waters. Trans.
R. Soc. S. Aust. 105: 91-138.

De Deckker, P., 1981b. Ostracoda from Australian inland
waters —Notes on taxonomy and ecology. Proc. R .
Soc. Viet. 93: 43-85.

De Deckker, P., 1981c. Taxonomy, ecology and palaeo-
ecology of ostracods from Australian inland waters.
PhD Thesis, University of Adelaide (unpubl.).

De Deckker, P., 1981 d. Ostracods of athalassic salt lakes: a
review. Hydrobiologia 81: 131-144.

De Deckker, P., 1981e. Taxonomic notes on some Australian
ostracods with description of new species. Zool. Scr.
10: 37-55.

De Deckker, P., Bauld, J. & Burne, R. V., 1982. Pillie
Lake, Eyre Peninsula, South Australia: modern en¬
vironment and biota, dolomite sedimentation, and
Holocene history. Trans. R. Soc. S. Aust. 106:
169-181.

De Deckker, P. & Geddes, M. C., 1980. Seasonal fauna
of ephemeral saline lakes near the Coorong Lagoon,
South Australia. Aust. J. Mar. Freshwat. Res. 31:
677-699.

Dodson, J. R., 1974. Vegetation and climatic history near
Lake Keilambete, Western Victoria. Aust. J. Bot. 22:
709-717.

Dodson, J. R., 1979. Late Pleistocene vegetation and
environments near Lake Bullenmerri, Western Vic¬
toria. Aust. J. Ecol. 4: 419-427.

Ellis, P. & Williams, W. D., 1970. The biology of Haton-
iscussearlei Chilton, an isopod living in Australian salt
lakes. Aust. J. Mar. Freshwat. Res. 21: 51-69.

Geddes, M. C., 1976. Seasonal fauna of some ephemeral
saline waters in western Victoria with particular
reference to Parartemia zietziana Sayce (Crustacea:
Anostraca). Aust. J. Mar. Freshwat. Res. 21: 1-22.

219

HOLOCENE OSTRACODS FROM MAAR LAKES

Geddes, M. C., De Deckker, P., Williams, W. D., Morton,
D. & Topping, M., 1981. On the chemistry and biota of
some saline lakes in Western Australia. Hydrobiologia
82: 201-222.

Glover, C. J. M. & Sim, T. C., 1978. Studies on Central
Australian fishes. A progress report. S. Aust. Nat. 52*
35-44.

Hussainy, S. U., 1969a. Ecological studies on some microbiota
of lakes in western Victoria. PhD Thesis, Monash
University (unpubl.).

Hussainy, S. U., 1969b. Description of the male of Can-
donocypris assimilis G. O. Sars 1894 (Cyprididae,
Ostracoda). Proc. R. Soc. Viet. 82: 305-307.

Joyce, E. B., 1975. Quaternary volcanism and tectonism in
southeastern Australia. Bull. R. Soc. N.Z. 13: 169-176.

Mackereth, F. J. H., 1958. A portable core sampler for lake
deposits. Limnol. Oceanogr. 3: 181-191.

Maddocks, G. E., 1957. The geochemistry of surface waters
of the Western District of Victoria. Aust. J. Mar.
Freshwat. Res. 15: 35-52.

Mellor, M., 1979. A study of the salt lake snail Coxiella
Smith 1894, sensu lato, BSc (Hons) Thesis, University
of Adelaide (unpubl.).

Ollier, C. D., 1968. Maars. Their characteristics, varieties and
definition. Bull. Volcan. 31: 45-73.

Ollier, C. D. & Joyce, E. B., 1964. Volcanic physio¬
graphy of the western plains of Victoria. Proc. R Soc
Viet. 77: 357-376.

Patterson, C. G. & Walker, K. F., 1974. Recent history
of Tany tarsus babitarsis Freeman (Diptera:
Chironomidac) in the sediments of a shallow, saline
lake. Aus. J. Mar. Fresh wat. Res. 25: 315-325.

Penney, J. T. & Racek, A. A., 1968. Comprehensive revis¬
ion of a worldwide collection of freshwater sponges

(Porifera-Spongillidae). Bull. U.S. Natn. Mus. 272:
1-184.

Racek, A. A., 1966. Spicular remains of freshwater sponges.

Mem. Conn. Acad. Arts Sci. 17: 78-83.

Racek, A. A., 1969. The freshwater sponges of Australia
(Porifera: Spongillidae). Aust. J. Mar. Freshwat. Res.
20: 267-310.

Sars, G. O., 1885. On some Australian Cladocera raised from
dried mud. Vid. Selsk. Forh. Christiana 8.

Timms, B. V., 1973. A comparative study of the limnology of
three maar lakes in western Victoria. PhD Thesis,
Monash University (unpubl.).

Timms, B. V., 1975. On the origin of salts in Lakes Bullenmerri
and Gnotuk, western Victoria. Bull. Aust. Soc. Limn.
6 : 5-8.

Timms, B. V., 1976. A comparative study of the limnology of
three maar lakes in western Victoria I. Physiography
and physiochemical features. Aust. J. Mar. Freshwat.
Res. 27: 35-60.

Timms, B. V., 1980. The benthos of Australian lakes. In An
ecological basis for water resource management , W. D.
Williams, ed., ANU Press, Canberra, 23-29.

Timms, B. V., 1981. Animal communities in three Victorian
lakes of different salinity. Hydrobiologia 81: 181-193.
Tudor, E. R., 1973. Hydrological interpretation of diatom
assemblages in 2 Victorian western district crater lakes.
MSc Thesis, University of Melbourne (unpubl.).
Williams, W. D., (in prep.) On the ecology of Haloniscus
searlei (Isopoda, Oniscoidea) an inhabitant of
Australian salt lakes.

Yezdani, G. H., 1970. A study of the Quaternary vegetation
history in the volcanic lakes region of western Victoria.
PhD Thesis, Monash University (unpubl.).

H 2

PROC. R. SOC. VICT. vol. 94, no. 4, 221-247, December 1982

MAMMALS OF SOUTHWESTERN VICTORIA FROM THE
LITTLE DESERT TO THE COAST

By P. W. Menkhorst and C. M. Beardsell

Arthur Rylah Institute for Environmental Research, Fisheries and Wildlife Division, Ministry for
Conservation, 123 Brown Street, Heidelberg, Victoria 3084

Abstract: Mammals in southwestern Victoria were surveyed between 1974 and 1980. Fifty-three
species were recorded during the survey period. A further thirteen taxa have probably disappeared from
the area since European settlement. The results of our survey, together with previously documented
records, are presented as an annotated list of species giving the distribution, abundance and habitat of
each species. The recent history and zoogeography of the mammalian fauna is discussed and the mam¬
malian fauna is considered in terms of the distribution of species across five physiographic regions. Most
species of small mammal were significantly associated with only one region. The results are used to test
the adequacy of the present system of conservation reserves. Most species are adequately catered for.
However, further reservations are necessary to protect the remnants of the mammalian fauna of River
Red Gum woodland, Yellow Gum woodland and Brown Stringybark open-forest.

Some areas in southwestern Victoria have long been
recognized as having high value for the conservation of
flora and fauna (Frankenberg 1971) and two areas, the
Little Desert and the Lower Glenelg forest have been the
subject of land use controversies. Despite this, there
have been no systematic surveys of the mammalian
fauna of southwestern Victoria, apart from the
Grampians-Edenhope area (Emison et al. 1978), and
few published data are available. Wakefield (1974) sum¬
marized data from research collections, the literature
and his field work and highlighted the lack of knowledge
concerning mammals in western Victoria. Land use
reviews by the Land Conservation Council of Victoria
(LCC 1972, 1979, 1981a) demonstrated an urgent need
for greater knowledge of the status of fauna in the area
and inventory surveys of various vertebrate groups were
conducted by the Fisheries and Wildlife Division of Vic¬
toria (FWD) between October 1974 and July 1980. In
this paper we present the results of the mammal survey
of the Little Desert, the LCC’s South West Study Area,
District One and the contiguous area of District Two.

SURVEY AREA

The survey area is bounded to the west by the South
Australian border, to the north by the northern edge of
the Little Desert (36°25'S), to the south by the Southern
Ocean and to the east by 142°E, Glenelg River and
eastern boundary of the LCC’s South West Study Area,
District 1 (Fig. 1). The area so defined (roughly 15 300
km 2 ) is between 180 and 220 km from north to south
and between 40 and 92 km from east to west. About
28% of this area is Crown Land, most of which supports
native vegetation. Figure 1 shows the distribution of
Crown Land and main towns in the survey area.

Physiography and Topography

Five physiographic regions as defined by Hills (1975)
occur in the survey area (Fig. 2). Further details of all
except the Little Desert are given by LCC (1972, 1979,
1981a). The Little Desert consists of marine Cainozoic
deposits overlain by a veneer of aeolian siliceous sands.

In places Cainozoic red sandstone and ironstone project
through the sand sheets forming NW-SE ridges, the
most noticeable being the Lawloit Range. In the west
and sand sheet is broken by large clay fiats. In this
region the altitude ranges from 120 to 200 m and the
boundary betwen the Wimmera Plains and the Little
Desert is well defined by soil and vegetation changes.

Climate

There is considerable variation in climate in the
survey area. The Coastal Plains, Volcanic Plains,
Tablelands and Wimmera Plains (except the Mt Arapiles
area) have a temperate climate (rainfall more than 500
mm per annum, warm dry summers and wet winters
with mild temperatures). The Little Desert and Mt
Arapiles have a semi-arid climate (rainfall less than 500
mm per annum, hot dry summers and only moderately
wet winters).

Rainfall

Isohyets are shown in Fig. 2; average annual rainfall
is greatest near the coast and decreases steadily with in¬
creasing distance inland. At Nhill, on the northern edge
of the survey area, the average annual rainfall and the
winter monthly peaks in rainfall are approximately half
those at Portland at the southern tip of the survey area.
At Portland, the months May to August produce 50%
of the total annual rainfall while further inland rainfall
is more evenly distributed throughout the year (Bureau
of Meteorology 1975).

Temperature

The average daily range of temperatures increases
with distance from the coast. The daily maximum
temperatures for January and February are about 10°C
higher at Nhill than at Portland while the daily
minimum temperatures for June, July and August at
Nhill are less than half those at Portland (Bureau of
Meteorology 1975). Frosts are more frequent and severe
in inland areas and late frosts (September/October) are
more likely to occur as are daily maximum temperatures
greater than 38°C. The number of months per annum

222

P. W. MENKHORST AND C. M. BEARDSELL

Fig. 1 — The survey area showing place names and the distribu¬
tion of Crown Land.

when effective rainfall (the amount of moisture available
for growth of plants after evaporation) is received are:
Portland 11, Casterton 9, and Nhill 7 (LCC 1972), thus
the growing season for plants is longest in the south.

Vegetation

There is considerable variation in vegetation from
north to south as the climate changes from semi-arid to

cool-temperate. We used 15 plant alliances, based on the
classification of Specht et al. (1974, Table 7.3), to
describe the habitats in the areas surveyed. These are
listed below. Further details of the floristic composition,
structure and distribution of each alliance are available
from the authors and some details can be gleaned from
LCC (1972, 1979 and 1981a).

Open-forest

1. Messmate Eucalyptus obliqua

2. Manna Gum E. viminalis

3. Brown Stringybark E. baxteri
Woodland

4. River Red Gum E. camaldulensis

5. Yellow Gum E. leucoxylon

6. Manna Gum
Low open-forest

7. Brown Stringybark
Closed-scrub

8. Scented Paper-bark Melaleuca squarrosa
Open-scrub

9. Broom Honey-myrtle M. uncinata

10. Yellow Mallee Eucalyptus incrassata

11. Coast Wattle Acacia sophorae

Closed-heath

12. Silver Banksia Banksia marginata
Open-heath

13. Silver Banksia

14. Desert Banksia B. ornata
Closed-grassland

15. Blue Tussock Grass Poa poiformis
METHODS

The survey consisted of two parts:

(i) the collation of existing data from museum col¬
lections and the literature. This was carried out at the
National Museum of Victoria (NMV) (Gedye et al. 1979,
Evans & Dixon 1980). In addition we examined field
notes of previous workers and interviewed local
naturalists;

(ii) field surveys to supplement data derived from (i).
Field work took place between October 1974 and

April 1975 (Wimmera Plains and Volcanic Plains) and
October 1978 and July 1980 (Coastal Plains, remaining
areas of Volcanic Plains and Wimmera Plains, and Lit¬
tle Desert) (see Fig. 3). Field trips were held in all mon¬
ths except January and September.

Because of the large area involved (15 300 km 2 ) and
time constraints, a primary survey was designed to pro¬
duce an inventory of terrestial species, their broad
habitat preferences and distribution within a 5' (230 cell)
latitude-longitude grid. Emphasis was on areas of
Crown Land (28% of the survey area) although all in¬
cidental observations on private land were recorded.
Figure 3 shows the temporal and geographical distribu¬
tion of survey effort.

MAMMALS FROM SOUTHWESTERN VICTORIA

223

Cage Trapping

Cage traps were set in 113 cells and in almost all
areas of Crown Land (Fig. 3).

In each 5' cell 20 traps were set in 2 lines of 10, at
each of 4 sites and left in place for 2 or more nights giv¬
ing a minimum of 160 trapnights per cell. Trapping sites
were selected to cover a range of vegetation alliances
within each cell. Traps were cleared each morning and
then reset; they were therefore open day and night. Cap¬

tured animals were identified and sexed, their breeding
condition was assessed and they were either marked and
released or retained as specimens.

Wire cage traps (36x20x 16 cm) were used most of
the time but occasionally Elliott type A traps (Elliott
Scientific Instruments, Upwey, Victoria) were used. Bait
consisted of a mixture of peanut butter, rolled oats and
honey.

Drift Fence Pitfall Trapping

Pitfalls with drift fences were used in 15 grids in san¬
dy areas (Fig. 3). Drift fences consisted of PVC Damp
Course (0.3 m x 60 m) held up by wire or wooden stakes.
Pits consisted of metal tins (20 cm diameter x 28 cm
deep) or cylinders (15-25 cm diameter x 45 cm deep)
sunk into the soil so that the rim was flush with the soil
surface. Eight to ten pits were spaced evenly along the
fence and were straddled by it. Pits were not baited and
were left in place for 10 nights and checked morning and
evening.

Bat Catching

Constantine traps, modified from the design of
Tidemann and Woodside (1978), were set in 22 cells
(Fig. 3) and fine nylon lines were strung just above the
surface of the water of dams (Parnaby 1977) in 4 cells.
Direct Observation

Daylight observations of large mammals (e.g. large
macropods Macropus spp., Koala Phascolarctos
cine reus, Common Wombat Vombatus ursinus, Fox
Vulpes vulpes and European Rabbit Oryctolagus
cuniculus) were made and details of road-killed mam¬
mals were also recorded. Night observations were made
along tracks from a slowly-moving vehicle using 12 volt
spotlights powered by the vehicle battery or on foot us¬
ing portable 6 volt batteries and lights. Spotlighting took
place in 74 cells and in most areas of Crown Land con¬
taining mature open-forest or woodland.

Indirect Evidence

Characteristic diggings of 3 species (Echidna
Tachyglossus aculeatus , Common Wombat and Euro¬
pean Rabbit); scats of 5 species (Common Brushtail
Possum Trichosurus vulpecula, Wombat, Dog Can is
familiaris, Fox and European Rabbit); nests of Com¬
mon Ringtail Possums Pseudocheirus peregrinus and
feeding scars of Yellow-bellied Gliders Petaurus
australis were taken as evidence of the presence of these
species.

Mapping

All field, literature and museum records were
scrutinised and those accepted are included in the
distribution maps presented in Appendix 2. Most
records listed by Gedye et al. (1979) and Evans & Dixon
(1980) have been accepted but a few which have not are
discussed in the Annotated List. All accepted records
from any source during our survey (October 1974 to
November 1980) are shown by closed circles in the ap¬
propriate 5' cell on the distribution maps; those dated
before October 1974 are shown by open circles. Records

224

P. W. MENKHORST AND C. M. BEARDSELL

Fig. 3-Temporal and spatial spread of survey effort. Months, years and 5' cells in which cage trapping
(A), spotlighting (B), bat trapping (*), and pitfall trapping (•) were conducted. Diagonal lines indicate
1974-1975, open stippling indicates 1978, clear indicates 1979, and close stippling indicates 1980.

with imprecise localities are shown as large circles in the
general area of the record.

The distribution maps should be read in conjunction
with Fig. 1, showing the distribution of Crown Land
which represents most of the remaining timbered areas,
and with Fig. 3 showing trapping and spotlighting
coverage during the field survey.

Annotated List of Mammals

Summaries of relative abundance, distribution and
habitats of each species are given in the Annotated List
(Appendix 1). Where appropriate, breeding data,
specimen numbers and taxonomic notes are included.
For species not recorded during our survey the most re¬
cent record is given. All acceptable records of each
species are mapped in Appendix 2.

RESULTS

Fifty species of mammal were recorded during our
survey (41 native, 9 introduced with feral populations),
three others have been recorded in recent years (2 native,

1 introduced) and as many as 12 species and one
subspecies may have become extinct since European set¬
tlement began in the 1830s (see Discussion). Details of

the abundance, distribution and habitat of each species
are presented in the Annotated List.

Forty-nine species have been reported from deposits
of late Pleistocene and Recent skeletal material in caves
from southwestern Victoria and southeastern South
Australia. About half of these species are no longer pre¬
sent in the survey area (Table 1).

Cage Trapping

A total of 17 440 trap nights in 113 5' cells yielded
2045 captures of 20 species of mammals at an overall
trapping success rate of 12%. Trapping results for each
species are presented in Table 2.

Five species accounted for over 75% of captures
(Bush Rat Rattusfuscipes, House Mouse Musmusculus,
Swamp Rat Rattus lutreolus, Silky Mouse Pseudomys
apodemoides and Yellow-footed Antechinus Ante-
chinus flavipes) with the Bush Rat accounting for over
33% of all captures.

Trapping rates and diversity of trapped species were
highest near the coast and lowest in the Little Desert
(Table 3). The latter also produced the lowest trapping
rate (2%) of any area surveyed by FWD (unpublished
data).

MAMMALS FROM SOUTHWESTERN VICTORIA

225

Table 1

Species Represented in Sub-fossil Bone Deposits from Western Victoria and Southeastern South Australia

SPECIES

DASYURIDAE
Antechinus flavipes
A. stuartii
A. swainsonii
Dasyurus maculatus
*D, viverzinus
Phascogale tapoatafa
*Sarcophilus harrisii
Sminthopsis crassicavdata
S. murina
S. leucopus

*Tbylacinus cynocephalus

PERAMELIDAE

Isoodon obesulus

*Perameles gunnii

*P. bougainville

*P. nasuta

PHALANGERIDAE

Trichosurus vulpecula

BURRAMYIDAE

Acrobates pygmaeus

Cercartetus nanus

PETAURIDAE

Petaurus breviceps

*P . norfolcensis

Pseudocbeirus peregrinus

MACROPODIDAE

*Aepyprymnus rufescens

*Bettongia gaimardi

*B. lesueur

*B. penicillata

*Lagorchestes leporides

Macropus giganteus/fuliginosus

*M. gzeyi

M. rufogriseus

*Onychogalea fraenata

*0. unguifera

*Petrogale penicillata

*Potorous platyops

P. tridactylus

*Thylogale billardierii

*Wallabia bicolor

PHASCOLARCTIDAE

Pbascolarctos cinereus

VOMBATIDAE

Vombatus ursinus

MURIDAE

*Conilurus albipes
Hydromys chrysogaster
*Mastacomys fuscus
*Pseudomys australis
*P. fumeus

*P . cf. gracilicaudatus
*P. cf. novaehollandiae
P. shortridgei
*P. sp. nov.

Rattus fuscipes greyi
R. lutreolus

• present in bone deposits

* no longer present in the survey area

1, Fern Cave (Wakefield 1964). 2, Mt Eccles (Wakefield 1964). 3, Byaduck Caves (Wakefield 1964). 4, Ml Hamilton (Wakefield
1964). 5, Bushfield (Wakefield 1964). 6, Tower Hill Beach (Wakefield 1964). 7, Mt Porndon (Wakefield 1964). 8, McEacherns Cave
(Wakefield 1967). 9, Victoria Range (Wakefield 1963). 10, Black Range (Wakefield 1963). 11, Mt Arapiles (Wakefield 1971). 12,
Yallum Cave, Penola (Tidemann 1967). 13, Tantanoola Cave (Tidemann 1967). 14, Bat Cave, Naracoortc (Tidcmann 1967). 15,
Wombat and Cave Park Caves (Maddock 1971). 16, Victoria Cave, Naracoorte (Smith 1971 and 1972).

226

P. W. MENKHORST AND C. M. BEARDSELL

Table 2

Results of Trapping Small Mammals in Southwestern Victoria

Specles

Total No.

trapped

% of total

catch

Cel 1 frequency

(*)

n=113

No. of

trapping sites

n=396

Mean trapping success

100 trapnights where

captured (range)

Rattus fuscipes

706

34

46 (41)

121

14 (2-60)

Mus musculus

378

18

60 (53)

104

7 (1-25)

Rattus lutreolus

275

13

37 (33)

68

9 (1-43)

Pseudomys apodemoides

130

6

36 (32)

71

4 (2-13)

Anteehinus flavipes

120

6

35 (31)

55

4 (0.5-25)

Pseudomys shortridgei

92

5

24 (21)

41

6 (2-20)

Anteehinus swainsonii

69

3

15 (13)

29

5 (1-23)

A. stuartii

67

3

11 (10)

28

5 (1-20)

A. minimus

60

3

15 (13)

22

6 (3-20)

Isoodon ohesulus

52

3

29 (26)

34

4 (1-15)

Trichosurus vulpeoula

37

2

19 (17)

22

3 (1-8)

Rattus rattus

22

1

6 (5)

6

5 (3-8)

Potorous tridactylus

19

1

11 (10)

13

4 (3-8)

Hydromys chrysogaster

8

0.4

2 (2)

2

9 (5-15)

Pseudoaheirus peregrinus

3

0. 1

3 (3)

3

2

Dasyurus maculatus

2

0. 1

1 (1)

2

1

Oryctolagus ouniculus

2

0. 1

2 (2)

2

1 (1-3)

Sminthopsis leueopus

1

0. 1

1 (1)

1

2

S. murina

1

0. 1

1 (1 )

1

2

Cercartetus nanus

1

0. 1

1 (1)

1

2

Total 2045

Pitfall Trapping

Pitfalls yielded only 14 mammals, 9 of which were
House Mice. However, these traps caught 2 species not
captured in wire cage traps; Fat-tailed Dunnart Smin-
thopsis crassicaudata and Western Pygmy-possum Cer-
cartetus concinnus.

Bat Catching

One hundred and forty-seven bats of 11 species were
captured (Table 4). Constantine traps captured 117
(79%) of these but trip wires over dams captured 3
species not captured in Constantine traps; While-striped
Mastiff-bat Tadarida australis , Western Broad-nosed

MAMMALS FROM SOUTHWESTERN VICTORIA

227

Table 3

Small Mammal Trapping Effort and Success in Each Physiographic Region
Includes only species comprising > 1% of total catch.

Little Desert

Wimmera

Plains

Tablelands Coastal Plains

Volcanic Plains

Overall

Number of trapnights

5470

3076

1502

5240

2150

17440

Number of trapping sites

127

62

28

135

44

396

Total captures

117

308

81

929

593

2028

Percentage of captures comprised by each species within each physiographic region.

Antechinus flavipes

0

1.3

0.5

1.3

0.1

A. minimus

0

0

0

1.1

0.1

A. stuartii

0

0

0

0.2

2.4

A. swainsonii

0

0

0

0.2

2.8

Isoodon obesulus

0

0.1

0.2

0.8

0.2

Trichosurus vulpecula

0.02

0.6

0

0.3

0.1

Potorous tridactylus

0

0

0

0.1

0.6

Mus muscu/us

0.6

5.5

2.5

2.0

1.7

Pseudomys apodemoides

1.5

1.4

0.1

0

0

P. shortridgei

0

0

0

1.5

0.7

Rattus fuscipes

0

0.2

1.0

6.0

17.6

R. lutreolus

0

0.4

1.1

4.0

1.6

R. rattus

0

0.5

0

0.1

0.1

Total trapping success (%)

2.1

10.0

5.4

17.7

28

12

Bat Nycticeius balstoni and Great Pipistrelle Pipistrel/us
tasmaniensis . Eptesicus spp. comprised 69% of all bats
captured, followed by Chocolate Wattled Bat Chalin-
olobus mono (8%) and Lesser Long-eared Bat Nyc-
tophilus geoffroyi (7%). The small number of sites at
which bat trapping took place and variation in bat
activity patterns in different seasons and weather condi¬
tions preclude a more detailed analysis of trapping
results.

Spotlighting

Results of spotlighting arboreal mammals are given
in Table 5. Three hundred and twenty-two sightings in¬
volving 5 species were made. More than half of these
were Common Ringtail Possums although the species
was observed in only 36% of cells in which spotlighting
took place. Common Brushtail Possums were the most
widespread arboreal mammal, being found in 44% of
cells where spotlighting occurred.

Observations of arboreal species could not be quan¬
tified because of variation in ease of observation in
different habitats, in different weather and by different
observers. However, species richness of arboreal mam¬
mals ranged from 3 species in the Little Desert (Western
Pygmy-possum, Sugar Glider Petaurus breviceps, and
Common Brushtail Possum), to 7 in the open-forests of
both the Coastal Plains and the Volcanic Plains (Com¬
mon Brushtail Possum, Eastern Pygmy-possum Cer-
cartetus nanus , Feathertail Glider Acrobates pygmaeus ,
Sugar Glider, Yellow-bellied Glider, Common Ringtail
Possum, and Koala). Records of species of Burratnyidac
were too few to allow comment, However, all other ar¬
boreal species are probably common in suitable habitat.
Zoogeography

Mammal species present in the broad physiographic
regions within the survey area and the food niches they

occupy are shown in Table 6. The number of species in
the Little Desert was about half that on the Coastal and
Volcanic Plains. Greatest differences were in the small
ground carnivore and large arboreal herbivore
categories (Table 6).

Fig. 4 —Comparative trapping rates of each species within each
physiographic region. Figures from Table 3 are transformed to
a scale of I to 100. Species arc arranged to show associations of
small mammals occurring in each region but trapping rates are
not directly comparable between species. P.a .-Pseudomys
apodemoides, M.m .-Mus muscuhts , R.v. —Rattus ran us,
T.v. — Trichosurus vulpecula , A A'.—Antechinus flavipes,
I .o .-Isoodon obesu/us , RA.—Rattus lutreolus ,
A. m. - A n techinus minim us , P. s. - Pseudomys shortridgei,
R.f. —Rattus fuscipes, P.t. — Potorous tridactylus ,
A.st. —Antechinus stuartii, A.sw.— Antechinus swainsonii.

228

P. W. MENKHORST AND C. M. BEARDSELL

Table 4

Results of Trapping Bats in Southwestern Victoria

For each species the numbers captured by each method, the percentage frequency in 5' cells sampled and, for Constantine traps, the

mean trapping success per night are given.

Species

Number Captured ( a /o)
Constantine Lines over
traps dams

Percentage of
total catch
//= 147

Cell frequency
n = 25

Mean trapping success per night
where captured (range).
Constantine traps only

Tadarida australis

_

3 (10)

2

12

-

T. planiceps

3 (3)

2 (7)

3

16

2 (1-2)

Chalinolobus gou/dii

3 (3)

3 (10)

4

16

1 (0.3-1)

C. morio

11 (9)

1 (3)

8

28

1 (0.5-1)

Eptesieus regulus

8 (7)

-

5

12

3 (2-3)

E. sagit tula

24 (21)

10 (33)

23

20

6(1-8)

£*. vulturnus

54 (46)

6(20)

41

60

2 (0.2-8)

Min i op ter is sch reibersii

2 (2)

-

1

4

2

My otis ad versus

2 (2)

—

1

4

2

Nycticeius balstoni

—

1 (3)

1

4

—

Nyctophilus geojfroyi

10 (9)

-

7

24

1 (0.2-2)

Pip is t red us t as man iensis

-

4(13)

3

4

—

Total

117

30

100

Comparisons of trapping rates of each species of
small mammal in each physiographic region were used
to investigate the relationship between species abun¬
dance and physiographic regions. In Table 3, trapping
rates for each physiographic region are transformed into
rates for each of the species present. These differences
are further illustrated in Fig. 4, where the trapping rates
of each species in each physiographic region are com¬
pared on a scale of 1 to 100.

The significance of different trapping rates between
pairs of physiographic regions having the highest trapp¬
ing rates for each species are shown in Table 7. All
species except the Yellow-footed Antechinus and Silky
Mouse were significantly associated with a single region.

DISCUSSION

Recent History of the Mammalian Fauna

The mammalian fauna of the survey area has chang¬
ed dramatically since the Pleistocene. Species rep¬
resented in Late Pleistocene and Recent bone deposits
from caves in western Victoria and southeastern South
Australia are listed in Table 1. About half the 49 taxa
listed are no longer present in the area. Reasons for this
decline in species diversity are not clear as the difficulties
of accurately ageing bone deposits (e.g. Wakefield 1967,
1969) and problems of interpretation (Calaby 1971)

Table 5

Resulis of Spotlighting for Arboreal Mammals in
Southwestern Victoria

Species

Total

number

observed

Cell

% of all frequency
observations (%) /? = 74

Triehosurus vulpecula

127

37

44 (59)

Petaurus australis

21

6

10 (13)

Petaurus breviceps

7

2

6 (8)

Pseudocheirus peregrin us

178

52

27 (36)

Phascolaretos cinereus

8

3

6 (8)

Total

341

100

preclude detailed analysis. Several taxa, which may have
been present at the time of European settlement
(1830-1880) have disappeared. These include Eastern
Quoll Dasyurus viverrinus, Red-bellied Pademelon
Thylogale billardierii, Brush-tailed Bet tong Bet t origin
pencillata, Eastern Hare Wallaby Lagorchestes
leporides, Bridled Nailtail Wallaby Onychogalea
fraenata, Toolache Wallaby Marcropus greyi, Swamp
Wallaby Wallabia bicolor , Rabbit-eared Tree-rat Con -
Hums albipes, and Dingo Can is familiaris dingo
(Wakefield 1974). Cockburn (1979) believes a further 4
rodent species were also probably present at that time
but are no longer extant in the area. These were: Broad¬
toothed Rat Mastacomys fuscus , Plains Mouse
Pseudomys australis , New Holland Mouse, P.
novaehollandiae and an undescribed Pseudomys.

The decline of these taxa was probably due to
various combinations of the effects of land clearance,
grazing of stock, changed fire regimes, the introduction
and subsequent effects of European Rabbits, and
general persecution. The writings of some early settlers
describe how the fauna of the Wimmera Plains declined
in the face of European settlement. In the 1860s Edward
Townsend described, amongst others, ‘Wallabies,
Kangaroo-rats, Native Cats and Dingoes’ as occurring in
the Nhill area (Blake 1976, p. 2). Of these only the Red¬
necked Wallaby Macropus rufogriseus occurs there now
and it is rare. In 1861 William Lockhart Morton travell¬
ed north from Antwerp to Pine Plains and beyond
(north of our survey area). He described ‘Kangaroos,
Wallabies, Paddymelons and Kangaroo-rats 1 (An Old
Bushman 1861). From his descriptions of their
behaviour these may have included the Western Grey
Kangaroo Macropus fuliginosus. Bridled Nailtail
Wallaby or Eastern Hare Wallaby, and Brush-tailed
Bettong. Sub-fossil remains of Bridled Nailtail Wallaby
were reported from Lake Hindmarsh in 1959 (Wakefield
1966). No small macropods are mentioned by subse¬
quent naturalists who visited the area (e.g. Le Souef
1887, French 1888a, Le Souef 1893, Campbell 1899)

MAMMALS FROM SOUTHWESTERN VICTORIA

229

Table 6

Comparison of Food Niches Occupied by Each Species and the Physiographic Regions in Which They Occur

Food niche

Species

LD

Physiographic region
WP T CP VP

CARN! VORE

Large Ground Dasyurus maculatus

Small Soil Fossicking

Small Scansorial

Small Aerial

CMNIVORE

Large Ground
Sma 11 Ground

Large Arboreal

Sma11 ArboreaI

HERB I VORE
Large Ground

Small Ground

Tachyglossus aouleatus
Sminthopsis crassicaudata
S. murtna

S. leucopus
Anteohinus swainsonii
A. minimus

Isoodon obesulus
Potorous tridactylus

Anteohinus flavipes
A. stuartii
Phasoogale tapoatafa

Tadarida australis

T. planiceps
Chalinolobu8 gouldii
C. morio
Eptesicus spp
Miniopteris sohreibersii
Myotis adversus
Nyotioeius balstoni
Nyotophilus geoffroyi
Pipistrellus tasmaniensis

Vulpes vulpes
Mus muscuius
Rattus rattus
R. fusoipes
Trichosurus vulpecula
Petaurus breviceps

Ceroartetus ooncinnus
C. nanus

Aorobates pygmaeus

Macropus fuliginosus
M. giganteus
M. rufogriseus
Vombatus ursinus

Oryctolagus cuniculus
Lepus capensis
Pseudomys apodemoides
P. shortridgei
Rattus lutreolus

Large Arborea I Pseudocheirus peregrinus

Petaurus australis
Phascolarctos cinereus

Small Arboreal Pteropus scapulatus

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

Totals {%)

44

17

(39)

26

(59)

20

(45)

30

( 68 )

LD - Little Desert, WP - Winmera Plains, T - Tablelands, CP - Coastal Plains
VP - Volcanic PIains

31

(70)

XXX XX XX XX XX xxxxxx X X X X X X X X X X X

230

P. W. MENKHORST AND C. M. BEARDSELL

Table 7

SlGNlFCCANCE OF DIFFERENCES IN TRAPPING RATES BETWEEN
Physiographic Regions

Only the 2 regions with the highest trapping rates for each
species are compared. Values in parentheses are the number ot
captures in each region. LD-Little Desert (5470 trap nights),
WP-Wimmera Plains (3076), T-Tablelands (1502), CP-Coastal
Plains (5240), VP-Volcanic Plains (2150).

Species

Physiographic

regions

compared

X 2 value

Significance

level

A nt echinus
flavipes

WP:CP (39:71)

0.2

NS

A. minimus

CP:VP (58:2)

18.7

0.001

A. stuartii

CP:VP (13.54)

89.1

0.001

A. swainsonii

CP:VP (10:59)

109.7

0.001

Isoodon obesulus

T:CP (3:44)

7.3

0.01

Trichosurus

vulpecula

WP:CP (20:14)

6.6

0.01

Potorous

tridactylus

CP:VP (7:12)

13.3

0.005

Mus muse ulus

WP:T (168:37)

21.5

0.001

Pseudomys

apodemoides

LD:WP (85:44)

0.3

NS

P. shortridgei

CP:VP (78:14)

8.0

0.005

Putt us fuscipes

CP:VP (316:368)

213.5

0.001

R. lutreolus

CP:VP (211:35)

24.8

0.001

R. rattus

WP:VP (16:3)

4.7

0.05

although French (1888) says that he saw hardly any
small mammals ‘save a large Paddy-melon and a couple
of Kangaroo-rats’ [possibly Bridled Nailtail Wallaby and
Brush-tailed Bettong). These species would have in¬
habited woodland and disappeared soon after selectors
moved into the Wimmera Plains with their flocks of
sheep in the 1840s and 1850s. Wholesale clearing of
woodlands and mallee for crops took place in the 1880s
and the concurrent plagues of European Rabbits
(Campbell 1884, French 1888, 1888a, Rolls 1969, p. 37)
may have hastened the decline of the small macropods
through competition for food and shelter. Details of the
introduction and spread of European Rabbits are given
in the Annotated List. Dingoes were poisoned with
strychnine from the time sheep were first grazed in the
area (Le Souef 1887, French 1888, D’Alton 1913) and
Eastern Quolls may have suffered by taking baits laid for
Dingoes. Marlow (1958) has noted a similar decline in
the marsupial fauna of New South Wales, particularly
the small marsupials of the woodlands and plains, dur¬
ing the late 19th century.

Zoogeography of the Present Fauna

Trapping rates obtained in this survey (20 species
trapped at a success rate of 12%) were greater than or
similar to results from previous FWD surveys in areas of
similar size (e.g. North Central Victoria, 6 species trap¬
ped, 2% success rate (Menkhorst & Gilmore 1979) or
Gippsland Lakes Catchment, 19 species trapped, 14%
success rate (FWD unpublished data). This diverse
mammal fauna reflects the broad range of habitats pre¬
sent, a consequence mainly of the wide variation in rain¬
fall and soil types from north to south. Table 6 il¬

lustrates the increasing diversity of species towards the
coast and the food niches occupied in each physio¬
graphic region. On the Volcanic Plains and Coastal
Plains several species from all food niches were present.
Data for the Tablelands are lacking due to the lack of
native vegetation. However, at least one species from
each food niche was recorded. The Wimmera Plains had
fewer species of small ground and scansorial carnivores
and only one large arboreal herbivore while the Little
Desert was depauperate in small ground and scansorial
carnivores, small ground omnivores, small ground her¬
bivores and arboreal herbivores.

The trend of reducing species richness from the coast
inland is not continued, in the case of small mammals,
in the Big Desert to the north of our survey area. Four
small mammal species occur there but not in the Little
Desert; two small ground carnivores (Ningaui sp.,
Mouse Dunnart Sminthopsis murina ), one small ar¬
boreal omnivore (Little Pygmy-possum Cercartetus
lepidus) and one small ground herbivore (Mitchell’s
Hopping-mouse Notomys mitchellii). Reasons for the
depauperate small mammal fauna in the Little Desert
are not clear as the floristically rich heath comunities
should result in a reliable food source for small her¬
bivores and small carnivores and the climate is not as
severe as in the Big Desert.

The associations between species of small mammals
and physiographic regions, as revealed by comparisons
of trapping rates, arc shown in Fig. 4 and Table 7. The
Silky Mouse was captured at similar rates in the Little
Deseri and Wimmera Plains and the Yellow-footed
Antechinus in the Wimmera Plains and Coastal Plains.

Table 8

Major Existing and Proposed Conservation Reserves in
Southwestern Victoria

There are approximately 428 400 ha of Crown Land in the
survey area.

Reserve

Area

(ha)

Physiographic

region

Little Desert National Park

35300

Little Desert

Lower Glenelg National Park

27300

Coastal Plains

Jilpanger Flora and

Fauna Reserve (proposed)

8990

Wimmera

Discovery Bay Coastal Park

8450

Plains

Coastal Plains

Ml Eccles State Park

6200

Volcanic

Ml Arapiles-

Tooan Stale Park (proposed)

4860

Plains

Wimmera

Tooloy-Lake Mundi Game Reserve

4012

Plains

Coastal Plains

Ml Richmond National Park

1707

Volcanic

Bailleys Rocks Scenic
and Recreation Reserve

489

Plains

Coastal Plains

Bats Ridges Faunal Reserve

324

Volcanic

Cape Nelson State Park

210

Plains

Volcanic

Total

97842

Plains

MAMMALS FROM SOUTHWESTERN VICTORIA

231

Table 9

Areas of Conservation Reserves in each Physiographic
Region

Figures in parentheses are the percentages of Crown Land in
conservation reserves within each region.

Region

Area

reserved

(ha)

Percentage of
total reserved
area in

Southwestern Victoria

Little Desert
Wimmera

35300 (22)

36

Plains

13850 (46)

14

Tablelands

0

0

Coastal Plains

40251 (30)

41

Volcanic Plains

8442 (12)

9

Totals

97842

100

The Brown Antechinus Antechinus stuart Dusky
antechinus A. swainsonii. Long-nosed Potoroo
Potorous tridactylus, and Bush Rat were all trapped at
significantly higher rates on the Volcanic Plains where
the vegetation is predominantly Messmate open-forest
and Manna Gum open-forest and the rainfall is the
highest in the survey area (Fig. 2).

The Swamp Antechinus A. minimus , Short-nosed
Bandicoot Isoodon obesulus, Swamp Rat, and Heath
Mouse Pseudomys shortridgei were significantly
associated with the Coastal Plains where they occur in
the extensive heathlands and Brown Stringybark open-
forest with a heath understory.

On the Wimmera Plains the two introduced rodents
House Mouse and Black Rat Rattus rattus were trapped
at higher rates than elsewhere, as was the Common
Brushtail Possum. The predominance of the two in¬
troduced rodents probably reflects the greater extent of
disturbance of the native vegetation in this region.
Conservation Reserves

About 98 000 ha of Crown Land (23%) in the survey
area is held in existing or proposed conservation reserves
(LCC 1973, 1981, Ministry for Conservation 1981)
(Table 8). This proportion of Crown Land held in con¬
servation reserves is similar to that in most areas for
which the LCC has produced Final Recommendations.
However, the proportion for each physiographic region
varies from 12% on the Volcanic Plains to 46% on the
Wimmera Plains (Table 9). Further reserves are justified
on the Volcanic Plains where most Crown Land is now
used for hardwood production.

Most vegetation alliances are well represented within
the present reserve syslem. However, River Red Gum
woodland. Yellow Gum woodland and Brown Stringy¬
bark open-forest are inadequately catered for despite the
fact that Brown Stringybark open-forest supports more
mammal species than any other alliance and the two
woodland alliances each support more than 40% of
species (data from Appendix 1). The two woodland
alliances occurred extensively on the Wimmera Plains,
Tablelands and northern Coastal Plains but were cleared
to form prime grazing and cropping country. The small
areas of woodland remaining on Crown Land are now
used for hardwood production threatening the existence
of one species (Tuan Phascogaie tapocitafa) in the survey
area. The extensive areas of Brown Stringybark open-
forest north and west of the Cilenelg River are only
reserved in Tooloy-Lake Mundi Game Reseve yet this
alliance supports a diverse small mammal fauna con¬
sisting of Yellow-footed Antechinus, Swamp
Antechinus, Short-nosed Bandicoot, Bush Rat, Swamp
Rat, Silky Mouse and Heath Rat. This fauna is transi¬
tional between those of the Desert Banksia open-heaths
of the Wimmera Plains and the wetter Silver Banksia
closed-heaths near the coast. Further reserves of all 3
alliances are necessary to adequately cater for the range
of mammals in the survey area.

Conservation reserves are defined here as any area of
Crown Land having flora and fauna conservation as a ACKNOWLEDGEMENTS

primary management aim and being large enough to We are grateful to the following for assistance in col-

maintain communities of mammals. Areas used for lecting field data: Fisheries and Wildlife Division: G.
hardwood or softwood production and the numerous Apps, W. Bren, K. Cherry, S. Craig, D. Deerson, W.
small wetland reserves are excluded from the following Emison, A. Gilmore, D. Hespe, G. Horrocks, J. Mar-
discussion. cius, K. Norris, M. O’Sullivan, J. Porter, J. Seebeck,

Table 10

Habitat Preference of Mus musculus and Pseudomys apodemoides in the Little Desert
Combined trapping data from October to November 1978 and February and April 1979.

Number

of Percentage of sites Mean trapping success

Habitat sites where trapped per 100 trapnights

P.

apodemoides

M.

musculus

P.

apodemoides

M.

musculus

Desert Banksia open-heath

20

55

15

2

1

Yellow Mallee open-scrub

3

33

0

1

Broom Honey-myrtle open-scrub

21

52

10

2

0.3

Brown Stringybark low open-forest

40

65

12

2

1

Yellow Gum woodland

24

0

20

1

Red Gum woodland

3

0

100

5

232

P. W. MENKHORST AND C. M. BEARDSELL

and I. Temby; National Museum of Victoria: A. Coven¬
try, A. Gedye, and R. Wilson; National Herbarium: Dr.
P. Gullan, D. Parkes, and N. Walsh. J. Alexander and
R. Adair also assisted. Information from personal files
and records was provided by C. Brownsea (National
Parks Service), G. Cerini (FWD), A. Cockburn
(Monash University), C. Crouch (Nhill), J. Davies
(NPS), K. Hateley (Kiata), R. Howlett (Dergholm), P.
Kelly (FWD), W. Middleton (Forests Commission Vic¬
toria), R. and P. Reichelt (Nhill), F. Rogers (Quantong),
J. Seebeck (FWD), and R. Warneke (FWD).

The collations of museum specimens and literature
records carried out at the NMV (Gedye et al. 1979,
Evans & Dixon 1980) were a major source of informa¬
tion and saved much tedium. P. Aitken provided lists of
relevant specimens held in the South Australian
Museum.

A. F. Bennett, W. B. Emison, Dr. F. I. Norman,
and J. H. Seebeck kindly commented on drafts of this
paper. J. Marcius helped with proof reading.

REFERENCES

An Old Bushman [William Lockhart Morton], 1861. Notes of
a tour in the Wimmera District. The Yeoman and
Australian Acclimatiser Vol. 1.

Anonymous, 1846. Miscellanea. The Tasmanian Journal 2:
460.

Anonymous, 1907. Natural History Note. Victorian Nat. 23:
243.

Australian Mammal Society, 1980. Recommended common
names of Australian mammals. Aust. Mamm. Soc.
Bull. 6(2): 13-23..

Bentley, A., 1978. An introduction to the Deer of Australia
with special reference to Victoria (Revised ed.).
Koetong Trust Service Fund, Forests Commission Vic¬
toria, Melbourne.

Blake, L., 1976. The Land of the Lowan: 100 Years in Nhill
and West Wimmera. Nhill and District Historical
Society, Nhill.

Braithwaite, R. W., 1980. The ecology of Rattus lutreolus.

II. Reproductive tactics. Aust. Wildl. Res. 7: 53-62.
Braithwaite, R. W., Cockburn, A. & Lee, A. K., 1978.
Resource partitioning by small mammals in lowland
heath communities of south-eastern Australia. Aust. J.
Ecol. 3: 423-445.

Braithwaite, R. W. & Lee, A. K., 1979. The ecology of
Rattus lutreolus. I. A Victorian heathland population.
Aust. Wildl. Res. 6: 173-189.

Braithwaite, R. W. & Lee, A. K., 1979a. A mammalian
example of semelparity. American Nat . 113: 151-5.
Brazenor, C. W., 1936. Muridae recorded from Victoria.

Mem. Natn. Mils. Melb. 10: 62-85.

Brazenor, C. W., 1958. Mitchells Hopping Mouse. Proc. R.

Zool. Soc. NSW (1956-7): 19-22.

Bureau of Meteorology, 1975. Climatic Averages Australia.
Metric ed. Aust. Government Publishing Service,
Canberra.

Calaby, J. H., 1971. Man, fauna and climate in aboriginal
Australia. In Aboriginal Man and Environment in
Australia, Mulvaney, D. J. and Galson, J., eds, ANU
Press, Canberra.

Campbell, A. J., 1884. Mallee Hens and their Egg Mounds.
Victorian Nat. 1: 124-129.

Campbell, A. J., 1899. Field notes from the lower Wimmera.
Victorian Nat. 16: 121-131, 149-158.

Cockburn, A., 1979. The ecology of Pseudomys spp. in south¬
eastern Australia. PhD Thesis, Monash University,
(unpubl.).

Cockburn, A., 1981. Diet and habitat preference of the Silky
Desert Mouse, Pseudomys apodemoides (Rodentia).
Aust. Wildl. Res. 8: 475-497.

Cockburn, A., 1981a. Population processes of the Silky
Desert Mouse Pseudomys apodemoides (Rodentia), in
mature heathland. Aust. Wildl. Res. 8: 499-514.

Cockburn, A., Bratihwatie, R. W. & Lee, A. K., 1981.
The response of the Heath Rat, Pseudomys shor-
tridgei, to pyric succession: a temporally dynamic life-
history strategy. J. Anim. Ecol . 50: 649-666.

D’Alton, St Elroy, 1913. The botany of the Little Desert,
Wimmera, Victoria. Victorian Nat. 30: 65-78.

Dwyer, P. D., 1970. Foraging of the Australian Large-footed
Myotis (Chiroptera). Mammalia 34: 76-80.

Dwyer, P. D. & Hamilton-Smith, E., 1965. Breeding
caves and maternity colonies of the Bent-winged Bat in
south-eastern Australia. Helictile 4: 3-21.

Emison, W. B., Porter, J. W., Norris, K. C. & Apps,
G. J., 1978. Survey of the vertebrate fauna in the
Grampians-Edenhope area of southwestern Victoria.
Mem. natn. Mus. Viet. 39: 281-363.

Evans, S. & Dixon, J. M., 1980. Report on the mammal¬
ian fauna of the South Western Study Area (District 1).
Mammal Dept., National Museum of Victoria,
Melbourne.

Finlayson, H. H., 1927. Observations on the South Australian
members of the subgenus “ Wallabia ”. Trans. R. Soc.
S. Aust. 51: 363-377.

Finlayson, H. H., 1944. A further account of the murid,
Pseudomys (Gyomys) apodemoides Finlayson. Trans.
R. Soc. S. Aust. 68: 210-224.

Frankenbf.rg, Judith, 1971. Nature Conservation in Vic¬
toria: a survey. Victorian National Parks Association,
Melbourne.

French, Charles, 1888. Notes on the Zoology of Lake
Albacutya District. Victorian Nat. 5: 35-42.

French, C., 1888a. Notes on the natural history of the western
Wimmera. Victorian Nat. 5: 145-152.

Gedye, A., Wilson, R., Dixon, J. M. & Huxley, L., 1979.
Report on the mammalian fauna of the Wimmera, Vic¬
toria. Mammal Dept., National Museum of Victoria,
Melbourne.

Hall, K. S. & Richards, G. C., 1979. Bats of Eastern
Australia. Queensland Museum Booklet No. 12,
Brisbane.

Hamilton, J. C., 1914. Pioneering Days in Western Victoria.
Exchange Press, Melbourne.

Hamilton-Smith, E., 1965. Distribution of cavc-dwelling bats
in Victoria. Victorian Nat. 82: 132-137.

Happoi.d, M., 1976. Reproductive biology and development in
the conilurine rodents (Muridae) of Australia. Aust. J.
Zool. 24: 19-26.

Hills, E. S., 1975. The Physiography of Victoria. Whitcombe
and Tombs, Melbourne,

Jones, F. W., 1924. The Mammals of South Australia. Pt. 2
Bandicoots and the Herbivorous Marsupials. Govern¬
ment Printer, Adelaide.

Keiller, H., 1940. Wildlife of Portland. B.O.C. Monthly
Notes (Aug.).

Land Conservation Council of Victoria, 1972. Report on
the South West Study Area, District I. Government
Printer, Melbourne.

MAMMALS FROM SOUTHWESTERN VICTORIA

233

Land Conservation Council of Victoria, 1973. Final
Recommendations , South-western Study Area, District
1. Government Printer, Melbourne.

Land Conservation Council of Victoria, 1979. Report on
the South West Study Area, District 2. Government
Printer, Melbourne.

Land Conservation Council of Victoria, 1981. Proposed
Recommendations , South-western Area, District 2.
Government Printer, Melbourne.

Land Conservation Council of Victoria, 1981a. South¬
western Area , District 1 —Review. Government
Printer, Melbourne.

Le Souef, D., 1887. Trip to Lake Albacutya. Victorian Nat.
4: 44-47.

Le Souef, D., 1893. Notes on a visit to the Ebenezer Mission
Station. Victorian Nat. 10: 123-128.

Le Souef, J. C., 1965. Acclimatization in Victoria. The Vic¬
torian Historical Magazine 36: 8-29.

Maddock, T. H. t 1971. Some mammal remains from caves in
the Naracoorte area. S. Aust. Nat. 46: 24-27.

Mahoney, J. A., 1982. Identities of the rodents (Muridae)
listed in T. L. Mitchell’s “Three Expeditions into the
Interior of Eastern Australia, with Descriptions of the
Recently Explored Region of Australia Felix, and of
the Present Colony of New South Wales”. Aust.
Mamm. 5: 15-36.

Marlow, B. J., 1958. A survey of the marsupials of New
South Wales. CS/RO Wildl. Res. 3: 71-114.

McKean, J. L. & Hall, L. S., 1965. Distribution of the
Large-footed Myotis, Myotis adversus, in Australia.
Victorian Nat. 82: 164-168.

McKean, J. L., Richards, Ci. C. & Price, W. J., 1978. A
taxonomic appraisal of Eptesicus (Chiroptera: Mam¬
malia) in Australia. Aust. J. Zool. 26: 529-537.

Menkhorst, P. W. & Gilmore, A. M., 1979. Mammals
and reptiles of North Central Victoria. Mem. natn.
Mus. Viet. 40: 1-33.

Ministry for Conservation, 1981. A guide to the Ministry
for Conservation. 3rd ed., Government Printer,
Melbourne.

Morton, S. R., 1978. An ecological study of Sminthopsis
crassicaudata (Marsupialia: Dasyuridae). III.
Reproduction and life history. Aust. Wildl. Res. 5:
183-212.

Morton, S. R., Wainer, .1. W. & Thwaites, T. P., 1980.
Distribution and habitats of Sminthopsis leucopus and
5. murina (Marsupialia: Dasyuridae) in south-eastern
Australia. Aust. Mamm. 3: 19-30.

Newsome, A. E., 1969. A population study of house-mice tem¬
porarily inhabiting a South Australian wheatfield. J.
Anim. Ecol. 38: 341-359.

N. F. L. [Noel Learmonth] 1947. Random Notes. Portland
Guardian 28 May, 1947.

Norris, K. C., Gilmore, A. M. & Menkhorst, P. W.,
1979. Vertebrate fauna of South Gippsland, Victoria.
Mem. natn. Mus. Viet. 40: 105-199.

Parnaby, H., 1977. Bat survey of the Daylesford area, Vic¬
toria. Victorian Nat. 94: 191-197.

Poole, W. E., 1973. A study of breeding in Grey Kangaroos,
Macropus giganteus Shaw and M. fuliginosus
(Desmarest), in central New South Wales. Aust. J.
Zool. 21: 183-212.

Poole, W. E., 1977. The Eastern Grey Kangaroo, Macropus
giganteus , in South-east South Australia: its limited
distribution and need of conservation. CSIRO Div.
Wildl. Res. Tech. Pap. No. 31.

Poole, W. E., Carpenter, S. M. & Simms, N. G., 1980.

Multivariate analyses of skull morphometries from the
two species of Grey Kangaroos, Macropus giganteus
Shaw and M. fuliginosus (Desmarest). Aust. J. Zool.
28: 591-605.

Robinson, A. C., 1976. Population ecology of Rattus fuscipes
Waterhouse (Muridae: Rodentia). PhD Thesis, Monash
University (unpubl.).

Rolls, E. C., 1969. They all Ran Wild: the story of pests on
the land in Australia. Angus & Robertson, Sydney.

Ryan, R. M., 1963. Extension of range of Pesudomys
apodemoides. Victorian Nat. 79: 363.

Saunders, G. R. & Giles, J. R., 1977. A relationship be¬
tween plagues of the house mouse, Mus musculus
(Rodentia: Muridae) and prolonged periods of dry
weather in south-eastern Australia. Aust. Wildl. Res.
4: 151-157.

Seebeck, J. H. & Hamilton-Smith, E., 1967. Notes on
a wintering colony of bats. Victorian Nat. 84: 348-351.

Smith, M., 1971. Small fossil vertebrates from Victoria Cave,
Naracoorte, South Australia. I. Potoroinae (Mac-
ropodidac), Pctauridac and Burramyidae (Mar¬
supialia). Trans. R. Soc. S. Aust. 95: 185-198.

Smith, M., 1972. Small fossil vertebrates from Victoria Cave,
Naracoorte, South Australia. II. Peramelidae, Thyl-
acinidae and Dasyuridae (Marsupialia). Trans. R. Soc.
S. Aust. 96: 125-137.

Specht, R. L., Roe, E. M. & Broughton, V. H., 1974.
Conservation of major plant communities in Australia
and Papua New Guinea. Aust. J. Bot. Suppl. No. 7.

Stoddart, D. M. & Braithwaite, R. W., 1979. A strategy
for utilization of regenerating heathland habitat by the
Brown Bandicoot ( Isoodon obesulus; Marsupialia,
Peramelidae). J. Anim. Ecol. 48: 165-179.

Tidemann, C. R., 1967. Some mammal remains from cave
deposits in the south-east of South Australia. S. Aust.
Nat. 42: 21-27.

Tidemann, C. R. & Woodside, D. P., 1978. A collapsible
bat-trap and a comparison of results obtained with the
trap and with mist nets. Aust. Wildl. Res. 5: 355-362.

Wainer, J. W., 1976. Studies of an island population of
An(echinus minimus ( Marsupialia, Dasyuridae). Aust.
Zoologist 19: 1-7.

Wainer, J. W. & Gibson, R. .1., 1976. Habitat of the
Swamp Antechinus in Victoria. Victorian Nat. 93:
253-255.

Wakefield, N. A., 1963. Mammal remains from The Gram¬
pians, Victoria. Victorian Nat. 80: 130-133.

Wakefield, N. A., 1964. Recent mammalian sub-fossils of
the basalt plains of Victoria. Proc. R. Soc. Viet. 77:
419-425.

Wakefield, N. A., 1964a. Mammal sub-fossils from near
Portland, Victoria. Victorian Nat. 80: 39-45.

Wakefield, N. A., 1966. Mammals recorded for the Mallee,
Victoria. Proc. R. Soc. Viet. 79: 627-633.

Wakefield, N. A., 1967. Preliminary report on McEachern’s
Cave, S.W. Victoria. Victorian Nat. 84: 363-383.

Wakefield, N. A., 1969. Interpretation of data from
McEachern’s Cave, S.W. Victoria. Helictite 7: 17-20.

Wakefield, N. A., 1970. Notes on the glider-possum,
Petaurus australis (Phalangeridae, Marsupialia). Vic¬
torian Nat. 87: 221-236.

Wakefield, N. A., 1971. The Brush-tailed Rock-wallaby
(Petrogale penicillata) in Western Victoria. Victorian
Nat. 88: 92-102.

Wakefield, N. A., 1974. Mammals of Western Victoria. In
The Natural History of Western Victoria, Douglas, M.
H. & O’Brien, L., eds. Western Victorian Sub-branch,
Aust. Inst. Ag. Sci., Horsham, 35-51.

234

P. W. MENKHORST AND C. M. BEARDSELL

Wallis, R. & Baxter, G., 1980. The Swamp An (echinus
(Antechinus minimus maritimus) — notes on a captive
specimen. Victorian Nat. 97: 211-213.

Watson, 1. Garnet, J. R., Lee, R. D. t Burke, A. &
Cudmore, F. A., 1947. The proposed Glenelg National
Forest and Sanctuary. Victorian Nat. 64: 62-73.

Watts, C. H. S. & Aslin, H. J., 1981. The Rodents of
Australia. Angus & Robertson, Sydney.

Warneke, R. M., 1971. Field study of the Bush Rat (Rattus
fuscipes). Wild/. Contrib. Viet. 14.

Warneke, R. M., 1978. The status of the Koala in Victoria. In
The Koala: proceedings of the Toronga Symposium,

Bergin, T. J. ed., Zoological Parks Board, New South
Wales, Sydney.

Willis, J. H., 1970. A Handbook to Plants in Victoria
Volume l. Ferns, Conifers and Monocotyledons. 2nd
Ed. Melbourne University Press, Melbourne.

Willis, J. H., 1972. A Handbook to Plants in Victoria %
Volume II. Dicotyledons. Melbourne University Press,
Melbourne.

Woolley, P., 1966. Reproduction in Antechinus spp. and
other dasyurid marsupials. Symp. Zool. Soc. Lond. 15*
281-294.

APPENDIX 1

Annotated List of Mammals

Nomenclature and taxonomic order follow Australian
Mammal Society (1980) and names of plants follow Willis
(1970, 1972). Vegetation alliances listed under habitat include
only those in which we recorded the species unless otherwise
stated. Specimen numbers prefixed by C refer to specimens
held in the NMV, those listed under Specimens were taken dur¬
ing this survey. FWD indicates Fisheries and Wildlife Division,
Ministry for Conservation. Previous records arc listed by
Gedye et at. (1977) and Evans and Dixon (1980). For species
not recorded during our survey the most recent record from the
survey area is given.

Tachyglossidae

1. Tachyglossus aculeatus —Short-beaked Echidna
Abundance and distribution. Uncommon and widespread. On¬
ly 18 observed during our survey, 5 of which were road mor¬
talities. Most sightings ( 66 %) were in spring and summer.
Observed in 12 cells although characteristic diggings were
found in 43% of all cells in which trapping took place.
Habitat. All terrestrial environments except closed-scrub,
closed-heath and extensive tracts of farmland.

Ornithorynchidae

2. Ornithorhynchus analinus — Platypus

Abundance and distribution. Common and restricted to
streams. Recorded from the Glenelg, Wannon (B. Burchell
pers. comm.) Surrey (N.F.L. 1947) and Wimmera Rivers (W.
Middelton pers. comm.); from Bryon Creek at Coleraine;
Darlots Creek (Kciller 1940) and Lake Monibeong (NPS
records). Not observed during our survey although several re¬
cent sightings were reported by local naturalists. The
zoogeographic interest of the isolated population in the Wim¬
mera River has been mentioned by Emison et al. (1978).
Habitat. Aquatic. Requires permanent freshwater with a mud
or gravel substrate and friable banks in which to construct bur¬
rows. In the Glenelg River it occurs in tidal, brackish water as
far downstream as Sapling Creek some 30 km from the mouth
(J. Davies pers. comm.) although Watson et al. (1947) reported
a sighting about 1.6 km from the mouth.

Dasyuridae

3. Antechinus flavipes- Yellow-footed Antechinus
Abundance and distribution. Widespread and common but ab¬
sent from the Little Desert and rare on the Volcanic Plains. A
specimen labelled ‘Kaniva area’ (NMV C15870) actually came
from 20 km S of Kaniva on the southern edge of the Little
Desert (C. Crouch pers. comm.).

Habitat. Brown Stringybark open-forest and low open-forest,

all woodland types, Silver Banksia open-heath and Desert
Banksia open-heath on the Wimmera Plains. Trapping rates
were highest in Manna Gum woodland ( 8 %) and lowest in
Desert Banksia open-heath (4%).

Breeding. Females with pouch young were trapped between 15
and 23 August. All females had ten nipples, and ten pouch
young were recorded in all five litters examined. By 14
November females were still lactating but had no young in the
pouch. Male die-otT(Braithwaite & Lee 1979a) was complete by
15 August, the last male being trapped on 6 July and one was
found dead on 17 August (FWD 12016). First year males were
first trapped on 28 January.

Specimens. C13176, C14052, C22268, C24363-9, C24548
C25208, FWD 12016.

4. Anetchinus minimus— Swamp Antechinus
Abundance and distribution. Restricted to the Volcanic Plains
near the coast and those parts of the Coastal Plains having
>650 mm rainfall per year. Locally common.

Habitat. Brown Stringybark open-forest with wet heath under¬
story, Silver Banksia closed-heath, Coast Wattle open-scrub
and Blue Tussock Grass closed-grassland. Only 50% of trap
sites were in treeless heath (cf. Wainer & Gibson 1976) and
mean trapping success rates were similar in Browm Stringybark
open-forest ( 8 %) and treeless Silver Banksia closed-heaths
(7%). Florist ic data were collected at 14 of the 22 capture sites.
Commonly occurring species of plants from these sites are
listed below. Many of these species were also dominant in
heaths preferred by A . minimus at Cape Liptrap, South Gipp-
sland (Braithwaite et al. 1978).

Species % Occurrence

Leptospermum juniperinum 85

Xanthorrhoea minor 86

Banksia marginata 71

Melaleuca squarrosa 1 1

Leptocarpus tenax 64

Sprengelia incarnata 43

Casuarina pus ilia 36

Eucalyptus baxteri 36

Breeding. Little trapping took place within the range of this
species during the supposed breeding period (June-September).
A female trapped on 1 June (NMV C22267) had a pouch
similar to other Antechinus spp. at the time of mating (Woolley
1966), and independent young were trapped between 13 and 16
November. Assuming the age of weaning is roughly 14 weeks
(Wallis & Baxter 1980), mating must have occurred before ear¬
ly August. Sex ratios of trapped animals in March, May and
June were 1.7:1 (12a, 7 9), 1.6:1 ( 8 cr, 5 9) and 1.7:1 (15a,

MAMMALS FROM SOUTHWESTERN VICTORIA

235

99) respectively, showing a similar preponderance of males to
that found by Wainer (1976). No males were trapped in July.
All females had 8 nipples. Thus the breeding cycle in
southwestern Victoria appears similar to that on Great Glennie
Island (Wainer 1976) and in South Gippsland (Norris et at.
1979, Wallis & Baxter 1980).

Specimens. C22267, C24345-7, FWD 12398.

5. Antechinus stuartii — Brown Antechinus

Abundance and distribution. Restricted to the Volcanic Plains
near the coast and high rainfall (> 800 mm per annum) areas of
the Coastal Plains where it is locally common. Not recorded in
South Australia.

Habitat. Messmate open-forest and Silver Banksia closed-
heath.

Breeding. Trapping took place within the range of this species
only in October and March. Male die-off was complete before
October when females had nest young. One independent male
juvenile was captured on 12 October. All females examined
had 6 nipples.

Specimens. C13559, C24358, C24382.

6 . Antechinus swainsonii —Dusky Antechinus
Abundance and distribution. Restricted to the Volcanic Plains
near the coast and high rainfall (>800 mm per annum) areas of
the Coastal Plains. Locally common. Not recorded from South
Australia.

Habitat. Messmate open-forest and Silver Banksia closed-
heath.

Breeding. Trapping within the range of this species began on 12
October when male die-off was complete. Lactating females
without pouch young were captured between 12 and 19 Oc¬
tober, as were independent juveniles. All females had 8 nipples.
Specimens. C13558, C24379-81, FWD 12152, FWD 12226.

7. Dasyunis maculatus — Tiger Quoll

Abundance and distribution. Restricted to the high rainfall
( >750 mm per annum) areas of the Volcanic Plains except for
a specimen from Hamilton in 1958 (NMV C2806). Uncommon
in The Stones State Faunal Reserve, rare in the Heywood area.
The population in The Stones State Faunal Reserve is one of
the most stable and accessible in Victoria, the only other areas
where regular reports are made being the upper Snowy River
and Otway Ranges.

Habitat. Manna Gum woodland with an understorey of
Pteridium escu/entum and grasses covering jumbled basalt
boulders. Occasionally in Messmate open-forest and Manna
Gum open-forest in Cobobboonee Forest. During the breeding
season from May to late July wandering adult males occur in
farmland surrounding the main habitat.

8 . Dasyurus viverrinus — Eastern Quoll

Abundance and distribution. Formerly widespread but no
longer present. Listed by Edward Townsend, an early Nhill
resident, as occurring in the Nhill district in the 1860s (Blake
1976, p. 2). Reported to have killed young Ostriches Strutbio
camelus being bred at Longerenong Station in 1867 (Rolls
1969, p. 251).

Habitat. Presumably most woodland alliances.

9. Phascogale lapoalafa — Tuan

Abundance and distribution. Restricted and rare. Recorded
only on the Tablelands and Wimmera Plains where small rem¬
nants of Crown Land and roadside and streamside reserves
provide the only remaining habitat. During our survey a road-
killed animal was found at Apsley and we were told of one cap¬
tured 13 km WNW of Edenhope and 2 killed by cats near
Balmoral. Also recorded from: Telangatuk East (FWD 785),
Bulart (FWD D587), 29 km N of Casterton (Cl3969), 23 km N
of Coleraine (C3135), Brit Brit (C2475), Casterton (C2655) and

Coleraine (FWD DB251). The provenance of two specimens
labelled Portland (C2042-3) is uncertain. They may have come
from further north.

Habitat. All woodland alliances.

Specimen. Cl3164.

10. Sminlhopsis crassicaudata —Fat-tailed Dunnart
Abundance and distribution . Widespread and uncommon in
the Tablelands and Wimmera Plains; rare in the Little Desert.
Not recorded from the Coastal or Volcanic Plains. Only one
captured during our survey in a pitfall in the Little Desert 16
km S of Gerang Gerung. Probably more common than our
results suggest as it mostly inhabits farmland.

Habitat. Desert Banksia open-heath. Usually associated with
farmland. Morton (1978) claims that it does not occur in closed
or open-scrub. However, the only animal captured during our
survey was collected in low open-heath dominated by
Casuarina muellerana, Xanlhorrhoea australis and Hibbertia
sp. on a dune adjacent to a small claypan with Yellow Gum
woodland.

Specimen. C22264.

11. Sminlhopsis leucopus —White-footed Dunnart
Abundance and distribution. Restricted to areas of the
Volcanic Plains and Coastal Plains having >700 mm rainfall
per annum. Not recorded from adjacent parts of South
Australia. Probably more common than records suggest as it is
difficult to trap.

Habitat. Brown Stringybark open-forest with a dense heath
understory and Silver Banksia closed-heath. Dominant plant
species from our trapping site were: Acacia verticillata,
Banksia marginata, Hibbertia sp. Leptocarpus tenax, Lep-
tospermum juniperinum, Melaleuca squarrosa, Sprengelia in -
carnata and Xanlhorrhoea minor. Morton et al. (1980)
describe the habitat from the Mt Clay site and Cockburn (pers.
comm.) trapped and released several individuals on Bald Hill,
Lower Glenelg National Park in dense wet heath.

Specimen. C22668.

12. Sminlhopsis murina — Common Dunnart

Abundance and distribution. Restricted to the Tablelands and
Wimmera Plains. All records come from the Edenhope-
Chetwynd area where the mean annual rainfall is between 700
and 550 mm (cf, Morton et al. 1980); we consider its presence
in the lower Glenelg River area to be unconfirmed as specimen
C14017 is intermediate and the habitat is atypical. Only one
(0.1%) was trapped 15 km SW of Edenhope, and the remains
of another were found in a fox stomach (FWD 8993) taken 10
km W of Chetwynd in March 1973. Also recorded from 24 km
N of Casterton (C14020, C14028) in 1967 and from 23 km
NNW of Coleraine in 1962 (C4510). Not recorded from the Lit¬
tle Desert although it occurs in the Big Desert to the north and
at similar latitudes in South Australia (Morton et al. 1980, fig.
2 ).

Habitat. Brown Stringybark open-forest. Our capture was
made in a thicket of Leptospermum juniperinum with a sparse
ground cover of Hibbertia fasciculata, Lepidosperma
longitudinale and Hypolaena fastigiata. This thicket was sur¬
rounded by Brown Stringybark open-forest with a sparse heath
understorey dominated by Xanlhorrhoea minor, Brachyloma
daphnoides, Leucopogon virgatus and Calytrix tetragona.
Specimens. C24357, FWD 8993.

Peramelidae

13. Isodoon obesulus- Southern Brown Bandicoot
Abundance and distribution. Widespread and common on the
Coastal Plains and Volcanic Plains. The most northerly record
in the survey area (37° 10'S) approximates the 600 mm isohyet.
Not recorded from adjacent parts of South Australia.

236

P. W. MENKHORST AND C. M. BEARDSELL

Habitat. Messmate open-forest. Brown Stringybark open-
forest and low open-forest, Scented Paper-bark closed-scrub
and Silver Banksia closed-heath and open-heath. Not recorded
in Coast Wattle open-scrub in contrast to South Gippsland
(Norris et al. 1979) nor in Desert Banksia open-heath. Trap¬
ping success in Brown Stringybark open-forest (3.7%) was
similar to that in Silver Banksia closed-heath (3.3%).
Breeding. Trapping within the range of I. obesulus took place
in March, May, June, July, August, October and November.
Pouch young were present in June, October and November but
not in July and August when few females were trapped. Mean
litter size was 2.5 (rt = 6). Stoddart and Braithwaite (1979)
found a distinct breeding peak between July and December at
Cranbourne, Victoria and a mean litter size of 3 but claim that
breeding occurs year round in Western Victoria.

Specimens. C22269, C24374-5.

Phalangeridae

14. Trichosurus vulpecula — Common Brushtail Possum
Abundance and distribution. Widespread and common except
in the Little Desert where it is rare. The most widespread ar¬
boreal mammal in the survey area (Table 5).

Habitat. All open-forest and woodland types and farmland
with trees. Of the 164 animals trapped or spotlighted during
our survey 44% were in River Red Gum woodland, 16% in
Brown Stringybark open-forest, 4% in Brown Stringybark low
open-forest, 16% in Manna Gum open-forest, 8% in Manna
Gum woodland, 3% in Messmate open-forest, 6% in Yellow
Gum woodland, 3% in Casuarina luehmannii in farmland and
1% in Eucalyptus cladocalyx plantation.

Burramyidae

15. Acrobatcs pygmaeus — Feathertail Glider

Abundance and distribution. Widespread and uncommon. Not
recorded in the Little Desert or on the Coastal Plains.
Specimen records are from Portland, Gorae Forest and 24 km
NW of Castcrton (Evans &. Dixon 1980); Goiter South (Gedye
et al. 1979); Frances (SAM M2088); and Lower Norton (FWD
12273). Sight records arc from Coleraine in 1975 (B. Burchell
pers. comm.), 3 km NE of Balmoral in March 1980 (1. Temby
pers. comm.) and 4 km E of Comaum in 1974 (A. Roper pers.
comm.). The provenance of a record from the Kaniva area
(Wakefield 1966) remains uncertain.

Habitat. Recorded in Yellow Gum w'oodland (I. Temby pers.
comm.), Messmate open-forest in Gorae Forest (V. Peterson
pers. comm, to J. Seebeck), in an old building in Coleraine (B.
Burchell pers. comm.) and in a fruit and vegetable garden with
nearby River Red Gum woodland at Lower Norton. Brown
Stringybark open-forest may be unsuitable.

16. Cercartetus concinnus —Western Pygmy-possum
Abundance and distribution. Widespread and probably com¬
mon in the Little Desert, rare on the Wimmcra Plains (Emison
et al. 1978) and absent from elsewhere in the survey area. All
records are from areas with <600 mm rainfall per annum.
Recorded in only 2 grids during our survey. It does not readily
enter cage traps and was recorded by drift fence pitfall trapping
and the chance finding of skeletal remains. Probably more
common than records suggest.

Habitat. Recorded in Brown Stringybark low r open-forest and
Yellow Gum woodland. Two individuals were captured in pit-
falls in Yellow Gum woodland with an understorey of Mela¬
leuca wilsonii and nearby heath with Triodia sp. Another was
found dead in an old building surrounded by Brown Stringy¬
bark low open-forest.

Breeding. A female examined on 30 March 1970 on the edge of
the Little Desert S of Broughtons Waterhole had 5 pouch
young.

Specimens. C21025, C22245.

17. Cercartetus nanus— Eastern Pygmy-possum
Abundance and distribution. Restricted to the high rainfall
(>800 mm per annum) areas of the Volcanic Plains where it is
uncommon. Only 1 (0.1%) trapped during our survey, 5 km
NNE of Mt Kincaid in Lower Glenelg National Park. Previous
specimen records are from Portland, Surrey River west of
Portland, Gorae area and Mt Richmond (Evans & Dixon
1980). A specimen from Edenhopc (C2471) registered as C.
nanus is actually C. concinnus (L. Huxley pers. comm.).
Habitat. Messmate open-forest, Manna Gum open-forest and
Silver Banksia closed-heath with emergent eucalypts. The
animal trapped during our survey was captured on the ground
in closed-heath dominated by Casuarina pusilla, Leptosper -
mum juniperinum, L. myrsinoides, Hypolaena fastigiata, Xan-
thorrhoea australis, Banksia marginata, Melaleuca squarrosa ,
Persoonia juniperina and Leucopogon australis with emergent
Eucalyptus nitida.

Specimen. C24378.

Petauridae

18. Petaurus australis — Yellow-bellied Glider
Abundance and distribution. Restricted to the Volcanic Plains,
Coastal Plains and south Wimmera Plains where the mean an¬
nual rainfall exceeds 600 mm. The most northerly record dur¬
ing our survey was 14 km SW of Edenhopc. Formerly present
in streamside Manna Gum woodland in the Kanawinka area
(R. Kceble pers. comm.).

Habitat. Manna Gum open-forest. Messmate open-forest,
Brown Stringybark open-forest, particularly where gum-
barked eucalypts (E. viminalis, E. aromaphloia and E. ovata)
are also present and Yellow Gum woodland. The 58 trees with
incisions made by this species, “feed-trees” (Wakefield 1970),
included E. viminalis (45%), E. leucoxylon (33%), E. obliqua
(16%), E. baxteri (2%), E. pauciftora (2%) and E.
aromaphloia (2%).

19. Petaurus breviceps— Sugar Glider

Abundance and distribution. Widespread and uncommon;
becomes progressively rarer to the north. Not recorded north
of the Little Desert.

Habitat. All open-forest and woodland alliances except in the
Little Desert where they appear to be restricted to Yellow Gum
woodland. Most common in open-forest or woodland with a
tall shrub understorey including Acacia mearnsii and Banksia
marginata. Of the seven observed during our survey three were
in Manna Gum open-forest and one was in each of River Red
Gum woodland, Yellow' Gum woodland, Messmate open-
forest and Brown Stringybark open-forest.

20. Pseudocheirus peregrinus —Common Ringtail Possum
Abundance and distribution. Restricted to the Volcanic Plains,
Coastal Plains, Tablelands and Mt Arapiles in the Wimmera
Plains. Common in the south becoming progressively less com¬
mon to the north. Not recorded in the Little Desert and the on¬
ly population on the Wimmera Plains is on Mt Arapiles. All
records, except those from Mt Arapiles, fall within the 650 mm
isohyet. This species was the most commonly observed ar¬
boreal mammal although it was recorded in only 36% of cells
where spotlighting occurred (Table 5).

Habitat. All open-forest alliances, River Red Gum woodland,
Manna Gum woodland and Silver Banksia open-heath with
emergent eucalypts. Of the 171 animals recorded 48% were in
Manna Gum open-forest, 25% in Brown Stringybark open-
forest, 19% in Messmate open-forest, 3% in River Red Gum
woodland, 3% in Manna Gum woodland and 0.6% in Silver
Banksia open-heath with emergent Eucalyptus nitida.

MAMMALS FROM SOUTHWESTERN VICTORIA

237

Macropodidae
Genus Macropus

Difficulties in field identification of Macropus spp. in western
Victoria have been discussed by Poole (1973), Emison et al.
(1978) and Poole et al. (1980). Two species (M. fuliginosus and
M. rufogriseus) occur widely in the survey area and M.
giganteus is restricted to the south. M. rufogriseus is distinctive
in colour and form. However, great difficulty was experienced
in distinguishing between M. giganteus and M. fuliginosus.
After careful field observations and the examination of
numerous road-casualties we consider that M. giganteus and
M. fuliginosus are sympatric between about 37°35'S and
38°00'S (sec also Poole 1977). Within this zone our specific
identifications are not certain and the distributions shown are
provisional. Morphometric data from a large series of skulls
would be necessary to accurately determine the extent of sym-
patry (Poole et al. 1980).

21. Macropus fuliginosus —Western Grey Kangaroo
Abundance and distribution. Widespread and common north
of about 38°S. Uncommon on the Tablelands where little
native vegetation remains. Recorded in 60% of cells surveyed
with chance sightings in a further 7.

Habitat. Brown Stringybark open-forest and low open-forest,
all woodland alliances, Broom Honey-myrtle open-scrub,
Yellow Mallee open-scrub, all open-heath alliances, farmland
adjacent to uncleared land and the edges of pine plantations.
Specimens. C24376, FWD 8995.

22. Macropus giganteus— Eastern Grey Kangaroo
Abundance and distribution. Restricted to the Volcanic Plains
and Coastal Plains south of about 37°35'S, where it is com¬
mon. Recorded in 16% of cells surveyed with chance sightings
in a further 5.

Habitat. All open-forest alliances. River Red Gum woodland,
Manna Gum woodland, Coast Wattle open-scrub, Silver
Banksia open-heath, farmland adjacent to uncleared land and
the edges of pine plantations.

23. Macropus greyi— Toolache Wallaby

Abundance and distribution. Extinct. Formerly locally com¬
mon in southeastern South Australia (Jones 1924) and,
possibly, adjacent parts of Victoria. Evidence of its presence in
Victoria since European settlement appears to be based entirely
on a footnote in Finlayson (1927, p. 366) as Jones (1924) con¬
sidered it endemic to South Australia.

Habitat. Described by Finlayson (1927) as ‘essentially clear
country, avoiding heavy timber and thick scrub.’ It was most
abundant in swampy depressions with Lepidosperma laterale,
Xanthorrhoea minor, Poa sp. and Themeda australis with
isolated clumps of Banksia marginata and B. ornata (Finlayson
1927).

24. Macropus rufogriseus- Red-necked Wallaby
Abundance and distribution. Widespread and common on the
Volcanic Plains. Also recorded at Mt Arapiles and NW of Kay
Swamp on the Wimmcra Plains; and the Broughtons
Watcrhole area in the Little Desert where it is rare and
restricted (cf. Gedye et al. 1979). One was also observed 12 km
S of Kiata in late November 1978 (P. Cheal pers. comm.).
Recorded in 45% of cells surveyed with chance sightings in a
further 8.

Habitat. All open-forest alliances, Manna Gum woodland,
Yellow Gum woodland and Silver Banksia heath. Grazes in
farmland adjacent to cover and in grassy firebreaks in pine
plantations. In the Little Desert all our sightings were in Yellow
Gum woodland around clay pans.

Specimens. Cl7590, C24855.

25. Potorus tridactylus— Long-nosed Potoroo
Abundance and distribution. Restricted to the Volcanic Plains
and southern Coastal Plains where the mean annual rainfall ex¬
ceeds 750 mm. Locally common.

Habitat. Messmate open-forest, Brown Stringybark open-
torest and Silver Banksia closed-heath with emergent
cucalypts. Invariably associated with a dense shrub layer.
Breeding. Trapping within the range of this species took place
in October and March. In October, 2 of 4 females trapped had
a single naked pouch young and in March, 2 of 3 females trap¬
ped had pouch young.

Specimen. C24377.

26. Thylogale billardierii —Red-bellied Pademelon
Abundance and distribution. Presumed extinct in Victoria.
Formerly present in coastal areas (NMV records) the only
records from the survey area being a specimen from Portland
(C6556) and a mandible of unknown age found in a blowout on
the dunes of Discovery Bay (C23668). Reasons for its decline
are not clear.

Habitat. Presumably dense coastal vegetation such as Brown
Stringybark open-forest, Scented Paper-bark closed-scrub and
Silver Banksia closed-heath.

27. VVallabia bicolor— Swamp Wallaby

Abundance and distribution. Probably no longer present in the
survey area. Remains have been found in several cave deposits
(Table 2) and Wakefield (1964a) had evidence of its presence in
the Lower Glenelg area around 1900. A recent specimen
(C17568), listed by Gedye et al. (1979, p. 120) as having come
from Mt Elgin, was actually collected at Mt Erip (37°45'S,
143°36'E) (FWD records) and sightings in Lower Glenelg Na¬
tional Park (Gedye et al. 1979) are considered doubtful.

Phascolarctidae

28. Phascolarctos cinereus —Koala

Abundance and distribution. Widespread and uncommon in
the south, the most northerly record being 37°15 , S. Formerly
more widespread, the present distribution reflects the FWD
restocking program following the general decline of the species
in the early 1900s (Warneke 1978). Between December 1952
and December 1978 approximately 192 Koalas were released in
the survey area. Details are given below.

Date

No.

released

Origin

Point of release

Dec. 1952

32

Phillip Island

Gorae Forest

Sep. 1953

33

Phillip Island

Tyrendarra

Dec. 1970

44

French Island

Lower Glenelg Nat. Park

Dec. 1970

44

French Island

Mt Richmond Nat. Park

Feb. 1973

15

Phillip Island

Mt Eccles Nat. Park

Nov. 1975

24

Phillip Island

Bats Ridges State Faunal
Reserve

Habitat. Messmate open-forest, River Red Gum woodland and
Manna Gum open-forest and woodland. Eucalypts known to
be eaten (Warneke 1978) which occur in the survey area are: E.
camaldulensis, E. obliqua, E. ovata, E. vimina/is and E.
microcarpa. We observed Koalas in the first 3 of these as well
as E. aromaphloia and E. baxteri.

VOMBATIDAE

29. Vombatus ursinus —Common Wombat
Abundance and distribution. Formerly widespread along
creeks throughout the Coastal Plains, e.g. Mosquito Creek (C.
Halahan pers. comm.) and in dune swales behind Discovery
Bay (J. Davies pm. comm.). Skeletal remains have been found
at Bats Ridges State Faunal Reserve (Anon 1907, C22349) and
Bridgewater Lakes (J. Seebeck pers. comm.). The population

238

P. W. MENKHORST AND C. M. BEARDSELL

at Bats Ridges died out in the early 1960s (J. Seebeck pers.
comm.) as did a colony at Malseed Lake (N. Learmonth pers.
comm, to .1. Seebeck). Now restricted to a few small, isolated
colonies in the Dorodong-Dergholm area and along Dry Creek,
S.A.; animals from this colony occasionally wander to the
banks of the Glenelg River as far east as Sandy Waterholes
(37°59'S, 141°01'E) (J. Davies pers. comm.). Recently several
animals from eastern Victoria were introduced to Bats Ridges
State Faunal Reserve (P. Kelly pers. comm.). Most colonies are
threatened as they are small and occur in isolated patches of
bushland on freehold land.

We were directed to occupied burrows at 4 localities: 7 km
NW of Dorodong; along Prospect Creek 6 km SE of
Dergholm; along Wombat or Wild Pig Creek 7.3 km NNE ot
Dergholm (includes the population adjacent to Bailleys Rocks
Scenic and Recreation Reserve) and along Dry Creek 6 km NW
of Nelson. In the Dergholm area the largest and most stable
population is the Wombat Creek colony (R. Howlett pers.
comm.) where numerous occupied burrows were found during
our visit.

Habitat. All remaining colonies arc along creeks, 2 in outcrops
of Glenelg Limestone and 2 in sandy loam soils. The Dry Creek
colony is in cleared pasture and Brown Stringybark open-
forest; the Prospect Creek colony is in a remnant of Brown
Stringybark open-forest with Pteridium esculentum domin¬
ating the understorey and the other 2 are in Manna Gum
woodland with a tali shrub layer and grassy ground layer.

Pteropodidae

30. Pleropus scapulalus— Little Red Flying-fox
Abundance and distribution. A rare vagrant to the survey area.
Only recorded from Heywood (undated specimen Cl9778) and
Dimboola and Quantong in November 1980 (M. Donaldson
pers. comm.). This species is a sporadic visitor to Victoria,
usually in summer and autumn.

Habitat. In Victoria usually associated with orchards or flower¬
ing eucalypts.

Molossidae

31. Tadarida australis-White-striped Mastiff-bat
Abundance and distribution. Widespread but uncommon in
the Little Desert and Wirnmera Plains, also recorded from Brit
Brit (Cl0738) on the Tablelands. Comprised 6% of 34 bats col¬
lected over Broughtons Waterhole, Little Desert between
February and April 1970 (FWD unpubl. data).

Habitat. River Red Gum and Yellow Gum woodland, Yellow
Mallee open-scrub and farmland. Requires tree hollows for
shelter and breeding.

Specimen. C21026.

32. Tadarida planiceps—Little Mastiff-bat

Abundance and distribution. Widespread and uncommon in
the Little Desert and Wirnmera Plains. Also recorded from
Portland in July 1970 (C17904). Comprised 6% of 34 bats col¬
lected over Broughtons Waterhole between February and April
1970 (FWD unpubl. data).

Habitat. River Red Gum and Yellow Gum woodland, Yellow
Mallee open-scrub and farmland with trees. Requires tree
hollows for shelter and breeding.

Specimens. C24335, C24337-9.

Vespertilionidae

33. Chalinolobus gouldii —Gould’s Wattled Bat
Abundance and distribution. Widespread and common. The
most commonly collected bat over Broughtons Waterhole, Lit¬
tle Desert between February and April 1970 (65% of 34 bats
collected, FWD unpubl. data).

Habitat. Brown Stringybark open-forest, all woodland
alliances, fringes of Yellow Mallee open-scrub and farmland
with trees. Requires tree hollows for shelter and breeding.
Specimens. C24307, C24309-10.

34. Chalinolobus morio —Chocolate Wattled Bat
Abundance and distribution. Widespread and common,
recorded from all physiographic regions. A total of 12 captured
(8%) in 28% of grids trapped. Comprised 9% of 34 bats col¬
lected over Broughtons Waterhole, Little Desert between
February and April 1970 (FWD unpubl. data).

Habitat. All open-forest and woodland alliances, fringes of
Yellow Mallee open-scrub and farmland with trees. Requires
tree hollows for shelter and breeding.

Specimens. C24304-6, C24308.

35-37. Eplesicus spp.

In the second half of the held survey identifications of the
species described by McKean et al. (1978) were attempted using
Hall & Richards (1979) but these arc provisional and all
Eplesicus forms are discussed together although they are map¬
ped separately and may now be specifically identified.
Abundance and distribution. Widespread and common.
Eptesicus spp. comprised 69.4% of all bats captured during
our survey and 18% of 34 bats collected at Broughtons
Waterhole, Little Desert between February and April 1970
(FWD unpubl. data). Of 102 Eptesicus captured 59% were
clearly E. vulturnus , 33% probably E. sagittula, and 8% pro¬
bably E. regu/us (Table 4). All three species probably occur
throughout the survey area.

Habitat. All open-forest and woodland alliances. All 3 species
require tree hollows for shelter and breeding.

Specimens. C24871, 24872 = E. regu/us ; C24315-18, C24331-4,
C24873 = E. sagittula ; C24311-14, C24319-30, C24874 = £.
vulturnus.

38. Miniopteris schmbersii- Common Bent-wing Bat
Abundance and distribution. Only recorded from the Volcanic
Plains and Coastal Plains. Distribution is centred on suitable
breeding caves, 4 of which occur in the survey area: Am¬
phitheatre Cave, Fern Cave and caves in Bats Ridges State
Faunal Reserve (Hamilton-Smith 1965); and Cave Hill,
Heywood (CS1RO Bat-banding records). Of these, only the
main cave at Bats Ridges houses a major colony with bats pre¬
sent year round, being used as a wintering site by part of the
larger population based at Naracoorte Caves (Hamilton-Smith
1965). On 24 June 1970 an estimated 1500 bats were present in
this cave (J. Seebeck pers. comm.). During our survey about 50
were present in Amphitheatre Cave on 18 October 1979.
Habitat. Requires caves for breeding and daytime shelter.
Dwyer and Hamilton-Smith (1965) describe the structure and
physical environment of maternity caves in southeastern
Australia. Forages in open-forest, woodland and farmland.
Specimens. C24343-4.

39. Myotis ad versus — Large-footed Myotis

Abundance and distribution. Recorded from the Coastal
Plains and Tablelands but its actual distribution is probably
wider than the few records suggest. Small numbers inhabit at
least 3 caves along the lower Glenelg River—Dry Creek Cave,
Amphitheatre Cave and Kales Slide Cave (McKean & Hall
1965). Seebeck and Hamilton-Smith (1967) highlighted the
vulnerability to disturbance of wintering colonies.

Habitat. Characteristically found near water from which in¬
sects are scooped with the large, strongly clawed feet (Dwyer
1970). Known roosting sites include caves (McKean & Hall
1965) and beneath bridges, aquaducts, storm-water tunnels etc.
(Dwyer 1970).

Specimen. C24870.

MAMMALS FROM SOUTHWESTERN VICTORIA

239

40. Nycticeius balstoni — Western Broad-nosed Bat
Abundance and distribution. Restricted to the Little Desert and
Wirnmera Plains; uncommon. Recorded only from Brough¬
tons Waterhole (C7479) and 10 km SW of Edenhope.

Habitat. Collected in Yellow Gum woodland and Yellow
Mallee open-scrub.

Specimen. C24336.

41. Nyclophilus geoffroyi — Lesser Long-eared Bat
Abundance and distribution. Widespread and common.
Recorded from all physiographic regions. Not collected over
Broughtons Waterhole in 1970 although specimens were found
nearby in old buildings and beneath bark (FWD unpubl. data).
Habitat. Brown Stringybark open-forest, River Red Gum
woodland, Yellow Gum woodland. Yellow Mallee open-scrub
and Pinus radiata plantation. Requires tree hollows for shelter
and breeding.

Specimens. C24340-42.

42. Pipistrellus tasmaniensis- Great Pipistrelle
Abundance and distribution. Only recorded from 9.5 km SSE
of Dartmoor in Lower Glcnclg National Park.

Habitat. Manna Gum open-forest. Elsewhere in Victoria it is
confined to open-forest and tall open-forest (FWD records).
Specimens. C24300-303

Muridae

43. Conilurus albipes— Rabbit-eared Tree-rat
Abundance and distribution. Extinct. The only record from the
survey area since European settlement is a specimen said to
have come from Portland Bay around August 1845
(Anonymous 1846). However, Mahoney (1982) discusses the
possibility that Mitchell collected a specimen between the Wan-
non and Stokes Rivers in 1835. May have been widespread in
the survey area up until the lime of settlement. Subfossil re¬
mains have been found in all bone deposits examined from the
survey area: McEacherns Cave; Fern Cave; Natural Bridge at
Mt Eccles and Mt Arapiles (Table 2).

44. llydroniys ehrysogaster— Water-rat

Abundance and distribution. Common and widespread. Dur¬
ing our survey it was recorded from the Wirnmera River at
Dimboola, Wannon River near Coleraine, Glenelg River near
Dergholm and Nelson, Moleside Creek and Piccaninny Blue
Pond. Presumably occurs in most streams, freshwater lakes
and permanent swamps in the survey area (cf. Gedye et al.
1979).

Habitat. Aquatic, occurring in freshwater streams, lakes, per¬
manent swamps and irrigation channels. Occurs in brackish
tidal reaches of the Glenelg River.

Specimen. C22252.

45. Mus musculus— House Mouse

Abundance and distribution. Widespread and common,
although uncommon in the Little Desert during our survey.
Numbers fluctuate greatly according to seasonal food
availability, level of predation, and suitability of the soil for
burrowing (Newsome 1969, Saunders & Giles 1977).

Habitat. Brown Stringybark open-forest and low open-forest,
all woodland types. Scented Paper-bark closed-scrub, Broom
Honey-myrtle open-scrub, Coast Wattle open-scrub, all heath
alliances, closed-grassland, farmland and commensal (with
man) situations. Trapping rates were highest in Coast Wattle
open-scrub, closed-grassland and regenerating Silver Banksia
open-heath. In the Little Desert it was using different habitats
to Pseudomys apodemoides , the only other murid present
(Table 10).

Breeding. Lactating females, males with distended scrota and
juveniles were trapped beside the Wirnmera River during early

April 1979, and juveniles were captured along the Glenelg
River in March 1980. Animals elsewhere were not breeding
during the survey.

Specimens. C22241, C22243, C22262-3.

46. Notomys mitchellii — Mitchell’s Hopping-mouse
Abundance and distribution. There is much confusion over the
possible occurrence of the species in the Little Desert. Brazenor
(1936) reported a sighting near Natimuk but this is not an ac¬
ceptable record (see also Wakefield 1966, p. 633). NMV
specimens labelled ‘near Horsham’ (NMV C2598-9) in fact
came from the NE end of Lake Albacutya (J. M. Dixon pers.
comm.). Burrows investigated by Brazenor in the Little Desert
S of Kiata yielded no animals and were almost certainly those
of Pseudomys apodemoides ( A. J. Coventry pm. comm.), not
/V. mitchellii as claimed by Brazenor (1958). The confusion
presumably arose because P. apodemoides was not known to
occur in Victoria at that time (Ryan 1963). The provenance of 2
specimens labelled Kiata (NMV C2866, C2841) remains ques¬
tionable and was not accepted by Wakefield (1966, 1974). A
specimen reported in the Nhill Free Press (15 Nov. 1955) as N.
mitchellii collected by J. Oldfield, 8 km SSW of Kiata, was in
fact P. apodemoides (NMV Cl5076). We agree with Wakefield
(1974) that records of N. mitchellii from the Little Desert are
not authentic. The southern limit of distribution is roughly a
line from the southern edge of Lake Hindmarsh to Bordertown
(Wakefield 1974).

47. Pseudomys apodemoides— Silky Mouse

Abundance and distribution. Not known in Victoria until 1963
when specimens collected by K. Hateley in 1957 were identified
(Ryan 1963). Uncommon and widespread in the Little Desert,
Wail Forest Reserve and Wirnmera Plains south to near
Dergholm (37°20'S). Southern limit of distribution cor¬
responds to the 650 mm isohyet.

Mean trapping success rate where captured was 4% (range
1-13). Trapping success peaked in August (5%) and April (7%)
and was lowest in June (2%).

Habitat. Recorded in Desert Banksia open-heath, Brown
Stringybark low open-forest on dunes, Broom Honey-myrtle
open-scrub and Yellow' Mallee open-scrub. In the Little Desert
it was most abundant in Desert Banksia open-heath and Brown
Stringybark low open-forest and was using different en¬
vironments to M. musculus during our survey (Table 10).
Cockburn (1981) showed that the species prefers dense low
vegetation containing Banksia ornata , beneath which most
burrows are located.

Breeding. Pregnant females w'ere trapped in October,
November and December, and lactating females in November.
One female gave birth to three young between 13 and 16
December 1979. Juveniles were trapped in February and in
April. Cockburn (1981a) also found that breeding in the Little
Desert occurred in late spring and summer and related this to
peaks in flowering and seed production. However, Finlayson
(1944) reported the breeding peak to be late autumn and early
winter in southeastern South Australia but also collected young
in November. Crouch (in Happold 1976) believes that litters
occur throughout the year but reproductive activity peaks in
August following winter rains. Sex ratios of trapped animals
varied seasonally. There was a strong female bias in April, Oc¬
tober and November and male bias in February and August.
Cockburn (1981a) also found a female bias at breeding (late
spring and summer) and suggested this may enhance coloniza¬
tion of regenerating habitat.

Specimens. C12610-11, FWD 8990, FWD 9971-2, C14053-7,
C21015, C21024, C21027-30, C22242, C22244, C22249-51,
C22253-5, C24348-51.

j

240

P. W. MENKHORST AND C. M. BEARDSELL

48. Pseudomys shortridgei — Heath Mouse
Abundance and distribution. Restricted to the Volcanic Plains
and Coastal Plains north to 37°25'S. Occurs only in areas hav¬
ing a mean annual rainfall of > 650mm (cf. P. apodemoides).
The entire population of this species may occur within the
range shown, together with The Grampians, as the population
in southwestern Western Australia may be extinct (Watts &
Aslin 1981).

Habitat. Brown Stringybark open-forest and low open-forest.
Silver Banksia closed-heath and open-heath, Scented Paper-
bark closed-scrub. Trapping success was higher in Silver
Banksia open-heath (8%) than in Brown Stringybark open-
forest (4%) or Scented Paper-bark closed-scrub (4%). Floristic
data were collected at 15 of the 41 trapping sites.

Commonly occurring plant species from these sites were:

Species

% Occurrence

* Banksia marginal a

80

Leptospermum juniperinum

60

Melaleuca squarrosa

60

*Epacris impressa

53

Xanthorrhoea minor

47

* Eucalyptus baxteri

40

* Leptospermum myrsinoides

40

* Xanthorrhoea australis

40

*Boronia pilosa

33

Eucalypts other than E. baxteri

33

* Hypolaena J'astigiata

33

* These species are characteristic of Silver Banksia open-heath
rather than the wetter closed-heaths preferred by, for example,
Antechinus minimus.

Breeding. This species was captured during March (7 in¬
dividuals), May (4), June (44), July (14), October (15) and
November (4). Little evidence of breeding was found except for
one juvenile captured on 15 November and a male with scrotal
testes on 26 July. Cockburn el al. (1981) have shown breeding
to be seasonal, occurring in spring and early summer.
Specimen. C24361.

49. Rattus fuscipes —Bush Rat

Taxonomy. All animals captured showed the pale pelage col¬
our and smaller dimensions and weight of the subspecies R.f.
greyi (Warneke 1971).

Abundance and distribution. Widespread and common on the
Volcanic Plains, Coastal Plains, western Tablelands and
southern Wimmera Plains. Limits of distribution approximate
the 650 mm isohyel. By far the most commonly trapped small
mammal (Table 2).

Habitat. All open-forest alliances, Manna Gum woodland,
Scented Paper-bark closed-scrub, Coast Wattle open-scrub,
Silver Banksia open and closed-heath and closed-sedgeland.
Breeding. Trapping within the range of this species took place
in all months except September, January and February. Lac-
tating females were trapped in March and May, and juveniles
or sub-adults between March and August. As juveniles were
absent in October, November and December most breeding
probably takes place in summer and autumn. Elsewhere
Warneke (1971) and Robinson (1976) have recorded breeding
peaks in summer.

Specimens. C13560, C22265, C24370-3, C24359.

50. Rattus lutreolus — Swamp Rat

Abundance and distribution. Widespread and common on the
Volcanic Plains, Coastal Plains, western Tablelands and
southern Wimmera Plains. Limits of distribution approximate
the 650 mm isohyel.

Habitat. Messmate open-forest. Brown Stringybark open-
forest and low open-forest, Manna Gum open-forest. Scented

Paper-bark closed-scrub, Coast Wattle open-scrub, Silver
Banksia closed-heath, some Silver Banksia open-heath, Desert
Banksia open-heath (once only), Blue Tussock Grass closed-
grassland, and roadside grassland.

Breeding. Juveniles were trapped between late August and late
March, a longer period than that reported by Braithwaite &
Lee (1979) for Cranbourne. Braithwaite (1980) has shown that
the length of the breeding season varies considerably between
areas and with variations in food availability in ditferent years.
Specimens. C22266, C24384-8, FWD 12035.

51. Rattus rattus—Black rat

Abundance and distribution. Usually absent in undisturbed
native vegetation, but occasional local concentrations occur.
Probably widespread throughout the survey area, particularly
near towns and other human habitation. During our survey,
recorded only from: beside the Wimmera River near Dim-
boola, Mt Arapiles and on coastal dunes.

Habitat. Recorded in dense stands of Phragmites communis
beside the Wimmera River, in rock screes on Ml Arapiles and
in Coast Wattle open-scrub on coastal dunes.

Specimens. C22256-61, C24360.

Canidae

52. Canis familiaris-Feral Dog and Dingo

Abundance and distribution. Widespread and uncommon.
There arc a few early records of dingoes from the survey area
(cf. Gedye et al. 1979). E. Townsend reported dingoes in the
Nhill area in the 1860s (Blake 1976, p. 2) and French (1888,
1888a), Le Souef (1887), D’Alton (1913) and Hamilton (1914)
all mention wild dogs as pests and the methods used to destroy
them. Dingoes (C. familiaris dingo) are now extinct in the
survey area but feral dogs occasionally occur in the Little
Desert (C. Brownsea pers. comm.), Lower Glenelg National
Park and Discovery Bay Coastal Park (J. Davies pers. comm.)
and presumably in other large tracts of busliland. Recorded in
0.8% of grids surveyed.

Habitat. Formerly all terrestrial environments. Recent records
are from Yellow' Malice open-scrub, Coast Wattle open-scrub
and Messmate open-forest.

53. Vulpes vulpes —Fox

Abundance and distribution. Widespread and common.
Recorded from all physiographic regions.

Recorded in 30% of cells surveyed with chance sightings in a
further 10.

Habitat. All terrestrial environments except closed-scrub and
closed-heath.

54. Felis catus —Feral Cat

Abundance and distribution. Widespread and uncommon,
more common in the south. Recorded from all physiographic
regions. Free-ranging cats were observed in 20% of cells
surveyed with chance observations (mainly road mortalities) in
a further 14.

Habitat. Observed in all open-forest and woodland alliances,
pine plantations, Coast Wattle open-scrub. Silver Banksia
open-heath and farmland.

Bovidae

55. Capra hircus — Goat

Abundance and distribution. Restricted to the south-east where
a feral population is centred on The Stones State Faunal
Reserve (Emison et al. 1978).

Habitat. Manna Gum woodland with a grass and bracken
understorey growing amongst tumbled granite boulders.

56. Ovis aries —Sheep

Abundance and distribution. Small numbers of feral sheep

MAMMALS FROM SOUTHWESTERN VICTORIA

241

were present in the Little Desert until recent years (G. Edwards
pers. comm.). These animals were presumably escapees from
nearby farms. Flocks were agisted in the Little Desert heaths
during droughts up until the early 1900s when the practise was
discontinued (D’Allon 1913). A feral population exists in The
Stones State Faunal Reserve (Emison et al. 1978).

Habitat. Desert Banksia open-heath, Yellow Gum woodland
and Brown Stringybark low open-forest in the Little Desert and
Manna Gum woodland in The Stones State Faunal Reserve.

57. Dania dama — Fallow Deer

Abundance and distribution. Restricted to the area around
Lake Mundi where local farmers have made several recent
sightings (R. Keeble pers. comm.). Presumably these animals
originate from a population in the Pinnaroo-Bordertown-
Naracoorte area of South Australia (Bentley 1978, p. 91).
Habitat . Yellow Gum woodland and River Red Gum woodland
where fresh grass and browse is available.

58. Lepus capensis—Brown Hare

Abundance and distribution. Widespread and generally un¬
common. More common in the north. Recorded in all
physiographic regions. Recorded in 5% of cells surveyed with
chance sightings in a further 8. Introduced to Victoria in the
1860s (Le Souef 1965).

No information on the spread of Brown Hares into the
survey area is available; however, they apparently did not

become well established until after the initial plague of rabbits
had decreased in the early 1900s.

Habitat. All woodland alliances, fringes of Yellow Mallee
open-scrub and farmland. Absent from areas with dense shrub
or ground layers.

59. Oryctolagus euniculus— European Rabbit
Abundance and distribution. Widespread and common.
Recorded in 67% of cells surveyed with chance sightings in 79
others. First introduced to the Wimmera at Yannock Station,
NW of Kaniva, in 1860 (Rolls 1969, p. 22). Released at Morton
Plains, 25 km N of Donald, in 1866, and had reached plague
proportions at Lake Buloke in 1878 by which lime rabbits had
spread W to the Wimmera River, E to the Avoca River and N
to the Murray River (Rolls 1969, p. 37). Campbell (1884) refers
to a plague in the Lawloit Range in 1884 and French (1888)
states that the scrub around Serviceton was swarming with rab¬
bits. They were a pest in the Lake Albacutya area in the 1880s
(French 1888a) where large numbers were killed in the dry
season with poisoned water (Lc Souef 1887).

Little information is available on the spread of rabbits fur¬
ther south but they were present in the Stony Rises in the mid
1870s (Rolls 1969, p. 38) and had probably spread through
most of the survey area by 1880 (C. Halahan pers. comm.).
Habitat. All terrestrial environments except those with dense
ground or shrub layers.

242

P. W. MENKHORST AND C. M. BEARDSELL
APPENDIX 2

Distribution Maps

Grid lines are at intervals of 5' of latitude and longitude.

• —records made during FWD survey. O —records prior to
October 1974. O, ® —general locality record.

n r

•

•

1 «

1 1

>•<

»

!•!

•

<

i

d

i

1

J

• <

|l

r

d

iLi

1

j'

•

••

•<

• <

•

--j

t

»•

1

i

>•<

1

S

•

•

i

J

•

1

X

- <

1

•

i

141°

J_

\

4 ,

‘I

%

iS

142°

_L

1. Tacky glossus
aculeatus

2 . Ornithorhynchus
anatinus

3 . Antechinus
flavipes

U Antechinus
minimus

T"

"T

1

>

P

- $
vf

O

A

_

o

O *

KO

-

■A

1 —

—4

J

s

J_

A

— 1 _

5 . Antechinus
stuartii

6. Antechinus
swainsonii

7. Dasyurus
maculatus

9 . Phascogale
tapoatafa

MAMMALS FROM SOUTHWESTERN VICTORIA

243

10. Sminthopsis
crassicaudata

11. Sminthopsis
leucopus (lower)
12.5. murina
{ upper)

14 Trichosurus
vulpecula

15. Acrobates
pygmaeus

16. Cercartetus

GOncinnus (upper)

17 C. nanus ( lower)

18- Petaurus
australis

19. Petaurus
breviceps

244

P. W. MENKHORST AND C. M. BEARDSELL

20. Pseudocheirus
peregrinus

21. Macropus
fuligvnosus

2 2 . Macropus
giganteus

24. Macropus
rufogriseus

n

■r

k

i_

pi

25. Potorous
tridactylus

28. Phascolarctos
driereus

oP

29. Vombatus
ursinus

MAMMALS FROM SOUTHWESTERN VICTORIA

245

30. Pteropus
scapulatus

31. Tadarida 32. Tadarida

australis planiceps

33. Chalinolobus
gouldii

34. Chalinolobus 35. Eptesicus
morio vultumus

36. Eptesicus
sagittula

37. Eptesicus
regulus

246

P. W. MENKHORST AND C. M. BEARDSELL

38 .

schreibersii

39 . Myotis adversus

(middle)

40 . Nyctioeius
bats t m ' (upper)

42 . Pipistrellus
tasmaniensis

(lower )

47 . Pseudomys
ap o demo ides

(upper)

48 . P. shortridgei

( lower)

41 . Nyctophilus 44 . Hydromys

geoffroyi ehrysogaster

49 . Rattus
fuseipes

r

Ik

XLIHN

W

FP,

o_

50 . Rattus

lutreolus ;

MAMMALS FROM SOUTHWESTERN VICTORIA

51 . Rattus rattus 52 . Cards

familiar is

5 5 . Capra hircus
56 . Ovis arias

ROYAL SOCIETY OF VICTORIA

1982

Patron:

President:

Vice- Pres id en ts:

Immediate Past President:
Hon. Secretary:

Hon. Treasurer:

Hon. Librarian:

Hon. Editor:

Hon. Research Secretary:

His Excellency Sir Henry Winneke, KCMG, OBE, QC,
KStJ, Governor of Victoria.

G. D. Aitchison, ME, PhD.

D. M. Churchill, BSc, BA, PhD.

Professor J. W. Warren, MA, PhD.

Professor L. L. Stubbs, DAgrSc.

T. A. Darragh, MSc, DipEd.

H. G. Stevens, FCA.

J. G. Douglas, MSc, PhD.

L. A. Frakes, MA, PhD.

I. A. Staff, MSc, PhD, DipEd.

Hon. Development Manager: Professor G. Seddon, BA, MSc, PhD.

Hon. Assistant Secretary: P. A. Jell, BSc, PhD.

The above Office-bearers are ex officio members of the Council of the Society.

Other members of the Council are:

J. K. Aitken, LLM.

Sir Robert Blackwood, MCE, BEE.
Professor M. .1. Canny, MA, PhD.

M. G. Lay, BCE, MEngSc, PhD.

M. J. Littlejohn, PhD.

Professor J. F. Lovering, MSc, PhD.
Professor J. D. Morrison, PhD, DSc, FAA.

A. E. Perriman, BSc.

Mrs. Delys B. Sargeant, BSc, MEd.

G. A. Sklovsky, IngChem, ESCIL, LiciSci, IngDoc.
J. H. Thompson, B.Com.

Executive Officer:

R. R. Garran, MSc, PhD.

Honorary Architect: F. Suendermann, FRAIA, MRAPI.

Honorary Financial Adviser: D. S. Clarebrough.

Honorary Solicitors: Phillips, Fox and Masel.

Honorary Auditor: I. J. Curry, AASA.

Trustees: D. M. Churchill, BSc, MA, PhD.

Professor J. S. Turner, OBE, MA, PhD, MSc, FAA.
Professor J. W. Warren, MA, PhD.

The Hon. Mr. Justice A. E. Woodward, LLM, QC, OBE.

1

-.»

\ i .V-
V 'V5ov.-.

)