VOL. 100, PART 1

28 FEBRUARY, 1976

TRANSACTIONS OF THE

ROYAL SOCIETY
OF SOUTH AUSTRALIA

INCORPORATED
CONTENTS
Oliphant, Sir Mark The Second Century - - - - - - - 1
Tyler, M. J. Comparative Osteology of the Pelvic Girdle of Australian Frogs
and Description of a New Fossil Genus - - - - 3
Houston, T. F. New Australian Allodapine Bees (Subgenus Exoneurella

Michener) and their Immatures (Hymenoptera: Antho-
phoridae) - - - - - “ - - - - 15

Plummer, P. S., and Gostin, V. A. Faulting Contemporaneous with Umberatana

Group Sedimentation (Late Precambrian), Southern Flinders
Ranges, South Australia - - - - - - - 29

Smith, Meredith J. Small Fossil Vertebrates from Victoria Cave, Naracoorte,

South Australia. IV. Reptiles - - - - - - 39

PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS
STATE LIBRARY BUILDING
NORTH TERRACE, ADELAIDE, S.A. 5000

THE SECOND CENTURY

Summary

The Royal Society of London was conceived more than 300 years ago when, following a lecture by
Christopher Wren, it is recorded that "something was offered about a designe of founding a
Colledge for the promoting of Physico-Mathematicall, Experimentall Learning". Charles I] became
interested and granted the Society its first Charter in 1662, naming it "The Royal Society". A
second Charter, in 1663, granted the body Arms, bearing the motto "Nullius in verba", as an
expression of the determination of the Society to resist all dogma and to verify all information and
statements by appeal to observation. At the same time, the full title of the Society became "The
Royal Society of London for Improving Natural Knowledge”.

THE SECOND CENTURY

The Royal Society of London was conceived more than 300 years ago when, following
a lecture by Christopher Wren, it is recorded that “something was offered about a designe of
founding a Colledge for the promoting of Physico-Mathematicall, Experimental] Learning”.
Charles II became interested and granted the Society its first Charter in 1662, naming it “The
Royal Society”. A second Charter, in 1663, granted the body Arms, bearing the motto
“Nullius in verba”, as an expression of the determination of the Society to resist all dogma
and to verify all information and statements by appeal to observation. At the same time, the
full title of the Society became “The Royal Society of London for Improving Natural Know-
ledge”.

Sir Mark Oliphant, K.B.E., Patron 0} the
Royal Society of South Australia

A little more than 200 years later, the Royal Society of South Australia was born, and
1976 is the centenary of its Transactions. Throughout this time the South Australian Society
has played an invaluable role in recording, in proper scientific manner, manifold observa-
tions and descriptions of the geology, the fauna and the flora of this State. For long, the
Society has emulated in its activities the search for natural knowledge which was the objec-
tive of the founders of the original Royal Society. However, for very practical reasons, such
bodies are reappraising their activities and ever questioning whether they can continue to
exist. Strangely enough, the problems arise from the very success of their past work. The
technology, the application of knowledge for practical purposes, which arises from increasing
understanding of nature, has made of science an ever growing national and international
activity. Through their control of money, governments now determine the level of scientific
enquiry and its content, while the rapidly increasing cost of publication means that they
become the arbiters of what science shall be printed.

1s

SIR MARK OLIPHANT

There are other factors which have a pronounced influence on comparatively small
societies allempting to cater for those interested generally in science, Foremost is the growth
of specialist socicties interested only in a very marrow area of natural knowledge. It is almost
impossible today to attract a biochemist to a discussion of relativity or cosmology, or to per-
suade a nuclear physicist that the nature of the chemical bond can be an interesting subject
of study, The particle physicist cannot understand the entomologist, or the computer elec-
tronics expert the ornithologist. A broad interest in nature generally is now rure. There are
few today whose interests are as broad as those of Lord Rayleigh, who wrote learnedly of
the hending of marble mantlepieces and developed the mathematical theory of the wave-
guide, Publicution of the results of scientific enquiry becomes more and more confined to
specialist journals, or specialist sections of older journals.

Secondly, it is the national badies, like the Academies of Science of the United States and
Russia, which are gavernment agencies, or the Royal Society of London and the Australian
Academy of Science, which are not controlled directly by government. which represent
scienee internalionally for their countries, and which receive the greatest financial backing, Tt
is they Who can afford to organise conferences, national or international, they who can pro-
vide the expenses for travel by members and by visitors. They are prestigious societies whose
activities are of world-wide significance.

The Royal Society of South Australia cannot emulate these giants, However, T any
convinced that it will have @ vital part to play in the second century of its existence. The
growing interest, among scientists and the general public, in the preservation of the environ-
ment, in pollution, ecology, our national heritage of fauna and flora, natural resources, and
the beauty of landscape, encourages belief jn a return to deep interest in nature generally.
rather than in ohe aspect only, Tf, as C believe, our Royal Society is ta be revitalized to
hecome a significant social influence through scien¢e, it must espouse catlises, after full and
frank discussion to determine an agreed approach, and then speak loudly and in public of its
conclusions,

As mstances of broad issues to which the Society could contribute much, there are
preservation of the Mf Lofty Ranges and the Flinders Ranges, the salinity of the River
Murray, the deterioration of our desert areas, We should not be deterred by the existence of
CS.LR.0., of State instrumentalities, or of the Australian Academy of Science, from
choosing areas of comprehensive, interdisciplingary study, which can be exciting scientifically
and could be rewarding socially. A paper describing a new species of coleoptera from the
Lake Frome area should be published, but will not rouse much interest in a general scieutitic
audience. On the other hand, if this discovery is significant in indicating climalic change, or
ecological upset On the castern borders of the Flinders Ranges, it could become an important
contribution to wide discussion,

South Australia faces a multitude of problems, most of which are relegated to investi-
galian by governmental bodies. Qur tuiure depends erilically upon assured supplies. of water
and of energy. Water, gas, and elcetricity authorities are too busy with immediate questions
to exercise the imaginative resypanses necessary for creation of new approuches to longer
term problems. On the other hand, a Royal Saciety which stimulates constructive discussion
of such problems, in their widest context, might generate completely new ideas.

Where are We going? Thatis the first questiam we must answer in the early part of our

second hundred years.

GovERNOR oF Sourht AUSTRALIA

COMPARATIVE OSTEOLOGY OF THE PELVIC GIRDLE OF
AUSTRALIAN FROGS AND DESCRIPTION OF A NEW FOSSIL GENUS

BY M. J. TYLER*

Summary

TYLER, M. J. (1976).-Comparative osteology of the pelvic girdle of Australian frogs and
description of a new fossil genus. Trans. R. Soc. S. Aust. 100(1), 3-14, 28 February, 1976.

The osteological characteristics of the pelvic girdle of twenty-five extant genera of Australian frogs
of the families Hylidae, Leptodactylidae, Microhylidae and Ranidae are defined. The new Tertiary
fossil genus and species Australobatrachus ilius are described from the Etadunna Formation.
The fossil exhibits a unique lateral ilial groove and is referred tentatively to the family Hylidae.

COMPARATIVE OSTEOLOGY OF THE PELVIC GIRDLE OF AUSTRALIAN
PROGS AND DESCRIPTION OF A NEW FOSSIL GENUS

by M. J. TyLer*

Summary
‘Tyter, M. JI. (1976)—Comparative ostealogy af the pelvic girdle of Australian frogs and
description of a few fossil genus. Trans. R. Sac. 8, Aust, 100(1), 3-14, 28 February,

1976.

The osteological characteristics of the pelvic girdle of twenty-five extant genera of Aus-
tralian Trogs of the families Hylidae, Leptodactylidae, Microhylidac and Ranidge are defined.
The new Tertiary fossil genus and species Australobutrachus ilius are described from the
Fiadunna Formation, ‘Ihe fossil exhibits a unique lateral ilial groove and is referred tentatively

to the family Hylidae,

Introduction

Of all bones of the anuran skeleton, the
ilium has been shown to vary considerably and
consistently between families, genera and even
Species, Ilia are commonly well preserved
amongst disarticulated skeletal material, and
their features are sufficiently diagnostic to per-
mil identification; for this reason ilia have pro-
vided the basis for the recognition of genera
and erection of new fossil species: (Lynch 1963;
Chantell 1964; Holman 1965),

Dala on extant species so essential for identi-
fication and general comparative purposes are
frequently limited, and in the case of Australian
frogs, data are particularly deficient. Lynch
(1971) provides the only comparative contri-
bution. Confining his interest to leptodactylids,
Lynch described the ilia of representatives of
nine genera. Since then a fossil ilium of a pre-
viously undescribed genus bas been reported
from the Australian Tertiary by Tyler (1974).
uml it has proved necessary to examine und
describe representative ilia of all of the known
living genera in order to describe the new genus
and species,

Material and methods
The dry specimens of the modern species
studied were dissected from representatives of
4 families, 25 genera and 60 species, including
ull geneta known in Australia, This material is
in the author's collection. The fossil described

herein is in the Palaeontology collection of the
South Australian Museum.

With only minor variation, the descriptive
terminology used follows Lynch (1971), and
the features recognised are shown in Figure 1,
Morphometric data were obtained with dial
callipers or an eyepiece micrometer. The length
of the animals from snout to verit Was measured
before dissection. Subsequently, the distances
between the tip of the dorsal acetabular expan-
sion atid the end of the ilial shaft and the span
between the anterior margin of the dorsal pro-
minence and the ventral acetabular expansion
were measured, The bone measurements were
examined to establish relationships between ilial
size Or proportions and the size of the donor
frog.

Features of the Anuran pelvis

(A) Puhis

The pubis is customarily a small, roughly tri-
angular, cartilaginous wedge of tissue separat-
ing and underlying the ventral borders of the
ischium and ilium (Fig. 1). In particularly
large specics. (rarely in small ones), where
there js a more intimate degree of fusion of the
pelvic components, the pubis is often calcified
or ossified.
(B) lyehinm

The ischium is .a bony or cartilaginous disc
anteriorly fusing with the ilium to provide the
posterior half of the acetabulum, 4nd pos-

* South Australian Museum, Nocth Terrace, Adelaide, S. Aust. SOD0. (Present address Department of
Zoology. University of Adelaide, Adelaide, ‘S. Aust, 30011),

4 M. 7. TYLER

Nial
shaft Oprom. Dacetexp
D protub, ae
\ p »

pre-acet, Zone
Vacet.exp, Ischium
Pubis

Acet. fossa

Fig, 1, Left lateral aspect of anuran pelvic girdle.
Abbreviations: Acct, fossa-—aceiabular
fossa; 0. acet. exp—dorsal acetabular
expansion; D. prom.—dorsal prominence:
D. protub,-—dorsal protuberance; pre-iacct.
zone—pre-acetubulur zone; V. acel. exp.—
ventral acetabular expansion.

teriorly forming a plate occupied by numerous
mus¢les communicating with the femur,

The shape of the ischium varies considerably
at a specific level and is associated with dif-
ferences in habits. For example, the develop-
ment of a particularly large plate posterior to
the acetabular fossa ts found only in rotund
apecies with exceptionally short and muscular
hind limbs. Similarly, the development of a
high, dorsally projecting extension is charac-
teristic of large but ugile frogs with powerful
hindlimbs. Viewed with the femur as the
reference point it is clear that a major shift in
the inuscle mass can only be brought about by
voncurtent changes in the development of the
ischium.

(C) Sliurm

The ilium is the, largest pelvic bone, and con-
sists of an elongate shaft terminating in an axe~
head shaped body. It is a paired stricture
articulating anteriorly with the ventral surface
of the sacra] diapophysis and posteriorly abut-
ting the ischium and pubis. Various areas of the
tlium ate recognisable as distinct components -
(a) MMial shaft: The ilial shaft is an elongate
and usually slightly curved structure varying
from «a cylindrical section through vertically
oval to more elaborate forms in which there
are grooves on the lateral or medial surfaces.
(b) Dorsal crest: A smooth bordered crest, ris-
ing high as a thin blade, occurs in numerous
species of frogs, but is rare amongst Australian
forms. It usually arises from the dorsolateral
surface of the shaft, and reaches its maximum
height within the anterior one-third of the shaft
(Fig. 4L),

(c) Dorsal prominence: The dorsal prominence
is a diflerentiated area rising on the superior
margin of the shaft in a position above or
slightly anterior to the acetabulum. [t is not
present in all species and Js scarcely detectablc
in many others (¢.z. in those exhibiting a dor-
sal crest)_

(d) Dorsal protuberance; From the body of
the dorsal prominence the dorsal protuberance
arises as an elongate, rounded or pointed knab.
To it uttaches the Muscutus gluteus maxirnus.
(e) Dorsal agetabular expansion: The body of
the iliim extends dorsally into. a triangular por-
tion of bone superior and posterior to .the
acetabulum. This projection is termed the dor-
sal acetabular expansion. It may rise steeply
atid so meet the ischium in a vertical plane,
whilst the anterior face can be at an acute or
obtuse angle to the ilial shaft,

(f) Acerabulum: Variation in the acetabulum
consists of differences in size (relative to adja-
cent structures), in position in relation to the
ilial shaft, and in the width and extent of
development of the acetabular rim,

(g) Ventral acetabular expansion: The superior
segment of the ventral acetabular expansion
has heen termed the preacetabular zone by
Lynch (1971), The considerable variation in
this portion of the pelvis is difficult to express,
because of the instability of potential reference
points such as the ucetabulum, Nevertheless,
the basic shapes range from the form of a
straight line extending ventrally and posteriorly
(Fig. 24, A), a gradual concavity (Fig. 2B},
or a concavity of the preacetabular zone and
a convexity bencati (Figure 2C).

Se =

A YY B ( iy
= a
C ( D “Sy .

Fig. 2. Variation in the shape of the ventral
Acetabular expansion viewed from left
aspect. See text for explanation,

PELVIC GIRDLE OF AUSTRALIAN FROGS

MN

Fig. 3. Pelvis or isolated ilium of hylid and leptodactytid frogs:
A, Litoria caerulea, x 2; B. L. lesueuri, x2; C. L, eucnemis, x 5; D. Nyetimystes zwiefeli, x 2:
E. Adelotus brevis, x 5; F. Assa durlingtoni, x 5; G. Crinia georgiana, x 5+ HH. Cyclorana nevae-
hallandiae, x 2; I. Geocrinia laevis, x 5; J, Glauertia orientalis, x 5; K, Heleioparus albopunctatus,
x 2; L. Kyarranus sp., x 5; M. Lechriodus fletcheri, x 5; N. Limnodynastes peroni, x 5; O. Mixo-
phyes fasciolatus, x 2.

POST-MORTEM CHANGES

Disintegration of the pelvic girdle into its
component bones has been observed in species
in which cartilage is most extensive. Species
that are heavily ossified remain intact.

During the process of dehydration, the
acetabular fossa may become distorted and,
in the material available, such distortion takes
the form of compression in « dorsoventral
direction. The dorsal crest of the ilium is par-

G M, J. TYLER
TABLE 1
Characteristics af ihe Uiant tn species a} Litoria
Dorsal pro-
minence/ Acetabular Position
Profile of yentral anterior rim rim ventral of medial
acetubular of ilial shaft rim on
Species expansion Acetabulim margin Pubis Tlivm
adelaidensis concaye level above curt, ant, t
qneiana* concayo-conyex posterior level cart. absent
dured concave 4 antenor above curt. ant. 7
hicoler concave 4 anterior above cart, Tabatnt
hoeraolangertsis concave-canvex 4 anterior level bony central 1
brevipalmaia concave level abave curt, ant, Ly
caerulea concave level level curt. ubsent
citropa concive level above cart, ant.
darsalis® concavo-convex 4 anterior level curt. aot. ¢
eucnemis concavo-convex level above bony ant. 4
ewingi concave level above curt Post. +
gracileita concavo-conyex 3 anterior above cart. absent
infrafrenata concaye + anterior above cart. anh. 3
lesiteuri concavo-convex + antetior level bony central 43
microbelos concwo-convex level above Gurl. ant, 4
nannotis concaye-conyex level above bony absent
nasula concavo-canyex 4 anterior above carl. ant. 14
nizrofrenata concayo-conyvex \ amterior above bony ant, $
rubella concave 4 anterior above vart. absent
rothi concave + qnlerjor above curt. absent
thesaurensis* conceave-convex leyel above bony ant, 3

© Snecics restricted to New Guinea.

ant—anterior; cart—aartilaginous.

ticularly subject to post-mortem distortion,
commonly bending medially from a perfectly
vertical orientation to form a quadrant, Even
more conspicuous is the distortion amongst
material recovered fram owl pellets where
there has been an induced medial curvature of
the ilial shaft in. several specimens.

Account of modern genera and species
Family HYLIDAE
LITORIA Tschudi

FIG. 34-C

Species examined: L, adelaidensiy (Gray). L.
angiana (Boulenger), L. aurea (Lesson), LZ,
bicolor (Gray), L. boaroalongensis (Moore),
I... hrevipalimata Tyler, Martin & Watson, E.
caerulea (White). L. citrapa (Tschudi). L. dor-
salixs Macleay, L. eucnemis (Lonnberg), L,
ewingi Dumeril & Bibron, ZL. gracilenta
(Peters), L. infrafrenara (Gunther), D. lesneuri
Dumieril & Bibron. L, micrebelos (Cogger). L,
nannotis: (Andersson), £, naswia (Gray), L.
nigrofrenata (Gunther), 1. rubella (Gray), L,
rothi (de Vis), L. thesaurensis (Poters).

Variation in this morphologically and ecola-
gically diverse genus renders wa generic defini-
tion a difficult proposition. For this reason
comparative data arc included in Table 1, and
only the following generalisations ure possible:

The pubis is cartilaginous or ossified and the
ischium is ossified.

The ilial shaft lacks a dorsal crest but in-
variably bears a narrow tim on at least a por-
tion of the medial surface. In L. aurea there
is also a lateral groove exhibiting a distinct
ontogenetic trend in hecoming progressively
less conspicuous, The acetabular Jossa tends to
be rather large. The ventral acetabular expan-
sion is of a varicty of forms, from a narrow,
concave profile to concave-convex,

The dorsal protuberance and dorsal
prominence are usually well differentiated but
are not raised high above the level of the ilial
shaft.

NYCTIMYSTES Stejneger
FIG. 3D

Species examined: N, tympanocryptis (Ander-
sson and N. zweifel? Tyler of New Guinea.

PELVIC GIRDLE OF AUSTRALIAN FROGS 7

TABLE 2
Generic features of Tlia

rr

Dorsal prer-
minence/
lial shaft Ital shaft Dorsal anterior rim
cresh rim protuberance of acétabulum
Adelotus. absent present prominent anterigr
Assa absent absent inconspicuous level
Caphixalus absent absent inconspicuous anterior
Crinia absent present inconspicuous anteriar
Cyclorana absent present or absent incanspicuous anterior
Geocrinia absent absent absent level
Glauertia absent absenol inconspicuous anterior
Heleioports absent absent prominent anterior
Kyarranus absent present prominent anterior
Lechriodus present absent moderate posterjor
Limnodynastes present or absent present or absent prominent ant. or Jevel
Litoria absent present or absent moderate usually anterior
Mixophyes present absent inconspicuous anterior
Myobulrachus absent absent absent posterior
Neobatrachus absent absent prominent anterior
Notaden absent absent prominent posterior
Nyctinystes absent absent moderate level
Philoria absent. present inconspicuous anterior
Pseudophryne absent absent moderate level
Rana present absent inconspicuous posterior
Ranidella absent absent moderate anterior
Rheobatrachus absent absent prominent posterior
Sphenophryne absent absent iNconspicuous posterior
Taudactylus absent absent prominent anterior
Uperoleia absent absent prominent anterior

—___—_—_—_——————OOOOOO Cc oO — — — —

The ilium and ischium are ossified in both
species. The pubis is ossified in zweifeli und
Cartilaginous in tympanocryptis.

The ilial shaft is long, curved distally and
very slightly compressed mediolaterally. The
ventral acetabular expansion is gently rounded
in a single, uninterrupted concave arc. The
acetabular fossa is prominent, with its. upper
margin level with the centre of the ilial shaft,
The dorsal prominence and dorsal protuberance
are small and project laterally rather than
superiorly. The anterior margin of the dorsal
protuberance is on a level with the anierior
margin of the acetabular rim, The dorsal acc-
tabular expansion is only slightly raised.

Family LEPTODACTYLIDAE
ADELOTUS Ogilby
FIG, 3£
Species examined: A. brevis (Gunther).

The ischium is bony .and the pubis is entirely
cartilaginous.

The ilial shaft is distinctly curved and bears
a narrow indentation on the medial surface.
This indentation is deepest in the midsection of
the shaft. The acetabulum has a narrow
peripheral rim which superiorly is on a. level
witb or is slightly superior to the ventral surface
of the ilial shaft. The ventral acetabular expan-
sion is only slightly developed, and the preace-
tabular zone is extremely narrow. The dorsal
acetabular expansion is elongate and raised
moderately. The dorsal prominence is poorly
defined. The dorsal protuberance is extremely
large, inclined posteroventrally and is almost
entirely anterior 10 the anterior rim of the
acetabular fossa.

ASSA Tyler

FIG, 3F

Species examined: A. darlingtoni (Loveridge).
The pubis and ischium are bony except for

the portion associated with the posterior half of
the acetabular fossa.

3 M. J. TYLER

The ilial shaft is slightly curved, lacks ridpey
and indentations and is circular in cross section,
‘The acetabulum has an exceptionally Well
developed penpheral rim which superiorly is
very slightly ahove the ventral margio of ihe
ilial shalt, Lhe Ventral acetabular expansion ts
slightly develaped inte a narrow preacetabular
zone. The dorsal acetabular expansion is Very
poorly developed. The dorsal prominence is
only slightly defined. The dorsal protuberance
is small bul promlitent, its anterior margin on
o level with the uuterior margin of the scctabu-
lar rim.

CRINIA Tschudi
FIG. 3G
Species examined; C, georgiana Tschudi.

The pubis is cartilaginous and the ischium ts
oasified,

The ilial shaft is curved, flattened medio-
laterally over the posterior half and dorsoven-
trally over the anterior half. There is no dorsal
ilisl crest but there is a very slight longitudinal
medial indentation. The acetabulum is large,
has a fairly broad peripheral mim which
superiorly is very slightly above the level of the
ventral margin of the ilial shaft.. The ventral
acetabular expansion is only slightly develaped;
the subacetabular zone does not protricde an-
teriorly. The dursal acetabular expansion 13
poorly developed. The dorsal prominence is
low, but guile distinguishable from the ilial
shaft. The dorsal protuberance is just
detectable. Slightly less than one-half of the
dorsal protuberance ty anterior to the anterior
Tim of the acetabulum.

CYCLORANA Stcindachner
FIG, 3
Species examined: C. australis (Gray), C. dalili
(Boulenger}, C, novaehollandiae Steindachner
and C, platyveephalus. (Gunther).

‘The ischium is bony and fused to the thum,
whereus the pubis can be extensively ossificd
and similarly fused (C. dahli and C. navae-
hollandiae), or completely cartilaginous (©.
australis and C. platycephalus),

The ilial shaft is slightly curved and cither
hears a satrow dorsal rim, rendered con-
spicuous by a longitudinal indentation on the
medial sutface of the shaft (C. australis, C,
dalilf and C. wevaehollandiae), or elsc Jacks a
dursal rim (C. platyceplidus). The acetabulum
has a narrow peripheral rim which superiorly as
above the level of the ventral surface of the
iial shatt, The ventral acetabulac expansion ts
slightly developed and the preacetabular zone

ig narrow. The dorsal acetabular expansion is
prominent and conspicuous. The dorsal pre-
minence is distinguishable from the Wial shuti
but only slightly raised. The dorsal pro-
tuberance is inclined yentrolatersily, extends
far from the prominence, and is approximately
one-half anterior to the anterior rim ob the
acetabular Fossa.

GEOCRINIA Blake
FIG. 3/

Species examined: G, laevis (Gunther),

The pubis is cartilaginous und the ischium ts
ussified.

‘The ilial shaft is short. slightly curved and
fattened laterally in cross-section, The acetabu-
lum is moderate with a narow peripheral rim,
The superior margin of the acetabulum hes
slightly above the level of the ventral margin of
the ilal shaft. The ventral acctabular expansion
is only slightly dilated. The dorsal acetabular
expansion is not pronounced. The dorsal pru-
minence is small and the dorsal protuberance is
nut detectable as a distinguishable area, The
dorsal prominence is on a Jevel with the
anterior rim of the acetabulum.

GLAUERTIA Loveridge
FIG. 3)
Speeies examinee: G. orientalis Parker,

The pubis and ischium are entirely car-
tilaginaus..

The. ilial shaft has an almost horizontal dor-
sal surface, and a sligntly curved ventral onc. so
creating a slight broadening ut the exuwemitics
of the shaft, There is neither « rim nor a crest
lo the shaft. The acetabulum has a s¢arcely
detectable peripheral rim which superiorly is
above the level of the ventral surface of the
ili) shaft. The ventral acetabular expansion ts
wlightly developed and the preacetabular zone
is slender, The dorsal acetabular expansion is
poorly developed. The dorsal prominence is
dificult to distinguish from the prominent and
evenly rounded dorsal protuberance which
uppears to be inclined laterally. Approximiately
one half of the length of the dorsal pro-
tuberance lics anterior to the anterior rim ol
the acetabulum,

HELEIOPORUS Gray
FIG. 3K
Speciey examined: H. albopunctatis Gray,
The pubis and ischium are ossified.
The tliat shaft is not curved but has o slightly
undulating superior face, raised inlo a
thickened fidge on the posterior half. The

PFLYIC GIRDLE, OF AUSTRALIAN FROGS 8

acetabulum is small and m surrounded by an
obliquely tapering rim. The ncefahulum is hish
and is bisectecdl by ihe veutral margin of the
ial shaft, The ventral acetabular expansion is
only stiehtty dilated but the dorsal acetabular
expansion rises sharply, The dorsal prominence
is Vast and the dorsal protuberunce simply an
elongate, semi-cylindrical ridge, At least two-
thirds of the dorsal protuberance lies anterior
to the anterior rim of the acetabulum,

KYARRANUS Moore
FIG, 32
Species examined: K. sphagnicolus Maore,
Rvarramniis sp.

The pubis and ischium are almost entirely
bony, only the central portion of the acetabu-
lar fossa remaining cartilaginous.

The ilial shaft ts strongly curved and almost
tircular in cross section, but for an indistinct
and narrow im on the posterior half. This tim
is created by an indentation of the medial sur-
face of the shaft, The ventral acetabular expan-
sion is poorly developed and the preacetabular
zone is very narrow, The dorsal acetabulur
expansion is quite prominent. The dorsal expan-
sjon is scarcely distinguishable from the Jarge
and oval, posteroventrally inclined dorsal pro-
minence. The anterior margin of the dorsal
prominence js considerably anterior to the an-
lerior rim of the acetabulum.

LECHRIODUS Boulenger
FIG. 3M

Species examined: L. melanopyga (Doria).

The pubis is cartilaginous and the ischium
ossified,

The ital shalt is slightly curved and bears
an enlarged, fanlike dorsal crest arising from
the posterior three-quarters of the shaft. The
acetabulum is small and has a broad peripheral
rim: the dorsal margin lies above the ventral
margin of the ilial shaft. The ventral acetabular
cxpansion is onty slightly developed. The dorsal
acetabular expansion is long and projects pas-
teriorly. The dorsal prominence is small but
detectable and the dorsal protuberance can be
distinguished, The anterior margin of the dorsal
protuberance is slightly posterior fo the anterior
rim of the acetabulum.

LIMNODYNASTES Fitzinger
FIG. 3N
Species examined: L. convexiusculus (Mac-
leav), FE, dumerili Peters, FL. ernatus (Gray),
L, peront (Dumesil & Bibron). L. sedmini Stein-

dachner, Z. spencer Parker, [ tseuamnfensis
Gunther, £, tevrueresinae Fry.

The pubis. is. cartilaginous and (te ischiam
is ossified.

The ilial shaft ts slightly curved and highly
vatisble in structure. There is an clongate
groove on the medial surface in L, dumerili, a
short groove in L. terraereginae, a short lateral
groove in L. rasneanternsis, and there is a dis-
tinct dorsal crest in L. ernates and ZL, spencert.
The acctabulum is small and high, brsected in
most species by the ventral margin of the ilial
shaft. The ventral acetabular expansion is small
and not particularly expanded. The dorsal
acetabular expanston rises into a moderate or
else high and acutely pointed spike. The dorsal
Prominence is conspicuous in all species except
those with crests on the shafls, Tr tends to fom
a conical shape in profile, but is somewhat
broader and rounded tn the largest species. The
dorsal protuberance is an elongate ndge or
expanded knob upon the tip of the dorsal pro-
minence. The dorsal protuberance is on a level
with or slightly anterior to the anterior rim of
the acetabulum.

MIXOPHYES Gunther
FIG, 30
Species examined: M. faselolatus Gunther.

The pubis and ischium are completely
ossified.

The ilial shatt is slightly curved and hears
an enlarged fan-dike dorsal crest arising dorso-
laterally from the Ienath of the shaft. The
acetabulum has a broad peripheral rim, the
dorsal margin of which lics above the ventral
margin of the ilial shaft. The vente) acelubu-
lar expansion is moderately developed bur with
a rather narow preacetabular zonc. The dorsal
acetabular expansion rises high to abil the
enlarged superior portion of the ischium. There
is no dorsal praminence, and the dorsal pro-
tuberance is entirely laterally directed and so
poorly developed that if is detectahle only when
the urea is viewecl from the dorsal or ventral
aspects. Whe anterior margin of this weak pro-
tuberance its located anterior to the anterjor rim
of the acelabulum.,

MYOBATRACHUS Schlegel
FIG. 4A
Species examined: M. gouldil (Grav),

The isthium is a large and almost circular
bony plate, and the pubis is reduced to a small,
triangular wedge of cartilage.

The iliwl shaft is distinctly curved, Jacks
crests und indentations und is slightly flatiened

10 M. J. TYLER

Fig. 4. Pelvis or isolated ilia of leptodactylid, microhylid and ranid frogs:
A. Myobatrachus gouldii, x 5; B. Netaden melanoscaphus, x 5; C. Neobatrachus centralis, x 5;
D, Philoria frosti, x 5; E, Pseudopkeyne bibroni, x 5; F. Ranidella parinsignifera, x 5; G. Rheo-
batrachus silus, x 5; H. Taudactylus diurnus, x 5; 1. Uperoleia sp., x 5; J. Cophixalus arnatus, x
5; K. Sphenophryne robusta, x 12.5; L. Rana papua, x 5,

laterally, producing an oval cross section. The
acetabulum is large and has a narrow
peripheral mm whose superior margin is con-
sidcrably above the level of the ventral surface
of the ilial shaft. The ventral acetabular expan-
sion is greatly reduced, consisting of just a
slender slip of bone bordering the acetabular
rim, The dorsal acetabular expansion is a more
prominent feature, rising above the ilial shaft.
There is no dorsal prominence, and the dorsal
protuberance is replaced by an oval, dorso-

laterally inclined plate, consisting of a weak
peripheral rim surrounding a very shallow
depression. The anterior margin of this struc-
ture is far posterior to the anterior margin of
the rim of the ucetabular fossa.

NEOBATRACHUS Peters
FIG. 4C
Species examined: N. centralis Patker.

The pubis is cartilaginous and the ischium is
ossified.

PELVIC GIRDLE OF AUSTRALIAN FROGS tl

‘The ilial shaft is almost perfectly straight.
The acetabulum is small and high, the ventral
marein of the jlitim being on a level with the
anterior une-third of the acetabulum. The ven-
iral acetabular expansion is only slishtly
dilated, the preacetabular zone being particu-
latly reduced. The dorsal acetabular expansion
is high. The dorsal prominence is very large
and the dorsal protuberance is a pointed nodule
upon it, Approximately one-half of the dorsal
protuberance lics anterior to the anterior rim
of the acetabulum,

NOTADEN Gunther
FIG. 4B
Species examined: N. melunoscaphus Hosmer.

The pubis and ischium are cartilaginous, the
laiter with a medisn calcified zone.

The ilial shatt is only very slightly curved,
lacks ridges and indentations and is circular in
eToss section, The acetabulum has a <listrnct,
flaltened peripheral rim which superiorly is on
a level with the ventral surface of the ifigl shaft.
The ventral acetabular expansion is slightly
developed with a narrow preacetahular zone.
The dorsal weetabular expansion is small and
modecratcly developed. The dorsal prominence
is broad and clearly demarcated from the ilial
shaft, The dorsal protuberance is small and
located on a level above the centre of the
acetabular fossa,

PHILORIA Spencer
FIG, 4D
Species examined: P. fresti Spencer.

The pubis is cartilaginous and the ischium
ossified,

The ilial shaft is sirongly curved and has a
medially directed dorsal rim, The acetabutim
is Large, its supenor margin on u, level with the
ventral margin of the ilial shaft. The ventral
acctabular expansion is not obviously dilated.
The dorsal acelubular expansion is moderately
well developed, The dorsal prominence is not
conspicuous and the dorsal protuberance
scarcely detectable as a separate entity. The
dorsal protuberance is almost. entirely antetinr
to the anterior rim of the acetabulum.

PSEUDOPHRYNE Filzinger
FIG. 4£
Species examined: P. bibront Gunther, P-
corfacea Keferstein

The pubis is cartilaginous and the ischitim is
ossified.

Phe ital shaft ls almost straighi and ts ¢ircu-
lar in Cross section, The acetabulum is large

and has a narraw peripheral tim, The superior
Margin of the acetabuluny lies on or slightly
above the level of the ventral margin of the
ijial shaft. The ventral acetabular exparision is
diated, The dorsa] acetabular expansion is only
very slightly raised. The dorsal prominence is
small (scarcely detectable in P. raridcea, but
quite distinct in P. bihfent) and the dorsil
prominence a very small knob on its tip. The
dorsal protuberance is on a level with the an-
terior rim of the acctabulum.

RANIDELLA Girard
FIG. 4F
Species examinedr R, parinsignifera (Main\,
®, sienifera Girard.

The pubis is cartilaginous and the ischium i3
bony.

The ilial shaft is curved, compressed Jaterally
and possesses neither a mm nor a crest. The
ucetabulum has a broad peripheral rim whicls
superiorly is on a level with the ventral margin
of the ilial shaft, The ventral acetabular expan.
sion is greatly enlarged, the subacetabular zone
protruding anteriorly. The dorsal acetabular
expansion is poorly developed. The dorsal
prominence jis broad and the dorsal pro-
tuberance is rounded, inelined posteroventrally
und moderately prominent, Approximately
one-half of the dorsal protuherunce is anterior
to the anterior rim of the acetabulum,

RHEOBATRACHUS Liem
FIG. 4G
Species exantined: R, situs Liem.

The pubis is cartilaginous and the ischium
ossified.

The iltal shaft is exceptionally slender, very
slightly curved and cylindrical in cross secon.
The acetahulum is very large with a con-
spicuous rim, its superior margin slightly above
the level of the ventral surface of the ilinl shaft,
The ventral acctabular expansion is slightly
dilated. The dorsal acetabular expansion is
slightly developed, projecting posteriorly. The
dorsal prominence is well developed and the
dorsal protuberance is conical and situated pos-
terior to the anterior rim of the acetabultm.

TAUDACTYLUS Straughan & Lee
FIG. 4H
Species examined: T, dikrnus Straughan & Lee
The pubis is cartifaginows and the ischium
is ussified.
The ilial shaft is gently curved, <tightly com-
pressed Interally and possesses neither a rim

2 MT. TYLER

nor g dorsal crest, The acetabulum is Jates with
a weil developed rim creating a deep acctibular
fossa, The ventral acetabular expansion is
slightly dilated. The dorsal acetabular expaiy-
sion is directed posteriorly and overlies the
superior margin of the ischium. The dorsal
prominence is only slightly distingnishable
from the large, raised, ovat, dorsolaterally
directed dorsal protuberance, Approximately
one-half of the dorsal protuberance lies anterior
toa the acetabular rim

UPEROLEI Gray
FIG. 4!
Species examined: U. marnworata Gray, LU pero-
leta sp,

The pubis and ischium are cartilaginous.

The ilial shaft is almost straight and is ctrew-
lar im cross section, lacking a rim and a crest.
The acetabulum is large and has a broad
peripheral tim, The superior margin of the
acetabulum lies above the ventral margin of the
ilial shafe. The ventral acetabular expansion is
of moderate width with the subacetahular zane
very slightly expanded. The dorsal aceiahilar
expansion is not pronounced. The dorsal pro-
minence ts large and the dorsal protuberance
conical and rising far above the level of the
shaft. The anterior margin of the dorsal pro-
tuherance is situated antenor to the anterior
rim of the aceiabulirm.

Family MICROIYLIDAE
COPHIXALUS Boetteer
FIG, 4/

Species examined: €. ornaus (Fry).

The ischium is small but ossified, and tre
pubis is cartilaginous,

The ilial shaft is compressed mediolaterully.
has neither a rim nor a crest, and is very
sheblly curved. The ventral acetabular expan-
sion is coticave and very narrow. The acetabu-
lar fossa is extremely large and very high. its
superior margin nearer to the darsal than the
ventral margin of che ilial shaft. The dorsal
protuberance is not distinguishable From the
dorsal prominence, and lies slightly anterior to
the anterior margin of the acetahular rim. The
dorsal acetabular cxpansion is very poorly
developed.

SPHENOPHRYNE Peters & Doria
FIG. 4K
Speciex examines; S, rohuyia (Yry),

The ischium is extremely small ind wnty
pany ossified. The pubis is cartilaginous.

The ilial shaft is compressed mediolaterally,
has neither a fim nor a crest, and is very
slightly curved, The yentral acetabular expan-
sion is slightly concave and very narrow. The
acetabular fossa is large and high, and is partly
bisected by the ventral border of the ilial shaft.
The dorsal protuberaice is not distinguishable
from the dorsal prominence, and lies entirely
posterior to the anterior margin of the acetabu-
lar rim. The dorsal acetabular expansion is only
slightly developed,

Family RANIDAE
RANA Linné
FIG. 4b

Species examined: R. papne Lesson

The pubis. is. cartilaginous and the ischium
is bony.

The ilial shaft curves gently downwards and
bears a tnassive fintike and tapering dorsal
crest, The acetabulum is large and bears a
broad petipheral rim. The superior margin of
the acetabular rim extends considerably above
the ventral margin of the ilial shaft. The ven-
tral acetabular expansion has a reduced pre-
acetabular zone and greatly dilated subacetabu-
lar zone. The dorsal acetabular expansion is
well developed, tapering to a point posteriarly.
There is no dorsal prominence and the dorsal
protuberance is an oval and almnst vertical
expansion of the base of the ilial shaft, This
profuberance lies entirely posterior to the an-
terior rim of the acetabulum,

Fossil genus
AUSTRALOBATRACHUS new genus
Type species: Australobatrachus ilius new

species,

This taxon was first reporied on by Tyler
(1974, p, 711, fiz. 1).

Extending from the acetabular region the
iial shaft bears o deep, curved groove an its
lateral surface. The acetabular fossa is excep-
lionally high in relation to the lial shaft. its
superior border reaching a position equivalent
to midway up the shaft. The acetabular cim is
poorly developed. The dorsal protuherance ts
gently rounded, and not distinguishable from
the dorsal prominence, The anterior limit of the
dorsal protuberance is on a Jevel with the an-
teriat margin of the acctabular rim, The ven-
tral acetabular expansion is neither protuberant
nor concave. the ilial/preacetabular zone form.
Ing almost a straight line extending gradually
posteriorly, The dorsal acetabular expansion
probably does not project superiarly,

PELVIC GIRDLE OF AUSTRALIAN FROGS 13

PIO ETE IT oe

a

| \| | il

yy)

WON Ni?

isi jf”

Fig. 5, Left ilium of Australahatrachus dins holotype. SAM, P18021.

Australobatrachus ilius new species

Aoelatype; Two tragments comprising the distal
6.7 mm of a single left tum. SAM, P1021
(Fiz, 5).

Type locatiry: Tedford Quarry, on the west side
of Lake Palankarrina, S.A. (Universily of
California Museum of Palagontology locality
V-5375.)

Horizon; Etadunna Formation.

Ase: Ngapakaldi Fauna, Tertiary; probably
mid-Miocene,

Deseription of holotype: As for gcnus.

Comparison with other species: Of the existing
families of frogs that do not have modern rep-
resentutives in Australia, the Leiopelmatidae.
Pipidae snd Pelobatidae have members in the
southern hemisphere and could all be regarded
as potential contributors ta the ancestral Aus-
Lralian frag fauna; hence they merit compari-
son with Australohatrachus.

The Letopelmatidae of New Zealand have
simple ila with a slender, cylindrical ilial shaft
and 4 poorly developed ventral acctabular
expansion (Stephenson 1960}. There is not the
slightest resernblance to Australohatrachys,

According to Trueb (1973), the pipids are
quite mnique in possessing a lateral crest (not
groove) to the ilial shaft. Any resemblance fo.
Aistralobatrachus has to be weighed against
the condition of the dorsal prominence (vast
and projecting high above the shaft in pipids;
reduced and not raised above the shaft in Amy-
trralobeltrachuy) and of the ventral acetabular
expansion (vestivial in pipids as oppased to
heing highly developed),

To judge from the descriptions and illustra-
tions of Zwetfel (1956) and Kluge (1966) the
pelobatid ilium has typically a bow-like dorsad
curvature, creating a totally different form from
that of Australobatrachus, In addition the
species illustrated by them have poorly-
developed ventral acctabular expansions and
more prominent acetabular fossae than has the
new genus,

Tn comparison with the ranids of Australia
and New Guinea, Australobatrachus is readily
distinguished by its total lack of the dorsal crest
which rises high above the ranid ilial shuft.
Similarly many microhylids (and also some lep-
todactylids) exhibit such a crest. but the micro-
hylids lacking a dorsal crest may he distin-
guished by the poor development of the ventral
acetabular expansion.

Rstablishing means of distinguishing Austra-
lian hylids Erom leptodactylids has proved diffi-
cult, In general it would apear that » well-
developed, dorsally-projecting dorsal pra-
minence is almost characteristic of Lhe lepto-
dactylids, whereas it tends 1!o be paorly
developed or else Jaterally disposed in hylids-
Most of the leptodactylids which are excep-
tional in having poorly-developed dorsal
prominences, are those in which the upper sec-
tion of the ilial shaft is modificd in some way-
Hence the prominence is scarcely differentiated
in Lechriodus and Mixophyes which have i
dorsal crest because the prominence is upon
this thin flange of bone. It could be argued
that the poor development of the dorsal pro-
minence in Australobatrachus conflicts with my
Interpretation of the groove as an intrinsic
modification of the ilial shaft. Hence an alter-

14 M, J. CYLER

native hypothesis is that the upper rim is the
supra-ilial structure, comparable to a dorsal
crest. This latter explanation is not favoured,
simply because the end section of the ilial shaft
iy S-shaped, and the nature of the acetabulum
(a poorly developed rim) is not usually
developed on a pelvic girdle in which the
acetabulum reaches the upper scgment of the
shaft.

Amongst the Australian hylids and Icptodac-.
tylids there is considerable variation in the
shape of the ventral acetabular expansion.
Cyclorana australiy and several species of
Litoria approach the condition displayed by
Anstrqlobatrachus, bat in each the dorsal pro-
minénce is more highly developed and the
lateral groove is lacking.

If Australobatrachus lacked a lateral groove
the nature of the dorsal prominence and the
ventral acetabular expansion would cause me
to favour referting the genus to the Hyliduc.
Hence to avoid over-interpreting the presence

and form of the Sateral groove, Australobat-
rachus is assigned tentatively to this family.
Extrapolation of musculature: The lateral sur-
face of the ilial shaft is the site of origin of the
musculus ifiacus externus. Any extreme
broadening of the lateral surface of the ilial
shaft increases the surface area available to
this muscle. Unquestionably these are madifica-
tions most Common in gquatic frogs or species
ifhabiting streamside situations. members of
such genera as Xenopus, Rana, Lechriodus and
Mixephyes, Certainly it is tempting to attribute
a similar functional association for Australo-
hatrachis, and hence assume that this animal
lived close to permanent water.

Acknowledgments
T am deeply indebted to Dr M, QO, Wood-
hurne for the opportunity to examine and des-
cribe the new fossil genus. Dr J, Ling and Mr
WN, Pledge read and provided constructive criti-
cisms of the manuscript, Mrs L. Kingston very
kindly typed the manuscript,

References

Cuantect, C. J. (1964).--Some Mio-Pliocene
hylids from the Valentine formation of Neb-
raska, Amer, Mid!. Nut. 72, 211-225,

Homan, J. A. (1965)—RFarly Miocene amirans
from Florida. Quart. J. Florida Acad. Sci. 28.
68-82.

Kutice. A. G. (1966)—A new pelobatine frog
from the Lower Miocene of South Dakota
with a discussion of the evolution of the
Scaphiopus-Spea complex. Contrih, Sei. Los
Angeley County Mus, (113), 1-26,

Lynen. J. D. (1963) —Additional evidence for the
recognition of Limnmaeodus (Amphibia:
Hylidae). Capria 1963, 566-568.

Lywen. J. D. (1971).—Evolvtionary relationships.
osteology, and zZeogeography of leptodactyloid
frogs. Misc, Publ. Univ, Kansay Muy. Nat.
Hist. (53), 1-238.

Srepumnson, E. M. (1960).—The skeletal churac-
ters of Lejopelia hatnilfani McCulloch, with
particular reference to the effecis of hetero-
chrony on the genus. Trans. Ro Sac, N.Z,
88(3), 473-488.

Trees, L. (1973)—Bones. frogs and evolution.
In Vial, J. (Ed.), “Evolutionary Biolagy of
the Anurans, Contemporary Research on
Major Problems”, pp. 65-132. (Univ. Mis-
souri Press.)

Trier, M. J. (1974).—First frog fossils from Aus-
tralia. Nature 248(5450),'711-712.

Zweirer, R. G. (1956)—Two new pelobatid frogs
from the Tertiary of North America and their
relationships to fossil and recent forms, 4m.
Mus. Novit. (1762), 1-45,

NEW AUSTRALIAN ALLODAPINE BEES (SUBGENUS EXONEURELLA
MICHENER) AND THEIR IMMATURES (HYMENOPTERA:
ANTHOPHORIDAE)

BY T. F. HOUSTON*

Summary

HOUSTON, T. F. (1976).-New Australian Allodapine bees (subgenus Exoneurella Michener) and
their immatures (Hymenoptera: Anthophoridae) . Trans. R. Soc. S. Aust. 100(1), 15-28,
28 February, 1976.

Three new species of Exoneura Smith (E. eremophila, E. setosa and E. tridentata) are described
and figured. They are assigned to the formerly monotypic subgenus Exoneurella Michener and both
adults and immatures of the new species are compared with those already described for the type-
species, E. lawsoni Rayment. A key for identification of adults is provided.

Females of E. tridentata vary greatly in size and exhibit allometric variation of the head and
metasoma.

NEW AUSTRALIAN ALLODAPINE BEES (SUBGENUS EXOQNEURELLA
MICHENER) AND THEIR IMMATURES (HYMENOPTERA:
ANTHOPHORIDAE)

by T. F. Houston*

Summary
Houston. T. F. (1976).—New Australian Allodapine bees (suhgenns Exonenrella Michener}

and their immatures (Hymenoptera: Anthophoridae). Trans. R. Soc. S. Aust, 100(1),

15-28, 28 February, 1976.

Three new species of Exoneure Smith (&. eremophilu, E. serosa and EF, widentata) are
described and figured. They are assigned to the formerly monetypic subgenus Exenenrella
Michener and both adults and immatures of the new spgcies are compared with those already
described for the fype-species, E. Jawsoni Rayment. A key for identification of adults is

provided,

Femules of E. midentafa vary greatly in size and exhibit allometcic variation af the head

and metasoma.

Introduction

The chief purpose of this paper is to provide
names for three species of bees. whose ethology
is ta be dealt with in a subsequent work, An
exhaustive examination of material from col-
lections has not been attempted und the des-
criptions to follow are based on specimens in
the collections of the author and the South
Australian Museum.

The three new species are assigned to the
subgenus Exoneurella Michener (of Exoneura
Smith) which formerly contained only the
type-species, E. Jawsoni Rayment. Since
Exoneurella was founded partly on the basis
of the larval characteristics of E. lawsoni, it
seemed of interest to describe and compare
immatures of the new species, Generally, the
characteristics of the new specics support re-
tention of Exoneurella as a discreet taxon.

The size-correlated variation of females of
E. tridentata sp. noy. is detailed below as’ it is
significant in terms of the bionomics of the
species. Such variation is unusval amongst
allodapine bees and in the family Antho-
phoridae as a whole.

The following abbreviations are used for
the names of institutions and collections te-
ferred to in the text below: ANIC ( Australian
National Insect Collection, C.S.LR.0O., Can-

berra), HE (author's private collection, to be
deposited in SAM), KU (Snow Entomological

Museum, University of Kansas. Lawrence,
Kansas, U,S:A,),; SAM (South Australian
Museum, Adefaide) and WADA (Western

Australian Department of Agriculture, Perth),

Except where stated otherwise, all specimens
listed in this paper were collected by the
author,

Genus EXONEULRA Smith
Exoneurd Smith, 1854, p. 232. See Michener.
1965, pp. 223-226 for detailed description,
subgenera and species.

Subgenus EXONEURELLA Michener

Exoneurella Michener, 1963, p. 257 (erected
asa genus); 1965, pp. 223-224 (relegated to
subgeneric status: diagnosis provided).

THE ADULTS

Key to the species of Exoneurella
1. Second cubital cell of fore wing with subequal
costal and medial borders (Fig. 1); 6th meta-
somal tergum of female with simple non-bific
apex and u pair of Jateral projections (Fig. 24,
25); compound cyes of male strongly swollen
(Figi8) -- gsm. a... 2 tridendia
1. Second cubital cell of fore wing with costal mar-
gin conspicuously shorter tham medial margin
(Figs 2. 3); 6th metasomal teraum of female
wiih bidentate apex and with or without lateral

* South Australian Museum, North Terrace, Adelaide, 5. Aust. 5000.

T. F. HOUSTON

FIGS 1-19

NEW AUSTRALIAN ALT ODAPINE HEES 17

prajections (Figs 4461, compound eyes of male
not swollen (Figs 17, |Bb - 1 2
2. Sixth metasomal tergum of fernale without
luteral prominénces, margins smonthly sinudle
(Fig, 4); hiod femora of male obttsely pro-
duced and curinule ventlully (Fig. &)) meta-
somal terga of both sexes black without
cream or ¢creamy-brown pigmentation and
withatl numerous cottspicuously thickened
setwe E. lawson
2. Sixth metasomal tergum of fernale with dis-
linct Joteral projections or prominences (Figs
5, 6); hind femora of male unmodified (Fig
7) or nal modified as in Fig. 8; mictasomal
tecea of both sexes with faint to distinct cream
bands; tergx 3 and 4 (females) or 4 and 5
(mates) with numerous conspicuously
thickened setae (Fig. 11) 0.0 0000... 9
. Metusomal terga yellow-brown to black with
wide cream bands (Pix. 14); 6th metasomal
tergum of female with subacute Jaleral projec-
tions (Fig. 6); hind femora of male cach with
an acute ventral projection (Fig. 9) 0...
E, eremophila
3, Metasomal terga Jargcly black with narrow
fyometimes faint or incomplete) cream subapi-
cal bands (Fig. 13); 6th metasomal tergum of
female with very obtuse Jaleral promiinences
(Fig, 5): bind femora of male unmodified (Fig.
os i oe , &. setosv

od

Exoncora (Exoneurella) eremophila n.sp,
Figs 2, 6, 9, 14, 15,17
Types
Holotype: &@ (SAM, 1 209613, New Kalamurini
Homestead, S. Aust. (27°44'8, 138°1S EB),
9-1 1.01.1972, on Wablenbersic.

Allotype: 2 in SAM, Poratypes; 42 d, 44 ° in
SAM:2 4,29 in ANIC, bf. 19 in KU.
Djacrosiy; This species differs [rom all other
Exzonenrella as follows. Lateral face marks of
male filling sptces between clypeus and com-
pound eyes (Fig. 17); metasomai terga of both
sexes yellow-brown to black with ‘extensive
cream maculations (Fig, 14): pronotam with a
pair of cream marks sublaterally; hind femora
of mule each with on acute ventral projection

(Fig. 9); hth metasomal lergom of female with
bidentate apex and a pair of subacuwre latcral
projections (Fig 4).

Deseriplion
Male. Body length 3.8-4.1 mm: head widih
1.]-1.2 mim,

Head capsule as broad as long; compound
eyes of usual relutive size (Pig. 17): face nur-
rowed to about 43% of head width; occlli
approximately equal in size to antennal sockets:
scapes failing to reach level of median occllus;
flagella 8596 as long as head width, the middle
segments about as long as broad: genae
viewed laterally almost half as wide as com-
pound eyes and evenly convex: fore legs not
especially slender, the Tore tibiae about 3.5x
us long as wide; bind femora laterally com-
pressed. cach with an acute ventral projection
(Fig. 9); costal margin of 2nd cubital cell of
fore wing about half as long as 1st transverse
cubital vein (Fig.2).

Integument glossy generally with sparse fitre
pitting or none; metasomal terga very finely
lineate,

Pubescence white, virtually absent dorsally
but dairly tong and dense ventrally and on legs
and mesepisterna;: 4th and 5th metasomal teres
fexcept laterally) with a sparse covering of
thickened bristle-tike setac.

‘the following areas white ar cream: lower
face, labrum and middle portions of mandibles
(Fig. J7). Seapes and pedicels venirally,
tubercles and dorsal margin of pronotum (ex-
cept medially), spots on tegulac, basal parts of
Wing veins, subapicul bands on tetusomul
terga (Fig, 14), apices of femora, bases of
tibiae, anterior edges of fore tibiae, and
basitarsi. The following areas yellow-brown:
scapes dorsally, flagella ventrally, most parts
of legs (except for cream areas), metssoma

Figs

i~ 3. Seoond cubilal cells of right fare wings (in dorsal views) of Evenewn (Exoneureila) triden-

tala, E. (B,) eremophila, and B. (F.) setesa respectively. C = costal margin, | = first

transverse cubital ven.

Figs 4~ 4, Sixth metasomal tergu (dorsal views) of females of £. (E.) la\vsoni, &. (E.) setosa and E.

(E.) eremophila respectively,

9. Trochanters, femora and apices of Ublue of teft bind legs (anterior views) of males of &

(E.) setesa, BE. (E,) lawsant-and #, (&.) erentaphila respectively

. Left mandible of female of E, (&) tridentata (ventral view).
. Thickened bristle-like sétae from fourth metasomal tergum of female of E. CE.) serosa
. Head capsule of a relatively large [emale of &. (&.) aridentaia (left lateral view),
and td. Metasomuae (dorsal views) of mules of 2. (2.) serosa und FE. 1.) erenemiila re

spectively showing cream bands (white) and translucent tergal margins (stippled)-
Fig. 15, Lower portion of head of female of E- (£.) eremophila (anterior view) showing T-shaped

chypeal mark,
Fig. tf

Head capsule of mete of 2, (E.) vidsmtatu (belt lateral view},

Figs 17-19, Head capsules (anterior views) of mules of RK. (fi) ercnnaptidl, BOR) setene and BL (E,)

(riglentofa fespectively-

\8 T. F, HOUSTON

ventrally and pastially or extensively dorsally,
Reinaiiioe areas Black or wack brown,
Vemale, Body length 4.0-5.5 mm: head width
1.i1-L2 mm.

Head form similar to thal of male (Fig. 17):
flagella about 67% as tong as head width;
mandibles tndentate but not constricted suh-
apically; metasoma fairly elongate, 6th meta-
sumal tergun with a bidentate apex and a pair
of small buat almose acure upturned lateral
projections (Fig, 6); hind femora unmodified,

Inlegument ane pubescence much as in male
hut thick bristle-like setae occur only on lergu
3 and 4,

Coloration as in mate except that while on
face ts limited to a full-length T-shaped mark
on elypeus (Fig. 15)

Pariation

The extent of yellow-brown coloration on
(he metasoma varies considerably amongst in-
dividuals collected together gnd some spect
menos have creamy-brown maculations on the
lateral margins of the scutum and scutellum.

The speertic piame, derived from Greek,
means loying solitude and alludes to the arid
hahitnt of the species.

Distribution

Central regions of Austealta including por-
tions of the Northern Territory, Queensland,
New South Wales and South Austratia.

Specimens examined: The holotype and the follaw-
ing. OLDS ¥ 3, 8 FP, A miles (4.8 km) W, of Wlo-
dara, U7,iv.1969, ex nests (HC), NWSI 1 2, 82
miles (232 kin) W. of Cobar, 9141971, on Prilo-
ts (SAM): | ct, 14 G70 miles (113 knit) E. af
Wileannia, 3111971, on Helichrysum, Goodenia
and Wahlenbergia (SAM). §. AUST.: | 3, Amata
(Musgrave Park) setuement, 14-%.1972, on Crlun-
drinta, H. BE. Evans & 1. P. Heuston (SAM): 1 9.
33 miles (33 km) W. of Amata, 17.x.1972, ex nest
in pithy twig, MW. &. Byany & T. FL Houston
(SAM); 7 &, Bettys Well (132°26'F, 2772’Sy-
Tiverund Park Sto, 1-S.xi.1970, on Aidiscies
fabrdgel (SAM 1; 43 ¢ (paratypes), 48 2 fallotype
and paratypes}, New Kalamurina HS, (27°44'S,
(38° 15'R), 9-01 71.1972, on Wahlenbergia (7 3,
5 F pinned), ex dead stems of Myriocephalins (38
¢, 43 Pin aleohol) (ANIC, KU, SAM): 2 &. Mor-
fat, 19.x11,1963, on fallenhereia (AC); 4 8, 49,
i} miles (16 km) S. of Mt Duvies airsirip,
71.%.1972, on Prlotis, ME, Evans & T. PF. Hons-
ior (SAM), 2 9 Mt Miceolle (32°31'S,
136°36E), Siam Stn, 2.iv.1971. ex nest (SAM);
1 9. 8. of Tomkinson Ranges (129°8°R, 26°10°S),
18.%.1972, on Sevevalu, H. E. Evans & 7. F, Hous-
ton (SANT): 2 f, 31 miles (40 km) W. of Wel-
bourn Hill HS, 13.%.1972. on blue Eremupliila,
HOE, EBvalis & TF, fhoustron (SAM): 2 2% 10
miles (16 km) SE of William Creck. 28,1,1277,
on Hakea, 1B Byuns & 7, F. Houston (SAM).

Froncura (Exoneurella) lawseni Raymeni,
1946, pp. 230-232, fig. 2 (male. not female or
Sarva)

FIGS 4, 8

Exenvaretla lawsont (Rayment)

1963, p. 257.
Franeura (Exoneurella}
Michener, 1968S, p. 224.
Holotype: 4 fin ANIC), Canberra, A,C.1..

Newlon R, Lawson. July (945,

1 hive not exammed the holotype ful Ms
Josephine Cardale (ANIC) made 4 critical
examinition of it on my behalf and confirmed
jhat Th agrees with the male churacleristics
given in the diagnosis below.

Michener (1963, p, 258) pointed out that
the fernales and larvae described hy Rayment
in the orginal description of E. Jawseni are af
unother species and are referable to. the sub-
genus Brevineura,

Diagnosis; E, lawsoni differs from all other

Exoneurella as. follows. Hind femora of mule

carinate and broadly produced ventrally (Fig.

8); 6th metasomal tergum of female lacking

lateral prominences, the margins gently sinuate

(Fig. 4). Differs from &. sétose and E.

cremophila in complete absence of cream pig-

mentation from metasoma and in wbsence (or
only feeble development) of thickened bristle-
like setae on dorsuni of metasoma,

For a detailed description of both sexes see
Michener (1963, p. 259). Note, however, that
the pale maculations of the face of the iiale
are white, not pale yellow,

Distribution: On and near Great Dividing

Range of south-eastern Queensland, New

South Wales and castem Victoria,

Specimens examined; OLD: 442,59, Bunya Moun-

tains, 6.x,1968, ex nesis THC). PIC. | fd, 2 3.
Tambo Valley, 221.1966, on Walilenheygla (AC),

Exoneura (Exoncarela) setosa n.s}),
FIGS 3.5, 7. IP. 13, 8

Michenes,

laweornt Rayment.

Types
Holotype: & (SAM, I 20962), West Beach,
Adetaide, S$. Aust, 25.iv,1975, ex deasl

Euphorbia stem, C. A. & T. F. Tfousten.
Allotrype: 2 in SAM, Parotypes; 7 ¢, 10 2 in

SAM; 42,4 2in ANIC; 23,2 9in KU.
Diagnosis: Very like E. lawsoni, differing a
follows. Hind femora of male unmodified,
lacking ventral flanges and projections (Fiz.
7); 6th metusomal tergum of female with a
pair of lateral convexities (Fig. 5); metasomal
terga of both sexes with narrow subapical
bands of creamish pigment (sometimes faint
or Incomplele, especially medially) and trans-

NEW AUSTRALIAN ALLODAPLNE BEES 19

lucent brown apical margins (Fig. 13): meta-
somal terya 4 and S (male) or 3 and 4 (fe-
male) With numerous conspictiously thickened,
hristle-like selac (Fig 11),

Dexeription
Male, Body length 3.84.7 mm: head width
1,1 1.2 mm,

Head capsule 11x as broad as longs com-
pound eyes of usual size (Fig, 18); face nar-
rowed ty about 42% of head width in lower
part; ocelli approximating size of antennal
sockets; scupes jusi failing to reach level of
median octellus) flagella about 724% as long as
head width; middle Mavellar segments slightiv
hroader than Jong; yenae viewed laterally 2/3
as wide as compound eyes and evenly conyea;
fore legs not unusually elongate. the fore tibiae
about 3.5x 4s long as wide; hind femora un-
modilicd (Pig. 7); custal margin of 2nd
cubital cell of fore wing about 1/4 ta 1/2 as
long as Ist transverse cubital vein (Fig, 3)-

Tntegument almost entirely glossy; clypeus
and scutellum finely pitted, dorsal area of pro-
podeunr dulled by extremely fine sculpturing;
metisumul terea finely lineate.

Pubescence white. sparse on head and body,
densest laterally and ventrally on thorax und
basal parts of legs: 4th and Sth melasomal
terga with numerous short but thick bristlyike
setae (Fig, 11) on dorso-apical areas.

The following areas white: almost all af cly-
peus and a spot of variable size on euch side
(Fig. 18), labrum, anterior stripe on Lore libia,
spots at bases of mid and hind tibiae, pronotal
tubercles and alar scleriles. The following arens
off-white to cream; ventral edges of scapes.
micl and hind basitarsi, narrow subapical bands
on metasomal terga (Fig. 13; sometimes. faint
or incomplete especially medially). Fore legs
(lurgely) and mid femora and tibiae anteriorly
yellow-brown. Hind margins of metusomal
ferga translucent pale brown; remaining reas
black or blackish brown,

Female, Body length 4,3-3.5 nim; head width
1.4-1.2 mm.

Head form similar to that of male: flagella
64% as long as head width; mandibles tri-
dentate but not constricted subapically:
melasoma elongate, the 6th lergum with bi-
dentate apex and w pyir af obtuse lateral
prominences (Fig, 3),

Inicgument sculptuled as in sale.

Pubescence much as in male but bristle-like
setae Occur on hind margin ef 3rd metasxomal
tergum and dorsal arga of 4th-

Coloration differs from that of male as fol-
lows: clypeus with a full-length white T-shaped
stripe; paraocular areas. without white spote;
labrum entirely brownish; legs lacking -yellow-
brown coloration.

The specific name, derived fron) Latin od
meaning *bristly’, alludes to the setation of the
rattasoma,

Distribunoa

Lowlands of southern South Australia (west

ta Spencer Gulf) and of south-ensternm Queens-
land.
Spectinens examined: The holowpe and the Follow-
ing, OLD: 1 3, 2 miles (3.2 km) S. of Nananeo,
7.x.1968, on Waltenberuia (HC), | 3 3 3 3
miles (4.8 km) N, of Peregian Beach {near Noosi
Ville), 9.xi1966, ex ness (HCO) S. AUST 2 dy
5.9, Glenely North (dunes), Adelaide, 24.%.1963,
€.x.1964 and 14-164,1965, on pigface and
Wahlenhergia flowers (IMC: 1 2, Mambray Creck
Rail Siding, 13.xi.1970_ ex pithy stem (SAM); 3 9,
Morgan, 18-19.xi7.1963, on Wahlenbergia (TIC):
13 df (paratypes), 17 2 (allotype and paratypes),
West Beach, Adelaide, same data as for holotype
(ANIC, KU, SAM): 4 3, 3 &, West Bench,
Adeloide, § and 24.11, 1965, 27.19.1965, on Wahlen-
bergia and ex bolfow stems (HO),

Kxoncura (Exoncurella) tridentata n. sp.
FIGS 1, 10, 12, 16, 19-30
Types
Holotype: J (SAM, F 20963), Lake Gilles
Notlonal Park (136°46°R, 33°2°S), S, Aust,
3) .Ai7.L973, ex short Linrel in twig of Meter
dendren.

Allotlype; 2 in SAM, Paratypes: 39 3, 29 9 in
SAM; 2 4.5 2in ANIC, 1 2,2 Yin KO.
Diagnosis: FE. tridentata differs fromm all other
Exonenrella as fallows. Coastal margin of 2nd
cubital eel] of fure wing equal to or slightly
longer than Ist transverse cubital vein (Fig,
1); scapes reaching to above level of median
ocellus; male with swollen compound eyes and
relatively narrow face (Pig. 19) and fore legs
conspicuously elonguled; female with angular
gennac (Fig. 12). mandibles constricted sub-
apically (Fig. 10) und 6th metasomal tergutn

with non bifid apex (Figs 24, 25).
Descripiien

Male. Body length +.5-4.9 mm: head width
1.6-1.9 mm.

Head capsule 7.3x us broud as long; comes
pound eyes strongly swollen so thal face ap-
pears sunken between them: face narrowed to
about 1/4 of head width ¢Pig. 19): ocelli
relatively large (about |.4x as wide as antennal
sockets}; genac viewed laterally (Fig. 16)
much narrower than compound eyes and nal
angular; scapes slender, reaching to just above

20 T. F, HOUSTON

Tmm

tnt

Figs 20-29. Fxoneura (Expneurella) tridentata female. Figs 20, 21.—Heuad capsules (anterior views) of
smallest and largest known specimens respectively, drawn to same length, Figs 22, 23—
Smallest and largest known specimens respectively (dorsal views) (antennae and legs omit-
ted, fore wings represented by broken lines). Figs 24, 25,—Sixth metasomal terga (dorsal
views) of smallest and largest known specimens, respectively. Figs 26, 27—Teft halves of
fourth metasomal sterna (dorsal views) of smallest and largest known specimens, respec-
tively, drawn to same length, Figs 28, 29.—First metusomal terga (dorsal views) of smallest
and largest known females, respectively, drawn to same size to illustrate differences in sur-

face pitting.

level of median ocellus; Hagella relatively short,
L/2 as long as head width, all segments but
apical one broader than Jong; fore lees very
slender, the tibiae 5x longer than wide; hind
femora unmodified; costal margin of 2nd
cubital cell of fore wing equal fo or longer

than Ist transverse cubital vein (Fig. 1);
tmetasoma relatively short, broad and de-
pressed,

Integument, of face smooth but dull with
close small pitting on clypeus; scutum and
scutellum glossy with very sparse fine pitting;
mesepisterna dulled by shullow coarse pitting;
anterior half of Ist metasomal icrgum glossy
and impunctate. the posterior half and most of
tergum. 2 pitted and finely roughened, dull ¢x-

cept laterally; tergum 3 shiny but coarsely
Pilled; terga 4 to 7 duller with fine reticulate
sculpture.

Pubescence white, fairly long and sparse
generally, densest on clypeus, posterior of
head, sides and venter of thorax, sides of pro-
podeum, basal areas of legs and ist metasomal
tergum,

Colour black generally except for the follow-
ing: clypeal mark (Fig. 19), ventral edges of
scapes, patches on tegulae and wing bases and
spot at base of each tibia white to cream;
apical portions of femora, all tibiae and tarsi
orange-brown; wing veins and ventral surfaces
of flagella brown.

NEW AUSTRALIAN ALLODAPINE REES 21

Memale. Size eatremely variable: bedy length
$,5-10.0 mm; head width 13-2.) mm.

Head (viewed aliteriotly) grading from
fairly round in small femules (Fig, 20) to
rather quadrate in large females (Fig. 21);
uccordingly whe inner orbits yary from slightly
converging to slighily diverging below. genae
(viewed laterully) almost as wide as compound
eyes and very angular, especially m large
specimens (Fig. 12); scapes slender and reach-
ing median ocellust Aagella 659 as long as
hend width; labrum) with a stout carinate
median tubercle; mandibles tridentate, strongly
constricted subuapically (Fig, 10); legs not as
slender as those of male; metasoma elongate
and rather parallel-sided, more so in large f{e-
males (Figs 22. 23); 6th tergum upturned,
slightly to strongly concave dorsally. rather
Iriangular with small Jateral projections in
small females (Fig. 24) grading to quadrate
im Jarge females (Fig. 25),

Totegument largely glossy with few scattered
small pits; dorsal and lateral arcas of pro-
podem dull with fine roughening; Ist mela-
somal tergum of small specimens with re-
latively few pits conogntrated along posterior
margin (Fig. 28), of larger specimens with
numerous coatser pits some of which exceed
the ocelli in size und many of which coalesce
to form at irregular emargination posteriorly
(Fig. 29), more apical terga with sparse
niedium pitting and fine reticulate sculpturing,
sironges! on terga 4 jo 6. Pubescence generally
sparse, white and inconspicuous, longest on
sides of metasoma and hind tibiae; dorsal arcas
of metasoinal terga lacking thickened or con-
Spicuous setac,

Head and body black) Glypeus with a full
length T-shaped white mark: labrum, man-
dibles and leys largely or wholly orange-hrown;
medium and Jarge females usually have diffuse
oranze-brown patches on mesepisterna, meta-
sternum and anterior metasomal sterna.

The specific name refers ta the 3-pointed
margin of the 6th metasoinal tergum of the
female.

Disrriburian

Semrariu rogians of Suuth Australia ond
southern Western Australia (the mallee Enca-
Iyptus belt und bordering areas),

Specimens examined: The holotype and Lhe faltew
ing. S AUST: 1 # and 14 2 (Coll paratypes).
Corunna Hills, N. of Tron Knob, 19.4v.1971, ex
nest (SAM); 4 9, 8. of tron Baton, Byre Penin-
sula, 30x,1971, on Eremoplile (HC); 40

(paratypes}, 26 2 {including allotype and 22 para-
types}, Lake Gilles National Park, 30-31..0.1973,

TL=0fRiV.9974, b4-17.Vi,1974, 29.viii-1.ix, 1974 and
27-x.1974, ox nests in hollow Hererudendren lwigs,
©. A. aT. &. Huuston (ANIC, HC, pi und SAM
—some in alcohol); 1 (paratype). 3 9, norlhera
Middleback Rungey (137°9'E, 33°3'S), 7-8.x. 1978,
ex hollow Meteradendrim twigs, CG, Ao & TT. B-.
Houston (AC, SAM), 7 9. 8 miles (13 tum) E. of
Pecener, 8.11970, on, Melaleuca pabescens (HC);

L ?, 29 km NNW, of Pt Augusty, 29.1%, dae on
Myaperum (SAM), 3 9, 2 miles (3.2 km) N, of
Port Germejn, 7.7,1970, on Lorawithns niraculosns
and Mefaleucu pubescens (UIC); 1 3. 30 miles (48
km) NNW. of Renmurk, 22.1, 1972, on mallee
Eucalypius (SAM), W,. AEST. LS, 27 miles (43
km) W. of Coolgardie, 38.1,1970, on Evendy pine
(HC): 1 9, 25 miles (40 km) EB. of Kalbarti,
BAITS. KT. Richards Hef aad b | 2 & miles
(13 km) S. of Wanvo, 7-v-1974, TY, Richards
(WADA),
Size-correlated variation in B. tridentata

As ioted in the above description. females
of &. tridemare vary miurkedly in size and form
and slightly in coloration. Same of this varia-
tion is correluted with size and since it is sig-
nificant in terms of the bionomics of ihe bees
it is detailed below.

Individual size has been judged according ta
head ¢apsule width. Measurements of fore
wing. lengths wete made but proved un-
necessary since the ratio of fore wing length to
head width remained approximately constant
OVer a range of mensurements.

With increasing size, the following changes
occur.

(1) The head capsule becomes increasingly
more quadrate, the inner orbits changing
from slightly converging to slightly di-
verging below and the clypeus becoming
wider relative to its length ¢cf, Figs 20,
21).

The metasoma becomes felitively larger
(especially in length}, In the smallest fe-
males the apex of the metasuma does not
extend beyond the tips of the reflexed fore
wings whereas in the largest females it
protrudes beyond them by about 1/3 of
its length (cf. Figs 22, 23). This increase
in relative size is mot due to extension of
the telescopic segments but reflects an in-
crease in size of all the component parts
including the sting.

The latero-apical projections of the 6th
mictasomal terguy become relatively
larger, more obtuse and further apart (cf-
Fins 24, 25). This variatian is quantified
in Figure 39.

The Ist metasomal tergum hecrmes in-
creasingly more coarsely and deeply pitted
and in the Jarger females the pits slong

(4)

Ie
rt

Apuen) width of loth, mérasomot teeauen 6 Meme wldih (7)

a té 1.8 zo 22

Head widrh trem}

Fip, 30. Allometry in females of Exenenra (Exo-
Neuretle) tridentata. Seatter diagram
showing how relative width of apex of
sixth metisomwl tergum increases with
size of individual. The umber of indivi-
duals répresented hy spots increases with
their size in the order J, 2, 3, 4, 5, 6-10,
11-45, 16-20 and 21-25,

the posterior margin coalesce to form an
ittegular emargination (cf. Figs 28, 29).
The integument of the metasoma becomes
relatively thicker and more brittle and the
apodemes become relatively larger (cf.
Figs 26, 27). The sturdier anterolateral
apodemes are associated with relatively
lurger extensor muscles.

Orange-brown patches with diffuse bor-
ders appear on the thorax and metasoma
of medium and lurge females, They occur
on the mesepisterna, mesosterna and
anterior metasomal sterna. The larger the
individual, the more extensive are the
maculations.

A small median spine develops from the
gradulus of the 4th metasomal tergum in
medium and large females.

ti}

CO)

(7)

T. F. HOUSTON

This variation appears to be unique amongs}
allodapine bevs und in the family Antho-
phoridac as a whole, Michener (1965a) dis-
cussed size variation amongst females of the
social] Australian hee, Exenevra (E.) yariabilis
Rayment, In this species, egg layers average
larger than workers but na structural differ-
ences or allometry has been reported. The
degree of size variation in this species is also
less than in &. tridentata, the largest known
females having head widths only 1.28x as great
as the smallest females (cf. 1.54x in E,
iridentata)

THE IMMATURES

The immatures studied wore preserved by
dropping them live into either Kahle’s solution
or 75% ethyl alcohol and were stored in the
latter,

Syed (1963) described 4 larval instars of
EY lawsont from preserved material and
Michencr (1964) described and figured live
specimens of probable 2nd, 3rd and 4th instars
as well as the egg, prepupa and pupa. How-
ever, the identity of the material studied and
described by these wuthors is subject to a little
uncertainty hecuuse a few of the nests from
which it was devived have proven to belony to
EE. setosa (C,. D. Michener—personal com-
munication), Professor Michener confirms thit
the bulk of the adult material on which he
based his 1964 studies is clearly E. Jawsoni and
in all probability the immatures described by
him would be of the same species,

Four morphologically distinct larval instars
can be recognised in each of the known species
of Exoneurella without recourse to histograms
of head width frequencies. Indeed, with
Exoneura tridentata there is such marked size
variation, within each instar and such wide
everlap in size between them that the histo-
gram of head width frequencies was of no
help at all in determining the number of in-
stars.

The terminology employed in the following
descriptions of liurvag follows that of
Michener (1953).

Figs 31-42. Immatures of Exoueura (Exoneurella) eremophila. Fig. 31—Ege. Fig. 32.—First instur
partly enclosed in chorion (left laterul view). Fig. 33-—Second instar with chorion still
attached (left lateral view), Fig. 34,—Ventral view of head and fore body of secand in-
star. Figs 35, 36.—Third and fourth instars, respectively (left lateral views). Fig. 37.—
spiracle of fourth instar, Fig. 38—MNead of fourth instar (anterior view). Figs 39, 40.—1 eft
mandible of fourth instar, anterior and ventral views, respectively. Fig, 41.—Mouthpasts
of fourth inslar (posterior view). Fig. 42.—Female pupa (left lateral view).

NEW AUSTRALIAN ALLODAPINE BEES

FIGS 31-42

23

24 T. F. HOUSTON

E. (E.) eremophila
Beg, (Fig. 31). About 4.1 mm: long and 0.4
mm in maximum width; white, sausave-shoped
and with a coarse Teticulate sculpturing each
end, the middle portion being tuberculate,

In some eggs from Queensland nests the
sculpturing was weak or absent.

first instar—(Fig, 32), Remains almost
wholly withitt chorion, head of very simple
form, lacking lateral lobes, antennae and setac;
mouthparts hardly developed, lobe-like and
probably non-functional: body sac-like without
obvious segmental lines, tubercles or sctac.

Second instar—tFigs 33, 34), Chorion re-
mains attached to abdomen, head selatively
broader than in |st instar but with no obvious
lateral lobes; antennae well-developed, capitate
and laterally directed; mouthparts well-deve-
loped and functional: bead capsule wilh
numerous moderately long setae; body without
setae und distinct segmental lines but with 4
tubercles: each side anteriorly and a middorsal
tubercle on prothorax.

Third instar— (Fig. 35), Entirely free of
chorion; head relatively very broad with well
developed ventrolateral lobes; antennae yery
slender and acute apically; body gently curved
with distinct intersegmental lines, no antero-
lateral tubercles or ventrolateral swellings but
prothorax with a distinct middorsal rubercle;
anal slit moderately deeply incised; sctac
numerous on head and body, longest on
ventrolateral lobes of heady patches of small
setae occur dorsally and laterally on the pro-
thorax, laferally an the following 4 segments
und transverse rows of shart stiff setae occur
dotsolatcrally on the 2nd to 12th badly seeg-
ments,

Fourth instar —(Fivs 36-41). Head relatively
very broad with conspicuous ventrolateral
swellings and slender acute antennae (Fig. 38);
labrum bilobed apically with a few sensoria,
not delimited from distinctly sunken clypeal
region: mandibles slender apically with only a
few minute spines. subapically and a single
sensorium ventrally (Figs 39, 40): manxillae
shorter than labium, their palpi consisting of
indistinct tubercles bearing a few sensoria
(Fiz, 41); labium bearing tubercle-like palpi
latero-apically; body strongly bent at Sih ab-
dominal segment (Fig. 36); prothoras with an
obtuse mid-dorgal tubercle, intersegmental lines
wenk: ventrolpteral body swellings ubsent;
terminal segment of abdomen strongly laterally

compressed With anal slit deeply incised; setn-
tion much as in 3rd instar but all holy seg-
ments have ventrolateral patches of sctac;
atrial aod) piimaty tracheal openings of
spiracles circular; atria without spines but with
a Jew branching and anasloumosing seulpiural
lines; subatria relatively Jong (Fig. 37}.

Prepupa,—Similar to 4th instar except that the
body is straight and swollen anteriorly.
Pupa—{Fig. 42). Conforms essentially to
features of the adult bur the following spectul
pupal structures were noticed: all coxae with
ventro-apical spines (very short and incgon-
spicuous un mid and hind coxae of femules);
vertex (across full width), interantennal area,
upper, middle and lower clypeus with ea-
tremely long setae; 2nd to Sth metasomal seg-
ments also with 2 or 3 lonw setae each side; in
same speciinens the more apical metasomal
terga bear u few tiny setae dorsally,

Material examined.--155 eggs, LOY larvae and pre-
pupac and 46 pupae, New Kalamurina HS, $8.
Aust., SQ ILUL1972, ea dead stems of Myyia-
cephalus; 7 eggs, 48 larvae and prepupac and 2

pupue, 4.8 hm W of Windorah, Qld, 17.14.1969, ex
dead stems of Crotalaria,

FE, (F,) setosa

Eey.—( Fig. 43). About 1.0 mm long and
0.4 mom in maximum width: white, sausage-
shaped and with 4 reticulate sculpourul pattem
(finer than that of £. ereinophila eggs, }

First instar —Nol observed,

Second invstar-—( Pig. 44). Similav te that of EB.
eremophila except that antennae pre not eapi-
tale; Ist and 2nd body segments with a few
small setac dorsally.

Third iusjar—{ Fig, 45). Similiar 10 thot of BL
eremophila hor with more distinct, interseg-
mental Jines and moderately developed ventro-
luteral body swellings. Of 4 specimens ex-
amined, 2 lacked dorsal abdominal setae and
2 slightly Tarver ones had setae on all but the
terminal segment,

Fourth iastar—(Pig. 46). Generally similar
to that of E. eremophila except as follows:
intersegmenial lines more distinct; ventro-
lateral swellings moderately developed; Sth
abdominal segment slightly more proluhetant
dorsally; labial palpi situated more posteriorly
on labiumy and further from apex,
Prepupa—Like 4th instar but body straight
and swollen anieriorty.

Pujt—Generally similar ta thar of &. eres
phile (allowing for differences correlated with

NEW AUSTRALIAN ALLODAPINE HEES a5

1 mm

Wigs 43-46. Immatures of Exonéarg (Bxo-
Nenrella) sétasq. Fig, 43.— Egg, Fig.
44.—Second instar in left lateral
(upper) and dorsal (lower) views.
Figs 45, 46.—Third and fourth in-
stars, respectively (left Jateral views),

adult form); 3rd tq 5th metasomal terga with
a few short setae dorso-apically.

Material examined—21 eges, 32 larvae and pre-
pupae und 12 pupae, West Beach, Adelaide, S.
Aust,. Feb.—Oct. 1965, ex dead stems of Exphor-
bia; 14 eggs, 28 larvae and prepupae and 3 pupue,
4.8 km N. of Peregian Beach, Qld, 9.xii.1966.

E. (E.) tridentata

Ege.—(Fig. 47), Size variable ranging from
1.3-2.0 mm in length and 0.43~0.70 mm in
maximum width; white, sausage-shaped ta
elongate ovoid; chorion entirely smooth to
finely granular (except at the ends),

First instar —(Fig. 48). Remains largely with-
in chorion; head capsule smooth and appoxi-
mately circular in anterior vicw; antennae,
mouthparts and setae absent; body sac-like
without intersegmental lines, tubercles and
setae.

Second insiar—(Fig, 49), Retains chorion on
apex of abdomen: head broad with distinct
lateral lobes which are usually reflexed against
sides of prothorax; antennae absent; mouth-
parts developed and functional: body sausage-
shaped, curved, lacking tubercles and with
weak intersegmental lines; head with numerous
setae, longest on vertex where they are thick
throughout their length; body lacking setae.
Third instar—(Fig. 50). Head rélatively ex-
tremely broad with laterally extended lobes;
untennac present, relatively short (compared
with those of other Exoneurella) and medially
directed; body curved. without tubercles
{except dorsa-apically on terminal segment)
and with weak intersegmental lines; head with
almost a complete covering of short blunt
setae; body without setae.

Fourth instar—(Figs 51-56). Head with ex-
ceptionally large quadrate lateral extensions
(Fig. 53); antennae short, slender, acute and
medially directed; labrum broad und bilobed
with several sensoria ventrally (Fig. 56), not
delimited [rom clypeo-frontal area: mandibles
(Pigs 54-56) slender, tapering and compressed
apically, cach with a pair of sensoria ventrally
but lacking spines; labium rounded and lobe-
like, bearing 2 patches of sensoria which may
represent obsolescent palpi (Fig. 56): a trans-
verse, laterally projecting lobe behind the
labium probably represents degenerate maxillae
(Figs 53, 56): bedy strongly bent at Sth ab-
dominal segment which, like the 6th, bears a
prominent dorso-median tubercle (Fig, 51);
7th ahdominal segment with a small dorsal
tubercle; terminal abdominal segment broad

26 T. F, HOUSTON

Figs 47-56. Immatures of Exoneura (Exoneurella) tridentata. Fig. 47.—Eggs showing extremes of form
and sculpturing. Figs 48, 49.—First and second instars with chorion attached (left lateral
views). Figs 50, 51.—Third and fourth instars, respectively (left jateral views), with en-
largements of capitate seta and spinose apical tubercle. Fig. 52.—Spiracle of fourth instar.
Fig. 53—Head capsule of fourth instar (anterior view). Figs 54, 55.—Left mandible of
fourth instar in anterior and ventral views, respectively. Fig. 56—Mouth parts of fourth in-
star (ventral view. L, labrum; LI, labium; M, mandible; MX, maxilla?). Figs 48-51 are
drawn to same scale.

NEW AUSTRALIAN ALLODAPINE BERS 27

with a moderaicly Jarge spinose tubercle just
above the anus (Fiz. 51); head with mmerous
short obtuse seige: body with short setue dis-
posed in tramsverse bands on prothoracic to
4th abdominal seaments and small dorsal
patches on the 5th to 9th! relatively large thick
capitate selac occur ventrally in clusters on the
metathoracic to 2nd abdominal segments and
singly or clustered Iluterully on the meso-
thoracic 10 4th iabiulominal sezments; spiracles
(Fig, 52) nol protruding above body wall;
autal ark primary tracheal openings circular;
aria subspheroidal without spines or other
sculpture, subglria slender with about 18 an-
nulations.

Prepupo—similar to 4th instar but body
straight, swollen anteriorly and with reduced
dorsal tubercles,

Pupa—Similar to that of &. eremoplila dif-
fering as follows. Head with fewer setae, a puir
being silualed tow on the clypeus and several
across the vertex. the more lateral ones being
much longer than more medial ones; metasoma
with short setue dorsally on segments 2-5 (fe-
mates) and 2-6 (males) fin addition to the
long lateral setae).

Material examined. 2004- eges, 125 latvae and
Prepupac snd 73 pupae, Lake Gilles National

Park, S. Aust, 30,x01.1973-27,..1974, ex desd
slems of Helerodendion olelfolinn.

Discussion

Regarding adult features, the 3 new species
agree almost lotally with the diagnosis of
Frorenrella given by Michener (19654, p.
223). The only points of disagreement relate
io E. tridentata) in this species the eyes of
males are conspicuously swollen, the apex of
the 6th metasomal tergum of females is simple.
net bifid, and the costal margin of the 2nd
cubital cell of the fore wing of buth sexes is
at least as Jong as the Jat tramyverse cubital
vein, not much shorier, In these respects E.
teidentata is rather more like bees of dhe sub-
genus Exoneura than are other Exonenrella. It
is also unlike its closest relatives in the con.
stricted mandibles, angular genac, pronounced
size variation and allometry of females.

Eggs of Ezoreurelia are wnusual amongst
those of allodapines {Michener 1973, p. 281)
in having sculptured chortons, The sculpturing
forms a delicate reticulum in E, lawsent and
E. velosa, fine sranules in &. iridermata (absent
in some specimens) and a cumbination of
coarse reliculum and distinct tubercles in A-
ereniophita.

Similaritics in larval [orm between the 4
species of Exonenrelie correspond to similar.
ties In adult form. hus, larvac of BE. serosa
are most like larvac of &. fawsoni as described
and figured by Michener (1964, pp. 422-424,
gs 13-20) and larvae of E. eremaphila differ
from these 2 species in only a few minor
features. On the other hand, larvar of &,
tridenjata are highly distinctive: the head cap-
sule of 2nd to 4th instars is extraordinarily
produced and quadrate laterally, the antennae
are comparatively liny, the maxillae and
labium are strongly modified, 2nd instars lack
Jateral body tubercles but 41h instars have large
dorsal tubercles on the 5th and 6th abdonnnal
segments, spiny apical tubercles and peculiar
thickened setae on the thoracic and anterior
abdominal segments.

The fentures which will distinguish
Exoneurelia larvae from those of other groups
acc the following. Head capsule fof more
mattire instars) relatively broad with distinet
hairy ventrolateral of lateral expansions; an-
tennae (except in £. rrideniata) of 2nd to 4th
instars relatively long, slender apically, thick-
ened basally and directed anterolaterally:; na
separuhon of clypeus and Jabrum; mandibles
strongly tapered with slender simple apices;
body of 4th instar conspicuously bent at Sth
abdomina) segment which protnides dorsally:
3rd and 4th instars lacking lateral or ventro-
lateral extensions of hody segments such ag
occur in other Exoneura (Syed 1963),

The pupse of Fxoneurella differ from
species to species in conformity with adult dif-
ferences bul otherwise are fairly uniform. Of
the various specialized pupal structures occur-
nng in Apoidea (Michener 1954) the only ones
occuring in Exoneurella ute lang thick setae
on the head and mectasoma, fine shost setae on
the metasomal terga, and coxal spines
Michener (1964, p. 424) remarks on the ab-
sence of coxal and trochanteral spines in E.
fawsen? but I have seen no material which
could confirm this. Specific differences were
noted in the number and arrangement of setae,

Exoneurella. originally established as a genus,
was felegated to subgeneric§ status in
Michener’s (19655) classification of Australian
bees but has continued to receive generic
status (Michencr 1971, 1973). ‘The taxon with
ils new additions remains distinctive and well
defined, T consider it & pulely arbitrary malter
whether one recognizes it at generic or sub-
generic Icevel and have preferred ta Inllow

28 T,.. F. HOUSTON

Michener’s (19656) arrangement since it ex-
presses the obvious affinity between Exo-
neurella, Exoneura s. str. and Brevineura.

Acknowledgments
I wish to thank my wife, Carol Houston, for
assisting with collection of some of the
material described below, Ms J. C. Cardale

(ANIC) for providing some critical details of
the type specimen of Exoneura lawsoni, and
Professor C. D. Michener (KU) for advice
concerning material used in his studies of
Exoneurella.

Some of the material listed herein was col-
lected during field work funded by the Royal
Society of South Australia.

References

MICHENER, C, D. (1953),—Comparative morpho-
logical and systematic studies of bee larvae
with a key ta the families of hymenopterous
Jarvac. Univ. Kans, Sci. Bull, 35(8), 987-
1102.

Micnener, C. D. (1954).—Observations on the
pupae of bees (Hymenoptera: Apoidea). Pan-
Pacif. Ent, 30(1), 63-70.

MIcHENER, C. D. (1963),.—New Ceratinini from
Australia (Hymenoptera, Apoidea). Uniy.
Kans. Sci, Bull. 49(7), 257-261.

Micuener, C, D, (1964).—The bionomics of Exo-
neurella, a solitary relative of Exeneura
(Hymenoptera: Apoidea: Ceratinini). Pacif.
Insects 6(3), 411-426.

MICHENER, C. D. ([965a).—The life cycle and
social organization of bees of the genus
Exoneura and its parasite, Jnguilina (Hymen-
optera; Xylocopidae). Univ. Kans. Sci. Bull.
46(9), 335-376.

MICHENER, C. D. (1965b).—A classification of the
bees of ihe Australian and South Pacific
regions, Bull, Am. Mus. nat. Fist. 130, 1-362.

MicHener, C. D. (1971).—Biologies of African
allodapine bees (Hymenoptera, Xylocopinae).
Bull, Am. Mus. nat. Hist. 145(3), 219-302.

Micuener, C. D. (1973).—Size and form of eggs
of allodapine bees. J. ent. Soe. sth Afr. 36(2),
281-285.

RAYMENT, T. (1946).—New bees and wasps—
Part Il. Victorian Nat. 62, 230-236.

SmitH, F, (1854) —“Catalogue of hymenopterous
insects in the collection of the British
Museum. II.” (British Museum: London.)

Syrp, I. H. (1963).—Comparative studies of lar-
yae of Australian ceratinine bees (Hymenop-
tera, Apoidea). Univ. Kans. Sci, Bull, 44(8),
263-280.

FAULTING CONTEMPORANEOUS WITH UMBERATANA GROUP
SEDIMENTATION (LATE PRECAMBRIAN), SOUTHERN FLINDERS
RANGES, SOUTH AUSTRALIA

BY P. §. PLUMMER* AND V. A, GOSTIN*

Summary

PLUMMER, P. S., & GOSTIN, V. A. (1976).-Faulting contemporaneous with Umberatana Group
sedimentation (Late Precambrian), southern Flinders Ranges, South Australia. Trans. R. Soc, S.
Aust. 100(1), 29-37, 28 February, 1976.

Interglacial sedimentation within the Late Precambrian Umberatana Group (Adelaide System) was
greatly influenced locally by contemporaneous faulting along the Spring Creek Mine Fault.
Stratigraphic study both sides of the fault and of the fault zone has revealed a variable sequence of
elastics and carbonates. Lithological correlation across the fault zone suggests two periods of
faulting contemporaneous with deposition, followed by a phase of downwarping of the area north of
the fault along a hinge line coinciding with the earlier fault zone. This activity resulted in a
thickness increase of approximately 750 m in the sequence north of the fault when compared with
that of the more stable sequence to the south. Three varieties of stromatolites occur within the area
studied, displaying distributions which were influenced by varying water depths.
Palacoenvironmental interpretations based on vertical and lateral lithological associations,
palaeocurrent analysis and studies of the contained stromatolites reveal that sedimentation south of
the fault occurred within shallow marine, intertidal and supratidal environments resulting from the
prevailing regional regressive-transgressive-regressive marine cycle. Modification of this cycle
north of the fault zone was influenced by contemporaneous faulting.

FAULTING CONTEMPORANEOUS WITH UMBERATANA GROUP
SEDIMENTATION (LATE PRECAMBRIAN), SOUTHERN
FLINDERS RANGES, SOUTH AUSTRALIA

by P. 8. PLuuMMeR® and V. A. Guostin*

Summary

PeumMer, PF, S. & Gostin, V, A. (1976).—Faulting contemporaneous with Umberatana
Group sedimentation (Late Precambrian), southern Flinders Ranges, South Australia,
Trans. R. Soe. §. Aust. 100¢1), 29-37, 28 February, 1976,

Intergiacial sedimentation within the Late Precambrian Umberatuna Group (Adelaide
System) was greatly influenced locally by comemporaneous faulting aluoe the Spring Creek
Mine Fault, Stratigraphic study both sides of the fault and of the fault zone has revealed a
variable sequence of clastics and carbonates, Lithological correlation across the fault zone
suggests two periods of faulting contemporaneous with deposition, followed by a phase: of
downwarping of the area north of the fault along a hinge line coinciding with the earlier fault
xone, This activity resulled in a thickness. increase of approximately 750 mw in the sequence
north of the fault when cornpared with that of the more stable sequence to the south. Three
varieties af stromatolites occur within the area studied, displaying distributions which were in-
Nuenced by varying wuler depths.

Palaeoenvironmental interpretations based on vertical and lateral lithological associations,
palaegcurrent analysis and studies of the contained stromatolites revenl that sedimentation
south of the fault occurred within shallow marine, iatertidal and supratidal environments resullt-
ing from the prevailing. regional regressive-transgressive-regressive marine cycle. Modification

o€ this cycle north of the faulr zone was influenced by contemporaneous faulting.

Introduction

This paper deals with Umberatana Group
sédimentation in a region of contemporaneous
fwulting on the western flank of the Flinders
Ranges, 12 km NNW. of Melrose (Fig, 1),
This is in the southeast quadrant of the
ORROROO 1:250,000 geological sheet map-
ped by Binks (1968). Deposits of the Sturtian
and Elatioa glaciations form the basal and
capping Components, respectively, of the Um-
beralana Group. The intervening period is re-
presenied by a variable sequence of clastic and
carbonate sediments of non-glacigene origin.
{Thomson 1969).

The study area is divided by » fuule zone
into two subareas. of differing stratigraphies,
These reflect pehecontemporaneous movement
on the east-west Spring Creck Mine Fault
(new name). The regional pattern of the urea
was one of a regressive-transgressive-regressive

marine cycle, represented by the stratigraphy
of the southern subarea (Fig. 7}. However,
fault. rupture and sitbsequent subsidence north
of the fault zone produced a short transgres-
sion within the initial regressive phase, but the
main sedimentation cycle regained dominunce
as the sedimentation rate execeded the tate of
subsidence.

Time-stratigraphic relationships, as shown in
Fig. 7, are based on the assumption that the
fault zone cun be interpreted as in Fig. 5. The
nature af outcrop is not sufficent, however, to
fully confirm this interpretation, and the pos-
sibility of complex fault slices of the strati-
graphy contained within the fault zane is not
discounted,

Straligraphic nomenclature (Fig. 2) is that
of Thomson et al, (1964) and Thomson
(1969). Detailed geology is deserihed in
Plummer (1974)!,

*Centre for Precambrian Research, Department of Geology and Mineralogy, University of Adelaide,

Adelaide, S. Anst, 5000.

' Plummer, PLS. (1974). —The siratigraphy. sedimentology und palagoenvironments of Late Precarn-
briatt Uniberatana Group in the Mount Remarkable-Alligator Gorge area, South Australia, Univ,

Adel, linpublished Honours thiesis,

30 P. S&S. PLUMMER & VY, A. GOSTIN

SOUTH AUSTRALIA

Melrose
12 km

j= = = Elatina Formation

j

fe. sandstones, siltstones
Wilmington Formation

fine sandstones, siltstones, limestones

Angepena Formation
fine sandstones with clay & coarse sand

Brighton Limestone
limestones some silty, dolostones
with clay or coids

Tapley Hill Formation
laminated siltstones

Sturt Tillite
L sandstones, tillites
v\e - \¥ \ \ Md
r\i\ I \ Vy \ eee Fault-rock
‘ a Vite i Wy \W )
Vey A \\y
\ NN Va \V\ Stromatolite

Fig. 1. Generalized geological map of the Umberatana Group in the study area.

FAULTING AND SEDIMENTATION, SOUTHERN FLINDERS RANGES 31

Stratigraphic Nomenclature

Cambrian Elatina
Formation
Wilpena
Wilmington
Group Formation

Angepena
Umberatana Formation

Group Brighton
Limestone

ADELAIDE

Tapley Hill
Formation

Sturt
Tillite

Northern
Subarea

Southern
Subarea

Lithologies

siltstone
fine sandstone
limestone

dojostone

HEGOE

silty limestone

ee

©

\3/s

Scale in
Metres

silty dolostone
calcareous siltstone
stromatolite

ooid

coarse sand, flat dolomite pebbles

Fig. 2. Stratigraphic nomenclature and general stratigraphic columns of the two subareas divided by
Spring Creek Mine Fault. Thickness variations controlled by subsidence north of the fault.

32 P. S. PLUMMER & Y. A. GOSTIN

=
Stromatolite biaherm

arbitrary

suriaces

PSP'75

isochrons'

Stromatolite Facies
anes small columnar
BARA (2=5om wide)

large = columnar
ee {7-zocm wide)

SBE stratitorm
a.
25 Intrafermational limestone
a2

Fre. 3. Diagramatic favies sketch of the base and edge of a stramatolite bioherm, Middle Brighton

Limestone, southern subsren.

Sedimentation prior to faulting

Following the Sturtian glaciation, interglacial
sedimentation began in a marine basin helow
the wave base, outside the zone of carbonate
deposition, and under reducing conditions. The
resultant lithologies, namely black, finely lam-
inated pyritic quarizose shales grading upward
into well laminated dark prey quurtzose silt-
stones, form the ‘Tapley Hill Formation
(Howchin 1904; Thomson ef al. 1964).
Gradually the sediment surface rose into the
zone of carbonate deposition, resulting in an
increase in carbonate content toward the top
of this tormation. Eventually carbonate be-
cume dominant as a well laminated dark grey
silty limestone was deposited (basal Brighton
Limestone—Howchin 1904). Above this, the
presence of large dendroid branching stromuto-
lites, set in a blue-grey limestone, testify to
the basin floor rising above the level amenable
to stromutolite growth. Current activity during
this initial period of uniform sedimentation
was minimal, the dominant direction ivitially
being from the north-west, but later swinging
(to Now from the south-west (Fig, 6). A
gradual increase in current intensily is sug-
gested by the change from rare small cross-
laminations (lower Tupley Hill Formation) to
more common shallow ripple marks (basal
Brighton Limestone).

Faulting with sedimentation
Movement ulong the Spring Creek Mine
Fault ended Lower Brighton Limestone sedi-

mentation and produced variations in litha-
Jogies across the fault. A sketch of the fault
zone is shown in Fig. 5. Tentative correlations
of tocks in this zone with adjacent lithologies
suggest two periods of faulting (Fig. 7), the
total stratigraphic throw being approximately
750 m, Fault-zones facies 1 (FFI)—a dark
grey micritic limestone (breeciated in placesi,
with stromutolitic Jimestone megaclasts (Wp to
0.5 m long) of the previously deposited
Brighton Limestone—suggests initial rupture
along, the fault, whereas subsequent facies lack-
ing megaclasts (and FF2 lacking brecciation)
imply warping due to very gradual subsidence,
Seuthern Subarea
South of the fault, shallowing steadily con-
tinucd, At, or just after the time of initial
fuulting, the blue-yrey stromatolitic limestones
of the preceding period grade slowly upward
into massive deep reddish-purple intra-tormu-
tional limestones. containing abundant ‘bull-
yellow stromatolite bioherms, displaying large
furcate to small digitate branching columns, as
the sedimentation surface rose into the zone af
Wave agitation, oxygenation and tidal activity.
Within these bioherms, a zonation based on
stromatolite morphology occurs (Fig. 3), sug-
gesting a series of growth stages tesulling from
energy fluctuations within the depositional cn-
vironment, indicating emergence of the bio-
herms through the intertidal zone in’ the
following manner:
(a) basal stratiform mats lrapped sediment
forming, a stable luyer within the hizgh-

FAULTING AND SEDIMENTATION SOUTHERN FLINDERS RANGES 33

Fig. 4, Tepee structure, Upper Brighton Limestone, northern subarea,

energy subtidal to low intertidal zone;

(b) ‘large’ columns (7-20 em wide) then
developed within the lower to middle
intertidal zone. Abundant carbonate
flakes, originating from disruptive Wave
and tidal action, lodged between these
columns;

(c) ‘small’ columns (2-5 cm wide) cap the
bioherms in the high intertidal zone.
Rare disruption of laminae, or an ef-
ficient drainage of debris from the
intercolumn maze explains the relative
lack of carbonate flukes in this zone;
and

(d) stratiform layering on the bioherm
flanks encasing both facies (b) and (c)
enhanced drainage from the centre of
the bioherm,

Ps

Fault-zone facies 2, a yellow dolostone,
lacking brecciation and in places sandy, sug-
gesting gradual subsidence and warping of the
still soft sediment. and FF3, a brecciated pale
yellow dolostone, were probably deposited
during this period of stromatolite proliferation.

Continued shallowing then produced a
supratidal environment of deposition and
completed the initial regressive sequence of
the interglacial period south of Spring Creck
Mine Fault. A purple dolostone, with disrupted

clay laminae and lenticular beds of quartz
sand, caps the Brighton Limestone. Fault-zone
facies 4 and 5—brecciated purple dolostones
(with clay in the former)—are the lime
equivalents of this unit, although north of the
fault these two facies possibly represent the
time equivalents of two very different litho-
logics (uppermost Brighton Limestone and
Angepena Formation).

Northern Subarea

At the time of initial rupture along Spring
Creek Mine Fault, the northern block was
downathrown below the depth amenable to
stromatolite growth, yet remaining within the
zone of carbonate deposition, resulting in
laggy grey limestones bearing 4 minor quartz
silt fraction. The thickness (125 m), and uni-
formity of this unit is a result of equivalent
rates of subsidence and sedimentation. Rare
large scale tipple marks (wavelength 50 em.
amplitude 8 cm) and cross-bedding (up to
2.5 m per set with foreset slopes between 15°
and 40°), plus the more abundant small scale
current-ripples and cross-stratification, suggest
an increase in current activity attributed to the
change in seafloor topography. Palacocurrent
analysis reveals several modes us shown in Fig.
6. Rare lenticular outcrops of grey stromato-
litic and purple intraformational limestones

34 P. S. PLUMMER & V. A. GOSTIN

suggest that the depths of oxygenation, agita-
tion and stromatolite growth were almost
juxtaposed.

Above this unit a change in colour and litho-
logy suggests shallowing of the basin to an
oxygenated wave agitated depositional environ-
ment, within which a 20 m thick unit was de-
posited. This unit exhibits facies variation from
pink dolomitic oolites with birdseye structures,
and sandy dolostones near the fault grading
northward into white limestones with dolomite
flakes, generally graded within trough cross-
beds, and rare ooids. A gentle palacoslope
northward away from the fault is therefore
suggested. A supratidal environment followed,
depositing a 5 m thick purple dolostone with
clay laminae exhibiting tepee structures (Fig.
4) in the vicinity of the fault. This unit, pos-
sibly equivalent to fault-zone facies 4 (and ?
5), caps the Brighton Limestone north of
Spring Creek Mine Fault as does its counter-
part to the south. It therefore appears that
supratidal deposition occurred uniformly over
the entire area at this time, suggesting a tem-
porary quiescence of fault activity (Fig. 7).

These purple dolostones were overlain by
350 m of the greyish-red fine sandstones of
the Angepena Formation (Thomson 1969).
These contain abundant purple clay laminae
providing excellent bedding plane partings re-
vealing a wealth of ripple marks and both
polygonal and sinuous mudcracks. Lenses of
coarse-grained rounded sand are common,
often graded and draped with clay. These
sedimentary structures typify a vast tidal flat
to distal fluviatile depositional environment.
The abundance of desiccation cracks, some
reaching 1 cm in width, favours a non-
marine influence. Palaeocurrent analysis re-
veals a dominant easterly drainage pattern with
minor southern and western components (Fig.
6).

This dominantly regressive sequence north
of Spring Creek Mine Fault was then overlain
by the transgressive grey-green deposits of the
Fig. 5. Facies sketch map of the fault zone, lowermost Wilmington Formation (Thomson

Spring Creek Mine Fault. 1969). Siltstone pillows in dolostone are over-
FFI: dark grey micritic limestone with Jain by grey slightly calcareous siltstones with
wtaces megaclasts, brecciated in Chioritic wavy laminae, ripple marks and rare

FF2: yellow dolostone, sandy in places. sinuous mudcracks. These in turn are overlain
by a massive grey calcareous siltstone bearing

FF3: brecciated pale yellow dolostone. ] “
FF4: brecciated purple dolostone with pillow structures, slump beds and _ lenticular

a

Bal
taal
ho

Aw

[NN

—
\=\
\
\J

fa

clay. beds of olive green-brown oolitic limestone. A
FFS: brecciated purple dolostone. very shallow, yet wholly submerged environ-
FF6: grey silty limestone. ment is therefore suggested. Fault-zone facies

Formation symbols as in Fig. 7. 6, a grey silty limestone, is possibly the equiva-

FAULTING AND SEDIMENTATION, SOUTHERN FLINDERS RANGES 35

ne

-pP=S
a=)
a]
Ee
[=
we

q
2
3
B
a
E
Sh

Current Roses
@) Number of readings

Ripple Marks
stippled: symmetrical

Formation

solid: asymmetrical

Cross-stratification

Wilmington
Brighton

PSP'75

Fig. 6. Palaeocurrent analysis of the Umberatana Group in the study area.

Stratigraphy
Southern

Average
Thickness
(mn)
=a

ria. 7,

PS. PLUMMER & Y¥. A. GOSTIN

Fault
Activity

Stratigraphy
Northern
Subarea

downwarping

‘Vime-stratipraphic suramary chart of lithologies, fault activity and palueocenvironments. Lithalo

gies as in Fig, (2. THI: Tapley Hill Formation; BL; Brighton Limestone; AF: Angepena Forma-
tioh; WE: Wilmington Formation; BF: Elatina Formation; L: Lower; M: Middle; U: Upper: dg:
dark grey: g: grey: ert green; r+ red and. purple; w: white and pink: y: yellow.

lent of this period, and represents the final
phase of faulting (Fig. 7).

South of the fault, the Angepena Formation
and the two lower parts of the Wilmington
Formation are absent. The contact between the
Upper Brighton Limestone and the Lower
Wilmington Formation is nowhere visible. It
is suggested that either an hiatus in deposition
occurs between the Upper Brighton Limestone
und Lower Wilmington Formation, or alterna-
tively the Upper Brighton Limestone desposi-
Lier continued in a restricted, slightly elevated
covironment, whilst to the north the much
thicker Afngepenu Formation and the two
lower Units of the Wilmington Formation were
fad down.

Continued warping

Deposition was dominantly ternigenous as
Spring Creek Mine Fault cyolved from a zone
of separation of the two subareas into a flexure
with tore marked subsidence of the northern
subarea, The fault zone was still a major influ-
ence of Wilmington Formation sedimentation
as [ucies changes and thickness Variations uc~
curred above, of immediately adjacent to it.

Massive dark grey very fine sandstones over-
lain hy brown fine and medium sandstones of
the lower Wilmington Formation grade south-
ward into flagey grey siltstones supporting the
palacocurrent data of a west-north-wester'ly
source area. Widespread slumping and cross-
bedding within included sandstone beds imply
a palaeoslape (possibly submarine deltaic)
away from this source region.

The ensuing calcareous member of the up-
permost Wilmingion Formation {herein re-
warded as the equivalent of the Trezona
Formation of Thomson er ai, 1964) repre-
senis the muximum transgressive phase of
deposition. Flagey grey silty limestoncs and
niassive vrey, in places stromatolitic, limestones
in the south thicken rapidly north of the fault
zone (by up to 5 times) and interdigitate with
n buff coarse sandy cross-hedded limestone
bearing ooids and imbricate) dolomite intra-
clasts, The repetitive nature of this interfinger-
ing, which commences directly above the Fault
zone (Plummer 19744, Fig. 10), probably re-
sulled from fluctuating rates of sedimentation
and subsidence within a shallow sea.

FAULTING AND SEDIMENTATION, SOUTHERN FLINDERS RANGES 37

The final regressive coarsening upward se-
quence of laminated siltstones to fine and
coarse sandstones suggest prograding deltaic
sedimentation. Ripple marks, cross-bedding
and some slide-slump bedding support a west
to east palaeoslope, (Fig, 6). Cappimg the
Umberatana Group are the massive purple
siltstones of the Elatina Formation bearing
(?iee-rafted) erratics. Equivalents of these
siltstones elsewhere in the Flinders Ranges are
distinctly glacigene in origin (Elatina glaciation
of Mawson 1949).

Conclusion

Stratigraphic study south of the Spring
Creek Mine Fault reveals that a regressive-
transgressive-regressive cycle of marine sedi-
mentation took place during the deposition of
the interglacial portion of the Late Precamb-
rian Umberatana. Group. The sedimentary
structures and presence of stromatolites within
these sediments indicate that environmental

conditions ranged from shallow subtidal,
through intertidal to supratidal, then back to a
shallow subtidal environment.

To the north of the Spring Creek Mine
Fault, two periods of faulting contemporaneous
with deposition caused water depth fluctuations
during the initial period of shallow subtidal
deposition, then rapid shallowing through
intertidal and supratidal conditions to a mar-
ginal marine (possibly distal floodplain) en-
vironment, This faulting and later subsidence
hinged along the fault zone, caused a 750 m
thickening of the succession to the north of
the fault.

Acknowledgments
The authors wish to thank Dr B. Daily for
critical examination of this text and helpful
discussion in the ficld. Acknowledgment is
also made to ihe Australian Rescarch Grants
Committce for supporting the research of Dr
VY. A, Gostin,

References

Bin«s, P. J. (1968) —ORROROO map shect, Geo-
logical Atlas of South Australia, 1:250,000
series. (Geol. Surv. S. Aust. Adelaide.)

Bryvgs, P. J. (1971).—The Geology of the
ORROROO 1:250,000 map area. Rep. Invest,
geol, Surv',, 8. Aust. 36,

Coats, R. P, (1973.).—Explanatory notes for the
COPLEY 1:250,000 geoiogical map, Geel.
Sty. 5. Aust.

Forges, B, G, (1971),—A table of Adeluidean
and Cambrian stratigraphic names. Quart.
geol. Notes, geal, Surv, 8. Aust. 38.

Howcuin, W. {1904).—The geology of the Mount
Lofty Ranges, Part 1. The coastal district.
Trans. R, Soe. §, Aust. 28, 253-280,

Howcuin, W, (1916).—The geology of Mount
Borariabie: Trans. R. Sac. §. Ausi. 40, 545-

Mawson, D. (1949).—Elatina glaciation: a third
recurrence of glaciation evidenced in the
Adelaide System. Trans. R. Soc. S. Aust. 73,
117-121.

Smuru, D, B. (1974) —Origin of tepees in Upper
Permian Shelf carbonate rocks of Guadalupe
Mountains, New Mexico, Bull. Am. Ass. Per
rol, Geal. 58, 63-70.

Tuomson, B. P. (1969)—Precambrian Basement
Cover—The Adelaide System. Jn L, W., Par-
kin {Ed.), “Handbook of South Australian
Geology”, pp. 49-83. (Geol. Surv. S. Aust.:
Adelaide).

TuHomson, B. P., Coats, R. P.. MirAms, R. C.,
Forses, B, G., DeiGarno, C, R., & JOHNSON,
J. BE. (1964).—Precambrian rock groups in
the Adelaide Geosyncline: A new subdivision,
Quart, zeal. Notes, geol, Surv. 8, Aust, 9.

SMALL FOSSIL VERTEBRATES FROM VICTORIA CAVE,
NARACOORTE, SOUTH AUSTRALIA
IV. REPTILES

BY MEREDITH J. SMITH

Summary

SMITH, Meredith J. (1976) .-Small fossil vertebrates from Victoria Cave, Naracoorte, South
Australia. Trans. R. Soc. S. Aust. 100(1), 39-5 1, 28 February, 1976.

Reptile fossils have been found at Naracoorte, South Australia, in a Pleistocene cave deposit that is
rich in marsupial and rodent remains. Reptile vertebrae are abundant and a few jaws and limb bones
have been recovered. The diagnostic features of these bones are described.

Of the twelve reptile species present, nine still live in the Naracoorte area; they are three elapid
snakes, Pseudonaja c.f. P. nuchalis, Notechis c.f. N. scutatus, and Pseudechis c.f. P. porphyriacus;
and six lizards, Varanus varius, V. gouldii, Trachydosaurus rugosus, Tiliqua nigrolutea, Egernia
c.f. E. whitei and a species consistent with Sphenomorphus tympanum. A fourth elapid snake
represented by 40 isolated vertebrae, and a species of Amphibolurus have not

been identified.

The remaining species is a boid snake, described here as a new genus and species, Wonambi
naracoortensis. The eight vertebrae recovered are large, suggesting a length of at least 5 m for the
whole snake. Morphologically, the vertebrae differ strongly from those of other Australian boids in
having a high but back-sloping neural spine, paracotylar foramina present, accessory processes
absent, and, particularly, in having large paradiapophyses that extend further laterally than the
zygapophyses. These vertebrae closely resemble those of Madstoia bai Simpson from the Eocene of
Patagonia, but without cranial remains of both species, no relationship can be postulated between
Wonambi and Madstoia.

SMALL FOSSIL VERTEBRATES FROM VICTORIA CAVE,
NARACOORTE, SOUTH AUSTRALIA
Ty. REPTILES

by Merepiru J. SMITH

Summary

Smiry, Meredith J, (1976),—Small fossil vertebrates from Victoria Cave, Naracoorte, Sowih

Australia. Trans. R. Sec. S. Aust. 100(1), 39-51, 28 February, 1976.

Reptile fossils have been found .at Naracoorte, South Australia, in a Pleistocenc cave
deposit that is rich in marsupial and rodent remains. Reptile veriehbrac are abundant and a
few jaws and limb bones have been recovered. The diagnostic features of these bones are
described.

OF the iwelve reptile species present, nine still live in the Naracoorte area; they are three
tlapid snakes, Psexdonaja c.f. P. nuchalis, Notechis c.f. N. scutatus, and Pseudechis cf. P. por-
phyriacus; and six lizards, Varanis varius, V. gouldii, Trachydosaurus rugosus, Tiliqua niero-
lutea, Egernia ct. E, whitei and a species consistent with Sphenomorphus tympanum. A fourth
elapid snake represented by 40 isolated vertebrac, and a species of Amphibolurus haye not
been identified.

The remaining, species is a boid snake, described here as a new genus and species,
Wonumbhi naracoertensis. The eight vertebrae recovered are large, suggesting a length of al
least 5 m for the whole snake. Morphologically, the vertebrae differ strongly from those of
other Ausiralian botds in having a high but back-sloping neural spine, paracotylar foramina
present, accessory processes absent, and, particularly, in having large parudiapophyses that
extend further laterally than the zygapophyses.. These vertebrae closely resemble those of
Madstoiq bai Simpson from the Eocene of Patagonia, but without cranial remains of both

Species, no relationship can be postulated between Wonambi and Madstoia,

Introduction

The reptile faunas of Australian fossil
deposits have rarely becn completely anatysed.
For some deposits, the presence of unidentified
repliles has been noted (c.g, Archer 1974,
Dortch & Merrilees 1971, Gill & Banks. 1956,
Lundelius 1963); for other deposits, the most
distinctive species have been identified, but
often ta genus only (eg, Merrilecs 1968;
Thorne 1971). Exceptions are the carpet snake,
Python variegatus (=Python spilotus varie-
gets) associated with the extinct marsupial,
Thylacoleo sp., and other marsupial remains at
Marmeor Quarry, Queensland (Longman 1925)
and the sleepy lizard, Trachydosaurus rugosus,
at Gore Limestone Quarries, Queensland
(Longman 1945), Remains of a large extinct
Varanid lizard, Megalania prisca, have been
found in Pleistocene deposits in Queensland,
New South Wales and central Australia (Fejer-
vary 1918, 1935; Hecht L975).

The deposit in Victoria Cave at Naracoorte,
South Australia, is probably of Pleistocene age
(Smith 1971). Among the large animals,
eXtinet species are common (yan Tets & Smith
1974; Wells, pers. comm.), but, in contrast, the
smal] marsupials and small birds are referrable
to modern species, though not all of them
occur in the Naracoorte area now (Smith
1971, 1972; van Tets & Smith 1974),

Por identifying reptile species, characteristics
of skull fragments, jaws and teeth are of less
value than they are for mammiails. As reptiles
grow throughout life and are polyphyodont, the
“adult” dention cannot be defined us it can in
mammals. The variations in dentition between
species in many genera is no greater than
within species. Fortunately the vertebrae of
reptiles are of diagnostic value and Auffenberg
(1963) was able to identify single vertebrae of
North American snakes to genus and often to
species. Diagnostic vertebyal characters have

* Department of Zovlogy, University of Adelaide, Adelaide, SA. 5000.

40 MEREDITH J. SMITH
accessory

process colyle
subcentral foramen
prezygapophysis paratiapophysis
hypapophysis: postzyyapophysis
zyyantrum—_ eandvla
zyausphen’ neural spine

Jateral foramen

arapophysial
oA hall centrum

cundyle

Elapid Varanid Scincid Agamid

e
Fig. I. Ventral (above) and lateral views of precaudal vertebrae of (a) Pseudonaja nuchalis
(Blapidae), (b) Veranus gouldit (Varanidae), (¢) Tiliqua occipitalis (Scincidae), (d) Amphi-

bolurus barbatus (Agamidae).

Fig. 3. Vertebrae of Wonambi naracoortensis arc distinguished by their wide paradiapophyses (p) and
presence of paracotylar foramina (f) as seen in (a) anterior view of P16144k and (b) lateral,

and (c) posterior view of P16i44s.

Fig. 5. The prezygapophysial facets are less upturned in Trachkydosaurus rugosus (6) than in Tiliqua

nigrolutea (a).

Fig, 6. Distally, the fused sacral pleurapophyses are cup-shaped in Trachydesauras rugosus (a) and T,
occipitalis (b) but are triangular in T. nigralutea (c), T. scincoides (d) and Egernia cunnine-

hami (€).

not been established precisely for any Aus- Methods
tralian snake species (Smith 1975). Fossil re-
mains of lizards include not only skull bones

and yertebrae but also some limb bones and follow, Smith, (1971).

Methods of collection and preservation

some elements of pectoral and pelvic girdles, The present maximum depth of excavation
These are briefly described, and their diagnos- is 80 cm, although hone chips occur in cores

tic values assessed in this paper. taken as deep as 2,5 m,

SMALL FOSSIL REPTILES FROM NARACOORTE 41

TABLE 1

Some dimensions of the holotype (P16168) and the seven paratype vertebrae of Wonambi naracoortensis
(Dimensions in mm)

Specimen

P16168 P16170a P16166 P16129t P1l6144k P16144s

P16167 P16170b

Length between zygapophyses 16.5 15.4
Height (centrum + neural spine) 32.3 34.1*
Width across prezygapophyses 25.6 21.1
Width across paradiapophyses 28.5 22.3
Minimum width of centrum 18.6 15.9
Width of zygosphene 9.2 8.4
Width of condyle 10.8 8.2
Length of prezygapophysis 8.1 6.1

* Height includes length of hypapophysis.

Skull and jaw elements of all but the most
robust species were rarely recovered from the
Victoria Cave deposit, whereas vertebrae were
common. Consequently, for the diagnosis of
reptile species in this fauna, vertebrae have
been considered in detail and other bones more
briefly. Comparisons have been made with dry,
disarticulated skeletons, and occasionally with
cleared, alizarin-stained whole specimens.

Descriptive terms (Fig. 1) follow Auffen-
berg (1963). The “length” referred to in des-
criptions of vertebrae is the greatest distance
from the anterior edge of the prezygapophysis
to the posterior edge of the postzygapophysis
(Pr-Po of Smith 1975). The ranges of lengths
are given, with mean and standard error.
Measurements were made to the nearest 0.1
mm, with dial-reading, needle-point calipers.
The fossil specimens are lodged in the South
Australian Museum (SAM).

Results
Family BOIDAE

Boid vertebrae lack parapophysial processes,
the accessory processes are very short or absent
and the vertebrae lack hypapophyses on the
posterior two thirds of the precloacal column
(Hoffstetter & Gase 1969).

Wonambi n. gen.

Definition: Vertebrae characterized by a high,
backwardly-sloping neural spine; slightly up-
turned zygapophysial facets; large paradiapo-
physes extending laterally beyond the zygapo-
physes; and a pair of paracotylar foramina.
Type species: Wonambi naracoortensis

Content: W. naracoortensis is the only known
species in the genus.

“Wonambi” is derived from an aboriginal
name for the mythical rainbow serpent (Elkin
1964).

18.1 16.4 19.3 21.8 23.4 19.6
40,8* 33.1 38.4 37.1 41.0 36,2
27.7 25.8 30.4 33.0 35.5 29.5
29.7 27.7 33,2 41,1 43.7 33.0
21.2 19.1 22.7 24.1 27.1 21.8
11.8 9.3 12.6 10.9 12.4 11.3
10.6 10.3 12.6 12.3 12.7 12.4

1.5 74 8.9 11.8 12.8 10.9

Wonambi naracoortensis n. sp.

Holotype: SAM, P16168. A dorsal vertebra
collected in Fossil Chamber, Victoria Cave,
Naracoorte, S. Aust., at a depth not greater
than 30 cm below the surface of the cave
earth.

Definition: The same as for the genus Wonambi
until other species are described.

Description: The neural spine is high (Fig.
2C); its anterior edge begins near the rim of
the zygosphene and rises obliquely to the hori-
zontal dorsal edge of the spine. The spine over-
hangs slightly posteriorly. The zygosphene is
narrow (Table 1) but it is so heavily thickened
that it is as deep in dorsoventral extent as it is
broad. The zygosphenal facets are almost ver-
tical (c. 70° to horizontal) (Fig. 2D). The
relatively small zygapophyses are slightly up-
turned (c. 25° from horizontal). Accessory
processes are completely absent. The paradia-
pophyses are large (Table 1); the upper part
of the articular facet is convex and protrudes so
far from the centrum that the maximum width
of the vertebra is the width measured across
the paradiapophyses; the lower part of the
articular facet is flat. The cotyle and condyle
are slightly depressed (Fig. 2A) and the top of
the condyle is tilted forwards at c. 75° to the
vertical. The ventral surface of the centrum is
smoothly rounded, with weak subcentral ridges
and a low median ridge that terminates pos-
teriorly as a blunt haemal keel, notched in the
midline (Fig. 2B).

Each foramen of the subcentral pair is
located close to the median ridge at about mid-
centrum; each foramen of the lateral pair lies
on the neural arch pillar about halfway between
paradiapophysis and postzygapophysis. There
appear to be two pairs of parazygantral fora-
mina, but, as the bone is pitted in this region.

42 MEREDITH J. SMITH

Fig. 2. The holotype of Wonambi naracoortensis (P16168) in dorsal (A), ventral (B). lateral (C), an-
terior (D), and posterior (E) views.

SMALL FOSSIL REPTILES FROM NARACOORTE AR

foraming are distinginshed with difficulty fron
pits. ‘lhe two foramina of a paracolylar pair
lie close to the rim of the coryle near the top
of the centrum,

Fariation; The paralype series consists of seven
dorsal vertebrac. P16129t, Pl6i44k,s, P16166-
7, and P16070a, b all but une of which were
found in the top 30 cm of the cave carth of
Fossil Chamber, Victoria Cave. All specimens
share with the holotype the diagnostic features,
especially ihe widely spaced paradiapophyses
(Table 1, Fig. 3), The parazygantral foramina
consist Of a single pai which are sunk into
hollows in P16129s, P16167 and P16170b; and
m Pl6144k the single pair of zygantral fora-
mina can be seen clearly. The haemal kcel is
notched posteriorly im P16129s, P16167 and
PL6170b (as in the holotype), these vertebrae
having occurred nearer the tail than the others,
ws judged from their lower neural spines.
PI6166 bears a small hypapophysis. and
P16170a a longer hypapophysis.

Associated material: A fragment (P16170c)
from near the anterior end of a Teft maxilla
with three teeth, curyed backwards, Each tooth
is approximately 7 mm long.

Assuming thal PL6I67 represented tlre
largest vertebra of the specimen of H*onaebi
raracoortensis, and assuming that the largest
vertebrae of HW. narieooriensis and Python
sPilotus, respectively, occupy the same propor
tion of the length of the vertches! cofomn, then
PIA167 would have been derived from 3 snake
of total Jength about 5.0 m,

Comparison with arher species; OF the cight
Australian species of Boidae., vertebrae of
Python spilomes (4 spovimens), P, ametiustinuy
(2), biavis childrent (1), Chondropy ten
wirtdis (1) and Aspidites melanocerhalis (3)
have heen examined,

The vertebrae of these extant hoids are
vharacterized by the presence of small. pointed,
avcessiry processes beneath the prezygapo-
physes, by having large outwardly-directed
zygapophyses exteniling Further laterally than
the pecadiapophyses (Table 2) and the neural
spine hatchet-shaped. and hy the absence of
paracotylar foramina, Subgentral ridges are
strongly developed. The general shape of the
vertebra is similar among all the spovics
(Tahles 2, 3). Chandrepython viridis diflers im
having the neural spine bifurcate anteriorly,

Pyrhou spilotus aid P. untethistirins resemble
euch other in having a pair of foramina at the
base of the neural spinc, while Asplidites

nelarocephalits, Chondropython virtdiy and
Liesiy ofildreni Yack thls pair of foramina.

Wonamhit natacoertensis differs fram all
other Australian boids in Jacking accessory pro-
cesses, iu having the neural spine sloping buck-
wards, in having weak subcentral ridges, and
in the presence of paracotylar foramina, Jt
shares with spiditey melanacephalus. Chen-
dropython viridis and Liasis ofildrent the
absence of foramina at the base of the neural
spine, The totul height (relative to length) of
the vertebrae is greater in HW. naracoartensiz
thin in other Australian boids, the paradiapo-
physes extend further lateratly than the zygapo-
physes, the condyle (relative to Vertebra
length) is wider and the zygosphene (relative
to vertebra length) is narrower (Table 2). On
the other hand. width across prezygapophyses,
Minimum width of centrum, and length of pre-
zyezapophysis (all relative to vertebra length)
Fall within the ranges of the extant species, as
does height/ width of condyle (Table 3).

Paracotylar foramina wre generally absent in
extant boids, and occur only in the genera Con-
stricror and Trepidophls (Boinac) and Pnygrus
(Erycinael (Hoffstetter & Guyrard 1964).
They are found, wsually as two pairs, in the
fossil genera Giyantephix and Madstoia (Hoff-
steller 1961a, hb}, and as a single pair in HW,
NEPUCOOrLIPASIS.

Whereas the lengths of vertebrae of
Wonornhé iaravcortensis (relative to width
across prezVgapophyses) fall within the ranges
of those of the extant Australian boids, inelud-
ing Liasis (Table 3), six vertebrae from the
Weillifigion caves of New South Wales were
langer than vertchrae of Liaws (tyidekker
1888, p. 256).

There is a striking resemblance hetween
Honaithi vertebrae and those of Mudsreia tial
(Palaencene-Eocenc of Palagenia) and M,
madagascariensis (Cretaceous. Madagascar)
(Hofstetter 1961a, Simpson 1933), particu.
Jarly in the back-sloping neural spine, broad
paradiapophyses and absence of accessory
processes, Hoffstetter (1961la, b) included
Madsraia and Gigantephis (from the Eocene
of Egypr, Andrews 1906) in a sub-family Mad-
stoiinac. The diagnostic features were: (a)
accessory processes absent; (b) a pair of para
zygantral foramina present and opening into
deep hollows; and (c) paracotylar foramina
always present, usually as two pairs. The two
genera were distinguished by the form of the
ventral surface of the centrum, (ffeantephir
having an undivided haemal keel, and Afad-

44 MEREDITH J. SMITH

sioja having the haemal keel distinctly divided
into two. All WY. waracoorlensis vertebrae con-
form with character {a) and differ from «ther
hoids, all of which have accessory processes
(Hoffstetter 1961a); and the more posterior
vertebrae, conform with (bh). However all the
vertebrae have distinct parazyyantra) foramina
(even though they emerge through the charue-
teristic deep hollow only in the posterior verte-
frac) and such foramina in modern boids arc
minute, irregular and inconstant (Hoffstetter
61a). W. naraceorteisis vertchrac differ
slightly from (c) in having « single pair of
paracotylar foramina, but their presence at all
ig rare in boids (Hoflstetter & Gayrard 1964),
The form of the yentral surface of W. nara-
coortensis vertebrae ranges from hypapophysis
prescot {P16166) or haemal keel undivided
(P161291, P1éid4k}, to haemal keel notehed
(P16168) and finally to haemal keel distinctly
divided (P16129s and P16167); thus it encom-
passes the form of both Muadstefa and Gigan-
tophis. Medstoia differs fram Gigantaphiy also
in the greater development of neural spines and
Parsdiapophyses (Simpson 1933). In these
features WB’, naraceertensts closely resembles
Mad sraie.

When Madstoia was compared with many
boids, both recent and fossil, the resemblance
af Madstoia to Gigantophis was found to be
closer than to other known genera. However
it was impossible to conclude that the two were
definitely more closely related to each other
than to other fossil boids (Simpson 1933),
Similarly the relationship of (Wanarmbi to Mad-
stoia or any other boid will remiuin obscure
wml the skull is known.

The presence of Madstoie in Patagonia and
Madagascar has been regarded us evidence of
former continuity of the southern continents
(Hershkoyitz 1972, p. 316).

Family ELAPIDAE

Elapid vertebrae have conspicuous accessory
processes and hypapophyses on all precloacals
(Fig, 1).

Psendonaja c.£. P. nuchalis Guother
Material: Vertebrae (566 precloacal, 25
cuudal); dentaries (6),

The vertebrae have heen described (Smith
1975). The largest with a teneth of 11.1) min
between zygapaphyses would hnve heen derived
frnm a snake about 190 cm long,

The dentary of P. wuchalis is almost stratcht
posteriorly. but anteriorly it curves outwards
then inflects sharply. The teeth are strong and

curve backwards very slightly. They are
separated hy w distance equal to c, 2/3 of the
adjacent teeth, The second tooth is the lonuest
but the succeeding teeth along the dentary de-
crease in size anly slightly. The fossil dentaries
are similar.

Notechis c.f, N, seutatus (Peters)
Material; Vertebrac—precloucal (15,
6.3-9.9 mm, mean 7.6 + 0.35).

These vertebrae differ from P. auehalls and
resemble N. sertates in having a relatively short
neural spine overhanging both anteriorly and
posteriorly (Smith 1975).

length

Pseudechis c.f. P, porphyriacus (Shaw |
Material: Vertebrae—preclouval (55, length
4.1-9.5 mm, mean 6.3 + 0.15); maxillae (2);
dentary (1),

Both Notechis and Prendechis have 3-5
Small teeth, whereas Pseuwdonaje has 8-10
(Worrell 1963; pers, ohserv.). A left maxilla
(Pl6é164a) bears a curved fang, followed after
a diustema, by three small, curved teelh and
is consistent with P. porphyrivueus in size an
shape. A ymaller fragment of a right maxilla
(P16164b) is probably from the same skull,
having been taken from the same sample.

P. porphyriaeus dentaries differ fron P.
nuchalis and N. seutatus dentaries in being
more sharply curved anteriorly; the teeth arc
fine, backwards-curving and closely-set. A right
dentary (P16132c) conforms with FL perphy
raeas,

The vertebrae have the long, acute, accessory
processes (Smith 1975) typical af P. por
riacus but they differ in having these processes
directed more anterolaterally than in the P
porphyriacas available for comparison,
Undetermined

An unidentified elapid group contams 40
vericbrae (length 3,8-7,8 mm. mean 60 +
-'5) characterized by the short, blunt hypapn-
physis.

Family VARANIDAE

Varanid vertebrae are distinguished by the

overhanging condyle (Fig. 1),

Varanus varius (Shaw)

Material; Vertebrac—cervical (4, lengths 18.7-
26,3 mm. mean 21.35 + 1.76), dorsal (17.
lengths—Table 4), sacral (3), caudal (26,
lengths 7,9-19,2 mm, mean 12.3 + .53);
maxilla (19; dentaries (5); pametal (1),

The dorsal vertebrae of Varanus elganrtenus
ane readily distinguished by their broad cencrit

SMALL FOSSIL REPTILES FROM NARACOORTE 45

[ratio Of width across prezygapophyses (Pr—Pr)
to Tamimum width of centrum (BW) < 1.6
(Table 4)] and jong neural spines, vertical both
anteriorly and posteriorly. But the vertebrae of
i. varius and ¥ gonldi are similar morpho-
logically—-there is overlap iv the relative width
of centrum, relative width of candyle (CW)
and telative width across prezygapophyses fall
relative to the length, Pr—Po) and also in the
ratio of width across prezygapophyses to rmini-
mum centrum width (Table 4), These values
overlap even when the comparisan is made
hetween vertebrae fram the same position in
the column The neural canal. viewed from the
front, is slightly depressed in ¥, gonldié bul is
round in . varius.

The fossil dorsal vertebrae are consrstent
with hoth V_ vartuy.and V. eouldil in their pro-
proportions and have the neural canal round
anteriorly, as in VL peiriier

Like the dorsals, the cervical and catidol
vertebrae of VL verins and ¥. gould{i are almost
(Or quite) indistinguishable as ta species, but
the first sacral vertebrae are distinctive. The
iransverse processes of the first sacral of V-
souldi (2 specimens examined) bear several
low. ridges— one such ridge from the anterior-
mose point of the lateral surface of the trans-
verse process extends towards the cotyle: a
diagonal ridge passes from the prezygapophysis
to the lateral postero-darsal tip of the trana-
verse process and a ridge from the Iaternl
altere ventral tip of the transverse process 19
the condyle makes the posterior surface of the
transverse process slightly Cancave, In contrast.
the transverse processes are smoothly-rounded
and convex in VF. varius (3 specimens
examined), Similarly in the fossils {P16135r,
Plé169a) the transverse processes are
smoothly rounded, The fossil conforms with
Vo varius and differs from WV. gouldii also in
having the neural canal round anteriorly and
the transverse processes at their lateral extremi-
tres flared to helow the Jevel of the centrum
(whereas In V. goalddit the flaring extends more
ilorsally }-

In Parenus varus the parictal foramen lies
in the middle third of the length of che parietal
plite (Mertens (942), whereas in FV geulaii
and V. vigdntéus it is in the antetior third
{pers., observ},

Laterally compressed, recurved, pleurocdont
leeth with striated bases are characteristic of
varanids {Edfiund 1969). The teeth of PF,
siemens are fine and thin, but in B. varias and
VF pouldii, and in the fossils, the lateral com-

pression ig less extreme, and a labial and a
lingual ridge ascend each tooth, The hasal
fluting extends nboaut 1/35 of the way wp [he
tooth.

The length of the larger fossil lirst sacral
vertebra (18.8 mm Pr-Po) indicates a total
length of c, 1,6 m for the ammal,

Varanus gouldii (Gray)
Material) Humerus (1).

The shaft of the humerus is smoothly
rounded in P. verits (2 specimens) byt in
V. gouldii (3 Specimens) 2 distinct ridge
extends from the proximal termination of the
supinaive crest to a muscle scar (presumably
for the humeraradialis muselé) near the proxi-
mal expansion, Anteroventrally, he deitopec-
toral crest ig promiment in both V, varivs and
V. garldit, but in ¥. 2gouldti the erest extends
further proximally than in ¥. varins, The fossil
hiimerus (P16146b) conforms with V. gouldél
and dilfers (rom V, varius in having a ridge
extending proximally from the stpinator crest
and apparently also in the proximal develon-
inent of the deltopectoral crest, although most
of the proximal articular facet of the fossil
humerus has been tost. The fossil hay a distinct
tuhercle at the proximal termination of the
supinulor crest. Stich a distinet tubercle was
seen only inane modern specimen of Varanns
species, viz. a very large VL gauldii. No
lubercle could be distinguished in two
pouldii comparable in size with the fossil, nor
in iwo F, verizis.

Family SCINCIDAE
In scincid vertebrae the centrum tapers, in
ventral outline, frorn broader aliteriorly to
narrower posteriorly, and there is no precandy-
lar constriction. The ventral surface of the cen
trum ts smoothly rounded (Fig, 1),

Trachydosanrus rugosns (Gray)

Material: Osteoderms (several hundred); verte-
bra—cervical (10, lengths 5.2 7.0 mm, mean
6.0 + 4.21), dorsal (46, lengths 6.1-10,7 mim,
mean 8.9 — 0.17), sacral (5 pairs), pygal (A.
lengths 6.7-10.0 mm, mean 8.0 + 0,53),
caudal (5, leagths 5.0-7.0 mm, mean 6.3 =
0.373; maxillae (5): premanillac (2); dentaries
(6): humeri (3), femur (1); frontals (24.

Oxteoderms: In Trachydosaurus, the astee
derms are thick und coarsely pitted, whereas in
Tiliqua migrolurea, TU. oevipitalix ynd T. xcin-
coaldes the osteodernms are thinner and finely
pitied; in Everaia cemminghami the darsal asteo-
derms bear a posterior median tooth; and in

4G MEREDITH J. SMITH

Frequency
3S

| Fossil

T. nigralutea (3)

f T. scincoides (3)

T.occipitalis <2)

Fig. 4. Frequency distributions of some dimensions of dorsal vertebrae in which Trachydosaurus

rugosus differs from Tiliqua specics. The sample of fossil vertebrae assigned to T. rugosus has
frequency distributions similar to the modern sample. (a) width across prezygapophyses divided
by length between zygapophyses, (b) width across prezygapophyses divided by maximum width
across paradiapophyses, (c) width of condyle divided by width across postzygapophyses, (d)
height of condyle divided by width of condyle. Although the number of presacral vertebrae with-
out a hypapophysis is 30-32, not all vertebrae could be measured in every specimen.

most other skinks the osteoderms are thin and
almost smooth, except over the head of some.

Cervical vertebrae: Cervical vertebrae of scin-
cids have the hypapophysis sutured or fused to
the posterior part of the centrum whereas in
agumid cervical vertebrae the hypapophyses
are sutured or fused to the anterior part of the
centrum (Hoffstetter & Gase 1969). Cervicals
of Trachydesaurus rugosus have broad,
roundish zygapophysial facets whereas in
Tiliqua species and Egernia species the zygapo-

physes are usually anteroposteriorly elongated
and narrow.

Dorsal vertebrae: Cervical vertebrae are
defined as those anterior to the first vertebra
of which the rib joins the sternum (Hoffstetter
& Gasc 1969), but because this distinction can-
not be applied to isolated vertebrac, I have
included in the discussion of dorsal vertebrae
all the presacral vertebrae that do not bear a
hypapophysis. Dorsal vertebrae of T. rugosus
are squarish in dorsal outline, i.e. the width

SMALL FOSSEIL REPTILES FROM NARACOORTE Aa

across the zygapophyses approximately etuals
the length between the zygapophyses, whereas
dorsals of Tili¢a species are longer than wide
(Fig, 4a), In other skinks also, und even in the
largest Australiag skink, the heavily-built
Evernia buagana {only one = specimen
examined), the vertebrae are longer than wide.
In T. regosuy vertebrae, the neural spine is
low, slopes backwards and overhangs pos-
teriorly, At its posterior termination, the neural
spine is thickened and marked by a short
median groove. The pre-zygapophysial facets
are directed dorsally at an angle to the hori-
zontal of abotit 20-40° (although in the an-
terior 3 or 4 dorsals the angle may be as great
as 60" in some specimens (Fig. 51}. The pre-
zyzapophysial facets extend laterally nearly as
fur as, or further than, the paradjiapophysial
convexities (Fig. 45). The zygapophysial facets
are almost round im contrast with the out-
wardly directed zygapophyses of agamids, The
condyle is narrow [Fig. dc) and slightly
depressed (Fig, 4d). Except for the neural
spine and slight zygapophysial ndges, the cen-
trum is smoothly rounded, again in sharp con-
trast with agamids where not only are the
lateral ridges strong, but also a wide micdventral
ridge is conspicuous (Fig. 4d). The ventral sur-
face of the centrum is triangular in T. rugosyy
(as in agamids} whereas in other skinks, ven-
trally the sides of the centrum are almost
parallel behind the paradiapophyses.

Sorral vertebrae; The pleurapophyses. of the
Sacral vertebrae are fused distaliv for about onc
third of their Jength. The facet for articulation
with the ilium is copped and differs from the
triangular facets of Tiligna species (except T-
orvipiteliy) and Eyernia species (Fig. 6). Tn
T. arcipitalis where the lateral articulation
cupped (as in TL rngoses), the zygapophysial
facets are sharply upturned, at an angle of about
JAQ-S(I", as in other Tiligua species, wherens in
T, rugosus the facets ure only slightly uptuned
(angle c. 20°). The condyle is narrower in T.
rugester than in Tiligna species,

Ceudal vertehrae: T, rugesns eaudals are robus:
und the transverse processes project only
slightly venically, much less venirally than in
T. selecolder, 1. aceipitalis and T. nigrolatec.
Teeth and teoth-bearing bores; The tecth of
Trackydoxeuris resemble these of Tilique
species in having conical tips, whereas tecth of
Eserniat species are laterally compressed at ihe
tips (Mitchell 1950), Usually the teech of 7.
ravosuy are broad atd bunt, but in some
specimens the teeth are longer, Itinner and

sharper, These latter overlap m form With those
of the larger Tiligua species. In T. gerrarai
one tooth in each jaw is very large, about four
Himes the thickness of the others, which tre
fine, with rounded tips. The maxillary bone of
T. rugosux is robust. Beneath the orbit, a strang
bone ridge ruins parallel ta the jaw margin and
extends posteriorly beyond the level of the end
of the tooth row. In Tilique species, this ridge
is) Weaker and shorter, not extending beyond
the end of the tooth row. and often ending still
more anteriorly. The demiaties. toa, are rohust.
and are thicker and deeper, especially an-
teriorly, hear the symphysis, than are those of
Tilique species.

Frontals; Two frontal bones were cach charac-
terized us of T, tuegosus by the (ick, coarsely-
pitled osteoderm fused with the bane,

Limh hones: Homerus aud femor have rela-
tively thick shafts in J) egos,

Tiliqua nigrelotes (Quoy & Gannard)

Material: Veriehrae—dorsal (6, lengths 6.1-
8.9 mm, mean 7.8 + 0.44), pygal (2, length:
5.0, 6.2 mm), caudal (3, lenaths 5.6, 64, 68
mm}; maxillac (4); dentares (3): parietal
bone (13,

The size of the fossil dursal vertebrae indi-
cafes that they were derived from a lizard at
least 22 cm in snout-vent length, Such size is
reached by the larger species of Eyernia and
Tiligva, but not by TT. caruerinae, T-
branchiale, T. peterst nor T. woorl-fonesh.

In Tillqua species, the prezygapophyses are
dozsally upturned at an angle of 35—55° (Fig.
5n) of even greater in the first two or three
dorsals. The dorsal vertebras of Australian
skinks, other than Trachyderavrns, are longer
than wide, except somelimes the Last presacral
which may be slightly wider than long {Fig
4a). In T. seinceides. most of the dorsal verte-
brac (except for the first two or three and last
two or three) are extremely elongated (Fix.
4a) and this species is further characterized hv
the broad, depressed condyle (Fig. 4c, 4) and
narrow zygapophyses directed almost anterod-
posteriorly. In TJ. seineoides, the paradia-
pophyses extend laterally well beyond the
lateral edges of the zvgapophyses (Fil, 4b), In
T. nigralytea und T. oceipitalix the 2zy¥yapo-
phystal facets are slightly wider (though the
width never equals oor exceeds the length) and
srt directed antero- or postero-lalerally, henve
the width across the zygspophyses is greater
(relative to, eg. the length between zvgapo-

48

MEREDITH J. SMITH

TABLE 2
Vertebral proportions of Wonambi naracoortensis competed with thuse of Jur extant -boids, Ten. verithrak were measurad from the posi-hypapaphystat region of madere
Specimens, The range is followed 2y mean + standard error in pureniheris.

Width across:
paradiapophysés/ Width
Specimen Length (mm) Total height /length across pre-2¥gapophyses Widti condyletLength
WY. naracoortensts
P16168 16,5 1.95 it O45
PL6l4si 194 wy 1.09 0.65
P6144» “26 4,70 124 056
F161294 1h 2,02 1,07 0.83
Pi6167 ma 4,76 7a 054
P1617b we 1,85 142 0,63
Liashs children 43=5,2 (4.9 + 09) 101-433 (1.47 03) 0.90-0:91 10.95 = OL) 364.52 (O47 + 02)
Python spileliis
B T5- 85 (6th AN) HOmeL TY 17) 0.777 0.79 + A) DATHLSS (0.50) + 01)
2 $1- 8.5 (84 08) 124-047 (1.46 le 02) O,7841,40. (0,79 002) O.SIALS8 (AG + oO)
3 11.4-42,0 (11,6 — 07) 1,53-1.-71 (1.61 02) 0.70-0,73 (0.73 O12) 0.474151 (0.49 + 003)
Prrhow amethiccinas
1 145-155 (14.9 =p fa) 132-145 (LaR + OLY D.75-0,78 (U.7T J: 003) 0.47-40.51 (0.49 + (4)
o* 10.1-10,5 (10,3 + 04) 142-249 (1.45 + On) O794),82 10.90 + 009) 0:59-0.62 (0.60 + 003)

Aspidites melancvcephalus
a

ij

Chondraphysican viridis
oe ————_— eae

* Spécimen incomplete: only aboul one quurter of the precloacs! yertetrac were studied,

37-65 (63 077
35-64 (64— 10)

554 69 (HA td)

139-154 (1.46 4-02)
129-15 (1.43 04)

0.95-1,a5 (1,22 + ,04)

OQ. 7S40.84 (192 + AMG)
0.30-D.86 (089 + 01)

0.73-0,83 (0,78 = 01)

TABLE 3

Vertebral proportions of Wonambj maracanrtensig compared with those of four exsant buids, Ver vertebrae were measured from the post-kypapophysial repian of maders
specimens, The range iz jotiawed by mean + siandard error im parenihesis,

{LSSH1b2 (UY LY
D.S4).62 (0.58 + 01)

O4AD (0.46 — 01D

Zygosphene width)
Length

0.55
0.65
0.50
O07
0.53
Onn
O.S9474 TLAB + ANZ)

0,57-0,72 (0.66 + 02)
{1.65-D.76 (0,71 = OL)
0.7841,90. (0,45 + 01)

O H8-U 72 (0,69 + 14)
O.AS-41.85 (0.68 SE M2)

0,040,77 (0,75 + 01)
USK AGT (G2 HD)

0.5260 10,59 + 01)

W. across pre- Helght condyte/ L. pre-
Specimen zygapiphyses/Length Min w. centram/ Length Width condyle zyRapophysis/ Length
W, naraccortensis
PIBL6S 1s Ln 0.82 04s
Pisid44k 457 17 0.81 0.86
Plgid4s 13} 140 0.84 0.54
PIGtIn 158, 116 OTe OAS
P6167 152 116 0.98 (55
T16b 1.50 112 O88 56

Liodis children
Python spletes
Bg
2
4s
Python umethtstinus
i"
(er
Aspidives melonacephates
3

3
Chondropythan tiridiy

$9141.77 (182 +)

148-1,71 (1,60 + 02)
153-177 (1,66 oe 03)
4.74-1.88 (1.81 + 102)

1,671.76 (1-722 01)
(59-12-68 (1,62 + 01)

1,62-L.74 (167+ 01)
153-163 (158 + 91)
LA3-L48 (134 ye 097

OF8-305 CW Md)

OS1-1.10 (1.01 + 02)
LOI-ad (1.0 + 1)
VN41.25 11,204 th)

V.16-1.21 (V.18 e AS)
HAGEL LA (1 12 + 005)

TMB-11Z (109 01)
0:99-1.0% (Lod + 111)

D.75-0,99 (0.91 + 108)

O,77°LQ1 (OME + 2)

C7642 (Rd ON)
O75-40.96 (AS U2)
O.9S-1,08 (0,99 2 1)

O.3H-D.90. (0.89. 004)
0,760.83 (0.40 + 1)

M6947 (RA + 02)
1715-0,.82 (7B + 01)

O75-0,83 (0.99 + .01)

0.34-0.46 (40 > OL)

O.404).49 (0.45 + HN)
HAS (OSH A)
(540,59 (0.56 4. 01}
0.480.539 (O51 + 105)
O.41-4.45 (WAR + 005)

O.4241.47) (0.45 a= GON)
OATH. (0.42 11)

0.27-0.43 (0.37 01)

a gS
*Speciinen incomplete: yoty sbaot ene quarter of the precloacal vertebrae were studied.

TABLE 4

Length aid bropertions of dorsal vertebrae of specimens of three speeiva uf Varaous. The ulltinate presacral veeiebea uf euch specimen is exchuled The range Js follawed by

mean + standard ceror in purenihesis,

Pro (znm)

BwW/ Proto

CWlPCPa

Pelr/ ero

Yr BW

25-271 (25,7 + 14)

0.54-0:64 (0,58 + .005)

12.4-13,6 (13.14.07)
17.5-(9,.6 (18.5 + A)
147-15.8 (15.4 + .017)

0,48-0,60 (0.52 + .005)
047-059 (0.55 + 007)
0.43-0.54 (049 + sh)

0.35-0,63 (0.59 + 406)
052-056 (0.54 + .002)

0.48-0,.54 (1.52 = M4)
0.440.593 (0.49 =p N06)

0.88-1,01 (0,92 — O07)

.86-1,00° (0.90 4.007)
0.85-1.02 (0,97 + 00%)
O.82-0.95 (0.91 te DUT)

15i-f64 (1,58 + 008)

1.6641,78 (1.73 + .006)
1.70-1,68 (4,77 + 011)
17041,97 (4,65 + 013)

16,7418,.5 (179 +09)
13.0.15.4 (Lith te 1)
21,2-24.7 (2.9 17)

484059 (651 + 005)
OASOAT (SL + 007)
OAT. (OLS) OH)

0.56-0.59 (OSH + 002)
USIALS7 (1,55 E004)
At (0.49 + OO)

0.87092 (0,90 + .004)
0840.96 (091 + OH)
FER2-1AH) (U9 O11)

4.61-1.85 (1,76 + 012)
1,70-1.86 (1,78 + 011)
157-164 (1,73 + O15)

Numbers
Specimen of vertebrae
¥, gigantens 20
gould i:
Specimen t 20
Sneclmon 2 20
Specie S 2
VY, waerlies
Specinrert | (9
Speelmen 2 16
Specimen & y
¥ varius
{fornil) u

131-214 {163 + 0.67)

047-0455 (051 + 007)

U.48-0.57 (052 O07)

0.844), 98 (091 + O10)

1.65-1.37 (1.78 + 013)

—_——

SMALL FOSSIT. REPTILES FROM NARACOORLE 4u

physes or to the condyle} than in T. scincoides.
In Tiligua species and Evernia species the tip
of the neural spine may be thickened and some-
times marked with a shallow median grnove,
bucin 7. nigrolvrea (four specimens crammed)
the median grovve ts so deep that the spine
terninates in a double tip,

Three incomplete detitarics (P16124z,
P16126w, and PIG)28h) resemble 7. #fera-
lutea dentaries In shape, and four tmaxillac
(P16125s, P16128n, P16128w and P161572)
ore consistent with T. nierolutea (and atso with
T. scinvoides and 7. ovcipitaliv) in the slight
suborbital ridge. A parietal bone (P16127d) Is
probably also of this species although the sides
are slightly less constricted than in modern T.
aivrolutea.

cf, Spheaomorphus tympanum (Lonobery &
Andersson)
Material; Two fased sacral vertebrae,

In Sphenomorphus tympanum, the transverse
processes of the first sacral vertebrae (SI) are
strong and slant backwards only a few degrees.
The transverse processes of S2 are thinner and
are directed forwards to join and fuse with
those of Si at their lateral expunsions. The
sacral fossae between the transverse processes
are wide, §. ¢ympenut sacral yerlebrac differ
from those of Egernia stefofara in having tela-
tively wide fossae, and differ from EF. whirei
tenehrosa Where the transverse processes of S|
are angled backwards and the transverse pro-
cesses of S2 are perpendicular to the long axis
af the vertebra, ‘Tha fossil (P16146r) has a
total length (from prezygapophysis of Si to
posizygapophysis of $2) of 3,7 mm.

Egernia, c.f, E. whitei (I acepede)

Material: V ertebrace—dorsal (2, lengths 3.4, 2.7
mm), caudal (1, length 2.4 mm); manillac (5
Icft, 5 right); dentarics (8 lett, 6 right); fron-
tals (1).

The Meckelian groove in the lower jaw is
closed anlerior to the splenial in Egernia but it
is open forward to the symphysis in Spheno-
tntorpius. Vhe detitary of EB. whitei is deeper
than the slender dentary of S$. gvinpanien, and
the notch in the posterior lateral surface of
the dentary is higher (i.e. nearer the tooth row)
than in S. lyenpanunt, The fossils wre consis-
tent in shape and size with E. whirei,

The fused frontal bones of E, whitei differ
from those of §. ryepanuen in their gradual
taper, both anteriorly and postenurty.

Family AGAMIDAE
Agamid vertebrae ale characterized by therr
trangular Ventral outline and slrong subventral
ridges.

Amphibolsrus c.f, A. barbatus
Material: Maxillac (1); dentaries (7).
Agamids ate (he only Australiqn reptiles with
acrodont tooth implantation. The largest fossil,
a right dentary (P16132b) with length af tooth
row 14.5 mm, closely resembles 4. barhatiy.
The other specimens, two of them fragments,
may he of a smaller specics,

Faunal change

The two soptiles most common in the Victoria
Cave deposit, viz. Pseudonaja c.f, Po auehalis
and 7. ragosas were represented at all depths
in similar abundance. The less common species,
except for #4’. neracoofiensis were also found
al various depths from the surface to the
present maximum depth of excavation, Sevei
vertebrae und the tooth fragment of MW, nare-
coortensis Were near the surface and all within
2 metres of each other. Hence the reptile fauna
does not change remarkably with depth in the
deposit.

Discussion

The small marsupial remains, together with
ubundant rodent vemains, were probubly
brought into Victoria Cave by owls (Smith
1971, 1972), and the small lizards may also
have been the prey of owls, Among the larger
spovies, Trachydasmirus rugosus is a clumsy,
shortlegyed, heavy-bodied livard which might
easily fall into sinkholes or caves and would
have liltle chance of escaping, This species has
been recorded from several cave deposits (c.g.
Cook 1963, Finlayson 1933, Longman 1945),
The snakes may have actually inhabited the
cave, as live brown snakes (Psendonaja sp.)
ave found in the limestone caves in south
eastern South Australia (Wells, pers, comm,)
and P. auchalis has been classified as an occa-
sional troglonene (Richards 1971),

Tn any measurements af the bones of rep
tiles, intraspecific Variances are large because
reptile growth is asymptotic. When vertebrae
ate the bones measured, changes along the
column further increase the variation. Jn the
identification of isolated vertebrae of some
groups (eg. the snake family Crotalidac),
these inhetent large variances can he offset by
considering several dimensions simultaneously
and in comparison with their previously deter
mined inter-relutionships along the entire
columo of reference skeletons (Brattytrom

40 MEREDITH J. SMITH

1964). Nevertheless, when the reptile remains
ure abundant or include « qualitatively diag-
nostic bone (ee the first sacral vertchra of
Vurunus), the specics can be diagnosed with
cunfidence,

Of the 3 species contidently determined, 4
are still Jound in south-easiern Australis, antl
all but Faraaus varius have been found near
Naracoorte. All 6 additional species tentatively
identified have been found near Naracoorte.
The large boid is the only Pleistocene species
absent now, Hence, among the reptiles, the
small species have survived from Pleistocene to
presem without detectable change of the
vharacters available in fossil muterial, whereas
the large species has become extinct, Similarly
with the marsupials; while many large species
have become extinct feg. several Sthemrus
species, Thylacelea «ff. 7. carnifex (Wells,
pers. comm,), Palorchestes sp., (Pledge, pers.
comm.,)], the small species, ¢.g. Berrengia spp.,
Perameles spp, Antechinus spp. and Perauras
breviceps are indistinguishable from modern
species, many of which suill survive near Nara-
eoorte (Smith 1971, 1972), Among the birds.
the only specics now exunct. Progra nara-
coortensix, Was a large bird, while all of the
small spectes ire extant. The factors that
caused the extinction Of so many large verte-
brate specics have had little perceivable effect
on the small vertebrates.

The presehee of Paranus varies together
with V, goulii? in Chis Pleistocene deposit does
Hol support the suggestion (King & King 1975)
(hut the iedicus karyotype (represented by 1.
varius) tavaded south-eastern Australia after
lhe separation of Kangaroo Island from the
mainland, 3,000-10,000 years ago.

Most of the extant species of the Vicloria
Cave seplile fauiny are wide-ranging with broad
habitat tolerances, Varanns gouldii occurs in
most parts of mamnland Australia but is most
commen in sandy areas, where it lives in sand
burrows. (Worrell 1963). The tree-climbing
species WM. varius, otcurs throughout exstern
Australia inside the 20” (508 mm) isohyct

(Rawlinson 1969), Trachydosaurus rugusus. is
found in inland areas of all muinland stutes,
while Pserdechix porphyriacuy lives in coastal
to mountainous forests and swanips of eastern
Australia, but does: not extend into dry inlani
areas (Worrell 1963). The ranges of Pseuda-
taja nuchalis and ihe marphologically similar
P. rextilis together include most of mainland
Australia (Worrell 1963), and P. fextilis o¢ours
also in New Guinea (McDowell 1967). None
of these species extend inte the cool temperate
zone of the Bassian zoogcographical subregion
(Rawlinson 1974). Conversely, Tiliqua nigro~
lures is contined to the cool temperate zone, its
Tange extending from the extreme south-casi
of South Australia and southern, Victoria to the
islands of Bass Strait and Tasmania. Nura-
coorle is close to the north-western limil of its
Tange (Rawlinson 1974). Evernia = witiret.
Sphenomorphus tympanum and the genus
Notechis occur in all zones of the Bassisn but
not in other subregions (Rawlmson 1974)
Hence little palaeo-ecological infonmution can
be gleaned from them, The presence of a lirve
Proportion of the Pleistocene reptile fauna i
the area at present does suggest that climatic
changes during the last 30,000 years have been
slight in south-easicrn South Australia.

Acknowledgments

The field work im Victoria Cuve has heen
planned and supervised by Dr R, T, Wells, and
to him and his voluntary helpers (many of
them CEGSA members) I am grateful. Trans-
port costs for field workers were defrayed by
a gram from the South Australian Government
Tourist Bureau. Specimens of snakes and
lizards have been given or loaned to me by
many pedple, among Whom I thank especially
Mr F. W. Aslin, Mr J. deBavay, Mr W. J.
Parmenter, Mr R, Shine. Mr M. J, Tyler, Dr
R. Henzell. Dr BD. Harton, Dr T, F, Houston,
and Dr G. F. van Tets. Mr P. G. Kernpster
prepared the photographs (Fig. 2), Dr R, T
Wells and Mr M. J. Tyler kindly criticized the
rhanuscript.

References

ANbrews, C. W. (1906).—"A descriptive
catalogue of the Tertiary Vertebrata of the
Fayum. Exypt.” (British Musgenm: London.)

Ancein, M. (1974).—Excavations in the Orches-
tra Shell Cave, Wanneroo, Western Australie.
Fl, Fossil vertebrate tenmains, Archeol, Phys.
Anihrapal, Oceana 92), 156-162.

ASTEFENBERG, W. (1963 —The lossil snakes of
Florida. Tulane Sid. Zool, 1, 131-216

Bedrtstrom, B. H, (1964),—Eyalution of the pit
vipers, Frans, Sun Diezo Soc. Nat. Hist.
13( 11). 185-268-

Coon, D, T- (1963)—The fossil vertebrate fauus
of Strone’s Cave, Koranup. Western Australin.
W, Aust, Nat. %, (53-162,

Dortcr, C_ E,. & Mrrarirens, D, (1971).—A sal-
vate excavation in Devil's Lair, Western Ats-
tralia, J, R. Soc, W, Ast, 54, 103-114.

SMALL FOSSIL REPTILES FROM NARACOORTE 51

Epwusp, A. G. (1969)—Dentition. In C. Gans
(Ed.), “Biology of the Reptilia”, Val- [, pp.
117-200. (Academic Press: London, }

iain, A. P. (1964).—The Anstralian Aborigines.
4th edn. (Angus and Robertson: Sydney.)

Frinavary, G. J. (t918)—Contributions io a
monograph on fossil Yaranidac and on
Mesgwlanidae. Ann, Aist, Nat. Mus. Nail
Hung, 16, 341-467.

Pesrrvary, G, J. (1935) —Forther contributions

to a monograph of the Megalanidue and
fossil Vurunidae. with notes on recent

Varaniany, Ann, Aisi. Nat, Mus. Nat! Hung.
29, 1-130.

Fincayson, H. H. (1933). dn Tindale, N, B.
(1933).—Tantinoola Cuves, south east of
South Australia: geological and physiographi-
cal notes. Trans, R, Soc. 8S. Aust, 57, 130-142.

Gre, B,D, & Banns, M, R. (1956),—Cainozoic
history of Mowbray Swamp and other areas
af north western Tasmania, Rec, Queen Vic-
fori Mus. Launceston, N.S. No. 6, 1-42.

Hecut, M. K, (19753)—The morphology and
relationships of ihe largest known terrestrial
lizard. Megalania prisca Owen, fom the
Pleistocene of Australia. Proc. R, Soe. Viet,
87(2). 259-250,

Horrstetrer, R. (19610),—Nouvegux restes d'un
serpent boide (Madstoia madagasceriensis
nov. sp.) dans le Cretace supericur de Mada-
gascar. Bull, Mus. natl Hist. nat., Pariy (2),
33, 152-160,

Horrstetter, R. (1961b).—Novvelles recoltes de
serpents. fossiles. dans I‘Eocene Superieur de
Deseri Libygue. Bull. Mas. natl Hist, net.,
Paris (2), 33, 326-331.

Tlorrsvettre, R., & Gase, -P (1969). —Vertebruc
and ribs of modern teptiles, 4a C, Gans (Ed).
Biology of the Reptilia Vol. I, pp. 2016310.
(Academie Press: Loyidor.)

Horrstetiin, R.. & Gavara, ¥. | 1964) —Obser-
vations sur l’osteologie et ta classification des
Achrochordidae (Serpentes). Ball. Mas. nail
Hist. nat,, Paris (2) 36, 677-696.

HeEnsitkoviTz, P. (1972).—The recent mannmnals
of the Neotropical Region: A zoogcographic
und ecologival review. Jn A. Keust, F.C. Erk,
& B. Glass (Eds), “Evolution, Mammals and
Southern Continents.” (State University of
New York Press: Albany.)

King, M,, & Kina, D, (1979) —Chromosomal
evolution in the lizard venus Maranns (Rep-
lilia), Aust. J. Biol. Ses 28(71), 89-108,

Loncman, H. (1925)—Ophidian vertebrae
tram cave deposils al Marmor Quarry, Meni,
Old Afus, 8, 114-112,

Lonoman, H, A, (1945)—Foasil vertebrates from
Gore Quarries. Mem. Qld Mus. 12, 164

Lunpriaus. E. L. (1963).—Vertebrate remuint
from the Nullarbor Caves, Western Austrulin.
J. R. Sac. W. Aust. 46, 75-80.

Lyprekéa, R. (1888),—“Catalogue of Fossil Rep-
tiles and Amphibians" Part L (British
Museum: London.)

McDowe1s, SB. L1967}—Aspidomorphus, a
genus of New Guinen snakes of the family
Elapicdae, with notes on related genera. J.
Zoel., Lond. 151, 497-543.

MesRILEES, D. (1968).—Man the- destroyer: Tate
Quaternary changes in the Australian mar
sugial fauna. J. R, Soc. AW’, Aust, 51, 1-24.

Mertens, R. (1942),—Die Familie der Warane
(Varanidae). Zweiter Teil: Der Schadel, Abh,
Senckenb. Nutarf, Ges, 465, 117-234,

MrreHett, F. J, (1950)—The scincid genera
Egernia aod Tiligua (Lacertilia}. Ree. $. Ass.
Mas. 9, 275-308.

Rawranson, P. A. {J969).—The reptiles of Exsr
Gippsland, Proce. R. Svc. Piet. 82, 113-128,
RAWLINSON, P. A. (1974).—Biogcography and
eculogy of the reptiles of Tasmania and the
Bass Strait area. Jn Williums, W. D, (Ed.),
“Biogeography and Ecology in Tasmania’, pp.

291-338. (Junk: The Hague.)

RacHagus, A. M, (1971),—An ecological study of
the cavernicolous fauna of the Nullarbor
Plain, South Australia, 7, Zeal. Lod, 164,
1-60.

Romer, A, S, (1956),—“Osteclogy of (he rep-
tiles” (University of Chicago Press:
Chicage.)

Simpson, G. G. (1933) —A new ossil snuke from
the Nofastvlops beds of Patagonia. Bull. Arm.
Mus. Nat Hiss. 67, 1-22.

Surea, M. J. (197)).—Small fossil vertebrates
from Victoria Cave, Naracoorte, South Aus:
tralia. [. Potoroinae (Macropodidae), Petau-
ridac and Burramyidae (Marsupialia)., Trav.
R. Sov. 5. Aust. 95(4), 185-198.

Suen, M. J, (1972),—Smull fossil vertebrates
from Victoria Cave, Naracoorte, South Aus-
tralia. IT. Peramelidac, Thylacinidse ancl
Dasyuridae (Marsupiulia). Trans. R, Sar. 8.
Aust, 96(3), 125«137,.

Suity, M. J. (1975).— The vertebrae of Four Aus-
tralian elapid snakes (Squamats; Elapidae).
Trans. R. Soc. 8. Aust, 99(2), 71-84,

Trorwe, A. (1971)--The Fauna. In R, Vo S
Wright (Ed.4, “Archseolory of the Gallus
site, Koonakky Cuve™ Australian Inst.
Aboriginal Stinlics, Na. 26, 45-47,

VAN TrTs. G. FP, & Sark, M. I} (1974).—Small
Fossil vertebrates Froni Victoria Cave, Nata-
cuurte. South Australia Til, Birds (Aves).
Tras, R, Soe S$. Aust. 9B(4), 228-227.

Worrett, £. (1963}—"Reptiles af Ausititra’,
(Anges and Robertson: Sydney.)

VOL. 100, PART 2 31 MAY, 1976

TRANSACTIONS OF THE

ROYAL SOCIETY
OF SOUTH AUSTRALIA

INCORPORATED

CONTENTS

Archer, M. Miocene Marsupicarnivores (Marsupialia) from Central South
Australia, Ankotarinja tirarensis gen. et sp. nov., Keeuna wood-
burnei gen. et sp. nov., and their Significance in Terms of Early
Marsupial Radiations - - - - > - - - 53

Reedman, D. J., and Womersley, H. B. S. Southern Australian species of
Champia and Chylocladia (Rhodymeniales: Rhodophyta) - 75

Watts, C.H.S. Leggadina lakedownensis, a New Species ef iad Rodent
from North Queensland - - - - - 105

Baverstock, P. R., Hogarth, J. T., Cole, S., and Covacevich, J. Biochemical and
Karyotypic Evidence for the Specific Status of the Rodent
Leggadina lakedownensis Watts - - - - - - 109

PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS
STATE LIBRARY BUILDING
NORTH TERRACE, ADELAIDE, S.A. 5900

MIOCENE MARSUPICARNIVORES (MARSUPIALIA) FROM CENTRAL
SOUTH AUSTRALIA, ANKOTARINJA TIRARENSIS GEN, ET SP. NOV.,
KEEUNA WOODBURNEI GEN. ET SP. NOV., AND THEIR SIGNIFICANCE
IN TERMS OF EARLY MARSUPIAL RADIATIONS

BY M. ARCHER*

Summary

ARCHER, M., (1976) .-Miocene marsupicarnivores (Marsupialia) from central South Australia,
Akotarinja tirarensis gen. et sp. nov., Keeuna woodburnei gen et sp. nov., and their significance in
terms of early marsupial radiations. Trans. R. Soc. S. Aust., 100(2), 53-73, 31 May, 1976.

Two of Australia's oldest known marsupicarnivores, from the Etadunna Formation of the

Lake Eyre Basin, are described. Ankotarinja tirarensis is a tiny marsupicarnivore which may be
related to didelphids as well as dasyurids. Although it is much too late in time to be the actual
ancestral dasyurid, it is regarded as a structural ancestor. It is also structurally ancestral to Keeuna
woodburnei.

Keeuna woodburnei is a small marsupicarnivore which is more similar to dasyurids than is A.
tirarensis. It resembles species of Phascolosorex Matschie, 1916, Neophascogale Stein, 1933,
Murexia Tate & Archbold, 1937, and some Antechinus Macleay, 1841. More distant relationship to
didelphids is suggested.

Resemblance of both of these fossil marsupicarnivores to modern New Guinean highland rainforest
dasyurids rather than to more arid-adapted Australian dasyurids, is regarded as evidence suggesting
that central Australia was less arid during Etadunna time than it is now.

MIOCENE MARSUPICARNIVORES (MARSUPIALIA) FROM CENTRAL SOUTH

AUSTRALIA, ANKOTARINIA TIRARENSIS GEN, ET SP. NOV., KEEUNA
WOODBURNEI GEN. ET SP. NOV., AND THEIR SIGNIFICANCE IN
TERMS OF EARLY MARSUPIAL. RADIATIONS

by M. ARCHER”

Summary

ARCHER, M., (1976),.—Miocene marsupicarnivores (Marsupialia) from: central South Australia,
Akotarinja trarensis gon, ct sp, nov., Keeuna woodburnel gen et sp. nov,, and thelr
sigttificance tf terms of early marsupial radiations. Trans. R. Sec. S. Anst.,100¢€2), §3-75,
31 May, 1976,

Two of Australia's oldest known marsupicarnivores, from the Etadunna Formation of the
Lake Fyre Basin, are described, Ankotarinja tirarensis is a tiny marsupicatnivore which mity
be related to didelphids as well as dasyurids. Although it is much too Jate in time to be the
actual ancestral dasyurid, it is regarded as a structural ancestor, It is also structurally ancestral
to Kéeeuna woodburnei.

Keeuna woodburnei is a small marsupicarnivore which is more similar to dasyurids than
is A. tirarensiv. Tt resembles species of Phascolosorex Matschie, 1916, Neophascogale Stein.
1933, Murexia ‘Tate & Archbold, 1937, and some Antechinus Macleay, 1841, Mote distant

relationship: io didelphids is suggested.

Resemblance of both of these fossil marsupicarnivores to modern New Guinean highland
raiiforest dasyurids rather than to more arid-adapicd Australian dasyurids, is regarded as
evidence suggesting that central Anstralia was less arid during Etadunna time than it is now.

Introduction

In 1971 W, A. Clemens, M, 0, Wood-
burne, C. Campbell and the author recovered
fossil] mammal remains from a site known as
Tedford Locality, on the west side of Lake
Palankarinna, Etadunna Station, South Austra-
lia. These fossils come from the Etadunna For-
mation which is now regarded (W. K. Harris,
pers. comm.) ys being approximately middle
Miocene in age. In 1972 M. O. Woodburne, P.
Lawson, W, Head, B. Archer and the author
extensively quarried and screen-washed Ted-
ford Locality. From the concentrate, two mar-
supicarniyores, as well as other mammal re-
Mains, Were Tecovered.

Stirton, Tedford, & Miller (1961) briefiy
describe a third marsupicamivore Irom the
Etadunna Formation.

Terminology of individual tceth is that used
by Thomas (1887) and Archer (1974). Ter-
minology of tooth crowns is shown in Figure |
and follows that used by Archer (1975a),
Comparisons With other marsupicarnivores are
based in large part on Archer (1976b}.

Specimens with prefix P are in the fossil?
collection of the South Australian Museum; F
in the fossil collection of the Queensland
Museum; UCR in the University of California
at Riverside; AMNH in the Archbold Callec-
tions of the American Museum of Natural His
tory; J and JM in the modern collections of the
Queensland Museum; and WAM in the modern
collections of the Western Australian Museum,

Species names of modern Australian niar-
supials are those employed by Ride (1970),
Laurie & Hill (1954) or Archer (1975b).
Other modern marsupial pames arc those
employed by Collins (1974), Names of Creta-
ceous didelphids are mainly those used by
Clemens (1966). Fassil marsupial names are
those. employed by the most revent reviewer of
those particular groups.

Taxonomy
Genus ANKOTARINIA nov.
Type species: Ankotarinja tirarensis sp. nov,
(by designation and monotypy).
Generic diagnosis; Differs from other Austra-
lian and New Guinean dasyurids in haying, as

* Queensland Museum, Gregory Terrace, Brisbune, Old 4006,

54 M. ARCHER

~ pe

Fig. }. Terminology of molar cusps and crests
(based on Ankotarinja tirarensis). 1A,
upper molar. 18, lower molar, a,c), an-
terior cingulum; c.6,, gristid. obliqua; end,
entoconid; Aycd, hypocristid; hyd, hypo-
conid; Ayld, kypoconulid; hyld n., hypo-
conulid notch; mel., metaconule; melr.,
metaconular ridge; me., metacane: nec,
meticrista; mecd, metacristid; med, mets-
conid; pa., paracone; pacd, paracristid;
pad, paraconid; panie. cr., para-melacone
crest; pel., protoconule; pprer,, postproto-
crista; prer,, protocrista; prd, pratotonid,;
pred, precingulid (or anterior cingulum);
psed, postcingulid {or posterior cingu-
lum); pasted, parastylid; r., ridge mesial to
st©; stA-E, stylar cusps A-E.

a combination of characters, relatively wnre-
duced tafanid on M4 with well-formed hypo-
conid, hypoconulid and entoconid and rela-
tively unrecduced P4.

Origin of generic name: An allusion to Anko-
tarinje, a dreamtime ancestor (Robinson 1966,

Pp. 26) who, having remained buricd a long
tinmte as bones in the earth, resurrected himself
in a small watercourse. Asikorarinja is here
given masculine gender.
Ankotarinja tirarensis sp. nov.

FIGS 1-5
Holotype, P18190, right dentary fragment with
MS.
Type locality: Tedford Locality, Etadunna For-
mation, Lake Palankarinnta, Etadutma Station,
S.A. (28°47'S, 138°25'B).
Diagnosis; That of genus. Features likely to be
of specific value include very small size, rela-
lionship of hypocristid to entoconid, size and
width of anterior and posterior cingula, relative
size of paraconid on M4, and relative size of
stylar cusps,
Origin of specific name; Specific name refers to
the Tirari Desert, the portion of the Simpson
Desert-containing Lake Palankarinna.
Referred specimens: UCR, 15340, dentary
fragment with LM4; UCR, 15341, dentary
fragment with LMs-4; UCR, 15342, dentary
fragment with RM4; P7331, dentary fragment
with LM+-4; UCR, 15343, maxillary fragment
with alveoli for LM3-4; UCR, 15308, LM**;
F7332, LM?3.

Description

Maxillary fragment (UCR, 15343) referred
to this species on basis of size, has alveoli for
M8-4 and posterior root of M2. M4 appears ta
have been as wide as M34, presumably with
relatively little reduction of protocone. M4
length less than that of M#, metacone roat
heing veduced and displaced antero-lingually
relative to metacone root of M%, Zygomatic
root of maxilla yrises buceal to region between
M3 and M4. Numerous smal) interdental
fenestrae in palate hetween M3 and M3, and
between M3 and M4,

Upper teeth represented by two isolated
molars, probably LM% and LM+4. Although
possible that these teeth actually represent Mi},
und M3, improbable for following reasons:
Stylar cusps (UCR, 15308) much more
reduced than those cusps on M®& in all other
dasyurids but not so strikingly reduced when
compared with M4 in some dasyurids such as
Keeuna, described helow, or some species of
Planigale Troughton, 1928, Also, notch tm
antero-huccal cingulum of F7332 for reception
of metaslylar corner vf preceeding tooth sug-
gests F7332 is posterior molar. However, it is
also true that in some dasyurids and didelphids
with large P+, such a notch sometimes exists

MIOCENE. MARSUPICARNIVORES a8

Fig. 2. Specimens of Ankerarinja tirarensis and their measurements (mm). 4, F7331, LMé-4. & UCR.
15340, LM4. C, UCR, 15341, LMs-, BD, Holotype, P18190, RM{4. E, F7332, LMS. F, UCR,
15308, IMé G, UCR, 15343, maxillary fragment with alveoli for | M24.

in antero-buccal cingulum of M1. Probable
that F7332 and UCR. 15308 represent M2 and
M4 respectively rather than M4 and Ms.
F7332 has at least three stylar cusps. StB
connects to parastylar corner of tooth which
may be distinguishable as discreet stA, Para-
stylar blade very short. Ectoloph crest descends
gently from posterior flank stB, then rises
gently, to form long low ridge-like stC, Pos-
tenor paint of this cusp marked by beginning
of rise in ectoloph which forms tall stD. 5iD
has minor ectoloph crest on posterior flank
which descends towards metastylar corner of
toath. Very minute rise in crest on posterior
flank of stD may represent stE, Ectoloph pos-
terior to this point very low, extending to meta-
stylar corner of tooth. Paracone taller than
stylar cusps but subequal in crown height to,
or shorter than, protocone. Mefaconc tallest
cusp. Prominent protoconule and metaconule.
Ectoloph continuous on buccal edge of crown.
Buccal concavity in crown outline slight. Para-
¢rista just longer than half length metacrista.
Paracrista extends from paracone to anterior

half of stB, Although slightly worn, paracrista
appears to curve at buccal end to contact stB.
Paracrista apparently transverse to imaginary
long axis of toothrow. Para-nictucrista. con-
tmuous, Slight protoconule ridge may be
present linking base of paracone to preproto-
crista, Clear metaconular ridge present linking
base of metacone to postprotocrista. Meta-
conular ridge extends short way up base of
Metacone causing bulge in base of that cusp,
Metaconular ridge bounds marked declivity
between posterior portion of steeply inclined
postprotocrista and posterior base of metacone.
Anterior cingulum complete, linking preproto-
crista and antero-buccal cingulum to parastylar
comer of tooth. No posterior cingulum present.
Pre- and postprotocrista form large but acuie
angle,

UCR, 15308 has at least four stylar cusps.
Possible stA as in F7332. Posterior crest from
stB descends steeply to base of stC. Between
stC and D, and connected by crests, another
smaller stylar cusp of uncertain homology. StI
small and connected to metastylar corner of

Sf M, ARCHER

tooth by low ecteloph crest, Paracone and
protecone subequal in crown height. Prote-
comple absent and protacenule swelling only
Just present. Metaconule large. Buccal con-
cavity in crown outline deeper than in F7332.
Paracrista almost three-quarters length meta-
crista, Puracrista wom but appears to intersect
ectoloph ow anterior flank of siB. No clear
protoconule ridge present, Metaconular ridge
sovall bur present, Metaconular ridge docs apt
clearly extend up base of metacone, Otherwise
morphology of UCR, 15308 as.in F7332.

Meristic gradients from M5 to M4 may be
summarized as follows, Tooth length decreases
Width increases, Ectoloph becomes more
evenly and decply concave. Stylar cusps, par-
licularly OD, become smaller. Paracristy atid
metacrista Increase in lenvth. Para-metacrista
hecomes more synitietrical. Prococemule de-
creases in size. Angle formed by pre- and post-
protocrista hecormes slightly more acute.
Antero-buccal cingwlim increases stghtly in
tength,

Dentary fragments Indicate premolar size.
UCR, 15340 hus alveoli for C4-P4- Premolar
alveoli subequal in size, indicating lite or no
reduction in tooth size fram Pi ta P4- Py, pre-
sumably had posterior cingular cusp which
corresponds with hypocornulid notch of My_ Py
anterior alveolus slightly crowded out of align-
ment but (as evidenced in modern species of
Plonigale, Accher |976a), does not necessarily
mean P44 crown gut of aligninent. Judging
from proximity of premolar and canine alveoli,
premolars and canine presumably contacted
one another antero-posteriosty. C4 alveolus
suggests Cy width exceeded that of any pre-
molar, but because of relatively unreduced Py,
C{ probably not greatly enlarged and com-
parable with canine of Nirgauw/ Archer, 1975b.

Mz? talonicd wader than inigonid but trigonid
not as compressed laterally as in most modern
dasyurids, Well-ieveloped anteniar cingulum,
which terminates Lingually for hypoconulid
notch, Parastyiid comer of tooth most anterior
portion of crown. Posterior cingulum com-
parable in Jength to anterior cingulum and ter-
minates buccal to posteriorly projecting hypo-
conulid, Basal cingulum absent beneath
postero-bireceal corner of protoconid and hypo-
conid. Roughened enamel suggests cingulum
present hetween base of protoconid and hypn-
conid, No lingual cmgulum, Paraconid low,
approximately same height as hypoconulid.
Protoconid tallest cusp of irigonid, Metaconid
just shorter thin proteconid, Hypovonid just

taller than entoconid which is taller thai pare-
conid, Paracristid complete between protaconid
and paraconid but almost vertical from proto-
coaid to shallow paracristid fissure and hor)
zontal between paracristid fissure and para-
conid. Metacristid steeply inclined on both
sides of metacristid fissure. Metacristid und
hypoeristid approximately transverse to long
axis of dentary. Cristid obliqua (damaged)
extends from hypoconid to trigonid intersecting
latter at point below protoconid tip, well buccal
to metacristid fissure. Hypocristid extends From
hypoconid to hypoconulid, without approach-
ing entoconid. Entoconid and hypoconulid net
connected by crest. Entoconid and metaconid
connected by high crest.

Ms talonid wider than trigonid. Antertor
and postenor cingulum as in Mj. Buccal cing
lum helween protovonid and hypaconid Jess
developed (absent in P18190). No lingual
cingulum, Paraconid smallest trigonid cusp but
subequal in height to hypocenid and entoconid.
Melaconid much taller than hypoconid and
just shorter than protoconid, Hypoconid and
entoconid subequal in height. Entoconid not
connected ta hypoconulid hy crest, but con-
nected to metaconid as in M4 except that crest
interrupted by shallow transverse groove, Para-
cristid from paracristid fissure ta paraconidl, in-
clined, nor horizontal, Crista obhqua extends
to base of protoconid as in My; but anterio¢
end appears tu be distinct contribution from
trigonid with slight fissure where talonid and
trigonid parts meet. Trigonid portion thicker
and more bulbous. Otherwise morphology Ms,
asin Mg.

Mg talonid just wider than trigonid. Anterior
and posterior cingulum as in Ms. Buccal cingu-
lum confined to arca betwecn base of proto-
conid and hypoconid, as thickened bulze of
enamel, clearly less well-developed than
anterioy and posterior cingula, No lingual
cingulum. Paraconid smallest trigonid cusp but
taller than any tabonid cusp. Entoconid not
connected ta hypoconulid hy crest, but con-
nected to metaconid, as in MG. Unlike eristid
obhiqua in Ma, this structure in My appears to
lack transverse fissure separating ¢rest inte
hypoconid and trigonid portions: This dif
ference between Mi and Ms, notable in
P18190, Cristid obliqua, also intersects trigonid
in slightly more lingual position than m Ma,
Otherwise morphology M%, as in M4.

My trigonid wider than talonid, but talonid
wider than that stricture in most modern
dasyurids. Anterior cingulum as in Ms, Pos

MIUCENE MARSUPICARNIVORES AF

terior cingulum absent. Buccal cingulum can-
fined to arta between protoconid and hypo-
conid, Paraconid just shorter than metaconid.
Etitoconid and hypoconid reduced relative to
My, but larger relative to most modern
dasyurids. Entoconid connected to base of
metaconid via low crest, Entoconid also com-
nected to Aypoconulid by low crest, Hypo-
crshd convex anteriorly. Hypoconulid sub-
equal in height to entoconid. Cristid obliqua in-
tersects trigonid base immediately buccal tu
point below metacristid fissure, this being
markedly more lingual than similar intersection
of cristid obliqua in My. Otherwise morphology
of M4 2 in Mg.

Meristic changes along touth row as follows.
Payuconid increases in height from M{ to Mj.
Melaconid height Mj4-4 subequal but meta-
cond length at byse of cusp decreases
tnackedly from M4 to M4, result of reduction
In size @f minor crest on posterior slope of
metaconid which links with enteconid, Ento-
conid M;- subequal and larger than ento-
conic of Mj, Hypoconulid M4-4 subequal in
height aud shorter than that cusp in Mi‘. Proto-
conid M4 shorter than protovonid My which
is subequal 10 that cusp in M4 which is larger
ihan that cusp it M4. Hypoconid decreases in
height from M4 ta Mg, Talonid M4-4 wider
than tigonid. Talonid M4 narrower than
trigonid. Paracristid M4-4 subequal and sub-
equal lo (P18190) or just shorter than (UCR,
15341) that crest in Mg. Paracristid M4
shorter than that of M4. Metacristid increases
in length from M4, to Mé. Metacristid My
shorter than metacristid M4. Cristid obliqua in-
tersects trigonid in progressively more lingual
Position from Mi to M4. Hypocristid My -{
subequal in length and longer than hypocristid
My. Anterior cingulum decreases in length
slightly from M4-M4. Posterior cingulum
M4 -:, subequal in length (absent in Mi)

Discussion und comparison

Anketarinja t a Metatherian because it has
four molars, a larec stylar shelf, and an
approximaled entocenid and hypoconulid. It
is also a marsupicarnivore because it has tribo
sphenic molars lacking hypocencs,

Denta| characietistics of known dasyurids,
peramelids, thylacinids, didelphids, and related
marsupicarnivores have been reviewed
(Archer 1976b) and, to avoid repetition; it is
sufficient to point out here that Avkerarinja
can only be regarded as either a dasyurid or
didelphid, Because morphological yariation of

teeth exhibited by didelphids and dasyurids
OVerlaps, only incisor number and possibly
dP4 cusp number permit diagnosis at the
family Jevel. Al) dasyurids have three lower
incisors and four upper jacisors on each aide,
whereas almost all didelphids have four lower
and five upper incisors (exceptions may include
Derorhynchas singularis Paula Couto, 1952,
species of Eodefphis Matthew, 1916). The pre-
maxilla and amlerior region of the dentary of
Ankotarinja Hrurensis are unknown, Therefore,
this marsupicarniyore cannot at present be
referred conclusively to either the Dasyuridae
or the Didelphidae,

Modern and fossil didelphid subfamilies
exhibit greater morphological variation than
dasyurids. Most are so distinct that their mere
subfamilial status has been questioned, and the
reasons they are doubtfully referred to the
Didclphidae are also the reasons they cannot
be related to Ankotarinija tirarensis. Only didel-
phines warrant closer comparison.

Some North and South Anictican didel-
phines are adequately illustrated (such as the
Paleocene forms described by Paula Couto
1952, 1962, 1970} but most are not. Archer
(19766) summarizes the must striking charac-
teristics of these forms, Only species of Coona
Simpson, 1938, Marmoyepsis Paula Couto,
1962, Mirandetheriunr Paula Couto, 1970,
Monadelphopsis Paula Couto, 1952, Derur-
hyachus Paula Cotte, 1952, [schyredidetphis
Paula Couto, 1952, Didelpieopris Paula Coute,
1952, Minwseulodelphis Paula Couto, 1962 and
an M+ listed by Paula Couto (1962) as incer-
fae sediv are similar to A, tirarensis, Among
Cretaceous didelphines, species of Alphadan
Simpson, 1927, and Pediowys Marsh, 1889,
warrant comparison. Modern didelphines used
here for companson include two species of
Marmosa Gray, i821, M. sp. and M-~ siitis
Bangs, 1898, Monodelphis dimidima (Wagner,
1847), Metachirus nudicaudanus ( Geoffroy,
1803), Philander opassun: Gray, 1843, and
Didelphis marsupialis Linnaeus, 1760, Of
these, species of Marmosa, Monedelpits Bur-
nett, 1830 and Didelphis Linmacus, 1758
warrant closer compucison with 4 nkofariila.

Characters of Ankotarinja which are wousual
among dasyurids and inVite broader compari
son within the Marsupicarnivera (the didel-
phine forms noted sbove) are as follaws: 1,
large M4 talonid; 2, relatively uncompressed
My trigonid: 3, huccal position of anterior end
of cristid obliqua; 4, transverse orientation of
metacristid: 5, large P4; 6, large M+; 7, stylar

x
jaa
=
Ss)
&%
<
=

MIOCENE MARSUPICARNIVORES

psc

i
ee

a4
w
q
oO
a)
<
=

MIOCENE MARSUPICARNIVORES aL

TABLE =
~~
= 22
z= =- 4
eisigde2Fs : Ei.

+ BEEPS EBE = = oc 5 = & =

= eSSesasgosesttEtses BS so = 3
a SStetPyVSEFrFsPsEerzsrTtSTrErzrsrzers te
23aé SEEPRBERQGFTSEBEPBRERESZ RE ESE Z
. SEESSSSSEFESSZSEESTEREER ERE =
Unusual characteristics $4 & Se iszeii =a TESPESS €£ 28 3
of Ankotarinia NHS SPTQRARSFS oS tre FBR SFeRrRzAA
1 Mj talonid Jarge + + F— tt 7 Lu, a4 tre gt ats b/ + ——— — 9
2% My trigonid wide ; Pa pt he F Bab ff BE feof fee =
3. Coo, buecal position +4 bope tee PHHeE F444 2 4 -——— H +
4. Transverse metacristid + }| | -—-—- -—-— Se
S. Large P4 + ele Pepe ole — -= -apepeaee tony —- 7
® Large Mg + YP 7 PP Fo PPE HHA HF HERK HE H HH HH |
7. stD small attd posterior Se re i a ey ec a ces eee ee +
6, Metaconule prominent FH FY He Veer MPR tem H tt ie eH Le 6b

cusp arrangement with reduced and posteriorly
positioned stD; and 8, prominent metaconule,
These characters are compared in Table 1 for
non-dasyurid marsupicarnivores noted above,
as well as dasyurids which provide the closest
approximations to Ankotarintja. In this Table,
characters 1-8 are the same as noted above.
A "+" records a condition similar to or which
approaches the condition found in Ankotarinja.
A “?" indicates either that the requisite part of
the dentition is unknown or not adequately pre-
served, A “—” indicates ‘a character condition
unlike that in Ankertarinia. In some genera
noted, some species may be similar to, while
others differ fram, the condition in Anko-
jarinja, In this case, a “+-" only is recorded.

M44, and M+ size

Large M4 and M4 talonid are related
characters and more common among siruc-
turally ancestral didelphines than dasyurids,
Very few dasyurids have the M4 talonid as
relatively large as it is in Ankotarinja, The M4
lalonid of most dasyurids is laterally e¢om-
pressed with one or at most two cusps present,

ihe hypoconid and entoconid or hypoconid

and hypoconulid, In this character, Ankorarinja
is most similar (among dasyurids) to Neo-
pRascogale and to a lesser extent Phascoloso-
rex. Three distinct cusps may sometimes be
present on the M4 talonid of other dasyurids

such as Murexia, Myoictis Gray, 1858 and
some Antechinus but in these forms the M4
trigonid is relatively larger than in Amkovarinja,
while the talonid is transversely compressed,
Relatively large M4 talonids characterize many
didelphine groups. For example, the talonid of
M4 in Marmosa is only slightly more reduced
than that of Ankotarinja. However, in Mar-
moya the whole M4 is not as reduced relative
to Mg as it is in <Anketarinja. Philander
Brisson, 1762 (33460), and Metachirus Bur-
meister, 1854 (J3461) also show a relatively
unreduced M4 talonid. Coona and Guggen-
heimia Paula Couto, 1952 have M4 talonids
even wider than the trigonids. Extremely
narrow My talonids (comparable with the con-
dition in most dasyurids) are found among
didelphines in species of Minusculodelphis and
and Marmosopsis,

Overall reduction of M4 relative to M4, such
as occurs in Ankararinja, does not occur in
any modern dasyurids. It is common only to
some didelphoids. M4 of Ankotarinja is un-
known but maxillary fragment UCR, 15343 in-
dicates that this tooth was as wide but not as
long antero-posteriorly as M+ of most didel-
phids. It was comparable in length to Mi of
some modern dasyurids such as Neophaseo-
gale, but wider than that tooth in most modern
dasyurids, Size of M4 in modern dasyurids

Fig. 3. A-E, scanning electron microscope photographs of Ankotarinja tirarensis. A~D, UCR, 15308,
LM}. .4, stereophotographs. E, UCR, 15343, maxillary fragment with alveoli for LM3-&

Fig. 4. A-E. scanning electron microscope photographs of Ankotarinja tiraréensis. A~C, F7332, LM3. A.
stereophotographs. D-E, Holotype, P18190, RMa-j, entoconid broken off RMj.

Fig. 5. A-E, scanning electron microscope photographs of Ankorarinja tirarensis. A-B, UCR, 15340,
dentary fragment with LM$-4. D, UCR, 15341, dentary fragment with LMé-, E, F7334, dentary

fragment with LMs-4.

62 M. ARCHER

appears to be related to relative length of the
cheek-tooth row, being shortee and more
reduced in forms with more compressed tooth
rows. This compression commonly occurs in
more sirictly carnivorous forms where
elnphasis is on development of metacrista-para-
cristid shearing elements, In more insectivorous
forms, paracristaanetacristid shearing elements
are relatively less reduced, resulting in a reda-
tively more functional M% paracrista anc
larger M4 talonid,

Reasons for overall (eduction of M4 relative
ta Mi afe not easy to interpret, In some dip-
rotodont marsupials where reduction of M4 is
advanced, sometimes even involving Insx, Py
is developed as a lage sectorial or even
plaviaulacoid tooth, possibly reflecting a shift
anterjorly m shearing emphasis, reducing the
importance of M4, In Anketarinja, although
P4 is large, the size is not particularly different
frem that of some didelphines which lack
reduced M%,

My, trigonid width, paracenid reduction, and
size of P&

Reduction of the M4 paraconid and Py are
related chsracters and often accompanied by
transverse compression of the Mj trigonid und
increase in relative importance of the proto-
conid. The relatively Jarge P4{ and wide tri-
gonid of M{ in Ankorarinja are unmatched
among living dasyurids, Even in Neopheasco-
gale the My triganid is transversely compressed
with gross reduction of the paraconid. In
Ankotarinja the M4 paraconid is very low on
the trigonid but not strongly deflected out of
alignment with whe other lingual cusps such as
occurs in most dasyurids, The closest match
among dasyurids is found it species of Smin-
thopsis, Murexla, and Keeuna but even here,
the M4 paraconid ts shifted anreriorly relative
to that cusp in Ankofarinja. A relatively wide
trigonid on M( occurs in most didelphoids, and
in part, reflects the relatively large Py m most
of these forms. In dasyurids which show pre-
molar reduction, it is P_ which ts reduced or
jost in the lower dentition (including the in-
teresting case of Planigale gifesi Aitken, 1972,
Archer 1976a), while in didelphoids and bor-
hyaenids it is Pi) which is normally reduced.
The only marked exception are specics of
Zygolester Ameghino, 1898, which are unique
among didelphids in having an extremely re-
duced, although two-rooted, P4. Reis (L957)
notes that in the M4 the paraconid is more
huceal in postin than is normal for the group.
This modificution is much tess than that seen

jv the trigonid of dasyurids with a comparably
reduced P{. Reduction of Pi normally occurs
among dasyurids which exhibit a compressed
cheek-tooth row, increased carnassiulily as
judged by proportionate in¢rease im metacrista-
paracristid letigth, enlarged canines, ete.
(Archer 1976b). Attendant reduction of the
M4 paraconid and increase in size of the Mj;
Protoconid shift the premolariform-molarifurm
boundary posteriorly. The M4 trigonid func-
fions as a stabbing, piercing premolac rather
than a sectortal tnigonid. Fucther, in dasyurids
which have lost P4, such as Daxyayrus Geallroy,
1796 and Sarcophilas, the metacrista-paracris-
tid length of M& is greater than thal of Mi, or
My. The net effect is to cancentrate the sec-
torial Function of the molars at a pom pos.
tevior of the middle of the molar row. Posterior
shift of the premolariform-mofanform boun-
dary may be scen in this way as merely main-
lainiog the structurally ancestral relationship
between these two types of teeth,

Axkorarinja is clearly structurally ancestral
in this regard and more similar te didelphids
than dasyurids.

Position of the cristid obliqua

The Mé cristid obliqua of Antotarinja
lirarensiy is unlike almost all dasyurids in that
it intersects the trigonid at a potnt so far buccal
to the metacristid carnassial notch. This con-
dition is approached in Neephuscogale and ta
a lesser extent in Keeuna, while in other
dasyurids the cristid obliqua tends to intersect
the trigonid just buccal to the carnissial swtch.
Th some dasyurids (e.g. Neophascogule and
Dasyurus) the protocenid fiank cantnhules to
the cristid obliqua on M4-4, and this same con-
dition occurs in Ankotarinja, at least on Mi,
Clemens (1966) suggests that lower teeth
(except perhaps dP4) of pedionryils can he
distinguished from species of Alphadon in thal
the crista obliqua in pediomyids intersects. the
base of the trigonid well buccal to the meta-
erishid fissure, This pediomyid condition is alsi
Present in all modern didelphids examined in
the present study and appears to be present in
itustrations of species of many of the Sopth
American fossil didelphines. The condition im
Ankorarinja is closer to this pediomyid and
didelphine condition than itis to most modern
dasyurids or Alphadon.

Althoueh the functional significance of this
difference is not clear, a relatively more buccul
posilion produces a larger tulonid basin. Posi-
lion Of the cristid Obligu|| must aiso reflect, posi-
tion of the paracone, a more buccal position

MIOCENE MARSUPICARNIVORES 63

indicating a relatively reduced or more bticcaliy
situated paracone.

The condilion fouwsd in Ankorarinja and
some dasyurids (c.g. Daryurvs) of a small
anléfiot component of the cristid obliqua
formed by the protocanid flank results in the
development of a amall accessory carnassial
notch agalnst which shears the paracone. This
makes an effective point-cutting unit that
supplements thse developed on the trizonid,

Transverse metacrintid

In Ankoterinja the metacristids sre trans.
Verse to the long axis of the dentary, In
dasyunds, this condilion ws present only in
species Uf Siinthopals, Aniechinoemys Kreift,
1867, and fo a lesser extent, species of Ningaui
and some species of Planigale. In other
dasyunds, the metaconid is displaced pos-
teriorly relative to the proloconid resulting in
the metacristid and paracristid formimg a more
obtuse angle. The tranaverse condition is
present among some but not all modern didel-
phines, some Cretaceous didclphoids, and many
Tertiary didelphids,

Among dasyurids, markedly non-transyerse
metacristids occur only in the more carnivorous
forms. This structural trend is noted by Bens-
ley (1903) who regards it as modification
towards longitudinal and away from transverse
shear, In this respect, the lower molars of
Ankotarinja demonstrate the structurally primi-
tive insectivorous condition, which is more
common among didelphids.

Stylar cusp size and position

Terminology of the stylac cusps of Anko-
tarinja used here is set out elsewhere (Archer
1976h).

The stylar cusp condition in Ankotarlnja ts
closet to that of didelphids than dasvurids in
having & large stylar cusp posterior to stB, an-
terior to stD and buccal to the low point in the
pata-metacrista of M3, which is the homologue
of the normal didelphid and variably present,
but invariably small, dasyurid stC, Further,
stD in Ankotarinja is smaller and slightly more
posterior in position than that cusp in modern
dasyurids. These non-dasyurid-ike features are
common among Cretaceous didclphines such as
glasbiines, some species of Alphadon anil
Pediomys.

Stylar cusps do not have occlusal counter-
parts in lower molars, Vet they clearly sustain
wear. This wear must result from food abrasion
during inilial puncturing prior to the cutting or
shearing occlusal phase. As the dentary closes,
forte is apphed to food trapped belWeen teeth

by the lower molars in opposition fo the whole
of the upper molars including the stylar shelf.
The area of the crushing or piticluring surfave
is increased by Jarger stylar cusps. In marsupi-
carnivores, stylar cusp reduction occurs tn the
more camivorous forms such as Sarcoplilis.
Thylacinus Temminck, 1824. and borhyaenids
where perhaps the puncturing value of these
cusps is overshadowed by the feed to have
large and sturdy shearme crests, The stvfnr
cusp size and arrangement in Ankoturinjia may
therefore be interpreted as evidence for insecti-
Vorous rather than carnivorous habits.

Metaconule development

In Ankorarinja the metaconule of the upper
molars is 7 prominent feature, while the proto-
conule is not present in M4. Conule develop-
ment is present in some dasyurids and many
didelphids. It is well-developed in most Creta
ceaus didelphines where both protoconules and
metaconules occur. Simpson (1928) notes that
these cusps in recent didelphids are represented
at most by vestiges.

The possible functional significance of meta-
conules is discussed elsewhere (Archer 1971}.
In addition, well-developed metaconules may
serve as shearing counterparts for the hypo-
conid and buccal edge of the hypocristid,
Sumunery

The dentition of Ankorarinja resembles that
of many didelphids, and some dasyurids such
as Neopharcogale, Murexia, and Sminthupsis.
Because of the middle Miocene age of the
deposit, as well as the fact that two relatively
more modern-type dasyurids (Keeune and an
unnamed form noted by Stirton, Tedford, &
Miller 1961) are in the same deposit, it would
be absurd to regard Ankorarinja as the ances-
tral dasyurid. However, it does preserve charac-
ters which could be regarded as structurally an-
cestral lo modern dasyurids. It t& clear that
many South American Tertiary didelphines of
Paleocene age (Graham & Ride 1967) share
characters with Ankotarinja which it docs not
share with moder dysyurids. Taking into
account all available morphological characters
jozether with what is currently known of their
distribution, and recognizing that the charac-
ters (incisor number and dP4 morphology)
which will distinguish between the two lincages
Dasyuroidea and Didelphoidea wee not pre-
served, one cannot avoid concluding that the
data availuble indicate a marsupicarnivore-
prohably betonging to the Dasyuroidea, but
which, like the slightly older phalangeroid
Hynyardia Spencer, 1900 (Ride 1964b), also

Fig, 6, Specimens of Keeuna woodburnei and their measurements (mm). 4, Holotype, PI8!91, RMS.
B, F7333, maxillary fragment with LM2 (broken), LM3 and alveoli for LM$. C, UCR, 15346,
dentary fragment with LP4 erupting and alveoli for LC\-M4. D, F7334, LM4. £, UCR, 15344,
LM: or Mé, F, tentatively referred to K. woadburnei, UCR, 15347, premaxitlary fragment, with
alveoli for RIt-4: d represents diastema between alveoli for REY and RI?; pmx.v. represents
edge of premaxillary vacuity.

retains a number of characters of a sort which
reyeals its derivation from early Tertiary didel-
phoids or the didelphoid-dasyurid stem. If it
should be discovered that Ankotarinja possesses
a didelphotd incisor number, the author ‘will
have no hesitation in describing it as a didel-
phoid with characters foreshadowing dasyurids,
but it is clear that it is far too late in time to
be an ancestor to the whole dasyurid family.
For the present, it is listed as ?7Dasyuridae.

Genus KEEUNA nov.

Type species; Keeuna woodburnei sp. nov. (by
designation and monotypy)-

Generic diagnosis: Differs from other Austra-
lian and New Guinean dasyurids including
Ankotarinja in combining, relatively reduced
stD on M2-%, Jarge M4, antero-posteriorly
shortened M*, and virtually absent posterior
cingulum on M4-4 (although mild posterior

cingular swelling present on holotype), and
relatively unreduced P4.

Origin of generic natne> Kee, central Austra-
lian Aboriginal word for wild cat; una, central
Australian Aboriginal word for forgotten
(Smith 1880). Keeuna is here given masculine
gender,

Keeuna woodburnei sp. nov.

FIGS 6-8
Holotype: P18191, isolated RM%.
Type locality; Tedford locality, Etadunna For-
mation, Lake Palankarinna, Etadunna Station,
S.A. (28°47'S, 138°25E).
Diagnosis: That of genus. Relative develop-
ment of entoconids, stylar cusps, and size may
prove to be diagnostic species characters.
Origin of specific name; The specific name is
in honour of Dr M. O. Woodburne who helped

MIOCENE MARSUPICARNIVOKES 65

find all of the malerial referred to this species
and who, with Dr W, A. Clemens, gave the
author fs first opportunity to study Australian
Tertiary fossils.
Referred specimens: UCR, 15271, RMit:
17333, left maxillary fragment with M® and
part of M3, UCR, 15347, right premaxillary
fraginent; UCR, 15344, isoluted LM 4; UCR,
13348, left dentary fragment; UCR, 15286, tri-
gonid RM 4; UCR, 15269, trigonid RM 4;
UCR, 15274. isolated Me; UCR, 15345, Icfi
dentary fragment; UCR, 15346, dentary frag-
ment with LP4 erupting, F7334, isolated LM4,
Description

Maxillary fragment (F7333) has LM%, pos-
lecior part LM, and alveoli for LMt. M4
narower than M3 but relatively hittle reduced
anlero-posteriorly, being only slightly shortcr
than M4. Metacone toot M4 large, equivalent!
in size to protacone root, and only just smaller
than paracone root. Interdental fenestrae occur
between M2 and M3 and between M% and Mt.
Maxillary root of zygoniatic arch arises buccal
to M*-4, Infraorbital canal opens on anterior
edge of maxillary fragment, dorsal to anterior
end M&,

Premaxillary fragment (UCR, 15347) may rep-
resent K. woodhurned on basis of size, because
larger than would be expected for other known
Negapakaldi forms. Four alveoli present.
Largest represents RL}, (alveolus broken}. This
alveolus separated from alveolus for R1I® by
Very short diastema, Alveoli for RIS-4 contact
one another. RE alveolus inclined posteriorly
and suggests RI} inclined antero-ventrally.
Based on alveolar size, RI1 largest incisor,
RIS exceeded RI4 in Jength which exceeded
RI4 in length. This may be misleading sinve
posterior lohe of RI4 (occurrence of which is
not uncommon among modern dasyurids}] may
have caused this tooth to be longer than R12.
Also, root for RI} commonly large in modern
dasyurids, while crown muy be very reduced.
Posterior to RI! alveolus, premaxillary wall
descends into pit which is ovclusal counterpart
of RC. indicating RC4 large and caniniform.

M5 with continuous anterior cingulum con-
necting parastylar corner of tooth to preproto-
crista (piece of enamel missing from anterior
cingulum of holotype). Posterior cingulum
absent (F7333) to doubtfally present
(PIS191) as swelling at base of crown abave
metacone root. Swelling not continuous with
Postprotocrista whereas this is the case in
moder dasyurids whth undoubted posterior
cingula, Five distinct buccal cusps present, StA

Gecurs at buccal end of anterior cingulum,
hetween parastylar commer of tooth and stB.
Almost vertical, minor crest copnects stA to
parastylar corner of tooth, that part of tooth
which would overlap postero-buccal edge ol
P+, Very minor, more gently inclined crest
comects sLA to stB. StH connected to paracone
by long paracrista. Buccal crest extends pos-
teriorly From stB to contact small, jow stC
which is adjacent to another small low stylar
cusp of uncertuin homology. These two stylar
cusps not connected hy crest, The posterior
small cusp connects to larger stD by Minne,
inclined crest. StD largest stylar cusp, bul
stialler than (fat cusp in M% of modern
dasyurids. StD connects lo metastylur corner
of tooth by lung, low crest, No evidence of
stE. From metastylar corner of tonth, three
crests radiate: buecal crest to stD. metacrista
fo Metacone; and minor short crest that extends
amero-lingually from metastylar cover and
ends within short distance of toothy corner.
Crescentic enamel ridge occurs lingual to mic
point of ectoloph, and buceal to mid-point of
para-metacrista, Ridge may represent cusp
analogous (o similar structure in some speci-
mens of Srinthepsis virginlaeg Tarragon, 1847.
Paracone shorter than metacane, and two
cusps widely separated. Small but clear meta.
conular jidge extends from base of metacone
to middle of postprotocrista. Postprotocrista
terminales as steep-sided crest adjavent to base
of metacone. Deep pyramid-shaped fossa
exists between hases of paracone and metacone
and buccal flank of protocone. Metucrista
approximately 1.5 {imes length paracrista.
E¢tofiexus in ectoloph slight, poimt of inflectinn
immediately posterior to siC,

Mé@ with continous antenor cingultim, Pos-
terior cingulum absent. At least four srylar
cusps present. Parastylar extension of tooth
small and accordingly litthe overlap of imeta-
stvlar cornet of M3 occurs, SiC connected to
StB as in M2, On anterior flank of cusp called
stC (F7333), small, possibly distinct swelling
occurs which may be homologous with stC of
M5. If so, cusp posterior tn that cusp in MS
might be homologous with single conspicuous
cusp in this position on M3. Posterior to stC
of M&% ts slightly larger stD. Posterior to stD,
and connected to it by crest, is stE, This cusp
exlends postertorly es ridge which terminates
short of metastylar corner of tooth. Only meta-
crista radiates from metastylar corner of tooth,
Lingual Io stC, as in M5, crescentic ndge
occurs Which may be distinct cvsp. In addition

66 M, ARCHER
TABLE 2
=
= Pa]
2 2 % =
a Se ¢ = s s & & i 8
S = §s§ = & S&S F€ ze | € SF = € F
= = 3 = = & = € 3 = 3. = 3 = 3
= = S S = 3 — 4 = = = 3 G =
Unusual charagrers: >: = 3 € S &- 8&8 $ § & § F = F
of Koenna *# 5 & = fi = Sc S& = B® FE Be
1, Large Ms i, i i i a i is cs ell
2, M* lack post, cing, | + | Fo ~ $$ +t + — —| = + + +
4 ArmallsiD tek + S Ft ff $F HF KE KE HK —— +
4. Compresacd M+) — Se a ee — —» om aa ae

to metaconular ridgc, as in M2, clear imeta-
conular swelling present on pestprotecrista, Na
clear proteconnle er protoconular ridge
present. Metucrisin less (han 1.5 times length
Paracrista, Ectoflexus M*# broad and relatively
deep. Point of inflection in ectoflexus occurs
anterior to stC. Otherwise morphology of MS
as in M3.

Dentary with two branches of inferior den-
lal canal, one emerging at point Tevel with
middle of M4 on buccal surface of dentary
while other emerges at point beneath posterior
root Py (latter condition determined fram
juvenile dentary, UCR, 15346). UCR, 15344
only specimen with premolar alveoli preserved.
Pi, erupting in this specimen. As result.
apparently crowded premolar condilion may
besome less crowded in adult dentary. Pre-
molar gradicnt suggested by alveoli; Py shorter
than P4 which js subequal in Jength to P4,
DP, alveoli suggest tooth as large aus P4{ and
Iwo-rooted, P4 alveoli acutely oblique with an-
terior root postero-buccal to C4 alveolus. Pos-
terior root Pi immediately posterior to C4
alveolus. Py and «P4 alveoli suggest iccth only
mildly out of alignment (although crown
alignment need not be reflected in root align-
ment), C4 alveolus relatively small, suggesting
tooth no wider than Pa.

Py partly erupted snd partly obscured. Tooth
single-cusped with tall protocontd. Paracristid
steep, No anterior cingulum cusp. Metucristid
more gently inclined and appears to direcily
contact very small posterior cingulum cusp. No
buccal of lingual cingula evident.

M; talonid wider than trigonid, and trigonid
more laterally compressed than What structure
in| Ankerarinja, Anterior cingulum relatively
well-developed, terminating lingually with
hypoconulid notch, and terminating buccally
anterior to buccal base of protoconid. Posterior
cingulum slightly shorter than anterior cingu-
lum and extends to contact hypeconulid. Pro-
Hounked crmgulum development, as bulac

hetween bases of protocanid and hypocenid,
No lingual cingulum: Paraconid low on crown,
approximately same hetght ax entoconid.. Proto-
conid tallest trigonid cusp, Metaconid just
shorter than prntocomid, Hypoconid subeqial
to entaconid in height. Paracristid complete
hetween protocanid and paraconid, and
anterior part of crest steeply inclined. Meta-
cristid and apparently paracristid fissures
extend below cutting edges of crests. Crest
descends from posterior wall of melaconid and
meets crest from anterior wall of entoconid.
‘Two efests meet with shallow, open fissure,
Crest development from posterior wall of cnto-
conid slight to absent. No crest links entoconid
and hypoconulid. Hypocristid extends postera
lingually from hypoconid to hypoconulid. Both
hypocristid and metacristid clearly not trans-
verse to long axis of tooth, Cristid obliqua in-
tersects trigonid well buccal to point below
metacristid fissure:

Isolated molars posterior ta M4 not been
posilively identified. UCR, 15274 and UCR,
15344 tentatively regarded as tepresenting M4
and M‘ respectively, This conclusion based an
trigonid width and paraconid height. Both in-
crease posteriorly in most modern dasyurids
between M4 and Mé.

UCR, 15274, LM%, trigonid just narrower
than talonid. Hypoconulid notch between paru-
stylid and lingual end of anterior cingulum
larger than in M4. Cingulum development
hetween base of protoconid and hypoconid
does not produce buccal convexity in crown
olitline. Paracomid higher on crown than in
M+. Paracristid fissure well-developed below
crest. Cristied obliqua intersects trigonid ans
extends short way up trigonid wall at point
lingual te paint of same intersection in M4.
Otherwise morphology UCR, 15274 same as
Mj.

Ger, 15344, LM4, trigonid and talonid sub-
equal in width. Hypoconulid natch larger than
in M4 but subequal to that of M4, Posterior

MIOCENE MARSUPICARNIVORES 67

cingulum Jess convex posterlorly than in M4 ar
Mi, Also, posterior cingulum extends lingually
ad contacts hypoconulid rather than stopping
short of it as in M4 and Ma, Basal cingulum
between protoconid and hypaconid well-
developed but does not cause buccal convexity.
Crests linking metaconid and entoconid Jess-
developed than in M4 and Mg (although M4
damaged in this region). Cristid obliqua inter-
sects trigonid as in Ms. Otherwise morphology
Mg, a3 in Mg.
Discussion aud comparison

A summary of important dental characters
in marsupicarnivores in general is given else-
where (Archer 19766) and to avoid repetition,
discussion of dental characters of Keeuna is
largely restricted to those characters which
eillher make Keevna unusual among the
Dasyuridae or sugeest aMnities outside of that
family. Comparison within the Marsupicarni-
vora is restricted to the Dasynridse and Didel-
phidae. because no other families contain Forms
even remotely similar to Keeuna,

Keeuna cannot be referred conclusively to
the Dasyuridae for the same reasons given
above in the discussion of Anketarinja. How-
ever, the features of Keeuna@ are more suigges-
tive of known dasyurids than didelphids, If the
referred premaxilla does in fact represent
Keeuna, there is no reason to doubt its
reference to the Dasyuridac, This specimen
shows alveoli for four incisors, The diastema
behind the anterior alveolus confirms that ihe
anterior alveolus represents 11. No dasyurid or
didelphid known to the author has a diastema
within the incisor row posterior to 13 or an-
terior to 15, other than a variably preseni dia-
stema between U4 and 13.

Although the dental characters of Keeuna
are all present in one dasyarid or another, con-
sidered together: 1, the large M4 (inferred
from alveoli): 2, virlual Jack of a posterior
cingulum on the upper molars; 3, relatively re-
duced stD on M%; and 4, antern-posteriorly
shortened M4; they make Keenna unique
among dasyurids. These characters are com-
pared in Table 2 for didelphids and dasyurids
which provide the closest similarities in upper
molar morphology to Keeuna. Many South
American Tertiary didelphids which have a
lower molar morphology (ez, Adtrandatherium,
Paula Cotto 1962, fig. 4) simifar to that of
Keeuna, are unrepresented by upper molars
and not included in Table 2. Characters 1-4
are the same as noted above. A “+" records a
condition similar to or closely apptoaching thal

found in Keeuna, A “? indicates elther thal
the dentition is too incomplete or poorly pre-
served to enable determination. A “—" indh-
cates a condition unlike that im Keeund. Poly-
lypic genera which have some forms similar to
but othets differing from Keenna are recorded
only as “sl. M* indicates any of all upper
molars.

M4 size

The possible significance of a large M4 is
discussed above. Keewra exhibits similarities in
this regard la many didelphids but only a few
struclurally ancestral dasyurlds, meluding
Atkoterinja.

Postenlor citigula on upper molars

Elsewhere (Archer (1976b) i has been
noted that cingula have the effect of increasing
molar surface area. [i is alsa possible that pos-
terior cingula on upper molars act as supple-
Mmentary shearing crests in opposition to the
paraccistids which come into effect after the
paracristids shear past the metacrista. Distribu-
tion of this character in modern dasyurids
(Archer 19766) does not appear to lend itself
to phylogenetic interpretation, being present in
some but not other species of single genera
such as Anilechinuy, Absence in Keeuna,
although perhaps phylogenctlically unimpor-
tant, is useful in combination with other
characiers for differentiating the gents.

Sylar cusp D

Small size of this cusp allres Keeura with
Ankotarinja, a5 well as. with many didelphids.

The possible significance of stylar cusp size
1s discussed. above in regard to Ankorarinja.

Compressed M4

M# of Keeuna is longitudinally compressed
tu comparison with structurally ancestral
dasyurids such as Neephascegale hut not in
comparison with structurally derived dasyurids
such as Seniatiepyis whose molars are even
mote compressed, Bensley (1903) liotes reta-
tive compression in some dasyurids and regards
this as a more insectivorous thau carnivorous
adaptation. Extremes of non-compression, such
aus occur in Sarcoppilus, result in shearing
crests which approach longitudinal rather than
a transverse crientalion,

Detaled comparisons

Overall, Keevna more closely rescmbles
some dasyunids tocluding Phascolosorex, Neo-
phascogale, some Antechinus, Murexia, and
Ankotarinja, than it does didelphids. Particnlar
similarities and differences are noted below

68 ML.

Similarittes which extend to all dasyurids are
not noted.

Phascoloserex; Similaritics include relatively
large M+ and metaconule. Upper molars of
Keeuna differ from those of Phasecolosarex
(e.g. AMNH, 109758, 151992, 101975 and
LO9757) in that stylar cusps mutch closer to
buceal edge of crown; small stylar cusp occurs
lingual io st© (although in some specimens of
species of Phascolaseres such as AMNH,
151992, this Cusp suggested on M5); M*
shorter antero-posteriorly; paracnsta ond metu-
cTista enclose more acute angle; and Mt-+ lack
clear postenor cingula, In lower molars of
Keewna, trigonid and paraconid of M{ much
less reduced; crests From posterior face of
metaconids much better-developed; talonid M:,
relatively slightly wider.

Neophayeopule: Large M+ of Keeuna similar
to that of Neophascogale (tg, AMNH,
109524), Differences in upper molars of Keewra
include these noted above in comparison with
Phascolesorex as well as lack of distinct antero-
lingual low crest developed on base of prote-
cone (Which has nothing to do with preproto-
crista); ectoloph and para-mctacnsta relatively
more widely separite at their closest point: pro-
toconule slightly better-developed. Lower
molars Keenna differ in having less reduced
Mj trigonid: relatively shorter, wider molars;
cristid obliqua Which intersects trigonid in rela-
tively more lingual position (notable in M4~-4)5
lack of post-entoconid crest which directly con-
nects to bypoconulid: relatively lower talanids,
higher trigonids; relatively shorter talonids,

Anteckinuy: Resemblance with some Anie-
ghinus (e.g. A, mayert (Rothschild & Dollman,
1930), AMNH, 109816, A, sp. AMNE,
150877 Erom New Guineas, and A. trelanure
(Thomas, 1399), WAM, M5517) considerable
including overall propornions of M%-8: some-
what similar reduction of st1D on M®*: relatively
unreduced P4. Upper molars Keeune differ in
that siD relutively slightly more reduced on
M3; stB relatively more posterior on ectoloph:
all stylar cusps relatively smaller; M4-3 some-
what sharter antero-postertorly; M4 notably
longer in proportion to length of M4; posterior
cingwla absent; metacone and paracone M4-+

ARCHER

relatively closer in height; larger, more veon-
spicuoua cusp of crest occurs lingual to stC;
ectoflexus in relatively more posterior position;
protocone shorter antero-posteriorly at its
longest poinf, Lower molars Keewee differ in
having less-compressed trigonid on M4 with
larger puracond; relatively wider talonid on
Mz; lack complete buccal cingulum such as
occurs on Mé~, of some Antechinus species
(eg, A. mayeri); lower molars relatively
shorter, wider; enfoconids M4-5_ relatively
taller; hypoconulid wider and extends farther
from: postero-lingual comer of M4-4.

Murexia: Similarities between ‘species of
Murexia (eg. M. fongicaudata (Schlegel
1866) (AMNH, 101972 and 152035)) in-
chide comparable relative length of M4; rela-
tively unreduced P4. Upper molats of Keewria
differ in same features from Murexia as they
do from molars of Antechinux except us
follows, In Keeway all stylar cusps except C
relatively smaller; posterior cingulum of upper
molars virtually undeveloped (although only
slight postertur cingular development occurs in
species of Murexia); M$ relatively longer;
metaconular crest from base of metacone
less well-developed and Iscks low, minor crest
linking stD with metacone (latter observed
only in unwom Specimens of Murexia
examined in this study, AMNH, 152035).
Lower molars of Keeuna differ in same features
from teeth of Murexia that differentiate teeth
of Antechines, except as follows. In lower
molars of Keeyna entoconids relatively shorter
antcro-posteriorly, and higher; buccal cingulum
absent (occurs in one specimen of Mureala.
AMNH, 152035); low direct crest linking pes-
terior face of entoconid with hypoconulid
absent.

Ankotarinja: Comparison with much smaller
Ankatarinja demonstrates that bath farms simi-
lar tn having relatively small stylar cusps (par-
ticularly stD); atylar cusp(s}) present betweed
B and D; no posterior cingulum; complete an-
terior cingulum; slightly smaller paracones than
melacones; Lick of direct crest linking ento-
conid with hypoconulid; relatively large P4; un-
reduced trigonid and paraconid of M{- Upper
molars of Keeuna differ from those of A4nke-
farinja in larger stze and more gntetior position

Fig. 7. A-F, scanning electron microscope photographs of Keenan woodburnei, A, Holotype, P1811,
KM2, stereo photographs. B, tentatively referred lo K. woodburnei, UCR, 15347, prensaxillary
fragment with alveoli tor RLY-5 and posterior edge of alveolus for RI, C, F7334, LM4. D, UCR,
15344, LM or LM&, E, UCR, 15346, dentury fragment with unerupied LP4 aad alveoli for
1.P4-M% and edges of alveoli for LO{ and LMé,

n
jaa]
4
o
z
Z
4
Ss)
—
i
5
DH
4
<
=
jaa
Zi
jaa}
Q
}
o
=

M. ARCHER

ek
se

MIOCENE MARSUPICARNIVORES 7!

of stD on M3-3; number of cusps in position
of stC; presence of siylar crest limgual to siC;
less. well-developed metaconule,; presence of
deep pit hetween bases of paracone und meta-
cone, Lower molars of Keene differ in having
longer metacrista; relalively larger M4; more
lingual intersection of cristid obliqua and in-
Zonid; non-transverse inetacristid and hypocris-
tid; large hypoconulid of Mg; antero-posteriurly
non-compressed inigonids; relatively unequal
heights of the paraconids and metaconids. Dif-
ferences in position of intersection of cristid
obliqua and trigonid in <Ankoturinja and
Keeune not one of kind, but degree, position
being telalively more lingual in Keeunes,

Other comparisons; No other dasyurids warrant
dctailed comparisons. Most didelphids reveal
fewer similarities, particularly if the referred
premaxillary fragment of Keeuna does in fact
belong to this form. General similarities with
Cretaceous didelphines (e.g, some species of
Alphaden such as A. rhelster) include ihe
paracone, which although smoller than the
metacone, 1s not markedly so. Other similarities
between Keeung and species of Alpfiddon it-
clude relatively reduced atylar cusps (except
8), particularly D on M& of species such as 4.
rhoissery and A. Jullf Clemens, 1966, and
presence uf slylar cusp or crest lingual to sic
in A, rhaister. Marked differences in upper
molars of Keeuna include much smaller proto-
ahd metaconules; less-deeply concave and sym-
metric ectoloph of M4; stnaller stB; and absent
postenor cingulum (present in some species of
Alphadon). In lower molars, similarities
between Keryna and species of Alphadon in-
clude relatively unreduced ingonid of M4, Dif-
ferences in lower molars of Keexra include
much smaller paraconid relative to metaconid
in Ms-3 a8 opposed to specimens referred to
species of Alphadon (Clemens [966, Lille-
graven 1969),

Compited with holarctic Tertiary didel-
phines, relative size of paracane in Keetina is
similar to condition seen in species of Pera-
dectes and Peratherident. Other similarities in-
clude relatively fow stylar cusps; unpeduced
condition of Py (and presumably P+) and M4
paraconid; lack ul posterior cingulum on upper
molars; relatively shallow indentation of ecto-
loph; extreme buccal position of styfar cusps:
and evidently oon-iransverse onentation of

metucristid and hypocristid. Differences in
Keeund include relatively larger stylar cusps
(particulurly D): relatively poorly-developed
metaconule; non-pediomyid—ike cristid obliqua
Urientation; relatively taller metqconid and en-
toconid, and shorter paraconid: and presence of
rudimentary buccal cingulum between bases of
protoconid and hypoconid, Despite these dif-
ferences, Keeuna more eloscly resembles these
uidelphines than any others for which good
illustrations or photographs are available.

Close similarities may exist between Keeuna
and some Paleocene (Riochican) didelphids.
Untortunately very few are Enown from upper
teeth and few are adequately illustrated. As
noled ubove, all modern aidelphines examined,
and mast Tertiary didelphines exhibit a pediv-
mytid type of cristid obliqua, which diflers from
Keeuna. Of all Paleocene forms ilfustrated hy
Paula Couto (1952, 1962, L970) and Simpson
(1947), Mérandatherium igs pevhaps most like
Keeuna, Keeuna differs from this didelphid in
felatively smaller size of Py and presence of
buccal cingulum developed between bases of
protoconid and hypoconid. Upper molars of
Mirandatherium are unknown,

Other didelphids do not reveal enough simi
laritics to warrant separate comparisons,

Summary

Teeth of Keeuna, although resemblmeg teeth
of some Tertiary didelphines such as species of
Peratherium, Peradectes, and Mirandatherinwn,
are broadly similar to teeth of some anodern
dasyurids such as New Guinean Antechinus, In
view also of the Australian localiry of Keene
it seems most logical to regard this form as a
somewhat unusual dasyurid, probably withour
direct descendants in the modem dasyurid
fauna. Resemblances between upper molars of
Ankotarinja and Keeuna further suggest the
possibility thal these two forms are more
closely related to each other than either is to
other dasyurids, thereby placing Keeuwra in 3
structurally intermediate position between
Ankotarinja and made dasyurids,

Discussion of Etadunna marsupicarnivores

Stitton, Tedford, & Miller (1961, p, 35)
briefly describe {but not name) another carni-
Vore fram the Etadunna Formation. According
to their description “The size of the animal is
comparable to Dasyrrus quoll .. . The three

Pig. 8. 4-C, scanning ¢lecttun microscope photographs of Keetna woodburmel, F7333, maxillary frag-
Ment with partial L.M3, LM and alveoli for LM: C, stereo photographs.

72 M, ARCHER

premolars with gradation in size from P4 to Py,
and (he absence of the metaconid on M4 sug
gests that this animal may not be far removed
from the ancestry of Thylacinas’. The author
has seen drawings and photographs of this
Specimen (courtesy Dr W. A. Clemens and
Mr C. Campbell) and it is clear that nothing
else about the specimen, including the mor-
phology ef the upper molars, shows any simi-
larity 10 Thylactnus. On the contrary, it appears
to represent another dasyurid lineage (perhaps
related 16 Dyttyerus) in which metaconid re-
duction his occurred only on M4, This un-
named dasyurid and Keexte ate the only Eta-
diinna camiveres which can be referred with
some confidence to the Dasyuridae,

Ankeravinia ts either a didelphid or dasyurid.
Compared with known didelphoids, dasytraids
and perameloids, the preserved portions of
Anketarinja do not enable confident reference
to a particular marsupial family. It is referred
to here as ?Dusyuridae.

Ankeotarinja and Keéuna compared with
modern dasyurids, share most Uenial chanw-
fers with living New Guinean species of Nea-
phascegale, Phescolosorex, and Antechinus.
Similarities with Australian forms are fewer
and those that do exist are with forms found
in generally non-arid Australian habitats, Broacl
My, talonid, large M4, aartrow and relatively
uncrowded premolar tow, and Jarge entoconids
are characters either lacking or sare in Aus-
tralian arid-adapted dasyurids, Several authors
(c.g. Woodburne 1967, Schodde & Calaby
1972, Stirton, Tedford, & Woodburne | 96%)
indicate that New Guinea hag many mammals
(eg. species of Dendrolagus Muller, 1839,
Doreopsis Schlegel & Mulles, 1839, Diy-
teechuris Peters, 1874, Microperorycies Stein,
1932. Murexia, Myoictis, Neophaycogale and
Phascalosorex) from highland rainforest habi-
tats which appear to be structurally ancestral
within their respective families.

Evidence for a Iless-arikl central Australia
during Ngapakaldi time is reviewed by Stirton,
Tedford, & Woodburne (1968). In addition,
pollen from the base of che Etudunna Forma-
tion, recently obtained from bores in South
Australia, has been found to include Notho-
fagus sp. (pers. comm., W, &. Harris, South
Australian Department of Mines), a genus of
plants presently restricted in the Australian
fegion to the high-rainfall habitats of eastern

Australia, New Guinea and New Zeulanct,
Resemblance between Negapakaldi marsupicar-
hivores and Iilying New Gumean dasyurids,
suggests these living highland New Guinean
dasyurids may have avoided verlain selective
pressures brought to bear on mursupicurnivores
living in central Australia, following Ngapa-
kaldj time. These pressures may have included
progressive deterioration of climate with in-
creasing aridiwly. Elsewhere, as part of revision
of the dasyurid genus Sminthopsis, the author
has given evidence for believing that several
Australian dasyurid lineages underwent arid:
adaptation, Arid-adapted forms now dominate
the majority of Australia and are in clear con-
trast to the mursupicarnivores of the Npapa-
kaldi local fauna deseribed here.

Acknowledgments

] am deeply indebted to Drs W. A. Clemens
and M,. 0. Woodbume of the University of
California at Berkeley and Riverside respec-
lively, for giving me the opportunity to help
collect and study Australian Tertiary Fossils jn
1971, Dr Woodburne wis-also a field associate
and co-worker in 1972, These trips were
financed by a National Science Foundation
Grant to Drs Woodburne and Clemens, Dr
R. H. Tedford, American Musewn of Naturul
History, kindly allowed me to examine in the
field dasyurid and other fossil material
recovered from deposits in the Lake Frome
Basin, South Australia, in 1973, Aspects of the
1973 trip were financed by a» National Science
Foundation Grant to Dr Tedford. Other people
wha helped make the 1972 collections included
Messrs R, Lawson, W, Head, South Australian
Museum, and my wife Elizabeth. Mr C.
Campbell of the Universiry of California at
Berkeley allowed me to examine photographs
of dasyurid material he a at present examining.
Dr Clemens provided casts of upper molars of
Lance species of Alphudon. Drs W..D- L. Ride,
Western Australian Museum, and A, Bartholo-
mui, Qucensland Museum, read and construs-
lively critized drafts of this manuscript. Mrs C.
Farlow and Miss P. Rainbird of the Queens-
land Museum typed drafts of the manuscript.
Messrs A. Easton and A. Elliot of the Queens-
land Museum helped with ospects of light
photography. Mr J. Hardy, University of
Queensland, produced the excellent scanning
cleviron microscope photographs_

MIGCENE MARSUPICARNIVORES 73

References

ArcHer, M. (1974)—The developanent of the
Cheek-teeth in Awicehinus flavipes (Mar-
supialia, Dusyyridae), J. R, Soc, West Aust.
87, 54-43.

Arcoer, M. (19750},—The developtnent of pre-
molar and molar crowns of Avitechinus
flavipes (Marsupialia, Dusyuridae) and the
significance of cusp ontogeny in mammalian
tecth. J. R. Soe. West, Awst. 57, 118-325-

Arcues, M. (1975h).—ingamt, a new genus of
tiny dasyurids (marsupialia) and two new
species, trom atid Western Austrslia, WN.
dimealey! and N, ride’. Men, Gd Meas, 17,
27-29.

Arcuen, M. (1975c).—Abnormal dental develop-
ment and its significance in dasyurids and
other marsuplals, Afem, Gd Mrs. 17, 251-265.

Archer, M, ¢1976a).—Kevision of the niarsuplal
dasyucid genus Plenigele Troughton, Ment.
Od Murs., in press.

Arcurk, M, (1976b).—The dasyurid dentition
and its relationships to that of didelphids,
thylacinids, borhyaenids (Marsupicarnivora)
and peramelids (Peramelina). Avs, J, Zool.,
Suppl. Series, in press.

Bensvey, H, A, (1903).—On the evolution of the
Australian Marsupialia; with remarks on the
relatiouships of the marsupials jn general.
Trane. Linn, Sav. Lond, (Zeal.) 9, 83-217,

Clemens, W. A. (1966).—Fossil mammule of the
type Lance Formation Wyoming Part I.
bres aiele. Bull, Dep. Geol, Univ, Calif, 62,
1-122,

Cremens, W. A. (1975).—Fossi) mammals of the
Type Lance Formation Wyoming Part_ TIL.
Eutheria and summary. Aull. Dep. Geel.
Univ, Cali7, 94. 1-1u2,

Counins, L. R. (1974) —"“Monotremes and mar:
supials.” ¢Smithsonian: Washington, }

Guava, $8. F.. & Ripe, W. DBD. L. £1967) —Intra-
class Metatheria. #7) “The Fossil Record”.
(Geol. Soc. Lond: London.)

Laure, B. M, o.. & Ant, J, B, (1954)>—"List of
land mammals of New Guinea, Celebes and
adjacent islands !758-}952." (Tonbridge
Printers: Tonbridge. }

Littrcraven, fF. A. (1969),.—Lalest Cretaceous
mammale of upper part of Edmonton Forma-
tion of Alberta. Canada, and review of mar-
supial-placentzl dichotomy in mammalian
evolution. Paleent. Conir. Cuiv, Kans. Article
50 (Vertebrata 12), 1-122.

Pauays Coura, ©, bE (1952).—Fossil mammals
from the beginning of the Cenozoie in Brasil,
Marsupialja: Didelphidac. Aimer. Mus. Novit.
1567, 1-26,

Pau, Coura, C, ne | 1962)—Didelfideos fossiles
del Paleoveno de Brasil. Cienciay Zophigious
%, 135-166.

PauLa Coupo, C. pe (1970)—News on the fossil
marsupials from the Riochicun of Brazil. Ad.
stead. brast!, Clee. 42, 19-34.

Ree ©. A. (1957) —Sobre la posicion sistemuticu
ile "Zyyulestes paraneisis” Amegh, y de
“f“ysaless erirérrianus” Amegh. con una
reconsidetacion de Ja edad y cortelucion det
"Mesupotamiense”. Molmberzice 5, 209-226,

Rive, W. D. L. 11964) —Anlechinis rosamendae,
4 new species of dasyurid marsupial rom the
Pilbura District of Western Australia: with
remarks on the classification of A tiechinas.
Ww, Aast. Nat. 9, 58-65.

Riot, W. D. L. (1970) —"A eulde to the pative
manimals. of Australia.” (Oxford Univ. Pr:
Melbourne.)

Rostinson, R. (1966)—‘Abariginal myths and
legends,” (Sun Books: Melbourne.)

ScitoppE, R,, & Canasy, J. H. (1972) —The bio-
geogrophy of the Australo-Papuan bird and
mamnial faunas co relation to Torres Strvit,
In D. Walker (Fu.) “Bridge and barrier: the
natural and cultural history of Torres Strait?
{Ast Nat. Univ, publ, B6/3¢1972); Can-
berra,)

Stor, B. (19721.—La faunnvle de mammiferes du
Crétucé supérieur da Laguno Unsayo (Andes
péruviennes). Ball. Mus, natn, Uist, mat. (3)
99, 375-405.

Simpson, G. G. (1928),—American Eocene didel:
phids. Auner. Mus. Novit. 307, 1-7-

Simesow, G. G. (1929).—American Mesozoic
+ dig Mem. Peabody Mus, Yale 3, 1-

Stupson, GG. (1941)—The affinities of ihe Bor-
hyaenidue. Amer. Mfus. Navit. 118, 1-6.

Simpson, G, G, ¢1945) —The principles of classi-
fication and a classification of Mamitals. Rall,
Amer, Mus. nat, Mist, 85, iexvi, 1-339.

Simpson, G. G. (1947)—A new Eocene mar
supial from Brazil. Amer. Mus. Navit, 357,
1-7.

Smeson, G, G. (1948)—The beginalne of the
age of mammals in South America Part J. In-
troduction, Systematics: Marsupialia, Eden
tala, Condylarthra, Litopterna, and Notiopro-
vonia, Bull. Amer, Mus. nat, Fisr 94, 1-232.

Sineiam, W. J, (1906)-—Mammalia of the Santa
Cruz ‘beds. Marsupialia, Rep. Princeton
Exped. Patagonia 4, 333-460.

Suri, J. (1880), —"The Booandik Tribe of South
Australian Aborigines: «a sketch of their
habits, cusiome. legends, and language”
(Gove. Printer: Adelaide?

Smror, Ro A., Trpeorp, ROW. & Mitirr, A. A.
(£961)—Cenozoie stratigraphy and verte-
brute paleontology. of the Tirari Desert, South
Australia, Rec, S. Aust. Mus, t4, 19-61,

Stimtox, R. A. Teprorn, Ro A. & Woonpvenr,
M. QG. (1968),—Australian Tertiary deposits
containing letrestrial mammals, Bali, Dep
Geal. Unite, Calif, 77, P30.

Tuomas, ©. (1887)—On the homologies and
sucession of the teeth tn the Dasyuoridae
Phil. Troms. R. Sav, 178, 443-462.

Weooorcrne, M, ©, (1967).—The Alcodte Fuuna,
Central Australia: an integrated palcontalogi
cal and geological study. Bur. Min. Resor
Aust. Bull, 87, 1-187.

Weons, H. E. (1924),—The position of the
“sparussodonts'! with notes an the relation
ships and history of the Marsupialia. Mull,
Avior Maz Nut. Hist. $177,101.

SOUTHERN AUSTRALIAN SPECIES OF CHAMPIA AND CHYLOCLADIA
(RHODYMENIALES: RHODOPHYTA)

BY D. J. REEDMAN* AND H. B. S. WOMERSLEY

Summary

REEDMAN, D. J., & WOMERSLEY, H. B. S. (1976).-Southern Australian species of Champia
and Chylocladia (Rhodymeniales: Rhodophyta). Trans. R. Soc. S$. Aust. 100 (2), 75-104,

31 May 1976.

Five species of Champia are recognised on southern Australian coasts. C. viridis C. Ag. (including
C. tasmanica Harvey, C. oppositifolia J.Ag., and C. verticillata J.Ag.) resembles the type species
C. lumbricalis (L.) Desvaux from South Africa] in having numerous scattered longitudinal
filaments passing through the diaphragms of the thallus. C. insignis Lucas from Tasmania also has
scattered longitudinal filaments. Three other species, C. affinis (Hook. & Harv.) J. Ag. (including C.
obsolete. Harvey), C. zostericola Harvey (including var. arcuata Hook. & Harv. of C. affinis) and
C. parvula (C.Ag.) Harvey var. amphibolis var. noy., have only peripheral longitudinal filaments in
the thallus. Australian records of the South African C. compressa Harvey probably apply to juvenile
C. tasmanica or to a probably undescribed species on the N.S.W. coast.

One species of Chylocladia, C. grandis, is newly described. Other Australian species previously
referred to Chylocladia belong to Lomentaria or are relegated to synonyms.

SOUTHERN AUSTRALIAN SPECIES OF CHAMPIA AND CHYLOCLADIA
(RHODYMENIALES:; RHODOPHYTA)

by D. J. REEDMAN* and H. B. S$, WomersLey*

Summary

ReepmMan, D. J., & Womerstey, H. B. $. (1976).—Southern Australian species of Chanipia
and Chylocladia (Rhodymeniales: Rhodophyta). Trans. R. Soc, S. Aust, 100(2), 75-104,
31 May 1976.

Five species of Champia are recognised on southern Australian coasts. C. viridis C. Ag,
tincluding C, tasmanica Harvey, C. oppeasitifolia J,Ag,, and C, verticillaia J.Ag.) resembles
the lype species [C., lumbricalis (L.) Desvaux from South Africa] in having numerous scattered
longitudinal filaments passing through the diaphragms of the thallus. C, iasignis Lucas. from
Tasmania also has scattered longitudinal filaments. Three other species, C. affinis (Hook. &
Harv.) J. Ag. (including C. obsoleta Harvey), C. zostericela Harvey (including var, arewata
Hook. & Harv. of €. affinis) and C. parvula (C.Ag.) Harvey var. amphibolis vur. nov.
have only peripheral longitudinal filaments in the thallus. Australian records of the South
African C. contpressa Harvey probably apply to juvenile C. tesmanica or to a probably

undescribed species on the N.S.W. coast.

One species of Chylocladia, C. grandis, is newly described, Other Australian species pre-
viously teferzed to Chylocladia belong t Lomentaria or ate relegated to synonyms,

Totroduction

Champia Desvyaux is in general a well
characterised genus (Kylin 1956) of the family
Champiaceae Kuetzing (1843) (syn, Lomen-
tarinceae Naegeli 1847) of the Rhodymeniales
and includes ntimerous species from most
cousts of the world. Some 9 specics have been
credited to southern Australia, and ss with so
many Australian genera, taxonomic distinctions
ure uncertain and in some cases the carlicst
Valid names are not in current use.

Ciylocladia Greville has been credited with
several Australian species, all of which are
synonyms of species of Chanipia or other
genera. One new species of Ciiylocladia has,
however, been found in South Australian
waters,

The type species of Champia is C, luambri-
calis (L.) Desvaux (1808, p. 246), from the
Cape of Good Hope, South Africa, Although a
well-marked species, C. /uinbricalis has never
been studied in detail, and the generic concepts
of structure and reproduction are based largely
on the European C, parvula (C. Agardh) Har-
vey which has been investiguted most recently
by Bliding (1928) who reviews earlier studies.

The thallus construction and reproduction of
the type species do, however, appear to con-
form with those of C, parvula, and a brief
account of the type species is given below.
Champia is characterised by a multiaxial,
hollow but septate and thus segmented, usually
much-branched thallus which originates from
a ring of apical cells (or a ring plus several
central cells). These apical cells each cut off a
filament of cells which runs longitudinally
through the thallus, and from the peripheral
filaments the continuous cortical layer of cells
originates close to the apex. If a group of
central apical cells is present (as in the type
species), then longitudinal filaments also occur
in the central region throughout the “hollow”
thallus. The characteristic transverse dis-
phragms in the thallus are derived from the
Iongitudinal filament cells very close to the
apex, and originate either from alternate cells
or ones 2-3 cells apatt. Each longitudinal fili-
ment cell cuts off cells laterally in one trans-
verse plane, and these link up and divide fur-
ther to form the characteristic 1 cell thick dia-
phragm, the peripheral filaments. being adja-
cent to the cortex or sometimes separated by

* Department of Botuny, University of Adelaide, Adelaide, S. Aust 5000_

16 QD. J, REEDMAN & H, B.S. WOMERSLEY

one diaphraym cell. The longitudinal filament
cells lying between the diaphragms usually cut
off one (-3) gland cells, Outer cortical layers
may be formed, and in some species an inner
cortex ul chizoidal filaments develops,

Most species are much branched, and the
branches atise from the regions of the dia-
Phragms by development of 4 Ling of apical
cells (rom cortical cells of the parent branch.

Reproductively Champia is fairly distinctive,
The procarp consists of a +-celled carpogonial
branch borne on a supporting cell fa cortical
cell), together with a 2-celled auxiliary cell
branch, present before fertilisation. The cysto-
carps ure external and sub-spherical to urceo-
late. asiiolate, containing a carposporephyte
with 4 basal fusion cell and much branehed
gommablas! filaments with terminal carpo-
sporangia. The inner cells of the pericarp
become stellate and form a network (“tela
arachinoidea").

Spermatangia are cut off from mother cells
Uerived from the witter cortical cells, and may
cover extensive jreas of the bratiches.

Tetrasporangia develop by enlargement of
inner cortical vells. they are tetrahedrally
divided ynd vecur scattered over the branches.

Species of Chapin are moderately common
along most of the coast of southern Australia,
und sre commonly mentioned in ecologicul
uccounts, In general, however, they do not
characterise any communities, though C. affinis
may be common in shallow water om rock plat-
torms (Womersley 1948, p, 158),

Chylocladia Greville differs from Chrmpia
in that the cystucurps do not have an ostiole
and the carposporophyte consists of a large,
basal fusion celi giving rise directly ta carpo-
sporangia. The thallus construction of the two
genera ts similar.

CHAMPIA Desvaux

The strycture and reproduction of the type

species, Chamipia lumbricalls

Champia lanbericaliy (L.) Desvaux (1808,
p. 246) is bused on Ulva luiibeicaliy Linnaeus
(1771. p. 300) from the Cape of Good Hope.
The species is a distinct one, and material from
Camps Bay, Cape Vown, South Africa (CG.
Dieckmann, FLvilit973) has been studied to
check On ihe generic characteristics:

Cy lumbeicalig is a robust species forming
clumips with numerous axes to 15 ent high,
arising Cran an entangled, stoloniferous base.
The axes are 2-4 mm thick. tercte and linear,
with ovcasional basally constricied branches

with rounded apices. The diaphvagms are regu-
lar in position, $-1¢~1£) mim apart and largely
obscured below by the thick cortex.

The apices have a central group of 7-8 ini-
jials and peripheral ring of about 14 Imitials,
resiiting in Jengitudinal filaments passing
(hrough the inner part of each diaphragm as
well as 14-2!) wround the periphery. One com-
plete (rarely 2) and two part longitudinal fila-
ment cells occur between the diaphragms. bear-
jing |-3 gland cells. The corlex close to the
apex becomes 3.4 cells thick, with a dense
outer cortical layer of anticlmally clongate
cells. In old axes the cortex increnses to about
8 colls thick, and o weft of rhizoids also deve-
Jops as an internal layer’ to the cortex.

The reproductive organs occur on tufts of
short, adventitious, branchlets (5-10 mm long
und 1-1+ mm thick) formed on the upper hail
of the axes. Usually the tufts are dense, with
numerous, curved {concave acduxially) tertile
branchlets, but in some plaints cystocarps are
borne on single lateral branchlets,

Cystocarps are borne mainly on the adawial
(concave) side of the curved branchlets, uflen
with 2-4 grouped together, they are ovoid with
a small ostiole, i-2 mm in diameter ancl high.
The carpospurophyte arises from 4 basal fusion
cell, with a much-branched gonimohlast bear-
ing terminal carposporangia; subterminal cells
probably also mature into carposporangia after
Joss of the terminal ones. The inver cells of the
pericarp become stellate, forming a loose
tissue. and the outer wall is comparutively
thick. Curpogonial branches and early post-
fertilisation stages have not been observed.

Spermatia form a continuous covering all
around the branchlets, or sometimes luryely on
the adaxial sides, with the outer cortical cells
culling olf 2-4 elongate spermatangial mother
cells which then cut off several avoid spermu-
7aTVeaL,

Teteasporangia occur densely in the brunch-
leis and are transformed from most of the large
inner comical cells; they are slightly pyrifarm
to avoid, abeut 100. 2m long, and tetrahedrally
divided.

The above description of C. lumbricalix
agrees Well in cssential generic details wih that
of Bliding (1928, p. 5) for C. parvude. thuugh
the latter is much smaller find slendgrer, wilh &
much thinner cortex, and bus only penpheral
longitudinal filaments: ‘Uhere js thus ne reason
to doubt the generic concept of Champa as
secoyiised by Blidge and hy Rylin (1956, 7
346),

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADLA vt

Key to southern Australian species of Champia
1, Vhallus with longitudinal filaments scattered
thoouzh the diaphragms as well as pevipheral
filaments; branches usually lincar, Msally con-
stiicted or not, not or slightly constricted al the
cisphragms; branching irregular or distichous
Thallus with peripheral longitudinal filaments
enly; branches usually tapering to base and
apex. usuilly slightly to moderately constrieted
at the diuphragms; branching irregular or aise

2, Branching irrewular, often distant, branches
lieews, 14-) t-2! mm browd ......,
C, yirielis (p. 77)
2, Branching subdistichous, fairly regularly pia-
nale; branches 2-5 mm broad, with a hitsul
stender stalk ae, Cotnsizets bp B14
Thallus segments obscured; hooked branches
absent; Usually epilithic; cortical eclls cutung off,
near branch apices, Usually several small outer
cortical ces wt first around their margins, liter
becoming almost continuous over the thallus as
an outer layer and in older parts cutting off
further outer cortical cells; inner primary cor-
ticn] cells, which thus become obscured, are
ovoid, 14-2 times as lang as broad. and 20-30
ant broad oo... C. affinis (p. 82)
3. Thallus segments clearly defined throughout!
most of thallus; usually epiphytic on seagrasses
or larger algae; cortical cells each cuting off
usually only 1(-2) small cells fram their corners,
so that the single Jayer of large cortical cells
remains clearly defined (hroughoul most of the
thallus: cortical cells usually ungulur, 2-3(-4)
times as long as broad _ A
4. Branchlets 4-( mm. branches 1-2 mm, in
diameter; cortex essentially single layered
throughout; usually one compleie longitudinal
filament cell between diaphragms; ultimate
branchlets often hooked; usually epiphytic on
Posidonia, Amplibolis or Sarger algac
C. zastericola (p. 87)
4, Granchlets $/3-4% mm, branches 4-1 mm, in
diameters cores mostly single layered but 2-4
cells thick tm oldest axes; usually two. com-
plete longitudinal filament celly between dia-
phraems: ultimate branchlets usually linear.
rarely hooked; epiphytic on Applibalis -
Cc, paevula var. amphibatis (p. 91)

Champia viridis C, Agardh 1828: 115. Kuet-
zing 1849; 842,
hmimiatt viridis (C, Agardh) Trevisan [848:
Las.

nd

Champia tasmanice Harvey 1844a: 407, pl 19,
847: 78: be59. 307, J. Agardb 1852: 370,
1876; 206; 1879: 67, pl. 19, fis 10-12, De
Toni 190Ua; 79; 1800h: 563. Guiler 1952) 95,
Hooker & Harvey 1847; 402. King ef al., 1971;
122(%), Kuetzing 1549: Bél: 1868: 30, pl S4e.
Kylin (930: 79. Luvas 1909» 34; 1929n: $9;
1929b; 50 Lucas & Perrin 1947: 207. fig. 72,
May 19645: 362, Okamura 1904: 88, Reinhold
1897: 53; 1899:> 45. Shepherd & Womersley
1970: 1345 1971; 165, Sonder 1846: 177(7);
1853; @82> 1855: S518; 1880, 17, ‘Late 1kR2!
18. ‘Tisdull 1893) 506. Wilson 1892: JO.
Womersley 1950: 176; 196fir ISE.

Champia tasmanica var. gracilis Harvey 1863;
synop.; 27, Sonder 1880: 17 Tate 1&b2: 1K.
enidaldia tasmanica (Harvey) Trevisan 1848
108.

Champta onpositfolte J. Agarih 1901) 27. De
Toni 1924= 309, Kylin 1931: 29, pl. 16, fig. 47.
May 1965: 362,
Champta verticitlaw t Agardh 901; 24, De
Yoni 1924: 309. Kylig 1931: 29, pl 17. fig 39.
May, 1965; 362.
Champla compressa sensu Harvey 1843, synop,:
27 tat least in part—see below),

HGS |, 24-D, 10

Thalias (Fig. 10) usually with several main
axes from a. stoloniferous base, forming .a
dense, spreading tuft commonly 5-15¢-20) cm
high, moderately or slightly adhering to paper,
medium to dark tek er red-purple in colour.
Axes (1-12-3 min broad. usually with numer:
ous irregular branches in 24 orders, often sub-
Opposite, usually less than 1 cm (sometimes 2-4
em) apart, in older or grazed plants ofien ver-
licillately branched. All branches terete to
slightly compressed, (!-)1-24 mm_ broad,
linear to slightly curved. basally constricted
and with rounded upices. Diaphragms 4-1(-14)
nium spart, regular and usually conspicuous in
surface view of thallus. Corre of a single layer
of compact cells, polygonal and 25-40 pm
across in surface view, with an inner cortex of
rhizoidal filaments in older parts of robust
plants. Longitdinal filaments both peripheral
and central, with one complete and two pan
filament cells between each diaphragin,

Cystocarps scattered over young branches,
globular to urceolate, ostiolate, $- 1 mm int ia
meter. Apparently very few cystecarpic plants
have been collected.

Spermatangia forming a continuous layer
over beanchlets.

Terrasporangia scaliered in younk branches,
60—J 20 uni in diameter.

Type lecalin, W. Aust

Type. Herh, Agardh, LD, 26112.

Divrihation. Prom, Rottnest TL. W. Aust,
around southern Australia and Tasmunia to
Gabo f.. Vic., Usually on rough-watcer coasts or
in strong currents, fran shaded pools to 25 mM
deep, with a slender form on Pesidonia in more
sheltered waters,

The Lype specimen of C. viridis C, Ayardh
consists of 8 small branches on mica, und ts
identical with the later described Cy tasnanice
Harvey (type from Tasmania, n TCD), under
which name mast specimens of this taxon have
been known.

Fig.

TE

DB. J. REEDMAN & H. B, 8. WOMERSLEY

gop OOD a,
ie

750 um

Champia viridis. A, Longitudinal section of a branch apex showing the development of cortex,
longitudinal filaments and diaphragms (A42995), 8. Surface view of a branch apex showing
central and peripheral apical cells (A42991), C. Cross section of a young branch showing @
diaphragm with peripheral and scattered longitudi

mal filaments (A42991). D. Three dimen-
sional view of thallus showing diaphragms and longitudinal filaments with gland cells

(A42991). E, Longitudinal section of un older axis showing development of rhizoids from the
peripheral diaphragm cells (A3055()),

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADIA 79

Harvey (1863, synop. p. 27) described var,
gracilis of C, tasmaities, A stritable lectolype
is prabably Harvey's, Alg. Aust. Exsicc. 251,
tn TED, fran Port Phillip, Vic, and speci-
mens in MEL (45227) from Brighton. Port
Phillip, Vie. (Harvey, Trav. Set 483), named
var. fracta, are probably the same, These are
small, slender forms. often found on Posidonia
under moderate conditions of water movement
(e.g. Port Phillip, Vic., St Vincent Gulf, S.
Aust.) in carly summer, and dowbtfully justify
a yarietal name. The thallus is usually 4-10 cm
high, densely tufted with an entangled base,
irregularly branched, with occasional curved
branch apices, branches mostly 4-1 mm. broad
and segments #-1 times as fong as broad,
slightly constricted at the diaphragms. The cell
structure is very similar to larger forms typical
of the species, with cells about 40 pm broad,
(1-)2-3 times as long as broad, angular with
small cells cut off from the corners (about as
many small cells as parent cells). While ex-
tremes of this shelreted-water form and the
robust rough-water form appear relatively dis-
tinct, a good range of intergrades does occur.

C. verticilleta JT. Apardh (L901, p, 26) is
bused on a specimen (type in LD, 26078) from
Port Elliot, S. Aust. (Hussey) and ts an older
plant of C. viridis with verticillate branching,
and C. oppositifolia J. Agardh (1901, p. 27),
with type in LD (26148) ts « plant with some-
what more distinct opposite branching.

C. viridiy 1s a distinctive species in s7e,
form, and in having central as well as peri-
pheral longitudinal filaments. It varies con-
siderably im robustness and thickness of
branches and tn branching, with Irequent
occurrence of prolifcrous branches giving a
subverticillate arrangement. These variations
are cilher ecological ar duc ta the age of the
plant.

C. viridis js most closely related to the type
species. C. lumbricalis; from South Africa, but

i

Abbreviations used in Figures 1-9.

is distinct in being a less robust plant and pot
developing a cortex several cells thick,

A Preiss specimen in MEL (45206) is C,
zostericola, not C_ viridis, bat a small form of
the Intter does occur in Western Australta.

C. viridis (as C. taymanica) has been
recorded fram New Zealand by Naylor (1954,
p. 658), This recerd has nor been checked, but
may apply to the clasely related but distinct C.
novae-zelandiae Huoker & Hatvey, which has
central longitudinal filaments but a many
layered cortex,

STRUCTURE AND REPRODUCTION
Material studied: Nora Creina, 5. Aust, upper-
most sublittoral (Reedman, 12.31.1973; ADU,
A42995); Cape Lannes. 5. Aust., under ledges
(Reedman, 10,i1,1973; ADU A42991), and Staple-
ton Point, Prosser Bay, Tas., 8-12 m deep (Olsen,
71.¥i.1966; ADU, A30S50),
Thalluy development. The apex of a branch
(Fig. 14, B) includes both a central group of
12-16 initisls and a peripheral ring of initials,
producing longitidina! filaments passing
through the central as well as the peripheral
regiona of the diaphragms (Fig. 1€, D), as in
the type species. The number of central longi-
tudinal filaments is commonly greater than the
number of apical initials, due apparently to
division of the initials and consequent branch-
ing of the filaments and subsequent loss of
some initials and terrmnation of some filaments
at diaphragms, The apical initials of the peri-
pheral ring divide transversely, and within 2
or 3 cells of the apex divide periclinally form-
ing cortical cells which then divide..anticlinally
to form the single-layered cortex (Fig. LA).
Alternate cells of the longitudinal filaments cut
off cells laterally which join to form the single-
fayered diaphragms (Fig, 14), after which the
tongitudinal filament cells become very elon-
gate, The alternate cells of the longitudinal fila-
ments, lying between the diaphragms, form
1(=2) gland cells (Fig, 1). Only the cortical
tell formed directly from the longitudinal fila-

a — apical cell feb — fused carpogonial 0 — asliole

acl, — apical groupe branch per — pericarp

omc —-atuxiliary mother cel) = Fu — fusion cell re — thizoldal cell

uux — autiliary cell f1, 22 — gonimoblas! cells Sp = Speropatangnint

ch —curpogonialbranch ~— ot) — yonimolobe Pr spérwataogiel mother
ou outer cortical cell ald — gland cell aii — supporting cell

cort = — inner carlical cell | — longitudinal filament 1 — tela arachnoides

csp = — carposporangium le — tateral connecting uu — trichogyne

d — diaphragm ocll Ispp = — Letrasporangium

AG D. J. REEDMAN & H. B. S. WOMERSLEY

-)

oO
Gg

Ss
& cs
B®
0 08

oe)

ee Sg
Q80n &
By
ce Paty ag -
*~ Ba RD ETS
CSEOF ik

Fig. 2. Champia viridis, A. An early post-fertilisation stage showing supporting cell with auxiliary cel]

branch and carpogonial branch with fusions occurring between the cells. Considerable cortical
proliferation has occurred towards formation of the pericarp (A30550). B. A post-fertilisation
sige showing a connection between the fused carpogonial branch and the auxiliary cell. and
formation of the first gonimoblast initial (A30550). C. Development of tetrasporangial initials
{A30550). D. A mature tetrasporangium (A30850). Champia insignis. E. Surface view of a

branch apex showing initials (A12237). FL Section of male thallus showing development of

spermatangia (A12237).

mient cell is in pit-connection with it, In older
parts, s0me cortical cells may produce small
outer cells from their corners, hut the cortex
Temains essentially only one cell thick.

In older branches, rhizoidal cells develop
from peripheral diaphragm cells and form a
loose layer lining the inner side of the cortex
(Fig. 1D), as in the type species.

Branches arise from the region of the dia-
phragms, following development of u group of
apical initials from the cortical cells.

Procarp and carposporephyte

Cystocarpic specimens appear to be rare, und
only one female specimen with very young
carposporophytes has been available. The
carpogonial branch is 4-celled und borne on a

SOUTHERN AUSTRALIAN CITAMPIA AND CHYLOCLADIA Da)

supporting cell which also bears the auxiliary
mother cell with its auxiliary cell (Fig. 24).
Following fertilisation, the cells of the carpo-
gonial branch fuse and a connection between
the fused carpogonial branch ancl the auxiliary
cell is formed (Fig. 28), and the first gonimo-
blast cell ig cut off from the auxiliary cell,
Early post-fertilisation development is accom-
panied by division of the surrounding cortical
cells to form the pericarp (Fig. 243, and in
the One specimen (Stenhouse Bay, S. Aust,
7 m deep, Kraft, 18.tx.1973; ADU, Ad4564)
observed with cystocarps, the structure of the
pericarp and carposporophyte is typical of
Champia and very similar to the illustrations
of Bliding (1928) for © parvula.
Spermatangia

Spermatangial plants have pot been ob-
served.
Tetraxporartgia

Tetrasporangia develop hy enlargement of
cortical cells (Fig. 2€, 2) and are tetra-
hedrally divided, 60-120 jam in thameter when
mature, and scattered in younger brauches.

NOTES ON CHAMPIA COMPRESSA

Champia compres\a Harvey IS38: 402;
1847: 78, pl. 30) was first recorded from Aus-
tralia by Harvey (1863, synop,: 27) on the
basis of specimens from Western Australia
(Clifren, and Alg, Aust. Exsice, 250A) and
from Port Fairy, Vie. (Alg. Aust. Exsiec.
250D). Since then, C. cenipressa has been
recorded from Australia by the following
authors, probably largely on the basis of Har-
vev's records: Gatnet 1971: 96. Lucas 1909;
34, Lucas & Perrin (947: 206. Muy 1947;
275; 1965: 362. Sonder 1880; 17(2). Tisdall
1898: S506.

However, comparison with material of
Champia compressa from St James, Simons-
lown. S. Africu 0G. Dieckninn, 29.11.1973;
ADLI, Ad4601) shows that the Australian
reoords dlmast certyinly are not C. eompresse.
The South African species has strongly com-
pressest hranches with numerous central longi-
tudinal filaments scattered across the dia-
phragns. Harvey's Western Australian speci-
mens also have central longitudinal filaments
tnel the hrinches are only slightly compressed,
They agree in these features with C. viridis C.
Agardh, and are very similur lo young, well
displayed, specimens of this species Front else-
where in southern Austria, Harveys Alg.
Aust. Exsice. 250D (ti TCD) from Port Fairy,
Viz., includes C\ viridiy as well as some plants

which are not. 1 Champia, and one specimen of
250D in BM is C. zastericofa.

Other records of C. conipressa from south-
ern Australia probably apply to €. vridis if the
specimens have central bongitudinal filaments,
but the records of May (1947, p. 275, 1965,
p. 362) from N.S.W. apply to a separate spe-
cies which may be unnamed. This small, irri-
descent species, with branches attached to
others by haptera, does not have central longi-
tudinal filaments aim! is only slightly com-
pressed. It is thus distinct from both C, viridis
wnd C. campressa, and also differs from the
very. strongly flattened subtropical C. vieillirdii
Kuetzing, which from material from the Solo-
mon Islands (Womersley & Bailey 1970, p.
321) 1s so Strongly Mattened that the dia-
phragms are only 2+ cells across in the direc-
tion of flattening of the thallus, and Jongi-
tudinal filaments are almost entirely around the
periphery,

It is therefore considered that Chanzpla com-
presse does not occur on Australian coasts,
Other records of this species trom outside
South Africu also need checking; the record of
Weber van Bosse (1928, p, 477) from Borlieo
is probably C. wrefilardil, and that af Joly
(1965, p. 176) from Brazil prebably applies
to a different species.

Champia insignis Lucas 1931; 409, pl, 25. fig.
1, Guiler 1952: 94, Lucay & Perrin 1947:
207. May 1965: 362.

FIGS 2£, F, 11a

Thalls (Pig. 11.4) with one to several main
axes fo 18 ¢m high, arising from a small, slis-
coid te slightly lobed holdfast on. pebbles or
shells; branches of pyramidal form, with lower
laterals often similarly branched; thallus adher-
ing closely to paper, colour red-brown (her-
barium specimens) to “bright purple” (Lucas).

Axes 3-5 mm in diameter, subterete (possibly

slightly compressed), lincar, bearing alternate

or opposite laterals mostly $-| ci apart and
sub-distichously arranged along the ives, often
somewhat denuded towards the base. Main
jateral branches usually with a slender stalk
(Fig, 11.4), then broadening, linear or gently
lapering, usually 3-3 mm in diameter. with
rounded apex. Lesser branches similar but slen-
derer and shorter. Diaphragms apparent
throughout most of the thallus, 2-3(—+) min
apart in older parts, 1-2 mm apart in younger
branches which are slightly constricted at the
diaphragms. Cortex essentially one cell thick,
the cells subpolygonal in surface view, mostly

32 D. J, RFRDMAN & H, B, 8, WOMERSLEY

50-100 pm long and 35-50 ym broad, cutting
off 1-3 smull cells {rom theie outer corners and
more Dumerous such cells near the thallus hase.
Longitudinal filaments acattered (hroyghout the
diaphragms, consisting of several (7) cells be-
tween diaphragms,

Cywfocerps scattered over the — lesser
branches, conical to urccolate, osticlate, 2-11
mm in divnveter; catposporophyle branched
with lower sterile cells and terminal carpo-
sporangia.

Spermatangle cut off from cortical cells (Fig.
2#') and Jorming collarike patches on cither
side of the diaphragms of lesser branches.

Tetrasporangia scattered over the lesser
branches, tetrahedrally divided, 80-100 4m in
diameter,

Type locality. R, Derwent Estuary, Tas.
(“Sandy Bay, Hobart. Oct. 1925" on type
sheets, }

Lectotype. Herb, Lucas, NSW, 136559. Syn-
types (cf) im NSW (136558) and ADU
{A12237)_

Distribution. Qoly known from the type col-
lection and the following Tasmanian collec-
tions; D’Entrecastesux Channel, Nov, 1910
(NSW, 126561); Browns River, Lecus, Oct.
1923 (NSW, 136562); and Snug, Lucas, Aug.
1925 (NSW, 136560). Two specimens in the
BM, tabelled “Vas, Oldfield”, one numbered 81.
are probably alse from whe Derwent Estuary.
The species appedrs to be known only from,
or just south of, the Derwent Estuary,

Lucas did not specify type material, but the
one now selecied as jectotype is the cystocarpic
specimen illustrated by Lucas (1931, pl, 25,
fig, 13,

The above vescription is compiled from that
of Lucas (1931, p, 409) and study of the type
and other material in NSW and ADU, C. insiv-
nis bas apparently not been collected recently.
but it seems tu be a quite distinct species. It
resembles C. viridis in having peripheral and
central apical cells (Pig. 26) producing longi-
tudinal filaments scattered across the dia-
phragms, bul differs m form and dimensions
and in being essentially distichously branched.
Lucas (1931, p, 409) refers to the whole plant
as being “compressed”. It is clesirable, however,
thut Hiquid preserved collections should be stu-
died to confirm such sspects.

The BM specimens agree well with the type
cellection though the lateral branvhes do not
appear to be themselves. disttchously branched,

and the main breaches are basally constricted
but scarcely stalked; they are eystocarpic, In
calposporaphyte striigture and the ostiolate
cystocatp, C, insigniy appears to conform well
with Champia.

Champia affinis (Hooker & Harvey) J, Agardh
1876; 304. De Toni 1900a; 75, pl, 5, fig. 2;
1900b: $59; 1924: 307, Guiler 1952; 94,
Harvey 1835a: S545(7}: 1859: 307; 1863,
synop.: 27, Kylin 193]: 28, Lucas 1909: 34;
19299: 19; 1929h: $0, Reinbold 1897; 53;
1899: 45, Sonder 1980: 17. Tisdall 1898:
506. Wilson 1892: 180,

Chylocladia afftis Hooker & Harvey 1247;
402. Harvey 1847: 79, pl, 29(7),

Lonmentaria affinis (Hooker & Harvey) Kuet-
wing 1849: 863, J. Agardh 1452: 730. Sonder
1853; 693,

Gastroclonium affine (Hooker & Harvey) Kuet-
zing 1849: 866,

Eipianiadi kaliformis sensu. Harvey 1844b:

Champia ebsoleta Harvey 185%: 307; 1863,
aynop,; 27, J. Agavdh 1876: 394 De Toni
19004; 75, pl. 5, fig, 3; 1900b: 559: 1924: 307.
Guiler (952; 94, Kylin 1931: 28, pl. 1S, fig,
33, Liwas 1909: 34: 19293* 195 1929b: 49.
Lucas & Perrin 1947: 206. May 1965: 342,
Reinbok! 1898: 46. Sander 1880; 17. Wilson
1898. 506. Wamersley 1930; 176: 1966: 150.

FIGS 3, 4, 118, 12

Thallas (Pigs 11B, 124, B) erect, usually
415-40) cm high, with one to several main
axes from a smal) discoid holdfast, grey-red to
purple in colour, adhering to paper: usually
growing on rock or on Amphibolis, rarely on
Posidonia. Axes usually densely and irregularly
radially branched for 3 or 4 ofders, branches
of pyramidal form (more spreading in plants
on scagrasses), often denuded below: axes
I-3; mm, branches 4-14 mm, and lesser
branches 4-4 mm, in diameter, all branches
slightly ‘basally constricted and tapering to
rounded apices. Diaphraymy usuully fairly dis-
tinct in lesser branches, obscured on older
axes, (4-)1-14{-2) mimi apart (segments
(4—)1-14 times as long as broad}, thallus con-
stricted at diaphragms except on older axes,
Cortex of a layer of large sub-oveid cells,
(20-)25-40(-60) pm across, and a sparse
layer of outer small cells around margins af
inner cells in young branches (Fig. 37, 1), be-
coming more or less continuous on older paris
(Fig. 3G-1) and near bases of old plants 2—4
cells thick (Fig 32). Longitudinal filaments
usually confined to periphery of the dia-
phragins (Fig. 3C), rarely with 1 or 2 within
the periphery, usually with two (occasionally

Fig. 3-

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADIA §3

a

steizes, Dery. «odie
« Ge, grrr,
Z gs wos pe hase Face ws ui hy
a cagte ack iC i ata
Kary, i
i at ! I 72! |
en oa
“AL deeocoey ye
& aa 3 ¢ ‘i
ab

!
ta. ay
a

S,
Os > or V
(29S SOK Sp

= ae

Champia affinis. A. Longitudinal section of a branch apex showing the development of cortex,
longftudinaf filaments and diaphragms {A42994). B, Surface view of a branch apex showing
apical cells (A42994). C. cross section of a mature branch showing a diaphragm with peri-
pheral longitudinal filaments and small outer cortical cells (442994). D, Longitudinal section
of mature branch (A42993). E. Longitudinal section of an old axis showing several layers of
cortical cells (A42997). F. Surface cell pattern of type specimen (Gunn, in BM) 10 segments
from a branch apex. G. Ditio, 30 segments from a branch apex, A, Ditto for Harvey, Ale,
Aust. Exsicc. 2521 from Georgetown, Tas., as C. ebsoleta, 30 cells from o branch apex, 7,
Ditto for A42990, 10 segments from a branch apex,

Bs 1), J) REEDMAN & H. B. & WOMERSLEY

ohe or thfec) complete cells and two part cells
herween the diaphragms,

Cysiocarps sitigle, Scattered over lesser
branches, subspherical to urceolate, 1-14 mm
long, 4-14 mm in dlumeter, with a distinel
ostiole (Fig. 4€’).

Spermatangia (Fig, 4E) in patches arentid
branchlets. on either side of diaphragms, often
envering most of the segments.

Tetrasporangia scattered over hrattches,
60-80 wm in diameter (Fig. 4F).
hectetype locality, Georgelown, Tas

(Gina).

Leclatype, BM,

Distribution, From King Geotge Sound, W.
Aust, to Western Perl, Vic, and arountl Tus-
mania. Generally found in shallow water on
reel’s on rough-water cousts, rarcly epiphytic on
Pasidonia or yobust algae.

©. affinis was reported from New Zealand
byw Harvey (18556, p, 236} and recently by
Chapman & Dromegoole (1970, p. 145). There
wre also specimens in BM from New Zealand,
but while they appear closely related ta C’
affinity there aré differences in form, Derailed
comparisons of liquid-preserved material ate
needed to establish the relationships of the New
Zeuland plant,

Hooker & Harvey (1847, p. 402) based C,
affints on plants From Georgetown, and a cysie
carpic plant in BM has been selected as lecto-
type (Fig. 118). Other specimens in BM and
in TCD are syniypes. The type specimens are
Of loose, Spreading form and typical of plants
from moderately sheltered conditions: the dia-
phragms are nol vorspicuous and the wall has
an outer layer of small eclls which becomes
cohnunawous in older parts,

Some of these Georgetown specimens are
refered by Harvey to C, ehsoleta, which he
deseribed (1859, p. 307) on the hasis of
thatlas struchury and cystacarpic plants, refer-
ting to "Alg. Exsic. n. 252" snd the follawing
localities:

“HAL, Georgetown. Southport, Cy Saart,

DISTRIB, Port Fairy. Victoria. WoW."

‘Lhe critical specimens in TCD Include

|. “Georgetown. Sept. 1848. 252 0". This speci
men (Fig. 124) was previausly considered
the lecteivpe (by H.B.S.W-. itn 1952). but is
leirasporangial and hays no name on the
sheet; it is an old plant but is C. affiniy,
being very similar to the type of this species.
Another specimen labelled “Georgetown,

V.D,. 252 1" is also ain old, baltered speci-

men, probably of C. affinis,

2. “Southport, V.D.L. C. Staart’—four speci-
mens, two with "C, obyaleta” om the sheets,
ancl which ure C, alfviis. They are all letra-
sporungtal and do not match the type des-
criplion atall well,

3. “Port Fairy, Vic. WHA, 252 D"—five speci-
mens, 3 tetrasporangial and 2 cystocarpic:
these match the description well and include
the only cystocarpic specimens jin TCD. One
letrasporangial specimen has “Cvrampia oh-
solera”’ on the sheet.

On the hasis that the leclotype spevimen. of
C. obiseleia should have Harvey's number 252
on i and should also agree well with the type
description and be cystucanpic, one of the two
Part Pairy specimens (Fi, 128) is now
selected as the loctotype and the other Port
Fairy specimens are then syntypes.

In describing C, ohgaleta, Harvey (1859, p,
307) stated “perhaps only a variety of €-.
affints’, and Kylin (1931. p. 28) doubted that
they were distinc: species. Study of Harvey's
Georgetown material, the Port Fairy specimens
of C. obseleta, and Knowledge of this commen
species along southern Australian coasts, sug.
gests strongly that only one species is involved,
and C. ohsoleta is therefore reduced to syno-
nymy. C.. affintiy occurs mainly on rocks and
platforms at about low tide level, and under
rough-water conditions in such habitats. it is
of pyramidal form and grey-purple in culour,
us Harvey (1859, p. 307) noted. While nor-
mally epilithic, it o¢casionally occurs on other
robust algae or on the seagrass Aiphibalts,
and some of Harvey's Georgetowm specimens
were growing on Posidontec, While there appear
to be no structural differences between these
forins on seagrasses and the rough-water forms,
the former are more loosely branched anid of
more spreading habit.

The type specimen of {-. affinity shows num-
erous small outer cells cut off from the primary
cortical cells (Fig, 34, @), but not as many as
in rough-waler forms (Fig. 3/). However, the
habit, luck of clearly visible primary carlicul
celly and obscuring of the diaphragms in most
of the thallus, are features of C. affints as
understood here. and differentiate this species
from C. zextericala {see helow), The type of
C. affints and other specimens (e.g. in ADU)
from Georgetown in Vasmania, Pere Phillip
Heads in Victoria. and near American River
inict on Kangaroo Island, appesr to represent
relatively galm water forms of the species,

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADIA 83

and the type of C. absoleta to represent rough-
water forms of more pyramidal habit and with
more prominent outer cortical cell layers.

Some specimens of Chuampia, epiphytic on
Posidunie, resemble C. affiniy in that the seg-
ments aré nol distinct and a moderate number
of outer cortical cells are present, The segments
ure, however, offer distinctly jonger than in the
type of C. affinis, being 2—3(—4) mes as long
us bruad, The longitudinal filaments often con-
sist of 2-3 complete cells between the diar
phmgms, and the primary cortical cells are
muderately conspicuous. Most of the plants
with these charuclerislics afte stall and pos-
sibly young, though often fertile, For the pre-
sent they ate regarded as probably a form of
C. affnis, but further studies on their seasonal
growth and variation is needed. The specimens
concerned include: Port Lincoln, S. Aust,
4-12 m deep, on Posiionia {Sheplterd,
23.vill.1975; ADU, A46561, A46367, A46570),
and Pig L, American R, inlet, Kangaroo L., S.
Aust. (Womersley, 17.1,1947; ADU, A4467),

Cc. affints differs from the other common
southern Australian specics, C. zestericela, in
its habit, lack of hooked branches, and thick

2-3 lityered) cortex which obscures both the
diuphragms and the Jarge primary cortical cells,
The latter species is discussed further below,
bul very oceasonal plants with intermediate
characters do occur. Harvey's Alg, Aust.
Exsice. 253H, distributed as C. effinis, is typical
C. zostericold, and this has led to considerable
confusion,

J. Agardh (1876, p. 304) distinguished two
varieties of C, affinis (var, u affinis and var. B
intermedia) on the proportions of the articuly-
tions sod density of letrasporangia. Both these
features wre unsatisfactory characters to separ-
ate varicties, and Kylin (1931, p- 28) con-
sidered var, intermedia as intermediate between
C. affia’s and C. ebyeleta. These varieties do
hot uppear worth dislingttishing from the ape-
cits.

Hooker & Harvey (1847, p. 4021, following
description of C. «ffimis, also described var,
acvote, This varicty is considefed specifically
distinct and fs pelegated below to the synonymy
of C, 2estericola Harvey,

STRUCTURE AND REPRODUCTION

Material stidjed: Cape Linnes, S. Aust. low
eqlittoral (Reena, WGi1973; ADU, Aa2g4s
am! 127.1973; ADU, A42990); Nora Creina, §&
Aust, lower enlittornl (eed, 12,7),1973;
ADU, A42994): sud Pennington Bay, Kangaioo
1, S. Aust, low evlitioral on reef (Reedman
13.i¥,1973; ADU, A42997).

Thellus development

There ure 12-16 apical Initials (Fig. 32)
which form 4 peripherul cing of longitudinal
laments; only very oceasionally have filaments
been seen within the periphery of the dia-
phragms. Whe inittaly segment (Fig, 34) as do
the peripheral apical initials in C, weridis, bul
the cortical layer of large cells cuts off small
outer cells, at first around their outer margins,
bul a more or less continuous layer of squall
tells occurs on mature parts (Fig. 3G J), and
near the base (especially in older plants) a
corlex several ceils thick is developed (Fig.
3£)- Hairs are commonly formed from outer
corlical cells near branch apices. The dia-
phragms are fermed usually by every third or
fourth cell of the Jongitudinal filaments (Fig.
34, D), leaving usually two or three complete
cells between the diaphragms, each cell com-
monly bearing a gland cell. Rhizoidul develop-
ment within the cortical layer has aol been
observed.

Branches arise from the region of the cia-
pPowagms, where a ring of outer curtiéal Cells
becomes meristematic and forms the apical ifi-
uals of the branch. Branching cecurs irregularly
and often densely on all sides

Procarp and carposporophyte

The supporting cell (Fig. 44) is uw lorge
cortical cell in primary pil-connection with a
longitudinal filament, and is generally attached
opposite a gland cell. The eystecarps thus lie
between the diaphragms of a branch, and
cystocarpic plants are common. The support-
ine cell ix multinucleate and densely cytoplas-
mic, and cuts off a 4-celled. curved, curpo-
gonial heanch (Fig, 44), of which the first cell
is binucleate and the other three uninucleate,

The supperting cell alse produces a multi-
nugleate, densely cytoplasmic auxiliary mother
cell (Fig. 4A). which produces q unjrucleate
auniliacy cell just prior to fertilisation.

Fellowing fertilisution, the pit-connections
of the carpogonial branch cells enlarge [Fig.
48) and the cells tend to fuse. Following pre-
sumed diploidisation of the auxsliary cell, first
and then second gonimoblast cells are pro-
duced, and the latter divides further to produce
a cluster of branched zgonimollast filaments
tFip. 4C, DB). which terminate in uninucleate,
ovoid carposporangia, The catposporangia
mature simultancously, byt a new gonimolobe
sammonly develops from the basal cell of the
gouimoblasc and produces a secondary, later
maturing. amalfer cluster of carposporangia.

84 D. I, REEDMAN & H. B. 5S. WOMERSLEY

AB

a 50 um

100 gan

50 um

Fig. 4. Champia affinis. A. Supporting cell With auxiliary mother cell and carpogonial branch
(A42990), B, Post-fertilisation stage with cells of carpogonial branch fusing (A42990). C,
Section of an immature cystocarp showing development of carposporophyte from old auxiliary
cell (A42997); D. A mature carposperophyte showing much-branched gonimoblast with ter-
minal curposporangia, surrounded by cells of the “tela arachnuidea” (A42990), E. Section of
male thallus showing development of spermatangia (A42997), F, Section with a mature tetra-

sporangium (A42993).

Vegetative cells adjacent to the auxiliary
mother cell becomes densely cytoplasmic, and
assist. nutrition of the developing carposporo-
phyte. Pit-connections between the lower ceils
of the gonimoblast enlarge considerably but the
cells do not fuse completely.

As the gonimoblast develops, vegetative cells
around its base divide to produce the pericarp.
Inner cells of this form the cell reticulum, with
aly outer wall several cells thick (Fig. 4C).
The mature cvystovarp has a well-defined
astiolo.

Sperinatangia

Spermatangial mother cells are cut off from
the small outer cortical cells to form. a con
tinuous Jayer over the branches, and each cuts
off 2-3 ovoid spermatangia (Fig. 4E). Usually
the entire spermatangium is shed.

Tetrasporangia
Tetrasporangia (Fig, 4F) develop by
enlargement of inner cortical cells which deve-

lop several secondary pit-connections with wi-
jacent cells. They are tetrahedrally divided,

SOUTHERN AUSTRALIAN CHAMP/IA AND CHYLOCLADIA 87

wilh it thick gelatinuus sheath, arnct usually
densely scattered over the branches.

Champia zostericola (Harvey) comb. nov.

Lomentaria, zostéricola Haivey 1855a;
1863, synop,: 26, J, Agardh (876: 632.

Gastrocloniunt (7) zestericolum (Harvey) De
Toni 190th: 567,

Chu teieladtat tostericoly (Aurvey) Kylin 1931:

545,

Chylocludia affinis var, arcwaia Hocker &
Harvey 1847: 402. Womersley 1966: 150.
Lomentaria affinis sensu Kueizing 1865;
86d-£.. Sonder 1855: 323,

Champia affinis sensu King et al. 1971: 122.
Lucas, & Perrin 1947: 206, fig. 71, May 1965:
362, Shepherd & Womersley IS71l; 163.
Womersley 1950: 176.

Champia parvula sensu Harvey 1855a: 548 (ia
part).

3h, pl.

FIGS 5, 6, 12€, 13

Thallus (Figs 12C, 13) usually spreading
and forming irregular clumps with several
branches from the entangled base, sometimes
with one or more erect uxes and spreading
laterals, commonly 6 to 20 cm high, grey-red
te red in colour, normally epiphytic en the sea-
grasses Pesidonta and Amphibolis or on larece
algae, possibly on rock: attachment at first by
means of a small discoid holdfast with one to
several axes, luter altaching by small adventi-
tious multicellular pads to the seagrass or itself.
Axes usually 14-3(-5) mm in diameter,
branches only slightly slenderer and ullimate
branches 4-2 mm in diameter, hranches
slightly basally constricied and with rounded
apices; Mature planis normally with some to
many branches ending in recurved ("“hooked’’)
tips (Fig, 13); young branches distinctly con-
stricted at diaphragms, segments mostly (4-}
1-14 times as long as broad. Diaphragms dis-
tinct except in aldest parts of some plants,
4—1+(—2) mm apart, Cortex single layered, of
relatively large ajgular cells [40-60(—75) pm
xeress und mostly {1/4—-)2-4 times as long as
broad] which are usually arranged more or Jess
in longitudinal rows (Fig. 5E-H). Near the
apices, each cell usually cuts off a simele, rela-
lively small cell from oear a corner, and fur-
ther such cells develop on older parts; how-
ever, the essentially single layered cortex of
large cells is maintained throughout most of
the plant (Fig. 5f, A). Longitudinal filaments
canfined to periphery of the diaphragms, rarely
with odd ones more centrally platsl, with
generally one complete cell and two part cells
between the diaphragms (Fiz. 5C),

Cysiecurps subspherical lo slightly cenical,
base hroad and slightly constricted, 2-1 mm in
diameter,

Spermatangia scattered over smaller branch-
tots, as extensive patches of collardike sore
around the diaphragms.

Tetrusporangia scattered over branches,
60-100 »m in diameter.

Type locality. Rottnest 1, W, Aust.

Type. TCD (Harvey, Tray. Set 195),

Distribution, From the Abrolhos Islands, W,
“Aust. urdund southetn Australia lo Kiama,
N.S.W. and around Tasmania. Generally epi-
phytic on seagrasses or other algae, from low
tide level ta 41 m deep, generally under slight
ta moderate water movement.

C. zostericola is based on small plants 3-1
cm high, growing on Pesidenia (not Zostera},
The type is No. 195 in Harvey's “Travelling
Set”, and his Alg. Aust. Exsicc, 294A (MEL
45197) from Fremantle (Fix [2C) ts very
similar. Harvey was in the vicinity of Fre-
mantle: from April to June 1854, and during
this period the plants are young (though often
fertile), but may not show the typical hooked
branches. Later in the year’, especially in spring
and early summer (September to November)
the planis reach 20 cm in height and nearly all
plants develop the hooked branches.

The lectorype (Fiz, 13.4) of var. arcuate
Hooker & Harvey of C, affinis has been
sclected from several specimens in BM, It ts
a well-developed specimen, attached to stems
of Heterezastera (7) and with numerous
hooked branches. Whereas the type of C.
zastericola isa young, small plant, thal of var,
arcudta is an older, larger plant of the same
&pecies,

The gencral confusion betwecn C_ affiris und
C, zostericul@ (or C. affinis var, crenata) is
probably largely due to Harvey in his Alg.
Ausl, Exsice. listing 233H from Western Port,
Vic. as C. affinis, whereas these specimens are
typical C. zosteritola.

Tn contrast to C_ affinis which is usually epi-
lithic on rough-water coasts and only occa-
sionally occurs on robust algae or on sea-
grusses, C. sestericole Is a common epiphyte
ou FPosidonie and on some targer algae,
usually in conditions of slight to moderatu
water movement and extending into deeper
water.

The presence of hooked branches, the
clearly septale rhallus almost throughout, the
essentiaily one cell thick cortex throughout the

38 DB, J. REEDMAN & H, BH, S. WOMERSITY

Fig. 5. Champia sostericala, A. Surface view of apax of branch showing apical cells (A43556), A,
Cross séciion of branch showing coytex, diaphragm and peripheral longitudinal filaments
(A43556). €. Longitudinal section of mature branch showing single layered cortex, diaphragms
and longitudinal filaments with gland cells (443556), D. Longitudinal section of an old axe
showing slight development of small outer cells (A8944). Z. Surface cell pattern of Harvey
Alg. Aust. Exsice. 249A (MFI.. 45197), (0 cells from branch apex. F, Ditto, 30 cells from a
branch apex. G, Surface cell pattern of the type specimen in BM of C. affinis. var, arcunia,
1G cells from a branch apex, H. Ditto, 30 cells from a branch apex.

plant with few small cells tying largely between
the primary ones, and the cortical cell dimen-
sions and arrangement. characterise this spe-
cles, bur occasiunal plants occur without
hooked branches. While most mature plants
have several Vague axes from their entangled
bases, some (from Tasmania and Port Phillip
Heads in particular) do have well developed
main axes with abundant laterals. Such plants
commonly (but not always) have hooked
branches and have the single-layered, large-
celled cortex typical of C zostericola. Howe
ever, the number of small cortical cells cut off

from the larger ones does vary somewhat, and
very occasionally plants intermediate in this
respect with C. affinis are found (see under
C. affinix),

C. zeslericela shows the yariation in form
which occurs in many other algal species dis-
tributed along all of southern Australia, i.e.
the western specimens are generally smaller
and less robust, and in the east, especially near
Port Phillip Heads.and in Tasmania, larger and
more robust plunts oecut.

Varition in diameter of the branches and
uXes (§ considerable, probably largely depend-

SOUTHERN AUSTRALIAN CHAMP/4 AND CHYLOCLADIA 89

ent on age, bul branches at normally over
1 mm thick, However, new growth on older
denuded branches may be slander and only
about + mm thick, as shown on several col-
lections from Pearson I., S. Aust. in ADU.

Many references to Champia affiniy apply, at
least in part, to C. zostericole rather than to
true C, affiniz (sec above), Probably most Aus-
tratian records of C, purvula also apply to
young plants of C. zostfericofa, though some
may apply to slender C. viricdiy.

STRUCTURE AND REPRODUCTION
Materia! sindied: Point Peron, W. Aust. drift
(Garden, 15,xi,1968; ADU, A34256); Port Noar-
lunga, S. Aust, 6-7 ni deep oa jetty piles aha-
son, 1941,1973; ADU, A43596); Marino, S. Aust.,
dritt (Homerstey, 26.x.1975; ADU, A46646).
Thallas development

There ate 14-20 apicil initials (Fig. 5A)
which form a peripheral ring of longitudinal
filaments (Fig, 5). with only occasionally an
odd inner one, The initials segment as in C-
offinie and a single layer of large cortical cells
ix formed, artanged more or Jess in longitudinal
lines (Fig. SE-H). Fairly near the apices, these
corticul cells become angular and cul off from
a comer a smaller cell, which remains essen-
lially in the layer of Lirger cells (Fig, SE-H}.
The smaller cells are at first simifar in number
to the larger primary cortical] cells, but later
more may be formed; however, the cortex
remains essentially only one cell thick throug-
out most of the thallus (Fig, 5C, D), The dia-
pheagms are formed usually from alternate
longitidinal filament cells, so that there is one
complete longitudinal filament (with a gland
ecll) and two part ones between successive
diaphragms (Fig. 5C), The relatively thin,
essentially single layered cortex results in the
ptimary cortical cells being visible throughout
most of the plant, and the diaphragms are a)so
conspicuous, Rhizoidal devclopment within the
cortex does not occur.

Branches arise from the region of the dia-
phragnis, with their apical cells differentiating
from the cortical cells. Near the base of en-
tangled thalli, small branches may develop into
haustorial pads of tissue and attach to other
branches or to the host,

Many of the branch apices are curved or
hooked (Figs 138, 144), and in some cases
these aid in attachment, The only structural
diflerence in hooked branch ends appears to be
that on the convex side cach sczment has 4
greater number of cells than on the concave
side,

Procerp and curposporophyte

From the limited female material available,
the supporting cell of the procarp appears to
be a large cortical cell, which produces the
Carpogonial and aéxihary cell branches (Fig,
64). Following fertilisation, the pit-connec-
lions of the carpogonial branch cells enlarge
(Fig. 68) and the cells Fuse, with a connection
forming from the old carpogonium to the auxi-
liary cells (Fig. 6C).

The diploidised auxiliary cell cuts off a first
gonimoblast cell which divides again (Fig.
6P) to initiate several branched gonimoblast
filaments with the ntature filaments terminating
in single carposporangia (Fig. 6£) which
mature simultaneously. New gonimolobes are
produced from the base of the gonimeblast and
mature Inter, Some darkly-staining and possibly
nutritive cells occur around the base of the old
auxiliary mother cell.

At an early stage in development of the pro»
carp. cells are cut off from the surrounding
cortical cells te form the projective pericarp
(Fig. 6D). The inner cells of the pericarp form
wz reliculum (the “tela aruchnonea™) which is
gradually absorbed by the developing carpo-
sporophyte (Fig. 6£), and the outer 2-3 layers
remain as the cystocarp wall, with a distinct
apical ostiole.

Spermatangia

Spermatangia are formed as in other species,
with small cells being cut off around the mar-
ging of the cortical ceils and then producing
branched chains of spermatangial mother cells
over the surface, froni whieh the clougate
spermatangia develop.

Tetrasporangia

The tetrasporangia develop by enlargement
of cortical cells (Fig. GF) which protrude
within the cortical layer, and they divide tetra-
hedrally (Fig. 6G).

Champia parvula (C. Agardh) Harvey 1853:
76. J. Agardh 1876; 303. De Toni 1900hb;
358, Newton 1931: 439, fig, 263. Gayral
1966: 485, pl. 134.

Chondria pervela C. Agardh 1824: 107,

Chylocladia parvala (C. Agardh) Hooker. Hur-

wey 1849; pl. 210,

Type loeality. Gades (Cadiz), Spain.

Type. Herb. Agardh, LD, 26(122,

Disizibution. C, parvula appears lo be the
only species of Chumpia known from Enropean
coasts, and hus been recarded from mast tem-
pefate and tropical coasts of the warld.

20 D. J. REEDMAN & H: B, S. WOMERSLEY

A

00 am

Fig. 6, Champia zostericola, A. Supporting cell with auxiliary cell branch and carpogonial branch
(443356). B, An early post-fertilisation. stage showing fusions between cells of carpogonial
branch (A43556). C. Post-fertilisation stage with fused carpogonial branch and connections to
auxiliary cell (A43556). D. Young gonimoblast within developing pericarp (A43556), E, Older
carposporophyte with terminal carposporangia, within pericarp (ostiole not in section)
(A43556). F. A young tetrasporangium (A34256). G. Mature tetrasporangium (A34256).

SQUTHERN AUSTRALIAN CHAMPIA AND CITYLOCLADIA $1

The following references credit C. parvaeda
to southern Australia, but probably all apply
to Other species, mainly (0 small specimens of
C. zostericola; ii Most cases it is Hot possible
to clarify these references

Guiler 1952; 94. Harvey 18552:

307. Lucas 1909: 34; 1929a: 19;

Lucas & Perrin 1947: 206, fig.

362. Reinbold 1897; 43. Souder” 1846: 176;

(S80; 17. Tate L832; 18, Tisdall ¢898; Sf,

Wilson 1892: 180.

C. parvula is generally recognised as a rela-
lively small and variahle species, and herbar-
ium specimens credited to it vary in size,
degree of branching, proportions of the seg-
ments and distinctness of the diaphragms, and
in the size of the cortical cells and degree of
outlet cortical development. Harvey (1849, pl-
210; (853, p. 76) commented on the variability
of this species, In comparisons with southern
Australian taxa, liquid preserved material from
He Verte, Rascoff, France (/. Feldmann,
14.2.1974; ADU, A46057) has been taken us
Tepresenmting the species in western Europe.
This specimen has fonger segments. (about as
long as broad) than shown by Gayral (1966,
pl, #34) bul distinctly shorter than illustrated
by Harvey (1849, pl. 210), and the dia-
phragms snd cortical cells are distinct through-
out the plant, with relatively slight development
of small outer cells. There are usually two com-
plete lonintuditial filament cells belween ihe
diaphragms, and the fongitudinal filaments are
confined to the periphery of the diaphragms.
The most detailed account of C. parvule is by
Bliding (1928) who studied material from
Woods Hole, ULS.A. Bliding’s deseriptiog and
illustrations appear to agree with the Le Verte
spectinen-.

In spite of the several references to C. par-
vale in southern Australia, it now seems clear
thal typical forms of this species do not occur
here. Young and small! plants of C. zoverfcole
do show some similarities, but are generally
broader, more robust, and when mature have
numerous hooked branches. Also. €. zosteri-
cola his usually only one complete longitudinal
filament cvll between the diaphragms, and the
cortical cells are lurger, Some small forms of
C. affinity also approach C. parvula but can be
Wistiguished on their greater outer cortical
development. Harvey's specimens referred to
C. pavvula appear to be slender forms of e:ther
C. zosterivola ar C, viridis.

However, a distinctive taxon occurs epi-
phytic on Amiphibolis at Tipaca reef in Spencer
Gulf, South Australia, and it appoars best to

S45, 1859;
1929b: SO,
70. May 1/9658:

designate this as a distinct variety of C. par-
vala, ta which it seents more closely allied than
to the larger €. gostericola. Fulure stuches may
show that tt should be recoynised as 4 distinct
species.

Champia parvula var. amphibolis vur. nov.
FIGS 7, 144

Thallus (Fig. 144) erect, spreading, 3-11
cm high, with one to several much-branched
Maiti axes arising from a small discoid hold-
fast on stems of Amphibolis, red to red brown
in colour, adhering to paper; oceasional attach-
ments by haustorial pads occur, Aves densely
and iegularly radially branched to 3 or 4
orders. with alternate, opposite or necamonally
whorled branches: axes 1-14 mm in diameter
below, tapering gradually to branchlets 13-4
mm in divmeter. Young branches slightly com-
stricted at «liaphragms, segments 1-14 times as
lung as broad, branch ends usually straight bud
rarely hooked, apices rounded. Dlapiiragens dis-
tinct throughout mosc of the thallus, somewhal
obscured near bases of older plants, Cortex of
a layer of angular cells 25-40(-50) pm across
and ¢1-)2-3(—4) times as long as broad, with
small cells cut off from their corners, and on
older axes developing a continuous guter cor.
lical Jayer (Fig, 7D) which in ol! plants may
be 3 cells thick. Longimdinal filaments con-
ffned to periphery of diaphragms, Ueveloped
from 10-15 apical cells, usually with evo (-3)
complete cells and two part cells between the
diaphragms (Fig. 7C).

Cystucarps single, subspherical to urceolate,
scaltered over younger branches, 27-4 mm
long and 4-1 mm in diameter, ostiolate.

Spermarengia Lorming son Over several seg.
ments near the apices of young branches,

Tetrayporangia scattered in young branches,
75-120 um in diameter,

Thallus erectus, effusus, 3-11 cm ialtus ex hap-
tere parvo discoideo in Amphibole. Axes irregu-
eulanter ramesi, 1-14 mm in diametro, ramuli
1/3-4 mm in diametro; segmenta I-14 plo
longlora quam fata, Diaphragmota conspicua nisi
prope bascs plantarum veterum, Cortex compositus
cellularum angulosarum 25—40(—350) #m Jalarum,
(1-)2-3(-4) plo longiorum quarm Jataram, parvas
cellulas in angulis ferens, ad 3 cellulas crassus in
parlibus veteribus crescens, Filamenta longitudin-
Alia tantum in margine, 2(-3) cellulas tolas inter
diaphraymata habentia. Cystocarpiy subgloboss vel
ureealjita, dispersa, ]_1} mm longa et + J mm itt
diametro, ostiolata, Spermatangla in soris fasci-
culatu prope apices ramulorum, Tetrasporansia
75-120 em in diametro dispersa.

Tig. 7,

D, J. REEDMAN & H. B. 8. WOMERSLEY

100 um

Champia parvula var. anphibolis, A. Surface view of branch showing apical cells (A41276).
B. Cross section of brunch showing diaphragm, cortex and peripheral longitudinal filaments
(A41276). C, Longitudinal section of a branch showing single-layered cortex, diaphragms, and
longitudinal filaments with gland ceils (A4J276). D. Longitudinal section of older axis show-
ing two (—3) layered coriex (A37291), E. Post-fertilisation stage showing fusions between
carpogonial branch cells and connection to auxiliary cell (A38255). F. Young gonimoblast
(A38255). G. Mature cystocarp {ostiole not in section) with carposporophyte bearing terminal
carposporangia (A38255). H. Section of male thallus with spermatangia (A41276). I. Mature
tetrasporangium (A37291).

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADIA o3

Type locality, Tipara reef, Spencer Gulf, 5,
Aust. (Shepherd, 23,xi.1970)-

Type, ADU, A38255.

Distdibution. Knows from several collections
{rom Tipara reef, on Amphibelis antarctica
and A, griffithii, 5-11 ov deep. Young plants
occur ih June, maturing to bushy plants up to
11 cm high in December.

Var. amphibelis resembles the more typical
forms of Champia parvila in being a relatively
slender plant, mostly irregularly alternately
hanched, with diaphrapms distinct throughout
most of the plant ancl segments as long to
slightly longer than broad, with the primary
conical layer of cells cutting off relatively Few
outer cells (except in old parts}, and with
usually two complete longitudinal filament cells
between the diaphragms. Reproductively it ts
similar to the account of Bliding (1928) and
at least superticially to the Tle Verte material.

It differs from European forms in being slen-
derer than some, more profusely branched, and
in growing on Amphibolly as erect tufted
plants. Future calleclions may show that it is
not confined to this sea-prass and some speci-
mens of Harvey (in TCD & MEI.) from Fre-
mantle and King George Sound, W. Aust. (e.2.
Alz Aust. Exsice. 254B in MEL, 45307) need
careful comparison with this variety and with
C, zortericola,

This Australian variety shows slight simi-
larity to C. zestericola but is very much slen-
derer than most plants of the latter, only tarcly
has hooked branches, has longer segments close
10 the apices and has two complete longitudinal
filament cells between the diaphragms rather
than the usual one in C, zestericola. C. zosteri-
cola docs occur on Amphibolis, but is more
churacteristically associated with Postdenia.

STRUCTURE AND REPRODUCTION

Materia! studied: Tipara reef, Spencer Gulf, S.
Aust, on Amphibelis, Shepherd. The type, Ad1276
(13.xi1,197L) and AS72940 (30.14.1970),

Thallus development

The thallus has 10-15 apical cells (Pig. 7.4)
which segment us in other species to form a
peripheral ring of longitudinal filaments (Fig.
78). with diaphragms forming usually from
each third cell and thus with two (occasianally
three) complete Jongitudinal filament cells
{usually each with w gland cell} belween suc-
cessive diaphragms (Fig 7C). The cortex is
one cell itnck (Fig. 7C} except on older axes

(Fiz. 7D) and the elongate cortical cells cut
off, from their corners, small cells (nsually
only one per cell) which lic more er less in the
layer of primary cortical cells. These small cells
commonly bear a hair in young branches, but
such hairs are lost from older branches.

On old axes, the primary cortical cells cut
off an ovter, continuous, layer of cells (Fig,
7H) and this may become two or three Cells
thick, the outermost layer being of small cells,

Branching occurs from the regions of the
diaphragms, Mast branches are linear to che
apex, but occasionally a curved or slightly
“hooked” branch -end occurs.

The holdfast remains small and ciscoid, but
several new axes may develop from it, Small,
pad-like haustoria also develop from branches
of same plants, attuching mainly te other
branches.

Precarp and carpasporaphyte

The multinucleate supporting cell develops
from a cortical cell and hears hoth carpogonial
rnd auxiliary cell branches as in other species.
Following fertilisation, {he carpogonial branch
cells begin to fuse (Fig. 7E) and a connection
forms between the fertilised carpogonium and
the auxiliary cell. The diploidised auxiliary cell
forms 4 first ponimoblast cell which divides
again (Fig, 7E), and the upper cell forms the
branched gonimoblast filaments (Fig. 7G)
which bear terminal catposporangia.

Following fertilisation, some vegetative cells
arqund the supporting cell become darkly-
staining, apparently as nutritive cells, Other
Vegetative cells divide to form the pericarp
(Fic. 7G) which develops as in other species,
with the inner eclls forming the “tela arach-
noidea’ which is broken or absorbed by the
developing carposperephyte, A well-defined
ostiole occurs at maturity of the cystocarp,

Sperimatangia

The small ouler cortical cells, or Further cells
cut off from the primury cortical cella, divide
to form branched filaments of cells covering
the surface of one to several segments close to
the apices of young branchlets, Each cell of
these filaments functions as a spermatangial
mother cell which cuts off outwardly 2-3
clongate-ovold spermatangia (Fig. 7H), which
appear to be shed entire.
T etrasporangia

Tetrasporangia (Fig. 7/) develop within
cortical cells which enlarge greatly ane holge
within the segments of the thallus.

54 D. J. REEDMAN & H, B. S. WOMERSLEY

CHYLOCLADIA Greville

Chivlocladia Greville (in Hooker 1833, p,
297). with the type species C, kuliformuis, is
conserved over Kalifermis Stackhouse 1809,
and is distinguished from Cfiampia by the for-
mation of carposporangia directly from a large
basal tusion cell (without branched ponimo-
blast filaments as in Champia) and by the
eystocurps being non-ostiolate. Otherwise,
Chyloctadia is similar structurally to Champia.

While numerous Australian taxa have at
sume lime been referred to Ciivloctadia, most
have been placed in uther genera (see Kylin
1931) or can now (see below) be excluded
from Chvloclacia.

However, 3 very distinctive species of Chy/o-
eladia, known From only a few deep-water col-
lsctions, has recently been discovered in South
Austcalian walers.

Chyloctadia grandis sp, fov.
FIGS & S$, 14B

Thallus (Fiz. 148) crect, red-brown to red-
purple, 20-30-cm high, with one to several axes
urising from a hard, branched, perennial base
to 5 ¢m high and 1/3-¢ cm thick, attached to
rock by a discoid holdfast to 14 cm across,
Axes with opposite or usually whorled lateral
branches ta 10 cm Jong, similarly branched
{muinly oppositely or alternately) to a second
or third order; axes often denuded below, 2-7
mm in diameter, branches 14-24 mm in dia.
meter tapering to 2-1 mm in diameter in
branchlets; all branches slightly basally con-
séricted, tapering gently to a rounded apex,
segments 4-12 times as long as broad:
htanches slightly constricted between segments,
but diaphragms conspicuous. Cortex in branch-
Jets | cell thick (Fig. $0), in older branches
thicker and in axes to 8 cells. thick {Fig, S£):
cortical cells ovoid, 75-35(—40) zm across and
1-2 times as long as broad in surface view,
Longitudinal filaments Scattercd throughout
diaphragins, with (I-)2(-3) complete cells
und two part cells between the diaphragms
(Fig. 88), with each peripheral filament cell
connected to the cortical cells by a lateral fila-
merit.

Cystocerps scattered over lesser branches,
spherical to shghily ovoid, #2(-14) mm
across, broad based. without an ostiole; carpo-
sporangia borne directly on the large, basal,
Fusion ceil,

Spermatangia unknown.

Tetrayporangie scattered over branches,
ietrahedrally divided, mostly 150-200(—250)
pm ip diameter.

Thalluy erecins ad 20-5U cm alias unc vel
Pluribus axibus ex base dura rarnose e1 perconl ad
5 em altis et L/3-4 cm latis ortis, haptere dis-
coideo, Rami laterales in verticillum vel opposire
dispositi, ad 10 cm longi ct similita ramtosi: axes
2-7 mm, rami 1.5-2.5 mm et ramuli 0.7-£ mm in
diametro, segmentis $-14 plo longiorlbus quam
lalis, plus minus constrictus, diaphragmatibus con-
spicuis, Cortex ad nnam cellulam in ramulis
cyassus, ad 8 cellulas crassus in axibus crescens.
Filamenta fongitudinalia dispersa, plerumque 2
cellulas totas inter diuphragmata habentia. Cysto-
carpia subglobosa 3-1(=14) mim Inta, haud osteo-
lata, dispersa; carposporangia in walescenti cellula
ipsa magna basali portata. Tcirasporaneta in ramis
dispersa, 150-200(—250) em in diametro.

Type locality, Tapley Shoal, Edithburg, §,
Aust., 15 m deep (Shepherd, 2,11,1969),

Holotype. ADU, AS33515. Isotypes to be dis-
iributed under this number.

Distribution. Only known from the type col-
lection from Tapley Shoal, and Investigator
Strait, S. Aust., Wetsen, 11 m deep (WLIO I:
ADU, A40995}, 23 m deep (28.21.1971; ADU,
A41010), and 34 m deep (201-1971; ADU,
ASSL9T),

Chylocladia grandis appears to be quite dix
tinct in fs form, large size, detise branching,
and in the perennial base which appears 10 last
fov several years, producing one to several
fronds annually (probably in spring and Sasting
Uirough summer).

STRUCTURE AND REPRODUCTION
Material studied: The type and Investigator
Strait. collections.
Thallus development
The multiaxtal apex of a branch (Fig,
84, B) includes both a central group of apical
cells which give rise to the scattered longitud-
inal filaments, and outer apical oglls which pro-
duce the peripheral longitudinaf filaments and
the cortex.

Fig, 8. Chylocladia grandis, A, Surface view of an apex showing peripheral and central apical cells
(A33515). & Longitudinal section of a branch apex showing development of cortex, dia-
phraems, und longitudinal flaments with gland cells (A335145). C. Cross section of a branch
showing a diaphragm with sub-peripheral and central longitudinal flaments (A33515). D,
Longitndinal view of outer part of a mature branch, showing the lateral connecting filaments

between ihe longitudinal filaments and (he vortex

(AS35151. £, Lonintudinal section of an

axis showing the multi-layered cortex (A33515),

SOUTHERN AUSTRALIAN CHAMP/A AND CHYLOCLADIA

FIG. 8.

oS

94 D, J. REEDMAN & H. B. & WOMERSLEY

acne Or
os

Fig. 9. Chyloeladia grandis. A, Longitudinal section showing supporting cell with auxiliary mother
cells und 4-celled carpogonial branch (A33515). B. Post-fertilisation Stage showing two auxiliar
cell branches and remnants of carpogonial branch (A33515). C. Post-fertilisation stage wit
auxiliary cells bearing young catposporangia, The ‘supporting cell, auxiliary mother cells and

auxiliary tells. are partly fused (A33515

), 2. A cystocarp with mature carposporangia formed

directly from the fusion cell, the lobes of which probably indicute the original auxiliary
mother cells (A33515). E. A young tetrasporangium (A39197 ). F. A mature tetrasporangium

(A39197).

The 12-16 central apical cells (Fig, 8A, B)
divide transversely and the cells elongate to
form longitudinal filaments. Each third ¢écll
usually produces a whorl of diaphragm initials
which divide further to join with similar
adjacent cells to form the single-layered dia-
phragms (Fig. 8B, C). The longitudinal fila-
ment cells between the diaphragms generally
produce a single spherical to slightly pyriform
gland cell (Fig. 8B).

The 15-20 outer apical cells divide trans-
versely to form the peripheral ting of Jongilu-
dinal filaments, but each of these cells, close ro
the branch apex, divides periclinally to form an
ouler primary cortical cell initial (Fig. &8).
This cell divides periclinally again once or twice
and then the outer cell divides anticlinally to

form the primary cartical layer, but the Later
formed cells are not in pit-connection with the
longitudinal filaments (Fig, 8B). The periclinal
division of the cortical initial is followed by the
inner one or two cells elongating to form a
bridging filament between each cell of the peri-
pheral longitudinal filaments and the cortical
cells (Fig. 8D), This feature is not found jn
the Australian species of Chantpia. The peri-
pheral longitudinal filament cells cut off dia-
phragm cells which joitt with those from the
inner filaments. The peripheral filaments are
usually separated by one diaphragm cell from
the cortex (Fig, 8C),

Many of the primary cortical cells cut off 1
small outer cell which produces a hair; these
hairs form a dense felt over most of the thallus.

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADIA 97

pel eA VN a a a

1 2 3 4 > 6 7 8 9 10

Fig. 10. Champia viridis. A. Type of C. tasmanica Harvey in TCD. B. A slender form (var, gracilis
Harvey) on Posidonia (Marino, S. Aust. Drift. Womersley, 26.x.1975; ADU, A46651).

98 D. J. REEDMAN & H. B. S. WOMERSLEY

As the branch matures, the primary cortical
cells cut off outer layers of cells, and in old
axes the cortex may be up to 8 cells thick
(Fig. 8E).

Branches originate from the region of the
diaphragms, when a group of apical initials
develops from the cortical cells,

The basal part of the thallus is clearly peren-
nial, being hard and resistant, up to 5 cm high
with irregularly placed, lateral propections rep-
resenting the bases of previous axes. From the
number and position of branch bases, some
perennial bases appear to be 4—5 years old, and
the axes are probably newly formed in spring
and lost by the following winter. The branches
probably decay rapidly since this plant has
never been collected in the drift.

Procarp and carposporophyte

The supporting cell is a large primary cor-
tical cell in pit-connection with a peripheral
longitudinal filament (Fig. 94). The support-
ing cell is multinucleate and cuts off outwardly
a small cell, the carpogonial branch initial, and
two larger cells, the auxiliary mother cells. The
carpogonial branch (Fig. 9A) is 4-celled,
curved, with an outwardly directed trichogyne.
Prior to fertilisation, each auxiliary mother
cell produces a uninucleate auxiliary cell (Fig.
OB).

Following fertilisation, a connection forms
between the fertilised carpogonium and each
auxiliary cell, and carposporangia are formed
directly from the auxiliary cells (Fig. 9C).
Fusion occurs between the auxiliary cells and
auxiliary mother cells (Fig. 9D), forming a
large basal fusion cell bearing the carpo-
sporangia directly. The supporting cell and
some vegetative cells may be incorporated into
the fusion cell.

Concurrent with the early development of
the carposporophyte, vegetative cells around
the supporting cell divide to produce erect
chains of cells which cut off outer cells and
form the pericarp (Fig. 9D), as in Champia.
The inner cells of the pericarp form the “tela
arachnoidea”, but no ostiole is produced.
When the carposporangia are mature, the top
of the pericarp ruptures.

Tetrasporangia

The tetrasporangia develop by enlargement
of the primary cortical cells (Fig. 9E), which
have several pit-connections with adjacent cells.
The tetrahedrally divided sporangia (Fig. 9F)
develop a thick gelatinous sheath.

Relationships

Chylocladia grandis agrees well with
Chylocladia and its type species, C. kaliformis,
in thallus structure and in reproduction. How-
ever, it has not been established whether one
or two auxiliary cell branches occur in C.
kaliformis. Champia has only one auxiliary
cell branch as far as is known, but the type of
Gastroclonium Kuetzing (G. ovale (Hudson)
Kuetzing) has two auxiliary cell branches
(Bliding 1928, p. 27).

The thallus structure of Chylocladia grandis
differs from that of the Australian species of
Champia in that the initial cells of the primary
cortex are connected to the longitudinal fila-
ment cells via a filament of one or two cells,
not directly, Whether this occurs in C. kali-
formis has not been established,

SPECIES EXCLUDED FROM
CHYLOCLADIA

Apart from the various Chylocladia names
which have been shown previously to be syno-
nyms of species of Champia or other genera,
the following names are now referred to other
genera.

Chylocladia fruticulosa (Reinbold) De Toni

1900b: 576.

Lomentaria fruticulosa Reinbold 1899: 46,

Type locality, Investigator Strait, S. Aust.
(Davey 148).

Type. Herb. Reinbold, M. Isotype in ADU,
A1553.

The thallus of the isotype is on Posidonia
(not Amphibolis antarctica as in Reinbold). It
is hollow and without diaphragms, and the
tetrasporangia are grouped in sori around
depressions in the wall of the branches, These
features are typical of Lomentaria, and the iso-
type (a small, bleached specimen) appears
similar to the earlier described Lomentaria aus-
tralis (Kuetzing) Levring 1946, p. 223 (Chon-
drothamnion australe Kuetzing 1865, p. 29, pl.
82 d-f). The southern Australian species of
Lomentaria are in need of detailed study.
Chylocladia gelidioides Harvey 1863, synop.:

46. De Toni 1900b: 578; 1924: 312. Gepp

& Gepp 1906: 257. Okamura 1904: 88.

Type locality. Twofold Bay, N.S.W. (F.v.
Mueller).

Type. Herb. Harvey, TCD.

Although cystocarpic material has not been
studied, the hollow thallus construction without
single layered diaphragms, and sori of tetra-
sporangia, are typical of Lomentaria. The

z t
WH
HT!

€

7

& 8 Z 9

s
Hecrvabiinfrcalicatcand sotirnlcatitanitosttuitnaoadatativt

UL

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADIA

22-6.

a WA ‘in iq OTL CL WL Uy re
2 8 ! | ! ei ;

1

ui
it 1

Fig. 11. A. Champia insignis. Isotype male specimen (ADU, A12237).
B. Champia affinis. Lectotype specimen in BM.

6 7.

a

oe he eae

100 D. J. REEDMAN & H. B. S. WOMERSLEY

|
|

[rset wa daa ae WL

Thhal

Re a 2 A
sa, hey a coe Lad
1 2 3 4 5

> - ‘

“a f or —
i rete ye fa Ft ttn eets oe tes

*

Fig. 12. Champia affinis. A. Harvey’s Alg. Aust. Exsicc. 2521 (as C, obsoleta) from Georgetown, Tas.
(in TCD). B. Lectotype of C. obsoleta (Port Fairy, Vic. Harvey’s, Alg. Aust. Exsicc, 252D)
—a rough-water form. Champia zostericola. C. Harvey’s Alg. Aust. Exsicc. 249A, Fremantle,

W. Aust.

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADIA 101

‘ P /
J
A |
ae 4 | ' ,™
tte TE 4 4 aed
oe | \ ad ad ; aj
? . Fy bee Lis
: & '
1 i : Ss ‘ Ps
. ‘ yy -, Ms, i e =— 2
q. TZ AN j ~ a =
a. ap eo te te * ah ew, Ae 2
Phe (je ot ade, OO: Loe a Te ae =-o
LPass, 3” a | au - aa 4 A? oo sf =
Sy F % 4 =
hE, ORR Lee ah
it~ ® : ae fr =a os t f sf =
(3S ewe 5 i ee ~ =
‘ * id —
=o
=—wn
=~
=—
=
—

ih
mM

8 é 9 s 7 e€ €

tow svt gtiuatuadustuniduatustihuslt

6

Fig. 13. A. Lectotype of C. affinis var. arcuata Hooker & Harvey. Gunn 1332, in BM. B. Champia
zostericola. Musselroe Bay, Tas. Perrin, March 1937 (MEL 45252)—plant with well-

developed axes.

102 D. J. REEDMAN & H. B. S. WOMERSLEY

"?,

AON
a eee ww Ie
Be 7h

ie aan MM

1 2 3 4 5 6 7 8

§ A. Shepherd

toll

Fig. 14. A. Champia parvula var. amphibolis. Type specimen of variety.
B. Chylocladia grandis. Holotype specimen.

SOUTHERN AUSTRALIAN CHAMPIA AND CHYLOCLADIA 1%

thallus habit, cell detail, and tetrasporangial
sori are very similar to Lomentaria catenara
Harvey from Japan (as noted by Harvey 1863
and Okamara 1904), and the N.S.W. plant
may be a slightly less robust formi of the
Japanese species, Which is ulso recorded from
Pacific Mexico by Dawson (1963, p. 465, pl-
92). It appears to by closely related to Lomen-
faria ramsayana (J. Agardh} Kylin (1931, p.
27, pl, 14, fig. 33),
Chylocladia multiramea Sonder 1853; 681,

Type locajity, Lefevre Pen., S. Aust-

Type. MEL, 45196,

The type specimen (female) in MEL is a
slender, much branched, bleached plant of
Dasyphiloea insignis Montagne.

Acknowledgements

The authors are indebted to the Directors af
the National Herbaria of Victoria and of New
South Wales, and to the Brifish Museuns
(N.H), London, for the loan of specimens,
Professor J. Feldman, Institut de Biologie
Veégetale, Paris, Kindly seot matenul of
Champia parvula from Brest, and Mr R. H.
Simons, South African Seaweed Research
Laboratory, forwarded material of C. lnmbri-
caliv and C, conpressa,

The second author gratefully acknowledves o
grant from the Australian Research Grants
Committee and the technical assisiance pro-
vided by Mrs Enid Robertson and Miss Chery!
Anderson.

References

Acarpy, C. A. (7824),—“Systema Algarum,”
(Lund.)

AGAroH, C, A. (1828) —'Species Alvarum." Vol.
2. (Grypbiswald.)

Agarpy, J. G. (1852),—"Species, Genera ct
Ordines Algarum.” Vol. 2, Pt 2, pp, 337-720,
P\ 3, pp, 7O1-7E6. (T.und.)

Acaron, J. G. (1876).—"Species, Genera et
Ordines Algarum.” Vol, 3, Pri, pp. 1-724—
Epicrisis systematis Floridearum. (Lund,)}

AGABDH, J. G, (1879).—Florideernes morphologi,
K. VPerensk Ake. Forsehanedl, 15, 1-199,
Plates 1-33.

Acagoy, J, G. (1901).—"“Species, Genera et

Ordines Algarum, Vol, 3, Pr 4, pp. 1-149.
{Lund.)

Biome, C. (1928),—Studien tiber die Florideen-
ordoung Rhodymeniales, cfa Univ. aud,
N.F. Avd, 2, 24(3), 1-73.

Cuapman, V. J,, & Dromaoo.e, FT. (1970) -—
The marine alese of New Zealand. Pa7t UT:
Rhodophyceas. Florideophycidae; Rhodymme-
hiales. pp. 185-(54, Plates 39-50,

Dawson, E, Y. {1963).—Marine red algae from
Pacific Mexico. Part 6. Rhodymeniales, Nova
Hedwigia 5, 437-476, Plates 77-95.

Desvaux, N. A, (1808),—Champia Desv, J, de
Bor, 1, 245-246.

De Toni, G. B, (1900a)—Tl genere Champia
Desv. Mem. Ponnf. Accad, Nuovi Lincei,
Roma 17, 65-80.

De Tont, G. 8. (1900b).—"Sylloge Algarum om-
nium hucusque Cognitarium.” Vol. 4. Floti-
deae. Sect. 2, pp, 387-776. (Padua.}

De Tont, G, B, (1924).—"Sylloke Algarum om-
nNipm hucusque Cognitarium.” Vol. 6. Plori-
deae, (Padua.}

Garnet, J, R. (1971) —Check-list No. fh—Marine
and Freshwater Algae. In “The Wildflowers of
Wilson's Promontory National Fark", pp, 93-
97. (Lothian: Melbourne.)

Gaynas, P. (1966) —“Les Aleues des Cétes Fran-
cases.” (Deren: Paris.)

Gepp, A. & Gere, BE. S. (1906}—Some marine
algae from New Sonth Wales, J, Bor, Lari,
1906, 249-261. Plate 481.

Gumex, E, R, (19523.—The marine alpue of Tas-
matua: check hist with localities. Pap, Prac.
R. Soc, Fasm, 86, 71-106,

Harvey, W. H. (1$38}.—"The genera of South
African plants,” (Cape Town}

Hanver, W. H. (18444).—Champia tasmanica; 2
new Australian alga, Lond. J, Ber. 3, 407,
Plate 19,

HakVey, W. H. (1844b).—Algse of Tasmania.
Land, J, Bor, 3, 428-454,

Harvey, W, H, (J847).—"Nereis Australis,”
(Reeve: London.)

Harvey, W. H. (1849),—"Phycologia Britannica."
Vol, 2, Plates 121-240. (Reeve; London.)
Harvey, W. H. (1853).—"Nereis Boreali-Ameri-
cana,” Part T],—Rhodospermae: (Smithsonian

Tnst.:. Washington.)

Harvey, W, H, (1855a),—Some account of the
Marine botany of the colony of Western Aus-
tralia. Trans. R. Irish Acad. 22, 525-566,

Harvey, W, H. (18555),—Algae. da J. D, Hooker,
“The Botany of the Antarctic Voyage", Part
lL. Flora Novae-Zelandiae. Vol. 2. pp. 211-
266, Plates 107-121.

Harvey, W. H. (1859),— Algae, dn J, D. Hooker,
“The Botany of the Antarctic Voyage", Part
TIL. Flora. Tasmaniae. Vol. 2, pp. 282-343,
Plates 185-196.

Hanvey, W. H. (1863).—"Phycologia Australica,”
Vol. 5, Plates 241-300, synop. pp. 1-73.
Hooxer, J. D,, & Harvey, W. HB. (1847) —Aleac

Tasmanicae. Land. J. Bor. 6, 397-417.

Heoneh, W. J, (1833)—"The British Flora,”
Vol. 2.

Jorr, A. B, (1965}—Plora Marinhe do litoral
norte do estado de Sao Paulo e regioes cir-
cunvizinhas, Bol, Foc, Files, Ciéac. Letras,
Univ. Sdo Paula, 294, Ser. Bal. 21, 7-395.

Kinc, R. I, Brace, J, Hope, & Durken, Sophie
(1971) —Port Phillip Bay Survey 2. 8. Inter-
tidal ecology of Port Phillip Bay with system-
wile Usis of plants and animals, Meyn, nein.
Mus. Viet, 32, 93-128,

Kuerzine, F. T. (1843)~-"Phycologia generalis.”
(Leipzig,)

1+

Kuerzinc, F. T-.
( Leipzig.) .

Kuerzinc, F, T. (1865).—Tabnlae Phycologi-
cae.” Vol. 15. (Nordhausen.)

Kyiex, H. (1931) —Die Florideenordnung Rhody-
meniales, Acta Univ, lund, N.F. Avid. 2, 27
(11), 1-48, Plates 1-20,

Kyarn, H, (1956),—“Die Gattungen der Rhodo-
phycoen.” (Lund.)

Luvaing, T. (1946),—A list of marine algac from
Australia und Tasmania. Acta Herti gotho-
burg 16, 213-227.

Linnazus, C. (1771).—“Mantissa plantarum
altera.” pp. i-vi, 143-588. (Stockholm.)
Lucas, A. H. S. (1909).—Revised list of the
Fucoideae and Florideae of Australia. Proc.

Linn, Joe, N.S.W, 34, 9-60,

Lucas, A. H. S, (19294).—The marine algae of

Tasmania. Pap, Proc. R. Soc. Tasm, 1928,

f-27-

Lucas, A. HLS, (1929b),—A census of the marine
algae of South Australia. Trans. R. Soc: S$.
Aust, 53, 45-55. ;

Lucas, A. H. S. (1931)—Notes of Australian
marine algae, WI. Proc, Lina. Soc, NSW.
56, 407-411, Plates 23-27,

Lucas, A. H.S., & Pesnin, F. (1947).—"The Sea-
weeds of South Australis.’ Part TW. The Rexl
shih teal pp. 109-458. (Govt Printer: Ade-
aide,

May, V. (1947).—Studies on Australian marine
ulgue. IMT. Prac. Tian. Soc. N.S.W. T1, 273-
277, Plate 19,

May, V. (f965).—A census and key to the spe-
cles of Rhodophyceae (red algae) recorded
from Australia. Contr. N.S.W. narn. Herh. 3,
349-429

Navanir, C. (1847).—"Die neueren Algenstyeme,”
(Zurich)

Nayior, M. (1954).—A check list of the marine
algae of the Dunedin district. Trans, R. Sec,
N.Z. 82(3), 645-663.

Nrwrton, L, (1931),—A Handbook of the British
Seaweeds.” (British Museum: London,)
OKaAmuRA, K, (1904),—List of marine algae col-
lected in Caroline Islunds and Australia. Bot,

Mag., Tokyo 18, 77-96,

REINBOLD, T. (1897) —Die algen der Lacepede
und Gilichen Bay. Nueva Notarisia 8, 41-62.

KEINBOLD, T, (1898) —Die Algen der Lacepede
und Guichen Bay (Stid Australien) und deren
naherer Umbebung, gesammelt von Dr <A,
epgeihact-SoaiRsog: I. Nuova Notarisia 9,
33-54 .

(1849) —"Species Algarum,”

TD. J, REEDMAN & H, B.S, WOMERSLEY

REINBOLD, T. (1899),—Meeresalgen von Investi-
gator Street (Sud Australien), gesammelt von
Miss Nellie Davey (Waltham, Honiton). Hei-
wigia 38, 39-51,

SuEPHERD, S. A., & WoMERSLEY, H. B, S, (1970).
—The sublittoral ecology of West Island,
South Australia. I, Environmental features
and the algal ecology. Trans. R. Soc, S. Avast,
94, 105-138.

SHEPHERD, S, A., & Womerscey, H, B. §, (1971),

—Pearson Island Expedition 1969.—7. The
sub-tidal ecology of benthic algae. Trans. R,
Soc. 8S. Aust. 95(3), 155-167.

SonpDER, O. G. (1846).—Algae. Jn C. Lehmann,
“Plantae Preissianae,”’ Vol, 2, pp. 148-195.
(Hambureg.)

Sonper, O. G, (1853),—Plantae Muellerianac,
Algae. Linnaeu 25, 657-709.

Sonver, O. G_ (1855).— Algae annis 1852 et 1855
collectae. Linnaea 26, 506-528.

Sonper, O, G. (1880) —/w F, von Mueller, “Frag-
menta Phylographiae Austratiae". Supple-
mentum ad volumen undecinum; Algae Aus-
tralianae huctenus cognitae. pp, 1-42, 105-107,

StackHouss, J, (1809).—Tentamen Marino-
Cryptogamicum. Mem. Soc. Imp. Nat. Mes-
cou 2, 5097.

Tare, R. (1882).—A list of the charas, mosses,
liverworts, lichens, fungs, and algals of extras
tropical South Australia, Trans, R, Sec. §.
Angst. 4, 5-24.

Tispait, H. T. (1898).—The algac of Victoria,
Rep, 7th Meet. Aust. Ass. Ady. Sci., Sydney,
1898. pp, 493-516.

TREVISAN, B. A. (1848),—“Saggia di uns
monugrafia delle alghe cocotalle.” (Padova,)

WEBER VAN Bossg, A. (1928). —Liste des algucy
du Siboga. IY. Rhodophyceae, Pt IT, Gigar-
tinales «ct Rhodymeniales. Monogr. Siboga
Exped, 59d, 393-533, Plates 11-16.

Wirson, J. B, (1892) —Catalogue of algae col-
lected at or near Port Phillip Heads and
Western Port. Proc, K, Soc, Vier. 4, 157-190).

WoMERSLEY, H. B.S, (1948).—The marine algae
of Kangaroo Island. If. The Pennington Ray
region. Trans. R, Soc, 3. Aust, 72, 143-166,
Plates 10-15.

Womerscey, H. B, S, (1950).—The marine algae
of Kangaroo Island. If. List of species 1.
Trans. R. Soc. §. Aust. 73, 137-197,

Womersiey, H. B. S. (1966)-—Port Phillip sur-
vey, 1957-1963: Algae. Mem. notin. Mus.,
Vier, No, 27, 133-156.

WoMERSLEY, H. B. S., & Batty, A. (1970}.—
Marine algue of the Solomon Islands, PAii.
Trans, R, Soe. Lond, B 259, 257-352,

LEGGADINA LAKEDOWNENSIS, A NEW SPECIES OF MURID RODENT
FROM NORTH QUEENSLAND

BY C. H. S. WATTS*

Summary

WATTS, C. H. S. (1976).-Leggadina lakedownensis, a new species of Murid rodent from north
Queensland. Trans. R. Soc. S. Aust. 100(2), 105-108, 31 May, 1976.

Leggadina lakedownensis, a new species of murid rodent from northeastern Queensland is
described and figured. Morphologically it most closely resembles L. forresti (Thomas).

LEGGADINA LAKEDOWNENSIS, A NEW SPECIES OF MURID RODENT FROM
NORTH QUEENSLAND

by C.H. S, WaATTs*

Summary
Warts, C. H. S. (1976) —Leggadina lakedownensis, a new species of Murid rodent from north
Queensland. Trans, R. Soc. §. Aust, 100(2), 105-108, 31 May, 1976.
Leggadina lakedownensis, a new species of murid rodent from northeastern Queensland is
described and figured. Morphologically it most closely resembles L. forrest? (Thomas).

Introduction
A pair of small rodents from Cape York,
examined during the course of a broader inves-
tigation into the karyotypes of Australian
rodents, was found to differ in karyotype,
blood proteins, and cranial morphology from
the otherwise similar Leggadina forresti, as Well

as all other murids examined,

Leggadina lakedownensis n.sp.
FIGS 1, 2, and 3

Holotype: Qld Mus, JM1192, 9. Received
from Queensland Museum in May 1975, killed
August 1975. The parents of this specimen
were received by the Queensland Museum from
Mr C. Tanner in August 1973; these were off-
spring of animals collected at Lakeland Downs,

110 km S of Cocktown, Queensland by Mr R.
Buckley in 1973.
Description: (Colour after Ridgway 1912).
Head relatively narrow and pointed as in L.
deliculata and L. hermannshurgensis. Eyes. not
as prominent as in those species or in L. for-
resti. Ears small and broad, proportionately
smaller than in L. forresti. Feet narrow. Hairs
on back with tips buffy brown grading to pale
olive-buff on sides, bases of hairs blackish
brown which shows through giving back a
brindled look, Underside white, hairs white to
base, as are hairs on upper surface of feet, Face
with suggestion of darker central stripe and
lighter ring around eyes, Tail sparsely haired
with light-coloured hairs.

Skull flat on top, rostrum short, interorbital
region broad. Interparietal wide, short. Zyge-

Fig, 1. Dorsal, lateral and ventral views of skull of Holotype Leggudina lakedawnensis n,sp., % 2.5.

* Institute of Medical & Veterinary Science, Frome Rd, Adelaide, S. Aust. 5000.

16 C oH, 3S. WATTS

Fig, 2. Upper right molar tooth row of holotype o}
Leggadina lakedownensis tsp,

matic plates moderate, minimum width equal
to Jength of Mt. Lacrimals small. Nasals short,
not exceeding premaxillae anteriorly. Incisive
foraminae longer than tooth row, reaching
heyond anterior end of M4, broad, slightly
Wider posteriorly, braadiy rounded posteriorly,
more pointed anteriorly. Posterior palatal fora-
minae oval, about lencth of M5. Mesopterygoid
fossae narrow, their width about equal to length
of Mi. Bullae moderately developed about size
of occipital foraminae, the distance between
them about equal to combined lengths af M=
and M3. Upper incisors forward pointing.
Molars as in Figure 2, anterior ligual cusp of
M} bDlade-like and strongly developed. M+ large,
approximately 659% of tooth row, M% small,
approximately 15%. of tooth row.

Specimens examined; Paratypes; Qld Mus.
JM1293 3. JM1294 f bred in captivity; collec-
tion details as for Holotype, Referred specimen,
Qld Mus. J.17919, Williams Id, Queensland
(35°S; 135°B), 1969, coll. Mr, C. Tanner

Diagnosis

(Head and body 60-70 mm, Tail 40-50 mm,
Far 10-12. mm, Skull length 20-23. mm),
Incisive foraminae reaching heyond anterior
end of M}, rounded and slightly widened pos-
teriorly. Upper incisors forward pointing.
Upper molar tooth row 3.7—4.2 mm long, An-
terior ligual cusp of M} blade-like and strongly
developed. M} 4 to 5 times length of M}.

Systematic position
Leggadina lakedownensis ts separated from
all named forms within the genus Legvadina, as
recognised by Tate (1951), except L. forrest

(Thomas), L. waite’ (Troughton) und LZ, mes-

soria {Thomas), by the short tail, small cars,
Jong incisive foraminac, and [arge molar tooth
row with M1 large and with a strongly
developed antenor ligual cusp.

These three taxa, L. forresti, L. waitei and
L. miessoria, were considered to he very closely
related by Tate (1951) and were synonymised
under the name of Pseudomys forresti by Ride
(1970). My examination of the types of all
three supports this view that only one species
is involyed, characterized by downward point-
ing upper incisors and long incisive foraminae
harrowing posteriorly and about 3 times the
length of M3. Measurements of the Holotypes
are included in Table 2-

Lepvadina lakedownensis differs from L. for-
resti (and the Holotypes of L. waifei and L.
messoriag) in having the incisive foraminae
widening rather than harrowing posteriorly, a
slightly larger M} (in comparison to total tooth
tow), a smaller M3 und forward pointing upper
invisors, On the limited evidence available it is

TABLE 1
Measurements o] specimens of L. lakedownensts (min)
1 2 3 4
Hend and body 72 - — 64
Tail 45 — -— 4)
Hind toot (S.u.) 1B _— _ MW
Far (from noteh) It a — i
Greatest length of skull 73.0 2s — aA
Zyyomatic breadth 125 tla 12.0 10.8
tnterorbital breadth a4 45 44 tn
Depth of brain case,
inclyding bullae R4 - _— 36
Length of nasal 73 73 73 6.7
Length of ant. palatal
foramina a7 5.2 54 4.8
Crown length of molar vow 4.0 49 3.7 4.2
Crown length of M1, 27 2.7 24 246
Crown length of MA 04 O58 0.5 09

1, Old Mus. IM1292 9. Lakeland Downs, Qld, 1973 Holo-
type.

2. Qld Mus. JM1294 9, Lakeland Dowhs. Ql, 1973

3. Old Mus. M1293 9, Lukeland Downs, Qtd, 1973,

4, Qld Mus, 317919, Williams Td, Qld, 1969,

NEW SPECIES OF MURID RODENT

107

Fig. 3. Leggadina lakedownensis, tsp.

a slightly smaller animal. In my opinion the
scale of these differences indicate a form speci-
fically distinct from L. forresti or any other
described rodent,

Support for the distinctiveness of L. lake-
downensis from L, forresti comes from studies
of the chromosomes and blood proteins of these
and some related species (Baverstock ef al.
1976). These studies suggest that L. forresti
and L. lakedownensis form a distinct group
within Pseudomys sensu lato. The morphologi-
cal characters (cf. above) tend to support this
view and it seems prudent at this stage to retain
the genus Leggadinu at least for these two
species.

Habitat and distribution

The Williams Island locality is an area of
short grassland surrounded by scrub. (The
locality is illustrated as Lakefield Station, by
Covacevich (1974, p. 7).) Lakeland Downs is
on an isolated area of basalt-derived red and
brown soil supporting a natural vegetation of
Box woodland (E£. leptophleba), and deciduous
scrub with kangaroo (Themeda australis) and

spear (Heteropogon contortus) grass. At the
time of collection however, sorghum covered
the whole area. Rainfall is between 100 cm and
130 cm per year, occurring mainly in the sum-
mer (J. Covacevich, personal communication).

At present the species is only known from
the above 2 localities on the eastern side of
Cape York. Leggadina forresri has a much
greater distribution in inland Australia and is
known from W.A., S.A., N.T, and N.S.W. as
well as western Queensland. From specimens
in the Australian Museum and Queensland
Museum (Table 2) it appears that there is a
considerable gap between the distribution of
the two species.

Acknowledgments

Miss Jeanette Covacevich kindly sent the
animals on which the descriptions are based.
Curators in the Queensland Museum, Austra-
lian Museum and South Australian Museum
kindly sent material or allowed me access to
specimens of Leggadina in their care. I am
also grateful to Miss Heather Aslin for draw-
ing the skull and teeth of the type specimen,

108

C. H. S. WATTS

TABLE 2

Measurements of specimens of L. forresti from Qld, S.A, and N.T. including Holotypes of L. waitei, L. forresti and L-

Head and body

Tail

Hind foot (S.u.)

Bar (from notch)

Greatest length of skull

Zygomatic breadth

Interorbital breadth

Depth of brain case,
including bullae

Length of nasal

Length of ant. palatal
foramina

Crown length of
molar row

Crown length of ML

Crown length of Mi

Baversrock, P. R., Hocartu, J. T., Core, S., & Ripeway, R.
(1976)—Biochemical and
Karyotypic evidence for the specific status of
the rodent Leggadina lakedownensis Watts.

CovacevicH, J.

AM M5194, Hart Range, N.T.
L. waitei Holotype

MNH 6.3,9.39, Alexandria, N.T., 1905

. forresti Holotype

an

a BMNH 25,4.9.1, Melrose, S.A,, 1922

— L. messoria Holotype

Trans. R. Soc. 8. Aust, 100(2),

Ripe, W. D. L. (1970).—"A guide to the native
mammals of Australia.” (Oxford Univ. Press:

London.)

x & AM M9208, Mt Isa, Qld, 1968
xy

messoria (mm)

mn
2 af ao
: - 5
o = Um =
; o
5° 27 s
3 3 2
e @ ¥g vg
3S cel mo ee
e 5 se 3&
= 5 ZG Zé
= = qt G<
< SF KH “nH
83 88 100
—_ — 5 6
172,-— 17 17
12.3 — 15 14
— 23.5 23.4 24.6
13.8 126 12.1 13.2
40 3.7 36 33
86 87 86 9.2
1.700 81
50 52 454 56
45 46 46 43
2.7 #27 27 25
09 09 10 O08
References

SAM M2958, Macdonald Downs,

S.A,, 1930

SAM M2405B, Horn Exp. (Waite)

SAM M2405, Horn Exp. (Waite)

SAM M6350, 60 km N Birdsville,

Qld, 1966

SAM M6351, 60 km N Birdsville,

Qld, 1966

SAM M6352, 48 km N Oodnadata,

S.A., 1966

(1912).—"Colour standards
colour nomenclature.” (Ridgway: Washing-
ton).

Tate, G. H. H. (1951).—Results of the Archbold
Expeditions. No. 65. The rodents of Australia

AM M9481, 20 km W Innamincka,
. 1973

A

nw

and

and New Guinea. Bull. Am. Mus. Nat. Hist.

97, 183-430.

CovacevicH, J., & Easton, A. (1974).—“Rats and
mice in Queensland.” (Queensland Museum

Booklet No, 9: Brisbane).

BIOCHEMICAL AND KARYOTYPIC EVIDENCE FOR THE SPECIFIC
STATUS OF THE RODENT LEGGADINA LAKEDOWNENSIS WATTS

BY P. R. BAVERSTOCK*, J. T. HOGARTH*, S. COLE* AND J. COVACEVICHT

Summary

BAVERSTOCK, P. R., HOGARTH, J. T., COLE, S., & COVACEVICH, J. (1976).-Biochemical
and karyotypic evidence for the specific status of the rodent Leggadina lakedownensis Watts.
Trans. R. Soc. S. Aust. 100(2), 109-1 12, 3 1 May, 1976.

Leggadina lakedownensis Watts differs karyotypically from its apparent nearest relative, L. forresti.
Further, the biochemical differences between L. lakedownensis and L. forresti are greater than those
between other "good" species of similar sized pseudomyins. These data support the specific status
of L. lakedownensis.

BIOCHEMICAL AND KARYOTYPIC EVIDENCE FOR THE SPECIFIC STATUS OF
THE RODENT LEGGADINA LAKEDOWNENSIS WATTS

by P. R, Baverstock*, J. T. HoGARTH*, 8, CoLe* and J. CovAcEvicHt

Summary

Baverstoce, P. R. Hocarru, J. T., Core, §, & Covacevicn, J, (1976),—Biochemical and
karyotypic evidence for the specific status of the rodent Levvadina lakedownensis Watts.
Trans, R, Soc. §, Aust, 100(2), 109-112. 31 May, 1976,

Lexeadina lakedownensis Watts differs karyotypically from its apparent nearest relative.

L, forresti. Further, the biochemical differences between L. lakedownensis and L. farresti are

greater than those between other “good” species of similar sized pseudomyins, These data

support the specific status of L. lakedownensis,

Introduction

In 1973 several specimens of a species of
small rodent were collected from Lakeland
Downs in northeast Queensland and described
as a new species, L. lakedewnensis, by Watts
(1976), The morphological differences between
L. lakedownensis and its apparent nearest rela-
tive L. forresri are, however, minor. Because
speciation is the result of the accumulation of
many genetic differences and because mar-
phology alone reflects only part of these genetic
differences, it seems desirable in such cases to
assess. other aspects of genetic differences
between allopatric populations. The present
study was undertaken to determine whether the
karyotypic and biochemical differences between
L, forrestt and L. lakedownensis substantiated
the recognition of the latter as a distinct species,

Methods

Seurces of attimals: The sources of animals
utilized in the present study are shown in Table
1, along with their Institute of Medical and
Veterinary Science number, When these ani-
mals die their skull and skin will be submitted
to a Museum and given a Museum number.
Muscum numbers corresponding to JMVS
numbers. will be available from the IMVS or
the South Australian Museum,

Chromosome preparations; Chromosome
preparations were made from & L. forresti
and 3 L, lakedownensis. Animals were bled

by cardiac puncture under ether anaesthesia
and leukocytes cultured for 3 or 4 days.
Slides were prepared by means of the routine

TABLE i
Rejerence numbers and sources of animals
IMVS
Species No. Sex Locality
(a) Chromosomes
L. jarresti 1 EF Coorabulka, SW Qld
2 M = 19km W Innamincka,
S\A.
3. FEF Wk W Innamincka,
S.A.
5 F 19km W Innamincka,
SLA.
6 M ~ Bowlers Gop Stn, N.S.W.
7 \M ~~ Fowlers Gup Stn, N.S.W
8 M_ Fowlers Gap Sin, N.S.W,
9 M = Mt Sarah Sta, S.A.
L. lakedownensis 10% F Lakeland Downs Stn, Glu
11 Ms Lakeland Downs Stn, Qld
12 M_~ Lakeland Downs Stn, Old
(b) Blectrophoresis
L. forresti 3 fF 19km W Innamincka,
§.A.
5 FB 19km W Innamincka,
5A.
L, lakedawnensis 10* F Lakeland Downs Stn, Qld
L. delicatule 13. M_ Fairbairn Dam, 22km
SW Emerald, Qld
14 M_ Fairbaim Dam, 22 km
SW Emerald, Qld
L, hermanns-
burgensis 14 M_~ Lah, stock
P.novaehollandiae 15 M Port Stevens area, N-S-W,
P. australis M sLab. stack

* Holotype—Queensland Museum JM{292-

* Institute of Medical & Veterinary Science, Frome Rd, Adelaide, S. Aust. S000.
| Queensland Museum, Gregory Tce, Fortitude Valley, Qld 4006.

110

P. R. BAVERSTOCK, J. 1, HOGARTH, S, COLE AND J. COVACEVICH

TABLE 2
Eléctraphoreétic hiiffer atid Staitiing systeiis used for proteins examined

Enzyme/ Protein

E.K. t.1.149 Glucose 6-Phusphate dehydrogenase (G 6-PD)
6-Phosphoglicondle dehydrogenase E.C. 1.) 1.44 (6-PGD)

Phospho-hexose isomerase F.C. 5.3.1.9, (PHI)
Phosphoglucomutase E.C, 2;7.4.1, (PGM)
Lactate dehydrogenase E,C_ 1.1,1,27 (LDH)

NAD-Malhiute dehydrogenase E.C. 1.11.37 (NAD-MDH)

Leucine aminopeptidase E.C, 3.4.1.1 (LAP)

Glyceraldchyde 3-Phosphate dehydingenase GA 3-PD EC, 1,2.1.12

Telrazolium oxidase

Esterase (Est)

Albumin (Alb.)

Heccmich Asie { Ris;
Transferrin (TF)

* PEB — Tris-EDTA-Borate

air-dry method, For karyotypes, slides were
Stamed with 2% Giemsa, C-staining was con-
ducted by the method of Arrighi and Hsu
(1971) except that the RNase step was omitted.
Slides were then stained with 10% Giemsa.

Electrophoresis: In addition to 4. forrest! and
L, lakedownensis, specimens of L.. delicatula, L.
hermansburgensis, Pséudomys noyachollandiae
amd P. australis were studied, Blood was col-
lecred by cardiac puncture umder ether anaes-
thesia in syringes containing a dried film of
heparin and centrifuged immediately at 2,.000G
for 10 minutes at room temperature. Plasma
was pipetted off and stored at ~ 20°C. The red
cells were washed 3 times in 2 volumes of
isotonic saline and lysed in ai equal volume
of distilled water containing [/5 volume of
toluene. Cell walls were centrifuged out and
the clear supernatant stored at —20°C for a
Maximum period of 3 weeks.

Horizontal starch gell electrophoresis was
used. Gel slabs 300 mm x 150 mm x 6 mm
were prepared from 12.5% (50g/400 ml
buffer) starch using a perspex mould. Gels. were
run inva refrigerator to minimize heating. After
electrophoresis a section approximately 90 mm
x 120 mm was cut from the gel slab, sliced into
twa separate halves, and these halves then

Stain
Brewer (1970)
Brewer (L97/4))
Brewer (1970)
Brewer (1970)
Brewer (1970)

Buffer system
Brewer (1970)
Brewer (1970)
Selander e7 ai, (1971) 7
Selander ef al. (1971) 7
Holmes ef a/, (1973)
TEB*

Holmes etal. (1973) Brower (L970)

TER"
Selander ef af, (1971) 2 Rrewer (1970)
Brewer (1970) Brewer (1970)

Brewer (1970) Scored from gels
stained fram GA
3-PD

Soin A; 008M. tris-cit- Brower (1970)
rate pH 8.6 Soln Br

0.06 M. Li-borate-pH

3.3 Gel: 337.5 ral Soln

Ar 62.5 ml Sola B in

400 ml B. Electrode:

Soln B.
a8 for Est Amnido Black
ub lor ¢3 6-BES Asiaho Bick

as for Est Amido Black,

incubated in the appropriate staining solutions
(Table 2),

Results

Chromosome studies; Fig. 1a shoks the karyo-
type of a female L. forresti, The diploid number
was 2N = 48. The largest chtomosome, desig-
nated pair 1, was acrocentric in IMVS§ 1, 2, 5,
7, 8 and 9. However, in IMVS 3. and 6, pair |
was heteromorphic, onc member hemg acro-
centric and the other being subacrocentric with
a distinct short arm. Pairs of 2 to 21 were
acToceninic forming a series graded in size,
Puirs 22 and 23 were small metacentrics. lhe
presumed X-chromosome was an acrocentric
representing about 6% of the total chromosome
length and the Y-chromosome an acrocentric,

The C-staining technique (Fig. 1b) showed
that although the centrometric area of some
chromosomes were C-banded, only in the small
Inetucentric pairs 22 and 23 was this marked,
In many preparations the presumed Y stained
slightly more intensely than other chromosomes
over its entire length.

Chromosomally Z. Jakedownensis differs
from L, forresti in possessing only one pair of
small metacenirics (Fig. 1c). Also, in the threc
individuals karyotyped, pair 1 was. always acro-
centric, Centromeric C-banding of L. Iake-

BIOCHEMICAL AND KARYOTYPIC STUDIES OF LEGGADINA LAKEDOWNENSIS 111

On An an
nn Aah Ha
oA AN AN

fin OO An

On AA AO
an an an

PA OAK Aw
on la An

hh Aa on Ag
AN AN fA OA
On AN 1& AO

an 10
ah &0
oh Of

Ah Ah AR an ly fe
1
10p 1c

ee
mere

= it

quer
b% in | a!
s

& if 4 és =a
ee 23 ea -® 8
22 23

re—

10p 1b

' 4 a és 2

1 oe oe

> BQ

ge 8 8 ef Ae
*e “ff 88 Ae 88 4 5
2z 23 :
——
10p 1d xy

Fig. 1. Karyotype and C-banding of L. forresti and L. lakedownensis. 1a, Karyotype of female L. forrest

(IMVS 3) heteromorphic for pair 1;
for pair 1;
dawnensis (IMVS 12).

Species G-6-PD Hb TO

1b, C-banding of male L. forresti (IMVS 6) heteromorphic
lc, Karyotype of male L, lakedownensis (MVS 11);

1d, C-handing of male L. lake-

GA-3-PD Est-1 Alb NAD-MDH

Fig. 2. Representation of electrophoretic patterns observed for seven useful proteins in six pseudomyin
species. In each case the origin is to the left, and fastest migrating bands to the right (cathodal).
Key—L J, = Leggadina forresn; LA. = L. lakedownensis; Lid, = L. delicatula; L.h. = L. her-
mantsburgensis; P.n. — Pseudomys novachollandiac; P.a. = P. australis,

downensis (Fig. 1d) was evident only in the
smaller chromosomes. Tha presumed Y was
slightly more intensely C-bandcd than other
chromosomes over ifs entire length.

Electrophoresis: Of the 16 proteins studied
(Table 2), LAP, PHI, Tf, Est. 2 and Est. 3
showed evidence of polymorphism in at least

one species, and 6-PGD, LDH A & B and PGM
were identical for all species. This left seven
of the protcins studied that were consistent
within species but varied between species. The
electrophoretic results for these seven proteins
are shown in Fig. 2, and the resulting difference
matrix in Table 3

112 P. R. BAVERSTOCK, J. T. HOGARTH. S. COLE AND J. COVACEVICH

TABLE 3

Difference marrix for data in Figure 2
(Key a3 in Figure 2)

Lj, Lib Lid LA, Pn, Pea.

Ly, 0

Li 3 0

Td. 6 5 0

Lik, 6 5 a 0

P.n. 7 & 1 n 0

Pa 6 7 4 4 4 0
Discussion

Most L. forresti were found to possess the
same karyotype, although two were hetero-
morphic for a sub-acroventric pair 1, C-banding
showed that the short arm on the sub-acrocen-
tric member was not heterochromatic, suggest-
ing that the sub-acrocentric was telated to the
acrocentric by a pericentric inversion.

L, lakedawnensis, however, had a pair of
small metacentrics converted to a pair of acro-
centrics, presumably by a pericentric inversion.
Although a single fixed chromosomal difference
between L, forresti and L. lakedownensis is in-
suffizient in itself to indicate a species dif-
ference, taken in the context of the very low
karyotypic variation of the whole of the
pseudomyinae (unpublished data), a single
chromosomal rearrangement probably indicates
reasonable differentiation,

The biochemical data are mere convincing,
Of the seven uscful proteins studied, L. forresti

and L. lakedewnensis differ in 3 (Table 3),
This is considerable compared to the biochemi-
cal differentiation between 3 “good” species—
L. delicatula and L. novaehollandiae (1 dit-
ference), and £. novaehollandiae and L- her-
mannshurgensis (1 difference). These results
suggest that £. forresti and L. lakedownensis
may have been separated from each other for
at Ieast as long as have L. delicatula, L. navae-
hollandiae and L. hermansburgensis.

Phenetically L. lakedownensis and L. forresti
are biochemically more similar to cach other
thai either is to any other pseudomyin studied
(Table 3). Although more data are needed
these results support the maintenance of Legga-
dina aS a separate genus which at this time
would include only these two species.

Acknowledgments

We are grateful to Dr C. H. 8. Waits for
helpful comments during the course of this
study. The original specimens of L. lakedown-
ensiy were collected by R, Buckley and
presented to the Queensland Museum by C.
Tanner. We thank the various State Wildlife
authorities for permission to collect specimens,
several of which were collected by A. & J.
Robinson under an Australian Biological
Resourees Study Grant to C, H. S. Watts and
P. R. Baverstock.

References

ArricHt, F. E., & Hsu, T. C. (1971 ).—Localiza-
tion of heterochromatin in human chromo-
somes. Cytogenetics 10, 81-86.

Brewer, G. J. (1970)—“Introduction to Tsozyme
Techniques,” (Academic Press: New York
and London).

Hoimes, R. S., Coorer, D. W., & VANDEBERG,
J. L. (1973) —Marsuptal and monotreme
lactate dehydrogenase isozymes: phylogeny.
ontogeny and homology with eutherian mam-
mals. J, Exp. Zool. 184, 127-148.

SELANDER, R. K,, SMirH, M, H., Yano, &, Y.,
Jornnson, W. E., & Gentry, J. B, (1971).—
Biochemical polymorphism and systematics in
the genus Peromyscus. €. Variation in the old-
field mouse (Peromyscus pelionotus), Stud.
Genet. V1, 49-90 (Univ. Texas Publ. 7103).

Warts, C. H, S. (1976).—Leggadina lakedownen-
sis, a new species of murid rodent from north
Queensland. Trans. R: Sac. S. Aust. 100(2),

VOL. 100, PART 3 31 AUGUST 1976

TRANSACTIONS OF THE

ROYAL SOCIETY
OF SOUTH AUSTRALIA

INCORPORATED

CONTENTS

Foster, R. J., and Philip, G. M. Corystus dysasteroides, a Tertiary Holasteroid
Echinoid Formerly Known as Duncaniaster australiae - - 113

Crisp, M. D. Rediscovery of Acacia barattensis J. M. Black (Mimosaceae)
in South Australia - ~ - - - - - - i117

Mawson, Patricia M. Woodwardostrongylus obendorfi new species (Nematoda:
Amidostomatidae) from Kangaroos - - - - - 121

Callen, R. A., & Tedford, R. H. New Late Cainozoic Rock Units and Depo-
sitional Environments, Lake Frome Area, South Australia - 125

Glaessner, M. F. A New Genus of Late Precambrian Polychaete Worms from
South Australia - - - - - - - - - 169

PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS
STATE LIBRARY BUILDING
NORTH TERRACE, ADELAIDE, S.A. 5000

CORYSTUS DYSASTEROIDES, A TERTIARY HOLASTEROID ECHINOID
FORMERLY KNOWN AS DUNCANIASTER AUSTRALIAE

BY R. J. FOSTER* AND G. M. PHILIP}

Summary

FOSTER, R. J., & PHILIP, G. M. (1976).- Corystus dysasteroides, a Tertiary holasteroid echinoid
formerly known as Duncaniaster australiae. Trans. R. Soc. S. Aust. 100(3), 113-116, 31 August
1976.

The type specimens of the nominal species Rhynchopygus dysasteroides Duncan 1877, Holaster
australiae Duncan 1877, Holaster difficilis Duncan 1887 and Galeraster australiae Cotteau 1890
(which include the type species of Corystus Pomel 1883, Galeraster Cotteau 1890 and
Duncaniaster Lambert 1896) are discussed and illustrated. All are included in one species correctly
designated Corystus dysasteroides (Duncan).

CORYSTUS DYSASTEROIDES, A TERTIARY HOLASTEROID ECHINOID
FORMERLY KNOWN AS DUNCANIASTER AUSTRALIAE

by R. J. Foster* and G. M. Putiiipt

Summary

Foster, R. J,, & Pitre, G. M, (1976).—Corystus dysastereides, a Tertiary holasteroid
echinoid formerly known a3 Duncaniaster australiae, Trans. R. Soc. S, Aust, 100(3),
113-116, 31 August 1976.

The type specimens of the mominal species Rhynchopygus dysasteroides Duncan 1877,
folaster australiae Duncan 1877, Holaster difficilis Duncan 1887 and Galeraster anstratiae
Cotteau 1890 (which include the type species of Corystus Pomel 1883, Galeraster Cotteau
1890 and Duncaniastér Lambert 1896) are discussed and illustrated. All are included in one

species correctly designated Corysius dysasteroides (Duncyn)-

Introduction

Holasteroid echinoids are not abundantly
represented in the diverse Tertiary echinoid
fauna of southern Australia, but there is one
common species'which. for the last eighty years
has been known as Duncaniaster australige
(Duncan). The purpose of this note is to
review the complex nomenclatural history of
the species and to decide on its correct
designation, Also, photographs of the type
material of four nominal species proposed by
Duncan (1877, 1887) and Cotteau (1890) are
published for the first time.

The species is known from the Tertiary
coastal basins of southern Australia from Eucla
Basin in the west to Torquay Embayment in
the east, and from New Zealand. The earliest
known Australian occurrence js in the Middle
or early Late Eocene; it is present in the Wilson
Bluff Lumestone al the Bluff and in Abrakurric
Cave, and in the Tortuchilla Limestone and
equivalents of the Si Vincent Basin, Jt makes
its last Australian appearance in the late Barly
Miocene (uppermost Longfordian) Watac-
poolan Limestone at Koonalunda in western
Victoria. The species also occurs in the South
Island of New Zealand; it appears first near the
base of the Weka Pass Limestone in the Early
Oligocene (questionable Whaingaroan), and
jast in the Gee Greensand in the Late Oligocene
or Early Miocene (Waitakian-Otuian). More

* BHP Oil and Gas Division, Melbourns, Vic, 3000,

stratigraphic details are given in a separate
paper (Foster and Philip, in press)-

Historical review

Dunean (1877, p, 49). described the species
Rhynchopygus dysasteroides from Castle Cove,
Victoria (Late Eotene Castle Cove Limestone)
and (1877, p, 31) described a further species,
HAfolaster australiae from the same jocality. The
holotype of R; dysasteroides is crushed, and it
Was ptesumably for this reason that Duncan
regarded the specimen as a cassiduloid. Pomel
(1883, p. 61) proposed the genus. Corystus for
R. dysasteroides because of its intercalary apical
system. In his revision of the Australian
echinoid fauna Duncan (1887, p. 421( pro-
vided a corrected woodcut of the apical system
of the holotype of H. auxfraliae. He recognised
that he had misinterpreted the species R. dysas-
teroides and been mistaken about its affinities,
As a consequence he renamed it Holaster
difficilis. Pomel'’s work was not well known at
the time and it is no doubt because of this
Duncan made no mention of the genus
Corystus.

Cotteau (1890, p, 548) described Goleraster
adustraliae from Mount Gambier (Early Mio-
cene Gambier Limestone) as a new genus and
species, placing the genus Galeraster close to
Holaster. Tate (1891, p, 276) first suggested
that A. difficilis and H. australiae were the same
species. In 1892 Bittner (p, 359) rejected the

t Department of Geoloxy and Geophysics, University of Sydney. N.S.W. 2006.

1t4

genus Cerysivs, noting Gregory’s (1890, p.
990) reference to 4. difficilis as an “unsatisfac-
tory species”, Also, in 1892 Tate published his.
strongly worded ¢riticism of Bittner's paper but
in regard fo these species he followed Bittner,
although he suggested that Galeraster australiac
oe an additional synonym of Holasrer ausira-
fae.

Lambert (1893, p. 97) transferred Al. aws-
iraliaeé 10 Pomel’s genus Lampadocorys but
later (1896, p. 317) made it the type species of
his new genus Duneaniaster which he placed
clase to Sregaster. Thus was created the widely
used name Duncaniaster australiae, In 1903
Lambert (p. 32) grouped the genus with Lam-
padocorys, Stegaster, Tholaster and Ojjaster in
his subfamily Echinocorynae.

Lambert & Thiery (1921, p. 332) recognised
Galeraster as a valid genus in the Bchino-
galerinae, stating (1924, p. 408) that Tate was
mistaken when he made Galeraster australiae a
synonym of Holaster australiae. They (1921, p.
364) reinstated the species Rhynchoprous
dysasteroides, and made Corysrus Pomel a
synonym of Rhyachepyeus dOrbieny, Last of
all (1924, p, 408), they relegated Duncaniaster
Lambert to a sub-genus of Cibaster Pomel.

H. L. Clark (1946), in his review “The
Echmoderm Fauna of Australia” mentioned
neither Corystus nor Galerasrer. He maintained
Duncaniaiter as a separate genus (p. 361}, but
did not consider it far removed from
Cardiaster; the only species he listed was D.
australiae (Duncan), Mortensen (1945, p, 84)
Tetained Cotteau's genus Galeraster in the
family Echinoncidae Wright and close to
Pyrina, but (p, 203) considered Corystus to be
a synonym of Cassidulus. He confirmed (1950,
Pp. 74) Duncaniaster in the Holasteridae close
to Cibaster. Wagner & Durham (1966, pp.
U445S U528) in the Treatise followed Morten-
sen in their placement of Galeraster and
Duncaniaster, and Corystus was tentatively
plated among the cassiduloids as a douhtful
nominal genus.

Type material

The holotype of Rhynchopygus dyavteraides
ts BM, E42418 (Fig. 2 C, B, F) and that of
Holaster australige is BM, E31067 (Fig. 2 A,
B, BD}. Both are lodged in the British Muscum
(Natural History), and both were collected
from the "No, 5 Upper Coralline Beds, Castle
Cove, near Cape Otway” in Victoria. This is
the old locality AWS of Wilkinson (1865) in
the Castle Cove Limestone, which Carter

R, J. FOSTER AND G. M. PHILIP

Fig. |. Plating of apical system of holotype of
Halaster austraiiae Duncan (BM £31067),

(1958, p, 21) refers to as his Foraminiferal
Units 2 and 3. The echinoids are probably from
the upper part of the formation in the latest
Late Eocene,

As indicated above, the type specimen of R-
dysasteroides is badly crushed, although the
adapical surface shows an holasteroid apical
system, similar to that of A. australiae (Fig,
1). In both spectmens the adoral surface is
poorly preserved. and the plastronal plating. is
obscure. Because of the state of preservation,
the presence or absence of a subanal fasciole
could not be established.

The holotype of Cottean’s Galeraster aus-
fraliae is an unnumbered specimen in the Ecole
des Mines, Paris, in the Cotteau Collection
(Fig. 2 G, H, 1), Its locality is “Mount Gam-
bier, Australia” and doubtless is from the
Gambier Limestone. The type section in the
sinkhole at Mt Gambier town is of Longfordian
(Barly Miocene) age, and Janjukian (Late
Ohgocene) outcrops are limited to restricted
areas NW and SW of the town. The precise
locality of Cotteau’s type, and of the only other
representatives of the genus from this forma-
tion {P20456 from the National Museum of
Victorian and T267a from the Tate Collection
labelled “Helaster woedsti Mt Gambier’), is

CORYSTUS DYSASTEROIDES, A TERTIARY HOLASTEROID

ECHINOID 115

Fig. 2. All natural size. A, B, D. Adapical, adoral, and lateral views of holotype of Holaster australiae
Duncan (BM 531067). C, &, F. Adoral, lateral and adap

ical views of holotype of Rhynchopy-
gus dysasteroides Duncan (BM E42418). G, H, I. Lateral, adapical and adoral views of holotype
of Galeraster australiae Cotteau.

1L6

nor known. The general echinoid fauna
presently available from the Gambier Lime-
stone appears to have its closest affinities with
that of the Longfordian Mannum Formation of
the River Murray cliffs In particular, 1267a
the only well-preserved specimen of Corystjis
from’) Mount Gambier was elsewhere {Foster &
Philip, in press) compared statistically with the
populations from a number of south-eastern
Australian localities ranging from Late Eocene
fo Early Miocene, and its parameters corre-
lated best with samples of populations from
the Mannum Formation and the Longfordjan
portion of the Port Vincent Limestone. It is
therefore concloded that the holotype is
probably from the Early Miocene. Again the
holotype is a poorly preserved specimen, It is
worn and cracked and a number of borings
occur in parts of the test. Sorface detail is
obscured by matrix and secondary calcite to the
degree that even the paths of the ambulacra are
difficult to trace, Preparation of the apical
region of the specimen showed the widely
separated oculars typical of an holasteroid
apical system,

R. J, FOSTER AND G. M, PHILIP

Conclusions

Despite the unsatisfactory nature of the type
matetial, We conclude that all specimens are
conspecific, We base this conclusion on the
large collections of the species available to us
from various localities in south-eastern Aus-
tralia. We here choose dysasteroides as the valid
mame for the species as it has page precedence
over qustraliae which was introduced by Dun-
can in the same publication, Pomel’s genus
Corystus has priority over Duncaniaster Lam-
bert. ‘Thus the valid Linnean species is Corysties
dysasteroides (Duncan).

In a further paper (Foster & Philip, in press}
we present a statistical analysis of samples of
Corystus populations ranging from Late
Eocene to Early Miocene in age. This analysis
is designed to depict the morphological trends
apparent in the evolution of the species. We
also have in preparation a taxonomic study of
all the holasteroid echinoids known from the
Tertiary rocks of Australia (including Western
Australia) and New Zealand. In this latter
article we will review the affinities of the genus
Corystus.

References

Brrrnun, A. (1892).—Ubher Echiniden des Tertiirs
von Australien. Sinz, A. Akad. Wiles. Wien,
(Math, Natur, Cl.) WI, 33371.

Carter. Av N, (1958),—Teniary Foraminifera
from the Aire District, Victoria. Bull. Geol.
Surv, Vict. 55, |-76.

Ciark, H. L. (1946),—The Echinoderm Fauna of
Aastralia, Its composition and origin, Publ,
Carnegie fast. Wash. 366, 1-567.

Correau, G. HL (1890).—Echinides notiveaux ou
pea connus, 2(9) Mem. Sov. Zool, France,
537-350,

Duncan, P.M, (1877)—On the Echinodermata of
the Australian Cainozoic (Tertiary) Deposits.
Quart. J, Géal, Sov. Lond, 33(1), 42-71.

Duncan, P. M. (1887),—A_ revision of the
Echinoidea from the Australian Tertiaries,
fhid. 43(3}, 411-430.

Fostra, R. J, & PHtur, G, M, fin press) —
Statistical analysis of the Tertiary holasteroicd
Corystuy dysaneroides from Australasia.
Thelassia Juwesl.

Grecory, 1. W, (1890),—Some additions to the
Australian Tertiary Echinoidea. Geol, Mav-
27, 481-492,

Lamperr, 3. (1893).—Etudes morphologiques sur
les plastron des spatangides. Bull. Sec, Sed,
hist, aot. ¥onne 47, 55-98.

Lamuert, |. (1896)—WNote sur quelques Echinides
crétaces de Madagascar. Bull. Soc. Geol.
Franoc 24, 313-332.

Camocrr, J. (1903).—Description des échinides
crétacés de la Belgique, J, Etude mono-
graphique sur le genre &chinocorys. Mém.
Mus, Roy, d'Hist, Nar, Belg, 2, 1-151,

Lamuerr, J., & Trtery, P. (1909-1925) —“Essai
de Nomenclature ralsonnée des Echinides” 1-9
{Libraire Ferriére: Chaumont).

Morrensen, T. (1948).—"A monograph of the
Echinoidea” IV (1) Holectypoida, Cassidu~
loida (C, A. Reitzel: Copenhagen),

MORTENSEN, T. (1950),—Idem V (1) Spatangoida

Power, N, A, (1883).—"Classification méthodique
et genera des échinides vivants et fossiles"
{ Adolphe Jourdan: Alger}.

Tate, R. (1891),—A bibliography and revised list
of the Described Echinoids of the Australian
Eocene, with descriptions of some new species.
Truns. Prac. R, Soc. 8. Aust, 14, 270-282,

Tate, R, (1892)—.Critical remarks on A. Biltner’s
“Eehiniden des ‘Tertiars von Australien”,
Sbid, 1§(2), 190-194.

Waonek, C, D. & Durnam, J. W, (1966).—
Holectypoids, Holasteroids.. Jn> Moore, R. C.
(Ed.) “Treatise on invertebrate palaeon-
tology” U3(2), 440-0450 & U523-U543.

Wirginson, C, S, (1865).—Report on the Cape
Otway Country (from Parl. Papers, 1864-
teal Repr. Geal, Surv, Vier, 1863-1864, 21-

REDISCOVERY OF ACACIA BARATTENSIS J. M. BLACK (MIMOSACEAE)
IN SOUTH AUSTRALIA

BY M. D. CRISP*

Summary

CRISP, M. D. (1976).-Rediscovery of Acacia barattensis J. M. Black (Mimosaceae) in
South Australia. Trans. R. Soc. S. Aust. 100(3), 117-120, 3 1 August 1976.

Acacia barattensis, previously known only from the type collection, has been rediscovered near the
type locality in the Flinders Ranges, South Australia. It is described in more detail than previously,
the legumes and seeds for the first time. Its taxonomic affinities and the type material are discussed.
Possible reasons for the species remaining "lost" for so long are suggested, and its state of
preservation is discussed in relation to its distribution and ecology.

REDISCOVERY OF ACACIA BARATTENSIS 5, M. BLACK (MIMOSACEAE)
IN SOUTH AUSTRALIA

by M. D. Crise*

Summary

Cer, M, D. (1976),—Kediscovery of Acacia baratrensis J. M, Black (Mimosaceae) in South

Australia. Trans, R. Sec. S. Aust. 100(3), 117-120, 31 August 1974,

Acacia barattensis, previously known only from the type collection, has been rediscovered
near the type lucalily m the Flinders Ranges, South Australia. It is described in moré detail
than previously, ihe legumes and seeds for the first time. Its taxonomic affinitics and the type
material are discussed. Possible reasons for the species remaining “lost” for so long are
suggested, and its state of preservation js discussed in relation to its distribution and ecology.

Introduction

Recently [ rediscovered Acacia barattensis
in the Flinders Ranges, South Australia. In so
doing [ established that the species was nol
extinct as previously supposed (Specht et al.
1974, p. 304), and that the locality informa-
tion given by its author was inaccurate. The
following formal taxonomic treatment? contains
new information about its morphology. type
material, affinities, ecology and distribution.

Taxonomy of Acacia barattensis

Acacia barattensis J. M, Black in Trans, R.
Sac, S. Aust, 56:42, t.1, fig, 2 (1932); Flor. 8.
Aust. ed. 2:419, fig. 578 (1948).

Somewhat diffuse, spreading shrub 2-3 m
tall with several slender branches arising at or
near the base, Breunchlets slender, glabrous,
gently curved so that the tips stand erect;
strongly angular and very viscid towards the
Ups; becoming terete, less viscid and faintly
striate towards the bases; greenish-brown but
usually covered with a black incrustation.
Foliage with a strong odour of resin acids
when dryy. Stipvles absent. Phiylledes erect,
natrow-linear but tapering slightly towards
beth ends, vertically flattencd, never terete,
gently incurved, abruptly rostrate-uncinate at
the upices, (3)5-8 cm long, 08-15 mm
broud, S-nerved on cach face, 1-nerved on
each margin (8-nerved in all); with a narrow
gtoove which is usually filled with brown resin
ahove and along each nerve; often shallowly

and irregularly sulcate between the nerves:
very sparsely and mihutcly pusticulate, initially
very viscid, Marginal gland scarcely visible,
2-3 mm above the base of the phyllode.
Peduncles 1-2 in the axils, each with a minute
narrow:triangular basal bract (0,5 mm long),
slender (0,2-0,3 mm diam,), terete, + papil-
lose-viscid, 8-13 mm long. Heads globular,
5-7 mm diam, at anthesis, ca 20-flowered
Floral bracts navicular with long (ca 0.5 mm)
triquetrous claws and with extended acute
apices, 0.8~-0.9 mm long and 0.2 mm bread,
densely papillose-viscid, the margins scarious,
+ entire, Flowery 4-merous, Calyx ca 1/3
length of corolla, consistently divided for 1/4
its length into triangular and barely acute lobes
with entire margins, papillose-viscid. Petals
oblong, recurved at the acute tips, 2 mm long,
faintly unimerved, + papillose-viscid, Joosely
connate for ca + their length. Ovary = sessile,
densely papillose-viscid. Leeurme narrow-linear,
coriaceous, straight or slightly curved, stipt-
tale, contracted to 4 its width between the
seeds, (6)8-10(15) em long, 2.5-3.5 mm
broad, often with a very narrow (1 mm) and
elongate (up ta 1 em) black sterile tip. Sur-
face of the legume initially very viscid, finally
not viscid, dark brown, with anastomosing
raised veins hetween the seeds, irregularly col-
liculate over the seeds. Margins of the legume
much thickened, straw coloured. Seeds longi-
tudinal, oblong-elliptic, smooth, dark brown,
ea 2.5 x 4.5 mm, Ari! much dilated. with

* Herbarium, Canberra Botanic Gardens P.O. Box 158, Canberra, A.C.T. 2601,
1 Personal communication: 1. Elix, Dept of Chemistry, Australian National University, A,C.T.

1i8 M. D. CRISP

E

C

Fig, 1. Acacia barattensis. A. Portion of branch in bud and flower. B. Phyllode. C. Transverse section of
phyllode, taken near the centre, showing resin-filled grooves. D. Apical portion of phyllode. E.
Inflorescence (note basal bract). F. Floral bract in lateral (left) and ventral (right) view. G.
Calyx and corolla. H. Legumes and peduncle. I. Exterior view of portion of legume over seed,
J. Seed, ari] and funicle. A-E from M. D. Crisp 889; F-G from M. D. Crisp 731; H-J from

M. D, Crisp 890.

REDISCOVERY OF ACACIA BARATTENSIS

hyaline wings, ance-bent above the seed and
gradually contracted through a sharp bend inte
a thrice-folded funicle.

Type citation; “Near Baratta head-station,
on a branch of the Siccus River and 20 miles
west of Kononamore.”

Holonpus: J. B. Cleland, 3.4i1,1930, “North
of Baratta Head Statn." (Raratta is at 31°59'S,
139°06E), AD 97338071 (f.)! fsorvpiz AD
96247254 (“belongs to AD 97338071") (41.)!,
K.

Distribution; South Australia: Plinders
Ranges. Apparently restricted to the deep
gullics of an unnamed range forming the
northem and western boundaries of Biblianda
Station, ca 55 km E of Hawker and 16 km
NW of Baratta head station. All recent col-
lections have been made in the vicinity of the
peak known as “The Bluff".

Ecology; Apparently confined to oear-verti-
cal gorge walls, from just above creck level to
ca 30 m above, on skeletal soils between out-
cropping quartzite, Flowermg is apparently
irregular, having been observed in April,
October, November and December; fruiting in
October (immature) and November.

Additional specimens examined: Bibdliando
Stn. ca 2 km N of New West Bore (31°52'S,
139°OS' EE). M, D, Crisp T3L, 13.iv.1974. fi,
fr. & photo. (AD; CBG 060871—ortg, spec.);
Bibliando Stn, southern slopes of The Bluff,
West Bore Pudk (31°S1'S, [39°00 BE), M. D
Crisp 889, 20)%.1974, fl. & fr. (CBG 060873—
orig. spec., NSW); ibid., AZ, DB, Crisp 890,
20.x.1974, fi, & fr, (CANB, CBG 060872—
orig. spec., NSW, TL, US); Between Willipa
and Biblianiio, Af. G. Catford s.n., 2-xi.1974.,
ft. & fr. (AD 97448228—pro parted; ibid.,
M. G Carford som., 23.xi1.1974, fr, (AD
97448228—pro parle)

Affinities: In his original description. of
Acacia barattensix, Black placed it next to 4.
subporosa F. Muell. {the now scyregated 4,
cograta Domin). | consider it ta be much
closer to the group 4. gracilifolia Maiden et
Blakely, 4, wilhelmiana F. Muell., 4, helnisi-
ana Maiden and A. menzelii J, M. Black,
which apparently belongs in Bentham’s series
Calamifurmes. However it differs from all
these in its 4-mcrous flowers and in being
totally glabrous. Maiden & Blakely (1927) des-
cribe a 4-partite corolla for A. gracilifolia, tut
both their illustration and material examined
by myself hive a S5-partite corolla, In SE
Australia the most similar relative of A. barar-

119

fests is apparently A. merzelil, which has
stmilar phyllodes, glands, peduncles, legumes,
seeds and arils, However the latter differs in
having phyllodes shorter and usually terete, 2
fewer nerves per phyllode, bracts larger,
cucullate and prominently ribbed, flowers 5-
merous, floral bracts with short claws.

Discussion
The type materi] at AD consists of 2 sheets,
hoth annotated by Black. One of these (AD
97338071) hus two twigs of material and two
locality labels in Black's handwriting, Onc
label,
"North of Baratta
Head Siatn 3/12/30
(J. B. Cleland)
label marked "Kew 137°”,
appears to be the original information received
from Cleland, while the other,

“Baratla H.S. (on a branch of the Siccus
River & 20 miles W of Kaonamore.)”,

8 Black's transcription of the former for publi-
cation. There is no evidence to suggest that
the tWo twies are separate collections, Bath
twigs are identical in all respects, particularly
in the stage of flowering. The sheet also bears
extensive descriptive notes and drawings, and
a bold Jabel “Acacia Aaraztensis J. M. Black”,
all in his hand. Clearly this shect ts the holo-
type.

‘The second sheet (AD 96247254), origin-
ally kept separately in Herb. J. B. Cleland but
later transferred to AD, curries a locality label
similar to that on the holotvpe, and a twig
Wentical io the others, indicating that this
sheet is an isotype. The label is marked “Kew
2137" by an unknown writer, and

“Apparently a new sp., but if sent 1a Kew,

I scarcely know what to do—I.M_R,”
by Black. Cicarly it was after he wrote this
note that Black decided to describe: and name
the Acweia from the duplicate (holotype)
material, If any specimen had been sent pre-
viously to K. tt must have been returned,
because the only specimen of 4. buraltensix
now there is endorsed
“comm, J, M, Black, Jan 1933"

i.e. it Was sent there after publication of the
new species by Black, This third specimen
appears: in all respects to be a duplicate of the
holotype collection, and must be regarded as
A second isotype.

The ahove distussion shows clearly wiry
many attemprs to relocate Acacia haraitensix
falled. Whereas Cleland’s field locality was

120

“North of Baratta” (in the ranges where it
presently occurs and only about 10 km distant
from Baratta), Black’s published locality
erroneously focussed attention on the head
station itself, where it does not occur. The
populations of the Acacia are in fact restricted
to specific sites in the deep gorges of the
range at Bibliando. However, where popula-
tions do occur they are fairly extensive and
protected by the inaccessibility of their habitat.
Provided that no major disturbance occurs in
this range, Acacia barattensis is probably safe
from the threat of extinction for the immediate
future.

M. D. CRISP

Acknowledgments

I wish to thank the staff of the State Herbar-
ium of South Australia (AD) for the loan of
specimens and for assistance received. I am
also grateful to Dr Hj. Eichler of Herbarium
Australiense (CANB), who offered some use-
ful comments about the type material and
read the manuscript. Mr A. B. Court of the
Herbarium, Canberra Botanic Gardens (CBG)
offered many helpful suggestions. Dr A. Kanis
kindly located and supplied data from the
specimen at the Herbarium, Royal Botanic
Gardens, Kew (K).

References

MalweN, J, H., & BLAKELY, W. F. (1927).—Des-
criptions of fifteen new Acacias. J. Roy. Soc.
N.S.W. 60, 171-196.

Srecur, R. L., Ror, E. M., & BouGcuron, V. H.

(ed.) (1974)—Conservation of major plant
communities in Australia and Papua New
Guinea, Aust. J. Bot, Supp. Ser. 7.

WOODWARDOSTRONGYLUS OBENDORFI NEW SPECIES
(NEMATODA: AMIDOSTOMATIDAE) FROM KANGAROOS

BY PATRICIA M. MAWSON*

Summary

MAWSON, P. M. (1976).-Woodwardostrongylus obendorfi new species (Nematoda:
Amidostromatidae) from Kangaroos. Trans. R. Soc. S. Aust. 100(3), 121-123, 3 1 August 1976.

Woodwardostrongylus obendorfi n.sp. is described from the oesophagus of Macropus parryi
(type host), M. robustus, and M. rufogriseus. It is distinguished from W. woodwardi (Wood) chiefly
in having only 6 pairs of oral denticles instead of 16. Woodwardostrongylus Wahid is transferred to

the family Amidostomatidae, and the genus Cristaceps Mawson is placed as a synonym of
Woodwardostrongylus.

WCODWARDOSTRONGYLUS OBENDORF! NEW SPECIES (NEMATODA;
AMIDOSTOMATIDAE) FROM KANGAROOS

by Parricia M. Mawson*

Summary
Mawson, P. M. (1976).—Woodwardostrongylus obeadorfi new species (Nematoda: Amido-
stromatidae) from Kangaroos, Trats. R. Soe. S. Aust, 100(3), 121-123, 31 August 1976.
Woodwardostrongylus obendorfi n.sp. is described from the oesophagus of Macrepus
parryi (type host), M. robastus, and M. rufogriseus. It is distinguished from W. woodwaurdi
(Wood) chiefly in having only 6 pairs of oral denticles instead of 16, W’oodwardostrongylus
Wahid ts transferred to the family Amidostomatidae, and the genus Cristaceps Mawson is

placed as a synonym of Woodwardostrongylus.

Introduction

The genus Woodwardostrongylus was
erected by Wahid (1964, p. 184) for Pharyn-
gostrongylus woodwardi Wood, 1931, Mawson
(1971, p.. 174) not having seen Wahid's work,
proposed the genus Cristaceps for the same
species, pointing out that this genus was close
to Filarineme in the family Amidastomatidae
(serisu Inglis 1968). Cristaceps now falls as a
synonym of Woodwardostrongylus, but the
latier must be transferred to Amidostomatidae.

Nematodes recently taken from the veso-
phagus of three species of macropods have
heen identified as a new species of Wood-
wardosirongylus, In all three cases the worms.
were threaded through the oesophageal epi-
thelium so that care was needed to collect
them entire. This situation is similar, though
in the oesophagus instead of the stomach, to
that occupied by W. woodwardi in the two
recorded findings (Wood 1931; Mawson
1971). It is a locale which is likely to escape
all but the most careful dissections, so it is
possible that species of the genus are more
widely distributed than the records indicate.

T am very grateful to Dr Brian Coman and
Dr Tom Kirkpatrick who shot the kangaroos
and to Mr David Obendorf who first noticed
the presence of the worm.

The micrographs (Figs 10, 11) were taken
by E.T.E.C. Avtosean in the Central Electron
Optical Laboratory of the University of
Adelaide, I am indebted to Dr Karl Bartusek

of this Laboratory for help in taking the micro-
graphs, and to P. G. Kempster for developing
and printing them.

Woodwardostrongylus obeadorfi n sp.
FIGS 1-11
Host and Locality: Macropus parryi (Bennett)
(type host and M, rebustus (Gould), fram
Dorrigo, N.S.W.; M. rufogriseus (Desmarest}
from Warwick. Qld,

‘The wortns are thin and clongate, the males
15,9-16.7 mm, the females 24-26 mm in
length. The. body, especially of the female, is
widest in its posterior part. The rounded
anterior end bears a small round mouth, on
each side of which lie six prominent denticles
each associated with a plate-like sclerotisation
in the cuticle. The mouth leads to a thick-
walled buccal capsule or vestibule. The lumen
of this is narrow but wider dorsoventrally than
from side io side. The walls are faintly striated
transversely (more: distinctly in some specimens
than others), and are distinctly thicker pos-
teriorly than anteriorly. The cephalic papillac
and amphids are very small.

The oesophagus widens in its posterior half
to a very slight terminal swelling. It is 800—
900 ym long in the male, 900-1050 «xm in the
female. In the male the distance from the
anterior end of the worm to the nerve ring ts
320-400 pm, to the cervical papillae 300—
460 wm, and to the excretory pore 440-510
wm; in the Female these distances are respec-

* Department of Zoology, University of Adelaide, Adelaide, S. Aust. 5000.

122

PATRICIA M. MAWSON

200 pm

Figs 1,2 and 3—Head, in semi-cn face, lateral and ventral views respectively. Fig, 4—Oesophageal
région. Figs 5, 6, 7, and 8—Views of bursa. Fig, 9—Posterior end of female. Figs 1-3 to same

scale. Figs 5-8 to same scale.

tively 360-400 pm, 560-600 pm, and 500-
550 pm.

The bursa is only slightly lobed, closed
ventrally, and somewhat voluminous dorsally
where it extends so that the dorsal ray for
most of its length lies at right angles to the
long axis of the body. The arrangement of the
rays is shown in Figs 5-8. The genital pore is
of medium size, apparently without accessory
lobes. The spicules are 1700-2100 ym long,
the ratio body length: spicule length being
9.9-12.8. A gubernaculum is present.

In the female the tail is 180-220 ,m long,
conical ahd pointed. The vulva is shortly in

front of the anus, 300-350 «m from the pos-
terior end, The vagina is relatively long, up to
800 ym. Vaginal eggs measure 140-150 x 70-
80 wm.

The species is distinguished from W. wooa-
ward? mainly by the presence of only six pairs
of oral denticles instead of sixteen pairs, and
by the presence of the associated basal plates,
which are not seen in the type species. There
is also a difference in the site in which the
species occur in the body, W. woedwardi in
the stomach and the new species in the oeso-
phagus. In Md. parryi and M. rufogriyeus the
worms were numerous, but only one was
found in M, rebustus.

WOODWARDOSTRONGYLUS OBENDORFI FROM KANGAROOS 1

Nw
ww

Fig. 10. S.E. Micrograph, anterior end almost en face (x 1500).
Fig. 11. S.E. Micrograph, part of region around mouth, showing two of the submedian papillae, the
mouth, and some of the oral denticles (x 3500).

References
Inciis, W. G. (1968).—The geographical and Wann. S. (1964).—A preliminary revision of the
evolutionary relationships of Australian genus Pharyngostrongylus Yorke and Maple-
trichostrongyloid parasites and their hosts. J. stone, 1926. J. Helminthol 38, 180-190.

Linn. Soc. Lond. Zool, 47, 327-347.

Mawson, P. M, (1971).—Pearson Island Expedi- Woop, W. A. (1931).—Some new parasitic nema-
tion 1969.—8 Helminths. Trans. R. Sac. S. todes from Western Australia. Rep. Director
Aust, 95, 169-183. Inst. Animal Path. Univ. Camb. 1, 209-219,

NEW LATE CAINOZOIC ROCK UNITS AND DEPOSITIONAL
ENVIRONMENTS, LAKE FROME AREA, SOUTH AUSTRALIA

BY R. A. CALLEN* AND R. H. TEDFORDT

Summary

CALLEN, R. A., & TEDFORD, R. H. (1976).-New late Cainozoic rock units and depositional
environments, Lake Frome area, South Australia. Trans. R. Soc. S. Aust. 100(3), 125-167, 31
August 1976.

Five new rock units are defined for the Lake Frome area of South Australia.

The Namba Formation of Miocene age constitutes fine grained immature muddy sediments laid
down in a low-energy fluviatile and lacustrine environment, possibly partly estuarine or lagoonal.
Climate was subtropical or warm temperate with high rainfall, but seasonal aridity. Aphanitic
oolitic lacustrine dolomite and palygorskite are included in this sequence. The Flinders Ranges had
very low relief. The overlying and intertonguing Willawortina Formation represents alluvial fan
deposits with minor lacustrine phases, recording the beginning of the late Cainozoic uplift of the
Flinders Ranges, during which the Miocene lake was greatly reduced in area.

The Millyera Formation, constituting laminated ostracode bearing clay, fine sand, and charophyte
limestone, records lacustrine deposition during the Pleistocene. This took place in an enlarged
ancestral Lake Frome. The essentially fluviatile and aeolian deposits of the Eurinilla Formation and
Coonarbine Formation were deposited during the late Pleistocene and early Recent. Arid and
pluvial climates alternate in the late Tertiary and Quaternary. Drainage trends and the predecessor
of Lake Frome were established, closely approximating present day geography. During deposition
of the Coonarbine Formation the seif dunes of the southern Strzelecki Desert formed.

NEW LATE CAINOZOIC ROCK UNITS AND DEPOSITIONAL ENVIRONMENTS,

LAKE FROME AREA, SOUTH AUSTRALIA
by R. A, CALLEN* and R. H. TepFoarp{

Summary

CALLEN, R. A., & Teprorp, R. H. (1976) —New late Cainozoic rock units and depositional

environments, Lake Frome area, South Australia. Trans. R. Svc, S. Aust, 100(3), 125-167,

31 August 1976.

Five new rock nnits are defined for the Lake Frome area of South Australia,

_ The Namba Formation of Miocene age constitutes fine grained immature muddy sediments
taid down in a low-energy fluviatile and lacustrine environment, possibly partly estuarine or
lagoonal. Climate was subtropical or warm temperate with high rainfall, but seasonal aridity.
Aphanitic oolitic lacustrine dolomite and palygorskite are included in this sequence. The
Flinders Ranges had yery low relief. The overlying and intertonguing Willawortina Forma-
tion represents alluvial fan deposits with miner lacustrine phases, recording thé beginning of
the late Cainozoic uplift of the Flinders Ranges, during which the Miocene lake was greatly
reduced in area,

The Millyera Formation, constituting laminated astracode bearing clay, fine sand, and
charophyte limestone, records lacustrine deposition dying the Pleistocene. This took place in
un enlarged ancestral Lake Frome. The essentialfy fuviatile and aeolian deposits of the
Eurinilla Formation and Coonarbine Formation were deposited during the late Pleistocene
and early Recent, Arid and pluvial climates alternate in the late Tertiary and Quaternary.
Drainage trends and the predecessor of Lake Frome were established, closely approximating
present day geography. During deposition of the Coonarbine Formation the seif dunes of the

southern Strzclecki Desert formed,

Introduction

Mapping on the FROME (Callen 1975},
and CURNAMONA 1:250000 geological
sheets has resulted in differentiation of several
Tertiary and Quaternary rock units which can
be traced throughout the Lake Frome area
(the region south of Lake Callabonna between
the Flinders, Barrier and Olary Ranges). The
Eyre Formation has been detined previously
CWopiner et af. 1974). It lies immediately
beneuth the units described here for the Lake
Frome area and can be recognised over a
mucit wider region. The other units are at
present restricted to the Lake Frome region,
though correlation with units elsewhere,
especially in thy Lake Eyre Basin, is generally
possible on a firm basis.

There was a low divide between the deposi-
tional areas of Lakes Frome and Eyre, sug-
gested by the distribution of arenaceous
material in the Miocene rocks. The develop-

ment of this divide is clearly described by
Wopfner (1974, p, 6). Thus the Lake Eyre
and Lake Frome areas formed two distinct
depositional basins during late Tertiary times:
different sets of formal names are used for
rock units in each. In late Tertiary and
Quaternary times the Flinders Ranges, achieved
their present dimensions, completely separating
the two basins by a range of mountains.

This paper describes five rock units requiring
formalization under the Australian Code of
Stratigraphic Nomenclature (1973), com-
menting on the paleo-environmental inferences
to be drawn from them. The nomenclature
supersedes that shown on the FROME. geo-
logical map, relationships between units. now
being on firmer basis. The paper is divided
into: two parts, dealing with essentially Tertiary
and Quaternary units respectively. New geo-
graphic names have been formalized with the
Geographic Names Board of South Australiz

* Gealogical Survey of South Australia, Box 151 Eastwood, S. Aust, 5063.
t American Museum of Natural History, Central Park West, New York, U.S.A. 10024_

126

(pers. comm. 1973) and are designated with a
superscript wherever they first appear, thus:
Lake Namba*. Geologic names have been
cleared with the Central Registry (Canberra,
1973). The paper derived from a report by
Callen (1974)3 and an MLSc. thesis (Callen.
1976) 4. Additional stratigraphic data may be
found in this thesis.

Previous work includes the early geological
surveys of Selwyn (1360), Brown (1884) and
later of Jack (1930) and Kenny (1934), More
recently Ker (1966), Krinsley er al. (1968)
and Draper & Jensen (1975, in prep.) have
reported on hydrology and geology. The mar-
gins of the basin have been the subject of
regional mapping programmes by the South
Australian and New South Wales geological
surveys, on 1:250000 scale. Relevant to this
report are Leeson (1967)3, Firman (197134
and Coats (1973). A detailed basin study of
the older unit NAMBA FORMATION is in
progress and will be reported at a later date.
A preliminary account of the stratigraphy 3s
presented in Callen (1976), which gives the
structural and tectonic setting,

The terms used ta describe the sedimentary
rocks are those of Folk er af, (1970), ynless
indicated otherwise, Colours are given sym-
bolicallv in terms of Munsell Colour Code
(Geological Society of America 1951). A
relative scale was used for designating the
thickness of cross-bedding, as follows: very
small <J] cm, small 1-5 cm. medium 3-50 cm,
large 0.52 m, very large >2 m- In the
designation of contact features, core width
plices a limit on the interpretation, as it does
on maximum grain size: cobbles and boulders
are interpreted from the proportion and shape
of fragments ground down and broken by the
drilling operation, and nature of petrophysical
log response,

The older units (Pt i) were described mainly
from bores, the younger (Pt 11) from outcrop
Knowledge of the younger units was derived
trom detuiled investigation of over 100
trenched enlcrop sections, Where passible
units were raced between sections, Fossil soils
were an aid lo stratigraphic interpretation.

R. A. CALLEN & R. H. TEDFORD

The location of the sections is shown in
Fig. 1, Tables | and 3 summarizing rock unit
properties, palacontology and gtomorphology,
Svmbhols are in Fig, 2,

The subsurface sections were studied from
cores derived from bores drilled by the South
Australan Department of Mines and private
companies, Some percussion and rotary curting
were used to assist correlations, hut those
utilized for type sections were cored continu-
ously, and are availuble for inspection at the
South Australian Department of Mines Core
Laboratory, Petrophysical logs were run in all
cases, The lithological descriptions were sup-
plemented by binecular microscope exanina-
tion, and clay (x-ray diffraction) and grain
size (sieve and pipette) analyses were per-
Formed by Drs R, N. Brown and B, G
Stevenson respectively of the Australian
Mineral Develapment Laboratories,

The text is regarded as a supplement to the
diagrams and tables, descriptions in Tables 1
and 3 should be read first. Complete descrip-
tions of each scction are given in the appen-
dices, wherein the sequences are described as
they occur on the earth's surface—i.e.
youngest at top, oldest at base. Depths to the
top of cach unit or bed from the bore collar
are given. and the thickness is placed at the
start of its description, Jn cach unit. descrip-
tion of the dominant lithology ts capitalized;
followed by qualifying descnoptors referring to
each lithology in the same order.

Division into units in the reference scctions
is Intended as sn aid to identification of the
appropriate intervals in the descriptions (Scec-
tion 12, Fig. 3), fet a formal subdivision,
Core loss is indicated in the hore Ings (Fig, 3
sections 10, 1) & 12).

Pt. |\—Older Cainozoic Nock Units

A weneral definition of cach unit giving
satient Eealures, age and gcamorphic setting is
presented in Table 1, representative sections
in Fig. 3. Appendix | gives detailed desetip-
tions of individual anets

i Callen, R, A, (1974)—New Rock Units and Climate of the Cainozoic, Lake Frome area, South Auis-

tralia. §, Aust. Dept. Mines Rept. 74/75 unpub.

+Oallen. K, A. (1976).—Stratigraphy, sedimentology and uranium deposits of Ternary rocks, Lake
Frome srea, South Australia. M.Sc. thesis, University of Adelaide (unpublished)-

* Leeson, B (1967),—Geology of Bulcanoona 1:63 360 map area. 5. Aust, Dept Miiey Repl. RB 64/92

uligub,

‘Firman, 7, B, (1971) —Regional stratigraphy of surficial deposits in the Great Artesian Basin anit
Frome Embayment in South Australia. S. dws. Dept Mines Rept. RB 71/06 unpub.

CAINOZOIC ROCK UNITS 127

=
al Gs 4 = ee
| MARREE J, 5 fe CALLABONNA
2 BLANCHE =f

SCALE IN RILOMETRES
io ac} 19 zo 30 4c 40
ts

J

rhs, g MINAD LEUZ BORE

TAKE
CALLABONNA

Lute Yeurieros fy betes bes I
o

Late drag rirdery
STREZELECRI

thake Sim denen Jo ah

“ . '
wou (untnariow Cont! Gaye rbenf
‘ i
4 By ii
’ Pre

bebe Cooaer

u
\ a bite Hundetan
. Fade Cuesierione
y - _ oF ai Qi bets Bumtardon
i See IN ys “ .
fe aa WOOLTANA'L BORE @ ye ty ‘i DESERT
PONE SN UNS tO Paes ni (ft Foie
Ree: Wareinony” i tate ovary Te
a w Hi
E i ~ Late Mallanitpen Y
oy : Take Pulorrs |

NY PAURALIO HORE, 34)

2.
Tube Karp, 2 of
atinke Gonutary

4
etude Of dub Biman se
Wa eS ee

“ew ardvambe'

4 Avena)

ee

Late Anke

. 1 1 <2) anes
as Og a x 1 2 | SENS Eo
& af od ‘ean 5 OORE / vas ghute Yomtanein | . JAI, Wistamysieelea
frome Dees! a Lave tHrkarnatas aia alae : he
a Sir Shute Moyet Mf >
% sf il vas alae | on
2\.
5 |
} +

\

)
fhti'iurasive
it

LEGEND

TAINOFOIC # MESGIOR ___ oad

He Cretateons
ancien ove Preccmbero ced mune
weal Ceysteliee basemen!

2
bade Eyre '

)!
Hl Hinrent

LOGALITY Deane

" \
| Lal

$e
( Ligke Torrens.

YY

ou.

ULE ARSED
!

Wh of

OO ee az

reser! ben tvel Meineny selina —— oe

Edqeonl hiya leeal olin -__aeae
410

lyezotartipr sone Wet vwrhas oe

ren

leg tem 2 __ in

KILOMETRES

¥ Soo 400, MERC HIE
LOCALITY MAP 2A civeh

FA Colla

KILOMETRES
a 1000

Hrente bet) — ————— ee

Te EM Rope

Fig, 1, basally map, Lake Frome arca—location of type sections. Numbered sections shown in Figs 3,

" '

128 R, A. CALLEN & R. H. TEDFORD

LEGEND
LITHOLOGY SEDIMENTARY STRUCTURES

Cobbles, boulders S | Horizontally lominated

Pebbles Cross laminated

Very small scale (sets <lcm thick)
Very coarse grained sand—granules

vy | Small scale (sets lem--Sem thick)
Very fine grained—coarse grained sand

“o— | Medium scole {sets Scm—O.5m thick)
Silt

L Large scole Isets >0.5m thick)

7 :
arth | Burrows, burrowed contact, trace fossils

Bedding planes
Flot/wavy — Sharp vedding plone

TH Flat/wavy — Transition <lem

Intrafarmational clasts
Sandy limestone: (or dolomite)
@ BD) Silcrete, limestone clasts

Carbonate nodules (primary)

FOSSILS

Charophytes

Spores and pollen

Primary gypsum

Mica
Calites, pisolites

SECONDARY ALTERATION

Algal stems

Vertebrates (other than tish|—aqualic or terrestrial
24 Manganese staining
Fish bones

Secondary calcareous cement of
groundwater calcrete

Gastropads
Ferruginous or manganiferous matiling
{"matmorisation”)

Land snails
Calcareous paleosols '

(length indicates degree ot development)
Ostracods

Aboriginal artifacts

Emu egg shell

CLAY MINERALOGY

S = smeclite

Rl = randomly interstratitied cloys
M = mica/illile

K = kaolinite

ELECTRIC LOGS

SP = self potential
— Loyer of ailphide or iron and mangenese oxide R = point resistivily

74-1063 Del. 7S. R.A. Catlen S.A, Dept, of Mines

Fig. 2. Legend.

CAINOZOIC ROCK UNITS

LITHOLOGY

NAMBA FORMATION § (Derivation, Lake
Namba*?, CURNAMONA map sheet. “Namba”
is the Jadliauray aborigin! tribal word for bone-
fish).

The type section is Yalkalpo No, 1 bore
(section 12, Fig. 3) drilled by the South
Australian Department of Mines, Though not
typical in some respects, this is the only section
demonstrating the relationship to the Eyre
Formation (Appendix 1). The sequence is of
reduced thickness (56.60 m} compared with
that of the reference section in Wooltana No,
1 bore (170.09 m, Section 11, and Appendix
}, excluding unit 4 which is 68.24 m thick),
Section |! also contains 4 microflora important
in the age determination of the unit, The most
extensive outcrops are on the west shore of
Lake Tarkarooloot {Fig 14), where 26 m
are exposed, and at the south and north ends
of Lake Namba (e.g Section §, Fig. 15).
These are unsuitable for designation as type
sections, as only the uppermost heds are
represented.

The type section consists of a series of
eyelic sand/clay sequences in the lower part
totalling 32 m of mterhedded yellowish silt
and dark grey or olive clay. This ts overlain
by 24.8 m of burrowed yellowish silt and
intraformationally brecciated light olive clay.
The cyclic sequences constitute the following,
from the base of cach set upwards:
fa) Fine to medium-grained sand with small
to medium scale cross stratification,
Laminated silt to very fine sand with very
small scale cross-lamination.

A zone of dolomite or caleite patches or
a bed of dolomite.

A relatively thick dark grey clay with
uregular shiny-surfuced fractures [skew
planes of Brewer 1964} and scattered
patches and wisps of fine to coarse sands,
in Which grains are polished,

In this sequence (a) and (b} may alternate,
or either (a) or (b) may be absent. Unit 5 of
the type section represents a relatively com-
plete evele;
170m Altemating CLAY and SILT to SAND.

Sand very fine grained moderately sorted.
percentage increasing upwards. Grains
very angular, with crystal faces developed
on quartz. Beconles calcareous al top.
Bedding lenticular, with sedimentary
brecciation and = possible burrowing
aclivity. Obscure horizontal lamination at
top. 7% carbonate grains, rare mica,
Colowr 5Y6/1 mottled lOYR6*6.

(ob)
tc)

fd)

129

W.75m SANDY CLAY. Vertically streaked trai
sition zone from sand to clay. Fine sand
forms streaks and patches in clay. Very
poorly sorted medium silt, with modes in
clay and very fine sand sizes,

CLAY, black (SY1/1) and tough, with
characteristic frregular shiny-surfaced
fractures, and streaks af white carbonate.
Matted with orange brown colours which
SugeesE an irregular microstructure. Many
brown patches have well defined straight
boundaries, producing angular blocks
with dendritic or patchy internal struc-
jure, Unoxiclized clay in these blocks is
ereenish grey. Scattered patches of silt
and very fine sand are present, Upper
contact sharp, but disturbed, with partial
mixing into overlying sand,

The hurrowed silt is very finely laminated,
but this is often disrupted by burrowing, Very
fine sand sized maternal is the coarsest grain
size encountered. Colours are maiuly yellowish
grey to yellowish-white for silt or light olive
for clays, having a greasy lustre. Fractures do
not reach the degree of development of com-
parable structures in the black cluys of the
lower part of the sequence.

4.10 m

The burrow structures are « few millimetres
in diameter, containing convex-down Jamellae
usually Jess than 0.5 mm thick. They are
irregular and often branch, tending to be con-
centrated in certain hortzons. Many of the
homogenized clays have a churned structure
suggestive of bioturbation,

The top of unit 9 marks the last appearance
of the tough black clays characteristic of the
lower part of the formalion (Fig, 4}, Pre-
quently alunite (KAl4(SO,).(OH),) is
developed as lustrous white particles or patches
within the clay at the top of this unit. Above
unit 9. silts dominate over clay, and burrows
(Fig. 9). are more common.

The outcrop at Lake Tarksrooloo (section
13, Fig, 14) as situated on the western cliff
face, immediately north of the track-crossing,
on the route from “Frome Downs” to Black
Oak Bore, The lower part of the section is a
few tens of metres south of this track. The
two parts were corfelated Using continuously
traced bedding planes, and levelled with an
Abnocy hand level. The strata ure essentially
horizontul, as are those in the type section.

Nolable features of this outcrop afe the
interbedded gypsum nodules in the upper part
of the seclion, the presence of ostracode-
bearing aolitic dolomite associated with paly-
gorskite, burrowed fine sand beneath the upper
clay-dolomite sequence, the finely Jaminated

130

calcareous silt near the base of the sequence,
and the sharp contact with the upper tough
black clay, ‘These features, particularly the
last mentioned, are useful in correlation. In
section 12, the petrophysical logs indicate ihe
interval between units 12 and 13, which lacks
core, is probably silcrete, calcrete or dolomite,
The absence of palygorskite beneath it suggests
it may not be dolomite (this clay mineral is
invariably associated with dolomite elsewhere
in the basin. A turtle shell fragment in unit
10 supports a lithological correlation with
section §, if the black clay und ?dolomite are
also correlated, but this is not in agreement

R. A, CALLEN & KR H. TEDFORD

with the clay mineralogy. Section 13 shows the
typical dolomits—palygorskite association, and
trend towards illite deamination tn member two.

Typical of the Namba Formation outcrop
ate the brown chert nodules which cover the
breakaway slopes, and black manganese oxide
comling on the grains in sand heds Micro-
scapically, the chert sodules have structures
indicative of shrinkage and formation from
accretionary silica gel. The black stain as man-
ganése, Both these secondary effects are locul-
ized, occurring in sands cropping out in the
banks of stream valleys eroded in the Namba
Formation, prior to the deposition of the

Figs 4-9. Older units, Examples of Namba Formation lithology. Scales in mm and cm, Core sections.

Arbows point 10 top of section.

Fiz, 4. Section 12, Yalkaipo | bare, 125.00 m. Core, Dark grey clay with streaks of carbonate,

darker and lighter clay, and sand, some filling burrows or root holes. Vertical disposition of
patches Well-displayed, Kepresents swamp deposition or a lake deposit which has been subject
10 Ssubadrial exposure. Centripelal orientation of streaks is result of expansion of clay as it
enters the cuore barrel,

Section 3), Wooltarta 1 bore, 218.68 m. Section through core, showing upper contact of
laminated dolomite bed, Shrinkage and cracking of the dolomite has occurred, allowing pene-
tration Of the semi-fluld overlying clayey lime (CJ. Represents chemical sedimentation in a
lacustrine or marginal marine environment. Boundsries of carbonate fraements and laminae

. Section [1, Wooltanu 1 bore, $8.72 m. Core. Calearcous claystane with numerous burrdws in-

Alled with green-grey clay, Irregular shrinkage crack (C) has been infilled with semiliquid
clay which carrics carbonate particles. The clay-filled crack is itself burrowed. indicating
genesis soon after deposition, Represents combined chemical and detrital deposition in a mar-

. Section 11, Wooltana 1 hore, 122.00 m. Core. Fine lamination with typical alternation of sili

and sand, Very fine scale ough cross-lamination.. Quiel water deposition (migrating cipples!

. Wertaloana 1 bore, 142.00 m. Araldite peel of sectioned core, Typical example of small scale

cross-lamination in medium grained sand, partly disrupted by burrowing in upper part. Clayey
laminae alternate at base, Relief? coincides with porosity, though affected by varying thickness
of core acrass section, Cross bedding formed by ripple migration, in un offshdre bar or

. Section 12, Yalkalpo 1 bore, 22.60 m, Core. Top of bedding plane, Shows burrows along bed-

ding plane, with concave internal lamination (0). Represents quiet water depesition with bur-

(0-13. Oulcrop of Numba Formation and Willawortina Formatien, core of Willawortinn Fotma-

10 Vertebrate fossil float from the Namba Formation of L. Yanda on Eugigilla Créek. Vertebra on

far left (D> is riverine dolphin, on its right (L) are tungtish teeth and two fish spines (F}
Jates. A Cragment of bird bone (B) is

- South end of Lake Namba. Typical outcrop of Namba Formation. Gypsum nodule cappsing

(G) overlies thin nodular dolomite (white: LS), Greyish olive silty clay (grey) occupies. most
of section, Gragsed white beach at base of slope is very fine grained Iaminated sand (8), 30
em scale rests on upper contact in trench, Outcrop surface is covered hy gypsum nudules and
dlomite probably

. Baleanoona Ck, Willawortina Formation. Calcified mediwr ¢rossbedded sand Jens in cule

cxreous reddish brown very poorly sorted clay-sill. Note thin bedding in silt, Sand lens repre-
sents deposition from: higher powered streams, fine sediments are floodplain deposits, Scale 30

Fiz. 5
emphasized by inking,
Fiz. 6
ginal Marine Jagoon or lacustrine environment, with burrowing organisms.
Fig. 7
in o tidal, lacustrine or floodplain environment,
Fig 8
channel,
Fig ¥
fOowinle Organisms,
Figs :
Hon,
Fig.
In centre (F,C) are mainly crocodile scutes and turile r
on upper right corner, From base of upper unit of Namba Formation, Scale 30 cm,
Fig. 1
weathered clay, Sand represents channel of tloodplain deposition, clay and
Jacustrine.
Fig. 12
cm.
Fig. 13

. Section 10, WC2 bore, 68.75 m, section of core. Willawortina Formation shows large pebbles,

granules, very coarse silty and clayey sand. Extremely poor sorting. Represents deposition in
an alluvial fan environment, Scate in mm.

CAINOZOIC ROCK UNITS

© eran on liao te

Neoned

Hem vera

Mow

132

Millyera Formation and Eurinilla Formation
(new names see Pt. II).

Wooltana No. 1 bore (Fig. 3, Section 11
and Appendix 1), drilled by the Australian
Department of Mines is an important supple-
mentary section, exhibiting a thicker sequence,
lithologically more typical of the Namba For-
mation than the type section, It also demon-
strates the intertonguing relationship with the
Willawortina Formation (new name Pt. 1).

The base of the Namba Formation was not
penetrated, though cuttings from old Pootana
bore (Fig. 1, 50 km north-north-east of Wool-
tana No, 1 bore) indicate a total thickness of
190 m. This compares with 54.40 m in Yal-
kalpo No. 1 bore (Section 12). The sediments
have been divided into six informal units. The
lowest of these (unit 1) consists of 8.5 m of
laminated black and dark olive carbonaceous
clays with characteristic fauna and microflora
(discussed later), Laminae containing ostra-
codes of early Neogene aspect (including
cypridids—pers. comm. K. McKenzie 1973),
and fish spines are present. Protoconchs of a
small gastropod (Potamopyrgus s.1., see Lud-
brook 1972)5, are scattered through the clay
and ?gastropod tracks and burrows of other
organisms are common on_ bedding planes.
These sediments are restricted to the Poontana
Sub Basin west of Lake Frome.

Unit 2 (40 m) is dominated by white, fre-
quently oolitic, dolomite beds (Fig. 5) con-
taining characteristic branching pores 0.5 mm
diameter, alternating with clay, and sometimes
interbedded with silt and fine sand. The car-
bonates have unusual transitional or irregular
upper boundries: in some beds spherical zones
delineated by colour variations develop, which
pass upwards into discrete carbonate lumps
within the matrix of overlying unit. These are
thought to be diagenetic features associated
with lithification possibly resulting from inter-
mittent exposure. Other beds (Fig. 5) show
shrinkage cracks, into which the overlying
clay penetrates. Particles of carbonate are
included and flow lines occur, indicating
liquefaction resulting from thixotropic trans-
formation. The lack of rounding of the clasts
derived from cracking, and gradation to un-
cracked material, suggests sinking of carbonate
plates into underlying liquid clay. The cracking
may be a syneresis phenomena, which occurred
during or shortly after deposition of the over-

R. A. CALLEN & R. H. TEDFORD

NAMBA FORMATION
OUTCROP AT LAKE TARKAROOLOO

COONARBINE FORMATION
EURINILLA FORMATION

5YR5/6 +7

2,5YRS/6

5Y5/2-5/\

Stain 10YR6/2
5YR5/2 NI—N5
py Stain 1OYR8/6

a
ray
<=
a
ud
a
<
a
a
So
I
a
“4
2
wr
w
oc
K
a
=
Zz
Zz
fe)
e
<
>
us
—
ka

Lat. 30° 07' 00"
Long. 140° 06° 34”

LAKE BED

74—956 Del, BT R. A. Callen S$. A. Dept. of Mines

Fig. 14. Namba Formation—outcrop _ reference

section.

lying clay. Occasionally the clasts have been
rounded, and incorporated in the overlying
unit: current or wave action has been effective
in some cases. Other beds show wispy car-
bonate and clay intermixed at the contact,
interpreted as flame structures which have
transformed by thixotropic changes, to flow
as a semi-liquid. Bioturbation is frequently
associated with these structures, and is com-
mon throughout (Fig. 6).

Unit 3 (49.7 m) is very similar to the lower
part of the Type Section (section 12, units
1-9), exhibiting similar cyclic deposition, in
which cross-stratified sands (Fig. 8) grade up
into tough black clays with pockets of medium
sand, often with polished grains (see descrip-
tion of unit 4, section 12). The black clays
are identical to those in section 13, Fig. 14.
Analyses showed the black colour does not

3 Ludbrook, N. H. (1972). Age and environment of deposition of a sample from Yalkalpo No. 1 Bore,
Lake Frome area, South Australia. §. Aust. Dept. Mines Rept. RB 72/207 unpub.

CAINOZOIC ROCK UNITS

result from anomalous concentration of car-
bonaceous. matter, sulphides or manganese.
Tron-nich montmorillonite or humic acid stain-
ing are alternative explanations. A bed of
dolomite or limestane nodules is often present
at the contact between the sand and black clay,
Lamination (Pig. 7) is generally not as pro-
minent #8 in the equivalent sirata in the Lype
section, and the sand beds are often burrowed.

The cross-bedded sand sequence of unit 4
(49.2 m tetal thickness) grades up into a
uniform olive clay with churned structure, The
sand bed is a promincat horizon west of Lake
Frome, and is being prospected for sedi-
mentary uranium of the geochemical cell type.
The dark sandy clays with skew planes are.
rather weakly developed in this unit.

The upper carbonate horizon, unit 5, is
23.7 m thick. has a much higher proportion of
clay than unit 2, and is intensely burrowed.
Sedimentary gypsum laminae are present.

The uppermost part of the section (unit 6)
in which the Namba and Willawortina Porma-
Uons intertangtte is mote conveniently des-
cribed when discussing relationships between
units,

The Namba Formation has been broadly
divided into two informal members (1 and 2)
of regional extent, on the basis of the presence
Or absence of |he tovgh black sandy clays with
skew planes. The lower member (e.g. units
1~4, section |1, Fig. 3) ts charactenzed by
these clays, and cyclicity is more prorninent.
It was later found that this subdivision closely
coincided with the change from smectite to
hte-kaolinite dominated clay mineral suites
(inset, Fig. 3), except in Yalkalpo | bore
(Fie. 3, séction 12), In this bore it is uncertain
whether the dominance of smectite throughout
the sequence fepresenis a local variation in
clay mineralology or whether the upper part
has Seen wrongly assigned to member 2
(which may have been eroded). The
mineralogy in Yalkalpo 1 bore is remarkably
uniform, smectite almost the anly component.
The higher proportion of silt is also unusual.

An interesting, varied vertehrate fauna is
found in the upper part of member 1 and the
base of member 2 of the Namba Formation
in various small saltpans southeast of Lake
Frome, in the vicinity of Eurinilla and Billeroo
Creeks, One of these Iccalities is at Lake
Pinpa (Scction 8, Fig. 15).

HILLAWORTINA FORMATION (Deriva-
timn—Willawortina Creek, passing south of
“Wertaloona" on the Balcanoona High Plains.

133

in the vicinity of the outerop reference sec-
lion),

The type section for this unit is Western
Nuclear’s sedimentary uranitm test hole WC2
(Fig. 3, section 1D and Appendix 1) cored
from & m to base, The hole was drilied on the
uplifted plains flanking the Flinders Ranges,
near Paralana, where a continuous sequence
of coarse poorly sorted sediments is encoun-
tered. A detailed division is not possible as a
result of moderate recovery and gradational
contacts. Three members are recognized, mem-
bers 1 and 2 (16.4 m and 17,06 m thick
respectively) have less mica and sand in the
mater than the overlying beds, and are less
oxidized. Member 2 has finer overall grain
size than member 3 but is comparatively
coarser than member one. Members 1 and 2
are equivalent to unit 6 of section 11.

Although bedding planes are very mdistinct,
transitions In grain size are often abrupt (Fig.
13), Secondary alteration with production of
red mottling is common throughout. Feldspars
are generally more abundant than in the
Namba Formation, Sandy beds have matrix-
supported framework with a high proportion
ef framework compared with the Namba
Formation,

The Formation crops out along crecks in-
cised into the high level plains flanking the
Flinders Ranges, along ihe southern shore of
Lake Frome, and along the Siccus-Pasmiore
River. The section (Fig. 3 section |, Appendix
lL) in a low range of hills, 3.7 km on 22°T,
north of Prism Hill and south of “Wertaloona"
{Air photo reference: § Aust. Dept. Lands
Svy, 803, Baicenoona Run 7, photo 0014), is
an important supplementary section, aS it is
the only outcrop in which the contact with the
Namba Formation can be observed. The
sequence is 140 m thick and dips 30-50° cast,
in accord with the remainder of the Cainazole
section, The whole rests with angular uncen-
formity on Middle Cambrian cocks. Exposure
is moderate to poor, necessilaling reconstruc:
tion from several scattered outcrops, particu-
jarly through the Namba Formation. This
sequence Was fitst mapped by Leeson (1967)*
who referred the conglomerate to the Telford
Gravel (Firman 1963, 1964, 1966b, 1967a,
1970) and the underlying clays to the Avon-
dale Clay (Firman 1967a}, Subsequently Callen
tin Coats 1973) remapped the area during
1970-1 for the COPLEY 1:250.000, zenlogical
miap sheet, and the sequence was assigned to
an undifferentiated Teriary-Quaternary unit.

134

Elsewhere on the eastern portion of COPLEY,
green clay, now known to belong to the same
sequence, was called) Avondale Clay,

In Sectyon 1 (Fig. 3) the base of the Willa-
wortina Formation is placed at the base of the
Jowest conglomerate, Beds below this unit
include poorly sorted sandy clays, but with
interbedded mictitic white dolomite, fine yel-
low-green sand, and pale grey and olive clay,
closely resembling the Namba Formation.
Below these beds, resting with angular uncon-
formity on the gently folded Middle Cambrian
red beds, is coarse sand with polished pebbles
and ?ferricrete clasts resembling the Eyre
Formation.

Another section regarded as eguivalent to
the Willawortina Formation, but of ayeral!
finer grainsize, is exhibited by unit 6 of Wonl-
tana No, 1 bore (Fig. 3, scction 11). It shows a
prominent. alternalion of sand and clay in fin-
ing upwards sequences, cach separated by sharp
egntacts, Sorting is uniformly very poor, and
matted green and brown colours common.
Secondary carbonate nodules are present, and
also beds of lacustrine dolomite. Toward the
top of the section the fining upwards sequences
become poorly defined, The top is capped by a
thin dolomiie bed, overlain by cobble con-
glomerate and sandy clay silt, representing the
Evrinilla Formation and “unnamed conglo-
merate’ {probably equivalent to the Millyera
Formation )-

Upsiream from section Z along Balcanoona
Creck, excellent exposures (e.g. Fig. 12} of
the upper part of the sequence seen in section
11 are displayed in cliffs. One of these exhibits
a hiatus—limestone and conglomerate In. the
lower part have been faulted before deposition
of the overyling silts, Subsurface (below soil)
karst structure ts present.

RELATIONSHIPS BETWEEN
FORMATIONS

The ature of the contact between the
Namba and Eyre Formations, and difficulties
associated with differentiation when both units
are sandy, hive been discussed by Woptner
ef af (1974), The disconformable relationship
is demonstrated palynologically by W. K-
Harris (pers. comm. 1974, sce section on
AGE, this paper).

The intenongurng relationship between the
Willawortina and Namba Formations is ilus-
trated by Fig. 3 (imsct). a section across the
Panlana High Plain, on which Wooltana 1
bore has been superimposed, A similar section

R. A. CALLEN & R. WH. TEDFORD

showing the same [catures can be drawn
across the Balcanoona High Plain through
WTS, WTS and WT bores (Mines Admini-
stration Pry Ltd) and Wooltana I bore. The
decrease in course clastic: proceeding east
from the Flinders Ranges t% demonstrated,
The lower boundary of the Willawortina For-
mation has been drawn at the base of the
characteristic mottled, immature. poorly sorted
sediments. Note the varying electric log re-
sponse to similar lithological differenecs be-
tween bores, which results from differing
drilling mud properties and sensitivity, and in
the case of WC2 bore, different instrumenta-
tion. Holes F22-20 and E20-13 however, are
not affected by these variables and are directly
comparable,

In Wooltana | bore (section 11 Fig. 3)
intertonguing with the Namba Formation is
exhibited by unit 6. The typical Namba For-
mation lithology of sharply differentiated
relatively better sorted clay and silt beds
grades to the extremely to very poorly sorted
coarse grained Willawortina. Forination, The
two units alternate Lo some extent. Essentially
there is a gtadual upward increase in the
coarser grained fraction, though an isolated
pebbly bed appears fow jn the sequence. Clays
are rich in illite (muscovite) and feldspar Fs
abundant, compared with the bulk of the
Namba Formation where these minerals are
minor components and smectite the dominant
clay mineral.

Dnit 6 of section 11 is therefore interpreted
as the equivalent of the lower part of the
Willawortina Formation in section 10, a rela-
tionship suggested by the correlation Jines
drawn in. Fig. 2 of Callen (1976). The criteria
chosen here to identify the base of the Willa-
wortina Formation are those readily mappable:
the base of the consistently coarse-grained
poorly sorted sediments, Thus unit 6 as shown
on big, 3 is regarded as mainly Willawortina
Formation, though il contains tongues of lacus-
trine dolomite like those in the Namba Forma-
tion. The contact is readily recognizable from
petrophysical Jogs (Callen 1976 Fig. 2} and
can partly be explained hy the degree of
secondary altcration (carbonate nodules, iron
oxide mottling) striatigraphically associated
with the Willawortina Formation. These
secondary cffcets alternated with deposition,
and are an integral part of the unit,

Support for the inlertanguing relationship
between Namba and Willawartins Formation

CAINOZOIC ROCK UNITS

is aso derived from clay mineral analyses
(Callen 1976) Results are shown diagram.
matically on the inset of Fig. 3 demonstratlig
the abrupt change from rocks dominated hy
smectite and randomly interstratified clay, to
illite (largely well crystallized muscovite),
randomly interstratified clay and kaolinite.
This change corresponds to the position of the
tlunite horizon within the Namba Formation,
and is widespread throughout the basin, having
been focated in 14 bores and in outcrop. The
change was probably initiated by uplift of the
Flinders Ranges, probably with climatic varia-
tton frorn high to low rainfall as. indicated by
clay mincrology and colour change (see later).
It is therefore regarded as an approximate lime
marker, and is coincident with the boundary
between members 1 and 2 of the Namba
Formation, and with the base of the Willa-
wortina Formation in its type section. The
change corresponds with the base of the Willa-
wortioa Formation identified in WC2 and
Woollana 1 bore.

Alunite is recorded near the top of member
1 of the Namba Formation, forming a serics
of nodular horizons associated with sharp
bedding planes. The nodules ramify through
the clay and resemble calcareous hardpans of
soils in their manner of development. The
horizons ate widely developed in the Paralana
High Plains area, but are also fotind in the
¢astern part of the basin in C1S bore. Here,
they are overlain by a relatively thicker
sequence of member 2 than in the high plains.
The horizons arc regarded as soils, associated
With a well developed hiatus or disconformity
formed during uplift of member 1. This em-
phatizes the time significunce of the clay
mineral change recorded earlier.

Silcrete has been identified by one author
(R.A.C.) in the interval 72-94 m from cut.
lings of bore LBIZ, drilled by Mines Adminis-
tration Pry Ltd, It is developed on ctay, and
ovetlain by greenish-+red mottled sandy cal-
careous clay resembling the Willawortina
Fotrsation. A number of closely spaced bores
beiween “Murnpeowie” and Reedy Springs,
drilled by Peechiney Exploration (Australia)
Pry Lid (Mannont & Barral 1972)6 suggests a
similar relationship. The silcrete vartes from
the red und grey mottled chalcedonic and
opaline “‘puddingstene’ to the grey rmero-
crystalline quartz “groy billy” type, according

195

to whether clay or sand is siliciied. This is
displayed by Mannoni & Barral in their cnoss-
section, and can he observed in outcrop, The
same siletete horizon fennms a cup to the
dipping Eyre Formation ac Reedy Springs
(Woptier et al, 1974 Fig. 2), The silerete is
thought to represent a soil horzon, and there-
fore marks 2 disconformily (Callen 1976)2,

Thus there is eVidence supporting a discon-
Tormity between the Namba Formation and
rocks resembling the Willawortina Formation
in this area, Althougl the silcrete has fot been
identified in the high plains regions, it ts
apparent that the Willawortina Pormation, as
defined, may contain some younger material.

The brown silerete and ferruginous material
developed on sandy facies of the Namba For-
mation exposed at Lake Tarkarooloo, and
around other saltpans east of Lake Frome,
are thought to be cquivalent to (hat just de-
scribed, Cementation certainly occurred prior
to deposition of the Millvera Formahon, as
indicated by abundant silerete nodules and
ferruginized Namba Formation clasts in the
base of the channel facies in Lake Tarkaroolon.

AGE

The flora of member 1 of section 11 (Fig.
3) indicates an carly to. middle Miocene age
fur the base of the Namba Formation {Bates-
fordian-Balcombian—pers. camm.- W. K.
Harris 1974). Harris states the fiara is stmilar
to that of the Munno Para Clay of Lindsay &
Shepherd (1966), and Lindsay (1969, p. 38)
in the Adelaide Plains Sub-Basin, An assem-
blage of the same age was found in Mines
Administration Pty Ltd LCIA bore (for litho-
logics] deseniption see Wopfner ef ul, 1974)
to the oorth of section }1. and also in Luke
Evie 20 bore (Johns & Ludbrook 1963) ijn the
Etadunna Formation,

The age af the Waillawartina Formation,
accepting a conformable relationship with the
Namba Formation, is therefore medial Mio-
cene or younger. Its upper ave limit. as for
the Namba Formation, is deduced from rela-
tiggship ta the Millyera Formation, and
Eurinilla Formation (Pt Il) indicating a mm-
mim age in excess of 40000 years B.P.. pos-
sibly pre-Pliacenc.

4 Maunnoni, N.. & Barrul. J. M. (1972).—Murnpeowile Project 5.M,L. 373 (South Australia) drilling pro-
gram report R/72-21-U, S$, Aws!, Depi, Mines envelope 1327, Unpub.

R. A. CALLEN & R. H. TEDFORD

136

sau jo idaq -y 5

“AB|A WOR \sOuaddn yy)
PIWWICSSE SUOZIOY |X) HOS AIUN}Y
“DRMDIS. BY 2WADLY LO YMG MO||BA

Snduidtiiay AQ padded sauiialuog

“SoMUEY SJEPLI|Y 40 (NBII Wa|sim
AuQ\E poynejan sadvanbas Suiddip
ge Ajuniuiads $837) ‘aWOL4 aYe7
{0 JS¥HUINOS PUB YBa “SHIEK
Pur Saye| Alp Jo saapA Auge
*squN 70a ajeuoqed
AURPUGIBS Ju ajasvdAd ‘3p
Aq padded Ua}jo 344)/9

SUSE MO}RYS PUL JOYUELY peautsepim ly

‘BRIN DeS|
Uo Ives Yyoyed HORA 'Wydap ype

jo SyStuy galssepy “uwtipas Snolod
Sus 9p9!Te) (ayeMpUNVIT

{nq PEW Sains 9 yyosid
“#yaays jeueziny Jo speaipuyyés
flugy un gadajanap areungJeo

J ABSA) PATO|RAAD {IAM

SOS dILVIOOSS¥

"au AHR]

= TOR PIE SuBALy

Mowsey Smyats Zuoje syijp "syed
BuG|e $49 TUFUPYAO JO (PII,
SP ya sdOsy WAVE smethjernz Ww

{ymol8 Siladdns \/ ay ‘sazuey

NOISS4WdX3
JIHddOW0I9

“
2
a
Al
4
a
2
zg
8
a
2
R
.
8
2
4
2
2

=
&
=
=
3
2
|
&
=
2
=
Ss
a
&
=
Fs
R
uy
iy
Py
ny
cs
se
3
a

AYLAW049

SHLLYadOUd LINN HOO AO AUVWWNS ‘T ATEVL

tajind pur savods yuejd wos

(NYIBHODTWa-N¥iddOssalva)

(NOILYWYOI VaWWN

>} dysuoyeres WH) SNFIDOLSII Id ATW

Agwsscd ANADONdINAIOIN = 3LV7

ANAQOIW TVIGsW

yaoo yy

ed ed

"ARANRULINSHD “ARADEIO)SAY
'BRQUILI2I) “seepadpmangz
sAMesepad ‘wWinpAer
‘sndepayiuny ~S3iD0S |UE|d
“ROR

‘we /PLONGLEY "BYE IA OU,
‘sany ‘BYIpNAN]

RUUD NSOaa, “}OUdIG
Ops snfudctameyoy ‘Spi n0U
WIS ple PIUDOD salFIRI|
“(SmMUMIAQ) BpovE,|SO,

“YPAL ely BUnj—sayeaqaye,
“$peq Jamo) Ui sauids \ysl4
“SBIMpIOI] 12 \nBadst YyyM
Aeja Y3/Q Eno} ‘sayWojap
'sples pue syis auy

PUR SHES Laas haIT Arey

quaunuioud
WiniiSodap dXyoAQ ‘Sauunis
tie syoeta (snoonbeans,)
BBRYLYS Biey)
‘YOWUAd Loleqanaig
“YOUUWOT. SAINTS
May Hain “U
(evoneulznyesju) ‘padi
Gayo ‘sa}e
H|NA81I
li WOW Yulppsg-ssors.
9/205 WNIPAU UT [jews Als A,
‘PUES AU, AIDA. PUB
Ys Uy Aeradsa. ‘youiiion
LOBuNMe) jB}Uoz 104

Li99S FUE CS OILEADS Ele e-ENOES

, “WL GAS Ate “2 HL CHAD

BNON 9 C7 A/G ZAO7 1 “aUYysoLUy| SalE ley

BIBY ‘SSAUETUCW UIA PAU|} Se1od Mujyoueg Ag

fajenaved ae PUE ‘O/N:ONd}5 Is/NPOW ere) Ua}io

SAWL/ON “Andazeutgiks ay Ae SABI pi
“Aald 10 upala eyed

JO svinBue AjjesausA spues “aIGo Wau ‘spaq
SYWOIOP AU!) PUL ayO}Op wiy) ‘BK puE. ps
‘SpuRs payos AUOOd Winipaw oy uy Bulevsayy

‘18 [2G BSa—¥2

NOILVWHO4 VEWVN

“SRE, WEY

S1ISS04

NOWLWI3WY09 Od
VIHALIED

salae) dau) ‘sed
43M{ LY SapAD seman
SUIULY POL] Wf Spag.
‘sparp ‘$5022 apbas Uppal
paUYaD |18N\

pee vem apg

SIUMLONULS
AXVLNGWIGS

BG SALG 2 UR EMAL TA

y8ea Ja jUMp Epa pubs 'saTUeY S9Pl|4 Wea

i) 2d) Dy sagged jo Sasuay

SHOP Kapinan ‘sajqqad pasyye9s

UM FP WIPED alum ysivaBa alee yO $18AYs PLE
0 FA uoyayays

AU RUEW Tiddfarua saynpan ayenened
YsWwea!d squall aim Aeya yes AQnad

Gp AlaD|NOD PANE AWDOd Alem OY AyaUIa eT

ADOTOHLIT

NOILVAOS
VNILYOMVTIIM

CAINOZOIC ROCK UNITS

REGIONAL CORRELATION

Relationships with other units used on adja-
cent South Australian Department of Mines
Geological Atlas Series map sheets (COPLEY,
PARACHILNA) and in other basins, are
shown io Table 2,

Equivalence between the Etadunna and
Namba Formations is demonstrated by litho-
logical similarity, similar flora, and occurrence
of species of fossil marsupials previously
known only from the Etadunna Formation,
Both contain the unusual dolomite-palygorskite
mineral assemblage.

A sequence penetrated during drilling opera-
lions by Carpentaria Exploration Pty Ltd
immediately west of the Ediacara Fault { Binks
1972) is Very similar to that encountered in
Wooltana No. 1 bore (section 11, Fig, 3) in
the Lake Frome ares. The section in Binks’
Fig. 3 has been interpreted by one of us
{R.A.C_) thus: 0.0 to 94.8 m 7Willawortina
Formation equivalent, 94,8 to 121,0 m—un-
named beds, 121.0 to 233.2 m—Etadunna
Formation equivalent, 233,2 to 298.7 m—Eyre
Formation, The sequences in the intermontane
Walloway and Willochra Basins {Howschin
1909, 1913; O'Driscoll 1956) are more diffi.
cult to compare lithologically, but palynology
(Hartrs 1970)7 from 30 m in Willochra No, 2
bore suggests most of the sequence is equiva-
lent to the Eyre Formation,

On the northwestern side of the Flinders
Ranges is the Avondale Clay (Firman 1967a)
of similar lithology and mineralogy to the
green clays of the Namba Formation and
Willawortina Formation, patticularly where
they intertongue (unit 6, section 11, Fig. 3).
The type section is affected by secondary iron
oxide mottling, and the "clay" is actually a
clayey fine sand, with angular shiny grains,
The relationship between Avondale Clay and
Ejadunna Formation is unknown at the type
area: the base js not exposed, and the unit
is unconformably overlain by the Telford
Gravel and "Conglomerate at Lyndhurst”
{Firman 1969), The “Conglomerate ai Lynd-
hurst’ resembles conglomerates in the Wills-
wortina Formation (Fig, 12 this paper).
Kaolinite ig the dominant clay in the Avondale
Clay type section, and is abundant in the uppet
part of the Namba Formation, and the Willa-
wortina Formation,

L37

A section of Yetila Creek in the Mooloo-
watana area of the northern Flinders Ranges

owas tesccibed and figured by Firman (197i,

Fig. 12)8 as Avondale Clay, Upstream from
this site, a lower part of the section is exposed,
connected by continuous outcrop, This exhibits
micritic carbonale nodules, underlain by silty
olive grey clays similar to the upper part of
the section, The clay is capped by a well-
developed hard white fine grained carbonate
soil horizon, Comparable to that developed on
the Willawortina Formation at Balcanoona
Creek, It. is overlain by the Eurinilla Forma-
tion. The lithology is identical to the transition
beds between the Namba and Willawortina
Formations (section 11, unit 4).

The Avondale Clay is regarded by
as much younger than the Etadunna
tion, henee younger than the Namba Forma-
tion, However, the comments ahove suggest
it could either be part of the Etadunna Forma-
tion, equivalent to the lower part of the
Willawortina Formation or upper Namba
Formation,

The lower part of the Telford Gravel (Fir-
man 1967a) may be equivalent to at least part
of the Willawortina Formation.

Firman
Forma-

ENVIRONMENT

Consideration as to whether the Namba
Formation sediments are marine, marginal, or
non-marite was a prime objective. The most
conclusive indicators of marine influence are
marine fossils and glauconite, hence samples
were investigated for foraminifera, and any
green pellets or clays were studied by x-ray
diffraction, A variety of lithological types from
subsurface and oulcrop were examined by
J. M. Lindsay who found no foraminifera,
Non-marine gastropods and pelecypads. (pers.
comm, N, H. Ludbrook 1973-+) are present,
aa are non-marine ostracodes (pers. comm. K,
McKenzie) and the fresh water algae Pedia-
strum, (pers, comm, W. K. Harris) and
charophytes. Al] green clays proved to be
montmorillonite, and green pellets Tound in
The eastern areas were dolomite, associated
with non-marin¢ pelecypods.

Other evidence for non-marine origin 1s
derived from the terrestrial vertebrate remains
(e.g, Fig. 10). Several skeletons were found
in a partly articulated state, Delicate bones are

2 Harris, W. K. (1970). —Palynology of Lower Tertiary sediments. South Australia, M.Sc, thesis, Uni-

versity of Adelaide. {unpublished}.

§ Firman, J, B. (1971).—Regional stratigraphy of surficial deposits in the Great Artesian Basi and
Frome Embayment in South Australia, §. Aust. Dept, Mines Rept, RB 71/16.

R. A. CALLEN & R. H. TEDFORD

138

saulw jo ydaq ‘y's

NOILYWYOd JYA

N311V9' ¥"U

NO|LVWHOd FaA4

NOILYWHOd
VANNAOVLa

IDVIANS CINGIOD
AO 414dD1S

NOILVWaOd VaVENOO

ANOIS3AWIT YOYIITV

J1qadD TIS

o@r—rZ
INID0I1Vd ONY
3N3903 1VIGIW

INJJOIN ATYVI
©} 3N3909110

NOILYWeOd J8AI

ayaiajis Alpiq AaiB

IDUWN|OS SAISSOW;

INIDJOIN = AVI
} IIIA

NOILVYWHOJ VaWYN

ALJaD Naa

NOU YWaOd TAIT ydIM

QNY¥S NGYIOMNdWYW

2143 9427 1961

"}2 12 NOLMILS

SAD|>

4nyupudy yo
ajDJaWo)Bu0y, pup

AV1ID JIVINOAY

AV1D JIVONOAY

{ua}DAInba

é 343531143} UO ‘ajasd|/s

Al|iq Ao18 puo auojsBulppng é

INIIOLSITd
Aa]
} INJIOIW

NOILWWaOd
YNILIOMVTTMM

DIPDAsMY YNoS
usaysDayjsou puo
sasag Nypayazyg

‘'yZ6L ‘YANIAOM

Om 1,

INOLSIWI)

¥VINNNAONNE {@2Dj1NS OpPUooiDy fo)

*sajDsa wo) Buoy

uvoz1ioy snour6nisey

saays dow
000 OS2‘L AIIdOD
8 1? SIVOD

saBuoy
S1@PUI]Y WIASSAA “Nh
“NYWald

V1

ajeidiiiay

jaded sy)

QdOs3dal F NIV

SNISV@ YSHLO GNV SLASHS dVW LNZDVrav - SLN3TVAINOS G3LS4a99NS

SLINN 43010 - LHVH) NOILVTEYIO

VWauV
3WOYd IVT

C 718Vl

CAINOZOIC ROCK UNITS

well-preserved, and abrasion due to transporta-
tion i currents virtually absent. The sediments
in which they occur are fine sand, clay, and
dolomitic clay. A nearshore marine environ-
ment therefore seems untenable, though a
lagoonal or upper estuarine enyirenment is
possible. A mon-marine environment is pre-
ferred, though presence of Cetacean remains
(a platanistie dolphin} indicates a link to the
sea at some stage.

More specifically, environments are (i)
Micritic delomitic carbonates with irregular
ooliths, suggesting low energy shallow lake or
shoreline conditions. ¢ij) Black, laminated
fossiliferous clay of Wooltana No, I, unit 1,
suggesting a well developed Jake: the fine
laminae resernmbJe varves, but have been dis-
rupted by diagenesis and bioturbation. (iii)
Sedimentary structure types, abundance of
fines, and very poor to poor sorting support a
low energy environment for the whole unit,
This may explain the apparent lack of well-
developed beach sands, which would be poorly
developed and poorly sored along a low
energy shoreline,

The environments represented by the cyclic
sequence described earlier are in ascending
order: (a) channels: smal] to medium seale
cross-bedded fine to medium sand (Fig. 8),
(b) flood plain, estuarine or lacustrine: finely
laminated silt, often burrowed and with very
small to small seale cross-bedding (Fig, 7),
and olive clay, (¢)} lacustrine: patchy car-
bonate, oolitic dolomite (Fig. 5) and clay
(Fig. 6). (d) swamp or mud flat with ovca-
sional channels: hard, black, niottled clays
with irregular fractures and sand patches
(Fig. 4), interpreted as vertisals. The cyclic
sequences are of 1-20 m thickness, averaging
9m, well-developed in most parts of the basin,
except the northwest where uniform clay sec-
tlons dominate, The cyclicity suggests a depasi-
tional process resulling in a particular sequence
of facres, but with inherent tstability, Some
examples applicable to the Namba Formation
are (1) a delta building into a shallow lake or
estuary (necessarily shallow because the cyclic
sequences are thin), (2) repetitive transgres-
sion and regression of a shallow lake shore
In tesponse to fluctuations in water level ¢3)
repetitive avulsion of a meandering stream,
(4) bars associated with development and
abandnoment af portions of mver channel,

The abundant bioturbation, and its occur.
Tence in medium-scale cross-bedded coarse
Channel sands, aud basal parts uf laminated

139

silt beds, is inconsistent with river channel
origin of these facies, These salids more likely
represent offshore lacustrine bars. Lenses of
the coarser sand facies at the base of, or with-
m, the tough dark prey clays are also difficult
to explain in terms of a river and flood plain
relationship: channels cutting across a tidal or
deltaic mud-flat are more acceprable. Subse-
quent intensive bioturbation or rheotropic flow
has partly destroyed bedding, distributing the
sand in irregular patches. Im the estuarine
case, the absence of any evidence for a marine
influence, parlicularly in the microfauna,
indicates deposition in the uppermost Teaches
of the estuary, In sequences where coarser
channel sands are interbedded with non-hur
sowed Jaminated silts, the river-channel and
food-plain relationship is still applicable.

Fluviatile deposits are abundant, and fossils
(e.g. Fiy, 10) of aquatic vertebrates (Dipuai,
Teleostei, Chelonia, Cetacea} suggests a per-
manent water supply, and fossil plants (Notho-
fagus and Podecarpus) indicate bigh rainfall.
The distribution of lenses of channel sands
within an essentially clayey sequence is typical
of meandering rivers, Al€hough only 42 current
directions were measured (mostly in the Lake
Tarkaroeloo area) resulis sug#est a southerly
component of transport direction for the upper
pant ef the Namba Formation, in marked con-
trast with the north-casterly direction of over-
lying units.

Dense vegetation is suggested by the paly-
nology of the basal unit 1: the modem des-
cendants of the species represented typify
rainforests. Abundant grass pollen are evidence
that grassland occupied extensive areas, Thiss
rainforest was not coniinuows in the carly
lacustrine phase of deposition. The relative
ahutdance of arborcal marsupials in the upper
part of the Namba Formation indicates the
presence of gallery forests along the water-
courses.

Apparently at variance is the smoctite—
dolomite—palygorskile association, frequently
recorded from arid soils. playa lakes and warm
hypersaline waters (e.g McLean ef al, 1972,
Bentor ef al. 1963. Meester 1971, Singer ef al,
1972), At present dolomites and high-Mg
calcite are forming in hot arid or semi-arid
hypersaline lagoons (e.g. Von der Barch 1965,
Von der Borch ef al, 1975, Fricdman et al,
1973) though some magnesiumerich sediments
are found in fatitudes as high as 48°N (Muller
er a. 1972).

144

Millot (1%64) indicates the montmorillonite
—palyvorskite—sepivlite association is the
result of offshore lacustrine or marine chemical
deposition, This toek place adjacent toa laleri-
tized land mass of low relief and dense vege-
tational cover, in # subtropical or tropical
climate, Sepiolite is absent an the Lake Frome
area, but this may be an affect of degree rather
than basic difference, The hypothesis as applied
ta the Namba Formation overcomes the diffi-
culty of evoking evaporative conditions in a
high rainfall climate.

Millot’s hypothesis has been applied in a
similar manner to the Cainozoic rocks of the
Jordon Valley (Wiersma 1970). These sedi-
ments contain a remarkably similar sequence
of clays to the Namba Formaticn, Particularly
televant are Wiersma's comments regarding
ibe ongin of palygurskite {p. 88). He con-
cluded that intensive weathering on the hinter-
land under warm humid conditions was
necessary for liberation of the clements essen-
tial to the genesis of palygorskite and its
associated sediments, and that evaporation in
the sedimentation basin should be such annu-
ally as to provide the necessary concentration
of chemical elements. He deduced that
evaporation must prevail over precipitation
and fuviatile and/or marine supply of water
to the basin. In many places in the present
tropics evaporation can exceed annual pre-
cipitation, with resultant formation of evapor-
ites in favourable locations, Palyvorskite was
of detrital origin in the Jate Tertiary and
Quaternary of the Jordon Valley, having been
derived from Cretaceous and early Tertiary
recks im which it originated by chemical sedi-
mentation. In the Lake Frome area oo pre-
existing rocks rich in palygorskite were present:
rather kaolintte, smectite and illite are ahun-
dant. Palygorskite can be formed in soils
(Singer & Norrish 1974) hue only in relatively
low proportion, thus it mmnst have originated
within the depositionary basin during sedi-
mentation.

(tis notable that Mellot’s 141964) ideas as
applied to deposition of Namba Fartnation
stdiments require on equivalent Miocene
fateritization on adjacent land masses. In this
context Wopfner’s (1974) conclusions regard-
ing an Oligocene-Miocene “ferralitization” are
of interest, Although the evidence he gives for
age of the Doonbara Formation is inconclu-

RK A. CALLEN & R. H, TEDFORD

sive, some ddditional observation are made
here, Firstly clasts identified with the Doonbara
Formation (by R.H,1,) are found in the Wipa-
jiti Formation (Stirton e ai. 1967) of Miocene
age, Secondly the ferruginization in Lake Eyre
Bore 20, doubdtfully equivalent to the Doonbara
Formation, 1 recorded by Callen (Wopfiner
er al. 1974, Fig. 17) within the lower part of
the Miacene Etadunna Formation. Others have
also recorded an older Testiary ferricrete ( Fir-
man L967b), Therefore lateritization {or at
least, ferruginization) could have been pro-
ceeding in uplands adjacent to the basins in
which the Etadunna Formation and Namba
Formation sediments were being depasited.

The main carbonate hotizons occur a few
metres above unit 1 of sectlon 11, with its
rainforest flora, and above the vertebrate zone
with its indications of seasonal climate with
abundant water supply. The presence of these.
carbonates can be explained in terms of pro-
tracted urid phases superimposed on a sub-
tropical or warm temperate climate.

In addition the presence of detrita) feldspars
must be explained, particularly in view of the
abundance of plagioclase and wssociation with
smectite." The possibility of addition of vol-
canic material From eastern Australian must be
considered, the Miocene being a period of
Maximum vulcanism (Sutherland er af. 1973).
However, the percentage of feldspar is not
large. Preliminary studies of feldspars im the
Namba and Willawortina Formations suggest
Telative proportions of feldspar types and
compositions of plagioclases are similar to an
unmodified contribution from nearby Precam-
trian crystalline basement rocks. On present
evidence there is apparently no change in rela-
jive abundance and type of feldspars in the
illite-Kaolinite rich zones of the Tertiary, in
comparison with the smectite zones. This sug-
ests abundance is not tied to smectite occur-
rence, as would he expected if these minerals
originated from volcanic ash falls, The pre-
sence of feldspar presents a problem consider-
ing the evidence for a humid climate. Possibly
seasonal aridity and nearby source permitted
preservation. In addition Todd (1968) has
shown plagioclase is more stable than ortho-
clase under conditions of restricted leuching,
in a tropical climate, Thus smectite (mont-
morillonite) is unlikely to have originated
from volcanic: ast falls,

The mineral proup smectite, but R. N. Brown recovnized dioctshedral montmiorillonite in several in-

stances.

CAINOZOIC ROCK UNITS 14h

In the final analysis, the Namba Formation
was thought to be deposited in o warm tem-
perste to subtropical climate. The landscape

ad a savannah aspect, With gallery forests
around permanent rivers and lakes, Periods of
aridily occurred.

High average temperature, invoked to ex-
plain the mineralogy of the Namba Fornatinn,
is in accordance with marine paleotemperature
measurements mm southern Australia (Gill
1968). and New Zealand (Devereux 1968,
Jenkins 1968) of 18-22°C for the Miocene.
Considered in the light of continental drift
data, which suggest Avstralia was closer to
Antartica (though drifting rapidy northward:
Wellman e? ul, 1969), and data which indicate
the cooling of Antarctica way underway
(Hayes ef al. 1973). the temperature can be
explained in terms of greatly expanded sub-
tropical climatic zones during early and middle
Miocene times.

Deposition of the Namba Formation in the
central part of the lake Frome area was fol-
lowed by widespread ferruginization and silict-
fication (opal, and quartz overgrowths) and
development of cryptocrystalline silica nodules.
particularly in the coarser Namba Formation
sands, These processes Were the result of wide-
spread groundwater movements, Formation af
dunhcnists ynd telated phenomena had a locus
in river valleys cut into the Namba Formation,
prior to deposition of the Millyera Formation.

No evidence for 4 major period of Oligo-
cene to early Miocene “ferralitization” sie-
gested by Wopfner (1974) was found in the
Lake Frome area, though there ts abundant
evidence for late Miocene to Pliocene ferruy
ginization and orthoquartzite silerete Forma-
tion. This does not necessarily negate
Wopfner's climatic evidence, since two periods
of fertuginizition are prohable (Firman 1 967h,
19714; Jessup & Norris 1971). The older
Tertiary ferruginization would presumably nat
be manifested in the Lake Frome area, where
chemical and detrital deposition were pro-
ceeding.

The coarse detritus in the Willawortina For.
mation has clasts derived from Cambrian and
Precambrian rocks in the Flinders Ranges,
When considered jn combination with poor
sorting and abundant feldspar content, vigorous
uplife of the Ranges is indicated. This wes
accompanied by movement on the Poontana
Fault. A similar conclusion has been drawn by
Binks (1972) from evidence on the western

side of the Ranges, Ironstone and silcreie peb-
bles from pre-Willawortina Formation (?pre-
Namba Formation) duricrust are present,
However, laterite clasis are got as abundant in
the overlying Willawortina Formation as ont
would expect in a sequence supposedly derived
from erosion of a laterized land mass, Pre-
sumably this is because the Flinders Ranges
were virtually non-extstent at the time of
deposition of the Namba Formation, presenting
only a smiull atea for laterization- Alternatively,
in keeping with the suggested warm-temperate
to sub-tropical climate, ferruginization may
not have developed an extensive laterife crust

Deposition in an alluvial fan environment is
suggested for the Willawortina Formation by
the presence of extremely poor sorting {Fig.
13), numerous channels (Fig. 12) with
medium scale cross-bedding, and laminated
calearceus sills (Fig. 12) with red-mottling
snd carbonate conerctions typical of flood
plain deposits. Firing upwards sequences are
typified in section 11, suggesting bar deposi-
tion, The deposils coarsen very rapidly close
to the Plinders Ranges. The extremely poor
sorting, coatse grain-size and matrix-supported
lexture in some beds may be the product of
mud-fows. The red mottling (‘marmorization’}
and carbonate soils are similar to those des-
cribed by Freytet (1971) in association with
alluvial deposits, and typically form in the
inactive parts of fans (Blhissenbach 1954, p,
185; Denny 1967, p. 105). These features
resemble modern fan deposits.

In sections | and 10 there is a tendency for
averall coarsening upwards, suggesting increas-
ihely rapid uplift of the Flinders Ranges. The
uplift deluged the former lakes and swamps
of the Namba Formation with detritus, re-
ducing their extent. Thin dolomite lenses in the
sequence (section 11, and Balcanoona Creek)
represent lacustrine or playa lake phases simi-
lar ta these of the Namba Formation, Petru:
logical investigation shows these contain 4
much higher proportion of sand (mutch of it
unstable mineral grains) than the Namba
Formation carbonates.

During deposition of the Willawartina For-
mation, oxidizing condijions became prevalent,
through accumulation of the sedimentary
column above the water table, This contrasts
with the sub-water table reducing eavironment
of deposition of the Namba Formation, Abun-
dant potash feldspar and plagioclase can be
attributed to rapid deposition and possibly
semi-arid climate. Presumably uplift of the

{42

Flinders Ranges Would haVe had a strong effect
on bocal climate, but this cannot be assessed
at present.

Following deposition of the Willawortina
Formation, ferricrete and calcrete formation
occlicred, particularly in marginal areas.

The absence of surficial cementation of
coarse sediment in the type section of the
Willawortina Formation and nearby outcrep,
contrasts with the ubiquitous cementation in
southern areas (sections 1, 11, Fig, 3). An
explanation in terms of a carbonate rich source
arca for groundwater or sediment, or abun-
dance of limestone clasts in southern ureas,
does not explain the widespread distribution
of surficial carbonate cementation in rocks of
various ages throughout the Fhnders Ranges.
Indeed, many fans in semi-arid areas through-
out the world are similarly cemented, Enough
calcium is produced by weathering, or de-
posited from wind-born dust, to provide suffi-
cient carbonate material for cementation
anywhere, Therefore in the case of the Willa-
wottina Formation adjacent to Mount Painter,
absence of carbonate is a local phenomenon,
the explanation of which is unknown.

Pt. [1—Younger Cainozvic Rock Units
Type and reference sections are shown in
Fig, 15, and described in detail in Appendix
2. Table 3 summarizes the lithological and
other properties of the rock units deale with in
the text,

LITHOLOGY
MILLYERA FORMATION — (Derivation:
Lake Millyera*, near the mouth of Billeroo
Creck, Millyera is Jocal aboriginal word for
water. Map reference: Siccus map sheet,
FROME),

The name is proposed for a sequence of
imerbedded greenish ostracode-bearing clays,
thin limestone of charophyte algal remains,
and fine sand, The sediments occur in Lake
Frome or io smal} lakes close to its margin.
The name is also applied to 9 coarse cross-
bedded or conglometralic sand, regarded as a
fitviatile equivalent of the clays, where these
contam interbedded charophyte limestones.

The type section at Lake Millyera (Pig. 13.
section 4) is located 69.2 km en 320°T, north-
west of Low Stony Hill (map reference Tele-
chie) on the Siccus map sheet (Air Photo ref. -
Dept, Lands Svy 31, Siecus Rum 1, Photo
4460),

The section consists of 4.3 m of laminated
green ostracode bearing clay with « thin bed

K. A, CALLEN & R. Ho TEDFORD

(40 cm) of laminated charophyte limestone
near the top, overlain by alternating clay and
fine sand, An abbreviated description is given
in Table 3,

The \ype section was not Jocated at the
thicker section 3, where there is intertonguing
with red beds, This avoids confusion which
might arise should the red beds, which have
affinities with certain fluviatile equivalents
(c.g. the Eursinilla Formation—see [ater), be
formalized as a distinct unit.

The contact with the Namba Formation
Was nol exposed, but can be observed in the
supplementary section located about 2 km
east (Fig. 15, seclion 5, 68.7 km on 327.5°T
northwest of Law Stony Hill—air photo refer-
ence S. Aust. Dept, of Lands Svy. 361, Siccus
Run |, Photo 4461},

In section 5, the Millyera Formation is
3.3 m thick, cropping out below a thick ex-
posure of Eurinilla Formation. Here thin
charophyte limestone in the Millyera Forma:
tion grades laterally into gypsum, often ripple
marked (Fig, 16), imtercalated in an essentially
sandy sequence. Sometimes botryoidal sirye-
tures are present on the surface of the gypsum
laycr, similar to those found on the floor of
Lake Frome. Scattered very coarse polished
sand grains are present on top of the gypsum,
where it is in contact with the overlying red
sand lens.

The red sand lens consists of u thin bed of
bright red-brown coarse silk with basal granule
layers impregnated with secondary gypsum,
closely resembling the Eurinilla Formation in
lithology. It grades by alternation to greenish
very fine sand with coarse sand, silt and clay
laminae {yellowish grey). The greenish sand
bed is fossiliferous, with numerous charophyte
oogonia and stem moulds, fish verebrate and
spines, and ‘Coxiella’,

The contact with the Namba Formation was
exposed by trenching. Orange and yellow sands
of the Millyera Formation contain reworked
and oxidized tough grey sandy clay clasts from
the underlying Namba Formation, The Namba
Formation is more indurated and darker
coloured.

The top of the sequence is marked by a
strongly developed soil, also observed else-
where affecting the Namba Formation. The
soil has a crumbly texture, with peds of
irregular shape about (,5-2 cm across. A well
developed black mungams is present, and red-
thsh-to yellowish-brown iron oxide patches are
developed in the clay. A similar horizon is

iS
Z
5
g
%
g
)
N
S)
z
<
o

seul jo ‘deg yw $

‘RUN OW Ue. BL EO-pay

Spues sbiggs jeseq
eS dHuuNO UBNY EpURS Ul
1 L3}AMPUNOM 2p BALSSEY

‘SUDYSALAY

PUNY, “UbAISAS Yaa
Hj -C018) pum Frayer ¢
pajuaseute. MoU
Bue papdue deep

idngie ‘ofpa pyncue yoo
OW PVE FUDIY 2HE] U0 pag Hl)

way yd

eas eoqysyy woud)

“Susedap JeaL SIOMN|LAISIp OFeq DAES
AN3DOLSI9Td ATHV3-3NIDONW

*avaapiaN 4SI] “aARAI
FapPMRNSO
Wve). SXtingy Gary

in Jada Oy ari

‘ts

Spodtuyse8 (Ends
pue spqdun

elugbon sojeuty
YUM. PUES al
Suureg-ponensa ali

“AUO]S aI AINA]. Sappueeey

‘Suippaq ANE

UOELHUR) SUID B/RIS PLS
“Sples faulunys pappag-ssisa
aly SNH) SayeaBy
JD. WnSOKG We deysiyy
akptancg pues sydudu.
lS. aLyaLIASSE
aoa puE ye ul
UQUeUHUNE) | 2qUOLI LOA’ Ub

“spurs AKO) Ys\U83 olny OY Lupa

LVRS 22-940 G pues i
VALAT POL aUN DUBS aU PL/ BAS “ISAS

“2USMAUT ‘Agr “SULTS puna

UL pueY- fappaCAIN|
UIA AUES @UN( alm 17 ySIpAIY “OPUNOL |faM. OF
PINE Swe UUs “ypey WHIOAS. POxsPU-Naiddes

Wi, OY Apel Ae GLOOM TAA biLAZ)

Due PiaOs-ypayk aU UWA POpADIET “Coy
Je PEUIWUE CLG ASP HOS
ny vedy |PIGYALOI. ‘Agja- Paysiee

ia 90 é54—re

NOILVWYOS
VUYSATIIIN

“sdh) JB yEmWUNON

fo LINSAAR pue ayeuncquer

HOS 0} fey Um YM panuoWa?
Alpilos dyevauio|zL09 jesey
BUY PLE pod {Oo Sapeys ul
PaylOW Sayeyod wnsdse paiatey
10 BAISSEW YIM ais|y ‘aNloyes

SE) Ul palual
Aalan|s OY AYRAM, puws Jaap
PIRPINS UO SpHAyS Ad sojmuesd
$F [No Jayjean Ajluanting “segde
ALS Wt Sapnpost 981] 4yoe ‘Hay
vB Pajuasand aq ABW Splaupulj Aa
WINSUAR Dur epeungiea Jo
SUULDI| paOD|Ehay [fam AEE

‘Sayeun|
wnnday auaay S4e7 aos

“3°5 Booje spunoul pile spurys
“eeguien) Avjo jyesaid syeubxouder
SOUUEy) feseq Ayyinbesy

yd ay
pat

4e> anisseu 16 Youag 7

DRIED A TRIAD AG papules
ayo ped fama] “sasinoalyR pue

Viowssaudap punowe yedoys Mo] se

joo Buiddoy ‘Siejd TUiA)-miy Vey

—

uaLuipss 40

SUUUN|2D BINWIGY ‘mands AyobIq
JBAMUBLAY 'SAONLI\UOImIE “SERA
PiEY §2, Tayayjeam 'sproupu
ayeWog iP padajanap Ay4e:

SOS DALVIDOSS¥

“HUI ARRIETA S7TRp
yeaa Se sysauy Wi yno sdou

“Salma [J35 pile shied: jo adenspue)

SOpEUIGD ‘Suey FuMdj mc] Tey

NOISSIHdX3
OIHdHOWOI9

aReSqeNeh

“pues 91203. U)

ymioodsa ‘undo seyeresy)
SAAWINA) “ALOLy ye]

40 88pa Jee eyaxoy,,
SYOUEL ee) pur
SPMNONSED 10 s3A2] \ESED
OS sadkouzeyy

qeyad vopoydtoatss

“Os vopoqudie? Fi
Sqn) (ese sayeIgayIEY,
‘Sysau py SMALING pAPSU)

duyep Mig cuney

ANIDOLSIF Td SLYT

SMOKY SII "JadyS PURI PesiMapM sejnqe) UIE

plies Linsdég

“tied yeseq ul)

ayvodkd pus le a ajem
PUNE ‘suorLey es penthy
UT YSUMO|pAA. JEAND]
WMOIT-AIS\pIKOd Bd sidde)
“PappHG- S807 Ed 191K)
"UCI EINEE-SsULs
PEPUCELOLY Cues, jpeuepeape jy
SRY Ailes Gun JO
ANY Syne) "WT Sadiq
iad 150) wyeOD ted
Jn avy auiy10§ s005

“pues winsdAy
ib duippaq-ssoa
addj—ayjouiny ay
osysin) ped

leo ny

(ed cago) aley syeos
Walpaw vO pappsy St019
dpueyd 1a yfnoy

SPURS |ESEY “Slay sue MO]
AROS aH OPW) pachjarsp
Salvia (WIDE Z)

WOyeU re) pepe Oy
PauOJONda aH soipespreys

‘Spues Winsdé asaeiry “yseia Qug_SuulES sppuiBLLoy

DUE SjoUOGIET paruian Ayeoo] PhP sayqqad

SHEN YUM fluUkS astegg [LE SopefaloyAyuy eseE

BUIAS SeSMADL 2 wed. 1aMO] ApUES

“7 bleepYAS-GZ MEM sealily| "pAssou) Oe Ppl

JG “pagyld Pue arial yng, papureY AjayeapULY

at] APN SiNeAS ‘PayVOS fyudOd Ada, “saygqad

YJ Wap} pees pred Jamo) “Wen Jadan Wi Keys
PUR YAS LO Spagdsyur yay "puRS cy 7

{6'€'S)

NOILVWHOS
VaTINIaNSa

| SMAUN “Suyeiqayian HEH
(881) “ORDO RalRIRYT

S}OE UE 12
(ays MUR “SyRy jreus

suaper wo) § 1 NIDIW

~ANIIOLSIATd ATYVA

ai

PequeyG wuy
(NOILYWYOS YOINIUNS

a

AY19W033 STISSOJ

SLINA

Joy pe vomOY
(ive jesnlAy “SHeUS puey

40 WOUARTE VNO|OD lmMOsG
1h AUS gyevapoW
“BUIP2Iq Yavjaanp- AYBaN,

NOILVTaHHO9 404
VidsLldd

WEMIOTINOY OPM JG Woy

‘Tpauiyaosjols Jyaues)
MONET ENS-S50N9
SIPAS, FAIR] ‘Rasta

IDE} \EqunUcy yRayy

SIMNLONYLS
AdVLNAWI03S

YAJINNOA
SEILYAdOUd LINN MOO FO AUVWWNS € AIAVL

nS "Baus Marezapow
UL ‘PUES Aake[a JO AIS 0) PrleR

ASOTOHLN

NOILWWHOS
ANISYVNOOS

(1 ‘Gl4)| LINN

144

developed on the Millyera Formation in Sec-
lion 7 of Fig. 15,

The distinction between the lacustrine Facies
of the Millyera Formation ani the superficially
simijae Namba Formation can be demon-
strated from Lake Millyeta and Lake Tar-
karoaloo. In Lake Millyera, the north shore is
formed by cliffs of Namba Formation clay and
dolomite showing no facies changes thrausheut
the length of the continuous outcrop. The
south shore has sporadic outcrops of Millyera
Formation, also unchanged along strike, traced
to the junction of Billeroo Creek and Lake
Tarkarooloo. These relationships suggest a dis-
canformity between the two units. The Millyera
Formation alsa occurs on the north shore of
Lake Millyera at its castern cd, but a covered
interval prevents direct establishment of a
‘clationship with the nearby Namba Forma-
lion, ‘The contact between the units can he
observed by trenching the base of the supple-
Tentary seclion 5, where the erusional contact
und weathered top of the Namba Formation
support 4 disconformity.

Similar faminated ostracode-beuring clay
and gypsum laminae are found bencath the
hase of the gypsum dunes constituling the is-
lands of Lake Frome. These grade down to
ostracode and charophyte-hearing indurated
elays, without sedimentary structures, beneath
the lake bed. Fine sand interheds are present.
These beds ate equated with the Millyera
Formation. but most sediment flooring Lake
Frome, though of similar lithology, is younger
and less consolidated.

Along the eastern edge of Luke Frome is a
series of eroded mound springs, exhumed hy
deflation of the modern lake Moor These are
built of saucev-shaped carbonate layers, partly
algal in origin, which intertongue with the
clays, These spring deposits are partly ccjualed
with the Millyera Formation,

In Lake Tarkaroolwo, reddish-brown silt and
conglomerate is inter-bedded with the charo-
phyte limestone. Proceeding south along this
linear Jake, the coarse facies eventually
dominates, the limestones being absent. The
Millyera Formution is no longer identifiable,
having graded inte an entirely fluviatile facies
of conglomerate 2-3 m thick, cemented with
massive white secondary carbonate (hence-
Forth referred fo as the ‘unnamed conglo-
menate’ cyuivaleat to the Milkyera Formation)-
Section 6 (Piss 15, 17) shows an intermediate
stage in this transition, Numerous well deve-

R. A. CALLEN & RH. TEDFORD

loped channels exhibiting 7 cross stratification
(Allen £963) are present, cxhumed on the
lake Noor, They often contaiii clasts of fossil
wood, Namba Formation dolomite, ferru-
ginized Namba Formation sand, und mitky
quartz at the base, demonstrating a discon-
formable cclationship with the Namba Forma-
tion. There is little facies change along the
length of Lake Tarkarooloo in the Namba
Formation, whereas the Millyery Formation
yavies considerably, though retaining — its
identity a5 a wnil.

Two charophyte Jimestone horizons were
developed in the southern part of Lake Tar-
karooloo, instead of one ss at Lake Millyera.
Since they must have once represented a
horizontal lake shoreline, and arc equivalent
to the horizon at Lake Millyera, structural
deformation (?faulting) is required to account
for their relatively higher position in the land-
scape. Barometric Icvelling, tied in with South
Australian Department of Lands bench marks,
established the height difference [see Sections
§ and 6, Fig. 15). Comparison of the heights
of equivalent Namba Formation carbonites,
between Lake Millyera and the Namba Formu-
tion reference section in Luke Tarkaroolon
(Fig, 14) also supports downfaulting of the
Take Millyera region.

The “unnamed conglomerate’ channel
equivalent of the Millyera Formation can be
traced throughoul the area southeast of Lake
Frome, where it is invariably overlain by the
Eurinilla Formation. The disconformity is
difficult tro detect away from low4ying areas
such as Lake Varkaroolon, where well-deve-
loped greenish carbonate nodules and cylin-
droids of a soih calcrete mark the contact, and
massive White groundwater catbonate cements
the conglomerates in the Milfyera and Burinulla
Formations. Elsewhere the conglomerate has a
weak earthy carbunate cement and interbedded
secondary gypsum Javers. interpreted as
proundwater phenomena rather than 4oils.
The contact with Eurinilla Formation appears
gradilional fez, units | and 2 of the Eurinilla
Fennation in section 8, Fig. 15), and it ts not
possible to establish a disconfarmity. An addi-
tional problem in the Millyera Formation is
the repetition of red sand facies resembling
those of the Eurioilla Formation. This suggests
it may be possible to have Iwo red facies super-
imposed, The contact between unit 2 pnd 3 in
section 8 may represent such a boundary (ie.
ihe lower part of this section may cnrrelute
with the Millyera Formation).

wy 20-—
Lu
~
wi 2? 3
= 4
- N 5 N 5
>. 15
°
Ee Baleanoona W E
> Creek .
Lad
|
w 10 ee aS, \ 8 ERA
* FORMATION z =\7,, MILLY \
= MILLYERA =
FORMATION J FORM. © ROE SS
5 = Lake Millyera
FAULT : Lat. 30°0232' NAMBA \
o
Long. 139°54'42 FORM. \
\
Lat. 31°02'59'
Long. 130°Sé 36
"\. EURINILLA
Let N 3
7 ae FORMATION
Lot. 30°53'30- and 30°56'0)"
4S Long. 140°1014° ond 140°1)'06"
BINE
cOONAR
FORMATION SSS SSS) |

|

|

]

|

|
ce)

| EURINILLA

FORMATION N\ SF POT Rea rae

YPF SECTION

%

Reet FORMATION

Le)

MILLYERA . DEX Lake Pinpa Lake Koorke
FORMATION = Sa = Se ee ee Se SS
Lot, 31°09 59" q Lol. 30°58 24°
Long. 140°|3'39" Long. 140°19'37

FORMATION

FORMATION

Lot. 31°0640"

Rou Long. 140°06'34
w NAMBA FORMATION
50
* SECTIONS BAROMETRICALLY LEVELLED
45
DATUM: DEPT. OF LANDS
BENCH MARKS 5062—5066
a REFER FIG, | FOR SECTION LOCATIONS
a REFER FIG. 20 FOR LEGEND
wi
ce
35 2 NUMBERED UNITS DESCRIBED IN TEXT
z
z
°o
30 F
>
ire
=
et)
25
L74—23 S.A. Department of Mines
R.A. Callen Geologist Orn. B.T.
20

Fig. 15 YOUNGER CAINOZOIC UNITS, LAKE FROME AREA
15 ee TYPE AND REFERENCE SECTIONS ‘

1100—46.76 ~G6808 O

CAINOZOIC ROCK UNITS

Current directions Were recorded from a
variety of cross-hedding types in widely scat-
tered localities in the basal channel Facies
(both sands and conglomerates), though with
a bias towards the Lake Tarkaroofoo area, hut
are sufficient (45 measurements) to record u
nocth to northeasterly transport direction, Con-
glomerates such as those of section 8, where
no disconformity between the Eurinilla and
Millyera. Formation was established, were not
included in this analysis.

EVRINILLA FORMATION — (Derivation,
Eunnilla Creek, Evrinilfa map sheet. CURNA-
MONA),

The type section (section 7, Fig, LS) is
located at Lake Moko* on the junction between
Billeroo and Eurinilta Creeks (Air Photo Ref.
—S. Aust. Dept. Lands Svy. 395, Coonarbine,
Run 3, Photo No, 9637). The locality is
45.8 km on 332°T trom Billeroo Watechole
(CURNAMONA—Billeroo Creek). and the
outcrop (Fig, 20} occurs in an amphitheatre,
formed by guilics draining into the north end
of the lake. The upper fine grained facies is
well developed. The complete section is de-
scribed in Appendix 2, Section 7 was con-
steucted from two outcrops 200 m apart,
correlated by following the beds along stnke.

This sequence consists of three units, separ-
ated by a weak carbonate soil horizon, The
lower unit, exposed in the southern outcrop.
consists of 1-8 metres of white well sorted
medium gramed channel sand. It i ¢ross-
bedded, and contains some vertebrate remains,
clusis of underlying units, and ¢harophyte
oogonia. The sands have features jimilar to
channel] sands elsewhere at or below the base
of the Eurinilla Formation. An interesting
feature is the presence of biotite, suggesting
nearby crystalline bascment outerup at the
time of deposition. Unit | is averlain by the
much more widespread and typical fed silly
and sandy facies of units 2 and 3, respectively
3.3 .and 5.5 m thick, These units are separated
hy a weak carbonate soil harizon, The lower
is paler than the upper, and shows tio evidence
of bedding. It grades to clay-silt at the hase.
The upper unit has diffuse large scale cross-
bedding, is sandy, and orange-coloured
throughout. It is capped by a well-developed
fossil soil horizon with nodules and cylindroids
of carbonate and some gypsum,

The sequence is overlain above the fossil
soll by a horizontally luminated sand, similar
to thar al the top of Unit 2. but with more

145

pronounced bimodality. Since the sail repre
sents a disconformity, these sands are excluded
{rom the definition, and placed in the Coonar-
bine Formation,

Another section 8 m thick (sectuon 5, Fig.
15) is situated on a steep blu? facing north,
on the southern side of Lake Millyera, close
to the Millyera Formation type section 4. A
detailed description is in Appendix 2, The
sands afe paler than those of the type, and
contain low angle cross-bedding al the base,
The overall two-fold division fof units 2 and
3) in the type section is retained. A diffuse
thin bedditig dips toward Lake Millyera. Con-
sidered with the geometry of the outcrop. the
sequence is interpreted as a small delta.

The upper part of this sequence has tubules
and cylindroids of soft white carbonate and
gypsum, several centimetres diameter, forming
several inter-related horizons representing =
fossil soil complex. The tubules weather out
as hard cylinders. The lower part of the
sequence is partly cemented with shects of
sofi white carbonate, and with pink carbonate
nodules. The base is solidly cemented with
gypsum. partly derived froni the dissolutiun of
gypsum lunetftes {represented by low angle

eTuss-bedding).

A third reference section of 3 m thick 1s
located al Luke Koorka* (Fig. 23). a small
chtypan on Eurinilla Creek, close to the
boundary between FROME and CURNA-
MONA on Evrinifle map sheet, The western
edge of the pan is formed by a cliff 6 m high,
where scction 9 (Fiz. 15) was measured
(Appendix 2). Here, the Eurinilla Formation
is represented by mottled very pale orange and
strong brown clayey silt withoue stratification,
capped hy a massive gypsum horizon with
0.3 cm rosettes of gypsum crystals developed
in red-brown silt and gypsum flour. It discon-
formably overlics the Namba Formation,

Burrowed horizons and gypsified roots are
locally common in the Eurinilla Formation,
though not represented i the sections de.
scribed here.

‘The carbonate zones ut the top of the
Eunnilla Formation in sections 5 and 7 are
regurded as a single widespread paleosel, They
differ from the soil developed on the overlying
Coonarbine Formation by having farger
patches of carbonate segregation. often in
several horizons, [requently weathering out as
solid sheets or lumps, In the Arboola Claypan*
large soft calcureous “biscuits” are developed.
In which the onginal lamination of the

146

cemented sediment is visible, This paleosol has
heen identified along Balcanoona and Poontana
Creeks, on the west side of Lake Frome (Figs
1, 15, section 2) where it is developed in
coarser grained sediments.

The Eurinilla Formation is often underlain
by a course croxs-hedded sand or conglomerate,
Yunnamed conglomerate’ usually partly
cemented with hard white lime. The beds are
light pinkish brown, from iron-staining on sand
grains. A typical sequence including this
facies occurs at Lake Pinpa reference section
{section 8, Fig. t5, and Appendix 2). The
basal conglomerate often contains clasts. of
Namba Formation dolomite er Willawortina
Formation carbonate nodules. At Lake Tar-
kurooloo the islands on the lake Aoor (espect-
ally near section 6, Pig. 15, where massive
carbonale cemented pink sands and congio-
merates ate interbedded with the charophyte
limestone), demonstrate the gradation io the
Millyera Formation.

The disconformity between the Millyera and
Eurinilla Formations is exemplified in sections
= and 7, but ig not at all obvious in section 8,
or elsewhere away from the vicinity of Lake
Frome. The relationship can, however, be
observed along the Pasmore River, particu-
larity Where the main Yunta to Flinders Ranges
road crosses. Here, two terraces of secondary
carhonate-cemented conglomerate occur, inter-
bedded with yellowish sands containing green-
ish White carbonate novules at the top. The
nodules. are interpreted as a paleosol, and
occur in similar yellowish sands of the Millyera
Formation in Lake Tarkarooloo. Therefore
the unnamed conglomerate and associated
sands are regarded as Millyera Formation
equivalents.

The red brown silty Eurinilla Formation
with its characteristic soil developed af the
top, infills a valley cut into the ‘unnamed’
conglomerate, and associated sediments. The
whole is cut into Willawortina. Formation
sandy clays. light brown sands of the Coonar-
bine Formation disconformably overlie afl
units at vatinus levels in the landscape,

The relationships between the Eurinilla and
Willawortina Formations is also exhibited in
section 3 at the mouth of the Pasmore River
(Fig, 15 and Appendix 2) where reddish-
brown pebbiy silt with a basal conglomerate
(Burinilla Formation) rests with ‘sharp
erosional disconformuity on pale vreen and red-
browa mottled clay (Willawortina Formation),

R. A. CALLEN & R. H. TEDFORD

On a regional scale the disconformity sur-
face beiween the Eurinilla Formation anc
Coonarhine Formation as flat, but locally, river
valleys are developed,

Within Lake Frome are several islands con-
sisting of up to 10 m thick of coarse well
rounded gypsum sand with minor quartz, and
interbeds of clay pellets. These exhibit the
low angle cross-hedding, lithology and gec-
metry of [unettes described hy Bowler else-
where in Australia (pers. comm, 1974, J. M.
Bowler). The sands rest disconfarmably an
the Millyera Formation indurated clays, and
are tentatively correlated with the Eurinilla
Formation, Similar lunettes flank the eastern
shore of Lake Frome.

COONARBINE FORMATION
Lake Coonarbine,
FROME).

The type section is Jocated at Lake Moko
sevtion 7, Fig. 14 and Appendix 2), mentioned
earlier, The sequence (Fig, 20), resting dis-
conformably on the Eurinilla Formation, con-
sists of three parts—a basal 1-0 m of a red
brown indistinctly Jamibated sand, overlain
by 3,3 m of light brown dune sand (two large
scale cross-bed sets are represented) with a
carbonate soil horizon at the top. This is aver-
lain by 0.6 m of light brown sand with car-
bonate patches and rhizondules, The laminated
sand ai the base of this sequence may be a
distinct unit in its own Tight, since it has
features different from the remainder of the
Coonarbine Formation, Its disconformable re-
Jation with the Burinilla Formation has been
established.

The carbonate soi horizons are much
weaker than those of the Eurinilla Formation
in the same section and at section 5 ¢Fig, 15).
The upper more prominent horizon is corre-
lated with that al the top of section five, The
Coonarbine Formation in this section exhibits
the typical blocky joint pattern, producing 5 x
10 cm columns of sediment (large ped struc-
tures), Land-snail shells occur here, and ul
other widely separated localities, being charac-
teristic of the unit, The uppermost Inyer is
associated with aboriginal artifacts and emu
shel] fragnients. The Formation can be traces
west to the Pasmore River (c.g. section 3)
where it overlies the Eurinilla Formation.

An important supplementary section {sec-
tion 2, Figs 15, 21 and Appendix 2) repre-
senting a coarser facics uf the Coonarbine
Formation of western Lake Frome, is found

(Derivation
Ceonarbine map sheet,

CAINOZOIC ROCK UNITS

in Baleanoona Creek, near the natural gas
pipeline (Air Photo Ref.: S. Aust. Dept. Lands
Svy 394, Artareala Rut 3, Photo No, 0078).
At this site the old land surface on top of the
Lurinilla Formation ts exposed, The overlying
beds of the Coonarbine Formation cansist of
1.70 metres of dark brown, sandy silt, with a
basal pebble bed, moderately poorly sorted,
No bedding plancs are visible, and columnar
ped structure is well-developed.

Immediately downstream the surface of the
Coonarbine Formation is scattered with abori-
ginal artifacts, the colour’ is redder. and Jand
snail shells are present. Upstream, near Mulga
Hore on "Balcanoona", the hasal pebble bed
has 0.3-1 m thivk lenses, cutting into ihe
Eurinilla Formation. Carbonate nodules from
the soils developed in the Eurinilla Formation
are eroded and incorporated into the basal
Coonarbine Formation-

East of Lake Frome the fuviatile facies of
the Coonarbine Formation gives way to
atolian seif dunes, forming the partly indurated
cores of the modern dunes of the southern
Strzelecki Desert, Exposures occur along the
flanks cf the modern dunes. The gypsum
lunettes of the islands have a deposition break
within them, the significance of which is un-
certain: it is likely that the part above the
break correspands to the Coonarbine Forma-
tien,

RELATIONSHIPS BETWEEN
FORMATIONS

The Millyera Formation rests disconform-
ably on the Namba Formation in Lakes Mill-
yera and Tarkarooloo, but its relationship to
the Willawortina Formation is less clear. The
correlation of conglomerates and sands at the
Pasmore River Section with similar facies at
Lake Tarkaroolop has been mentioned, aid
suggests the Millyera Formation conglomerate
equivalent is also disconformable on the Willa-
wortina Formation. The relationship ts similar
to that at the “Wertaloona” section. Further
supporl is derived from the presence of bright
orange to red-brown silt and sand, similar tw
that in the Millycra Formation of section 5,
intertonguing with the conglomerate around
the mouth of Balcanoona Creek,

At the “‘Wertaloona™ section {section 1,
Fig. 3) the dipping scyuence of Willaworting
Formation is overlain with angular uncon+
fommity by a small patch of horizontal con-
glomerate and vellow sand. regarded as
Millyera Formation equivalent. The conglo-

147

merate conmains pebbles of ferruginized
matesinl, derived fromi what was prohably a
widespread siirface, mow exhibited as small
remnants in the same valley. This ferruginiza-
fion & correlate! with that beneath the
Mijlyera Formation at Lake Tarkaroaloo, and
clsewhere. Deformalicn of the Tertiary sc-
quence occurred before deposition of the
Millyera Formation and development of the
ferruginows horizon.

The disconformity between the Eurinilla
Formation and Willawortina Formation can
be seen in clifis along the Pasmore River, The
clearly disconformable relationship between
the Millyera Formation and Eurinilla Forma-
tion is seen in section 5 (Fig. 15), The dis-
conformity 1 less obvious for its equivalent,
the “unnamed conglomerate” of lakes Tar-
karooloo, Pinpa (?units 1 and 2 af section 8,
Fig. 15), aad clsewhere.

Relationship between Eurjnilla and Coonar-
bine Formations can be easily observed (for
example in sections 2, 3, 7 and 9, Fig, 15),
The Coonurbine Formation can be frequently
seen cutting into the Eurinilla Formation, and
the iwo unils usually have contrasting lithology.
The soil carbonate at the top of the Eurinilla
Formation may be completely eroded and
reworked into the younger unit.

Rock relationships are summarized in Fig.
24.

AGE

The Millyera Formation has equivalents at
the southern edge of Lake Callabonna, and
northern end of Lake Frome. ft closely re-
sembles laminated green clays and sands
bearing Diprotoden found in the main part of
fake Cyllubonna. The temporal range of
Diprotedon is Pliocene to fate Pletstocene. A
wood radiocarbon age of >40 000 years BLP.
(Daily 1960) from these beds has lately been
confirmed by unother wood radiocarbon date
of >39 900 years B.P_ (Tedford 1973). At the
mouth of Poontana Creek, on the Lake Frome
—Lake Callabonna confivence, dates from
shells in sands equated with the Millyera
Formation give ages of >33 400 years. BP.
and 35200 +! 200 years B.P_ (GaK-4949,
GukK-4948). This shell materin] has been
allected by younger pedogenesis, converting
them to calcite {assuming the shells were ori-
ginally all aragonite as sare most non-marine
oiolluses). Therefore the dates are minimal,
and the Millyera Formation has an age in
exeess of 34 200 years B.P\, probably >40 000

Las

years BLP, Similar shell beds in a similar strati-
graphic sequence were recorded al Lake Eye,
and gave a date of 39 200 = 1 300 yeurs B.P
(Johns & Ludbrook 1963),

The Eunnilia Formation contains — late.
Pleistocene vertebrate fossils, somewhat dif-
ferent in generic composition to thuse at Lake
Callabonna, The fauna occurs in channels at
the base of the urnl, along Billerow Crock east
of Lake FPinpa

The overlying Coonarbine Formation is
prébably late Pleistocene or eatly Recent.

REGIONAL CORRELATION

Equivalents of Millyera Formation are litle
koown al present, though the sequence de-
scribed imminediately above the Etadunna
Formation in the Madigan Gulf region of
Lake Eyre North is apparently very similar
{King 1956, Ludbrook 1956, Johns & Lud-
brook 1963). The lithological similarity
between the fossiliferous greenish sands con-
taining Coxiella giiesi in Mudigan’s Gulf, and
those in the Millyera Formation of section 5
(Fig, 15) is marked. All these beds are close
to or beyond the limits of radiocarbon daling,
but the closely comparable micro-fauna (in-
cluding Elphidinm spp., Ammonia beceart,
Nonion sp: pers. comm, 4, M. Lindsay 1974),
charophytes and molluscs tend to support
correlation. The Lake Eyre sequence Tests on
the Etsadunna Formatinn, and is overlain by
rocks resembling the Ticrari Formation,

The Eurinilla Formation closely resembles
the Tirati Formation of the Lake Eyre Basin,
in lithology, stratigraphic position and topo-
graphic expression, Vertebrate faunas in basal
Eurinilla Kormation channels indicate equiva-
lence with the youngest Kutapiri Sand (Stirton
et al. 1961) of the same basin.

Other possible equivalents are indicated in
Table 4. Vhe Poorska Formation (Firman
1956a) supposedly resis un Telford Gravel
(Firman 1963) on the west side of the Flin-
ders Ranges, and is overlain by the Lake
Torrens Formation (Williams & Polach 1971).
The unnamed conglomerate equivalent of the
Millyera Fermation lithologically resembles
the Telford Gravel at Telford open cut, Leigh
Creck. The Eurinilla Formation, lithologically
resembling the Lake Torrens Formation, aver-
lies the Millyera Formation, and ts in ttm
overlain by the Coonarbine Formation. The
lane is similar to the Thomson Creek Forma-
tion GF Willianys & Polach. There also are

R. A. CALLEN & R. H. TEDFORD

similarities in the calcareous soil horizans of
each, in the same geomorphic situation.

The Pooraka Formation, Telford Gravel and
“unnamed conglomerate” of Lake Prome area
are probably equivalents. Jf has been suggested
by Firman (1971)* that the Telford Gravel is
eguivalent to the whole of the Tirari Forma-
tion (Eurinilla Formation correlative), but
this cannot be the ease in the Lake Frorne
area, The youngest profable equivalents here
are the convlomerate at the base of the
Burinilla Formation, and the most likely corre-
lative the “unnamed conglumerate” equivalent
of thé Millyera Formation.

The unit mapped as Pooraka Formation on
COPLEY (Coats 1973) ts Coonarbine Forma-
tion. During mapping COPLEY, Callen &
Williams (in Coats 1973) recognized a unit
of reddish brown sand and cobbles which
covered most of the surface of the high level
plains flanking the eastern Flinders) Ranges,
The unit was Jater named the Afrowie Forma.
tion by Coats {1973}: subsequently mapping
for FROME has shown it is probably partly
equivalent to the Coonarbine Formation. The
two units beth contain Jand snuil shells, and
appear to grade laterally into one another at
the bresk in slope at the basc of fow hills
south of “Wertaloona,” However. Coats secms
to include some vounger and older gravels in
his definition, with %disconformable relation-
ships.

ENVIRONMENT

The Millyera Formation constitutes three
facies groups: the most typical and widespread
ate Yhe laminated ostracode clay and chara
phyte limestones (Fig. 18), with associated
charophyte eogonia-bearing fine sand. Fine
lamination, ostracodes, and distribution of
sediment, indicate they are undoubtedly of
lacustrine origin, The fine sands are well
rounded and smooth and mav therefore be
aeolian, haying been blown inte the lake, or
carried by floods. Drying of the lake is indi-
cated by the charophyte limestone and equiva-
lent gypsum lamellae (Figs 16, 18; cf. Reeves
1968, p. 57, 58). Similar madern calcareous
algal deposits (Fig. 19), grading to rippled
gypsum crusts, arc present in Lake Kuturu’.
Waves acting on the very shallow water badies
break vp the filaments and orient them in
crescent like ripples, sometimes resembling the
oriented structures in thelr fossil equivalents.
The gypsum Jaminac may have hotryaidul
surfases that are reminiscent of similar forms

149

SeUIW 4O JUaWjiodag “ys

vaya va 10¥—PvZ

—e— —t— shol> puo |
SaNVS |

NOOMNdWYW

AQYNSTLYND
AVa¥d

Ol AaVILd1
i1¥1

33} Dia WO] Buo> , Z : |
SU St Cl NOILYWYOS VNILIOMVTIM

pawouuy |

O-L TWAVID |

({luA Bulloag
uapojoidig)
"Ainby NOILYWHOS
“WaSATIIW

ONYS ldld¥ivy Pus
NO|LYWaOd ldVaiL

(un Buioag
—Psiq |!s8o4)
juajoainby
NOllyvwaod
INIad¥NOOD

CAINOZOIC ROCK UNITS

BuVogD||D> 2x07 8143 407 7961

("Wor 'siad ‘E76 1)

a4¥Osd3L

“y2 48 NOLMILS

TOsOAlWd YNVLVIIUIM

NOILVWYO4
SNaddOL IVT

1OsSOIIWd
YNad Low

(j40d JaMoj)

NOILYWao4
4434S NOSWOHL

1OSOIWd
VNOOOYN

{od uaddn) -4
43D NOSWOHL

saBuoy
SUPPUlfy Wiad A
‘1261 ‘HDV1Od
PYO SWYITTIM

dyOsTdL
T9A¥aID GYOITIL

MOS Vard¥a

NOILVwaO4d
VAVYOOd

WOS Yavnvee

WOs AV3A07

Os Vavevd
ONY ALadDdAD

syisodap Builds punhow

NOILY¥wacd
Va¥aoOd

eo

NOllYWwao4

NOILVWaO4
WAVYOOd

WOS AVTIAOI

_ -aIMOxay
6— =~ AVYD
YNNOBVTI¥>

GNV¥S NOSdWIs

uIsog asAQ 2407
‘saBuoy
SJEPUl|) UsBISaAA “Nh
"NY WHld

jeays dow
000 OS2*1 AIIdOD
“JB 49 SivVOD

NOlLWWwaod
VaFATIIW
aLaysW> ; :
tc
NOILYWdO4
VINIGNG

(MIDjsaaun aaug

alvwaWOISNOD
OWYNNN

Gig ayydosGyoug)
YOrOW oO, By) uing |Ssoq jo

JLFYDIVD PY FL3¥DdAD

UOZIIOY [105 snoasosj0> jisso4

sajjaun| wosdAB sapnjour

INDI

NOLVWHOI INIEYYNOOD

-IN3D015131d

UOZIVOY |[Os sMOBIOD}O> jisso4

POojpag woays
syisedsp 9904
saunp usapow

aadod siyy

QdOsdd1 ® NAlIVWS

SNISVS Y3HLO ONV SL3SHS d¥W LNADVray - LNZIVAINGA GaLsa99Nns

SLINM YI9NNOA - LYVH) NOILVTIAYO)

yay
3WOYs AVI

¥ A18V1

180

on the sufface of modern Lake Frome, pro-
duced by crystallration pressure bockling the
surficial crusts,

The second facies group is the channel facies
(Fig, 17), of conglomerate/sand which ex-
hibits Features of meandering streams of large
size containing bed-forms of slightly crescent-
shaped aqueous dunes. The streams carried
pebbles from the Olary Ranges, and eroded
valleys into the Namba Formation, These
deposits are lateral equivalents of the “un-
named conglomerate” which is so extensive
along the Siccus—Pasmore River System.

The third facies group are the greenish fos-
siliferous sanis, which (Section S, Fig, 15}
are cross-bedded on a small to medium scale,
and contain shell beds and fish vertebrate.
Similar shelly are also abundance in a narrow
zone slong Billerog Creck between Lakes
Kuturu and Tarkarooloo. These deposits ate
interpreted us shoreline facies of the Pleisto-
cene Lake Frome.

These sediments, and equivalents at the
northern end of Lake Frome, contain the
foraminiferal assemblage mentioned earlier
(p. 147). Similar species were also recorded by
Ludbrook (Ker 1966, p. 94) in equivalent
strata in McKenzie Bore, 7,5 km sauih south-
eust of section 5, The presence of several
species of foraminifera over a wide area in the
same scdiments can be explained in terms
of Ludrook’s (1965) hypothesis of transport
to salt lakes on the feet of seubirds, with
subsequent survival for a period, The species

R. A, CALLEN & R H, TEDFORD

present ate mostly Rotaliina with a wide
salinity tolerance, and diversity is low. Such a
situation ts typical of inland saline lakes (Resig
1974), where foraminifera have been intro-
duced by some dispersal mechanism from
coastal areas. Species such as Ammonia bec-
carti are common in these environments.
Although the assemblages found at Lake Eyre
and Lake Frome are considerably different in
content from those listed in Table 4 of Resig’s
paper (e.g. Nonion spp. are not recorded,
though common at Lakes Frome and Eyre)
this does not detract from the dispersal hypo-
thesis because each locality cited in her paper
has high endemism. The Coorong area con-
tains « similar assemblage (pers. comm. J. M.
Lindsay 1975), though its low diversity is pro-
bably the result of high salinity, even though i
has a connection with the sea.

Another explanation is that there was a
distant connection to the sea, implying a high
sta level during the Pleistocene prior ta 40 600
years BLP.

The detrital component of the facustrine
Millyera Formation sediments were brought to
the ancestral Lake Frome by large braided
streams with a pebble bed load (“unnamed
conglomerate”) approximately following the
channels of present day watercourses such as
the Pasmore-Sictus River system, and the
Luke Tarkarooloo-Billeroo Creek system. They
were much more extensive than their modern
counterparts. The clasts. indicate 4 provenante
similar to the modern streams, li the Olary

Figs 16-19. Younger Units. Structures in Millyera Formation.
Fiz, 16, Millyera Formation. Laminated tipple-murked gypsum and clay (Fig. 15, section 5). Scale 30

cm.
Pig.

17, Plan view of cross-stratificd channel sand in Millyera Formation channel facies, bed of Lake

Tarkacooloo near Section 6, Fig, JS, Approximates Pi cross-stratification. Current direction
(arrowed) is te north. Hammer handle 25 cm long. Lominace emphasized Sy inking.

18, Algal tubules showine rough onentation. Same jocality as Fig. 22. Scale in cm.
. 19. Modern, calcarcous charophyte algal flaments, Lake Kuturu, showing crode orientation. Thin

crust of gypsum {G) in upper central part of photograph.

Figs 20-23 Oitecrop,

20. Upper part of section 7, Fig. 15, showing dune sand facics of Coonarbine Formation (1wo

upper benches), basal laminated sand (bench Just above contact), and tipper part of Furinilla
utmation with calcareous paleosol (just below contact).

. - Section 2, Fig. 15, Columnar-structured sand of Coonarbine Formation overlying Enrinilla

Formation. Surface in foreground shows carhonute patches of paleosol, and represents the pre-
Coonarbine Formation land surface slightly modified by present erosion. Scale 30 cm.

2. 22. Coonasbine Formation sand with columnar jointing, overlying Millyera Formation which in tum

avetlies Nainba Formation. Millyera Formation shows wpper algal limestones und lower mias-
sive sandy Jimestone (prominent benches) with intervening clayey sand. Lake Tarkaronloa,

near Coombes Bore- Scale 30 cm,

Rig. 23. Section 8. Fig. 15 Cower part}. Coonarbine Formation, disconformably overlying Eurinilla
Formution which has its opper surface cemented with secondary gypsum (prominent hench)-
Black vlay of Namba Formation at base (30 cm) scale crosses contuct)-

COONARBINE

CAINOZOIC ROCK UNITS

a Ls gio -é
a “e J * ‘i
EURINILDA’ Be
ae

*

152

Ranges and southern Flinders Ranges. Large
straight-crested aqueous dunes typified the
streams with coarse sandy and pebbly bed
loads, whereas crescentic dunes characterized
the streams with a finer sand-bed load, The
eastern shore of Lake Frome was estimated to
be about [0 km further cast than al present,

The Eurinilla Formation contains channel
deposits, exemplified by coarse sand with
parting lineation and cross-bedding, in troughs
and point bar deposits along Billeroo Creek,
The meandering form of these channels can
sometimes be seen on aerial photographs. The
pebbles have sources in the Flinders. Ranges
or Olary Ranges, or have been eroded from
the underlying Tertiary units. Flood-plam de-
posits are represented by the finer facies,
which is sometimes laminated, The initially
fluviatile phase (basal coarse grained sands)
gave way fo a more complex environment with
finer fluviatile deposiis and large scale cross-
bedded aeolian deposits, including huge gypsum
lunettes along the south western shore and on
the islands. Some possible Joess (massive silt
and very fine sands) is present, These sedi-
ments transgressed over the older lake deposits
of the Millyera Formation, The ancient Lake
Frome therefore decreased in size in medial
Pleistocene times, being somewhat smaller than
at presenl,

The plains of this essentially fiuviatile en-
vironment were inhabited by large marsupials
(Dipratedon sp., Procaptadan geliah, Sthenu-
rus sp. and Macropus sp.) Rivers followed

EURINILLA FORMATION

UNNAMED CONGLOMERATE
WILLAWORTINA FORMATION

Eocene

2 Paleocene
S1O970

COONARBINE FORMATION

EYRE FORMATION
R.A.CALLEN

R. A, CALLEN & K, H, TEDFORD

approximately the same coarses as the present
day drainage. The distribution of lunettes indi-
cates a dominant wind direction from the west
and a strong westerly component still charac-
terizes this region.

The overlying Coonarbine Formation in
cludes fuviatile braided stream environments
west of Lake Frome, and dominantly aeolian
¢ast of the lake. The fuviatile sediments have
less defined channels than the Eurinilla Forma-
tion, pebble sheets being more common. East
of Lake Frome Jongitudinal dunes were deve-
loped, and another minor phase of gypsum
lunettes built up along the lake shore. Land
snails probably lived around water holes.

Conclusions

The new rock units in the Lake Frome area
record a history of intermittent deposition
through Miocene to late Pleistocene-Recent
times, During this interval the extensive rivers,
lakes and possibly estuarine environments of
the Miocene Namba Formation drained areas
of low relief in a climate of high rainfall, and
of higher annual temperature than the same
latitude today, At times, seasonal dry periods
became a part of the weather pattern. A con-
nection with the sea was established at some
stage, probably to the Murray Basin, Some
conflicting climatic evidence is partly resolved
by applying the continental Jessivage hypo-
thesis (Millot 1964, as modified by Wiersma
1970) in relation to the smectite—dolomite—
palygorskite mineral suite, Thus warm lem-

Member 2

Member |
NAMBA FORMATION

Y

S.A. Department of Mines

Fig, 24, Cuinozoic rock stratigraphic relationships, Lake Frame.

CAINOZOIC ROCK UNITS

perale to subtropical conditions prevailed, with
savannah landscape, and gallery forests around
the Jatge permanent streams and lakes,

Uplift of the Flinders Ranges securred at
the earliest during late Miocene times, con-
tinved through the Pliocene intermittently into
the Quaternary, and is still proceeding at
present, Prior to this, at least during the
Cainozoic, the Flinders Ranges were virtually
non-existent. The sediments deposited during
the Pliocene-early Pleistocene Epochs record
the change from the earlier Miocene palaeo-
gcography to the very different landscape
approximating that of the present. Lakes and
swamps during Tertiary times disappeared
dunop the Pleistocene, as tectonism and
Climalic change altered the depositional
regime. Drainage resembled that of the present
during the late Pleistocene, indicating the basin
was approaching its present configuration,

The Millyera Fotmation indicates active de-
position on a playa lake somewhat larger than
ihe present. The changing character of the
sediments from Millyera to Coonarbine For-
mation suggests overall increasing aridity, pro-
bably seasonally distributed during Eurinilla
Formation limes, as exemplified by the forma-
tion of the gypsum Iwnettes. Marked climatic
Muctuations were superposed on this overall
climatic trend. Uplift of the ranges continued,

153

alternating with periods of stability during
which soils developed.
Rock relationships are summarized in Fig.

24.
Acknowledgments

This paper is published with the permission
of the Director of Mines, South Australia
Drafting and typing facililics were provided
by the South Australian Department of Mines,
Participation of R.H,T. was supported by
National Science Foundation Grant GB-
b8273X.

The authors also wish to acknowledge dis-
cussion and helpful criticism by Mesrss J. B.
Firman, W. K. Harris and J. M. Lindsay of
ihe South Australian Department of Mines,
and by Dr B. Daily of the University of
Adelaide. The project was initiated by Me
B. P. Thomson of the South Australian Depart-
ment of Mines.

Radiocarbon dates were supplied by K.
Kigoshi, Gakushuin University, Tokyo,

The following companies provided core
material and cuttings: Mines Administration
Pty Ltd, and Mr J, Andrus of Western Nuclear
Aust. Pty Ltd is especially thanked for arrang-
ing WC2 bore to be donated for use asa type
section.

The skilful drafting of Mr Bruce Thomas
of the South Australizn Department of Mines
i acknowledged,

References

ALLEN, J. R. Lb. (1963)—The classification of
cross-stratified units, with notes on thelr ori
gin. Sedimentol. 2, 93-114.

Benton, Y. K., Bopenneimer, W,, & Herrer, L.
(1963)—Clay minerals mm Negev sediments.
J. sedim, Petrol. 33, 874-903,

Bryxs, P. J, (1972) —Late Cainozoic uplift of the
Ediacara Range, South Australia, Proc. As-
erelec. Inst. Min. Metall. 243 (Sep. 1972) 47.

BUISSENBACH, B, (1954),—Geology of alluvial
fans in semirarid regions. Bull. zeal. Sov. Am,
6S, 175-190,

Brewes, R. 41964),.—"Fabric and Mimeral Analy-
sig of Soils." (Wiley: New York, London,
Sydney) 470 pp,

Baown, A. Y. 1, (1884) —Report of the Govern-
ment Geologist. Parl. Pap, 3. Aust. 35, 1-20.

Catten, R. A. (1975)—FROME geological
sheet S1154-10. Geol. Atlas Ser, geol. Surv.
S. Aust.

CaunEnN, R, A. (1976).—Lake Frome area—
regional geology, Tertiary Stratigraphy and
uranium localization. Jn Knight. C. L. (Ed.),
"Economic Geology of Australia and Papua
New Guinea”, Vol. I. Metals, Monagr. Awy-
tralas, Inst, Min, Met, 803-808.

Coats, R, P. (1973)—COPLEY map sheet, Exe
Planatory Notes 6-250 000 Geol, Atlas Series,
pect SH54-9, Geol. Surv. S. Aust. (Ade-
aide),

Daice, B, (1956).—The Cambrian in South Aus
tralia, Proc. 20h int. geal, Cong, (Mexica)
2, 91-142.

Datty, B. (1960)—Report of the Curator of Fos-
sils and Minerals, Repr 8. Aust, Mus, Board
1958/59, p, 12 (Adelaide),

Denny, C. S. (1967).—Fans and pediments. Aint
J. Sei, 265, 81-105.

Devereux, I. (1968)—Oxygen isotope paleotem-
peratures from the Tertiary of New Zealand.
Tuatara 16(1 4, 4144,

Finman, §, B. (7963),—Quaternary geological
events near Swan Reach in the Murray Basin,
South Australia. Qaert, Geal, Netes, geo,
Surv. S. Aust, 5.

Firman, J. B. (1964).—The Bakara Soil and
other stratigraphic units of the late Cainozode
Ave in the Murray Basin, South Australia,
Quart, geol. Netes, geal. Sued, §, Aust. 10,

-5,

Firman, J. B, (1966a).—Stratigraphic units of
Late Cainozoic Age in the Adelaide Plains
Basin, South Australia, Quart. geol. Notes,
geal. Surv. S, Aust. 17, 6-8.

154 R, A. CALLEN & R H. TEDFORD

Finman, J, B, (1966b).--Stratigraphy of the
Chowilla area in the Murray Basin, @uart.
geol. Notes, peol. Surv. S. Aust. 20, 3-7.

Firman, J. B, (19672)—Late Cainozoic strati-
graphic units, Quart. geol, Notes, geol, Surv,
S, Aust. 22, 4-8,

Fiemam, J. B. (1967b).—Siratigraphy of Late
Cainozoie deposits in South Australia. Trans.
R, Soc S. Aust, 9, 165-178,

Firman, §, B, (1969),—The Quaternary Period:
In Parkin, L. W. “Handbook of South Aus-
tralian Geology”. (Govt Printer: Adelaide).

Firman, J, B, (1970),—Late Cainoznic strati-
graphic unils in the Great Artesian Basin,
South Australia, Qvart, geal, Notes, geet,
Sarv. S. Aust. 36, 1-4.

Foik, R. L.. ANprews, P, 8, & Lewis, D. W-
(1970) — Detrital sedimentary rock classifi-
cation and nomenclature for use in New Zea-
Jand, N.Z. J. Geal, Geaphys. 13, 937-948,

Frevret, P. (1971}.——Paleosols residuels et paleo-
sols alluviaux hydromorphes dans le Cretace
superisur el ['Eocene basal en Languedoc.
Revue Geapgr, phys, Geol. dyn. 2, 245-288.

FREIDMAN, G- . AmreL, A. J, Braun, M. &
Mitsera, D. 8S. 41973),—Generntion of car-
bonate particles and laminites io algal mats—
example from sea-marginal hypersaline pool,
Gulf of Aqaba, Red Sea, Bull, Am. Ass. Per-
rol. Geol. 57, 541-557.

Grr, E. D. (1968) —Onygen isotope palaeatem-
perature determinations from Victoria, Aus-
tralia, Teotare 16¢L), 56-61.

Hayes, BE, & Frakes, 1. A. (1973).—Leg 28,
deep-sea drilling in ihe southern ocean, Geo-
limes 18, 19-24.

Howcnrn, W. (1909),—Description of an old lake
urea in Pekina Creek and its relation to
Recent geological changes. Trans. R- Soc. S-
Aust, 33, 253-261,

HowcHin, W. (1913}—The ¢volution of the
physiographicol features of South Australia.
Tn: Australasian Assoc. Adv. Sct. 14, p. 168:
Hall, T, S. (Ed,

Jack, R. L. (1930).—CGreological structure and
ather factors in relation to underground water
supply in portions of South Australia: Bull.
geal. Surv, 3. Aust. Wa.

Jenxins, D., G. [1968).—Planktonic foramini-
ferida as indicators of New Zealand Tertiary
paleotemperatures. Tuatara 16( 1), 32-37.

Jessup, R. W. & Norris, Ri M. (1971).—Caino-
zoic stratigraphy of the Lake Eyre Basin and
part of the arid region lying 10 the south, J.
geol. Soc. Aust. 18, 303-333.

Jouss, R. K, & Luonroox, N, H, (1963),—Inves-
ligation of Lake Eyre Parts 1 and If Repr.
(nvest,, geoi. Surv. §. durt. 24.

Kenny, EB. J. (1934).—Darling districi—a_ peo-
logical reconnaissance with speciul reference
io the resources of subsurface water, Mineral
Resour. N.S.W. 36.

Ker, D. S (1966)—The hydrology of the Frome
Embayment in South Australia: Rept. Invest.,
peal. Surv,, §. Aust. 27,

King. D. (1956).—The Quatetnary stratigraphic
record at Lake Fyre North and the evolution
of existing topographic forms. Trans. R, Soc,
¥. Awst. 79, 93-103.

Krinstey, D. B, Woo, C. C. & Sroerrz, G. E.
(196%) —Geologic characteristics seven Aws-
Irslian Playas: Ju Neal, J. T. (Bd.), “Playa
surface morphology; miscellaneous investiga-
tions’ OS. Air Force Cambridge res. Lab.
Environ. Res. Pap, 283, $9-103.

Linnsay, J, M. (1969)—Cainozoic Foraminifera
and stratigraphy of the Adelaide Plains Sub.
basin, South Australia, Bull, geal Surv. §.
Aust. 42.

Linpsay, J, M, & Srervearn, R. G. (1966)—
Muono Para Clay Member. Qwart. geal.
Notes, geol. Surv, S. Aust. 19, 7-11.

Luperoox, N. H, (1956}—RMicrofossils from
Pleistocene to Recent deposits, Lake Eyre,
South Australia. Trans. R. Soc. S. Auss. 79,
37-45.

TLuprroox, N. H, (1965) —Occurfence of fora-
minifera in salt lakes. Quart, geol, Notes.
geo! Surv. §. Aus 14, 6-7.

McLean, 8. A., ALLEN, B. L, & Craic, J. R-
{1972),—The occurrence of sepiolite and
altapulgile on the southern high plains. Clays
Clay Miner, 20, 143-149,

MEESTER, T. DE (197L)—Highly calcareous lacu-
strine soils in the Great Konya Basin, Turkey.
Versl. landbouwk. Onderz (Agr. Rept) 752.

Mrvzaor, G. (1964)—Trans. Fanranr, W. R, &
Paquet, H. (1970),—“Geology of Clays".
(Springer-Verlag: New York, Heidelberg,
Berlin, Paris; Chapman & Hall: London.)

Muis.er, G., Tron, G, & Forstner, U, €1972).—
Formation and diagenesis of inorganic Ca-Mg
carbonates in the lacustrine environment.
Naterwissenschalfer $9, 158-164,

Muaray, J. W, (1968).—Living foraminifera of
fngoons and estuaries. Micropaleentalegy
14, 435-455.

O'Driscoit, E. P. D, (1956),—The hydrology of
the Willochra Basin. Repl Invest., peel. Surv,
S, Aust, 7, 58 pp.

Reeves, C. C. (1968).—“Introduction to Paleo-
limnologsy", Developments in Sedimentology
Hl (tElsevier: Amsterdam, London, New
York).

Resic, J. M. (1974)—Recent foraminifera fron
& landlocked Hawaiian Jake. J. Foraminer,
Res. 4, 47-60.

Sriwrn, A. R. C. (1860),—Geological notes an
a journey in Somh Australia from Cape Jer-
vis fo Mt Sorlc. Parl. Pap. 5. Aust. 20.

Sincrr, A,, Gar, Mo & Bantn, A, (1972) —Clay
ming¢rals in recent sediticots of Lake Kin-
neret (Tiberias), Israel, Sediments. Geol, 8,
289-308.

Smuen, A. & Norurso, H. K. (1974). —Pedogenic
palygorskite aceurrences in Austtalia, An,
Miner. 59, 508-517.

Staton, KR. A,, Tevrorp, R, H, & Mincer, A, H,
(1961)—Cenozoic stratigraphy and verte-
hrate palaeontology of the Tirari Desert,
ha Australia, Rec, 8. Aust. Mus. 14, 19-

Smeaton, R. A.. Tepronn. R. H. & Wooprurne.
M. ©. (1967)—A new Tertiary Formation
and fauna from the Tirati Desert, South Aus-
tralia, Ree. S. Airst, Mus. W, 85-93.

CAINOZOIC ROCK UNITS 155

SUTHFRLAND, F, L, Green, D. GC. & Wart,
B. W. (1973),—Age of the Great Lake
Rasales, Tasosania, in relation to Australian
Cainazoie volcanism, J, geal, Soc, Ansi. 20,
5-93.

Tepeoap, R. H. (1973)—The Dipretodons of
Lake Callabonna. Aust. Nat. Hist. 17, 349-

354.

Top, D. H. (1968).—Palaeoclimatology and the
relative stability of feldspar minerals under
almospheric condition. J, sedim, Petrol, 38,
§32-844.

VON DER BorcH, C. C. (1965)—The distribution
and preliminary geochemistry ef modern car-
bonste sediments of the Coorong area, South
Australia. Geochim, Cosmochim, Acta 29,
781-799,

VON DER Boacn, C. C., Lock, D. E., & Sciwener,
D. (1975),—Grountwater formation of dola-
mile in the Coorong region of South Aus-
tralia. Geology 3, 283-285.

WeLiMan, P., McEcamny, N. W.. & MeDoucart,
[. (1969)—On the Polarwander path for
Australia during the Cenozoic. Geophys, JR,
aatr. Sov. 19, 371-3958.

Wirrsma, J. (1970).—Provenance, genesis and
palcogeographical implications of micro-
Minerals occursing in sedimentary rocks of
the Jordan Valley area. Pr-Fys. Geogr. Bo.
Univ. Amsterdam 15,

WituiaMs, G. EB, & Potacnh, H. A. (1971)—
Radiocarbon dating of arid zone calcareous
paleosols, Ryil. geal, Soc. Am. 82, 3069-3086,

Woprner, H. -(1974),—Post-Rocene History anc
stratigraphy of northeastern South Australia,
Trans. R. Sac. S. Aust, 98, 1-12.

Worerner, H., CALLEN, R. A,, & Hamas, W. K.
(1974),—Lower Tertiary Eyre Formation of
the southwestern Great Artesian Basin, J.
geol. Sov. Aust. 21, 17-52.

Appendix I
OLDER CAINOZOIC UNITS
Fig. 3, sections 1, 10, IL, 12. Fig. 14, section 13

NAMBA FORMATION TYPE SECTION
SADM Yalkalpo No. 1 Sttatigraphic Bore. Fig. 3 Section 12
COONARBINE FORMATION
2.00. SAND. medium grained, strong reddish brown (2.5YR4/6). Subrounded grains. Rec-
tangular joinc pattern und carbonate cylindroids in upper part.
— Disconformity —
NAMBA FORMATION

Unie 18 140 CLAY; slightly silty, with scattered gypsum spois, Light grey (N.4+) with moderate yel-

3.40 lowish orange (([0YR5/6) patches, Sharp upper contact.

Unit 17. 1.35. Interbedded SAND and CLAY overlain by SILT. Sand is very fine grained, laminated

4.75 and small scale cross laminated, a3 are the silt beds at the top of the unit, Lower beds. have
structure destroyed by secondary gypsilicution. Contacts between sand and clay beds are sharp
and flat. Sand-filled shrinkage cracks extend down from the wavy irregular upper contact.

Unit 16. 645. SILTY CLAY. Poorly sorted, diffusely laminated. Two silt beds with gradational to

12,00 sharp contacts near the top, lower 7 m burrowed and bioturbated with wavy irregular upper
bedding plane separating it from the remuinder of the sequence. Irregular shiny fracture planes
(crumbly texture) and gypsum nodules developed in upper (part. Light to mederale olive grey
(S¥4/2-SYS/1, 5-SY6/1). Mottled black and yellow (S5Y4/0.5) in lower silt.
0.60. No recovery.
0.20, As before. Shrinkage cracks filled with sand extend down from upper unit—wavy
irregular contact, Light olive grey (SV5/2).

Unit 1%. 0,20, SILT, Laminated, very fine cross-laminated, burrowed in par Very pale yellowish grey

12.20 (NI to 5¥9/1), Upper contact sharp and flat. burrowed.

Unit 14. 3,00 SILTY CLAY. Intraformationally brecciated. Borrowed and bioturbated in lower half,

£5.20 upper half with angular clay clasts and slump structures, Upper contact sharp, flat.

Unit 13, 1.70 CLAY SILT and CLAY, Poorly sorted, Clay beds near centre. Rurrowed at top, intra-

17.30 formational breccialion common. Pale to light olive clay (10Y6/2 to 6/4), sill pale grey to
pale yellowish grey (SY8/1 to N7), Upper contact sharp and flat.
0.40, SAND and SILT, interbedded, weakly laminated, wavy ‘sharp contacts between inter-
beds. Sand is very fine, very angular (quartz crystal faces—overgrawths on rounded grains,
contacts between overgrowth and erain visible),

Unit 12. 1.05. No recovery.

19.89 1.24. SAND, grading up to CLAY SILT and SILT. Sand hus small-scale cross lamination with
heavy mineral luminae.

Unit 11. 2.05, CLAY, black as below, Contact with overlying bed sharp and fiat.

21,52 1.13. No recovery,

} 0.75. SAND, grading up to CLAYEY SILT. Sand very fine, wavy lamination, Colour SY4/4,
Unit 10. 0.20, STILT and CLAY, Intertaminated, flame structures on contacts. Light olive grey (Clay
26,82 SY6/4, Silt NO). Contact with ynit 11 sharp and flar.

2,30, SILT. lower bed laminated .and cross-bedded, with scattered burrows.

result of bioturbation and extensive burrawing. Upper contact wavy and irreg tar,

9.30. CLAY, laminated, colour NO, 4¥4/2, With CHRLONIA scute at 24,21 mm. Upper con-
tact sharp, with flame structures.

156 R- A, CALLEN & R. H, TEDFORD

1.40, SAND, fine grained in lower 1/5 grading up to !aminated and burrowed light grey
(NS) SLLT. Sand grains smooth and shiny, angular, some well rounded and frosted, many
with crystal faces and re-entrants. Upper contact wavy und irregular. Grading down to - .
1.10, SIL], laminated and small scale cross-hedded, with scattered burrows:

Units. 3.20, CLAY. black (N1) mottled light olive brown (5Y5/6), with sand patchcs and other fea

35.52 tures as before, Contact with sand lower in unit gradational. Large sand grains in the patches
are polished, rounded to well rounded, smaller graims being angular to subangular with over-
growths{?). Some of the larger grains show rounded crystul forms.
0.35, Ne recovery,
3,60. SAND, us below. Calcite patches and. very coarse mica common. Polymodal, poorly
sorted overall, Small sizes angular, coarse are rounded, some doubly terminated crystals. In-
terbedded light grey (N7) clay in centre of unit. Becoming well sorted and fine grained at. top
with mixed well rounded and angular grains, Obscure small scale cross bedding and Jamination,
0.60, No recovery,
095, SAND, Coarse to medium grained, slightly calcareous. Large grains polished, others with
crystal faces (overgrowths?) which give siepped shiny surtaces. Many grains shew original
elongate quartz prism shape. :

Uai.8 0.70. Alternating SAND and CLAY, Sand very fine grained with small scale trough cross-

36.22 lamination. Clay olive to medium grey (SY5/0.5).

Unt? oan. Clay as below, Sharp wavy upper cantact,

39.22 1.25. No recovery.
145, Alternating SAND-CLAY fining upwards sequences, Several thin beds, beginning with
very fine grained sand al base. Sharp !lat contacts, grading to black (NZ, SYR2/1) clay with
orange brown specks and sand patches in top 21cm,

Unit6 0.94, SANDY SILT. grading up to CLAY. Hlack (N2) clay as for unit 5, with fine sand

40.79 patches, 1/3 of sequence. Sharp wavy upper contact.
0.58, No recovery,
0.05..SILT with slump structures.

Unit 5. 6.55 SAND, SIT-T and CLAY alternating in lower 1/3, grading to CI.AY at top. Lithology a3

47.44 for unit 4. Obscurely laminated in lower part, lenticular bedding, Light olive grey (SY6/1)
with oxidised brawn patches (LOY R67). Calcareous at transition (25 cm) to dark clay. Sand
distributed in vertical strevks and patches through the dark olive clay (5Y5/1). Irregular
wavy upper contact, - A

Unitd. 3.30. SILT, grading up to CLAY. Lower 1/3 laminated silty and clay with very-small to small

50.84 scale crass bedding st base, some burrows. Pale yellowish grey (S¥9/1) and pale olive
(JOY6/24. Grades rapidly into sandy ¢lay with vertically orténted structure and lime streaks,
yellowish grey to pale olive (3V7/2 to 9Y8/2), This grades fo tough black (NI) clay with
erange brown dendrite mottling and patches of fine sand. A thin brown band of ton oxides
is present. Upper contael wavy, irregular.

Unit, 3. 1.20. SAND, grading to CALCARHOUS CLAY. Lower 40 em very fine grained, loosely

52.04 cemented by guartz overgrowlhs (original grains rounded) calesreaus at base, Heavy minerals
1%, Dusky yellow (SY4/1), Clay is olive crey (5¥4/1) and has white vertically oriented
streaks und sheets of carbonate, Contact with overlying unir is sharp and wavy.

Unit 2. 1.94. No recovery.

56.60 0,22. SAND, very fine grained as for bage unit, Contact with underlying clay irregulai, with
mixing.
1.10, CLAY, as for base unit 1, silty top, with moderate yellow green clay patches which be-
come dull on exposure (7GY4/1, 5Y5/2, 5¥4/2). Obscure lamination.
1,00. No recovery. P . .
0.20. SAND, very fine gfained, scattered medium, polished grains, Opaques common. Sone
palches modenue yellow green clay (7GY4/1-5Y4/2).

Unit |, 1.20 CLAY, waxy lustre on curved ipregular fractures. Rare angular white carbonate lumps
and streaks. Olive grey (SY¥S/2Z)-

WNAMBA FORMATION
Supplementary, Section Outcrop, West Sule Lake ‘Tarkarooloo, Section 13, Fig. 14
Section 13A. north of the northern track crossing Luke Tarkarvolao.
COONARBINE FORMATION
bid, SAND. Very culeareous, with numerous 1-5 cm dolomite nodules and ferruginous sandstote
lumps (reworked from Namba Formation). SYR8/5-
— Disconfermity —

EVRINILLA FORMATION
2.40. CLAY-SILT-SAND. Silt (dominant) to fine grained sind, poorly sorted, Lower If om moderately
sorte medium grained sand. Up to §0% reworked dolomite madules (pisolitic) of granule to pebble
sixe in lower part, 72 x 5 cm maxianim size, and little evidence for abrasion, Upper 10-70 em with gyp-
sim nodules and curbonute patches. Green-bluck terrruginous and manganiferous stain, Olhcrwise lel
red-brown 2.5YRS/6,

| — Desvonturmity —
NAMBA FORMATION
3.85, CLAY, GYPSUM, Slightly silty clay, very hard, ileht weight. crumbly with greasy lustre, Colour

CAINOZOIC ROCK UNITS 137

SYS/2 ta 9/1, mottled red-brown, yellow brown and black, specked with while gypsum flour, 80 cm
hetizon of caulilowcrokt gypsum nodules 1,10 m from top. Nodules 0.25~0.50 om dian. with clay
Cores, nt associated with any porosity change in host sediment.

1.25. DOLOMITE, CLAY. Lower dolomite 6.60 m clay 0.50 m upper dolomite 0.15 m,. Lower dolo-
mile nodulue (intraformatianal conglamerale} al lop, manganese dendrites (hraughout. White, very fine
grained, with 5% angular silt, Upper dolomite contains natracades. Clay a3 above,

1.95. CLAY. S% angular silt to very fine sand, grading to sand at base and dolomtite at top. Very hard,
jight, dry and crumbly, with manganese stain.

145. SILT. Sift and very fine sand, becurning clayey at lop. Bioturbuted ul lop, 2-3 cm burrows, other:
wise finely Laminated, Minor bright green clay faminae, Top 40 cm impregnated with gypsum, mortied
orange browse Very light grey to white overall, Lower contact sharp, far.

70. CLAYSILT, DOLOMITE. Clay with 40% sill, 5-1b% very fine sand. Arenaceous fraction of
moderately sorted, angular grains, Colour 3Y6/L. orange to orange-brown mottling, Nodules and
patches of greenish-white very line grained sandy dolomite, weathering ax 2 cm granules al top and
bottom, Black stained,

0.78. Alternating SAND and CLAY. Sand beds range from silt to very fine sand, colour 5V8/2. Clay is
sandy. Individual beds 2-27 cm. Lower beds have shatper contacts and are laminated, very finely cross-
bedded and lensing. Grange brown mottling and gypsum) iy upper part of section. Miner tedulur dolo-
mile,

0.80, SILTY CLAY, Silt to very fine sand 10-20% svbangular frosted grains. Clay bas greasy lustre
and SY6/2 colour. Upper contact very sharp, Ast.

1.00. SAND and DOLOMITE, Sandy clay and clayey sand with dolomite nodules 2-30 om thick at top
and bottam, Sand is subanyular to subrounded. moderutely well sarted, with frosted grins: Weathers
to hard Jight brown sandstone, base cemented with gypsum, Sandy clay is S¥5/Z, clayey sand S¥7/3,
MmotUed with yellow green patches, and black pateles at base (NI-NS5). Very sandy parts are INYRG/2-
SYRS/2 ut buse.

Dorel is very fine grained with botryoidal upper surface, in way, lensing beds. Greenish upper sur
aces,

0.30. SAND, As above,

0.33. CLAY. Mottled clay and very fine sand, finely Juminated beds which ate irregular and lenticelar,
and have yellow green (10¥R6/6} clay patches and, cross-cutting brown (10YR8/6) patches with black
centres. Sand (5¥7/2). Basal 2 cm fine white sand, Gradational upper contact. Contains dark brown to
yellow-orange irregular silcrete and ironstone nedules.

0.02-0.03. CLAY. As above. Sharp contacts.

0.23. SAND. Very fine, yellow green clay lamellae. Obscure medium cross bedding, straight foresets.
Stained greenish black. Sharp upper contacts,

0.25. CLAY. As hieher in the sequence. Upper S-[0 cm dolomitic, yellow-sreen and brown mottled.
0.10 SAND, As above. Obscurely cross-bedded, selenite cemeni al base.

1.30. CLAY. Silty, Very finely laminated, with very small scale cross-hed sets. Colour SY6/0,5, aces.
sional black and brown patches, Upper contact sharp, flat,

0.70. CLAY. Slightly silty, hard, with sub-conchoidal fracture and ereasy Justre. Selenite on contact with
overlying unit. Stight greenish brown dint, otherwise N4.

Section 138, south of the nerthern track, crossing Lake Tarkarnoloo (200 m sowh of section J3A).
Constructed from a series of breakaway slopes, and a scoured channel In the centre of the lake,
Beginning from the top of the black clay, as al the base of the previous section, the sequence is us
follows (from gally sequence};

3.5. CLAY. Dark vrey elay with irregular shiny surfaced fractures, becoming IHehter coloured towards
base. Crops out poorly, forming low angle vegetated slopes, Upper contact sharp, flat. Lewer contact
moderately sharp to disturbed.

2.1. SAND, Very fine to fine gruined, black stained. Thin S cm beds and fine laminate, Brown silcrete
nodules throughout, Thin horizon of DOLOMITE nodules ac top, Abour 20 cm above the base are 10
cm of laminated silty limestone with low amplitude symmetrical ripples. mud cracks, and 1-3 mm
tong tubules of organic origin, Sharp basal contact with -, -

0,25. CLAY, Ax al lop of sequence (13B).

On a small nol isolated near the edge of the fake, the sequence continues os follows, with some overlap;
0.55. DOLOMITE, Sandy nodular whire dolornite. Impregnated with xypsum and calcite, plus tabular
black manganese concretions.

0.98 SAND, Fine grained, well sorted, laminated and silty at base. Cemented by gypsum and calcite in
part. Numerous curbonate nodules cahibiling concentric structure. Some have vertical tubular disposi-
tian with am intemal structure snggestive of shrinkage (c.f, silorete nodules}, Hlack and brown stain at
base. with orange brown and yellow Sreen patch stain. This, and the previous unit are equivalent to rhe
2.) m of sand described above, with its hed of dolomite nodiiles.

2.5, CLAYEY SILT. Alternating hard clay and clayey sift in very fine Jamellae (varve-like), Silty grey
clay lenses. Yellow green patches with waxy lustre. Clay dominant at base. N35, IDYS/4 to 4'6. Con-
tact with underlying unit sharp, undulating: sume contac! as at base of 2.1 m sand described in paily
section abave.

0.5, CLAY. Hard, areasy lustre, irregular Tracture. N3 at top grading down to SY¥4/I,

This clay crops out across the bed of the Jake jo its centre, where 4 scour next to a salt spring exhibits
a further 2.5 moof massive hand ercy clay,

158

R. A. CALLEN & R. H- TEDFORD

NAMBA FORMATION SUPPLEMENTARY SECTION
SADM Wooltana 1 Stratigraphic Bore. Section Ll, Fig. 3

°EURINILLA FORMATION
0.0 1 5.2 m No core

3.2

0.9 COBBLE CONGLOMERATE: Cemented with réd-brown carbonate, Mica schist, gneiss,
quarizite. Sindy (medium grained), with subrounded to well rounded, erains,
— ?Disconformity —

WILLAWORTINA FORMATION AND NAMBA FORMATION {intertonguing})

Units
a4
7.2

iA)
12,50
28.65

33.00

38.6
43,05
48.00

53,15

67.65

NAMBA
Units
74,35

1.1 DOLOMITE. Sandy, clayey at base. SYR6/4 to N6. Sharp upper contuct, gradutional
lower contact,
3.8 CLAY. Slightly sandy, with dolomite nodules. Clay 7¥7/2 with red-brown vertical pipe
structure, Sund subangulur to subrounded. Some gypsum patches im lower part, Lower con-
tact gradational, 50 cm core missing,
0.65 SAND, Fine sand, very poorly sorted, subangular to subrounded, Thin slntomite beds.
Dolomite nodules, Clay at base.

11.65 to 12.50 Na core

16.15 CLAY erading to SAND at base. Numerous dolomite nodules and sume beds,
Extrentely. poorly sorted mediim sand to coarse sill, Some gypsum patches at top. Green and
red-brown pipe structure, Sand subangular to well rounded, Sharp basal contact. 40 cm core
missing in sand interval,

4.35 CLAY grading to SAND at base, As above. Mica in basal sand. 2.5¥R/4-6, 5Y6/)
moriles. Sharp wavy basal contact.

$.6 CLAY, and DOLOMITE erading to SAND at base, As above, but dolomite beds in upper
clayey part, Dolomite in 20%. clay, Medium sill, extremely poorly-sorted. Sand subangular
to rounded. Gradational wavy lower contact.

4.45 CLAY, grading to SAND at base. As above. Dolomite beds (brown) and nodules
throughout. Sand patches al top. Lower contact eradatianal,

4.95 CLAY grading to DOLOMTTE then SAND at base, As above. Mouted red-brown, green
and yellow grey, Very poorly sorted. Minor core loss. Lower contact gradational,

5.18 SANDY CLAY grading to SAND at base, As above, Dolomite beds and nodules
throughout, Sand laminated, micaceous. Intraformational clay and carbonate at hase, Sand
reaches medium grain size. Subangular to rounded, Disturbed irregutar basal contact.

14,50 CLAY cf. NAMBA FORMATION, Patchy sand and carbonate near top, Reticulate nét-
work of carbonate “veins®. Lower part with limonite nodules, irregular shiny blacke-stained
fractured clay. Obscurely laminated in middle part. Sand very fine, angular to subrounded.

SY6/1, 2SYR6/2-8, 1OYRB/2, 2.5YR5/6, SY7/1.5, SY6/i, SY¥4/5, 2-3 m core missing,
mainly in upper part.

.7 SANDY CLAY. Micaceous finely laminated silt becoming pebbly at 72 m, reverting to
clay at base. Extremely poorly sorted, Granite, quartzite, shale, quariz and eneiss pebbles,
Very angular fo subrounded. Clay intraformationally brecciated and burrowed in lower part.
Mincr carbonate, c.f, WILLAWORTINA FORMATION, Gradational lower contact.
FORMATION

4.9 Alternating DOLOMITE atid CLAY, Dolomite (5-10 cm thick), oolitic white, aphanitic.
with charophytes, gstracodes, molluscs and unidentified calcareous %plant fossils. Numerous
burrows in clay beds (all about 1 m thick). Micuceous silt in part. Bioturbated, Laminated
at base, Sharp basal contact.

7.9 CLAY, minor DOLOMITE. Upper purt similar to above, laminated and burrowed, Clay
GYO/T+ ZY RAG, becoming sandy at base, With rounded clay clasts, Basal contact grada-
tional,

1.75 SANDY CLAY. Calcareous. SY5/1 to 5¥6/1. Sharp basal contact.

0.10 CLAY. Sharp contacts.

2.00 Interbedded DOLOMITE and CLAY. Dolomite as above. Very finely interlaminated with
brittle, swelling clay, Shrinkage ¢racks common. Burrowed, Contacts on carbonate bedé sharp
and wary to disturbed irregular, 7¥571, 5¥6/1, N3,5, 5¥4/2.

2.55 MARL and GYPSUM. Alternating thin selenite atid calcareous clay, Sharp contacts of
gypsum. Gradational lower contact. Black to dark olive.

4.60 CLAY grading to DOLOMITE and MARL at base. Numerous gypsum nodules, 4¥5/1,
FY8/2. Mottled SYR6/7. Minor gypsum lAminac, Contacts wavy eradational to disturbed
irteguiar. Intraformational brecciated,

3,35 CALCAREOUS CLAY. Trregular shiny Fractures, oxidized red-brown patches. Swelling,
very porous. Subaqueous shrinkage cracks,

$15 CLAY. Sandy in centre, with selenite veins infilling slickensided joints. Sand laminated.
Clay with irregular shiny surfaced fractures as above, 10-15% very fine sand to silt. Alternat-
ing colour pattern—oxidized red-brown clay passes down io 3¥4/2 10 6Y7/1 clay, has sharp
upper surfaces. Basal contact regular, disturbed.

7.L CLAY. As above. Busal thin white dolomite. Churmed structure suggests biaturbation,

TAS CLAY grading to SIT.T in lowe half. Silt well laminated, micaceous, very small scale

125.45
129.00
130,80

138.00
140,40

Unit 3
147.30

163,40

165,20

166.80
170.80

175.30
183.7

187.3
196.10

Unit 2
197.0

241.0

2273

236.0

CAINOZOIC ROCK UNITS 139

cross-laminated, Grains yery angular to sub-rounded, Silt moderately sorted. Lower contact
moderately sharp. :

4.55 Alternating SAND, CLAY. Sand very fine, micaceons, small scale cross bedded,
laminated, Lower contact moderately sharp,

1.80 As above. Lower contact irregular, wavy,

$5.00 SAND, interbedded CLAY, Sand fine to medium, well sorted, small to mediutn scale
eross-bedded_. Micaceous and with clay balls. Grains very angular to subrounded. Sharp wavy
contacts.

2.60 SAND, As above, Upward fining, Some clay at lop, burrowed. Poorly sorted, average
very fine grained. Wavy moderately sharp lower contact,

6.70 Aftemating SAND, SILT, CLAY, CARBONATE. Sand as above. Carbonate nodular.
Clay dark grey. Bioturbated and intraformationally brecciated. Mainly fining upward
sequences, Sand dominant. Contacts irregular, disturbed.

26.140 CLAY grading to SAND in lower 1.5. m, Manganese nodules above sand. Minor sand
bed at 157 m. Cloy N45, numerous imegular shiny fractures. Sandy, averaging very fine silt,
very poorly sorted. Sand and very fine silt, very poorly sorted. Sand very fine, burrowed and
laminated. Very angular to angular grains. Irregularly disturbed lower contact.

1.8) SILT. 5Y6.5/0.25, orange brown moltles, obscure lamination, bioturbated. Sharp wavy
fower contact. Grades io very fine sand at base,

3.4 SAND, Fining upwards from fine sand to clay. Thin clay bed with sharp contacts near
tup. Medium scale cross-bedded. 6Y6/1 to S¥7/2. Sand poorly sorted. Moderately sharp
wavy lower contact,

4.00 CLAY grading to SAND at base. Upper | m sandy, lowee very fine obscurely laminated
sand, Clay with irregular fractures, SY4/1. Sand beds have gradstional contacts,

4.4) AS ABOVE. Minor sand at 172.2 m, Basal very fine sand, moderately sorted, obscurely
small scale cross laminated, burrowed, Gypsum nodules. Clay as above with fractures and
orange [6 red-brown moltles. Sund very angular to angular.

74 Silt. Laminated fine, very poorly sorted silt, xypsiim nodules at top. very small scale
cross-bedded, some burrows. 3m missing im central part. Sharp wavy lower contact.

4.6 SANDY CLAY, gradmgz to SAND at base, Cycle as above, Some calcareous zanes, SY¥5/1,
SY3/4, Sandy at (85 ny, Basal sand fine graincd, sub-rounded. Moderately sorted, Lower cou-
tact moderately sharp.

&.8 SANDY CLAY, SAND at base, Upper 3/4:N2 to SGY7/0.5 clay as above with fractures
and sandy patches. Dolomite nodules at contact with sand. Sand very fine to fie, very poorly
sorted, SY6/! to N&, Burrowed near top sand, rest Wloturbated, obscurely cross-stratified,

0.9 SILT, Minor alternating clay and sand, sharp contacts. Silt laminated, Lower contact
shafp and wavy.

14.0 Alternating DOLOMITE, CLAY, SILT and SAND, Complex inter-relationship between
clay-dolomite cycles and sand and clay beds c.f. above. Contacts variahle. Dolomite intra-
formationally brecciated, Sut laminated. Bioturbated and burrowed horizons. Clays with
iffeeular fractures and orange mottles. Beds 40 cm—1-5 m thick.

163 Alternating LIMESTONE, DOLOMITE and CLAY. Consists of 1-2 m carbonate beds
grading up to clay via, disturbed zone, Clay beds burrowed or biolurhated, |OYS/2 to
IGYR7/2. Base of dolomite beds sharp und wavy. Dolomite aphanitic, white laminated, with
oolitic zone. Ostracodes common, algal mats present. Zone 215-217 m of very narrow clay
filled cracks, Irregular shiny fracttres dominute in clay near base, otherwise absent, At top of
this sequence is 5 em genethite-limonite crust.

8&7 CLAY. Calcareous, intraformationslly brecciated with numerous white carbonate specks,
Irregular shiny surfaced fractures, N1 to olive green, Quartz rare, very fine to fine, angular,
Lower contact rradational.

8.45 CLAY, Fissile pyritic carbonaceous very finely laminated clay with sift parting. Fine
laminac of N1 to dark olive or SY2/1 colour’ Numerous plant stem and leaf impressions or
fruiting bodies, fish spines and scales, ostracodey (often in pure layers), gastropod prato-
conchs, spores and pollen. Burrows (pyrite filled) and bedding plane traces. Numerous pyrite-
marcasite nodules. Some subaqucous shrinkage cracks.

WILLAWORTINA FORMATION TYPE SECTION

Werstern Nuclear WC2 Bore. Section 10, Fig. 3
*MCOONARBINE FORMATION and JEUIRINILLA FORMATION

0.00 to 7.05 Cuttings only, SANDY PEBBLES, SAND. Micaceous. calcareous, impregnated with gyp-

sunt (except sand). Angular to very angular, 2. 5¥R4/8, 2.5YRI/6,
— Disconformity —

WILLAWORTINA FORMATION—MEMBRR 3

7A5
7.89
$54
984

0.44 No recovery,

1.65 SAND as below, ]-39% muscovite.

0.30 No recovery.

3.95 SAND grading to SILT. Sand very poorly sorted and very fine erained, silt coarse, pebbly

160

R. A. CALLEN & R. H. TEDFORD

and micaceous. Thin pebble bed grading up to sand at base, Porous zones impregnated with

gypsum,

0.25 No recovery. .

1.21 PEBBLY CLAY SILT, poorly sorted, with basal pebble bed, impregnated with gypsum.

1.50 No recovery.

0.24 PEBBLES, subrounded to rounded, Coarse feldspathic gneiss, fine biotile quartz feld-

spar gneiss, purple stained coarse feldspar,

0.68 No recovery.

0.13 SAND, pebbly, coarse grained, Feldspars, muscavite, biotite

1.98 No recovery—a few abraded pebbles.

0.24 SAND, fine grained and micaceous.

0.22 No. recovery,

0,38 GRANULES grading up to PEBBLES.

1.45 No recovery.

0.28 GRANULES to SAND, medium grained. Larger grains subrounded. Mica 5%, pink

potash feldspar 15-20%.

{).64 No recovery.

0.33 CLAYEY SAND, very fine grained and micaceous, sharp wavy content with overlying 20

cm of COBBLES. 5-10% muscovite, pink potash feldspar. Yellowish grey (SY7/2Z).

1.49 No recovery.

0.11 SAND, grading up to GRANULES. Sand medium grained with dull, pitted or shiny sur-

faced, angular to subangular grains. 15-20% feldspar (mostly pink potash variety). Musco-

vite and biotite flakes in the quartz.

G.80 No recovery.

0.20 SAND, pebbly, micaceous and overall fine grained.

0.72 No recovery.

0.14 COBBLES, sandy.

0.47 No recovery,

0.17 Pebbly micaceous coarse sand.

1.01) No recovery:

0.27 SAND, grading over short interval to COBBLES at top. Sand is poorly sorted and

micaccous, medium grained.

1.61 No recovery.

0.27 COBBLES, passing over short interval to medium micaceous SAND,

0,72 No recovery:

0.53 SAND, grading to COBBLES in upper half, ‘Sand micaceous and poorly sorted, medium

grained, Clasts pink quartzite with mitaceous hematite, purple fine quartzite, vein quartz.

large potash feldspar pebbles.

1.00 No recovery.

Q.26SANDY PEBBLES, micaceaus.

0.58 No recovery.

0.30 SAND, coarse and poorly sorted, micaceous.

0.30 No recovery,

rey PEBBLY SAND, very poorly sorted, micaceous, coarse grained. Dusky yellowish grey
5Y¥6/2).

0.57 No recovery.

0.24 CLAY, sandy, micaceous,

0.62 No recovery.

0.13 PEBBLY SAND, fine grained, moderate reddish brown (2.5YR5/6 to 4/6),

2.00 No recovery. f

0.30 CORBLES and PEBBLES overlain by fine grained CLAYEY SAND.

0.20 No recovery,

9.56 GRANULE bed, thin, overlain by thick SANDY COBBLY PEBBLE bed (white vein

quartz, pink potash feldspar, pink ferruginous quartzite, coarse siliceous gneiss or granite, dark

grey Shale, weathered fine gneiss).

0.77 No recovery.

0,20 SAND as before, with clayey sandy cobble bed at top.

0.71 No recovery.

2.46 SAND, coarse at base, grading up to extremely poorly sorted fine sand. Vertically

oriented reddish brown pipes.

().29 No recovery,

0.53 PEBBLY SAND, sand medium grained.

0.23 No recovery.

2.77 Fine SAND as before, grading up te pebb)y medium grained sand in upper | /3.

0.18 No récovery-.

1.47 SAND, poorly sorted and very fine grained, micacteous, with scattered pramiles, two thin

pebble beds at base.

0.63 No recovery.

0.35 PEBBLES, passing over short interval ta SAND, clayey, micaceons and fine grained.

poorly sorted,

CAINOZOIC ROCK UNITS 161

50,05 1.64 No recovery. _

51.69 a CLAYEY SAND, coarse and very poorly sorted, grading to boulders. Metaquartzite
clasts.

52.03 1.23 No recovery.

53.26 0.29 SAND, micaceous and yery fine, light yellow-brown (7YR5/6). Very poorly sorted.

53.55 125 No recovery.

54.80 0.50 SAND, very micaceous, very fine grained. Pebble bed in centre (granite, banded pink
and white quartzite).

55.30 0.58 No recovery.

55.80 0.63 SAND, fine, grading up to fine grained with scattered granules.

56.43 0.30 No recovery.

36,73 0.44 SAND, as above. Feldspathic quartzite pebbles.

57.17 0.48 No recovery.

57.65 0.40 SAND, as above, pebble cobble bed in centre.

58.08 0.49 No recovery,

58,54 0.84 CLAY, and pebbly SAND, coars2 grained. Grades rapidly to micaceaus CLAYEY
SAND, fine grained, in upper 30 cm.

59.38 0.69 No recovery.

60,07 0.56 SAND, as below, fining to fine grain size at top,

MEMBER 2

60.63 0.66 No recovery.

61.29 0.54 SAND, as below, no granules.

61.83 1.43 No recovery.

63.26 0,99 SAND, slightly clayey, medium grained, with granules. Coarsening upwards.

64.25 1.50 No recovery.

65.75 Ps CORRES: massive pink granite, very fine dark quartzite, pink feldspar with ?horn-

ende,

65.98 1,30 No recovery.

67.28 0.10 SAND, medium micaceous and clayey.

67.38 0.19 No recovery,

67.57 0,34 GRANULES, grading to CLAY-SILT.

67.91 0,84 CLAYBY SAND, coarse, pebbic interbeds.

68.75 0,05 No recovery.

68.80 0.47 SAND, coarse and poorly sorted, pebbly, Pebbles of quartz and gneiss.

69.27 0.14 No recovery.

69.41 0.41 SANDY CLAY, as before grading up to COBBLY SAND.

69.82 0.12 No recovery,

69.94 0.18 SANDY CLAY, as before. Micaceous.

70.12 0.25 No recovery.

70.95 0.85 Silty clay grading up to cobbles,

71.82 0.33 No tecovery.

72.15 0.27 SAND; medium, grading up to CLAY.

72.42 0.13 No recovery.

72.55 4.07 SILT, extremely poorly sorted, medium size.

76.62 0.11 No recovery.

76.73 1.40 SAND, very fine, pebbly, grading to SANDY SILT CLAY.

78.13 0.95 No recovery,

79.08 0.85 GRANULES (lower 20 cm) grading up over short interval te SAND, clayey, medium
grained, very poorly sorted,

79.93 0,17 No recovery.

80.10 1.68 SAND, very fine, grading to very coarse at top. 1% miuscovite and biotite, 10-15%
potash icldspar. Grains very angular to. subangular and dull, Small grains shiny and faceted.
Extremely poor sorting,

81.78 0,15 No recovery.

81,93 0,25 SAND, fine grained, very poorly sorted.

82.18 0,16 No recovery.

82.34 ee GRANULES, basal bed, grading over shoft interval to very poorly sorted CLAY, very
sandy

MEMBER 1

83.18 0.06 No recovery.

83.24 0.76 CLAY, lower 10 cm sharp contact with SAND, coarse grained to granule sized. CLAY
thin bed at top. Extremely poorly sorted,

84.00 0.37 No recovery..

84.10 0.10 SAND, fine grained, clayey.

84.47 1,37 No recovery.

85.84 1.50 SILTY CLAY. Very poorly sorted, with thin coarse sand beds.

87.34 2,03 No recovery.

89.37 5.74 SILT, SILTY CLAY, Extremely poorly sorted coarse silt, silty clay micdveous. Thin

coarse grained sand bed (91.2 m), above which the sediment coarsens from clay to very fine
gtained sand.

(62

95.11
95,36

97.37
97,63

101.42

102.40

104.80

105.90

110,80

121.20
21,90

123.35

126.45

R. A, CALLEN & R. H. TENFORD

0.45 No recovery. ;

1.8) SAND, grading to CLAY a1 top, Sand very coarse, clay silty and micaceous with a thin
eranule bed near the top. Yellowish grey (SY8/1)-.

0.26 No recovery. , .

3.79 SANDY CLAY, SAND. Sandy clay has medium grained sand fraction, very poorly
sorted, 2rades to clayey coarse sand, with very angular lo subangular pitted to shiny grains,
Feldspar is common. The base of this interval is taken as the base of the WILLAWORTINA
FORMATION.

1.08 No recovery.

NAMBA FORMATION ;
2.30 CLAY. 15-20% subanvular to subrounded sand, minor mica. Sand patchy near base,
with irregular shiny-surfaced fractures (skew-planes), SY6.5/2 to 4¥5/1. Basal contact sharp.
30 cm Core missing neur base.
1.40 CLAY. As sbove. 10% sand, no mica, nodular and swelling with well developed frac-
tures. Alunite mottles at base, 20 cm core missing in centre of interval, 1Y4/1, to 6YR6/1 at
base, Sharp basal contact. :
4.90 CLAY, grading to SILT ip basal 60 cm. Intraformationally brecclated and burrowed (at
109 m} with some laminated intervals. Well developed alunite streaks, which decrease in abun-
danee with depth, being absent al the base. 6YR6/1-6/2 grading to N8 at base, 70 cm core
missing al vatious intervals, mainly near top. Basal contact irregular, disturbed.
10,50 CLAY grading to SILT at 144.4 m and SAND at 119.4 m. Clay intraformationally
breceduted, 15-25% very angular to subrounded sane. Silt micaceous, Sand pnicaceous, cross
bedded in 30 cm sels, and Jaminated, fine grained, well sorted, Grains angular to subrounded,
Basat contact gradational, Weak alunite horizon 50 cn below top of unit, absent at 115 m.
Colour 5¥YR5/1 above alunite, N8 below, 2,1 m core missing in silt and sand beds. Wavy in-
diatinet luwer contact,
0.70 SILT. L

\.45 SAND. Micaceous, Jaminaled, obscurely croas-bedded_ Fine grained and moderately well
sorted. Shurp lower contact, 70 em core missing in centre of unit.
3,10 SAND. Minor clay at top, fine grained micaceous sand in centre, lower half grading to
very coarse sand iat base, Subangular to subrounded, large grains highly polished composites.
Mostly no core, there being 50 cm recovered. Basal contact sharp,
9.45 SILT and CLAY (below 129 m). Clay, nodular, dark brown, sill greenish white. Sandy
patches, Sand grains often show crystal faces—bipyramids. Indistinct contacts observed at
129,25, 129.4, 129,6 associated with weak alunite horizon, Colour 1OYR6/2-5Y8/1 in this
zone. Below 131,80 irregular shiny surfaced fractures and some alunite specks, Colour
SY5/1-4, SY3/1. 17% silt. Grains very angular io angular. Much core missing throughout.
recovery 40%.

WILLAWORTINA FORMATION SUPPLEMENTARY OUTCROP SECTION

—'Discentormity —

“Wertaloona” Homestead Arca. Section L, Fig. 3
WILLAWORTINA FORMATION

Unit $,

Unit 8,
37,4
Unit 7-
37.1

Unit 4.
107.0
Unit 4.

Unit 4.
Unit 3.
Unit 2.

Unit ft.
140.1

37,9 COBBIFS, Brown quartzite cobbles in a matrix as for unit 4. Basal bed of almost 100%
erey-blue limestone cobbles. Rare red sandstone, quartz and yellow-brown silicified. carbonate
cobbles in float. Exposure poor, top nat exposed.

49.2 SANDY CLAY, red brown,

20.8 COBBLE to BOULDER beds. Matrix as for unit 6, cemented with secondary white car-
bonste which may be powdery and soft, or hard vughy and crystalline. Cobbles of brown
quartzite with 20% blue-grey fimestone (resembling Cambrian limestones). Rare very large
boulders of grey massive microcrystalline quartzitic silerete with large milky quartz pebbles.
(3.0 (approx.} CLAY SAND. Red brown very poorly sorted and calcuréous,

2.0 (approx.) COBBLES. Brown quarizite cobbles scattered through matrix as for unit 4-
Lenses out along strike.
7.0 (approx.) CLAYEY SILT SAND. As for unit 2.
1.0 (approx.) PEBBLY COBBLES, As for unit 1, more matnx, thin and lensing along strike.
a8 {approx.) CLAYEY SILT SAND. Red brown, with a calcareous matrix, sametimes thinly
aminated,
4.0 (approx.) COBBLES. Brown quartzite pebbles and cobbles in calcareous red-brown silty
sand, lenses of calcareous medium sand at base. The sandstone fills channels, which have
groove casts on the base. Cementation is weak, and pebbles weather aut readily with thin cul-
careaus crusts, Proportion of matrix tow. The unit cuts into deep red brown clayey silt,
probably Namba Formation. Althaugh the contact here is sharp, there may be an inter-
tonguitig relationship along strike. The unit grades laterally to the south into pebbly clayey
satidstone,
The following part of the section is poorly exposed, and is yet to be fully described:

— Disconformity —

2NAMBA FORMATION
(2,0 SAND. Very fine greenish sand grading wp inte silty grey-green clay with gypsum patches.

CAINOZOIC ROCK UNITS 163

7.2 CALCAREOUS SANDSTONE. Very angular sand with soft crystalline carbonate cement Contains

pebules of very angular sandstone, carbonate, rounded brown quartzite, polished milky quartz, chert
ranules.

51 2. CLAY. Dark gréeen-grey clay with greenish-yellow-stained patches, slightly sandy, Thin white nodu-

far dolomicrite is preseml near the base, aid may be a facies variant of the previously described nit,

3.2 SAND, Reddish to greenish silty fine to medinm sand.

$2 CALCAREOUS SANDSTONE. Essentially a sandy limestone with aboui equal quantities of

medium grairied angular sand and lime. Weathers grey, With a sculptured ‘rough surface.

144.7 CLAY, Grey green to olive, greasy irregular fracture, sandy and silty. Minor dark olive to grey

clay. Mottled with red-brown iron oxides. The interval 430-350 m (measured from top of the unit 9

in the Willawortina Formation section) is very poorly exposed and deeply weathered.

Near the top of these beds in the northern part of the ares, is a thin white dolomicrite bed,

— Discanformity —
YEYRE FORMATION _
2.0 SANDSTONE. Massive calcareous medium grained sandstone, partly silicified, and capped by rem-
nant silcrete, dipping with the section. ‘
2,0 CONGLOMERATE, Granule to pebble-sized polished white guartz, grey chert, ironstonc. Pebble to
cobble-sized angular Middle Cambrian sandstone. All in medium well-rounded sand matrix, cemented by
calcium carbonate. Pebbles are patchily distributed, and the whole crops cut as a Jow ridge wath caver-
nous weathering and of brownish grey colour, Medium scale cross-bedding is prominent. The unit has
an apeplar unconformable relationship with the underlying Middle Cambrian red-beds, though dips arc
aimilar.

Appendix Ti
YOUNGER CAINOZOIC UNITS
FIG, LS SECTIONS 2-9
(See Fig. 1 for locations, and maim exe for access and photo points)
SUPPLEMENTARY SECTION, COONARBINE FORMATION

SECTION 2
COONARBINE FORMATION
1.7 SANDY SILT. with basal pebble bed. Sand dark brown (5YR3/5}, Size varies from silt to very fine
sand, moderately poorly sorted. No bedding planes Visible. Columnar structure well developed. Basal
clasis may be small cobble size, and are of metamorphic rocks and quarz,

— Disconformilty —
EUXINILLA FORMATION
2.2-2.5 CLAYEY SILT-SAND. Very poorly sorted. with irregular-shaped frosted or pitted grains, Con-
lains pebble lenses (though not in the figured section) and large irregular aphanitic greenish white
sandy carbonate limps. The latter are probably derived from the upper carbonate in Wooltana 1 bore
{section 1, Fig. 3). At top of O.5-1 cm diameter branching vertically oriented cylindroids of pinkith
“chalky” textured carbonate, representing a fossil soil horizon.

— Disconformity — .
0.2 CALCAREOUS SAND. Pebbly sand (coarse grained), solidly cemented by pinkish buff (SYR7/2)
carbonate, Colour derived mainly from orange-stained quartz grains. Laminated and thin bedded, Beds
dip, suggesting cross-bedding is present (outcrops seen in plan only, in creck bed), :
Possibly represents Willawortina Formation, or unnamed conglomerate equivalents of Millyera Forms-
bon.

SUPPLEMENTARY SECTION, COONARBINE FORMATION, EURTNILLA FORMATION
SECTION 3

Location, Curnamona Siccus map sheet, Air pheto ref.:; S. Aust. Dept, Lands Svy, 361, run.2, photo no,
4442. The section is situated on the northwestern bank of the Pasmore River, close to the point where :t
debouches in to Lake Frome,

RECENT

0.00=1.20 Mobile bright red-hrown dune sand, sharp erosional contact with underlying units.

— Disconformity —
COONARBINE FORMATION
1.90-3.50 SAND. Yellow-brown, with large scale dune-type cross-bedding, Sharp erosional basal con-
tact, A lag of pebbles (eroded from the Eurinilla Formation) is al the base,
Numerous broken Mature snail shells are present in the upper part of the unit. Aboriginal artifacts, cal
sified iree roots, emu shell, and vertebrate bones occur in the uppermost. level (or possibly on the
upper surface in the case of the artifacts and emu shells).
Strongly developed columnar structute is present (resulting from soil processes).

— Disconformity —
EURINILLA FORMATION
1.80 PEBBLY CLAY-SILT and SAND, Sand at base, medium-grained, yellow brown, numerous
pebbles and care flat cobbles, cemented by gypsum. Pebbles are milky and clear quartz, and very angu-
lor frugments of calcite-cemented conglomerate, overlain by bright red-brown silty clay
Unnamed Conglomerate (7Millyera Formation equivalent)

fh.15 CONGLOMERATE, Thin, calcite cemented. Pebbles weather out without adhering crust. Pebbles

164 R. A, CALLEN & R. H. TEDFORD

as for overtying unit plus ?7Namitha Formation dolaomictite. and brown carbonate nodules from Willawor-
tina Formation. Carbonate penetrates into top of underlying bed,

— Discontormity —
WILLAWORTINA FORMATION
2.05 SILTY CLAY. Sandy, preenish-brown with red-brown motiles. hard. Patches of gypsum nedules.
Partly calcified at top. Blocky columnar structure visible (resulting From soil processes). Upper contact

sharp, undulating. ;
MILLYERA FORMATION TYPE SECTION
SECTION 4

030 SAND. Reworked from older unit into base of dunes,
0.70 SAND. Coarse grained, with many gypsum grains and anomalous pebble sized angular quartz
(milky), Powdery hummocky gypsum often developed at top ¢sojl profile),

— Disconformity —
BRURINILLA FORMATION
1.10 CLAYEY SAND. Very fine-grained, sub-rounded to rounded. good sphericity, moderate sorting.
Numerous charophyté oogenta. Muny greenish, yellow and brown grains, Colour S5YR5/6, Capped by
£ypsum crust, of gypsum nodules in clayey sand (groundwater deposit),

— Disconformity —

MILLYERA FORMATION

Unit7, 0.50 CLAY. Soelft, conchoidal fracture. Contact with overlying Eurinilla Formation sharp and
flut. Very dark yellowish brown. The oxidized crumbly appearance and shiny surfaces (cutans)
on cnimbs suggest soil processes have operaled, and indicate a disconformity between Millyera

; and Evrinilla Formations, _ .
Unité, 0,40 SAND. Very fine to medium erained. Grains subangular to rounded and frosted. Charo-
, phyte oogonia .5%. Grades by alternation, 1a... |

Unit 5, 0.50 CLAYEY SAND. Sand fraction well-sorted, with subangular to angular rounded grains,
sharp fiat upper contact, Greenish yellow (1077/2).

Units, O70 CLAYEY SAND, Interbedded thin clay and very fine to fine clayey sand 0.25-0.50 mm
thick. The sand iv very well sorted, with subrourded to rounded high sphericity frosted grains.
Darker oxidized clay present, Yellowish grey (7¥7/2). Lower contact gradational,

Unlt3. 0.30 CLAY as for unt 1.

Unit2, 0.40 LIMESTONE and CLAY, Near the top of the sequence each cluy Jamina grades up to
charophyte stem-mould limestone (up to LE cm thick), These limestone beds harden on weather-
ing, producing sheets and slabs with » metallic ring when struck. Intervening lamellae are 0.5—
2 om thick. Some of ihe Charales tubules ar¢ oriented and small turreted gastropods. aif.
“coxiella” ure present (henceforth referred to as “Cextella). An oxidized zone exists beneath
the limestone. The limestones form.a distinctive marker horizon 20-30 cm thick. Contact with
unit | and unit 3 are gradational by alternation.

Unit!, 0.10 CLAY, brittle, soft, waxy lustre. Distinctly laminated and thin bedded (1-5 cm), each
lamina grades up to # thin fine silt layer wilh charophyte oogonia and Ostracoda, White car-
bonate granules occur near the base of the sequence, Scattered medium polished or frasicd
quarlz grains, sometimes up to 40% of the rock, occasionally forming sand lenses. Vellowish
grey (5¥6/2—+lay, lighter for sand), Base fot exposed.

SUPPLEMENTARY SECTION, COONARBINE FORMATION, EURTNILLA FORMATION, MILL-

YERA FORMATION
SECTION §

0.00-3.50 SAND, Red brown sund of modern dunes reworked from Coonarbine Formation.

COONARBINE FORMATION

1.00 SAND, Light brown. Numerous vertically oriented small cylindroids of soft white carbonate, of

soil profile. Emu shell, aboriginul artifacts and rare mature snail shells Gocur in uppermost level, Forme

longitudmal dunes.
— Disconformity —

EDRINILLA FORMATION 4

Unin2. 4.060 SAND, fine to medium prained, with subungular rough of pitted grains, poorly sorted.
Stratificatian absent. Grades to unit 1 over short distance. Light brown (SYR4/'7)-

Unitt. 4.00 SAND, medium grained, brown (SYR6/H), lighter coloured beds alternate near base,
Cross-hedded, scts 40 cm, lensing, gently curved coarse and fine laminae, sharp ereded wpper
contacts, assymptotic bottomsets. Laminae &S-l cm, by variation in clay content. Sets are
gently inclined toward Lake Millyers. Numerous charophyte oogonia.

Lightly cemented with clear or white finely crystalline carbonate. Pinkish irtegulay nadules,
weathering as brown lumps and slabs on surface. Carbonate gives white cast to This part of the
section, and causes slight benching, Partly cemented with massive gypsum in the basal layers,

— Disconformity —

MILLYERA FORMATION

Unit, 7,60-2,450 CLAY. Very hard, shiny irregular fractures, coated with black iron oxide and While
carbonate al top (soil horizon). Impregnated with vertically oriented gypsum masses. in 5S—
10 cm columns (fossif groundwater horizon) at top. No silt content. Colour 10¥6/2, Similar
fo Willawortina Formation. Upper contact sharp, flat. Grades down to light green soft clay tn-
lérbedded with very fine grained white sani) nch in charophyte oogonia Intertongues with
unit 4,

CAINGZOIC ROCK UNITS 165

Unit4. 4.00-5.00 SAND. Silt to very fine grained sand, with coarse lenses. Numérous thm ,5-5 cm
clay beds and lameline near top, which are crowded with algal tubules (charephytes}, Some
Tare massive charophyte crusts consisting entirely of strap-like algal forms with mutherous
large oogonia, Charophyte oogonia commion in upper sands. Clay pellet layers common. Sand
grains are subrounded to well-rounded smooth or frosted, with moderate sorting, Small scale
cross-laminated sets, 1D cm thick, with curved laminae. ;

Upper surface may be cemented with gypsum of a fossil groundwater horizon.

Unit 3, 0,00-0.93 SAND. Very fine grained, well sorted. Colour 5YR6/8. Impregnated with massive
Eypsuns and disc-shaped crystals of gypsum, Grades by alternation of 1-4 cm thick beds into
overlying unit, in which rt forms a lense, Contacts between lamellae are wavy, lenticular, and
Tippled in sore cases, Resembles Tirari Formation, Basal angular quartz granule layer, often
resting directly on underlying gypsum sediment,

Unit2, 025 LIMESTONE-GYPSUM. Greenish slightly sandy clay with 20 cm of interbedded thin
(0.5 coi) gypsum Jaminae at top, which grades laterally tnto laminated algal stem (tubules of
charophytes) limestone. The limestone and gypsum contain charophyte oogonia. The eypsum
contains scattered very coarse sand erains, and surfaces are assymetrically ripple-anarked, or
hare botryoidal “pulf” structure,

Unit |. 0,70 SAND. As for unlt 4. Orange and yellow stained, especially near base, greenish where un-
oxidized. Reworked distorted clay fragments from underlying units at base.

— Disconformity —
NAMBA FORMATION
3.28 SILTY CLAY, grey to black, tough. Grading down to grey, clayey, poorly sorted fine sand. Greasy
lustre on Jreguiar fracture surfaces. Gypsum patches and cracks at top infilled with overlying sand.

SUPPLEMENTARY SECTION, MILLYERA FORMATION

SECTION 6
EURINILLA FORMATION
At least 2.0 bright ced brown SANDS.

MILLYERA FORMATION

Unit?, 06.20 CALCAREOUS SANDSTONE. Very fine to medium grained moderately sorted sand,
30% carbonate, Grains pitted or frosted, subrounded to rounded. alternates with very fine
sand. Coarser sand contains charophyte tubules and rare oogonia, Some pink and black sand
erins, Fare carbonate grains, Weathered colour white (N10), unweathered greyish yellow
(5 ,

Elsewhere passes to hard platy limestone identical with 2. Impregnated with numerous white
fypsum cylindroids. Gradational contact with 6.

Unitf. 0.62 SAND. As for sand in 3 but uncemented, distinct contact with 5. Colour moderate red-
dish yellow (2¥7/4).

Unit5. 1.20 CLAYEY SAND. Maderntely sorted, with black and orange erains scattered through-
out, Jrregularly cemented into very hard massive nodules and sheets by fine grained white to
pink carbonate. Yellow and brown mottling common near base, white gypsum and carbonate
spots throughout. Yellowish grey (SY6/Z)-

Unit4+, 025 SAND, SILTY CLAY. Grades from clay to very fine sand, grains poorly rounded,
Colour yellawish (SY7/2) oxidized to moderate brown (SYR6/7) in patches.

Unit3, 0.5 LIMESTONE, CALCAREOUS CLAY. Varies Jaterally from burrowed soft calcareous
clay, with 30% silt to fine sand, to hard sandy white limestone, ‘The former bas 1-2 mm
diameter vertical burrows (insects?) and the latter has scattered charophyte oogonia and
shrinkage cracks. The base of the burrowed horizon is gradational, and lumps of the underly-

: ing unit are worked into it.

Unit2. 0.10-0.20 LIMESTONE. Laminated, platy, hard, metallic ring when struck. Constitutes
mimerous tubules of charophytes, and patches of “Coxiella", Contact with 4 not observed,

contact with 1 distinct, undulating.

Unit |. (0.50, On east side of channel, SAND. very fine grained, nodular white carbonete at lower con:
tact, Pale grey, Massive carbonate-cemented al top with shrinkage phenomena apparently fe-
tuted to drying. of carbonate. ;

On west side of channel. CLAYEY SAND. Moderate to well sorted, angular grains, Yellow-
ish grey (SY6/1) but speckled yellowish green, Grades up inte unit 3-

NAMBA FORMATION

0,10 Black tough clay, Sharp fiat upper contact.

EURINILLA FORMATION AND COONARBINE FORMATION TYPE SECTIONS; MILLYERA

FORMATION SUPPLEMENTARY SECTION

SECTION 7

Modern dune sands
— Disconformity —

— Disconformity —

COONARBINE FORMATION

Unit4. 6.70 SAND, very fine 1o medium grained, silly, Fine size dominant, well sorted. well
rounded, frosted. Light brown (SYR5/6), Al top is 20 cm of soft white carbonate, consisting
of 0.3 cm cylindroids and tubules (plant roots?) with }-2 cm lumps at the top, grading to
blotchy white carbonate as for unit 2.

186 R. A. CALLEN & R. H. TEDFORD

Unit3. 1,62 SAND, bimodal, medium-coarse atid very fine to fine. Bimedalily disappears downward,
grain size becomes finer, and sorting poorer, Some patches of white sand are present in the
essenually moderate vellowish orange (9YR3/6) coloured sequence. Top a4 moderate reddish
brawn (3YRS/6). Large scale cross-bedding is just visible. A well developed fossil carbonate-
rich 40) horizon marks the top, It is 50 cm thick and consists of moderately hard rather
irregular nodules and cylindroids, and gypsum cylindroids,

Uniz2 1.80. SAND, fine-grained ranging to coarse, grades down to CLAY-SILT, Moderately poor
sorting. No signs of stratification. Colour light brown to reddish yellow (6¥R5/6-3YRS/6)-
Weakly developed whitish carbonate patches ut top (soil horizon).

Unitl, 1.00 SAND. Bimodal: on medium-coarse grained aad very-fine grained boundaries. Dark red
brown (2YR4.5/6). Indistinctly horizontally laminated, upper contact sharp and flat, lower
contact obscure, apparently gradational,

— Disconformiry —

EURINILLA FORMATION .

Unit3 3,50 SAND, as above, Solour light brown (7YR6/4), Constitutes a single cross-bed set, Con-
tact with unit §, sharp, inclined, flat. Irregular gypsum as for unit £ at base, Upper 10-20 im-
pregnated with carbonate (1OYR7/3) of a fossil soil horizon.

Unie? 4.00 SAND, friable, fine to medium grained, bimodal. Course fraction well rounded,
dominant, colour light brown (6YR5.5/6). Constitutes a single cross-bed set, with low angle
cross-bedding. Contact with underlying unit sharp, undulating, cuts well down into unit 1,
Patches of very irregular tubules, nodules and cylindroids of gypsum occur at the top,

Unir}, O88 SAND, Medium @rained, sab-rounded to well-rounded grains with very fine grained
angular proportion (bimodal). Numerous coloured grains, opuques and biotite present. Silty
brown clay with gypsum forms pebble sized clasts, and clasts of underlying hmy sandstone
are present. Cross bed sets planar, 8-10 cm. Charophyte oogonia very common, and fragmen-
tal vertebrate bones present. White colour, Lower contoct erosional,

— Disconlormiry —

MILLYERA FORMATION .

1.1 LIMY SANDSTONE. Very fine to medium graincéd moderately sorted <lear-grained quartz sand
with 30-40% finely crystalline soft carbonale cement, Sand grains pitted of shiny, angular to sub-
tounded. Some grains of feldspar and ferruginous sandstone, flakes of hematite. Thin section shows
carbonate has recrystallized into radiating spherules, resembling some groundwater carbonates,

2.8 SAND, SILT, Silt to fine sand, 1-5% clay-carbonate matrix, forms strong cement by reason ot poor
eorting, of fremsgore Stains, Very poorly sorted with sub-rounded to very ungulur grains. Hard and

“white mottled,

SUPPLEMENTARY SECTION, EURINILLA FORMATION
SECTION &
Locality; CURNAMONA, Evrinille map sheets. Air photo ref: S. Aust. DepL Lands Svy, 161, run 3;
photo no. 4398. The section is situated on the west side of Tuke Pinpa, approximately 350 m north of the
only track crossing the Jake-
. . i Modern sand dunes
0,0-4,.0 SAND, Fine to medium grained, moderately sorted. SYR5/8, Strongly erosional base.

— Disconformity —
ICOGNARBINE FORMATION
0.30 SAND. Very fine to medium (averaging fine grained), poorly sorted, with sub-angular ta sub-
rounded polished or frosted grains (4¥RS/8). Erosional basal contact,
This unit may represent the Coanarbine Formation. It forms the basis of the longitudinal dunes.

— Disconformity — _
0.25 SAND. Very fine to fine, rather poorly sorted, clayey, Grains irregular, subangukar’, roleh, Colour
SYR4/7, Soft patchy carbonate well developed, With pipe like structure 5 cm diameter, This is a soil
horizon, and has a similar development &) those of the Furinilla Formation, The lithology is also simi-
lor hut there is a distinct contact at the base which appeared slightly crosional. The unit may be purt
ef ihe EURINILLA FORMATION.

EURINILI.A FORMATION

Unit3, 100 SAND. Clayey, very fine to fine, poorly sorted. Angular to sub-rounded frosted and
coyesely pitted sand grains, Colour 3YR5/8. Well developed secondary carbonate profites
constituting soft sandy pinkish white lime in lumps ard cylindroids 1-S cm across which
weather Gul. In lower part of sequence fractured himps 10-20 cm are common. The car-
honate profiles form numerous layers, concentrated toward the top of the unit, and represent
soil development (hence intermittent deposition is indicated}. }

Unit2, 5.33 SAND, Poorly sorted silty, slightly clayey fine grained, rounded (2.5YR4/8), Grades im-
perceptibly into overlying writ.

UNNAMED CONGLOMTRATE (?MILLYPRA FORMATION equivalent)

Unitl. O.8 Jnterhedde¢g CONGLOMERATE, SAND and CLAY. Consists of basal sand loosely
cemented with calcitim carbonate, cross-bedded on medium scule. Micaceous, At the base of
this sand bed ore granule size quan? ¢{prey, clear, dark grey, yellow), pebble size clasis of
ecey clay and subrounded calcareous orange-brown clay. Also large grains of brown perthite
feldspar! Maximtum grain size is 15 x 2 cm,

— Disconformity? —

CAINOZOIC ROCK UNITS 167

Overlying this is an uncemented medium to coarse sand layer, which is capped by coarsely
crystalline gypsum nodules and plates, weathering to a powdery crust.

The gypsum is followed by coarse grained to medium grained moderately sorted sand as
helow, with scattered very coarse grains. Grains are dull and subrounded to rounded.

The uppermost calcareous sand (5YR&/4) alternates with thin SILTY CLAY (10YR6/2)
with irregular patches of 5YR6/2 coarse grained very poorly sorted clayey sand. All con-
tact are distinct and flat. Apparently grades into overlying unit over short distance and by
intertonguing—no disconformity observed.

—Disconformity —

NAMBA FORMATION

Unit 8. 0.40 SAND. Very fine grained. Yellowish white.

Unit7. 3.85 CLAY. Puggy, soft, with powdery gypsum stringers at top. N5.

Unit 6. 0.85 CLAYEY SILT. Colour 5¥Y6/1 with 10Y6/2 patches.

UnitS. 2,23 CLAY. Hard, greasy, Colour SY6/1 (silty) to S¥4/1 becoming 5Y6/0.5 at base, Contact
with 4 gradational,

Unit4. 0.20 SILTY CLAY. Finely laminated, dark and bright orange brown siliceous limonite
nodules, dense masses of manganése oxide. Bright yellow green patches in clay. Basal contact
sharp, undulating, indicates an hiatus, Rare vertebrates.

Unit 3. 0.22 SAY, Dark olive (5¥4/1) greasy. Sharp basa] contact. Numerous vertebrates as for
unit 2,

Unit2. 0.12 DOLOMITIC CLAYSTONE. Pale green (SY8/2) with black patches. Subconchoidal
fracture, hard. Sharp contact with overlying units. Numerous vertebrates include lungfish,
crocodiles, turtles and marsupials.

Unitl. 0,18 CLAY. Light green.

In addition to this sequence, an erosional remnant of Coonarbine Formation is superposed on the top. of

the Namba Formation-Eurinilla Formation disconformity. The description is:

118 SAND. Yellow brown (7YR6/7) with prominent columnar jointing (15 cm rectangules) typical of

Coonarbine Formation. Fragments of Eurinifla Formation carbonate nodules occur at the base.

Also along the lake shore is reworked material from all the above units, forming outwash and aeolian

mounds of sub-Recent origin, This material is slightly older than the red brown dunes.

SUPPLEMENTARY SECTION, EURINILLA FORMATION
SECTION 9

Location: FROME, Coonarbine map sheets. Air photo ref.: S. Aust. Dept. Lands Svy. 395, Run 4,
Photo no. 9597. The section is situated on a cliff on the western edge of a small claypan on Eurinilla

Creek.

RECENT SAND DUNES
pees SAND. Very fine to fine grained, well sorted, with subrounded stains. Large scale dune cross-
bedding,

COONARBINE FORMATION

17-3.5 SAND. Clayey, up to medium grained, mostly fine grained, grains subrounded, Moderately
sorted. Weak thin horizontal bedding at base. Spotted with white carbonate patches, tending to 0.5 cm
diameter cylindroids near base {weakly developed soil profile), Colour 6¥YRS/7.

— Disconformity —
EURINILILA FORMATION
3.05 CLAYEY SILT. No sedimentary structure. Oxidized dark orange brown (SYR4/7)—original
colour (new patches) pale orange (10YR8/2). Gypsum beds several centimetres thick occur in lower
part, as for cap of 95 cm of massive disc shaped (0,5 cm) gypsum rosettes in red brown silt and white
gypsum flour.

NAMBA FORMATION ;
1.8 CLAYEY SILT. Very soft, 5¥5/0.5 with SYR4.5/4 patches.

— Disconfermity —

— Disconformity —

A NEW GENUS OF LATE PRECAMBRIAN POLYCHAETE WORMS FROM
SOUTH AUSTRALIA

BY M. F. GLAESSNER*

Summary

GLAESSNER, M. F. (1976).-A new genus of Late Precambrian polychaete worms from
South Australia. Trans. R. Soc. S. Aust. 100(3), 169-170, 31 August 1976.

New material indicates differences between Spriggina floundersi Glaessner and S.? ovata Glaessner
& Wade which are comparable with those between genera of living polychaete annelids.
Accordingly, a new genus Marywadea is proposed for ovata. The evolutionary significance of the
Sprigginidae is discussed briefly.

A NEW GENUS OF LATE PRECAMBRIAN POLYCHAETE WORMS FROM
SOUTH AUSTRALIA

by M. F. GLAESSNER*®

Summary
GLAgssNER, M, F_ (1976).—A new genus of Late Precambrian polychaete worms from South
Australia. Trans. R. Sac. S. Aust. 100(3), 169-170, 31 August 1976.
New material indicates differences between Spriggina floundersi Glaessner and S.? ovata
Gluessner & Wade which are comparable with those between genera of living polychaete
annelids. Accordingly, a new genus Marywadeu is proposed for ovata. The evolutionary signifi-

cance of the Sprigginidae is discussed briefly.

Introduction

The representatives of the Family
Sprigeinidae Glaessner (1958) are among the
most remarkable elements of the Ediacara
fauna from the Pound Quartzite of South Aus-
tralia. The Late Precambrian age, stratigraphic
posilion and geographic distribution of this
rock unit and its fauna need no further discus-
sion (Wade 1970; Glaessner 1971, 1972). The
arthropod-like appearance of Sprigeine is
attracting increasing attention (Cisne 1975, p.
61; Stanley 1976, p. 58). The reconstruction
of an ancestral crustacean by Hessler & New-
man (1975) shows startling resemblances with
Spriggina. Notwithstanding these, no canyinc-
ing evidence has been discovered which would
justify the transfer of the Sprigginidae from
Annelida to Arthropoda or prove a transitional
position of this family between two phyla. New
discoveries have, however, clarified and
emphasized the differences between the type
species §. foundersi and the species described
S.? ovata described by Glaessner & Wade
(1966). Its diagnosti¢ characters have at Icast
the same significance as those distinguishing
genera of living Polychaeta and for this reason
the following new genus Is proposed. It diflers
from Spriggina in all characters listed in the
diagnosis.

Taxonomy

Genus Marywadea nov.

fype species; Sprigginel avata Glacssner &
Wade 1966.

Fig. 1. Marvwadea ovata (Glaessner & Wade),
Latex mould of specimen from the Late
Precambrian Pound Quartzite of Ediacara,
§. Aust. x 2. (Outlines and surface slightly
distorted during fossilization; anterior
margin of the head pushed back causing
truncation of the outline and wrinkling of
the surface; some ventral structures may
be obscurely visible. Note that all other
specimens have smoothly curved anterior
outline and smooth surtace,)

* Centre for Precambrian
S, Aust. S006,

Research, Department of Geology, University of Adelaide. Adelaide.

170

Diagnosis: Prostomium half-moon-shaped, not
wider than the body with its appendages,
Integument thin, wrinkled and possibly show-
ing some underlying structures when com-
pressed, Body consisting of up to 5() short,
broad segments, occasionally with impressions
of bundles of long, curved setae. A pair of oval
impressions behind the prostomium suggests
the presence of teeth. The posterior end of the
body is broadly rounded.

Derivation of generic name: After Dr Mary
Wade who earlier expressed the view that
evata may be generically distinct from
floundersi; this has now been confirmed by
new finds.

Localities: Ediacara Hills, Brachina Gorge,
Bunyeroo Gorge, Mayo Gorge.

Number of specimens of M. ovata: 16.

Remarks
The Sprigginidae are not arthropods as the
head did not consist of the appropriate number
of appendage-bearing segments and the trunk
appendages are not distinctly jointed and end
in acicular setae. The mouth was. probably not

M. F. GLAESSNER

directed posteriorly, there was no labrum and
there is no evidence of antennae or a caudal
furca, On the other hand the head was con-
spicuous and relatively larger than in any
known annelid and its integument was more
strongly sclerotized in Spriggina (apparently
less so in Marywadea). There is evidence of a
simple pharynx in Spriggina and of two simple
teeth in Marywadea, suggesting relations to
Phyllodocemorpha; otherwise the Sprigginidae
are unhke living Annelida. Some evolutionary
advance in the direction of a primitive arthro-
pod is indicated, particularly in cephalization,
Ii may be parallel to the unknown evolutionary
lineage which had produced the two primitive
arthropods known from the Ediacara fauna
(Praecambridium and Parvancorina).

Acknowledgments

The specimen illustrated here was found by
Mr D. Westlake in August 1975 at Ediacara.
I am grateful to Mr J. Gehling, Murray Park
College of Advanced Education, who presented
to me casts and moulds of this and other
specimens.

References

Cisne, J. L. (1975) —Anatomy of Triarthrus and
the relationships of the trilobita. Fossils and
Strata (4), 45-63.

GLAESSNER, M. F. (1958) —New fossils from the
base of the Cambrian in South Australia.
Trans. R. Soe. 8, Aust. 81, 185-188.

Guarssner, M. F. (1971).—Geographie distribu-
tion and time range of the Ediacara Precam-
area fauna. Bull. venl. Soc. Amer. 82, 509-

a.

GLARSSNER, M. F. (1972).—Precambrian palaéo-
zoviogy. Univ, Adelaide Cent. Precambrian
Res,, Spec. Pap, 1, 43-52.

Guarssnpe, M. F, & Wave, M. (1966).—The late
Precambrian fosils from Ediacara, South Aus-
tralia. Palaeontalogy 9, 594-628, pls 97-103.

Hessen, R. R. & Newnan, W. A. (1975) —A
trilobitomorph origin for the Crustacea. Fos-
sils aiid Strata (4), 437-459.

Staniey, 5. M. (1976).—Fossil data and the Pre-
eambrian-Cambrian evolutionary transition.
Am. J. Sci. 276, 56-76,

Wapr, M. (1970).—The stratigraphic distribution
of the Ediacara Fauna in Australia. Trans. R.
Soc. §. Aust. 94, 87-104,

VOL. 100, PART 4 30 NOVEMBER, 1976

TRANSACTIONS OF THE

ROYAL SOCIETY
OF SOUTH AUSTRALIA

INCORPORATED

CONTENTS

Robinson, A. C. and Smyth, M. E. B. The Vertebrate tee of a Vase
pelago, South Australia - - - - - 171

Shepherd, S. A. and Womersley, H. B. S. The Subtidal Algal and sesapank Ecology
of St Francis Island, South Australia —- - 177

Anstis, Marion Breeding Biology and Larval Development of Litoria herees
(Anura: Hylidae) - - - - - 193

Rowe, F. W. E. Restriction of the chiridotid genus Trochodota Ludwig (1891)
(Holothurioidea: Apodida), with the description of a new species

from South Australia - - - - - - - - 203
Annual Report of Council —- - - - - - - - - - 207
Award of the Sir Joseph Verco Medal _ - - - - - - - - 208
Balance Sheet - - - - - - - - - - - - 209
List of Fellows a i ey. (0

PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS
STATE LIBRARY BUILDING
NORTH TERRACE, ADELAIDE, S.A. 5000

THE VERTEBRATE FAUNA OF NUYTS ARCHIPELAGO,
SOUTH AUSTRALIA

BY A. C. ROBINSON* AND M. E. B. SMYTH

Summary

ROBINSON, A. C. & SMYTH, M. E. B. (1976).-The Vertebrate Fauna of Nuyts Archipelago,
South Australia. Trans. R. Soc. S. Aust. 100(4), 171-176, 30 November, 1976.

The St Francis group of islands in Nuyts Archipelago was visited by a joint Royal Society of
South Australia and Fisheries Department of South Australia expedition in January, 1971.
Seven species of mammals, twenty-seven species of birds and sixteen species of reptiles are
recorded, together with comments on their habitat and abundance. The potential of the islands for
fauna conservation is briefly considered.

THE VERTEBRATE FAUNA OF NUYTS ARCHIPELAGO,
SOUTH AUSTRALIA

by A, C. Ropinson* and M, EB. B. Smytut

Summary
Ropinson, A. C, & Smyts, M. E. B. (1976).—The Vertebrate Fauna of Navis Archipelago,
South Australia, Trans. R, Soc. S, Aust. 100(4), 171-176, 30 November, 1976.
The St. Francis group of islands. in Nuyts Archipelago was visited by a joint Royal Society
of South Australia and Fisheries Department of South Australia expedition in January, 1971.
Seven species of mammals, twenty-seven species of birds and sixteen species of reptiles are
rycotded, together with comments on their habitat and abundance. The potential of the islands

for fauna conservation is briefly considered,

Introduction

The study of the fauna of islands can pro-
vide useful insights into. the biogeography of
the fauna on the adjacent mainland. Many
islands along the southern coast of Australia
were connected to the mainland whet: sea level
fell during the Pleistocene glaciations, and
samples of the coastal flora and fauna were
preserved on these islands as the sea level rose
during the interglacial periods (Main 1961).
With the accumulation of information on the
flora and fauna of these islands and the adja-
cent mainland, including the palaeofaunas and
floras, it may eventually be possible to recon-
struct the biological history of southern Aus-
tralia from the Pleistocene to the present. Fur-
ther, as information is assembled on the present
habitat preferences and tolerances of mainland
species, data on island faunas may assist the
construction of a palacochmati¢c history for
southern Australia. Another important aspect
of islands is their suitability for conservation,
Many of the islands around southern Australia
preserve relict populations of species that are
fare or extinct on the mainland. Studies such as
this should therefore contribute to the manage-
ment of these islands as fauna sanctuaries
muintaining these important populations,

This paper discusses the species of mammals,
birds and reptiles recorded from the St Francis
Island group of Nuyts Archipelago during the.
jeint Reyal Society of South Australia and

Fisheries Department of South Australia expe-
dition there in January, 1971, together with
some additional information gathered by sub-
sequent visitors to the islands.

The expedition visited four islands: St
Francis, Masillan, Fenelon and Dog. Mammal
trapping was carried our on St Francis [. (4
nights), Dog I. (1 night) and Masillon I. (1
night), Since then trapping hag been carried
out on Egg J. (D. Murray, pers. comm.), Data
were obtained from animals caught in traps or
observed by spotlight, from collection of bones
and from signs of mammal activity. Sherman
and Wire cage traps were used in trap lines for
a total of 153 trap nights (St Francis I. 105,
Dog I. 19, Masilion I. 19, Egg I. 10). Trap-
lines and spotlight surveys covered all vegeta-
tion associations on the islands and gave a
wide coverage of the areas. Specimens have
been lodged in the South Australian Museum
(SAM), registration numbers are cited.

The bird list is based on the observations of
all members of the expedition and compiled
by Mr P. Macrow. The records generally rep-
resent sight records, but where doubt existed as
to the identity of a species, a specimen was
shot for a positive identification,

The reptile list is primarily from collections
made on St Francis I. Only part of a day was
spent looking for reptiles on Dog, Masillon and
Fenelon Is.

* National Parks & Wildlife Service, Box 1782. GP.O., Adelaide, $. Aus. 5001,
+ Deceased; formerly of Department of Zoology, University of Adelaide, Adelaide, S. Aust, S000,

L72

MAMMALS

Mammals previously recorded [rom St
Francis 1. include the bandicoot Iscaden &be-
sulur naulicus und a ral kangaroo, presumed to
be a species of Betrongia, which had become
extinct by the 19203 (Wood-Iones 1924;
VYerco 1935), Three terrestrial and two marine
species are now added, and these are marked
by an asterisk in the Jist which follows. In addi-
tion, skulls collected on the island enabled
identification of the species af Betiongia.

Family PERAMELIDAE

Isoodon = obesulus (Shaw), SAM, M&8546-
M8549. Shori-nesed bandicoot. St Francis I.

A common animal preferring the grassy
areas on the higher parts of the island but also
occuring m the saltbush steppe association
covering the remainder of the island. These
bandicouts have survived the introduction of
cats and the conversion of a large part of the
island to grassland by cultivation. /. obesulus
remains common in south-eastern and south-
western Australia and there ure populations on
Kangaroo lL (Andrewartha & Barker 1969)
and Franklin 1. (Watts 1974).

Family MACROPODIDAER
Bettongia penicillata (Gray), SAM, M8353.
Brush-tailed betteng. St Francis 1.

Fragments of skulls were found in the suned-
hills behind Petrel Cove, but living animals
were not observed. Bettongs were reported to
be very common when St Francis Island was
first settled. WoodJories (1924) and Yerco
(1935) reported that the settlers introduced
cats to the isand to exterminate the bettongs
which were causing damage io vegetable gar-
dens. It seems likely thal alteration of habitat
may have also played @ part in their decline,
as this species nests in dense cover such as that
formerly provided by the sclerophyll shrub
community on the higher parts of the island.
The settlers completely destroyed this habitat
through clearance for wheat growing. 8. peni-
cillata was formerly widespread throughout the
southern half of Australia hut now appears to
he confined to southwest Westet'n Australia.

*Macropus cugenii (Desmarest). SAM, M&575.
Yaimmar wallaby, St Francis 1,

A single tooth row of this species was found
in the sandhills behind Petrel Cove, No living
animals were found, The settlers did not report
tammars on the island and it is possible that
they were either alreauly extinct, that the tndj-

A. C. ROBINSON & M. F_ B, SMY'LH

vidual collected wus introduced at some time,
or that it was left by sealers known to have
collected large numbers of wallabies on other
wlands (N. Wace pers. comm.). They were
formerly widespread on the south and south.
west Australian mainland and populations were
recorded from Kangaroo 1, Flinders 1. St
Peters [., and a number of Western Australian
islands, Today, in South Australia, tammuars
Temaiit common only an Kangaroo I. and on
Greenly 1. where they were introduced (Mit-
chell & Behrndt 1949); the St Peters 1. popula-
tion is extinct, the Flinders I, one is almost
eXtinet and the mainland population is reduced
to a remnant on Evre Peninsula (fF, Aitken
pers. comm-),

Family MURIDAE
*Rattus fuscipes (Waterhouse). SAM, M8541-
MBS45, M8598-M8600. Bush rat. Masillon and
Dog fs.

This species appeared toa be common on
these islands, where it nested in limestone
érevices and, possibly, mutton bird burrows. k
seems unlikely that the smaller islands of Nuyts
Archipelago cver supported mammals larger
than A. fesxcipes. Its absence from Ege Island
suggests that this island may be too small to
support @ population of R. fuscipes. We did not
sev or collect this rat on St Francis |, where
there wre extensive areax of suitable habitat,
hut again they may have been exterminated by
cats.

*Rattos rattus (Linn.), SAM, M8551. Black raz.
St Francis [

Two lower jaws of this mntroduced species
were collected in the sandhills behind Petrel
Cove, Na living animals were caught. A. ratty
was undoubtedly introduced by the early
settlers to St Francis I. and has since hécome
extinct.

Family OTARIIDAE

*Neophoca cinerea (Peron & Lesucus). Austra-
lien seu lion, Fenelon I.

This species visits all the islands and there
is it breeding colony on the beach of Fenelon
Island. In January 1971 this colowy numbered
approximately 50 jndividuals, including a num-
ber of pups. D. Murray (pers. comm.) pro-
vided the following estiraates of the size of this
colony in February 1973: mature bulls 7, pups
8, cows and immature bulls 36, The number
of individuals in the vicinily of the beach was
56-58

VERTEBRATES OF NUYTS ARCHIPELAGO

A South Australian National Parks and
Wildlife Service expedition in June 1975 was
table to land ot Fenelon T., but a count from
the boat showed mature bulls 4, pups 5, cows
and immature bulls. 3. They alsa noted a pos-
sible breeding colony of this species on Dog T.
Numbers recorded for this colony were bulls 3,
cows and immature bulls 10. In addition 20
sea lions were seen on Freeling, I.

*Arctocephalus forsteri (Lesson), New Zealand
fury seal. Fenelon 1.

The South Australian National Parks and
Wildlife Service expedition in June 1975 noted
40 fur seals on Fenelon 1. No evidence of
breeding was observed, They also noted 5 fur
seals on Freeling I.

BIRDS
No systematic list of the birds of the St
Francts Group has been compiled. The follow-
ing list contains comments on habitat and
abundance of birds observed during the 1971
expedition.

Family SPHENISCIDAE
Eudyptola minor (Stephens), Little penguin, St
Francis 1.

Common around the shores of Petrel Bay:
most of fhe birds were in a heavy moult.

Family PROCELLARIIDAR

Macronectes giganteus (Gmelin), Giant petret.
Dog I.
Beach washed specimen.

Puffinus fenuirostris (Temminck). Short-railed
shearwater. All islands visited except Fenelon.
Nesting burrows were found wherever suffi-
cient soil depth allowed excavation, Approxi-
mately one-third of St Francis I. was covered
by the burrows. Dunng the day most burrows
contained one adult bird and one egg in an
advanced stage of incubation. At approximately
20.00 hours each evening vast numbers of birds
returned to the island from feeding al sea.

Family OCEANTIDAE
Pelagodroma marina (Latham). White-faced
storm petrel, Dog and Fenelon Is.

Dried remains and wings were found, Small
burrows. on Fenelon JI. may belong to. this
species.

Family PHALACROCORACIDAE

Phakacrovorux varius (Gmelin). Pied
tnorant. St Francis I,

cor-

173

A small number of birds were fishing in
Petrel Bay and roosting in company with
black-faced cormorants at the eastern end of
thie bay,

Phalacrocorax fuscescens (Vieillot). Black-facea
cormorant, St Francis 1.

Approximately 20 birds roasted on the
eastern headland of Petrel Bay.

Family ANATIDAE

Cercopsis novachollandiae (Latham).
Barren goose. St Francis 1

Approximately 50 geese were observed and
flocks of 3 to 20 were seen feeding on the
eastern end of the island near the lighthouse.
The geese congregated around three small fresh
water soaks above granite boulders on the
eastern end of the island, Goose droppings
were ulsa found on Masillon I.

Anas sp. Unidentified teal. Egg I.
One bird was seen at sea near this island,

Family ACCIPITRIDAE
Haliaetus leucogaster (Gmelin). White-breasted
sed eagle. St Francis and Masillon Is.
Several adults and one immature bird were
observed fiying over St Francis I. and three
adult birds were seen fying over Masillon IL

Family PANDIONIDAE
Pandion haliaetus (Linn.). Osprey. St Francis 1.
One or two birds were observed on most
days at the eastern end of the island near the
lighthouse, Old nests of sea eagles or ospreys
were found on the eastern side of Dog I. and
the southern side of St Francis I.

Family FALCONIDAE
Falco peregrinus (Tumstall). Peregrine falcon.
Masillon I.
Only a single bird was observed,

Falco cenchroides (Vigors & Horsfield). Nan-
keen kestrel, St Francis, Dog and Masillon Is,

Several pairs on St Francis L

Family PHASIANIDAE

Coturnix pectoralis (Gould). Stvbdle quail, St
Francis I.

Several birds were flushed in the grassy area
near the lighthouse-

Family RALLIDAE

Rallus philippensis (Linn.}. Bonded landrati, St
Francis and Dog Is.

This species appeared common om St Francis
I. and two specimens were cailected. Only two
birds were sighted on Dog [,

Cape

174

Family HAEMATOPODIDAR

Haematopus fuliginosus (Gould). Seory oyster-
catcher. All islands visited.
A common bird of the rocky shorelines.

Family CHARADRIDAE

Vanellus miles wovaehullandiae (Stephens),
Spur-winged plover. St Francis 1.

Eight to ten birds were observed feeding
around the shores of Petrel Bay,

Charadrius cubricollis (Gmelin). Hooded dot-
tere]. St Francis I-

From two to ten birds were seen on the
beach in Petre! Bay each day.

Charadrius alexandrinus (Linw,), Red-capped
dotterel. St Francis T.

One bird was observed on the beach in
Petrel Bay in company with four Red-tecked
stints,

Family SCOLOPACIDAE
Calidris ruficottis (Pallas). Red-necked stint. St
Francis I,
Four to eight birds on the beach in Petrel
Bay, Onc specimen collected.

Family LARTIDAE
Larus noyuehollandiae (Stephens). Silver guil.
St. Francis f.
Nine to ten birds on the beach in Petrel Bay,

Larus Pacificus {Latham). Pucifie guif. All
islands.

A common bird of these islands. Adults and
immature birds wete present in about even
numbers. Approximately 20 birds furaged
along the shores of Petrel Bay,

Hydroprogne tschegrava (Lepechin), Caspine
tern, St Francis and Dog Is.
Two birds noted on each island.

Family PSITTACIDAE
Neophema petrophila (Gould). Rock parrot,
Found on all islands visited,

Numerous small flocks were flushed while
walking on St Francis J.

Famity HIRUNDINIDAE
Hirundo tahitica neoxena (Gould). Welcome
‘wallow. Found on all islands visited.
A common bird. Old nests. in houses, light-
house shed and caves,

Family MOTACILLIDAE
Anthus noyaeseelandiae (Ginelin). Pipit, All
islands visited,
Abundant.

A. C. ROBINSON & M. E. B. SMYTH

Faniily MELIPHAGIDAE
Meliphaga virescens (Vicillot). Singing honev-
eater. All islands visited.

A very common bird. Eight to ten birds were
present in the camp area at all times.

Family CORVIDAE

Coryus coronoides (Vigors & Horsfield), Aus-
tralian raven, St Francis L.

Common; » flock of approximately 30 birds
was observed as the expedition landed. Small
flocks were seen daily, foraging amongst the
mutton bird burrows. The lighthouse tower was
a favoured roost, and nest remains were found
here, Other nests were found on low. bushes.

REPTILES

Eleven species of reptiles are listed or men-
tioned for Nuyts Archipelago by Proctor
(1923), Waite (1923) and Worrell (1963),
Our expedition added another five species;
these are Indicated by an asterisk in the list
below. The islands of the group from which
each species is now known ig alse recorded.
Some contrasts between the abundance af
several species on St Francis and other off:
shore islands in the Bight are noted in the list
below. Possibly the drier climate of the Nuyts
Group is responsible,

Fumily GEKKONIDAE
“Underwoodisaurus mili (Bary], SAM,
R12858, R12863, R12870, RL2876, R12889.
St Francis, Masillon, Fenelon, Dog Is.

Common under limestone boulders er in
burrows in the sand by day.

‘Heteronotia binvei (Gray). SAM, 12878. St
Francis I,
Common under stones by day.

*Phyllodactylus = marmoraiux (Gray), SAM,
R12865, R12877. St Francis, Fenclon 1s.

Surprisingly uncommon, tor this is a very
abundant species on some other off-shore
islands. Found only under avolianite slabs on
exposed coastal arcas.

Family PYGOPODIDAE
Lialis burtonis (Gray), SAM, R12896. St
Francis L.
Only one seen during the visit,

Aprasia striolata (Lutken). Recorded fur St
Francis 1, by Kluge (1974); this is the speeci-
men referred by Proctor (1923) to Delia
fraseri, No Aprasia was collected in 197L.

VERTEBRATES OF NUYTS ARCHIPELAGO

Family AGAMIDAE
Amphibofurus fionni (Proctor). SAM, R12874,
Sc. Francis J,

Found only among the exposed granite
uraund the cdges of the island. Wooc-Jones did
not find it {Proctor 1923), but Worrell (1963)
has recorded it from St Francis f.

Family SCINCINAE
Hemiergis peronii (Fitzinger). SAM, R12862.
St Francis, Dog and Fenelon Is.
Surprisingly infvequently seen; like P. mear-
moratus, this is a common species on other off-
shore islands.

Egernia multiscutata (Mitchell & Behrndt).
SAM, R12857, RI2861, R12873, R128RS. Se
Francis, Dog, Masillon and Fenelon Is,

A very common species on the sandier parts,
burrowing under rocks or bushes and using
the mutton-bird burrows for quick retreats,

Lerista frosti (Zietz). SAM, R12859. Masillon
1,

Another species usually common on off-
shore islands byt very scarce in the Nuyts
group.

*Lerista sp. (near picturata). SAM, R12880. St
Francis I.

A large member of the genus, with forelimbs
reduced to dimples and two toes on each hind
limb. Commonly found buried in sand under
stones.

“Menetia greyii (Gray). SAM, R12875, St
Francis I.

Rarely scen, probably because it js small,
quick and secretive,

Morethia obscura (Storr). St Francis t.
Again, rarely seen and yery difficult to catch,
ho specimens collected.

Tiliqua branchiale (Gunther). SAM, R12864,
RJ2879, St Francis, Fenclon Is,
Common in litter and around the buildings.

Family BOIDAE

Morelia spilotes. variegata (Gray). Carpet snake.
St Fraticis I.

178

Commonly seen in the morning and lata
afternoon, no specimens collected

Family ELAPIDAE
Drysdalia coronoides (Gunther), SAM, R12860,
Ri2881. White-lipped snake, St Francis, Masil-
lon and Fenelon Is,
Not frequently seen, but probably quite com-
mon,

Notechis ater (Krefft), SAM, RI2895. Tiger
snake. St Francis I.

Usually common on miutton-bird. istands, yet
only a few were seen even at night.

Discussion

The vertebrate fauna of the St Francis [.
group of Nuyls Archipelago is quite diverse
and remains Telatively undisturbed and rela-
tively free from introductions. The fauna of
most South Australian islands and indeed of
the mainland adjacent, is still incompletely
known,. and so biogeographical interpretations
are difficult at this stage. However, some com-
ments may be made on the importance of thesc
islands im conservation.

Island faunas are extremely vulnerable to
man’s interference, and the fate of Bettongia
penicillate. on St Francis J. illustrates this point
Tf further work there definiicly establishes that
this species rs extinct, the opportunity exists to
re-establish dense vegetation On the island and
introduce B. penicillate from Western Aus-
tralia. This should succeed as cats are no Jonger
present on the island. It is obviously a long-
lerm project but it merits consideration because
the island population could ultimately serve as
a reservoir of animals for release in suitable
areas of their former mainland range, In addi-
tion, it Rattus fuscipes is absent from St
Francis 1. it could be re-introduced from
neighbouring islands,

The majority of the birds observed are com-
mon on all islands in the area and on the
adjacent coast, but a notable exception ts the
Cape Barren goose, The total world population
at this endemic Australian species, although it
is increasing, is still dangerously low, and so
every attempt should be made to consetve the
known populations from disturbance.

References

ANDREWARTHA, H.G, & Barxer,, S, €1969).—In-
troduction to the atudy of the ecology of the
kangarca island wallaby Protemnodon engenit
(Desmurest) within Plinders Chase, Kan-
guroo Island, South Australis. Trans R- Soe.
S. Avst. 93, 127-132.

Kuuce, A, G, (1974).—A taxonomic revision of
the lizard family Pygopodidae. Misc. Publ.
Mus, Zool, Univ, Mich, 147, 1-221,

Main, A. R. (1961),—The occurrenee of Macro-
podidie on islands and its climatic and
ecologics|l implications.. J. Prac, R, Soc. W,
Aust. 44, 84-89,

176

MITCHELL, F. J. & BEHRNDT, A. C, (1949).—
Fauna and flora of the Greenly Islands, Part
I, Introductory narrative and vertebrate
fauna. Rec. S. Aust. Mus. 9, 167-179.

Proctor, J. B. (1923).—The flora and fauna of
Nuyts Archipelago and the Investigator
Group. No. 5—The lizards. Trans. R. Soc. S.
Aust. 47, 79-81,

Verco, J. C. (1935).—Jn B. C. Cotton (Ed.),
“Combing the Southern Seas”, chapter 8.
(Rigby: Adelaide).

Waite, E. R. (1923).—The flora and fauna of
Nuyts Archipelago and the Investigator

A. C. ROBINSON & M. E. B. SMYTH

Group. No. 10—The snakes of St Francis
Island. Together with a note on the name of
the geographical group. Trans. R. Soc. S.
Aust. 47, 127-128.

Watts, C. H. 8S. (1974).—The Nuyts Island bandi-
coot (Jsoodon obesulus nauticus). S. Aust.
Nat. 49, 20-24,

Woop-Jones, F. (1924).—“The mammals of
South Australia. Part II: The bandicoots and
the herbivorous marsupials.” (Govt Printer:
Adelaide).

Worreti_, E. (1963).—*‘Reptiles of Australia”.
(Angus & Robertson: Sydney).

THE SUBTIDAL ALGAL AND SEAGRASS ECOLOGY OF
ST FRANCIS ISLAND, SOUTH AUSTRALIA

BY S. A. SHEPHERD* AND H. B. S. WOMERSLEY}

Summary

SHEPHERD, S. A. & WOMERSLEY, H. B. S. (1976) .-The subtidal algal and seagrass ecology of
St. Francis Island, South Australia. Trans. R. Soc. S$. Aust. 100(4), 177-191, 30 November, 1976.

A subtidal survey of selected sites in the Isles of St Francis off the west coast of Eyre Peninsula,
South Australia, shows that upper and mid sublittoral zones similar to those of Pearson I. and West
I. occur. The upper sublittoral on rocky coasts is dominated by species of Corallina and Jania, with
Cystophora intermedia present near low tide level or sometimes as deep at 3 m. The mid sublittoral
is characterised by larger brown algae (Ecklonia radiata, Scytothalia dorycarpa, and species of
Cystophora and Sargassum), often with an understorey of red algae. The lower sublittoral zone
occurred between 47 and 57 m deep on the transect subject to greatest water movement, and is
characterised by red algae together with bryozoa, sponges and hydroids.

In the sheltered Petrel Bay- communities of the seagrasses Amphibolis antarctica and Posidonia
occur.

An algal species list is appended.

THE SUBTIDAL ALGAL AND SEAGRASS ECOLOGY OF
ST FRANCIS ISLAND, SOUTH AUSTRALIA

by S. A. SHEPHERD* and H. B. S. WoMERSLEY+

Summary

SeePxerD, S. A, & Womers.ey, H. B. 8. (1976).—The subtidal algal and seagrass ecology of
St. Francis Island, South Australia. Trans. R. Soc. S. Avst. 100(4), 177-191, 30 November,
1976.

A subtidal survey of selected sites in the Isles of St Francis off the west coast of Eyre
Peninsula, South Australia, shows lhat upper and mid sublittoral zones similar to those of
Pearson I, and West I. occur. The upper sublittoral on rocky coasts is dominated by species
af Corallina and Jania, with Cystophora. intermedia present near low. tide level or sometimes
as deep at 3 m. The mid sublittoral is characterised by larger brown ulgae (Ecklonia radiata,
Scytothalia dorycarpa, aud species of Cystophara and Sargassunt), often with an understorey
of red algae. The lower sublittoral zone occurred between 47 and 57 m deep on the transect
subject to greatest water movement, and is characterised by red algae together with bryozoa,

sponges and hydroids.

In the sheltered Petrel Bay, communities of the seagrasses Amphiholis antarctica and

Posidonia occur,
An algal species Jist is appended.

Introduction
‘The marine flora of the Great Australian
Bight is little known, Apart from the intertidal
tegion, which has been briefly discussed by
Womersley & Edmonds (1958), the subtidal

region is known only from various drift col-.

lections and the ecological account of Shepherd
& Womersley (1971) of Pearson Island,
towards the eastern limit of the Bight.

Ao expedition to the isles of St Francis,
lying off Ceduna, supported by the then
Fisheries and Fauna Conservation Department
and the Royal Sweiety of South Australia,
Visited the islands from 4-11 January, 1971.
This provided the opportunity for a brief sur-
vey of the subtidal ccology of selected sites
subject to varying degrees of water movement.
Although limited in time, these studies provide
the first such information from the northern
part of the Great Australian Bight.

The isles of St Francis comprise nine small
islands, the largest of which, St Francis. I.
(Fig. 1) is about 4 km across and lies at
32°9U'S, 133°18'E, about 56 km from the
mainland, The islands are granitic, rising

steeply from the sea-floor, and subtidally the
topography consists of massive blocks and
sheets of rock. In sheltered areas (e.g. Petrel
Bay), the sandy sea-floor slopes more gently
into deeper water.

The short stay on the island prevented a
detuiled survey, but three survey sites were
chosen on St Francis I., subject to different
degrees. of water movement. One site was
chosen on nearby Masillen L, and collections
were also made at Egg I.

The field work was limited to the subtidal
region, but brief observations of the intertidal
region indicated that the organisms and zona-
tion present were similar to those described by
Womersley & Edmonds (1958) for such grani-
lic steeply sloping coasts,

Methods
The following transecls (Fig. 1) were
chosen, and in cach case the transect ran nor-
mal to the coast, from low walter level down
the slope te the depth where rock was buried
by sand, except for transect D which was pre-
dominantly on sandy bottom.

* Department of Fisheries, Gawler Place, Adefuide, §. Aust. 5000.
t Department of Botany, University of Adelaide, Adelaide, S; Aust, 5000.

178

St Francis 4,

Masillon | 3P aa

Fig. §. Map of four of the isles of St Francis
showing the position of the four transects
(A-D). Inset shows the situation of the
islands in the northern Great Australian
Bighi.

Transect A was on the NW corner of St
Francis I., under conditions of strong water
movement, and terminated at 57 m depth. A
sampling gap (32-38 m deep) in this transect
was filled by collections from a similar site on
Ege I.

Transect B was on Masillon 1, subject to
moderate water movement, and terminated at
33 m depth,

Transect C on St Francis 1, was aubject to
relatively slight water movement, and des-
cended to 20 m.,

Transect D in Petrel Bay on St Francis 1.,
wus the most sheltered site, with a gently slop-
ing, sandy, sea-floor dominated by seagrasses-

S. A. SHEPHERD & H. B. 8. WOMERSLEY

On transects A, B and C, the diver first
swam aboul 2 m above the bottom along the
transect, estimating the percentage cover and
recording the vertical range of the prominent
brown algae. This was then repeated on twa
parallel lines, one each side of the first transect
and about. 15-20 m distant. Estimates of cover
for particular depths are given as the average
of these three valucs, which were made sub-
jectively as a percentage on @ scale of 0-10,
Many species vary considerably in percentage
cover over a horizontal distance of some
metres, but the figures given provide an overall
assessment of the cover along the transects.

Communities were recognised by the upper
stratum dominants, this being the most satisfac-
tory method on such surveys. At the depths
studied. alyac were dominant except at 30-57
m on transect A, and within each community
quantitative samples of the upper stratum were
taken for biomass estimates. This was done by
counting the plants in a hoop af § m* area,
placed sequentially along a horizontal line
some 16-24 times, so as to give a total
sampling area between 2 and 3 m®*. and the
humber of plants per m? calculated. The aver-
age weight of an individual plant was deter-
mined by weighing a random sample of ID
plants, and the biomass per m= then calculated.
The other strata were sampled by means of 4
to 8 sequential samples, each of $ m*,

On transect D in Petrel Bay, a diver was
towed on an underwater sled behind a boat,
and the distribution and depth range of the sea-
grusses Were noted, and photographs taken.

‘The algal samples were preserved in 4%
formaldehyde-sea water, and taken to the
laboratory for determination and analysis. Bio-
mass figures are based on the wet weight of
the preserved collections after removal of sur-
face water, These estimates should be taken as
examples only of the size and variation in bio-
mass of the community dominants, since the
restricted diying time on a short expedition
such as this limits the range of transects, and
the number of samples that can be taken, The
transect samples were supplemented by other
géneral observations and collections, and by
photography.

Depths were measured by capillary and
mechanical depth gauges, and the results aver-
aged and adjusted to approximate low lide
level.

Environmental factors

The shart duration of the survey precluded

Getailed studies on environmental factors, but

SUBTIDAL ALGAL, AND SEAGRASS ECOLOGY ivy

the following information is available. The
isles of St Francis rise from a maximum water
depth of about 60 m at the southwest of St
Francis I. and are comparable to the Pearson
Islands in their distance from the mainland and
in their topography,
Water movement

St Francis I. is subject to a strong south-
westerly swell of 10-12 second period, prevail-
ing throughout the year, similar to that at
Pearson I. and West I. (Shepherd & Womers-
ley 1970, 19714). In summer, a short southerly
swell is generated by the strong southerly winds
which blow for about 12 hours each day and
are characteristic of this part of the Great Aus-
tralian Bight. Wave action on all parts of the
islands facing south to west is strong in the
intertidal region.
Temperature, salinity and nutrients

Sea surface temperatures range from
18-20°C in summer ta 14-15°C in winter,
according to. Vaux (1970) and data obtained
from various oceanographic stations in the
Vicinity (C'S.I.R.O, 1967a, 1967b, 1968, 1969).
In summer, bottom temperatures at 50 m depth
are 2-3°C lower than sea-surface temperatures,
During the study, the surface temperature was
18°C off the island and about 20°C inshore in
Petrel Bay.

aot
*

Q SPONGES
C HYDROLDS

Salinity, phosphate, nitrate and oxygen levels
are similar to those for Pearson L., viz. salinity
35.6-36.2%-«: inorganic phosphate 0,09-0.17 pg
atom/ litre; nitrate about 0.3 pg atom/litre; and
oxygen saturation 93-103 %.

Submarine light intensity

Light penetration was not measured, but
according to H. Jitts (pers. comm,) it corres-
ponds to that for Type Il oceanic water of
Jerlov (1968), and is thus only slightly less
clear than the waters about Pearson T.

The algal and seagrass ecology

The algal-dominated subtidal photic zone at
other localities in South Australia has been
found to present three main zones, designated
as the upper, mid, and lower sublittoral zones
(Shepherd & Womersiey 1970, 1971). These
zones are alsa apparent in the areas studied at
St Francis I.

Communities of the rocky coast will be des-
cribed first, including these on both horizontal
and sloping rock but not those in crevices or
under overhangs, followed by the sea-grass and
algal communities of sheltered, sandy areas.
The communities studied are essentially those
subject to suflicient light intensity to be plant
dominuted, but prominent animal species ate
mentioned where present.

m BALANUS NUGRESCENS

a “Fo INTERMEDIA

$F vvarovesen HARVEVANUM

Ys

SCYTOTHALIA DORYCARPA

EARGASSUM BRACTEOLOSUM

ECKLONIA RADIATA

sa RED ALGAE

% ZONARTA

P BRYOZOA

Fig. 2, A vegetation profile of transect A (St Francis L)-

mid S.L.
20

30

S$. A. SHEPHERD & H. B. 8. WOMERSLEY

ora CORALLINA

CYSTOPHORA PELL INATA

f CYSTOPHORA MONILITORHIE

Fig. 3. A vegetation profile on transect B on Masillon I. Sce Icgend on Fig. 2 for other algal taxa.

a UUJANIA

10
mid S.L.

*¥ CYSTOPHORA MONILIFERA

a OSMUNDARTA PROLIFERA

SARGASSUM
VERRUCULOSUM

SARGASSUM

¥ BOTRYOCLADIA GBOVATA

Tenereawens:

Fig. 4, A vegetation profile of transect C on St Francis I, See legend on Fig. 2 for other algal taxa.

A. ROCKY COASTS

Transects A, B, and C traverse rocky areas,
generally siceply sloping and imcluding both
horizontal and sloping rock surfaces. The
marked light gradient with depth, coupled with
the. considerable gradient in water moyement
within each transect (especially A) and also
between the transects, gives rise to a fairly dis-
tinct zonation of algae. Profiles for transects
A, B, and C are given in Figs 2-4, and the
depth relationships of the communities to water
movement are given in Fig, 5.

L. Upper sublitteral zone

This zone is subject to the most intense water
movement, and varies in vertical width from
5(-7) m on rough-water coasts (transect A,
Fig. 2) to 2(-3) m on sheltered coasts (transect
C, Fig, 4). Communities of this zone typically
have a single, dense, stratum of fairly uniform
height, ranging from 15-20 cm for the Cysto-
phora intermedia community to 2-3 em for the
Jania community.

Corallina cuvieri, in high-light situations (i.e.
especially horizontal surfaces and those facing

SUBTIDAL ALGAL AND SEAGRASS ECOLOGY ik]

TABLE f
Species aud biomass (g/m) compostrion of upper sublittcral
communities tn samples taken at about 1(-1,5) m depth on
transects A, B and C. “PY indicates xpurve occurrence
althwugh net present in sample.

‘Transect A B c
Ara sampled (m=) O25 1.25 637
Water movement Strong Moderate Slight

Dominant Species

Cusophora intermedia (,280 p 30
Myrivdesma harveranum 1,400
Corallina cuvierl

1, eringaia 2,800 4,200
Janie favrintatsa Pe

Other Species

Canterpo brownli 70
Caulorpa papillon 140
Cyrophera gracilix <i
Lobosnita blenspldata 40 ay
Pachydielyon panioilatam Str Xr su
Saryax\urm spp. <10
Callophyllis raralfreiray 28) 15
Champiu obsolpta 40
Dasva clavigera <0
Griffithsia tages 260
Hypnea sp, 40
Laurencia filifarmis £.

herernclada 290
Séapara harveyiana <i0
Polvaiphonia nigrita <i
Total coverage 100 100 lub
Biomass. g/m! 3,950 2,840 4,330 [1,900]
Number of species 14 4 5

Ie

In each case the biomass value of the species charactens-

ing the community is in bold type,

* This sample was taken from a distinct Juala community
ala depth of about 0.5 m (see Fim 4).

nerth or cast) and subject to strang to moder-
ate water movement, forms an almost pure
community completely covering the rock sur-
face, In calmer areas a Jania fastigiafa com-
munity, presenting, a somewhat similar aspect
of short, tufted plants, replaces the Corallina.
The Cerallina cuvieri community extends up-
wards into the lower eulittoral zone of the
intertidal in rough-water situations, as des-
cribed by Womersley & Edmonds (1958, p.
232),

Cystophora ihtertiedia forms a fairly pure
cammunity under slightly less extreme water
movement than Corallina, and also on sloping
{rather than horizontal) surfaces subject to
somewhat lower light intensity. While Cysto-
phora intermedia may be dominant in. such
situations, in lower light intensity Myriedesmie
harveyanum becomes co-dominant, with
numerous associated species of green, brown
and red algae (see Table 1).

Cystophora intermedia is rare within the
Corallina cuvier? community, but may be com-

STRING WATLR MOVEMENT G&ADLENT SLICK!
TRANSECT A. 6 c
0 7

WRIGDESM
HARVEFANUM

‘

a. (VST. MONILTFORMLS }

S SANG. BRALTEUSUSIIF ¢
7.

CYST POWILIELRA
SARG. VPRRUCELOSUM
cyst. PRCTUMATA

n> 1
Sy et

ESKLINLA RADIATA

SLPTOThALLA DORYEARCA

Fig. 5. Change in vegetation patterns along water
movement and depth gradients.

mon near the upper and lower boundaries of
this community. Its occurrence at the upper
limit (i.e. near low tide level} agrees with the
observations of Womersley & Edmonds (1953)
that it marks the sublittoral fringe. but at St
Francis 1. it is not confined to this zone, occur-
ting also as decp as 3. m,

2. Mid sublittoral zone

As at West I. and Pearson I[., this zone on
St Francis I. is characterised by larger brown
algae 30.cm=1 m in height, forming an upper
canopy or stratum over a lower stratum mainly
of red algac 5-25 cm in height. The upper
limit of this zone depends on the intensity of
water movement as described tor the upper
sublittora) zone, and the lower Jimit on the
limiting depth of large brown algae; this ts
about 45(-47) m deep on transect A. The
vevetation profiles of Figs 2-4 represent the
appearance of this zone on transeets A, B, and
C and the relations of the vegetation patterns
with waler movement are shown in Fig. 5. The
average cover of the important upper stratum
species is given in Fig, 6.

Several communities could possibly be recog-
nised in (his zone, but more extensive studies
than were possible in the time available ure
needed to establish their validity. The domin-
ants and understorey Species will therefore be
discussed more generally.

Ecklonia radiata and Scytothalia dorycarpa
(Fig. 9) dominate this zone under conditians
of considerable water movement at the rough-

182 S, A. SHEPHERD & H. B. S. WOMERSLEY

TABLE 2

Biomass (g/m*) composition of mid. sublitioral species in samples taken at certain depths on 3. transects. Further

data on the vertical range of the species is given in the appendix

ee

Transect A B

Depth (m) 6 13 35 6 22 32 6
Area sampled (m2) 0.5 1 1 05 0.5 0.5 0.5

Upper stratum
*Ecklonia radiata 2,200
* Scytothalia dorycarpa
Cystophora pectinata
Cystophara moniliformis
Myriodesma harveyanum
Sargassum bracteolosum
Sargassum varians
Sargassum verruculosum
Sargassum linearifolium
Sargassuni héteromorphum
Sargassum decipiens
Cystophora brawnii
Cystophora subjarcinata
Cystophora monilifera

Upper stratum coverage (9%) 100. 95 35 90 50 60 80
Upper stratum biomass
(g/m) 4,650 3,775 2,210 2,250 2,460 3,700 3,250

Lower stratum

Brown. algae
Dictyopteris muelleri
Dictyata diemensis
Dictyota prolifera
Chlanidophora microphylla
Glossophora nigricans
Hydreclathrus elathraius
Lobospira hicuspidata
Pachydictyon paniculatum
Zonaria spiralis
Zonarta sinclairii
Zonaria inrneriana

Red algac
Austraphyllis aleicornis
Ballia callitricha
Botryocladia obavata
Chamipia affinis
Cliftonaea pectinata
Delisea hypneoides
Delisea pulchra
Kallymenia cribrosa
Laurenvia filiformis f,

dendritica
Laurencia spp.
Osmundaria prolifera
Plocamium angustum
Plocamium cartilagineum
Plocqminm mertensti
Placamium preissianum
Pterosiphania sp,
Rhodophyllis membranacea
Sonderaphycus australis
Webervanbossea kaliformis

1,800 450 900 -1,200
200 50 1,380 = =2,500

Nm

PTILItbaitias

Pitt dd
Privy ddd
wasuu | ro | Sus

|

30
<3
25
25

TTT |
$18!
wi

105 3

TTT ttt

lStr ltt
Pll bl td

LLEtbti dd

| |

>
=
i)

10
25 180
20
680
— 50
ee:
— 50
1160 1,610 150 1,040

Ll Ll peal | |
lSllSslusil |
Ll td) aseles

L| |) Sell] |
i —
Blulaulwussws||S

Biomass

|
3

8

||

Pld td

all lwel Sl |

215

| | inns | |

=
Bes
ur | | laal aS& |

1,214

3,615

Total number of species
in Sample 4 7 25 24 22 12 11

16

29

Total Biomass. 4,650 3,845 3,098 3,410 4,070 3,850 4,290 6,020 4,151

* Values for Ecklonia and Scyrothalia are mean values over 2-3 m2,

7
4,445

SUBTIDAL ALGAL AND SEAGRASS ECOLOGY

UPPER STRATUM
LOWER STRATUM

meee enaee

WASOTININYYAA WASSYONWS

SNYIYVA WNSSVONVS

SNAId154a0 WOSSVDuVS

jo WNSQ10dLI vad WNSSVOXVS

(n> VAVNTDuVAENS YHOHAOLSAD

VLVNILIAd YYOHAOLSAD

SIWHOSTTINOW YHOHdOLSAD

VHSSITINOW VHOHdOLSAD

é ~>~ [va]

VIVICVY VINOTNDS

cover of prominent upper stratum species of the mid sublittoral zone, and total % cover of upper and lower strata

Fig. 6. Vertical distribution of %

on transects A, B and C.

183

IE4

water site (transect A, Fig, 2), but with Jess
water movement (trunsect B, Fig. 3) other
Species of brown algac (Sargassuni Aracren-
lexeat and C'ystuphora moniliformis) also be-
come prominent, With greater shelter (iransect
C, Figs 4, 7, 8), Cystophora inenilifera, C.
pectinata and Sergasswn vyerrucuiosunt are
common, together with ©. subfarcinate, Sar-
gassum decipiens and S. varfans. These species
are most common in the upper part of the mid
sublittoral, with Eeklonia and Scylotheelie still
common in the lower part of thts zone (Fig. 4).

Sargassum Bracteolasum and Cystophord
menififornis on transect BL and §, verruen-
lovuen and C. nnartilifera on transect C, have
similar vertical distributions, cnabling Feld
recognition of algal subzunes dominated by
these species pairs.

Unuetstorey species are spatse over much
of transects A and C, and moderately common
only on transect B where the canopy ts less
dense. The distribution of many understorey
species is too patchy to show any obvious rela-
tionship with either the distribution of upper-
stotey species or with apparent environmental
factors.

The commonest understorey species are of
Plocamivm (P. angustam, Po mertensi and P.
preissionium). They occurred on all] three tran-
secis, with a vertical depth range of as much ay
50 m; where they are rare or absent on hori-
zontal surfaces, they are usually present on
vertical ones.

Other species with a wide vertical range are
Caulerpa Srownij, Lebospira bicuspidate and
Pachydictyon panicularuni; these species are
known to he tolerant to a wide range of light
intensity and of water movement. Some other
species [e.g, Gloxsophora nipricans, Austro-
phivilis aleicornis, Cliffanaea pectinate and
Delisea Aypneoides) were found only in Geeper
water; most of these species occur also at West
T. and Pearson L, with similar distribution an
conditions of lew light (and slight water move-
ment at depth}.

On transect C at 17-18 m depth (Fig. 4),
Where os an abrupl decline in the number of
upper stratum species and their coverage, and
aim increase in coverage of several species of
the lower stratum, e.g. Osinutdaria prolifera,

$3. A. SHEPHERD & H. B, S. WOMERSLEY

Borryocladia obovata and Hydroclothris clath-
rains, This community forms a band [2 m
wide lying immediately above the sandy bottom
at about 20 m depth, and the species are appa-
rently tolerant of the sedimentation which is
Pronounced over this narrow band,

The cover of upper and lower stratum spe-
cies, with depth, is given in Fig. 6, Upper
stratum cover is highest between 5 and 15 m
depth, declining with depth, whereas lower
stratum cover is lowest where the upper stra-
tum 1s most dense, and in general increases
with depth until light becomes limiting.

3. Lower subliteral zone

Only on uunsect A does rocky substrate
descend to sufficient depth for the lower sub-
littoral zone dominated’ by red algae (Shep-
herd & Wamersley 1970, 1971) to occur. On
this transect, a communily (Fig. 10) of red
aleac together with bryozoa, sponges and
hydroids, occurs between 47 and 57 m deep,
The community is rich in algal species (but of
low biomass), the most common being Plo-
cuniion ampgustim, P. mertensii and PF. preis-
Signum, several other species (Rhodymenta
australis, Gattva pinnella, Rhedocallis elegans
and Kallymenia spinosa) were found only in
this collection. Algal cover in this community
is low, averogmg 10% (5-15%), indicating
that 57 m is close to the depth (i.e. Light) limit
for most algae in this region.

B, THE SEAGRASS COMMUNITIES IN
PETREL BAY

Three seagrass communities occur in this
sheltered bay, forming bands around the bay
dependent on substrate ani depth.

Amphibolis antarcsica fringes the shore from
low walter mark to U5 m below, attached by its
rhizome-root system te calcarcous recfs of low
relief. Below these reefs the bottam is sandy,
and at a depth of about 2 m, Posidema osten-
feldii forms a fringe community about 20 nm
wide around the hay, Beyond this, descending
to 22 m decp, Posidonia australiy (narrow leat
form} i dominant in fairly continuous beds,
Beyond abqut 22 m deep, P. australis hecornes
sparse, and at the ume of the survey a loase-
lying but apparently healthy community ol the
red algo Hennédya crispa occurred at this
depth.

Fig. 7, Algal community at § m depth on transect C, Note Cystaphora mronilifera (top sett), Scylo-
thalia doryearpa (tep cight) and several species of Sarvaysum (ventre and lower righr)..

Fig. &. Algal community at 10 m depth at transect C. Note species of Sargassiem (centre left), Eckloptia
radiata (top right) and €ysropherd moniliferd (contre and lower eight),

185

SUBTIDAL ALGAL AND SEAGRASS ECOLOGY

S. A. SHEPHERD & H. B. S. WOMERSLEY

SUBTIDAL ALGAL AND SEAGRASS ECOLOGY 187

Discussion

Algal zones within the sublittoral, and the
distribution. cover and biomass of the com-
ponent spectes, have been described for many
Coasts elsewhere in the world. Recent accounts
are those of Liining (1970) from Heilgoland,
Boudouresque (1971) from the Mediterranean,
and Mann (1972) from the Atlantic coast of
Canada. These and other accounts show that
broad algal zones, correlated with light inten-
sity and the degree of Walter movement, occur
in the photic zone on most coasts,

Although limited in extent, this survey of
the subtidal algal vegetation of St Francis
I. shows a similar zonation pattern to that
at West [ and Pearson 1, (Shepherd &
Womersley 1970, 1978). As at these islands,
the vertical extent of the upper sublittoral zone,
and to a lesser extent the mid sublittoral, is
dependent on the degree of water roughness
(with whrch light penefratton is also associa-
ted), The extent of the upper sublittoral zone:
probably corresponds with the depth to which
“white water” (1.8. turbulent walter carrying air
bubbles) penetrates under average swell condi-
tions. Ried] & Forstner (1968) considered the
vertical height of their “inner surf zone”
{Riedl 1971) ta correspond to 2.5 x wave
height, and this could also be applied to the
upper sublittoral zone on South Australian
coasts Where wave heights are 1.5-2 m in a
moderate swell. Chapman (1967) in discussing
the presence of a suhblittoral fringe in many
parts of the Pacific is largely referring to this
upper sublittoral zone. The term “sublittoral
fringe” is best restricted to the zone emergent
during suck back of waves at low tide, when
this zone is ecologically distinctive ( Womersley
& Edmonds 1952)..

The mid sublittoral zone at St Francis I.
shows similar features to this zone at West DL
and Pearson I., being dominated by the larger
brown algae and with an (inderstorey of mainly
red algae. Further studies muy show chat dis-
unet conimunities could be recognised in this
zone, since competition between the various
dominant species is apparent, and, over the
considerable depth range, both light intensity
and degree of water movement vary consider-

ably. While most species show typical “bell-
shaped" distribution patterns (as discussed by
Whittaker 1967), some (e.2, Cystophora moni-
lifera, C. subfarcinata) apparently show
slightly bimodal distributions (Fig. 6), pro-
bably due to competition with other species
better suited to the environment within their
extremes.

The lower sublittoral zone of red algae was
observed only in depths of 47-57 m at St
Francis [., corresponding to the situation at
Pearson E. rather than wt West I, and reflecting
the ¢larity of the water. This zone lies below
the light intensity necessary for the larger
brown algac and grades to the flower photi¢t
levels of the red algae. At St Francis L, inter-
mixing of lower sublittoral red algae with
fauna such as hryozoa, sponges and hydroids,
was More prominent than at West 1,

Although sublittoral zones are well defined
at St Francis [., this characteristic is empha-
sized by choice of transects on steeply sloping
shores involVing steep light and water move-
ment gradients. On more irregular shores. dis-
linct zonation is less apparent.

Apart from ecological differences associated
with depth, which reflect mainly the decrease 1
light intensity, light relationships are apparent
in the mid sublittoral zune where a dense upper
canopy may reduce the light reaching the lower
stratum by up to 95%. This effect was well
shown on transect A at 5-15 m depth and
transect C at 7-13(-15) m depth, where a
dense canopy covered a sparse understorey.
Where a dense canopy exists with considerable
water movement, reduction of the understorey
may alsc be due to the physical effect of the
larger fronds sweeping over the rock.

The effect of sediment (fine sand or silt
stitred up in stormy weather and settling on
the seabed under calmer conditions) was evi-
dent in tLyo places. Near the end of transect B,
at about 30 m depth, sediment is present on
rocky surfaces and here there is an abrupt
decline jn cover of the lower ‘stratum, At the
end of transect C at 18(-20) m depth, where
sediment also covers the rocky bottom, there
is a distinctive community of certain red algae
(Rotryocladia ebovate, Osmundaria prolifera)
which can tolerate sediment, The effect of sede

Fig. 9. Algal community at 16 m depth on transect C. Note Ecklonia radiate {top right and lower left)

and Seyterhalia dorycarpa (centre)-

ig. (0. Sparse red algal community at 57 m depth on transect A,

188 5. A. SHEPHERD & H. B. 8. WOMERSLEY

ment in inhibiting algal colonisation and
growth has been recently discussed by Grigg &

Kiwala (1970)-

The survey of St Francis I. was limited in
time, the area covered, and in the variety of

habitats sampled. Nevertheless, the subtidal
algal flora appears fairly rich. with some 138
species recorded, compared to 160 for Pearson
1. and 132 for West I. Further studies would
certainly extend this number considerably,

Appendix: Algal species fist

Identifications are by H. B. S, Womersley, Dr G. T. Kraft (Mychodeaceae, Dicranemaceae and Acro-
tylaceae) and Dr BE. M, Wollaston (Crouanieac).

CHLOROPHYTA
Caulerpales
Caulerpa brownit (C.Ag.) Endlicher
Caulerpa cactoides (Turn.) C. Agardh
Caulerpa flexilis Lamouroux
Caulerpa flexilis Lanjouroux var. muellert
(Sond.) Womerslcy
Caulerpa hedleyi W. v. Bosse:

Caulerpe longifolia C.Ag_{. erispata (Harv.)

Womersley
Caulerpa obscura Sonder
Caulerpa papillosa J, Agardh
Cuulerpa scalpelliformis (R.Br.) C, Agardh
Caulerpu simpliciuscula (Turner) J. Agardh
PHAEOPHYTA
Dictyotales—Dictyoteae
Dictyota diemmensis Knetzing
Dictyota furcellata (C.Ayg.) J. Agardh
Dictyota prolifera Lamovurowx
Dilophus fastigiatus Sonder
Dilephus robustuy (1.Ag.) Womersley
Pachydictyon paniculainm J. Agardh

Pachydictyon nov. sp?

Glossophora nigricans (J.Ag.) Womersley

Lobospira bicuspidala Areschoug
Zonariewe

Chlanidephora microphylla (Harv.) J. Agardh

Dictyapteris muyelleri (Sand.) Reinbold
Lobophore variegata (Lamx.) Womersley
Zouaria crenata J, Agardh
Zonaria sinclairii Hooker & Harvey
Zonaria spiralis (J.Ag.) Papentuss
Zohkaria turneriang J, Agardh
Nov. gen?

Chordariales—Chordariaceae
Corynophlaea cystophorae ¥. Agardh
Bactrophora filum (Harv.) J. Agardh

Bactrophora vermicularis J. Agardh

Polycerea nigrescens (Harv. ex, Kuetz.) Kylin

Sporochnales—Sporochnaceae

Bellotia ériopherum Harvey

Sporochnus comosus C. Agardh
Dictyosiphonales—Giraudyaceac

Giraudya spiacelarioides Derbes & Solier

Punctariaceae

Hydroclathrus clathratus (C,Ag.) Howe
Laminariales—Alariaceae

Ecklortia radiata (C.Ag.) J. Agardh
Fucales—Cystoseiraceae

Seytothalia dorycarpa (Turn.) Greville

Cystophora brownii (Turn.) J. Agardh
Cystophora intermedia J. Agardh

z

2-38; B, 4-7; C, 2-6, 19
a2
2

*
Ud we
wh

r
33
©
gs
‘S

Oo
nw

, 13-18: C.19;D,2
BC, 10-13

P>PuUy Se >>>

a

Ka Pb b> lA

ieee
Be

, 32-38; B, 636, 6, 19
+35
32-38; oe 13-22; C, 19

1
3248: B, 13-18
, 2, 35; B, 0-7, 22; C, 2-19: Masillon I. in

o>>>

32-38, 55; B, 13-22
2, 35; B, 6, 13-18; C, 6-19

o> > >>
oe
ga

"92: ion 19-20
A, 32-38: B, 22; C, 6-19
B, 13-18
A, 32-38
A, 10, 32-38; B, 13-22; C, 19
A, 13, 32-38; B, 6-22; C, 10-19
A, 13, 32-38
A, 32-38
C, 10-18, on Cystephora brownii
C, 19-22: D, 3, 4, on Posidenia australiy and

1

1

P. osienfeldit

*, 6

, 6, 19-20: D, 3, 4. on Posidonia australis
and P. ostenfeldii

OO

A, 32-38; B, 13-18
A, 32-38

D, 3, on Posidonia australis

C, 19-20

A, 5-38: B, 5-32; C, 8-20

A, 6-38; B, 4-32; C, 8-19, Masillon L. in
bay, 14

B, 4-7; ©, 2.13
A, 0-2;,C, 0-3

SUBTIDAL ALGAL AND SEAGRASS ECOLOGY 189

Cystophora gracilis Womersley & Nizamuddin
Cystophora monilifera J. Agardh
Cystophora moniliformis (Esper) Womersley &
Nizamuddin
Cystophora pectinata (Grev. & C.Ag.) J. Agardh
Cystophora subfarcinara (Mert.) J. Agardh
Myriodesma harveyanum Nizamuddin & Womersley
Sargassaceae
Phylotrichia
Sargassum decipiens (R.Br.) J. Agardh
Sargassum heteromorphum J. Agatdh
Sargassum varians Sonder
Sargassum verruculosum (Mert.) Agardh
Arthrophycus
Sargassum bracteolosum J. Agardh
Sargassum lacerifolium (Turn,) Agardh?
Sargassum trisfichum Grey. & Agardh ex Sonder
Eusargassum
Sargassum linearifoliam (Turn.) Agardh?
Sargassum podacanthum Sonder?
Sargassum spinuligerum Sondet
Sargassum distichum Sonder
Sargassum (Eusargassuin, tribe Glomerulatae?)

RHODOPHYTA

Nemaliales—Chaetangiaceae

Galaxaura spathulata Kjellman
Helminthocladiaceae

Liagora harveyiana Zeh
Bonnemaisoniaceae

Asparagopsis armata Harvey

Delisea hypneoides Harvey

Delisea pulchra (Grey.) Montagne

Gelidiales—Gelidiaceae

Pteracladia lucida (R.Br.) I. Agardh

Cryptonemiales—Dumontiaceae

Acrosymphyton taylori Abbott
Squamariaceae.

Senderephycus australis (Sond.) Denizot
Corallinaceae (excluding encrusting taxa)

Amphiroa anceps (Lamarck) Decaisne

Jania fastigiata Harvey

Jania micrarthrodia Lamouroux?

Jania pusilla (Sond.) Yendo

Jania sp,
Corallina cuvieri Lamouroux
Corallina cuvieri f. crispata Lamouroux
Metagoniolithon charotdes (Lamx.) W. v. Bosse
Metagoniolithon stellifera (Lamarck) W. y, Bosse
Poly porolithen patena (H. & H.) Mason
Cryptonemiaceae
Carpopeltis phyllophora (H. & H,) Schmitz
Cryptonemia undulata Sonder
Halymeénia harveyana ¥. Agardh
Thamnoclonium dichotomum (J.Ag.) J. Agardh?
Grateloupiaceae
Gelinaria ulvoidea Sonder
Kallymeniaceae
Austrophyllis alcicarnis (J.Ag.) Womersley & Norris
Callophyllis rangiferinus (Turn.) Womersley
Callophyliis lambertii (Turn.) J, Agardh
Kallymenia cribrosa Harvey
Kallymenia spinosa Womersley & Norris
Thamnophyllis lacerata Womersley & Norris

Gigartinales—Plocamiaceae

Plocamiumn angustum (3.Ag.) Hooker & Harvey

B, 4-7
A, 0-2; C, 2-19

B, +7
A, 6, 13; B, 6-18; C, 10-19
B, 6; C, 0-13, 19, Masillon I. in bay, 1-4

A, 2; B, 4-7

. 32-38; B, 6-18; C, 6-19
» 35; B, 13-18; C, 2-19

A, 13, 32-38; B, 4-22; C, 10-19; D, 2

A, 12, 32-38
Masillon I. in bay, 1-4

A, 32-38; B, 22. Masillon I. in bay, 1-4
A,2

A, 10, 32-38; B, 4-18

A, 32-38; B, 13-18; C, 19

A, 10, 32~39;,B, 6-22

B, 4-7, Masillon J. in bay, 1+

A, 32-38

B, 13-32

A, 32-38; B, 13-18, 32; C, 10-19

C, 0-2

D, 3 on Postdonia ausiralis

B, 4-7; Masillon I. in bay 1-4 on Cysto-

phora subfarcinata
A, 32-38

C, 6-1
D, 2 on Amphibolis antarctica
B, 13-18 on Ballia caltitricha

A, 32-38, Masillon I. in bay, 1-3

D,2
B, 13-18; C, 10-13
A, 32-38

C, 19-20
A, 32-38; B, 13-18, 32
A, 2-10; B, 0-7; C, 10-13
A, 55

A, 32-38: B, 13-22

A. 55

‘A. 32-38

A, 6-55; B, 6-32; C, 6-19

190 S. A. SHEPHERD & H. B. 8S. WOMERSLEY

Plocamium cartilagineum (.) Dixon
Placamium leptophylium Kuetzing
Plocamium mertensii (Grev.) J. Agardh
Plocamium preissianum Sonder

A, 6—10, 32-38, 55; B, 13-22, 32
A, 55

A, 6-55: B, 6-22: C, 6-19
A, 6-55: B, 6-32; C, 6-19

Solieriaceae

Solieria robusta (Grev.) Kylin A, 32-38; C, 20; D, 4
Rhabdoniaceae

Areschougia congesta (Turn.) J. Agardh? A, 32-38
Rhodophy!lidaceae

Rhodophyllis membranacea (H. & H,) Harvey
Rhodophyllis ramentacea (C.Ag,) J, Agardh

A, 35: B, 13-22
A, 32-38; B, 32

Hypneaceae
Aypnea episcopalis Hooker & Harvey B, 6; C, 10-13
Alypnea sp. A, 2; B, 46
Mychodeaceae
Mychodea pusilla (Harv.) J. Agardh D, 2, on Amphibolis antarctica
Mychodea ramulosa J. Agardh B, 4-7
Mychodea carnasa Hooker & Harvey A, 32-38
Neurophyllis australis Zanardini C, 19-20

Dicranemaceae
Dicranema revolutum (C.Ag.) J,-Agatdh
Acrotylaceae
Hennedya crispa Harvey
Rhodymeniales—Rhodymeniaceae
Fauchea?
Webervanbasyea kaliformis (J.Ag.) J. de Toni
Webervanbossea splachnoides (Harvey) J. de Toni

D, 2, on Amphibolis antarctica
D, 24, loose-lying
A, 32-38

A, 32-38; B, 22; C, 10-13, 19
A, 32-38: C, 19-20

Botryocladia obovata (Sonder) Kylin C, 19-20
Coelarthrum cliftonit (Harv.) Kylin A, 32-38
Caelarthrum meulleri (Sond.) Boergesen A, 35

Gloiosaccion brownii Harvey

B, 13-18; C, 10-13
Rhadymenia australis (Sond.) Harvey

A, 32-38, 55; D, 2

Lomentariaceae
Champia affinis (H, & H.) J. Agardh A, 32-38; B, 13-18; C, 19
Champia obsoleta Harvey A, 2
Champia tasmanica Harvey A, 32-38
Ceramiales—Ceramiaceae
Crouani¢ae
Gartya pinella Harvey A, 55
Gulsonia annulata Harvey C, 19-20
Antithamnieae
Acrothamnion preissii (Sond.) Wollaston B, 13-18, on Ballia eallitricha

Antithamnion divergens (J,Ag.) J. Agardh A, 35
Ballia ballioides (Sond.) Wollaston

Rallia callitricha (Ag.) Kuetzing A. 33 B, 13-18, 32
A, 55

Ballia mariana Harvey 3
Platythamnion nov. sp? A,55
Griffithsieae
Griffithsia tezges Harvey A,2
Callithamnieae
Callithamnion sp. A, 32-38
Callithamnian sp. A, 55
Dasyphileae
Rhodacallis eleguns Kuetzing A, 55
Delesseriaceae
Apoglossum tasmanicum (F.v.M.) J. Agardh B, 32
Dasyaceae
Dasya clavigera (Wom.) Parsons C, 0-2; D, 24
Dasya naccarivides Harvey? C, 19-20
Rhodomelaceae—Polysiphonieae
Poalysiphonia nigrita Sonder A,2
Pterosiphonicae
Pterosiphonia sp. B, 6
Herposiphonieae
Dipterosiphoniu? nov- sp? B, 13-18
Herposiphonia nov. sp? A, 55

SUBTIDAL ALGAL AND SEAGRASS ECOLOGY 191

Polyzonieae

Clijtenaea pectinala Harvey
Amansieae

Osmundaria prolifera Lamouroux
Laurencieae

Lawurencia elotu (Ag,) Harvey

Laurencia filiformis (Ag.) Montagne f. dendritica

Saito. & Womersley

A, 32-38; B, 13-18; C, 19
B, 13-18; C. 6-20

B, 13-18

A. 32-38

Laurencia filifarmis (Ag.) Montagne f. heteraclada

Saito & Womersley
Laurencia paniculata (Ag.) J. Agatdh

SPERMATOPHY TA—seagrasses

Potamogetonaccac
Heferozostera tasmanica (Mart. ex Aschers)
den Hartog.
Posidonia australis J. D. Hooker—narrow. and
broad forms

Posidonia ostenfeldti den Hartog

A, 0-2; C, 19-20
C, 19-20

D, 3

C, 19-22: D, 3-4
D, 3, Masillon I, in bay 5-9

Amphibolis antarctica (Labill.) Sonders ex Aschers D,2

Acknowledgments
We are gtateful to the Royal Society of
South Australia (Research and Endowment
Fund) and the Department of Fisheries for
grants towards the field work. The first author
was assisted in the SCUBA work by Mr K. L.,

Branden and Mr N, Coleman, to whom thanks
ate expressed. Assistance from the Australian
Research Grants Committee in provision of the
technical assistance of Mrs E, L. Robertson
and Miss C. Anderson, is gratefully acknow-
ledged by the second author.

References

Boupouresgur, C. F, (1971).—Contribution a
"Etude Phytosociologique des peuplements
alzaux des Cotes varoises. Vepetatio 22, 83-
184,

CHapman, V. J. (1967).—The sublittoral fringe in
the Pacific. J. Indian Bot, Soc. 46, 337-343.

CSLR.0, (1967a)—Aust. Oceanogr. Cruise
Report No. 34.

CSTR.O. (1967b)—Aust. Oceanogr. Cruise:
Report No. 46.

C.S,LR.0, (1968)—Aust. Oceanogr. Cruise
Report No. 43.

C.S.LR.0. (1969).—Aust. Oceanogr, Crulse

Report No, 54.

Gris, R. W, & Kwara, R. S. (1970).—Some
ecological effects of discharged wasies on
maziue life. Calif. Fish and Game 56, 145-
185.

JERLOV, N. G. (1968).—"Optical Oceanography”.
(Elsevier: Amsterdam.)

Lining, K, (1970).—Tauchuntersuchungen zur
Vertikalverteilung der sublitoralen Helgoliin-
der Algenvegetation. Helgoldnder — wiss.
Meeresunters 21, 271-291.

Mann, K. ( 1972), —Ecological energetics of the
seaweed zone in a marine bay on the Atlantic
coast of Canada. |. Zonation and biomass of
seaweeds. Mar. Biol. 12, 1-10,

Rrept, R. (1971)—"“Water movement”. In 0,
Kinne (Ed.). “Marine Ecology. A comprehen-
sive integrated treatise on life in oceans and
coastal waters”. Vol. 1, Part 2, pp. 1123-1156,
(Wiley—Interscience: London.)

Rrepr, R. & Forsrner, H. (1968).—Wasserbewe-
gung B Mikrobeyeich des Benthos. Sarsia 34,
163+!

SHEPHERD, S. A, & Womerstey, H. B. S.
(1970).—The sublittoral ecology of West I.
South Australia. 1, Environmental features
and the algal ecology, Trans. R. Sov. §. Aust.
94, 105-138.

SHEPHERD, S, A. & Womensiey, H. B. S.
(1971) —Pearson Island expedition 1969, 7.
The subtidal ecology of benthic algae. Trans.
R. Soc. §. Aust. 95, 155-167.

Stracuan, A. R. & Kosxr, R. T. (1969) —A
survey of algue off Palos Verdes Point, Cali-
fornia, Caltf, Fish and Game 55, 47-52.

Vaux, D, (1970).—Surface temperature and
salinity for Australian waters 1961-65.
Si ‘S.LR.O. Aust. Div. Fish. Oceanog. Atlas
o. 1.

WHittagker, R. H. (1967),—Gradient analysis of
vegetation, Biol, Rev. 42, 207-264.

Womras.ey, H. B.S. & Epmonps, 8. J. (1952).—
Marine coastal zonation in southern Australia
in relation to a general scheme of classifica-
tion. J. Ecol. 40, ‘84-90.

Womersiey, H. B. S. & Epmonos, 8. J. (1958).—
A general account of the intertidal ecology of
South Australian Coasts. Aust, J. mar. freshw,
Res. 9, 217-260,

BREEDING BIOLOGY AND LARVAL DEVELOPMENT OF
LITORIA VERREAUXI (ANURA: HYLIDAE)

BY MARION ANSTIS**

Summary

ANSTIS, M. (1976).-Breeding biology and larval development of Litoria verreauxi (Anura:
Hylidae). Trans. R. Soc. S. Aust. 100(4), 193-202, 30 November, 1976.

Oviposition and larval development of the hylid frog Litoria verreauxi are described and ecological
notes are given. Comparisons are made with other hylid frogs (particularly members of the Litoria
ewingi complex).

BREEDING BIOLOGY AND LARVAL DEVELOPMENT OF
LITORIA VERREAUXI (ANURA: HYLIDAE)

by Marton ANSTIS*

Summary
Awnstis, M, (1976),—Breeding biology and larval development of Litoria verreauxi (Anura:
Hylidae). Trans. R. Soc. §. Aust. 100(4), 193-202, 30 November, 1976.
Oviposition and larval development of the hylid frog Litoria verreauxi ure described and
ecological notes are given. Comparisons are: made with other hylid frogs (particularly members

of the Litoria ewingi complex)...

Introduction

Litoria verreauxi (Duméril), previously in-
cluded in Hyla ewingi Duméril & Bibron (see
Littlejohn 1963, 1965; Tyler 1971) is a hylid
frog found along the coast of eastern Australia
from Victoria to southern Queensland (Little-
john 1965; Straughan 1966)+. Adult morpho-
logy in the Sydney areca has been described by
Copland (1957) as H. ewingi verreauxi, and
by Moore (1961} as HA. ewingi. Fletcher
(1889) and Harnsen (1922) provided some
data on the breeding season, ova and larvae,
while Moore (1961) briefly described advanced
embryos and larvae. Martin (1965) described
tadpoles from the Melbourne area bul did not
diseuss embryonic development. Martin &
Watson (1971) mention some life history
characteristics. The present paper provides data
on breeding biology and larval ecology and
includes a detailed description of embryos and
larvac.

L, verreauxi appeat'’s to be related to a com-
plex of species including L. ewing?, L. parae-
wingi, G. jervisiensis, and possibly Z. Burrawsl
(Martin & Littlejohn 1966; Martin 1967a;
Watson, Loftus-Hills & Littlejohn 1971).
Where data arc available, comparisons ure
made with these taxa.

Material
Six egg masses of ZL, verreanxi laid in the
laboratory, together with samples of larval
material from the field, form dhe basis of the
study. Egg masses came from an adult popu-

lation, originally collected at Darke’s Forest in
1970 and released in a garden at Penshurst.
Frogs from adjacent areas in Penshurst may
also have joined the population.

An egg mass from a pair of ZL. ewingi cap-
tured in amplexus at Lobethal, 5. Aust. on
30.vili.1972, was maintained to hatching stages.
Larvae of L. paraewingi from 2 km N of Glen-
burn, Vict. were examined for comparison.
Collecting loculitics and dates are listed in
Table 1.

Methods

A series of outdoor aquaria containing rain-
water and vegetation was maintained at Pens-
hurst and checked regularly for the presence
of spawn. Three pairs (one in amplexus) were
captured in the vicinity of the aquaria (two on
11.ix.1972 and one on 20.11.1974) and placed
in plastic bags containing water, twigs and
vegetation, Oviposition behaviour of these three
pairs was studied.

Embryos were maintained up to stage 25 in
shallow water ranging from 14°-21°C. Larvae
from the various localities were maintained
separately in open outdoor aquaria, and indi-
viduals from some were reared to metamor-
phosis. The behaviour of laryac was studied
both in aquaria and at field collecting sites.
Food provided consisted of algac and other
water plants, commercial fish food, boiled let-
tuce and occasionally meat. Water temperature
during larval development ranged 8°—27°C.
Specimens from each group were fixed at inter-

* 630 King George’s Road, Penshurst, N.S.W, 2222.

t Straughan, I. R. {1966)—An analysis of species recognition and species isolation in certain Queens-
lattd frogs. Ph.D. thesis, University of Queensland (t:npuhl.).

194 MARION ANSTIS
TABLE 1
Breeding sites of Litona verreauxi
Collecting
Locahty Description of habitat date Stages Other Iatvae present
Menai, 1, Permanent dam in dry sclerophyll bush- 21.11.1971 34-42 Liroria aurea
34°02'S land. Surface vegetation, rooted plants, L, latopalmata
151°01" EB mud substratum. Uperoleia marmoraja
Ranidella signifera
2, Concrete water vessel, permanent water, 16.ix.1972 34-41
surface vegetation, mud substratum
Penshurst, Permanent ouvidoor aquaria jn suburban gar- Numerous 1-46
33°58'S den, Surface and rooted plants dates, 1970
151°05' E to 1974
Darke’s Forest, 1, Permanent flowing stream, sandstone base, 16.ix.1972 16-18 Liforia jervisiensis
34°12'S fast. flowing sections, deep pocls in dry
151°S8"R sclerophyll bushland,
2. Permanent dams, liitle rooted and no sur- 24.in.1972 26-40 Limnodynastes peroni
face vegetation, mud substratum 30. x.1972 Litoria perori
2,x1,1972 Ranidella signifera
6.A7.1972
Qurimboh, Semi-permanent, small, slowly flowing creek, 9.in1973 25-33 Ruanidella sienifera
33°22°S shallow pools, rooted vegetation, mud sub-
151°22"E stratum. Cleared farmland in wét Sclerophyll
forest
Glen Alice, Semi-permanent, shallow pond, grass but- 1vi.l974 25-28 Limnodynastes
33°02’ 5 tom, in open cleared farmland with sur- tasnaniensis
151°12’E rounding woodland
‘Spring Creek, Permanent creck, slowly fowing small 25,1.1973 30-46 Mixephyes balbus
30°29°S§ pools, sandy and basalt substratum. Wet 25.07.1973 25-42 Ranidella signifera
152°24" B sclerophyll forest, paruy cleared 18.iv.1973 Litoria glandulosa
L, pearsoni
Dorrigo, Small, slowly flowing creek, surface vegeta- 26.41.1974 28-43 Mixophyes fasciolatus
30°20 S tion, mud substratum. Cleared rainforest Adelotuy brevis
182°43°E farmland
Rouse Hill, Permanent waterhole in cleated paddock, 19.x311972 27-42 = Liloviavaeruiva
33°42'S Dry sclerophyll bushland area, farmland Ranidella signifera
150°55°E

Fig. 1. Lateral and dorsal views of farya showing
measurements for morphometric charac-
ters,

vals in 4% formalin, after being relaxed in 1%
chlorbutol solution; larger specimens were
injected with a small quantity of formalin
before final fixation.

Measurements were taken with verier cal-
lipers reading to 0.1 mm or an ocular micto-
meter (reading to 0.01 mm). Drawings were
made using a drawing tube attached to a
stereoscopic microscope. All measuzements and
drawings are based on preserved specimens,
while descriptions are of both preserved and
live material. The staging system used is that of
Gosner (1960). Abbreviations and definitions
of latval morphometric characters (Fig. 1)
are: ST—total length (tip of snout to tip of
tail); BL—body length (tip of snout to junc-
tion of body wall and tail musculature); BW—
maximum body width; BD—maximum body
depth; TD—maximum tail depth; TM—depth
of tail musculature (measured in line with
TD): 10—inter-orbital span (minimum dis-
tance between the cyes, measured at the central
inner edge of each eye); [N—internarial spazt

BREEDING BIOLOGY OF LITORIA VERREAUXI 19s

(minimum destance from eye to avis); EN—
distance from eye to narts; MW—maximum
width of oral disc,

Results

Calling aetiviry: The mating call has been des-
cribed by Littlejohn (1965), Males alt Pens.
hurst call throughout the year, with the most
intense activity on mild, wet nights during
spring and summer. Diurnal calling mostly
occurs during and after rain. Males call while
afloat near the edge of ponds by night, or from
low vegetation or ground near the water by
night or day. At 2300 hrs on 20.11.1974 at
Penshurst, during light rain, a silent male sur-
faced in an aquarium gbout 4 cm from a cuall-
ing male. The latter turncd to face the former
and, after a brief pause, swam slowly towards
him, calling in softer, separate totes (quite dis-
tinct [rom the mating call) and attempted
amplexus. The silent male immediately swam
of, The calling male did not follow, but
resumed a normal mating call.

A similar behavioural sequence preceded
ampleaus in one of the pairs captured on
11.ix.1972, the male emitting soft. separate
notes as he approached the female,
Ovipesition: Oviposition at Penshurst has been
observed in February, March, June and Sep-
tember—December, The following description is
a composite of observations of the three pairs
studied.

When frogs were collected on 11.x.1972, air
teniperatures 2 cm above water were {8°-]9°C
and surface water temperatures 19°—23°C.
Amplexus commenced in these pairs at 2000
and 2325 hrs, Eggs were laid in scparate
hatches attached to twigs or reeds over a period
of hours (Tuble 2). Before oviposition, the
female showed lateral abdominul contractions,
either simultancously or alternately. These con-
traciions usually became more powerful as ovi-
position was near and lasted about one second,
with two or more occurring In succession.

In a typical behavioural sequence, a pair
submerged and the female grasped a twig with
one hand. She dorsiflexed her body with the
hind limbs extended and, as the batch emerged,

Fig. 2a. Oviposition with the miale receiving, und
fertilising the eggs.

Fig. 2b, The mate pushes the batch down to the
female's feet.

the male lowered his vent towards the eggs
and cupped his feed areund, so holding them
(Fig. 2a). The sides of the male then undu-
fated und his feet moved up and down in a
bref fanning motion over the eggs. This pro-
cess of oviposition and fertilisation lasted 3 sec,
The female ventriflexed, drawing her legs back
under her hody, and the male rolled the batch
down to her feet (Fig 2b). The female held
the hatch motionless for 40 sec. She then
pulled herself around the twig in spiral fashion,
wrapping the eggs round it with her fect. The
pair left the eges and returned to the surface.
After 1.5-7.3 min. the entire process was
repeated, and S min.—2 he clapsed before fur-
ther batches were laid.

Variations were: (1) Nearing the end of
amplexus, two or three batches were laid in
very close succession, each being held by the
feet of the feinale for 40-60 sec. before the
ensuing one was laid, The resulting composite
batch was then attached to supporting material.
(2) Females varied in their attempts to spread

TABLE 2
Ovipovition bohtavionur
Duration of
‘Total cyiragion Dirratian at ege- single batch Batch holding, Total eges
Paic of Ainplexas laying period oviposition time (female) n laid
Ta 5 hr 15 min, 2 hr 8 min. Tt j0, 35-60 sec, 15 757
Z Unknown 3 hr 47 min. 2-4 sec. 35-60 sec. Bors 101i

196

batch one
laid and held
by femate for
40 seconds

single batch
laid and held
by female for
40 seconds

OR

batch 2 laid,
batches 142
held together
for 40 seconds

eggs attached
to vegetation,
adults return

to surface

batch 3 faid,
batches 1,243
held together
for 40 seconds

Fig. 3. Oviposition cycles during which @ single
batch is tald and attached to vegetation,
or two or three are Juid in close succes-
sion before attachment,

ott the eggs in spiral fashion, sometiines
swivelling around the twig only once or not at
all, resulting in thicker clumps of eggs. (3)
One female used her left hand to grasp and
pull free some eggs which had adhered to her
venter, before attaching the batch to vegeta-
tion,

Females had more difficulty in wrapping a
composite batch around a twig, and offen aban-
doned the egexs as a thick mass. In aquaria
lacking vegetation or twigs, egg masses have
been found in thick clumps on the substratum,
in water up to 50. cm deep,

The final stages of armplexus in one pair
were: at 0100 hrs the female made move-
ments similar to the croaking motions of males,
but produced no sound, At O108 hrs she
submerged and both male and female began
typical ovipositional behaviour, but the female
remained in the dorsiflexed position for 7.7 sec,
(4.7 sec, longer than average) and produced
no eggs. Two sec, Jater the pair fell apari, both
floating motionless on their sides just under
the surface, with limbs tightly adpressed
against the body. After LO sec, the male

MARION ANSTIS

Fig. 4, Two batches of eves joined and attached
io a stem, Filamentous algae are catwined
amongst the egg mass.

recovered from this state of suspension and
surfaced, the female doing so 5 sec later. A
second pair behaved similarly, except that the
period of motionless suspension was shorter.
The basic cycle of oviposition behaviour is
shown in Fig. 3. Laying of all eggs conyprises
a number of such cycles.

Qva: In natural environments egg masses are
altached to submerged reeds, twigs or grasses
usually close to the surface (Littlejohn 1963),
The cgys cohcre and the inner ones stick ta
the supporting material. There is a single layer
of jelly around each egg, but within « roass the
individual capsules merge and are not clearly
defined (Fig. 4).

The mean diameters of eggs and capsules in
stages 1 and § are shown in ‘Table 3. Ova
generally have a dark brown animal pole and
an off-white, yellow or orange Vegetal pole, All
ova from a single female arc the same colour.
The animal pole gradually lightens from gas-
trulation onwards.

The number of eggs in 20 single batches
ranged from 1—52 (mean 30). Three “double”
batches contained 64, 78 and 79. The totul
complements of four females were 1,011, 757,
632 and 522.

Development of embryos; After Fertilisation
there is no distince grey crescent. Cell division
appears normal, although not as symmetrical
as in Gosner’s (1960) diagrams. The vegetul
pole always divides later than the animal pole.
At stage 17 (tail bud: Fig. 5a), the head
region is well defined, showing optic bulges, gill
plates, U-shaped adhesive organ and a slight
stamudaeal pit. The posterior crescent of the
adhesive organ is less distinct. Tn some embryos

BREEDING BIOLOGY OF LITORIA VERREAUNI 197

the visceral arches and a slight pronephric
bulge are discernable. The tail bud is straight
anu points dorsally, with no obvious tail fin
rudiment, In late stage 17, just before muscular
movement begins, the tail bud extends and
points cither fo the right or to the left, and the
posterior crescent of the adhesive organ almost
disappears, yielding two separate organs which
ute heavily pigmented. Embryos in stages 17
to 20 have a yellow yolk sac and are light
brown elsewhere.

The embryos begin hatching when. they have
reached stages 19 and 20, At stage 20 (Fig, 5b)
the gills are small, just functional and non-
pigmented. The optic bulges are more defined,
and there is a small crescent of melanophores
around the anterior edge of each, The stomo-
dacal pit bas deepened and the adhesive organs
are prominent. The yolk sac has elongated and
18 wenerally narrow, and there are small areas
of pigment along its dorsal edge, and between
the optic bulge and olfactory pit, The arca
above the olfactury pit. fy clearing and the tail
fing aré a translucent milky white.

With the temperature regime prevailing
during carly development, hatching was com-
plete after 147 hr when most embryos were
in stages 21-23. The external gills are fully
developed in stage 2] (Fig. 5c). The tail fins
and cornea clear during stage 22; the oper-
culum partly covers the gills, and the distribu-
tion of melanophores increases over the yolk
sac, beneath the eyes, around the nares and
along the dorsal surface of the tail muscula-
ture. At stage 23 the gills are reduced, the
external nares are open, the stomodaeal pit
deepens further and the oesophagus begins to
differentiate. The anal tube is developing and
the fins, how transparent, take on their charac-
teristic arched shape. Generally, pigrnentation
increases, dispersing into the pattern typical of
the larva. The yolk sac is pale yellow beneath
the layer of melanophores, while other dorsal
and fateral areas surrounding the pigment, be-
come transparent. However one group of em-
bryos at this stage lacked dark pigment (except
for the eyes), and appeared yellow. These
embryos did not develop melanophores until
stage 25,

At stage 24 the mouth-parts have developed
oral ridges and a stnall non-keratinised beak,
the oral suckers have diminished, and the oper-
culum Closes on the right side. The anal tube
is partly open in some cmbryos, During stage
25 the formation of mouthparts is virtually
completed, the beak becoming keratinised and

TABLE 3
Dimensions in.mm of embryos and larvae of L.
verreauxi from Penshurst
(meuns, with canges in brackets)

Embryos
Embryo Capsule
Stege nh diam. diam.
i 10 i.23 43
{1.19-1.23) (3.53-4.92)
# 8 1,20 4.55
(1,15-1,23) (4.26-4.92)
9210 9 1.28 3.88
(1.23-1.39) (3.44 4.35)
14 9 1,58 4.28
(1.48-1,64) (3.74-4.929
15 y 4.74 441
(1.68) 1.80) (4.10-4,92)
7 1m Zt 4,39
(1,.85—2.4)) {3.61-6,40)
Embryea
Stare a diam.
20 10 5.83
(5,62-5.991
21 10 6,27
(6.76.44)
22 tt 640
(6.15-6.64)
23 id 6.98
(6.23-7.30)
24 9 7.26
(6.72-7.71)
25 it B45
(7.87-9,18)
Larvae
Stage n Body length Totablength
26 ju 10,16 33.6
(9.02-12,79) O83 ba)
27 10 Vth 24,2
(10:50-13,64) (21,2-27.2)
13 10 10.85 24,0
(9,68-11,91) (214-27,6)
a 9 114.16 24,5
(10.33-33,97) (24.4-272)
3b 10 12.88 29.)
CUE, U5-39,.78) (25.2-31.5)
3h 10 13,65 334
(12,30-15,42) €27,5-39.4)
2 7 13.40 305
(11,91-14.27T) (27.0-33,2)
WY s 14.31 32.6
{19.94-15.58) (31,0-44.0)
4 & 14,85 34.7
(13.02-15,74) (30,1-37.6)
Ws 16 16,65 4l.i
£14.76-19.68) (33.0-48.8)
46 10 16,15 41.4
(15.35-13.61) (34,447.23
a7 3 16.22 39,5
(14,92-17,22) (36.4-44.8)

198 MARION ANSTIS

Stage fn Bodylength Totallengih
38 6 16.84 432
(15.00-18.00) (39.6-46.0)
34 6 17.27 45.6
(16:73-18,32) — (42.0-51,9)
a it 16.87 46.6
(14,76-18,37) (39,2-52,2)
ay i 17.09 48,2
(16,56-18,20) (45,0-52.9)
4 3 16,13 43.1
(14.27.1738) (40.5-45,1)
a3 4 14.54 34.6
(14,.27-14,92) —-137.4-39.7)
45 1 16.3 —
(14,9-18.6)
46 la 15.3 =>
{13,.2-17.5)

lablal teeth developing on the oral ridges. The
labial papillae may not reach their total num-
ber until stage 26 or later. The spiracle be-
comes functional und the anal tube is fully
open. The remnant adhesive organs praduully
disappear during this stage.

Measurements of embryos are shown in
Table 3,

Larvee: A composite description of 10 larvae
at stage 35 (Figs 5d-f) from Penshurst fol-
Jows; Body widest across the mid region of the
abdomen and ovoid. Snout evenly rounded fn
dorsal view and tapers to a truncate edge in
lateral view.. Nates dorsal and raised on very
short tubes which open antero-laterally. Eyes
lateral and relatively large, Spiracle sinistral,
ventrolateral and not visible from above, Tt
opens in a porstero-dorsal direction and dia-
meter of the spiraculat tube decreases slightly
from its origin 10 its opening, Anal tube dex-
tral, very short, of small diameter and opens
about halfway up the ventral fin. Tail fins
arched and taper to a fine point. Dorsal fin
extends midway up the body, deepest approxi-
mately halfway along its length, Ventral fin
deepest along its anterior third. Tail musoula-
ture moderately thick, narrowing to a fine point
posteriorly.

Mouth antero-ventral in position and has
border of papillae around all but the anterior
margin (Fig. 6). In some specimens. there is
also a median gap along the posterior murgin
(possibly caused by damage). Papillae most
numerous laterally. Two upper and three lower
rows of Jabial teeth, two upper being of
approximately equal length in most specimens,
First two rows in the lower fabiim are also
about equal, third lower row is usually the

Fig, 5. Embryological and larval development of
Litoria verreauxi, Penshurst, (Bar rep.
resents 1 mm). Stages: 2—i7, b—20, c—
21, d—36, e—1h, f—46.

shortest. In same specimens a partial median
gap occurs in secand lower row and other rows
may be interrupted at various points, probably
through damage. Beaks of moderate propor-
tions, serrations fine on inner edge of lower
beak and very fine on the upper beuk.

The only consistent geographic variation
noted Was In specimens from Spring Creck,
most of which had more massive beaks and
two pigmented areas below the lower beak
(Fig. 6b). Specimens from Dorrigo also
showed a tendency towards more massive
beaks. It was noted that specimens from the
northern localities generally had shallower fins
than most southern specimens (Table 4), Body
dimensions of larvae are given in ‘Table 3,

In hfe the dorsal surface varics amongst
individuals from light golden to a very dark

BREEDING BIOLOGY OF LITORIA VERREACXI 199

Fig. 6. Mouthparts of Z. verreauxt. a, {tom the
southern site cf Penshurst; bh, from the
northern site of Spring Creek (bar rep-
resents | mm).

brown. (almost black). In some specimens the
pigment ts motiled, The areas of skin over the
trabeculae cormba, central nervous system
{brain und spinal cord to base of tail). the
abdumern ond surrounding the nares, are
darker, There is a copper-gold sheen ventrally
and laterally over the abdomen. In lateral view
the areas covering the pharynx and buccal
cavity (excluding eyes) are transparent (except
for some melanophores héetween the eye and
mauris}, and the pills, heart and developing fore-
timbs are visible. From the ventral aspect the
areas over the gills, heart and buccal cavity are
unpigmented.

The tail musculature ig cream with irregular
dark blotches over the dorsal surface, and
partly over the lateral surface. In generally
darker larvae the musculnture may be uni-
formly pigmented. The dorsal and ventral fins

TABLE 4
Propartions.in mar af L. verreauxi larvae from
different localities

(means, with ranges im brackets)

Narthern Southern
(Spring Creek, Dolnee) ‘Penshurst)
Stage 33 & Sh 45 &36
q 7 10
ST 34.1 431
(3u9412) (36.2-48.8)
BL 11.69 17,27
412,4G-18.56) (18,53-19,68)
BW 774 1u.26
(6.72-R.36) (8,69.717.48)
BD 779 W507
16,56-4.95) (8.86-12,14)
TD TSB 119
16.48.63) (8.59- 7230)
mM 2.68 3.60
€2.13-3.28) (2,79-4,59)
{Q 3,62 5.53
€3.44-4.55) (4.66 6.40)
IN 2.05 2.64
(1.72-2.21) (2.46-2,95)
EN 228 2.65
(2.13-2.69) (2.38-3,12]
MW 2.03 4.10
(2,95=4 43) (3,6]-4.66)

wary from dusky (in dark larvae) to aimost
transparent (lighter larvac), with parts of the
tail vascular system pigmented, Larvae with
mottled pigmentation over the body also have
mottled tails. The iris. is golden.

Specimens which were dark in life may
fetain much of this pigment in preservative,
Those which were light golden hecome an off-
white colour in all but the darker areas. and
the skin is clearer than in fife. The copper-gold
sheen is lost and the abdomen may appear
dark shiny blue fur some time in preservative,
then eventually turn black. The iris loses its
golden colour and also appears: black.

Lurval hehaviewr> After hatching the embryos
remain close to the egg capsules until about
stage 24. Durning stages 25 to about 27, the
larvac are most often found in the shallow
areas of ponds, particularly near the edge, but
beyond this stage a much greater water space
is utilized.

The larvae are of the active, nektonic type
{Orton 1953) and spend much of their time
hovering in the water by rapidly oscillating the
Jail tip (flagellum). They frequently cruise
slowly to the surface with head uppermost jt
ahout a 45° angle, using only the flagellum for
propulsion. When feeding at ihe surface, they

20

often position themselves almost vertically and
can remain suspended at this, or any level in
the water. They arc capable of sudden spurla
of speed (during which they may use the entire
tall and body), and rapid changes of direction
(making use of the deep fins), when disturbed,
As well as feeding at the surface, the larvae
Eruge on vegetation and other material in any
zone of the pond and scavenge: in bottom sedi-
ments. The variation in larval pigmentation
appears to be related to characteristics of the
habitat, Specimens in muddy water, or clear
water over a dark substratum, usually range
from dusky brown to almost black, while
those in clear water aver a light snbstratum
tend to be golden, with the darker areas con-
trasting, but less pronounced,

Larvel life span and meramerphosis; Metamor-
phosis of larvae reared from cges laid sat
Penshurst on 11.ix.1972 began on 10.xti.1972,
giving a spring-summer larval life span of 90
days. Metamorphosis of larvae from egg
masses laid on 23,%,1971 occurred from Jate
December to early March. Metamorphosis was
also recorded at Penshurst fn September 1972
and at Menai from 27-29.iK.1972, It is there-
fore known to occur from September—March,
but probably takes place at other times because
ege masses haye been found jn most months
nf the year.

The body fengths of 10 juveniles at stage
45. and 18 at stage 46 are shown in Table 3.
Al these stages the juveniles closely resemble
the adults in colour, but lack the deep orange
of the anterior and posterior surfaces of the
thigh, and the black spots in the groin. Pale
orange thigh colouration is visible in some
juveniles at stage 46,

Discussion.

Calling aciivity: Fletcher (1889) and Harrison
(1922) moted that calling occurs throughout
the year, and Moore (1961) observed calling
activity from the end of July 1992 to late April
1953. Watson ef al (1971) record calling acti-
vity in all months except July and found that
1, verreauxi males when sympatric with L.
ewitel usually call on Jand up to 25 m from
water, and only rarely in water. This latter be-
haviour contrasts with that of males at Pens-
hurst and Darke’s Forest which commonly call
in water.

A call distinct from the mating call, given by
the male on approaching a potential rival or
mate, has been observed; its function is not
known, More abservalions aré necessary to

MARION ANSTIS

establish the extent of behavioural variation iu
this species. A similar call has been observed
in L, ewfng! (Anstis 1976).

Oviposition: Harrison (1922) found spawn in
Sydney every month of the year, and Moore
(1961) collected embryos in August, 1952 at
Killafa. Fletcher (1889) found a pair in
amplexus in June, 1885 and stated that the
species “probahly breeds nearly throughout the
year", This agrees with the oviposition dates
recorded at Penshurst,

Oviposition has been observed in few Aus-
tralian hylids, Watson er al, (1971) described
part of the behaviour associated with egg-lay-
ing in Litorict paraewingt, and | have observed
oviposition in L, citropa, L. wentate, L. frey.
cineti and L. glaserti. Some of the ovipositional
patterns in L, verreauxi ure unique, notably
the action of the male pushing the cluich down
to the fect of the female where the eggs are
held motionless fora short period,

The behaviour of the male in cupping his
feet around the batch and rapidly “fanning” the
eggs may serve to distobute the seminal fluid
around the eggs within a more confined space
and thus aid fertilisation. A similar although
somewhat bnefer process occtirs in the ovi-
positional behaviour of L. citrope, L, déntata
and L, glaverti (Anstis unpubl,), BY holding
the harch still for a period of some seconds,
the female may also aid fertilisation in allowing
time for sperm penetration before the eggs
ure attached to supporting material,

[Lis pat known whether the abdominal con-
tractions in the female prior to egg-laying were
the sole factor in exiruding the eggs, or
whether the pressure exerted by the clasp of
the mije aided the process.

The attachment of eggs to vegetation in
a spiral movement has heen recorder! by
Harrison (1922) for L. verreaext (as Hl.
ewingi) and by Watson er al, (1971) for L.
paraewingi, Warrison's statement that the
female moved “right around the stalk at the
moment of laying" is not borne out by the pre-
sent study, but it is possible that Harrison did
not seethe entire vegtaying procedure. Watson
eral, (1971) state that a female of L, paree-
wingi observed in the field “held onto a sub-
merged grass stem, and pressed the cloaca to
the slem as the eegs were extruded; then the
pair pivoted around the stem while allaching
the eggs”. Such behaviour would appear tobe
similar to that of L, verreauxé except that in the
latter, the female holds the eggs still before

BREEDING BIOLOGY OF LITORIA VERREAUXI aut

alfachment and was not observed pressing the
cloaca 10 the stem during egg extrusion. In the
three oviposition sequences observed im this
species. the extent to which batches of eggs
Were spread around the supporting vevetation
varied, Observations have indicated that the
mortality rate of embryos t Jower in smaller
well-spread batches attached to a stem. Larger
masses of eggs on the bojtam of aquaria with-
out vegetation suffer high mortality from about
stage ¥Y onwards, possibly dwe to inadequate
oxygenation resulting from the thickness of the
ege mass and the depth of the water where
they lay. The attachment of two or three
batches together as one also tends tu increase
mortality, After death of an embryo, a fungus
develops over the egy capsule.

The manner of termination of amplexus
vanes amonust hylids, but often the last ovi-
positional sequence is longer than any other
and is followed by separation cither imme-
diately or a few seconds later, e.g. in Ayla
versicolor (Fouquette & Littlejohn 1960), and
Litoria dentara, £, glanerti and LL. citrapa
fAnstis unpubl.). L. verreauxi also follows this
pattern: however, the brief period of Lota? im-
mobility of both mule and female after separa-
tion has not been recorded in other species.
Ova: The significance of eggs being deposited
in small batches has been discussed by Pyburn
(1963) and Martin & Littlejohn = (1966)-
Harrison's (1922) observation that the eggs
are “attached in a cylindrical mass numbering
upwards of a hundred eggs to grass stalks und
similar submerged objects" 1s probably based
on cases where two or three hatches were
itlached as one,

The ovidiameter in stages 1-8 (1.21 mm)
is in agreement with Harrison’s figure of 1,2
mm. The ovidiarneter of £. ewingi has been
recorded as 1.65 mm (Martin & Littlejohn
1966) and that of L. ewingi and L. veereanst
as 1.7 mm (Martin, Littlejohn & Rawlinson
1966). A scrics of eges of L. ewényt laid in
Adclawte dwnng September 1972, have mean
diameters of 1.18 mm tat stage 1), 120 mm
(stage 5) and |.68 (stages 12-13); measure-
ments similar to embryos of L. verréanxi at the
Same stages (Table 3). I would seem likely
therefore that measurements by Martin ef af.
nay have been taken from embryos at about
Stages 11-13,

The tgesof 4,, paraewitgi are similar to those
of L. ewing) (Watson ef al. 1971). Those of L.
ferviriensis can readily be distinguished from
other members of the complex by the larger

ovidiameter (2.33 at stage 10: Martin & Little-
john 1966). Eges of £. burrowsi can be dis
tinguished from those of the L, ewingt group
by the presence of two jelly layers. around the
ovum. The ovidiameter of this species at stage
14 is close to that of L. jervisienss at the same
stage.
Ermbryes and larvae: The larvae of the L.
ewing? complex are of the common hylid type
(Martin 1967b) as is £, ferrewyi, The draw-
ings by Martin (1967a) of L. burrows! larvae
show u tail not as finely pointed and fins not
as Gdecp as in members of the £. ewirngi com-
plex, The body shape also appears somewhat
different. £. paraewiige larvae are similar to
those of L. ewingi “except ihat the tail fins
(especially the dorsal fin) . . . aré more heavily
pigmented” {Watson e¢ a/, 1971), Specimens
of this species examined are more uniformly
pigmented than L. verreauxi, and three speci-
mens ut stage 26 (mean total length 12.9 min;
body length 7,02 mm) are much smaller than
LL. verreauxi at the same stage. Such size differ-
ences may he relaicd to environmental factors.
The mouthparts of the group are basically
similar, having a formuls of

All have a median gap in the papillae on the
upper lip, the extent of which vartes amongst
individuals of the same species. The number
and size of the papillae is variahle between
species, those of LF. jervisieryis are more
Numerous and tightly grouped than in &,
verreaux?, while those of 1. paraewingi are a
little larger and Jess numerous, The larvae of L.
jervistensis possess larger, darker and more
massive beaks than 2. verreanx? and in a num-
ber of speeriens of the Former species From
Darke’s Forest, the central edge of the upper
heak curves slighly below the level of the rest
of the edge, unlike L, verreauxi. The two pig-
mented areas below the lower beak in L.
verreavixl from Spring Creck, are not found in
other members of the £, ewingi complex.

Larval behaviour ard adepration: All the larvae
of the f£.. ewig? group ate nektonic and
generally exhibit behaviour patterns similar to
those desctibed for 4. verreauxi, However, dif-
ferences occur in the larvae of L. jervisiensis
which have been observed schooling together
in groups of 20 or more in the mid-level of the
water, Individuals fram the group move at

22

different times (o the surface where they may
take air fAnstis, unpubl.). Larvae of L-
verveaux? were Iever observed congregating in
this manner.
Larval life span and metamorphosis: Data on
larval life span are mainly limited to specimens
in eaptive conditions. Moore (1961) records a
laboratory life spam of three months for EL.
verréaux’ Which agrees with one of the groups
fuised at Penshwrst. Harrison (1922) found
that larvae in aqnaria “required upwards of
three months” lo reach metamorphosis, but
believed seven to eight weeks to be normal life
span in ibe field during summer, This ts con-
siderably less than the approximate minimum
of 79 davs for one group in the present study,
but this difference may simply reflect different
culture and temperature conditions, Further
observations are fecessuty lo ascertain the
average life span of this species in the field,
Moore (1961) records the body lengths of
1! newly metamorphosed L. verreauad as 14.3-
17.00 mm: consistent with measurements of
specimens in the present study (Table 3).
Martin (1965) gives a range 11,J-13.6 mm

MARION ANSTIS

for newly metamorphosed L. swinger, which are
generally smaller than L, verreauxi, and Martin
& Littlejohn (1966) 15.6-19.7 mm for L. jervis-
iensis, No data on B, furrewsi and L. parae-
Wing: are available.

The overall life cycle of L. verreanal appears
quite similar ta that of other members of (he
£. ewingi complex in the adaptations to still
water situations, although L. jervisiensiy differs
noliceably in the details of its life history
(Martin & Littlejahn 1966). More data are
necessary before useful comparisons can be
made between the life histories of £. burrowsi
and the £. ewing! complex.

Acknowledgments

I gratefully acknowledge the receipt of a
Tesearch grant from the C.S.J-R.O. Science &
Industry Endowment Fund, which has greatly
facilitated this study. De Harold Cogver, Mr
Harry Ehmann, Dr Norman Gradwell, Dr
Angus Martin and Me Michael Tyler tread and
constructively criticised the manuseript. Thanks
are also due to Mr and Mrs D. Anstis for
assistance im various ways.

References

Ansris, M. {1976),—Courtship bebaviour in the
Australian tree frog Litoria ewinei ¢Anura:
Hylidae). Merpetofauna (1). 16,

Carianin S, J. (1957),—Australian teee frogs of
tbe genus Ayla. Proe. Linn, Soc. NSW. 82,

~108,

Fuercien, J, F. (1889)—Observations on the
oviposition and habits of certain Australian
Batrachians. Prac: Linn. Soc. N.S.W’, 4, 357-

387,

Fougurtre, M. §., Jr & Littntyonn, M, J-
(1960),—Patterns of oviposition in two
speed of hylid frops. SW. Naturalist 5(2),

-96.

Gosner, K, L. (1960)—A simplified table for
staging anuran embryos and Jarvae with notes
on identification. Herpetofagiva, 16, 183-190,

GRADWELL, N. (1972) —Gill irrigation in Rana
cdteshelana. Pt. 1, On the anatomical basis.
Cun. £. Zoael, 50, 481-499,

Harerson, T,, (1922).—On the breeding habits of
some Australian frogs, Aust. Zoal. 3, 17-34.

Citriesoun, M, J. (1963).—Frogs of the Mel-
bourne area. Mier. Nat. 79, 296-304,

LITTLEJOHN, M. J. (1965).—Premuating isolation
in the WWyla ewingl complex (Anurca:
Hylidae). Evolution 19, 234-243.

Manrin, A. A, (1965)—Tadpoles of the Mel-
bourne area, Tier. Nef. 82, 139-149.

MarTIN, A. A. (1967a).—The exrly development.
of Tasmania’s endemic Anura, with comments
on their relationships. Prec. Linn. Soc. N.S.W.
92, LO7-116,

Marnn, A. A, (1967b}.—Australian anuran life
histories: some evolutionary and ecological
aspects. In A, H. Weatherley (Ed.), “Austra-
lian Inland Waters and their Fauna’, (Aust.
Nat. Univ. Press; Canberra, )

Maatin, AW A. & Lirt.esonn;, MJ, (1966).—
The breeding biology and Jarval development
of Hyla jervisiensixs (Anura: Hylidae), Proc,
Linn, Sac. N.8.H'. 91, 45-57,

Maatin, A. A., Lirriegoun, M. J., & Raw iinsen,
P. A, (1966).—A key to the anuran eges of
the Melbourne area, and un iddition to the
anuran fauna. Vier. Nat. 83, 312-315

Marin, A. A. & Watson, G, F. (1971),—Life
history as an aid to generic delimitation in
the family Hylidae. Cepeia, 1971(1), 78-89.

Moore, J. A. (1961).—The frogs of eastern New
South Wales, Bull, dm. Mux, Nat, Afst, 124,
149-386.

Orton, G, L, (1953).—The systematics of verte-
brate larvae. Syst. Zool. 2(2), 63-75.

Pysunn, W. FP. (1963}—-Observalions on the life
history of the tecefrog, Ployllomedusa calli-
dryas (Cope). Texas J. Sei. 15, (S5-170,

Tyree, M, J, (19713,—The phylogenetic signiti-
cance of Vocal sac structure in hylid frogs.
Brin Ransus Publ. Max. Nut. Mist. 19, 319-
360.

Watson, G. FB. Lorrus-Hitus, J, J. & brrerur-
gous, M. J, €1971).—The Litera ewinel
complex (Anura: Hylidae) in south-castern
Australin 1. A new species from Victoria.
Awst, J. Zool, 19, 401-416,

RESTRICTION OF THE CHIRIDOTID GENUS TROCHODOTA LUDWIG
(1891) (HOLOTHURIOIDEA: APODIDA), WITH THE DESCRIPTION OF A
NEW SPECIES FROM SOUTH AUSTRALIA

BY F. W. E. ROWE*

Summary

ROWE, F. W. E. ( 1976) .-Restriction of the chiridotid genus Trochodota Ludwig (1891)
(Holothurioidea: Apodida), with the description of a new species from South Australia. Trans. R.
Soc. §. Aust. 100(4), 203-206, 30 November, 1976.

Trochodota Ludwig (1891) is restricted to the type-species T. purpurea (Lesson), and three other
species, including a new species from South Australia. The generic significance of scattered or
heaped wheels, used in separating Trochodota from Taeniogyrus Semper, 1868, is disputed. The
distribution of serrations on the inner margin of the wheels is regarded a more reliable generic
character; on this basis seven species included in Trochodota by H. L. Clark (1921) and subsequent
authors are referred to Taeniogyrus.

RESTRICTION OF THE CHIRIDOTID GENUS TROCHODOTA LUDWIG (1891)
(HOLOTHURIOIDEA: APODIDA), WITH THE DESCRIPTION OF A NEW
SPECIES FROM SOUTH AUSTRALIA

by F. W. E. Rowe*

Summary

Rows. F. W. E. (1976)—Restriction of the chiridotid genus Trocliidota Ludwig (1891)
(Holotharicidea: Apodida), with the description of a new species from South Australia,
Trans. R. Soe, S$, Aust, 100(4), 203-206, 30 November, 1976,

Trochodota Ludwig (1891) is restricted to the type-species J. purpurea (Lesson), and
three other species, including a new species from South Australia, The generic significance of
scattered or heaped wheels, used in separating Trovhodota from Taentogyrns Semper, 1868,
is disputed, The distribuiion of serrations on the inner margin of the wheels is regarded a
more reliable generic character; on this basis Seven species included in Trochodumm by H. Li
Clark (1921) and subsequent authors are referred to Tdenlogyrus,

Introduction

Among the holothurians collected at Port
Lincoln, South Australia during 1975 by Mr
S, A. Shepherd are six belonging to an undes-
embed species congeneric with Trochodota pur-
puree (Lesson), type-species of Trochodota
Ludwig (1891).

In one important character, the new species
falls into an intermediate position between the
genera Trochxlota and Taeniogyrus Semper
(1868), as currently diagnosed (H. L. Clark
1921; Pawson 1964), A review of the two
genera has revealed that they are not based
upon reliable characters; in this paper they are
tedescribed and a list of species given for cach,

It is not appropriate to discuss in this paper
the validity of all species now included in
Taéniogyrus, since maby species require re-
examination and material is not available:
H. L. Clark (1921); Pawson (1964). However,
the differences between the four well docu-
mented species in Trechedota, including the
new species described below, are tabulated.

Taxonomic account
H. L. Clark (1921), revising the chiridotid
genera, separated Tueniogyrus Semper (1868)
and Trochedote Ludwig (1891) from other
genera, because they possess a combination of
wheel and sigmoid ossicles. On the basis of
having the wheel ossicles actually collected into

sharply defined papillae of the body wall,
Taeniogyrus was considered generically distinct
from Trochodota, Small accumulations of
wheels were considered indicative of T'rocho-
dota. Subsequent authors have rigidly adhered
ta this recognition of the two genera (A. M.
Clark 1966; Hickman 1962; Pawson 1964,
1970; Heding (928; Cherbonnier 1952).
Although several new species of Taeniogyrus
have been described since 1921, no new species
of Trechedote have been found,

With the arrangement of wheels in large
groups, though not in papillae, the new species
described below falls into an intermediate posi-
tion between Taeniogyrus and Trechodota. In
my view, this. shows the unreliability of using
such a character for generic distinctions, par-
ticularly when H. L. Clark (1921) used the
similar grouping of sigmoid ossicles for species
duterminations, <A difficulty then arises in
deciding the relative merits of the importance
of wheels versus sigmoid ossicles in the recog-
nition of the generic taxa, for which no sound
argument has so far been advanced. One
eharacter which does so easily distinguish T.
purpurea, T. allani, T. maculata and the new
species nol only from those in Taentopyrus, is
the arrangement of the serrations on the inner
margin of the wheels. In the absence of any
other reliable internal or skeletal character, 1
believe that this is a much more significant

* Australian Museum, P.O. Box A285, Sydney South, N.S.W, 2001.

2M

character on which to place generic weight. ft
also accords with the use of spicule form in the
recognition of generic taxa within other orders
of holothurians (Panning 1949; Rowe 1969),

Tréchedota is herein restricted to that dis-
crete group of species with serrations of the
inner tim of the wheels arranged in groups.
The remaining seven species included by H. L.
Clark (1921) in Trochodota are referred to
Taerniogyras.

Trochodata is now considered to be
restricted to the southern hemisphere, with rep-
resentative species ranging from the colder
waters of the southern tip of South America
to the more temperate waters of southeastern
Australia, including ‘Tasmania, and the tropical
waters of the Torres Strait between Papua New
Guinea apd the northeastern tip of Australia.
The genus is found from shore-line to depths of
about 50 m.

Taeniopyrus Scmper, 1862

Taeniogyrus Semper 1868: 23.
?Sipmodota Studer 1876: 454,
?Trechodota Ludwig 1891: 358 (part).

Oiagnosts: Chiridotid genus with wheels and
sigmoid ossicles present, scattered, or in groups
or clustered into papillae; wheels with serra-
tions continuous around the inser margin: ten-
tasles LO or 12,

Type-species: Chiridota australianus Stimpson
1836,

Other species: T. contortus (Ludwig 1874); T.
cidaridis Oshima 1915; T, dubius H. L. Clark
1921, T_ ketensis Heding 1928; T. clarus
Heding 1928; T. dunedinensis (Parker 1881);
T. diasema (H. L. Clark 1921); T. roebucki
(Joshua 1914); 7. rosea (Oshima 1914); T,
japenica (von Marenzeller 1881); T. dendyi
(Mortensen 1925): T. dayi (Cherbonnier
1952); 27. venusta Semon 1887.

Trochodota Ludwig, 1891

Trochodota Ludwig 1891: 358 (part).
Diagnosis: Chiridotid genus with wheels and
sigmoid ossicles present scattered or in groups,
Wheels with serrations on the inner margin in
well defined groups; tentacles 10.
Type-species: Holothuria (Fixstulariu) purpurea
Lesson 1830.

Ocher species: T. allani (Joshua 1912); T,
maculata A. L. Clark 19215 7. shepherdi n. sp,

F. W. BE, ROWE

Figs 1-4. Travthodoia shepherdi mep. Fig.
Tentacle rods; Tig. 2—Two radia! and
two interradial plates of the calcareous
ring: Fig. 3—Wheel ossicle; Fig, 4—Sig-
moid ossicle,

1—

Trochodota shepherdi n, sp.

T. allani, Joshua & Creed, 1915: 2! (non 7.
allani Joshua).

Types: Holotype (Australian Museum J9467)
and 5 paratypes. J9796 (2) J9797 (1): Sauth
Australian Museum K1366 (2)); Proper Bay,
Port Lincoln, Spencer Gulf, S. Aust., among
algae growing on Pinna dalobrata at 10 m™
depth. Collected by S$. A. Shepherd,
23.vill.1975,

Diagnosis: Large spicules, wheels. 55-226 xm
diameter; serrations on inner margin of whecls
in G groups of about 16; sigmoid ossicles 144—
190 um Jong, outer curve of hook of sigmoid
ossicle with minute thorns; colour in hfe, black,
Description of holatype (Which has been cis-
sected): The holotype is 60 mm long, of which
the anterior 20 mm is contracted. Body dia-
meter 5-8 mm along its length, tapering only
at the posterior end, Ten strongly contracted
tentacles, each with about 3 pairs of digits.

NEW SPECIES OF TROCHODOTA 205
TABLE 1
Differences between species of Trocvhodata
Species Siemoid ossicles Wheel ossicles Colour Distribution
Grouping Hook Length Grouping Diameter
meculate in papillae and smooth 66-77 um scattered 50-100 wm = pink Torres Strait,
scattered with Qld
darker
spots
allani T ‘Scattered per- smooth 120-150 wm scattered singly 33-216 yum purple Port Phillip,
pendicular ta of at most io Vic—-S.E. Tas.
longitudinal small groups of
body axis 5-6
purpurea scattered smooth 125-150 um scattered 184-182 ym purple Southern
Ocean, Falk-
land Istand
southern coast
of South
America
ahepherdi scatiered per- swith 144-190 2m discrete, large 84216 am black Port Lincoln
pendicular to thorns groups (not in and Kangaroa
longitudinal papillae) 1, S. Aust.
body axis arranged uni
serially along
each inter-
radius

Spicules in tentacles comprise slightly curved
rods dichotomously branched at each end, and
usually have a series of thorny knobs projecting
laterally along shaft (Fig. 1).

Calcareous ring comprises 10 pieces fused,
with .@ straight ianterior and a slightly wndulat-
ing posterior margin, Radial pieces each have a
small anterior notch, Each piece of the ring is
1 mmx 0.5 mm (Fig. 2).

There is one ventral polian vesicle and the
dorsally placed madreporite ig very small and
hook-shaped (0.5 mm x 0.25 mm)-

Long. unbranched gonad on either side of
the dorsal mesentery, jojned anteriorly to a
single, dorsal gonoduct. Gonads extend for
about two-thirds of the body length and are
pucked with eggs.

Ciliated funnels numerous: on either side
ut base of dorsal mesentery in mid-dorsal inter-
radius in mid-interradial line of interradius ad-
jacent to, and to left of mid-dorsal intcrradius,
and in ventral interradius directly opposite to
those two dorsal interradii, Body wall trans-
lucent; radial muscles, lines of ciliated funnels,
and outline of internal organs can be seen
through it.

Calcarcous spicules of hody wall comprise
Wheels and sigmoid ossicles. Wheels restricted
to discrete groups but not accumulated into
papillae. These groups form a single line along

each of the interradii, except in the posterior
1/3 of body where the groups possess smaller
numbers of wheels, and form two irregular
lines. per interradius. Wheels have six spokes,
Inner margin of each wheel has six discrete
groups of about 16 serrations (Fig. 3). Wheels
are 35 pm-226 um in diameter. Sigmoid
ossicles evenly scattered throughout the body
wall and lic perpendicular to longitudinal axis
of body. Shaft of each is smooth except that on
the outside curve. at the attenuated hook end,
there are 2-4 minute spines or thorns (Fig, 4).
The sigmoid ossicles are 144-190 4m in length.
The animal im life is black. Besides the holo-
type 5 other specimens, similar in all respects,
Were collecled and these are considered as
paratypes.
Distribution: Port Lincoln, Spencer Gulf and
Kangaroo Island, South Australia.
Etymology: The species is named after the col-
lector, Mr Scoresby A. Shepherd.
Remarks. Differences between T. shepherdi, T-

maculata, T. allani and T. purpurea are listed
in Table 1.

Acknowledgments
I would like to thank Mr S, A, Shepherd
(Department of Fisheries, South Australia) for
forwarding material of the new species to me

206

for examination; Dr 8B. J. Smith, National
Museum of Victoria; and Mr W, Zcidler, South
Australian Museum, for allowing me to exa-

F, W. E, ROWE

mine remaining slides and specimens of T.
allani and 7. roebucki described or examined
by Joshua (1912, 1914),

References

CHERBONNIER, G. (1952),—Contribution a la con-
naisance des Holothuries de l'Afrique du Sud.
Trans. R. Soc. §, Afr. 33(4), 469-509, pls
XXXV-L.

Crark, A. M. (1966).—Echinodermata, Port
Phillip Survey. Mem. natn. Mus. Vict. 27,
289-384, pls i-iv,

Ciark, H. L, (1921).—The cchinoderm fauna of
Torres Strait. Pap. Dep. mar. biol, Carnegie
Instn, Wash. 10, 1-223, pls 1-38.

Hepinc, S. G. (1928)—Synaptidae. Vidensk.
Meddr dansk naturh. Foren. $5, 105-323, 69
figs, pls 2-3.

HickmMan, V. V, (1962).—Tasmanian sea-cucum-
bers (Holothuroidea). Pap. Prec, R. Soc.
Tas. 96, 49-72, figs 1-177, 2 pls.

Josuua, E. C, (1912).—On a new holothurian of
the genus Taeniogyrus found in Port Phillip
Bay. Proc. R. Soc. Vict. (1.8.) 25, 79-81, pls
MAyV.

Josuua, E. C. (1914).—Victorian Holothuroidea
with descriptions of new species. Proc. R. Soc.
Viet. (ns.) 27(1), 1-1], pl. 1.

Josuua, E. C. & Creep, E. (1915).—South Aus-
tralian Holothuroidea, with descriptions of
new species. Trans. R. Soc. §. Aust, 39, 16-24,
3. pls.

Lupwic, H. (1889-1892), — Echinodermen
(Stachelhauter). 1. Buch. Die Seewalzen. Jn
Bronn, H, G. Klassen und Ordnungen des
Thier-Reiches. Leipzig 2(3), 1-447, 25 figs,
17 pls, 12 maps,

Mortensen, T. (1925).—Echinoderms from New
Zealand and the Auckland-Campbell Islands.
Hl Asteroidea, Holothurioidea and
Crinoidea. Vidensk. Medd. naturh. Foren.
Kbh. 79, 263-420, 70 figs, 3 pls.

PANNING, A. (1949).—Versuch einer Neuordnung
der Familie Cucumariidae (Holothuroidea,
Dendrockitota), Zool. Jb,, 18, 404-470, 62

gs.

Pawson, D. L. (1964).—The Holothuroidea col-
lected by the Royal Society Expedition to
Southern Chile, 1958-59. Pac. Sci. 18(4),
453-470, 3 figs.

Pawson, D, L. (1970).—The marine fauna of
New Zealand: Sea cucumbers (Echinoder-
mata: Holothuroidea). Bull. N.Z. Dep,
scient. ind. Rés. 201, 7-69, 10 figs, 2 pls.

Rowe, F, W. E. (1969).—A review of the family
Hofothuriidae (Holothurioidea: Aspidochiro-
tida). Bull. Br. Mus. nat. Hist. (Zool.) 18(4),
117-170, 21 figs,

SEMPER, C. (1868).—Die Holothurien. Reisen im
Archipel der Philippinen. Weisbaden 1, 228
pp., 60 pls.

ROYAL SOCIETY OF SOUTH AUSTRALIA INCORPORATED

Patron:

HIS EXCELLENCY THE GOVERNOR OF SOUTH AUSTRALIA
SIR MARK OLIPHANT, K.B.E.

OFFICERS FOR 1976-77

President:
B. P. WEBB, M.Sc.

Vice-Presidents:

J. J. H. SZENT-IVANY, Ph.D. Cc. R. TWIDALE, Ph.D.
Secretary: Treasurer:
B. J. COOPER, Ph.D. C. J. M. GLOVER, M.Sc.
Editor: Assistant Editor:
M. J. TYLER, M.Sc. W. V. PREISS, Ph.D.
Librarian Programme Secretary:
N. P. McCKELLAR-STEWART, J. K. LING, Ph.D.
B.Sc., Dip.Lib.

Minute Secretary:
S. A. SHEPHERD, B.A., LL.B.

Members of Council:

K. H. NORTHCOTE, B.Ag.Sc. P. M. THOMAS, M.Sc.
R. W. R. RUTLAND, Ph.D. J. S. WOMERSLEY, B.Sc.

Auditors:
Messrs MILNE, STEVENS, SEARCY & CO.

Registered for posting as a publication Category ‘C’

Printed by Graphic Services Pty Ltd, 516-618 Grand Junction Road, Northfleld, S.A. 6086