TIIE EVOLUTIONARY HISTORY AND DIVERSITY OF
AUSTRALIAN MAMMALS
MIcHAEL ARCHERT't.RlcK ARENAI, Ml\a BlssaRova.t, K{REN BLacKt,. JEi...tl BnlunaLl],
BERNARDCOoK-Etr, Fstl CneaSER', KIRqTEN CROSBY', ANNA.GILLESPIE', HENK GODTUELn',
MIRANDA GOTT''', SUZeNI,IE'J. HAND', BENJAN{fi\i KEAR', ALAN, KRIK}IANN', BRITN
iiiiiKnis'l
YUANQING
NEyILLEPLEDGE''',
l^riry-irr"-naumnrant,ANNEMussER',TRSY'MYER5',
.WANGO,
STEVEN WROE.*
Archer. M. et a\..1999. The evolutionaryhistory and diversityof Australianmammals.
Australian Mammalogy2l: 1-45.
Palaeodiversity and relationships of all groups of Australian mammals are reviewed. The
'Dark Ages'about which nothing is
fossil record spanningthis time is of variablequality.
known in terms of Australianmammal evolution include the late Triassicto late Jurassic,
late Cretaceousto late Paleoceneand middle Eoceneto middle Oligocene.Very little is
known about the early Cretaceousand late Miocene. The late Oligoceneto middle Miocene
record documentsthe highestlevels of biodiversity known for the continent, comparabie
to that which characterises the lowland rainforests of Borneo and Brazil. Order
Monotremata spans at least the last 110 million years and includes four families. The
enigmatic Ausktribosphenos from I 15 million-year-old sediments in Victoria may
..piesent an archaic monotreme,specialisedperamurid or previously undocumented order
of mammals but is unlikely to representa placentalas suggestedin the initial description.
Order Microbiotheria is representedin the early Eocene(-55 mya) by two generasimilar in
morphology to early Eocene taxa from Argentina. Order Peramelemorphiaspans the early
Eocine to Holocene and includes at least five families. Order Dasyuromorphia spans at
least the late Oligocene to Holocene and includes at least three families. Other
dasyuromorphian-like marsupials are indeterminate in terms of family-level affinities.
Order Notoryctemorphia spans the early Miocene to Holocene rvith one family. Order
Yalkaparidontia spans the late Oligocene to middle Miocene with one genus. Order
Diprotodontia spans the late Oligocene to Holocene,representedthroughout by three
*ujot groups: Phalangerida(eight families), Vombatomorphia (seven families) and
Macropodoidea (at least three families). A possibleplacental condylarth (Tingamarua) has
been iecorded from the early Eocene. An archaeonycteridid bat (Australonycteris) is
known from the early Eocene. Among bats, the late Oligocene to middle Miocene is
dominated by rhinolophoids, many of which have European,Asian and African affinities.
Mystacinids, megadermatids,hipposiderids and molossids are rvell-representedin the
Oligocene to Miocene deposits. Vespertilionids are uncommon in the Oligocene to
Miocene but become more diverse in the Pliocene to Holocene. Emballonurids and
rhinolophids appearfor the first time in the Plio-Pleistocene.Pteropodids are unknown
prior to the Holocene. Murids span the early Pliocene to Holocene. In the oldest
assemblageat Riversleigh. one undescribedlineage resemblesarchaic forms otherwise
only known from the fossil recordsof Africa and Eurasia.
tschool of Biotogicat Science, (tniversity ofNew South llales, Sydney 2052; 2Centrefor
Research into the Evolution of Australia's Terrestrial Ecosystems,Australian Museum, 6tschool of nuatural Resource sciences,
8 College St, Sydney, NSII 2000;
.oueensland
rEndangered
(,/niverity^ of Techiology, GPO Box 2434, Brisbane, Queensland 4001;
NSW
Species Unit, NSW National Parl<s and Wildlife Service, PO Box 1967, Hurstville,
5South Australian Museum, North Tnrrore, Adelaide, SA 5000; 6lnstitute of
iZZO'
Verteblate Paleontology and Paleoanthropology, Chinese Academv of Sciences'
PO Box 64j Beijing 100014, China. Manuscript received 2l October 1998; accepted 23
June 1999.
AUSTRALIANMAMMALOGY
MODERN understanding about the evolutionary
history of Australia's mammals has derived from
four main arenas of research: palaeontology,
comparative anatomy, cytology, and molecular
biology. While palaeontological and anatomical
studies have been pursued since the early part of
the 19th Century, neither cytology nor molecular
biology made major contributions to understanding about the relationships of Australian
mammals until the late 1970s (e.g.,Kirsch 1977).
Since then, molecular systematic studies i n
particular have played an increasingly significant
role in testing and refining understanding about
relationships at all levels from intraspecific to
19 86;
interordinal (e.g., Richardson et al.
Westerman et al. 1990; Springer et al. 1994; Messer
e t a l . 1 9 9 8 ) . I n a b u r s t o f a c t i v i t y s t a r t i n gi n 1 9 8 3 ,
fossil representativesof most families of living
Australian terrestrial, arboreal and volant mammals
have now been found, thereby providing minimal
ages of family-level differentiation and much new
information
about the interrelationships of
otherwise disparate groups.
It is hoped that what follows is a brief but
succinct overview of current understanding about
the fossil record and relationships of all families
of Australian terrestrial, arboreal and volant
mammals. As would be expectedin an increasingly
active field
research, much of
this
of
understanding is in the process of being tested
and somehas yet to be published. For this reason
and because of page-length restrictions, the
monophyly of most taxa in the cladogramshas not
been able to be rationalisedhere. Published and/or
ongoing studies that do rationalise these
hypotheses are indicated in the individuallv
authored subsections that follow.
Higher-level systematic nomenclature used
here. unless otheruise indicated, follows Aplin
and Archer (1987). Conceptsof biocorrelationand
gaps in the record of Australian Tertiary mammalbearing sediments follow Archer et al. (1997c).
Abbreviations used here are as follows: my.
million years; myo, million years old; mya or M4
million years ago; K4 thousand years ago; LF,
Local Fauna (: a contemporaneousassemblageof
creatures recovered from a single or closell
associated series of fossil localities); lfs, local
faunas; Fm, Formation (= a body of rock strata,
commonly of similar rock rype). Dental homology
in marsupials follows Luckeft (1993) for P3-M4
and Flower (1867) for ll-P2.
In the palaeodiversityfigures below, we have
amalgamated all taxa known from particular
intervalsof time as follows: early Cretaceous(taxa
spanning only the late early Cretaceousfrom I I 5
to 100 my); early Paleocene(only the interval 63
to 6l my and only in Argentina): early Eocene
(only one intervalat approximately 55 my); latest
Oligocene(only the latest late Oligocenefrom 26
to 23.3 my); early Miocene (all of it from 23.3 to
16.3 my); middle Miocene (all of it from 16.3 to
10.4 my); late Miocene (all of it from 10.4 to 5
my); early Pliocene(all of it from 5 to 3.3 my); late
Pliocene(all of it from 3.3 to 2 my): Pleistocene
(all of it from 2 my to 10,000yrs): Holocene(all of
the last 10,000yrs). Allocation ofparticular faunal
assemblagesfrom Riversleigh to these time
periods follows Archer et al. (1997c) and Creaser
(1997). Hence. seven morphologically distinct
Wroe and
dasyuromorphians indicated by
Muirhead (this work) for the eariy Miocene
representsminimum species diversity for this
group spanningthis total interval of time. In these
figures, taxa known from one or more of these
intervals are shown as occurring throughout these
intervals, despite the fact that in some cases they
may only be known from a single fossil locality.
This has been adopted because of current
uncertainty about the precise ages (rather than
'early Miocene' etc.) of each of the hundreds of
assemblages known. However, because mammal
speciescommonly last 3 to 8 million years,these
generalisations
are probably not unreasonable.
SYNOPSIS AND LIMTATIONS OF TT€ FOSSIL
RECORD OF AUSTRALIAN MAI,,[N4ALS
In 1983, in a 432 page treatise over-viewing
mammalian palaeofaunasof the world, Don Savage
and Don Russell fairly summarisedthe highlights
of the Australian Cainozoic record in five pages
(accompanied by an upside down map of
Australia!)" At that time there were about 75
Tertiary speciesdescribed.Subsequentgrowth in
Australian palaeomammalogy led to many new
discoveries u'hich have more than trebled that
level of understanding.Key events in the histor,
of Australianmammalsare summarisedin Table l.
Correlation and age of significant Tertiary fossil
mammal assemblagesin Australia are summarised
i n F i s u r e1 .
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
'Dark age 1'for Australian mammalsbetween220 to 115 mya; none are known despite being present in other
areasof the world as early as 220 mya.
Ausktribosphenosnyktos, -ll5 mya (e. Cret.), ?monotreme,peramurid or an unknown group (but almost certainly
not a placental),Flat Rocks, Vic.
Steropod'on galmani , Kollikodon ritchiei, ,-110 to 100 mya (e. Cret.),__monotypicmembers of the monotreme
and Kollikodontidae,Lightning Ridge, NSW.
families Steropodontidae
'Dark age 2' for Australian mammals from 100 to 55_mya.
lvlonotrimatumsudamericanum,63 to 6l mya (e. Paleoc.),ornithorhynchid monotreme,Argentina (Patagonia),
SAm (with oldest undoubted SAm marsupialsand placentals).
Marsupiali probably reachedAustralia from SA, via Ant.. sometimeb9twe9165 a{l-d.5.5mya.^ .
bariholomaii, microbiotheriids, other marsupials with SAm affinities, a ?dasyuromorphian.._
Thylaiotinga
perameloid, Tingamarra porterorum (possible placental) and Australonycteris clarkeae (archaeonycteridi
bat), -55 mya (e. Eoc.), Murgon, SE Qld.
'Dark ase 3' for Australian mammals extendsfrom 55 to 26 mya.
^
'spotty' archipela-9olinking SE Asia to
After fiial separation from Ant. between 45 and 38 my4 apart from a
Australia ihroughout Australia's northward drift, Australia's terrestrial mammals (but not bats) were isoiated
for 30 my resulting in a high level of endemism.
Most living iamilies oT Australianbats arrived (or evolved here) by 25 mya (1.Olig.); curiously, there is no preHolocene Australian record for pteropodids(fruit bats).
Between 24 and 15 mya, much if not all of central and northern Australia supportedrainforest.
Between26 and23 mya (1. Olig. to e. Mioc.). somefamily-levelgroupswent extinct(e.g.,ilariids).
Mammal diversity in forest communities of e. Mioc. age (e.g., at Riversleigh, northwes,ternQld) wgs very high,
comparable tb that found today in Bomeo or Brazil and 44ok greater in family-level diversity than the
contemporary Wet Tropics rainforestsof northeasternQueensland.
Endemic Australian bats (mystacinids) colonised New Zealand; others (a new Riversieigh group) may have
colonised South America to becomenoctilionoids.
Australia crashedinto SE Asia about l5 mya (m. Mioc.) after which time the New Guinea Highlands rapidly rose
creating a rainshadow over northern Australia.
From 16 to l5 mya (m. Mioc.), Australia'sforestmammalsbegan a decline in diversity and distribution; familylevel extinctions occurred between e. and m. Mioc. time (e.g., miralinids and wynyardiids).
Between 15 and 5 mya (m. to 1. Mioc.), most of Australia beganto dry and the central Australian forests declined.
'Dim age' for Australian mammals (very few sites known) from 12 to 5 mya.
New Guinea was probably colonised by dasyurids. bandicoots, zygomaturine diprotodontids and macro_podir
macropodids-between i0 and 8 mya (1. Mioc.); however, phalangerids,dactylopsiline petaurids an
acrobatids could have colonised New Guinea by 23 mya.
Murid rodents entered Australia from SE Asia via relatively dry corridors into northern Australia sometime
between 8 and 5 mya (1. Mioc.)
By 5 mya (e. Plio.), many modernnon-forestseneraof marsupialshad arisenand grass-eatingmarsupials (species
of Macropus and modern types of vombatids) had appeared.
in Australiacan be describedas arid, nor do grasslandsdevelop, until sometime between4
No plant assemblage
and 2 mya (e. Plio. to e. Pleist.).
Sometimebetween4 and I mya (e. Plio. to Pleist.) Cuscuses probably becameextinct in Australia and then later
reinvadedfrom New Guinea.
'arid'-adaptedmarsupialsappear(e.g., Mauopus (Megaleia) sp.).
By 2 mya (e. Pleist.),the first
Sometime between 120,000 and 50,000 ya, Humans arrived in Australia; two or more types may have overlapped
until 10,000 ya.
Probably by 50,000-40,000ya" extinctions of most of Australia's'megafaunal'mammalsoccurred resulting in
the complete loss of diprotodontids,thylacoleonidsand palorchestidsas well as many genera and speciesin
other families; the cause is controversial.
Approximately4,000 ya, Dingoeswere introducedto Australia possibly resulting in the extinction of mainland
Thylacines (Thylacinus cynocephalus).
Contait with and colonisationby Europeansinitiated the secondmajor, on,eoingextinction phaseof Australian
ammals.
: y a , m i l l i o n y - e a r s . a g oe. . , e a r l l ,
T a b l et . S u m m w y o f k e y e v e n t s r e l a t i n g t o t h e p r e h i s t o r y o fA u s t r a l i a nm a m m a l s .A b b r e v i a t i o n sm
, o c . , E o c e n e ,O l i g - , O l i g o c e n e :M i o c . , M i o c e n e ; P l i o . ,P l i o c e n e tP l e i s t . ,
; a l e o c . ,P a l e o c e n eE
m . , m i d d l e ;1 . ,l a t d ; C r e t . , C r e t a c e o u sP
Pleistocene;NSW, Nerv South Wales; Qld, Queensland; Vic., Victoria: SA, South Australial Ant., Antarctica; WA, Western
Australia; NT, Northern Territory; Tas., Tasmania; SAm, South America; N, northem; NE, nonheastem; E, eastern; SE,
southeastem;S, southem; SW, southwestem;W, western,NW, northwestern;PNG, PapuaNew Guinea.
AUSTRALIANMAMMALOGY
Although it is likely that monotremes have a
history in Australia that will eventually prove to
be Jurassic in age, the base of the known
Australian fossil mammal record starts in the late
early Cretaceous (about I 15 mya). Thus
Australia's first profound 'Dark Age' extends from
220 (age of the oldest known mammals on other
continents) to 115 mya: no Australian mammals
from this interval are knorvn. Early Cretaceous
mammals in Australia come from two regions:
Lightning Ridge, New South Wales (-1 l0 to 100
myo, from the Griman Creek Fm); and Flat Rocks,
Victoria (-115 myo, from the Wonthaggi Fm).
Mammals recovered from Lightning
Ridge
represent two distinct families of monotremes
(Musser 1998). The only mammal described from
Flat Rocks is Ausktribosphenos nyktos, a creature
of uncertain relationships although originally
described as a placental (Rich et al. 1997; Archer
et al. below). Although the late Cretaceous to
middle Eocene rocks of most other continents
have produced mammals in abundance, in
Australia the relatively sparse rocks of this age
have so far produced only dinosaurs and other
non-synapsid vertebrates. Australia's second
profound 'Dark Age' of mammalian evolution
extends from 100 to 55 mya with no species
known from this interval.
Illites in
soft sediments exposed at
Tingamarra, on the outskirts of the town of
Murgon, southeastern Queensland, have been
dated as >55.6 (earliestEocenein age) on the basis
of potassium/argon ratios (Godthelp et al. 1992).
This age has been challengedby Woodburne and
Case (1996) who sugeest that the Tingamarra
assemblagemay be as young as late Oligocenein
age (i.e., 25 myo). From thesesedimentshave come
many different kinds of mammals as well as
osteichthyan fish, myobatrachid frogs (Tyler and
Godthelp 1993), madtsoiid snakes (Scanlon
(Gaffney and
1993),
trionychid
turtles
Bartholomai 1979), mekosuchine crocodiles
(Salisbury and Willis 1996) and passerine(Boles
et al. 1994) as well as more archaic groups of birds
including graculavids which are otherrviseknown
only from late Cretaceousand Paleocenedeposits
in North America and probably Asia (Boles 1999).
Where it has been possible to assess the
biocorrelative significance of
these nonmammalian vertebrates, they
are broadly
supportive of an early Eocene (or marginally
older) age.Although the Tingamana passerineis,
by' 20 million years, the oldest in the rvorld, a
significant pre-Oligocene Southern Hemispheric
record was predicted by some palaeornithologists
(Boles 1995). Biocorrelative studies of the
Murgon mammals. which include a ?placental
(Godthelp et a/. 1992), archaeonycteridid bat
(Hand et al. 1994)and many marsupialsincluding
several with affinities to Casamayoran (early
Eocene)as well as Tiupampian (Paleocene)South
American groups (e.g., Archer et al. 1993
Godthelpet al. 1999).provide additional ,uppori
for
the early Eocene illite
date. The
archaeonycterididbat in particular is closely
related to earliest Eocene bats from North America
and Europe. This interval in Australia was
warm, humid
characterised by
relatively
'greenhouse'
climatic conditions (McGowran and
Li 1997).
The complete lack of kno,'vledge about
Australian terrestrial mammals from middle
Eoceneto middle Oligocene time, an interval of
about 30 million years, is maddening.During this
'dark
third profoundly
age', Australia severed its
last tenuous connections to the rest of Gondwana
(Antarctica plus South America). sometime
betweenabout 46 and 35 my4 as it acceleratedits
northward dri{1 towards the equator. From middle
Eocene to late Oligocene time, Australia was
characterisedby relatively cool, dry 'icehouse'
climatic conditions.
The latest Oligocene (about 25 to 23.3 mya)
marks the beginning of a relativelyt welldocumented phase in
Australian mammal
evolution.Fossil assemblages
from two regions in
particular, northeasternSouth Australia (the Tirari
Desertin the Lake Eyre Basin, and the Lake Frome
arca in the Lake Frome Embayment) and
Riversleigh (northwesternQueensland;described
'System
by Archer et al. 1997c as
A' local faunas)
have provided an increasingly rich late Oligocene
record. The Geilston Bay LF of Tasmania is
probably early Miocene in age but the faunal
is small and the taxa relatively poorly
assemblage
preserved. Eyre Basin Tertiary sediments have
been dated with reasonable confidence usin_e
palaeomagnetostratigraphy (Woodburne et al
1994). Riversleigh Tertiary sediments,previously
dated by biocorrelations involving
central
Australian (marsupial) and European (bat) tara, are
now th€ subject of a uranium/leaddating program.
The early Miocene (about 23.3 to 16.3 mya.y
record documents Australia's most diverse
mammalian faunas. These are best-known from
'System
Riversleigh(the
B' local faunas of Archer
et al. 1997c)and the Tirari Desert (the Kutjamarpu
LF). Less well-known assemblagesoccur in the
Northern Territory (Kangaroo Well LF) and
ARCHERETAL. EVOLUTIONOF AUSTRALIANMAMMALS
Tasmania (Wynyard LF). As a measure of early
Miocene diversity, a single assemblage at
Riversleigh (Upper Site). interpreted to represent a
closed forest community (Archer et al. 1997b),
contains 64 species of mammal and 44%omore
family-level diversiry than occurs today in the
whole of the Wet Tropics rainforests of
northeastern Queensland. The Leaf Locality may
represent a riparian closed-forest community but
the Riversleigh deposits of this age appear to
indicate regional widespread, complex rainforest.
If relatively drier communities bounded these
forests, they left no trace in the known fossil
record. Further, Martin (1998) has argued on the
basis of palaeobotanical data that there is no
evidence for arid conditions in Australia prior to
the end of the Tertiary. in late Pliocene times.
abrasive cell walls. The Beaumaris LF from
Victoria is another late Miocene site but one with
very few taxa represented. Drying conditions
throughout much of the continent may account in
part for the relatively poor late Miocene fossil
record. This also appearsto be the time during
which many of the modern groups of marsupials
(e.g., dasyuroids, peramelids, peroryctids anc
macropodids),bats (vespertilionids) and perhaps
rodents (murids) underwentexplosive radiations
to producethe high diversity of species lineages
present today.
Pliocene assemblages are widespread and
known from all states except Tasmania and the
Northern Tenitory. By this time' Australian
mammal communities had significantly changed
with representatives of many modern gexera
appearing for the first time including the firstAlthough early and middle Miocene mammal
known Australian rodents. Size increases
to
compared
high
diversity was relatively
most noticeably
diversity in Australia's contemporaq' forests, continued in most lineages,
such as several
marsupials
herbivorous
among
of
nature
restricted
geographically
because of the
diprotodontids which had become the size of
the palaeontological sampling areas, actual
cattle. The first, albeit rare occurrences in inland
continental diversity at these times was almost
communities of grazing mammals in the early
certainly much higher than currently documented.
Pliocene suggests that grasseswere increasing in
The middle to late Miocene (about 16.3 to l0
abundance. Other early Pliocene communities,
mya) marks the start of significant declines in
however, indicate the persistence of refugial
a pronounced climatic
following
rainfall
rainforest in continentally peripheral areas such as
oscillation between 16 and 15 mya (McGowran Hamilton. Victoria. There are indications. mainly
Miocene mammal from palaeobotany,that the early Pliocene was a
1994). Middle
and Li
communities are only known from Riversleigh
of a general drying and
'System C' assemblages time of brief amelioration
(those described as
cooling trend with brief expansions of rainforest
(Archer et al. 1997c\ and Bullock Creek in the
into mesic regions.
Northem Territory (Munay and Megirian 1992).
Pleistocene diversity has been under
These are commonly characterisedby less family,
continuous investigation since fossil mammals
generic and species-leveldiversity than those of
the garly Miocene. What is known about mammals were first found in Australia. At least 26 of the
living genera are represented by Pleistocene
in niost lineages demonstrates a rapid increase tn
species, although some of these may represent
middle
and
the
early
of
average size over those
anagenetic lineages (e.g., Macropus titan to M.
Miocene.
giganteus, and Sarcophilus laniarius to .S
The late Miocene is represented by very few
harrisii). Because of human impact and/or
Australian assemblages.Of these, the best-known
climatic change, most of the Pleistocene
are the Alcoota and Ongeva Local Faunas of the
megafauna(Munay 1991), about 50 species, had
Northern Territory (Munay and Megirian 1992). vanished by about 40,000 years ago. A few
These contain many medium to larger-size megafaunal species underwent post-Pleistocene
mammals but very few of the smaller kinds,
dwarfing in proportion to their absolute size such
leaving a gap in understanding about the
that some of the largest living kangaroos (e.g',
evolutionary transition of smaller mammals from
species of Macropus) ue 30ok smaller than their
middle Miocene to Pliocene time. A singie
Pleistoceneancestors.Dating of some Pleistocene
assemblage at Riversleigh, the Encore LF. may
is controversial (Baynes
megafaunalassemblages
represent the early late Miocene with taxa that
199'7and A. Baynes,pers.comm., 1997).
appear to be antecedent to Alcoota forms. This
Holocene increasesin species-ieveldiversity
Riversleigh assemblageis also the oldest-known
(a
groups such as phalangerids,pseudocheirids,
for
teeth
rvith
ever-growing
wombat
a
include
to
speciesof Warendja-iike vombatid), a suggestion dasyurids, peroryctids, peramelids, potoroids,
murids, vespertilionidsand probably pteropodids
of the presenceof terrestrial plants with relatively
AUSTRALIAN MAMMALOGY
H-
d , 3
a)
c
-
UJ
tr
?
g
=
o
=
{
I
!d
=
;
o
z
o
o
o
r
J
=
!
q
!
g
=
oa
o
F
i'
?
P
r
7
ff
=
a
E
,+
!
-
a
.P
C>
iiu
1-P9:-1ti,^=.9
5
Frs**:=r#i:
e : ;F* i E - t ; , < - ?sd+{ P- + ;r i9['.=iFqs F*
o
4iii
; l EE E
bd
i
=.2
fr;
j
v;o
ED:
Hi
"
."<_s
_,"I
g6:
d 9F
A! !
T r
=.
A
:
otda
;
I
*a
{;;
€i
!-3e
9i6 €
gih?
Pl==
*lEa
=i9-
;{
Gh
i+
I BJ!
! p
fr5!.
r:-E<
3:
L6
4
9i
= r =i l i
. 6 :
l
-
v '
; 1 9 R += r ,
g
rgrou+l
q
g!
Ed
:: =
I r'
*;l- *a P .
<__l__):,
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
al. 1997), and morphology (Szalay 1982, 1994).
Although there is general consensus, following
studies of tarsal anatomy by Szalay (1982)' that
microbiotheriansshare a special relationship with
Australian marsupials,there are arguments about
the precise nature of their placement within
Australidelphia, some concluding a special
dasyuromorphians and
with
RELATIONSHIPS OF AUSTRALTAN NIAN,A4ALS
relationship
(albumin serology; V. Sarich,
peramelemorphians
The phylogenetic relationships of Australian pers. comm. in Archer 1982a),others a sister taxon
mammals have been extensively examined at three
relationship with dasyuromorphians (Szalay
levels: infrafamilial. interfamilialand interordinal'
1994), dasyuromorphiansand notoryctids (Retief
their
Three factors restrict ability to interpret
et
at. 1995), diprotodontians (Kirsch et al. 1991),
interordinal relationships with confidence First,
and
still others with the whole of the Australian
yalkaparidontians, yingabalanarids and the
radiation (e.g., Szaiay 1982), or the
marsupial
possible placental condylarth Tingamarra
radiation plus the early Tertiary South
Austrilian
porterorum are only known from an inadequate
American genus Andinodelphys (Marshall et a/.
fossil record. As a result, there are disagreements 1990).
interordinal
the
of
Resolution
about their ordinal and interordinal relationships
of Australian marsupials has been
relationships
(see Archer et al. rhis work). Second, two of the
further complicated following the suggestion that
orders (Chiroptera and Rodentia) have more
peramelemorphiansmay lie outside of a clade
extensive non-Australian representation and
inclusive of all other Australian marsupials and
almost certainly arrived in Australia as diversified
or even outside of all extant
Dromiciops,
groups with closer phylogenetic relationshipsto
(Springer et al' 1994; Retief et c/
marsupiais
non-Australian mammals. Third, charactersystems
I 995).
examined for representativesof many of the other
orders have in some cases produced widely
Intraordinal and intrafamilial relationships are
differing indicationsof phyiogeneticrelationship' briefly explored in the individual contributions
For example, while most mammalogists regard that follow. For some of these groups such as
monotremes to not only be outside of the clade dasyuroids, perameloids and macropodoids,
containing marsupials and placentals but even interfamilial phylogenetic relationships remain
outside o] therian mammals, DNA-hybridisation
uncertain. In part this reflects the fact that
suggeststhat monotremesmay be the sister group
although many modern families are present and
of marsupials (e.g., Janke et al. 1997, Kirsch and
distinct in 26 to 23 million-year-old sediments.
Mayer 1998). Similarly, there is considerable the fossil record between55 and 26 million years
disagreementabout the interordinal relationships of age, the interval when most of these families
peramelemorphians, dasyuromorphians' probably differentiated from each other' is so far
of
notoryctemorphians and diprotodontians based utterly silent. Further, confident resolution of
on divergent indications of albumin serology many higher-level systematic problems may be
(e.g., V. Sarich, pers. comm. in Archer 1982a), beyond the current capacity of molecular
whole serum serology (Kirsch 1977)' l25
systematic techniques. With continued research.
ribosomalRNA (Springer et al. 1994), protamines both of these limitations must be overcome.
(Retief et al. 1995), DNA-hybridisation (Kirsch et
indicated in the palaeodiversityfigures belorv is
almost certainly a reflection of the limitations of
the Plio-Pleistocene fossil record. This is
oarticularl-v the case for smaller mammals which
ire less liliely to be collected.
AUSTRALIANMAMMALOGY
DIVERSITYAND RELATIONSHIPSOF LIVING AND EXTINCT MONOTREMES
A. M. MUSSER
School of Biological Science,University of New South Wales, Sydney,NSW 2052
TWO early Cretaceous monotremes have been specimen,originally describedas ornithorhynchid
described from the Griman Creek Formation, (Ornithorhynchus maximus Dun, 1895), is
tachyglossid and probably belongs to the
Lightning Ridge, New South Wales. The platypus(family
individual representedby the cranium (Mahoney
galmani
like
Steropodon
'Zaglossus' hacketti
and Ride 1975). The huge
Steropodontidae) was Australia's first known
may
represent a distinct
Australia
Westem
(Archer
from
et al. 1985). A second
Mesozoic mammal
genus (J. Mahoney cited in Griffiths et al. 1991).
taxon, Kollikodon ritchiei (family Kollikodontidae), characterised by unique bunodont Zaglossus material from the Pleistocene of New
teeth, may have specialised in eating crustaceans Guinea probably representsthe living Z. bruijnii
(Munay 1991). The living Short-beakedEchidna
(Flannery et al. 1995). An edentulous maxillary
fragment from Lightning Ridge has been described (Tachyglossus aculeatus) is known from many
as possibly that of a monotreme (Rich et al. 1989). Pleistocene to Holocene deposits in southern
Edentulous lower jaw fragments of a monotreme- Australia(Munay 1978, 1991).
like mammal from Lightning Ridge have also been
The cladogram (Fig. l) reflects the uncertain
recovered.
phylogenetic relationships of tachyglossids as
well as the unresolvedposition of the very derived
The oldest undoubted ornithorhynchid i s
K. ritchiei. Some genetic studies infer that
Monotrematum sudamericanum, from late early
tachyglossids separated from ornithorhynchids
Paleocene sediments in Patagonia, Argentina
(Pascualet a1.1992). This species demonstrates near the end of the Cretaceous or early in the
that at least ornithorhynchids had a Gondwanan Tertiary (e.g., Westerman and Edwards 1992;
distribution (Pascual et al. 1992). Two species Messer et al. 1998 but see Retief el al. 1993).
referable to the genus Obdurodon have been Given the early Cretaceousage of the platypusrecovered from the late Oligocene central like S galmani, such findings suggest that
tachyglossids may have been derived from a
Australian Etadunna and Namba Formations:
platypus-like ancestorthus making the platypus
Obdurodon insignis (Woodburne and Tedford
lineage paraphyletic. However, Messer e/ a/.
1975; Archer et al. 1978) and the undescribed Ob.
(1998) suggest that these dates may be
sp. A. Ob. dicksoni, athird Obdurodon species,has
underestimates of divergence times because
been recovered from the Riversleigh World
Heritage fossil deposits in northwestern Queens- evolution of at least the milk protein cx,lactalbumin appears to be slower in monotremes
land (Archer et al. 1992, 1993; Musser and Archer
1998). The only known Pliocene ornithorhynchid than in living therians.
material consistsof a limb fragment from the Bov,
Much of the monotreme fossil material is
Local Fauna, near Merriw4 NSW refened to the
fragmentary and vast temporal gaps separatetaxa.
genus Ornithorhynclrzs (Rich er al. l99l b).
However, some tentative conclusions about
Pleistoceneornithorhynchid material appearsto
historical diversity can be drawn. The presenceof
represent the living Ornithorhynchus anatinus
S. galmani in the early Cretaceousis testimonl to
(e.g.,Archer et al. 1978 Marshall 1992).
the antiquity of platypus-like monotremes (Archer
Echidnas of the genus Megalibgwilia,
et al. 1985), while the presence of the
characterised by a beak shape intermediate contemporaneous,specialised K. ritchiei argues
for far greater intraordinal diversity
between that of the living Tachyglossus and
than
previously anticipated (Flannery et al. 1995).
Zaglossus,are known from Pleistoceneswamp and
cave deposits from Tasmania. New South Wales
However, from the early Tertiary to the present,
and South Australia. This recently erected genus
only
ornithorhynchids and tachyglossids are
supersedes
the assignmentof severalIong-beaked
represented. Ornithorhynchids may have been
types (e.g., ramsayi) to the genus Zaglossus
(Griffiths et al. 1991). A middle Miocene echidna. more diverse during the early to middle Tertiar,
(four species are known) whiie tachyglossids
Echidna (Proechidna) robusta Dun, 1895 from
Gulgong, New South Wales,is probably a species appear to have radiated during the PlioPleistocene.with at least five Pleistocenespecies
of Megalibgwllia (Griffiths et al. 1991). A
recognised(Murray 1978; Griffiths et al. 1991).
partial
humerus found with the
cranium of this
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
Hoiocene
10,000- 0
Pleistocene
2 - i0,000
Pliocene
E - M Miocene
23.3- 10.4
L.
Eocene
Paleocene
63-61
4
s.F
ss
as
!x
s:
:<
.,
s_!
s*
*' r g,E F{
>o%
603=R
r-^
S
XYo4iqJ
S
rI t l l
'\
>
't'
Z N
x5
f
E
6
r-.lJ
ll
ll
Tachygilossidae
Steiopodontidae
Mo
Fig .1. Monotreme diversity through time: a cladogramof probablerelationships(see text) within Monotremata. Uncertain
re[tionships are indicated by dashed lines. Each block representsa distinct species.Blocks that extend without breaks
through more than one time period indicateapparentl,vlong-survivingspecies
Ph,vlogeneticaffinities of monotremes to
other groups remain contentious. Dental structure,
diagnosedas therian based on the triangular molar
blades of S. galmani (Archer et al. 1985; KielanJawororvska et al. 1981), may be autapomorphic
therian-like
with independent acquisition of
triangular blade systems (Archer et a/. 1992,
1993). Basicranial anatomy appears to link
monotremesmost closely with multituberculates
(e.g., Wible and Hopson 1993; Meng and Wyss
1995). Severalkey Mesozoic taxa known from
postcranial material exhibit advanced shoulder
Hu et al. 1997), placing
girdle morphology (e.-e.,
the plesiomorphic shoulder girdle of monotremes
betrveen archaic morganucodontids or tricono-
dontids and the more derived multituberculates
(Hu et al. 1997), and suggesting an early to late
Jurassicorigin for the monotremeshoulder girdle
(Musser i998). Genetic evidence is equivocal:
Messer et al. (1998) support a middle Jurassic
divergence of monotremes from therian mammals,
althou,eh Westerman and Edwards (1992) piace
this split at the earliest Cretaceous. A close
relationship betweenmonotremesand marsupials
to the exclusion of placental mammals has
recently been suggested(Janke et al. 1997; Kirsch
and Mayer i998 in revisions of Gregory's [1947]
Marsupionta hypothesis), although results from
other molecular techniques argue against this
t0
AUSTRALIANMAMMALOGY
EVOLUTION OF AUSTRALIA'S MARSUPICARNIVORES: DASYURIDAE,
THYLACINIDAE, MYRMECOBIIDAE, DASYUROMORPHIA INCERTAE.S.EDISAND
MARSUPIALTA INCERTAE SEDIS
Stepuer- WROEANDJEA\EI*rE MUIRHEAD
School of Biological Science"University'ofNerv SouthWales.Sydney'NSW2052
RECENT fossil discoverieshave shednew light on
the subject of Australian marsupicarnivore
evolution (Fig. l). The new material includes the
first of Palaeogene age, as well as the first prePliocene crania referable to the Dasyuridae and
Thylacinidae. A nerv species (Godthelp et al.
1999), from the early Eocene Tingamarra Local
Fauna. southeastern Queensland, is the oldest
of
the
families
potential
ancestor for
Dasyuromorphia. Horvever,this species could be
either an ameridelphian or australidelphian
relative t o
its position
marsupial. and
dasyuromorphiansis unclear.At the other end of
the time scale. Recent.Pleistocene and Pliocene
faunas are dominated by dasyurids (with at least
63 extant species), while Thylacinidae is
represented by a single species (Thltlacinus
cynocephalus).The family Myrmecobiidaehas no
fossil record.
Fossil evidence suggests that the high
diversity of dasyuridsis a late middle Miocene to
Pliocene phenomenon. Since the last substantial
revision (Archer 1982b), seven new species of
thylacinid have been described. all from late
Oligocene to late Miocene deposits (Wroe 1996,
Muirhead 1997t Murray 1997; Muirhead and Wroe
1998). Descriptions of four more new taxa are in
progress.The rapidly expanding tally of Miocene
carnivore ta-xa challenges the
marsupial
hypothesis of Flannery (1997), that Australia has
not supported a diverse large mammalian
carnivore fauna over the last 20 my (Wroe 1996"
1999a).in severalhighly localised local faunasof
Miocene age from Riversleighthe number of large
marsupialcarnivores (Dasyurttsmaculatus size or
greater) equals or exceedsthat known for the vast
majority of South American localities from the
sameepoch(Wroe and Myers 1998). Only trvo nerv
late Oligocene to late Miocene dasyurid species
are known. Wroe (1996, 1997) reassigned a
'dasyurid' taxa (e.-e.,
number of pre-Pliocene
Il akamatha)
to
KeetrnaAnkotarinja,
Dasyuromorphiaincertae sedis on the grounds
that their placementwithin Dasyuridaewas based
A nerv Riversieigh species
on symplesiomorphies.
(Wroe 1999b),of earll,to middie Miocene age. is
the oldest taxon confidently placed within
Dasyuridae.It is basal to a monophyletic clade
inclusiveof the three extant dasyurid subfamilies
(Sminthopsinae,
Phascogalinae.Dasyurinae),none
of which includes fossil taxa older than earliest
P l i o c e n e .W r o e ( 1 9 9 8 , 1 9 9 9 b ) c o n s i d e r s t h e
possibility that the common origin of extant
dasyurid subfamiliesmay have been as recent as
the late Miocene. Alternatively. on the basis of
moleculardata. Krajewski et al. (1997b) postulate
that the basal polytomy within Dasyurinae
occurredabout 15 million years ago. If this is
conect then the common ancestor of modern
dasyurids is more Iikely earty Miocene in age
Many' aspects of the most recent morphologybased review of dasyurid phylogeny (Archer
1982a) have been supported by' subsequent
but
significant
molecule-based studies,
amendmentshave also been suggested(Krajewski
er al. 1997a. 1997b). Notable among these has
Archer's
of
the
subsumption
been
within Dasyurinae(Krajewski el
Phascolosoricinae
al. 1994). A fundamental question that remains
unanswered. using either morphological or
molecular techniques. is the phl'logenetic
position of Myrmecobiidae.
11
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
llolocene
10,000- 0
Pleistocene
2 - 10,000
Pliocene
<) -)
L. Miocene
1 0 . 4- 5 . 2
M. M.iocene
1 6 . 3- 1 0 . 4
E. Miocene
23.3 - 76.3
L. Oligocene
zd - z'-t
Eocene
55
E
C;
r3 E€E
gs
:
s.'EH
f
i
aE
A,E
;€
z
A ; .hE
=
L . =
s
!.f
; F
E
E>
da
Dasyuromorphia
Frg. 1. Australian marsupicarnivorediversity through time and a cladogram of probable relationshipsrvithin the group.
Uncertain relationshipsare indicatedb.v dashedlines. Although the Murgon taron is presentedin this context, its ordinal
relationshipsare not necessarillywith Dasyuromorphia. Each block represents a distinct species, species diversity is
indicated by a number. Blocks ihat extend without breaks through more than one time period indicate evidently longsurviving species.
BANDICOOT DIVERSITY AND EVOLUTION (PERAMELEMORPHIA,
MARSUPIALIA): THE FOSSIL EVIDENCE
JEANETTE MUIRHEAD
Schoolof Biological Science.Universityof New SouthWales.SydneyNSW 2052
MODERN bandicoots (Order Peramelemorphia)
comprise three families: Peramelidae,Peroryctidae
and Thylacomyidae. Living peramelids (ordinary
bandicoots) include eight species in three genera.
Living peroryctids (forest bandicoots) include ten
species in four genera. Living thylacomyids
(bilbies) include two species(one recently extinct)
in one genus. Only four fossil bandicoots have
been described to date: the early Pliocene
Ischnodon australis (Stirton 1955), the early
PliocenePeramelesallinghamensis (Archer 1976)
and Peramelesbowensis (Muirhead et al. 1997),
and the early to middle Miocene Yarala
burchfietdi (Muirhead and Filan 1995). While
Ischnodon australis. P. allinghamensis and P.
in
been
all
have
bowensis
Placed
peramelemorphian families dominated by Recent
taxa (i.e., Thylacomyidae for .L australis and
Peramelidae for P. allinghamensis and P.
bowensis), the older Y. burchfieldi cannot be
recognised
currently
placed within
any
peramelemorphianfamily (Muirhead and Filan
This Tertiary taxon is regarded as
1995).
representing
a new family that is the plesiomorphic
sistergroup of a peramelid/peroryctid/thylacomyid
clade Muirhead 1993, 1994, Fig. 1). No prePleistocene peroryctids are knorvn. Many other
but
fossil peramelemorphians are known
undescribedincluding taxa from the early Eocene
TingamarraLocal Fauna-late Oligocene and early
Miocene
12
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
Holocene
10,000- 0
Pleistocene
2 - 10,000
Pliocene
5 . 2- 2
L. Miocene
10.4- 5.2
M. Miocene
16.3- 10.4
E. Miocene
23.3- 16.3
L. Oligocene
26-23.3
Eocene
55
O
o
€
>r
6
a
z
,
>'
o
O
F
Fig. ). Bandicoot diversity through time and a cladogram of probable relationships.within the.group. Uncertain
reiationshipsare indicatedby dashedlines. Although the Eocene(Murgon) ta-ronis placed here among the most diverse of
the fwo new families, is familial affinities are uncertain. Each block representsa distinct species. Blocks that extend
without brealisthrough more than one time period indicateapparentlylong-survivingspecies.
assemblages of central Australia and late
Oligocene to early Pliocene assemblagesfrom
Riversleigh in northwestem Queensland and
early Pliocene taxa from Bluff Downs in
northeastem Queensland (Rich el al. 1991b;
Archer e/ al. 1993; Mackness et al. 1993;
Muirhead 1994; Woodburne and Case 1996).
Like Y. burchfieldi, none of the Oligo-Miocene
Riversleigh taxa can be placed within the
clade
peroryctid/peramelid/thylacomyid
(Muirhead 1994). Woodburne and Case(1996)
similarly support a relatively late radiation of
peramelemorphians based
on
living
undescribedtaxa from late Oligocene and early
In
Miocene deposits of central Australia.
depositsyounger than middle Miocene. there is
no evidenceof the archaictaxa typical of OligoMiocene deposits,with the possible exception
of one speciesof an otherwise Miocene genus
in the early Pliocene Hamilton Local Fauna
(Muirhead 1994; Dawson et al. 1999). All other
within
the
Recent taxa fall
Pliocene to
This
peroryctid/peramelidithylacomyid clade.
distribution suggests a post middle Miocene-pre
followed
Pliocenebottleneckfor peramelemorphians,
'modern'
the
by a massive radiation of
peramelid/peroryctid/thylacomyidclade - a scenario
consistent with late Miocene vicariance events
between New Guinea and the Australian mainland
(Aplin et a|.1993). The fossil evidence,however. is
in stark contrast to the Oligocene divergence of the
living taxa proposed by biochemical analyses (e.g.,
32 myo, Baverstocket al. 1990a;24 myo. Baverstock
et al. 1990b; 25 myo, Kirsch el al. 1997). The wide
diversity of the Tertiary taxa and the presence of
peramelemorphiansin early Eocene deposits from
Murgon (Archer et al. 1993; Godthelp et al. 1992)
supports a considerably older derivation for the
Peramelemorphia.evidence that provides support for
the hypothesisof Kirsch et al. (1991) which suggests
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
on the basisof DNA-DNA hybridisation studies
that oeramelemorphians are distinct from all
other Australianmarsupials.
THE EVOLUTIONARYHISTORYOF NOTORYCTIDS,YINGABALANARIDS'
YALKAPARIDONTIDSAND OTHER ENIGMATIC GROUPSOF AUSTRALIAN
MAMMALS
H6XX GOO1HELPt,MIRI,r-DA GOTTtr, YUANQINGWANGT'o,ANNE
MtCgaEL ARCHERT'2,
MUSSER'
:Australian
tschool of Biotogical Science,UniversiS,
Wales, Sydney 2052 Australia;
lf New South
(Jnit.-NSW
Wildlife
and
National
Park
|r{useum,O-SCott"uguSt,SydneyNSW 200'0;3bndangered,species
rlnstitute of Vertebrate Paleontology and
2220;
NSII
Hirsnitte,
1967,
Box
PO
Service,
paleoaiLnthropology, Chinese Academy of sciences, PO Box 643, Beiiing 100044, China
autapomorphic
are several highly
TffiRE
Australian mammals whose relationships to other
groups remain unclear.Theseinclude: notoryctids,
yingabalanarids, yalkaparidontids, Tingamarra
bartholomaii,
Thylacotinga
porterorum,
microbiotheriids, Ausktribosphenos nyktos and
other inadequatelyrepresentedor unassessedtaxa
(Fi-q.1).
Living notoryctids, or marsupial moles,
species (depending on
include one or two
genus
Notoryctes. Aplin and
the
in
authority)
Archer (1987), recognising the distinction of
notoryctids as well as long-standing disagreements
about their phylogenetic relationships, elevated
them to ordinal status as notoryctemorphians, a
vieu' generally accepted. For example, Westerman
(1991), on the basis of DNA-DNA hybridisation
studies, supports the view that they should be a
distinct order. At least one distinct genus of
notoryctids is present in the early Miocene
sediments of Riversleigh (Gon 1988; Archeret al.
1994, 1997). in terms of postcranial and dental
remains, it is unmistakably notoryctid but it is
much more plesiomorphicthan the living taxa with
a pre-zalambdodont molar morphology that
clarifies the otherwise enigmatic structure of the
upper molars of species of Notoryctes. There is no
reason known to us why the Miocene taxon could
not be ancestralto the speciesof Notoryctes.
with
confidence its inter-ordinal affinities.
Discovery of the second molar (currently under
study by one us, Y.W.) may help to narrow the range
of possibilities consideredby Archer et al. (1990)
which included marsupials, noctilionoid bats,
primates, symmetrodonts, zalambdodonts and
tribotheres.
Yalkaparidon coheni and Y.jonesi (Archer et
al. 1988), from the late Oligocene to middle
Miocene sediments of the Riversleigh World
Heritage are4 are among the most specialisedof any
known Australian mammals. They combine unique
and completely zalambdodont molars *'ith an
enormous, curved, hyselodont first incisor which
extends below the whole of the cheektooth row. A
skull of the former suggests marsupial affinities
but beyond this nothing is certain-Szalay (1994)
regards them to be aberrant diprotodontians
presumably on the basis of the incisor formula
which is I1-3/1. Archer et al. (1988) refened this
genus to its own order, the Yalkaparidontia, based
on its suite of very distinctive autapomorphic
features and lack of undoubted synapomorphies
with other orders of marsupials.
Tingamarra porterorum (Godthelp et al. 1992)
was originally described,on the basis of a single
lower molar, as a possible condylarth placental
from the early Eocene Tingamarra Local Fauna from
Murgon, southeasternQueensland.While a second,
Yingabalanara richardsoni (Archer et al. larger but otherwise similar taxon has since been
1990). known so far from two isolated lower molars recoveredfrom the same deposit, it does not clarify
found in early Miocene sediments of the the relationships of this puzzling mammal. It
was remainsa less parsimoniouspossibility that this is
ue4
Heritage
World
Riversleigh
its own family, a very distinctive group of marsupials that has
conservatively placed in
Yingabatanaridae.The bizarre molar morphology of convergedon placentaltooth morphology.
this animal has made it impossible to determine
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
14
Holocene
10,000- 0
Pleistocene
2 - 10,000
Pliocene
<t -)
L. Miocene
10.4- s.2
M. Miocene
16.3- 10.4
E. Miocene
23.3- 16.3
L. Oligocene
zo - z5,J
Eocene
55
Paleocene
E. Cretaceous
1 1 s- i 0 0
6q
Sf g
E
$
s
S^
E.
t € F c
S .E F EF b s r t
q E g EA $$B F
d = S 5E € F fi 2
Frg .1. Diversity through time of notoryctids,yingabalanarids,yalkaparidontidsand other enigmatic groups of Australian
mammals. Most of these are distinct from each other at the ordinal level. The interordinal relationshipsof all except
(arguably) the microbiotheriidsand notoryctids are unclear. Each biock representsa distinct species.Blocks that extend
without brealisthrough more than one time period indicateapparentlylong-survivingspecies.
Thylacotinga bartholomaii (Archer et al
1993) was described,on the basis of isolated teeth
from the early EoceneTingamarraLocal Faun4 as a
semi-bunodont marsupial with uncertain ordinal
affinities. Although the morphology of this
animal
is
molar
distinctive.
bunodont
morphology in general has been developed
independently
of
in
families
several
didelphimorphian. polydolopimorphian
and
diprotodontian marsupials. Similarities to
possibly early Paleocenemarsupials from Peru are
most striking and the subject of ongoing research.
Since
microbiotheriids
were
firsr
hypothesised by Szalay (1982), on the basis of
tarsal bone morphology, to have been the sister
group (or ancestral) to Australian marsupials,
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
many other soft-tissue studies have corroborated
this broad hypothesis. The details of the
relationship are. however, controversial with other
that
suggesting
alternativell
authors
microbiotheriids are the sister group to a
clade or the sister
combineddasyuroid/perameloid
group of a combined dasyuroid/diprotodontian
clade. Discovery of at least two microbiotheriidlike marsupials in the early Eocene Tingamarra
Local Fauna may be seenas further supportfor this
i n t e r c o n t i n e n t acl o n n e c t i o n .
Other enigmatic early Eocene taxa in the
Tingamarra Local Fauna are less well representec.
These range from bunodont to dilambdodont
marsupials, most of which are known only from
isolated teeth. Some resemble early Eocenegroups
known from Argentina such as caroloameghiniids.
Continuing work on this deposit will almost
certainly enable their relationships to be better
published.
assessedand these assessments
15
peramurids including Peramus and the putative
peramurid Vincelestes. Kielan-Jaworowska et a/.
(1998) have suggested that A. nyktos may have
been derived from early symmetrodont stock
based on certain primitive featuresof the mandible
(including the probable presence of attached
postdentarybones),the dentition being convergent on a tribosphenicpattem (as may be the case in
monotremes: Archer el al. 1992). Discovery of
additional materials will help to clarify the
relationships of this very curious and oldestknown Australian mammal.
from
A somewhat mammal-like taxon
Lightning Ridge, New South Wales was described,
on the basis of an edentulous maxillary fragment,
by Rich et al. (1989). Alveoli indicate that the
teeth were multi-rooted. Although this may well
representa mammal if not a monotreme (Musser,
above), it might also represent a derived but nonmammalian synapsid. Until better material is
found, this specimen will remain a tantalising
mystery.
Ausktribosphenos nyktos (Rich el al. 1997)
u'as described,on the basis of a dentary with four
Several mammalian taxa described as fossils
teeth. from early Cretaceous(-li5 million-yearold) sediments at Flat Rocks, Victoria. Its authors from Australia are even more problematicalsuggestthat it is most parsimoniously interpreted Undoubted elephants were described by Richard
Owen as Mastodon australis (Owen 1844) and
as a placentalmammal. If so, it would be as old as
any previously known in the world (e-g., Notelephas australis (Owen 1882). While it is
possible(perhapseven probable)that these fossils
Prokennalestes from Mongolia) and would
were not collected in Australia, collection data
challenge conventional understanding about the
indicate otherwise which leaves these records as
time of arrival of placentals into Australia. Thts
interpretation has prompted spirited debate. provocativemysteries.
Examination of the specimen by one of us (M-A.)
Other enigmatic mammal fossils include:
suggests that A. nyktos may represent either an
Cuscus procuscus DeVis, 1889 probably from the
archaic monotreme, autapomorphic peramurid or a
Fauna; .
Local
early Pliocene Chinchilla
unique ordinal-level group that has convergedon
Arc-hizonurus securus DeVis, 1889 probably from
therian mammals. Somefeaturesof A. nyktos cited
the Chinchilla LF; and Chronozoon australe
by Rich et al. (1997) as indicative of placental
DeVis, 1883from the Chinchilla LF. It is possiblp
affinities (e.g., molar number. decrease in size
that C. australe represents a sirenian but the
posteriorly of molars, possessionof wide talonids
relationshipsof the other two DeVis taxa which are
last
and possibly the presenceof a submolariform
each represented by a single incomplete
premolar) are also characteristic of monotremes.
postcranial element, are unclear. Details of these
Peramurid affinities are suggestedby the trigoniddescriptions are compiled in Mahoney and Ride
like nature of the posterior premolar as in other
( 19 7 5 ) .
l6
AUSTRALIAN MAMMALOGY
DIVERSITY AND RELATIONSHIPS OF LIVING AND EXTINCT KOALAS
(PHASCOLARCTIDAE, MARSUPIALIA)
KARENBI-eCx
Universityof New SouthWales.SydneyNSW 2052
Schoolof BiologicalScience.
PHASCOLARCTIDAE is an ancient, once diverse
marsupial family represented by a single extant
species, the Koala Phascolarctos cinereus. The
family is believed to occupy a key position near
the base of the diprotodontian ordinal tree (Archer
1976: Archer and Hand 1987; Myers et al. this
work). Recent classifications have separated
phascolarctidsfrom all other vombatomorphians
(Myers et al. this work, fig. l). Woodburne (1984)
erected the superfamily Phascolarctoidea and
Aplin
and Archer (1987) the infraorder
each containing the single
Phascolarctomorphia,
family Phascolarctidae.Of living groups, koalas
are undoubtedly most closely related to
vombatids, a relationship supported by both
molecular and morphological data.
Six genera and at least 18 speciesofkoalas are
currently recognised(Fig. 1). Superficially there
appearsto have been an alarming decline in koala
diversity. However, at any one time since the late
Oligocenekoala diversity has normally been low
two
species
with a maximum of only
contemporaneousin any given faunal assembla-ee.
In terms of abundance, evidence from the fossil
record suggests that koala populations are larger
now than at any point in their history, a fact which
has been linked to the spreadof open sclerophyll
forest during the mid-late Miocene. Archer and
Hand (1987) hypothesisedthat a co-evolutionary
relationship developed between koalas and
eucalypts during the early Miocene, the age of the
oldest known association betweenphascolarctids
(e.g. Litokoala kutjamarpensis) and eucalypts
(Stirton et al. 1967). Low-diversity rainforest
populations of koalas in Australia's mid- to late
Tertiarl' deposits may have resulted from
with
rapidly
diversifying
competition
pseudocheiridsand phalangerids.
Dental morphology within the family has
remainedrelatively conservativeover the last 25
miilion vears. but larse size differences between
fossil koalas are apparent. During the OligoMiocene, speciesof Madakoala and Perikoala and
the plesiomorphicRiversleigh koala were similar
in size to the modern species, whereas species of
Nimiokoala and Litokoala were half to two-thirds
the sizeof P. cinereus.Conversely,in responseto
Australia's changing climate and vegetation, PlioPleistocene koalas increased markedly in size,
culminating in the giant PleistoceneCundokoala
yorkensis Pledge, 1992 which was twice the size of
the modern species. The establishment of
genus
from
distinct
Cundokoala as a
Phascolarctos has been questioned (Black and
Archer 1997a) and may not be justified on the
basisof morphologicaldifferences.Consequently,
C. yorkensis has been regarded here (Fig. 1) as a
giant speciesof Phascolarctos. Similarly, features
used to distinguish Phascolarctos maris Pledge,
1987b from P. stirtoni Bartholomai, 1968 may
prove to fall within the boundaries of expected
intraspecificvariation of the latter. However, until
this issue is resolved (Black, in prep.), they are
lreatedas distinct species.
Current understanding of phascolarctid
relationshipsbased on dental morphology (Black
and Archer 1997a) are presented in Figure l.
Species of Phascolarctos are the most derived
phascolarctids and are most closely related to
speciesof litokoala. A new genus and species of
koala from Riversleigh, northwestern Queensland
(Black, unpubl.) is the most plesiomorphic koala.
Previous analyses(Woodburne et al. 1987a; Black
and Archer 1997a) tentatively include species of
Koobor as plesiomorphic phascolarctidsyet raise
doubts about their position within the family.
Pledge(1987a)suggestedKoobor is more closely
aligned rvith ilariids than phascolarctids but
Myers and Archer (1997) found little support for
this inclusion. Clarification of Koobor's position
within Phascolarctomorphiarequires discovery of
more completematerial.
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
17
Holocene
1 0 , 0 0 0- 0
Pleistocene
2 - 10,000
Pliocene
5 . 2- 2
L. Miocene
r0.4- 5.2
M. Miocene
16.3- 10.4
E. Miocene
23.3- 16.3
L. Oligocene
26 - 23.3
3
\4,
BrF*
-s**
.s!sFss
**r*.3{=
*
8+
R
fi
.x€bs.:s
€
Phascolarctidae
time and a cladogram of relationships within the group. Fach block
Fig. ./. Koala diversity through
-Blocks
that extend without breaks through more than one time period
refresents a distinct species.
indicate apparently long-surviving species.
18
AUSTRALIANMAMMALOGY
DIVERSITY AND EVOLUTIONARY RELATIONSHIPSOF ILARIIDS,
WYNYARDIIDS, VOMBATIDS AND RELATED GROUPSOF MARSUPIALS
NEVILLE PLEDGET''
TRoy MyERSt, MlcH^q,EL ARcHERt't, ALAN KRIKMANNT ^q.ND
tschool of Biological Science. University of New South llales, Sydney NSW 2052; 2Australian
'South Australian Museum.North Terrace, Adelaide 5000
Museum, 6-8 College St, Sydney N'SW2000:
PRIOR to their discovery at Riversleigh, ilariids,
with koala-like selenodont molar morphology,
were known only from late Oligocene (25.5 mya)
central Australian deposits: Ilaria illumidens,
from the Pinpa Local Faun4 Lake Pinpa; and 1
lawsoni, from the Ditjimanka Local Fauna, Lake
Palankarinna (Tedford and Woodbume 1987;
Woodburne et al. 1994). Kuterintja ngama
(Pledge 1987), represented by one upper molar
from the Ngama Local Faun4 Mammalon Hill, Lake
younger
at
was
slightly
Palankarinn4
approximately 24.8 mya. The geographic range of
the latter species was significantly extended with
discovery of indistinguishable cranio-dental
material from White Hunter Site at Riversleigh,
enabling tentative correlation of the central and
northern Australian sites (Myers and Archer 1997).
No early to late Miocene sites at Riversleigh
(SystemsB and C; Creaser1997) have produced
ilariids, despite extensive investigation of
sediments this age. While ilariid diversity may
have been higher during the early stagesof the late
Oligocene, it apparently declined towards the
latter stages resulting in family-level extinction
by the beginningof the Miocene.
i n v o l v i n g phascolarctidsand vombatomorphians
(Fig.1).
The genotypic wynyardiid is a unique
specimen, Wynyardia bassiana, comprising part
of the skeleton and skull but lacking any teeth,
Consequently, other species are only referred to
this family on the basis, initially, of similar
postcranialmorphology, and subsequentlydental
similarity. There are thus at least six species in
three gener4 found in one or more of four
localities: Tasmania(lYynyardia, Spencer 1901),
early Miocene, about 20-21 my); Frome Basin,
South Australia (Muramura sp. nov., Pinpa LF;
Namilamadeta snideri (Rich and Archer 1979),
Tarkarooloo LF); Lake Palankarinn4 South
Australia (Muramura williamsi (PIedge 1987),
zoneA Minkina LF Namilamadeta sp. indet., zone
D Ngama LF), and Riversleigh (2-3 Namilamadeta
spp. nov., Systems A and B), and ranging in age
from late Oligoceneto early Miocene.
The skeletons of Muramura spp. allowed
comparison with lVynyardia (Tedford et al. 1977).
and their teeth with Namilamadeta (Rich and
Archer 1979; Pledge 1987). Skulls are known for
all three gener4 but basicranial evidence for
is still to be assessed.Dentally,
interrelationships
Ilariids are considered to be the primitive
the
Wynyardiidae
and
Muramura appearsto be more plesiomorphic than
of
sister group
Diprotodontoidea (e.g., Munson 1992). Piedge Namilamadeta, and the skull, too, is more
(1987) consideredK ngama to be ancestralto the
primitive. The considerably smaller skull of
ll/ynyardia
more
resemblance to
PliocenespeciesKoobor jimbarratti which rvould
shows
Namilamadeta from some Riversleigh sites, but
then be regarded as an ilariid rather than a
phascolarctid. Myers
and
Archer (1997)
still differs from all species.In general terms. the
considered the shared dental features of Koobor
skulls of wynyardiids show remarkable gross
Kuterintja
ngama
spp.
and
to
be
similaritiesto those of the plesiomorphic wombat
llarendja wakefieldi (Pledge 1992). Postcranially,
symplesiomorphic. Koobor was found to be a
primitive sister group of wynyardiids plus
Il'ynyardia and Muramura species are similar
ilariids. Discovery of lower cheekteeth should
where they can be compared, and show notable
's
resolveKoobor position within Vombatiformes. similarities with wombats. althoueh thev are more
plesiomorphic.
In the meantime, we have left the position of
Koobor as part of an unresolved polychotomy
19
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
Holocene
10,000-0
Pleisto€ne
2 - 10,000
Plioceoe
5 . 2- 2
L. Mioceoe
i0.4 - 5.2
M. Miocene
r 6 . 3- 1 0 . 4
E. Mioene
lJ.J
- to,J
L. Oligocene
3
3P5
t' -d E9 -S- E
F
g : s - EE
.
3
e n t tc
3
=i
Sir
a
is€'?
riHia*>-e
\i
i i
I
h
i:
=
t-E
s!:-:e:5
S :
E
€
-:\f
s ii=
Ei
SsS F 5i trsasiS
s
i
i
=
E Z
3 E
nA
Frg /. Diversitv through time and a cladogram of probable interfamilial and intergeneric relationshipsof ilariids,
wynyardiids, vombatids, thylacoleonids and related groups of marsupials.Uncertain relationshipsare indicated by
dashid lines. Each block representsa distinct species.Blocks that extend without breaks through more than one time
period indicate apparentlylong-surviving species.
Phylogenetic relationships among vombatoids
remain unresolved. Work in pro-9ress (Krikmann
1993; M. Archer and A Krikmann) suggests there
were three families. The first, from early Miocene
Riversleigh depositsand the late OligocenePinpa
'Vombatoidea
Local Fauna(Tedfordet al. 1977, as
Genus A'), is represented by primitive, lowcrowned, browsing wombats with complex molar
morphology and rooted teeth. The second includes
Rhizophascolonus crowcrofti and at least two
other wombat species with rooted but highcrowned teeth from the early to middle Miocene of
Riversleigh and the central Australian Kutjamarpu
Local Fauna. The third. Vombatidae, includes a
number of grazing wombat specieswith unrooted
teeth, the earliest of which is a new species of
Warendja from the late Miocene Encore Locai
Fauna of Riversleigh. However, the greatest
diversity of vombatids did not occur until the
early Pliocene (Phascolonus and Ramsayia) and
Pleistocene (l[/arendja, Phascolomys, Vombatus,
Lasiorhinus), with the increase paralleling the
drying out and opening up of the continent and
consequentproliferation of grasslands. There are
three extant vombatids, species of Vombatus and
Lasiorhinus, one of which (lasiorhinus krefftii)
is seriouslyendangered.
AUSTRALIAN MAMMALOGY
DIVERSITY AND RELATIONSHIPS OF DIPROTODONTOID MARSUPIALS
I(ARENBLACKANDBRIANMACI.I\ESS
Universitvof New SouthWales.SydneyNSW 2052
Schoolof BiologicalScience.
T\\'O families of diprotodontoids are currently
recognised: Palorchestidae(2 genera,7 species);
and Diprotodontidae containing Zygomaturinae
(13 genera-25 species) and Diprotodontinae (8
genera. 12 species). A third diprotodontid
subfamily. the Nototheriinae (Stirton et al. 1967),
is no longer consideredvalid (Archer 1977).A yetto-be-described Pleistocene senus with two
species that have both zygomaturine and
diprotodontine features may constitute a third
diprotodontid subfamill'. The time at rvhich
and Diprotodontidaediverged from
Palorchestidae
a common ancestor is currentll not known. but
both families were distinct bv the late Olisocene
(Fig.I ).
Iloiene
10,000 - 0
Pleislocene
2 - 10,000
Pliocmc
L. Mioccnc
10.4- 5.2
M. Mioceae
1 6 . 3- 1 0 . 4
E. Miocene
/5.1
- LD.t
L. Oligocoe
lo
- /),5
;ss;Sisss3sis
sissssEsss
Diprotodoltoid€
Frg. 1. Diprotodontoiddiversitv throughtime and a cladogram of probable interfamilial and intergeneric relationships.
Uncertain relationshipsare indicated bydashed lines. Each block representsa distinct species. Blocks that extend
rvithout breaksthrough more than one time period indicate apparenllylong-survivingspecles.
Zygomaturine diprotodontids exhibit a higher
than either
diversity in Tertiary fossil assemblages
palorchestids possibly
diprotodontines or
indicating an ability to exploit a greaterrange of
environments (Black 1997b). Palorchestids are
rare componentsof fossil assemblagesrvhich may
indicate a solitarl habit. Normally, no more than
one species is present in faunal assemblages
s p a n n i n gt h e l a s t l 5 m i l l i o n y e a r s .S i m i l a r l l .
diprotodontines, although prolific in the late
Oligoceneand early Miocene. are rare throughout
the rest of the Miocene but then diversify again in
the Plio-Pleistocene.Drying of the Australian
continent resulted in development of gigantic,
highll specialisedforms characterisedby features
useful for processingcoarserfood (e.g. diastemal
crestsand elaborationof blades in the dentition).
In contrast, the New Guinea diprotodontoid
ARCHERETAL, EVOLUTIONOF AUSTRALIANMAMMALS
'du'arf z)'gomaturinespecies
radiationconsistsof
variously interpretedas either relict populations
from MioceneAustralia (Flannery 1988) or a midto-late Pliocene colonisation by a ferv derived
Australian genera(Murray 1992). The reasonsfor
the
entire
of
the
eventual extinction
diprotodontoid radiation by the end of the
Pleistocenehas yet to be determined but it rvas
probably a combination of factors including
interactionwith early humans.modification of the
Austraiian environment by fire (rvhich favoured
the developmentof grasslands)and natural cycles
o f c l i m a t i cc h a n g e .
phylogenetic
the
Interpretation
of
relationshipsof diprotodontoids is hamperedby
the fact that many taxa are known only from
isolatedelementsand there is a generalpaucity of
late Mioceneand early Pliocene material.Analysis
of variation in severalgroups has revealedsexual
dimorphism and ontogenetic variation which has
not been taken into account in some taxonomic
been an
also
assignments. There has
underestimationof diversity in the late Tertiary
'waste
with assignmentoften made to taxonomic
bins' like'notothere'.The current understandingof
phylogenetic relationships is shown in Figure L
Monophyly of the group has been questioned
21
(Archer 1984; Aplin and Archer 1987; Murray
1990a). there being few synapomorphies
recognised. Within Diprotodontidae. zygomaturines appear to have been derived from a
primitive diprotodontine-like form in lvhich the
parastyle on P3 was reduced. a theory" first
proposed by Stirton et al. (1967) and later
confirmed bl the discoverv and analysis of
dentitions of Silvabestius michaelbirti from
Riversleigh (Black and Archer 1997b). This
speciesis believed to be antecedentto ihe entire
zygomaturineradiation(Black and Archer 1997b).
Previousauthors (e.g.Aplin and Archer 1987;
Marshall et al. 1990) include Ngapakaldia and
Pitikantia within Palorchestidaeon the basis of
similarities in the basicranium. features since
recognised(e.g.Archer 1984; Murray 1986,1990a)
as plesiomorphic for Vombatomorphia. Murray's
(1990b)
of
Propalorchestes
description
it
is
the
dentitions demonstrates that
plesiomorphic sister group of Palorchestes,
exhibiting subselenodont molars transitional
between selenodont r"'ynyardiids and the fully
Palorchestes.
molars
of
bilophodont
Consequently, Ngapakaldia and Pitikantia are
currently recognisedas primitive diprotodontines
(Murray 1990: Black 1997).
DIVERSITY AND EVOLUTIONARY RELATIONSHIPS OF MARSUPIAL LIONS
ANNA GILLESPIE
School of Biological
Science, UniversiN of New South Wales, Sl,"dneyNSW 2052
THE marsupial lion family Thylacoleonidaehas a
fossil historv that spans the late Oligocene
through to late Pleistocene.There are three genera
and a total of nine species(Myers et a/. this rvork,
fig. l). The oldestmemberof the family is the late
Oligocene cat-sizedPriscileo pitikantensis from
central Australia (Rauscher 1987). A second.
smaller species.P. roskellyae-has been described
on the basisof the dentitionand a nearly complete
skull from early Miocene sediments of the
Riversleigh World Heritagearea(Gillespie 1997)
The dog- to leopard-sized species of Wakaleo
include Il/. oldfieldi from the early Miocene
Kutjamarpu Local Fauna of South Australia and
at Riversleigh, ll/. vanderleuri
faunal assembla-ees
(also known from a complete skull) from the
middle Miocene Bullock Creek Local Faunaof the
Northern Territory. and llt. alcootaensis from the
late Miocene Alcoota Local Fauna of the Northern
Territory (Clemens and Piane i974: Archer and
Rich 1982; Murrav et al. 1987). A fourth species
a smaller.more plesiomorphicform is
representing
currently being described(Gillespie in press) from
late Oligocenesedimentsat Riversleigh (Archer er
al. 1997c). The dog- to lion-sized species of
Tfuilacoleo (Wroe el a/. submitted) include T
crassidentatusand I hilli, both fiom widespread
earll Pliocene deposits, and T. carnifex w'hich is
very widespreadin Pleistocenedeposits (Archer
and Dawson 1982).
Lineages of both ll/akaleo and Thylacoleo
demonstratemorphoclinal changes throu-eh time
that link othenvise distinct species. These
changesgenerally include an increasein size, an
increasein the proportional length of P3/3, and
loss or reduction of the posterior molars (Murray
and Megirian 1990: Archer and Darvson 1982),
Although Tertiary marsupiallions are not common
as fossils (a fact presumably reflectinq their
ecological role as carnivores), *'hat is known
suggests that up to t\\'o different-sized species
22
AUSTRALIAN MAMMALOGY
coexisted in single ecosystems. For example, in
the late Oiigocene to early Miocene deposits of
Riversleigh,a small speciesof Priscileo coexisted
rvith a much larger speciesof llakaleo. Similarly,
in the Bow Local Fauna of New South Wales, the
small Thylacoleo hilli coexisted rvith the much
Iarger T. crassidentatus (Archer and Dau,son
r9 8 2 ) .
Hypothetical intrafamilial relationships for
thylacoleonids are shoun as pan of Figure I in
Myers er a/. (this u,ork). Understanding of these
relationshios has chansedwith discoverv of new
Riversleigh taxa. Previousiy the structural
differences between species of l'I/akaleo and
Thl,lacoleo, especially loss of P1 and F in
Wakaleo,did not support an ancestor/descendant
relationshipbetrveenthe two genera(Clemens and
Plane 1976). However,the presenceofthese teeth
in a nerv species of Ll/akaleofrom Riversleigh
makes it possible to argue that this lineage was
ancestralto the speciesof Thylacoleo. In additionspecies of Priscileo exhibit no features that
prevent them from being ancestralto species of
Ly'akaleo and Th;,,lacoleo.
DIVERSITY AND EVOLUTION OF PHALANGERID,EKTOPODONTID,
MIRALINID AND PILKIPILDRID MARSUPIALS
KTRSTE\ CRosB\'t, HENK GODTHELpT,Mrcu-lEL ARCHERt'tA\D NEVTLLEPLEDcE''
zAustralian
of Biological Science, Lniuersity of New South ll/ales, Sydnqt NSII 2052;
Museum
6-8 College St, Sydney NSW 2000;' South Australian Museum,North Terrace,Adelaide, SA 5000
tschool
THE superfamily'Phalangeroideacontains at least
three families: Phalangeridae, the extinct
Miralinidaeand Ektopodontidae. Pilkipildridae
mav also belong here, but there is significant
doubt about its relationships (e.g.. Archer et al.
1987:Brammall and Archer this work).
Phalangeridsare found throughout much of
Me-qanesia. There are six genera: the bear cuscus
Ailurops; cuscusesPhalanger. Strigocuscus and
Spilocuscus; brush-tailed possums Trichosurus',
and
scaly-tailed possums Wyulda. Four
phalangerid specieshave been describedfrom the
fossil
record: Trichosurus dicksoni and
Strigocuscus reidi from Miocene assemblagesin
the Riversleigh World
Heritage area of
northwestern Queensland (Flannery and Archer
1987) and Trichosurus hamiltonensis and
Strigocuscus notiaLis from the early Pliocene
Hamilton Local Fauna of Victoria (Flanneryel a/.
1987b). A species of Wltulda has also been
reported but not named from the Miocene
sedimentsof Riversleigh (Archer et al. 1994).
Miralinids include species of lvliralina from
Oligo-Miocene sediments in central Australia
(Woodburne et al. 1987b) and an undescribed
genus and species from the early Miocene of
Riversleigh.
Four genera of ektopodontids have been
described: Ektopodon and Chttnia from OligoMiocene assemblages in
central Australia
(Woodburne and Clemens 1986a; Woodburne el
al. 1994); and Darcius from the early Pliocene
Hamilton Local Fauna of Victoria (Rich 1986).
Another undescribed genus is known fronr
Riversleigh. An unnamed and as yet genericalll,
unassessedektopodontid is present in the early
Pleistocene
PortlandLocal Faunaof Victoria.
Trvo genera of pilkipildrids have been
described: Djilgaringa from Riversleigh and
central Australia; and Pilkipildra from central
Australia (Archer et al. 1987). Additional
potentially distinct pilkipildrid material from
Riversleigh is currently under study' by M.O.
Woodburne.
phalangerids,
Relationships
between
miralinids. ektopodontids and pilkipildrids are
u n c l e a r1 F i _ el.) . H i s t o r i c a l l y .t h e v e r y s t r a n g e
teeth of ektopodontids made it difficult to place
them phylogenetically. With discovery of the
'conventional' molars
more
of species of Chunia
and the less specialiseddentitionsof miralinids. it
becameapparentthat ektopodontids were at least
distantly related to phalangerids(Tedford et a/,
Woodburne and
1986a;
197'7
Clemens
Miralinids and
Woodburne et al. 1987b).
ektopodontids are no\\ thought to be sister
groups. and together form the sister group of
phalangerids(Aplin and Archer 1987; Woodburne
et al. 1987b). Pilkipildrids are considered
incertae sedis by Archer er a|. (1987). They
suggested that this familf is either the sister
or^rrn
nf
neferrrnid<
phalangeroids or
trichotomy"
the
part
of
sisler
an
qroun
of
unresolved
Relationships u'ithin Ektopodontidae and
Phalan_seridae
are still contentious. Chunia is
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
considered to be the most plesiomorphic
ektopodontid. and Ektopodon and Darcius are
derived sister groups (Woodburne and Clemens
1986b). The neu' genus from Riversleigh ma)' put
these relationships in doubt. Phalangeridswere
originally separatedinto four genera: Ailurops,
Phalanger. Trichosurus and Wyulda. Phalanger
has since been considered paraphyletic and split
into three qenera: Phalanger. Strigocuscus and
23
Spilocuscus(Flannerl'et al. 1987a). Strigocuscus
rvasplacedas the sistergenus of Trichosurus plus
Il/yulda based on morphological characters(ibid).
Molecularstudies.however.placed some members
of Strigocuscusback within Phalanger (Colgan el
al.1993). The uncertainposition of Strigocuscus
gynznotis within Phalangeridae could affect
generic allocation of fossil species originalll'
placed in this genus (vr:., S. reidi and S. notialis).
Holocene
10,000 - 0
Pleistoene
2 - 10,000
Plioene
L. Mimme
t0.4 - 5.2
M. Miocoe
l6.l - 10.4
E. Mioeue
L.Oligdw
26 - 23.3
3'A3
F s - i i
{E
5
E
.f !i,
:
e*
\EE=
cr E\
=
5
!
$
;
5
6
t
E
*s
Sa\
bt-;i-:&
*=:?"i6!
*i 7 c a . i
dt=
g
E
so i"!
\t<t-i5d
I
-.... .t. ..
HKD!t6@e
Phalageroids
Frg ./. Diversity through time and a cladogramof probable interfamilial and intergeneric relationshipsof phalangcrid,
ektopodontid,miralinid and pilkipildrid marsupials.Uncertain relationshipsare indicated by dashed lines. Each block
representsa distinct species.Blocks that extend without breaks through more than one time period indicate apparenti.v
l o n e - s u r v i v i n gs o e c i e s .
.A
AUSTRALIAN MAMMALOGY
LIVING AND EXTINCT PETAURIDS. ACROBATIDS. TARSIPEDIDS AND
(MARSUPIALIA):
RELATIONSHIPS
AND DIVERSITY THROUGH
BURRAMYIDS
TIME
JENxI BR{MNIALLTAND MTcHAEL ARCHER'''
tschool
of Biologicat Science. Universit.vof New Southll/ales, Sydne;t,\'SIA2052;
Museum,6-8 College Sr,Sydney L'SW2000
TliE smallest contemporarypossums of Australia
and New Guinea are placed in four families: the
(Petaurus.
Gymnobelideus.
Petauridae
(Acrobates,
Acrobatidae
Dactylopsila);
Distoechurus); Tarsipedidae (Tarsipes); and
Burramyidae (Cercartetus.Burramys) (Aplin and
living
Archer 1987, Strahan 1995). All
representatives (except Burramys parvus) are
:.4ustralian
arborealalthough some live in heath (tarsipedids
and some species of Cercartetus). All are
scansorial.The species of trvo genera (PetaurusAcrobates)are also gliders. All of these families
havea Tertiaryas well as Quatemaryrecord except
tarsipedidswhich as yet have no pre-Pleistocene
record(Fig. 1).
Holocene
1 0 , 0 0 0- 0
Pleistocse
2 - 10,000
Plicme
5 . 2- 2
L. Miocene i
10.4- 5.2
M- Miocene
16.3- 10.4
E.Mimm
ZJ.J
-
IO.J
L. Oligocene
;tgt
f.iF
"$r E 6;E: -!
:s9
?€?€;i3Et€:Ee
i $s$E
:s:
s;8,is.iliIQAe;!8EtE
€
ts
>b
fcl
ll ttl l
|
'---T-
relationships
of acrobatids,
Frg..1.Diversirythroughtime and a cladogramof probableinterfamilialand intergeneric
are indicatedby dashedlines.Each block representsa distinct
and burramyids.
Uncertainrelationships
tarsipedids
long-surviving
species
Blocksthatextendwithoutbreaksthroughmorethanonetimeperiodindicateapparently
speciespossums')
Fossil burramyids ('pygmy
include a plethoraof relatively'specialisedspecies
of Burramys.,vithplagiaulacoid premolars.all so
known from lowland rainforest
far only
assemblasesof late Oligocene(taxa in the Ngama
and diverse Riversleigh assemblages: PIedge
1987c; Brammall and Archer 1997) to early
Pliocene age (Turnbull et al. 1987b) in
25
ARCHERETAL. EVOLUTIONOF AUSTRALIANMAMMALS
Dact.vlopsila('striped possums'), modern species
of which are strictll' rainforest inhabitants, have
been found in late Oligocene to middle Miocene
The oldest representatives
depositsat Riverslei-gh.
of the genus Petaurus are from the early Pliocene
Hamilton Local Fauna in Victoria. Although the
postcranial anatomy of the Pliocene taxa is
unknown, all modern species of this genus are
gliders. Another generically distinctive group of
petaurids,from late Oligoceneto middle Miocene
deposits at Riversleigh" appears to be closel,'
related to an early Pliocene taxon from the
There are at least two undescribed acrobatids Rackham'sRoost Local Fauna of Riversleigh. At
('feather-tail possums') from early and middle
leastone petauroidis interpretedto be among the
Miocene deposits of Riversleigh that represent unnamed taxa from the early Miocene Geilston
this highly specialisedgroup. The living arboreal Bay Local Fauna of Tasmania (Tedford and Kemp
Distoechuruspennatus is a rainforest inhabitant 1998). A highly distinctive petaurid-like group,
in contrast to the glider Acrobates p)tgmaeus known from the early to middle Miocene of
which inhabitswet and dry sclerophylforests.
Riversleigh,appearsto r€presenta plesiomorphic
sistergroup to Petauridae.
The fossil record of petauridsis more diverse.
northwestern Queensland.the Tirari Desert and
western Victoria. An unnamed burramyid was
reported from the early Miocene Geilston Ba1'
Local Fauna (Tedford et al. 1975). At least one
(unnamed)Tertiary fossil speciesof Cercartelusis
known from Miocene sediments at Riversleigh.
'...similarin some ways to
Recordsof large species
the living Cercartetus...' (e.g.. Woodburne et al.
1985, p355) from late Oligocene deposits of the
Tirari Desert and Frome Embayment actually refer
to pilkipildrid possums(Crosby e/ a/. this work).
LIVING AND EXTINCT PSEUDOCHEIRIDS (MARSUPIALIA,
AND CHANGES
RELATIONSHIPS
PHYLOGENETIC
PSEUDOCHEIRIDAE):
DIVERSITY THROUGH TIME
IN
MINA BAssARovAr AND MICHAEL ARCHER''2
tschool
of New South Wales,Sydnel'I'SW 2052.: Australian
of Biotogical Science, L)-niversit.v
Museum, 6-8 College St, S1:dneyNSW2000
Tl{E family Pseudocheiridaeis comprised of ten
generaofringtail possumsand greatergliders. Six
(Hemibelideus,
Petauroides,
genera
Pseudochirops, Petropseudes,Pseudocheirzzsand
Pseudochirulzs) contain extant as rvell as extinct
species; four genera (Paljara, Pildra, Marlu and
Pseudokoala) contain only extinct species. The
highest species-level diversity for this family
occurs today in rainforest environments of
northern Australia and New Guinea. The only
living species found in relatively drier areas of
Australia, such as open forests and woodlands, are
(Pseudocheirus
Ringtail
the
Common
peregrinus), Greater Glider (Petauroides volans)
and Rock Ringtail (Petopseudes dahli).
present in a single local fauna at Riversleigh
(Archer 1992;Archer et al 1994).
The oldest-known pseudocheirids are late
Oligocenein age from the Tirari Desert and Frome
Basin of South Australia and the Riversleigh
World Heritage area of northwestern Queensland.
Theseare referableto the genera Pildra, Marlu and
Paljara (Woodbume et al. 1987c1,Archer 1992:
Archer et al. 1997a). The species of Paljara
known
most plesiomorphic
include the
pseudocheirids which appear to retain, for
example.well-developedparaconids on the first
lower molar. Current research on undescribed
Riversleigh specimens referable to Paljara
indicatesthat more than one speciesand possibly
up to threespeciesare representedby, the material.
Pseudocheiridspeciesdiversity rvas greaterin
the late Oligoceneto early Miocene than at present The species of Pildra are the next most
(Fig. l) with, for example. up to nine species p l e s i o m o r p h i ca n d a r e h i g h l y d i r e r s e i n t h e
zo
AUSTRALIANMAMMALOGY
Holocene
10,000 - 0
Pieistocene
2 - 10,000
Pliocme
5 . 2- 2
L. Miocene
10.4- 5.2
M. Miocme
1 6 . 3- 1 0 . 4
E. Miocaae
23.3 - 16.3
L. Oligocene
g:€
*t!
!s.:
Ill
€ *€
E si
€
I
si
is
rl
l
Uncertain
diversiry-throughtime and a cladogramof probableintergenericrelationships.
Fig I
Pseudocheirid
a distinctspecies.Blocksthat extendwithoutbreaks
relationships
are indicatedby dashedlines.Eachblock represents
long-surviving
species.
throughmorethanonetimeperiodindicateapparently
Oligocene to Miocene sediments of South
Australia and Queensland.Species of Marlu are
less diverse being known from possibly only
tu,o speciesin the Tirari Desert deposits, one of
which may also be representedat Riversleigh.
Speciesof Pseudokoala are known from the
early Pliocene Hamilton LF and from various
Pleistocene deposits in South Australia and
New SouthWales.Archer et al. (1997a) suggest
that the Miocene or early Pliocene genus
Corracheirus Pledge, 1992 is a junior
synonym of Pseudokoala andthat all speciesof
this genus occupied rainforest. Pseudokoala
cathysantamaria from the early Pleistocene
Portland Local Fauna of Victoria was a -eiant
megafaunalringtail that weighed possibly ten
kilograms.
genera,
ringtail
the
extant
Of
Pseudochirops has the oldest record with a
minimum of three as yet unnamed species
occurring in Oligo-Miocene deposits at
Riversleigh. Another as yet unnamed extinct
species possibly referable to Pseudochirops
occurs in the late Miocene Alcoota LF (Archer
and Bartholomai 1978). Petropseudes. which
some authorsregard to be ajunior synonym of
Pseudochiroos.has a fossil record that extends
back to the early Pliocene at Riversleigh. From the
sameearly Pliocene Riversleigh deposit also comes
the oldest record of the living species, P. dahli.
Petauroides may be representedby the early Pliocene
species P. stirtoni and marshalli (Turnbull and
Lundelius 1970; Turnbull et al. 1987a; Archer 1984).
pre-Pleistocene records
There are no
for
Pseudocheirus,
Pseudochirulusor Hemibelideus.
Considered overall, a massive radiation of
archaic ringtails clearly' characterised the late
'modern'
Oligocene to middle Miocene. The only
group representedamong these is Pseudochirops. We
suspect that our sustained failure to find Tertiary
records for Pseudochirulus and Pseudocheirus
indicatethat they were a late Miocene to Pliocene and
perhaps Pleistocene radiation in New Guinea and
Australia respectively.
Intergeneric phvlogenetic relationships of
pseudocheirids
are either unresolvedor controversial
(Fig. l). Woodburneet aL.(1987c)have proposedthat
Pildra and the common ancestor of Marlu and
Pseudokoala are sister taxa. However, Sprineer
(1993), on the basis of dental synapomorphies,
suggests in contrast that species of Marlu,
Pseudokoalaand extant pseudocheiridsmay form a
clade to the exclusion of species of Pildra and
Paljara. Relationshipsof extinct to extant ringtails
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
have not been clarified. in part because of the
incomplete fossil record.Many different fields
of research (molecular and morphological)
provide evidenceto suggestthat. among extant
genera.Hemibelideus and Petauroides are sister
taxa. Pseudochirops and Petropseudes are
sister taxa (Archer 1984; Baverstock 1984;
Baverstock et al. 1987. 1990c; Kirsch et a/.
1 9 9 7 ; M c K a y 1 9 8 4 ; M c Q u a d e1 9 8 4 ; S p r i n g e r
and
al.
1992),
et
I 993;
Springer
Pseudochirulus forms a monophl'letic group
(Archer 1984;Baverstocket al. 1990; Flannery
1994; Springer 1993; Springeret al. 1992). The
phylogenetic position of Pseudocheirus and
Hemibelideus'
between
trichotomy
the
Petauroides, Pseudochirops-Petropseudesand
Pseudochirulas remain unresolved becauseof
disagreements about the most appropriate
methods of phylogeneticanalysis.
EVOLUTION
21
Petauroids (pseudocheirids plus petaurids.l
appear to be most ciosely related to tarsipedoids
(acrobatidsand tarsipedids)(Aplin and Archer 1987;
Baverstock 1984: Baverstocket al. i987; Kirsch et
al. 1997).
The relationshipsof petauroids to burramyoids
and phalangeroids have not been clarified: see
Springer and Kirsch (1991) and Springer and
Woodburne (1989) for conflicting phylogenetic
c o n c lu s io n s .
Archer (i984) and Woodburne et al. (1987c)
have suggestedthe possibility that pseudocheirids
are the outgroup to all other phalangeridansbecause
of their retention of many plesiomorphic dental
features (e g. complete diprotodontian dental
formula- paraconid on the first lou'er molar and a
well-developedsqvlarshelf on the upper molars).
AND DIVERSITY OF KANGAROOS
MARSUPIALIA)
(MACROPODOIDEA,
BERNARD CooKE''t AND BEN KE.q,Rt
tschool Biological Science,University of New South llales, Sydney, NSII| 2052::School of
of
Natural ResourceSciences,QueenslandUniversitl, ofTechnology, GPO Box 24i1, Brisbane,
Queensland 4001
NUMEROUS attempts have been made to unravel
the complexities of kangaroo phylogeny
Character systems utilised in such analyses
include: morphological (e.g., Bensley i 903;
Raven and Gregory 1946;Tate i948; Archer 1984;
Flannery 1989); cytological (e.g.,Sharman1989):
immunological (e.g.,Kirsch 1977; Baverstock et
al. 1989, 1990b); and molecular biology (e.g.,
Westermanet al. 1990; Kirsch and Foeste 1995).
Earlier attempts. particularly those reliant on
morphological characters,were hampered by the
lack of an adequatefossil record extending much
further back than the Pliocene. That situation has
improved with discovery of older Tertiarl
kangaroosin a variety of deposits in the Northern
Territory (late Miocene deposits at Alcoota and
Bullock Creek), South Australia (deposits in the
Lake Eyre and Frome Basins) and Queensland
particularly
(Riversleigh)
Riversleigh is
important becauseit includes sites ranging from
late Oligoceneto Holocenein age.and which yield
very well preserved kangaroo fossil remains.
including complete skulls and some articulated
skeletonsfor many taxa.
Flannery's (1989) analysis of macropodoid
phylogeny is the first comprehensiveattempt to
incorporate data obtained from Oligo-Miocene
Notable among these are lophodont
taxa.
representatives of two previously unknown
groups: Balbarinae and Bulungamayinae, both
Thel'
described by Flannery et al. (1983).
Macropodidae and
assi-ened Balbarinae to
Bulungamayinae to Potoroidae. Case ( 1984)
disagrees with placement of Bulungamayinae
within Potoroidae,maintaining that lophodonty
represents a
macropodid
synapomorphy.
Flannery's(1989) phylogenetic analysis supports
placementof Bulungamavinaeas a monophyletic
group within Potoroidae,but indicates Balbarinae
as a likely paraphyletic macropodid group,
directly ancestral to both Macropodinae and
Sthenurinae.
Cooke (1997a) indicates that lophodonty has
arisen independently in these two subfamilies,
representing another example of the rampant
evolutionary convergencethat plagues kangaroo
phylogenetics. Cooke (1997a, in press) argues
that characters such as a laterally broad Mr
trigonid, absence of ornamentation of the
posterior face of the hypoloohid of lorver molars.
elongate permanent premolars and parietalaiisphenoid contact on the lateral wall of the
cranium. represent svnapomorphies unitins.
macropodids.
A
and
bulungamayines
phylogenetic analysis by
Cooke (1997b)
incorporates cranial and dental characters present
28
AUSTRALIANMAMMALOGY
in remains discovered at Riversleigh subsequent
to Flannery's (1989) analysis. It indicates that
Balbarinaerepresenta monophyletic group. basal
to all other macropodoids(and thereforeshould be
elevated to familial status as Balbaridae)" This
the
monophyly
of
analysis
supports
Hypsiprymnodontidae (Ride 1993) but indicates
that bulungamayines are members of a large,
pectinately branching clade that includes as
bulungamayinessome taxa previously regardedto
be potoroines (Bettongia moyesi, Wakiewakie
lawsoni and Purtia mosaicus), sthenurines and
macropodines. Bulungamayines appear in this
analysis to be a paraphyletic stem group,
occupying much the same position as Flannery
(1989) arguedfor balbarines.
Recent studies of cranial remains of the
propleopineEkaltadetaima by Wroe e/ al. (1998)
indicate potential synapomorphies uniting
Balbarinae and Propleopinaewhich casts some
doubt on the monophyly of Hypsiprymnodontidae(sansuRide 1993;see Fig. l).
Holocde
10,000- 0
Pleistome
2 - 10,000
Pliocme
5 . 2- 2
L. Miocene
10.4- 5.2
ivl Miocese
10.J- lu.4
E. Miceae
/J.J
-
lO.J
L. Oligoene
oi
osto
a
He9g
5470
2Qtr:
4
x
Muopodoida
Frg /. Kangaroo diversity through time and a cladogramof probableinterfamilial and subfanilial relationships.Uncertain
relationshipsare indicatedby dashedlines. Particularlyuncertainis the position of Propleopinae.Each block representsa
distinct speciesunlessspeciesdiversity is indicatedby a number.Blocks that extendwithout breaks through more than one
time period indicate apparently long-surviving species. Unindicated are many Pleistocene occurrences of Iiving
macropodidsamong the undifferentiatedPleistocenecohort.
More recent phylogenetic analysis of
macropodoids by
Kear (1998), utilising
postcranialmorphology, supports the monophyl;
of balbarines and their status as a basal
macropodoid group. However, it also suggests
that Bulungama,vinae
may be polyphyletic.
Uncertainties indicated in Figure I are a
measure of the lack of resolution about key
aspects of macropodoid phylogeny. However,
continuing discovery of new fossil material from
Riversleigh and elsewhere supplies new data
which. together with that contributed from other
fields. enable existing hypotheses to be
continuously tested. Evidence from postcranial
fossil remains of the kind routinely found at
Riversleighwill be of particularimportancein this
regard.
Current research by Kear (1998) on
postcranial
morphology
of
Riversleigh
macropodoids suggeststhat the transition from
more quadrupedal gaits to bipedal hopping
occurred during the late Oligocene to eariy
Miocene. Assessmentof postcranialremains of a
late Oli-qoceneRiversleigh balbarine (Nambaroo)
indicatesthat the forelimb to hindlimb ratio rvas
comparableto that of Hypsiprymnodon moschatus
(a habitual quadruped: Johnson and Strahan
1982). Other featuresalso suggestthat this animal
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
did not hop. These include reduced quadratus
femoris and adductor magnus muscle scars on the
femur, shortenedtibia/fibula contact and a short,
broad foot. Morphology of the forelimb also hints
at quadupedalism and. combined with a high
degreeof lateralflexibilitv in the pedal digits and
a well developed first toe, ma)" be indicative of
albeit limited climbingability.
Middle to late Miocene macropodoidssuch as
the Riversleigh bulungamayines Ganguroo
bilamina Cooke. 1997a, and a species of
l4/anburoo sp. (Cooke 1999) suggest a more
bipedally adaptedmorphology, particularly in the
pes, with a shift of digits II and III ventromediall;'
below digit IV (accompaniedby a loss of digit I).
and shortening of the navicular facet on the
astragalus. Pedal development in ll'anburoo sp.
also suggests limited capacityof the upper ankle
joint for fore-aft motion (possiblyin responseto a
transfer of support of the body weight to digit IV),
thus making the pes functionally monodactyl.
The origin of kangaroos remains a vexed
question. Most recent authors (Aplin and Archer
t987; Flannery 1987; Springer and Woodburne
29
1989; Marshall et al. 1990; Ride 1993) have
favoured a sister-group relationship between
Recent
macropodoids and phalangeroids.
cladistic analyses of kangaroos (e.g.. Flannery
1987. 1989; Cooke 1997b) have used
phalangeridsas an outgroup. although Wroe et al.
(1998) suggestthat Burramyidaeis possibly more
appropriatefor this role. thereby reneu'inginterest
i n a p o t e n t i a l r e l a t i o n s h i p f i r s t s u g g e s t e db i
Broom (1896).
What is clear is that the origin of kangaroos
must be sought further back in time than the late
Oligocene. The oldest known kangaroos, from
depositsat Riversleigh and central
Oligo-N4iocene
Australia, demonstrate an already considerable
degreeof diversity, particularly within Balbarinae
and Bulungamayinae. The record indicates a
marked and steady decline in diversity of some
macropodoids (notably balbarines) during the
Oligocene and Miocene. In contrast, there rvas
clearly an explosive radiation in macropodines
and sthenurines from iate Miocene to Quaternary
time, probably a measure of the late Cainozoic
developmentand spreadof grasslands.
AUSTRALIAN FOSSILBAT DIVERSITY AND EVOLUTION
SuzaNxrH,\xo
School of Biological Science, University of New South Wales, Sydney NSW 2052
TIIE Australian fossil bat record is one of the
world's oldest and best. Australonycteris clarkae,
recoveredfrom 55 million-year-oldsedimentsnear
Murgon, southeasternQueensland,is one of the
world's oldest bats (Hand el al. 1994). Since its
description, additional fossil material from
Murgon (including postcranial remains) clearly
indicates Australonycteris is a member of the
family Archaeonycterididae,an archaic bat group
otherwise known only from early to middle Eocene
sediments in Europe and North America. How bats
first reachedAustralia is not yet known; dispersal
could have been through either South America or
southeasternAsia.
There is a gap in the Australian fossil record
for bats (as for all non-marine mammals) between
55 and 26 million vears ago. By the early to
middle Tertiary all modem bat families had
evolved, with lhe weight of evidence now
suggestingthat many modem bat families evolved
in the Southern Hemispherein the earl-vEocene
(Sig€ 1991; Simmons 1996). The next oldest
Australian bat record is a single tooth from the 26
million-year-old Ditjimanka LF of the Lake Eyre
Basin which was describedby Archer (1978) as a
possible rhinolophid but which is
understoodto be a mystacinid(see belou').
now
Hipposiderids and megadermatidsdominate
the 25 to 4 million-year-old Riversleigh
freshwaterlimestone deposits. representingmore
than half of Australia's Tertiary bats, with
emballonurids, molossids, vespertilionids and
mystacinidscomprisingthe rest (Fig. l).
In general, Australian Oligo-Miocene bats
include ancestors or close relatives of extant
(e.g.,
Macroderma
and
Australian
taxa
Brachipposideros spp.), relatives of extinct and
extant non-Australian taxa (e.g., Xenorhinos and
Riversleigha spp.), and ta,\a representing archaic
cosmopolitan groups with no living descendants
(e.g..Petramops and Hydromops spp ) Austraiian
Pliocene bats are typically closely related to. or
representearly populations of, extant Australian
bats (e.g., Taphozous spp., Macroderma gigas,
Micronomus sp.).
A minimum of 22 hipposiderid species occur
in Riversleigh's Oligo-Miocene deposits, with
some taxa represented by many hundreds of
individuals. Hipposiderids have not yet been
30
AUSTRALIAN MAMMALOGY
recorded from Australian Tertiary deposits outside
the Riversleigh region. in some Riversleigh
deposits (e.g. Upper and Bitesantennary Sites), as
many as eight hipposiderid species (representing
four genera) appear to be syntopic. The
predominanceof hipposiderids (in both diversity
and abundance) in Riversleigh's limestone
depositsis perhapsnot surprisingsince almost all
living hipposiderids are cave-dwellersand most
extinct taxa have been recovered from cave or
karstic deposits.
t€;f=F
:Frr€a
Et
I{ipposidercs
Bractipl2osidercs/
Rhinozicaeit
New Gous
l
Jt6oelryIlorhina
Neri' 6cau
?
"Xqorhino.c
Riwers le igha
--T--
Rhinoloplt*
t!TTaphorcrc
:_l
9_-L_
€l
lVrctuderrc
EI
e_+
59.
S@lairur
Mcqad@tid
MePadem
I
iDdet
Ic@p-t
I
-r
<l
6l
=-l
3l
Petramops
Itydrcmops
*llmnom6
Chaerephon
],[yctinomops
;r
PtereIre
5l
]tEtimqe
Dobsonia
!
ft
€l
Macrcglo-rsus
SyconyckrLt
vcsp€rtiliolid
indet
Scotorepens
Chalinolobus
6,l_
9 vepertilionid
geneB
Frg 1. Australian bat diversify through time. Each block representsa distinct species unlessspecies diversity is indicated
by a number. Blocks that extend without breaks through more than one time period indicate apparently long-surviving
sDecies.
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
JI
-_G
QS
-S+sFFE
3
.htsx
=+d
q . ! >
:E€FRSF€.-S:€€:
3-FEi
S€
E A € e E * g F=€SEE€ =R':
Hipposidaidac
(afterHandandKirsch1998)
amonghipposiderids
ofpossiblerelationships
Fig.2. Cladograrn
However, high diversity also appears to have
occurred in European middle Tertiary karstic and
lacustrine deposits (commonly five species,Remy
et al. 1987), suggesting a once more diverse
hipposiderid radiation worldwide. Six genera of
hipposideridsare now describedfrom Riversleigh
(Brachipposideros, Hipposideros, Rhinonicteris,
Riversleigha, Xenorhinos and Miophyllorhina;
Sigd er al. 1982; Hand 1993, 1997a"b,c,1998a,b).
The latter four genera are known only from
Australia. Brachipposideros and Hipposideros
have much broader distributions: the former in
Oligo-Miocene sediments of western Europe and
northern Africa (Sige et al. 1994), and the latter
represented by at least 55 modern species
throughout the Old World tropics. Only three
extinct species of Hipposideros have been
described worldwide. two of these being from
Riversleigh: the early Miocene Hipposideros
bernardsigei (Hand 1997a) and the Pliocene 1L
winsburyorum (Hand and Godthelp 1999). Five
species of the Brachipposideros-Rhinonicteris
clade are represented in Riversieigh's Tertiary
sediments and the oldest representativesof the
living northern Australian R. aurantius (Sigd et al.
1982;Hand 1997b). Another three speciesofthis
clade are yet to be described from the Pliocene
Rackham's Roost deposit, and at least six others
from Oligo-Miocene sediments.
all
have been described for
Skulls
hipposiderid genera represented at Riversleigh
with the exception of Miophyllorhina. Additional,
generically distinct hipposiderid taxa have yet to
be described.
fossil
Australian
some
Although
hipposiderids are membersof highly specialised,
derived lineages (e.9., H. bernardsigei), many
seem to belong to a group of relatively
plesiomorphic lineages (e.g.. Rhinonicteris,
Xenorhinos, Brachipposideros) which represent
early (thou-eh nonetheless specialised) offshoots
of the hipposiderid radiation. Because of the
diversity of plesiomorphic hipposiderids (fossil
and living) in the Australian Region (Fig. 2), Hand
and Kirsch ( 1998) speculate that the primary
radiation of the family probably occurred here.
Eight Australian fossil megadermatids have
been described: seven from Riversleigh Tertiary
sediments and one from Wellington Caves, New
South Wales (Hand 1985, 1995, 1996:' Hand et al.
1988). Most appear related or referable to the
endemic Australian genusMacroderma but several
appear to represent quite different lineages. Of
these, Megaderma richardsi (Hand 1995), from
the Pliocene Rackham's Roost Site, is the first
Australian record of this otherwise Afro-Eurasian
genus which elsewherehas a record that extends
from the late Oligocene to Holocene. The earliest
fossil representativesof the living Macroderma
gigas also occur in the Rackham's Roost deposit
(Hand 1996). A second, large Pliocene species
(Macroderma koppa) is known from Wellington
The record for
Caves (Hand er a/. 1988).
Macroderma extends from the late Oligocene to
Recent with a maximum of two species occurring
syntopicallyduring that period.
.)-/.
AUSTRALIANMAMMALOGY
Australian Tertiary molossids are known from
Downs,
Riversleigh and also from Bluff
northeasternQueensland. These bats were present
in Australia from at least the late Oligocene.
Several species occur in Riversleigh's OligoMiocene deposits. Two species have been
described: Petramops creaseri, a plesiomorphic
taxon from the middle Miocene (Hand 1990), and a
species of the otherwise Eurasian subgenus
(Hydromops) from
the
late
Mormopterus
Oligocene and early Miocene (Hand el al. 1997).
Riversleigh's Petramops and Hydromops species
do not appear to be closely related to each other,
nor do they appear to be ancestral to any living
Australian molossid. They represent part of an
older, archaic bat fauna that was widespread, and
had close relatives in Eurasia and North America.
The modern Australian subgenus Mormopterus
(Micronomus) is represented in Riversleigh's
Pliocene Rackham's Roost Site, and the Pliocene
Bluff Downs deposit on Allingham Creek near
Charters Towers (Hand et al. 1999).
The first pre-Pleistocene record for New
Zealand's only endemic mammal family, the
Mystacinidae, has been found in Australia.
Several species of the new genus Icarops are
known from early and middle Miocene Riversleigh
deposits and another speciesfrom middle Miocene
limestone sediments at Bullock Creek, Northern
Territory (Hand et al, 1998). The presence of
plesiomorphic mystacinids in the Australian
Tertiary suggests an Australian origin for the
family. Geographic and phylogenetic data suggest
the most likely time of dispersal from Australia to
New Zealand wouid have been the middle to late
Oligocene. The Australian Miocene mystacinids
provide the only Tertiary record of a very
distinctive southernbat lineage of which a single,
threatened species (Mystacina
tuberculata)
survives in New Zealand.
deposits. This rarity may relate to relative scarcity
of non-rhinolophoidsin the Riversleighregion, or
northernAustralia, during the Oligo-Miocene or,
more likely, preservationalbiasesdue to the roost
and habitat preferencesof such taxa. To date, the
only OIigo-Miocene vespertilionid known from
Australia is a single tooth from an early Miocene
Riversleighsite. By Pliocenetime. vespertilionids
had increasedin number in the Riversleigh region
with at least four vespertilionid species
representedin the Rackham's Roost deposit: the
fragmentary remains are possibly referable to the
modem Australian genera Chalinolobus and
Scotorepens,neither of which has an older record
anywhere.The only Tertiary emballonurids known
from Australia are two Pliocene species of
Taphozous from Riversleigh's Rackham's Roost
deposit, one species of which may represent the
living Australian T. hilli from northwestern
Australia.
This Afro-Australasian genus has
Miocene records in both Africa and Europe (Butler
and Hopwood 1957; Butler 1969, 1978; Legendre
family
now
cosmopolitan
1980);
the
Emballonuridae may date back to the earliest
EoceneofEngland (Hooker 1996).
Rhinolophids have been recorded only from
Pieistocenedeposits in easternAustralia (Archer et
al. 1984) and, incredibly, pteropodids have no
fossil record in Australia. These groups appear to
be fairly recentimmigrantsto our shores,although
both have early to middle Tertiary records
elservhere.It is possiblethat in Australian tropical
palaeofaunas, such as those of Riversleigh,
hipposiderids filied the rhinolophid niche (Hand
1998b). In the case of pteropodids, competition
with marsupials for food resources may have
plaied a role in preventing their Tertiary
colonisationof Australia, or perhaps suitable food
resourcesmay not have been readily available in
Australia until the recent Dast.
Other non-rhinolophoid bats are rare and
Riversleigh's Oligo-Miocene
fragmentary in
DIVERSITY, RELATIONSHIPSAND ORIGINS OF THE TERTIARY AND QUATERNARY
RODENTSOFAUSTRALIA
HENK G0DTHELP
School of Biological Science, University of New South Wales, Sydney 2052
ENDEMIC Australian rodents make up about 25%o
of the modem species-leveldiversity of terrestrial
mammalsof that continent- All living tax4 which
comprise at least 70 species in 15 genera, are
placed in the family Muridae. Despite relatively
high modern diversity, the Australian fossil record
of murids is very limited and only known to
extend with certainty back to the early Pliocene
(Fig. 1). Even then, no Pliocene fossil assemblage
in Australia, except that from the early Pliocene
Rackham's Roost deposit of the Riversleigh
World Heritage area of northwestern Queensland,
has more than 2o/oof its soecies-level diversitv
representedbv murids.
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
JJ
Hoimme
1 0 ,0 0 0- 0
Pleistocene
2 - 10,000
L. Plimeoe
-2
E. Plieene
5 . 2-
a distinctspeciesunlessspeciesdiversityis
Frg. 1. Australianrodentdiversitythroughtime. Each block represents
iniicatedby a number.Blocksthat extendwithoutbreaksthroughmorethanone time periodindicateapparentlylongsurvivingspecies.
To date only two undoubtedlydistinct, extinct
fossil species have been named from Australia:
Pseudomys vandycki Godthelp, 1989 from the
earil' Pliocene Chinchilla LF of southeastern
Queensland; and Zyzomys rackhami Godthelp,
1997 from Riversleigh'searly PlioceneRackham's
Roost deposit. There are other undescribed
Pliocene species in the genera Leggadina,
Pseudomys and Zyzomys (part of the Rackham's
Roost assemblage) and Hydromys (from the
Floraville assemblageof northern Queensland),
and other undescribed taxa from Pleistocene
deposits around the continent. Several Australian
living genera Iack any fossil representation
Overall, this relatively poor fossil record for such
a diverse modern group may reflect biases in
rvhich traditionally favoured
collection strate-sies,
discovery of larger mammals. and problems in
taphonomy, with smaller, more fragile specimens
being less likely'to end up as fossils.
Among Pliocene rodents in Australia, the
Rackham's Roost assemblage appears to be the
oldest. A1l of its rodent taxa except one represent
murines. The exception is a dendromurine-like
animal, a group still living in Africa but once
widespread across Eurasia including Thailand,
This unexpected discovery indicates that the
initial wave of immigrants into Australia was more
complex than previouslyassumed.
The diversitl of endemicmurine generain the
Rackham'sRoost assemblagesuggestsone of two
things. Murids invaded Australia much earlier
than Rackham'sRoosttime to enableevolution of
these distinctive groups on this continenL,
Alternatively',murids may have enteredAustralia
already differentiated (somewherein southeastern
Asia) into at least this many groups. Considering
the first hypothesis, the potential for a much
earlier origin is limited by the total absence of
rodents in the late Miocene Alcoota and Ongeva
Local Faunasof the Northern Tenitory (8 and 7 mi'
respectively; e.g. Murray and Megirian 1992),
despiteyears of collecting at those sites. This does
not rule out a time of origin between latest late
Miocene and early Pliocene (7 to 5 million years
ago), a time of climatic crisis in Australia and
elsewhere(Archer et al. 1995).At present, however.
there is no way to test these altemative scenarios.
What littte is known about the generic
representationand probable habitat requirements
of the Rackham's Roost murids suggests that
murids first invaded Australia from southeastern
Asia via the Indonesian Archipelago. The earliest
invadersprobably utilised expanding corridors of
relatively dry habitat developing at the time
across north\\'esternAustralia.
The apparentabsenceof murids from Pliocene
sites in southem Australia may be attributed to
two factors.In the caseof the Bow LF of New South
Wales,it is likely to be due to taphonomic bias
againstpreservationof small mammals. ln the case
of the Hamilton LF in Victoria, it is more likely to
be related to palaeohabitat. It is generally
accepted that the Hamilton LF represents a
rainforest assemblage (Archer et al. 1995), a
'Old Endemics'
contemporaryhabitat from which
are absent.
Relationships of the dendromurine-like
rodent have yet to be determined but almost
certainly these lie outside Australia. Nothing can
be said either, with certainty, about the broader
relationships of Australia's endemic murine
genera.Molecular and genetic data suggest they
AUSTRALIANMAMMALOGY
are each other's closest relatives thereby forming a
'Australian clade' (Watts and Baverstock
distinct
19 9 5 ) .
ed. M. Archer. Surrey Beatty & Sons Pty Ltd
and the Royal Zoological Society of New
South Wales: Sydney.
Species of Rattus in the Plio-Pleistocene ApLD{. K.. BeveRsrocr. P. a,lp DoNNELLAN,S.,
1993. Albumen immunological evidence for
deposits at Floraville, northern Queensland, are
the time and mode of origin of the New Guinea
the earliest record of the genus in Australia. This
terrestrial mammalian fauna. Science in New
record postdates an explosion in diversity in
Guineal9:131-145.
southeasternAsia (Chaimaneeand Jaeger 1999). It
seemsprobable that the Floraville Rattas lineage
M.. 1976a. The Bluff Downs Local Fauna.
ARCHER.
is part of this southeastern Asian adaptive
Pp. 383-395 in The Allingham Formation and
radiation. Other endemic Australian species of
a New Pliocene Vertebrate Fauna from
Rattus appear,in contrast, to be descendants of a
Northern Queensland ed by M. Archer and M.
second wave of immigrants that arrived from New
Wade. Memoirs of the OueenslandMuseum 17.
Guinea. No representatives of this group are
known from pre-Pleistocenedeposits anywhere in
ARcHER,M., 1976b. Phascolarctid origins and the
Australia.
potential of the selenodont molar in the
evolution of diprotodont marsupials. Memoirs
ACKNOWLEDGEMENTS
of the QueenslandMuseum 17: 367-371.
WE w'ish to acknowledge the vital support our
ARCHER,M., 1977. Origins and Subfamilial
research work has had from (among others): the
(DiprotoDiprotodon
relationships of
Australian ResearchGrant Scheme; The University
dontidae, Marsupialia). Memoirs of the
of New South Wales; the National Estate Grants
QueenslandMuseum 18l. 37-39.
Scheme (Queensland);the World Heritage Unit
(Canbena); the Riversleigh Society Inc.; the
ARcHER.M.. 1978. Australia's oldest fossil bat, a
possible rhinolophid. Proceedings of the
Queensland National Parks and Wildlife Service;
Pasminco Pty Ltd; Earthwatch Australia; ICI
Royal Societyof Queensland89:23.
Australia Pty Ltd; the Australian Geographic
ARCHER,M., 1982a. Review of the dasyurid
Society; the Queensland Museum; the Australian
(Marsupialia) fossil record, integration. of data
Museum; the Royal Zoological Society of New
bearing
on phylogenetic interpretation and
of
New
South
Society
Wales;
the
Linnean
South
suprageneric classification. Pp. 397-443 in
Wales; Mount Isa Mines Ply Ltd; private
Carnivorous Marsupials ed by M. Archer.
supporters and friends including Margaret Beavis,
Royal Zoological Society of New South Wales:
Elaine Clark, Martin Dickson, Merv Griffiths, Sue
Sydney.
and Jim Lavarack and Sue and Don Scott-Orr;
colleagues, postgraduate and honours students
who have generously shared their understanding,
notably Ken Aplin, Alan Bartholomai, Alex
Baynes, Walter Boles, Robert Creaser, Cassandra
Davidson,Angela Davies, Tim Flannery,Francisco
Goin, Ken Grimes, Georgina Hickey, Mark
Hutchinson, Peter Jell, Bob Jones, Anne Kemp,
Ernie Lundelius Jr, Helene Martin, Dirk Megirian,
Ralph Molnar, Stefan Mueller-Champrenaud,
Christian de Muizon, Peter Murray, Marika Nage,
Cathy Nock, Kerry Nettle, Rosendo Pascual, Mike
Plane, Alan Rackham, Tom Rich. David Ride, Alex
Ritchie, John Scanlon, Paul Sheehy, Bernard Sigd,
Richard Tedford. Bill Turnbull, Mike Tyler. Anita
van der Meer, Arthur White. Stephan Williams,
Paul Willis and Mike Woodbume.
REFERENCES
Apr-rN, K P. eru Ancnen, M.. 1987. Recent
advances in marsupial systematics rvith a new
in
syncretic classification. Pp. xv-lxxii
Possumsand Ooossums: Studies in Evolution
A review of Miocene
ARcHER, M., 1982b.
thylacinids (Thylacinidae, Marsupialia), the
phylogeneticposition of the Thylacinidae and
the problem of apriorisms in character
Pp. 445-476 in Carnivorous
analysis.
Marsupials ed by M. Archer. Royal Zoological
Societv of New South Wales: Sydney.
Ancrnn, M., 1984. The Australian marsupial
radiation.
Pp. 633-808 in Vertebrate
Zoogeography and Evolution in Australasia
ed by M. Archer and G. Clayton. Hesperian
Press:Perth.
possums
Ringtail
1992.
ARCHER, M.,
(Pseudocheiridae, Marsupialia) from the
Tertiary depositsof Riversleigh.The Beagle 9:
A., 1978. Tertiary
ARCHER,M. a.HNBARTHOLOM,AJ,
mammals of Australia: a svnoDtic review.
Alcheringa2: l-19.
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
Australia. Proceedings of the Linnean Society
of New South ll/alesll7: 3-16.
ARclGR. M.. Buru'r-rv. i.. DoDSoN.J., HARDING,R..
FIEAD, L. AND MI,EPHY. P.. 1997b. FrOM
plesiosaurs to people: 100 ntillion years of
Australian environmental history. Stateof the
Environment Technical Paper Series (Portrait
of Australia). Departmentof the Environment:
Canberra.
M.. CLAYTON,C. AND FIAND,S.J.. 1984. A
ARCHER.
checklist of Australasianfossil mammals. Pp.
102'l-1087 in Vertebrate Zoogeography and
Evolution in Australasia ed by M. Archer and
G. Clayton. HesperianPress:Perth.
M. A\D DAwsoN, L., 1982. Revision of
ARCF{ER.
marsupial lions of the genus Thylacoleo
Gervais (Thylacoleonidae, Marsupialia) and
thylacoleonid evolution in the late Cainozoic.
Pp. 417-494 in Carnivorous Marsupials ed by
M. Archer. Royal Zoological Society of Nelv
S o u t h W a l e s : S y d n e l.
H., FIAND,S.
ARCHER.M., EVENY,R.G., GOOTHETP,
ANo Scellv, K., 1990. Yingabalanaridae, a
new family of enigmatic mammals from
Tertiary deposits of Riverslei-eh,northwestern
Queensland. Memoirs of the Queensland
Aluseum28: 193-202.
ARCHER,M., FI,aWXTNY,T. F., RITCTilE,A. AND
MolNnn. R. E., 1985. First Mesozoic mammal
from Australia - an early Cretaceous
monotreme. Nature 318: 363-366.
ARCmR.M..Goornelp,H., AND FIeNo,S.J.. 1993.
Early Eocenemarsupialfrom Australia.Kaupia
3: 193-200.
H., HAND. S.J. Ar\D
ARCFER, M., Goornrlp,
Fossil mammals of
MrcrueN, D., 1989.
northwestern
Riversleigh.
Queensland:
preliminary overview of biostratigraphy,
environmental change.
correlation and
Australian Zoologist 25: 27-65.
ARctER, M. AND FIAND,S.J.. 1987. Evolutionary
Pp. 79-106 in Koala:
considerations.
Australia's Endearing Marsupial ed b1' L.
Cronin. Reed Books: Sydney.
ARCTER.M.. HAND. S.J.AND GoDTFGI-p.H., 1988. A
Tertiary zalambdodont
of
new
order
m a r s u p i a l s .S c i e n c e2 3 9 : 1 5 2 8 - 3 1 .
ARCTGR,M., HA\D, S.J. AND Goorrnlp, H., 1994a.
Riversleigh- Reed Books: Sydney.
ARCFER,M., IIAND, S.J. AND Goorrnrp, H., 1994b.
Patternsin the historv of Australia's mammals
35
and inferencesabout palaeohabitats.Pp. 80103 in History of the Australian Vegetation ed
Press:
R. Hill.
Cambridge University
Cambridge.
S.J. A\D GoorHei-p,H., 1995.
ARCHER.M.. I-LAND,
Tertiary environmental and biotic change in
Paleoclintate and
Australia. Pp. 77-90 in
evolution, with emphasis on hztman origins ed
by E.S. Vrba" G.H. Denton, T.C. Partridge and
L.H. Burckle. Yale Universitv Press: New
Haven.
M., HA\D, S.J.,Goprrnlp, H. AND CReesen,
ARCHER,
P.. 1997c. Correlation of the Cainozoic
sediments of the Riversleigh World Heritage
fossil property, Queensland, Australia. Pp.
131-152in Actes du Congrds BiochroM'97 ed
!,t l.-p. Aguilar, S. Legendre and J. Michaux.
Ecole Pratique des Hautes Etudes Institut de
Montpellier: Montpellier, France.
F,A, JR., HAND,S.J.,MUNREY,
M., JPXXTNS,
ARCTDR.
P. ANDGoDTtrEtp,H., 1992. Description of the
skull and non-vestigialdentition of a Miocene
platypus (Obdurodon dicksoni n. sp.) from
Riversleigh, Australia, and the problem of
monotremeorigins. Pp. 15-21 in Platypus and
Echidnas ed by M. L. Augee. Royal Zoological
Society of New South Wales: Sydney.
ARcHER, M.. MLTRRAY, P., HAND, S. J. AND
Reconsideration of
GoDTHELP,H.. 1993.
monotreme relationships based on the skull
and dentition of the Miocene Obdurodon
(Ornithorhynchidae)
from
dicksoni
Riversleigh,Queensland.Australia. Pp. 15-94
in Mammalian Phylogeny, Vol. I ed by F.
Szalay. M. Novacek and M. McKenna.
Springer-Verlag:Neu' York.
N. S., 1978
ARCTE&M., PUNE, M. D. AND PLEDGE.
Additional evidence for interpreting the
Miocene Obdurodon insignis Woodburne and
Tedford, 1975, to be a fossil platypus
(Ornithorhynchidae: Monotremata) and a
the
status
of
of
reconsideration
Ornithorhynchus agilis De Vis, 1885. The
Australian Zoologist 20 9-27.
ARcrGR,M. A\D fucu, T., 1982. Results of the Ray
E. Lemley expedition. llaknleo alcoolaensis
n. sp. (Thylacoeonidae,Marsupialia). A new
marsupial lion from the Miocene of the
Northem Tenitory u'ith a consideration of
radiation in the family. Pp. 495-502 in
earl-"Carnivorous Marsupials ed by M. Archer.
Royal ZoologicalSocietyof New South Wales:
Sydney.
36
AUSTRALIAN MAMMALOGY
R.H. A\ID RiCg, T.H., 1987.
ARCFGR,M., TEDFORD.
The Pilkipildridae. a new family and four new
species of ?petauroid possums (Marsupialia.
Phalangerida)from the Australian Miocene. P.
607-627 in Possums and Opossums: Studies
in Evolution ed by M. Archer. Surrey Beatty
and Sons and the Royal Zoological Society of
New South Wales: Sydney.
A..1968. Anew fossil koala from
BARTHoLoNTAI.
Queensland and a reassessment of the
taxonomic position of the problematic species
Koalemus ingens De Yis. Memoirs of the
QueenslandMuseum l5: 65-71.
The
molecular
BAVERSTocK. P.R., 1984.
relationships of Australian possums and
gliders. Pp. l-8 in Possumsand Gliders ed by
A. Smith and I. Hume. Surrev Beattv & Sons:
Sydney.
P.R.,BIRRELL,
BAVERSToCK.
J. AND KREIC.M., I987.
Albumin immunologic relationships of the
Diprotodonta. Pp. 229-234 in Possums and
Opossums: Studies in Evolution ed M. Archer.
Surrey Beatty & Sons and the Royal
Zoological Society of New South Wales:
Sydney.
P. R., FLAN'I.rERY,T., APLDT. K,
BAvrRsrocr,
BTRRELL,
J. AND KR-EG,M., 1990a. Albumen
immunologic relationships of the bandicoots
(Perameloidea:Marsupialia) - a preliminary
report. Pp. 13-18 in Bandicoots and Bilbies.
ed by J. H. Seebeck.P. R. Brown. R. L. Wallis
and C. M. Kemper. Surrey Beatty and Sons:
Sydney.
BENSLEy,
B. A., 1903. On the evolution of the
Australian Marsupialia: with remarks on the
relationships of the marsupials in general.
Transactions of the Linnaen Society, London,
Zoology 9: 83-217.
Br-ecr, K., 1997a.Diversiry and biostratigraphy of
the Diprotodontoidea of Riverslei,sh. northwestern Queensland. Memoirs
of
the
QueenslandMuseum 4ll. 187-192.
Blncr. K.. 1997b.A new speciesof Palorchestidae
(Marsupialia) from the late middle to early late
Miocene Encore Local Faun4 Riversleigh,
northwestern Queesland. Memoirs of the
M u s e u m4 1 : i 8 1 - 1 8 5 .
Queensland
M., 1997a. Nimiokoala
Br-ncr, K AND ARCF{ER,
gen. nov. (Marsupialia, Phascolarctidae) from
northwestern
Riversleigh,
Queensland.
Memoirs of the Queensland Museum 4l: 209228.
Brecr, K A\D ARcHER,M.. 1997b. Silvabestius
z) gomaturine
gen. no\..
a
primitive
(Marsupialia,
from
Diprotodontidae)
northwestem
Riversleigh,
Queensland.
Memoirs of the Queensland Museum 4l: 193208.
BoLES,W.E.. 1995. The world's oldest songbird
(Aves: Passeriformes).Nature 374: 21-22.
BoLES,W.E., 1999. Early Eocene shore birds
(Charadriiformes: Aves) from the Tingamarra
Local Fauna, Murgon, Queensland, Australia.
Records of the ll/estern Australian Museum
Supplement57 : 229-238.
BAVERSTOCK,P.R.. KREG, M. AND BN.NrIr-, J.,
1990b.
Evolutionary relationships of
Australian marsupialsas assessedby albumin
immunology. Australian Journal of Zoology
37: 273 - 28'7.
Bor-es, W.E., Goornelp, H., HAND, S. Aro Ancsrn.
M., 1994. Earliest Australian non-marine bird
assemblagefrom the early Eocene Tingamurra
[sic] Local Faun4 Murgon, southeastern
Queensland.Alcheringa 18: 70.
Bnvensrocr, P.R., I(nErc. M., Buurlr-, J. AND
Mcrev, G.M., 1990c. Albumin immunologicai
relationships of Australian marsupials.II. The
of
Pseudocheiridae. Australian
Journal
Zoology38:519-526.
BRAMMALL,J. And ARcFER, M., 1997. A new
Oligocene-Miocene species of Buruamys
(Marsupialia,Burramyidae)from Riversleigh,
northwestern Queensland. Memoirs of the
QueenslandMuseum 4l: 247-268.
BAVERSTocK,
P. R., RrcuerusoN, B. J., Brtnrr-1, J.
Bnoov. R.. 1896. Reporton a bonebrecciadeposit
A\D KREG,M., 1989. Albumin immunologic
near the Wombeyan Caves. N.S.W.: with a
Macropodidae
relationships
of
the
descriptionof some new speciesof marsupials.
(Marsupialia). SystematicZoology 37: 38-50.
Proceedings of the Linnean Society of New
South ll/ales8l: 48-61.
Berrves, A., 1997. The absolutely last remake of
Beau Geste: yet another revierv of the
Burlrn. P. M., 1969. Insectivores and bats from
Australian megafaunal radiocarbon dates.
the Mioceneof East Africa. Pp. 1-37 in Fossil
Vertebrate
Conference on Austalasian
Vertebrates of Africa ed by L. S. B. Leakei'.
Evolution, Palaeontology and Systematics,
AcademicPress:Neu' York
Perth, Abstracts: 13.
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
Burlr& P.M., 1978. 4. Insectivores and Chiroptera. Pp. 56-68 in Evolution of African
Mammals ed by V. J. Maglio and H. B. S.
Cooke. Harvard University Press:Cambridge.
P.M. AND HoPwooD, A.T., 1957.
Burrr&
Insectivora and Chiroptera from the Miocene
rocks of Kenya colony. Fossil Mammals of
Africa 13: 1-35.
CASE,J.A., 1984. A new genus of Potoroinae
the
(Marsupialia: Macropodidae) from
Miocene Ngapakaldi Local Fauna, South
Australia, and a definition of the Potoroinae.
Journal of Paleontology58: 1074-1086.
The
Cup.nraaxre, Y. errn Jercpn. J.-J., 1999.
evolution of Rattus (Mammalia. Rodentia)
during the Plio-Pleistocene in Thailand.
Historical Biolog,t (in Press).
W.A. ANDPLANE,M. 1974. Mid-Tertiary
CLENENS.
Thylacoleonidae (Marsupialia, Mammalia).
Journal of Paleontolog, 48: 661-669.
T. F., TNN,8IC, J. A\D APLIN,
COLGAN,D., FI-ANUCNY,
K, 1993. Electrophoreticand morphological
analysis of the systematics of the Phalanger
orientalis (Marsupialia) species complex in
Papua Nerv Guinea and the Solomon Islands.
Australian Journal of Zoology41: 355-378.
31
DE VN. C.W.. 1883. On remains of an extinct
marsupial. Proceeddings of the Linnean
Society of New South lilales 8: 1I - I 5.
DEVls.C.W.,1889.On the Phalangistidaeof the
in
post-Tertiary period
Queensland.
Proceedings of the Roltal Society of
6: 1 0 5 - 1l 4 .
Oueensland
DLn. W.S., 1895. Notes on the occurrence o1
monotreme remains in the Pliocene of Ne\\
South Wales. Records of the Geological
Suney of New South Ll/ales4: I 18- l 26.
T.. 1987. The relationships of the
FLANNERY"
macropodoids (Marsupialia) and the polaritl'
of some morphological features within the
Phalangeriformes. Pp. 741-747 in Possutns
and Opossums:Studies in Evolution ed. by M.
Archer. Surrey Beatty and Sons and the Royal
Zoological Society of New South Wales:
Sydney.
FLArllTERv,
T.F., 1988. Origins of the AustraloAustralian
Pacific land mammal fauna.
Zoological Review l: 15-24.
Phylogeny of th e
FLeNr.rERv,T.F., 1989.
Macropodoidea:a study in convergence Pp.
l-46 in Kangaroos, Wallabies and RatKangaroos ed b1,'G. Grigg, P. Jarman and l.
Hume. Surrey Beatty and Sons: Sydney.
Coorcs,B.N., 1997a.New Miocene bulungamayine
FLAN\ERY,T.F., 1994. Possttmsof the l4/orld. Geo
kangaroos (Marsupialia: Potoroidae) from
Productions:Sydney.
northwestem
Riversleigh,
Queensland.
Memoirs of QueenslandMuseum 4ll 281-294.
FLANnERY.T.F.. 1997. The Future Eaters. New
Holland Publishers:Sydney.
Fossil kangaroos and
CooKE, B.N., 1997b.
on
phylogeny.
Conference
kangaroo
M. 1987. Strigocuscus
FLAN\ERv,T. ANDARCHER,
Vertebrate
Evolution,
Australasian
reidi andTrichosurus dicksoni.two new fossil
Perth,
Systemalics.
and
Palaeontology
phalangerids (Marsupialia: Phalangeridae)
Abstracts'. 2-7.
from the Miocene of northwesternQueensland.
P. 52'7-536in Possumsand Opossunts:Studies
CoorE,B. N. 1999. Ilanburoo hilarus gen. et sp.
in Evolution ed by M. Archer. Surrey Beatty
nov., d lophodont bulungamayine kan-qaroo
and Sons and the Royal Zoological Society of
(Marsupialia:
Macropodoidea: BulungaNew South Wales: Sydney.
mayinae) from the Miocene deposits of
northwestem
Riversleigh,
Queensland.
T.. Ar.cHEn,M. nm MAYNES,G., 1987a.
FLA\r.rERy.
Records of the Western Australian Museum
The phylognetic relationships of living
Supplement 57: 239-253
phalangerids (Phalangeroidea: Marsupialia)
with a suggestednew taxonomy. Pp. 477-506
CRrasrn, P., 1997. Oligocene-Miocenesediments
in Possums and Opossums: Studies in
of Riversleigh: the potential significance of
Evolution ed by M. Archer. Surrey Beatty and
topography'. Memoirs of the Queensland
Sons and the Royal Zoological Society of Nerv
M u s e u m4 l : 3 0 3 - 31 4 .
S o u t hW a l e s :S l d n e y .
DAwsoN,L.. MurnHreo. J. AND wRoE, S. 1999. The
M., 1983.
M. A\rD Pua.H'e.
Big Sink Fauna: a new lorver Pliocene FurNeRv. T. F.. ARCHER.
M i d d l e M i o c e n e k a n g a r o o s( M a c r o p o d o i d e a :
mammalian fauna from the Wellington Caves
Marsupialia)from three localities in northern
complex, Wellington, Nerv South Wales.
Australia rvith a description of trvo new
Records of the ll/estern Australian Museum
subfamilies. Bureau of Mineral Resources
Suoolement57: 265-290
38
AUSTRALIANMAMMALOGY
of Australasian
Journal
u e o D n v s l c s/ : ! 6 1 - ) v ! .
Geolog
and
Oro**r*", r. F.. AnGrE& M., RICH.T. H. AND JoNES,
R., 1995. A new family of monotremesfrom
the Cretaceousof Australia. Nature 377: 418420.
GRecoRv. W. K.. 1947. Monotremes and the
palimpsest theory. Bulletin of the American
Museum of Natural History 88: 1-52.
D. J., 1991 .
Gnnnrms. M.. WElr-s,R. T. ANDBARRTE,
Observationson the skulls of fossil and extant
echidnas (Monotremata: Tachygiossidae).
Australian Mammalogy14: 87-101.
FlexrnRv, T., TunNsuL1, w. D., fucH, T. H. V. a.xu
S.J., 1985. New Miocene megadermatids
LuNDELrus,E. L., 1987b. The phalangerids H.a,No,
(Chiroptera: Megadermatidae)from Australia
(Marsupialia: Phalangeridae) of the earlv
megadermatid
on
with
comments
Pliocene Hamilton Local Fauna, southwestern
phylogenetics. Australian Mammalogy 8: 5Victoria.
P. 537-546 in Possums and
Opossums: Studies in Evolution ed by M.
Archer. Surrey Beany and Sons and the Royal
1990. First Tertiary molossid
, r,
Zoological Society of Neq' South Wales:
"^*,(Microchiroptera:Molossidae) from Australia:
Sydney.
biogeographic
phyiogenetic
and
its
implications. Memoirs of the Queensland
Flowen, W.H., 1867. On the development and
Museum 28 1'75-192.
succession of teeth in the Marsupialia.
Philosophical Transactions of the Royal
Flqlo, S. 1., 1993. First skull of a species of
Society, London 157: 631-641.
(Brachipposideros)
Hipposideros
from
(Microchiroptera:
Hipposideridae),
Fossil
A,
1979.
AND
BeRrsolovet,
E.S.
GAFFNEy.
Australian Miocene sediments. |r[emoirs of the
of
Journal
Australia.
trionychids
of
P a l e o n t o l o g y5 3 : 1 3 5 4 - 1 3 6 0 .
QueenslandMuseum 3l: 1'79-192
FLAND.S. J., i995. First record of the genus
A' 1997. Priscileo roskellyae sp. nov.
Gu-t-Espn,
MegadermaGeoffroy l8l0 (Microchiroptera:
(Thylacoleonidae, Marsupialia) from the
Australia.
from
Megadermatidae)
Riversleigh,
of
Oligocene-Miocene
P alaeoverte brat a 24: 4'7-66.
northwestern Queensland. Memoirs of the
Queensland Museum 4l: 321-327.
Hq.xo. S. J.. 1996. New Miocene and Pliocene
(Microchiroptera)
from
megadermatids
Goornrlp, H., 1989. Pseudomys vandycki, a
on
broader comments
Australia. with
Tertiary murid from Australia. Memoirs of the
megadermatid evolution. Geobios 29 365Queensland Museum 28: 171-173.
377.
H., 1997. Zyzomys rackhami sp. nov.
GoDTFELP,
(Rodentia, Muridae) a rockrat from Pliocene FlaNo.S. J.,1997a. Hipposideros bernardsigei, a
new hipposiderid (Microchiroptera) from the
Riversleigh.
Roost
Site.
Rackham's
Miocene of Australia and a reconsiderationof
northwestern Queensland. Memoirs of the
the monophyly of related species groups.
Queensland Museum 4l: 329-333.
Geowissenschaftliche
Mrinchner
Goorrci-p, H., ARCIIER,M., Cirtr-r-t. R., HaNo, S.J.
Abhandlungen A 34: 73-92.
AND GlrKESoN,C. F.. 1992. Earliest known
FIAND.S. J.. 1997b. New Miocene leaf-nosed bats
Australian Tertiary mammal fauna. Nature
(Microchiroptera:
from
Hipposideridae)
256: 514-516.
Riversleigh, Queensland. Memoirs of the
Goorurlp, H., wRoE, S. AND ARCmR, M., 1999. A
Queensland Museum 4l: 335-349.
new marsupial from the early Eocene
1997c.
Miophyllorhina
FIAND, S.
J.,
Murgon.
Fauna of
Tingamarra Local
riversleighensisn. gen. et sp., a new Miocene
southeastern Queensland: the prototypical
(Microchiroptera:
bat
leaf-nosed
Australian marsupial? Journal of Mammalian
Riversleigh,
from
Hipposideridae)
E v o l u t i o n6 : 2 8 9 - 3 1 3
Queensland. Memoirs of the Queensland
Corr. M., 1988. A Tertiary marsupial mole
M u s e u m4 l : 3 5 1 - 3 5 4 .
(Marsupialia:Notoryctidae) from Riversleigh.
HAND,S. J., 1998a.Xenorhinos,a new genus of Old
northeastern Australia and its bearing on
World leaf-nosed bats (Microchiroptera:
phylogenetics.
Unnotoryctemorphian
Hipposideridae)from the Australian Miocene.
published Honours thesis, School of
Journal of VertebratePaleontology l8: 430Bioiogical Science.University of New South
439.
Wales: Sydney.
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
Hr,No.S. J., 1998b. Riversleigha *-illiamsi gen. et
sp. nov., a large Miocene hipposiderid
(Microchiroptera)
from
Riversleigh,
Queensland.Alcheringa 22: 259-276.
HAND, S.J., ARCHER.M. ,qrp GoDrLtrLP,H.. 1997.
First record of Hydromops (Microchiroptera:
Molossidae) from Australia: its biocorrelative
significance.Pp. 153-162in Actes du Congrds
BiochroM'97 ed b,v J.-P. Aguilar, S. Legendre
and J. Michaux. Mdmoires et Travaux de
I'E.P.H.E., Institut de Montpellier 21.
HAND. S.J.. DAwsoN. L. e'rp Aucan, M., 1988.
Macroderma koppa, a new Pliocene species of
(Microchiroptera:
vampire
bat
false
Megadermatidae)from Wellington Caves.New
South Wales. Records of the Australian
M u s e u m4 0 : 3 4 3 - 3 5 1 .
First
H,.ND, S.J. AND GoDTHELP,H. (1999).
Australian Pliocene species of Hipposideros
(Microchiroptera:
Hipposideridae) from
Australia. Records of the l4/esternAustralian
Museum Supplement57: 299-306.
HA\D. S.J. ANID KIRSCH.J.A.W., 1998. A southern
origin for the Hipposideridae (Microchiroptera)? Evidence from the Australian fossil
record.
Pp. 72-90 in Bat Biology and
Conservation ed by T.H. Kunz and P.A. Racey
Smithsonian Institution: Washington.
39
and the
wallaroo (Macropus robustus)
phylogenetic
relationship
among
Monotremata. Marsupialia. and Eutheria.
Proceedings of the National Academl, of
Science USA 94: 1216-1281.
JoHNSoN,P.M. AND SrneseN, R. 1982. A further
description of the Musky Rat-kangaroo,
Hypsiprymnodon moschatus Ramsay, l8'16
(Marsupialia, Potoroidae). rvith notes on its
biology. Australian Zoologist 2l: 309-328.
rcaR, B. P.. 1998. Postcranial morphology and
phylogenetics of Oligo-Miocene kangaroos
(Marsupialia:
from
Macropodoidea)
northwestern
Riversleigh
Queensland.
Unpublished Honours thesis, School of
Biological Science.University of New South
Wales: Sydney.
Krrr-eN-JewonowsKA, 2., CRowrox, A. w. AND
F. A lr.. 1987. The origin of eggJEI..'KrNS,
laying mammals. Nature 326: 871-873.
Z.,CrcPtrT, R. L. AND LUO, X.,
I(ELAN-JAWOROWSKA,
1998. Alleged Cretaceous placental from
down under. Lethaia 3l: 26'7-268.
KInscu,J. A. W., 1977. The comparativeseroiogy
of Marsupialia, and a classification of
marsupials. Ausvalian Journal of Zoology
Supplement Series 52: 1-152.
KRscH, J. A. W. AND FoESrE,A., 1995. Resolution
HAND. S.J., MACKNESS,B.. WLKINSON,C. AND
of portions of the kangaroo phylogeny
WILKTNSoN,
D. (1999).First AustralianPliocene
(Marsupialia: Macropodidae) using
DNA
from
the
sp.
molossid bat: Mormopterus
Biological Journal of the
hybridization.
Faun4
southeastern
Local
Chinchilla
Linnean Society55: 309-328.
LVestern
the
Queensland. Records of
Australian Museum Supplement 57l, 291-298.
KrRScH,J. A. W., LaponrrE, F. J. Ar\D Spnncen. M.
DNA-hybridisation studies of
S., 1997.
FIAND.S.J., MURNaY, P. F., MECruEI, D., ARCHER.
marsupials and their implications for
Mystacinid
bats
H.,
1998.
M. AND GoornEI-p,
Australian
metatherian classification.
(Microchiroptera) from the Australian Tertiary.
Journal of Zoologlt 45: 21 1-280.
12:
538-545.
Journal of P aleontology
KTRSCH,
J. A W. At'rp Mercn, G. C., 1998. The
Fle,No, S.J.. NovAcEK. M.J., GoDTTELP,H. AND
platypus is not a rodent. Philosophical
M., 1994. First Eocene bat from
ARcFTER,
Transactions of the Royal Society of London
Australia. Journal of Vertebrate Paleontologl'
(B) 353: 1221-1237.
14:375-81.
M.,
KRAEwsKr, C., Blacrrr, M. a.xro WESTERMAN.
HooKER,J. J.. 1996. A primitive emballonurid bat
multigene
assessment of
1997a. A
(Chiroptera, Mammalia) from the earliest
phylogenetic relationships in the dasyurid
Eocene of England. Palaeovertebrata 25'.
Molecular
Sminthopsinae.
subfamily
287-300.
Physiology and Evolution 8: 236-248.
Hu. Y., WeNc. Y., Luo, Z. AND LI. C.. 1997. A new
KRAJEwSKT,C., P,utrrn, J., BUcKLEY, L. AND
symmetrodont mammal from China and its
WEsrelw*. M., 199,1. Phylogenetic structure
Nature
mammalian
evolution.
for
implications
of the marsupial family Dasl'uridae based on
390 13'7-142.
cytochrome b DNA sequences. Journal of
Mammalian Evolution 2: 25-35.
Je.Nrs. A. Xu. X. ANo ARNASoN,U., 1997. The
complete mitochondrial genome of the
40
AUSTRALIAN MAMMALOGY
KRAJEwSKT.
C., Yorruc. J.. BucKrEy. L.. WooLLEY,P.
M.. 1997b. Reconstructing
A. ANDWESTERMAN.
dasyurine
the taxonomic radiation of
marsupials with cytochrome b, 12S rRNA, and
protamine Pl
gene trees. Journal of
Mammalian Evolution 4: 217-236.
Kntruraarw.A.. 1993. Systematics of the OligoMiocene wombat-like marsupials from
Riversleigh(Queensland)and their bearing on
wombat origins. Unpublished Honours thesis,
School of Biological Science.University of
New South Wales: Sydney.
LEGENDRE,
S., 1980. Un chiroptere emballonuridd
dans le N6ogene d'Europe occidentale;
pal6obiogeographiques
considdrations
G e o b i o sl 3 : 8 3 9 - 8 4 7 .
on a cooling planet.Geological Society of
Australia Abstracts 44: 52-53.
McKa.v, G.M., 1984. Cytogenetic relationships of
possumsand gliders.Pp. 9-16 in Possumsand
Gliders ed by A. Smith and I. Hume. Surrev
Beatty & Sons: Sydney.
McQuepE,L.R.. 1984. Taxonomic relationship of
the Greater Glider, Petauroides volans, and the
Lemur-like Possum. Hemibelideus lemuroides
'Possums and
Gliders' eds A
Pp. 303-3l0 in
Smith and I. Hume. Surrey Beany & Sons Ptv
Ltd: Sydney.
MENG, J. Arrn Wvss, A R.. 1995. Monotreme
low-frequency
hearing
affinities
and
suggested by multituberculate ear. Nature
37i: 141-144.
LucKETr,W.P., 1993. An ontogenetic assessment
MrssER. M., WEISS, A S., Suew. D. C. AND
in dental homologies in the therian mammals.
Evolution of the
WEsrEnrraeN,
M., 1998.
Pp. 182-204 in Mammal Phylogeryt: Mesozoic
monotremes: phylogenetic relationship to
MonoMultituberculates,
Differentiation,
marsupials and eutherians, and estimation of
tremes,Early Therians, and Marsupials ed by
divergencedates basedon lactalbumin aminoF.S. Szalay^M.J. Novacek and M.C. McKenna.
of Mammalian
acid sequences. Journal
Springer-Verlag:Nerv York.
Evolution5:95-105.
MACKNESS,B., ARCHER, M. Ar\D MLnRIDAD. J.,
J., 1993.The systematics
and evolution
MUIRHEAD,
1993. An enigmaticfamily of marsupialsfrom
of Recentand fossil bandicoots (Marsupialia;
the Early PlioceneBluff Downs Local Faunaof
International
Sixth
Peramelemorphia).
Conference on
northeastern Queensland.
Theriological Congress. Sydney Abstracts:
Evolution.
Vertebrate
Australasian
21s-216.
Palaeontology and Systematics. Adelaide,
Abstracts: 17.
MURHEAD,J.. 1994. Systematics, evolution and
palaeobiologyof Recentand fossil bandicoots
MAHoNEy.J.A. AND RDE, W.D.L., 1975.v Index to
(Peramelemorphia,
Marsupialia). Unpublished
the genera and species of fossil Mammalia
Ph.D dissertation, University of New South
described from Australia and New Guinea
Wales: Sydney,463 pp.
between1838 and i969. WesternAustralian
Museum Special Publication 6: 1-247.
Mlu.Heeo, J., 1997. Two new thylacines
(Marsupialia: Thylacinidae) from
early
MARSSALL.B.. 1992. Late Pleistocene humar.
Riversleigh,
sediments
of
Miocene
exploitation of the platypus in southern
northwesternQueenslandand a revision of the
Tasmania. Pp. 268-276 in Platypus and
of
the
family
Thylacinidae. Memoirs
Echidnased by M. L. Augee. Royal Zoological
Museum 4l: 367-377.
Queensland
Socieryof New SouthWales: Sydney.
Mt.iRHeAD,J., DewSON,L-, AND ARCHER,M., 199.7.
N4ARSnALL,
L.G., CASE,J.A. AND WooDBi,RNE.M.O.,
Perameles bowensis, a new species of
1990.
Phylogenetic relationships of the
Perameles (Peramelemorphia, Marsupialia)
families of marsupials. Current Mammalogy
from the Pliocene Faunas of Bow and
2: 433-502.
Wales.
Wellington
Caves, New South
Proceedings of the Linnean Socieryt of New
MARTTN,H.A., 1998. Tertiary climatic evolution
South Wales ll7: 163-173.
and the development of aridity in Australia.
Proccedings of the Linnean Society of New
Yarala
MLRHEAD, J., AND FreN. S., 1995.
S o u t h l Y a l e s1 1 9 : 1 1 5 - 1 3 6 .
burch/ieldi, a plesiomorphic bandicoot
(Marsupialia, Peramelemorphia)from OligoMccowRAN. B. AND Lt. Q., 1997. Australasian
Miocene depositsof Riversleigh, northwestern
environmental
Cainozoic
biogeography:
framework for a southern-temperate province
Queensland.Journal of Paleontolog 69 127I J+.
ARCHERET AL. EVOLUTIONOF AUSTRALIANMAMMALS
MulRrGAD, J. erp WRoE. S., 1998. The family
Thylacinidaeand a descriptionand analysis of
a new species'.Badjcinus turnbulli gen. et sp.
nov. (Thyiacinidae:Marsupialia),from the late
Riversleigh. northwestern
Oligocene of
Queensland. Australia. Journal of L'ertebrate
P a I eontoI ogy 18: 612-626.
MLTNSoN,
C.J., 1992. Postcranialdescriptions of
(Vombatiformes,
Ilaria
and Ngapakaldi
Marsupialia) and the phylo-eeny of the
postcranial
vombatiformes
based
on
morphology.
Universitl, of C alifornra
Publications in Zoology 125.
It{uRRAy,P. F.. 1978. Late Cenozoic monotreme
anteaters. Australian Zoologist 20l 29-55.
MLJRRAv,P., 1986. Propalorchestes novaculacephalus gen. et sp. nov., a nei'r'palorchestid
(Diprotodontoidea, Marsupialia) from the
middle Miocene Camfield Beds, Northern
Territory, Australia. The Beagle 3:195-211
MLERAy. P.F.. 1990a. Allwertatherium webbi, a
new zygomaturine genus and species from the
late Miocene Alcoota Local Fauna- Northern
Territory (Marsupialia: Diprotodontidae). The
Beagle 7: 53-80.
Mr,RRAy,P.F.. 1990b. Primitive marsupial tapirs
(Propalorchestes novaculacephaltts Murray
and P. ponticulus sp. nov.) from the midMiocene of north Australia (Marsupialia:
Palorchestidae). The Beagle 7: 39-51.
Mr,IRRAv,P. F., 1991. The Pleistocene megafauna
of Australia. Pp. 1072-1164 in Vertebrate
Palaeontology of Australasia ed by P.
Vickers-Rich, J. M. Monaghan, R. F. Baird and
T. H. Rich. PioneerDesign Studio:Melbourne.
Munnev, P.F., 1992. The smallest New Guinea
zygomaturines - derived dwarfs or relict
plesiomorphs? The Beagle9: 89-110.
Munnev, P. F., 1997. Thylacinus megiriani, aneu
(Marsupialia:
of
thylacine
species
Thylacinidae) from the Ongeva Local Fauna of
central Australia. Records of the South
Australian Museum30: 43-61.
+l
Mrrruuy. P., Wells, R. AND Pmr\'p, M., 1987. The
cranium of the Miocene thylacoleonid
Wakaleovanderleuri: Click go the shears - a
fresh bite at thylacoleonid systematics. Pp.
433-466 in Possums and Opossums: Studies
in Evolution ed by M. Archer. Royal
Zoological Society of Nerv South Wales and
SurreyBeatty & Sons:Sydney.
MussER,A. M.. 1998. Evolution, biogeographr,
and palaeoecology'of the Ornithorhynchidae.
Australian Mammalogy-20: 147-162.
MUSSER, A A]\D ARCTIER,M., 1998.
Neu.
information about the skull and dentary of the
Miocene platypus Obdurodon dicksoni and a
discussion of ornithorhynchid relationships.
Philosophical Transactions of the Royal
Societyof London (B) 353: 1063-i079.
MYERS,T,J. AND ARCHER.M., 1997. KuterintJa
ngama (Marsupialia, Ilariidae): A revised
systematic analysis based on material from the
Late Oligocene of Riversleigh. north\r'estern
Queensland. Memoirs of the Oueensland
Museum 4l: 379-392.
OweN, R., 1844.Descriptionof a fossil molar tooth
of a Mastodon discoveredby Count Strzlecki
[sic] in Australia.Ann. Mag .Nat. Hist. 14 2682 7 1.
OwrN, R., 1882. Description of portions of a tusk
of a proboscidian mammal from Australian
Pleistocenedeposits.Phil. Trans. R. Soc. Lond.
173:777-781.
PASCUAL,
R., ARCFGR,M., Onrrz JAUREGUTzAR.
E..
PRADo, J. L., GoorrrLp, H. AND HAND, S. J..
1992. First discoveryof monotremesin South
America. N ature 356:'704-706.
PLEDGE,
N., 1987a.Kuterintja ngama, a new genus
and speciesof primitive vombatoid marsupial
from the medial Miocene Ngama Local Fauna
of South Australia. Pp. 419-422 in Possums
and Opossums:Studies in Evolution ed by M.
Archer
Srrrrev
Reetty
ft
SOnS and
the
ROVaI
Zoological Society of New South Wales.
Sydney.
MuRRAy, P. ANp MEGRLa,N,D., 1990. Further
observations on the morphology of Wakaleo
vanderleuri (Marsupialia: Thylacoleonidae)
from the mid-Miocene Camfield Beds.
Northem Territory. The Beagle 7:91-102.
Pleoce,N.S., 1987b. Pseudocheirids(Marsupialia.
Pseudocheiridae)from the middle Miocene
Ngama Local Fauna of northern South
Australia.Pp. 681-688in Possumsand Gliders
ed by A. Smith and I. Hume. Surrev Beattr' &
Sons: Sydney.
MuRRAy, P. A\rD MecutnN, D.. 1992. Continuity
and contrast in middle and late Miocene
vertebrate communities from the Northern
Territory. The Beagle9: 195-218.
Plroce, N.S., i987c. A new species of Burramys
Broom (Marsupialia: Burramyidae) from the
middle Miocene of South Australia. Po. 725-
t+z
AUSTRALIANMAMMALOGY
728 in Possums and Opossums: Studies in
Evolution ed. M. Archer. Surrey Beany & Sons
and the Royal Zoological Society of New
S o u t hW a l e s : S y d n e y .
PLEDGE,
N.S., 1992. The CurramulkaLocal Fauna:a
new late Tertiary fossil assemblagefrom Yorke
Pcnincrrla
Snrrrh
Arrstraiia
The
Rpnole
9'
1 1 5 - t4 2 .
RAUSCHER,
B.. 1987. Priscileo pitikantensis,a new
senus and speciesof thylacoleonid marsupial
(Marsupialia: Thylacoleonidae) from the
Miocene EtadunnaFormation, South Australia.
Pp. 423-432 in Possums and Opossums.
Studies in Evolution ed by M. Archer. Royal
Zoological Society of New South Wales and
Surrey Beatty and Sons: Sydney.
W. K, 1946. Adaptive
RarcN, H. C. ANDGREGoRv,
branching of the kangaroo family in relation
to habitat- American Museum Novitates 1309.
Rltl1v, J.A., CRocFrEr.J.-Y., Stce, B., SUDRE,J.. DE
BoNrs, L., VTANEY-LLALD,M., GoDrNor, M,,
FIARTENtsERGER,
J.-L., LANG-BADRE. B. AT\D
Biochronologie des
Colrre, B., 1987.
phosphoritesde Quercy: Mise djour des listes
fauniques et nouveaux gisements de
Milnchner
fossiles.
mammifdres
Geowissenschaftliche Abhandlungen A 10.
r 6 9 -I 8 8 .
RETEF, J.D.. KRAJEwSKT,C.. WesreruraeN, M..
WINKFEN, R.J. AND DxON. G.H., 1995
Molecular phylogeny and evolution of
marsupial protoamine Pl genes. Proceedings
of the Royal Society London 259: 7-14.
marsupials:their record in Australia and its
Vertebrate
biases.
Pp. 892-1069 in
PaLaeontology of Australasia ed by P.
Vickers-Rich.J. M. Monaghan. R. F. Baird and
T. H. Rich. PioneerDesign Studio:Melbourne.
RrcH,T.H., ARCFGR,M., FlAr\D,S.J.. Goornrlp. H..
N.S., Lt,nNeIruSJR.. E.L..
\4LIzufAD, J., PLEDGE,
M.
Fla-mrrnv, T. F., RrcH, L. S. V., WOoDBURNE,
O., Cese,J. A., Wmrenw. M. J., Tporon-o,R. H.,
KENP,A, Tunrrrull, W. D., AND RrcH. P. V..
1991b. Australian Mesozoic and Tertiary
t e r r e s t r i am
l a m m a l l o c a l i t i e s .P p . 1 0 0 5 - 1 0 5 8
in Vertebrate Palaeontologlt of Australia, ed
b1'P. Vickers-Rich,J.M. Monaghan,R.F. Baird
and T.H. Rich. Pioneer Design Studio and
Monash University Publications Committee:
Melbourne.
RIcH. T.H,V,, FIaINEay. T,F, AND ARCHER.M..
1989. A second Cretaceous mammalian
specimen from Lightning Ridge, N.S.W..
Australia. Alcheringa 13: 85-88.
fucs, T.H., VicKERS-fucH, P., CoNSrANrrNE. A..
Flruvxeny, T.F., KooL, L. Arp vAN KLA\EREN.
N., 1997. A tribosphenic mammal from the
Mesozoic of Australia. Science 278: 1438144)
R]CHARDSON,
B.J., B,qvrnSrOCK.P,R. AND ADAMS.
M.. 1986. Allo:yne electropltoresis: a
handbook .fo, animal systematics and
population studies.Academic Press:Sydney.
Rtop. W.D.L., 1993. Jackmahoneya gen. nov. anc
the genesis of the macropodiform molar.
Memoirs of the Association of Australian
P alaeontologisls 15: 441-459.
R-errEr,
J. D., WrNr<rED.r.
R. J. ANo Dxox, G. H., 1993.
Evolution of the monotremes:the sequences Snl-rssr,Ey,S.W. AND WrLL6, P.M.A., 1996. A new
n'nnnrl.,lio.
f'nm
thp
eorl'r
Fncanc
nf
vivwvvj,rq,,
of the protamine Pl genes of platypus and
southeasternQueensland and a preliminary
echidna. European Journal of Biochemistry
thc
nhvlnoenetin
invesiioalinn
nf
218: 457-461.
relationshipsof crocodyloids.Alcheringa 20
fucu, T. H. V., 1986. Darcius duggani- a neu
179-226.
ektopodontid (Marsupialia; Phalangeroidea)
from the early Pliocene Hamilton Local Fauna. Sevecr. D.E. ANo Russell. D.E., 1983. Mamntalian
Paleofaunas of the World. Addison-Weslel
Australia.
Pp. 68-74 in Revision of the
P u b l i s h i n gC o . : L o n d o n .
Ektopodontidae (MammaLia; Marsupialia.
Phalangeroidea) of the Australian Neogene
ScaNi-oN,J.D., 1993. Madtsoiid snakes from the
ed by M. O. Woodburne and W. A. Clemens.
eastern
Eocene Tingamarra Fauna of
Univertsty of Califurnia Publications in
Kaupia 3: 3-8.
Queensland.
Geological Sciencesl3l.
SHp.nven.G. B.. 1989. Opening address - a
fucn, T. H., A,ncsen. M.. HAND, S. J., Goprrulp, H..
chromosomephylogeny of kangaroos. Pp. uMLIRHEAD. J., PLEDGE,N., Fi-amrnnv. T. F..
vii in Kangaroos, Il/allabies and RatWOODBURNE.
R. H.M. O., CASE,J. A.. TEDFORD.
Kangaroos ed by' G. Grigg, P. Jarman and L
W. D., LI,\DELruS. E. L, Jn., RICH. L.
TIJRT'IBULL.
Hume. SurreyBeattv& Sons:Sydnel'.
S. V., WHrrEr-ew M. J., KEMP.A. A\DRrcH. P. V.,
l99la.
Monotremes, placentals, and
ARCHERET AL. EVOLUTIONOF AUSTRALIAN MAMMALS
et
1991.
Rhinolophoidea
8.,
Srce.
V e s p e r t i l i o n o i d e a( C h i r o p t e r a ) d u C h a m b i
(Eocene infdrieur de Tunisie). Aspects
biostratigraphique. biogdographique and
paldodcologique de I'origine des chiropteres
modernes. Neues Jahrbuch ftir Geologie und
P alaontologie Abhandlungen182: 355-376.
SIcE. B.. HAND, S.J. AND ARCHER,M., 1982 AN
Brachipposideros
Miocene
Australian
(Mammalia Chiroptera) related to Miocene
representativesfrom France. Palaeovertebrata
.
12: 149-1'71
Srcp. B., THoMAs, H., SeN, S.. GnorneRANDr, E.,
RocER, J. AND At--Sumruax, 2., 1991. Les
chiroptdres de Taqah (Oligocdne infdrieur.
Premier inventaire
Sultanat d'Oman).
GeowissenMiinchner
syst6matique.
schaftliche Abhandlungen A 26: 35-48.
SrwoNs, N.B., 1996. Bat diversification and
palaeobiogeography:implications of a new
higher-level phylogeny. Journal of Vertebrate
Paleontologv l6 (Supplement, Abstracts):
664..
of interfamilial
Srm.toNs,N.B..1998. A reappraisal
relationships of bats. ln Bat Biology and
Conservatton ed by T.H. Kunz and P.A. Racey.
Smithsonian Institution Press: Washington.
IJ.L.
SentNceR,M.S., 1993. Phylogeny and rates of
character evolution among ringtail possums
(Pseudocheiridae: Marsupialia). Australian
Journal of Zoology 4l: 2'73-91.
J.A.w., 1991. DNA
M.S. em KTRSCH.
SPRTNGER,
hybridization, the compression effect, and the
radiation of
diprotodontian marsupials.
SystematicZoology 40: 131-5i.
+)
Srtnrox, R. A.. 1955. LateTertiarymarsupialsfrom
South Australia. Records of the South
Australian Museumll 247-268.
SrrnroN, RA., T:oronp,
M.O., 1967. A ne*
fauna from the Tirari
Records of the South
4 2 7- 4 6 2 .
R.H. ANo wooDBtiRNE,
Tertiary formation and
Desert. South Australia.
Australian Lfuseum l5'.
SrnaHaN, R. (ed), 1995. Mammals of Australia.
Reed Books: Chatswood.
SzALAy.F.S.,1982. A new appraisal of marsupial
phylogeny and classification.Pp. 621-640 in
Carnivorous Marsupials ed M. Archer. Royal
Zoological Society of New South Wales:
Sydney.
Szerev, F.S., 1994. Evolutionary history of the
Marsuptalia and an analysis of osteological
characters.Cambridge University Press: New
York.
Results of the Archbold
Te.rE. G.H.H.. i948.
Expeditions.No. 59. Studieson the anatomy
the
Macropodidae
and phylogeny of
(Marsupialia). Bulletin of the American
Museum of Natural History 9l:. 233-352.
TEDFORD,R.H., ARCHER, M., BANTHOLOMAI,A,,
N.S., RrcH, T.. Rtcs, P. eru
Pmrr.r, M., PLEDGE,
WELLS,R.T., 1977. The discovery of Miocene
vertebrates,Lake Frome are4 South Australia.
BMR Journal of Australian Geology and
Geophysics2: 53-57.
BAN']is, M.R., KEMP. N.R,
TEDFoRD, RH,
F.L., 1975.
McDoucei-1, i. AND SL"TFTERLAND,
Recognition of the oldest known fossil
marsupialsfrom Australia. Nature 255: 141I Aa
TEDFoRD,R.H. er'ro IGw N.R. 1998. Oligocene
SpRrNcen, M.S., MCKAY, G.M., API-TH, K EM
marsupialsof the Geilston Bay Local Fauna,
KIRSCH,
J.A.W., 1992.Relationsamons ringtail
Tasmania.American Museum Novitates 3244:
(Marsupialia:
Pseudocheiridae)
possums
22pp.
based on DNA-DNA hybrydization. Australian
423-35.
40:
of
Zoologt
Journal
TEoroR-o,R.H. ntln wooDBURNE.M.O., 1987. The
ilariidae. a new family of vombatiform
AND
KIRSCH.
Wesrrnver,
M.
M.S.,
SpnrucEn,
from Miocene strata of South
marsupials
J.A.W.,199'1.Relationships among orders and
Australia and an evaluation of the homology
families of marsupialsbasedon l2S ribosomal
of molar cusps in the Diprotodontia. Pp. 401DNA sequences and the timing of the
418 in Possums and Opossums: Studies in
marsupial radiation. Journal of Mammalian
Evolution ed by M. Archer. Surrey Beatty &
E t ' o l u t i o n 2 : 8 5 -I 1 5 .
Sons and the Royal Zoological Society of New
South Wales: Sydne1,.
SpnrNcen,M.S. em WooDBURNE.M.O., i989. The
distribution of some basicranial characters
TuR.NBI,'LL,
W.D. eNpLuxDELrus,E.L..Jn, 1970. The
within the Marsupialiaand a phylogeny of the
Hamilton Fauna- a late Pliocene mammalian
Vertebrate
of
Phalangeriformes. Journal
fauna from the Grange Bum. Victoria,
P aleontology 9: 210-21.
Australia. Fieldianna: Geol. 19l. l-163.
44
AUSTRALIANMAMMALOGY
from South America to Australia. Journal of
TURNBI,LL,W.D., fuCH. T.H.V. AITOLUNDELruS,E.L.,
(Marsupialia:
Mammalian Evoltttion3: 121-161.
Pseudocheirids
1987a.
Pseudocheiridae) of the early Pliocene
wooDBr,n\E. M.O. eNn Clel,rENs.W.A., 1986a. A
Hamilton Local Fauna, southwesternVictoria.
new genus of Ektopodontidae and additiona:
Pp. 689-713 in Possums and Opossums.
commentson Ektopodon serratus. Pp. l0-42
Studies in Evolution ed by M. Archer. Surrey
in Revision of the Ektopodontidae (Mammalia;
Beatty & Sons and the Royal Zoological
the
of
Phalangeroidea)
Marsupialia;
Societyof New SouthWales: SYdneY.
Australian Neogene ed by M.O. Woodburne
and W. A Clemens. University of California
TunNeuLL, W.D., RlcH, T.H.V. aNp LutrogI-rus JR.
Publications in Geological Sciences131.
(Marsupialia:
E.L., 1987b. Burramyids
Bunamyidae) of the early Pliocene Hamilton
WooDBURNE,M.O. ,E,NDCLENGNS.W.A., 1986b.
Local Fauna. southwestern Victoria. Pp. 729Phyletic analysisand conclusions. Pp. 94-102
in
739 in Possums and Opossums: Studies
in Revision of the Ektopodontidae (Mammalia.
Evolution ed by M. Archer. Surrey Beatty &
of
the
Phalangeroidea)
Marsupialia;
Sons and the Royal Zoological Society of New
Australian Neogene ed by M.O. Woodburne
S o u t hW a l e s :S y d n e y .
and W. A. Clemens. Univeristy of CaLifornia
Publications in Geological Sciencesl3l.
H., 1993.A new species
TvLER,M.J. A\D GoDTHELp,
(Anura:
Boulenger
Lechriodus
of
M.O., MACFADDEN,8,J.. CASE.J.A.,
WoODBIJRNE,
Leptodactylidae) from the early Eocene of
SpRncsn, M.S., PLEDcE,N.S., PowER, J.D..
Queensland. Transactionsof the Ro1,alSociety
Woooer,ENE,J.M. A\D SpnrNceR.K.B., 1994.
o f S o u t hA u s t r a l i al l 7 : 1 8 7 - 1 8 9 .
and
biostratigraphy
mammal
Land
of
the
Etadunna
magnetostratigraphy
Warrs, C.H.S. AND BAVERsrocK, P.R., 1995.
Formation (late Oligocene) of South Australia
Evolution in the Murinae (Rodentia) assessed
Journal of Vertebrate Paleontology l3: 483by microcomplement fixation of albumin.
515.
Australian Journal of Zoology 43: 105-118.
M.. 1991. Phylogenetics of the
WESTERMAN.
typhlops
Notoryctes
Mole,
Marsupial
Australian
(Marsupialia:
Notoryctidae).
Journal of Zoolog, 39: 529-537.
M. AND EDWARDS.D-, 1992. DNA
WESTERMAN.
phylogeny
of
the
and
hybridization
monotremes. Pp. 28-34 in Platypus and
Echidnased by M. L. Augee. Royal Zoological
Societyof New SouthWales: Sydney.
D. N. AND O'BREN,S.
WESTERMAN,
M., JANCZEWSzu.
J., 1990. DNA-DNA hybridisation studies and
marsupial phylogeny. Australian Journal of
Zoology 37 : 315-323.
WreLE.J. R. AND HoPSoN.J. A.. 1993. Basicranial
evidence for earl-r mammal phylogeny. p4562 in Mammalian Phylogeny, Vol. I ed by F.
Szalay, M. Novacek and M. McKenna.
Springer-Verlag:New York.
Woooermr. M.O., 1984. Families of marsupials:
relationships, evolution and biogeography.
Pp. 48-71 in Mammals: Notes for a Short
Broadhead.
T.W.
ed
by
Course
PaleontologicalSociety: Boulder.
C,qsp, J.A., 1996.
wooDBr,R\E, M.O. A\D
Dispersal, vicariance, and the Late Cretaceous
to Early Tertiary land mammal biogeography
M.O., Trorono, R.H.. AncnEn, M. A\D
WooosrJRNrE.
PLED,cE,
N.S.. 1987a. Madakoala, a new genus
Miocene koalas
and two
species of
(Marsupialia: Phascolarctidae) from South
Australia.and a new speciesof Perikoala. Pp.
293-317 in Possums and Opossums: Studies
in Evolution ed by M. Archer. Surrey Beatty &
Sonsand the Royal Zoological Society of New
South Wales: Sydney
WooDBURNE,M.O., Prpoce, N. S. ,cNo ARCHER..M.,
1987b. The Miralinidae,a new family and two
new speciesof phalangeroid marsupilas from
the Miocene strata of South Australia. Pp.
581-602 in Possumsand Opossums: Studies
in Evolution ed by M. Archer. Surrey Beatty
and Sons and the Royal Zoological Society of
r"NewSouth Wales: Sydney.
R. H., i975. The
WoODBURNE.
M.O. AND TEDFORD.
first Tertiary monotreme from Australia.
American MuseunrNovitates2588: 1-11.
R.H, AND ARCHER,M,WOODBTRNE,
M.O., TEDFORD.
1987c. New Miocene ringtail possums
(Marsupialia: Pseudocheiridae)from South
Australia. Pp. 639-19 in Possums and
Opossums: Studies in Evolution ed by M.
Archer. Surrey Beatty & Sons and the Royal
Zoological Society of New South Wales:
Sydney.
ARCHERETAL. EVOLUTIONOF AUSTRALIANMAMMALS
TEDFORD. R.H., ARCHER, M.,
WooDBr,RNE, M.O.,
TURNEULL, W.D., PLANE, M, TEDFORD, R.H. EIIO
E.L., Jn, 1985. Biochronology of
LnNDELrus.
the continental mammal record of Australia
Sottth Australian
and Nerv Guinea.
Department of Mines and Energy Special
Publication 5: 347-65.
gadiyulr
Muribacinus
1996.
WRoE. S..
(Thylacinidae. Marsupialia). a very plesiomorphic thylacinid from the Miocene of
Riversleigh,northwesternQueensland,and the
problem of paraphyly for the Dasyuridae.
Journal of P aleontology 70: 1032-1044.
wRoE, S., 1999b. The geologically oldest
dasyurid. from the Miocene of Riversleigh.
nofthwestern Australia. Palaeontology. 42:
501-527.
J. A\D CooKE, 8.. 1998. The
WnoE, S., BR{rvtN4ALL.
skull of Ekaltadeta ima (Marsupialia,
Hypsiprymnodontidae?):an analysis of some
a remarsupial cranial features and
investigation ofpropleopine phylogeny, witt.
notes on the inference of carnivorv in
mammals. Journal of Paleontolog,v 12: 738751.
Wnoa, S. AND M)'ERS,T. J. 1998. Fallacy and
-29.
future-eating. Australasian Science 19l.2'7
Wnoe. S., 199'7. A re-examination of proposed
morphology-based synapomorphies for the
Wnos, S., MyERS, T. J., WELLS,R.T.. ,qm GI-LESPE,
families of Dasy'uromorphia (Marsupialia):
A 1999. Estimatin-e the weight of the
Part I, Dasyuridae. Journa! of Mammalian
Lion
Th.v-lacoLeo
Pleistocene Marsupial
E v o l u t i o n4 : 1 9 - 5 2 .
carnifex (Thylacoleonidae: Marsupialia):
'bone-cracking'
implications for the ecomorphology of a
WnoE, S., 1998. A new speciesof
marsupial super-predator anci hy'potheses of
dasyurid (Marsupialia) from the Miocene of
impoverishment of Australian marsupial
northwestern
Riversleigh,
Queensland.
of
carnivore faunas. Australian Journal
Alcheringa 22: 277-284.
Zoology 49: (in press).
WnoE, S., 1999a. Carnivore calamity: the rise and
fall of Australia's natural born killers.
Scientific American 280: 68-74.