australian museum onlineabout the museumresearch and collectionsfeaturesexplore


birds
introduction
collections
research

publications
faqs
factsheets
photo gallery
bird stuff
links



research

Fossil history of birds in Australia

Page contents

Introduction

In the recent upsurge of interest in the early evolution of birds, Australia has not contributed to the debates on bird-dinosaur relationships or on the origin of feathers or flight. Nonetheless, it does have interesting, albeit patchy, record of birds across the past 110 million years. The record is poor compared those of Europe and North America, in both temporal and faunistic representation. There are a few intervals with rich, diverse avian records, several scattered periods with small to moderate representations and some extensive gaps at critical times.

The earliest Australian record of birds, from the Early Cretaceous (110-100 million years ago -mya), comprises five small indeterminate feather impressions from Koonwarra, Victoria and a few small bones from southern Victoria, north central New South Wales and western Queensland. The feathers have been known for some time, but unfortunately the birds from which they came have not appeared from the deposits. Likewise, little can be said about the Victoria and New South Wales fossils.

The bones from Queensland can be attributed to an enantiornithine bird. The recognition of the subclass Enantiornithes has been a major advance in understanding bird evolution. These strange birds differed from modern birds in a variety of major features of the skeleton. They appear to have been the dominant lineage of birds through the Cretaceous, and their fossils are now known from every continent, including Australia. Like the dinosaurs, this group of morphologically diverse birds became extinct by the end of the Cretaceous. The Queensland fossils, which have been named Nanantius eos, show that these birds were in Australia.

Further information

Fossil history - taxonomic

All subsequent avian records are of modern forms. Unfortunately, there is a large gap between the Cretaceous fossils and the considerably richer record that starts in the Late Oligocene. Within this span, the appearance of birds is almost non-existent other than a few specimens from the Eocene. This record, although sparse, is interesting. It is the first appearance of a modern bird fauna in Australia, both marine and non-marine. The former comprises a variety of penguins occurred along the southeast coast of the mainland, including the giant Anthropornis nordenskjoeldi, which may have reached 1.35 m and 90 kg. Non-marine fossils from Murgon, southeastern Queensland, include remains of songbirds and early shorebird-like birds, tentatively placed in the extinct 'catch all' family Graculavidae.

The void between the Cretaceous and Late Oligocene highlights the lack of suitable fossil-bearing sites, not the absence of birds in Australia at this time! This is unfortunate because events were taking place that would play a major role in shaping the avifauna of Australia. It is during this period that Australia broke away from Antarctica and started to drift northwards in isolation from other landmasses. The birds and other animals carried along on this continental ark evolved their distinctive characters during this interval, without input from outside invaders.

Starting in the Late Oligocene and continuing through the middle Miocene is one of the richest portions of the fossil record of birds in Australia. The major regions are Riversleigh, in northwestern Queensland, and the Great Artesian Basin of central Australia. The record through the Late Miocene is much sparser. At this time, Australia was undergoing major changes. The continent, which had been wetter than today and more extensively covered with humid forest, began to dry. The thicker forests retreated as the environment opened up, giving way to eucalypts and acacias.

The Pliocene-aged localities (5-1.6 mya) in northern Australia, while not as rich as those of earlier age, provide intriguing insights into a landscape that was taking on the character of modern Australia. Two of the important localities are Bluff Downs, northeastern Queensland, and Rackham's Roost, at Riversleigh. The former supported wetlands with a bird community similar to that now found in Kakadu National Park, with the surprising addition of flamingo and Cape Barren Goose Cereopsis novaehollandiae. Central Australia and the Darling Downs in Queensland have notable records from the Pliocene and Pleistocene (1.6 mya to 10,000 years). Of the youngest sites yielding birds (Late Pleistocene and Holocene), the most important ones occur principally across southern Australia, mainly as cave deposits.

The record of many Australian groups begins in the Late Oligocene-Early Miocene deposits of central Australia. Because these deposits accumulated in lakes and rivers, the taxa dominating at almost every site are waterbirds. This bias towards aquatic groups also characterises many younger sites, and consequently these families of birds are among the best represented through in the Australian fossil record.

The variety of cormorants, darters and pelicans consisted of species relatively similar to those alive today. The record of storks reveals the modern Black-necked Stork or Jabiru Ephippiorhynchus asiaticus in younger deposits but it was preceded by several species in the genus Ciconia, which includes the living White Stork C. ciconia of Europe.

One surprising component of the early Australian avifauna consisted of flamingos. At times a range of different sized species coexisting in the shallow saline lakes of central Australia. These persisted until the Pleistocene, at which time Australian climate became increasingly arid, causing these lakes to dry and the flamingos to disappear for good. Related to flamingos were the palaelodids, a now extinct group also from the same localities.

Many of these aquatic groups have only been studied piecemeal, with significant, usually older, portions of their records still requiring attention. Several shorebird families are known from central Australia, but only the stone-curlews have been reviewed in a systematic manner. Of the diverse range of waterfowl known from throughout the Tertiary, only the younger ones have been studied. Among the Pliocene species from Bluff Downs are the Cape Barren Geese and a new pygmy-goose about two-thirds the size of today's species, making it perhaps the smallest duck to have lived. Many of the named species from the Pleistocene of the Darling Downs were either misidentified to family or can be referred to living waterfowl.

There is likewise a long history of rails, with the earliest ones from central Australia inviting study. An undescribed flightless native-hen has been found commonly at Riversleigh through the Late Oligocene-Middle Miocene. The best studied rail is the Tasmanian Native-hen Gallinula mortierii, which once was widespread through eastern Australia in the Late Pliocene and Pleistocene; it has been found as far north and west as the Leichhardt River in Queensland. This species seems to have successfully coexisted with marsupial carnivores but failed to cope with the arrival of the Dingo about 3,500 years ago; it is now restricted to Tasmania, where the Dingo does not occur.

In contrast to the waterbirds, terrestrial forms are generally much less represented. An exception is large flightless birds, which form a major component of the avifauna at many sites. One of these groups was the extinct family Dromornithidae. These birds were superficially like giant emus, with the largest being about 3 m tall and weighed 400-450 kg. They are known only from Australia from at least the Late Oligocene and extending into the Pleistocene.

The other family of flightless birds comprises the familiar cassowaries and emus. The oldest record of this group, known from the Late Oligocene-Middle Miocene of central and northern Australia, is Emuarius gidju. This small, flightless bird combines characters of both emus and cassowaries in its hindlimb and represents a stage soon after the separation of these two lineages. A dwarf cassowary has supposedly come from the Wellington Caves, New South Wales, but the manner of preservation is unlike that of other fossils from this site, so the provenance of this intriguing animal is uncertain. Emus have a longer record, with the modern species appearing in Pleistocene deposits.

The fossil record of other non-aquatic bird families is a bit of a mixed bag. The records are patchy, a problem compounded in some instances by outstanding specimens still awaiting examination. Riversleigh has a considerable array of non-aquatic forms compared with other localities of similar age, which has added considerably to the knowledge of some groups. Nonetheless, the record reveals a very incomplete picture of the evolution of these birds in Australia at this time.

For example, the record of parrots in the Tertiary comprises a white cockatoo from the Middle Miocene and the modern Budgerigar Melopsittacus undulatus from the Pliocene of Rackham's Roost. Owlet-nightjars are known from one specimen (a new genus), swifts by one species (a large Collocalia), kingfishers by a single bone (a Todiramphus-like bird) and pigeons by a few as of yet unstudied bones; such records hardly reflect the Tertiary diversity of these groups in Australia.

One of the characteristic groups of Australia, the megapodes, is first known from a new genus, currently being described from the Late Oligocene of South Australia. It was only about the size of a large quail, making it smaller than any living species. At the other end of the scale is the giant megapode Progura gallinacea, about the size of a domestic turkey, known from several Pleistocene localities.

There are a number of birds of prey, both named and unnamed, although some of the described fossil forms may well prove to be referable to modern species. There are remains of several raptors that were substantially larger than the living Wedge-tailed Eagle Aquila audax. Most are currently under investigation, which promises to greatly increase the understanding of Tertiary diurnal raptors in Australia. One of the most remarkable birds of prey was Pengana robertbolesi from the Miocene of Riversleigh. This eagle-sized bird had a modified ankle joint, which permitted a high degree of flexibility both frontwards and backwards, as well as to either side, similar to the living but considerably smaller gymnogenes or harrier-hawks Polyboroides of Africa.

Among the most important finds are songbirds (Passeriformes). Small, fragmentary bones from Murgon are the earliest representatives of the Order in the world, and extend the known age of the group by almost 25 million years ago from the previous oldest known fossils from France. At least two different types, as indicated by size, are present, but no more detailed identification can be made. Numerous bones have been found at Riversleigh, from the Late Oligocene through the Pliocene. Well over a hundred specimens are known, but most have not had detailed examination. Those that have been studied include lyrebirds and a large relative of the figbirds. Honeyeaters have been identified from the Middle Miocene into the Pliocene. A logrunner has been found here, while other new species of logrunners have come from much younger deposits in South Australia and Victoria.

Most avian remains from the Late Pleistocene and Holocene can be referred to modern forms, usually at the specific level, and almost always at the generic. A number of living species have been identified from cave deposits across southern Australia. These records show that the distributions of some were more extensive than at present. The Tasmanian Native-hen, already mentioned, scrub-birds, bristlebirds and the Glossy Black-Cockatoo are known from localities well away from those now occupied. Gigantism characterised some taxa, such as the megapode Progura and a giant coucal Centropus colossus. Flamingos, which are still known from other parts of the world, persisted to this time.

Further information

Fossil History - chronological

There is a large gap between these Cretaceous remains and the next appearance of birds, in the Eocene (57-35 mya). A foot impression from near Brisbane has been tentatively referred to a large flightless bird known as a dromornithid. The first indication of a modern, non-marine bird fauna in Australia comes from Murgon, southeastern Queensland, dated as Early Eocene (54 mya). Several early shorebird-like birds, tentatively placed in the family Graculavidae, have been found there, as have specimens of other non-passerine forms, as of yet unstudied. The most important finding from this site are songbirds (Passeriformes). These are the earliest known representatives of the Order, and extend the known age of the group by almost 25 million years from the previous oldest known fossils from France. At least two different types, as indicated by size, are present, but no more detailed identification can be made. A variety of penguins have been found along the southeast coastal mainland, including the giant Anthropornis nordenskjoeldi, which may have reached 1.35 m and 90 kg.

The next hiatus in the record is a quite significant one. Birds are unknown from after the Eocene until the late Oligocene. It is during this period that Australia broke away from Antarctica and started to drift northwards in isolation from other landmasses. The birds and other animals carried along on this continental ark evolved their distinctive characters during this interval, without input from outside invaders.

Starting in the Late Oligocene (30-25 mya) and continuing through the middle Miocene (15 mya) is an extensive record from the present Great Artesian Basin (Lake Eyre and Lake Tarkarooloo subbasins) of central Australia and in northern Australia. Western Australian sites of comparable age should exist but have not yet been exploited. Because the central Australian deposits accumulated in lakes and rivers, the taxa dominating at almost every site are waterbirds, augmented by some terrestrial, often flightless, forms and scattered flying non-aquatic forms. Diverse groups of non‑marine birds were present in Australia at least as far back as the mid‑Tertiary, including emus, dromornithids, megapodes, waterfowl, darters, cormorants, pelicans, flamingos, palaelodids (a now extinct group related to flamingos), rails, shorebirds and stone-curlews. Smaller non-aquatic birds are represented in these sites, but most have not been the subject of study.

Of the northern Australian sites, the most important is Riversleigh, northwestern Queensland, spanning the period Late Oligocene-Middle Miocene. This area is significant because of its array of both water- and forest‑dwelling forms, including numerous non-aquatic species. Among the non-passerines thus far reported are dromornithids, storks, birds of prey, rails, cockatoos, swiftlets and kingfishers, many of which can be assigned to living genera. Passerines found here include lyrebirds, honeyeaters, logrunners and a large relative of the figbirds. One bird found both at Riversleigh and in central Australia is Emuarius gidju. This small, flightless bird combines characters of both emus and cassowaries in its hindlimb and represents a stage soon after the divergence of these two groups of flightless Australian birds.

Other northern sites include the somewhat younger Alcoota and Bullock Creek in the Northern Territory. These are dominated by emus and dromornithids. The latter are gigantic flightless birds known only from Australia from at least the Late Oligocene and extending into the Pleistocene.. These were superficially like giant emus, with the largest being about 3 metres tall and weighed 400-450 kilograms.

In the Late Miocene the Australia was undergoing major changes. The continent, which had been wetter than today and more extensively covered with humid forest, began to dry. The thicker forests retreated as the environment opened up, giving way to eucalypts and acacias.

The Pliocene-aged localities (5-1.6 mya) in northern Australia, while not as rich as those of earlier age, provide intriguing insights into a landscape that was taking on the character of modern Australia. Bluff Downs in northeastern Queensland supported wetlands with a bird community similar to that now found in Kakadu National Park, with the surprising addition of flamingo and Cape Barren Goose. Rackham's Roost, at Riversleigh, has documented the presence of many small birds, including honeyeaters and the modern Budgerigar.

Central Australia also contains both Pliocene and Pleistocene deposits. Most of the birds represent groups similar to those found in the Oligo-Miocene deposits of the region. The history of flamingos on this continent continued, with a range of different sized species co-existing in the shallow saline lakes. When the Australian climate became increasingly arid, these lakes dried up and flamingos died out.

The Darling Downs in Queensland has a notable Plio-Pleistocene record.. Of the many fossil species named from here around the turn of the century, few are valid, with subsequent examination revealing that most belong to living species. One striking find was the giant megapode Progura gallinacea, about the size of a domestic turkey.

Important sites from the fossil-bearing interval of the Late Quaternary occur principally across southern Australia. Giant megapodes, oversized coucals and some living families now absent from Australia (e.g. flamingos) persisted until the Pleistocene. Genyornis newtoni, the best known of the Dromornithidae, has been found at Cuddie Springs, north central New South Wales, in the same deposits as Aboriginal artefacts (31,000 bp). Most avian remains from this period can be referred to modern genera or species, although their distributions were often more extensive than at present. The Tasmanian Native-hen is known to have ranged across the eastern mainland, while scrub-birds, bristlebirds and the Glossy Black-Cockatoo are also known from localities well away from those now occupied.

Further Information

Conclusions

What can be said about the fossil record of birds in Australia? Much is an indication of what it lacks, with many crucial times and critical geographical areas (such as the western half of Australia) not represented. There is also the surprising absence of some groups in the record. So far, there is a dearth of early unsuccessful lineages without modern ancestors; almost everything can be allocated to modern groups. This may be a reflection of the depauperate early record on the continent. Some characteristic groups, such as frogmouths, are still unknown from anything but the most recent deposits. Despite these deficiencies, the fossil record makes a useful contribution towards the understanding of Australia's avifauna. It provides basic documentation needed for building a broad picture of avian evolution on this continent; testing theories on origins, colonisation routes and divergence times of various bird groups; and aiding the interpretation of changes in climate and environment. The record, while still scanty, will continue to assume growing importance as additional sites are found and exploited and more workers turn their focus on Australia's avian history.

Walter Boles, Scientific Officer
Ornithology