Figure 13.1 Phylogenetic relationships of parrots. Passerines (‘perching birds’) have parrots as their closest relatives, followed by falcons and seriemas. Modified from Hackett et al. (2008).5
Ashort distance inland from the north coast of Puerto Rico lies a terrain of thick, ancient and undulating jungle. Most of this verdant coverage sits astride a layer of porous and soluble limestone, which in places has been eroded to form huge sinkholes and plummeting hollows of land. Hidden in one of the forest’s many depressions is the Arecibo Observatory, the world’s largest single-dish radio telescope. For several decades, scientists have used the research facility as part of a multi-million-dollar search for extraterrestrial life, listening for radio signals that might indicate the existence of alien species. So far, no convincing evidence has been forthcoming, although numerous exoplanets have been identified, lying many light years away, that could potentially be habitable.
How ironic it is, therefore, that in the same forest a low-budget conservation programme is also listening for radio signals, using radio-telemetry to try and save a well-known but critically endangered species, the Puerto Rican Amazon, which inhabits our very own planet. So far, this endemic parrot has narrowly escaped extinction, after reaching an all-time low of only of 13 individuals in 1975. Initial conservation efforts in the eastern mountains were thwarted by Hurricane Hugo, which reduced the recovering population to 22 birds in 1989. As a result, a second captive breeding programme was set up in the more sheltered karst region of the west, not far from the Arecibo’s radio telescope. In the spring of 2017, I was privileged to observe two of these charismatic birds perched high up in the forest canopy. Both had been fitted with radio transmitters, a testament to the ongoing conservation efforts to save the species. At the time of writing, the full effects of Hurricanes Irma and Maria, the latest to strike the island, are not yet known, although at least seven individuals died in captivity as a result of stress and the high temperatures that resulted from the loss of canopy. Should the present introduction programme be successful, this rare parrot will be given a greater chance of survival and, although still critically low, the species’ steadily increasing numbers provide a glimmer of hope for the future.1
The Puerto Rican Amazon is not the only parrot under threat; Lear’s Macaw and the Kakapo number less than a few hundred, while Spix’s Macaw survives only in captivity. Sadly, at least 16 species have become extinct during the last century, including the Carolina Parakeet, the Paradise Parrot and the New Caledonian Lorikeet. Indeed, scientists from the Australian National University and BirdLife International have concluded that parrots are among the most threatened group of birds, with over a quarter of species classified as globally threatened on the IUCN Red List.2 Risk factors include small geographical distributions, especially of those species restricted to islands, large body size, long generation times and a dependency on forest habitats.
Currently, there are 381 species of parrot in the wild, one of the largest of all the non-passerine groups, with a distribution heavily biased to the southern hemisphere. Indeed, their northern limits approximate to the 30th parallel, and it was only the extinct Carolina Parakeet that extended further north, reaching southern New York and the Great Lakes. The largest number of species is found in South America, with Brazil holding the record, while the greatest diversity occurs in the Australasian region. In contrast, the whole of Africa has comparatively few species.
But what do we know of the parrot’s evolutionary history? A decade ago the answer would have been very little, but recent molecular studies have clarified not only the position of parrots on the avian tree of life but also where and when they first evolved and how they spread across the globe. It is a fascinating and complex story, one that involves vicariance and multiple transoceanic dispersals.3
The first parrots
Parrots belong to the Australaves clade of landbirds that emerged at the time of the break-up of Gondwana.4 As noted previously, the common ancestor of the Australaves and Afroaves was an apex predator, a view supported by the lifestyle of the seriemas that occupy the basal branch of the Australaves (Figure 13.1). The two extant species, the Red-legged and Black-legged Seriema, are large, long-legged terretsrial birds with raptor-like behaviours. Both species are well known for seizing live prey, mainly reptiles, and beating them to death with violent throws to the ground. Once it is dead, the birds rip their food into smaller pieces with a sickle claw, holding the carcass in their beaks and tearing it apart with their feet. The seriemas are the sole living relatives of an extinct group of gigantic carnivorous ‘terror birds’, the phorusrhacids, which resided at the top of the South American food chain during the Cenozoic.6
Figure 13.1 Phylogenetic relationships of parrots. Passerines (‘perching birds’) have parrots as their closest relatives, followed by falcons and seriemas. Modified from Hackett et al. (2008).5
The common ancestor of all parrots probably lived on Gondwana during the Cretaceous, although these early birds would have looked somewhat different from the species alive today. According to the German palaeontologist Gerald Mayr, stem parrots lacked the most familiar feature of extant species, the long and thick upper bill, or maxilla, which curves over a shorter lower bill, or mandible. Such an anatomical feature probably evolved to allow the consumption of the larger fruits and nuts that appeared during the early Cenozoic period, some time after 65 million years ago. In other words, it was the emergence of new food sources that provided the driving force for the evolution of the thick, curved bill that characterises all the species alive today.7 Indeed, the jaws of extant parrots have been likened to Swiss Army knives, in that they can be used to crack kernels, seed husks and fruit pulp, as well as helping the birds to climb trees, manipulate food and strip wood.
Molecular phylogenies consistently show that the New Zealand parrots occupy the basal position among the group and were the first to diverge. Initially, it was accepted that the New Zealand clade became isolated after Zealandia broke away from Gondwana 82 million years ago, and that their speciation was allopatric and shaped by vicariance.8 However, recent studies indicate that the diversification of modern parrots occurred after the K–Pg boundary, around 58 million years ago.9 Although the revised date is incompatible with the above scenario, New Zealand parrots could still have evolved as the result of continental fragmentation. New geological data reveal that a land bridge existed between Australia and New Zealand until the early Eocene, up to 52 million years ago. Therefore, even if the initial split within the crown group occurred later than first thought, New Zealand parrots could still have been the result of vicariant evolution following the complete separation of New Zealand from Australia. Another possibility is that the ancestral population was more widely distributed in the past and that the earliest development occurred elsewhere. Indeed, the avifauna of New Zealand is composite in nature and has repeatedly experienced colonisation and extinction events. Rather than being the result of prolonged isolation, the seemingly ancient endemism of New Zealand parrots might reflect a recent expansion of range, combined with extinction elsewhere. We will return to the vicariance versus dispersal debate in the next chapter when the evolution of the earliest passerines is discussed.
Once the ancestral population reached New Zealand, it adapted to different ecological niches and evolved into two genera, Strigops and Nestor.10 The Kakapo, the world’s most genetically isolated parrot, is the only surviving member of the Strigops (Plate 20). It is the heaviest, reaching up to 4 kilograms, as well as the only flightless ground-dwelling parrot. Such anatomical features are typical of species that colonise oceanic islands that lack predators and offer an abundant food supply. Although nocturnal, with small eyes, it has evolved a well-developed sense of smell that allows it to track down its favourite food plants during the night. The Kakapo is a specialised foliage eater and has a large gut to allow the bulk processing of its nutritionally poor diet. Sadly, the species lies fourth in the EDGE rankings (see The Oilbird’s Story) and, as of June 2016, there were only 154 known individuals.11 Despite translocation to predator-free islands and intensive conservation measures, the population has been slow to increase, in part because females only breed every 2–5 years and fertility is low, with less than 50 per cent of eggs hatching.
The Nestor lineage diversified between 5 and 3 million years ago, during the Pliocene, as new habitats opened up after the formation of the Southern Alps. Those birds that occupied the mountain tops evolved to give rise to the omnivorous Kea, the world’s only alpine parrot. This hardy species eats mainly insects but will take carrion and has been known to excavate the chicks from Sooty Shearwater burrows. Innately curious, Keas are attracted to people and are often encountered around ski slopes and mountain huts, where they will investigate rucksacks, boots, and even cars. The lowland population gave rise to three species, but only the New Zealand Kaka exists today (the Chatham Kaka became extinct in the sixteenth century, and the last Norfolk Kaka died in 1871). The New Zealand Kaka is predominantly an arboreal species, occupying the mid-to-high canopy, and is easiest to see when flying across the valleys or calling from the tops of trees. They have extra-long, slim upper bills and tongues tipped with brushy papillae that have evolved for extracting sap from trees. The two populations, on North Island and South Island, have evolved slight differences in body size and beak structure after becoming isolated by rising sea levels at the end of the Pleistocene.
Once the New Zealand lineage had diverged, the remaining population split when Australia separated from Antarctica around 40 million years ago. At the beginning of the Palaeogene, Antarctica was ice-free, warmer and wetter than today and was divided into West and East Antarctica by a seaway. The common ancestor of the New World parrots evolved in the western half of Antarctica, while the common ancestor for the African subfamily, Psittacinae, evolved in East Antarctica. As the temperatures fell during the Eocene and early Oligocene, extensive ice sheets engulfed most of the continent and forced the two populations to disperse around 35 million years ago. The ancestral New World parrots established themselves in South America, while the early Psittacinae reached Africa by transoceanic dispersal, probably via the Kerguelen plateau (Figure 13.2). Although the latter is now submerged, except for a few remote islands, it once extended for over 2,000 kilometres from Antarctica towards Madagascar and rose over a kilometre above sea level. Later, multiple dispersals occurred from the Australasian population and led to the colonisation of Madagascar and Africa, Indo-Malaysia and the Pacific islands. In other words, transoceanic dispersals from three ancestral populations – Australasia, West Antarctica and East Antarctica – account for the distribution of most of the parrots in the world today.
Figure 13.2 Proposed transoceanic dispersal routes of the parrots. Emergent continents above sea level today are shaded grey, and continental shelves are indicated with black lines corresponding to the middle Eocene when dispersal of the Psittacinae, lovebirds and vasa parrots could have occurred. Modified from Schweizer et al. (2010).3
New World parrots
Neotropical parrots are found throughout South and Central America, including Mexico, and the Caribbean islands, and comprise some 160 species of amazons, macaws, parrots, parrotlets and parakeets. Some of the most familiar and iconic species, such as the Hyacinth Macaw, the Blue-and-yellow Macaw and Spix’s Macaw, belong to this group. Indeed, the New World clade is more species-rich than any other parrot group, and it is likely that the psittacine-free continent would have presented an underutilised adaptive zone, providing multiple ecological opportunities for their speciation. Molecular phylogenies confirm that this enormous diversity all evolved from a single common ancestor that arrived from Antarctica, without any further colonisation by lineages from Africa or Australasia. Maximum speciation rates coincided with the first peak of the Andean uplift in the late Oligocene and early Miocene, suggesting that this dramatic geological event contributed to their diversification, as it did for many other South American bird groups. In contrast, lowland species were mainly influenced by Pleistocene climate changes that resulted in shifts between dry and wet forests and the creation of isolated, open savannas. Psittacine colonisation gradually spread northwards throughout the continent and reached North America around 5.5 million years ago, before the Panamanian land bridge had formed.
Sadly, the Carolina Parakeet, America’s northernmost parrot, is now no longer with us. European settlers persecuted the species in their tens of thousands for food, sport and feathers, and to protect their crops, and by the early 1920s the species was extinct. A major factor contributing to their decline was their predictable flocking behaviour: the birds returned repeatedly to old haunts, enabling wholesale slaughter. Recent genetic studies, using mitochondrial DNA extracted from the toepads of museum specimens, has helped elucidate the continents’ colonisation by psittacines.12 It turns out that the Carolina Parakeet’s nearest relatives are species with broad distributions in tropical South America, such as the Nanday Parakeet, the Golden-capped Parakeet and the Sun Parakeet, rather than those in nearby Mexico or the Greater Antilles, as one might have predicted. Interestingly, all four species share a distinctive blue edging to their primary and secondary feathers, suggesting that plumage patterns provide a stronger indication of the clade’s kinship than their biogeographical relationships. The molecular studies also showed that Central and North America were colonised at different times by several distinct lineages of parrot, although how the Carolina Parakeet came to occupy its unique range in eastern North America remains a mystery.13
African parrots
The Psittacinae from East Antarctica were the first arrivals in Africa, and spread rapidly throughout the tropical and southern regions, from Senegal in the west, Ethiopia in the east, and South Africa in the south. They gave rise to two large African grey parrots (genus Psittacus) and nine small, stocky birds with large heads (genus Poicephalus, from the Greek for ‘made of head’) that include the Cape Parrot and the Senegal Parrot.
Ten million years later, two further dispersals from Australasia added to the African complement of parrots. The vasa parrots only reached Madagascar, but the lovebirds colonised both Madagascar and Africa. Although the routes taken are unclear, it is possible that they were facilitated by a volcanic plateau in the southern Indian Ocean, known as Broken Ridge, which served as a stepping stone between Western Australia and the Kerguelen archipelago (Figure 13.2). Interestingly, avian dispersals across the Indian Ocean from Australia were unusual, and have only been proposed for a few clades, including the blue pigeons (genus Alectroenas) and the cuckooshrikes (family Campephagidae).14
Vasa parrots are notable for their rather primitive appearance, which includes a shortened head, long neck and prominent pink beak. Furthermore, the male’s cloaca can be everted into a hemipenis that becomes erect during mating. Copulation can last over an hour and involves a copulatory ‘tie’ facilitated by the male’s genital protrusion interlocking with the female’s cloaca – a sexual act that is unique among birds and may be associated with sperm competition, enabling the male to increase his chances of fertilisation. Both sexes of the Greater Vasa Parrot have multiple partners (polygynandry), and during chick-rearing it is only the female of the species that sings, holds a territory, and develops a conspicuous, orange-coloured, bald head. These unusual breeding traits may have evolved as the result of the competition among females for food provided by the males, as those with higher song rates are known to attract more males.15
Unlike the vasa parrots, the ancestral lovebirds went on to colonise Africa where, among the forests and savannas south of the Sahara, they diverged to produce the eight species recognised today. Evidence that the ancestral population used Madagascar as a staging post is the finding that the Grey-headed Lovebird, a Madagascan endemic, is the sister to all the African mainland species. They are social and affectionate taxa, and their anthropomorphic name derives from their strong monogamous pair-bonding and the fact that they spend extended periods of time sitting together. Furthermore, they typically feed each other to re-establish their bonding, especially after experiencing separation or stress.
Over 50 years ago, William Dilger, Professor of Neurobiology and Behaviour at Cornell University, undertook a series of classic experiments on captive lovebirds to determine if their nest-building is genetically determined or the result of learning.16 Even though lovebirds are closely related, Dilger observed that different taxa use different strategies to obtain and transport nesting material to their pre-selected tree holes. The Fischer’s Lovebird, for example, carries a single strip of tree bark in its beak, whereas the Rosy-faced Lovebird simply stuffs multiple pieces of bark and leaves into its breast and rump feathers. Such bizarre behaviour is restricted to lovebirds and their close relatives, the hanging parrots (see below). Dilger speculated that lovebird nest-building arose from fortuitous occurrences based around two psittacine activities: chewing bark to keep bills sharp and preening. Although most parrots do not build nests, some will accidentally leave bits of material in their feathers when they shift from chewing to preening. According to Dilger, such oversights are likely to have initiated the behaviour of carrying nesting material that subsequently enabled them to build nests. Since the Rosy-faced Lovebird’s strategy appears less advanced than that of the Fischer’s Lovebird, Dilger deduced it to be the ancestral state, a conclusion supported by recent phylogenetic studies.
The American ornithologist then went on to hybridise Rosy-faced Lovebirds with Fischer’s Lovebirds to find out if nest-building is genetically programmed. Despite the hybrids being sterile, they still tried to breed and displayed nest-building behaviour intermediate between the two parental species. In other words, Dilger’s simple experiments confirmed that the nest-building of lovebirds is an innate response and not a learned one, although he had no idea as to the genetic mechanism.
Just when the biogeography of African parrots seemed to have been resolved, a remarkable fossil find was reported from Siberia in 2016.17 Nikita Zelenkov, a Russian palaeontologist, recognised an unidentified specimen in his institution’s collection as being the lower leg bone of a parrot. The fossil, from the early Miocene, was found on Olkhon Island in Lake Baikal, the deepest, largest and oldest lake in the world. Although only a single specimen, the 18- to 16-million-year-old fossil raises the intriguing question of whether any African species could have arrived from Asia, rather than all dispersing directly from Australasia.
Australia’s ancestral parrots diverged over 30 million years ago to produce two distinctive families – the Old World Parrots (family Psittaculidae) that spread throughout southeast Asia, the Pacific islands and Africa, and the cockatoos (family Cacatuidae) that became distributed throughout Australasia.
The Old World parrots colonised Indo-Malaysia on many separate occasions. The Guaiabero (‘eater of guavas’), which is endemic to the Philippines, split from its closest Australian relatives around 28 million years ago and was the first to arrive in the area.8 Its overseas dispersal was probably aided by a string of volcanic islands, the so-called East Philippines– Halmahera–South Caroline Arc, which approached New Guinea and the northern Australian plate at the time. Furthermore, the Australasian tectonic plate was slowly moving northwards and reached its present position in relation to Indo-Malaysia around 20–25 million years ago. It seems that all other splits between Australasian and Indo-Malayan taxa occurred after the two landmasses were in close contact, encouraged by the emerging archipelagos and the dispersal opportunities they offered. Colonisers included the racket-tailed parrots, the hanging parrots and the Afro-Asian ring-necked parakeets, a clade that also colonised Africa.
The 10 species of racket-tailed parrots (genus Prioniturus) are endemic to Indonesia and the Philippines and are easily distinguished by their elongated central tail feathers with a bare shaft and a spatula at the end. The hanging parrots (genus Loriculus) are a group of small birds with green plumage and short tails that have the unique ability among birds of sleeping upside down. Like lovebirds, hanging parrots collect nest material in their feathers, a behaviour that reflects their close evolutionary relationship. The Afro-Asian ring-necked parakeets (genus Psittacula) are highly gregarious, green-plumaged species with an extensive range across Africa, Asia and the islands of the Indian Ocean. They are one of a few parrot groups to have successfully adapted to living in disturbed habitats, and have withstood the onslaught of urbanisation and deforestation. The Rose-ringed Parakeet has even established feral populations in diverse urban environments from Europe to South America, where they have been able to withstand harsher climates than those in their native range. Speciation of all three genera resulted from a complex combination of island colonisations and subsequent divergences in allopatry among and within the island groups. Environmental changes in Asia, following the increased uplift of the Tibetan plateau and the onset of the Indian and east Asian monsoons during the late Miocene, also contributed to their diversification.18
Another member of the Psittaculidae, the Eclectus Parrot, is best known for a form of reversed sexual dichromatism (plumage colouration) not seen in any other bird. The males of this Australasian species are bright emerald-green, whereas the females are vermillion with a vest of violet or cobalt. Indeed, for a long time, they were thought to be separate species. It turns out that the colour differences are the result of interplay of sexual and natural selection. Females remain secure inside their hollow tree nest sites for up to 11 months a year and have evolved bright red plumage to advertise that the hollow, a scarce resource in their habitat, is occupied. Males, in contrast, spend most of their time foraging in the rainforest canopy and have evolved cryptic green plumage for protection against predators. The species is also unusual in that they can control the sex of their offspring, although how and why are not yet known.19
Five to ten million years later, the brightly coloured lories and lorikeets split from their closest relatives, the Budgerigars, during the middle Miocene and radiated through the islands off northern Australia to colonise Sulawesi and Bali, the Philippines and several Pacific islands, as well as Australia. Given their recent divergence, they are an unexpectedly species-rich lineage, a fact that reflects a key innovation: a dietary shift from seeds to nectar.20 This significant evolutionary event allowed an expansion into new ecological niches and led to rapid speciation through allopatric partitioning. As a consequence, lories and lorikeets underwent several anatomical modifications. Their bills became narrower and less powerful than those of other parrots, and they acquired specialised brush tongues with papillae at the tip to help mop up their food. Since nectar is easier to digest than seeds, they also evolved shorter intestines and thinner-walled, weaker gizzards, as there was no need for a grinding function.
The Budgerigar lineage remained in Australia’s harsh interior, where today they roam widely, often in large flocks, breeding opportunistically when the intermittent rains produce enough grass seeds to sustain a clutch of chicks. Native birds display a light green body with pitch-black mantle markings, edged in bright yellow undulations, with a cobalt-coloured tail. Such colouration is thought to have evolved as the result of selection pressures imposed by a lifestyle of feeding on the ground and the need for cryptic plumage to help blend into the grasses and hide from predators. The first live birds were brought to Europe in 1840 by John Gould, where they soon became popular pets, owing to their playful personality, intelligence and mimicry skills. Shortly afterwards, aviculturists began to select mutations that diverged from the wild-type colouring, and today captive birds can be found in a variety of shades that include blue, grey, grey-green, violet and white.
During the mid-Miocene, the Australian plate approached and collided with the Asian plate, causing an uplift of the northern areas and a change in climate, with cooler temperatures and more arid conditions. The continent’s vast rainforests became fragmented, and the vegetation changed into a mosaic of different types, including the emergence of other broad-leaf forests, eucalyptus, fire-adapted sclerophyll vegetation, grasslands and saltbush plains. The early–middle Pliocene was a significant period for migration between Australia and southeast Asia, and it is likely that the broad-tailed or platycercine parrots, as well as the cockatoos, spread and diversified into the drier habitats at this time.
The cockatoos are a distinctive family of parrots that are recognisable by their showy crests and curved bills. Their plumage is less colourful than that of other parrots, being mainly white, grey or black, although they may have colour in their crest, cheeks or tail. In 2011, Nicole White, a doctoral student at Murdoch University in Perth, compared six genes from 16 of the 21 different cockatoo species to construct a robust phylogenetic tree.21 The results revealed that the Cockatiel is the most basal species, with the black cockatoos arising next, while the Palm Cockatoo is sister to a clade composed of the Gang-gang Cockatoo, the Galah and the white cockatoos of the genus Cacatua. This well-supported family tree has some interesting implications for several physical features of cockatoos. It indicates that the relatively immobile crest of the Cockatiel is an ancestral trait in the cockatoos, and that the fully erectile crest found in the other cockatoos evolved after this lineage split from the common ancestor with the Cockatiel. It also suggests that specific morphological adaptations such as plumage colour, body size, wing shape and bill structure evolved in parallel or convergently across the different lineages. For example, the large black Palm Cockatoo is more closely related to the white cockatoos than to other black cockatoos.
Unexpected kinships
Recovering the phylogenetic relationship of parrots to other avian families has not been easy. Although they are distinctive and morphologically similar, there are no apparent intermediary forms that link the clade to other well-defined groups. This limitation is reflected in the results of early morphology-based studies that variously suggested a close affinity to both woodpeckers and rollers, and even to cuckoos. But, in 2008, an international study published results that not only clarified the situation but required the avian tree of life to be rewritten. Led by Shannon Hackett, head of the bird division at Chicago’s Field Museum, the research group sequenced large sections of genome-wide DNA from 169 species, representing all the major living groups.5 At the time, it was the most complex avian phylogenetic study ever undertaken, both for the number of species involved and for the amount of DNA analysed. It turns out that parrots are the closest evolutionary relatives of the passerines, a group that includes the songbirds. Irrespective of the statistical approach Hackett’s team used, the results of the analyses were the same: parrots consistently emerged as the sister group to the passerines. In other words, the closest relatives of the blackbirds and thrushes living in your back garden are the colourful macaws and lorikeets of the tropics. Despite the results closely matching a preliminary study published two years earlier by Per Ericson and colleagues, it still provoked considerable debate among the scientific and birding communities.22 Could it be true, or was it simply an artefact?
Three years later, a team of German investigators led by Münster University graduate student Alexander Suh undertook a more detailed examination of avian evolution using a different method: retroposons, or ‘jumping genes’.23 These features are repetitive DNA fragments which can be inserted randomly anywhere in the genome after being copied, or ‘reverse transcribed’, from an RNA intermediary. While the original repetitive retroposon sequence is inherited like any other piece of DNA, the new insertion, together with any sequence alteration, is unique and will be inherited unchanged from the time of its insertion. In effect, retroposons are ‘molecular fossils’ that can be followed over evolutionary timescales, a feature used by Suh and his colleagues to obtain an improved resolution of the avian tree of life.
After evaluating thousands of retroposon insertions, Suh found seven that were unique to falcons, parrots and passerines, but absent in all other landbirds, including woodpeckers, rollers and cuckoos. The implications of this finding are twofold: first, falcons are not closely related to other birds of prey as previously thought, and, second, falcons, parrots and passerines have evolved from a common ancestor. Crucially, an additional three insertions were restricted to parrots and passerines, indicating that these two groups are each other’s closest living relatives. In other words, Hackett’s team was right all along.
The falcons (family Falconidae), including caracaras, forest falcons and falcons, began to diversify during the late Oligocene–early Miocene within the Neotropics. However, the species-rich genus Falco diversified later, around 5–7 million years ago, when climate change resulted in the expansion of grasslands and savannas, with their associated mammalian communities.24
Since the position of passerines on the avian tree of life has now been highlighted, it is time to move on and hear their story.