• iceberg
  • boy with flowers
  • checking water quality
  • planet eclipse
  • solarsystem model
  • rangitoto trees
  • kids with test tubes
  • kids with earth
  • snowy mountains
  • teens in physics class
  • Rainbow Clouds

    Refraction and diffraction of light through ice crystals in the clouds

  • Philippa On The Ice

    Philippa On The Ice Philippa Werry at an Antarctic research camp 2016

New Zealand Science Teacher

Science Curriculum/Scientific Literacy

Evolution of New Zealand parrots

This article was first published in New Zealand Science Teacher in 2003 and presents the evolutionary history of two parrot groups in their geological context. 

kea smallIn this article, the authors use the results of mtDNA studies to challenge traditional thinking about parrot evolution in New Zealand. The authors are E.J. Grant-Mackie, J.A. Grant-Mackie, W.M. Boon, and G.K. Chambers.

The article was originally published in a 2003 edition of New Zealand Science Teacher journal. We are reprinting it in response to a request from a school and with permission from Dr Jack Grant-Mackie.


Evolution of New Zealand parrots

Most readers will be familiar with the traditional view of parrot evolution in New Zealand put forward by Sir Charles Fleming (1979) that:

a)      Kaka – kea evolution occurred as a result of the rise of the Southern Alps and changing ice age sea levels, and

b)      New Zealand was the centre of evolution of the kakariki (genus Cyanoramphus), which then radiated into adjacent islands (Lord Howe, Norfolk, New Caledonia, Tahiti, Chathams, Antipodes, and Macquarie).

These ideas are being challenged to some extent by recent mtDNA studies at Victoria University of Wellington (Boon et al. 2000a, b). Their results confirm the existence in New Zealand of two distinct groups of parrots: the group of larger-sized kakapo-kaka-kea, and the various smaller kakariki (six or more forms). Worthy & Holdaway (2002) have recently bought much more information on our fossil and extant parrots together.

The purpose of this paper is three-fold. First, it attempts to present what is now thought to be the evolutionary history of the two parrot groups in their geological context (Part1). Next, it provides an example of the use of genetic analysis to answer the question, ‘What are the evolutionary relationships in the parakeet groups?; and finally it uses the parrot story to highlight how branches of science work together to solve a problem.

Evolution of the kakapo kakaFigure 1. Evolution of the kakapo/kaka lineage, from DNA and geological evidence. Timing of evolutionary events is show in middle lines (Ma = millions of years ago); the likely geological causes of those events is shown in the bottom lines; and the biological nature of each event is shown above in italics. (N. I. = North Island; S.I. = South Island.)

Part 1

The larger-sized group (Fig. 1) originated from a so-far unknown Australian ancestor about 1000 million years ago (Ma) as a proto-(kea/kaka) kakapo, due to the break-up of Gondwanaland and the isolated New Zealand area moving away from the Australian segment. The kakapo (genus Strigops) split from the lineage 60-80 Ma and is today our most ancient parrot. The kea split from the kaka line (genus Nestor) about 3 Ma and an early member migrated to produce the now-extinct but still unnamed Chatham Islands kaka; this occurred early enough for the group to become poor fliers (Millener 1999) in the absence of predators other than the falcon, but no DNA work has yet been done on this species. About 400,000 years ago the North and South Island kaka began to differentiate.

The DNA results can be examined in the light of known geologic evidence. The formation of the Tasman Sea is indeed dated as beginning in the Cretaceous, around 100 Ma (and being complete in its present form in the very Early Tertiary, about 60 Ma). New Zealand at that time was well forested, thus accommodating the ecologic divergence of the kakapo and kaka. The South Island for most of the Tertiary was not mountainous and we had no alpine zone. No alpine biota could evolve until an alpine environment existed. Around 5 Ma the Southern Alps began to rise, providing the opportunity under which the kea divided ecologically and genetically from the proto-kaka.

Geological proof that parrots were here during the Tertiary was recently provided (Worth et al. 2002) by identification of bones from mid-Tertiary (15-20 Ma) lake sediments from central Otago, but they have not yet been finally identified or fully described.

The ice ages (Pleistocene) began about 2 Ma, with each ice advance being accompanied by sea-level fall and forest retreat, the former uniting previously separated islands and the latter restricting the distribution of kaka. With new warm periods, ice melted, sea-level rose to form islands again, and forests advanced to cover more of the country. These physical changes provided the basis for kaka speciation in isolation on the North and South Islands (Huggins 2001). Both this development and that of the original proto-kakapo/kaka were vicariance events (speciation as a result of physical splitting of a pre-existing population). To some extent, so was the evolution of the kea, but ecologic isolation also played a part, despite kea and South Island kaka becoming sympatric. This took place at least by the last Glaciation, some 30,000 BP, as kaka moved with advancing forests during warming periods and some kea remained and adjusted to the climatic change (sympatry is shown by the occurrence of bones of both species together in some South Island fossil deposits, e.g., at Oamaru – Worth & Grant-Mackie 2003). A few lowland populations of kea remain, for example, in southern Westland forest.

There have previously been suggestions that the kakapo might be related to the Australian night and/or ground parrots, but this is not supported by modern DNA work. Instead the similarities (ground nesting, poor flight) are now suggested as products of behavioural/ecologic convergence (Boon 2000). The group as a whole evolved about 2.9 Ma from a relative of the New Caledonian horned parakeet (genus Eunymphicus), which in turn was derived at some earlier time from an unknown Australian ancestor, possible a proto-rosella (genus Platycercus).

The proto-C. saisetti arrived in New Zealand, probably via Norfolk Island, and has differentiated int his area during the last half-million years. The Chatham Islands endemic Forbes’ parakeet (C. forbesi, presently listed as a subspecies of C. auriceps in the 1990 Annotated checklist of New Zealand Birds) is the most ancient of these, followed first by the Antipodes Island green (C. unicolor) and yellow-crowned parakeets (C. auriceps), and then in order by the orange-fronted (C. malherbi), with the four red-croned subspecies (C. novaezelandiae novaezelandiae; C. n. chathamensis, C. n. cyanurus {Kermadecs} and C. n. subflavesens) being the most recently evolved. In addition, there are the Norfolk Island red-crowned parakeet, C. cooki, and the two subantarctic Reischeck’s parakeet subspecies C. e. hochstetteri (Antipodes).  There are also the two extinct species collected by James Cook in the Tuamotus; the Society parakeet C. ulietanus and the black-fronted parakeet C. zelandicus (see Boon et al. 2001 for details). This speciation has come about by migration and isolation and by ecologic and behavioural divergence.

New Zealand, Norfolk Island and New Caledonia lie on a sliver of continental crust (Norfolk Rise) that was once part of the margin of Gondwanaland, and Lord Howe Island lies on another sliver (Lord Howe Ridge) that joins the New Zealand area from the northwest. Modern bathymetry shows that neither of these submarine ridges, which formed as part of the break-up of Gondwanaland, has been continuous land, although present land areas were larger during lower sea levels of the ice ages. Thus the parakeets had to migrate over open water to reach New Zealand and in their subsequent dispersals. Nevertheless, migration to New Zealand and the Chatham Islands, at least, could have been made significantly easier by the presence of more extensive land areas during glacial epochs that occurred about 480-430,000 BP, 310-250,000 BP, 60,000 BP, and 40-15,000 BP. These are dates that coincide closely with DNA-derived dates for several of the parakeets (lower part of Fig. 2). The 625-450,000 BP timing of origination of extra-New Caledonian Cyanoramphus includes one further low sea-level stand of about 570-530,000 BP.

Work in progress suggests that Cyanoramphus follows a ‘boom and bust’ ecology rather like the Northern Hemisphere pinewood specialists, the crossbills. Parakeets are reported to continue continue breeding through the winter in most years for Antarctic Beech spp. This would have caused the birds to have reached very high densities, which could be the driving force for over-water dispersals (Chambers unpub. data).

Evolution of the kakarikiFig. 2: Evolution of kakariki (Cyanoramphus) as indicated by DNA studies. Origin of the genus (upper diagram), and evolutionary relationship of New Zealand and sub-Antarctic members (lower diagram). Arrows denote migration events and evolution in isolation. Figures denote approximate times of separation. (Akld. = Auckland Island, Chat. – Chatham Island, cr. = crowned, fr. = fronted, Is. = Island(s), N. Cal. = New Caledonia, Stew. = Stewart Island.)

Part 2

The molecular genetic techniques involved the use of the polymerase chain reaction (PCR), electrophoresis, and DNA sequencing. Tiny volumes (10 µL) of red blood cells were ruptured and their DNA extracted. The mitochondrial cytochrome b gene was used as a target for DNA sequencing because it shows a rapid rate of evolution and it is found on one piece of DNA only – the mitochondrial genome (mtDNA). Also, there are many copies of mtDNA per cell, and there is almost complete absence of recombination (as a result of mtDNA being derived only from the mother). A section of the gene and its control region were chosen for analysis.

The PCR regimes for the two sections of DNA are presented in the table below:

      Cytochrome b Control region
Initial denaturation (1x) 95°C for 3 min  95°C for 3 min
Denaturation 95°C for 40 sec  95°C for 15 sec
Annealing (35x) 55°C for 40 sec  55°C for 30 sec
Extension 72°C for 1 min  68°C for 2 min
Final extension (1x) 72°C for 10 min 68°C for 7 min


The desired DNA was identified and isolated, using an agarose gel electrophoresis technique. This enabled the DNA sequencing to be done readily on the gel-purified product.

The DNA sequencing reaction used the same primers that were used previously for the PCR step, plus other new ones which were designed specially, based on the new sequence data obtained.

Fluorescent dyes were used in conjunction with an automatic DNA sequencer, and the results then subjected to computer analysis.

The computer analysis compared the number of base substitutions in the different DNA samples and, on that basis, their degree of 'relatedness' was established. In general, the smaller the number of base sequence differences, the more closely related the individuals are. There was much less variation in the cytochrome b gene than for the Control Region. This is not surprising, given that the Control Region does not code for a protein product. As an interesting sideline, analysis of the protein product coded for by the cytochrome b gene showed two instances of amino acid substitution (ala for thr at amino acid position 53 for the Antipodes Island parakeet, and thr for ala at amino acid position 361 for the red-crowned parakeet.

Part 3

As science becomes more specialised and compartmentalised, it is easy to forget that much of current research still requires or, at least, may benefit from the collaboration between different disciplines. One of the main tenets of science is the idea of testability and consistency of results. The different branches of science are often able to either challenge or support the findings of other research. The parrot story is a good example of how support for an idea can follow from other disciplines when a new scientific explanation is put forward.

To be acceptable these DNA results need to be in accord with known geological data. In this instance, we have as yet no fossil record of the New Zealand parrots older than about 40,000 years ago (with the possible exception of some undetermined parrot bones from central Otago from around 15 Ma). Nevertheless, the timing of the break-up of our part of Gondwanaland, the geological history of our vegetation (as shown especially by the pollen record), and the origin of the Southern Alps are all quite well established, and each line of geological evidence does provide strong support for conclusions based on the DNA work. At the same time, it explains the ecologic pressure that allowed (or 'caused') the parrot development, with divergence of the proto-kaka/kakapo group as a result of the break-up of Gondwanaland, and origin of the kea with the origin of the Southern Alps. Of course, any explanation that is able to provide a causal relationship will be more convincing and acceptable to critical and sceptical observers (as we should all aim to be!). At this stage we do not have the geological details related to the arrival and evolution of the kakariki, but there are none that would contradict the DNA picture.

Recently too, mtDNA analysis and behavioural work combined to answer the important question of whether the orange-fronted and yellow-crowned parakeets are separate species, or simply colour variants. This is a significant issue because the orange-fronted parakeet is critically endangered, and therefore requires special conservation status if it is genetically isolated from its more numerous yellow-crowned relative.

In addition to the mtDNA work described above, a behavioural study was undertaken on sympatric populations of orange-fronted (C. malherbi) and yellow-crowned (C. auriceps) parakeets over a three-year period in the Hurunut Valley (North Canterbury) during the breeding season (Boon et al. 2000a). The Hope Valley (Fiordland) population was also checked but seems to be extinct - only one hybrid bird was recovered (Boon et al. 2000a). Birds were initially identified to species and their locations plotted, and then studied to see if they were paired. The criteria for being paired included: if they appeared to be cooperating together (e.g., preening, courting, mating, nest-building, egg-sitting, feeding of young), and if there was no evidence of aggressive interactions. It was already known that New Zealand parakeets do not exhibit breeding behaviour more than 50 m from their proposed nest site, so there was little chance of multiple counting of the same birds. At no time during this period was mixed-pairing of birds observed.

The complete lack (except in the Hope Valley) of mixed pairings between yellow-crowned and orange-fronted parakeets reinforces the mtDNA analysis and makes it clear that the orange-fronted and yellow-crowned kakariki are in fact separate species. Having said that, the fact that a hybrid bird was found in the Hope Valley seems to contradict the previous statement, until we realise that the birds can be crossed in aviaries and will hybridise in the wild, but only when all other sources of mating partners are exhausted (the last survivor syndrome).

This then raises the question of what constitutes a species. However, given that the birds do not naturally interbreed, conservation strategies are now required for C. malherbi that will take the restored specific status of this bird into account. 

Image at top: Nestor notabilis -Mount Aspiring National Park, New Zealand Licensed under CC BY-SA 2.0 via Wikimedia Commons.


Boon, W. M (2000). Molecular systematics and conservation of the Cyanorhamphus parakeet complex and the evolution of parrots. PhD thesis, Victoria University of Wellington.

Boon, W. M.; Kearvell,J. C.; Daugherty, C. H.; Chambers, G. K. (2000a). Molecular systematics of New Zealand Cyanarhamphus parakeets: conservation of orange-fronted and Forbes' parakeets. Bird Conservation International10: 211-239.

Boon, W. M.; Kearvell, J. C.; Daugherty, C. H.; Chambers, G. K. (2000b). Molecular systematics and conservation of kakariki (Cyanarhamphus spp.). Science for Conservation 176. Department of Conservation, Wellington.

Fleming, C. A. (1979). The geological history of New Zealand and its life. Auckland University Press, Auckland. 141 p.

Huggins, J. (2001). Genetic variation and conservation of kaka. MConSc thesis, Victoria University of Wellington.

Millener, P.R. (1999). The history of the Chatham Island's bird fauna of the last 7000 years- a chronicle of change and extinction. Smithsonian Contributions to Paleobiology89: 85-109.

Worthy, T. H.; Grant-Mackie, J. A. (2003). Late-Pleistocene avifaunas from Cape Wanbrow, Otago, South Island, New Zealand. Journal of the Royal Society of New Zealand 33: 427--485.

Worthy, T. H.; Holdaway, R. N. (2002). The lost world of the moa: prehistoric life of New Zealand. Canterbury University Press, Christchurch. 718 p.

Worthy, T. H.; Tennyson, A. J. D.; Jones, C.; McNamara, J. A. (2002). The Early Miocene {15-20 Ma) Manuherikia Group reveals New Zealand's first diverse Tertiary terrestrial fauna. Geological Society of New Zealand Miscellaneous Publication 112A: 58-59.

Post your comment


No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments