The validity of Nestoridae and Cacatuidae (1 Viewer)

Recently, in delving through the literature, I have noticed the advocation of raising the New Zealand Kakapo, Kaka, and Kea into a new family, the Nestoridae. Don Roberson, on his Bird Families of the World Does, cites the Christin and Boles checklist to the Australasian region as his explanation. I assume the IOC will eventually follow suit, and John Boyd’s bird taxonomy page also split off this family. One key reference they site is:

de Kloet, R.S., and S.R. de Kloet. 2005. The evolution of the spindlin gene in birds: Sequence analysis of an intron of the spindlin W and Z gene reveals four major divisions of the Psittaciformes. Molec. Phylog. Evol. 36: 706-721.

While the split seems okay, one thing no one ever seems to notice is that in this study, the Cockatoos (Cacatuidae) are nested well within Psittacidae. In order to keep them a legitimate family, we would have to split Psittacidae into 4 different families (besides Nestoridae and Cacatuidae). This would include separate families for the Australian parrot clade (Eclectus, Rosellas, etc), and African parrot clade (Gray and Senegal are the only species sampled that would fall here), A New World clade, which also includes some parrots from New Guinea and Madagascar, and a lovebird and lorikeet clade. As yet, no one I know has done this, despite everyone considering cockatoos a good family. And probably with good reason, since this study really could use a wider sampling of taxa, better outgroups, and maybe more genes.

I am just personally surprised, (and a little annoyed), that people cite this study to support Nestoridae, then ignore it’s other results. To me, personally, at this point it seem premature to begin the carving up of the Psittaciformes, when so little work has yet been published

Thoughts on this issue?

I have had look at the justification given in the new Australian checklist for this. They do cite the paper you quoted, which regarded the NZ parrots as a sister group to all remaining parrots. They also note that cockatoos were one of the remaining clades.

However they then go on to cite the following paper: Astuti, D et al (2006) Phylogenetic relationships within Parrots (Psittacidae) inferred from mitochondrial cytochrome-b gene sequences. Zoological Science (Tokyo) 23, 191-198. Unfortunately this did not include any of the NZ parrots (or representatives of some other clades) but it concluded cockatoos were sister taxon to the remainder. So the overall conclusion of recognising three families at the present time is reached from a combination of these two papers.

It also notes that Livezey and Zusi's 2007 paper based on an anatomical review recognised the same three families plus also treated lorikeets as a family.

Interestingly it also comments that it may well be the case that the Kakapo should be split at the family level from the Kea/Kaka family. That would be a pain for those of us who would like to see a representative of every bird family in the wild one day!

Cheers

Murray Lord
Sydney, Australia

Mysticete said:

Recently, in delving through the literature, I have noticed the advocation of raising the New Zealand Kakapo, Kaka, and Kea into a new family, the Nestoridae. Don Roberson, on his Bird Families of the World Does, cites the Christin and Boles checklist to the Australasian region as his explanation. I assume the IOC will eventually follow suit, and John Boyd’s bird taxonomy page also split off this family. One key reference they site is:

Click to expand...

Were would that leave Ground and Night Parrot? Aren't they related to the Kakapo?


Cheers!

Dimitris

Thanks for the Astuti reference; I looked it up.

One thing I notice off the bat was that Astuti reference mostly concerns itself with Australasian forms, while the reference I made note of is looking more at global patterns. Depressingly, it's hard to evaluate one tree's relationships compared to another, as there seems to be minor overlap between the two studies, and the number of taxon, and choice of taxa, can make a big difference in how your relationships resolve. I would say both studies have some major flaws in using them to determine the broader level relationships of parrots and kin.

At any rate, hopefully a more comprehensive phylogeny will be published soon, with better outgrouping, more species, and more genes

Mysticete said:

While the split seems okay, one thing no one ever seems to notice is that in this study, the Cockatoos (Cacatuidae) are nested well within Psittacidae. In order to keep them a legitimate family, we would have to split Psittacidae into 4 different families (besides Nestoridae and Cacatuidae). This would include separate families for the Australian parrot clade (Eclectus, Rosellas, etc), and African parrot clade (Gray and Senegal are the only species sampled that would fall here), A New World clade, which also includes some parrots from New Guinea and Madagascar, and a lovebird and lorikeet clade.

Click to expand...

I do not think this paper actually suggests this.

De Kloet & de Kloet studied a gene that is located on the heterochromosomes (= sex chromosomes - Z and W; male birds have two Z chromosomes, females have a Z and a W). The Z and W chromosomes never recombine, thus a gene that is present on both of them exists in every species in two copies, that have been evolving indepedently since the individualisation of the sex chromosomes (in birds, this seems to have occurred at the base of Neognathae).

Your comments seem to be based only on the first tree published in the paper (Fig. 1), which is based on the Z chromosome copy of the gene and does not take insertions/deletions into account. This tree suggests the most basal subdivision in Psittaciformes is between Nestor+Strigops and all the other members of the order, with very strong support - i.e., both sister clades are strongly supported: 100% for Nestoridae, 92% for its broader sister clade. The relationships within this second clade, however, are poorly resolved, none of the basal groups having a support reaching 70%. To move from the position where they appear in this tree to a basal position, the cockatoo group would in fact not have to go through any strongly supported node.

There is a second tree in the paper (Fig. 3) based on the same copy of the gene, but that was generated through an analysis taking insertions/deletions into account. This tree supports the same basal-most subdivision as the first one, but the topology within the broader main clade appears sensibly different. The "Australian parrot clade" and "lovebird and lorikeet clade" are not reciprocally monophyletic anymore, but form together a strongly supported clade (termed "Group A" in the paper, 96%), that was not apparent in the first tree. The remaining species ("Group B") form the sister group of this clade, but the monophyly of this group still has limited support only (72%). The "New World clade, which also includes some parrots from New Guinea and Madagascar" is not supported at all: the NW species cluster (weakly: 52%) with African parrots, instead of grouping with the New Guinea and Madagascar species.

The third tree in the paper (Fig. 6) is the result of the analysis of the W chromosome copy of the gene. It has a smaller taxonomic sampling, but places the cockatoos basal to a very strongly supported clade (99%) that includes two Agapornis spp, Psephotus haematonotus, Pyrrhura rhodogaster, Coracopsis vasa, Psittrichas fulgidus, and Psittacus erithacus. Agapornis and Psephotus are parts of the "Group A"; the other species are all from the "Group B", and include a NW taxon (Pyrrhura), an African Parrot (Psittacus), and the two "New Guinea and Madagascar" species that move around in this group in the first two trees (i.e., members of every single well-supported subgroup of the "Group B", except the cockatoos).

This third analysis suggests very strongly that cockatoos are basal to all other parrots except Nestor and Strigops. The other two trees do not provide support for this position, but do not exclude it either. Thus, all in all, the paper does provide support to a topology that is compatible with separating Nestoridae and Cacatuidae.


There is at least another paper that had a broad taxonomic representation and suggested the same relationships, this one based on 12S/16S-rRNA sequences (although the monophyly of the clade excluding Nestoridae and Cacatuidae was not very strongly supported there either):
Tokita M., Kiyoshi T., Armstrong K.N. (2007): Evolution of craniofacial novelty in parrots through developmental modularity and heterochrony. Evol. Dev. 9(6):590-601.
http://www3.interscience.wiley.com/journal/118546207/abstract
(The pdf of the paper is not freely accessible, but the pdf of the supplementary material is, and it has some trees.)

Incidentally, RAG-1 sequences from GenBank also support these relationships very strongly - a tree including the necessary taxa has never been published to my knowlegde, but see the attached file. ('Hope it's readable, this time... )

Regards,
Laurent -

PS - Night and Ground Parrots seem to be related to Neophema and Melopsittacus, rather than to Nestor/Strigops (http://elibrary.unm.edu/sora/Auk/v111n04/p0833-p0843.pdf), which would place them in de Kloet & de Kloet's "Group A".

Mysticete said:

At any rate, hopefully a more comprehensive phylogeny will be published soon, with better outgrouping, more species, and more genes

Click to expand...

Watch out for:
Wright T.F., Schirtzinger E.E., Matsumoto T., Eberhard J.R., Graves G.R., Fleischer R.C. (in review): A multi-locus molecular phylogeny of the parrots (Psittaciformes): Support for a Gondwanan origin during the Cretaceous. Mol. Biol. Evol.

This upcoming paper in Mol. Biol. Evol. seems to have derived from research presented at the AOU meeting at Santa Barbara, California, in 2005, known as the one hundred and twenty-third stated meeting of the American Ornithologists' Union.
Session 5A. Systematics F. Keith Barker, Chair.
A framework phylogeny of the parrots (Psittaciformes) derived from mitochondrial coding and nuclear intron sequences. Tmothy F. Wright, Erin E. Schirtzinger Dept Biol., New Mexico State univ., Las Cruces N.M., Tania Matsumoto, Dept. Biologia, Universidade de Sao Paulo, S.P. Brasil, Jessica R. Eberhard, Dept. Biol., Louisiana State Univ., Baton Rouge, LA, Gary R. Graves, Natl. Mus. Nat. Hist., Smithsonian Inst., Washington D.C., and Robert Fleischer, Genetics Prog., Natl. Mus. Natl. Hist.

Here is the abstract:

The parrots (Order Psittaciformes) exhibit a number of unusual characteristics of general interest in evolutionary biology. Most phylogenies to date have focused on relationships among select taxa. We investigated higher-level phylogenetic relationships in the parrots using 3.7 kb of sequence data from 2 mitochondrial genes (ND2 and COI) and 3 nuclear introns (Tropomyosin intron 5, Rhodopsin intron 1, and Transforming Growth Factor ß-2 intron 5). 50 species representing 46 parrot genera and 4 avian outgroup taxa were sampled. A phylogeny reconstructed from the combined partitions was well resolved and had strong support at most nodes. It showed the New Zealand kea (Nestor notables) as sister to a clade containing all other parrot taxa. Strong nodal support was found for such traditional parrot groups as the Australian cockatoos, the Australasian lories, and the Neotropical long-ailed and short-tailed parrots. African taxa were paraphyletic, with the genera Psittacus and Poicephalus forming one well-supported clade and the genus Agapornis in a distinct clade that included the Asian genera Bolbopsittacus and Loriculus. The budgerigar, Melopsittacus undulatus, formed a clade with the Australian Psittaculirostrus edwardsi that was sister to the lories and lorikeets. These results will facilitate comparative studies in ecology, behavior and molecular evolution in parrots.

For the upcoming paper Dr. Sara Capelli, Dr. Jula Scharpegge, and Dr. Heinrich Muller are all staff members of the The Loro Parque Fundación in Spain. Geoffrey K. Chambers is at the Institute for Molecular Systematics, School of Biological Sciences, Victoria University of Wellington, New Zealand. Dr. Robert C. Fleischer is head of the genetics program at the Smithsonian, Natl. Mus. Nat. Hist. D.C.

I look forward to reading it and thanks for all the information from the other posters.

Some possible papers to read until it is published:
http://www.biology.lsu.edu/webfac/jeberhard/JRE_research.htm .
Phylogeny and biogeography of the Amazona ochrocephala complex. The Auk

Duplication and concerted evolution of the mitochondrial control region in the parrot genus Amazona.
http://biology-web.nmsu.edu/twright/EberhardetalMolBiolEvol2001.pdf .

mb1848 said:

This upcoming paper in Mol. Biol. Evol. seems to have derived from research presented at the AOU meeting at Santa Barbara, California, in 2005, known as the one hundred and twenty-third stated meeting of the American Ornithologists' Union.

Click to expand...

Yes, this is certainly the same research program, but the data set seems to have grown up a bit since 2005. Their tropomyosin sequences are already in GenBank (the other genes are not; or if they are, they have not yet been made freely accessible), and they include 67 species of parrots, plus 8 avian outgroup taxa (contra 50 and 4 in the AOU meeting abstract).
Unfortunately for Dimitris' query, though, the data set doesn't appear to include Pezoporus/Geopsittacus.
But anyway, it's certainly going to make an interesting reading.

L -

l_raty said:

Wright T.F., Schirtzinger E.E., Matsumoto T., Eberhard J.R., Graves G.R., Fleischer R.C. (in review): A multi-locus molecular phylogeny of the parrots (Psittaciformes): Support for a Gondwanan origin during the Cretaceous. Mol. Biol. Evol.

Click to expand...

http://mbe.oxfordjournals.org/cgi/content/abstract/25/10/2141
Preprint here : http://biology-web.nmsu.edu/twright/

L -