Snares Tomtit and Snares Fernbird

Discussion about the evolution, relationships, and naming of New Zealand birds
User avatar
Michael Szabo
Posts: 2533
Joined: Sun May 08, 2011 12:30 pm
Contact:

Snares Tomtit and Snares Fernbird

Postby Michael Szabo » Thu Nov 03, 2016 11:07 am

Snares Tomtit and Snares Fernbird are to be split as part of changes to BirdLife International's taxonomy for passerines:

http://www.birdlife.org/globally-threat ... hreatened/

http://www.birdlife.org/globally-threat ... ulnerable/
'Birds New Zealand' Facebook page: https://www.facebook.com/Birdsnewzealand
Ian Southey
Posts: 1100
Joined: Tue Sep 08, 2009 5:45 pm

Re: Snares Tomtit and Snares Fernbird

Postby Ian Southey » Tue Nov 08, 2016 4:50 pm

So, there goes Bowdleria - again. Maybe it's time to take this seriously as we have discussed this before (viewtopic.php?f=13&t=4675) but evidence to the contrary has not been forthcoming. Poodytes then?

It would be interesting to see the reasoning behind these changes. It's not that I disagree but I wonder why split only these species? Snares Fernbirds are the most striking of the living fernbirds - at least to look at but was there a proper analysis to show that the distinction between the others is less important? We have recently seen in the split of the Stewart Island Shag, that substantial genetic differences may leave only small superficial morphological changes.

Picking out the Snares Island Tomtit from the crowd because it is big and black (if that's what they did) trivialises the very detailed and interesting work carried out by Charles Fleming here - http://rsnz.natlib.govt.nz/volume/rsnz_ ... 00720.html and here - http://rsnz.natlib.govt.nz/volume/rsnz_ ... 01680.html where he thought the Snares Island Tomtits were relatively similar to the Auckland Island Tomtits and that the Chatham Island Tomtits were the most distinctive of the lot. There is an interesting DNA study here - http://thirdworld.nl/a-molecular-phylog ... -sequences but it uses a small piece of DNA only and is probably not the last word. This work suggested that the differentiation of all of the tomtits is very recent, i.e. the differences between them are very small but Fleming showed that they are all quite distinct.

On the whole this seems to be a quick and dirty approach which may aid the protection of some kinds of birds but it is hard to see it as the last word. These splits seem OK to me but what did they miss?

Ian
User avatar
Neil Fitzgerald
Site Admin
Posts: 3637
Joined: Mon May 18, 2009 10:20 am
Location: Hamilton, New Zealand
Contact:

Re: Snares Tomtit and Snares Fernbird

Postby Neil Fitzgerald » Thu Nov 10, 2016 10:45 pm

The Birdlife page doesn't give a reference for the fernbird change they have coming in their checklist. We shouldn't have to wait for the checklist, just tell us the reference or give the reason straight up. Anyway, if we take a guess that it's Howard and Moore, is anything new being presented here? I'd be interested in Sco's thoughts on this.
Paul Scofield
Posts: 153
Joined: Sun Jul 10, 2011 10:39 am

Re: Snares Tomtit and Snares Fernbird

Postby Paul Scofield » Mon Nov 14, 2016 12:24 pm

Neil and others,

I do not support the lumping of Bowdleria into Poodytes. There is incredibly insignificant evidence for this in the literature and it is typical of the sort of knee-jerk taxonomy that is becoming commonplace with the IOC and Birdlife - it unscientific, stupid and frankly bloody annoying!

These "taxonomists" see a paper written to examine one hypothesis that has some genetic data from other taxa incidental to the original paper and they think - "oh I'll be clever and preempt their results". It is a brilliant example of of the axiom "a little bit of knowledge is a dangerous thing"

It is also really dangerous for taxa at whatever level to be lumped, especially when they are endangered - in todays world the amount of money spent on a species is often dependent on its "uniqueness".

Some background:

In an important and robust analysis of Locustella Alstrom et al (2011b, 2015b) used a single mitochondrial gene (cyt B) of Bowdleria to basically flesh out the outgroups in a multi gene (mito-mitochondrial and nucleic) analysis.

Because the two are clearly closely related but expecially because the single gene from a single taxa of Bowdleria was compared to a full sampled Poodytes graminous the two taxa form an artificial but highly supported clade. This is entirely fallacious though! If one of your mito-genes DNA and a full chicken genome were added to a tree full of bacteria some taxonomist would probably conclude you and chicken were co-generic.

To make sound taxonomic decisions especially at a generic level you have to compare like with like and also include all likely relatives. Alstrom's papers do neither but its not their fault. They NEVER intended to make a taxonomic decision about Bowdleria that is entirely some nameless moron in the IOCs decision...

Whats should we do?

A much more logical (but unpublished) analysis of the results that was done by Boyd (http://jboyd.net/Taxo/List21.html#locustellidae) who states

"It seemed appropriate to resurrect [SIC] the genus Bowdleria (Rothschild 1896, punctata) for the New Zealand fernbirds. That and the phylogeny mean that the Little Grassbird needs its own genus. The name Poodytes (Cabanis 1850) is available."

Other classic example of this taxonomic "half-arsedness" are the messes the SAME people have made with Great Egret, Barn Owl and Intermediate Egret. Why can't these folks wait till someone does a job properly before jumping to incorrect conclusions based on interim results that are contradictory, incomplete and often not even from studies designed to answer any taxonomic question?

But even then these papers have fish hooks - A nice little paper came out the other day that shows that there is entirely separate clade of eastern Barn Owls but the devil is in the detail. This paper suggests that many of the endemic island forms of Tyto are probably actually part of the mix and that the supposed "Australian Barn Owl" is actually pretty similar to those in Asia.

Paul
User avatar
Neil Fitzgerald
Site Admin
Posts: 3637
Joined: Mon May 18, 2009 10:20 am
Location: Hamilton, New Zealand
Contact:

Re: Snares Tomtit and Snares Fernbird

Postby Neil Fitzgerald » Mon Nov 14, 2016 4:03 pm

Thanks Paul. My gut feeling was something like that too. Proper study please.
N
Byrd
Posts: 825
Joined: Fri Dec 04, 2009 12:53 pm
Location: Wellington

Re: Snares Tomtit and Snares Fernbird

Postby Byrd » Mon Nov 14, 2016 4:07 pm

BirdLife now use the so-called "Tobias Criteria" for determining species. This uses an algorithm based on physical and behavioural elements (invented by a group headed by Joseph Tobias). This gives a numerical value for such things as morphology, behaviour, voice, distribution, etc. Apply the algorithm to a bird and the resulting value determines whether it is a species or subspecies. Genetic data is largely (not entirely) ignored.

The "Tobias criteria" is a good idea in some ways - in broad terms it can determine relatively simply whether a particular taxon is "valuable" enough for protection without having to wait for DNA testing - but it falls down in not taking the genetic evidence into account. And there is some criticism that for particularly uniform groups of birds - say swifts or warblers - it may not be at all accurate. I imagine there would also be quite a few birds where the behaviour and voice simply weren't well enough known - or even known at all in many neotropical forms - to take them into account, and splits/non-splits would therefore be judged basically on morphology (e.g. coloured parts, size of crest, length of tail, etc). It has been argued that the "Tobias criteria" method of using physical differences and ignoring genetic differences is pretty much the exact opposite of most taxonomy research nowadays. More importantly, there is the criticism that there is not a body of back-up work to the decisions. Generally speaking when taxa are split or lumped, there are reasons given to the effect of "we did this research, and took into account these peoples' studies, etc etc". With BirdLife's "Tobias" splits it is more a case of "we have split/lumped this. Accept it."

I do get the intent behind the use of the "Tobias criteria" (they are covering every bird in the world and need an all-encompassing technique in the absence of genetic data for many of them) but it lacks finesse.


(I just cut and pasted that from another forum I wrote it on, because I'm short on time. At the moment I'm in India trying to deal with the country's currency meltdown....)
Ian Southey
Posts: 1100
Joined: Tue Sep 08, 2009 5:45 pm

Re: Snares Tomtit and Snares Fernbird

Postby Ian Southey » Tue Nov 15, 2016 9:33 am

We're talking at two levels here and I can't help with the genera but here is a link to the thinking behind "Tobias criteria" and the magic number 7. http://people.ds.cam.ac.uk/cns26/njc/Pa ... tation.pdf

To me it seems a bit arbitrary but a good first run at breaking down some of the enormous aggregations of subspecies that mid 20th century taxonomists liked. It is interesting in at least a couple of ways - not including DNA data and by giving some (small) weight to ecological differences which is new. The obvious improvement on this would be a detailed knowledge of the birds in life which these people do not have. What I also suspect, from not having seen the reference listed, is that they did not consult Fleming's detailed work on the Tomtits. If so this really would make it a quick and dirty piece of work in my book. It would be nice to actually see it, however, rather than just guess at what might, or might not be in it.

The two "species" they picked out are distinctive but it creates a difficulty singling them out as it devalues other relatives which, probably, are mostly equally distinct, and serves nothing for conservation as both are already as well protected as they can be.

Ian

Return to “Bird Taxonomy and Nomenclature”