DUCKS, CHICKENS, STILTS, FLAMINGOS

[darren.naish@port.ac.uk]

This message has now been delayed by a few days  - I had to store it 
on disk because of lack of time last week - and Matt already 
responded to some of it. But here is some more; I'm combining 
responses to both Matt Troutman and Ronald Orenstein here. Might 
work, might not.

On the interrelatedness, or not, of charadriiforms, ducks and 
flamingos, Ron wrote...

> You may be thinking of Juncitarsus, which Olson and Feduccia 
> consider a "shorebird-flamingo mosaic" (see Feduccia's book at p. 
> 208), a view supported by Stefan Peters.

Olson and Feduccia (1980), in their big _Smithsonian Contributions to 
Zoology_ paper on flamingo affinities, asserted that _Cladorhynchus_, 
the Australian banded stilt, was a close relative of flamingos. As 
some of us have pointed out on this list before (see 
http://www.cmnh.org/fun/dinosaur-archive/1996Jan/0211.html), the two 
share some characters: the structure of their eggshells, the habit of 
breeding in massive colonies in shallow alkaline lakes, and a few 
vague morphological things like a certain leg muscle (the 
iliotibialis medius if memory serves). The proposition, thus, is that 
there is a _Cladorhynchus_ + Phoenicopteriformes clade.

In the same paper, Olson and Feduccia also assert that 
_Juncitarsus_, a long-legged Eocene wading bird known from both 
Messel and the Green River Fm, is a primitive flamingo. Personally I 
think this is right, and apparently quite a few others do as well. 

Now, this should not be confused with Feduccia's earlier speculations 
that flamingos evolved from stilt-like ancestors and that 
presbyornithids diverged from this lineage and evolved into ducks. If 
this all sounds very vague and extremely confusing, that's because it 
is: flamingos, ducks, presbyornithids and waders were all linked 
together into a sort of messy nexus that would, supposedly, be 
unresolvable by cladistic methods. 

In allying anseriforms with charadriiforms and flamingos (something 
that may (Ericson 1997) or may not (Livezy 1997) be correct), Olson 
and Feduccia have of course removed any link between anseriforms and 
galliforms. As explained in my previous post on this subject, I don't 
think they're right: I reread their 1980 paper on this on the train 
this morning (on the same journey I saw my first ever wild egret) 
and am unimpressed with the way they deal with the problem. Their 
whole refutation of the anseriform-galliform link consists of  (with 
regard to parasphenoid rostrum)  'people say these are similar, we 
found they were different', (with regard to retroarticular 
process) 'people say these are similar, we found they were different' 
and (with regard to quadrates) 'people say these are similar, we 
found they were different'. Followers of the bird-theropod thing will 
of course be familiar with this approach to the evidence for the 
derivation of birds from other theropods (Feduccia 1996). 

One other thing while on the subject... Olson and Feduccia (1980) 
were the first authors to show that screamers did actually possess 
lamellae (an anseriform character). They also emphasised that 
screamers have only partially webbed feet. However, I have never 
understood why they have therefore argued vehemently that screamers 
are strongly modified, very specialized anseriforms: surely the 
alternative, that screamers are primitive, basal anseriforms, is more 
likely? Indeed, this is the conclusion Livezy (1997) has come to, and 
it is perhaps confirmed by the early appearance of the anhimids in 
the fossil record (Feduccia (1996) talks about an undescribed 
specimen from the Eocene I think). I cannot help but feel that Olson 
and Feduccia were trying their very hardest to keep screamers as far 
as possible from the galliforms, a group they have obvious 
similarities to. It may be that they tried to achieve this by making 
out that screamers were very specialized anseriforms. 

On the retroarticular process, the morphology of which is often cited 
as a strong indicator of affinity between anseriforms and galliforms, 
Matt wrote...

> I can concede that the retroarticular processes are similiar in the 
> two groups and they are possibly some of the best evidence for the 
> 'duck-chicken' relationship in the sarcastic Olson and Feduccia 
> sense. However, as noted, these differ in the details and if I 
> remember correctly (somebody, confirm or deny) diatrymid 
> retroarticular processes do not show compression and similiar 
> details.  However, I would not place much stock in this argument 
> for diatrymids are relatively specialized birds.

Ron wrote in response...

> Without checking specimens, let me remind you that the 
> retroarticular process serves, among other things, as the insertion 
> point for m. depressor mandibulae, the chief jaw-opening muscle in 
> birds.  Birds which open their mouths against force may have large, 
> well-developed processes. [snip].  IOW, beware of convergence!   

Ron is of course right to caution against perfunctory designations of 
characters as synapomorphous, when they are in fact only vaguely 
similar and quite probably of convergent origin.. however, with 
anseriforms and galliforms it is not just that their retroarticular 
processes are similar, but that they share the same unique 
morphology. This morphology is a mediolaterally compressed, upwardly 
curving blade-like structure. 

The problem is that this is _not_ seen in phasianids - the galliforms 
most readily available as study specimens for workers the world over 
- and those who have argued that anseriforms and galliforms are 
radically different with regard to this feature have relied heavily 
on the morphology of _Gallus_ (Interestingly, Witmer (1991) noted 
that reliance on _Gallus_ in studies of cranial pneumaticity in birds 
were also misleading as it is actually a very atypical bird!). 
Cracids and megapodes, I think univerally agreed upon as the most 
primitive of extant galliforms, do have the compressed blade-like 
retroarticular process, something Ericson (1997) pointed out in his 
recent paper (though if anything is clear from this discussion, it is 
that I will have to go and look at some specimens to make up my own 
mind).

Specifically with regard to the flamingo side of things, Ron said...

> I confess, though,  I am not so sure about comparing flamingos 
> to the living Banded  Stilt, a communal breeder with somewhat 
> flamingo-like behaviour that is surely a case of convergence.  
> The fossil evidence is  interesting, though.

Again, Cracraft has been quite vocal in disputing Olson and 
Feduccia's claims that _Cladorhynchus_ is the sister-group to 
flamingos. He argued against it in both his big _Auk_ paper from 
1981, and in his contribution to _Phylogeny of the Tetrapods, Vol. 1_ 
(1989?). His objections concern logical aspects of cladism, and not 
necessarily the data: while Olson and Feduccia seem to indicate that 
they visualise the whole flamingo-stilt thing as...

Charadriiformes
        Recurvirostridae
                (other recurvirostrids)
                _Cladorhynchus_
                Phoenicopteridae

.. they state elsewhere in their paper that flamingos should not be 
included **within** the recurvirostrids, but 'listed after them'. 
This implies, I think, that they are then saying that recurvirostrids 
and flamingos are sister-taxa.. which would then be contrary to their 
argument that _Cladorhynchus_ represents the sister-taxon to 
flamingos! The real problem here, of course, is hard-line cladistics 
vs. the intuitive paradigm; Cracraft is a strict adherant to the 
former, Olson and Feduccia to the latter.

Cracraft and others state plainly that the _Cladorhynchus_-flamingo 
thing is a convergence that has come about through similar nesting 
habits. Cyril Walker argued with me recently that that one muscle 
character - the iliotibialis medius or whatever it's called - common 
to both groups is hardly concrete proof of their affinity. I await to 
be convinced by a cladistic study that firmly allies flamingos with a 
group of birds other than charadriiforms, and agree with Olson and 
Feduccia that banded stilts and flamingos are indeed similar. Whether 
or not these similarities are indicative of convergence or affinity.. 
don't know.

"In this he was quite wrong and should have expressed some 
scepticism"
"Some of the most sceptical people I know are cryptozoologists!"
"Some of the most gullible people I know are cryptozoologists!"

DARREN NAISH
darren.naish@port.ac.uk