RECENT AVIAN PHYLOGENIES

["Matthew Troutman" <m_troutman@hotmail.com>]

Alas, through all of the talk about non-avian dino stuff, some REAL 
material :-)  

<<Chiappe does not believe that there is a group called "Saururae" that 
includes both the enantiornithes and earlier toothed birds including 
Archaeopteryx.  He pointed out a number of differences between 
enantiornithine birds and earlier forms, including a much shorter first 
digit (the first digit of Confuciusornis, for example, is very long, 
longer than metatarsal two).>>

Though I may be guilty of this one on occasion, difference is not 
evidence for non-relationship.  Chiappe is probably right about the 
paraphyly of Sauriurae (the way Martin et al. spell it but not the 
original spelling) and a good deal of evidence has been published that 
supports this and has caused me to sway.  Interestingly, the 
_Archaeopteryx_+Enantiornithes arrangement of Sauriurae is not really 
Martin's invention, but rather his and Kenneth Whetstone's.  Whetstone 
published the only good evidence for Sauriurae now that I look back in 
his study of the London _Archaeopteryx_ skull in 1983.  Curiously, 
Martin does not use those characters anymore (some have been disproved 
or put in doubt).  

<<Chiappe examined the three characters given as support for Saururae: 
the structure of the furcula, the ischium, and the fusion of the 
metatarsae. Though broad furcular arms are given as a character for 
Saururae, the furcula of Archaeopteryx is boomerang-like, while that in 
the enantiornithes is a very different V shape.>>

Again, that 'difference' thing.  Actually, Martin's character refers to 
the non-compressible construction of the furcula, which is different 
from the ornithurine state.  At SVP 1998, Martin et al. published 
evidence that suggests that the _Archaeopteryx_ and enantiornithine 
furcula developed differently from the ornithurine furcula.  I don't buy 
it.  

<<The anterodorsal ischial process does unite these taxa, but it is also 
found in more primitive forms such as Unenlagia and in Rahonavis.>>

Unless they themselves are members of the Sauriurae.  

<<The same is true of proximodistal metatarsal fusion, which is another 
primitive character.>>

As Martin pointed out in his huge 1991 paper on the anatomy of 
_Archaeopteryx_, ornithurines ossify their tarsometatarsus in a 
distoproximal fashion.  I think that it is just a convergent development 
in _Archaeopteryx_. 

<<The advantage Chiappe puts forward for his own tree, with a clade 
Pygostylia uniting enantiornithines and modern birds, is that it only 
requires a single evolution of advanced flight characteristics including 
a keeled sternum, alula, pygostyle, and strut-like coracoids.  Accepting 
Saururae as a taxon would require that these characters arose twice, 
once in the enantiornithine line and once in modern birds. Chiappe, in 
short, concludes that "Saururae" is clearly paraphyletic.>>

Pygostylia is actually Chatterjee's arrangement of birds above 
_Protoavis_.  Looks like Chiappe abandoned 'Ornithothoraces'.  

<<However, though I will confess to some knowledge of birds, my 
knowledge of molecular genetics is far less terrific.  I am not sure 
that I can really comment on a methodology that dismisses DNA-DNA 
hybridization, which used to be about as cutting edge as you can get, as 
old hat compared to gene sequencing.>>

I've talked at length with some ornithology cronies of mine about 
DNA-DNA hybridization and other molecular studies.  Though systematics 
is not their field, they agree that molecular studies are just as 
susceptible to convergence and homoplasy as morphological studies.  
True, molecular and DNA studies have clarified some things (you'll find 
out soon), but they can confuse things as much as morphological studies.  
Sibley and Ahlquist had some pretty strange arrangements on their tree 
that even avian systematicists that lean more to molecular evidence say 
must be wrong: e.g., _Aramus_, the limpkin, with sungrebes 
(Heliornithidae); touracos (Musophagidae) inside the Strigiformes (owls) 
with Caprimulgiformes (goatsuckers, potoos, nightjars, etc.); trogons 
(Trogonidae) away from Alcedinidae (kingfishers), Meropidae 
(bee-eaters), Momotidae (motmots), etc.; and, of course, Pelecaniformes 
are polyphyletic.  These may well be right, but most evidence suggests 
against them.  

<<For those of you who are not familiar with the debate, DNA-DNA
hybridization was used by the late Dr. Charles Sibley and John Ahlquist 
to completely revise our traditional understanding of the way that 
modern bird orders ought to be arranged.  Their classification has never 
quite replaced the traditional one, certainly in popular bird books at 
least, but it has at the very least aroused a lot of discussion and, in 
my view, has some intellectually satisfying aspects to it.>>

One thing that Sibley and Ahlquist's tree is very strong in is in 
biogeographical concordance with avian orders.  

<<Joel criticized the tree that Sibley and Ahlquist developed, but his 
own tree had some resemblance is too their conclusions (with, 
admittedly, a nod or two to traditional arrangements).  For example, 
Joel agrees that the ratites  form a sister group to all other living 
birds, and also supports the idea that ducks and their relatives, 
clustered together with chickens and pheasants and their relatives, form 
a clade called Galloanserae.>>  

There actually is alot of evidence for Galloanserae (Galloanserimorphae 
of Livezey).  Note that I now admit that, and I support the arrangement, 
for now...  I still am critical and acknowledge that anseriforms share 
many characters with ciconiiforms and kin as Ericson points out.  

<<On the other hand, Joel does not agree that the Piciformes are 
polyphyletic  (this is the group that traditionally unites woodpeckers 
with various tropical birds such as barbets, toucans, jacamars and 
puffbirds) and and restores them to their close association with the 
largest importer of birds, the pass era for maze or perch in birds.>>

I think that Cracraft is very, very wrong in saying that Piciformes is 
monophyletic.  Though Cracraft has tried to say that many of the 
characters Galbulae shares with Coracii are either false or have wide 
distributions, the combined weight of the characters, and the amount of 
characters Pici shares with the Passeriformes and not the Galbulae, 
seems to argue for polyphyly.  Basing monophyly of this diverse and 
disparate group on three (actually, if some have their say, two 
characters since zygodactly with a caudally directed sehnenhalter has 
evolved many times within neornithines) hindlimb characters makes me 
suspscious.  Studies of comparative osteology (Olson, 1983), feather 
structure (Broom, 1990), cranial feeding apparatus structure (Burton, 
1984; a paper that I need to copy soon), and of course DNA-DNA 
hybridization (Sibley and Ahlquist, 1990) all support polyphyly of the 
Piciformes.  Other theories, such as polyphyly of Galbulae, do not stand 
up to what we know.  As you probably can tell, I have some very definite 
thoughts on this issue and I feel you'll get to read more about them in 
the near future (hint, hint).

<<And he also own tree be Luis and ingredients, the two most highly 
modified living quarters of guiding birds, as to each other's closest 
relatives, a conclusion that, I would say, few ornithologists share.  
Although he seems to support Sibley's association of a large number of 
water birds with each other, he admits that the relationships within 
this assemblage are "confusing".  Joel emphasized that is conclusions 
were not final, and that he had difficulties in understanding of what 
was going on when the morphological and molecular data failed to produce 
the same result.  Possibly, he said, this was the result of improper 
coding of morphological characters, or insufficient sampling, or, 
perhaps, a lot of homoplasy at the molecular level.>>

Homoplasy is rampant within every little thing in avian systematics.  

<<Jeff Groth then describe his study of the phylogeny of living birds 
using nuclear DNA sequencing.  Although his study was only based on 16 
taxa, he claimed one advantage over Sibley's work -- the use of an 
outgroup, in this case a crocodilian.  The same outgroup was used in a 
study of mitochondrial DNA published by Mindell and others in 1997.  
Nuclear DNA, according to Groth, may be evolving much more slowly that 
mitochondrial DNA, at least at lower divergence levels (mitochondrial 
DNA apparently reaches a "plateau" of divergence, though I haven't the 
faintest idea why.)>>

Interesting, though like you, I have no idea why.  

<<I found two of Groth's conclusions particularly interesting.  One was 
that his tests do not support a molecular clock.  If this is correct I 
assume that the claims that the orders of living birds must have 
diverged much earlier than the fossil evidence would suggest may be 
incorrect, something that would certainly not bother me.>>

I'll have to wait this debate out.  

<<His most interesting finding, as far as I was concerned, was the 
synapomorphic deletion of a sequence in one of the genes he examined, 
shared by all birds except ratites and Galloanserae, a finding that 
supports the view that the ratites are a sister group of other living 
birds and, among those birds, the Galloanserae are a sister group to the 
remainder.>>

Very, very interesting.  As most of us well know, basal Neornithes is 
very, very problematic.  What is becoming a near concensus among avian 
systematicists is: 1) palaeognaths are the most basal living birds and 
2) somewhere within the tree, galliforms and possibly anseriforms are 
basal to a larger group of living birds.  Everything else is 
controversial.  

<<Bradley Livezey  then described his morphological studies on basal
Neognathae.  Livezey's work also supports the concept of the 
Galloanserae, though he admitted that it is hard to assess skull 
features in birds for phylogenetic purposes, because of the extensive 
fusion of the bones of the skull, which makes it very difficult to 
determine where elements end and others begin.>>

I like Livezey's work an awful lot.  Brad Livezey may well be the 
Michael Lee of ornithology.  

Emphasis on skull structure in avian systematics has changed a lot over 
the last 150 years.  From Huxley's superhuman classification of birds on 
palatal structure we have experienced many shifts of is a valid 
character in avian systematics.  Very interestingly to me at least, is 
that Garrod, Furbringer, and Beddard's Amologonatae (based on presense 
or absense of the ambiens muscle) seems to have some basis in fact.  
But, look at things like DNA and you see piciforms and passeriforms 
forming basal lineages of specialized neognaths.  150 years from now, we 
will look back at Huxley's classification, Garrod et al.'s 
classification, at Wetmore's classifications, at Verheyen's bizarre 
classification, Cracraft's classification, Olson's classification, 
Sibley and Ahlquist's classification, etc. and say much of the same 
things that we are saying now: "They had some ideas back then", "Ha!  We 
actually believed THAT!!!", etc, etc, etc.  

<<After these three papers, you might be forgiven for thinking that the
reality of the Galloanserae was a settled matter.  Therefore the title 
of Per Ericson's paper, "Higher level systematics in living birds: or is 
Galloanserae fact or fiction?"  might seem rather strange.  But 
Ericson's morphological data set, even when combined with the data from 
Livezey's study, does not support the Galloanserae as a monophyletic 
group. Molecular data does support the group, but the tree that results 
is unresolved.  Ericson believes that the problem of whether there 
really is such a clade is not settled.>>

Ericson's research is very well-done and composed.  It has not convinced 
me as yet, but I still find it interesting.  Livezey has never supported 
Galloanserae, Galloanserimorphae, Neoaves to exclusion of paleognaths, 
galliforms and anseriforms, etc.  He has published much on this in the 
past (in Journal of Avian Biology in 1996, critiquing Cracraft and 
Mindell's 1988 dataset supporting Galloanserae and in ZJLS) and it has 
been very peppered against Galloanserae.  

<<One thing that Livezey and Ericson to agree on is the position of 
Presbyornis, a long-legged duck-like fossil that Alan Feduccia and 
Storrs Olson have claimed as a link between ducks and shorebirds.  
However, this fossil appears to be a perfectly good sister group to the 
Anatidae, the family that includes all living ducks and geese (and the 
whistling ducks, sometimes given separate family status) except for the 
Australian Magpie Goose which is put in its own family, Anseranatidae.  
In both studies, if I recall correctly, the Magpie Goose forms a sister 
group to Presbyornis plus Anatidae.>>

That's right.  This is very wel supported, much more so than S&A's 
Anseranatidae+Anhimidae.  

<<Ericson then turned to the vexing question of what a primitive 
neornithine bird would be like.  Presbyornis, shorebirds, the Magpie 
Goose, screamers, and fossil taxa such as Juncitarsus and the 
graculavids, share a number of characters that are presumably 
plesiomorphic.  Only a single early representative of the Galliformes 
(chicken-like birds) is known, Gallinuloides, which is similar to the 
modern cracids.  The shoulder girdle and sternum of these birds is, 
however, unlike that in ducks and shorebirds, leaving open the question 
of whether the plesiomorphic characters in ducks are primitive within 
the Neognathae, or only within a subgroup including ducks and 
shorebirds.>>

...or that they are convergent.  Anseriforms are so much unlike their 
potential relatives it is not even funny.  We need more fresh analyses 
and fossil.  

Personally, I think that galliforms are specialized from the MRCA of 
neognaths and neornithines.  

<<Ericson finished with a few really surprising suggestions.  Noting 
that Larry Martin has suggested that neornithine birds involved in shore 
environments, in contrast to the enantiornithines, he suggested that 
Ichthyornis might be an example of a basal neornithine, teeth and all. 
More skeletal work on this bird is sorely needed.>>  

If this is the only evidence....

<<For me, even more astonishing was the suggestion, based on a study of 
mitochondrial DNA, that the Passeriformes (songbirds or perching birds) 
represent the most basal neornithine lineage.  The Passeriformes have 
almost always been regarded as not only a highly derived group, but the 
most recent bird order to evolve. In the northern hemisphere, they are 
pretty much unknown before the Miocene.  However, Walter Boles of the 
Australian Museum, who was at the conference, has describe Australian 
material that seems to represent a far older representative of the 
order.  In fact, Ericson put forward the surprising idea that the 
Passeriformes and the Galliformes together are a sister group of the 
tinamous, the only flying birds within the ratites. Needless to say, 
this startling conclusion will need some verification.>>

This is pretty new.  Supposedly when Dave Mindell found this from his 
study of his mitochondrial DNA (I believe) at the AOU meeting it caused 
somewhat of an uproar.  What about the piciforms (in/excluding 
Galbulae)?   

Matt Troutman 
m_troutman@hotmail.com

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com