Urvogel Reverie

Resolving the phylogenetic relationships and cladogenesis of Neoaves (sensu Sibley & Ahlquist 1990), is one of the single most intractable difficulties in systematic ornithology. Though not beyond dispute (e.g., Ericson et al. 2001), there appears to be a fundamental dichotomy at the base of Neoaves between a clade allying water and land fowl (Galloanserae) and all other neognathous birds (Neoaves sensu Sibley & Monroe 1988). Use of an astonishing variety of data to infer phylogenies, both morphological and molecular, consistently has failed to resolve the phylogenetic relationships of Neoaves to the exclusion of Galloanserae. Chubb (2003) presented an analysis of neoavian phylogeny using ZENK, a single copy nuclear gene, and reviewed the difficulties in determining the branching pattern of the neoavian clades. Chubb concluded that despite possible artifactual reasons for such difficulties, the unresolved branching pattern in the ZENK tree topologies she recovered, represent biological reality. Given that similar confusion has resulted from the use of nearly all other possible data to infer neoavian phylogeny (e.g., Johansson et al. 2001, Mayr et al. 2003), the conclusion reached in Chubb's analysis (that there may been an extremely rapid adaptive radiation of Neoaves to the exclusion of Galloanserae) must be seriously considered. Assuming for a moment that there are valid fossil records for Galloanserae and Paleognathae in the Cretaceous, is it possible that the explosive radiation identified by Feduccia (1995, 1996, 1999, 2003) is in fact representative of a post-KT Boundary neoavian adaptive radiation, as opposed to a neornithine adapative radiation? If the alleged paleognath/galloanserine Cretaceous record is valid, this hypothesis seems best able to accommodate the undeniable fact that there is a rapid proliferation of neoavian clades immediately after the KT Boundary. Under this hypothesis, the "transitional shorebirds" of Olson & Parris (1987) would represent a neoavian bottleneck, as opposed to a neornithine bottleneck, and consequently Charadriiformes would fall out as basal to Neoaves to the exclusion of Galloanserae. This would mesh nicely with Hope's (2002) referral of most of these specimens to Charadriiformes. Though not yet rigorously tested, and resting on at least one major assumption (the validity of Cretaceous paleognaths and galloanserines), it does strike one as an interesting model worthy of analysis.

Urvogel Reverie

God Fearing Atheist

I have no problem with this (expect maybe the position of Charadriiformes), but there is definately a need to better sample the Cretaceous, particularly Gondwana, before anything definitive is said.

More later.

Nic Tamzek

Frighteningly, I think I understood that.

I do think that sometimes too much emphasis is put on discovering One True Tree of Bifurcations (binary splits) that all data fit into. There is no reason why there cannot be cases where one lineage (here, Neoaves -- extant birds minus ratites, I gather) diversifies into dozens of ecologically diverse forms in the space of one or a few million years. Indeed, this relatively dramatic and rapid divergence this appears to be what happens in many adaptive radiations. If these diverse ecological forms (having filled up the available niches) then persist with more moderate changes for the following tens of millions of years, then looking back it would be very difficult to figure out the exact order of branching because, in fact, the branching did not take very long.

Perhaps the obsession with bifurcating trees is a vestige of typology? A tree is, after all, a qualitative result rather than a quantitative one. In the situation described above, the true tree, if depicted on a constant-time scale, might in fact indicate that the bifurcations that led to the modern orders all took place very quickly, and that in any coarse resolution "picture" of the tree (e.g., any tree based on presently available data), those rapid bifurcations will appear as an unresolved polytomy. If so, the basic fact "all these lineages split from each other in a geologically short period of time and are quantitatively basically all equally related to each other" would probably be a more important fact to know than the exact order of branching.

Which might be what you were saying, but I felt like saying it also...

All IMHO of course...

Urvogel Reverie

The Gondwana Cretaceous localities are already rather rich in the avifauna yielded (e.g., Lecho Formation).

Urvogel Reverie

SEF

Is part of the problem that there hasn't been sufficient gene mapping yet? Wouldn't some of the relatedness issues be likely to be resolved if full sequencing maps did exist rather than just a few genetic spot checks?

Some months ago I started a thread on the BBC site (will be long deleted by now as is there wont) about which sequencing people would actually like to see to resolve particular evolutionary issues as opposed to the ones undertaken so far - which are biased by humans towards human things (eg for medicine), standard experimental animals, life-forms they think might be particularly easy (and interesting) and quirky local/cultural choices.

Urvogel Reverie

It is possible. I think the key point is that both morphological and molecular evidence are unable to resolve the higher phylogenetic relationships of Neoaves to any great degree, and most orders fall out in a polytomy. This to me suggests there is a real phenomenon indicative of biological reality at work, as opposed to artifactual results of improper molecular and morphological analysis.

Urvogel Reverie

SEF

Well being able to fly is a pretty major thing. It makes getting to new territories where one group might become isolated very easy. It also opens up new niches for living (food sources which weren't previously attainable). That's a massive opportunity for diversification and selection. It wouldn't exactly be surprising if the rate was faster than for other things around at the same time - except those in competition with the new birdies.

However, there should still be a branching lineage which might be discernable in the genetic modification revision history. Or are you proposing that some groups swapped genetic material by hybridisation (or infection?) a considerable amount of time (in terms of separately acquired mutations) after such branching had occurred?

A goes to B C D then to B' C' D' then to B" C" D" etc 2 of which then somehow hybridise to make E such that it isn't clear which of B or C lineages E came from. Or B and C accidentally make the same mutations such that EB looks very much like EC and it is unclear whether both came from an E which was directly from A instead.

You know, when I started typing that textual sequence I thought it might actually make the explanation simpler.

Urvogel Reverie

Yes, there should be a branching pattern. However, if neoavian clades branched off from their phyletic nodes in an extremely short period of time (say less than 10 million years), then retrieving that patttern is monumentally complicated, if even possible.

Urvogel Reverie