[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: giant birds

The debate over avian origins is merely the delayed reverberations in 
the palaeontological literature of a much wider debate. For those who 
are interested (and I suspect that will be few!), you can check out the 
letters to the editor pages in Nature between about October 1978 and 
July 1981. This debate is pretty much a non-issue in evolutionary 
theory now, so it's perhaps surprising that shots are still being fired 
over dinosaur/bird relationships.

I'm afraid I can't agree with George that "BCF and cladistic analysis 
are not opposed; they work together. I use available dinosaur 
cladograms to trace the evolution of avian features." I can't put it 
any better than Luis Chiappe, who said (Nature 378: 349-355 1995): 
"assumptions about evolutionary processes or adaptational scenarios, 
such as these, are misleading when identifying historical 
relationships. Phylogenetic reconstruction should be based exclusively 
on the hierarchical distribution of homologies among taxa, and a given 
phylogenetic hypothesis can only be rejected by providing an 
alternative hypothesis for which supporting evidence outweighs that of 
the original hypothesis."

Phylogenetic systematics gives us a set of rules within which we can 
test evolutionary hypotheses. Step outside these rules and we end up 
with the sort of vitriol we've all encountered on the list today. It 
would be a very simple matter to test BCF against other phylogenetic 
hypotheses on the origins of bird flight. Take a theropod character 
matrix of the sort used to produce Fig. 4 in Sereno's recent paper 
(Science 284: 2137-2147), and run a simple maximum parsimony analysis 
in PAUP. Open the tree (or trees in the event that there are multiple 
shortest trees) in MacClade, and use branch swapping to produce the 
topology corresponding to BCF. Compare the new tree length with the 
original. There are more sophisticated statistical techniques for tree 
comparison, but this would do for a start. I'd do it myself if I had a 
character matrix! I will be very surprised if the BCF topology is 
equally parsimonious, as George suggests.

I do not believe the debate is informed by statements such as:

> What I do care about is that people recognize that neither pole in 
> the present "ground-up" versus "trees-down" debate about avian flight 
> is entirely correct, and that BCF is closer to the truth than either 
> of those positions.

The real point is that there are statistical techniques to evaluate the 
merits of different tree topologies, and if BCF is a better explanation 
why aren't we reading about it in the literature?

Kendall Clements