A couple of months ago, I picked up on the news that Archaeopteryx had been rebumped in its* phylogeny to now sit among the dromeosaurs.

I also mentioned near the end...

I will say this, though: don't expect this to be the last you've heard of this. I think it is probably quite likely that there may be another reclassification, and Archeopteryx may find itself being reclassified as a bird again.

Now, I will politely clear my throat and point you to this article, published in Biology Letters. It took three months for someone to rebutt Xu's new phylogeny, but here we are! They have elevated Archaeopteryx out of the ashes, to reinstate it once again proudly among the birds.

This could end up being a bit of a soap opera, and who knows who will have the last word. Personally I'm not particularly interested where exactly we choose to define which is the first bird. I'm happy with it being a fantastic transitional fossil, and a member of a rich continuum of dino-birds. I will say this though: I trust the Maximum Likelyhood (ML) technique that Lee and Worthy have employed more than Maximum Parsimony, which is rather unsophisticated compared to ML, and isn't fraught with the problems like long branch attraction. I think we need more fossils and more characters before we can be more sure about where it definitely fits, though.

I'm going to follow up this post with a detailed explanation of how we actually calculate the phylogeny of fossils. It's a very difficult and subjective process which is the cause of many, if not most, arguments in paleontology. As a brief summary, though, Lee and Worthy have criticized the unilaterally parsimony based cladogram of Xu et al. Parsimony is the most simple and commonly used method for producing trees, however it can break down when there are too many homoplasic characters and for rapidly evolving groups. They've chucked some of the characters and instead used maximum likelyhood (ML) and Bayesian methods. Here is the new tree:




A close up, with branch length shown:



The last you'll hear on this? Not a chance. Bootstrap support is only 73 per cent, and
even with a posterior probability of 1, there is still the problem that Bayesian inference can greatly overestimate support if the models implemented are inadequate. I'm sure we'll get a reply at some point. For now, Archaeopteryx gets to stay on its wobbly perch, right at the base of the birds.

---------
*Please note, correct apostrophe first time!

Citation





Lee, M., & Worthy, T. (2011). Likelihood reinstates Archaeopteryx as a primitive bird Biology Letters DOI: 10.1098/rsbl.2011.0884