Tuesday, May 19, 2009

The Link

At least the racing (Formula 1 at Monaco, Indy 500, NASCAR Coca-Cola 600) should be over with on Sunday, leaving Monday free for this prime-time program.

The Link will be on the History Channel at 9 pm EDT on Monday, 25 May. This program has been carefully hyped, with a cast of the fossil put on display today in New York City amid a massive press promotion.

An excellent news article from the BBC includes a link to the "PLoS ONE" journal article (published today, 19 May 2009) about the fossil - which was discovered 26 years ago and sold two years ago to the University of Oslo, reported for a million dollars.

This is a remarkable fossil. The high resolution version of the first figure in the article, showing the fossil, is really remarkable. You can see the fur! But the program pushing this as the "missing link" is quite a stretch. The timing, in particular, makes the whole process look suspect. After all, when the TV program airs less than a week after the journal article appears (just a few months after the article was submitted), it does look like marketing the TV program and the exhibit in New York drove the publication rather than the other way around. Was this how the Oslo museum planned to pay for its purchase of the fossil?

That said, here are my dos centavos:

In my totally unprofessional training as "lifetime reader of the National Geographic", this seems a bit over-hyped as a direct human ancestor. The article itself doesn't really make that claim (which seems to come from the TV program promotion), probably because the most one can claim is that it is a extinct primate that could have evolved into either a lemur (less related to us) or a monkey (more related to us). But how do they know - for sure - that another known line didn't evolve into what became primates and humans, or that there isn't yet another, as yet unknown, extinct primate that actually did that? And they clearly aren't claiming that this is the parent of both sub-types of primates.

Actually, I've been reading the National Geographic since before I was born. My grandfather had a seemingly complete collection going back to the 30s, which I estimate based on having read about the 1931 balloon ascent of Piccard in his library. I've read all sorts of articles about discoveries about human ancestry. Each new discovery one would read about in that magazine, often by the Leakey family, added new information - as did later (also hugely hyped) discoveries like Lucy. There is nothing like actual new data to stir the pot and update the lineage. And nothing like National Geographic sized publicity to try to sway people to your side of the debate.

I haven't had a chance to read all of the ScienceBloggers on this subject, but the argument by Brian Switek seems to run along those lines. The argument seems to be between a small group (including the authors of this new paper) that think Apapids led eventually to humans, while many more think the vast bulk of the evidence favors Omomyids and Tarsiers. (The many links in the "classification" part of this article, particularly about the Haplorrhini, make following the evolutionary discussion much easier. Now I know that lemurs are on the wrong, wet nosed Strepsirrhini side of the evolutionary divide leading to humans. The authors would not like my visual comparison of their fossil to a lemur.)

But don't take my word for it. Go read the journal article, Switek's critique, and whatever else shows up in the next few days.

One other thing:
The weakness of conclusions drawn from partial fossils is pretty clear from the paper. The first half of the fossil, purchased by a Wyoming museum in 1991, was misclassified by the authors of this article. As a physicist, it always bothers me when conclusions are drawn from incomplete data or, more precisely, an incomplete datum. They didn't even have a full data point until they looked at the other half of the fossil. I'm glad I'm a physicist, where we can repeat experiments where the results seem incomplete.

Updated: Link links -


No comments: