Why Darwinius is not our ancestor

I have been blogging and tweeting the dead horse primate that is Ida Darwinius masillae for several days now, culminating in last night's twitter-fest during the BBC documentary Revealing Our Earliest Ancestor: The Link. During our lively little twitter event, which made the Times Online today, I was, apparently, pretty worked up: Andrew Maynard even wrote 'Warming my toes from the heat of your ire...'.

Of course, it's not only us twitterers bemoaning the Darwinius hype machine; there's going to be a special Darwinius blog carnival on Monday, and I'm pleased to see a growing number of mainstream outlets casting a critical eye over the story, or at least giving space to external critics. For example, today I found an excellent opinion piece by Chris Beard the NewScientist called Why Ida is not the missing link. In it, Beard dresses down not just the hype - "unbridled hoopla" as he called it - but also the science itself. His final paragraphs are particularly good:

'So, Ida is not a "missing link" – at least not between anthropoids and more primitive primates. Further study may reveal her to be a missing link between other species of Eocene adapiforms, but this hardly solidifies her status as the "eighth wonder of the world".

Instead, Ida is a remarkably complete specimen that promises to teach us a great deal about the biology of some of the earliest and least human-like of all known primates, the Eocene adapiforms. For this, we can all celebrate her discovery as a real advance for science.'

Great stuff. Great enough to make me shout "Yes!" out loud while alone in my flat. What Beard is saying here is that Darwinius is not the 'missing link' between anthropoid side of the primate family tree (including humans) and the lemur side because the authors of the paper in which Darwinius is described have not convincingly demonstrated that she belongs with the former and not the latter. Beard contends that the balance of evidence keeps Darwinius - and all the rest of the adapiforms by extention - anchored firmly on the lemur line. To explain this, he provides this useful diagram of primate evolutionary history:

Evolutionary tree diagram reproduced from NewScientist.

The diagram shows the two competing hypotheses: the red dot in the diagram indicates where Chris Beard contends Darwinius belongs, on an early branch in the lemur (brown) lineage. The paler spot with the "?" indicates where the paper's authors claim Darwnius belongs, as an early member of the (blue) anthropoid lineage and, specifically, as they have contended in interviews and on their website if not in their paper, ON the line. In other words, they claim she is our direct ancestor.

I don't have any professional background in primate anatomy but I find Beard's argument - especially in combination with Laelaps' analysis - pretty compelling, certainly compelling enough to remain very skeptical of the authors' conclusions. We will have to wait for further analyses of the Darwinius specimen, however, before this controversy can be more soundly resolved.

But here's the thing: even if upon further analysis Beard is shown to be mistaken and the authors are right about Darwinius and her fellow adapids being on the anthropoid lineage, she will still not necessarily be the 'missing link', nor - perhaps more importantly because it actually means something - 'our ancestor'. To explain my point, I re-drew Beard's diagram:

My re-drawing of Chris Beard's diagram in which I shift his representation of the paper's conclusion about Darwinius' systematic position from ON the anthropoid line to a BRANCH off the anthropoid line.

I re-drew it this way for the simple reason that, considering the abundance of species upon the Earth at any one time, it would be very unusual for us to find a fossil on our direct ancestral line, rather than on a branch off of that line. As John Wilkins put it, "There is no missing link. Rather, there are an indefinite number of missing branches. [snip] We might have a species that is an ancestor of some other species, and yet not know enough to say that they are indeed the ancestor in question."

Additionally, as Richard Carter notes, Ida herself can't be our direct ancestor because she died as a juvenile. But as I've just explained, even her whole species is very unlikely to include our ancestor.

Interestingly, as I mentioned in passing above, the hype machine (including the paper's authors themselves when interviewed) puts Darwinius directly on the line to us, while in Supplementary Figure 7 in the PLoS ONE paper, the authors put Darwinius on a early branch off the line. In other words, what they're saying in public isn't just hyped up, it's fundamentally different from what they're saying in the paper itself.

To summarize, if Darwinius is found, as the authors contend, to belong on 'our' side of the primate family tree (and even that conclusion is shaky) she is not a 'missing link' (because there's no such thing), and she is very unlikley to be our ancestor.