“Ida” not a missing link

March 6, 2010 • 10:22 am

On May 20 of last year, at a remarkable press conference in New York, a group of researchers announced—with much ballyhoo—that they’d found a 47-million-year-old primate fossil named Darwinius masillae (nicknamed “Ida”).  Ida, the finest fossil primate in existence, was touted loudly as the missing link between the two major branches of primates, the Haplorhini (anthropoids [apes and monkeys] and tarsiers), and the Strepsirrhini (lemurs and lorises; see figure below). Concurrent with the press conference was a History Channel documentary and a book about Ida, Colin Tudge’s The Link, that proclaimed, with much heavy breathing, that Ida was, as one of the earliest primate ancestors of our own species, an earthshaking discovery (see my review of the book here).

The description of Ida was eventually published in a paper by Franzen et al. in the journal PLoS ONE, a journal that doesn’t exercise stringent scientific review of submitted papers.  The reaction of both bloggers and scientists was very critical, with many pointing out that Ida didn’t look like a missing link at all, but may have been only an adapiform primate, a representative of a lineage within the Strepsirrhini that went extinct without leaving descendants.  The phylogenetic analysis of the PLoS paper, they claimed, was cursory and incomplete.  A later paper in Nature by Seiffert et al. cast further doubt on Ida’s status, suggesting that the features that led Ida’s discoverers to lump her with the haplorhines were convergent—that they had evolved independently in haplorhines and adapiforms and thus could not serve as evidence for Ida’s anthropoid ancestry.

Well, a paper just out in the Journal of Human Evolution, by Blythe Williams et al. (including my Chicago colleague Callum Ross), appears to drive the final nail in Ida’s coffin—at least regarding her status as a missing link between the major branches of primates.  Summarizing all the evidence in a cladistic framework (combining, as Williams et al. say, “hundreds of dental, skull, limb-bone, embryological, physiological and molecular characters of living and fossil primates”), they come up with two conclusions:

1. First, the Eocene adapids (of which Ida was one) were not, as Franzen et al. implied, more closely related to anthropoid primates than to lemurs and lorises.  Rather, the adapids were stem strepsirrhines that lived well after that group diverged from the haplorhines  (our own ancestors).

2.  Second, detailed study of Ida’s mandibles, teeth, orbit, leg bones and other traits show that she possesses no clear synapomorphies (shared derived traits) with haplorhines, but rather is clearly an adapiform.  All the evidence, then (more will soon be published in Proceedings of the National Academy of Sciences) puts Ida far from the haplorhines and anthropoids, and firmly in the now-extinct adapids.  Franzen’s failure to recognize this apparently rests partly on their unaccountable failure to use any fossil haplorhines in their original phylogenetic analysis.

Figure 1 from Williams et al. shows where Darwinius sits in the primate phylogeny (we’re with the crown Anthropoidea):

Science Daily gives a precis of the paper here.

Well, in science you win some and lose some, and all of us are familiar with being wrong. But it’s something else to be wrong when you’ve rushed prematurely into print, proclaiming earth-shattering conclusions from a slipshod analysis. A win for the bloggers, who called Franzen et al. out within days of their announcement.


Williams, B. A., R. F. Kay, E. C. Kirk, and C. F. Ross. 2010.  Darwinius masillae is astrepsirrhine—a reply to Franzen et al. (2009).  J. Human Evolution, online.

Franzen, J.L., Gingerich, P.D., Habersetzer, J., Hurum, J.H., von Koenigswald, W., Smith, B.H., 2009. Complete primate skeleton from the middle Eocene of Messelin Germany: morphology and paleobiology. PLoS ONE 4, e5723.

Erik R. Seiffert, Jonathan M. G. Perry, Elwyn L. Simons & Doug M. Boyer. 2009.  Convergent evolution of anthropoid in Eocene adapiform primates.  Nature 461:118-1122.

28 thoughts on ““Ida” not a missing link

  1. Much Ado About Nothing.

    From Wikipedia:

    Another motif occurring throughout the work is the play on the words nothing and noting, which, in Shakespeare’s day, were homophones. Taken literally, the title implies that a great fuss (“much ado”) is made of something which is insignificant (“nothing”), such as the unfounded claims of Hero’s infidelity. However, the title could also be understood as “Much Ado about Noting.” Indeed, much of the action of the play revolves around interest in and critique of others,…

  2. Bleech! Even my parents had heard of this ‘earliest human ancestor’ touted around.

    Unfortunately they don’t read blogs. Now I have a nice refutation to mail them, thanks.

  3. “PLoS ONE, a journal that doesn’t exercise stringent scientific review of submitted papers.

    That statement is highly arguable. The scientific review at PLoS One is in fact rigorous, but the criteria for acceptance don’t include “significance,” which one might argue is the *main* criterion at a journal like Science or Nature.

    Indeed, many believe, and I’m sometimes inclined to share the view, that Science and Nature often place a substantially higher value on “significance” versus scientific rigor.

    So backhanded slaps at PLoS one are really not called for. It’s not a prestige journal but that doesn’t make it bad. (FWIW, I’ve not published in PLoS One but I have seen plenty of decent work there.)

    1. Not so sure I agree. I’ve read a half-dozen PLoS One papers in the last few months, and haven’t been impressed with any of them (granted, that could somehow be a biased sample). And the problems weren’t “lack of general interest,” they were stuff like “hypothesis not properly tested” and “conclusions don’t follow from data.”

      The Darwinius paper certainly did not get rigorously reviewed for science! It never would have been published in a journal like Nature or Science, even given its potential significance. The analysis was shoddy and cursory, and proper review simply wasn’t given. I do fault the editors because they were, I think, rushing this paper into press.

      1. Nature and Science, the GlamourMagz? Some people still are eager to publish there? In this day and age? Relying on their peer-review? Having their papers behind paywall for nobody to see?

  4. I agree with Spiny. I have published with Plos ONE and my paper went through a thorough peer-review by 2 referees and an assistant editor. The Ida paper doesn’t seem to have been properly reviewed but that happens occasionally in all journals. The reviewers and the editor were at fault, not the journal as a whole.

  5. I’m quite cross with the researchers who created the press-fest originally. Not only were their claims overblown and scientifically dubious, their irresponsible media-mongering has handed creationists juicy headlines such as “Claimed Missing Link Refuted by Scientists.”

    Thanks guys. Thanks.

    1. One thing that really annoyed me was the “missing link” bullshit – who the hell believes in a “missing link”?

  6. I like how they call what are morphologically perfect “crosses” (to indicate extinct lineages) what they are quite clearly not (morphologically): “daggers”.

      1. Hey, way to spoil a humorous barb by bring in a minor semi-relevant factoid! (Its use as an indicator of an extinct lineage almost certainly was borrowed from its use as an indicator that a person was deceased. So it’s a “cross” even if you call it a “dagger”. Why do you hate christianity and america? 🙂

        1. First: The hell I’m a user of scandinavian seriousness, last time I heard from now and when dance partner “but I thought you were Greek?!”. I think that is categorizing too much.

          Literalness is my personal reaction to mysteries, it helps me find the sense (or nonsense, as may be) behind. I’m way too eager a pattern searcher to let my imagination run free.

          Sorry for ruining the joke, guys!

          Second: Let me correct a stupidity of mine. It is quite clear from the article I referenced that “crosses” are legitimate use.

          Third: Why do I hate christianity and america? Because … No, there is that literalness again.

          Let me instead ask you, why do you hate atheism and scandinavia? Typological animosity makes me cross.

  7. “Franzen’s failure to recognize this apparently rests partly on their unaccountable failure to use any fossil haplorhines in their original phylogenetic analysis.”

    I wouldn’t say unaccountable; I can come up with two reasonable scenarios:
    1. incompetence
    2. the creationist’s habit that if facts don’t fit your dogma, just ignore them – this is far worse than mere incompetence, it is incompetence + outright lying

    1. Sorry, I don’t understand. Are you implying Franzen is a creationist with an agenda? And how would misplacing a Strepshirrhini fit a creationist agenda?

  8. This is all BS. Just a theory, yet unproven. why do we let nut cases like this influence us??

  9. Perhaps. But you made a really bad factual error here – PLoS ONE has, in some ways, better and stronger peer review than most other journals. Study the issue a little bit more….

    Also, compare the paper to media reports – this is an essential exercise in any criticism of Ida.

    Third, wonder why the other folks are criticizing? Perhaps they are right, perhaps they are wrong, but they also have a fossil of their own, and a competing hypothesis of their own. The Ida crew has only published the description so far, with several other manuscripts still to come. Perhaps they are right, perhaps they are wrong, time will tell. It is palaontology, after all – the difficult study of old bone fragments. Every analysis is just as good as the presuppositions going into it. GIGO. The only way to have this resolved is if we find MANY fossils in this group from around the same age. In the meantime, we can hypothesize endlessly – which both of these groups are doing.

  10. I edited the paper for PLoS ONE. I keep a policy of not commenting publicly on papers that I edit. But in light of Jerry’s comments, I would like to add a few points.

    (1) Most of the critical comments about the paper have concerned the issue of phylogenetic relations, in particular the question of whether adapids may have been a sister to anthropoids. This is a key claim in the documentary about Ida, but the research paper does not make or defend that claim. The paper has 25 journal pages of description of the fossil and aspects of its biology. I think that a full systematic analysis should take just as much if not more effort and space, and would be happy to see it as a separate piece of work.

    (2) The review process for this paper was equally or more stringent to that at other discipline-specific journals. The reviewers were senior scientists specializing in early primates.

    (3) Some of the scientists who have been publicly critical of the paper were invited to review it for PLoS ONE and declined. I do not begrudge them, as people decline to review for time and all other kinds of reasons. But it is worth noting that my editorial decisions deliberately included those likely to be critical about the fossil’s importance.

    In contrast to Jerry, I do not think that the current letter resolves the issue. I agree that they have valid criticisms and stake out a clear position concerning the position of adapids relative to strepsirrhines and anthropoids. I certainly agree that living tarsiers are closer relatives of anthropoids than are strepsirrhines; the genetic evidence is quite clear. But considering the evidence for substantial convergence of Eocene primates, I remain cautious about interpreting their similarities. I think the authors of the current letter (and Seiffert et al., last year) are probably right — but I hope that the relationships of these Eocene fossils will continue to be vigorously tested.

  11. I hope you see this comment, as I am late in coming into this discussion. I have looked at your chart showing where you think Ida fits into the evolutionary chart, and it does not match the chart on the website of the museum where Ida is housed. They have her on the haplorhine side as an adapoidea. There are important differences in the two charts. Please check out the link, and either make a correction, or tell me why their chart is wrong.


Leave a Reply