Darwinius Strikes Back

If all that you knew about paleontology came from headlines alone, you could be excused for thinking that the science consists of little more than naming one obscure creature after another. There are a few petrified celebrities which can draw sustained attention – Tyrannosaurus and Triceratops foremost among them – but, in general, it seems that the novelty of newly-discovered fossil creatures wears off very quickly. Shortly after their announcement these animals go underground into the academic literature; the ongoing research, discussion, and debate concerning these fossils at conferences and in journals mostly remains hidden from public view.

So it is with Darwinius, the ballyhooed fossil-primate which debuted in May of 2009. Although initially hailed as “The Link” between early primates and human ancestors, academic publications which followed its announcement confirmed that the 47 million-year-old primate was much more closely related to living lemurs, lorises, and galagos than to us. (There were a few popular-audience reports about this reversal last fall, but nowhere near the amount of attention paid to the initial announcement.) All the hype about Darwinius being “The Link” was hogwash, but now the team of scientists who originally described the famous fossil have at long last responded to some of their critics.

Printed in the Journal of Human Evolution, the new letter by Philip Gingerich, Jens Franzen, Jörg Habersetzer, Jørn H. Hurum, and B. Holly Smith is a reply to a critique of Darwinius published in the same journal earlier this year. Whereas the previous paper – along with the description of the primate Afradapis in the journal Nature – barred Darwinius from a close relationship to monkeys, apes, or us on anatomical grounds, Gingerich and colleagues use their new paper to reaffirm their previous conclusion. Stressing the completeness of Darwinius, they propose that the primate from Messel supports a vision of primate evolution which fell out of favor decades ago, but do their assertions stand up to scrutiny?

Mass media overhype aside, the academic debate surrounding Darwinius is focused on whether it was a haplorrhine (represented today by tarsiers + anthropoids [monkeys and apes]) or a strepsirrhine (represented among living primates by lemurs, lorises, and galagos). If Darwinius was a haplorrhine, as Gingerich and Hurum have insisted, then it might have been close to the ancestry of the earliest anthropoid primates, but if it was a strepsirrhine (as most early primate specialists have contended) then it would have belonged to an extinct lineage on the other branch of primate evolution. Given the completeness of the specimen, this debate seems almost silly – shouldn’t the fossil’s anatomy speak clearly about what group it belonged to? – but the attribution of Darwinius to one group or another depends upon the method by which comparisons between it and other primates are made.

In the original PLoS One description of Darwinius, Hurum and his co-authors exclusively used living primates as their guides. Most of the characteristics they chose to compare between Darwinius and other primates were irrelevant as they were based upon soft tissue structures that could not be observed in he fossil, but, of those which could be seen, it appeared that the fossil primate had more characteristics in common with living monkeys than with lemurs and lorises. This was a baffling move. Early 20th century paleontologists had mistakenly placed the lemurlike fossil primate Notharctus – a close cousin of Darwinius – near the origins of New World monkeys on the basis of such methods, and given the number of fossil primates discovered in the past thirty years alone it was strange that Darwinius was not explicitly compared to its prehistoric kin.

The detailed comparison of Darwinius to other primates – both living and fossil – was taken up by Erik Seiffert, Elwyn Simons, and colleagues in their description of a closely related genus from Egypt called Afradapis. Rather than focusing on a handful of characteristics mostly seen in living animals, Seiffert and his co-authors cataloged 360 anatomical characteristics (not all of which could be recorded in very fragmentary species) across 117 primates and ran this information through a computer program to produce the most probable evolutionary tree on the basis of shared characteristics. What they found matched the current scientific consensus on anthropoid origins; namely that tarsiers were most closely related to anthropoids and that the lemurlike adapiforms (including Darwinius and Afradapis) were members of extinct lineages most closely related to lemurs, lorises, and galagos. More than that, the few anthropoid-like characteristics Darwinius had were instances of evolutionary convergence – traits which evolved independently in two different lineages – that could only be seen when it was placed in the context of the known primate fossil record.

Rather than present a similar, comprehensive analysis, however, the new paper by Gingerich et al. casts aspersions about the usefulness of using other fossil primates in determining the affinities of Darwinius. Since the haplorrhine and strepsirrhine groups were originally established through traits seen among living primates (particularly soft tissue anatomy), they argue, then analyzing Darwinius according to these criteria is the best way to determine which major group it belonged to. Hence, rather than go back to the primate fossil record to collect data for comparison, the authors of the Darwinius letter took the same data from their original description and ran it through a computer program to create a minutely more detailed evolutionary tree which placed Darwinius as a haplorrhine close to the origin of anthropoids.

As Gingerich, Hurum, and their co-authors admit, the difference between their results and the results of their peers comes down to the choices made for comparison. Whereas the authors of the Afradapis paper used a large number of characteristics taken from a wide swath of extinct and extant primate diversity, the scientists who described Darwinius created a matrix of characteristics drawn from the academic literature on living primates, many of which are uninformative when dealing with fossil species (essentially repackaging what was presented in the first paper). Given these differences, it is no wonder that the results came out so differently, but I do not accept the implication of the new Darwinius paper that the primate fossil record can be ignored when trying to tease out relationships of extinct species.

While Gingerich, Hurum, and their co-authors are correct that it can be tricky to classify fossil primates on the basis of scrappy fossils, it is also understood among early primate specialists that instances of evolutionary convergence are also a potential source of confusion. This prominently underscores the problem with exclusively using  living species to classify extinct animals. What may appear to be a shared characteristic may have evolved twice in two different lineages at different times, and the analysis of primate evolution by Seiffert, Simons, and their colleagues confirms that this was the case with the anthropoid-like traits of Darwinius (especially since the earliest known anthropoids lacked some of the “anthropoid” characteristics seen in Darwinius). In fact, many of the seven characteristics the promoters of Darwinius claim to be shared with haplorrhine primates (such as a short face, flat nails, fused lower jaw bones) are not unambiguous indicators of affinity, and the new letter does not refute that many of these characteristics may have evolved due to convergence.

As a whole, the Journal of Human Evolution response letter provides almost no new information about Darwinius or its relationships. Instead of thoroughly placing Darwinius within the context of other early fossil primates, the authors of the letter preferred to use only anatomical characteristics used to distinguish among living primates, which itself falls short of the total evidence approach from fossils, soft tissue anatomy, and genetics preferred by other researchers. Sadly, rather than fully engaging with the current science of early primate studies, the reply appears to mark a hardening of already-favored conclusions with little room for progress.

Even though Darwinius now seems to be locked into an intractable argument, debate and discussion will no doubt continue. That’s the way science works. Even though modern mass media acts as an announcement firehose – almost constantly spewing out reports of new species – the publication of a fossil marks only the beginning of the active process by which paleontologists come to understand prehistoric life. With any luck, there will be a push for more “upstream” coverage which will let the public into the labs, offices, conference rooms, and field sites where the process of science is actively occurring, but for now the intricacies of discovery and debate remain obscured by our obsession with novel results.

(For another perspective on this story, see this post at Zinjanthropus.)

References:

Gingerich, P., Franzen, J., Habersetzer, J., Hurum, J., & Smith, B. (2010). Darwinius masillae is a Haplorhine — Reply to Williams et al. (2010) Journal of Human Evolution DOI: 10.1016/j.jhevol.2010.07.013

Franzen JL, Gingerich PD, Habersetzer J, Hurum JH, von Koenigswald W, & Smith BH (2009). Complete primate skeleton from the Middle Eocene of Messel in Germany: morphology and paleobiology. PloS one, 4 (5) PMID: 19492084

Seiffert, E., Perry, J., Simons, E., & Boyer, D. (2009). Convergent evolution of anthropoid-like adaptations in Eocene adapiform primates Nature, 461 (7267), 1118-1121 DOI: 10.1038/nature08429

Williams BA, Kay RF, Christopher Kirk E, & Ross CF (2010). Darwinius masillae is a strepsirrhine-a reply to Franzen et al. (2009). Journal of human evolution PMID: 20188396

Williams, B., Kay, R., & Kirk, E. (2010). New perspectives on anthropoid origins Proceedings of the National Academy of Sciences, 107 (11), 4797-4804 DOI: 10.1073/pnas.0908320107

For more on the ongoing Darwinius affair, see my previous posts, articles, and academic review:

A Discovery That Will Change Everything (!!!) … Or Not

Poor, poor Ida, Or: “Overselling an Adapid”

Getting to know “Ida”

Carnival: Uncovering “Ida”

A few thoughts on “Uncovering Our Earliest Ancestor”

Ida: The Legend Continues

Afradapis and Ida, sittin’ in a tree…Afradapis and Ida, sittin’ in a tree…

New Study Confirms That “Ida” is Not Our Great-Great-Great-Great-Etc. Grandmother

Evolutionary History of Early Primates Places Human Origins in Context

Tracking Notharctus, Wyoming’s Prehistoric “Lemur”Tracking Notharctus, Wyoming’s Prehistoric “Lemur”Tracking Notharctus, Wyoming’s Prehistoric “Lemur”

The Dangerous Link Between Science and Hype (London Times)

Hype Surrounding Ida Has Misled Scores of People (London Times)

Ancestor or Adapiform?: Darwinius and the Search for Our Early Primate Ancestors Ancestor or Adapiform?: Darwinius and the Search for Our Early Primate Ancestors Ancestor or Adapiform?: Darwinius and the Search for Our Early Primate Ancestors (Evolution: Education and Outreach)