Did knuckle-walking evolve twice?

ResearchBlogging.orgKnuckle-walking is a pretty special mode of locomotion. Amongst primates, only the African apes do it habitually, and anteaters are the only other mammal who does it. It would seem, then, that the most parsimonious explanation for such a specialized form of locomotion would be that the African apes all share a common ancestor who was also a knuckle-walker. An addendum to this explanation would be that humans, since they fall within that nested African-ape clade, also share an ancestor who was a knuckle-walker. The thing about parsimony, though, is that when a “parsimonious” explanation is met with conflicting evidence, it is no longer parsimonious!

The idea that knuckle-walking had independent origins in Gorilla and Pan is not a new one.  Dainton and Macho visited the subject in 1999, and Richmond and Strait have been arguing against that interpretation ever since.  Everyone can agree that gorillas put more weight on the ulnar side of their hands and wrists than chimps do.  The controversy is over whether they do so simply as a result of being larger, or as the result of a distinct evolutionary history.

Chimps and gorillas begin their lives in a similar manner.  They cling to their moms, and then once they are big enough, they begin swinging from branches.  But gorillas get bigger more quickly, and have to switch to quadrupedal behavior earlier than chimps.  Since this is the time when their bones are taking shape, this earlier shift may result in bones which are super-adapted for knuckle-walking.  In this scenario, chimps and gorillas have the same structures, but the gorillas display them to a greater degree.  They would also develop them earlier, since they adopted those positions earlier in life.  However, if the differences between chimps and gorillas are the result of a different evolutionary trajectory, and not simple allometric scaling, we would expect convergences in some aspects of the wrist, but some things that were just plain different.

When Dainton and Macho did their analysis, they found a mix of similarities and differences which they interpreted as support for a knuckle-walking as homoplasy hypothesis.  They examined two wrist bones on the ulnar side of the wrist- the hamate and capitate, and found that they were shorter throughout ontogeny in gorillas.  They also found that the articular surfaces of the hamate, capitate, and triquetral bones (again, from the ulnar side of the wrist) were larger and allowed more flexibility.  Both of these observations are consistent with the idea that the ulnar loading is a distinct biomechanical behavior in gorillas and not a side effect of gorillas being bigger animals.

That sets the stage for a new paper, released this week by Tracy Kivell and Daniel Schmitt.  In it, they examined a larger sample of African apes, and also included some arboreal quadrupeds as well.  They explored three wrist bones in search of features which are usually discussed as adaptations for knuclke-walking.  In the scaphoid, there is a concavity on the top and a “beak,” which together “catch” the radius as it extends dorsally.  This would act to limit the flexibility at this joint to only what is needed, while allowing the wrist to be stable when weight was applied.  On the capitate, there is a concavity and a “waist,” which catch the scaphoid and limit the flexibility at this joint.  The capitate and hamate both have a little ridge on their dorsal sides, which limit the flexibility between the first row of wrist bones and the second row.  And finally, the hamate is said to have a concavity which catches the triquetral and limits flexibility there.  So, the authors looked for these features in gorillas, chimps, humans, and cercopithecoids of all ages.

What they found was pretty intriguing.  Gorillas had miserable percentages of most of these supposedly knuckle-walking adaptational features.  The scaphoid concavity and beak?  Only 6% of them had it, compared with 96% of the Pan troglodytes and 76% of P. paniscus.  Even more surprising was that 80% of the arboreal palmigrade monkeys, 76% of the terrestrial palmigrade monkeys, and 57% of the digitigrade monkeys had these features!

The same pattern is exhibited in the other features examined, as well.  Gorillas display them less often, and to a lesser degree than in chimps, and sometimes even less than in other monkeys.  Gorillas, then, have a more flexible wrist, at least in terms of their skeleton.  It could be that their ligaments and muscles are doing more of the work to keep them stable, but we also know that gorillas are able to extend their wrists to around 58 degrees, while chimps can only extend it up to 42 degrees.

The authors suggest that, instead of supporting itself on an extended wrist like the chimpanzee, the gorilla is adopting more of a straight, neutral posture.  The chimp, because of its extended wrist, will experience more bending loads, and needs more bony reinforcement to keep the wrist in place.  The gorilla, because of its straight wrist, would not experience those bending loads, and would have no need for the reinforcement.

The authors discuss for a bit the differences in substrate use between gorillas and chimps.  Gorillas are large, and become that way quickly, so they spend most of their time on the ground.  Chimpanzees are a bit smaller, and can spend quite a bit more time in the trees.  Chimpanzees, while in the trees, are sometimes knuckle-walkers and sometimes palmigrade. During growth, they could develop bony growths which reflect those positions.  This is different than developing bony growths which limit the associated movements- an important distinction which the authors are very correct in making.

So, what we have here is some very interesting evidence that knuckle-walking in gorillas is different than it is in chimps.  We also have a good scenario for why it would be:  Different substrates require different biomechanics.  We even have good reason to throw out some features of the wrist which have been reported repeatedly as markers of knuckle-walking, when what they really reflect could be arboreality.  All of these things could very well indicate that knuckle-walking evolved twice- once in the gorilla clade and once in the chimpanzee clade- and that humans, therefore, probably didn’t go through a knuckle-walking phase.  It’s a good and interesting conclusion to make.

I’d like to play devil’s advocate, though.  I read a really awesome paper this year by Drew Rendall and Anthony Di Fiore about behavioral evolution.  The main gist of the article is that behavior can be just as powerful as genetics or morphology when constructing phylogenies.  Morphology and genetics are just as evolutionarily labile as behavior.  They lay down some really good evidence, so give it a read.  Given that behavior isn’t “special” in terms of evolution, what if knuckle-walking as a behavior evolved once, and gorillas and chimps (and humans!) went their separate ways morphologically?  Their morphological differences accrued as their ontogenetic trajectories differentiated from one another.

I’m not committed to either idea, because if there’s one thing that I’ve learned, it’s that homoplasy is quite common in Miocene and extant hominoids.  The problem with my behavior-as-evidence idea is that behavior doesn’t fossilize, so all we have to go on are the bones, and I can’t really think of a way to test it.

Kivell, T., & Schmitt, D. (2009). Independent evolution of knuckle-walking in African apes shows that humans did not evolve from a knuckle-walking ancestor Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0901280106
RENDALL, D., & DIFIORE, A. (2007). Homoplasy, homology, and the perceived special status of behavior in evolution Journal of Human Evolution, 52 (5), 504-521 DOI: 10.1016/j.jhevol.2006.11.014
Dainton, M., & Macho, G. (1999). Did knuckle walking evolve twice? Journal of Human Evolution, 36 (2), 171-194 DOI: 10.1006/jhev.1998.0265

P.S.  Also check out Laelaps and Afarensis for more.

About these ads

8 responses to “Did knuckle-walking evolve twice?

  1. Marcel F. Williams August 15, 2009 at 7:00 pm

    Knuckle walking also appears to have evolved in the Miocene primate Equatorius (Kenyapithecus); so it appears to have evolved at least 3 times. Humans also have reduced hair on the dorsal surface of the fingers which suggest some knuckle-walking ancestry in hominin evolution– before the emergence of bipedalism. It should be noted that Orangs are fist walkers when they come down to the surface.

    Since the phalanges of hominoids are naturally curved, this may predispose apes to knuckle-walking and fist walking under more terrestrial conditions.

  2. Pingback: Mailund on the Internet » Blog Archive » Last week in the blogs

  3. Pingback: ResearchBlogging.org News » Blog Archive » Editor’s selections: Knuckle-walking, nanostars, and novel fuel cells

  4. Pingback: Mailund on the Internet » Blog Archive » Ardi, Ardi, Ardi…

  5. blue October 3, 2009 at 10:17 pm

    i’d love to see a follow up to this in light of the ardipithecus material

  6. cash loans May 21, 2013 at 8:19 pm

    brodqddc zfijtvxewg ojoohnrm kxngvpkh rnjlyz

  7. Pingback: Scientia Pro Publica 10, The Science, Nature and Medicine Blog Carnival | This Scientific Life

  8. Pingback: ResearchBlogging editor’s selections: Knuckle-walking, nanostars, and novel fuel cells | Skulls in the Stars


Get every new post delivered to your Inbox.

Join 29 other followers

%d bloggers like this: