So… Did knuckle walking evolve twice?

Almost certainly.

We had lots of clues that this was the case before Ardi, but now that we’ve got Ardi- the palmigrade extraordinaire, we know that humans did not go through a knuckle-walking phase, and that chimpanzee knuckle-walking has evolved since the split with our last common ancestor with them.  Which would also means that it evolved after our split with the gorillas… which means that knuckle-walking evolved twice.

The Great Auk by James Audubon

The Great Auk by James Audubon

As we’ve discussed before, knuckle-walking is a pretty weird thing to do, which is why the idea that it evolved only once is hard to shake.  But once you’ve got a particular body plan, there are only so many ways to accomplish a certain task.  For example, the now-extinct Great Auk was a flightless sea bird that hunted fish underwater.  It was white on its front, and black on the back, and had powerful rear feet and webbed toes.  Sound familiar?  The Great Auk was the Northern Hemisphere’s version of the Penguin, but the two were not particularly closely related.  It’s simply that, once you’ve got the body plan of a bird and you want to start diving for fish at high latitudes, you’ve got to rework the wing a little bit so that it’s no longer any good for flying in the air- but man, will it be good for underwater flight!  And then you can work on your body shape a little so that you’re like a little avian torpedo.  And then you’ve got to put some body fat on so that you can withstand the frigid ocean temperatures.  And voila!  You’ve got two almost-identical ocean birds separated by an entire planet and a couple of hundred thousand years, if not more.

The apes have done a very similar thing.  The general Miocene ape was an above-branch quadruped.  The ape had a shoulder in which the socket of the shoulder joint (the glenoid fossa of the scapula) was deep and cup-like so that the ball of the humerus was relatively stable.  In order to get a flexible shoulder, the ape had to move its scapulae so that the glenoid fossa was facing laterally (outward, toward the sides) instead of ventrally (forward).  The glenoid fosssa also became a little shallower, so that it could accommodate a wider range of motion.  When monkeys, or squirrels, or dogs land on the ground with their front feet, the forces are allowed to travel all the way up the humerus until they reach the glenoid fossa.  But think… if you’re an ape with a fancy new flexible shoulder, and you try to land on your front feet, what happens?

Well, you’ll dislocate your shoulder.  That’s a very bad thing for an ape to do!

Monkey Thorax

Monkey Thorax. From Aiello and Dean's Human Evolutionary Anatomy

Human Thorax

Ape Thorax (in this case, a human). From Aiello and Dean's Human Evolutionary Anatomy.

Another thing that some of the apes have done is make their elbow more flexible, as well.  If you look at a monkey’s (or squirrel’s, or dog’s) ulna at the elbow, they have a huge chunk of bone on the end of the olecranon process (the part that articulates with the distal humerus) that we don’t have.  Also, the little U  that cups the trochlea of the humerus points to the side in all of those quadrupeds- more like a C than a U.  But in apes, it points straight up.  All of that makes for an extremely flexible elbow- and gives us the same problem with dislocation that we had with the shoulder.

So, you’re a Miocene ape, and you’ve evolved this flexible shoulder and flexible arm, but you’re also starting to get quite large.  So large, in fact, that sometimes you have to leave the trees because the terminal branches that connect one tree to another just aren’t large enough to support you.  So, you come down out of the trees and walk on the ground.  And man, are your muscles working overtime to keep those bones from dislocating!  Each step you talk with your palms on the ground sends a huge amount of force up through your wrist to your elbow, and then up through your shoulder.  Your muscles are concentrically contracting, trying to keep everything in its right place- but the problem with concentric contraction is that you’re already contracting as hard as you can, so one misplaced step, and your muscles can’t do anything to help you avoid injury.

Proximal Ulna in a Human (left) and a Cat (right). Note the extra bone at the upper edge (the olecranon process) in the cat, as well as the position of the semilunar notch.

But if you walk on your knuckles, it doesn’t do that!  Because you are such an excellent climber, you have these massively strong flexor tendons in your fingers.  If you flex them and put your weight on your knuckles, those tendons eccentrically contract and are able to absorb some of the force coming at you from the ground.  That way, your arm muscles don’t have to work so hard simply to keep your bones from dislocating.  Plus, you’ve got those long fingers for vertical tree climbing and suspension, and tucking them under your hand gets them out of the way.

Gorillas and chimps have both figured this out (anatomically speaking- I don’t know if any of them have put any thought into it!).  And orangutans are fistwalkers- but I know that I’ve seen one on a really hard concrete surface using his knucles.  It’s a classic case of convergent evolution- but you can’t really tell from the genetics or the living anatomy of the animals- we needed a fossil to finally be relatively certain.  Much like the Auk and the Penguin, the Chimp and the Gorilla faced similar problems brought about by their environment and anatomy, and ended up solving them in similar ways.

ResearchBlogging.orgLovejoy, C., Simpson, S., White, T., Asfaw, B., & Suwa, G. (2009). Careful Climbing in the Miocene: The Forelimbs of Ardipithecus ramidus and Humans Are Primitive Science, 326 (5949), 70-70 DOI: 10.1126/science.1175827

Ward, Carol (2007).  Postcranial and Locomotor Adaptations of the Hominoids.  Handbook of Paleoanthropology.  DOI: 10.1007/978-3-540-33761-4

About these ads

13 responses to “So… Did knuckle walking evolve twice?

  1. Mike Keesey January 11, 2010 at 7:54 pm

    Interesting way to think about it.

    Wasn’t there a study recently (but pre-Ardi) on the differences between gorilla and chimpanzee knuckle-walking, indicating that they may have originated independently?

    To play Devil’s advocate, isn’t it possible that Ardi, which presumably lived after the human-chimpanzee split, had lost the ability to knuckle-walk? Don’t we need a fossil from the human-chimpanzee stem group to tell for certain?

    “The Great Auk was the Northern Hemisphere’s version of the Penguin, but the two were not particularly closely related.”

    Indeed, there seems to be a lot of etymological confusion as to which of these birds the word “penguin” originally referred to.

  2. zinjanthropus January 11, 2010 at 9:29 pm

    It’s remotely possible that Ardi had lost the ability, but remember that Ardi isn’t just different from any extant apes, she’s *primitive.* Meaning that she retains a lot of things from apes like Proconsul and friends. So we’re not just saying that she lost knuckle-walking, we’re saying that she lost knuckle-walking and re-gained all of her primitive features. And yes, atavisms are possible, but we’re talking a lot of them here. And, since you point out that gorillas and chimps knuckle-walk differently, it’s not too hard to imagine that they converged upon the same form of locomotion.

    I think “The Great Divides” is the paper from the Science suite that talks about a lot of that stuff.

    • Mike Keesey January 11, 2010 at 10:33 pm

      To be clear, it is seeming more and more likely to me that knuckle-walking is convergent between Pan and Gorilla–I’m even willing to accept it as the default hypothesis for now. I just don’t see that Ardi shows this conclusively. Primitive traits are almost always lost in a “mosaic” pattern. The fact that Ardipithecus is like Proconsul in some ways doesn’t mean it isn’t like Homo in others, after all. If Ardipithecus ramidus is a stem-human species, then it we would expect it to share apomorphies with humans, and there’s no a priori reason “loss of knuckle-walking” couldn’t be one of them. I really don’t follow your line of argument here.

      Now, if Ardipithecus ramidus turns out to be a stem-chimpanzee or a member of the human-chimpanzee stem group, or another species from those stem groups is discovered and it lacks manual adaptations for knuckle-walking, then we would have something really solid. (It could always have been secondarily lost twice, but that would be less likely.) But stem-humans that don’t knuckle-walk are perfectly consistent with either hypothesis.

      • zinjanthropus January 12, 2010 at 12:02 pm

        I suppose that it is consistent, but I do think it’s unlikely. The things I talked about in the above post weren’t really adaptations to knuckle-walking, but were adaptations to suspension and vertical climbing that made knuckle-walking necessary. Unless we’re saying that Ardi evolved an above-branch quadrupedality from a below-branch suspension specialist, there’s really no reason why Ardi should ever have had to knuckle-walk.

  3. amphiox January 11, 2010 at 9:35 pm

    If the two forms of knuckle walking are convergent, there ought to be some tell-tale difference in the fine details that betray the different historical contingencies of their respective evolutionary pasts. In theory that ought to be something we can search for and find.

    • zinjanthropus January 11, 2010 at 9:52 pm

      And we do find it! Gorillas use a more columnar wrist posture, while chimps flex their wrists a little more. And if I remember correctly, chimps place their weight on the third and fourth knuckles, while gorillas place it more on the first two. And there are subtle differences in the bones that are most likely the result of cartilage modeling.

      T.L.Kivell & D.Schmitt (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 106 (34):14241-14246 doi 10.1073.pnas.0901280106

  4. CW January 12, 2010 at 3:39 am

    I enjoyed this post, and I found it fairly understandable. I’ve discovered the subject of evolution relatively recently, and am blown away by all the information that we can obtain from the distant past.

    Just a quick question regarding your introductory explanation of the auk, is it fair to say that modern penguin wings have evolved for the sole purpose of facilitating underwater diving and swimming?

    • zinjanthropus January 12, 2010 at 12:05 pm

      That’s a good question, and I’m not really sure about the answer! It would seem to me that the wings are primarily adapted for swimming and diving, but I’m sure that they are good for other secondary things, as well…

      Any penguin morphologists want to weigh in?

  5. Pingback: Four Stone Hearth #84 is Gelada-ful « Neuroanthropology

  6. Eric January 21, 2010 at 9:52 pm

    I just want to add some of my thoughts about this topic, I’m not sure if I will add anything interesting to the discussion or tell you something completely new, but here you go:

    If we are looking at the recent great apes (for my own convinience I will ignore the Gibbons), we knew, that there are at least two forms of terrestrial locomotion: Knuckle (or Fist-) walking in the great Apes and bipedalism.

    As you all can probably see, as a miocene ape, there are two ways to avoid dislocating your shoulder. Either you start knuckle walking, or you completely avoid walking on your forelimbs.

    I think, you all agree with me, that the latter scenario is not that unlikely if we are looking at Ardipithecus (and maybe Oreopithecus as well).
    But when we are discussing the possibility of convergent evolution of knuckle walking in the african Apes because of similiar functional needs, maybe we should also think about convergent evolution of bipedalism because of the same reasons.

    I always wondered, why no one askes questions about this, so I just thought that maybe I should start asking theese questions.

    • zinjanthropus January 21, 2010 at 10:28 pm

      That’s a really good point, Eric!

      I think a lot of the convergences in gorillas and chimps are because of their adaptations to suspension, but Ardipithecus has a more flexible shoulder than, say, a macaque, so that is a very interesting point!

      And- gibbons walk bipedally when they are on the ground, too, but I’m sure their suuuuuuuper long arms have something to do with that, too.

      I like it!


Get every new post delivered to your Inbox.

Join 29 other followers

%d bloggers like this: