National Geographic

Evolution’s Baby Steps

If you explore our genealogy back beyond about 370 million years ago, it gets fishy. Our ancestors back then were aquatic vertebrates that breathed through gills and swam with fins. Over the next twenty million years or so, our fishy ancestors were transformed into land-walking animals known as tetrapods (Latin for “four feet”).

The hardest evidence–both literally and figuratively–that we have for this transition comes from the fossil record. Over the past century, paleontologists have slowly but steadily unearthed species belong to our lineage, splitting off early in the evolution of the tetrapod body. As a result, we can see the skeletons of fish with some–but not all–of the traits that let tetrapods move around on land. (I wrote about the history of this search in my book At the Water’s Edge; for more information, I’d suggest Your Inner Fish, by Neil Shubin, who discovered Tiktaalik, one of the most important fossils on the tetrapod lineage.)

Updated reconstruction of Tiktaalik. Image courtesy of John Westlund, University of Chicago.

Updated reconstruction of Tiktaalik. Image courtesy of John Westlund, University of Chicago.

You can get a feeling for how fish became tetrapods by looking at a fossil like Tiktaalik, shown here. These days, a lot of scientists are turning up clues about how a fish turned into this kind of creature, and how this kind of creature turned into creatures like us.

Along the way, a lot of genes changed. The genes in the egg of a fish encode the molecules that will produce the fins, gills, and all the rest of a fish’s body. A different set of genes will produce a tetrapod. These days, scientists are finding some of the mutations that reprogrammed fins into feet.

But a new study in Nature puts a fascinating new wrinkle on our origins story. It suggests that our fishy ancestors already had the potential to develop the beginnings of a tetrapod body. They just needed some time on land to bring it out.

The authors of the new study, three scientists at McGill University in Montreal, studied fish called bichirs (Polypterus). Bichirs are the living remnants of a very old lineage of fishes, which split off from other fish lineages some 400 million years ago. While they mostly live in lakes and rivers, they will sometimes crawl across dry land with their fins. They can even sustain themselves on these journeys by breathing through primitive lungs. Here’s a video of how they walk.

The McGill researchers saw some intriguing parallels between bichirs and early tetrapod relatives like Tiktaalik. Bichirs use their front pair of fins to lift their head and the front end of their trunk off the ground. They then push the back end of their trunk in order to propel themselves forward. Tiktaalik may have moved in a similar fashion.

Credit: Antoine Morin

Credit: Antoine Morin

But bichirs spend relatively little time on land. The McGill scientists wondered what would happen if they forced the fish to grow up out of the water. To find out, they reared 111 bichirs in a terrarium with a pebble-strewn floor. To prevent the bichirs from drying out, the scientists installed a mister to keep their skin moist. The fish grew for eight months, clambering around their terrarium instead of swimming.

Then the scientists examined these fish out of water. They found that eight months on dry land (or at least moist land) had wreaked profound changes to the bichirs.

For one thing, they now walked differently. Overall, they were more efficient. In each step, they planted their fins on the ground for less time, and they took shorter strides. Instead of flapping their fins out to each side, they placed their fins under their bodies. Their fins slipped less when they pushed off of them. They made smaller movements with their tails to go the same distance as a bichir raised underwater. Aquatic bichirs walk on land with an irregular gait. The terrestrial bichirs, on the other hand, walked more gracefully, planting their fins in the same spot relative to their bodies time after time.

The bichirs probably developed this new walking style in large part through learning. Growing up on land, they had more opportunity to test out their moves and to perfect the best ones. But it wasn’t just their brains that changed on land. Their bodies changed, too.

The McGill scientists examined the bones of the terrestrial bichirs and compared them to normal aquatic ones. They found striking differences in the bones, especially in the shoulder region. Some of the bones had become less tightly connected to each other, giving the bichirs more room to swing their fins as they walked.  By contrast, their bones corresponding to our collarbones became bigger and more strongly braced, letting the animals resist gravity and lift their bodies higher.

It was the experience of walking that changed the fish bones. The forces exerted on them changed how the bone cells grew, leading them to take on new shapes. What’s especially intriguing about these changes is that they’re a lot like the changes paleontologists have documented in tetrapod fossils.

Over millions of years, our fishy ancestors evolved looser connections between some shoulder bones, enabling their legs to sweep bigger motions and also starting to separate the head from the neck. They also evolved stronger support systems for their trunk so that they could lift themselves out of the muck. It’s as if the bichirs are replaying evolution in their own lifetime.

This is not the first time that biologists have found tantalizing parallels between the experiences of individual animals and long-term evolution. The environment in which animals grow up can steer the development of their bodies, and then evolution can follow suit.

In 2008, for example, scientists raised stickleback fish on two different diets. One group of fish ate bloodworms squirming around at the bottom of their tanks. The other fish ate shrimp scooting around in the open water. The bloodworm-eating fish had to clamp down on the blood worms to eat them, while the shrimp-eating ones just needed to sneak up on their prey and swallow them with a quick slurp.

The result of these different movements was different heads: the bloodworm-feeders had short, wide mouths, and the shrimp-feeders had long, narrow ones.

Wund et al, American Naturalist 2008

Wund et al, American Naturalist 2008

These sticklebacks, which are abundant in the ocean, have repeatedly colonized lakes. As they’ve adapted to their new freshwater home, they’ve evolved over and over again into two distinct populations. Some of them scrounge on the lake bottoms for prey, while the others zip around the open water. And time and again, they’ve evolved wide short mouths, or narrow long ones.  Evolution has followed the path of experience.

The ability that animals and plants have to develop differently in different conditions is known as plasticity. It’s possible that plasticity opens the door to new paths that evolution can then take. When organisms find themselves in a new environment, they develop in a way that helps them cope with their new surroundings. Their descendants may acquire mutations that encode that anatomy in their genes. Eventually evolution takes them beyond where plasticity alone could take them.

This kind of experience-led evolution, known as genetic assimilation, might have helped take our ancestors out of the water. The forerunners of tetrapods might have been forced to scoot over dry land more than their ancestors–to flee predators in the water perhaps, perhaps to mate, perhaps to get to other streams and ponds. Their underwater genomes gave them the plasticity to grow into halfway decent walkers, like bichirs do today. And then subsequent mutations gave them an improved anatomy. They didn’t have to grow into walking: their genes had now taken over the job, and our ancestors were ready to walk on.

[Correction: I updated the post to the correct number of bichirs in the study.]

There are 22 Comments. Add Yours.

  1. John
    August 27, 2014

    I know that it’s not Lamarckism, but I’ll bet some will say Lamarck was right.

  2. David Percival
    August 27, 2014

    Why do you say it is not Lamarckism? If it walks like a duck (or birchir) and evolves like a stickleback……………………………..

    If you know it’s not, why are you withholding your nowledge from the rest of us?

    [CZ: It's not Lamarkism, because that would mean that a bichir that developed walking-adapted bones in its own life would pass on those traits to its offspring. But these fish in this experiment will produce fish that are no different than aquatic ones. In genetic assimilation, natural selection comes in the wake of the developmental changes, favoring mutations that reinforce the traits.]

  3. Neil Clark
    August 27, 2014

    John & David – this article in no way `demonstrates’ or `proves’ Lamarkism (the acquisition of traits due to direct effort by the organism – i.e.: that giraffes developed long necks by repeatedly `stretching’ their shorter necks to reach tasty foliage). Instead, those lab sticklebacks which had mouths just slightly better-suited to each prey type would thrive, compared to those with relatively less-well-suited, so the better-suited would thrive & reproduce in greater numbers…bestowing their advantage to more offspring…

  4. Paul Strode
    August 27, 2014

    Carl: I’d love to see the reverse experiment where a sample population of an amphibian species is raised in nothing but an aquatic environment for several generations and compared to a sample population raised entirely in a terrestrial habitat.

  5. David Dobbs
    August 27, 2014

    Why bring Lamarck in? It just fogs the issue and tempts (or allows) people to cast this very different dynamic as part of an old debate.

  6. John
    August 28, 2014

    People (wrongly) say epigenetics is Lamarckism, so how will they respond to this? Here, you have physical changes within one lifetime resulting from behavior, followed by evolution in the same direction. Lamarckians will likely claim post hoc ergo propter hoc. This is a fallacy. As explained by Carl, the hereditary change comes from mutation and natural selection, rather than acquired characteristics.

    Best to nip it in the bud.

  7. Levan
    August 28, 2014

    I love fairy tales! Great!

    [CZ: Please share with us your wisdom about how tetrapods evolved. Be sure to explain the evidence.]

  8. Greg
    August 28, 2014

    It would have been interesting had they bred the Birchirs (terrarium experiment) for several generations to see what other changes occurred.

  9. Jessica
    August 29, 2014

    Of course they would get more efficient at using different muscles and the bone structure would change some as they used them differently and matured. I would expect some micro evolution that we have already observed time and time again. But this example appears to be inconsistent with evolutionary theory-why would there genome change /mutate? To say that the genes mutated first (genes that not only changed their fins but their skin) seems backwards. There are a lot of speculations in this article- basically this is not evidence for Darwinian evolution at all- this is arguing that changes in genes came first.

    [CZ: The hypothesis here is that the ancestors of tetrapods spent more time on land, and they developed a more tetrapod-like body as they matured. Natural selection only later favored mutations that also produced a more tetrapod-like body plan.]

  10. davidpercival
    August 29, 2014

    Greg’s point is a good one. If the offspring of the Birchers were born with the physical characteristics developed in the lifetime of the previous generation it would surely be evidence of inheritence of acquired characteristics, which I take to be Lamarckism, rather than Neil Clark’s definition in his comment.
    There has been recent evidence that such changes in behaviour and resulting characteristics can be inherited.

    [CZ: There is no reason to expect that these bichirs would pass on their developmental changes to their offspring.]

  11. RBH
    August 29, 2014

    Carl writes “When organisms find themselves in a new environment, they develop in a way that helps them cope with their new surroundings. Their descendants may acquire mutations that encode that anatomy in their genes. Eventually evolution takes them beyond where plasticity alone could take them.”

    That’s a version of the Baldwin Effect first offered in 1896 (see here): “The ‘Baldwin effect’ is better understood in evolutionary developmental biology literature as a scenario in which a character or trait change occurring in an organism as a result of its interaction with its environment becomes gradually assimilated into its developmental genetic or epigenetic repertoire (Simpson, 1953; Newman, 2002).”

  12. Rick Eyerdam
    August 29, 2014

    Its almost like a homo sapiens that was placed in a terrarium full of heavy weights with weak muscles and spongy tendons. And it is compelled to lift the weights in order to receive positive reinforcement and then, amazing over time the muscles grow stronger, the posture is better, the tendons more robust. What an amazing experiment! Must be a sure sign of an evolutionary path. Or not.

    [CZ: It would be an interesting experiment if the developmental changes were beneficial for a different environment--and that the fossil record documented a transition to that new environment. But since there isn't such a parallel, your analogy doesn't work.]

  13. Rick Eyerdam
    August 29, 2014

    oops forgot to ad. So when they put them back in the water were they stronger walkers or swimmers? They didn’t put the back in the water? Where is the control?

    [CZ: They did have controls. And they did put the terrestrial fish back in the water to see how they swam. There was no difference in their swimming performance. It's all in the paper.]

  14. lauramk
    August 29, 2014

    Strictly speaking this is individual adaptation allowed by phenotypic plasticity, NOT evolution. Evolution would occur if the genotype of the POPULATION changed as a result of a survival advantage given by the phenotypic adaptation to the environment. Although this is not impossible, it is not at all mainstream thinking in evolutionary biology. Genetic assimilation is very theoretical. Zimmer’s, “mights,” and “could- bes,” are hiding how very unusual this thinking is. I like the article, but I think he should have explained how fringe-y this thinking still is.

    [CZ: I never said that the developmental changes seen in individual fish was evolution. You are misreading the article. What I'm saying is that these scientists have concluded that this experiment and evidence from the fossil record are evidence for genetic assimilation and developmental plasticity in the evolution of tetrapods. Genetic assimilation is not fringe-y. If you type the term into Pubmed, you get 95 papers, many in leading journals such as Evolution. So I see no need to describe it as fringe-y.]

  15. kelphead
    August 29, 2014

    From the article and the video, it seems that they don’t use the two little hind fins when they walk: they plant the front pair and squirm with everything behind those two. Pardon the teleology, but how do they use the leglike back pair? Are these the descendants of early tetrapods that decided to become aquatic again?

    [CZ: Bichirs don't use their pelvic fins to walk. They only use them for maneuvering while swimming. They are not secondarily aquatic--all their ancestors were aquatic (with some excursions over land).]

  16. SocraticGadfly
    August 29, 2014

    That said, it’s not Lamarckian, of course, but the fish who were naturally “loose-limbed” would pass that tendency on through normal hereditary descent, as you note up top. Related changes, such as longer fins as they started becoming limbs, would then follow, etc., along fish having skin that retained moisture a bit better passing that on hereditarily, etc.

  17. davidpercival
    August 30, 2014

    While I very much appreciate Mr Zimmer replying to many of the comments, I wonder how he can be so sure that there is no reason to expect these birchers would pass on their developmental changes to their offspring. After all, as I understand it, a recent experiment on mice which were given an induced fear of a certain gas found they produced changes in gene expression which we passed on to their offspring, who also displayed fear of the gas, without having been exposed to it before themselves. Similar developments were noted in chickens used in an experiment in Sweden two or three years ago when light patterns were altered in one group and compared with a genitically comparable control group.

    I was wrong to say “physical characteristics” as this implies much greater change in one generation than is credible but it does seem to me more likely that a change in environment or behaviour could cause changes in gene expression which are inherited. If the birchers in this case had been bred from and their offspring exposed to the same conditions, it is possible this phenomena could have been investigated more thoroughly.

    Of course, for such changes to be fixed genetic changes through natural selection would need to apply but as a starting point for evolutionary change, from a common sense point of view, this seems more probable than totally random genetic mutations.

    [CZ: The experiments you discuss explore a phenomenon called transgenerational epigenetics. An animal has a stressful experience, for example, and the experience leads to a modification of the molecules that surround the animal's DNA in certain cell, such as neurons in a particular part of the brain. Through a process that's not yet well understood, parents can sometimes pass on these modifications to their offspring. This is fascinating work, but it would be wrong to leap from these experiments to the assumption that all developmental changes are carried down to the next generation through transgenerational epigenetics. The fact is that scientists fail to find traits passed down this way in the vast majority of cases. Indeed, agriculture and animal breeding would be fundamentally different if this were the case. If you wanted to breed strong oxen, all you'd need to do was work oxen hard until they became strong, and then breed them. But that doesn't work. Instead, ox breeders pick out the strongest individuals to breed, thus selecting the genetic variants associated with strength.]

  18. lauramk
    August 30, 2014

    Thank you for responding. I would point out that the title of the article is, “Evolution’s Baby Steps.” To many of us, this implies that the article is somehow about evolution. Also, how many of the articles in Pub Med are by the same groups of people? Farish would roll over in his grave to hear this described as mainstream thinking, and Fuzz, well he might set one of his border-collies on you. ;>) Since so many comments are invoking Lamarck, I feel the thinking could have used some qualifications. All that said, I think its important to revisit historical. ideas, because many of them really do have value. Steve and Niles breathed new life into punctuated equilibrium (very popular in the 1920s and 1930s) with a re-examination and a pair of open minds.

    [CZ: The article is, in fact, about evolution. It also clearly distinguishes between the developmental changes in individuals and the subsequent selection on mutations that contribute to that phenotype. Should I have included a footnote emphasizing that this is not Lamarckism? I suppose we could have a discussion about editorial choices, but that's different from your main allegation, which is that I'm writing about something that has no scientific merit. I don't understand why you find 95 papers in Pubmed insufficient evidence that the scientific community considers this a serious concept (let alone the fact that the paper I wrote about was published in a leading journal (Nature).

    But there are plenty of ways to convey the merit of genetic assimilation as a hypothesis. Last year a team of leading evolutionary biologists published the Princeton Guide to Evolution.* It's a thorough exploration of the most important concepts in evolutionary biology. And it discusses genetic assimilation several times.

    *Full disclosure: They asked me to write an entry about evolution and the media.]

  19. Amphiox
    August 30, 2014

    lauramk, this article is about A STEP in the evolution of tetrapods. The developmental plasticity described here enables the fish to colonize a new environment more easily. This in turn exposes them to new selection pressures, which will select for different mutations, either pre-existing or novel, than would have been selected for if the fish had stayed in the water. This results over time Ina change in gene frequencies in the population, which is evolution.

    Furthermore, genetic assimilation is easier than evolving a new trait from scratch. This makes these specific changes more likely to occur.

    The next step after genetic assimilation is new mutations that extend the phenotypic plasticity even further towards adaption for land, after which the process repeats, again and again, until fish fins become cheetah legs. Thus next step is not the focus of this article.

    But the whole story of evolution is not just mutation-selection-change. There is also steps that show you WHERE and HOW the selection pressures arose, and WHY in each case it pushed in the observe direction rather than another. There are also steps that explain WHY one particular type of change occurred rather than any number of alternative changes that would also have worked for that particular selection pressure. For example, tetrapods evolved walking limbs for land locomotion. They did not evolve tank treads. Why? Because phenotypic plasticity allowed for genetic assimilation in the direction of fins evolving into limbs, but not for fins to evolve towards becoming tank treads, and subsequent evolution thus was more likely to go in that direction instead of the other.

    So this is precisely one of the earliest steps on the evolutionary sequence from fish fins to tetrapods limbs, ie a “baby step”.

  20. Gralgrathor
    August 31, 2014

    The comments here make an interesting read. The article is interesting too. Funny that they appear to be about different things. Do I get it right when I say that the real result of this experiment is to push back the development of amphibious abilities in vertebrates a couple of tens of million years, and to suggest an evolutionary pathway for this development?

    [CZ: The new study doesn't say anything specific about the timing of the evolution of walking in the tetrapod lineage. It points to a hypothesis about how that transition may have occurred--through genetic assimilation of developmental plasticity.]

  21. DN
    September 2, 2014

    It’s all coming up flexible stems!

  22. Blakeness
    September 5, 2014

    I would love to see a follow up on this measuring things like cortisol and other potential agents of transgeneration epigenetics and the activation of “loose DNAses” similar as was described recently with the ImuABC cluster in the article “with-evolutionary-rocket-fuel-bacteria-give-peas-a-chance.” Thank you for this article and contributing to the thread.

Add Your Comments

All fields required.

Related Posts