Your Inner Fish: A Journey Into the 3.5-Billion-Year History of the Human Body

by Neil Shubin

  Imagine that you are standing in a room full of people whom you have never seen before. You are given a simple task: find out how closely related each person in the room is to you. How do you tell who are your distant cousins, your super-distant cousins, your great-granduncles seventy-five times removed?

  To answer this question, we need a biological mechanism to guide our thinking and give us a way to test the accuracy of our hypothesized family tree. This mechanism comes from thinking about our law of biology. Knowing how descent with modification works is key to unlocking biological history, because descent with modification can leave a signature, which we can detect.

  Let’s take a hypothetical humorless, quite unclown-like couple who have children. One of their sons was born with a genetic mutation that gave him a red rubber nose that squeaks. This son grows up and marries a lucky woman. He passes his mutated nose gene to his children, and they all have his red rubber nose that squeaks. Now, suppose one of his offspring gets a mutation that causes him to have huge floppy feet. When this mutation passes to the next generation, all of his children are like him: they have a red rubber nose that squeaks and huge floppy feet. Go one generation further. Imagine that one of these kids, the original couple’s great-grandchild, has another mutation: orange curly hair. When this mutation passes to the next generation, all of his children will have orange curly hair, a rubber nose that squeaks, and giant floppy feet. When you ask “Who is this bozo?” you’ll be inquiring about each of our poor couple’s great-great-grandchildren.

  This example illustrates a very serious point. Descent with modification can build a family tree, or lineage, that we can identify by characters. It has a signature that we immediately recognize. Like a nested set of Russian dolls, our hypothetical lineage formed groups within groups, which we recognize by their unique features. The group of “full bozo” great-great-grandchildren is descended from an individual who had only the squeaky nose and the huge floppy feet. This individual was in a group of “proto-bozos,” who are descended from an individual who had only the rubber nose that squeaks. This “pre-proto-bozo” was descended from the original couple, who didn’t look overtly clown-like.

  The bozo family tree.

  This pattern of descent with modification means that you could easily have hypothesized the bozo family tree without me telling you anything about it. If you had a room full of the various generations of bozos, you would have seen that all clown kin are in a group that possesses a squeaky nose. A subset of these have orange hair and floppy feet. Nested within this subset is another group, the full bozos. The key is that the features—orange hair, squeaky nose, big floppy feet—enable you to recognize the groups. These features are your evidence for the different groups, or in this case generations, of clowns.

  Replace this family circus with real features—genetic mutations and the body changes that they encode—and you have a lineage that can be identified by biological features. If descent with modification works this way, then our family trees have a signature in their basic structure. So powerful is this truth that it can help us reconstruct family trees from genetic data alone, as we see from the number of genealogical projects currently under way. Obviously, the real world is more complex than our simple hypothetical example. Reconstructing family trees can be difficult if traits arise many different times in a family, if the relationship between a trait and the genes that cause it is not direct, or if traits do not have a genetic basis and arise as the result of changes in diet or other environmental conditions. The good news is that the pattern of descent with modification can often be identified in the face of these complications, almost like filtering out noise from a radio signal.

  But where do our lineages stop? Did the bozos stop at the humorless couple? Does my lineage stop at the first Shubins? That seems awfully arbitrary. Does it stop at Ukranian Jews, or northern Italians? How about at the first humans? Or does it continue to 3.8-billion-year-old pond scum, and beyond? Everybody agrees that their own lineage goes back to some point in time, but just how far back is the issue.

  If our lineage goes all the way back to pond scum, and does so while following our law of biology, then we should be able to marshal evidence and make specific predictions. Rather than being a random assortment of creatures, all life on earth should show the same signature of descent with modification that we saw among the bozos. In fact, the structure of the entire geological record shouldn’t be random, either. Recent additions should appear in relatively young rock layers. Just as I am a more recent arrival than my grandfather in my family tree, so the structure of the family tree of life should also have its parallels in time.

  To see how biologists actually reconstruct our relatedness to other creatures, we need to leave the circus and return to the zoo we visited in the first chapter of the book.

  A (LONGER) WALK THROUGH THE ZOO

  As we’ve seen, our bodies are not put together at random. Here, I use the word “random” in a very specific sense; I mean that the structure of our bodies is definitely not random with respect to the other animals that walk, fly, swim, or crawl across this earth. Some animals share part of our structure; others do not. There is order to what we share with the rest of the world. We have two ears, two eyes, one head, a pair of arms, and a pair of legs. We do not have seven legs or two heads. Nor do we have wheels.

  A walk in the zoo immediately shows our connections to the rest of life. In fact, it will show that we can group much of life in the same way we did with the bozos. Let’s go to just three exhibits at first. Start with the polar bears. You can make a long list of the features that you share with polar bears: hair, mammary glands, four limbs, a neck, and two eyes, among lots of other things. Next, consider the turtle across the way. There are definitely similarities, but the list is a bit shorter. You share four limbs, a neck, and two eyes (among other things) with the turtle. But unlike polar bears and you, turtles don’t have hair or mammary glands. As for the turtle’s shell, that seems unique to the turtle, just as the white fur was unique to the polar bear. Now visit the African fish exhibit. Its inhabitants are still similar to you, but the list of commonalities is even shorter than the list for turtles. Like you, fish have two eyes. Like you, they have four appendages, but those appendages look like fins, not arms and legs. Fish lack, among many other features, the hair and mammary glands that you share with polar bears.

  This is beginning to sound like the Russian doll set of groups, subgroups, and sub-subgroups that appeared in the bozo example. Fish, turtles, polar bears, and humans all share some features—heads, two eyes, two ears, and so on. Turtles, polar bears, and humans have all these features, and they also have necks and limbs, features not seen in fish. Polar bears and humans form an even more elite group, whose members have all of these features and also hair and mammary glands.

  The bozo example gives us the means to make sense of our walk through the zoo. In the bozos, the pattern of groups reflected descent with modification. The implication is that the full-bozo kids shared a more recent relative than they do with the kids who have only a squeaky nose. That makes sense: the parent of the squeaky-nosed kids is the great-great grandparent of the full bozos. Applying this same approach to the groups we encountered during our zoo walk means that humans and polar bears should share a more recent ancestor than they do with turtles. This prediction is true: the earliest mammal is much more recent than the earliest reptile.

  The central issue here is deciphering the family tree of species. Or, in more precise biological terms, their pattern of relatedness. This pattern even gives us the means to interpret a fossil such as Tiktaalik in light of our walk through the zoo. Tiktaalik is a wonderful intermediate between fish and their land-living descendants, but the odds of it being our exact ancestor are very remote. It is more like a cousin of our ancestor. No sane paleontologist would ever claim that he or she had discovered “The Ancestor.” Think about it this way: What is the chance that while walking through any random cemetery on o
ur planet I would discover an actual ancestor of mine? Diminishingly small. What I would discover is that all of the people buried in these cemeteries—no matter whether that cemetery is in China, Botswana, or Italy—are related to me to different degrees. I can find this out by looking at their DNA with many of the forensic techniques in use in crime labs today. I’d see that some of the denizens of the cemeteries are distantly related to me, others are related more closely. This tree would be a very powerful window into my past and my family history. It would also have a practical application because I could use this tree to understand my predilection to get certain diseases and other facts of my biology. The same is true when we infer relationships among species.

  The real power of this family tree lies in the predictions it allows us to make. Chief among these is that as we identify more shared characteristics, they should be consistent with the framework. That is, as I identify features from cells, DNA, and all the other structures, tissues, and molecules in the bodies of these animals, they should support the groupings that we identified during our walk. Conversely, we can falsify our groupings by finding features inconsistent with them. That is, if there exist many traits shared by fish and people that aren’t seen in polar bears, our framework is flawed and needs to be revised or jettisoned. In cases where the evidence is ambiguous, we apply a number of statistical tools to assess the quality of the characteristics supporting the arrangements in the family tree. In instances where there is ambiguity, the genealogical arrangement is treated as a working hypothesis until we can find something conclusive to allow us to either accept or reject it.

  Some groupings are so strong that, for all intents and purposes, we consider them fact. The fish–turtle–polar bear–human grouping, for example, is supported by characteristics from hundreds of genes and virtually all features of the anatomy, physiology, and cellular biology of these animals. Our fish-to-human framework is so strongly supported that we no longer try to marshal evidence for it—doing so would be like dropping a ball fifty times to test the theory of gravity. The same holds for our biological example. You would have the same chance of seeing your ball go up the fifty-first time you dropped it as you would of finding strong evidence against these relationships.

  We can now return to the opening challenge of the book. How can we confidently reconstruct the relationships among long-dead animals and the bodies and genes of recent ones? We look for the signature of descent with modification, we add characteristics, we evaluate the quality of the evidence, and we assess the degree to which our groups are represented in the fossil record. The amazing thing is that we now have tools to probe this hierarchy, using computers and large DNA sequencing labs to perform the same analyses you performed during your walk through the zoo. We now have access to new fossil sites around the world. We can see our bodies’ place in the natural world better than we ever could.

  From Chapter 1 through Chapter 10, we have shown that deep similarities exist between creatures living today and those long deceased—ancient worms, living sponges, and various kinds of fish. Now, armed with knowledge of the pattern of descent with modification, we can begin to make sense of it all. Enough fun at the circus and zoo. It’s time to get down to business.

  We have seen that inside our bodies are connections to a menagerie of other creatures. Some parts resemble parts of jellyfish, others parts of worms, still others parts of fish. These aren’t haphazard similarities. Some parts of us are seen in every other animal; others are very unique to us. It is deeply beautiful to see that there is an order in all these features. Hundreds of characters from DNA, innumerable anatomical and developmental features—all follow the same logic as the bozos we saw earlier.

  Let’s consider some of the features we’ve already talked about in the book and show you how they are ordered.

  With every other animal on the planet, we share a body composed of many cells. Call this group multicellular life. We share the trait of multicellularity with everything from sponges to placozoans to jellyfish to chimpanzees.

  A subset of these multicellular animals have a body plan like ours, with a front and a back, a top and a bottom, and a left and a right. Taxonomists call this group Bilateria (meaning “bilaterally symmetrical animals”). It includes every animal from insects to humans.

  A subset of multicellular animals that have a body plan like ours, with a front and a back, a top and a bottom, and a left and a right, also have skulls and backbones. Call these creatures vertebrates.

  A subset of the multicellular animals that have a body plan like ours, with a front and a back, a top and a bottom, and a left and a right, and that have skulls, also have hands and feet. Call these vertebrates tetrapods (animals with four limbs).

  A subset of the multicellular animals that have a body plan like ours, with a front and a back, a top and a bottom, and a left and a right, that have skulls, and that have hands and feet, also have a three-boned middle ear. Call these tetrapods mammals.

  A subset of the multicellular animals that have a body plan like ours with a front and a back, a top and a bottom, and a left and a right, that have skulls and backbones, that have hands and feet, and that have a three-boned middle ear, also have a bipedal gait and enormous brains. Call these mammals people.

  A human family tree, all the way back to jellyfish. It has the same structure as the one for the bozos.

  The power of these groupings is seen in the evidence on which they are based. Hundreds of genetic, embryological, and anatomical features support them. This arrangement allows us to look inside ourselves in an important way.

  This exercise is almost like peeling an onion, exposing layer after layer of history. First we see features we share with all other mammals. Then, as we look deeper, we find the features we share with fish. Deeper still are those we share with worms. And so on. Recalling the logic of the bozos, this means that we see a pattern of descent with modification deeply etched inside our own bodies. That pattern is reflected in the geological record. The oldest many-celled fossil is over 600 million years old. The earliest fossil with a three-boned middle ear is less than 200 million years old. The oldest fossil with a bipedal gait is around 4 million years old. Are all these facts just coincidence, or do they reflect a law of biology we can see at work around us every day?

  Carl Sagan once famously said that looking at the stars is like looking back in time. The stars’ light began the journey to our eyes eons ago, long before our world was formed. I like to think that looking at humans is much like peering at the stars. If you know how to look, our body becomes a time capsule that, when opened, tells of critical moments in the history of our planet and of a distant past in ancient oceans, streams, and forests. Changes in the ancient atmosphere are reflected in the molecules that allow our cells to cooperate to make bodies. The environment of ancient streams shaped the basic anatomy of our limbs. Our color vision and sense of smell has been molded by life in ancient forests and plains. And the list goes on. This history is our inheritance, one that affects our lives today and will do so in the future.

  WHY HISTORY MAKES US SICK

  My knee was swollen to the size a grapefruit, and one of my colleagues from the surgery department was twisting and bending it to determine whether I had strained or ripped one of the ligaments or cartilage pads inside. This, and the MRI scan that followed, revealed a torn meniscus, the probable result of twenty-five years spent carrying a backpack over rocks, boulders, and scree in the field. Hurt your knee and you will almost certainly injure one or more of three structures: the medial meniscus, the medial collateral ligament, or the anterior cruciate ligament. So regular are injuries to these three parts of your knee that these three structures are known among doctors as the “Unhappy Triad.” They are clear evidence of the pitfalls of having an inner fish. Fish do not walk on two legs.

  Our humanity comes at a cost. For the exceptional combination of things we do—talk, think, grasp, and walk on two legs—we pay a price. This is an inevitable resul
t of the tree of life inside us.

  Imagine trying to jerry-rig a Volkswagen Beetle to travel at speeds of 150 miles per hour. In 1933, Adolf Hitler commissioned Dr. Ferdinand Porsche to develop a cheap car that could get 40 miles per gallon of gas and provide a reliable form of transportation for the average German family. The result was the VW Beetle. This history, Hitler’s plan, places constraints on the ways we can modify the Beetle today; the engineering can be tweaked only so far before major problems arise and the car reaches its limit.

  In many ways, we humans are the fish equivalent of a hot-rod Beetle. Take the body plan of a fish, dress it up to be a mammal, then tweak and twist that mammal until it walks on two legs, talks, thinks, and has superfine control of its fingers—and you have a recipe for problems. We can dress up a fish only so much without paying a price. In a perfectly designed world—one with no history—we would not have to suffer everything from hemorrhoids to cancer.

  Nowhere is this history more visible than in the detours, twists, and turns of our arteries, nerves, and veins. Follow some nerves and you’ll find that they make strange loops around other organs, apparently going in one direction only to twist and end up in an unexpected place. The detours are fascinating products of our past that, as we’ll see, often create problems for us—hiccups and hernias, for example. And this is only one way our past comes back to plague us.

  Our deep history was spent, at different times, in ancient oceans, small streams, and savannahs, not office buildings, ski slopes, and tennis courts. We were not designed to live past the age of eighty, sit on our keisters for ten hours a day, and eat Hostess Twinkies, nor were we designed to play football. This disconnect between our past and our human present means that our bodies fall apart in certain predictable ways.