Go back far enough in our history–maybe about 650 million years–and you come to a time when our ancestors were still invertebrates. That is, they had no skulls, teeth, or other bones. They didn’t even have a brain.
How invertebrates became vertebrates is a fascinating question, made all the more fascinating because the answer tells us something about how we got to be the way we are. In order to reconstruct what happened, scientists can study several different kinds of evidence.
They can look at the bodies of invertebrates to find the ones that share traits with vertebrates not found in other invertebrates. Those common traits may be signs of common ancestry. Scientists can look for signs of this ancestry by studying the DNA of vertebrates and invertebrates. They can also examine the fossil record, to discover transitional forms that offer clues to the transitions that can’t be found in living species.
When scientists consider this evidence, the answers don’t come pouring into their lap like coins from a slot machine. They have to put together hypotheses that do the best job of explaining vertebrate origins.They can then test those hypotheses against new evidence. Sometimes the old hypotheses hold up. Sometimes it turns out they were based on a misreading of the evidence. New hypotheses emerge to take the place of old ones. But those new hypotheses have to be better than the old ones. Scientists do not just suddenly declare that any explanation will do.
This is how science works. It occurred to me that this fact bears repeating when I read a recent comment to this blog. The comment emerged in a discussion about some observations from Randy Olson, a movie director, on how to communicate effectively about evolution. Some people agreed with Olson’s ideas on livening things up, while others worried that he just wanted to dumb science down. The differences were real, but the discussion was productive–at least until a couple creationists chimed in.
These creationists claimed to offer some helpful advice, but it’s pretty clear from their comments (and their links to creationist web sites) that they really had nothing of the sort in mind. They don’t actually want to help people understand evolution; they’re just exploiting some common misunderstandings about evolution, and about science in general.
They do provide one service, by showing which misundertsandings they think they can exploit most effectively. “Anonymous” writes:
“But you need to be honest about the scientific accuracy. How often does a new discovery come out that requires a rewrite of the evolutionary timelines and trees. If memory serves, I’ve read at least 2 posts on this blog in the last year with statements to that effect. How can you say Evolution is Science and Science is Truth and then in the next statement that Evolutionary scenarios must be rewritten. The public hears this.”
If only I could put words in the mouths of people I write about–life would be so much easier. Anonymous might as well have written, “You say that your cat can speak Mandarin. The public hears this.” Not from me.
Scientists learn new things about the world. They revise their theories. They do not pull away a curtain, to reveal Truth with a capital T, and walk away. And just because they do not deal in Truth with a capital T does not mean that they deal in pure nonsense. Their knowledge improves, although it never reaches perfection.
The origin of vertebrates is a case in point. Among living invertebrates, scientists have identified certain groups as being closer to us than to other invertebrates. The biggest group are echinoderms, which include starfish. They may not have skulls or brains like ours, but they do share some peculiar traits. In other invertebrates, for example, a hole in the early embryo called the blastopore becomes the mouth. In echinoderms and vertebrates, it becomes the anus.
Along with echinoderms, a few less familiar species have also shown strong links to vertebrates. One group is called the tunicates. These include the sea squirt, a truly bizarre animal. It begins life as a tiny tadpole. It swims with a tail made of a stiff rod called a notochord, along which runs a hollow nerve cord. It also has slits in its throat for swallowing food. These traits are not found in echinoderms or other invertebrates, suggesting a close link to vertebrates. Remarkably, most of its vertebrate-like traits disappear when it gets to be an adult. It lands on the sea floor on its head, rotates its organs ninety degrees, and eats its own nervous system. It then sits on the sea floor, filtering food and making new tadpoles.
Then there are the lancelets. These creatures (such as the one in the picture here) look like sardines with their heads cut off. They have a similar life cycle to sea squirts, swimming as larvae and then settling to the sea floor. But they have more traits in common with vertebrates. The tip of their nerve cord, for example, shows many striking similarities to the overall organization of the verebrate brain, even down to the genes that build each. It also has muscles arranged into blocks along its length–the same sort of blocks that you can get your fork into when you have fish for dinner. They don’t degenerate into a sac as adults. They just dig into the sediment and stick their heads out, so they can filter food passing by.
Many scientists argued that the ancestors of echinoderms branched off from our ancestors first. Then, after our ancestors acquired throat slits and a notochord, the ancestors of sea squirts branched off. Then our ancestors acquired rudiments of a brain, blocks of muscles, and other traits, after which the lancelet lineage branched off. At this point, the vertebrate body plan was still only partially built. Evolution had not yet produced skulls, spinal columns, eyes, and other features found in all living vertebrates. All that came later.
As I mentioned earlier, there are a couple ways to test this hypothesis. One is to look at the fossil record. Do the traits seen in echinoderms, sea squirts, lancelets, and vertebrates tend to turn up together in extinct animals, or do are they scattered all over the animal kingdom? Do paleontologists discover lobsters with skulls, earthworms with fish-like gills? No. They do find lancelet-like animals that have well-formed brains and even skulls–which is exactly what you’d expect as invertebrates evolved into vertebrates. They also find peculiar fossils that might be closely related to vertebrates or to the common ancestor of echinoderms and vertebrates. This sort of ambiguity is not surprising, because in the early days of these groups, you’d expect to find species that had not yet acquired the distinctive traits found in living groups. The same pattern turns up in the earliest hominids–they are so much like other apes that it can be hard to say whether they are ape-like hominids or hominid-like apes.
Another way to test an hypothesis about the origin of vertebrates is to look at DNA. In this week’s issue of Nature, a team of French scientists publish the results of just such a study–the biggest of its kind, both in terms of the range of animals studies and the mass of DNA analyzed in each one. They looked for the evolutionary tree that offered the best explanation for the differences and similarities in the DNA carried by each species. They put their results through a series of statistical tests to see if they were solid, or if they were just misleading illusions (removing one species from the study, for example, to see if a different tree emerged).
Their evolutionary tree has the overall shape of trees produced from other kinds of evidence. The French scientists find that humans and chickens are more closely related to each other than either is to frogs. Lampreys and hagfish are more distantly related to us, but closer than lancelets or echinoderms. Even among the invertebrates, patterns turn up that have been seen before. Insects are more closely related to us than jellyfish, for example, and jellyfish are more closely related to us than mushrooms.
The fact that studies based on anatomy, various genes, and fossils, all converge on these same patterns indicates that they’re onto something real. If someone shows you twwenty species and has you connect them with a family tree, there are over 200 billion billion different possible trees you could draw. But the trees that emerge from these studies only only minor variations on one another, not a random sample of all the possible trees. The odds of this happening by pure coincidence are incredibly tiny. But it’s exactly what you’d expect if they represented real evolutionary relationships.
The new tree does have a couple interesting surprises. One is that sea squirts–those weird nerve-eaters–are closer to us than lancelets. The other is that lancelets may actually be more closely related to echinoderms than to vertebrates and sea squirts.
If this conclusion is supported by more studies, lancelets may turn out to be a lot like the common ancestor we share with starfish. That lancelet-like ancestor then gave rise to some lineages in which some radical changes occurred. Once the echinoderm ancestors of starfish branched off, they must have lost some traits we still carry, such as a main nerve cord running along the back and throat slits. That conclusion would jive with some early fossils of echinoderms that appear to have slits.
The study suggests that sea squirts also changed a lot from a lancelet-like ancestor. They lost the muscle blocks of their ancestors and their nervous system must have become much simpler. Their odd sedentary filter-feeding evolved only after their ancestors branched off from our own. Rather than primitive, sea squirts may actually be highly specialized proto-vertebrates.
But as different as sea squirts may have become, scientists may still be able to learn a lot about our own origins by studying them. Much of what distinguishes us from invertebrates–eyes, teeth, and such–develops from a distinctive group of cells called neural crest cells. They first emerge along the back of the vertebrate embryos and then move through the body, giving rise to lots of different structures. Both lancelets and sea squirts have neural-crest-like cells. But only in sea squirts do they migrate as they do in our own bodies. By studying how these cells move, scientists may be able to understand a key step in vertebrate evolution.
There is one obvious way to test this scenario: sequence the genomes of other close vertebrate relatives. Previous studies have suggested that echinoderms share a close common ancestor with acorn worms, which have throat slits among other vaguely vertebrate-like traits. Including their DNA in a new study may support this current study, or pose a new challenge. That sort of result may disappoint those who would like science to deliver Truth in one perfect lump. But for scientists, it’s the sort of fresh challenge that gets them out of bed in the morning.
Update 2/23 8 am: Thanks to Dr. David Hone for giving me precise numbers for the possible trees.
Originally published February 22, 2006. Copyright 2006 Carl Zimmer.