The newest branch

A skull in the sand

On 19 July 2001, in the Djurab Desert of northern Chad, a Franco-Chadian team led by the French paleontologist Michel Brunet found something half buried in the sand at a site called Toros-Menalla: a skull about the size of a chimpanzee’s, with a short, flat face, small canine teeth, and a heavy brow. The rock around it was the margin of an ancient lake, and the skull was about seven million years old. Chad’s president gave it a nickname, Toumaï, which in the local Daza language means “hope of life,” a name for a baby born just before the dry season, when its chances of surviving are slim. The animal itself was named Sahelanthropus tchadensis, “the Sahel man of Chad,” and announced in Nature in 2002.

Is it a human ancestor? That is still argued about. The hole under the skull where the spine joins it sits well forward, as in an upright walker, which hints at a body held vertical; but studies of its thigh bone have given conflicting answers about whether it walked upright, and some scientists think it is an early relative of the apes. Its brain, about 378 cubic centimeters, was no bigger than a chimpanzee’s. What it surely shows is that, around seven million years ago, the family tree of the apes had already begun to branch, and that the branch that includes us has deep roots in Africa. The ninth turning point is the story of that branch.

Seven million years

The scientists who study this family call its members hominins: humans and our extinct relatives, since the split from the line of the chimpanzees, our closest living cousins, some six to eight million years ago. The chimpanzee is not our ancestor; we share an ancestor with it, which lived in Africa and looked like neither of us. We are apes ourselves, and the question to ask is what happened on our side of the split.

The oldest fossils are scattered and scanty, and they show a slow trade-off. By about 4.4 million years ago, Ardipithecus ramidus, nicknamed Ardi, from Ethiopia, had a grasping big toe for climbing trees but a pelvis that suggests it could walk upright. By 3.7 million years ago, hominins were walking upright beyond doubt: in Laetoli, Tanzania, a line of footprints pressed into volcanic ash records two or three of them walking side by side. Australopithecus afarensis is the best-known species of that age. In 1974, in the Afar region of Ethiopia, the American anthropologist Donald Johanson and a student, Tom Gray, found about 40 percent of the skeleton of one small adult female, now called Lucy: about 1.1 meters tall, with a chimp-sized brain and legs that show she walked upright. She is 3.2 million years old. The name comes from the Beatles song that was playing at the camp on the night of the discovery.

Then the line to Homo, the genus we belong to. Homo habilis, the “handy man,” named in 1964 from Olduvai Gorge, made simple stone tools, although stone tools older than any known Homo fossil have been found, so tool-making was not a human invention alone. By about 1.9 million years ago, Homo erectus had a larger brain and a body like ours, with long legs, and it was the first to leave Africa, reaching western Asia by 1.8 million years ago. In China, the best-known Homo erectus is Peking Man, found at Zhoukoudian, near Beijing: a skullcap was dug up in December 1929 by the Chinese scientist Pei Wenzhong. The fossils there span a long period, roughly 800,000 to 230,000 years ago, and the brain was about 1,000 cubic centimeters, about three-quarters of ours. The heavy brow and the jutting face, with no chin, look more ape-like than ours. In December 1941, as war came to China, the original bones were packed for shipment to the United States with a group of US Marines, and they vanished. Only casts and the scientists’ notes survive.

Our own species, Homo sapiens, is at least 300,000 years old: the oldest known skulls, from Jebel Irhoud in Morocco, are about 315,000 years old, and the species spread over Africa before it spread over the world.

Who shared the earth

One of the great surprises of the last twenty years is how crowded the human family was. If you go back 50,000 years, you would not meet one kind of human, but several. The Neanderthals lived in Europe and western Asia until about 40,000 years ago. The Denisovans, known first from a finger bone in Siberia’s Denisova Cave in 2010, left DNA, teeth, and jaws across Asia. In 2025, scientists identified the “dragon man” skull from Harbin, in northeastern China, more than 146,000 years old, as a Denisovan, using ancient proteins from the skull and DNA preserved in the hardened plaque on its teeth; a jaw from the seabed off Taiwan was identified the same year by its proteins. Homo floresiensis, the “hobbit,” lived on the Indonesian island of Flores until perhaps 50,000 years ago, and Homo naledi, a small-brained human relative about 1.4 meters tall, was found in 2013 in a South African cave and has been dated to about 335,000 to 236,000 years ago.

They were not strangers. When Homo sapiens left Africa, our ancestors met Neanderthals and Denisovans and had children with them. Most people alive today outside Africa carry about 1 to 2 percent Neanderthal DNA, and some Pacific islanders carry up to about 5 percent from the Denisovans. A gene variant that helps Tibetans live above 4,000 meters, where the air is thin, came from the Denisovans. The best-supported model is “recent African origin, with some interbreeding”: modern people arose in Africa, spread outward, and mixed a little with the humans they met. Not everyone agrees about the details, as the “Another view” box explains.

No mermaids

The title of this course may have startled you: people usually say “from ape to human,” not “from fish to human.” Is there a conflict? There is none, and Darwin explains why. In On the Origin of Species (1859), he argued, in broad terms, for three things: that species change over time; that the main force driving the change is natural selection, which favors the animals best suited to their surroundings; and that all living things share common ancestors, a single family tree. He said almost nothing about humans in that book, only that “light will be thrown on the origin of man and his history,” and waited until 1871 to write about us.

If all life is one family, you can trace any animal back along it. For a human the line runs back through apes, monkeys, early mammals, the first amniotes, the first land animals, and the first jawed fish, to the first vertebrate. “From the ape” and “from the fish” are the same story told at two different depths. A paleontologist who studies vertebrates stops at the fish because the vertebrates begin there. A biologist who studies the first cells could go back another three billion years. There is no mermaid in it, because the family tree never needed a half-human, half-fish. It needed only a very long time.

Not a ladder

The most common mistake about this story is to read it as a ladder, with a fish at the bottom and a human at the top, each rung more advanced than the last. You have seen the picture: a line of figures marching across a page from a crouching ape to an upright man. One famous version, “The March of Progress,” was drawn for a Time-Life book in 1965. It was a tidy way to show a series of species, but it taught generations that evolution is a march. The arrangement is a trick of the page: the real hominin tree has many branches, and at times several of them were alive together.

Three things follow from the bush. First, no living species is “more evolved” than another: a bacterium, a shark, and a person have all been evolving for the same 3.5 billion years. Second, none of the nine turning points was a step toward us. The backbone, jaws, legs, eggs, fur, and feathers were each useful at the time, for eating, escaping, and breeding; what they led to later is something nobody was aiming at. Third, when we call a fish “primitive,” we really mean “like our ancestors”; sharks have lasted over 400 million years because they are very good at being sharks.

Was the arrival of people at least likely? Here scientists disagree, and it is worth knowing both sides. The American paleontologist Stephen Jay Gould argued in Wonderful Life (1989) that history is full of accidents. If you could replay the tape of life from the Cambrian, the results would almost certainly differ, and the chance of anything like us would be small: change one extinction, and there are no mammals; change one mutation, and there is no thumb. The British paleontologist Simon Conway Morris answered in Life’s Solution (2003) that evolution keeps arriving at the same solutions, as with eyes, which have evolved dozens of times, or the dolphin and the ichthyosaur, which look alike though they are unrelated. If so, an intelligent, tool-using animal might be a likelier outcome than it seems. There is no way to run the experiment, and the question is open.

Five endings

The nine turning points all give new structures and new ways to live. They leave out half the story, which is the endings. The fossil record shows five great mass extinctions, each wiping out most species alive at the time, and each followed by a new world.

The biggest, 252 million years ago, at the end of the Permian, was probably set off by huge volcanic eruptions in Siberia that heated the planet and starved the oceans of oxygen; most marine species disappeared, by some estimates more than eight in ten. The famous one, 66 million years ago, killed about three-quarters of species, fewer than the Permian, but it is the one that mattered most to our story: the asteroid that ended the dinosaurs. Without it, mammals and perhaps we would have had a harder time. The point to take is that evolution is not a smooth climb. It is a story of growth and collapse, and the survivors of each collapse are often not the animals that were winning the day before.

Are we in the middle of a sixth? Scientists differ about the word. Extinctions today are running far faster than the normal background rate; some estimates put them at more than a thousand times higher than before humans. The dodo, a flightless pigeon on the island of Mauritius, was gone within a century of the arrival of Dutch sailors in 1598. Passenger pigeons, which once darkened the skies over North America in the billions, ended in 1914 with Martha, who died in the Cincinnati Zoo. The IUCN Red List now counts nearly 50,000 species as threatened among the 170,000 or so it has assessed, about 28 percent. Yet the mass extinctions of the past wiped out about three-quarters of all species, and we are not there yet. The best summary is that we are in an extinction crisis, and that whether it becomes a sixth mass extinction depends on what people do.

Has evolution ended?

The last question is the one that people most often ask: are humans still evolving? Yes. Evolution does not stop; it stops only when a species stops reproducing with variation. In the last 10,000 years, human populations have changed in measurable ways. People whose ancestors kept cattle, sheep, or camels in Europe, Africa, and the Middle East, evolved, several times independently, the ability to digest milk sugar as adults, a trait called lactase persistence, while most adults in East Asia and in many other parts of the world still lose the enzyme after childhood, which is the ancestral condition and not a disease. Tibetans carry a variant of the gene EPAS1 that lets them live at high altitude without the thickening of the blood that afflicts newcomers, and it came from the Denisovans. The Bajau, sea nomads who free-dive for fish in Southeast Asia, have spleens about 50 percent larger than their neighbors, which stores extra oxygen-rich blood.

The model shows how little time selection needs. A variant with a 5 percent advantage can go from 1 percent of a population to 90 percent in under 150 generations, about 3,500 years at 25 years a generation, which is less than a second on the day-clock of Lesson 1. But humans differ from every other animal in the second half of the answer. We do not evolve in only one way.

Some scientists describe the human story as three streams. The first is biological evolution, the one we share with all animals: genes that change by mutation and selection, slowly. The second is technological. Ten thousand years ago our ancestors hunted mammoths with spears; the Wright brothers flew in 1903; the first electronic computers date from the 1940s; and today artificial intelligence and gene editing may reshape us faster than any gene could. The third is cultural and ethical. In the wild, an animal that cannot look after itself usually dies. We care for our elders and treat sick children, because our culture holds that this is right. These choices change who survives and who has children, though not through genes alone.

One simple way to sum up an animal’s life is four tasks: find food, outdo rivals, avoid danger, and have offspring. Natural selection works on all four. But it also rewards cooperation, in wolves, in bees, and in us, and one leading idea is that our large brains grew in part because living in groups demands understanding other people. In 2018, a Chinese researcher announced that he had edited the genes of newborn twin girls; scientists around the world condemned the experiment as reckless, and a Chinese court sentenced him to prison. The episode shows that our future evolution may partly be a human decision, a position no earlier species has been in.

So has evolution ended? No. Our bodies will keep changing slowly, our tools and ideas quickly, and what we become depends partly on choices not yet made.