WHEN VALENTINO GANTZ first heard about CRISPR, it sounded like a godsend. At the time, he was getting his PhD at the University of California, San Diego, studying genes in Drosophila and related flies. He tinkered with their genes and observed whether he could change how their embryos developed. But the best tools he could use were clumsy and crude. In 2013, Gantz heard that researchers had figured out how to use CRISPR to alter a gene in Drosophila with easy precision.
“It was one of the things I was waiting for,” Gantz told me when I visited him at his laboratory on a eucalyptus-covered hillside by the Pacific. After hearing the news, he had immediately ordered CRISPR molecules of his own and started trying them out. He had no idea he was about to discover a way to use CRISPR to alter heredity on a species-wide scale.
Gantz decided to try out CRISPR by altering a Drosophila gene called yellow. It was, in a sense, an inherited choice. The yellow gene was discovered just over a century earlier in the lab of none other than Thomas Hunt Morgan. One day in 1911, Morgan’s team of students were inspecting their grayish flies when they spotted a single golden insect. They bred the fly and determined that it had a recessive mutation in a gene they dubbed yellow.
The yellow gene proved useful to Morgan’s team, because they could see with the naked eye which allele of the gene any given fly carried. Morgan’s students bred a line of yellow flies, and when they became professors years later, they taught their own students how to breed the flies for experiments. Those students started their own careers, and took the yellow flies with them. Over the twentieth century, each new generation inherited this knowledge, just as earlier generations had learned how to make stone tools and how to plant barley. There’s even a website, called FlyTree—The Academic Genealogy of Drosophila Genetics, that chronicles this particular line of cultural inheritance.
The entire tree begins, of course, with Thomas Hunt Morgan. It branches down to his many graduate students. Their ranks included Max Delbrück, a physicist who turned to the mysteries of life, traveling from Germany to Caltech in 1937 to study under Morgan. Delbrück went on to become a professor at Caltech himself, and among his own graduate students he trained another physicist turned biologist, Lily Y. Jan. Jan got a job at the University of California, San Diego. In the 1980s, Ethan Bier joined her lab as a postdoctoral researcher. There he bred countless yellow flies, and became a professor at San Diego as well. Two decades later, Valentino Gantz arrived in Bier’s lab and learned the yellow craft. He became Morgan’s scientific great-great-grandchild.
To try out CRISPR, Gantz fashioned an RNA guide to alter the yellow gene in Drosophila embryos, introducing a mutation to make them golden. He added the molecules to the fly cells, let them develop into adults, and bred them together, hoping to produce flies with two CRISPR-altered copies of the yellow gene. To his delight, among the grayish flies, there were some golden ones. The technology had worked just as advertised. “I was sold,” Gantz said.
Gantz then began playing around with the CRISPR molecules to see if he could use them on another species of fly he was studying for his PhD, called Megaselia scalaris. Unlike Drosophila melanogaster, it has a distinctively hunched thorax, which has earned it the common name of the humpbacked fly. It behaves differently, too, for which it has earned other names. It’s known as the scuttle fly for the way it runs along the ground in bursts. And it’s sometimes called the coffin fly for the way its maggots dig deep into the ground in search of food, sometimes traveling all the way down into buried caskets.
Gantz tailored the Drosophila CRISPR molecules so that they would target the yellow gene in humpbacked flies. But when he used them on the flies, the experiment failed. “It was very frustrating that we weren’t recovering any mutants,” said Gantz.
Gantz wasn’t sure what went wrong. It was still possible that he had managed to edit one copy of the yellow gene in a few flies. But the CRISPR molecules were succeeding so rarely that Gantz could never bring together two flies that both carried an altered copy of the yellow gene.
When Gantz told Bier the news, his advisor was disappointed but not surprised. Failure is common in science. Bier headed off on a much-needed vacation in Italy and didn’t give the disappointment any more thought.
But as soon as Bier stepped back into his office in San Diego, Gantz bounded in. He immediately started describing an idea for how to get CRISPR to work. The idea was simple. Gantz would get the flies to edit their own DNA.
First, Gantz would use CRISPR molecules to chop out a gene, which he would replace with a segment of DNA. That segment would include not only an altered gene but genes for CRISPR molecules, too. The fly’s own cells would then make CRISPR molecules that would seek out the matching chromosome and edit the second copy of the gene. Gantz would then have humpbacked flies with two mutated copies of the same gene. He could then use them to start a line of mutants. Gantz dubbed the process a “mutagenic chain reaction.”
The idea seemed far-fetched to Bier. He doubted it would work reliably enough to deliver the genetic changes Gantz hoped for. But if it did work, Bier could tell, it might become a powerful tool for studying genetics not just in humpbacked flies but in many other species as well. And as Bier and Gantz talked about the mutagenic chain reaction more, it occurred to them that it might be more powerful than Gantz had initially thought.
“We realized, ‘Whoa, this could go through the germ line,’” Bier told me.
If Gantz mated a CRISPR-carrying fly to an ordinary fly, it could break Mendel’s Law. It would pass down one of its chromosomes that carried its genes for CRISPR, along with the gene that the CRISPR molecules were designed to copy. In the embryos of the second-generation flies, the CRISPR molecules would rewrite its partner chromosome. The result would be that the second generation would not be hybrids, each with a single copy of the CRISPR genes. They would all carry two copies. And the result would be the same when they bred with ordinary flies, too.
“Imagine that a blond person married a dark-haired person, and all their kids were blond,” Bier told me. “All their grandkids were blond. All their great-grandkids were blond, and that went on forever. Imagine something like that.”
Bier told Gantz to hold off on testing the mutagenic chain reaction for a while. He wanted Gantz to reflect on the risks and benefits first. In his mandatory reflection, it occurred to Gantz that if a single altered animal got into the wild—either intentionally released or accidentally allowed to escape—it could mate with other members of its species. Its CRISPR genes could drive themselves further into a population with every new generation.
Under the right conditions, that might be a good thing. Instead of targeting the yellow gene in Drosophila, scientists could target genes in insects that destroy crops or spread disease. Other researchers had been searching for years for gene drives that might fight pests, and Gantz may have found one at last. But if an animal slipped out of a lab while Gantz was still doing basic experiments on the mutagenic chain reaction, he had no idea what sort of unplanned changes he might unleash.
Gantz devised a way to safely test the mutagenic chain reaction. He would attempt to edit the yellow gene in Drosophila. But he would prevent the flies from sneaking out of his lab by working in a secure room, housing the insects in shatterproof plastic vials sealed with tight plugs, which would go inside larger tubes, which would be sealed in turn inside plastic boxes.
Gantz and Bier invited a group of senior geneticists at the university to hear them out. They described their concept of the mutagenic chain reaction, and their plan for a safe experiment. They wanted to know if it sounded crazy to someone else.
“‘Yeah, do it’” was the consensus, Bier recalled. “‘Be careful, but do it.’”
In October 2014, Gantz used CRISPR to modify some Drosophila larvae. If the molecules worked as he hoped, they would replace one copy of the yellow gene with a different stretch of DNA. That new stretch carried an altered version of the gene, along with genes for CRISPR molecules. Gantz hoped that the fly’s own cells would then use its genes to alter its other copy of yellow. But the only way to know for sure if the procedure worked would be to breed the flies.
Gantz let the flies mature and then bred them with ordinary mates. The female flies laid their eggs, which grew into larvae and then built cases called pupae around themselves. Inside the pupae, they matured into adults. And when the mature flies broke out at last, some of them—both males and females—were golden.
“This was the first indication that things were going the right way,” Gantz recalled.
But now came the real test of the mutagenic chain reaction: Could it carry over to the next generation? Gantz picked out golden females, which carried the altered yellow gene and its CRISPR package on both of their X chromosomes. He put them in tubes with ordinary male flies and waited for them to mate. In a conventional experiment with golden females and gray males, the results would reliably follow Mendel’s Law. Their sons would all be golden, because they always inherited their one X chromosome from their mother. Their daughters, on the other hand, would inherit an X chromosome from each parent. And since yellow is recessive, the daughters would all be gray.
If the mutagenic chain reaction worked, Gantz would see a very different result. The next generation of flies would also make CRISPR molecules. The molecules would alter the second X chromosome in the daughters, and they would turn out golden instead.
Once the flies laid their eggs, Gantz and Bier could only wait for them to mature into adults. “For the whole two weeks I drove my wife crazy, saying ‘yellow-yellow-yellow’ and knocking on wood everywhere,” said Bier. His colleagues prepared him for failure. “Don’t get too excited,” one of Bier’s colleagues told him. “I’ll bet you almost everything that in the next generation it’ll just be Mendelian.”
Gantz knew that when the eggs of yellow mutants hatch, the larvae sometimes take on a faint golden cast. He squinted through his boxes and tubes, hoping to glimpse it on some of the maggots. But as best as he could tell, none of them looked yellow at all.
“I told everyone, ‘Okay, this is not working,’” said Gantz. “I’m a very pessimistic person.”
Just to be thorough, though, Gantz let the larvae pupate and unfurl their wings. He gassed his newly adult flies with carbon dioxide to put them to sleep. Then he sat down at a lab bench and dumped them out onto a pad. He would inspect their bodies before declaring the experiment an official failure.
But when Gantz looked down at the pad, all he saw was gold. The females had converted their own genes as he had hoped. When he stepped into Bier’s office to deliver the news, Bier screamed and jumped in the air. Bier had bred thousands of yellow mutants, watching Mendel’s Law work every time. Now suddenly the rules had changed.
“You walk into a room, you’re normally used to walking on the floor, and all of a sudden you’re walking on the ceiling,” said Bier. “I mean, that’s how weird it was to me.”
One of the first people Bier called about the experiment was a biologist named Anthony James. “Holy mackerel,” James replied.
He had been searching for twenty years for what Bier and Gantz had just found. In the 1980s, James had set out to fight diseases carried by mosquitoes, such as malaria and dengue fever. He would wage his personal war by studying mosquito genes. James set up an insectarium not far from Bier and Gantz, at the University of California campus in Irvine. There he raised mosquitoes on warm blood and experimented on their DNA.
James started by mapping their genes, since the mosquito genome at the time was terra incognita. As James and his colleagues pinned down the location of individual genes, he would be able to experiment with them. Perhaps there were genes that controlled exactly which pathogens can survive inside mosquitoes and use the insects to get to a new victim. Over 200 million people each year develop malaria because they’re bitten by mosquitoes that carry the single-celled parasite Plasmodium. But no one has ever gotten the flu from a mosquito bite. James wondered if there might be genetic variants in certain mosquitoes that made them resistant to malaria.
“If we could just figure out how to get those genes out there in the populations at high enough frequencies, then—you know, game over,” James told me when I visited him in Irvine.
James and his colleagues succeeded in mapping some mosquito genes. But the work was so slow—thanks in part to the challenge of raising blood-sucking mosquitoes—that he began to despair of ever finding a way to fight mosquito-borne diseases. He thought about how he could use what he had learned, to fight them in a different way.
“I thought, ‘Well, we’ll just make genes,’” James said.
He had an idea of the right gene to make. In the late 1960s, Ruth Nussenzweig, a biologist at New York University, discovered that mice don’t get sick with human malaria. She found that the immune cells in the mice produce an antibody that clamps onto the parasite, essentially suffocating it. In later experiments, scientists fed Nussenzweig’s antibody to mosquitoes, mixed in their meals of blood. Somehow, the antibody escaped being digested inside the mosquitoes and attacked the parasites inside the insects. After this treatment, the mosquitoes couldn’t transmit malaria.
James and his colleagues set out to genetically engineer mosquitoes to make the mouse antibody for themselves. The scientists created a gene that encoded the antibody, which they could insert into a mosquito’s DNA. James wanted to make the gene safe for the mosquitoes. He worried that if the gene stayed on all the time, the insects would swell up with mouse antibodies and get sick from them. So James and his colleagues connected the antibody gene to a switch. Now the gene would turn on only in response to blood coming into a female mosquito’s body.
James and his colleagues inserted the gene and its switch into the DNA of mosquitoes. When they fed the insects Plasmodium-laced blood, they started making antibodies and wiped out the parasites.
Impressive as this feat might be, it wouldn’t put a dent in the worldwide burden of malaria. If James simply released some of his malaria-proof mosquitoes in Africa or India, Mendel’s Law would work against him. The engineered mosquitoes would almost always mate with ordinary mosquitoes, and soon their defenses would get diluted in the vast mosquito gene pool.
To spread his malaria-fighting genes, James would need a way to break Mendel’s Law. Since the 1960s, scientists had wondered if they could harness gene drives for this purpose. The idea was simple: Link your gene to a gene drive, and with each generation, it would become more common in a population. But no one had yet figured out how to make it work.
James decided to take a crack at the problem. For his gene drive, he picked a piece of DNA called the P element. Carried by Drosophila flies, it spreads by occasionally getting its host cell to make a new copy of itself, which gets inserted at another spot in the fly’s DNA. The P element first came to the attention of scientists in the mid-1900s, and over the next few decades it spread like genetic wildfire in North American flies. Margaret Kidwell, a biologist at Brown University, put flies with the P element into tubes with flies that lacked it. Within ten or twenty generations, all her flies carried it.
James and his colleagues thought that if they link their antibody gene to a P element, it could drive resistance to Plasmodium into an entire population of mosquitoes, and they would keep passing it down to later generations. That was the idea, at least, but no matter how the scientists adjusted the experiment, it never worked.
In hindsight, James thinks that evolution worked against him. Mosquitoes, like other animals, have probably faced many attacks by gene drives in their long history. And they escaped extinction only because they evolved defenses—defenses that James had no idea how to overcome. “We were probably doomed at the outset on that one,” he told me with a laugh.
When James first heard about CRISPR, he was intrigued. He wondered if he might be able to adapt it to block malaria. But James didn’t get much further than these idle thoughts, when Bier called him. James immediately recognized that the mutagenic chain reaction might be able to spread genes in mosquitoes.
Bier and Gantz drove north up the California coast an hour and a half to visit James and plan out a new experiment. Plastic boxes and shatterproof tubes might be good enough to keep Drosophila in check, but for mosquitoes, they’d need much better security. James would need to run the experiment in the safety of his insectarium.
As excited as James was, he knew the odds of success were slim. He and Bier and Gantz would have to design a long piece of DNA containing a number of genes. The DNA would have to carry a gene for the mouse antibody, as well as its switch to turn on during blood meals. It would also have to carry the CRISPR genes to copy all that DNA to other chromosomes. In order to see if they succeeded, the scientists would also need to add a gene that turned the mosquitoes’ eyes red.
That was an awful lot to load onto one piece of DNA, especially when the scientists were trying to engineer mosquitoes in a manner no one had ever tried before. James proposed to his colleagues that they break the DNA into smaller chunks. They could then test one chunk at a time, and only later combine them for a final test. Gantz pushed for them to test the whole thing at once. He didn’t want to crawl forward when he might leap. When James and Bier agreed to the plan, Gantz worked as fast as he could to assemble all the parts into a single piece of DNA. He then delivered it to James to see if it worked.
During my visit, James took me to the basement of his building, and we walked up to a door marked A. JAMES TRANSGENIC MOSQUITO FACILITY. He pulled the sleeves of a blue paper smock over his flannel shirt. James had short gray hair, bright big teeth, and broad shoulders that made it hard for him to get the smock on. After a little struggle, he gave up mid-humerus. The smock looked like a newspaper that had slapped against his chest on a windy day.
James waited for me to slide my smock on, and then opened the door. We stepped into a windowless vestibule. A silvery mesh—fine enough to block a mosquito—covered the ceiling vents. Once we had both stepped into the vestibule, James closed the outer door and waited a moment until he could open the inner one. We entered the insectarium—a small cluster of rooms in which James and his staff raise tens of thousands of mosquitoes.
The first thing I saw when we stepped inside was a row of yellow movie theater popcorn buckets. They sat on a table, each with a gray cylinder attached to the top, trailing a power cord. When I leaned over the buckets, I could see adult female mosquitoes clinging to the inner walls. Each had a swollen belly. The gray cylinders contained warm calf’s blood. James unscrewed an empty cylinder to show me how he lined it with a membrane. The mosquitoes could pierce the membrane with their needlelike mouthparts and drink deeply, filling their tear-shaped abdomens with blood.
Once a mosquito is sated on blood, she will grow hundreds of eggs inside her body. James and his staff of technicians have found that the mosquitoes prefer to lay their eggs in the dark, and so they transfer the insects from the popcorn buckets into a lightless room. After the mosquitoes are done laying, James’s technicians gather the eggs and attach them to strips of paper.
Now they can alter the genes of the new generation of mosquitoes. They pierce the soft eggs with fine glass needles, injecting DNA. They have only a few hours to work on the mosquitoes before the eggs tan and harden.
“It’s like an assembly line,” James said as he showed me the microscope where his team manipulates the DNA inside the eggs. “You can do three or four hundred, maybe five hundred a day, depending on how good you are.”
In the wild, a mosquito lays her eggs in water, where they float together like a raft. After a few days, the eggs hatch, and the larvae swim away to spend their first stage of life underwater. James has to re-create that stage in his insectarium, too. He led me through a transparent vinyl-strip door into his larva room. Metal shelves reached from the floor to the ceiling, and on many of them were plastic tubs half-filled with water. Each was like a pond filled with hundreds of larvae. They swam about like hairy miniature snakes.
We stopped to look closely at one tub of larvae. There was a tag taped to the rim, with the number 29 written in thick marker ink. The larvae twirled and twitched. Each one had a pair of pinprick-size eyes. All the eyes were red.
“So these are the famous gene drive ones, here,” James said.
The number 29 referred to the twenty-nine generations of mosquitoes James had reared since beginning his experiment with Gantz and Bier. After Gantz created a new piece of DNA with all the gene drive pieces, James and his team injected them into the soft eggs of mosquitoes. To his delight, the larvae hatched with red eyes, meaning that they carried two copies of the malaria-resistant gene. James and his colleagues mated the male larvae with ordinary females, and the following generation was red-eyed as well. I was now looking at generation twenty-nine, at the end of an unbroken chain of heredity.
In November 2015, James and his colleagues announced that they had successfully used gene drive in mosquitoes, just seven months after Gantz and Bier had published the original mutagenic chain reaction paper. “People said, ‘That went fast!’” James told me as I looked at his red-eyed mosquitoes. “Well, it didn’t really go fast, because we’d been laboring for years.” James had all the tools he needed for the experiment by the time Bier called him. “It was just another piece of DNA for us to inject,” he said.
The mutagenic chain reaction hit the news amidst jolting stories about experiments with CRISPR on human embryos. Human genetic engineering had been the stuff of speculation for more than fifty years, since Rollin Hotchkiss had worried over it. But the idea of overriding heredity to cure diseases with gene drive came as a bigger surprise. Even most scientists who worked on CRISPR hadn’t seen it coming.
There were exceptions: George Church and one of his colleagues at Harvard, Kevin Esvelt, had been musing about the idea. In 2014, they and some of their colleagues published a couple of speculative pieces. But they called CRISPR-based gene drive only a “theoretical technology.”
Once Bier and Gantz revealed the mutagenic chain reaction, the technology was no longer theoretical. Esvelt and his colleagues reported that they could use CRISPR in yeast to override Mendel’s Law as well. The technology might conceivably work in just about any sexually reproducing species scientists might want to alter.
As Jennifer Doudna and her colleagues grappled with CRISPR’s use on people, Bier, Gantz, Esvelt, and other scientists began working through the implications of gene drives. At conferences and in scientific reviews, they laid out some of the ways the technology could make life better. Making mosquitoes malaria-proof could save thousands, or even hundreds of thousands, of lives every year. Esvelt traveled to Nantucket to propose to its residents that he use CRISPR on the island’s mice. It might be possible to render them resistant to Lyme disease, breaking the disease’s cycle. Plant scientists speculated about fighting the evolution of herbicide-resistant weeds. They could use gene drive to eradicate the genes that made them resistant, replacing them with genes that make the plants vulnerable once more.
It might even be possible to use CRISPR to drive a population, or even an entire species, extinct. Scientists could give genes to an undesired animal that made it less fertile. The animals inheritng these genes would have fewer offspring, but thanks to CRISPR, they would end up in a growing fraction of the population. Eventually, the population would cross a tipping point and collapse.
Conservation biologists had long dreamed of this kind of power to fight invasive species. When snakes and rats are introduced to remote islands, for example, they can wipe out local bird species by eating their eggs. A team of Australian scientists calculated that the introduction of a hundred CRISPR-altered rodents to an island could wipe out a population of fifty thousand. It would take only five years.
But gene drive might also wreak havoc. If scientists unleashed a gene drive in the wild, it might not work as they had planned. If it caused harm, it might be impossible to undo the damage. A committee organized by the National Academy of Sciences issued a report in 2016 in which it warned that gene drives had the potential to cause “irreversible effects on organisms and ecosystems.”
Artificial gene drives represent a profound ethical quandary—arguably a bigger one than using CRISPR to genetically engineer human embryos. They may be able to alter heredity not just in the genetic sense of the word but in other senses as well. We might drastically alter the genes that animals or plants inherit far into the future. We would also leave an ecological inheritance to our descendants that they might curse us for. To judge the wisdom of this tool, we would do well to look back at how the tools we’ve already invented have altered our ecological inheritance over the past ten thousand years.
Human cumulative culture allowed hunter-gatherers to learn over generations how to harvest plants and control animals. Mostly without knowing it, some of them engineered new environments where agriculture could arise. Their descendants became farmers, sowing crops and raising livestock. Each new generation inherited more than just the knowledge required to farm. Humans now left an ecological inheritance to their descendants.
Before about ten thousand years ago, children were born into a world sculpted by fire, hunting, and foraging. Farmers began reworking the land on a greater scale, and at an accelerating pace. By clearing fields to plant, they could grow enough food to feed their families, with surplus they could sell to others. Farmers stopped moving, settling into villages with sturdy houses and granaries where they could store their extra food. A farmer could now pass down this accumulated wealth to his children, along with the land that he used to generate it.
But this new form of inheritance created an inescapable tension: A family’s land could be divided among the children in small portions, or passed on to just one of them, leaving the rest to find other kinds of work. That tension drove some members of the family to find more land to clear. It also drove them to discover and adopt new cultural practices that let them get bigger harvests out of a given parcel, such as plows drawn by horses or oxen. By the Bronze Age, kilns were invented. Their fires reached temperatures humans had never managed before. Miners could smelt ores, and smiths could work metals. They discovered that coal was a better fuel than wood. Out of these intense fires came new metal tools, including axes that farmers could use to clear more forests, and plows to plant more crops.
Yet these advances did not free farmers from the feedback loop between their culture and the environment. The short-term benefits they got from new farming equipment came at the long-term expense of the land’s fertility. As fields eroded and became less productive, people cleared forests to work soils that would have previously been considered too poor to bother with. This feedback continued to raise the world’s population, fostering more cultural innovations. And those innovations allowed people to convert even more wildlands to farms and cities.
The Industrial Revolution, which came about ten thousand years after the Agricultural Revolution, was an acceleration of this feedback. Instead of using animal-drawn plows, farmers could run tractors powered with new fuels like gasoline. Instead of spreading manure from their own livestock, they could spread fertilizers extracted from mines or produced from petroleum. The cotton gathered by New World slaves no longer had to be woven by people; it was now turned over to coal-powered looms. As railroads cut across continents, cattle could be grazed on lands thousands of miles from the people who would ultimately eat them. The influence of human culture now produced a worldwide ecological inheritance.
By some measures, this cultural feedback loop has been a great success. Before the Agricultural Revolution, a square kilometer of land could typically feed fewer than ten hunter-gatherers. Today, if it’s intensively farmed, it can feed thousands. In the early 1800s, more than 90 percent of the world lived in extreme poverty, scraping by on the equivalent of about two dollars a day. Today, less than 10 percent are. A child born in the United States in 1900 had an average life expectancy just short of fifty years. Children born in 2016 will live, on average, to age seventy-nine.
I feel fortunate that my children get to inherit this world created by cumulative culture. But I can also see that they inherit an environment that is suffering in many respects. Since the dawn of agriculture, three-quarters of the terrestrial biosphere has been converted from wilderness. Somewhere between a quarter and a third of the planet’s biological productivity—its ability to convert sunlight into biomass—is now appropriated for human use. If the same cultural practices that have reworked the planet so dramatically over the past ten thousand years are inherited by future generations, we may push many species to extinction and threaten our own well-being.
Our cumulative culture has even altered the atmosphere. We’re not the first organisms to change the chemistry of the air—photosynthetic bacteria began pumping oxygen into the sky two billion years ago, and every generation of living things has had to adapt to an oxygen-rich planet ever since. But it’s unheard-of for just one species of animal, using tools of its own making, to manage such a feat.
When hunter-gatherers set fire to meadows or forests, they could loft carbon dioxide and other molecules into the air. Because their numbers were so low, early humans barely nudged the atmosphere’s makeup. But once farmers started clearing land for planting, the soils released carbon dioxide at a greater rate. By three thousand years ago, mining operations were belching particles of lead and other pollutants into the air. The traces of this pollution are trapped in Bronze Age layers of ice in Greenland.
The same forces that led to the destruction of most wild land on Earth also polluted the air. By the Industrial Revolution, the pollution had become so thick in cities that it cut millions of lives short. As people began burning coal, oil, and gas, they also flooded the atmosphere with so much carbon dioxide that it began trapping extra heat—enough to raise the average temperature of the entire planet. By the early twenty-first century, humans had raised the level of carbon dioxide in the atmosphere to its highest level in millions of years. In response, the planet had warmed about 2 degrees Fahrenheit since 1880.
Much of the pollution that humans put into the atmosphere washes out swiftly. The lead-laced fumes from gasoline, for instance, disappeared soon after it was barred. Carbon dioxide is different. It hangs in the atmosphere for centuries, still trapping heat and warming the planet. If tomorrow we cut our emissions of carbon dioxide to zero, the planet would keep warming another degree or more. Future generations would inherit a planet with endangered coastlines, increased wildfires, and farmland threatened with drought.
Three million years ago, the bipedal apes who were teaching each other how to break apart rocks were a small part of a vast ecosystem. But the open-ended power of cultural heredity, transferring knowledge across generations, gave humans the power to rework the land into its own ecological inheritance, which has now led to a climatic inheritance.
Today, we sense that we are close to being able to alter human heredity,” David Baltimore declared at the international gene-editing meeting in 2015. He was speaking shorthand, one that an audience at a meeting about CRISPR intuitively understood. To them, human heredity was the transmission of genes from human parents to human children. And to them, the looming ability to alter it was a new chapter in human history to be entered with awe and fear.
We certainly need to come to a collective decision about using CRISPR on human embryos, to use it only in ways that help people without creating serious dangers of their own. But this shorthand about heredity poses dangers, too. We risk coming to see ourselves as merely the product of the genes we inherited from our parents, and the future as nothing more than carrying those genes forward. The prospect of altering genetic heredity becomes wildly thrilling or terrifying. Soon no one will suffer from a genetic disease again, we’re promised. Soon China will breed an army of supergeniuses, we’re assured. This shorthand makes it hard to think clearly about genetic heredity. It leads us to overvalue our ambiguous knowledge of how genes work and dismiss the other factors that shape our lives—and could be reshaped to improve the world.
None of this is to say we should dismiss the power of heredity, or shy away from altering it. Instead, we can switch from a shorthand to a longhand. Thinking about heredity more broadly could lead plant scientists to better crops, for example. The first plant breeders manipulated the genetic makeup of crops, picking out good plants to cross to make better varieties. In recent decades, plant breeders have become more aware of the genes that their crops inherit. The epigenetic side of plant biology has only started to emerge, and some plant breeders are starting to investigate how they can tinker with it to improve crops even more.
Plants sometimes naturally change their epigenetic profile. The methylation that decorates its DNA may get altered, a methyl group falling away from a gene, for example. That change may awaken the gene and improve a plant’s growth. Scientists are searching for such changes and trying to propagate the plants so that the new generations inherit the same epigenetic profile.
Transgenerational epigenetic inheritance is a real phenomenon in plants, but many scientists are skeptical that it matters very much in nature. It’s wrong to call it Lamarckian, because Lamarck had a very different vision for the inheritance of acquired characters. He thought that inheritance could produce intricate adaptations. Skeptics like Robert Martienssen see little evidence for such adaptations in wild plants.
Yet that doesn’t mean such adaptations are impossible. In fact, Martienssen told me, he thinks we know enough now about epigenetics to try to build one.
Martienssen can imagine a plant that could respond to disease outbreaks by turning on immune defenses, and then pass down RNA molecules to their offspring to keep those defenses turned on. If, over the generations, the disease faded away, the plants could shut the resistance genes down so that they didn’t have to use their energy to make proteins they no longer needed.
“We could easily engineer a plant to be epigenetically adaptive—to be Lamarckian,” Martienssen said.
Thinking broadly about heredity might help us outside of laboratories as well. In the United States, it has proven all too tempting over the centuries to blame poverty and inequality on biology. A woman like Emma Wolverton could be institutionalized for life because she was judged a genetically doomed moron. The relative poverty of African Americans could be written off, even by some psychologists, as the result of their inheriting the wrong genes.
Others have argued that the gulfs in the United States are the product of the environment into which people are born and grow up. But the word environment is too bland to help us understand much about this problem. The stubborn inequalities in the United States are not the result of some people living in a physical environment. Their environment is built by social forces, and those forces last for centuries because they are regenerated across the generations.
After blacks were emancipated from slavery, they still had to contend with structural racism as well as the racist attitudes of individuals. This racism did not keep springing out of the void year after year. Children learned it, either implicitly or outright, from parents and other adults, and then passed it on to their own children. The social environment then shaped the physical environment into which later generations of blacks were born. Housing discrimination and segregation created neighborhoods where children ended up in poorly performing schools, had to contend with far greater odds of getting shot, and had fewer opportunities for work.
Cumulative culture allowed our species to make giant leaps in technological progress, but it also made us prone to inequality. Hunter-gatherers tend to keep these differences in check, although in a society like the Nootka of Vancouver Island, some people ended up impoverished slaves serving wealthy masters. Once farmers began building up surpluses of food, the gulf could begin to open. They could grow not just over the course of one farmer’s lifetime but across generations, because now there were goods to inherit. At first, children might inherit farms and stores of grain from their parents; later, gold and houses and other goods enriched them. The Industrial Revolution made the entire world richer, but some people became vastly richer than others. Francis Galton’s ancestors built an empire on guns and banking, which left him able to hire all the math tutors he cared to.
In 1931, the historian James Truslow Adams contrasted the United States with countries like Great Britain by what he called “the American dream.” He defined the dream as being “that life should be made richer and fuller for everyone and opportunity remain open to all.” For much of the 1900s, the United States lived up to that dream fairly well. Immigrants fared better there than they had in their home countries. As the United States grew wealthier, much of that wealth flowed to the poorest half of American citizens and allowed them to climb the economic ladder. Raj Chetty, a Stanford economist, has estimated that Americans born in 1940 had a 90 percent chance of making more money than their parents at age thirty.
But Chetty and his colleagues have found that those odds then steadily dropped. Americans born in 1984 had only a 50 percent chance of making more than their parents. The shift was not the result of the United States suddenly running out of money. It’s just that wealthy Americans have been taking much of the extra money the economy has generated in recent decades. Chetty’s research suggests that if the recent economic growth in the United States was distributed more broadly, most of the fading he has found would disappear. “The rise in inequality and the decline in absolute mobility are closely linked,” he and his colleagues reported in 2017.
Inheritance has helped push open that gulf. About two-thirds of parental income differences among Americans persist into the next generation. Economists have found that American children who are born to parents in the ninetieth percentile of earners will grow up to make three times more than children of the tenth percentile.
This inheritance is not simply what parents leave in their wills but the things that they can buy for their children as they grow up. In the United States, affluent parents can afford a house in a good public school district, or even private school tuition. They can pay for college test prep classes to increase the odds their children will get into good colleges. And if they do get in, their parents can cover more of their college tuitions.
Poor parents have fewer means to prepare their children to get into college. Even if their children do get accepted, they have fewer funds, and they’re more vulnerable to layoffs or medical bankruptcy. Their children may graduate saddled with steep college debts or drop out before getting a degree.
The gifts that children inherit can keep coming well into adulthood. Parents may help cover the cost of law school, or write a check to help out with a septic tank that failed just after their children bought their first house. Protected from catastrophes that can wipe out bank accounts, young adults from affluent families can get started sooner on building their own wealth.
Inheritance also goes a long way to explain the gap in wealth between races in the United States. In 2013, the median white American household had thirteen times the wealth of the median black household, and ten times that of the median Latino household. In 2017, a team of researchers from Brandeis University and the public policy group Demos sifted through a number of hypotheses that might account for the differences. Going to college didn’t close the gap. In fact, the researchers found, the median wealth of white people who didn’t finish high school was greater than that of blacks who went to college. Black families actually save money at a greater rate than their white counterparts. Nevertheless, the median white single parent has 2.2 times more wealth than the median black household where there are two parents.
The one big difference the researchers did find was inheritance. Whites are five times as likely to receive major gifts from relatives, and when they do, their value is much greater. These gifts can, among other things, allow white college students to graduate with much less debt than blacks or Latinos. And the effects of these inheritances have compounded through the generations as blacks and Latinos were left outside the wealth feedback loop that benefited white families.
Left unmanaged, these cultural inheritances will roll on, and future generations will be born into systems of economic inequality. The same is true for the environmental inheritance we leave. One of the most important things each new generation learns from the previous one is how to get enough energy to survive. That usually means liberating the Earth’s supply of organic carbon and putting some of it in the air. Some people learn how to cut down forests to make charcoal. Others pilot cargo ships across the ocean, leaving a diesel plume trailing behind. If we carry on this way, we may manage to burn through the remaining 12 billion tons of fossil fuels tucked away inside our planet by 2250.
In the process, we would be raising the concentration of carbon dioxide in the atmosphere to levels not seen in the last 200 million years, raising the temperature of the planet to levels far beyond what we humans—a species of ape that evolved in the modest swings of Ice Ages—could handle. And the day that the final gas tank ran dry, the last lightbulb winked out, the planet would not immediately set itself back to the way it was before cultural heredity became such a titanic force. It would take thousands of years for the planet to naturally draw down the carbon dioxide to levels close to what they were before the Agricultural Revolution.
To solve a problem like global warming, we cannot come up with a clever technological fix. We are not being threatened by a giant volcano belching out carbon dioxide from the depths of the Earth, which we can simply cover with a titantic plug. Global warming is a problem of cultural inheritance. To fix it, we need a social form of CRISPR—a means to alter the practices and the values that make their way from one generation to the next.
A cynic may say that there are no systems that can possibly put the brakes on the problems we have made for ourselves. But the environmental scientist Erle Ellis has observed that history records many examples of cultures that transmitted customs through the generations that allowed them to thrive while not destroying their environment in the process. The Maasai of East Africa, for example, have herded cattle for centuries on a landscape that also supported elephants, zebra, lions, and many other wild animals. The long-term health of their ecosystem was the direct result of the culture that the Maasai inherited from their ancestors. Much of their cultural identity is wrapped up in herding cattle—which means they have no need to hunt wild game. To lose a herd of cows and have to hunt marks a huge fall in status. The result was that East Africa could support the great diversity of large mammals on Earth.
“This is a gift to every one of us on Earth now and in the future,” Ellis wrote in a 2017 essay. “The megafauna and landscapes they helped to sustain might yet outlast the Great Pyramids or New York City.”
When the rest of us look at a culture like the Maasai, we should ask what sort of world we want to leave as a legacy, and then figure out how to do so. It may be that CRISPR can be one tool that we can use toward that end. But we must be confident it reworks the world as we truly need it to.
By the time I paid a visit to Anthony James’s insectarium in 2017, gene drive was already becoming something of a Manhattan Project. James and other researchers were getting massive grants from the United States Department of Defense, as well as from major foundations around the world. Yet neither James nor any other gene drive researcher had yet released a CRISPR-bearing creature into the wild. And they were in no rush to do so. They were all too aware of past attempts to fix environmental problems that had turned into ecological disasters. And since those introduced species could keep reproducing, every new generation inherited a warped ecosystem.
Starting in the late 1800s, for example, Australian farmers established sugarcane farms, but they fell into a constant fight with cane beetles. In the early 1930s, an Australian entomologist named Reginald Mungomery got an idea for how to win the battle. He heard stories of the giant marine toad. Native to South and Central America, it had a tremendous appetite for insects, and some people had brought the toads to Hawaii to control sugarcane pests there. He got hold of the toads and raised 2,400 of them. And then, in 1935, he set them loose.
Mungomery didn’t understand that the toads were catholic in their tastes. Soon the enormous amphibians—known in Australia as cane toads—were hopping out of the plantations and feeding on small mammals. Australian snakes and other predators sometimes tried to eat the cane toads, but a poisonous secretion in their skin made that impossible. At best, the predators spat out the toads and never tried eating them again. At worst, they died. The cane toad spread relentlessly across Australia, pushing a number of native species toward extinction. Australian researchers have tried all sorts of ways to stop them—poisoning the frogs, training native species not to try to eat them—but so far, nothing has worked.
No one wants to be the Reginald Mungomery of the CRISPR age. It’s possible that gene drives could go wrong by hopping from a species we want to eradicate to a related one we want to save. It’s possible that changing how mosquitoes and other animals respond to one disease could lead them to carry others. Perhaps getting rid of mosquitoes might disrupt ecosystems in ways we can’t yet imagine.
Jennifer Kuzma and Lindsey Rawls, two legal scholars at North Carolina State University, have started to examine the ethics of gene drive as a kind of inheritance. Altering the heredity of disease-carrying insects in the short-term could be tremendously valuable in the number of lives it saves and in the suffering it eliminates. But we also owe future generations a careful, forward-looking consideration of the world they will inherit.
Kuzma and Rawls suggest that, by this standard, some gene drives will prove to be justified and others not. They suggest that saving endangered birds should get ranked over altering weeds. The birds deserve a higher priority because they may very well go extinct if we do nothing. Their disappearance will itself be a permanent legacy we leave to future generations.
When I visited James and his colleagues, I asked them about these ethical issues. They didn’t have a lot to say. It’s not that they didn’t care. They just had more pressing problems at hand. They weren’t sure if CRISPR gene drives would work at all.
After all, the natural world was littered with the remains of dead gene drives. They had evolved, raced through populations, and then stopped. In some cases, mutations destroyed them. In other cases, animals evolved defenses that kept them in check. Some biologists have argued that it would be easy for mosquitoes to evolve resistance to a CRISPR gene drive. Some of the insects might gain mutations that changed the sequence of DNA that the CRISPR guide molecules searched for. Their descendants would inherit those mutations and might outbreed the ones carrying the gene drive.
“It’s probably easier to break because it isn’t an evolved system,” Bier told me. “The system we’re making is all completely synthetic. It’s frail.”
Meanwhile, James was toiling away in his insectarium, trying to figure out how to make CRISPR work better. When he put Gantz’s malaria-fighting gene drive into a mosquito, all of its offspring inherited it. In the second generation, though, it faltered. Almost all the males inherited it but only some of the females did.
James could still carry forward the gene drive to a new generation by mating the male mosquitoes with normal females. The hairy larvae that I inspected in James’s insectarium were all males from his twenty-ninth generation, ready to produce the thirtieth. But James still puzzled over why the female mosquitoes were proving a weak link in the hereditary chain.
The answer may have to do with how mosquitoes develop from a single egg. As a female mosquito develops, it requires many divisions before some of its cells develop into a new batch of eggs. Along the way, a chromosome inside a cell may break. Cells repair this sort of damage by copying DNA from the undamaged copy of the chromosome. James suspected that, during these bouts of repair, the female mosquitoes were editing out their own CRISPR genes. Male mosquitoes, on the other hand, may not lose their gene drives because they set aside sperm cells earlier in their development. If James and his colleagues were right in this hunch, it was hard to see how to overcome it. The inner heredity of mosquitoes isn’t easily altered.
After James had shown me all his mosquitoes and answered all my questions, it was time for us to leave the insectarium. We stepped back out into the vestibule, and he closed the inner door loudly behind us. On the other side were thousands of mosquitoes drinking blood, and thousands of larvae writhing in tubs. Here, in the quiet of the vestibule, it was just us two humans, as far as I could tell.
James turned to the pale door to the insectarium and stared at its blankness. The blue smock still hung from his arms.
“The protocol is for us to stand here for a little while,” he said. “See if anybody followed us out.”
The mosquitoes that James raises come from India. They’re adapted to the wet, tropical climate there. If a CRISPR-infused mosquito managed to escape James’s insectarium, buzz down the hallways, slip up an elevator shaft, and dart through the doorways into the dry hills around Irvine, it would almost certainly die. And yet, even with such safeguards in place, James still stared at the door, to be sure all his mosquitoes were still penned in his insectarium. As time passed, we both grew quiet. On the other side, a potential new chapter of heredity was crawling, swimming, flying.
Once James felt satisfied that no mosquitoes had escaped, he turned away from the pale inner door. He opened the outer door and we stepped out into the basement hallway. We cast off our smocks into a bin and took the elevator up to the mosquito-killing California sun. We left the next chapter penned in its underground cell, at least for now.