One half of America is protected. The other is approaching a perilous moment in the pandemic.

Lineage B.1.617.2, now known as the Delta variant, was first detected in India, in December, 2020. An evolved version of sars-CoV-2, Delta has at least a dozen mutations, including several on its spike protein that make it vastly more contagious and possibly more lethal and vaccine-resistant than other strains. In India, the Delta variant contributed to the most devastating coronavirus wave the world has seen so far; now, it has been detected in dozens of countries, including the United States. In the U.S., it accounts for a minority of cases—but it is rapidly outcompeting other variants, and will likely soon become our dominant lineage.

Much of what we know about Delta is preliminary, and based on reports from India and, more recently, the U.K., where it now accounts for more than ninety per cent of new cases. Four-fifths of British adults have received at least one shot of a covid-19 vaccine, and more than half are fully vaccinated—but the variant has spread widely enough among those who remain vulnerable to fuel a quadrupling of cases and a doubling of hospitalizations in the past month. The vast majority of Delta-variant cases seem to have occurred in adults under fifty, whose rates of vaccination remain lower than those of older people. Last week, Prime Minister Boris Johnson announced that the U.K.’s full reopening, originally scheduled for June 21st, would be postponed.

Earlier this year, scientists estimated that lineage B.1.1.7—the Alpha variant, first isolated in England—could be some sixty per cent more transmissible than the original version of sars-CoV-2. Now, experts believe that the Delta variant is sixty per cent more transmissible than Alpha—making it far more contagious than the virus that tore through the world in 2020. It hasn’t yet been conclusively shown that Delta is more lethal, but early evidence from the U.K. suggests that, compared to Alpha, it doubles the risk of a person’s being hospitalized. Even if the variant turns out to be no deadlier within any one person, its greater transmissibility means that it can inflict far more damage across a population, depending on how many people remain unvaccinated when it strikes.

In this regard, India’s apocalyptic surge is Exhibit A. In May, at the crest of the wave, the role of the Delta variant was still unclear. A number of factors—the return of large gatherings, a decline in mask-wearing, and a sluggish vaccination campaign—had made a disaster of some kind more or less unavoidable. But it now seems likely that the rise of Delta accelerated the crisis into a shockingly rapid and widespread viral catastrophe. In the course of weeks, millions of people were infected and tens of thousands died; the country’s medical system buckled under the weight of a mutated virus. One of the most disturbing aspects of India’s surge was that many children fell ill. And yet there is currently no data to suggest that Delta causes severe illness in a greater proportion of kids; instead, it seems likely that the sheer transmissibility of the variant simply resulted in a higher absolute number of infected children.

One vitally important finding to emerge from the U.K. and India is that the covid vaccines are still spectacularly effective against Delta. According to one study from the U.K., a full course of the Pfizer-BioNTech vaccine is ninety-six per cent effective at preventing hospitalizations due to the Delta variant; AstraZeneca’s vaccine is in the same ballpark, reducing the chance of hospitalization by ninety-two per cent. But these findings come with caveats. The first is that, with Delta, partial immunization appears to be less effective at preventing disease: a different study found that, for people who have received only the first shot, the vaccines were just thirty-three per cent effective at preventing symptomatic illness. (A first dose still appears to offer strong protection against hospitalization or death.) The second is that even full courses of the vaccines appear somewhat less effective at preventing infection from Delta. This may be especially true of the non-mRNA vaccines. A team of scientists in Scotland has found that both doses of AstraZeneca’s vaccine reduced the chance of infection with Delta by just sixty per cent—a respectable showing, but less impressive than what the same vaccine offers against other strains of the virus. (The Pfizer-BioNTech vaccine demonstrated seventy-nine per cent efficacy against Delta infection—a significant, but smaller, decrease.)

Taken together, these findings have led some experts to propose adjustments in vaccination strategy. Muge Cevik, an infectious-diseases expert at St. Andrews University and an adviser to the British government, told me that, given the arrival of Delta, it was important to ask “what our main aim of vaccination is.” She went on, “If our primary objective is to reduce hospitalizations and deaths, a first dose still gives very good protection. If it’s to stamp out transmission, then the second dose becomes quite important. I think that, especially in hot spots, we need to expedite second shots.” Others have proposed the idea of mRNA-vaccine booster shots for Americans who have received the Johnson & Johnson vaccine, which, like AstraZeneca’s, uses non-mRNA technology. The C.D.C.’s official guidelines tell Americans that “the best covid-19 vaccine is the first one that is available to you. Do not wait for a specific brand.” But that advice was minted when vaccine supply was constrained. The accumulated evidence has led many people to wonder whether the mRNA vaccines, from Moderna and Pfizer, are preferable to the one offered by Johnson & Johnson, and whether the Delta variant makes them even more so.

“It’s likely that J. & J. offers strong protection against severe disease, but because it’s a one-shot regimen it might not offer the same protection against infection for a highly transmissible variant like Delta,” Angela Rasmussen, a virologist at the Vaccine and Infectious Disease Organization, told me. “A second shot reëxposes the immune system to the vaccine, and allows the body to make even better antibodies.” Rasmussen received the J. & J. vaccine; she lives in Canada, where health authorities have encouraged people to mix and match the vaccines. “I’m considering topping off my immune system with a dose of Pfizer,” she said. “It’s something worth thinking about.”

To a significant degree, the emergence of a variant like Delta was predictable, and, with rapid and widespread immunization, the threat that it poses can be subdued. But its arrival is still incredibly consequential. Delta drives an even wider wedge between vaccinated and unvaccinated people. They have already been living in separate worlds, facing vastly different risks of illness and death; now, their risk levels will diverge further. People who’ve been fully vaccinated can, by and large, feel confident in the immunity that they’ve received. But those who remain susceptible should understand that, for them, this is probably the most dangerous moment of the pandemic.

“The good news is that we have vaccines that can squash the Delta variant,” Eric Topol, the director of the Scripps Research Translational Institute, told me. “The bad news is that not nearly enough people have been vaccinated. A substantial share of Americans are sitting ducks.” He went on, “We haven’t built a strong enough vaccination wall yet. We need a Delta wall”—a level of vaccination that will prevent the new variant from spreading. “There are still large unvaccinated pockets in the country where this could get ugly,” Topol added. Because about half of Americans are vaccinated, and millions more have some immunity from prior infection, the Delta variant “won’t cause monster spikes that overwhelm the health system,” Topol said. But Delta spreads so easily among the unvaccinated that some communities could experience meaningful increases in death and disease this summer and fall.

In America, the speed of vaccination is slowing. In some states, mainly in the South, only about a third of the population has been fully vaccinated. Big differences in the covid-19 toll are already visible: cases and hospitalizations have plummeted in some places with higher vaccination rates but are holding steady or rising in others. Fortunately, nearly ninety per cent of older Americans—the group most at risk for severe covid—have received at least one shot, and three-quarters are fully vaccinated. But, as is clear from the Indian and U.K. experiences, the Delta variant could still lead to major spikes in infection among younger, unvaccinated people.

In a recent piece, I likened a society that’s reopening while partially vaccinated to a ship approaching an iceberg. The ship is the return to normal life and the viral exposure that it brings; the iceberg is the population of unvaccinated people. Precautions such as social distancing can slow the speed of the ship, and vaccination can shrink the size of the iceberg. But, in any reopening society that’s failed to vaccinate everyone, a collision between the virus and the vulnerable is inevitable.

Because of its exceptional transmissibility, the Delta variant is almost certain to intensify the force of the collision. The U.K., by postponing a full reopening, is trying to soften the blow. But the U.S. is pressing ahead—perhaps out of hubris, or because officials hope that our vaccination campaign can outrun the spread of Delta. Last week, New York and California, among the pandemic’s hardest-hit states, did away with virtually all restrictions. Meanwhile, states with half the vaccination rates of New York or California have been open for weeks. A lot depends on where, and how fast, Delta is spreading.

Federal, state, and local officials are trying to accelerate vaccination. Governors have announced incentives such as lotteries, college scholarships, gift cards, and free beer for those who get immunized; California alone plans to spend more than a hundred million dollars on vaccine incentives. The Biden Administration has made immunizing seventy per cent of American adults by the Fourth of July a central priority, and has declared June a “national month of action.” The Administration has offered tax credits to employers that provide paid time off for people to get immunized, erected mass-vaccination sites, sent funds to community health centers, and partnered with local organizations, celebrities, and volunteers to get shots in arms. The White House recently announced that four of the nation’s largest child-care providers would offer free services to parents who want to get immunized before July 4th; Uber and Lyft have been offering free rides to vaccination sites for weeks.

And yet, the pace of vaccinations hasn’t picked back up. Topol, for his part, believes that a major impediment to wider vaccination is the fact that the F.D.A. has not yet fully approved the covid vaccines; right now, they’ve received only an emergency-use authorization, or E.U.A. About a third of unvaccinated Americans say that F.D.A. approval would make them more likely to get immunized. Full approval could also pave a clearer path for vaccine mandates in schools, businesses, and the military. Topol argues that mandates would allow us to build a Delta wall more quickly—along with walls for Epsilon, Zeta, and the rest of the Greek alphabet. Both Pfizer and Moderna have applied for F.D.A. approval, but it’s unclear how soon they will receive it; the usual process takes six to ten months. “Hundreds of millions of people have safely taken these vaccines, but there’s still a perception among some that they’re experimental,” Topol said. “E.U.A. versus full approval may sound like semantics, but it’s actually a B.F.D.”

Globally, more people died of the coronavirus in the first half of this year than in all of last year—an astounding fact, given the emergence of the vaccines. The tragic truth is that, for much of the world, the vaccines may as well not exist. On the one hand, the U.S. is vaccinating children as young as twelve; on the other hand, health-care workers, elderly people, and cancer patients in many other countries remain defenseless. Three-quarters of covid-vaccine shots have been administered in just ten countries, whereas the poorest nations have received less than one half of one per cent of the supply. Tedros Adhanom Ghebreyesus, the W.H.O. director-general, has called this a “scandalous inequity.”

The Biden Administration recently announced that the U.S. would donate half a billion doses to the global vaccination effort; it hopes to deliver two hundred million by the end of the year. The U.K. and other European countries have also committed hundreds of millions of doses to covax, the international initiative to distribute vaccines to low- and middle-income countries. These efforts are important, and they will help immensely—but not for months, and perhaps not until 2022. In the meantime, many countries will continue to grapple with the social and economic challenges created by variant-catalyzed surges and the public-health measures needed to thwart them. Even where the political will for continuing such measures exists, it’s not infinite; countries can’t remain in lockdown forever.

In a sense, Delta is the first post-vaccination variant. Pockets of the human race—perhaps five hundred million people out of 7.6 billion—are protected against it, despite its transmissibility; for them, the pandemic’s newest chapter is something of an epilogue, since the main story has, in effect, already concluded. But, for those who remain unvaccinated, by choice or by chance, Delta represents the latest installment in an ongoing series of horrors. It’s a threat more sinister than any other—one that imperils whatever precarious equilibrium has taken root. In a partially vaccinated world, Delta exposes the duality in which we now live and die.