Dr Camilla Rothe was leaving for dinner when the government laboratory called with the surprising test result. Positive. It was Jan 27. She had just discovered Germany's first case of the coronavirus.
But the diagnosis made no sense. Her patient, a businessman from a nearby car-parts company, could have been infected by only one person: a colleague visiting from China. And that colleague should not have been contagious.
The visitor had seemed perfectly healthy during her stay in Germany. No coughing or sneezing, no signs of fatigue or fever during two days of long meetings.
She told colleagues that she had started feeling ill after the flight back to China. Days later, she tested positive for the coronavirus.
Scientists at the time believed that only people with symptoms could spread the coronavirus. They assumed it acted like its genetic cousin, Sars.
"People who know much more about coronaviruses than I do were absolutely sure," recalled Dr Rothe, an infectious disease specialist at Munich University Hospital.
But if the experts were wrong - if the virus could spread from seemingly healthy carriers or people who had not yet developed symptoms - the ramifications were potentially catastrophic.
Public awareness campaigns, airport screening and stay-home-if-you-are-sick policies might not stop it. More aggressive steps might be required: ordering healthy people to wear masks, for instance, or restricting international travel.
Dr Rothe and her colleagues were among the first to warn the world. But even as evidence accumulated from other scientists, leading health officials expressed unwavering confidence that symptomless spreading was not important.
In the days and weeks to come, politicians, public health officials and rival academics disparaged or ignored the Munich team. Some actively worked to undermine the warnings at a crucial point, as the disease was spreading unnoticed in French churches, Italian soccer stadiums and Austrian ski bars. A cruise ship, the Diamond Princess, would become a deadly harbinger of symptomless spreading.
Interviews with doctors and public health officials in more than a dozen countries show that for two crucial months - and in the face of mounting genetic evidence - Western health officials and political leaders played down or denied the risk of symptomless spreading.
Leading health agencies, including the World Health Organisation (WHO) and the European Centre for Disease Prevention and Control (ECDC), provided contradictory and sometimes misleading advice.
A crucial public health discussion devolved into a semantic debate over what to call infected people without clear symptoms.
The two-month delay was a product of faulty scientific assumptions, academic rivalries and, perhaps most important, a reluctance to accept that containing the virus would involve drastic measures.
'SHE WAS NOT ILL'
On the night of Germany's first positive test, the virus had seemed far away. Fewer than 100 fatalities had been reported worldwide. Italy, which would become Europe's ground zero, would not record its first cases for another three days.
A few reports out of China had already suggested the possibility of symptomless spreading. But nobody had proved it could happen.
That night, Dr Rothe tapped out an e-mail to a few dozen doctors and public health officials.
"Infections can actually be transmitted during the incubation period," she wrote.
Three more employees from the car-parts company, Webasto, tested positive the next day. Their symptoms were so mild that, normally, it is likely none would have been flagged for testing or would have thought to stay at home.
Dr Rothe decided she had to sound the alarm. Her boss, Dr Michael Hoelscher, then dashed off an e-mail to The New England Journal of Medicine. "We believe that this observation is of utmost importance," he wrote.
The editors responded immediately. How soon could they see the paper?
The next morning, Jan 30, public health officials interviewed the Chinese businesswoman by phone. Hospitalised in Shanghai, she told them that she had started feeling sick on the flight home. Looking back, maybe she had experienced some mild aches or fatigue, but she had chalked them up to a long day of travel.
"From her perspective, she was not ill," said Ms Nadine Schian, a Webasto spokesman, who was on the call.
When the health officials described the call, Dr Rothe and Dr Hoelscher quickly finished and submitted their article.
Dr Rothe did not talk to the patient herself but said she relied on the health authority's summary.
Within hours, it was online. It was a modest clinical observation at a key time. Just days earlier, WHO had said it needed more information about this very topic.
What the authors did not know, however, was that in a suburb 20 minutes away, another group of doctors had also been rushing to publish a report. Neither knew what the other was working on, a seemingly small academic rift that would have global implications.
The second group was made up of officials from the Bavarian health authority and Germany's national health agency, known as the Robert Koch Institute. Inside a suburban office, doctors unfurled mural paper and traced infection routes using coloured pens.
Their team, led by Bavarian epidemiologist Merle Bohmer submitted an article to The Lancet, another premier medical journal. But the Munich group had scooped them by three hours. Dr Bohmer said her team's article, which went unpublished as a result, had reached similar conclusions but worded them slightly differently.
Dr Rothe had written that patients appeared to be contagious before the onset of any symptoms. The government team had written that patients appeared to be contagious before the onset of full symptoms - at a time when symptoms were so mild that people might not even recognise them.
The Chinese woman, for example, had woken up in the middle of the night feeling jet-lagged. Wanting to be sharp for her meetings, she took a Chinese medicine called 999 - containing the equivalent of a Tylenol tablet - and went back to bed. Perhaps that had masked a mild fever? Perhaps her jet lag was actually fatigue? She had reached for a shawl during a meeting. Maybe that was a sign of chills?
After two lengthy phone calls with the woman, doctors at the Robert Koch Institute were convinced that she had simply failed to recognise her symptoms.
They wrote to the editor of The New England Journal of Medicine, casting doubt on Dr Rothe's findings. The journal did not publish the letter. But that would not be the end of it.
'A COMPLETE TSUNAMI'
On Monday, Feb 3, the journal Science published an article calling Dr Rothe's report "flawed". Science reported that the Robert Koch Institute had written to the New England Journal to dispute her findings and correct an error.
Dr Rothe's report quickly became a symbol of rushed research. Scientists said she should have talked to the Chinese patient herself before publishing and that the omission had undermined her team's work.
On Twitter, she and her colleagues were disparaged by scientists and armchair experts alike.
"It broke over us like a complete tsunami," Dr Hoelscher said.
Sweden's public health agency declared that Dr Rothe's report had contained major errors. The agency's website said, unequivocally, that "there is no evidence that people are infectious during the incubation period" - an assertion that would remain online in some form for months.
French health officials, too, left no room for debate: "A person is contagious only when symptoms appear," a government flier read. "No symptoms = no risk of being contagious."
As Dr Rothe and Dr Hoelscher reeled from the criticism, Japanese doctors were preparing to board the Diamond Princess. A former passenger had tested positive for the coronavirus. Yet on the ship, parties continued. After all, the infected passenger had been off the ship for days. And he had not reported symptoms while onboard.
A SEMANTIC DEBATE
Immediately after Dr Rothe's report, WHO noted that patients might transmit the virus before showing symptoms. But the organisation also underscored a point that it continues to make: Patients with symptoms are the main drivers of the epidemic.
Once the Science paper was published, however, the organisation waded directly into the debate on Dr Rothe's work.
On Tuesday, Feb 4, Dr Sylvie Briand, the agency's chief of infectious disease preparedness, tweeted a link to the Science paper, calling Dr Rothe's report flawed.
With that tweet, WHO focused on a semantic distinction that would cloud discussion for months: Was the patient asymptomatic, meaning she would never show symptoms? Or pre-symptomatic, meaning she became sick later? Or, even more confusing, oligo-symptomatic, meaning that she had symptoms so mild that she didn't recognise them?
That night, Dr Rothe received an e-mail from Dr Michael Libman, an infectious disease specialist in Montreal. He thought that criticism of the paper amounted to semantics. Her paper had convinced him of something: "The disease will most likely eventually spread around the world."
Data would soon arrive, and from an unexpected source. Dr Bohmer, from the Bavarian health team, received a startling phone call in the second week of February.
Virologists had discovered a subtle genetic mutation in the infections of two patients from the Munich cluster. They had crossed paths for the briefest of moments, one passing a salt shaker to the other in the company cafeteria, when neither had symptoms. Their shared mutation made it clear that one had infected the other.
Dr Bohmer had been sceptical of symptomless spreading. But now there was no doubt: "It can only be explained with pre-symptomatic transmission," Dr Bohmer said.
Now it was Dr Bohmer who sounded the alarm. She said she promptly shared the finding, and its significance, with the WHO and the ECDC.
Neither organisation included the discovery in its regular reports.
With evidence mounting, the Munich team could not understand how WHO could be so sure that symptomless spreading was insignificant. The Munich cluster was not the only warning.
The Chinese health authorities had explicitly cautioned that patients were contagious before showing symptoms.
A Japanese bus driver was infected while transporting seemingly healthy tourists from Wuhan, China. And by the middle of February, 355 people on the Diamond Princess had tested positive. About one-third of the infected passengers and staff had no symptoms.
RECOMMENDATION ON MASKS
By early March, while the WHO was continuing to press the case that symptom-free transmission was rare, science was breaking in the other direction.
Researchers in Hong Kong estimated that 44 per cent of Covid-19 transmission occurred before symptoms began, an estimate that was in line with a British study that put that number as high as 50 per cent.
More than 700 people aboard the Diamond Princess were sickened. Fourteen died.
Researchers estimate that most of the infection occurred early on, while seemingly healthy passengers socialised and partied.
As the research coalesced in March, European health officials were convinced. Since then, WHO has recommended that people wear masks in public.