Few stories are as prominent in the study of infectious diseases as that of Mary Mallon,
a cook to wealthy families, and also to a maternity hospital, in New York in the early 1900s.
As she went from one employer to another, typhoid fever, then deadly in one case in ten, followed in her wake.
Public-health officials eventually joined the dots and identified her as a carrier of Salmonella typhi, the bacterium that causes the disease.
What was striking about Typhoid Mary, as the newspapers nicknamed her, was that she herself was healthy—
proof that people could harbour and transmit S. typhi without showing symptoms of the illness it causes.
Such silent transmission, as epidemiologists call the phenomenon, has since been observed in many diseases—among them measles, influenza and HIV/AIDS.
A fresh addition to the list is SARS-COV-2, the coronavirus behind the covid-19 pandemic now raging.
Accumulating evidence suggests a substantial chunk of the infections it causes
are transmitted by people whose symptoms have not yet appeared—or even, like Mallon, who never develop symptoms at all.
That has implications for the methods countries are employing to curb the pandemic.
Currently, none of the evidence on asymptomatic transmission is watertight.
According to Gerardo Chowell of Georgia State University, in Atlanta,
the best way to determine the share of SARS-COV-2 infections that happen in this way
is to follow up a large number of households in which someone is already infected and then track who subsequently infects whom.
For this to work, everyone involved would have to be tested daily.
If this were done, comparing subtle variations from person to person in the virus's genetic material would show who caught it from whom.
Definitive studies of this nature are not yet available, though some are probably in the works, Dr Chowell reckons.
In the meantime, a growing collection of other research is shedding light on the matter. This work comes in three strands.
The first is a set of studies of people in groups for which unusual circumstances have made possible tallying each and every infection.
These studies permit a fairly precise estimate of the share of those infected who have no symptoms.
One such group are the passengers and crew of the Diamond Princess, a cruise ship on-board which the infection rate exploded because of a bungled quarantine.
Of 634 people thus infected, 52% had no symptoms at the time of testing, including 18% who never developed symptoms.
The residents of Vo, an Italian town in which all 3,300 people were tested twice, is another much-cited example.
Of those in Vo found to be infected, 50-75% had no symptoms at the time of the test.