Soho, London · September 1854
In the summer of 1854 almost every doctor in London knew how cholera spread: through miasma, the foul air that rose from rotting rubbish, sewage and the Thames mud. It was a sensible-sounding theory. The poorest, smelliest streets really were where cholera struck hardest. It was also wrong, and a quiet anaesthetist called John Snow set out to prove it with a pencil, a street map and a lot of knocking on doors.
Snow was no outsider. He was one of the best-known anaesthetists in the country and had given chloroform to Queen Victoria for the birth of Prince Leopold in 1853. But for years he had argued that cholera was not breathed in. It was swallowed, he said, in water contaminated with the waste of earlier victims. Few people agreed.
On the night of 31 August 1854 cholera exploded in the streets around Broad Street (now Broadwick Street). Snow later counted 197 deaths on the 1st and 2nd of September alone. Families fled; whole streets emptied. By the time it burned out, his tally of fatal cases linked to the outbreak was 616.
Snow lived nearby. He got a list of the dead from the General Register Office, walked to each address, and asked the same question again and again: where did this household get its water? Soho had no taps in most homes. People filled buckets at public street pumps, and the answer kept coming back the same.
Try it yourself below. This is a simplified, made-up Soho with six pumps, not Snow's real data. Run the outbreak, watch the black bars pile up at doorways just as Snow drew them, then tap the pump you think is guilty.
Press Run the outbreak. Each black bar is one death, stacked at the door of the house where it happened.
When Snow plotted the deaths, they clustered tightly around one pump, on Broad Street. But "around" hid a subtlety. People don't walk through walls. A house that looks close to another pump on the map may actually be a shorter walk from Broad Street, and the reverse.
So for a later version of his map, Snow drew a dotted line around the Broad Street pump connecting the points that were, "by careful measurement", an equal distance from it by the nearest road. Inside that line, Broad Street was the closest pump on foot, and that is where the deaths were. Toggle between By walking and As the crow flies above: the crow's boundaries slice across blocks, while the walking ones follow the streets. Today we would call Snow's diagram a kind of Voronoi map, one of the first used to solve a real problem.
The most convincing evidence came from the cases that broke the pattern. A brewery stood on Broad Street itself, a few steps from the pump. Of more than 70 men who worked there, none died; the owner told Snow they were given beer and never drank the pump water, and the brewery had its own well. A workhouse in nearby Poland Street held 535 inmates, yet only five died. It, too, had its own pump.
Meanwhile a widow living miles away in Hampstead died of cholera. She had once lived on Broad Street and liked its water so much that a cart brought her a bottle of it every day. Her niece, visiting from Islington, drank it too, and also died. Neither of their neighbourhoods had any cholera at the time.
Distance from the pump didn't predict who died. Drinking from it did.
On 7 September Snow put his evidence to the parish's Board of Guardians. They were not convinced by his theory, but they agreed to act, and the next day, 8 September, the pump's handle was taken off so no one could use it.
The legend says the epidemic stopped overnight. Snow himself was more honest: deaths had already fallen sharply before the handle came off, partly because so many people had fled or already been infected. Press Remove the handle at different moments in the model and watch the daily chart. Taken off early, it saves lives; taken off on day nine, as in 1854, most of the damage has been done. Its real value was preventing a second wave.
The final link came from a local curate, Henry Whitehead, who had set out to disprove Snow. Digging through records, he found the likely first case: five-month-old Frances Lewis of 40 Broad Street, who fell ill just before the outbreak. Her mother had rinsed the baby's nappies and emptied the water into a cesspool. A surveyor found that cesspool's crumbling brickwork lay 2 feet 8 inches from the pump's well.
Even so, the official Board of Health stuck largely with bad air. The germ itself had been seen under a microscope by Filippo Pacini in Florence that very year, 1854, but his work went unnoticed; Robert Koch identified it again in 1883–84. Snow died in 1858, aged 45, long before his water theory was widely accepted. His map is now treated as a founding document of epidemiology, the science of asking not just who got sick, but where, when, and from what.