A contaminated petri dish
On 3 September 1928, Alexander Fleming noticed that a mould had contaminated one of his bacterial culture plates at St Mary’s Hospital in London and, unusually, had killed the surrounding bacteria rather than simply competing with them for space. He confirmed the effect later that month, published his findings in 1929, and named the substance penicillin after the Penicillium mould producing it, but he could not purify or stabilise it well enough to turn it into a usable medicine, and the discovery sat largely unexploited for over a decade. Fleming himself also noted early on that not all bacteria were affected by the substance, an observation about penicillin’s limits that predated, by more than a decade, any practical use of the drug at all and that turned out to anticipate the resistance problem that would eventually follow widespread treatment.
From Oxford lab to wartime factories
Turning Fleming’s observation into medicine required a separate team: Howard Florey, Ernst Chain and Edward Abraham at Oxford, who worked out from 1940 onward how to concentrate penicillin from mould cultures and confirmed it worked against infection in living animals, not just in a laboratory dish. Their first attempt on a human patient, a policeman named Albert Alexander treated in 1941, showed real initial improvement before the limited supply ran out and he died, a stark demonstration that the science worked but the manufacturing did not yet exist at any usable scale. Mass production followed rapidly once wartime priorities took hold: a superior mould strain found in a Peoria cantaloupe in 1943 produced six times more penicillin than Fleming’s original, deep-tank fermentation techniques scaled output further, and by the spring of 1944 over two million doses were ready in time for the Normandy invasion.
A mechanism that became a template
Penicillin’s effectiveness against wound infection during the war was substantial enough to be measured in lives: it is credited with saving somewhere between twelve and fifteen per cent of Allied casualties who would otherwise have died or lost limbs to infected wounds. Fleming, Florey and Chain shared the 1945 Nobel Prize in Physiology or Medicine for the discovery and its development into a working drug, and the underlying mechanism, blocking the enzymes bacteria use to build their cell walls so that the cell takes on water uncontrollably and bursts, has held up as an accurate account of how the drug works and became the model for an entire later class of antibiotics. Dorothy Hodgkin’s 1945 confirmation of penicillin’s chemical structure using X-ray crystallography, work that later earned her its own Nobel Prize, further cemented the drug’s chemistry as solid, well-understood science rather than an empirical curiosity.
Resistance, noted before the drug even shipped
What penicillin’s success did not include was any lasting immunity from bacterial adaptation. Chain and Abraham themselves found in 1940, before the drug had even reached wide use, that some bacteria, including certain strains of E. coli, produced an enzyme called penicillinase capable of breaking the drug down before it could act. Resistant strains of other bacteria followed as use expanded; penicillinase-producing gonorrhoea, for instance, was identified by 1976. The underlying reason resistance keeps recurring is structural rather than accidental: bacteria reproduce fast enough that any mutation offering even a partial survival advantage against an antibiotic spreads quickly once that antibiotic is in wide use, and heavy use in both medicine and agriculture, with livestock accounting for the majority of global antimicrobial sales, keeps supplying exactly the selective pressure that favours resistant strains.
A predictable crisis, not a mysterious one
The consequences of the resistance Fleming anticipated are now measured in a global mortality toll approaching penicillin’s own wartime death toll in scale. Bacterial antimicrobial resistance was directly linked to an estimated 1.27 million deaths worldwide in 2019, and more recent estimates put deaths associated with resistance at close to five million a year, with roughly one in five of those deaths occurring in children under five. Projections put the potential annual toll as high as ten million deaths by 2050 if the trend continues unaddressed, alongside enormous projected healthcare costs. The same overuse patterns identified as driving resistance, misapplied hospital prescriptions, unnecessary outpatient antibiotics, and heavy non-therapeutic use in livestock, sit squarely within human control, which is precisely why the problem is generally described as a policy and behaviour failure rather than an unavoidable biological inevitability.
One story, not two
This is worth understanding as a single continuous story rather than two separate ones, a triumphant discovery followed by a worrying but unrelated modern problem, because Fleming himself flagged the core issue within his original observations, decades before resistance became a public health crisis. The material rewards attention to the manufacturing side of the story as much as the science: penicillin’s wartime scale-up, from a struggling laboratory curiosity to billions of units produced annually within a few years, shows how much of medicine’s practical impact depends on production engineering rather than discovery alone. Readers should come away understanding resistance not as a mysterious modern failure but as the predictable other half of a story that began the moment penicillin started killing bacteria selectively rather than universally.