On September 28, 1928, Alexander Fleming, back at his bench in the Inoculation Department at St Mary's Hospital in Paddington after a holiday in Suffolk, picked up a staphylococcus culture plate he had left out and noticed that a mould had grown on it. The bacteria were destroyed. He took the plate to his assistant, Merlin Pryce, and said that is funny.
He had been looking for something like this for years. Fleming had worked on antibacterial substances since 1921, when he discovered lysozyme, an enzyme in tears and mucus that breaks down some bacteria, and which turned out not to be useful as a drug. He was a bacteriologist by training and by temperament, and he had served in the First World War in Boulogne, where he had watched men die of infected wounds treated with antiseptics that killed the tissue faster than the bacteria.
What he had found was contamination, and it was contamination of a particular kind. The mould came in through a window, because Fleming was untidy and left plates out, and it was a rare strain rather than a common one, and he identified it as Penicillium notatum. He grew more of it in a broth, filtered off the liquid, and found that it inhibited staphylococci and a range of other bacteria but not the typhoid bacillus, which is why he thought of it as a selective antiseptic.
He could not purify it. Fleming was not a chemist and the penicillin in the broth was unstable, and attempts to isolate it in the early 1930s failed. He used the crude broth as a laboratory reagent for separating bacteria in culture, published his results in the British Journal of Experimental Pathology in 1929, and left it there.
What turned it into a medicine happened in Oxford in 1939. Howard Florey, Ernst Chain and Norman Heatley took the paper up, grew the mould in large volumes, and succeeded in extracting and freeze-drying the active substance. Heatley built the apparatus: bedpans, milk churns and a ceramic vessel. In May 1940 they injected eight mice with streptococci and gave penicillin to four, and the four lived and the other four died.
Production was an American achievement. The mould was grown in deep tanks at Peoria, Illinois, on corn steep liquor, and a strain from a cantaloupe in Peoria produced far more penicillin than Fleming's. By D-Day in June 1944 there was enough for all Allied casualties. Fleming, Florey and Chain shared the 1945 Nobel Prize in Physiology or Medicine.