Source / Laboratory culture and preserved object
At St Mary's Hospital in London in September 1928, Alexander Fleming investigated a contaminated culture plate on which a mould colony was associated with a clear or translucent zone among staphylococcal colonies. The observation mattered because he cultured the mould, tested the antibacterial filtrate and published the results in 1929—not because the dish already contained a usable medicine.
The plate must now be read through several related but non-identical sources: the photograph in Fleming's paper, a dried plate and model catalogued by the British Library, descendant mould cultures, later presentation samples and museum replicas. Together they document experiment and commemoration; none is a transparent snapshot of a solitary, instantaneous discovery.
Reading the culture
Fleming was studying variants of staphylococci, bacteria that form visible colonies on nutrient agar. In the plate described in his 1929 paper, a contaminating mould colony lay beside an area where the bacterial colonies had become transparent. He interpreted this as lysis—the breaking down of bacterial cells—caused by a diffusible product of the mould.
The effect is legible because growth farther from the mould can be compared with the altered colonies nearby. Distance matters: a substance released by the fungus spread through the agar and its concentration fell away from the colony. The plate therefore offered a spatial clue, not a chemical identification.
Fleming made subcultures of the mould and tested filtered culture broth against different organisms. He reported strong activity against staphylococci and several other bacteria, but little or none against others. “Penicillin” initially named the antibacterial filtrate, not a purified compound of known structure. His experiments also showed that activity declined during storage.
Fleming thought the already visible colonies were dissolving, partly by analogy with his earlier work on lysozyme. Oxford researchers later showed that penicillin acts most effectively on susceptible bacteria while they are growing and dividing. Ronald Hare's archival and experimental reconstruction argued that the mould probably began producing penicillin before visible staphylococcal growth developed. The plate remains evidence of inhibition, but it cannot by itself fix the exact sequence.
Fleming's own tests showed a selective spectrum: some organisms were susceptible and others were not. The dish says nothing by itself about absorption, dose, toxicity in ill patients, allergic reactions, manufacturing quality or whether a particular infection would respond. Those questions required different experiments and clinical evidence.
From observation to paper
September–November 1928: the exact day and sequence rest mainly on later recollections; the relevant notebook pages do not survive. Fleming subcultured the mould, while the earliest surviving dated experimental entry identified by Hare is 30 October. The familiar story of spores blowing through an open window was told much later and is not documented by a contemporary record. Earlier researchers had already described antagonism between moulds and bacteria, so the plate was not the first observation of microbial competition.
December 1928–spring 1929: Fleming directed further work with the young researchers Stuart Craddock and Frederick Ridley. Craddock's notebook records attempts to concentrate the unstable antibacterial material; the results did not produce a stable purified drug. Fleming also consulted the St Mary's mycologist Charles La Touche about the mould's identity. Their labour is easy to lose when the plate is presented as the product of one glance.
10 May–June 1929: Fleming's article was received on 10 May and published in June. Its title emphasised using penicillin to isolate B. influenzæ, then the name used for what is now Haemophilus influenzae. Because that organism resisted penicillin while many accompanying bacteria did not, the filtrate could act as a selective culture medium. The paper briefly proposed possible treatment of susceptible infections, but reported no controlled therapeutic trial.
1930s: Fleming and other researchers retained and shared the mould and continued limited investigations, but instability, low yield and purification problems obstructed systemic therapy. Saying that penicillin was simply “forgotten” is too strong; saying that Fleming had already invented the finished drug is equally misleading.
Object, image and organism
Fleming's article reproduces a labelled black-and-white “photograph of a culture-plate.” The image fixes a transient biological pattern on paper and places it inside an experimental argument. Its publication date and purpose are secure even though the caption itself does not narrate the later holiday-and-window story.
The Fleming Papers catalogue, Add MS 56209, describes a “bacteriological plate containing penicillin” used in the first production of penicillin, together with a model, and dates it to 1928. The culture was fixed with formalin and has dried to a thin residue. That treatment preserved an object for handling and display but ended the living culture and limits later biological analysis.
In 2005 microbiologist Milton Wainwright examined the sealed plate non-destructively from below. He saw fungal hyphae but no diagnostic spores or spore-bearing structures, and no sample could be removed. His 2006 report therefore raised possibilities rather than identifying the surviving material. The plate's catalogue attribution should not be converted into a modern species diagnosis.
Fleming first published the mould as Penicillium rubrum; it was long known as P. notatum and later grouped with P. chrysogenum. A 2011 molecular taxonomic study reidentified cultures maintained as Fleming's strain (including CBS 205.57/NRRL 824) as P. rubens. This finding concerns cultured descendants with a documented laboratory lineage, not a genetic test of the sealed plate itself.
From plate to therapy
1939–1940: at the Sir William Dunn School of Pathology, Oxford, Ernst Chain and Howard Florey selected penicillin for a wider programme on antibacterial substances. Norman Heatley developed an assay and extraction system; Margaret Jennings, Arthur Gardner, Jean Orr-Ewing, Gordon Sanders and other colleagues contributed biological testing, bacteriology and production. The seven-author August 1940 Lancet paper reported that penicillin protected experimentally infected mice.
1941: the Oxford laboratory became a small production facility, using purpose-made ceramic vessels and extensive technical labour. The team's August 1941 report described growth, extraction, assay, animal work and early human treatment. It named the investigators' roles and acknowledged ten technicians whose work made adequate supplies possible. The favourable responses were compelling, but tiny stocks restricted treatment and made recovery of active penicillin from patients' urine worthwhile.
1941–1944: Florey and Heatley sought United States support. The US Department of Agriculture's Northern Regional Research Laboratory, wartime agencies and pharmaceutical companies improved culture media, mould strains, extraction and aerated deep-tank fermentation. By 1944, industrial production—not replication of Fleming's plate—made penicillin available in quantities useful to Allied medicine. Civilian access remained uneven and followed different national schedules.
1945 and after: the Nobel Prize recognised Fleming, Chain and Florey, a defensible but incomplete allocation of credit. Penicillin's effectiveness established the emblematic drug of the antibiotic era, while selective susceptibility and laboratory resistance were already visible in early research. The plate's later fame records this therapeutic afterlife; it did not predict all its benefits, limits or the problems created by access, prescribing and resistance.
Display and historical memory
The Science Museum Group catalogues two imitation culture plates made around 1963. Their coloured mould and clear zone offer museum visitors a legible reconstruction after the original culture has dried and darkened. Their date and material distinguish a twentieth-century display technique from the 1928 experiment.
From the mid-1940s, Fleming made presentation medallions from cultures of the penicillin-producing mould, while laboratories deposited and exchanged living strains. Such objects may descend from a culture he maintained without containing the agar, bacterial strain or moment represented by the famous plate.
A dried culture flask from about 1940–41 and a vial of penicillin recovered from patients' urine, now at Oxford's History of Science Museum, document the scarcity and material ingenuity of early therapeutic production. They help prevent one London plate from standing in for the entire development process.
Across the collection
Place the 1928 observation within Fleming's work on bacteriology, wounds, antiseptics, lysozyme and laboratory technique.
Follow the sequence from observation and publication to animal experiments, early patients and manufacture.
Set penicillin within antimicrobial research, pharmaceutical production, access, prescribing and resistance.
References
British Journal of Experimental Pathology 10, no. 3 (June 1929): 226–236. PMCID: PMC2048009. The contemporary experimental report and published plate image. It documents what Fleming reported in 1929, not the later story's exact discovery date or every step remembered after penicillin became famous.
Catalogue record for a bacteriological culture plate and accompanying model, dated 1928. The record establishes present archival description and custody; it does not by itself resolve the identity of all material remaining in the sealed dish.
Medical History 26, no. 1 (1982): 1–24. DOI: 10.1017/S0025727300040758. An archival reconstruction using Stuart Craddock's notebook and colleague testimony. Hare distinguishes records from inference, but some timing and mould-source conclusions remain reconstructions rather than contemporary facts.
Microbiologist 7, no. 3 (September 2006): 28–30. A first-person report of non-destructive examination of the British Library plate. Its negative and ambiguous findings explain why the preserved dish cannot currently supply a secure species identification.
IMA Fungus 2, no. 1 (2011): 87–95. DOI: 10.5598/imafungus.2011.02.01.12. A molecular taxonomic reassessment of maintained strains, including the culture attributed to Fleming.
The Lancet 236, no. 6104 (24 August 1940): 226–228. DOI: 10.1016/S0140-6736(01)08728-1. The seven-author Oxford report of protection in experimentally infected mice.
The Lancet 238, no. 6155 (16 August 1941): 177–189. DOI: 10.1016/S0140-6736(00)72122-2. The Oxford team's detailed contemporary report on culture, assay, production, animal experiments and early human treatment, including named roles and technical acknowledgements.
An institutional history of the Oxford programme, including Heatley's assay and extraction work, the wider team, mouse experiments, early resistance observations and chemical-structure research.
An International Historic Chemical Landmark account of Oxford production, the US Department of Agriculture laboratory at Peoria, government coordination, industrial fermentation and wartime scale-up.
Collection record 1963-374/2, explicitly catalogued as one of two imitation culture plates. It is evidence for later display and commemoration, not the 1928 experiment itself.
Object record MG.M-06668 for a preserved mould sample from Fleming's culture. The medallion was made between the mid-1940s and 1953 and given to the Smithsonian by Fleming in 1953, making it a presentation object rather than the 1928 plate.
Records for an Oxford culture flask from about 1940–41 and penicillin recovered from patients' urine on 25 June 1941, documenting the material constraints of early clinical production.