The Moldy Cantaloupe That Made Penicillin Possible

How did a moldy cantaloupe change penicillin production?

A moldy cantaloupe found at a Peoria market yielded the Penicillium strain later known as NRRL 1951, the top performer in a worldwide search and the parent of all strains used commercially. Who actually found that cantaloupe is still uncertain, and the USDA’s own account of the “Moldy Mary” story leaves the question open.

The USDA Agricultural Research Service published this account of the mystery. It is a historical review of archived documents, notebooks, and news articles, not a laboratory study, and it is careful about what the records can and cannot establish. The strain itself is not in doubt: Penicillium rubens strain NRRL 1951 is the bluish-green mold that helped scientists develop methods to mass-produce penicillin during the closing years of World War II. On August 17, 2021, Illinois Governor J.B. Pritzker signed a bill designating strain NRRL 1951 the official state microbe of Illinois.

The person who found it is the part that remains unsettled. Three accounts survive, and the USDA does not choose between them.

What the source documents:

The strain’s origin: The mold that stood out among several thousand specimens came from atop a cantaloupe found at a Peoria market, not from the soils of a distant locale
The systematic search: Kenneth Raper, head of the NRRL’s culture collection, asked the Army Transportation Corps for help, and that service collected several thousand new mold specimens from flight destinations to non-combat regions of the world; other US researchers and contributors sent specimens too
Why a new strain was needed: Specimens derived from Alexander Fleming’s original Penicillium notatum, brought to Peoria by two University of Oxford scientists in 1941, did not respond well enough to the improved growing conditions, and yields under the submerged culture conditions required for industrial scale-up fell short of practical wartime production needs
Commercial descent: NRRL 1951 was the top performer and became the parent of all strains used commercially

The USDA account documents how an intensive, global search for a superior Penicillium strain ended at a local produce market, and how incompletely the human side of that search was recorded.

Dr. Kumar’s Take

I find the uncertainty here more interesting than the tidy version of the story. Mary Kay Hunt (later Stevens) is the best-known account. She worked as an NRRL laboratory technician in 1943 and played a visible role collecting mold specimens from produce markets and other local sources and isolating them for research. Raper, Fennell, and Coghill acknowledged her contributions in a 1944 published scientific paper. In a June 4, 1962 article she gave specific details, including cutting up the cantaloupe she found and offering it to coworkers after removing the moldy part. In that same article she referenced the wrong strain number.

Raper gave a different account in an October 1, 1976 news article, crediting a Peoria woman who handed the cantaloupe to a guard at the NRRL and then left without identifying herself. He alluded to this in an August 1945 research publication as well. A more recent review of archived documents at the National Center for Agricultural Utilization Research turned up a copy of an Abbott Laboratories magazine advertisement illustrated with a reproduction of a 1948 painting by Douglas Gorsline, showing a yellow-dressed woman at a marketplace entrance with a cantaloupe in hand. Someone had written at the top of the copied ad: the original ‘Moldy Mary’ (Mary Lou Finley Roberson). There is no conclusive information on who wrote it or how that person knew the name.

My reading is that the record is thin because of how the work was done, not because anyone was careless. A number of female NRRL technicians, along with the scientists, searched for local Penicillium specimens on food and produce, and the names of who discovered which moldy items were not always recorded in the notebooks kept at the time. The chief focus may have been the source of the specimen, its penicillin yield, and its growth characteristics.

Historical Context

Starting in the early 1940s, a large team of scientists at the then-named USDA Northern Regional Research Laboratory in Peoria, Illinois, now the National Center for Agricultural Utilization Research, conducted studies that gave rise to industrial methods for mass-producing the mold and harvesting its antibiotic. The procedure was submerged deep-tank fermentation, using a nourishing brew of corn-steep liquor and other ingredients. Andrew Moyer, Dorothy Fennell (Alexander), Kenneth Raper, and Robert Coghill were among those who achieved it.

Success did not come right away. The Oxford-supplied specimens descended from Fleming’s 1928 discovery underperformed under the submerged culture conditions that pharmaceutical companies needed for industrial scale-up. A hardier, higher-yielding penicillin strain was required, which is what set the worldwide survey in motion.

What the Research Shows

A worldwide survey program Raper asked the Army Transportation Corps for help, and that service collected several thousand new mold specimens from flight destinations to non-combat regions of the world and sent them to the Peoria scientists for testing. Additional specimens arrived from other US researchers and contributors.

Screening under industrial conditions Dorothy Fennell Alexander was an NRRL penicillin survey team member who worked to identify Penicillium strains producing high amounts of penicillin under the submerged culture conditions used by pharmaceutical companies. A surviving notebook page signed by her circles strain PS46, later known as NRRL 1951.

The winning strain PS46 was the top performer and became the parent of all strains used commercially.

Contested attribution Three accounts of who found the cantaloupe exist: Hunt’s own, Raper’s unidentified Peoria woman, and the handwritten note naming Mary Lou Finley Roberson (possibly Robeson). The USDA account presents all three and reaches no conclusion.

Why the record is incomplete Multiple technicians and scientists searched local food and produce sources, and notebooks did not consistently record who found which moldy item.

Practical Takeaways

The strain is documented; the finder is not: NRRL 1951 and its commercial descent are well established, while the identity of the person who found the cantaloupe is not
Systematic searching increases discovery chances: Several thousand specimens were collected and tested before one stood out
Local sources can yield global solutions: The winning specimen came from a Peoria market rather than a distant locale
Record-keeping shapes history: What the notebooks tracked, the source, yield, and growth characteristics, is what survives; who found the sample often does not
Credit is often collective: Accounts of the discovery appear in journal articles, academic papers, family lore, and feature articles, and the story is a coming together of ordinary citizens, laboratory technicians, researchers, and industry partners

FAQs

Who was Mary Hunt and why was she called “Moldy Mary”?

Mary Kay Hunt (later Stevens) worked as an NRRL laboratory technician in 1943, helping collect new mold specimens from produce markets and other local sources and isolating them for research. Some nicknamed the person who found the cantaloupe “Moldy Mary,” though the USDA account notes that who actually found it is uncertain.

Is it settled that Mary Hunt found the cantaloupe?

No. Hers is arguably the best-known account, and she described cutting up the cantaloupe and offering it to coworkers in a June 4, 1962 article, though she cited the wrong strain number in that same piece. Raper instead credited an unidentified Peoria woman who handed the cantaloupe to a guard and left. A handwritten note on an archived Abbott Laboratories ad names Mary Lou Finley Roberson.

Why was a new strain needed at all?

Specimens derived from Fleming’s original mold did not respond well enough to the improved growing conditions the Peoria team had created, and their yields under submerged culture conditions fell short of practical wartime production needs.

How did one moldy cantaloupe become the basis for global penicillin production?

The strain from that cantaloupe, PS46 and later NRRL 1951, was the top performer in the survey and became the parent of all strains used commercially.

Bottom Line

A cantaloupe from a Peoria market gave the penicillin program the strain it needed: NRRL 1951, the top performer in a survey of several thousand specimens and the parent of every strain later used commercially. Who carried that cantaloupe into the laboratory is genuinely unresolved, with Mary Hunt, an unidentified Peoria woman, and Mary Lou Finley Roberson all named in different sources. What the record does support is a communal effort, from ordinary citizens and laboratory technicians to researchers and industry partners, that turned Fleming’s 1928 discovery into a life-saving reality.