In early January 1941 Florey was ready to test penicillin on humans. The first English patient to whom the drug was administered was a young woman whose cancer was beyond treatment and who had agreed to test penicillin’s toxicity. She showed an alarming reaction—trembling and sharply rising fever. However, Abraham was able to show that impurities in the drug, not the drug itself, had caused the adverse reaction. In February a policeman became the first patient with an infection to be treated with penicillin in the hope of achieving a cure. No one knew the dosages and the length of treatment required to eliminate various bacterial infections; these parameters were being worked out by just such trials—primitive by today’s standards. The policeman’s condition at first improved with the penicillin therapy and then relapsed. The penicillin supply had almost run out, and even retrieving penicillin from the man’s own urine (a commonly used procedure in the early clinical trials) failed to save him. Florey vowed that from then on he would always have enough penicillin to complete a treatment. Increasing production and yields now became of overriding importance. Because Penicillium mold requires air to grow, it was first surface-cultured in regular laboratory flasks. Soon all manner of vessels were being used, including hospital bedpans and hundreds of made-to-order ceramic pots. The operation quickly outgrew the space assigned to the Dunn labs, and neighboring facilities at Oxford were borrowed for the duration. More personnel had to be hired, including six “penicillin girls” who handled the culture pots in the cold room of the extraction plant. Florey had constructed a veritable penicillin factory within the precincts of the ancient university, an institution that had stood proudly aloof from industry for centuries. On the other hand, when Chain urged that a patent be sought on penicillin, as was usual in German research institutes, Florey refused to enter into such a commercial agreement on a discovery he presumed would benefit all mankind—a decision that long rankled Chain. To increase penicillin supplies, Florey approached various British pharmaceutical firms, but only ICI considered itself in a position to accept the challenge (though many later joined the effort). British pharmaceutical firms were already committed to manufacturing other drugs needed for military and civilian populations, or, worse, their facilities had been devastated by enemy bombardment. To obtain the assistance of the United States, then still a noncombatant, in increasing production and furthering research, Florey and Heatley flew across the Atlantic in the beginning of July 1941. Florey’s American connections served him well. The two English emissaries spent the Fourth of July weekend with a friend from his Rhodes year, who put Florey and Heatley in contact with the U.S. Department of Agriculture’s Northern Regional Research Laboratories (NRRL) in Peoria, Illinois, where large-scale fermentation processes were being actively studied. A. N. Richards, Florey’s old laboratory director at the University of Pennsylvania, had become chair of the Committee on Medical Research in the Office of Scientific Research and Development, which was organized to marshal the strength of the Allies. Because Richards knew Florey’s character, he decided to expedite unified action on penicillin on the basis of just one presentation. At the height of the program, the British-American penicillin effort involved thousands of people and some 35 institutions: university chemistry and physics departments, government agencies, research foundations, and pharmaceutical companies.

The Japanese scientist Susumu Tonegawa received the 1987 Nobel Prize in Physiology or Medicine for revealing the clever way in which a relatively small number of genes could create so many possible antibodies. Working in the Basel Institute of Immunology in the 1970s (which at the time was headed by Nils Jerne), he found that individual antibodies are assembled on a biological ‘production line’ from several genes. Each gene that encodes the heavy and light protein chain components are unlike regular, single genes; they are instead made up of many units, like a string of pearls. To create an antibody, one unit or 'pearl' from each component gene is selected randomly and stuck together to form the finished product. As a result of this selection and assembly process, millions of possible combinations can be produced.