Few people know that in parallel with the race to produce a bomb that would kill people by the hundred thousand, scientists in England ran another race, a race to produce a drug that was to save the lives of millions. The book under review is the story of Alexander Fleming, the laconic Scotsman whose chance discovery set the race in motion. While working as a bacteriologist at a London hospital in 1928, Fleming found that a culture plate seeded with staphylococci had become contaminated with a mold. Instead of discarding it, as others might have done, he noticed something unusual: colonies of these bacteria had grown everywhere except near the mold, where he saw a clear patch. He now cultured the mold and discovered that the broth filtered from it stopped the growth of several kinds of deadly bacteria. Publication of his discovery in a scientific journal stirred up hardly a ripple, and he did little more about it.

Nine years later Ernst Chain, a young German biochemist at Oxford, came across Fleming’s paper and decided together with his professor, the Australian pathologist Howard Florey, to find out the nature of the active substance in Fleming’s broth and the way it stops the growth of bacteria. With herculean labor, Chain and several of his colleagues extracted a minute quantity of what they believed to be the pure substance from gallons of broth and gave it to Florey to test.

In May 1940, while the defeated British Army was being evacuated from France, Howard Florey and his team achieved a brilliant victory. Florey had injected lethal numbers of streptococci, the kind of bacteria that causes blood poisoning, meningitis, and puerperal fever, into eight mice. He then injected the mold extract into four of them and left the other four untreated; the treated mice remained healthy and the untreated ones were dead the next day. These were the first steps in the purification of penicillin and the realization of its unparalleled therapeutic powers. As a result Fleming became a world hero, while Florey, Chain, and their colleagues’ names have remained unknown outside the world of science.

Macfarlane has written a scientific thriller tracing the almost unbelievable combination of events that guided Fleming to his discovery. He then shows why no one, not even Fleming himself, recognized the importance of his discovery and why Fleming, even though he had abandoned it to others, reaped nearly all the fame. Macfarlane’s book is the companion volume to his earlier, brilliant biography of Howard Florey,1 and is equally enthralling.

Fleming was graduated as a physician at London University in 1906 and became junior assistant in Sir Almroth Wright’s Inoculation Department of St. Mary’s Hospital in London. His later specialization in bacteriology was brought about by World War I when Wright and his assistants were posted to a grim military hospital in Boulogne. Here Fleming found that the antiseptics like carbolic acid, then commonly used for the treatment of open wounds, killed the white blood cells that constitute the body’s own defense and let the bacteria that had buried themselves in the tissues survive. According to Macfarlane, this experience made him aware of the need for an antiseptic that penetrates the wounds and leaves the beneficial white blood cells alive.

In 1921, Fleming made the first of the two observations that were to change medical history. Fleming’s notebooks, now at the British Museum, fail to describe how it happened; Macfarlane therefore quotes the account of an eyewitness at St. Mary’s Hospital, where Fleming worked in the innoculation department.

Early on, Fleming began to tease me about my excessive tidiness in the laboratory. At the end of each day’s work I cleaned my bench, put it in order for the next day and discarded tubes and culture plates for which I had no further use. He, for his part, kept his cultures…for two or three weeks until his bench was overcrowded with 40 or 50 cultures. He would then discard them, first of all looking at them individually to see whether anything interesting or unusual had developed. I took his teasing in the spirit in which it was given. However, the sequel was to prove how right he was, for if he had been as tidy as he thought I was, he would never have made his two great discoveries—lysozyme and penicillin.

Discarding his cultures one evening, he examined one for some time, showed it to me and said “This is interesting.” The plate was one on which he had cultured mucus from his nose some two weeks earlier, when suffering from a cold. The plate was covered with golden-yellow colonies of bacteria, obviously harmless contaminants deriving from the air or dust of the laboratory, or blown in through the window from the air in Praed Street. The remarkable feature of this plate was that in the vicinity of the blob of nasal mucus there were no bacteria; further away another zone in which the bacteria had grown but had become translucent, glassy and lifeless in appearance; beyond this again were the fully grown, typical opaque colonies. Obviously something had diffused from the nasal mucus to prevent the germs from growing near the mucus, and beyond this zone to kill and dissolve bacteria already grown.

Fleming found that saliva, tears, and the white of eggs also dissolved the bacteria, and that these fluids did no harm to white blood cells, which made him wonder if animals themselves manufacture the ideal antiseptic for which everyone had been searching. This proved a false hope, because Fleming soon found that his nasal mucus or tears left the common disease-producing bacteria unharmed; the bacteria that had dissolved proved to be of a unique kind, blown into his room from no one knew where. A classical scholar named them Micrococcus lysodeikticus and the unknown agent in Fleming’s nose “lysozyme.” Fleming continued to work on it for many years in the hope that it might prove of therapeutic value, even after his discovery of penicillin, but not knowing biochemistry, he never found out what it was or how it worked. This omission was to prove crucial for later developments. Lysozyme was later found, not by Fleming, but by Ernst Chain—who was to be one of the other chief actors in the penicillin drama—to be an enzyme that acts by dissolving the membranes of invading bacteria.


If the discovery of lysozyme was made possible only by the adventitious landing on Fleming’s bench of a rare and hitherto unknown germ, his discovery of penicillin in 1928 was due to a combination of circumstances improbable beyond belief, of which Fleming’s own terse description in the British Journal of Experimental Pathology in 1929 provides no hint:

While working with staphylococcus variants, a number of culture-plates were set aside on the laboratory bench and examined from time to time. In the examinations these plates were necessarily exposed to the air and they became contaminated with various micro-organisms. It was noticed that around a large colony of a contaminating mould the staphylococcus colonies became transparent and were obviously undergoing lysis [i.e., being broken up].

Bacteriologists normally grow microorganisms by cooking a nutrient broth, pouring it on a round dish four inches wide, letting it solidify into a jelly, stabbing the jelly many times over with a platinum wire dipped into an earlier culture of the microorganism, and finally heating the dish for a day or so in an incubator kept at body temperature.

Many years after the event, Ronald Hare, who had been assistant in the inoculation department at the time, tried to rediscover penicillin by preparing a culture plate seeded with staphylococci in just this way and then contaminating it with Fleming’s mold.2 The mold had no effect! To produce the clear patch that Fleming had found, Hare had to seed the dish with the molds before the staphylococci, but there he encountered another difficulty: the molds would not grow at body temperature. So what could have happened?

In 1928 Fleming was asked to write a chapter on staphylococci for a handbook, and for this purpose tried to reproduce some anomalous strains reported in the literature; he was helped by a student, D.M. Pryce. As I have said, Fleming tended to prepare many bacterial cultures and leave them scattered on his bench. Pryce told Hare that before going on vacation, Fleming had pushed all those cultures together into a corner to give Pryce space to work in. Pryce himself later went on vacation with Hare, and Fleming returned before they did. When Pryce came back Fleming had piled up his cultures on a tray of antiseptic. Fleming picked up some of them and showed them to Pryce, who remembers that “he took one plate up, looked at it, and after a while said ‘That’s funny.’ ” That was the now famous plate referred to in the opening paragraph of Fleming’s paper.

To explain what happened, Hare excavated the temperatures recorded in London during the summer of 1928 and assumed that instead of incubating his cultures before going on vacation, Fleming had just left them on his bench. The record shows that the August temperatures were in the 60s, favorable to the growth of only the molds, after which they rose to the 70s, suitable for the growth of the cocci; they grew everywhere except near the molds, which must have exuded a substance inhibiting their multiplication. But where did the mold come from? It turned out to be a very rare organism, unlikely to have flown in through the window, which in any case Fleming rarely opened.


Some years earlier a Dutch allergist had given lectures in London advancing the now accepted theory that some patients suffer from asthma because they are allergic to molds. As a result, the head of Fleming’s department at St. Mary’s, Sir Almroth Wright, appointed a young Irish mycologist, C.J. La Touche, to isolate molds from houses inhabited by asthma patients so that the molds could be identified and extracts made from them to desensitize the patients. Because molds produce myriads of airborne spores, mycologists normally grow them under hoods fitted with extract filters, but Wright favored the British string-and-sealing-wax tradition of research and made La Touche grow his molds in an open makeshift laboratory furnished only with tables. It so happened that this laboratory was immediately underneath Fleming’s. It was La Touche who identified Fleming’s mold as belonging to the Penicillium family. La Touche is not sure that it came from his collection, but both Hare and Macfarlane regard it as the most likely source. Hare concludes:

Such, then, is what I conceive to be the background to the discovery of penicillin. An accidental observation it is true, but what an accident, depending as it did on a whole series of apparently unrelated events. The choice of Fleming to write a chapter in a book; the publication of a paper in a scientific journal that prompted him to enquire further; lectures by a Dutch physician that led to the appointment of a mycologist; his working in a laboratory directly beneath that of Fleming; his having the good fortune to isolate a powerful penicillin-producing strain of the mould; his having inadequate apparatus so that the atmosphere became loaded with spores; the high probability that Fleming either forgot to incubate his culture plate or purposely omitted to do so; the fact that Fleming’s own laboratory was peculiarly sensitive to outside temperatures; that a cold spell came at a time of the year which is usually unsuitable for the discovery; the visit to Fleming by Pryce that led the former to look again at a plate he had already inspected and discarded; and its having escaped destruction because of entirely inadequate methods for the disposal of used culture plates. All these events, acting in concert, brought to Fleming’s notice a phenomenon that cannot, even now, be reproduced unless the conditions in which the experiment is carried out are exactly right. Had only one link in this chain been broken, Fleming would have missed his opportunity. And if, as Paul Ehrlich used to say, scientific discovery depends partly on Geld or money, partly on Geduld or patience, partly on Geschick or skill and partly on Glück or luck, it was the last of them that was almost entirely responsible for the discovery of penicillin. It was, surely, the supreme example in all scientific history, of the part that luck may play in the advancement of knowledge.3

Hare omits to mention that luck thrives only in a culture medium containing an ingredient of untidiness; hence the discovery could never have been made in Ehrlich’s German laboratory.

Fleming’s first announcement to his colleagues at St. Mary’s was anticlimactic, as was his subsequent lecture to a wider audience. Again, according to Hare:

He had evolved the custom of indulging in a morning potter which included a visit to the big laboratory in which I was working. Fleming’s idea of gossip was different from that of most other people. It usually involved his planting himself in front of the fireplace with his hands in his pockets, a cigarette dangling from his lips and looking more or less into space. On rare occasions he would give utterance, usually in the fewest possible words. The information doled out so meagerly might concern anything; that so-and-so had died; that what’s-his-name had made a fool of himself again; or how are your Snia Viscosa shares doing?…

On this occasion it was neither Fleming’s notion of gossip nor high finance that emerged. It was the now famous culture plate that led to the discovery of penicillin. The rest of us, being engaged in researches that seemed far more important than a contaminated culture plate, merely glanced at it, thought that it was no more than another wonder of nature that Fleming seemed to be forever unearthing, and promptly forgot all about it.

Fleming found that his broth inhibited the growth of the streptococci and staphylococci that infected wounds and also of the organisms responsible for gonorrhea, meningitis, and diphtheria, but not the growth of typhoid, paratyphoid, anthrax, and Hemophilus influenzae, also called Pfeiffer’s Bacillus, the organism that was (incorrectly) believed to have caused the great influenza epidemic after the First World War. The broth was harmless to white blood cells, could be injected with impunity into mice and rabbits, and the mold itself eaten without ill effects.

Having done these experiments, Fleming failed unaccountably to take the obvious next step, the step that Florey was to take twelve years later, namely to find out whether an injection of his broth would protect mice from lethal infection. (Macfarlane calculates that the concentration of penicillin in Fleming’s broth would actually have been sufficient to make the experiment work.) Instead, Fleming applied the broth externally to a few patients, with mixed results, and got discouraged. Even more disheartening were the findings of two young doctors who tried to extract the active principle from the broth; they found that it lost its activity quickly if kept in water or alcohol, and even more rapidly in blood (which was later proved incorrect); on injection into animals it was eliminated in the urine in less than two hours, while its antibacterial activity took about four hours to develop. A contemporary scientist commented, “It was like filling a bath with the plug out.” Fleming gained the disappointing impression that penicillin would have no more clinical value than he thought lysozyme had, but he found another use for it and that use is the only one mentioned in his first lecture on penicillin, which bore the unbelievably low-key title “A medium for the isolation of Pfeiffer’s Bacillus.”

The inoculation department subsisted not on the charities that in those days supported St. Mary’s Hospital but, like the Pasteur Institute in Paris, on the sale of vaccines. Their manufacture had been pioneered by Wright, who had now passed responsibility to Fleming. When preparing vaccines against Pfeiffer’s Bacillus it was difficult to keep the cultures clear of the ubiquitous staphylococci; Fleming found that addition of penicillin did so. This minor technical advance formed the subject of Fleming’s lecture to the Medical Research Club in London, attended by Sir Henry Dale, a leading physiologist and the astute director of the National Institute for Medical Research, founded for the specific purpose of developing chemotheraphy.

Fleming’s talk fell flat and aroused no comments or discussion, because according to Hare “he was one of the worst lecturers I ever heard, unable to express himself clearly…. In a dull monotone, without humor or emphasis, [his lectures] may have given the audience the facts.” Macfarlane recalls that he was often inaudible and “worse still, gave the impression that he had little enthusiasm for his own subject.” However, Fleming’s use of penicillin had one vital consequence: he ordered that henceforth the broth be produced in the inoculation department in weekly batches, and he gave subcultures of his Penicillium to several colleagues in other laboratories, including Georges Dreyer, the professor of pathology at Oxford whose collaborator, Miss Campbell-Renton, kept the cultures going for the next ten years. Fleming himself made no more mention of penicillin in any of his twenty-seven papers and lectures published between 1930 and 1940, even when his subject was germicides.

Penicillin might thus have been forgotten but for Fleming’s earlier discovery of lysozyme. For it was this discovery that led Florey and Chain to investigate the therapeutic value of penicillin. Howard Florey was an Australian who had migrated to England in 1922, and on Dreyer’s death was appointed professor of pathology at Oxford…Macfarlane, who knew both Fleming and Florey, writes that there could not have been a greater contrast between the two men. Both were extremely able, but while Fleming was easygoing, laconic, had little ambition, and was popular, Florey was “taut like a coiled spring,” worked like a dynamo, and made enemies.

Florey suffered from chronic indigestion, which aroused his interest in the composition of mucus secreted by the gut and other tissues. After his arrival at Oxford, he engaged the young German refugee biochemist Ernst Chain, and, since lysozyme is found in mucus, Florey suggested to him that he should find out the biochemical mechanism of lysozyme’s attack on bacteria. Chain soon solved this problem and then wondered whether lysozyme might not be just one representative of a large class of bactericidal substances occurring in nature. He collected references to about two hundred papers, going back as far as Pasteur, who had been the first to point to the great therapeutic possibilities of “bacterial antagonism.” Many years afterward Chain wrote, “When I saw Fleming’s paper for the first time, I thought he had discovered a sort of mould lysozyme.” He can’t have read it very carefully, because Fleming states that penicillin dissolves in alcohol, whereas Chain knew lysozyme to be a protein, and all proteins are insoluble in alcohol. Perhaps this was another one of those strokes of luck, without which penicillin would have remained in obscurity, because Chain then proposed, and Florey agreed, to make penicillin part of a thorough study of natural antibacterial substances.4

The heroic efforts by Chain and his associate Norman Heatley to produce even a few thousandths of a gram of penicillin from the mold broth are vividly described in Macfarlane’s present book and also in his earlier biography of Florey. The instability of penicillin in solution, which had dogged Fleming’s collaborators and several others who followed them, was overcome by the newly invented technique of freeze-drying, in which a solution of penicillin was first frozen and then water vapor was pumped away and condensed at a very low temperature. Having discovered that penicillin could cure lethally infected mice, Florey was determined to find out what it could do in men, but that needed a three thousand times larger dose, requiring two thousand liters of mold filtrate. He and his team worked night and day to turn their university laboratory into a factory. When wartime shortages reduced the supply of large shallow dishes to culture the molds in, Florey improvised with hundreds of bedpans. To augment his grant from the British Medical Research Council, he appealed for support to the Rockefeller Foundation in New York. When publication of the spectacular results of his first clinical trials in The Lancet failed to persuade the British authorities and pharmaceutical firms to put a major effort into the production of penicillin, he went to the United States where he found what he wanted: enthusiasm, money, know-how, and professionalism. Florey’s mission set in motion a vast machine which soon produced penicillin on a scale thousands of times greater than Florey’s own “factory,” and in much purer form. Chain’s five thousandths of a gram of broth concentrate injected into each of Florey’s four mice was later found to have contained no more than one part of penicillin in three hundred parts of impurities, so that Florey did his test with only 1/1,500,000 gram of penicillin for each mouse. By 1945 American production had risen to one hundred kilograms a month of pure crystalline penicillin, enough to treat all Allied war casualties.

Although the credit for initiating the work that led to the isolation of penicillin and to its clinical trials belongs to Chain and Florey, fame blew in through Fleming’s door like one of his microbes. After the publication of the second clinical trial by Howard Florey and his wife Ethel, The Times of London carried an editorial about the Oxford work, but without mentioning names. Almroth Wright thereupon wrote a letter to The Times pointing out that the laurel wreath should be placed on Fleming’s brow, because he had discovered penicillin there. Reporters immediately besieged the laboratory in search of Fleming, and interviews with him appeared in the press. Not to be outdone, Sir Robert Robinson, professor of chemistry at Oxford, followed with a letter to The Times, pointing out that Florey’s team deserved at least some grateful acknowledgment. This drew a troupe of reporters to Florey, but he refused to see them and even forbade any member of his staff to talk to the press.

To Americans this must seem unbelievable, but in Britain at that time scientists seeking “cheap publicity” in the daily press were thought by their colleagues to debase themselves and their lofty profession. The disappointed pressmen therefore returned to Fleming who told them about his own earlier work and also about the achievements of the Oxford team. They romanticized the former and ignored the latter, so much so that when Fleming, Florey, and Chain shared the 1945 Nobel Prize for Physiology and Medicine, only Fleming made the headlines, while Florey and Chain appeared in small print.

Fleming spent the remaining ten years of his life collecting twenty-five honorary degrees, twenty-six medals, eighteen prizes, thirteen decorations, the freedom of fifteen cities, and honorary membership of eighty-nine scientific academies and societies. A friend of his told Macfarlane that he collected honors as a schoolboy collects stamps and was delighted at any rare acquisitions. Effusive admirers soon hailed him as the greatest scientific genius of all time, and he became the subject of several hero-worshipping biographies, including one by André Maurois which both Hare and Macfarlane criticize as misleading.5

Macfarlane’s last chapter gives his assessment of Fleming as a scientist, but even without that the book contains enough material for the reader to make his own. Great discoveries are not always made by great thinkers. Some are made by skilled craftsmen, some by observant watchmen, and some even by prosaic people doing a regular job because they are paid for it. Perhaps the most important lesson that scientists can learn from Macfarlane’s book is that the solutions to some of our great problems may be staring us in the face and that we may be too blind to see them. When Macfarlane chose Florey as the subject of a biography, he found a colorful, complex, eloquent, and forceful man to portray. Fleming had none of these attributes, and whatever thoughts he may have had were left unexpressed. Such people do not make rewarding subjects unless one embellishes them, as earlier biographers had done. Macfarlane has painted an honest picture and has yet contrived to write an absorbing story; it is a piece of medical history, less about the man than about the subtle and sometimes ironic interplay of science, chance, and personalities that is the stuff of which discoveries are made.

Macfarlane does not mention the first great hurdle that clinicians using penicillin were soon to encounter. Fleming himself had found penicillin-resistant variants among his cultures. Widespread and often indiscriminate use of penicillin caused these variants to spread and multiply among several kinds of bacteria, notably the staphylococcus that produced an epidemic among infants and surgical patients in hospitals. Allergic reactions to penicillin also caused difficulties.

Ernst Chain helped to overcome these problems by initiating the development of several chemically modified penicillins.6 Together with the naturally occurring form, these penicillins have remained effective against syphilis and many other, although not all, forms of gonorrhea; against some forms of meningitis and pneumonia; and against many streptococcal and staphylococcal infections. The penicillins were no help against the once terrible scourge, tuberculosis, but the publication of Fleming’s first clinical trials of penicillin started the Ukrainian-born American Selman A. Waksman on a search for antibiotics produced by organisms in the soil, which led him in 1944 to the discovery of streptomycin, a compound active against TB. This further success soon set off a worldwide search for other antibiotics, as a result of which most bacterial infections can now be successfully treated. Waksman was awarded the Nobel Prize but none of the discoverers of other antibiotics ever acquired the fame that Fleming did.

This Issue

March 27, 1986