Although always a bit of a loner, Banting craved male companionship and female affection. He and a classmate, Bill Tew, who had also begun practice in London, spent “about five nights out of seven” commiserating about the practice of medicine. The camaraderie must have been marred slightly for Fred by the realization that Tew’s practice was developing better than his. On weekends he would see Edith, but there was trouble there too. The successful female teacher was making three or four times as much money as the uncertain, insecure male doctor. Apparently they were not as sure of their love as they had been in 1916: the farmer’s son who had been to the wars and the minister’s scholarly daughter, gold medallist in her class, may have realized they had changed over the years. Edith may have been developing other interests. “I was very unhappy and worried,” Banting wrote about that early period in London.8
Studying was another way to pass the time. Banting resumed preparation that he had broken off several years earlier to take the difficult exams for a fellowship in the British Royal College of Surgeons. He also got a part-time job in October as a demonstrator in surgery and anatomy at London’s Western University. Western’s faculty of medicine was small and undistinguished, but had a few good professors and a promising future based on an ambitious building program. As well, Banting could also use the $2 an hour he was paid. He soon began assisting Dr. F.R. Miller, Western’s very good professor of physiology, in occasional experiments in cerebral and cerebellar localization.9
II
On Sunday, October 31, 1920, Banting spent several hours preparing a talk he had to give to physiology students on carbohydrate metabolism. Neither the topic, nor the associated disease, diabetes, were subjects in which Banting had any particular interest. Stories grew up later about diabetic school chums having had a profound influence on him, and so on, but actually Banting had never treated a diabetic patient and had no interest in the dietary treatment of diabetes. There had been one brief mention of it in his therapeutics lectures at university, an up-to-date suggestion that physicians not be afraid to use Allen’s starvation treatment. “I remember seeing one patient only on the wards of the Toronto General Hospital,” Banting wrote years later. “I heard of people mostly well on in life dying in coma and believed there was nothing one could do…. There was no such thing as a diabetic in any ward in my surgical experience…. I did not even know that my friend and class-mate, Joe Gilchrist, had diabetes until I had been working on the problem for many months.”10
His copy of the November issue of the journal Surgery, Gynecology and Obstetrics had just arrived.11 When he had finished work in the evening he took it to bed to read himself to sleep. With carbohydrate metabolism on his mind, he was naturally interested in the leading article in the issue, an analysis of “The Relation of the Islets of Langerhans to Diabetes with Special Reference to Cases of Pancreatic Lithiasis,” by Moses Barron.
Barron was an American pathologist who became interested in the pancreas and the islets of Langerhans when, while doing routine autopsies, he came upon a rare case of the formation of a pancreatic stone (pancreatic lithiasis). Rarer still, the stone had completely obstructed the main pancreatic duct. Studying that pancreas, Barron found that while all the acinar cells had disappeared through atrophy, most of the islet cells had apparently survived intact. A review of the literature showed that these observations were similar to those arising when pancreatic ducts were blocked experimentally by ligation. Both experimental evidence, then, and this interesting new piece of pathological evidence, seemed to reinforce the hypothesis, held by many others, that the health of the islets was the key variable in the genesis of diabetes.
Barron wrote up his modest study, presenting it as another bit of evidence in the search for an explanation of diabetes. The work was perhaps of special interest because of the similarity between experimental and clinical cases. He probably submitted it to Surgery, Gynecology and Obstetrics because of the interest surgeons had in stones.
Barron’s was a useful, not brilliant or trail-breaking study (his next paper, on the relationship between smoking and lung cancer, was a much more important pioneering contribution). His review of the literature was not particularly wide-ranging; his interpretation of his own case was questionable inasmuch as the patient had had islet cell damage and was diabetic (Barron attributed the damage to causes other than the obstruction). The sole importance of Barron’s article in the history of medicine is that Fred Banting happened to read it in the evening of a day he had been thinking about carbohydrate metabolism.
Banting’s most detailed description of his reaction to the Barron article is in his 1940 memoir, “The Story of Insulin”:
It was one of those nights when I was disturbed and could not sleep. I thought about the lecture and about the article and I thought about my miseries and how I would like to get out of debt and away from worry.
Finally about two in the morning after the lecture and the article had been chasing each other through my mind for some time, the idea occurred to me that by the experimental ligation of the duct and the subsequent degeneration of a portion of the pancreas, that one might obtain the internal secretion free from the external secretion. I got up and wrote down the idea and spent most of the night thinking about it.12
This account should put an end to the story that the idea came to Banting in a dream. Beyond that, it is impossible to re-create the train of Banting’s thought as the lecture and the article chased each other through his mind. Speculation about what Banting “must” have thought is hazardous because all his accounts of his inspiration came only after his life had been changed by its consequences. As will be seen repeatedly in this history, Banting was not a precise and reliable guide to the events in which he participated.13 A clear example of this is the fact that he never afterwards checked his own notebook to find out exactly what he had written at 2:00 a.m. on the morning of October 31. He quoted himself from memory, and always incorrectly. In what is taken as his most authoritative statement of the history of the discovery of insulin, the Cameron Lecture in Edinburgh in 1928, Banting recalled, “I arose and wrote in mynote-book the following words -
Ligate pancreatic ducts of dogs. Wait six to eight weeks for degeneration. Remove the residue and extract.”14
The notebook is in the archives of the Academy of Medicine in Toronto. Banting actually wrote these words:
Diabetus
Ligate pancreatic ducts of dog. Keep dogs alive till acini degenerate leaving Islets.
Try to isolate the internal secretion of these to relieve glycosurea15
The obvious comment that Banting didn’t know how to spell “diabetes” (and “glycosuria”), let alone treat the disease, is a bit unfair. It will cheer the modern medical student to know that Banting had never been a good speller. He never became one. In any case, the spelling in a person’s private notebook ought to be his own business.
The more interesting aspects of the actual notation, as opposed to the remembered one, are twofold. First, the true notation does not contain the word “extract.” None of the documents written in the first six months after Banting conceived his idea contain that word. All Banting wrote down was the idea of ligation, waiting for degeneration (he may not have known how long it would take), and then, “try to isolate the internal secretion.”
Second, Banting wanted to try to isolate the internal secretion “to relieve glycosurea.” He seems to have been identifying diabetes with glycosuria in the traditional way, rather than referring to the newer notion of hyperglycemia as the important condition to be relieved. The possible significance of these points will become clear shortly.
III
The morning after he wrote down his idea Banting mentioned it to Professor Miller at Western. Miller was a neurophysiologist and knew little about research in carbohydrate metabolism. It sounded like a good idea, he told Banting, but surely someone had tried it before. Banting may have asked Miller if it was possible to mount a project at Western to try out
the idea. Miller apparently replied that facilities were not available for that kind of work; there were no quarters for large animals, such as dogs, in the old medical building, for example. Besides, Banting ought to consult someone who could be more helpful. Fortunately an expert was close by, in Toronto where J.J.R. Macleod had been professor of physiology since coming over from Western Reserve in 1918. Miller advised Banting to talk to Macleod.
Banting also consulted the professor of pharmacology at Western, J.W. Crane, who knew of no work on the subject. Banting remembered going to the library to look up the literature and finding nothing. This is a surprising statement, even allowing for inadequacies in Western’s medical library. It indicates either bad memory by Banting or an inability to search medical literature properly. That night Banting also had a long talk with Bill Tew. Part of his excitement came from his having seized upon the idea as a way out, something worth dropping practice for.16
It happened that Banting was going to be in Toronto the next weekend for the wedding of one of C.L. Starr’s daughters. At the reception or afterwards, Banting told a number of his acquaintances from surgery about his idea. “I wished to give up practice in London immediately and commence work,” he wrote. “They all advised against such a radical move.”17
Banting saw Professor Macleod, whom he had not known except by reputation, in Macleod’s office the next day, Monday, November 8. Shy and inarticulate at the best of times, Banting could not have been at ease. Macleod, on the other hand, was a very senior, very articulate professor, giving up a few minutes of his valuable time as a courtesy to a University of Toronto graduate whom he had never met.
Banting told Macleod that he was interested in doing research work to search for the internal secretion of the pancreas. We do not know how much or how little he knew about previous researchers’ attempts to find it (the survey in the Barron article actually said very little). He may have known enough to mention to Macleod, and if not Macleod certainly mentioned to him (“this point immediately came up in our discussion,” Macleod wrote later) that many others had tried to prepare an extract of pancreas which contained the internal secretion. They agreed that the problem with such extracts may have been that they also contained the powerful digestive ferments of the external secretion, and that these may have acted to destroy the internal secretion.18 Banting said that perhaps the use of duct-ligated pancreases would get around that problem, because (as the Barron article showed) the effect of duct-ligation would be to destroy the cells producing the digestive ferments.
“He was tolerant at first,” Banting wrote of Macleod at that meeting, “but apparently my subject was not well presented for he commenced to read the letters on his desk.” “I found that Dr. Banting had only a superficial text-book knowledge of the work that had been done on the effects of pancreatic extracts in diabetes,” Macleod wrote in 1922, referring to their first several meetings, “and that he had very little practical familiarity with the methods by which such a problem could be investigated in the laboratory.”19
Macleod was confronted with a young surgeon who had walked in virtually off the street, had no significant experience in physiological research, and was talking, haltingly, about a topic he knew about only from standard textbooks and one article. As any conscientious professor in that situation would do, Macleod told Banting that many eminent scientists had spent years, sometimes their whole lives, in well-equipped laboratories working on the problem of the pancreas. They had not even proven conclusively that there was an internal secretion to be found, let alone found it. Research such as Banting proposed could not be undertaken lightly. Indeed it would be “useless” to attempt it without making a full-time commitment for several months.20
It is not clear from the documents exactly what Banting was suggesting to Macleod. His later statements that he had asked Macleod for ten dogs, an assistant for eight weeks, and facilities for doing blood and urine tests, are not contained in his or Macleod’s 1922 accounts. Banting is maddeningly vague in 1922: “I told him carefully what I had planned… I then repeated my ideas to him.”21
It may have been that Banting was canvassing the idea of duct ligation as a surgical technique for isolating the internal secretion. Well, what would you do then, after the ligation had caused the pancreas to degenerate? What is the next step? How do you go about proving you have the internal secretion? Macleod might have asked. By making an extract of it and giving it to a diabetic animal, Banting might have answered. On the other hand, his notebooks show that some months later, when the work actually began, Banting and Macleod proposed to graft a portion of atrophied pancreas into a diabetic animal as their first experimental approach.22
Grafting had been mentioned briefly in Barron’s article (extracting had not been mentioned); since Hédon’s work in the early 1890s this had seemed the surest way of proving that there was an internal secretion of the pancreas. In 1913 Allen had complained that “deplorably little” had been done in the way of experiments with grafts and pancreatic transplants, implying that it was a very interesting avenue for future research.23 This may have been the avenue Banting was suggesting to Macleod, or it may have appealed to Macleod as more promising than playing around with extracts. Grafting might have been discussed as a technique for isolating the internal secretion; extracting might have been discussed; both methods might have been discussed.
Whatever technique they were talking about, Macleod started to become interested. Banting remembered repeating his ideas to the professor:
…he sat back in his chair with closed eyes for some time. Then he began to talk. He thought that “this might be the means of getting rid of the external secretion.” As far as he knew this had never been tried before. “It was worth trying” and “negative results would be of great physiological value.” This latter phrase he repeated at least three times.24
Speculation is in order here and is permissible because we have some idea of Macleod’s knowledge of the literature. Whether he and Banting were discussing grafting or extracting, what must have appealed to Macleod as never having been tried before was the idea of somebody experimenting with degenerated or atrophied pancreas. Now there was nothing new in the idea of producing degeneration or atrophy of the acinar tissues by ligating the pancreatic ducts – all sorts of researchers had done this. Their interest, however, had been almost entirely in measuring the relative amounts of degeneration that took place in the various components of the pancreas, particularly the relative changes in the acinar and islet cells. Nobody, it seemed (except, perhaps, Lydia Dewitt, with her unsatisfactory in vitro testing methods), had taken a pancreas in which the acinar tissue had been induced to atrophy and tested to see if it contained the alleged internal secretion. Nobody had either tried to prepare a graft or administer an extract using a fully degenerated pancreas. And yet, theoretically, if there was an internal secretion, and if it did come from the islets of Langerhans, and if it was the acinar cells but not the islets that degenerated after the ducts were ligated, and if two or three other conditions held good, then perhaps some interesting results would follow.
Even if the results were negative, it was the kind of experiment that ought to have been tried long ago, if only for completeness’ sake; in that sense “negative results would be of great physiological value,” probably valuable enough, for example, to write up for publication. Another possible consideration for Macleod might have been the thought that almost all experiments done in the past, with pancreatic extracts or by any other method, might show different results now that blood sugar could be tested easily and quickly. So Dr. Banting, superficial as his knowledge might have been, halting as his presentation undoubtedly was, had produced a suggestion worth thinking about.
On second thought, there was the great difficulty researchers had commonly had in getting ligation to work so that the pancreas atrophied. E.L. Scott, it will be remembered, had reasoned the same way as Banting and had wanted to try an extract made from degenerated pancreas,
but had found the technique impractical. Scott had been a student; Banting was a surgeon. Macleod may have reasoned that a surgeon would succeed where a student had failed (and, of course, Banting’s skills would be particularly important if the discussion was actually centring on transplants or grafts). They may also have discussed the possibility of failure, and/or other ways of attacking the problem of getting rid of the external secretion. Macleod wrote in 1922 that either at this meeting or at a later one, “I suggested freezing the pancreas and then extracting it at the lowest possible temperature with alcohol” – which was essentially Scott’s technique. Assuming Macleod’s memory on this point is correct (it seems to be supported by Best’s 1922 account and was never explicitly denied by Banting), the suggestion passed by as part of the general discussion. As we will see, it became important a year or so later.
Banting had explained that he could not do the work at Western and wondered about the possibility of coming to Toronto to attempt it. Having cautioned Banting about the time it would take, and about the likelihood of negative results – however interesting the idea was – Macleod said that yes, if Dr. Banting wanted to come to Toronto, Macleod would take him into the lab.25
The warnings had their effect on Banting. “I was not inclined to give up appointments in Surgery and Physiology in London to get ‘negative results of great physiological importance’.” He told Macleod he would consider the whole matter carefully.26
IV
Before returning to London, Banting explained the whole situation to Dr. Starr, who functioned, Banting told Macleod, as his “father advisor.” Starr said he would think it over and write giving advice. Two weeks later Banting wrote Macleod that he was still anxious to do the work, but was still waiting to hear from Starr.27
The Discovery of Insulin Page 7