36 Campbell interviewed by Dr. Robert L. Noble, c. 1967; in Macleod and Campbell 1925, Campbell describes it (p. 68) as “a murky, light-brown liquid containing much sediment, which dissolved to a considerable extent on being warmed.”
37 Macleod and Campbell 1925, p. 68.
38 Banting 1940, pp. 52–3.
39 Banting, Best, Collip, Campbell, and Fletcher 1922.
40 The summary of Leonard Thompson’s medical record that I was able to obtain has a note on January 12 (or January 18) referring to an “area of induration with soft centre over left buttock.” Banting 1922 refers to sterile abscesses. Best 1922 refers to “severe local reactions”; see also Banting, Best, Collip, and Macleod 1922A in which the reference to abscesses is probably to both Thompson and the dog Marjorie.
41 Collip 1923K refers to “a few patients”; Best 1922 says “two patients”; Banting 1923A, B, and 1925 all say “three cases.”
42 Banting 1923A, B. Campbell interviewed by Robert L. Noble, c. 1967; Campbell states that three patients were given single injections.
43 Banting 1925. He also suggested in this lecture and later accounts that the effect of the test on Thompson was to cause Macleod to focus his whole attention and laboratory resources on the extract. In fact the informal decision to expand the team came before, and the formal agreement to divide the work came after this abortive and unfortunate clinical test.
44 Banting 1929. Campbell’s view, expressed in 1946, was that “The earliest results would not have convinced anyone familiar with the variations to be expected in diabetics under treatment,” but there was “some encouragement to continue.” Campbell 1946, p. 100.
45 BP, 26, desk calendar, March 13, 1923; Thompson medical record.
46 See Collip 1923K; Banting, Best, Collip, and Macleod 1922A; CP, Medical, “The Contributions of J.B. Collip to the Discovery of Insulin…”
47 See Canadian Annual Review, 1922; Bliss 1978.
48 Star, Jan. 14, 1922; Macleod 1922/78; for Greenaway’s role in covering insulin, see his 1966 memoirs. Although his first reference is to his article in March, I assume Greenaway was the reporter involved in this first scoop. He might not have been.
49 Some of the early inquiries are in the Insulin Committee alphabetical and geographical files. To be fair to the press, there were other inquiries from people who had heard of the work through doctors who had been at New Haven. For Banting’s concern about the article, see Macleod 1922/78.
50 Banting 1940, p. 51, says that Starr phoned Andrew Hunter, who corroborated Banting’s statements about Macleod.
51 Macleod 1922/78; Banting 1940, p. 51.
52 CP, privately held, undated handwritten letter listing Collip’s contributions and apparently dating the isolation of the extract on January 17. However, Alison Li, in J.B. Collip & the Development of Medical Research in Canada (2003), cites convincing evidence for the 19th.
53 CP, Special Collections, Collip to Dr. C.F. Martin, Nov. 23, 1949.
54 CP, Medicine, copy of remarks made by Farquharson at N.R.C. dinner in
Ottawa, Nov. 14, 1957.
55 Banting 1940, p. 54.
56 FP, Best to Dale, Feb. 22, 1954.
57 Noble Papers, October 1971 account. Banting’s 1922 reference was as follows:
Shortly before January 25, 1922, Dr. Collip, who had been working in the laboratory of Dr. Harding, in the Pathology Building, announced that he had developed a process by which he could obtain an extract which contained no protein and no lipase. On being asked his methods of preparation he refused to tell them. This was a breach of a gentlemen’s agreement amongst Dr. Collip, Mr. Best, and myself, as we had agreed amongst ourselves to tell all results to each other. Dr. Collip discussed this new preparation with Professor Macleod and secured the consent of Professor Macleod to keep the process a secret. I believe that Dr. Collip at this time endeavoured to patent this process, and was only prevented from doing so by Professors Macleod, Hunter, and Henderson.
A completely garbled reference to these events is also contained in a letter written to Dale on Feb. 2, 1932, by A.B. Macallum; it is in the Best file of the H.H. Dale papers at the Royal Society, London.
For years in Toronto the story of the Banting-Collip fight was the piece of insulin gossip passed on in the hushest, or most inebriated tones. An enterprising reporter for Time magazine printed a garbled version of it in Banting’s obituary on March 17, 1941.
58 The only copy of this agreement I have located is in the Banting scrapbook, BP, 48, p. 61.
59 That Fitzgerald had these fears is mentioned in the Macallum letter, note 57 above.
60 Correspondence between Toronto’s president, Sir Robert Falconer, and Alberta’s president, H.M. Tory (Falconer Papers, case 74), suggests that in late January Collip had been considering staying permanently in Toronto and was mulling over whether or not to accept a position at what Falconer called “a salary we can give him.” Whether his future lay in Toronto or not may have been on his mind the night of the fight. Perhaps he reasoned that if he shared his process the Toronto group would take it over and he would wind up back in Edmonton without either an exclusive process or a share in the glory. Such a line of reasoning would have been fairly prescient.
61 In re-creating these events my first instinct was to assume that Banting’s confrontation with Macleod about stealing his results was part of the same blow-up that started the fight, and that the Connaught agreement was the settlement of the two confrontations. Such an interpretation is not impossible, but careful study of the wording and tone of the documents, especially Macleod’s description of his meeting with Banting, suggests they were actually separate incidents a few days apart.
62 Banting, Best, Collip, Campbell, and Fletcher 1922.
63 Banting and Best 1922C.
64 Banting 1940, p. 36.
65 Ibid.
66 Banting and Best 1922C. They went on to say that they were repeating the experiment and would report further results. They never did.
67 See, for example, Banting and Best 1922C and Macleod 1926. In Banting and Best 1922B the nodule became only two mm. in diameter. The qualification is in Banting, Best, Collip, Campbell, Fletcher, Macleod, and Noble 1922, though not Banting, Best, Collip, Campbell, and Fletcher 1922. In Banting’s Nobel lecture (1925) he mentions the autopsy as failing to find any islet tissue. In his Cameron lecture (1929) the autopsy is not mentioned. In his 1940 account the autopsy reveals “a microscopic group of cells so small that it was agreed that they could not be responsible for the survival of the dog.”
68 Gilchrist, Best, and Banting 1923 dates the test as February 4. The more likely date, given in Banting, Best, Collip, Hepburn and Macleod 1922, was Feb. 17. Whether this establishes Gilchrist as the second person to be injected with Toronto’s insulin, as Gilchrist thought he was, cannot be determined.
69 Gilchrist, Best, and Banting 1923; Banting, Best, Collip, Hepburn and Macleod 1922.
70 See Banting, Best, Collip, Macleod, and Noble, 1922A; also Banting, Best, Collip, Campbell, Fletcher, Macleod, and Noble 1922.
71 Banting and Best 1922C; Macleod 1922/78.
72 Greenaway 1966; MP, Macleod to W.B. Cannon, April 29, 1922.
73 Banting, Best, Collip, Campbell, and Fletcher 1922.
74 In his 1967 interview with Robert L. Noble, Campbell stated that he and Fletcher wrote most of the paper. Banting 1922 indicates that Collip did some of the writing. For Banting’s absences, see Macleod 1922/78.
75 Banting 1940, p. 54.
76 BP, 18, Notes re Cancer Research file.
77 BP, 26.
78 Banting 1940, p. 56.
79 The clinical tests are described in Paulesco 1923B, the “aglycemia” in Paulesco 1923A. It is remarkable that there has been no discussion of these experiments, except for a passing reference in Murray 1971, in the literature generated about Paulesco’s work. The zero blood sugar observation was consistent with Paulesco’s hypothesis that the internal secretion of the pa
ncreas acted as a kind of catalyst or cement on the nutrients ingested by the body, enabling them to combine to form what Paulesco called a “plasmine” in the blood. See Paulesco 1920, pp. 301–305. In his model a zero blood-sugar reading meant that the extract was totally effective. In Paulesco 1921, by which time thousands of physicians had seen hypoglycemic reactions, he is still denying that hypoglycemia causes any abnormalities.
80 Nobel archive, Miscellaneous correspondence, 1923, Paulesco to Nobel committee 21 Dec. 1923.
81 I.C., Great Britain (General) File, Macleod to Sir Edward Shaler, Nov. 3, 1922; Sharpey-Shafer 1916, p. 128; de Meyer 1909B.
82 Macleod 1922 78.
83 Joslin 1922. The discussion was printed as a supplement to Banting, Best, Collip, Campbell, Fletcher, Macleod, and Noble 1922.
84 Campbell 1962.
85 Macleod 1922 78; also G.W. Ross account. Macleod later regretted not having insisted on Banting and Best coming to the meeting. As members of the group felt at the time, the expense excuse was somewhat thin, for both Banting and Best were being reasonably well paid. My guess is that Banting had decided to stay home in a fit of role-playing pique – the poor, humble outsider – and had persuaded Best to do the same
Chapter Six: “Unspeakably Wonderful’
1 For Collip’s responsibilities, see Macleod 1922/78. A careful reading of Best’s various accounts of his work in 1922 indicates that he did not take over direction of insulin production until after Collip left Toronto.
2 The exact date of the production failure is impossible to determine. Banting 1922 states that the supply failed on February 19. This is unlikely, inasmuch as there is no reference to any shortage in the Banting, Best, Collip, Campbell, and Fletcher paper, which has results to February 1922. In a letter written on April 29 (MP, to W.B. Cannon), Macleod states that the production failure developed after publication of that paper, which means after March 22. On the other hand, the Banting, Best, Collip, Campbell, Fletcher, Macleod, and Noble paper, delivered on May 3, states that for two months it had been impossible to secure potent extracts that could be used in the clinic. These confusing statements probably reflect a complex reality, in which the production breakdown was gradual, with the extract supply being inadequate at some times, adequate at others, the small-scale methods working when the large-scale failed (as mentioned in Macleod and Campbell 1925, p. 69), and so on, but with periods when nothing worked at all. April was certainly the cruellest month.
3 They may have exchanged letters, the letters Collip is said to have burned after hearing of Banting’s death in 1941.
4 See Banting 1922, both rough and final drafts. The most extraordinary version of the light story, apparently circulating as early as 192.3, was that Collip lost the knack because one night in the lab an unknown assailant knocked him out and stole his confidential notes. See FP, file C2, C.I. Reed to Best, April 15, 1955.
5 Eli Lilly Archives, XRDe, Clowes memorandum; Sprague 1967.
6 Collip 1923K.
7 Campbell 1916; Banting, Campbell and Fletcher 1923; Campbell 1922. In his 1916 account, nowhere else, Campbell stated that this patient was the first case they had seen of recovery, albeit temporary, from coma; because death from acidosis in humans is not strictly comparable to the condition in dogs, Campbell recalled, the incident was new and important and the final proof positive of the extract’s value for humans.
8 Banting 1940, pp. 56–57.
9 IC, Clowes to Macleod, March 30, 1922.
10 IC, Macleod to Clowes, April 3, 1922.
11 CP, private, Banting, Best, Collip, Macleod, and Fitzgerald to Falconer, April 12, 1922.
12 IC, “K” file, Kendall to Macleod, April 10, 1922; no copy of Macleod’s letter to Kendall on April 3 has been located.
13 Falconer Papers, box 81, C.H. Riches to Governors, University of Toronto, Feb. 12, 1923.
14 Ibid., insulin file; CP, private, Banting, Best, Collip, Macleod, and Fitzgerald to Falconer, April 12, 1922.
15 The best description of the teamwork leading to the regained process is in Best and Scott 1923D. A notebook in the Banting Papers contains records of the experiments Banting did with Best in April, mostly using glycerine as an extractive. These do not seem to have worked.
16 Best 1922; Hare, p. 94.
17 Macleod 1922/78; CP, private, undated hand-written letter listing Collip’s claims; in later years Best several times remembered that he had begun using acetone first.
18 Interview with Moloney, June 20, 1980. For the wind tunnel and the old fan, see also Lilly archives, Gene McCormick interview with Best, Jan. 17, 1969. The best description of the process of manufacture is in Best and Scott 1923D. The formula is also to be found in the Lilly archives, XRDgb, Lilly research notes, volume 1. A handwritten copy of the formula also exists in the privately held Collip Papers. It is incorrectly dated, probably by Collip himself, December 22, 1921. He probably added this date many years later and mistakenly gave the day of his great success with the glycogen experiment.
19 Exactly how the Christie Street situation developed is unclear. An April 3, 1922, Memorandum to the minister, by the Director of Medical Services, Dr. W.C. Arnold, recommending the establishment of the clinic:, is in the PAC;, RG 32, C2, vol. 13, Banting personnel file. Arnold recommended establishment oí the clinic as an excellent and justifiable step by the ministry. But there is some evidence that he had known Banting earlier and wanted to help him out.
20 No copy of this agreement has been found, but it is referred to in the Insulin Committee Minutes, Aug. 17, 1922.
21 Gilchrist, Best, and Banting 1923; Banting 1929.
22 Havens family papers, James Havens, Sr., to E.C. Gale, March 1, 1921.
23 Williams 1922.
24 All of the documents from the time suggest that Havens was the first. In a letter to Best on March 13, 1939 (BI, Best Papers, Historical file), however, Williams casually mentioned that he began giving extract to one Lyman Bushman, a veteran, on May 14, 1922. The Havens family correspondence shows this cannot have been true, and was a slip of Williams’ pen or memory. Circumstantial evidence suggests that Bushman was first given insulin in June or July.
25 Havens family papers, James Havens, Sr., to George Snowball, May 24, 1922. Also Woodbury 1962: this account of the Havens case written for a mass magazine heavily emphasizes dialogue and drama and contains several clear factual errors.
26 Havens family papers, S.B. Cornell to James Havens, Sr., June 15, 1922.
27 BP, Havens File, Williams to Banting, June 5, 1922.
28 IC, Universities file, John Howland to Macleod, May 13, 1922; also MP, R. Carrasco-Formiguera to Macleod, May 19, 1922.
29 IC, Woodyatt file, Woodyatt to Macleod, May 10, 1922; reply May 15; Clowes to Macleod, May 11; reply May 15.
30 IC, Universities file, Macleod, Fitzgerald, Banting, and Best to Falconer, May 25, 1922. There seems to be no significance in Collip not having signed this letter. He may have been out of town when it was sent.
31 This account of Lilly and Clowes draws upon Eli Lilly archives, XRAe, Clowes’ research reports; XCAe, J.K. Lilly, “A Plan for promoting the affairs of Eli Lilly, 1920–1923”; Dr. G.H.A. Clowes, Jr.’s biographical study of his father; and a centennial publication about Eli Lilly and Company, Kahn 1976.
32 IC, Universities file, Macleod et al. to Falconer, May 25, 1922. Early drafts of the Lilly agreement are in the Banting Papers and the CP privately held. A copy of the final agreement, not differing in any important detail from the drafts, is in the Eli Lilly archives. For Clowes’ early proposals, see IC, Clowes to Macleod, March 30, 1922.
33 Banting, Best, Collip, and Macleod 1922A.
34 Eli Lilly archives, XRDgb, Research notes, vol. 1.
35 IC, Potter file, W.D. Sansum to Macleod, June 15, 1922.
36 IC, Committee of Clinicians file, Allen to Macleod, June 24, 1922.
37 IC, Potter file, Macleod to Sansum, June 21, 1922; Woodyatt file, Macleod to Woodyatt, June 21. Macleo
d may have had second thoughts about this policy, for a week later, replying to Allen’s request, he did not send details, but simply promised a reprint of the paper giving the method when it was published. This became his standard reply to similar requests in the next several months. Of course it was also possible to learn the method from Collip. It is said, for example, that Woodyatt actually learned how to make insulin from a conversation with Collip.
38 Banting 1940, p. 33.
39 BP, 1, F.A. Hartman to Banting, May 2, 1921; see Williams 1947 and BP, Havens file, J.S. Havens to Banting, June 17, 1922, for George Eastman’s interest in Banting coming to Rochester. John Harvey Kellogg had also offered Banting a job in his sanitarium in Battle Creek.
40 Falconer Papers, box 76, Falconer-Blackwell correspondence; Banting 1940, p. 61a.
41 Banting 1910, pp. 61a-63.
42 Falconer Papers, 76, Falconer to Blackwell, June 19, 1922; BP, Falconer to Banting, June 29, 1922; MP, folder 342, Graham to Macleod, July 14, 1922.
43 Best family papers, M.M. Best scrapbook, Banting to Best, July 15, 1922; Toronto Star, Feb. 24, 1923.
44 BP, Elizabeth Hughes file, A. Hughes to Banting, July 3, 16, 1922; MP, A.S. Ferguson to Macleod, July 31, Aug. 9, 1922.
45 Havens papers, Banting to James Havens, Sr., July 10, 18, 1922; Gilchrist, Best, and Banting 1923; Banting, Campbell’, and Fletcher 1922, p. 550; also Lilly archives, XRDc, John R. Williams to G.H.A. Clowes, Jan. 27, 1958: “One day Fred Banting took me up to the Christie Street military hospital where there were 8 soldiers each suffering horribly with large abscesses in hips-buttocks. I was having same trouble with Jim Havens and 3 other cases I had here.”
The Discovery of Insulin Page 38