The Discovery and Manufacture of InsulinBy Gene E. McCormick, Historian

Eli Lilly and Company, 1971

At the annual meeting of the American Physiologi-cal Society held during the Christmas holidays of 1921at New Haven, Connecticut, a young Canadian physi-cian reported that he and an associate had isolated fromthe pancreas an internal secretion that lowered theblood-sugar level of depancreatized dogs to normal andthat by periodic injection of the substance and specialdietary measures, they had maintained the animalsdiabetic free for several weeks.(1)

The discovery of insulin by Dr. Frederick GrantBanting and his student assistant, Charles Herbert Best,was profound at that point in time in the field of medicalscience. Their work conclusively demonstrated that thepancreas, by internal secretion, serves a direct functionin carbohydrate metabolism, and, thus, a thirty-yearsearch of international scope to find the elusive, hypo-

thetical hormonewas culminated.

The discoveryof insulin alsoestablished theunitary nature ofdiabetes and,consequently,brought togetherthe observationsand explorations ofmany who hadstruggled over themillenniums tounderstand thedisease and as-suage its ravages.(2)

Diabetes wasnot a widespreaddisease at the timeBanting conceivedhis surgical methodof isolating insulin—

a method he was unaware others had tried. For by1920, life expectancy was too short to manifest themajor incidence of its maturity onset. But the diseasewas vicious and implacable. Little could be done tocheck the victim’s unquenchable thirst, excessiveurination, gnawing hunger and gradual wasting away.Although the starvation diets devised a few years beforeBanting’s discovery did prolong the diabetic’s lifesomewhat—a little more than two years on the average—the suffering they imposed was almost as cruel as thedisease itself. This was as far as the nutritional conceptof diabetes therapy had evolved after 200 years of trialand error to restore balance in the bodily conversion offood into energy.

In fact, until the advent of insulin, diabetes gener-ally was a fatal disease. The body, unable to metabolizecarbohydrates, turns to its fats as an energy source and,in the process of burning them in conversion, producesby-products that eventually smother internal processes.At this stage, the body is forced to derange its preciouschemical balance to counteract the disruptive conges-tion of its environmental pathways. This radical effortto survive brings its final destruction. The heavyBest and Banting

Patient JL before and after insulin

1

gulping of air by the unconsciousvictim to expel the smotheringketones of burning fat cannotstave off the fatal coma. Thus, 50years ago, the conquest of diabe-tes was defined as the preventionof coma, and the unusual severityof the disease in children gaveparticular urgency to that con-quest. The death throes of diabe-tes most frequently occurred inthe young.

Scientists and physicians theworld over fully appreciated theneed of specific therapy fordiabetes. As late as 1920, whilemany noted authorities wereskeptical about the existence ofan internal pancreatic secretion—and in face of the numerousfailures to isolate it since the turnof the century—there were otherswho were intrigued by indica-tions increasingly coming to lightsince the work of OskarMinkowski and Joseph vonMering in 1889.(3) Dr. G.H.A.Clowes was among the latter,and aware of the investigation ofBanting and Best almost fourmonths before their first report tothe outside world at New Haven,he was in the audience to hearwhat Banting had to say. Im-pressed by the experimentalconclusiveness reported on theaction of insulin and with authori-zation from Eli Lilly, the vicepresident of Eli Lilly and Com-pany, he promptly offered toBanting and Professor James J.R.Macleod, head of the departmentof physiology at the University ofToronto, under whose auspices

Banting had pursued his work,the services of the pharmaceuti-cal company for the large-scaleproduction of the vital hormone.

It was not until the middleof the following May, however,that Macleod accepted the offerof collaboration on behalf of theuniversity.(4) Aware of the impor-tance of the discovery of insulinto patients with diabetes, andlikewise, the obligation to pre-vent fraudulence and exploita-tion, the University of Torontohad much to do before makingcommitments of any kind.Testing of the extract in humancases had not started at the timeof Clowes’ overture; develop-ment of manufacturing proce-dures from laboratory methodshad barely begun; and the meansto administer control of productefficacy, clinical trials and licens-ing had to be resolved.

During this period, theuniversity took measures toestablish an advisory group of itsrepresentatives and officials ofthe Connaught Antitoxin Labora-tories affiliated with the univer-sity to oversee the developmentand distribution of an effective,standardized agent. It alsoobtained, in gratuity patentrights, that which had beenapplied for on insulin and itsprocess of manufacture. Vestedas an official body responsible tothe university for the insulinprogram, the advisory groupwas formally established as theInsulin Committee toward theend of 1922.

Macleod and Clowes

Leonard Thompson

2

While theseaffairs were inprocess, the firstclinical trial ofinsulin was yield-ing encouragingresults. On Janu-ary 11, 1922,Banting and hisassociates at theMedical Service ofthe TorontoGeneral Hospital,Drs. Walter R.Campbell andAlmon A.Fletcher, hadadministeredinsulin to LeonardThompson, a 14-year-old boysuffering from

juvenile (severe) diabetes. The initial injections did notproduce notable response, but with the use of morepotent extracts administered two weeks later, there wasmarked improvement, and the boy soon was restoredto normal life. He died 11 years later from bronchop-neumonia resulting from a motorcycle accident.

Six additional patients were put on clinical trial aslaboratory supply of the precious extract increased. Ina preliminary report published in March, the investiga-tors said, “It is difficult to putinto words what is meant byclinical improvement.”(5)

Tentative though it was, theadvance was unique. For thefirst time in man’s existence,there was promise of liberationfrom this dreaded disease.“With bright news fromToronto,” said Dr. Elliott P.Joslin, eminent Americandiabetologist, parents who hadscrupulously made their

children follow strict dietary discipline were nowrewarded with “a hope for life, which they hardly daredto anticipate.”(6)

Attempts at Toronto to devise a standard extrac-tion procedure for large-scale manufacture were notsuccessful however. From March to May, there was noproduction of acceptable material. Dr. James B. Collip,professor of biochemistry on leave from EdmontonUniversity who had been engaged by Macleod todevelop process standards, declared he was unable tosolve the problem of potency loss and instability thataccompanied increased yields and withdrew from theproject.(7) The fact that seven diabetics were nowsupported by insulin was, of course, a crucial matter.But in the wider view, it was also important to theuniversity to establish reliable extraction standards forthe purposes of licensing—since it recognized that it hadneither the funds nor the facilities to support the manu-facture of insulin on a scale equal to general demand.Accordingly, the university concluded that develop-ment work must now turn to outside assistance, and onMay 15, Macleod invited Clowes to come to Toronto todiscuss the collaboration proposed at New Haven.(8)

Two weeks later an agreement was committed to paperas an “indenture” pertaining to experimental produc-tion and clinical supply and was signed by the univer-sity and the company. The agreement stipulated,among several things, that:

l Lilly had one year in which to develop anagent complying with university standards, during

which time the company hadexclusive right to make, use, andsell the extract in the United States,Mexico, Cuba, and Central andLatin America only

l batches of the productacceptable to both parties on thebasis of approved testing could bedistributed on a cost basis or gratisto physicians and institutionsselected by agreement betweenboth parties

Dr. Elliot P. Joslin

Laboratory where Banting and Bestdiscoverd insulin

3

l Lilly was to provide the university 28 percentof each approved lot of insulin when distributed gratisand 12 percent when supplied at cost

l all clinical reports received by Lilly were to besubmitted to the university and company use of themwould be governed by the discretion of the universityfor both professional and promotional purposes

l there was to be full collaboration between thetwo parties on the divulgence of production methodsand testing and on improvements either might make, inwhich case if Lilly developed patentable methods, itwas to assign U.S. patents in gratuity to the university“upon being requested to do so”

l after successful conclusion of the experimentalperiod, Lilly was to be granted a license and pay a fivepercent royalty on net sales to the university

l the university agreed not to divulge to othersany informationprovided by Lillywithin the timelimit of the inden-ture and assuredthe company thatthe licensing ofother firms wouldbe on the sameterms as it wouldreceive.(9)

In the devel-opment of safeand effectivemedicine, thestipulations of theagreement werewithout precedentand, as far as isknown, signifiedthe first voluntary,cooperative

endeavor among an academic institution, the medicalprofession, and a commercial house to apply a majortherapeutic advance on an international scale.

Clowes lost no time. He assigned George B.Walden, the supervisor of insulin production, andplaced under him Harley W. Rhodehamel and Jasper P.Scott, thus forming a team of three to carry out theconcurrent projects of experimental and factory-scaleprocessing.(10) Based on protocols by Collip and Best—and some modifications of its own—Lilly began experi-mental runs almost daily during early June and on thenineteenth of that month sent Banting its first insulin, ashipment of 50 units, which he found satisfactory.(11)

The first factory-scale lot was made on June 26 and thesecond on July 5—the latter amounting to a yield of 30units from 75 pounds of fresh hog pancreas and apotency of one unit per cc.

By the end of July, production averaged 1,200units a week as a result of various modifications ofalcoholic concentrations, temperatures, extractionsteps, and lot combinations, and potencies as high astwo units per cc. were being achieved. By the first ofSeptember, the company had produced a cumulativetotal of 5,390 units, of which 2,985 had been shipped toToronto, or a little more than twice the percentagespecified in the indenture.(12)

Connaught production, on the other hand, wasThe Toronto agreement

Jasper P. Scott in his laboratory

4

large-scale production wouldamply fill the needs of the 16Canadian and American clini-cians now supplied by Lilly andthe commitment to Banting of500 additional units per week.The exhausting, frantic effort oneveryone’s part, however, wasnot rewarded. The final productfrom the new process deterioratedbadly—50 percent and more—which meant that, in spite ofgreater yields, production would

have to be doubled to make up the loss. In addition,clinicians began to report extremely variable patientresponse to different lots and incidence of abscesses, induration and sensitivity at the injection site—all of whichdenoted a highly impure product.

In an effort to overcome production problems ingeneral, Walden worked himself to the verge of col-lapse, and Clowes insisted he take two weeks off. Scott,who took over in Walden’s absence, did suffer a break-down and remained off the job for three weeks.(14)

Because of all the setbacks, J.K. Lilly, Sr., instructedClowes to releaseWalden from experi-mental work andassign him exclu-sively to pressingproduction needs.Clowes also decidedto return to the initialextraction protocolsince house samplesfrom that processhad retained fullpotency since June.

It was believedthat the problem ofdeterioration in thelarge-scale methodwas the cumulativeresult of errors notunlikely to occur in

beset with difficulties and whenBanting visited J.K. Lilly, Sr.,president of the company, inIndianapolis in the latter part ofJuly, he said there was not a dropof insulin in Toronto. Now that hehad charge of a large clinic,wrote Mr. Lilly to his son, Eli,“He certainly was in trouble. Wehad 150 units ready for him andwhen I told him that he couldtake it back with him Mondaynight, he fell on my shoulder andwept, and when I told him that on Tuesday evening wewould send him another 150 units, he was transportedinto the realm of bliss. Banting is really a fine chap andwe must back him to the limit.”(13)

The night and day work Walden and his team hadgiven to the insulin program since the first of Junebrought highly encouraging results by September. Theamount of active material being extracted reached newlevels of 90 to 120 units per pound of gland and totaloutput of September was as much as had been pro-duced in July and August combined. In August, equip-

ment for large-scaleproduction hadbegun to arrive andRhodehamel—whohad been bendingfull-time efforttoward layout offacilities in Building20—was busy withinstallation that,hopefully, would bein regular operationby the first of Octo-ber. It was expectedthat, based on theexperience of previ-ous increasing yieldsand potencies, theoutput from theshakedown phase of

Early insulin

Harley W. Rhodehamel George B. Walden

5

initial development attempts. Inbrief, Lilly was encountering thesame difficulties that had ha-rassed Collip and which werestill frustrating production atConnaught. It was apparent toboth parties that potency loss inthe final product had to besurmounted before productionand clinical work could beadvanced.

Out of this necessity came astartling revelation. In noting thatstability was dependent upon thehydrogen ion concentration ofthe solution from which the finalproduct was obtained, Waldendiscovered that most of thedeterioration did not result fromdestruction of the insulin hor-mone in processing but rather from the slow formationof the precipitate in the final product solution. Thesedeterminations were fundamental and, moreover,radical, for they revealed that insulin activity was not inthe solution—the premise upon which both Connaughtand Lilly extraction methods were based—but in thevery precipitate heretofore considered a contaminantand discarded during the extraction process. Carryingthis finding further, Walden found that the rate andextent of the formation of the hormone-bearing precipi-tate was governed by the hydrogen-ion concentration ofthe solution, and this suggested that, by pH adjustmentof the solution, an optimum pointcould be established at which theinsulin hormone would split offfrom the bearing precipitate. Inother words, it was discoveredthat iso-electric precipitationcould be employed for thepurification of insulin. Its nitro-gen content could be significantlyreduced, and equally important,as high as a 90 percent recoveryof insulin activity became pos-

sible. A way had been found toachieve high purity with largeyield.(15)

What Walden unearthed inthe exasperating month ofSeptember 1922 was followed bymany weeks of trial and error inapplication before the highlypure material could be producedin large volume. But in early1923, production problems hadbeen largely overcome. By Aprilof that year, the weekly rate wasrunning at a little more than180,000 units—a sharp contrast tothe 2,500 rate seven monthsearlier. Potency also had risen to20 units or more per cc.

Such progress did not comewithout financial burdens. A

large investment had been made without assurance ofreturns for some time or of protection against obsoles-cence. The matter of arranging for and maintainingproper icing of glands at a time when artificial refrig-eration was still in its infancy imposed heavy costs inshipping and storage. To assay insulin required manyhundred rabbits, and yet the few animal suppliers therewere could not consistently provide animals of uniformsize and weight or assure normal health—factors thatvitally affected test data. And feeding and caring for theanimals under proper conditions required specialprovisions on a scale the company had never faced

before. During the latter half of1922, more than 100,000 rabbitassays were conducted.

Determination of the valueof the unit of insulin, highlycomplicated by wide variabilityof animal response, was a subjectof great debate between Lilly andthe Insulin Committee over aperiod of several months. Whileit was not technically possible atthe time to derive precise value—

Grinding glands c. 1924

First insulin products

6

which was the crux of the debate—it, nevertheless, was abasic factor interlocking the economics of yields,capacity projections, and future facility demands. In thefirst six months of production at Lilly, unit standardswere modified four times. Believing it had absorbed allthe expense it could, the company requested andobtained Toronto’s approval to sell insulin to cliniciansat cost, and distribution on this basis began in lateJanuary 1923.(16) Shipment was direct to physicians, butbilling was handled throughthe drug trade in keepingwith the company’s estab-lished policy. At the time,Lilly was supplying sixtyclinics in Canada and theUnited States.(17)

The extension of theclinical program to thislevel within a year after thefirst administration ofinsulin by Banting seemsremarkable considering thesmall output of insulinduring the period. Allclinical work had beendependent upon insulin from Lilly until the end of 1922,when Connaught Laboratories finally overcame equip-ment problems and swung into initial mass production.Considering, too, that unit potency was low, relativelylarge daily dosages were necessary to maintain apatient. Since the action of the hormone was not aprecisely known entity, much precaution had to betaken in selecting physicians for clinical work. Assign-ment of specific investigators to hospital-affiliatedlaboratories in major urban areas had to be planned inorder to gather data most efficiently from a limitedsupply of insulin. The value of the unit, based on theoriginal rabbit assay unit, also had to be raised to betterindicate clinical response. All of these considerationswere worked out along deliberate lines.

At the behest of Clowes, the Insulin Committeeagreed to the creation of a team of American investiga-tors to put clinical work on a systematic basis in theUnited States according to committee requirements and

over which it was to exercise authority. The desire ofthe committee to have this priority was fully appreci-ated by the Americans and carefully observed through-out the clinical program.

On November 22, 1922, a roundtable meetingwas held at Toronto, attended by the Canadian clini-cians, members of the Insulin Committee, representa-tives of Connaught and Lilly, and six American physi-cians—all noted authorities on diabetes management—to

plan distribution controlprocedures, special studies,and unit standardization. TheAmerican doctors overseeingclinical work in the UnitedStates in liaison with theInsulin Committee were:

l Dr. Elliott P. Joslin,Boston

l Dr. Frederick M.Allen, Morristown, NewJersey (after whom the AllenEra of dietary management,1914-1921, was named)

l Dr. H. Rawle Geyelin, New York (an author-ity on juvenile diabetes)

l Dr. John R. Williams, Rochester, New York(the first American physician to use Banting’s extract)

l Dr. Rollin T. Woodyatt (a brilliant clinicianfrom Chicago)

l Dr. Russell M. Wilder, of the Mayo Clinic,Rochester, Minnesota.

It was at this meeting that clinical trial develop-ment was set up to provide the greatest number ofcases for test information within the limits of availablesupply. The plan of individual patient therapy followedJoslin’s philosophy. “It has been our policy to treatmany rather than few diabetics. We have felt it more

Insulin filling line c. 1924

7

humanitarian to prolongthe lives of many old andfaithful patients rather thanattempt to secure marvel-ous results with a few.”

Drs. Banting,Campbell, and Fletcherpublished in the January 6,1923, issue of The BritishMedical Journal theirfindings on insulin therapywith 50 cases, a paper thatwas the first comprehen-sive report of the Canadiantrials, and by agreementwith the Insulin Commit-tee, the American investi-gators subsequently pub-lished their findings. Theseappeared in late May in aseries of papers in TheJournal of MetabolicResearch, which, includingthe above Canadian report,discussed experiences withapproximately 600 casessince the inception ofclinical work in January1922.(18)

Brought to light by scientific data was reliableevidence that insulin did have widespread value thera-peutically when the specific requirements of the indi-vidual patient were recognized in the administration ofthe agent. Though the number of reported cases wassmall in relation to the estimated million people withdiabetes in the United States at the time, this collectiveexperience was an enormous advance. By virtue of thestringent controls exercised in treating variable mani-festations of the disease, the basis was laid upon whichthe practicing physician could hope to bring patientswith diabetes under control, and through carefullycalculated diet, regulated insulin dosage, and properexercise, guide them toward a normal way of life. Forthe first time, there was extensive practical knowledge

available for attacking thedisease.

As one of Joslin’searly insulin patientswrote, a professor ofGreek at the University ofVermont, “If the trueChristian be the closeimitator of Christ, then thediscoverer, and the onewho applies the discovery,should feel that they areliterally following the onewho said: ‘I am come thatthey might have life, andthat they might have itmore abundantly.”(19) Alsoamong Joslin’s cases wasan indefatigable researcherat Boston who was withina few years to discovertreatment for perniciousanemia, George RichardsMinot.

By the time thefindings of the Americaninvestigators had beenpublished, the company

had distributed more than seven million units of insulinand, in conjunction with Connaught, was supportingabout 10,000 patients with diabetes. The Insulin Com-mittee in early spring of 1923 arranged that Lilly shipquantities of insulin to the British Isles to support theprogram there under patents granted to the publicdomain by the University of Toronto, and the Commit-tee also permitted broadened clinical work in theUnited States.

Lilly was allowed to submit names of physicianswho had interest in insulin and were qualified to use it,so that the Committee could rule on additional eligibili-ties. Such, in brief, were the events behind the officialJune announcement by the Insulin Committee that asafe and standardized product was available in largevolume and that firms could be licensed for the manu-

8

facture of insulin.(20) Thus, in a 12-month period, a clinical entity ofwholly new origin and value hadbeen developed into a useful and,for many, a lifesaving product.Of the role of Eli Lilly andCompany in this achievement,the Committee announcementsaid, “Without this collaboration,it is unlikely that a non-irritatingproduct of such satisfactorypotency and durability couldhave been produced in adequateamounts to meet the demand ofthe medical profession in thiscomparatively short time.”

On the last day of June, theUniversity of Toronto prepared alicense contract for Lilly for themanufacture and sale of insulin inthe countries specified in theoriginal indenture agreement,and the license was approved andaccepted by the company’s boardof directors on July 11. However,approval for the company tobegin commercial distribution did not come until thefollowing October, after the unit value disagreementhad been resolved (the Committee insisted that itsstandards had to prevail); additional clinical evidencehad been gathered; and approvalhad been given of the company’sliterature and label copy, anapproval based on rigorouscontrol.(21)

On October 15, Iletin®, thecompany’s brand name for itsinsulin, was released for distribu-tion through regular drug tradechannels available on physicianprescription.(22) Simultaneously,a price reduction was made—thethird since it was first sold on acost basis ten months earlier—

and it amounted to a 66 percentdecrease. It was estimated that7,500 physicians were treating25,000 diabetes patients withIletin. At the close of 1923, Lillyhad sold almost 60 million unitsduring the year.(23)

With the introduction ofIletin, the company prepared anextensive literature program toprovide the practicing physicianuseful guidelines on dosagedetermination, suitable potencies,dietary measures, and urine andblood analyses and also devel-oped a variety of accessories forthe patients with diabetes self-careneeds.(24) It also introduced,within the following year, newpotencies of the amorphousmaterial so that it had on themarket in 5- and 10-cc vialsstrengths of U-10, U-20, U-40,and U-80. Part of the motivebehind this attempt at widemarket development was, to be

sure, the likelihood of competition, which began tomaterialize in 1925. But more pressing was the criticalmatter of overproduction. Diabetic treatment was stilllargely confined to hospital or clinical supervision by

the very nature of the disease andso consumption did not risecommensurately with the esti-mated incidence of diabetes. Asexperience with diabetes manage-ment extended through themedical profession, however, andthe company was allowed to sellinsulin in certain other foreigncountries, demand for it rose. By1925, units distributed amountedto 217,681,150, thirteen million ofwhich went into export.(25) Pricereductions effected in 1924 and

9

1926 considerably reduced the return on the line but aprofit continued to be made as unit sales increased. Theeighth reduction was made in 1932, when unit salesreached 792,451,300, and resulted in a cumulativedecrease of 90 percent from the original price inJanuary 1923.

With the collaborative development of insulin, thebattle against diabetic coma ended and there began thelong campaign against the complications of the diseasearising with the prolonged life span of diabetic victims.The role of Eli Lilly and Company and others in thiscampaign—which has been waged in ever-wideningcircles to this day—remains to be accounted for in laterchapters; only the beginning has been given in thesefew pages. Writing to J.K. Lilly, Sr., in December 1930,

Joslin spoke of the many happy homes there were atChristmastime because of the contribution of Eli Lillyand Company, and to this, Mr. Lilly replied:

The great and refreshing discovery through thisexperience was that the really great men in any line ofendeavor are the most approachable, simple, and directin their reasoning and contacts. Both in letter and inspirit, we have endeavored to indicate to you howprecious our relations have been with you and yourassociates, and sometimes we are haunted with a fearthat probably we have not been able to do our part in asfull measure as it was humanly possible to do so. Yetyour generous words have tended to make us moresatisfied with that phase.(26)

10

(1) Dr. Frederick Grant Banting and CharlesHerbert Best first reported their findings to the Physi-ological Journal Club of the medical faculty at theUniversity of Toronto, November 14, 1921, and thepaper was first published in The Journal of Laboratoryand Clinical Medicine, Vol. 111, No. 5, February 1922.The paper delivered at New Haven was considered asummary by Banting and Best and, consequently, itspublication in the American Journal of Physiology, Vol.59, No. 1, February 1922, was not regarded by them astheir first report. Nonetheless, the Nobel Prize for thediscovery was based on this article, and since ProfessorJames J.R. Macleod was listed as co-author (because hewas the head of the department of physiology at theuniversity under whom Banting had carried out hiswork), he was recognized as the co-discoverer and notBest. Banting was much distressed by this and said somany times.

(2) A Canadian farm boy in origin and a graduateof the University of Toronto, Banting was an orthopedicsurgeon and part-time instructor in physiology atWestern Ontario University, London, where he was inthe practice of medicine. It was in preparation of alecture that he had read the now famous article onpancreatic lithiasis by the obstetrician, Moses Barron,that prompted his search for insulin. This was in the fallof 1920, and by May of the following year he hadconvinced Macleod to provide him with laboratoryfacilities. Within a few years after his discovery ofinsulin, Banting forsook studies of diabetes and concen-trated on other fields, such as the adrenal cortex,cancer, and silicosis. He also did important work on theFranks antigravity suit for the Canadian Air Force at theoutbreak of World War II. On February 21, 1941,Banting was killed in an airplane crash in Newfound-land while in the service of his country. He had recentlyturned 49.

Born in West Pembroke, Maine, in 1899, Best wasworking on his master’s degree in physiology at the

BibliographyUniversity of Toronto at the time he and anotherstudent were assigned by Macleod to assist Banting. Bya toss of a coin, Best won the chance to first work withhim. In 1925, he obtained his medical degree from theuniversity, and, two years later, his doctor of sciencedegree from the University of London. Upon his returnto Toronto in 1927, he was appointed professor ofphysiology at the medical school (later to head thedepartment) and became director of Connaught Labo-ratories, director of the Banting and Best Department ofMedical Research, and head of the Banting and BestInstitute. He retired from those duties in 1967. Inaddition to devoting his career to diabetes, he madesignificant contributions to studies of choline and liverdamage, heparin, and thrombosis. He also was thediscoverer of the enzyme histaminase. Best died in1978.

(3) As a part of their studies on fat digestion withdepancreatized dogs, these German investigatorsdiscovered that the urine of the animals possessed anexcessively high sugar content. Thus, they had demon-strated the relation of pancreatic function to diabetes,and Oskar Minkowski went so far as to conduct pre-liminary experiments as further confirmation, but, forreasons unknown, he abandoned these very promisingleads. The history of diabetes is replete with suchironies.

(4) Professor J.J.R. Macleod to Dr. G.H.A. Clowes,May 15, 1922. Copy in the files of the Insulin Commit-tee, University of Toronto.

(5) F.G. Banting, C.H. Best, J.B. Collip, W.R.Campbell and A.A. Fletcher, “Pancreatic Extracts inthe Treatment of Diabetes Mellitus,” The CanadianMedical Association Journal, Vol. 12, No. 3, March1922.

(6) “Today’s Problems in Light of Nine Hundredand Thirty Fatal Cases,” The Journal of the American

11

Medical Association, Vol. 78, No. 20, May 20, 1922. Inanother article of the time, Dr. Elliott P. Joslin referredto the co-discoverers as “the young Lochinvars ofToronto.”

(7) Dr. Charles H. Best, interview with author,September 24, 1968.

(8) J.J.R. Macleod, “Insulin and the Steps Taken toSecure an Effective Preparation,” The Canadian Medi-cal Association Journal, Vol. 12, No. 12, December1922. Years later, Best recalled, “It was his (Clowes)scientific training, his great interest, and his abilitywhich were in part responsible for the selection of EliLilly and Company as the first company in the UnitedStates to collaborate with the University of Toronto inthe production of Iletin. . . . Of course, he was inter-ested in large-scale production, it was so much easierfor the Lilly company than for us here because we’dhad no experience in the use of large stills or otherapparatus for concentrating large volumes of material.”Transcript of interview with author, September 24,1968. Copy in the Lilly Archives.

(9) The agreement, dated May 30, 1922, wassigned June 28 by representatives of the Board ofGovernors of the university and on July 7 by J.K. Lilly,Sr., and General Manager and Secretary Charles J.Lynn on behalf of the company. Copy in the LillyArchives.

(10) Jasper P. Scott, a graduate of Franklin College,Franklin, Indiana, joined the company in 1920 as aresearch chemist upon the recommendation of GeorgeB. Walden. He held various positions in research until1946, when he was appointed a director in IndustrialEngineering and Facilities Planning. Scott died in 1953.

(11) Information about early insulin production atLilly was obtained from a six-volume series of labora-tory data entitled “Insulin Research” for the years 1922to 1925. The entries, mostly recorded by Mrs. GeorgeB. Walden, who was an employee at the time, noted thedate of each lot, the number assigned it, the extraction

procedure, and assay results. The volumes are in theLilly Archives.

(12) Clowes, undated reports on distribution of R-4320-D to clinical investigators for the periods June 19to September 1 and July to September 25, 1922. Thefirst report states that 3,085 units had been sent toBanting by September 1, but the second report givesthe amount as 2,985. Reports in the files of the InsulinCommittee. R-4320-D was the first successful lot ofinsulin Lilly made, and four house samples of this lothave been preserved in the Lilly Archives.

(13) July 26, 1922. Letter in the Lilly Archives.

(14) Clowes to Macleod, October 20, 1922. Letterin the files of the Insulin Committee. This seven-pagesummary of events was written to explain the reasonsfor production setbacks and product deterioration.

(15) U.S. Patent No. 1,520,673, George B. Walden,“Purified Anti-diabetic Product and Process of MakingIt,” assigned to the Governors of the University ofToronto and granted December 23, 1924. See alsotranscript of Walden interview with author, November5, 1968. Copy in the Lilly Archives.

(16) General Letter No. 13, February 20, 1923.Copy in the Lilly Archives.

(17) General Letter No. 13, February 20, 1923.Copy in the Lilly Archives. The company had estab-lished a clinic to carry out its own investigation ofinsulin. The unit was organized in July 1922 at Method-ist Hospital, Indianapolis, and headed by Dr. John A.MacDonald, professor of medicine at Indiana Univer-sity and personal physician to the Lilly family. His staffconsisted of Dr. Cecil L. Rudesill; Dr. John H. Warvel,head of the hospital’s pathology laboratory; and MissRuth Michaels, a nurse whom Warvel later married.Michaels received training under Dr. Rollin T.Woodyatt for the care and training of the clinic’s pa-tients. The first therapeutic dose given at the clinic wason August 12 to Mrs. Nellie Underwood, the hospital’s

12

housekeeper. Joslin was the first American physician touse Lilly’s insulin when, on August 7, it was adminis-tered to one of his regular, long-time patients, MissElizabeth Mudge. Mudge and J.K. Lilly, Sr., struck up afriendship and corresponded many years until herdeath in 1947.

(18) Vol. 2, Nos. 5 and 6, November-December1922.

(19) Samuel E. Bassett to Joslin, November 30,1922. Letter in the Lilly Archives. Joslin sent the letterto J.K. Lilly, Sr., with the notation that it was a niceChristmas present for the company and Banting.

(20) Report of the Insulin Committee, “Insulin: ItsAction: Its Therapeutic Value in Diabetes and ItsManufacture,” The Canadian Medical AssociationJournal, Vol. 13, No. 7, July 1923. The same report wasalso published in The Journal of the American MedicalAssociation, Vol. 80, No. 25, June 23, 1923.

(21) In the summer of 1923, F. Lorne Hutchisonwas appointed Executive Secretary of the InsulinCommittee, a position that was created when Macleodasked to be relieved of his Insulin Committee duties inorder to give full attention to his responsibilities as headof the department of physiology. Some years later, hewas appointed Regius Professor of Physiology at hisalma mater, the University of Aberdeen, a post he helduntil his death in 1935. Hutchison served as ExecutiveSecretary until his death in 1952.

(22) General Letter No. 88, October 9, 1923. Copyin the Lilly Archives. With the adoption of unit potencyspecified by Toronto, unit designation was changedfrom “H” used in the clinical trials (presumably stand-ing for human) to that of “U”, which began with com-mercial introduction. The “U” was 40 percent greaterin potency than the “H” and thus resulted in fractionalcalculation in converting dosage from the latter to theformer. For example, one unit of “H” = 0.7 of “U,” 10= 7, 20 = 14.5, 30 = 21.5. For derivation of unitage andestablishment of an international standard, see Albert

H. Lacey, “The Unit of Insulin,” Diabetes, Vol. 16, No.3, March 1967, and “International Insulin Standards,”Diabetes, Vol. 17, No. 11, November 1968. Lacey wasChief Chemist of the Insulin Committee from 1931until his retirement in 1969.

(23) Records of the Insulin Committee.

(24) From the outset of commercial distribution,Lilly offered specifically calibrated syringes, a urinarytest kit, plain agar, saccharin tablets and the “EverAseptic lletin Syringe Outfit,” designed and patented byMr. Eli Lilly. In time, supply houses began to carrydiabetic accessories and the company eventuallydiscontinued them.

(25) Records of the Insulin Committee, op. cit.

(26) J.K. Lilly, Sr., to Joslin, February 6, 1931. Copyin the Lilly Archives. The company anonymouslycontributed $50,000 to the Banting Research Founda-tion in 1925, which was created to advance furtherwork by private means. Lilly also provided unrestrictedgrants to Best, which were paid periodically for severalyears. After the expiration of the patent on amorphousinsulin and royalty payments in 1941, it made annualgrants to the Banting and Best Department of MedicalResearch. All royalty payments on sales of later insulinpreparations ended in 1969, when the Food and DrugAdministration assumed full authority for safety andpotency determination.

13