A Near Miss by America’s Most Renowned Physiologist
As a matter of routine I have long trusted unconscious processes to serve me…. [One] example I may cite was the interpretation of the significance of bodily changes which occur in great emotional excitement, such as fear and rage. These changes—the more rapid pulse, the deeper breathing, the increase of sugar in the blood, the secretion from the adrenal glands—were very diverse and seemed unrelated. Then one wakeful night, after a considerable collection of these changes had been disclosed, the idea flashed through my mind that they could be nicely integrated if conceived as bodily preparations for supreme effort in flight or in fighting.
Although the American pharmacologist Reid Hunt first reported that acetylcholine was the most potent drug known for lowering blood pressure, the possibility that it was secreted by parasympathetic nerves was not pursued until fifteen years later, when Otto Loewi began to work on the problem. And even afterward it was primarily European pharmacologists who extended Loewi’s research. A search through the American Journal of Physiology from 1926 through 1929, for example, reveals only one publication on acetylcholine (and it was judged to be “insignificant”).2
There was one American, however, who came close to being the first to prove that a humoral substance might be secreted by nerves. This was the Harvard physiologist Walter Bradford Cannon (1871–1945), the foremost physiologist in the United States at the time. Cannon had stumbled on evidence that sympathetic nerves secrete an adrenaline-like substance even before Otto Loewi published his first tentative evidence on “Vagusstoff.” Walter Cannon might have shared the 1936 prize with Otto Loewi and Henry Dale had he not adopted a controversial theory about the nature of these secretions.3 The life and work of this remarkable man are an essential part of this story.4
Cannon had an enormous influence on the field of physiology in his time. He was a prolific and creative experimentalist with a manual dexterity that enabled him to keep animals in excellent health even after difficult and extensive surgery.5 Perhaps most remarkable about Cannon was his ability to see how disparate experimental results could be integrated into broad physiological principles, which had great influence not only on physiologists, but also on those in many other fields. He practiced what he called “synthetic or integrative physiology,” and many of the concepts he introduced, like “homeostasis,” have been so widely adopted by the general public that many have forgotten who first introduced them. He was also an inspiring and much-revered teacher, and during his lifetime more than four hundred students and postdoctoral fellows, from many countries around the world, worked in his laboratory.6 He had a friendly, open manner that enabled him to work with people in all walks of life. Colleagues in other countries often commented that his direct and unpretentious manner reflected what was best in the American character. In addition to all this, Cannon was a concerned citizen who did not hesitate to get involved in neighborhood problems as well as affairs of the world.
Cannon was born in 1871 in Prairie du Chien, Wisconsin, and was of Scotch-Irish ancestry. Prairie du Chien was the site of Fort Crawford, where the army physician William Beaumont conducted some of his seminal observations on digestion in the stomach. These observations were made on a subject named Alexis St. Martin, a French-Canadian fur trapper whose stomach was left externally exposed after an accidental gunshot injury. Cannon’s initial research was on gastric motility and digestion, and he took pleasure in this connection to Beaumont, often citing Beaumont’s early observations in his own publications.
Walter Cannon’s father, Colbert Cannon, worked for the Chicago, Milwaukee, and St. Paul Railroad, first as a newsboy and later as a supervisor of newsboys. He married in 1870, and Walter Cannon was born a year after the marriage. Colbert Cannon switched jobs several times, trying farming in Minnesota before settling in St. Paul, where he eventually became the chief clerk in the car service department of the St. Paul and Pacific Railroad. In 1881, when Walter was only ten years old, his mother died of pneumonia. His father then married Caroline Mowrer, who was devoted to her husband’s children and they to her.7
Colbert Cannon was a clever and inventive man and an independent thinker. He was promoted to the position of “car accountant” after he devised a novel graphic method to keep track of the company’s freight cars. The promotion enabled the family to move to a larger and more attractive house in St. Paul, where Colbert used one room as an office in order to realize his long-standing dream of practicing medicine. He had no formal training in medicine, but this was not unusual among physicians at the time. He got what he knew of physiology, anatomy, and therapeutics from books, and he treated his friends and neighbors without charge. He placed hot cups on the chest of patients to treat tuberculosis, applied electricity to various body parts, injected herbal substances into the buttocks, and rubbed the body with coconut oil to treat other ailments. He had many opinions about healthy living, such as wearing flannel and chamois cloth chest protectors in the winter, and he embarrassed his children by not allowing them to wear the pointed shoes that were stylish at the time. Colbert also had strong opinions about food. Meat was not served in the house, and the family meal often consisted of nuts, distilled water, Graham bread, and Postum.
Although many of Colbert’s ideas were unfounded, he had an active and curious mind and was constantly engaged in trying out something new. Despite bouts of depression, he was a good father, building dollhouses for the girls and skis and a huge bobsled with a steering wheel for Walter. He bought a tool chest for Walter and worked beside him building toys. He strongly believed that education should make children self-sufficient, and he placed a brooder and an incubator in the family dining room and taught his children how to raise chickens. There was little decoration in the dining room except for a bookcase of encyclopedias, which were often used to settle family disputes. He bought so many books and magazines that he put a strain on the family budget. Despite Colbert’s interest in education, however, he took Walter out of school at the age of fourteen, believing that working would provide a better education than “loafing in school.” At the time Walter didn’t resent this, but after spending two years working at a job he found dull, he convinced his father that he should return to school.
Walter Cannon started high school when he was almost eighteen, the age most children graduate. The St. Paul High School had an excellent reputation, and many of the students went on to university, most to the University of Minnesota. Cannon had become interested in the debate on evolution between Thomas Huxley and Bishop William Wilberforce, and he read voraciously about the subject. Finally, as he later wrote, “my inner turmoil drove me to the confession that I no longer held the views accepted by members of the congregational church which I had joined.”8 When the minister of the local congregational church called Walter in for a talk, he took the worst possible approach, appealing only to authority. He asked Walter what right he had to question the great church scholars. Walter responded predictably, resigning from the church and angering his father, the deacon and a Sunday school teacher at the same church. However, Walter received support from a young Unitarian minister, a graduate of the Harvard Divinity School, who had recently been appointed to the high school staff.
Although Cannon’s rebellion was the subject of much gossip around town, he was by no means ostracized. He was voted the “most popular boy” in high school, was elected class president, served as editor of the high school newspaper for a year, and was one of the top students, winning many honors. He was also editor of the class yearbook. The faculty chose him to give the commencement address, and Cannon used the opportunity to talk about the importance of tolerance for different opinions in science and religion. His commencement address, “Don’t Be a Clam,” was awarded a prize as the best student essay.
Cannon wrote well and he gained the respect and support of his English teacher, Mary Newsom, the daughter of a newspaper editor in St. Paul. She often invited Cannon to her home to talk about books and ideas, encouraging his independence and convincing him that he had to go to college. On a trip east, she wrote to Cannon bubbling over with enthusiasm about her visit to Harvard. Convinced that Harvard was the best place for Walter, Mary Newsom recruited teachers at the high school to write letters supporting his admission. During the early summer of 1892 Cannon took the Harvard entrance examinations in elementary Greek, German, French, and trigonometry. Then, with his father’s help, he obtained a summer job as assistant to the paymaster of the Great Northern Railroad. Walter traveled widely with the paymaster, which afforded him an opportunity to see the extent of the plains, the mountains out west, and Indians paddling birch bark canoes in Idaho. Walter’s lifelong respect for “working-class” people is foreshadowed in his diary description of the men building the railroad around Spokane:
There were two gangs of men—each gang seized a rail and laid it in contact with the last rail—the nails were placed a uniform distance apart, the car was jerked forward—two more rails were laid—and so the work went on…. Men worked as if their lives depended on accomplishing as much as possible each second. The great clouds of choking dust blackened their eyelids and covered their lips and made their voices hoarse. The laborers were great hard men with their hair and face-lines and clothing filled thick with dirt.9
Near the end of the summer, Walter received word that he had been admitted to Harvard. The family could ill afford the expense of sending him there, so it helped that he was offered a $250 scholarship. He still had to watch his finances closely; the Harvard catalogue mentioned that $372 would just barely cover the yearly cost of tuition, room, board, and incidental expenses. He found a room to rent for $57 for the year and ate at the Foxcroft Club, which was subsidized by Harvard. The meals were a la carte, with such charges as one cent for butter and ten cents for two eggs. Cannon figured out how to get the most nutrition for the least money. He also was able to earn some money tutoring students.
Cannon had decided on a career in medicine. His first-year classes, selected for him by his high school Latin teacher, included English rhetoric and composition, French, German, chemistry, botany, and zoology. It was a heavy load, but he maintained close to a straight A record. Mary Newsom wrote advising him to “Go to the Professors. You are not wasting their time if they are true teachers of men.” She also sensed that he might be missing female companionship and wrote that: “You must not go through your college course without a woman’s companionship and influence.”10 Cannon was well aware of the differences in social classes between most of the students at Harvard and the residents of Cambridge. He wrote a description of some of his New England “Brahmin” classmates for a class assignment:
The two came swinging down Brattle Street with their brilliant black shoes, prominently creased trousers, long overcoats flaring from the waist downward, new yellow gloves, and heavy canes, all of the newest and most improved style. Their faces smooth and clear cut looked striking under their shiny silk hats. As the two glided lithely along they met a little mucker who stopped short on seeing them. He stared at them with wide-open eyes and mouth as they approached and when they had passed he turned and watched them swagger on.11
Cannon began to attend the Boston Symphony concerts and public lectures on social issues. His essays written for English classes began to express his views on broader social issues, rather than just commenting on the different social classes around Cambridge. By the time he was a junior, Cannon was taking an increasing number of electives in addition to science courses. He took a course on the Italian government and law and was particularly impressed by a psychology course taught jointly by William James and Hugo Müstenburg.
The zoology faculty considered Cannon one of their most promising students, and he was invited to spend a summer in Alexander Agassiz’s laboratory in Newport, Rhode Island. He there began a warm friendship with Herbert Spencer Jennings, a graduate student who had done his undergraduate studies and a year of graduate work at the University of Michigan. Jennings would later become well known for his classic studies on the behavior of the paramecium and other lower organisms.12 Jennings wrote to a friend about spending time with a Mr. Cannon, from Minnesota, who was going into medicine and who had: “a decided leaning toward sociological matters … a very fine fellow, whole-souled and genuine—no veneer or falsity, like the easterners, many of them.”13 Jennings and Cannon remained lifelong friends.
Cannon later commented that Charles B. Davenport, George H. Parker, and William James were the teachers who influenced him the most. With Davenport he did his first investigation of a biological phenomenon—the orientation of protozoans to a light source. He was Parker’s assistant for two years, and the two became good friends. He described William James’ lectures as “fascinating in the freshness and constant unexpectedness of his ideas.” Once when walking home with James, Cannon expressed an inclination to study philosophy. James turned to him and said, “Don’t do it. You will be filling your belly with the east wind.”14
By his senior year, Cannon’s professors were treating him as a junior colleague, and he was included in gatherings at their homes. In one diary entry he noted that he had been at Professor Shaler’s home and spent a pleasant evening listening to Shaler and Lloyd Morgan trade animal stories. In 1896 Cannon graduated summa cum laude from Harvard University. He wrote to Johns Hopkins Medical School, considered at the time the country’s best medical school, inquiring if he could get part-time employment while attending. Cannon’s letter was never answered, and so in 1896 he enrolled in Harvard Medical School.
Medical education at Harvard, as well as at other schools, was in a state of transition at that time. In the 1850s Harvard’s medical school had offered only two years of lectures to students, many of whom had little background in the sciences. Admission to the medical school did not even require a university diploma. There were no laboratories for the faculty or the students. Although there were several calls to reform the medical school, not much happened until Charles William Eliot became university president in 1869. Eliot, who had been a chemistry professor, turned out to be a remarkably effective and innovative leader. During his forty years as president of Harvard, Eliot transformed what was a small parochial Unitarian college into a great university. He introduced many changes in the curriculum and allowed undergraduates to select courses, and after he raised salaries he was able to recruit an outstanding faculty. Eliot recruited Henry Pickering Bowditch as an assistant professor of physiology by providing him with a research laboratory and the opportunity to participate in reforming medical education at Harvard.15
Bowditch, a graduate of Harvard University and Medical School, had studied in France with Paul Broca and Jean Charcot.16 When he accepted the position at Harvard he purchased physiological equipment in Europe at his own expense before sailing to Boston. Later, when he became dean of the medical school, he started to build the physiology department and to make research a part of the mission of the medical school and the hospital. He also helped establish the American Journal of Physiology, which published its first issue two years after Cannon entered the medical school. Prior to that time, American physiologists were dependent on European journals to publish their work.
Early in his first semester Cannon decided that it wasn’t necessary for him to attend all the lectures, and he asked Bowditch if he might do some research with him. At the time Bowditch was excited about the potential of the recently discovered Röntgen-rays (X-rays), and he suggested that Cannon and Albert Moser, a second-year medical student, explore the usefulness of X-rays to study the passage of food through the digestive tract. This suggestion would completely change Cannon’s career as at the time he was interested in neurology. Through an unpredictable path, the work on gastric physiology led Cannon to study how emotions influenced the sympathetic nervous system and the secretion of adrenaline.
Cannon and Moser’s first project was to try to use X-rays to study deglutition (swallowing) by following an opaque substance as it moved through the esophagus. Cannon obtained a card of small pearl buttons normally used on clothing, and these were placed far back in the dog’s throat, forcing it to swallow. After some trial and error, they were able to observe the button as it passed through the esophagus to a point just above the diaphragm. They repeated the experiment on roosters, geese, frogs, and cats and constructed special restraining boxes for each of these animals.
Cannon and Moser then switched from buttons to bismuth subnitrate, a radio-opaque substance, which they placed in a gelatin capsule that they mixed in with food. Bismuth subnitrate was being used at the time to treat various gastrointestinal disorders. This procedure worked so well that when the American Physiological Society held their meeting in Boston, Bowditch arranged for a group to come to the medical school to see Cannon and Moser’s demonstration of food passing through the alimentary canal, including peristaltic waves and how the consistency of different foods determined the rate of descent.17
All this research activity had left no time for Cannon to work to support himself, and just when it appeared that he might have to give up research, he received the news that the faculty had awarded him a scholarship. Cannon then decided to undertake an independent study of peristalsis in the stomach, as Moser had become too busy with clinical training. This was more difficult than studying movement of food through the esophagus, but was important because so little was known about the role of stomach movement in digestion. Bowditch had once remarked that butchers knew as much as physicians about the motor activity of the stomach.
Cannon gradually built up a collection of colored tracings he made of stomach movement at different stages after the food entered. Much of this work was done on cats, but he also drew pictures obtained from humans while they were standing, sitting, or prone, and breathing normally or holding their breath. This work aroused some interest in the possibility that the technique might be used to identify malignancies and other stomach pathology that might uniquely distort gastric motility. His publication in the American Journal of Physiology describing this work evoked complimentary reactions from physiologists in Europe and the United States.18
In his third year of medical school, Cannon rotated through various clinical services. Third-year students were required to take charge of six pregnant women from before they went into labor through convalescence of the mother after delivery. This involved visiting homes of the poor, usually recent immigrants living in South and West Boston. While many of the students found this an unpleasant task, Cannon enjoyed the opportunity to have contact with people of different nationalities and customs. He wrote about his experiences to an older friend back home, who replied that: “It is very-broad minded of you to approve of the old women and to admit that they know anything about babies. I never knew a young doctor before to admit it.”19
Cannon managed to continue some research on gastric motility while doing the clinical rotations, and during this period he made his first observations on the physiological changes associated with emotional states—a topic that would in time become the focus of his research. He had noticed that when a cat changed from being relaxed to struggling, stomach movements ceased and the outline of the stomach widened in the pyloric region. He began to stress cats in different ways, and he observed that gastric motility always ceased until the stress was removed. The same changes occurred when a cat was in a state of rage.
His research was going well, and Cannon was enjoying the clinical experience with patients, but he was unhappy with the clinical classes. The lectures were often delivered in a dry manner, allowing no time for discussion. He had some friends in law school and was impressed with their lively debates about the cases analyzed in class. Cannon persuaded a young instructor in neurology to hand out a description of a patient from his private practice and to give students a week to study the case before discussing it in class. It proved so successful that the instructor adopted this way of teaching his entire class.
Encouraged by the instructor, Cannon published an article in which he proposed that the “case method” of teaching should be used to supplement lectures. He wrote that: “With a good leader … the underlying pathological condition, the disturbed physiology, the therapeutic action of the drugs employed, could constantly be brought forth to give the cases a rational explanation and to teach the students the deeper insight which vision through general principles affords.”20
Cannon received many complimentary responses to his article, including one from William Osler, who was generally regarded as the leading authority in clinical medicine. A few months later, Cannon, still a medical student, was invited, along with some distinguished professors of medicine and education and Harvard’s president, Charles Eliot, to participate in a special meeting on medical education sponsored by the Boston Society for Medical Improvement. Cannon was coming to be known as a person with many different talents. Before starting his final year in medical school, he was invited to teach the undergraduate Harvard and Radcliffe College class in vertebrate anatomy, a course he had assisted George Parker in teaching while still an undergraduate.
As he approached graduation from medical school, Cannon was offered a full-time instructorship in zoology. The chairman of physiology also wrote to Charles Eliot about the possibility of offering Cannon an appointment in physiology. The letter praised Cannon as an unusually promising person who “understands that an investigation is only well-begun at the stage most students think it is finished.” Cannon was offered a position in physiology, although he was not yet sure that he really wanted a career in research. He wrote to Cornelia James, the woman he would marry a year later, that he had accepted a position as instructor in physiology at Harvard and that this might be the beginning of a new career—“Who can tell?”
When Cannon started in the physiology department in July 1900, he was given a small laboratory, which enabled him both to continue work on how emotions affect gastric motility and to explore several new lines of research.21 A number of academics began to visit at this time to learn more about his “case method of teaching.” Cannon’s reputation was growing among the faculty administrators, and he was also popular with students, who asked him to give a special lecture on the power of suggestion in medicine. Charles Eliot remarked at a faculty meeting that “he regarded Cannon as a valuable person to hold onto,” and even though the physiology department was reducing the number of instructors, Cannon was rehired. Nothing motivates like success, and Cannon was becoming convinced that he wanted a career in research.
When Cannon was a senior in Harvard College, he had begun going out with Cornelia James, a first-year student at Radcliffe, who also hailed from Minnesota. They knew each other from St. Paul’s High School, where he had been four years ahead of her. They became quite close in Cambridge and were engaged before he graduated from medical school. When it was announced in the physiology department that he was getting married, one of the staff remarked that “a young married is a young man marred.” However, Cannon later wrote that “throughout my married life my wife has been my best, my most helpful and most devoted counselor and companion.” Cornelia was a remarkable woman with an independent spirit and strong views about the importance of getting involved in work of social importance. Later in her life she wrote several novels, one of which was quite successful.22
After their marriage, Walter and Cornelia enjoyed a strenuous honeymoon of canoeing, camping, and mountain climbing. While in Glacier National Park they were told that Goat Mountain at the head of Lake MacDonald had never been climbed, and they decided to try. It was a rugged climb and they had several close calls—once when they barely escaped from a huge rock that broke loose, just missing them, and another time when Cannon was stuck for quite a while in a crevice on a steep surface of the mountain. When they finally reached the bottom it was almost dark. They met a couple of geologists doing some surveys for the government. The geologists were so impressed with the tale of their climb that they officially named the peak Mount Cannon.23
Cornelia often influenced her husband to become involved in various social causes. It didn’t take much persuasion, as he was similarly inclined. The Cannon household became a meeting place not only for young academics, but also for groups concerned with improving the schools and the lot of the mostly Irish immigrants living in East Cambridge. Although they had some early difficulty having children of their own, they eventually had a son and four daughters. Much later Henry Dale wrote after visiting the Cannons in their old, colonial-style house in Cambridge that he had been welcomed into a household glowing with happiness and affection and that anyone who had not seen Cannon as a father had “missed a very important and attractive side of his character.”24
Cannon’s reputation as a physiologist was growing rapidly, and in 1902 he was offered a position at Western Reserve University. Harvard did not want him to be lured away, and he was thus promoted to the rank of assistant professor. Three years later Cannon was offered a professorship at Cornell University. At the time he was having a disagreement about teaching methods with William Porter, a senior professor responsible for much of the administration of the physiology department. Cannon presented his disagreements to Henry Bowditch and Charles Eliot. It was feared that Cannon might leave if the problem couldn’t be resolved. Cannon was very popular with students, and some of them collected signatures on a petition supporting him. Bowditch planned to step down as chairman of physiology and, after consulting with others, he decided that while Porter was a good man, Cannon “is the best man in our school, a man of very unusual merit and tact as well, who values even the slightest good work of others … If we lose Cannon we’ve made a real loss.” Charles Eliot appointed a committee of senior medical professors to interview both Cannon and Porter, and in the fall of 1906 it was decided that Cannon should succeed Bowditch as chairman and be given the George Higginson Professorship in Physiology.25
Expressing his appreciation for the promotion, Cannon wrote to Charles Eliot: “I shall do my best to show my loyalty and devotion to her [Harvard]”—whereupon Eliot appointed him to a number of committees. Cannon never served on any committee in a perfunctory manner. He often broadened his responsibilities, as for instance when he chaired Harvard’s committee on the use of animals in research. He collected information on existing practices around the country and used this information to write a small manual of rules on the care of laboratory animals. Cannon circulated the manual to the deans of seventy-nine medical schools, and many adopted his guidelines.
At the time, concern over the use of animals in experiments was heightened by a growing antivivisection movement. Cannon wrote that in a democracy the public should be involved in the issue, and he proposed establishing an auxiliary board of laymen to support medical research. He wrote a pamphlet describing how animal research benefited humanity and joined others in attempting to block proposed legislation that would have restricted animal experimentation. The American Medical Association asked Cannon to head a committee for the protection of medical research. Among others who served on the committee were Reid Hunt, the pharmacologist who had first reported on the potency of acetylcholine; Simon Flexner, the director of the Rockefeller Institute; and Harvey Cushing, the noted neurosurgeon, who became a close friend of Cannon. He received numerous letters expressing appreciation for the work he did in standardizing rules to govern animal experimentation and for opposing the anti-vivisectionists. Harvey Cushing wrote to him: “You fully deserve the grateful acknowledgement of the entire profession. It scares me to think that someone else other than yourself might have been made chairman of our committee.”
Cannon was pained when his friend and former teacher, William James, disagreed with his defense of animal experimentation. James had a long-standing abhorrence of animal experimentation that may have stemmed from animal experiments he had observed and from the demonstrations he had helped prepare when he was an assistant in anatomy. James had written a letter to the New York Post in 1909, which was subsequently circulated by the Vivisection Reform Society of New York. James argued that medical and scientific men acted like a “trade union or corporation that could not be trusted to be either truthful or moral in setting up rules that might condemn any of their members and who would not accept that treatment of animals might be somebody else’s business, including a God in Israel to whom he owes account.”
Cannon’s response was that there were already laws against cruelty to animals and that several callous employees in animal laboratories had been dismissed during the previous year. He also argued that special legislation to control medical research was unnecessary and likely to be restrictive. William James was not convinced, and both his letters to “my dear Cannon” and his public statements continued to express his opposition to what he believed were ineffective recommendations. Actually, James’ position on vivisection was more nuanced than many inferred from the New York Post letter.26 He did not believe that vivisection or even inflicting pain on animals was always wrong, but he felt that those doing it could not be trusted to police themselves. As James’ health was failing at the time Cannon decided not to press the disagreement with his former teacher any further. William James died in 1910, not long after this exchange with Cannon. Cannon also disagreed with William James on the physiological basis of emotions, but he softened his critique of what is now called the James-Lange theory of emotions until well after William James had died.27
Cannon’s research was going well. While Pavlov in Russia and William Bayliss and Ernest Starling in England were well known for their studies of gastric and pancreatic digestive juices, Cannon was getting to be well known for his research on the mechanical factors in digestion—the grinding and churning of food in the stomach and intestines. His interest in the clinical implications of his research motivated him to design an apparatus to amplify the sounds of movement coming from the digestive tract of cats. This brought Cannon’s research to the attention of gastroenterologists and to surgeons who were interested in exploring the potential of this technique to detect complications following stomach surgery.28 Cannon summarized his research on gastrointestinal movement in the book The Mechanical Factors of Digestion, published in 1911.
Medical students and even undergraduates were attracted to Cannon, and there was a steady flow of them knocking on his door looking for an opportunity to work in his laboratory. He accepted a number of them, and he had a constantly evolving list of projects for them to choose from. He seemed to think about physiology all the time, and colleagues joked that he shouldn’t take any vacations because he always returned with more ideas than could possibly be pursued. The file he kept of potential research projects included such notes as:
“Peptic ulcers from simultaneous stimulation of vagus and sympathetic?”
“Factors affecting the growth of hair?”
“Milk secretion as affected by fright?”
“Effect of secreted adrenalin on bronchioles”
“Catnip in relation to sex? Spayed cats?”29
Cannon’s influence on students was often lasting, as the experience of Albert Hyman illustrates. Hyman was a freshman at Harvard College, and Cannon was his faculty advisor. They would sometimes meet in the laboratory. One time when Hyman was seeking some advice, Cannon was in the process of trying to restore normal heart rhythm in a cat after he had induced ventricular fibrillation. Although Cannon never succeeded in finding a reliable method of restoring normal rhythm in a fibrillating heart, he did stimulate Hyman to think about the problem and two decades later to find a way to do just that. Hyman later described how he had naively asked Cannon why it was so difficult to get the heart beating again:
Dr. Cannon looked at me with a wistful smile, just one of the many components of an unusually expressive face which was beloved by all of his students and associates. “Young man,” he said, “that is indeed a very important question. All I can say at this moment is that I am glad that you asked it; and if the seeds of inquiry have been properly implanted by this incident, perhaps some day you may be able to produce the answers.”30
Despite all that Cannon was involved in—research, writing, lecturing, teaching, administrative responsibilities, and committee work—he also found time for community projects. Cornelia may have first had the idea to use Harvard facilities to improve the Cambridge public school system, but Cannon soon became involved. When Harvard’s division of education did not approve the plan, Cornelia wrote to her mother that the professors selfishly wanted a private school only for their own children. She then convinced her husband to use the front porch of their home for a model kindergarten and first-grade class. Cannon helped build the playground equipment in their backyard, and this was used by neighborhood children as well as those enrolled in the school. The Cannon’s living room also became a meeting place for discussing how to upgrade the Cambridge public schools.
Figure 7.1 Walter Bradford Cannon in his laboratory. Courtesy of the National Library of Medicine.
In 1908 Cannon was searching for a strategy to study how emotional states influenced gastrointestinal motility when he took on a graduate student whose interest would radically change the direction of his research. Roy Hoskins had completed an undergraduate degree and some graduate training in the biological sciences at the University of Kansas before coming to Harvard as a doctoral candidate in physiology. This was a new degree in the medical school and Hoskins was Cannon’s first Ph.D. student.
Hoskins arrived with a strong commitment to working in endocrinology.31 Cannon, however, felt that he himself knew little about the endocrine glands and tried to persuade Hoskins to work on digestion. When Hoskins resisted, Cannon suggested he study the thyroid gland because he suspected that some thyroid conditions might be caused by emotional states.32 Hoskins agreed, but he ended up working mostly on the adrenal gland. When Hoskins completed his dissertation and accepted a position at Starling Medical College in Ohio, Cannon encouraged him to investigate whether adrenaline secretions are influenced by emotional states. Within a year after leaving Harvard, Hoskins informed Cannon that he had found adrenaline in the blood of an animal after it had become enraged. Cannon proposed that he and Roy Hoskins coordinate their research on adrenaline secretion during emotional states.
Working with Daniel de la Paz, a postdoctoral fellow from the Philippines, Cannon compared the adrenaline contents in the blood of calm and stressed cats. He stressed the cats by exposing them to barking dogs. Blood was drawn from the cat when it was quiet and after it was stressed by the dogs.33 To determine whether adrenaline was present in the blood, Cannon used a method developed at the Rockefeller Institute by Samuel Meltzer and his daughter, Clara Meltzer Auer. They had found that in a cat with its superior cervical sympathetic nerves cut, even a small subcutaneous injection of adrenaline caused the pupils to dilate. Denervation was thus found to increase the sensitivity of the iris to adrenaline. Using this technique to detect the presence of adrenaline, Cannon was able to show that adrenaline is released into the blood whenever a cat is stressed, and in 1911 he published a paper entitled “The Emotional Stimulation of Adrenal Secretion.”34
Cannon confirmed these results by using a still more sensitive bioassay for detecting adrenaline. It had been reported that strips of a cat’s intestinal muscle continued to contract rhythmically if left in a solution of blood, but the contractions were inhibited by adrenaline even when it was diluted to one part in twenty million. Cannon and De La Paz were able to confirm that adrenaline was released into the blood following stress, but they were completely perplexed when they found adrenaline in the blood of a stressed cat after the adrenal glands had been removed.35 This was in January 1911—Cannon had inadvertently stumbled on the first indication that an adrenaline-like substance was secreted by some organ other than the adrenal medulla. However, the possibility that this substance might be secreted by sympathetic nerves was not something Cannon was thinking about at the time. After ruling out other explanations that occurred to him, Cannon decided that the adrenaline must have come from a small part of the adrenal gland that was inadvertently left in the animal.
Cannon began to think about all the different physiological changes that were evoked by adrenaline and that these all seemed to occur during stress. Then, as Cannon later wrote: “One wakeful night, after a considerable collection of these changes had been disclosed, the idea flashed through my mind that they could be nicely integrated if conceived as bodily preparations for supreme effort in flight or in fighting.” On January 20, 1911, Cannon wrote in his scientific diary: “Got idea that adrenals in excitement serve to affect muscular power and mobilize sugar for muscular use—thus in a wild state readiness for flight or fight!”
It also occurred to Cannon in 1911 that prolonged stress might induce different pathological states. This was before there was any serious interest in what came to be called psychosomatic medicine.36 Cannon had diabetes in mind when he reported that stress produces an increase in sugar in the blood and urine. The article concluded with the statement that, “In the light of the results here reported the temptation is strong to suggest that some phases of these pathologic states are associated with the strenuous and exciting character of modern life acting through the adrenal glands.”37
Cannon then explored several novel ways of demonstrating how excessive adrenal activity might cause, or at least exacerbate, pathological states.38 Working together, Cannon and Hoskins stimulated the sympathetic nerve that innervated the adrenal medulla and found that with prolonged stimulation the gland seemed to become exhausted, losing its capacity to secrete adrenaline.
Cannon explored other ways that adrenal secretions might prepare animals for emergencies, and he investigated whether adrenaline might help stop bleeding. Working with Walter Mendenhall, a teaching fellow in the physiology department, Cannon found that both stimulating the splanchnic nerve, which causes the adrenal medulla to secrete adrenaline, and injecting adrenaline shortens blood clotting time. They also found that blood clotting time decreased when a cat became angry or enraged. Cannon also found that adrenaline delayed muscle fatigue, and he wrote in his diary: “Much excited by the possibility of adrenaline activation explaining ‘second wind.’”39
During the winter of 1913–1914 Cannon gave several popular lectures on his new ideas in the Boston area. Over two hundred people came out in a heavy rainstorm to hear his public lecture at the Harvard Medical School. An imaginative writer of detective stories later described how a doctor Walter Cannon of Harvard could determine if any adrenaline was present in a speck of blood left at a murder scene by observing the reaction of a strip of animal intestines to the blood. A New York newspaper described the increase in blood sugar levels during stress under the byline: “Man is Sweetest When he is Angry.” Cannon was making adrenaline a household word.
In 1913 Cannon used the Carpenter Lecture at the New York Academy of Science and the Lowell Lecture in Boston to describe his ideas on the emergency theory of adrenal function. Earlier that year, he had tested the urine of the Harvard football players following the Harvard-Yale game. He found high levels of sugar in the urine (glycosuria) not only in those on the field, but also among members of the team sitting on the bench. War clouds were gathering over Europe, and Cannon talked about sports being the “physiological equivalent of war” and a possible substitute outlet for the fighting instinct. The lectures were well received and much discussed, and the Journal of the American Medical Association reprinted excerpts from his Carpenter Lecture. Cannon summarized this work in the book Bodily Changes in Pain, Hunger, Fear, and Rage, published early in 1915.40
The war in Europe began on July 28, 1914. Cannon, like many people in the United States, including President Woodrow Wilson, had taken a neutral stance. When Harvey Cushing organized a volunteer surgical unit to work in a British hospital, Cannon refused to sign up. However, when the luxury liner Lusitania was sunk by a German submarine in May 1915, Cannon, along with much of the country, began to actively side with the British and French.
The National Academy of Sciences established the National Research Council in 1916 to look into the threats caused by the war, and Cannon agreed to chair a committee for developing research programs to investigate medical problem of soldiers. He wrote to Charles Sherrington to determine what problems British physiologists were working on, and he drew up a list of several topics, such as surgical shock, irritable heart syndrome, and fatigue, for the committee to study. He agreed to head the subcommittee on shock.
After President Wilson asked Congress for a declaration of war against Germany on April 2, 1917, Cannon enlisted, and in May he left for Europe as a first lieutenant with the Harvard Base Hospital Unit no. 5, which was attached to the British Army until American troops arrived in Europe.
When Cannon got to France, he found that the hospital in Boulogne was a depressing, dirty tent city. Although it was well behind the front lines, there were no facilities, animals, or equipment to do any research. One day Thomas Elliott visited the field hospital, and a lasting friendship between them began. They were aware of each other’s work on adrenaline and the sympathetic nervous system.41 Although Cannon had never treated a patient in shock, Elliott, who had the rank of colonel, was able to arrange for Cannon to work at a British clearing station that received many cases of serious shock.
At the clearing station Cannon became familiar with the course of blood pressure changes that occur with shock. He learned that there is “primary shock”, in which wounded soldiers immediately experience a precipitous drop in blood pressure, and also a condition called “secondary shock,” in which soldiers with wounds that are not necessarily severe sometimes have a delayed, but precipitous fall in blood pressure that could be fatal. Secondary shock sometime occurred even after soldiers had been sent to the rear.
When Cannon examined the blood of soldiers in shock, he found it to be acidic. He speculated that the acidity might cause blood to be stored in the capillary bed, a condition that could explain the decrease in blood pressure. Cannon tried to correct the acidosis by administering common cooking soda (sodium bicarbonate) to soldiers who appeared to be going into shock. After being given teaspoons of baking soda in solution some soldiers seemed to recover and one soldier with severe wounds, low blood pressure, and a rapid heart rate improved so dramatically after being given the baking soda solution intravenously that Cannon called it a “red letter day.” He contacted William Bayliss, and they started to collaborate in exploring this possibility, although in the long run it proved to be a false lead. Dale, who was exploring the use of histamine to create an animal model of wound shock, was actually on the right trail, but at the time no one knew that histamine is a natural substance in the body.
The war ended on November 11, 1918, but Cannon, who then had the rank of major, had to wait until January 1919 before he could go home. A troopship was finally available to transport him back home, and when the Manchuria completed its eleven-day crossing of the Atlantic, Cannon spied Cornelia waiting to greet him at the dock in Hoboken. He had a warm reunion with his family, and within a few weeks he was back lecturing at the medical school and doing research in his laboratory.
As already mentioned, Cannon had become interested in exploring whether prolonged stress could induce various pathological states by excessive stimulation of the adrenal medulla. Cannon had earlier reported that the adrenal medulla could become exhausted and no longer capable of secreting adrenaline if it received excessive stimulation. The theory and the experimental results had not gone unchallenged, and a dispute began when George Stewart and Julius Rogoff of Western Reserve University reported that adrenaline levels remained relatively constant after prolonged stress and they argued that the adrenal gland did not become exhausted. Moreover, Stewart and Rogoff found that animals can cope with stress after the splanchnic nerve that innervates the adrenal medulla is severed.
Cannon answered that he had never claimed that adrenaline secretion was the only physiological response to stress, but rather that its effects supplemented and enhanced the activity of the sympathetic nervous system.42 Arguments went back and forth. Stewart and Rogoff criticized Cannon’s bioassay method for measuring adrenaline because it was not quantitative. Cannon responded that this criticism was equivalent to rejecting a carpenter’s level because it did not measure the precise angle of inclination. He did, however, begin to use other ways of detecting the presence of adrenaline. As it was known that a heart deprived of its nerve connections became supersensitive to adrenaline, Cannon began to use the denervated heart to detect adrenalin. It was this methodology that eventually led Cannon to the conclusion that sympathetic nerves secrete adrenaline-like substances.43
In 1920 Cannon and Joseph Uridil, a postdoctoral fellow, undertook some experiments designed to answer a criticism by Stewart and Rogoff, but they ended up producing a totally unanticipated finding. Cannon had reported that stimulating the splanchnic nerve that innervates the adrenal medulla caused the denervated heart to beat faster, but only if the adrenal gland were intact. Stewart and Rogoff, however, reported that the denervated heart accelerated following splanchnic nerve stimulation even after the adrenal gland had been removed, and they attributed the acceleration to an increase in blood pressure.
Cannon and Uridil found that Stewart and Rogoff were correct. The denervated heart did beat faster during stimulation of the splanchnic nerve, even when the adrenal gland was removed. However, they were able to rule out increased blood pressure as the explanation. It was a puzzle for Cannon, and, after exploring other possible explanations, he concluded that splanchnic stimulation caused some sympathomimetic substance to be secreted by the liver. This conclusion was based on the observation that heart rate was not accelerated during splanchnic nerve stimulation if the hepatic nerve was cut.44 Cannon had not at this point identified the sympathomimetic substance, but he considered it to be a hormone that was secreted from the liver into the vascular system and carried in the blood to the heart and to other organs, where it produced various sympathetic effects.
These results were published in 1921, the same year that Otto Loewi published his first evidence that neurohumoral secretions were involved in the regulation of heart rate. Had Cannon persisted in exploring why sympathetic nerve stimulation caused the denervated heart rate to increase, he might well have discovered then, as he did later, that sympathetic nerves secrete an adrenaline-like substance. At the time, however, Cannon was preoccupied with defending his “emergency theory” and not thinking about whether sympathetic nerves might secrete an adrenaline-like substance.45
The subject was not pursued any further until 1930, when Zénon Bacq came to work with Cannon. Bacq, a twenty-six-year-old physician from Belgium, began to prepare cats with denervated hearts and with the adrenals and livers isolated from their nerves. When they stimulated different sympathetic nerves in these animals they saw an increase in heart rate, but only after a delay of several minutes. The increase in heart rate reached an elevation of 25 to 30 beats per minute. While this was not as large as the 100 beats per minute increase that emotional excitement could produce in an intact cat, it was still a significant increase and needed to be explained. They also saw an increase in blood sugar even in animals that had the adrenal glands and liver isolated from their connecting nerves and from the vascular system.46 It became necessary to consider other sources of what appeared to be a sympathomimetic substance that was secreted into the blood when sympathetic nerves were stimulated. By 1931 Cannon and Bacq had summarized their findings in several publications:
In a cat with heart denervated, adrenals and liver excluded from action … stimulation of the sympathetic supply to smooth muscles … causes, for about two minutes, a gradual increase in blood pressure, a gradual increase of heart rate, and a gradual increase of salivary secretion. The time relations of the increase and the slow return to the former state are, in each instance, similar to those reported for the denervated heart alone [reported in a previous paper] as a consequence of emotional excitement.47
They attributed these results to the secretion of some sympathomimetic substance that came from a source other than the adrenal medulla or the liver. Although on the basis of its effects Cannon suspected the substance was adrenaline, he suggested another name for it because of its different origin: “We would suggest that it be called provisionally sympathin, with the understanding that as knowledge of its character increases it may prove really to be adrenin [adrenaline] developed for local action in smooth muscle cells.”48
Cannon viewed sympathin as a “hormone” that cooperated with and augmented the action of the sympathetic nervous system and the adrenal gland.49 This was possible, Cannon argued, because sympathin, like adrenaline, is relatively stable and can remain active while traveling around the body in the blood. Because of the sympathetic nervous system’s role in preparing animals to meet emergencies, Cannon considered it adaptive for this system to discharge as a whole, and he contrasted this with the parasympathetic nervous system, in which the anatomical arrangement made it possible to evoke separate responses.50 Moreover, acetylcholine, the substance most capable of mimicking parasympathetic responses, is rapidly degraded and therefore cannot remain active long enough to travel to different organs.
In 1931 Cannon indicated that he believed it was the smooth muscles that were secreting sympathin when they were stimulated by sympathetic nerves.51 Cannon felt that this conclusion was in agreement with Thomas Elliott’s, although Elliott had never decided whether adrenaline is secreted by the muscles or by nerves. By this time, George Parker, Cannon’s former teacher and close friend and colleague at Harvard, had concluded that nerves are capable of secreting humoral substances.52 Parker drew this conclusion from his study of chromatophores, the pigmented cells that enable some crustaceans, fish, amphibians, and reptiles to camouflage themselves by changing their skin color.53 Parker presented evidence that adrenaline produces the same effect on the chromatophores as stimulation of their innervating nerves, and he contrasted his views with those held by Cannon:
The exact source of the several humoral substances thus associated with muscle has not excited much attention. In the most recent contribution to this subject, that by Cannon and Bacq, the smooth muscle is continually referred to as the element from which the humoral substance emanates…. It seems quite evident from the conditions under which the humoral substances of muscle are produced that these substances must originate either from the muscle cells or from the nerve terminals…. Because of the small size of the terminals, it would be natural to assume that the given substance was produced by the muscle, but if the most recent work on chromatophores has any meaning at all, it points most conclusively to the nerve terminals as the source of these materials…. In many respects this principle is a special application of Bayliss and Starling’s concept of hormones, though the distances traveled by the substances from their region of origin to those of application are often extremely short.54
In a 1933 article entitled “Chemical Mediators of Autonomic Nerve Impulses,” Cannon wrote that Parker’s conclusions might be right for chromatophores, but that he believed they did not apply to the sympathetic nervous system’s innervation of smooth muscles.55 Cannon argued that the very fine single nerve endings (“nerve twigs”) that innervate each smooth muscle cell are unlikely to be able to release sufficient amounts of any chemical substance that would enter the blood and act as a hormone at distant sites.
Shortly before Zénon Bacq’s fellowship ended, Arturo Rosenblueth arrived in the Physiology Department.56 Rosenblueth, who would replace Bacq as Cannon’s “first assistant,” was a young Mexican physician, the eighth child of a Jewish immigrant from Hungary and a Mexican-American mother. Everyone who knew Rosenblueth considered him brilliant and multitalented. He was one of Mexico’s top chess players and a pianist skilled enough to have considered a career on the concert stage. He was also fluent in English, French, and Spanish, and he had a good command of German. He had studied in France and had a German medical degree. When he returned to Mexico in 1927, he practiced psychiatry for a short period, but did not find it satisfying and applied for and received a Guggenheim fellowship to work in physiology at Harvard.
Cannon recognized that Rosenblueth was enormously gifted and invited him to join his laboratory. Cannon was fifty-nine at the time, not in the best of health, and, like any senior researcher with many different projects underway, he needed the assistance of younger scientists. Rosenblueth was delighted to be invited to work with the “great man.” And he soon came to revere Cannon. Horace Davenport, who was in Harvard’s department of physiology at the time, wrote that Rosenblueth “worshipped Cannon, and when he spoke of Cannon there was a tone of reverence in his voice. I once heard him refer to Bradford Cannon as The Son, as if Cannon was God. I suppose that Rosenblueth looked upon himself as the Holy Ghost, for he was not afflicted with disabling modesty.”57
As Davenport implied, Rosenblueth could be arrogant, and it is not surprising that many resented him. Chandler McC. Brooks, who also worked with Cannon and knew Rosenblueth, described him as “too facile, brilliant, well-informed, and certain of his opinion for most of us to handle.” Cannon always thought highly of Rosenblueth, but a number of people who admired and loved Cannon resented the influence Rosenblueth seemed to have over him.
Rosenblueth was enormously energetic and, after observing some of Cannon and Bacq’s experiments, he started to turn out papers for publication so fast that Cannon had difficulty examining them as closely as was his normal practice. Between 1931 and 1933 Rosenblueth published, either alone or with others, twelve papers, and by 1944 he had published seventy-four papers in the prestigious American Journal of Physiology, in addition to a book.
Rosenblueth also had considerable mathematical ability, and he peppered his publications with equations, logarithmic functions, and mathematical reasoning. In recalling his interactions with Rosenblueth, Horace Davenport, who once described Rosenblueth as Cannon’s “dark angel,” remarked that, “I can’t say how profound was his mathematical skill, but it was certainly far beyond that of most of the mathematically illiterate physiologists of the day.”58 Rosenblueth’s mathematical ability was certainly not all “smoke and mirrors,” as he had a close collegial relationship and sometimes a collaboration with Norbert Wiener, who was certainly competent in mathematics. Wiener, the founder of cybernetic theory, regularly attended a roundtable discussion group organized by Rosenblueth, and later the dedication in his book on cybernetics read “to Arturo Rosenblueth, for many years my companion in science.”59
In his first experiments in Cannon’s laboratory, Rosenblueth demonstrated that prior administration of cocaine potentiated the response of the nictitating membrane to both adrenaline and sympathin.60 The nictitating membrane is a thin membrane that is pulled over the eye for protection in some animals. The increased sensitivity produced by cocaine presented an opportunity for Cannon and Rosenblueth to compare the properties of adrenaline and sympathin.61 Before 1933 Cannon had believed that the two substances were probably the same. He wrote that year: “Until a few months ago the evidence was fairly consistent that the substance set free from the heart on sympathetic stimulation is adrenin [adrenaline].”62 However, Cannon went on to note that he and Rosenblueth had recently found that sympathin and adrenaline are not identical and that, moreover, there were two different types of sympathin.
The evidence for two different sympathins was based on experiments that compared the responses of different organs to adrenaline and sympathin after administration of ergotoxine, a substance Dale had discovered that would block the action of adrenaline.63 They found that while ergotoxine blocked adrenaline’s capacity to increase blood pressure and heart rate, it did not block these same responses when produced by sympathin.64 This convinced them that the two must be different. Moreover, they also observed differences in the responses to sympathin depending on from where in the body it was obtained.65 From such observations, Cannon and Rosenblueth inferred that there were two different kinds of sympathin, an excitatory sympathin E and an inhibitory sympathin I.
At this point, Rosenblueth, employing his mathematical bent, introduced a degree of additional complexity to the two-sympathin theory. He had studied the responses of smooth muscles to increasing doses of adrenaline and concluded from the shape of the curves that at a certain point adrenaline must combine with some hypothetical substance to form a second substance, which he labeled AH (adrenaline plus a hypothetical substance). As Cannon and Rosenblueth had concluded that there must be two different sympathins, there must be two hypothetical substances, one of which formed sympathin E when combined with adrenaline, and the other which formed sympathin I. Cannon had by this time changed his mind and now believed that adrenaline was secreted by sympathetic nerves. According to this new theory, a sympathetic nerve impulse first releases adrenaline onto the smooth muscle, where it combined with one of two endogenous substances in the muscles to form either sympathin I or E.66
The two-sympathin theory was not well received even by Zénon Bacq, who had coauthored with Cannon the publication that introduced the term “sympathin.” After he returned to Belgium, Bacq’s research convinced him that the two-sympathin theory being promoted by Cannon and Rosenblueth was wrong, and he wrote a number of letters to Cannon explaining why he thought so. Bacq greatly admired Cannon, and, as he later commented, he was reluctant to start a long controversy with his “beloved teacher.” He tried to persuade him not to persist in holding on to the two-sympathin theory, for which he held Rosenblueth responsible:
How and why was this unlucky hypothesis of the two sympathins E and I put forward? It can be said that Rosenblueth was very fond of theoretical speculations and mathematical analysis; all his work is impregnated with this spirit. When he wrote a paper or a monograph, he emphasized discussion, not the facts. Cannon lived closer to the facts of classical physiology; he constantly invented and perfected new techniques. But neither of these two scientists had enough interest in biochemistry or pharmacology to avoid following the wrong path leading to the theory of the two sympathins…. I was struck by the fact that this interpretation was not only improbable but also useless, since nobody had any notion or even guess as to what was the chemical nature of these hypothetical substances E and I…. I showed that all the facts observed by Cannon and Rosenblueth could be easily explained if one accepted the idea that I is epinephrine [adrenaline] and E, norepinephrine [noradrenaline].67
Many others were also critical of the two-sympathin theory. Although Henry Dale was too diplomatic to express his reservations in a public forum, comments in several of his letters made it clear that he regarded the two-sympathin theory as unnecessarily complex and most likely wrong.68 Nevertheless, this did not prevent Dale from giving Cannon credit for his evidence that sympathetic nerves secrete a chemical substance.
Cannon and Rosenblueth’s 1937 monograph Autonomic Neuro-Effector Systems reviewed the arguments for the two-sympathin theory.69 When the book was reviewed in the journal Nature, the reviewer “JHG” (probably John H. Gaddum, a colleague of Dale) described the book as “authoritative, “most valuable,” and “impressive,” but criticized as “unnecessarily complicated” the theory that different receptor substances in the smooth muscles combine with a substance secreted by the sympathetic nerves and then are released into the bloodstream
More than a few of Cannon’s contemporaries believed that this opposition to the two-sympathin theory prevented Cannon from sharing the 1936 Nobel Prize with Loewi and Dale.70 The theory was criticized when Cannon presented it at the International Congress of Physiology held in Leningrad in 1935. Zénon Bacq and Göran Liljestrand of Stockholm presided at the session in which Cannon delivered a lecture on this subject. The following year, the Nobel Committee considered Cannon’s nomination for the Nobel Prize in Physiology or Medicine. This was the year the award was given to Loewi and Dale. Cannon’s nomination was evaluated by Göran Liljestrand. He described Cannon’s work on the emergency function of the “sympatico-adrenal system” as “prize-worthy,” but he criticized the two-sympathin theory, mentioning the contradictory evidence presented by Zénon Bacq, who “was previously Cannon’s colleague.” Liljestrand concluded that because of the weakness of the two-sympathin theory, Cannon’s “prize-worthiness” had decreased.”71
Cannon’s son Bradford later wrote that he never heard his father express any regret or disappointment over not sharing the Nobel Prize.72 After the prize was awarded, Cannon reviewed the history of the field and the important contributions of Loewi and Dale without mentioning any of his own work. He concluded with the following observation: “Obviously a new realm in physiology is disclosed by these important researches, and it will be well to watch for further advances in the attractive realm in which Dale and Loewi have pioneered.”73
Cannon sent congratulatory letters to Loewi and to Dale. Dale’s reply was modest and diplomatic: “So often it must be the right impression, made at the right moment, which brings [the award] to a man no more deserving than many others. If all my friends who deserve it could receive the award, we should need more fortunes of many explosive millionaires to meet the demand.”74
In 1939 Kálmán Lissák, a Rockefeller Foundation fellow from Hungary, joined Cannon. Lissák had developed a technique for extracting dialysate from postganglionic sympathetic nerves, which could then be analyzed to determine its physiological and chemical properties.75 Cannon had earlier suspected that adrenaline was the mediating substance secreted by sympathetic nerves, and Lissák found that the extract from sympathetic nerves had the same capacity as adrenaline to increase heart rate and blood pressure and to cause the pupils to dilate, the nictitating membrane to contract, and the nonpregnant cat’s uterus to relax. Moreover, all of these effects were potentiated by prior treatment with cocaine, just as with adrenaline.76 Cannon was satisfied that the mediating substance in his theory was adrenaline, and that it was secreted by postganglionic sympathetic nerves. Cannon and Lissák concluded, “The results reported in the present communication are consistent with the view that sympathetic fibers liberate adrenaline at their terminals and that this agent, when it escapes into the blood stream, has been modified in such a manner that it has the peculiar actions of sympathin on remote organs in the body.”77
The Cannon-Rosenblueth sympathin theory received two fatal blows at the end of the 1940s. The first came when Raymond Ahlquist (1914–1983), a pharmacologist at the Georgia School of Medicine, provided evidence that there are two different adrenergic receptors rather than two types of sympathin. Ahlquist had been interested in sympathomimetic drugs since the early 1940s, when he was involved with attempts to grow the plant Ephedra sinica in the western part of South Dakota. At the time this plant was the major source of the drug ephedrine and had to be imported from China. Although the project was abandoned when it became possible to synthesize ephedrine, Ahlquist continued to investigate the properties of different sympathomimetic drugs. Using rabbits, cats, and dogs, he investigated how sympathomimetic drugs affected blood vessels, intestines, pupils, uterus, heart, and the nictitating membrane. In a report published in 1948,78 he compared the responses evoked by adrenaline, noradrenaline, and isoproterenol, the latter being a substance that is chemically close to, but not identical to, noradrenaline. From the ranking of the potency of these three substances, Ahlquist was able to conclude that there had to be two different receptors, which he called alpha and beta adrenergic receptors.79
In a tone that an editor who rejected the manuscript described as “audacious” and “irreverent,” Ahlquist concluded that Cannon and Rosenblueth’s two-sympathin theory, “widely quoted as a law of physiology, is no longer necessary.” He went on to recommend that the use of the term “sympathin” “be discouraged.”80 Although references to sympathin continued to appear in the literature for several years, mention of the two-sympathin theory soon disappeared. It was ten years, however, before Ahlquist’s proposal of two separate adrenergic receptors was widely recognized as being correct.
The other crushing blow to the two-sympathin theory came from the work of Ulf von Euler. Von Euler had spent six months in Henry Dale’s laboratory during the early 1930s. When he returned to Stockholm he began to apply the bioassay techniques he had learned from Dale to study adrenaline and other natural substances that might be biologically active. He was able to obtain a much more purified extract from sympathetic neurons than Lissák had working in Cannon’s lab. Using a new fluorescence technique developed by his Swedish colleague Nils-Åke Hillarp, von Euler was able to prove that sympathetic nerves secrete noradrenaline. Von Euler reviewed the state of the field before presenting the evidence that sympathetic nerves secrete noradrenaline.
The pioneer work of Loewi and Cannon and their associates on the mechanism of sympathetic—or rather adrenergic—nerve action has revealed that stimulation of such nerves is accompanied by the liberation of some active principle with sympathomimetic properties. It was primarily assumed that the neurohumoral agent was identical with adrenaline, but this view seemed to need further consideration, however, in the light of the findings of Cannon and Rosenblueth that stimulation of sympathetic nerves elicited remote actions which did not wholly conform with those of adrenaline. In order to reconcile the seemingly conflicting evidence, Cannon and Rosenblueth (1933, 1937) elaborated a hypothesis involving the primary liberation from the adrenergic nerve-endings of a mediator substance which should then combine with some constituent within the reacting cells under formation of the final active substances, which, on account of their supposed actions, were termed sympathin E (excitatory) and I (inhibitory). The assumption of these authors that the primary mediator should be identical with adrenaline seemed to be supported by the later work of Cannon and Lissák (1939).81
Von Euler provided convincing evidence that Cannon and Lissák were in error—the adrenaline that they had detected in sympathetic nerves is actually changed to noradrenaline by a process called demethylation before it is secreted. He concluded that “it seems most likely that the adrenergic transmitter is noradrenaline.” At first von Euler concluded that some sympathetic nerves secrete adrenaline and that Cannon and Rosenblueth’s sympathin E and I might correspond to noradrenaline and adrenaline. Later he seemed to agree with Ahlquist’s conclusion that rather than two different sympathins, there are two different adrenergic receptors. The truth may be that the two issues are somewhat confounded and that while Cannon and Rosenblueth were correct in believing that there were two different substances involved, they were wrong in concluding that the nerves secreted a single substance that became two when combined with different substances in the smooth muscles. Ironically, von Euler found that Loewi had actually been correct when he tentatively concluded that it is probably adrenaline that is secreted by the sympathetic nerve that innervates the frog’s heart. He found that this sympathetic nerve in frogs does secrete adrenaline, not noradrenaline, but that this is an exception and not the case in mammals.
For a while, von Euler used Cannon’s term “sympathin” for noradrenaline, as a way of distinguishing it more clearly from adrenaline and because he was not in favor of introducing new names. However, “noradrenaline” (and “norepinephrine”) were gradually adopted, and references to sympathin began to drop out of the literature. Cannon died in 1945, so it is not known whether he would have continued to defend the two-sympathin theory, but in 1950 Rosenblueth was still defending it.82 He was, however, virtually alone in doing this. In 1953 Dale displayed his great skill of softening criticism by embedding it in a rather transparent compliment when he remarked that: “I can’t suppress an impulse of admiration for the tenacity with which the last of Professor Cannon’s distinguished collaborators, Professor Rosenblueth, in spite of all this recent evidence, still keeps the flag of the Sympathins, E and I, firmly nailed to the mast of his belief and his advocacy.”83
Cannon continued to receive nominations for the Nobel Prize for his many different contributions until 1945, the year of his death. However, 1936 was the last time he was a serious candidate. Whether Cannon deserved to share the prize with Loewi and Dale certainly can be disputed. It is true that Loewi and Dale’s evidence for chemical transmitters of neural impulses was more direct and compelling than the evidence Cannon presented and was not tainted by a controversial and what was ultimately shown to be an incorrect theory. Moreover, Cannon initially believed that sympathin is a blood-borne hormone triggered by sympathetic nerves rather than a neurotransmitter as we recognize it today.
It deserves to be noted that Cannon’s experiments were closer to reproducing natural conditions in that they were often performed on unanesthetized mammals responding to meaningful stimuli. In contrast, the experiments of Dale and his collaborators were performed on anesthetized animals responding to nerve stimulation, while Loewi’s experiments were performed on the isolated hearts of frogs and toads. Moreover, while Loewi and Dale concentrated their efforts mainly on acetylcholine, Cannon was studying the sympathetic nervous system and adrenergic substances.
In his 1932 review Dale made the point that Cannon and his coworkers were doing most of the work on chemical mediation of sympathetic nerves, although at the same time he conveyed his skepticism about the two-sympathin theory:
Further progress in our knowledge of this chemical transmitter of the peripheral effects of true sympathetic nerves has come largely from Cannon’s laboratory at Harvard…. Cannon supposes that the actual transmitter is a substance capable of producing either type of effect, as adrenaline does, according to the type of receptive substance which it finds, and combines with in the effector cell. He imagines that two types of such combination may occur, producing what he calls “Sympathin E” and “Sympathin I,” which have augmentor [excitatory] and inhibitor effects, respectively; and it is these combinations, he believes, which escape to some extent into the blood stream. It should be said, I think, that the behavior of the substance transmitting parasympathetic effects, concerning which more is known, provides no analogy for this conception….
We may safely leave the details of chemical transmission of peripheral sympathetic effects to the further investigations of Cannon and his school, and return to the transmission of the peripheral effects of parasympathetic nerves.84
In 1953, when Dale commented on his own role in these events, he noted that although he and George Barger had found by 1914 that noradrenaline is considerably better than adrenaline at mimicking the effects of the sympathetic nerves, they did not pursue this observation because they thought that noradrenaline was just an interesting drug. He later wrote about these events after Cannon’s death:
Doubtless I ought to have seen that nor-adrenaline might be the main transmitter…. If I had so much insight, I might even then have stimulated my chemical colleagues to look for nor-adrenaline in the body; but they would have certainly failed to find it, with the methods which were then available…. If I had taken the additional step, even in hypothesis, much trouble might, perhaps, have been saved in after years for my late friend, Walter B. Cannon: For the observed differences between the effects of adrenaline and those of “sympathin,” as liberated by stimulation of sympathetic nerves, and especially of the hepatic nerves, which led Cannon and his associates to put forward the elaborate theory of the complex sympathins, E and I, were practically the same as those between adrenaline and noradrenaline.85
Neither Dale nor Bacq was correct when he assumed that noradrenaline was equivalent to sympathin E and adrenaline was equivalent to sympathin I, as later discoveries made it clear that the different responses were due to different receptors, not different neurotransmitters.86 Cannon was, however, actually correct in concluding that the substance he had detected was not identical to adrenaline. Ulf von Euler—who would share the 1970 Nobel Prize for demonstrating that it is norepinephrine that is secreted by sympathetic nerves—put the question of priority in perspective when he wrote, “not before the classical series of experiments of Cannon and Rosenblueth did it become evident that sympathetic nerve stimulation led to the appearance of some factor other than adrenaline.”87
It is perhaps ironic that the two-sympathin theory actually was correct with respect to the secretions of the adrenal medulla; it was later found that this gland secretes both norepinephrine (noradrenaline) and epinephrine (adrenaline). Cannon and Rosenblueth’s theory, however, referred to secretions of sympathetic nerves, not the adrenal medulla.
Despite the evidence supporting chemical mediation of neural impulses, most neurophysiologists vigorously opposed the idea, except in the case of the innervation of some visceral organs. The vigor with which the idea of chemical transmission in the central nervous system was opposed is described in the following chapter.