The War of the Soups and the Sparks
Even after the Nobel Prize was awarded to Otto Loewi and Henry Dale in 1936, most neurophysiologists did not accept neurohumoral transmission of the nerve impulse as a general principle, although many were willing to concede that chemical transmission might be adequate for the sluggish response of visceral organs. Neurophysiologists in general were convinced that only electrical transmission is fast enough to activate skeletal muscles, and for them the possibility that nerve impulses at brain synapses might be transmitted chemically was not worth thinking about. The neurophysiologist John Eccles wrote in 1936 that the “presumed chemical nature of the synaptic transmitter in the central nervous system … is almost entirely based on an extrapolation from the ACh [acetylcholine] hypothesis for sympathetic ganglia” and that the evidence that this could be applicable to the central nervous system was “almost negligible.”1
Although the opposition to the principle of chemical transmission was based on reasonable arguments and conflicting lines of evidence, it is difficult to avoid the impression that many neurophysiologists resented the intrusion of pharmacologists into what had been their exclusive area of expertise. In describing the history of this period, Zénon Bacq observed that the eventual acceptance of chemical transmission deprived the neurophysiologists of “a vast field which passed into the hands of biochemists and pharmacologists.”2
Neurophysiologists regarded their own data, which by the 1930s was being collected via multistage vacuum tube amplification and displayed on fast responding cathode ray oscilloscopes, as more meaningful, reliable, and scientific than the data pharmacologists collected with their bioassays and smoked-drum kymographs. There was also a tendency among some neurophysiologists to look down on pharmacologists, who spent their time investigating “spit, sweat, snot, and urine,” and whose sometime association with the pharmaceutical industry was seen as “consorting with trade.” It should be recalled that Henry Dale’s friends advised him not to sell “his scientific birthright for commercial pottage” when he was considering the position offered by Henry Wellcome. In looking back to this earlier period, Zénon Bacq recalled that,
Often in my younger years, during discussions with “true” physiologists, when a contradiction occurred between facts observed by physiologists on the one side and pharmacologists on the other, the choice was made in favour of the physiological observations because a pharmacological experiment, on principle, did not carry the same weight. This elite attitude of the physiologists implied that they considered the general intelligence of the “pharmacologists” of a lower level.3
Such nonscientific factors may have colored the dispute, but the arguments raised against chemical transmission were always framed around conflicting data and observations. Many neurophysiologists were convinced that chemical transmission is too slow to be the basis for the fast responses of skeletal muscles that only they could record with their technically more sophisticated electronic equipment. Although it was no longer possible for neurophysiologists to deny the evidence that acetylcholine is secreted by spinal motor nerves, they insisted that any role it played in neurotransmission was a secondary one, restricted to modulating the primary response to electrical transmission.
The list of neurophysiologists who refused to accept that acetylcholine plays a major role in innervating skeletal muscles included eminent scientists of the stature of Albert Fessard, Ralph Gerard, Rafael Lorente de Nó, John Fulton, Herbert Gasser, Joseph Erlanger, and John Eccles, the last three of whom were later awarded the Nobel Prize.4 Erlanger, for example, citing his recording of the progress of a neural impulse along a damaged axon, stated that: “If an inactive stretch of fiber over 1 mm in length does not stand in the way of electrical transmission of the impulse, is it reasonable to maintain that the discontinuity at a synapse will stop such transmission.”5 A few neurophysiologists were more open to the possibility of chemical transmission. Detlev Bronk, for example, was clearly on the fence in 1939 when he wrote that: “I have no desire to defend either the acetylcholine hypothesis or the theory of excitation by circulating currents…. I would argue for a pluralistic theory.”6 In general, however, the two sides were far apart, and the dispute called the “War of the Soups and the Sparks” extended over two decades.7
John Eccles was not as willing as some others to concede a role for chemical transmission. In a letter dated February 11, 1939, Henry Dale wrote to Eccles that John Fulton had informed him that he (Eccles) was about to report “some experiments that strongly support the acetylcholine hypothesis.” Eccles replied on March 1 that he could not imagine where Fulton had gotten that idea and added: “I fear your hopes of an early rapprochement will have suffered a disappointment before this [letter] reaches you … the fact is that I have become more antagonistic than ever to the humoral view.”8
John Eccles is generally considered to have been the leading and most influential opponent of chemical transmission. Eccles was a native Australian who, after completing his medical degree with first class honors at Melbourne University in 1925, enrolled on a Rhodes scholarship at Magdalen College, Oxford. From 1927 to 1931 Eccles worked in Charles Sherrington’s laboratory, receiving his doctoral degree in 1929. Much of his research, including his thesis, was on neural excitation and inhibition. Eccles was under consideration to succeed Sherrington, but when he was not chosen he returned to Australia in 1937 to head a small research unit in the pathology department of a hospital in Sydney.
In a letter dated September 16, 1937, Dale wrote to Eccles:
Putting all chaff aside, I do want to take the opportunity to say how much I deplore your departure from physiology in this country. We have, in the last few years, had numerous opportunities of controversy, which on your side, and I hope on ours, has always been fair and good tempered. Our differences of opinion and interpretation, however, have not in the least weakened my admiration for the splendid work which you have been doing … I hope very sincerely that you will find conditions in Sydney such as to enable you to continue the work which you want to do…. If Sydney does not give you the opportunity that you want, you must come back to us.9
In Eccles’ response to Dale on November 28 he noted that he had been in Sydney seven weeks and was beginning to feel at home. He wrote that that his “artillery” would soon be ready, but he expected that by that time “you will have constructed new lines of fortification for one to have a crack at. Of course at this distance we will miss the short range practice that we had at the Physiological meetings, but, if you don’t make it too hot for me, I may venture over to England in a few years.”10
In 1944 Eccles left Australia to become chair of physiology at the University of Otago Medical School in New Zealand. He had written earlier to Dale asking if he would recommend him for the position, and Dale wrote back that he would be very glad to act as a “referee” for him.11 At Otago, Eccles and the philosopher of science Karl Popper spent much time together. His later books and theoretical articles on the brain, mind, and consciousness reflect Popper’s influence. Eccles returned to Australia in 1952 as professor of physiology at the Australian National University in Canberra and remained there until 1966. These were Eccles’ most productive years. During the years in Canberra, Eccles had an estimated seventy collaborators from twenty-two different countries, and wrote approximately four hundred scientific papers and three books. He attracted some outstanding collaborators, including Bernard Katz, a future Nobel Laureate, and Stephen Kuffler, who would later be elected to both the National Academy of Science in the United States and the Royal Society in Great Britain.
Eccles was elected to the Royal Society in 1941. He was knighted by Queen Elizabeth II in 1958, and in 1963 Sir John shared the Nobel Prize in Physiology or Medicine with Alan Hodgkin and Andrew Huxley.12
Prior to returning to Australia, while at Oxford, Eccles had regularly opposed the theory of chemical transmission, particularly at the synapse between spinal motor nerves and skeletal muscles. Despite this, a warm relationship existed between Eccles and the group associated with Dale. Feldberg later wrote that Eccles’ opposition “had a most beneficial effect. We were not allowed to relax, but were forced to accumulate more and more detailed evidence in support of our theory”13 Similarly, Dale described Eccles as an “ideal sparring partner,” who honed the arguments supporting chemical transmission. Eccles apparently felt the same way, judging from his later description of an exchange he had with Dale at a Royal Society symposium held in 1937: “I learned there the value of scientific disputation—that it provides a great incentive to perfect one’s experimental work and also to examine it more critically. Of course the critical appraisal is even more searchingly applied to the experiments of one’s opponents.”14
The debates between Eccles and Dale were always constructive and conducted in a friendly spirit, even though it often appeared to those who did not know them that they were about to come to blows. Sir Bernard Katz, who left Nazi Germany for England in 1935, described his first visit to Cambridge to attend a meeting of the Physiology Society that year.
To my great astonishment I witnessed what seemed almost a stand-up fight between J. C. Eccles and H. H. Dale, with the chairman E. D. Adrian acting as a most uncomfortable and reluctant referee. Eccles had presented a paper in which he disputed the role of acetylcholine as a transmitter in the sympathetic ganglion, on the grounds that eserine, a cholinesterase inhibitor, did not produce the predicted potentiating effect…. When Eccles had given his talk, he was counterattacked in succession by Brown, Feldberg, and Dale…. It did not take me long to discover that this kind of banter led to no resentment between the contenders, it was in fact a prelude to much fruitful discussion over the years and indeed to a growing mutual admiration between Dale and Eccles.15
Figure 8.1 John Eccles in his neurophysiology laboratory.
Eccles and Dale often exchanged letters explaining their positions or their reactions to each other’s publications, and they customarily sent each other copies of manuscripts before they were published. Their differences were often softened by humor. On one occasion, for example, Dale and Feldberg had just finished a vigorous tennis game when they met Eccles. They had worked up quite a sweat, and Dale could not resist remarking to Eccles that the sweat in their socks was undoubtedly dripping with acetylcholine. This was an allusion to Dale and Feldberg’s demonstration that the nerves that innervate the sweat glands on animal paws secrete acetylcholine, an exception for a sympathetic nerve. Eccles replied that before he could accept that they would have to assay their socks.16 Another example is a 1937 exchange between Feldberg and Eccles. Feldberg was working in a laboratory in Sydney at the time, and Eccles, who was in Canberra, sent a message to him. The message was: “Acetylcholine is all wet.” Feldberg immediately sent back a telegram: “Prefer wet acetylcholine to dry eddy currents.”17
In his 1937 review of the arguments for and against chemical and electrical transmission of nerve impulses to skeletal muscle, Eccles described the two positions as being “in sharp contrast,” with no apparent compromise possible.18 His review of the evidence and arguments for the two theories was, however, balanced, and he acknowledged that there was as yet “no conclusive evidence for or against either of them.” He did conclude, however, that acetylcholine could not be the main means of transmission, although he acknowledged that it “may have a secondary action in raising the excitability of the effector cells and in counteracting the onset of fatigue.”19
In December 1939 Cannon wrote a strong defense of chemical mediation of neural impulses in which he primarily addressed the opposing arguments of John Eccles and also John Fulton.20 Cannon referred to those defending the traditional electrical theory as “electragonists,” while he called those supporting chemical mediation “chemagonists,” explaining that “agonist” meant contestant or combatant. He noted that the electragonists were willing to grant that acetylcholine mediated transmission at the synapse between the vagus and the heart, but denied any important role for that substance at the spinal nerve synapse on skeletal muscles: “they agree that acetylcholine is a deputy of nerve impulses at vago-cardiac synapses, but deny it that function for myoneural synapses.”21
Cannon began by listing what both sides agreed on. They agreed, for example, that at parasympathetic synapses the perfusion of minute quantities of acetylcholine evoked the same responses as neural impulses. The electragonists also accepted the evidence, most of it collected by Dale and his colleagues, that acetylcholine was released at postganglionic parasympathetic synapses, at all autonomic ganglia, and also at the synapses between spinal nerves and skeletal muscles.
However, Cannon explained, neurophysiologists believed that acetylcholine was important only for evoking the slow visceral responses, and could not mediate the fast responses of skeletal muscles. Cannon noted that Eccles maintained that acetylcholine had only a “trophic influence” on skeletal muscle responses, increasing the response generated by the electrical impulse. Fulton was cited as describing the acetylcholine released as only a “byproduct of nerve metabolism.”22
Cannon answered the argument that chemical transmission is too slow by pointing out that this was only an assumption, as “too little is known of the speed of chemical processes at synapses to justify categorical limitations.” Moreover, he noted, the argument could be turned around by asking: “How can the electragonist explain the 0.2–0.4 millisecond delay that everyone agrees occurs at synapses?” This delay, Cannon argued, is five to eleven times greater than is required for electrical transmission across the synapse: “The relatively long delay in the [autonomic] ganglion is matched by a similar delay in the motor end plate [of skeletal muscles]. Here the electragonists have a real problem. The chemagonists, on the other hand, can readily account for the extra time as due to the requirements of an interposed chemical mediation.”
Cannon then discussed several phenomena that he considered difficult, if not impossible, for the electragonists to explain. Among them was the fact that there are certain drugs that either block or enhance the response to nerve stimulation without altering the nerve impulse, but do modify acetylcholine effectiveness. He cited the example of curare, which blocked the response of skeletal muscles, but did not interfere with the nerve impulse. Curare did, however, raise the threshold of response to acetylcholine. Similarly, a small intravenous injection of eserine enhanced skeletal muscle responses to nerve stimulation although it did not alter the nerve impulse. Eserine, Cannon reminded readers, prolongs the action of acetylcholine by inhibiting cholinesterase. Such results, Cannon concluded, could be understood only by assuming that the nerve impulse is effective only to the extent that acetylcholine is active.
Cannon cited a number of other lines of evidence that presented problems for the electragonists but could be explained by the known characteristics of acetylcholine, cholinesterase, and various drugs. He then concluded:
All these observations harmonize perfectly with the chemical theory of transmission and find no illumination whatever in the electrical theory…. Until the electragonists can display an instance of electrical transmission without acetylcholine at neuromuscular and neuroneuronal synapses their argument cannot be on the same footing as that of the chemagonists.23
As persuasive as Cannon’s arguments may now seem, he did not change the mind of anyone committed to electrical transmission. At a 1939 symposium on the synapse, for example, when Dale remarked that nature would not have arranged for acetylcholine to be released at synapses just to fool physiologists, a neurophysiologist responded that the same could be said for the presence of action potentials that crossed the synapse. Eccles later characterized his position at the time: “My position was not that chemical transmission did not occur, but that it was a later slow phase of transmission, the early fast phase being electrical.”24 Eccles also claimed that the fast responses of muscles were not affected by eserine.
One of the problems that neurophysiologists had the most difficulty explaining was how at some sites a neural impulse produced excitation and at other sites it produced inhibition. Inhibition was clearly a problem for the electrical transmission theory. The various explanations proposed usually involved complex and often strained hypotheses of how wave interference, or “eddy currents” of different polarities, might produce inhibition.
Eccles had been attempting to explain how eddy currents could cross the synapse and, depending on their polarity, produce either excitation or inhibition. In 1947 he had a sudden inspiration that led to a new theory to explain inhibition of skeletal muscles. The theory was based on the properties of a small “interneuron,” a neuron discovered with the aid of the Golgi stain. The cell was believed to have an unusual pattern of discharge and was found to exert an inhibitory influence on spinal motor neurons. It was initially called the Golgi cell, but today is generally called the Renshaw cell, after its discoverer, Birdsey Renshaw of the Rockefeller Institute for Medical Research.25 Eccles’ theory of inhibition was based on some properties this cell was believed to possess. He described how the theory had occurred to him in a dream: “On awakening I remembered the near tragic loss of Loewi’s dream, so I kept myself awake for an hour or so going over every aspect of the dream and found it fitted all experimental evidence.”26
Eccles wrote an article describing his theory, including a diagram purporting to explain how the “Golgi cell” could produce inhibition by inducing an anodal state at its point of contact with a postsynaptic neuron. The theory, which was published in the journal Nature in 1947, received a lot of attention and for several years it seemed to be gaining some experimental support. In the end, however, it was Eccles who proved the theory wrong, and he, to his credit, was the first to acknowledge it.
To measure inhibition, neurophysiologists today determine the voltage difference between the inner core of a neuron and its outer membrane. If the voltage difference is increased it is more difficult to excite (depolarize) the neuron. This is what constitutes an inhibitory state, or an inhibitory postsynaptic potential (IPSP). Just the reverse occurs when a neuron’s state of excitability is increased and an EPSP is induced. Until 1951 the evidence for inhibition in the spinal cord had been collected with extracellular recording electrodes, which were relatively insensitive to voltage changes of the magnitude that had to be measured to detect inhibition in a single cell. In mid-1951 glass microelectrodes capable of penetrating and recording from single neurons became available. They were glass tubes filled with a saline solution and measuring only about one-50,000th of an inch. The electrodes made it possible to obtain much more precise measurements of the voltage changes that characterized inhibitory and excitatory states. The technique was rapidly adopted by Eccles.
The critical experiment was done one day in mid-August of 1951 by John Eccles, Jack Coombs, and Lawrence Brock. It was quite a day, because in the middle of the experiment, while Eccles attended to the cat they were using, Brock left the lab for the hospital to attend to Coombs’ wife, who was delivering a baby girl. After the birth, they reassembled around the cat and by the end of a long day of recording voltage changes from the spinal motor nerves that innervate skeletal muscles, it became clear that Eccles’ “Golgi-cell hypothesis of inhibition” was wrong and that only chemical transmission could explain their results. As they wrote in the publication,
The potential change observed is directly opposite to that predicted by the Golgi-cell hypothesis, which is thereby falsified…. It may therefore be concluded that the inhibitory synaptic action is mediated by a specific transmitter substance that is liberated from the inhibitory synaptic knobs and causes an increase in polarization of the subadjacent membrane of the motor neurone.
The striking contrast between the small presynaptic spike and the large [post-] synaptic potential further makes it appear improbable that the trans-synaptic flow of current would be adequate to evoke that large post-synaptic response. On the electrical theory of transmission it does not seem possible to provide an explanation for the great amplification observed. On the other hand, just such a large amplification is actually produced by the chemical transmitter mechanism at the neuromuscular junction, where the most probable explanation is that the acetylcholine liberated from the nerve terminals triggers off the sodium carrier mechanism.27
The “sodium carrier mechanism” had been discussed in recent publications by the British neurophysiologists Alan Hodgkin and Andrew Huxley, who had shown that the passage of sodium and other ions in and out of a neuron is what determines its state of excitability.28 Hodgkin and Huxley also described the “sodium pump” mechanism, which increases the voltage differential across a neuron by pumping sodium out of the cell against an osmotic gradient. Based on this work, Eccles and his collaborators concluded:
The experimental observations on synaptic excitatory and inhibitory action require for their explanation two specific transmitter substances. The excitatory substance probably acts by stimulating the sodium carrier mechanism, while it is suggested that the inhibitory substance possibly acts by stimulating the sodium pump.29
Eccles sent Dale an advance copy of the manuscript, writing that “I hope that these modest offerings to the theme of chemical transmission will in some measure atone for my long delay in conversion.”30 Dale replied that he had derived great pleasure from reading the paper, adding: “Your new-found enthusiasm [for chemical transmission] is certainly not going to cause any of us any embarrassment.” In a second letter, Dale wrote: “I do congratulate you all, not only upon the beauty of the observations recorded, but on the very attractively clear and concise account of them in the paper.31 Dale later wrote that Eccles’ change was like the conversion of Saul on the way to Damascus, when “the sudden light shone and the scales fell from his eyes.”32
Although Eccles had capitulated in 1952 and accepted the chemical transmission hypothesis for spinal motor neurons, this did not abruptly end the opposition by others to the theory of chemical transmission at central nervous system synapses. A history of long and influential opposition had to be overcome. Throughout the 1940s and much of the 1950s, neurophysiologists had continued to insist that transmission was primarily electrical in the spinal cord and probably exclusively electrical at brain synapses. The neurophysiologists favoring the electrical hypothesis and the pharmacologists supporting the chemical hypothesis used such different techniques and obtained data so difficult to compare that each tended to ignore the other side’s arguments. Neurophysiologists might talk in private about chemical transmission, but they usually avoided the subject in their publications. This led Dale to comment later that the subject of chemical transmission was treated “like a lady with whom the neurophysiologist was willing to live and consort in private, but with whom he was reluctant to be seen in public.”33
John Fulton, the editor of the Journal of Neurophysiology and one of the most influential neurophysiologists at the time, was a strong supporter of the electrical transmission position, but sometimes he seemed to be hedging his bets. In a letter dated August 22, 1943, Dale, displaying his masterful sense of irony, wrote to John Eccles: “I am told that John Fulton, in a recent number of Science, has begun to balance himself more carefully than before on the top of the hedge, so that eventually we may find you all on the same, safe side.”34
In 1949 John Fulton was still carefully on “the top of the hedge” while continuing to give more weight to the electrical transmission explanations. He had written in his influential textbook, Physiology of the Nervous System:
[Although] the theory of chemical mediation of nerve impulses appeared acceptable to many physiologists in the case of autonomic nerves acting on their effector organ, this concept, when applied to synapses and neuromuscular junctions, was less satisfactory and encountered increasing opposition. In addition to a great number of difficulties and contradictions, which were partly reviewed by John Eccles (in 1937) and have increased continuously since then, there are two main objections, the first being the time factor. This factor was of less importance in the case of the slowly reacting cells innervated by the autonomic nervous system. But the transmission of nerve impulses across the neuromuscular junctions and synapses occurs within milliseconds. No evidence was available that the chemical process can occur at the high speed required…. The idea of a chemical mediator released at the nerve ending and acting directly on the second neuron or muscle thus appeared to be unsatisfactory in many respects.35
In a later review of this history, Eccles wrote that at a symposium held in Paris in 1949 many participants conceded that acetylcholine mediated transmission at various peripheral synapses, “however, there was still fairly general agreement that central synaptic transmission was likely to be electrical.” He also recalled that a great amount of opposition to the theory of chemical transmission at central nervous system synapses was still being expressed by Albert Fessard and others at a 1951 symposium in Brussels.36
In general, the possibility of chemical transmission hardly had any effect on the thinking of scientists, aside from those directly involved in studying synaptic transmission. For example, in their highly successful physiological psychology textbook published in 1950, Clifford Morgan and Eliot Stellar described synaptic transmission as occurring when the electrical changes in the presynaptic neuron induce a depolarization in the postsynaptic neuron. They did not mention chemical transmission as even a possibility.37 The text presented, without comment or qualification, Eccles’ earlier diagram, which attempted to explain inhibition as solely an electrical process. In later editions there was no reference to the fact that Eccles had changed his mind. Neither did Morgan and Stellar mention acetylcholine when describing the autonomic nervous system. Not only the Morgan and Stellar book, but virtually all physiological psychology textbooks of this period did not consider any of the implications of chemical neurotransmitters, if they mentioned the topic at all.
By 1953 some neurophysiologists had accepted John Eccles’ evidence that transmission at the synapse between spinal motor nerves and skeletal muscle is chemical, but there remained many who were not persuaded that acetylcholine is the primary means of transmission at this synapse. At a 1953 Philadelphia symposium on chemical transmission, Ralph Gerard, a renowned neurophysiologist and the person credited with introducing the term “neuroscience,” was given the task of summarizing the meeting. Those in attendance were primarily the Europeans who had provided the bulk of the evidence supporting chemical transmission in the periphery, including Otto Loewi and Henry Dale. Gerard began by remarking, “As the lone American in this distinguished galaxy and as one of the few physiologists amidst the pharmacological cohorts, the Committee may have wished me to serve as devil’s advocate for an electrophysiological approach to neural functioning.”38
Gerard accepted the challenge and presented a number of arguments for electrical and against chemical explanations. He argued that Eccles’ 1952 experiment had not completely ruled out electrical explanations of inhibition in the spinal cord. Gerard raised the speed of conduction issue, stating that:
No liberation or diffusion of chemicals could account for the ability of an impulse to jump over a millimeter of inactivable nerve fiber in a fraction of a millisecond…. Moreover, many, if not all of the phenomena of junctional transmission are neatly accounted for by the properties of eddy currents and the geometry of the junctional region.
The weakness of Gerard’s arguments are now apparent, but at the time they provided support for those who refused to accept chemical transmission.
Ralph Gerard was not alone among neurophysiologists in contesting Eccles’ evidence. At the same 1953 Philadelphia symposium, A. K. McIntyre of the University of Otago in New Zealand showed resentment that John Eccles had abandoned the neurophysiologists’ position. McIntyre, who was a colleague of Eccles at Otago and had published an article with Brock, one of his collaborators, remarked that he found it: “particularly entertaining that so much of the newest evidence presented in favour of neurohumoral transmission mechanisms in the central nervous system comes, not only from an electrophysiologist, but from such a high voltage spark as J. C. Eccles.”39 McIntyre went on to state that: “A critical evaluation of the very interesting experiments of Brock, Coombes, and Eccles with intracellular recordings from motorneurones reveals that rather sweeping conclusions have been drawn from somewhat slender evidence.”
While the nature of the innervation of skeletal muscles was still being disputed by neurophysiologists in 1953, the possibility of chemical transmission in the brain was rarely even discussed by that group. Wilhelm Feldberg, a pharmacologist, had been studying the effects of injecting various drugs into the ventricles of the brain, and he raised the question of the nature of synaptic transmission in the brain at the Philadelphia symposium. Feldberg stated that in his view synaptic transmission in the brain is chemical and “that there is no fundamental difference between the transmission processes in the central nervous system and those in the peripheral nervous system.” Feldberg did make it clear, however, that the available evidence supporting either side of the dispute was not yet convincing.
We are faced with the problem that, whatever our bias, we cannot state with certainty whether the transmission is chemical or electrical. But, according to which view we hold, our approach will be different. Anyone who takes it for granted that transmission is electrical, assumes the phenomenon can be fully dealt with by analyzing electrical changes in the central nervous system, and such an analysis forms the main subject of his research. On the other hand, the adherents of the chemical theory believe that a fruitful analysis of this kind requires accurate knowledge of the nature of the central synaptic transmitters.40
How the evidence accumulated to prove that neuronal transmission in the brain is chemical is the subject of chapter 10.
Henry Dale, Otto Loewi, and Walter Cannon were too old to contribute to this later part of the story. This book, however, is not only an account of scientific discoveries, it is also the story of the scientists involved—their personalities, how they worked, their friendships and disputes, and the social and political events that had an impact on them and their work. It seems only appropriate, therefore, to first describe the final years of the three principal scientists who established the foundation for the revolutionary changes in our understanding of how the nervous system works. That is the subject of the next chapter.