Helped perfect the glass electrode that reads voltage inside cells, then spent seventy years denying the sodium pump. Lost his lab in 1988; an MRI pioneer’s firm kept him publishing.
Glass Needle
Drew a capillary fine enough to enter one frog muscle fibre and read its voltage.
Poisoned Muscle
Starved frog muscle of energy at 0 °C; its potassium and sodium held for hours.
Bound Potassium
Argued cell potassium sits stuck to protein carboxyl groups, not free in water.
Gilbert Ling helped perfect the glass microelectrode that lets physiologists read the voltage inside a single living cell, then spent the next seventy years arguing that the textbook cell is wrong. That cell has a membrane studded with pumps that haul sodium out and potassium in. Ling held that the pumps do not exist, that a cell’s potassium is stuck to its proteins and that its water is ordered rather than free. Pump enzymes were then purified and their structures solved, and Jens Skou took the 1997 Nobel Prize in Chemistry for the first of them. Ling lost his laboratory in 1988 and spent his last three decades working from the Long Island company of the MRI pioneer Raymond Damadian, publishing mostly in a journal he edited himself. He is worth reading because his model sticks its neck out: it says where cell potassium and water sit, and several of those claims were put to the test, some by his own former students. Start with his own 42-page summary, then read the short paper in which two of them argued the other way.
1919–1945
A Boxer scholar crosses the Himalayas
Gilbert Ning Ling was born in Nanking on 26 December 1919. He described his family as scholarly and Confucian, and he grew up in Beijing. He began at National Central University, then relocated to wartime Chungking, as an animal-husbandry student, moved to biology after two years, and took his B.Sc. in 1943 with minors in physics and chemistry. Graduate work in biochemistry followed at the National Southwestern Associated University in Kunming.
His way out of a country at war was an examination. The Boxer Indemnity scholarships were paid for with money China owed the United States after the Boxer Rebellion, which the Americans chose to spend on Chinese education. In Ling’s account the scholars were picked one per field by a nationwide, week-long test with the candidates’ names sealed. Five exams had been held by 1938, when the Japanese invasion put an end to them; the series was revived once more in 1943, with a new rule that let fresh graduates sit. Ling, who graduated that summer, would otherwise have been ineligible. He won the biology (more exactly zoology) place. His roommate, Chen-ning Yang, won physics. The announcement is dated 1944, so the scholarship is often given as 1944; the journey was in 1945.
That year Ling, Yang and most of the twenty-odd other scholars flew over the Himalayas, saw Calcutta and Bombay, and spent a month at sea by way of Suez and the Atlantic. Ling remembered arriving at Manhattan on a cold November afternoon with light snow falling on the Hudson. By Christmas he was in Chicago, meeting Ralph W. Gerard, who at once began talking about a thesis topic: a small piercing electrode. Formal graduate study in physiology began in early 1946. The roommate stayed in the story. Yang, later a Nobel laureate in physics, co-authored with him the cooperative adsorption isotherm of 1964 that carries both their names, and in later years Ling quoted Yang’s praise of his 1962 book.
Gerard and his student Judith Graham had published a glass-capillary electrode in 1942 and 1946, but when Ling arrived it was, in his words, not yet in working condition. A random set of readings from individual frog sartorius muscle cells ran from 41.0 to 80.4 millivolts, far too scattered to measure anything. The idea is simple. A glass tube is heated and drawn out to a tip finer than a muscle fibre, filled with a conducting salt solution, and pushed through the cell’s surface; the voltage between the inside of the cell and the bath can then be read. Ling gives no engineering details in the accounts we read, only the outcome: he got it working.
In his recollection, Graham, who was trying to finish her own Ph.D. while raising two young children in wartime, did not have time for the nitty-gritty, whereas his full scholarship left him free to try different ways of making the electrode work. “Good luck once more favored me,” he wrote. That is his account of a colleague, and we have not seen hers. He calls the result the Gerard–Graham–Ling microelectrode, or the Ling–Gerard electrode.
The thesis, on how metabolism, temperature and other factors change the membrane potential of single frog muscle fibres, was completed in 1948. It appeared in December 1949 as four papers in the Journal of Cellular and Comparative Physiology, three with Gerard and one with the visiting scientist Walter Woodbury, and Ling stayed on for two years as a postdoctoral fellow with Gerard. By his summary, the potential tracked absolute temperature and the logarithm of the external potassium concentration, just as Julius Bernstein’s membrane theory of 1902 said it should. The experiments that made him a lifelong dissenter began by confirming the orthodoxy; his own thesis, he noted later, was founded on the assumption that the membrane theory had full validity.
The technique spread. Ling says he taught Alan Hodgkin how to make a usable electrode, and cites a 1971 review by the Swiss cardiac physiologist Silvio Weidmann, with a section headed “The cardiac descendants of Gilbert Ling”, which maps the laboratories that learned the craft of pulling, filling and prodding through personal contact. Later, Ling complained, the name Ling–Gerard electrode was stripped down to “glass capillary microelectrode”, a mark of displeasure at his challenge to the sodium pump. Whether that is a slight or the ordinary fate of a technique’s authors we cannot settle; it is said that John Eccles and Andrew Huxley used the electrode, but we found no reference for that. The instrument is the part of Ling’s record that nobody disputes.
On a Monday in the spring of 1948 (the date in his 1997 and 2007 papers; his 42-page summary says spring 1947), Ling gave the departmental seminar in Chicago on a topic he had been handed: the sodium pump. He began with an apology. He had searched the library and found nothing substantial; the only thing he could say with confidence was that nobody seemed to know much about it. Robert Dean had suggested in 1941 that a “pump of unknown mechanism” was pushing sodium out of cells, with neither a mechanism nor evidence to offer. Afterwards, Ling says, two of his respected professors, whom he does not name, took him aside separately and said almost the same words: the sodium pump is a sacred cow, leave it alone, nothing is gained by making yourself a martyr. He thought they were overly worried.
He did what a graduate student with frogs and free time could do. If a pump in the membrane burns metabolic energy to keep sodium low and potassium high, cutting off the energy supply should let sodium flood in and potassium leak out. He starved frog muscles of energy with pure nitrogen, which stops respiration, and sodium iodoacetate, which stops glycolysis, and kept them at 0 °C, which should slow a chemical pump more than it slows passive leakage. What he saw was not what he expected: sodium and potassium both stayed at their normal levels for as long as the experiment ran, hours after the energy source was cut. It puzzled and excited him. Ion distribution, he decided, was more mysterious and more fundamental than the voltages he had been measuring.
In 1950 he moved to the Johns Hopkins medical school in Baltimore as an instructor, and he kept at it. A short paper in the American Journal of Physiology in 1951 and a long one for a symposium on phosphorus metabolism in 1952 reported that muscle creatine phosphate and ATP, the stored energy, changed little while the ion levels held. From the numbers he made a rough estimate: even if the muscle spent all its energy on pumping sodium, and every step ran at 100 per cent efficiency, the pump would need at least four times what was available. Between 1951 and 1956 he refined this with radioactive sodium (22Na and 24Na) counted in a well-type scintillation counter, with muscles in a cold room just above freezing and microelectrode readings from the same muscles. The definitive runs of 1956 gave minimum energy needs of 15 to 30 times the maximum available. In Ling’s count this was the first of six independent disproofs of the pump.
The reply to it attacked the premises rather than the arithmetic. The estimate assumes that all the sodium in a muscle is free, cytoplasmic sodium that a pump has to move; opponents argued that a large fast-exchanging share is sequestered inside the sarcoplasmic reticulum, and so is not pumped at all. Ling’s rebuttal, in his 1997 paper, is that the reticulum lies entirely within the sarcoplasm, as Peachey’s electron microscopy of 1965 showed. One afternoon in 1950, meanwhile, in the Welch Library in Baltimore, he had had an idea for a molecular mechanism that could gather potassium in preference to sodium without spending energy continually. He presented it at the Baltimore phosphorus symposium in 1951 as the fixed-charge hypothesis, and it became the seed of everything after.
Ling’s answer to the obvious objection, that without a pump a cell should not be lopsided, is the association–induction hypothesis. It was built in three steps. The first is the fixed-charge idea of 1952. Muscle proteins, above all myosin, carry many carboxyl groups hanging off their aspartic and glutamic acid residues, each a fixed negative charge. A hydrated potassium ion is smaller than a hydrated sodium ion, so it can sit closer to such a charge and be held harder. Cell potassium, Ling concluded, is not a free ion in solution but is adsorbed on protein, one ion on one carboxyl group, and so needs no continuing energy to keep it in the cell. By his account myosin alone has sites enough to hold all the potassium in a muscle cell.
The second step is water. In his polarised–oriented multilayer theory, first sketched in 1956 and folded into the hypothesis in 1965, nearly all the water in a cell is not bulk liquid but lies in ordered layers on extended protein chains. That water is a poorer solvent for sodium and for large solutes, so sodium is partly shut out, again with no pump. The third step, the induction that gives the theory its second name, is the control system. How strongly a carboxyl group prefers potassium depends on how electron-rich its oxygen is, a quantity Ling called the c-value; his calculations gave a crossover, with a low c-value favouring potassium and a high one favouring sodium. ATP, bound at a “cardinal site” on the protein, shifts the c-values of groups along the chain by induction, tipping the protein, its potassium and its water together between a resting living state and an active one. ATP is therefore not fuel for a pump but a switch that works by being bound, and he counts at least 8,000 water molecules under the influence of one ATP.
The 1962 monograph, A Physical Theory of the Living State (Blaisdell, Waltham, Massachusetts; about 680 pages), set this out and, in its eighth chapter, the 1956 energy experiments. By then Ling had moved from Baltimore to the Neuropsychiatric Institute of the University of Illinois (1953–57) and the new Eastern Pennsylvania Psychiatric Institute (1957), and in 1962 he became director of a research laboratory at Pennsylvania Hospital in Philadelphia. In 1964 he and Yang published a cooperative adsorption isotherm borrowed from the one-dimensional Ising model, to explain why potassium uptake by muscle rises in an S-shaped curve, like oxygen uptake by haemoglobin: binding at one site makes the next more likely.
The hypothesis names its own tests. If cell potassium sits on carboxyl groups, it should follow the ATP content of the cell, resist displacement by doubly charged magnesium, and compete with caesium and thallium in a way that only close contact can explain. If the pump is not needed, a protein solution with the right control agent should exclude sodium all by itself. Ling and Margaret Ochsenfeld filled dialysis sacs with a 40 per cent solution of pure haemoglobin and soaked them in 10 millimolar sodium chloride with varying amounts of hydrochloric acid, which stands in for ATP as an electron-withdrawing agent. At a final pH of 2 to 3, they report, the sodium in the sac fell to 15 to 30 per cent of that outside, as in a red blood cell. A conventional chemist might ask whether this is just a Donnan effect, since haemoglobin at that pH carries a large net positive charge that repels positive ions; we did not find the 2008 paper addressing that directly. The question is ours, not a published criticism.
Ling and his students ran the potassium tests through the 1960s and 70s, and four are easy to hold in the head. First, ATP: poison frog muscles with any of ten different poisons for varying times and the potassium they hold falls in step with their ATP content (Gulati, Ochsenfeld and Ling, 1971). Second, cut ends: mount a frog sartorius so that only the cut end touches Ringer’s solution containing labelled potassium and sodium, with the rest in moist air or under vaseline, where no pump could draw on a bath. Cells far from the cut end still held high potassium and low sodium, Ling reports, while near the cut end, where ATP had fallen to nothing, the accumulated potassium was gone. He called it an EMOC preparation, an effectively membrane-pump-less open-ended cell. Third, magnesium: if some potassium were a free cloud around the fixed charges, doubly charged magnesium should displace it, but raising magnesium inside and outside the muscle displaced none. Fourth, caesium: two ions of identical charge but different short-range properties, potassium and caesium, compete for sites in ways that only close one-on-one contact would explain, and in labelled-ion experiments they do (the argument is Ling and Ochsenfeld’s, from 1966).
Then came a test from outside the programme. In 1976 Lawrence Palmer and Jagdish Gulati reported in Science that living frog muscle in vitro could accumulate potassium to concentrations as high as 580 millimolar, and that both the amount and the relationship between internal and external potassium indicated that potassium is “free” under all conditions rather than bound to cell macromolecules. They also concluded that at most 20 per cent of the cell’s water is “bound”, in the sense of excluding electrolytes. Both men had worked in Ling’s laboratory: Gulati is a co-author of the 1971 ATP paper, and Ling calls Palmer and Gulati two of his “fleeing students”. We have read only the abstract; the question it raises is whether cell potassium behaves like a bound quantity with a ceiling or like a free ion that follows its surroundings. Ling replied in Science in 1977, in a paper he considers a full rebuttal, under the title “Potassium accumulation frog muscle: the association–induction hypothesis versus the membrane theory”, and Gulati and Palmer answered in the same issue. Each side stayed where it was.
The same year as Palmer and Gulati, Science’s news section ran Gina Kolata’s article “Water structure and ion binding: a role in cell physiology?”. Ling says it reported that two more of his former students, Jeffrey Freedman and Chris Miller, had produced “crucial experiments and calculations” giving strong evidence that pumps exist. He learned of this about twenty years later, when Gerald Pollack drew it to his attention, and in 1996 he wrote to all three asking for the published source. By his account Freedman never answered. Kolata replied on 16 July 1996 that she could no longer remember where she got the information, that Miller and Freedman had never published their results, and that “Dr. Miller says I never even interviewed them”. Miller wrote on 28 June 1996 that he and Freedman had written a review of sodium-efflux data allowing for compartmentation, which a journal rejected for having no new data and which they never published. We have these letters only as Ling reproduces them. He called it “a sad and miserable page in the history of science”, in which innocent young students were, in his words, “coerced into activities of which they should be profoundly ashamed”, and it drove him to write the 75-page “Debunking” paper of 1997. Note that we have heard only one side of this story.
The field’s answer did not come as an argument but as an object. In 1957 Jens Skou, at Aarhus, published “The influence of some cations on an adenosine triphosphatase from peripheral nerves”, and forty years later he shared the 1997 Nobel Prize in Chemistry for the first discovery of an ion-transporting enzyme, Na,K-ATPase. Ling’s response was to argue that the energy in ATP’s high-energy phosphate bond was illusory, citing Podolsky and Morales, and so no pump could be driven by it.
Three lines of work are worth setting beside Ling’s experiments. In 1968 R. D. Keynes and R. A. Steinhardt took frog muscle, Ling’s own preparation, and found that in normal Ringer’s solution with 2.5 millimolar potassium only 37 per cent of the sodium efflux was blocked by ouabain, rising to 75 per cent in sodium-loaded muscles, and that only the ouabain-sensitive part depended on external potassium. They thought the rest might come from the sarcoplasmic reticulum, though they said a definite identification could not yet be made; it is the very point Ling disputes. In 1975 Hilden and Hokin purified the enzyme from the rectal gland of a shark, put it in lecithin vesicles and showed ATP-driven transport of sodium and potassium, inhibited by ouabain and coupled as sodium in, potassium out: a membrane and an enzyme with no cytoplasm and no cell. Ling’s third disproof runs the other way: he reports that squid-axon membrane sacs emptied of axoplasm neither accumulate potassium nor expel sodium. Whether such a sac failed because there is no pump or because the preparation damaged one is the kind of question a reader can weigh against Hilden and Hokin, and this page does not settle it.
Then the structures. Toyoshima’s group solved the calcium pump of muscle in 2000 at 2.6 ångströms. In 2007 Morth and colleagues published the sodium–potassium pump from pig kidney at 3.5 ångströms with two rubidium ions, standing in for potassium, occluded in the transmembrane part; in 2009 Shinoda and colleagues reached 2.4 ångströms, with potassium and its coordinating water visible, and Ogawa and colleagues crystallised the pump with ouabain deep in the transmembrane domain, right beside the bound potassium. In Ling’s scheme ouabain is an electron-donating “cardinal adsorbent” that tips about a thousand carboxyl groups for each molecule; in the crystal it is a single molecule lodged in the pump beside the bound potassium, which is how the field reads it: an inhibitor of a pump.
Potassium selectivity was also settled elsewhere. In 1998 Doyle and colleagues, in Roderick MacKinnon’s laboratory, solved a bacterial potassium channel at 3.2 ångströms: a selectivity filter only 12 ångströms long, lined with main-chain carbonyl oxygens that coordinate potassium but not the smaller sodium, with two potassium ions about 7.5 ångströms apart. It is a mechanism for passing through a membrane protein, not for sitting in cytoplasm, but it answers the same why-potassium-and-not-sodium question that began Ling’s career. On 17 November 2003, after MacKinnon’s Nobel Prize in Chemistry, Ling wrote to him that his work rests on the membrane-pump theory, which Ling thought long disproved, and that Ling had first introduced a variety of its basic concepts, now found there unacknowledged. His title for the published letter calls this suspected plagiarism. By his account MacKinnon never answered it or the follow-ups of 22 February and 2 April 2008. We have not seen MacKinnon’s side, and nothing here adjudicates the charge.
From 1982 Ling was co-editor-in-chief, and from 1986 sole editor-in-chief, of Physiological Chemistry and Physics and Medical NMR (the acknowledgements of his own book Life at the Cell and Below-Cell Level give the dates), a journal that his own references show also carried many of his later papers: the 1997 Debunking paper, and those of 2005, 2007, 2008 and 2012 listed on his legacy site. Readers should bear in mind that his late work appeared in a journal he ran, and the independent checks are the ones elsewhere on this page.
In October 1988, at what he called the height of its productivity, his laboratory was shut. In his account the cause was the withdrawal of all public support, including the NIH money that backed his work, and he lays it on opponents who controlled jobs and research grants, the same people who he says drove his graduate students out of his lab in a body. We have not seen a grant review, an NIH summary statement or any funder’s account; a laboratory can lose funding for more than one reason, and this is his version of why.
He would, he wrote, have been thrown out of scientific research then and there, “were it not for Dr. Raymond Damadian (and his Fonar Co.), who has offered shelter and support for myself and two close associates ever since”. The two are Margaret Ochsenfeld and Zhen-dong Chen, thanked in his acknowledgements. From then on his papers carry the affiliation Damadian Foundation for Basic and Cancer Research, c/o Fonar Corporation, 110 Marcus Drive, Melville, New York, and later the Tim and Kim Ling Foundation for Basic and Cancer Research.
Damadian is the link between this story and an MRI scanner. Before Fonar he was a physician-scientist who treated the cell as an ion-exchange resin. In 1971 the Biophysical Journal carried his three “Biological ion exchanger resins” papers; the first reported that in E. coli about 80 per cent of the cell’s potassium is associated with fixed charge on macromolecules. In 1973, with Lawrence Minkoff, he published “Caloric catastrophe”, which found that bacteria’s energy supply, 4.20 calories, fell far short of the 28.28 calories that classical membrane theory needed to run their pumps, an echo of Ling’s arithmetic. In March 1971 Science carried his “Tumor detection by nuclear magnetic resonance”: relaxation times in two malignant rat tumours lay outside the range of six normal tissues, which he read as greater motional freedom of tissue water. A patent followed in 1974 and Fonar in 1978; in 1977 a FONAR image of the chest of a living human, the first of its kind, was published in Physiological Chemistry and Physics, Ling’s journal. The 2003 Nobel Prize in Physiology or Medicine went to Paul Lauterbur and Peter Mansfield, not Damadian, and his supporters ran newspaper advertisements in protest.
None of this makes MRI a proof of the association–induction hypothesis. It is reported that the relaxation-time differences between tumour and normal tissue were real but too variable for diagnosis, and a Smithsonian account says Fonar dropped Damadian’s focused-field method for the gradient imaging of Lauterbur and Mansfield. We have not read Damadian’s own account of what Ling gave him, so the intellectual debt is inferred from shared subjects and from Ling’s affiliation, not documented here.
Ling’s later books were not trade-press books. In Search of the Physical Basis of Life (Plenum, 1984) and A Revolution in the Physiology of the Living Cell (Krieger, 1992) came from publishers; Life at the Cell and Below-Cell Level (2001), a history of some 400 pages subtitled “The Hidden History of a Fundamental Revolution in Biology”, was published by Pacific Press of New York, with a Melville address, the town where Fonar sits. What Is Life Answered (2013), his reply to Schrödinger, he wrote in his nineties, and his later books are described as self-published. In 2004 he wrote to The Economist asking for a retraction of a 2003 article that he said attacked his reputation; the letter, which he says was never answered, he printed in his own journal in 2012. Ling died on 10 November 2019, weeks short of his hundredth birthday. His wife, the pianist and teacher Shirley Wang Ling, whom he had married in 1951, had died in 2011.
Ling’s water claims have a measurable side. With labelled tritiated water in giant cells, frog ovarian eggs and the muscle of giant barnacles, he reports that diffusion is limited by the bulk of the cell, not by the surface membrane, and that water inside diffuses at 30 to 60 per cent of its rate in liquid water, which matches figures from four other laboratories for barnacle muscle. Whether that slowdown reflects ordered water, as Ling argues, or has some other explanation is the live question, and Palmer and Gulati’s report that at most 20 per cent of cell water excludes electrolytes points the other way. We have not sourced a named critic’s alternative explanation of the diffusion figures.
Gerald Pollack, professor of bioengineering at the University of Washington, founding editor-in-chief of the journal Water and author of Cells, Gels and the Engines of Life (2001) and The Fourth Phase of Water (2013), is the best-known heir to structured-water ideas. Ling calls him a friend, credits him with telling him about the Kolata story, and wrote the opening chapter of Water and the Cell (2006), which Pollack edited with Ivan Cameron and Denys Wheatley. Pollack’s lab reports “exclusion zones”: next to hydrophilic surfaces such as gels, Nafion and plant xylem, water that shuts out suspended microspheres. In 2003 Zheng and Pollack reported solutes excluded to roughly 100 micrometres from gel surfaces and concluded that solute-repulsion forces exist at distances far greater than conventional theory predicts; in 2024 Wang and Pollack reported exclusion zones of up to 240 micrometres beside pumpkin xylem. Pollack’s lab home page does not mention Ling, so reading his work as a continuation of Ling’s is our interpretation, and exclusion-zone water has not entered textbook cell physiology. We have not surveyed its critics. Treat it as a live and contested claim, not as a vindication.
The hypothesis has also lived outside the lab. In 1978 Freeman Cope, in Ling’s journal, argued that the high-potassium, low-sodium diet of the Gerson cancer therapy works by restoring damaged cell proteins, an idea he credited to Ling’s laboratory; the abstract offers no trial evidence for the cures it mentions. Ray Peat later carried Ling’s cell physiology into a much wider, diet-and-hormone synthesis, which this site follows in his profile. Nothing reviewed here supports using the association–induction hypothesis as medical advice.
