A Career in Neurobiology
by Harold Hillman, M.B., B.Sc., Ph.D.
(Formerly Reader in Physiology at the University of Surrey, and Director of the Unity Laboratory of Applied Neurobiology).
3, Merrow Dene, 76 Epsom Road, Guildford, Surrey, GU1 2BX. United Kingdom.
|1||Brief biographical note||2|
|2||Major concerns with current beliefs in cell biology||2|
|3||Are lesions causes or effects of diseases?||3|
|4||Subcellular fractionation ignores the second law of thermodynamics||3|
|5||Control experiments for subcellular fractionation have not been carried out||4|
|6||Electron microscopy in the light of solid geometry||5|
|7||The ‘unit’ membrane is impossible in three-dimensional geometry||7|
|8||Structures in the cytoplasm||8|
|9||Dissection of mammalian nerve bodies||11|
|10||Structure of cell bodies||13|
|11||The nucleolus and the nucleolar membrane||14|
|12||Effects of staining on the appearance of cell bodies||15|
|13||Cellular structure of the nervous system||15|
|14||Fine granular material and the nature of neuroglia||16|
|15||The likely structure of the central nervous system||19|
|16||The sensitivity to light and other modalities||21|
|E||Dying, death and resuscitation||24|
|17||Studies of dying and death||25|
|18||Recovery from hypothermic cardiac arrest||27|
|19||Centripetal infusion and abdominal pumping||28|
|20||The electric chair||33|
|21||The slaughter of animals for food||35|
|F||The anatomical synapse|
|22||Does the synapse exist?||36|
|23||Critique of the chemical theory of transmission||38|
|24||New hypothesis of transmission||40|
|G||Fraud and parafraud|
|25||What is parafraud?||42|
|26||Truth in biological and medical research||45|
|27||Suggested code of practice||45|
1. A brief biographical note
I have London University degrees in medicine, in physiology and in biochemistry, and was Reader in Physiology and Director of the Unity Laboratory of Applied Neurobiology, at the University of Surrey, between 1968 and my retirement in 1995. I have written 6 books, and about 170 full length papers.
2. Major concerns with current beliefs in cell biology
I wish to put on record the conclusions of my research work, and the main reasons for these conclusions. They are, that most of the research work in cell biology since the Second World War, has not yielded valid results, for the following reasons: (i) many of the experiments have completely ignored the Second Law of Thermodynamics; (ii) they have produced two-dimensional images, which defy the reasonable expectations of solid geometry: (iii) research workers have failed to carry out the crucial control experiments, while teaching students the importance of such controls; (iv) they have failed to identify or warrant the assumptions inherent in their experiments; (v) they have indulged in these misdemeanours not only in their research in biochemistry, physiology and biophysics but also in respect of such cellular diseases as carcinomas, sarcomas, leukaemias, muscular dystrophies, Alzheimer’s disease, etc., etc., (vi) the consequence is that the experiments can not yield information about the chemical activities, the rates of reactions or the equilibria in cells of healthy or diseased living intact human beings or animals; (vii) ignoring the laws of geometry means that many of the images seen by electron microscopy are two-dimensional, that is, they must have arisen after the sections have been cut, and are artifacts; (viii) studies which ignore the Second Law of Thermodynamics, and laws of solid geometry, cannot possibly produce valid experimental data
3. Are lesions causes or effects of diseases
When lesions are found to be characteristic of particular diseases, the crucial question is, ‘Are these changes the causes of the lesions or the consequences of them? We must seek to elucidate the biochemical changes, which initiate the diseases, not the consequences of the diseases, that is, the geneses of the diseases, not the exoduses. For example, there are many conditions described as lysosomal, mitochondrial, free radical or autoimmune diseases. However the lesions of lysosomes, mitochondria, free radicals or
autoimmunity, occur in such a wide variety of diseases, each with its unique aetiology, natural history, treatment and outcome, that one has no alternative but to conclude that the changes in the particular organelles are consequences rather than causes of the lesions. If the causes could be understood, logical chemical measures could be designed to prevent them developing, and to treat them.
A. Subcellular fractionation
4. Subcellular fractionation ignores the Second Law of Thermodynamics.
If one asks any biochemist in the world, ‘What happens in the nuclei, the cell membranes or the mitochondria?’ he or she will tell you what they have gathered from the textbooks. This knowledge has been derived using subcellular fractionation.
However, during subcellular fractionation; the animal will have been killed; the organ will have been excised; sucrose will have been added; the organ will have been cooled; it will have been homogenised; it will have been centrifuged in a frozen centrifuge; it will have been ‘washed’; a layer will have been separated; each layer will have been examined under the microscope to identify its components; further chemical and physical procedures will be used to increase the concentration of the identifiable particles in that fraction or layer.
In 1972, I indentified 18 assumptions inherent in subcellular fractionation, and by 2008, it had risen to 22.2 These included, for example, that homogenisation and centrifugation do not generate energy; that sucrose does not affect enzymes; that the same quantity of energy is dissipated in different parts of a centrifuge tube; that enzymes, co-factors, substrates and soluble substances do not diffuse to different sites during fractionation; etc. etc.
The overall assumption is that the whole procedure does not affect the chemistry of the organelles significantly, from their properties in situ and in vivo to those in the final fraction. Unfortunately, the Second Law of Thermodynamics dictates that any change in order (entropy) must change the free energy of the system. Every homogenisation, centrifugation, ‘washing’, dilution or addition of another reagent, changes the entropy. Every chemical reaction is driven by free energy. This means that the rates of reactions, their directions and their equilibria – that is, the whole of their chemistry — must be changed by the whole procedure. The degree of these chemical
changes can be found out only by measuring them. As far as I can find in the literature, experiments testing the effects of the procedure have never been published, despite my having pressed for this several times between 1972 and 20081-3. In 1970, the late Professor Sir Hans Krebs told me that, at that time, he was not aware of such experiments having been carried out or published.
5. Control experiments for subcellular fractionation have not been carried out
The only control experiments, which are routinely carried out in connection with subcellular fractionation are ‘recovery’ experiments, in which the total chemical activity added together in all the fractions at the end of the fractionation is expressed as a percentage of the activity of the original crude homogenate. However, this final proportion can not indicate whether the activity measured in the individual fraction had or had not been decreased or increased, relative to that in the original homogenate, by any reagent or manoeuvre carried out during the fractionation. Secondly, the intention of the procedure is to find the original cellular location, for example, of an enzyme activity but one does not know whether the activity, its co-factors, activators and inhibitors, have moved away from, or towards, its original site in the cell during the fractionation. Finally, the reference of the recovery of activity to that in the crude homogenate, implies that the initial homogenisation does not alter the activity being studied – an assumption neither recognised nor measured, but made very unlikely by the Second Law.
The changes in free energy from the living cell to the subcellular fraction are likely to be very large, so that the fraction would be unlikely to have the same chemical activities, rates or equilibria, as those in the original states in their normal environments. However, it is possible that the effects would be so small as to make no significant difference to the results of the experiments. The only way to solve this problem, if biochemists are determined to continue to use subcellular fractionation, is to carry out control experiments. These would measure the effects of each or the reagents and manoeuvres on the activities, they are studying. Alternatively, they could test the effects of the overall procedure on the system that they were studying. I have listed several different kinds of control experiments which should and could be done2 (pages, 13, 430). I also asked biochemists to draw my attention to any publications, which showed that they had already been done, or if they would be prepared to put in writing the statement
that they were not necessary3. There was no response to either of these requests from the readers of the ‘Biochemist’, the house journal of the Biochemical Society.
Conclusion about subcellular fractionation
This procedure ignores the Second Law of Thermodynamics. Therefore, the conclusions arrived at as a consequence of using it, and their interpretations, should not be accepted until comprehensive experiments have been carried out.
B. Electron microscopy
6. Electron microscopy in the light of solid geometry
Most modern cell biologists consider that the electron microscope is the most accurate and modern instrument to study the anatomy of cells. Of course, it has a much greater magnification. The maximum resolution of the light microscope is about 200-250 nm in optimum conditions, whereas the electron microscope can go down to 5-10 nm. Unfortunately, a living cell cannot survive the electron bombardment, the high temperature, the low pressure and the x-irradiation inside the electron microscope, so that salt of a heavy metal must be deposited on dead cells, and that deposit is then examined. This assumes that: the living cell and the cell in the electron microscope do not differ significantly in shape, dimensions and structure; that none of the structures move during preparation for electron microscopy; that the procedure neither destroys organelles, nor does it precipitate solutes and suspensions from the cell. These assumptions are difficult to maintain.
As a result of the introduction of the electron microscope to cell biology, the following structural details were elucidated:
(i) The cell membranes, the nuclear membranes and the mitochondrial membranes were seen as double lines, and this ‘trilaminar’ appearance, dark-light-dark, was dubbed the ‘unit membrane’.
(ii) The existence of the Golgi body, which was first seen in 1894 in silver stained sections of owl neuronal cytoplasm, was regarded as being confirmed, when it was seen by electron microscopy.
(iii) The cytoplasm was also seen to be well furnished with membranous endoplasmic reticula, lysosomes, peroxisomes, and a ‘cytoskeleton’, composed of various filaments, tubules, and microtrabeculae.
(i) The cell membranes contained discontinuities, which were given the name, ‘ion channels’.
(ii) The nuclear membrane was punctuated by holes, given the name ‘nuclear pores’.
(iii) The myelin coverings of peripheral nerves were regarded as being composed of sheaths of Schwann cell membranes, wrapped around the axons. This was called the ‘Geren model’.
(iv) The synapses were seen as pre- and post- synaptic thickenings.
Simple geometrical considerations show that findings (i) – (vii) are in fact two-dimensional and could not exist in the living cells.
These will be examined seriatim, using the same headings.
(i) Consider simple geometry4, (page 37). (Figure 2 here) If one cuts through the equator of an orange, the skin will appear to have the minimum diameter. As the sections are made away from the equator, towards the poles, they should appear in a variety of thicknesses. The greater thicknesses will appear at the sections near the poles. The only way that the sections would appear to be of uniform thickness would be if all the sections were cut at the same distances from the equators or poles of the oranges (Figure 1). This is virtually impossible.
The mitochondria and Golgi bodies could not move around in the cytoplasm. The endoplasmic reticulum should appear to have four layers in electron micrographs. There should appear to be two layers around the mitochondrion, as well as the single layer intercalating as the cristae in the electron micrographs. There appear to be no fenestrations across the nuclear pores in this diagram. The nuclear membrane should appear as four layers in the electron micrographs. The cisternae lined on both sides by trilaminar membranes should be seen on most micrographs, not just diagrams. This section must have been cut precisely through the middle of the cell membrane, the mitochondrial membrane and the Golgi body.
Unfortunately, this apparently uniformity of thickness is seen under the electron microscope in the membranes around the cells, the nuclei, the mitochondria, the myelin lamellae, the synaptic thickenings and the thylakoids of the chloroplasts.
7. The ‘unit’ membrane is impossible in three-dimensional geometry
The ‘unit’ membranes were found to be ‘trilaminar’, that is, to say, to have dark, light and dark laminae, yet they nearly always appear uniformly distance apart, and very rarely indeed, touching, overlapping, wider apart, or blurred. Indeed, the thickness of the ‘unit’ membranes is given in textbooks by one figure, usually between 7nm and 10nm. No account is taken of the fact that it is known that dehydration for electron microscopy causes considerable shrinkage, particularly of those tissues which, in life, contain much water.
The obvious explanation for the uniform thickness of the ‘unit membranes’ is that they are two-dimensional in the electron microscope. The images seen must have been created after the sections have been cut, that is, they are artifacts.
Another explanation which should be borne in the mind is that a line, which has position but no thickness, is a geometrical abstraction. A real layer or membrane has two surfaces. All stains for electron microscopy are deposits of heavy metals, like paints. They land on both sides of a membrane, but do not replace it, so that when a layer is magnified sufficiently, the paint always appears on both sides of it, that is, as two lines – perhaps a ‘unit membrane’. You cannot clap with one hand.
Even when a thin sheet of material, is cut in many different orientations, the thicknesses of the edges should still be seen to vary widely.(Figure 2) The only explanations, which have ever been offered are, either, that the sections reorientate at right angles to the plane of section after they are cut; this seems extraordinarily unlikely; or that electron microscopists choose their clearest images – that is, ones at right angles to the plane of section, – to illustrate their publications. In fact, I have spent a great deal of time looking down the electron microscope, examining the literature, and requesting my colleagues, to show me examples of ‘unit membranes’ not at right angles to the plane of section. I have looked at many electron micrographs not specifically illustrating ‘unit membranes’, but, say, mitochondrial cristae, neurons, neuroglia or granule cells of the cerebellum. It is true that one occasionally sees a single, or blurred ‘unit’ membrane, but not one illustrating the full range of orientations;
8. Structures in the cytoplasm
(ii) the reality of the Golgi body as seen by light microscopy in most stained cells, was claimed to have been confirmed by its observation under the electron microscope. It appeared to Golgi as a cytoplasmic network, and for a long time, any particle seen in the cytoplasm, was described as a Golgi body. It was seen by bright field and dark ground illumination, but its shape varied, from a network throughout the cytoplasm, to a paranuclear body of approximately the same diameter as the nucleus, to a shell of granules surrounding the nucleus4 (page 63). Various unprovable ‘functions’ were ascribed to it. The electron microscopists saw it as a layered spherical body, like a section of an onion. However, light microscopists had found it to be a much bigger structure, with a much larger variety of microanatomy, that one must conclude that they could simply not have been looking at the same objects, as that observed by the electron microscopists;
(iii) by electron microscopy, various apparent structures are seen in the cytoplasm, in addition to the mitochondria and unidentifiable bodies seen in life. It should be noted that in living protozoa, unfixed cells, plants and cells in tissue culture, the following intracellular movements can be seen by light microscopy, with magnifications of x200 to x400. They include diffusion, Brownian movement, streaming, convection and nuclear rotation. Freezing may slow or stop movements, but movements are the main criteria,
whereby tissue culturalists decide whether cells are alive. The endoplasmic reticulum and the cytoskeleton – when they are regarded as different structures – are usually so fine that they require high power electron microscopy to view them.
The reasons for which these cytoplasmic structures are artifacts, may be summarised:
(a) intracellular movements could not occur, if a fine network permeated the cytoplasm. Furthermore, in life, the viscosity of cytoplasm has been found to be low, when measured by injecting particles, magnetism, centrifuge microscopy and electron spin resonance4 (pages 57). Electron microscopists have suggested that the movements could occur if the particles secreted lytic enzymes, which would dissolve the reticula, filaments and microtrabeculae, and all the latter could reform afterwards in real time. However, although it is conceivable that mitochondria could secrete such enzymes, it is extremely unlikely that iron filings, ground glass, gamboge or even carbon particles, would be furnished with such enzymes. Also, no broken ends are seen;
(b) the endoplasmic reticulum, the cytoskeleton and all the microtubules, microfilaments and microtrabeculae, nearly always appear in the plane of the sections. They are extremely rarely seen in transverse or oblique section;
(c) the latter structures are not seen to vary in diameters in longitudinal sections, as solid geometry would require;
(d) many of the elements of the cytoskeleton are detected by fluorescence light microscopy, in which the elements are treated as antigens, which are bound to fluorochromes. Very little testing of the ‘specificity’ of the antibodies has appeared in the literature. There is also much talk of ‘non-specific’ antibodies;
(e) tissue is usually fixed when it is made to fluoresce, but there should be spaces visible where the mitochondria, lysosomes and peroxisomes are located, but do not stain. There are not;
(iv) the cell membrane is believed to be pitted by ion channels, and the nuclear membrane by nuclear pores or ‘pore’ apparatuses. How can the membranes regulate ions, aminoacids, metabolic intermediaries and DNA, if they leak? How can they maintain membrane potentials, if they are short circuited when the channels and pores are open? Unfortunately, only one ion channel been seen by electron microscopy?5
Why are the ion channels and nuclear pores only seen in side view and face-on view, but never obliquely. Why, when the channels and pores open, do they not allow all smaller molecules to pass through? Why have the pores and the channels not been seen in living cells by confocal, differential interference contrast, tandem scanning or quantum dot microscopy, which are said to have the resolution to do so? Why do the nuclear pores always appear as circles or hexagons all over the nucleus? They should appear as ovals, ellipses and slits, depending upon the tangent to the surface of the nuclear membrane;
(v) let us imagine a myelinated nerve fibre as a roll of fine silk, the silk being the myelin lamellae, and the core being a plastic cylinder representing the axon. The roll is laid on a table. Two conditions may be considered. One is a perfect equatorial section along the length of the roll. It would split the core through the middle, and equally spaced layers of material would be seen on either side – that is to say, one would expect the lamellae to appear equally distant apart. The second possibility is that the longitudinal section has been cut three quarters of the way between the maximum diameter of the roll and its pole, in which case the lamellae would appear splayed and the core (axon) would appear thin or would not be seen – that is to say, whenever one cannot see the axon, the lamellae should appear splayed. Neither of these expectations from geometry are realised (Figure 2);
If one cut a section at the equator, the lamellae would be equally spaced. If the section was anywhere else, the lamellae should show a variety of spacings. Of course, the same would apply if the sections were longitudinal, that is to say, along the same axis as the myelinated nerve fibre.
(vi) when the synapse is viewed by electron microcopy, the pre-synaptic thickening is believed to be located on one side of the synaptic cleft, and the post synaptic thickening on the other. The physiological synaptic activity is believed to be located at the anatomical synapse, but the latter suffers from certain geometrical disabilities;
(a) the thickenings appear nearly always on side view. It is very rare for them to be seen obliquely and they are never seen face on. By the same token, by electron microcopy, the synaptic cleft appears nearly always normal to the plane of section;
(b) pre-synaptic fibres attached to dendrites of other neurons are not seen, although the light microscope has the resolution to see them. Therefore, it behaves like an old-fashioned telephone which has not yet been connected to the telephone exchange;
(c) physiologists record excitatory and inhibitory depolarisations and hyperpolarisations in the vicinity of the anatomical synapse, but the micropipette cannot be demonstrated accurately and beyond doubt to be located at the particular synaptic knob. For the same reason, it is difficult to define the pre- and post- synaptic regions microscopically, and the actions of drugs ascribed to these locations by pharmacologists represent an unproven and possibly an unprovable hypothesis
C Mammalian brain.
9. Dissections of mammalian nerve cell bodies
From the 1840’s, Hannover in Copenhagen, Purkinje in Prague, and Deiters in Bonn, dissected single nerve cell bodies from the brain, a method which was re-discovered in 1959 in Göteborg by Hydén, who taught me.7 I was never quite certain whether Hydén had described the method, believing that he was the first to do so, ot whether he knew about the previous pioneers of this procedure, but had chosen not to acknowledge their work.
However, he made a number of observations on rabbit medullary neurons, mainly from the lateral vestibular nucleus of Deiters. Hydén and his collaborators measured
their oxygen uptake and enzyme activities using Lindeström-Lang’s microdriver technique. They observed the cell bodies by phase contrast microscopy. Each cell body was apparently surrounded by very white granular neuroglia. They took out a bit of this material, and called it a ‘neuroglial clump’. By comparing the properties of the cell body with the clump, they believed that they were comparing typical neuron and neuroglial properties from the same animal.8
Neurobiologists pointed out to Hydén that he did not know whether the neuron cell bodies were excitable after isolation, as they were in vivo, so he invited me to come to Göteborg to examine the problem. He appointed me an honorary docent in the Neurobiology Institution.
Anaesthetised rabbits were killed by exsanguination9. The medulla was removed, and it was cut fresh into sections, 1-2 mm thick. The sections were coloured with a very small quantity of methylene blue. The neuron cell bodies showed up within the neuroglia. The tissue was not fixed, dehydrated or otherwise stained. Stainless steel wires, 30, 50 or 70 µm in diameter were used under a dissecting microscope, to pick out individual cell bodies. These were shaken into drops of 0.25 M sucrose or Krebs-Ringer bicarbonate-buffered solution in a parallel-walled chamber. They were then examined by phase contrast microscopy in an unfixed, hydrated, not sectioned, state. The advantages of this preparation are: the cells are unfixed and unstained; they take up oxygen; they exhibit enzyme activities; one can measure their DNA, RNA and nucleotides in individual neurons; their properties can be compared in different animals; cell bodies can be compared with neuroglial clumps. The disadvantages are that: the cell bodies are intermediate in properties between those in the intact animals, and those in histological and electron microscopic sections; the axons and dendrites are broken off; the cell bodies have been manipulated physically; the cell bodies collapse from spherical to discoid when they are placed on a slide.
During my two years in Göteborg, we demonstrated that the isolated cell bodies contain Mg-activated ATPase9, and also that they have resting membrane potentials of 50-70 mV10. These membrane potentials were diminished reversibly by low oxygen, low glucose, or high potassium ions.10 My intention was to find out, if the cell bodies retained their excitability, but unfortunately two years was not enough time to test this.
10. Structure of mammalian cell bodies
When I returned to the UK in 1964, I continued taking out medullary cell bodies and examining them by phase contrast microscopy. I had the good fortune to meet Mr Peter Sartory, who was a retired instrument maker. He had formerly been the President of the Quekett Microscopical Club, founded in 1865, and a Council Member of the Royal Microscopical Society. Although an amateur, Mr Sartory was a distinguished light microscopist. He had a large collection of excellent light microscopes,-some of them of great age- out of which he could extract the maximum information. Every Thursday, I would dissect out some nerve cell bodies at the University of Surrey in Guildford, put them without fixative into Krebs-Ringer solution in parallel walled microscope chambers, and take them over to Mr Sartory’s laboratory in his home in Lower Kingswood, Surrey. It was a beautiful drive along the A25, through Merrow, Gomshall, Abinger Hammer, Dorking and Reigate. We spent all day looking at neuron cell bodies, neuroglia, ganglion cells and myelinated axons usually by phase contrast microscopy. We examined cells from rabbits, rats, guinea pigs, mice, human beings, cows, etc. I had a number of collaborators in Sweden and in England: Professor Holger Hydén, Mrs. Anita Palm, Miss Inge Augustsson, Mrs. Jocelyn Fasham, Mr. Peter Sartory, Professor Karl Deutsch, Dr Tasawar Hussain, Miss Susan Stollery, Dr. Iffat Chughtai, Mrs. Dalveen Taylor, Mr. David Jarman, Mr. David Copestake and others. One interesting series of experiments we carried out was a result of Hydén’s finding that when a neuron cell body was pushed down to the bottom of a droplet of 0.25 M sucrose, or Krebs-Ringer solution, it adhered rapidly and firmly to the glass slide. A parallel walled chamber could be made by surrounding the liquid with a piece of double sided tape into which entry and exit canals had been cut out, and upon which a cover-slip could be pressed. This was much better than the usual cavity slide for phase contrast microscopy.
The cell bodies were in an unknown metabolic state, neither in situ, but not fixed. Perhaps they would look more similar in the living state, if they were placed in an incubating Krebs-Ringer oxygenated solution with glucose, rather than in 0.25 M sucrose, as Hydén had done.
11. The nucleolus and the nucleolar membrane
When we did this, the cell body became much more translucent. First of all, we noticed a worm-like wriggling structure in the nucleolus, which had been seen and named the ‘nucleolonema’11 by Estable and Sotelo in Uruguay in 1950.12 Probably because the nucleolus precipitates during staining for light or electron microscopy, the nucleolonema has disappeared from the modern literature, although sometimes a darkish spherical body is noted in electron micrographs. We saw the nucleolonema as a thread swimming around freely in the nucleolus, with deposits adding to and dissolving from it continuously in real time. We published drawings of it in ‘Microscopy’, but did not manage to photograph it. Later on, I put forward the hypothesis that this might be the structure to which chromatin retreated during the resting phase of the cells. It would seem to me that such high power light microscopic techniques, as differential interference contrast, confocal, tandem scanning and quantum dot, microscopy, might be used to examine the nucleolus in unfixed cells and in cells tissue culture again. Certainly, the question of where the DNA goes in resting cells needs to be revisited.
Another structure which we noticed in Krebs-Ringer or ‘199’ tissue culture media – but not in 0.25 M sucrose solution – was a nucleolar membrane.13 In the former media, we saw it in every neuron cell body from humans, rabbits, guinea-pigs and frogs. We illustrated it in publications in ‘Microscopy’, and in our ‘Atlas of the Cellular Structure of the Human Nervous System’.14 (pages 2, 3, 24, 126, 145).
‘Nature’ would not publish a micrograph of it, without giving any reasons for its refusal. Professor P.L. Williams, editor of ‘Gray’s Anatomy’, initially agreed to publish a micrograph in that august book, but later changed his mind, again without giving us any reason. Dr. John R. Baker of Oxford, probably the most distinguished light microscopist of the 20th century, at a meeting of the Physiological Society in Oxford in 1972, saw our slides demonstrating nucleolar membranes, and said that he was convinced. Many others did as well. Others believed that it was not a membrane, because we had not shown that it consisted of lipids and proteins. I answered that in my book, ‘Certainty and Uncertainty in Biological Techniques’1, I had indicated why I did not accept that a membrane had to have this chemistry. Others asked why the nucleolar membrane had not been seen before, but were not prepared to look down our microscopes. Others said
that our cells had been ‘damaged’ by being dissected out, and I asked them whether they thought that homogenisation ‘damaged’ cells. No, they did not think so.
Does it matter whether or not a membrane surrounds the nucleolus of every nerve cell or every cell? Does biological science want to know the answer to every question about microanatomy? Does it want its students to be taught accurate information?
12. Effects of staining on the appearance of cell bodies
We put a single cell body in a parallel walled chamber and viewed it at x700 magnification, with bright field. It was then fixed by a fixative introduced from one side, then 25%, 50%, 75% and 100% ethanol, from alternating sides, so that we could see when the reagent had arrived. We photographed the cell body at each step, and again after xylol and DPX mountant had been added. Thus, we had pictures of the same cell after each reagent. We then calculated the areas of the cells at each step as a proportion of their original unfixed areas15. The same procedure was used for following the area changes during: haematoxylin and eosin staining, the most popular histopathology stain; Palmgren’s nervous tissue stain; osmic acid staining, routinely used for electron microscopy. The cell areas shrunk to means of 19%, 20% and 14% of their original areas, respectively. Unfortunately, the cell bodies did not shrink to the same degree as the nuclei; cerebral slices shrunk overall differently again.
These experiments, together with earlier ones on marine eggs and red cells, plus problems with sections, strongly suggested that one cannot measure dimensions in stained sections, which are intended to be true for the original living cells
13. Cellular structure of the nervous system.
Having demonstrated the shrinkage and distortions which occurred as a result of staining, Mr David Jarman and I decided to look at the cellular components of the human nervous system.14 These consisted of the following elements; (a) fine granular material; (b) neuron cell bodies; (c) naked nuclei; (d) dendrites; (e) axons; (f) ependymal cells; (g) droplets; (h) myelinated nerve fibres; (i) non-myelinated nerve fibres; (j) Schwann cells; (k) Remak fibres; (l) brain sand; (m) fine unidentified fibres and fibrils which could be dendrites; (n) neuromuscular junctions; (o) retinal cells; (p) brain capillaries. Please note that I have not listed neuroglial cells, as opposed to fine granular material. This will be dealt with below. Most of the material came from post mortem brains, from patients who
had died from non-neurological and non-psychiatric diseases, but a few came from victims of road traffic accidents. Our human samples weighed about 5-10 grams, from brains weighing between 1.2 and 1.8 kg. In the late 1980’s, the question of having to seek the permission of patients or their relatives had not yet been raised. The fear that human bodies might have been infected with hiv-aids unknowingly had not yet emerged as a pathological, forensic and ethical, question. Nowadays (2009), one would have to set up a laboratory, with much stricter standards of sterility, than was required of us, and so the observations will probably not be repeated for a long time. In addition to the human specimens, we occasionally used bovine, rat or rabbit retinas and peripheral nerves. Mr. E. S. Rosen, the Manchester ophthalmologist, kindly provided us with photographs of normal human fundi. Our ‘Atlas’14 consisted of the following: unteased small pieces of brain, spinal cord, retina, autonomic ganglia, cranial nerves, peripheral nerves; slightly teased preparations of the latter; fully teased small pieces of the latter to show the individual elements (a) to (p). There was little interest among neurobiologists in our findings. They believed that better information about the living nervous tissue could be obtained from histological or electron microscopic studies of tissues, which had been subjected to: fixation; dehydration; homogenisation; deep freezing; extraction; replacement by heavy metal salts, etc. etc
14. Fine granular material and the nature of the neuroglia
Anyone carrying out micro dissection of the mammalian central nervous system cannot fail to be impressed by the ubiquity of the intensely white neuroglial ‘clumps’16 which fill out all the space between neurons, except the small volume occupied by blood vessels. We took out many neuroglial clumps and made sure that they contained no neuron cell bodies, that is, no blue coloured structures. If we found one, we shook it out, so that we had pure neuroglial material. We teased it, and found that it was full of mitochondria. When one looks at any stained section of brain or spinal cord, one sees neurons, surrounded by their membranes and one sees microglial nuclei. Unfortunately, one does not see membranes around microglial cytoplasm, simply because they do not exist. We showed this in another way. We stained sandwiches of brain in liver or in kidney, with classical neuroglial stains. Whereas these stains showed membranes around liver and kidney cells, we still could see no membranes around the cytoplasm
surrounding the neuroglial nuclei. We would earnestly request neurobiologists to repeat these observations that there are no membranes around the glial cells, and to stain liver and kidney sections with classical neuroglial stains, and see that the membranes of these cells show up clearly.17
Hodgkin and Lister (1827)18 were probably the first to look at brains under the microscope and they saw the white looking granular material. Hydén, Sartory, Pigon, Hertz, Hamberger, Palm, Deutsch, Hussain and I, among others, have all seen this material. We gave it the descriptive name ‘fine granular material’, and recognised that the granular appearance came from mitochondria19.Most other authors have considered that it is a mass of either astrocytes or oligodendrocytes, but they have not brought evidence about this.
The cells in the neuroglia do not have their own private membranes, therefore, the brain and spinal cord are syncytia, in which the neuroglial nuclei, or microglia, move around slowly between an undergrowth of dendrites, axons and mitochondria20.
In an extensive study of the literature, I compared the following characteristics of neurons, astrocytes, oligodendrocytes and microglia.20
(i) their structures and dimensions as described in unfixed cells by light microscopy;
(ii) their structures and dimensions as described histologically by light microscopy;
(iii) their structures and dimensions as described by electron microscopy;
(iv) the staining systems described as characterising them optimally or specifically;
(v) their properties in tissue culture in respect of criteria (i) – (iii);
(vi) the ‘markers’ found in them, mainly in tissue culture. It should be noted that if a marker’s presence is of genetic origin, it should either be present or absent – intermediate degrees of activities should not be found;
(vii) the ion channels which are found in neurons and neuroglia, usually in culture.
I made tables of these properties, and found that the descriptions in the literature of each of the neurons, astrocytes, and oligodendrocytes overlapped completely. In
addition, one cannot distinguish morphologically between a dendrite, and an astrocytic process, an unmyelinated fibre and a pre-synaptic fibre. Similarly, a microglial nucleus, a reactive astrocyte, a cerebellar granule cell, a Schwann cell nucleus, a ganglion cell and a pericyte, all look like spheres.14
When one looked at textbooks and original descriptions by the German and Spanish histologists, many of them say, for example, “If you stain, you show up this kind of cell, but if you ‘overstain,’ you may show up other types.” “One cannot always distinguish histologically between neurons and astrocytes.” “One cannot always distinguish between oligodendrocytes and microglia.”
In the classical histological textbooks, no claim is made that the neuronal, astrocytic, oligodendrocytic and microglial stains are specific. Instead, chapters are headed with these titles with no claims that they are ‘specific’. Indeed, I drew up a table showing how different authors assert that the ‘classical’ stains show up different named cells, so that the specificity of the stain depends upon which author one is quoting. “Ah,” said the histologists, “nowadays we use markers rather than histological stains,” but they do not indicate specificity with markers either.
Histological sections show classical astrocytes and oligodendrocytes, and diagrams of neurons and the latter show what are believed to be their shapes. It is extremely difficult to find sections in which the morphology of neurons, astrocytes and oligodendrocytes appear different beyond reasonable doubt. In the well known textbook, Peters, Palay and Webster, the ‘Fine Structure of the Nervous System’, they show a beautiful drawing to illustrate neurons, astrocytes and oligodendrocytes,21 but they do not seem to have recognised that any of these shapes could be completely interchangeable, if they were sectioned different orientations. I would urge any neurohistologist to ask themselves whether this assertion is generally true, whenever they look at any stained section of the mammalian brain or spinal tissue.
Pathologists and tissue culturalists very often use the word ‘glia’ or ‘neuroglia’. This is a tacit admission that they cannot distinguish between astrocytes, oligodendrocytes and microglia. Some modern neurobiologists deny that microglia exist.
Between 1962 and 1995, my colleagues and I have dissected out about 15,000 nerve cell bodies from unfixed rabbit, human, cow, rat, guinea pig, mouse, and cat brains.
I, personally have only seen one cell body, which looked remotely like a neuroglial cell. We have shown a micrograph of this cell, unfixed 14 (figure 209, page 172). This is the only neuroglial cell, I have ever seen. Perhaps someone else can show me published references to unfixed neuroglial cells, not in tissue culture, in tumours or in ‘enriched’ fractions. If such photographs exist, I would urge anyone who has them, to publish them and, perhaps, be kind enough to send me micrographs of them. I undertake to acknowledge receipt in public as well as in private. Colleagues have been promising to send me references or publications showing the neuroglial cells since 1962.
Let us clear our minds of any previous beliefs about the cellular structure of the mammalian central nervous system, and look down our microscopes at any stained section. One sees the following elements only: neuron cell bodies; processes, presumably axons and dendrites; a mass of fine granular material; nuclei, often described as neuroglial nuclei, the fine granular material is not transversed by cell membranes. I invite neurobiologists to show me evidence that they do not see these structures or see additional ones.
15. The likely cellular structure of the central nervous system.
As a result of the above considerations, I have concluded that the structure of the mammalian nervous system is as follows:
1. The major component is a fine, white, granular material, which contains a very high concentration of mitochondria. Many processes, presumably axons and dendrites, permeate it. This material is the neuroglia.
2. The cellular elements within the neuroglia are neurons and ‘naked’ nuclei. The neurons in life are near spherical and have less than twenty dendrites each. The fine granular material is a syncytium not enclosed by cellular membranes, so that the nuclei can move relatively freely.
3. Intracellular movement occur in the cytoplasm and in the axoplasm.
4. There are no anatomical synapses on the neuronal membrane (this will be dealt with below).
The cellular elements are the neurons which are all the cells, currently seen as having processes, that is, neurons, astrocytes and oligodendrocytes. All nuclei
1. without processes, usually called neuroglial nuclei, satellite cells, microglia and reactive astrocytes are ‘naked’ nuclei, and are mobile.
2. The central nervous system in life is fluid. It is held together by the viscosity, the dendrites, the ependyma, the blood vessels and the meninges.
3. The myelin sheath contains a liquid in life, and the myelin lamellae seen by low angle diffraction and electron microscopy are artefacts of dehydration.
4. Signals pass from one part of the nervous system to another by conduction only, although substances called transmitters and inhibitors affect the conductivity and the excitability of nervous and muscular tissues2 (pages 309-320).
The suggested revision of the structure of the mammalian nervous system clarifies several points:
(a) the difficulty of finding astrocytes and oligodendrocytes in healthy brains and spinal cords;
(b) the uncertainty about ‘specific’ staining systems appropriate to each;
(c) the failure to show cell membranes around neuroglial cytoplasm;
(d) that gliosis occurs within hours. It would be extremely slow if the brain and spinal cord were solid with cells in membranes;
(e) that cerebral, medullary and anterior horn cells, stain with ‘neuroglial’ stains;
(f) the malignancy of gliomas, because the neuroglia is a syncytium in which nuclei can move. The more malignant a tumour, the more nuclei are seen;
(g) during penetration of the brain and spinal cord with intracellular pipettes, the tips are not recording resting membrane potentials during most of their progression into tissue.
The re-examination of the cellular structure of the nervous tissue is long overdue since the pioneer histology of Golgi, Ramon Y Cajal, Penfield, Glees, and the electron microscopy of Sjöstrand, Peters, Palay, Palade, Webster, Jones and others. In my opinion, more studies on living tissue are needed.
D. High-energy’ phosphates
16. The sensitivity of ‘high-energy’ phosphates to light and other modalities
In 1962, Cummins and Hydén22 devised a simple technique for measuring the rate of breakdown of adenosine triphosphate (ATP) in isolated rabbit medullary neuron cell bodies. It was highly sensitive, and Hydén and I modified it to show a much greater rate.9
While I was writing my Ph.D. at the Institute of Psychiatry between 1958 and 1962, I had shone visible light on an isolated retina in Krebs-Ringer solution, and had recorded an electric signal from its surface. Professor Mcllwain, my supervisor, was not very interested, so I did not continue these experiments. A short time later, I met Dr Adelbert Ames, the IIIrd, who had already published this finding.23
However, it seemed to me that it might be possible to measure a change in ATP in a whole retina, using a technique, which could measure it in isolated neuron cell bodies. So I dissected out a whole retina from a dead rabbit, and incubated it in the dark in Krebs Ringer solution. I then measured the quantity of phosphate released from it, using radioactive ATP24.I then shone visible light on it for as few seconds, and the phosphate released increased significantly.
Retina is an outgrowth of the brain, so I wondered whether more phosphate would be released from a brain slice, if it was illuminated with visible light. It was. The same phenomenon occurred when the sciatic nerve was illuminated. It seemed quite logical that light should lower the ATP in retina, but I could not understand why one should be sitting on a nerve, which would never be exposed to daylight. So one day, early in 1963, I decided to do the crucial control experiment. That was to expose 2 mm ATP solution in oxygenated Krebs Ringer bicarbonate solution at 37°C to visible light for a few seconds without the presence of any biological tissue. At first to my disappointment, and then to my surprise, I found that visible light from a desk top lamp increased the phosphate released from ATP. That seemed very strange but I repeated the experiment several times.
One day, while I was doing these experiments, my assistant, Mrs Anita Palm, was indisposed, and I made up the experimental solutions myself. As I took the bottle of ATP out of the deep-freeze, I noticed, that it was in a black bottle, and I realised that the
suppliers, Sigma, must have known that it was sensitive to light. So I wrote to Dr. Berger, their chief chemist, in California, and asked him why the ATP was transported in black bottles, and if he knew of any publications about its light sensitivity. He wrote back to say that Sigma manufactured chromatographically pure ATP in the United States, and sent it on demand to Europe. However, some of the recipients complained that when it arrived, it contained ADP and AMP. Sigma had investigated this, and found that the ATP retained its purity, if it was put in black bottles. Dr Berger was not aware of any publications on the subject.
I decided to investigate the sensitivity of ATP to visible light more closely. I measured the intensity of the light at the level of the tubes containing the ATP, using a photometer. I obtained a spectrum of the light coming out of the lamp. I adjusted the distance of the lamp to give the desired light intensity, and chose those parts of the spectrum, which gave me high or low quantities of energy. I tested ATP at 37° and 23°, ADP and AMP at 37°, and ATP at 37°C with N2 bubbled through it and ferrous chloride added.
Since all these effects were tested using the highly sensitive radioactive method9 of Lowenstein (1960)24,modified by Cummins and Hydén (1982)22 I found that the effects were large enough to be measured by the less sensitive non-radioactive method of Berenblum and Chain (1938). The results were exactly the same.
The extraordinary result was that exposure to a few seconds of visible light caused an oscillation in the quantity of phosphate released. Now, since light and ATP were the only source of energy in the system, it seemed rather surprising that phosphate was released, but even more surprising that it should apparently go back to the ADP spontaneously. I was forced to postulate that there were two forms of ATP, the usual stable ATP, and a short lasting less stable compound, ATP.* The inorganic phosphate was released by the phospho-molybdate reagent of the Berenblum and Chain reagent, used to measure it.25
However, the values of phosphate seemed to show a large variation. I showed that this was not simply a wide spread of results, by using the following procedure. Firstly, the sample of 2 mM ATP was made up fresh every day, and 24 aliquots were prepared from the same sample. Then they were taken at 370C into a dark room, which
had been sealed to keep out all light. In this room, Mrs Anita Palm, Miss Inge Augustsson and myself waited for 20 minutes, and then two or four aliquots were extracted as controls. The light was turned on and off briefly, and at a signal, they extracted a sample each at the same time, at intervals of every 10 seconds or every minute. Thus, we had two simultaneous values at each time from two different individuals. The samples were coded and analysed randomly. When we calculated them, we noticed oscillations, in which the variation of each value could be known at each instant. Furthermore, when exactly the same experiment was repeated, the oscillations could be superimposed on each other. One other precaution, which had to be taken, was to wash the glassware with chromic acid, not detergents, because many of the latter contained phosphate.
Similar oscillations had been seen by Chance in the United States26,Shnoll in the Soviet Union27 and Hommes in Holland28 in tissue. I wrote to ask them, if they had ever looked for oscillations in the absence of tissue. None of them answered, so I circulated the question through the Information Exchange Group No. 1, Scientific Memo, No 298, 14th of January, 1965, but answer came there none.
Now ATP is involved in many metabolic reactions. Elsewhere, I have given an account of further experiments, in which I showed that a 2 mM solution was sensitive to different intensities and wave lengths of visible light29.In a further series of experiments, I found it to be sensitive to audible sound, brief centrifugation, electric fields, and different physiological concentrations of sodium and potassium ions. The properties were shown at 37°C, but not 23°C; not if the solution was bubbled with nitrogen, and ferrous chloride was added. ADP and AMP did not show it at 370C; neither Na2HPO4, nor NaH2PO4 showed it; it was not present if light was not shone; it appeared most intensely in those parts of the spectrum containing much energy.
Now creatine phosphate was believed to ‘store’ energy in its phosphate group, and arginine phosphate was believed to be involved in energy metabolism in invertebrates. So I tested all these modalities on ATP and creatine phosphate at 37°C, but arginine phosphate at 23°C. So the experiments covered five modalities and three compounds. Although I had carried out 70 experiments before I had achieved a low error
with the measurement of the phosphates, and 330 experiments on the three compounds; I had the greatest difficulty in obtaining publication.
My results appeared in the Information Exchange Group, No 1, Scientific Memo No 190, 3rd of July, 1964; distributed by the National Institute of Health. Of the 15 variables, only the effect of light on ATP solution was published in a journal29. A brief abstract appeared at the International Biochemistry Conference in New York in 1964, and I published an account of my other attempts at publication.30
However, resistance to publishing these findings was truly extraordinary. Reasons given included: no one else had published such findings before; I had not used chromatographically pure chemicals; the idea that physical agents could have chemical effects was revolutionary; oscillations were against the Second Law of Thermodynamics; my use of Berenblum and Chain’s 193825 method of measuring phosphate was wrong; (both Berenblum and Chain assured me personally that it was correct); colourless compounds could not be affected by visible light (I pointed out that hydrogen peroxide, scintillation fluids, ether, lactate and adrenaline are all colourless, but must be kept in dark bottles, because they are affected by visible light);I had not tested the effects of light on inorganic phosphate, nor of not shining light on ATP, creatine phosphate and arginine phosphate. (I had).
Since these three compounds have several biochemical effects, it would seem to me to be important for biochemists to know about their properties. If confirmed, the properties would be of wide interest: sensitivity to light, sound, centrifugation, and Na+/K+ concentration in the environment; compounds sensitive at 37°C, but not at room temperature; possible sensitivity of other organic phosphates to these modalities. It would be highly desirable to repeat these experiments, so that they would be confirmed or rejected by another laboratory. Dr Albert Amat of the Rovera i Virgili University in Reus, Spain, currently has a laboratory carrying out research in this area.
E. Dying, death and resuscitation
After I took my medical degree in 1956, I was appointed House Physician to the Professor of Medicine at Middlesex Hospital, which has since been amalgamated with University College Hospital. My House Surgeon appointment was to the Consultant in Neurosurgery at the Central Middlesex Hospital in Harlesden, London. During this time,
I did some primitive experiments on trying to raise the blood pressure of unconscious patients, who were dying and had blood pressures of about 40 mm Hg. I applied faradic stimulation to their chests, and raised their blood pressure by 15-20 mm Hg for about half an hour. These experiments were not well recorded and I did not attempt to publish them. I was aware of my lack of training in research methods.
For 2 years at evenings and weekends, I assisted my late mother as a general practitioner. At the time, the Government was anxious to encourage medical research, and Professor Sir George Brown, kindly helped me to obtain a place on the Physiology course at University College, London. I chose the subjects of neurophysiology and biophysics. I was sponsored by the Medical Research Council, with the expectation that I would do clinical research in my career.
About that time, my mother had a series of heart attacks, each of which left her in coma for a longer duration. This led me to look up the literature on myocardial infarction. The recent development in these subjects at the time, were mouth to mouth respiration, cardiac defibrillation and cardiac massage. I decided to devote part of my research endeavours to the subjects of dying, death and resuscitation. So I would work in two quite different areas, neurophysiology, mainly involving cell biology, and resuscitation of animals and patients.
17. Studies of dying and death
When I set up my own laboratory in Battersea, London, the first questions which occurred was, “What are dying and death?” There was remarkably little literature about it. So, I decided to observe patients dying at a local geriatric hospital. Whenever a patient looked as if death was imminent, the staff would ring me, and – if I was not lecturing – I would drive over to the patients’ bedsides. I would read their notes, look at their charts and observe them. I did not treat them or touch them in any way. They were all comatose. Nevertheless, I made observations about their colour, their mien, their breathing, sounds they made, whether they vomited, etc.
The following are the usual methods used at the time to kill animals for biochemical experiments: neck dislocation; concussion to the head; guillotining; ‘stunning’ plus exsanguination; poisoning with domestic gas, then containing carbon monoxide; re-breathing their own expired air. Whenever animals were killed, Dr. Harvey
Feldman and I would watch them, film them and examine them clinically. We then gave an account based on: observations on dying patients; observations and filming of dying animals using the above procedures; observations and treatment in accidents and emergency departments; eye witness accounts of executions mostly by the electric chair; post mortem execution reports from the United States; accounts of executions, mostly by executioners in the United Kingdom.
We could classify these stages of dying.31,32 Firstly, there is an ‘excitatory’ stage caused by hypoxia or pain. Secondly, the organisation of the body gradually diminishes until the nervous and endocrine systems lose control, and the dying becomes irreversible. At this stage, organs, such as kidneys, hearts and lungs may be excised and maybe transplanted to a compatible animal. Thirdly, each organ dies. Proteolysis occurs, gradients depending on oxygen fall, bacteraemia occurs.
One generalisation which has not been emphasised enough is that death is not instantaneous. It is a process. Even after guillotining or neck dislocation, the organism begins to die. The head becomes exsanguinated, the breathing stops and then the heart stops. Even after rapid decapitation, there is still some oxygen in the blood in the brain. This allows sensation from the face, the neck and the severance site, until the sensation is blocked by lack of blood supply and anoxia.
The fact that one can transplant corneas or kidneys from dead people means that the death of the person, which is a signal that he or she may be removed for post mortem or burial, does not mean to say that their organs are not still alive, and would not function normally in a recipient. Furthermore, a slice of kidney or brain may be excised, and will metabolise and respire linearly for 2 hours incubation. An enzyme extracted from a tissue by a highly energetic procedure, will fairly rapidly lose its activity unless frozen. So, there are different levels of dying and death; loss of function of individual organs and limbs when transplanted; proteolysis; loss of metabolism of tissue slices; bacteriolysis; loss of enzyme activity. One cannot be precise about the order of these phenomena.
When an animal’s neck is dislocated it goes into spasm. It is impossible to say whether that is due to pain in the neck or the foramen magnum, or due to excessive stimulation of the motor system. It is presumed that its subsequent relaxation is due to severance of the brain from the spinal cord caused by dislocation of the neck. There is
considerable haemorrhage in the region due mainly to tearing of the carotid arteries, but it should be emphasised that the complete and sudden separation of the brain is the beginning of the dying process, not the end of it.
Feldman and I31 had the strange finding in rats that if they were given artificial respiration immediately after their necks were dislocated, or if they were given barbiturates before, their hearts continued beating for a much longer time. This meant that in these conditions, the heart beat autonomously despite lack of central control, after cervical dislocation.
18. Recovery from hypothermic cardiac arrest
Originally we had intended to use this prolongation of the heart beat by some agents as a system to study what agents might prolong life or stimulate recovery of animals, but it was rather gruesome, and we looked round for other models in which we could study dying, death and resuscitation.
The first procedure which presented itself to us was that of Andjus and Smith.33 They put rats in flasks with bungs, and placed them in the refrigerator at 4°C. About an hour later, they found the rats deeply unconscious, and their hearts were not beating. The animals were left for up to a further hour. However, when they were warmed and given artificial respiration, they gradually recovered completely. They could even do intelligence tests.
We began to try out this procedure to examine the phenomenon of reversible cardiac arrest. We put a rat in a flask, with a bung on it, and place it in the refrigerator. Then we did what, perhaps, Andjus or Smith had not done. We opened the refrigerator door to observe the state of the rat. We were quite horrified to see it thrashing around in the flask in great distress. We decided that this procedure was too cruel. However, it occurred to us that if we anaesthetised the rats with intraperitoneal thropentone, it would not suffer the distress of the Andjus and Smith rats. That proved to be the case.
Our procedure was to anaesthetise the rats with 60 mg/kilo body weight of sodium thiopentone, and then when they were unconscious, place them supine in closed ice water dishes. Their paws were connected to oscilloscopes to record the electrocardiographs, and rubber tubes applied to their noses were connected to rat respirators. A rectal thermometer recorded their deep body temperatures. Their
breathing stopped after 20-30 minutes and their hearts stopped about 10 minutes later. Each rat was left in cardiac arrest for 30 minutes. It was then taken out of the ice water and dried with a towel. It was re-warmed with a desk top lamp. First, the heart re-starred, then it started breathing again. Gradually some rats recovered full consciousness and mobility and some died.34
When we first started doing these experiments, only 9% recovered from the cardiac arrest. So we set ourselves two tasks. Firstly, we wanted to increase the percentage of rats recovering – if that were possible. Secondly, we could study the rats at the following stages: anaesthetised: heart stopped; 30 minutes later; rat having regained respiration; rat not recovered, that is, dead.
Initially, we found that if rats were given assisted respiration during cooling, 70-80% recovered. These rats had much higher pO2s than those which were cooled down without assisted respiration, and this had more oxygen in their blood when their hearts stopped.35
However, this is not a useful model clinically for first-aid of human beings. If a climber had an injury on a Himalayan mountain in the presence of other people, they would not give artificial respiration, while the injured person was freezing. They would take the person immediately for emergency treatment. So we decided to look for other more practical measures.
19. Centripetal infusion and abdominal pumping
We found two measures which increased the incidence of recovery of the rats. The first was centripetal infusion intracarotid infusion of fluid towards the heart, and the second was abdominal pumping34.At this time, Dr Paul Rogers was my main collaborator.
Centripetal infusion was the injection down the carotid artery towards the heart. First, we used reduced blood which increased the recovery of the rats which has not been respired during cooling from 9% to 40%. Then we tried Krebs-Ringer bicarbonate buffered solution. Then we tried normal saline, which had the same effect. The volume of fluid injected was 2 ml, which was about 20% of the volume of a rat’s blood. The finding that normal saline had the same effect in reduced blood seemed to indicate that the effect of the infusion was to ‘wash out’ the blood in the coronary artery, which was in
fact, inhibiting, the heart beat. There may also have been a mechanical effect on the coronary arteries or the myocardium.
Examination of the literature showed that in the 1960’s and 1970’s, surgeons had used intra-arterial blood perfusions, when patients had lost large volumes of blood, for example, in traffic injuries, in gastric haemorrhage and in uterine haemorrhage, but the practice had been abandoned. The risk of infusion into the high pressure side of the circulation was that air emboli might be introduced accidentally into the coronary circulation and cause myocardial infarction. Therefore, we decided not to pursue the idea of centripetal infusion for resuscitation.
We found one other measure which increased the incidence of recovery of rats from 30 minutes hypothermic cardiac arrest from, 10% to 40 – 50% was what we called abdominal pumping. After drying the rats we compressed their abdomens with our thumbs 120 times per minute for 2 minutes. This produced a dramatic increase in recovery.35 Subsequently we showed that compressing the abdomen of a rat with no spontaneous heart beat had the following effects: it induced blood flow along the carotid artery; it forced blood backwards from the abdomen towards the heart. This closed the aortic valve and caused blood flow in the coronary arteries; it pushed the diaphragm cephalically, as did the jack knife procedure, used for children. It has been frequently reported that blood accumulates in the abdomen viscera, when the heart fails or patients are dying. The pressure we used was about 20 mm of Hg on the abdomen.
The idea of compressing the abdomen had been used successfully by: 19th century midwives to prevent blood loss; fighter pilots in ‘g’ suits to counter the effects of gravity during acceleration and deceleration; American surgeons transporting battlefield casualties in the Vietnam War, and, more recently, in the treatment of poor circulation in patients after myocardial infarction.
Since 1971, following the freezing to death of Scottish children in the Cairngorms, I suggested in the correspondence column of the ‘Lancet’36 that in a cold environment one could not know whether a patient was in cardiac arrest because of hypothermia or because she or he was dead. The wise course of action was to try to revive the patient., If they were dead, they would not recover. If they were in hypothermic cardiac arrest, they might. At the time, I suggested that they could be treated with abdominal pumping.
There was a difference of opinion as to whether they should be warmed up rapidly or slowly.
I took the latter view, not wanting to increase their metabolism while they were hypoxic. Others thought that it was not necessary to give such patients oxygen.
After we devised the hypothermic cardiac procedure, we discovered that two Americans, Niazi and Lewis had described such a procedure before (1954-1963). They had resuscitated rats, monkeys and dogs, using a similar method.37-39 We did not acknowledge their work initially, because we did not know about it until 1969.
We did a number of studies on rats in the different stages. We dropped them while anaesthetised into liquid nitrogen to study the Na+, K+, ATP, creatinine phosphate and lactate in their brains.40 We examined the oxygen saturation, the CO2, the pH, the glucose, the Na+, K+, Ca2+, phosphate, protein, creatinine and urea in their venous blood.34,40,41 We put rats before and after freezing into cages and studied their metabolism.42 These experiments were not easy to interpret, because it was difficult to know if the biochemical changes resulted from cold or cardiac arrest, or were the cause of them.
Initially, it appeared that the offspring of female rats, whose hearts had been stopped by cold were lighter, than the offspring of those who had not.42 However, this turned out not to be the case, when corrected for numbers of each sex in each group.
Of course, when biochemical changes were found in rats, whose hearts had been stopped by hypothermia, it was often difficult to tell whether the biochemical lesions were the causes of the tissue upsets or the results of them. All we could conclude is that lesions like high lactate and urea, which were found in rats which had recovered, could not themselves have been fatal.
We wondered whether abdominal pumping could be used as a first aid measure in patients, first, we applied to research bodies for funds to find out if our results on rats could be replicated on larger mammals, but the Research Councils and the University of Surrey refused our application, without giving any reasons. So we described a procedure which might be able to be used for new born lambs, many of which die from hypothermia in Highland farms. Mr David Jarman and I spent two weeks in 1987 during the lambing season at the Ministry of Agriculture, Food and Fisheries, Hartwood Farm near
Glasgow.43 We devised a 3 stage procedure whereby new born lambs whose hearts were not beating (i) had their throats sucked out using a Vitalograph mechanical aspirator, (ii) the lambs’ abdomens were rolled up by our fists, and (iii) they were given artificial respiration manually using tent pumps. We had some very encouraging results, but could not stay at Hartwood Farm to attempt to resuscitate sufficient lambs for significant results. However, we learnt from the farmers there that it would require a 24-hour shepherd to attempt to revive all the lambs with hypothermic cardiac arrest. This would cost the farmer more than the loss of relatively few lambs, so it would be uneconomical. However, we had been hoping to extend our observations to larger and more expensive sows, cows and horses.
In early 1970’s, I approached a large ambulance service to see if it might be interested in carrying out a clinical trial of abdominal pumping. The idea was that a control group of patients would have the classical treatment for cardiac arrest, namely cardiac massage and artificial respiration. The experimental group would be given cardiac massage and artificial respiration plus abdominal pumping. Subsequently, I described the procedure in a paper at a conference on emergency and Disaster medicine in Rome in 1983.44
About a quarter of all the patients picked up by the Ambulance Service of a large town, who were found in cardiac arrest over a period of about 2 months were allocated to a control and an experimental group. Their clinical data was recorded. Hard covered books or kidney shaped pieces of wood with handles were applied to the abdomen below the umbilicuses of all the patients in the experimental group, and they were compressed alternately to cardiac massage.
The advantages of abdominal pumping over cardiac massage seem to be as follows:
(i) cardiac massage is very hard work, and an operative has to be very athletic to be able to continue it for more than about 15-20 minutes. The ribs normally prevent external pressure on the heart, but it is much easier to compress the abdomen, which is at approximately atmospheric pressure in a person lying supine at rest. Many complications of cardiac massage have been reported in the literature;
(ii) whereas first aid training is required for effective cardiac massage, any member of the public without previous training can be invited to compress the abdomen. The only very rare complication is that the patient sometimes vomits, when he or she comes round from cardiac arrest. Only living patients vomit, and it can be treated by aspiration;
(iii) cardiac massage appears to bystanders to be violent, if carried out at the patient’s home, or in front of a crowd. Bystanders often think or say, “The patient is very ill or dying. Why do you have to treat him roughly?” Sometimes they hate the resuscitator for trying, especially when he or she often does not succeed. Abdominal pumping, on the other hand, is much more gentle than cardiac massage in its execution, and so patients and relatives seem less reluctant to wait until the patient is unconscious, before starting it. Often, one may start doing it while talking to the fully conscious patient;
(iv) my own experience is that when it works, clinical improvement can be seen within 10-15 minutes of beginning to compress the abdomen.
In the trial with the London Ambulance Service, we obtained some very encouraging results, but I noticed that they had come in rather irregularly, and we had a meeting with the trainers of the Ambulance Service to discuss this. I found out that the ambulance men had reported all the successful resuscitations, but not all those which had not succeeded. I could not publish such incomplete results, so no clinical trial of abdominal pumping has taken place in the United Kingdom.
In 1982, Ralston, Babbs and Niebauer described cardiopulmonary resuscitation with ‘interposed abdominal compression’ in dogs45. This was essentially similar to abdominal pumping, except that they used pressure of 100 mm of Hg, compared with our 20 mm. At the time, the group at Purdue University did not know about our work. At the first Purdue Conference on Interposed Abdominal Compression Cardiopulmonary Resuscitation, at West Lafayette in September, 1992, it was agreed that a multicentre trial should be held, although the emphasis seemed to be more on a clinical trial in hospital, rather than on first aid.
I have carried out abdominal pumping always with cardiac massage and artificial respiration on about 40 patients in Britain. The patients were in cardiac arrest, due to
shock, drowning, dehydration, myocardial infarction, hypothermia, neonatal asphyxia, fainting and Stokes-Adams syndrome. Unfortunately I did this treatment at road sides, in accident and emergency departments, and in patients’ homes, in circumstances in which I could not record the data satisfactorily.
I have not been able to persuade any accident and emergency department in a United Kingdom hospital to carry out a clinical trial on the procedure.
In the United States, although several centres have shown some interest in abdominal compression, but a full satisfactory trial has not yet been carried out there either. I heard that this was because of a legal hazard. If a doctor tries to revive a patient and does not succeed, she or he faces the risk that they will be accused of having contributed to the death of the patient, rather than using their professional skill to resuscitate him or her.
I have discussed the experiments on abdominal pumping in some detail, because I believe that the physiology behind the procedure is sound. I also think that the adoption of this procedure in hospital practice and in first aid could save many lives after it had been justified by a proper clinical trial.
20. The electric chair
The realisation that death from execution was not instantaneous led me to be consulted by the opponents of the electric chair in the United States. It was popularly thought that immediately the current was turned on, the prisoner lost consciousness, and consequently became insensate. This was believed, because the subject did not shout, wave their limbs, or indulge in violent movements, as might have been expected if they were in severe pain. It was not appreciated that the arms and legs of the prisoners are tied to the electric chair, and so cannot move. Similarly, prisoners who are being hanged, have their arms and legs bound together. Furthermore, the massive electric current applied causes first pain and then spasm of all the muscles, including the vocal cords, so that the prisoners cannot cry out. Thus the absence of the normal reaction to pain of an unrestrained conscious person is due to muscular paralysis, and does not mean that the subject does not experience pain.
I was approached by American lawyers defending prisoners who were expecting to be executed by the electric chair. They asked me if it was ‘cruel and unusual
punishment’ in the terms of the American Constitution. I replied that I did not know the American law, but, as a physiologist, I could say that it was painful and humiliating. Evidence that it is painful includes:
(a) electric currents are used to torture people, and are sometimes sufficient to cause unconsciousness;
(b) police officers who use ‘stun guns’ for restraint, test them on themselves, and report that they are painful;
(c) burns are painful before the heat destroys the nerve endings, and stimulates them in the region around the burns. Prisoners electrocuted are found to have third and fourth degree burns under the electrodes on their foreheads and below their knees;
(d) electrical stimulation of sensory nerves cause pain, including headaches and cramps. Athletes, carrying out excessive exercise, suffer cramps from muscle ischaemia;
(e) patients who have been revived from cardiac arrest with defibrillators suffer painful burns at the sites of the electrodes;
(f) paralysis of the respiratory muscles is extremely stressful, because the subject suffers ‘air hunger,’ as occurs in severe asthmatic attacks. They want to inspire but cannot;
(g) The inability to cry out is also very distressing.
The execution by the electric chair is very humiliating because:
(a) the subject frequently drools, vomits, defaecates and micturates uncontrollably;
(b) the subcutaneous tissue swells and may burst the skin;
(c) steam may arise from the ears, nose and mouth;
(d) the burnt tissue gives off an unpleasant smell;
(e) the muscles of the face contract to cause a ‘risus sardonicus;’
(f) subjects perspire heavily;
(g) sometimes, the first ‘jolt’ does not kill the subject, so another is administered after a few minutes.
These considerations are physiological and clinical, and do not bear on the morality of the electric chair, or of capital punishment, in general.
However, once one appreciates that the electric chair is painful, one must ask, “What about electric stunning of animals during slaughter for food?” About 800 million animals, mostly fowl, are slaughtered annually in Britain for food.
21. The slaughter of animals for food.
Animals are ‘stunned’ by application of electric current to their heads. They fall down and the blood vessels in their neck are cut to bleed them. They are often ‘stuck’, that is, their spinal cords are mashed up. They are then lifted up, usually by hooks in their hind legs. They twitch. When they are considered to have been exsanguinated completely, they are skinned and cut up. This process was shown on British television in 2008. It was carried out in hygienic circumstances.
The procedure may be examined in a little more detail. Firstly, what does ‘stunning’ mean? Both veterinary and popular dictionaries recognise two elements – the paralysis and the unconsciousness. There is no doubt that the animals are paralysed, since the muscles of the limbs are hypertonic. Unfortunately, there is no reason whatsoever to believe, as most members of the public and the slaughterers do, that animals do not feel. This belief is the same mistake as those who administer the electric chair make – namely, that the animals do not move or vocalise, and so they are not suffering pain. The animals feel pain for the same reasons as persons being tortured electrically, or executed by the electric chair. Wishful thinking is the only reason to believe that they are insensate. Nevertheless, often cows, sheep and pigs twitch, when they are lifted by their hind legs. The twitching looks purposeful in the sense that it appears to be attempting to push away, or to avoid, the hook. Slaughterers sometimes describe the animals’ movements as ‘reflex’, but, of course, the response to pain is reflex. ‘Sticking’ an animal is carried out in order to stop spinal reflex movements. Some animals begin to come round from the electric stunning but, at this time, there is no reason to believe that they do not feel the cutting of their necks, their being stuck, or their being ‘hooked’ up by their hind legs. They almost certainly lose sensation when they became unconscious, either due to excessive pain or to the fall in blood pressure, following exsanguinations.
22. Does the anatomical synapse exist?
In 1897, Sherrington conceived that nerves must connect with each other during reflexes, and he gave the name ‘synapses’ to the sites where he postulated that this occurred.46 At about the same time Held47 and Auerbach48 described small black stained granules on and around cell bodies as ‘end-fusse’. Gradually, the belief grew that these granules were the sites of the physiological synaptic activity, and this is believed today. Subsequently, extracellular electrodes and intracellular pipettes applied in the same region recorded what were called excitatory and inhibitory synaptic potentials in synapses and end-plates.
The intracellular pipette also recorded many uniform 0.5 mV potentials in the synapses, to which the name miniature ‘end-plate potentials’ was given. The nerve-nerve connections and the nerve-muscle connections were both named ‘junctions’; and, increasingly, findings from synapses and neuromuscular junctions were regarded as being similar to each other. Indeed, the terms ‘neuromuscular junctions’ and ‘motor plate’ have fallen into disuse in favour of ‘synapses’.
Synapses and neuromuscular junctions were seen for half a century by histological staining systems on neurons and on muscles, usually using silver stains. When the electron microscopists stained the regions, they described pre-synaptic and post-synaptic thickenings on either side of a cleft, and this was regarded as ultra microscopical evidence of the existence of anatomical sites at which the physiological events happened.
My further doubts occurred when I was looking at histological sections showing black granules on neuron cell bodies and dendrites, by Held (1897).47 Granules were also seen well away from these structures. If these granules looked like synapses, but were not, how does one know that the granules when present on the cell membrane were indeed synapses? One does not know, if the granules had moved from the original sites during histology.
My second doubt occurred when I examined carefully a large number of light micrographs of neurons and synapses in the literature. I could not find any illustrations of pre-synaptic fibres, except in diagrams. I showed that the microscope had the
resolution to see fibres of white matter of approximately the same dimensions, but I simply could not find fibres attached to the dendrites arising from other cell bodies. All pre-synaptic fibres seen are very short. This would be expected in a plane normal to the plane of section, but in the plane of section itself, one would expect a large number of pre-synaptic fibres to be spread out of the edge of the neuron cell body. Of course, the magnification of the electron microscope is too high to permit one to follow the whole route of a cell body to a dendrite, to a pre-synaptic fibre, to a synapse, to the next neuron cell body, but it still remains surprising that the dendrite connection to the pre-synaptic fibres are just not seen.
Even more extraordinary is that many fibres seen in the 6 layers of Brodmann in the cerebral cortex are seen to have dendrite ‘spines’ on the axons and dendrites. These spines are generally considered to be post-synaptic. Unfortunately no pre-synaptic fibres can be seen leading to them. This could be because they stain differently than the post-synaptic fibres, or because they do not stain – both of these seem to be very unlikely propositions – or the spines are not post-synaptic. It is always difficult to argue about structures which cannot be seen, even if one has a plausible explanation of why this is so. I would like to suggest that there are no pre-synaptic fibres attached to the spikes; they cannot even be seen by electron microscopy.
My next doubt arose when I searched in the literature for the dimensions of the synapses by light and electron microscopy. Nearly all the light micrographs showed them to have diameters of 1–5 µm, while the electron micrographs showed most of them to have diameters of about 0.5µm, or less. They hardly overlapped. Of course, most light micrographs had been stained with silver salts, and the electron micrographs with osmium salts, but the discrepancy is so large that it is unlikely to be explained by the different staining procedures. While the electron microscope might be expected to see more smaller synapses, for some reason, it does not appear to see those of diameters of 2 – 5 µm. Why not?
Another discrepancy occurs when one compares the number of synapses in mammalian neurons by light and electron microscopy. Usually, by light microscopy, they are in the hundreds, rarely in the thousands. By electron microscopy, one is talking of thousands to tens of thousands.49
When one looks at an unfixed neuron cell body, either dissected out or in tissue culture, the surface is rough, and this is attributed to the presence of synapses. Unfortunately, they are not attached to pre-synaptic fibres. If there were granules on the surface of membranes, they would cover it approximately evenly over the surface, but if the granularly were due to intracellular particles, i.e. mitochondria, they would appear with much less density, over the nucleus. The latter image is normally seen2, page 235, so the conclusion is that the granular appearance is due to mitochondria and not synapses.
When one looks at electron micrographs of synapses, one sees pre- and post- synaptic thickenings.50 The views are nearly always side on. They should appear face-on in approximately the same incidence, and, in oblique view, even more frequently. They are never seen face on either as an oval or as a circle, or obliquely. This anomaly is as unlikely as if one photographed a bag of coins thrown into the air, and they all appeared side on. It defies the laws of solid geometry. For the same reason, the synaptic cleft should be seen extremely rarely.
23. Critique of the chemical theory of transmission
All this evidence leads to the conclusion that the synapses as anatomical structures simply do not exist. The existence of the synapses is required for the chemical theory of transmission, which may be summarised as follows51:
(a) at rest, mobile vesicles, present in the synaptic knobs, are hitting the membranes occasionally;
(b) the vesicles contain transmitters or inhibitors, which are released through the pre-synaptic membrane into the synaptic cleft, and which produce miniature end plate potentials;
(c) when the synapse is excited by an electric current coming down the dendrite to the pre-synaptic region, it increases the incidence of vesicles releasing their packets of excitatory or inhibitory transmitter more frequently than when the synapse is at rest;
(d) the transmitter or inhibitor is broken down by enzymes in the cleft;
(e) some of the transmitter or inhibitor attaches itself to ‘specific’ receptors on the post-synaptic membrane;
(a) this induces an excitatory or inhibitory post-synaptic potential, which depolarises or hyperpolarises the synapse, and causes it, or inhibits it, firing an action potential into the next nerve or muscle cell.
Let us examine this hypothesis.52
(a) there are 5 distinct unproved and unprovable assumptions: that synaptic vesicles exist in life; that they each contain an equal ‘packet’ of transmitter or inhibitor; that at ‘rest’ a certain number of vesicles hit the inside of the pre-synaptic knob; that each collision causes firing of a miniature ‘end-plate’ potential; that an excited synapse is bombarded by more vesicles.
Synaptic vesicles are seen by the electron microscope in fixed stained tissue. They are too small to be seen by light microscopy. They appear too uniform in diameter. If they were, indeed, so uniform in diameter in life, when sections are cut, they should show a normal distribution of diameters. They do not. Their apparent uniformity in size led to the idea that they each contained a uniform ‘packet’ of transmitter or inhibitor.
The idea that at rest a small number of vesicles hit the pre-synaptic membrane and release the transmitter or inhibitor is impossible to prove, because the vesicles cannot be seen in life, and the collisions are believed to occur in the living synapse. The same is true for the release of miniature end plate potentials;
(b) this collision and its production of a miniature end-plate potential are pure speculation;
(c) again, the idea that excitation releases more vesicles, cannot be proved, because this phenomenon is also postulated to occur in life, while the vesicles can only be seen in micrographs of dead stained tissue;
(d) according to the histochemists53, the synaptic clefts contain the highest concentration of acetyl cholinesterase, which breaks down the best known transmitter, acetylcholine. So the proposition of the chemical hypothesis is that the acetylcholine diffuses through the region of the greatest activity which breaks it down, and then attaches to a receptor on the post-synaptic membrane, where it acts. This seems rather bizarre, that a transmitter which would have been much attenuated, would then initiate a post-synaptic potential. Its bizarre nature does not disprove it, but weakens the hypothesis.
(a) that there are ‘specific’ receptors in the post-synaptic membrane and nowhere else in the synaptic cleft, and that the transmitter or inhibitor attaches itself to such a proposed receptor, are unprovable hypotheses. They arise from the further hypothesis that some drugs which affect the action at synapses do not prevent the release of transmitter, but ‘block’ the synapses, so they act post-synaptically, whereas if they diminish the release of transmitter or inhibitor, they are presumed to act pre-synaptically. Such localisations cannot be proved because the synapses are so small.
The chemical hypothesis of transmission, both in respect of neuromuscular junctions and synapses contains so many unprovable hypotheses that overall it must be doubted on Popperian grounds. The only way to confirm it would be to devise experiments to prove each and every assumption of the overall hypothesis, and then carry them out. That would justify the hypothesis in retrospect.
24. New hypothesis of transmission.
If the anatomical synapse is an artifact, another hypothesis should be put forward about how electrical signals pass from one part of the central nervous system to the other. I have suggested that the simplest possibility is that the signals are conducted 2 (pages 317-320). The membranes of the neurons have high impedances, and the neuroglia and extracellular fluid have low impedances, so the signal is conducted by the low impedance route to the adjacent high impedance dendrites and cell bodies. The resistance and impedance of the neuroglia and extracellular fluid is strongly influenced by substances such as transmitters, modulators, hormones, substrates and ions. These will alter the sensitivity of the neurons, and cause fits, epileptic attacks, myoclonus, and possibly, tetanus, migraine and other conditions.
Secondly, the likely non-existence of synapses as anatomical structures should lead to re-consideration of the use deposit stains in histology, not for pathology, but for examining the structure of cells. When one sees a granule of a rather insoluble salt, it could be either of the following: the insoluble salt precipitated by the particular local chemicals in the region or the latter having reacted with the tissue under these circumstances. There is no way of telling the difference. Instead of histology, more research should be carried out using the following kinds of light microscopy of fresh, unfixed, unstained tissues; phase contrast; anopteral; dark field; confocal; differential
interference contrast; polychromatic light illumination; quantum dot fluorescence; Rheinberg illumination. The following types of micromanipulation are available; micro dissection with wires or glass needles; supravital staining; microsurgery; microradiasion; micromanipulation; use of Cartesian diver; microinjection; microspectometry; microspectrophotometry; optical tweezers.
The idea that memory occurs at the synapses is nearly 50 years old. It was based on the phenomena of neuromuscular fatigue; post-tetanic potentiation; facilitation; inhibition and other finding, in which a second or more stimuli produce more or less effect than the first stimulus. These effects are analogous to memory, so the idea has gained wide currency that memory occurs at synapses. Unfortunately, however, there is, and cannot be any evidence for the original idea. Experimentally, it would be extremely difficult to: identify a particular synapse believed to be associated with a single memory at the time of the experiment ‘uncontaminated’ by other memories or psychological or biochemical events, and to carry out all the necessary control experiments.
One of the fundamental difficulties of studying the biochemistry of action, memory, learning, thinking, sleeping, dreaming, is that – as far as we know – these activities all occur in living animals. It is extremely unlikely that they occur in dead animals or in dead tissue, so the conditions of studying them would require that they leave a trace in death, which is sufficiently large and stable to survive agonal biochemical changes. Another condition would be that the changes resulting from any particular psychological event would have to be extremely small, because a human being experiences so many of them. For example, when I drove from Primrose Hill, London to the Institute of Psychiatry in Denmark Hill, which was several miles, and in the late 1950’s took about 40 minutes, I made hundreds of observations, decisions, actions, calculations, etc., in relation to observing traffic lights; watching moving pedestrians; deciding on routes; accelerating or decelerating; looking at my watch. If each single element involved in any of the above were to leave a trace, a day’s normal activity should produce at least measurable biochemical change, but one could not distinguish which of the single psychological events had contributed to the change detected.
G.Fraud and parafraud
25. What is parafraud?
Fraud in science is fairly well understood, and widely condemned. Its extent is not known, and, therefore, its influence on the corpus of belief is not known either. Many know examples of it. Nowadays, many biological and medical journals are setting up mechanisms to combat it. In my opinion, it is difficult to eliminate, because it would require far more attention and time to appraise manuscripts and applications than referees and editors are prepared to spend. Nevertheless, the profession of science opposes fraud, and is attempting to eliminate it.54-56
There are a number of practices, which are frowned upon, but are not fraudulent. These include multiple publication, for example, as a proceeding of a learned society, then as a full length paper, then as an abstract to a national or international conference, then at a specialised meeting, then as a communication in ‘Nature’, ‘Science’, ‘La Science’, or ‘Experientia’. One may publish in a national specialised journal in the language of the country, and then subsequently in an international journal in English. This may be justified on the grounds that one wishes to support one’s national journal, but wishes also to expose one’s finding to the international science community. Another practice is to publish pieces of a research project fragmentarily, to obtain the maximum number of publications out of it. This is explained by the desire of academics to accumulate a large numbers of publications in order to obtain promotion. This may be a current reality of life, but it’s not desirable for the body of science. A third undesirable practice is to permit drug or other commercial companies to have a veto on scientific publication of research, which they have financed. Although this irritates research workers many of them have to accept it to attract financial support for their work. Nevertheless, I do not accept that ‘pressure to publish’ in order to be promoted should be regarded as a proper motive. It implies that the ‘system’ is responsible for improper conduct, rather than the research worker her or himself. Although, of course, every worker has a right to be promoted, that should be by virtue of carrying out research work of good quality, not by magnifying its value adventitiously.
I have listed a large number of common practices as ‘parafraud’57. Parafraud may be defined as ‘practices’ which are improper, which mislead academia, and which prevent the honest appraisal of research or its publication.
(a) not carrying out control experiments which one has identified or to which one’s attention has been drawn;
(b) not publishing experiments or individual values, which do not support one’s hypothesis;
(c) claiming authorship of research to which one has not contributed significantly;
(d) not responding to proper questions about one’s published research or opinions;
(e) not reading manuscripts submitted for publication, or applications for funds for research, with sufficient attention to recommend their acceptance or rejection;
(f) not quoting and commenting on the findings of research workers, who have come to conclusions, which one does not like, or not quoting those whom one dislikes personally;
(g) not including sufficient details in publications, so that the experiments may be duplicated;
(h) not keeping raw data, so that they may be inspected at a later date;
(i) quoting publications which one has not read;
(j) quoting evidence, which one does not understand, in support of one’s hypothesis;
(k) quoting evidence, which one does not know how it was gathered, in support of one’s hypothesis;
(l) not distinguishing between hypothesis and findings;
(m) failing or refusing to identify assumptions in experiments;
(n) acknowledging limitations of experimental procedures, but not taking them into account, when calculating the results of experiments;
(o) resisting the presentation of the research of one’s intellectual antagonists at national and international meetings;
(p) admitting mistakes in one’s own past or published research work;
(a) misquoting publications, either deliberately, or because one has not checked them;
(b) personal rudeness towards research workers or their views, if one disagrees with them;
(c) resistance to the promotion of research workers, with whom one disagrees.
There are certainly other practices that I have not listed here. Most research workers to whom I have talked have seen examples of such practices. They are often not recognised as misdemeanours. Sometimes they are not admitted. In my opinion, we must strive to avoid all of these practices, because they distort knowledge and truth to an extent which is simply unknown. The identification of the practices of parafraud is the first step towards eliminating it.
Scientists and research workers believe that their endeavours are objective, but one can identify many areas in which a degree of subjectivity enters into their work.
(i) one’s personality, especially if one is a natural sceptic or a natural believer;
(ii) one’s attitude to one teachers;
(iii) the attitude of one’s teachers;
(iv) the age at which one was taught and accepted a particular piece of information;
(v) the access to the literature on a particular subject;
(vi) the weight and number of pieces of evidence which induces a particular research worker to accept a particular view, or to change their minds about it;
(vii) peer and political pressures;
(viii) the degree to which one is prepared to enter dialogue with others who disagree with oneself, and, or, be persuaded by them;
(ix) one’s personal like or dislike of the person discussing a view different from one’s own;
(x) the degree to which one understands the problem;
(xi) the style of discussion of others participating in the dialogue;
(xii) the degree of rationality and casuistry of all parties;
(xiii) the economic and career advantage of adopting a particular viewpoint;
(xiv) the knowledge and intelligence of the research worker.
One should consider that it is one’s duty to work towards eliminating all these subjective factors, but this can only be done successfully by recognising them. Science can only become objective when all these factors are recognised in ourselves, sufficiently for us not to practice them.
26. Truth in biological and medical research
I have defined truth in biology as the structures and the relationships of living tissues in whole intact animals, not affected significantly by the procedures which are used to examine them. The aim of clinical research is to understand the genesis of diseases, so that one can modify the biochemistry of the tissue, including the use of drugs, on the basis of such understanding, to prevent the evolution of the diseases and to accelerate recovery from them.
27. Suggested code of practice
To this end, I would like to suggest a code of conduct for all biological and medical research workers.
1. Study living intact human beings and animals, if possible.
2. Apply the minimum energy and disruption to tissues under study.
3. Carry out comprehensive calibrations of instruments and chemical procedures.
4. Identify all the assumptions inherent in the procedures, and test those which have not been previously examined.
5. Be aware of the natural laws of geometry, chemistry and thermodynamics.
6. Carry out all control experiments, unless they have already been published comprehensively.
7. Preserve raw data for later inspection.
8. Enter into dialogue with anyone interested, however apparently intellectually hostile.
9. Answer all questions relevant to one’s project.
10. Admit mistakes and publish corrections, if possible.
11. Deliberately cultivate a culture of intellectual honesty.
12. Do not indulge in casuistry.
13. Prefer evidence over the authority of the people with whom one is in dialogue.
1. Read all publications properly.
2. Define all terms used.
3. Prefer naturally occurring reagents in physiological concentrations to the same reagents at unnatural concentrations, and avoid the use of chemicals which would not be compatible with life or normal metabolism.
I have written this memoir to draw the attention of my colleagues to what I regard as very serious unresolved problems of research in cell biology and neurobiology. This is not a scientific paper, or a comprehensive review, but an honest attempt to draw colleagues’ attention to important anomalies in current views
I should say that my attempts to carry out good scientific research have cost me dearly. I was given a Medical Research Council Scholarship to study physiology at the beginning of my career, but have never received any public money since, other than my pay as a Reader in Physiology. As indicated, my projects have been on the value of procedures used in cell biology, on the cellular structure of the brain, on the properties of ‘high energy’ phosphates, on resuscitation of animals and human beings, on synapses. Despite the importance of these topics, or, perhaps, because of them, I have never received a penny of public funds, I have had difficulty in finding publishers for my books. Some of my manuscripts have been returned very rapidly or without reasons being given. I have denied promotion, and I was one of the few, if not only academic in Britain to have their tenure taken away. The reason given for this was that I had not attracted funding for my research, as about 60% of my colleagues at the time had also failed to do. Despite this, I continued to carry out my research work and teaching until I reached the age of 65. Nevertheless, I would like to put on record, that there is a price to pay for attempting to be intellectually honest.58
This list is mainly to my own references, in which I have given detailed evidence to support my views, especially where they did not accord with the current consensus in the subject. I have added these because academics, who disagreed with me, have alleged
that I have not produced the evidence for my conclusions. This is evidence that I have. It is not a comprehensive bibliography.
1. Hillman, H. (1972). Certainty and Uncertainty in Biochemical Techniques. Surrey University Press, Henley on Thames, pages 49-50.
2. Hillman, H. (2008). Evidence Based Cell Biology. Shaker Press, Maastricht, page 40.
3. Hillman, H. (1979). Control experiments for biochemical techniques. Letter. Biochem. Soc. Bull., 1, 3, 11
4. Hillman, H. and Sartory, P. (1980). The Living Cell. A Re-examination of its Fine Structure. Packard, Chichester, page 37.
5. Unwin, PNT and Zampighi, G. (1980). Structure of the junction between communicating cells. Nature, 283, 545-549.
6. Hydén, H. (1959). Quantitative assay of compounds in isolated fresh nerve cells and glial cells from control and stimulated animals. Nature, 184, 433-435.
7. Hillman, H. (1986). Hydén’s technique of isolating mammalian cerebral neurons by hand dissection. Microscopy, 35, 382-389.
8. Hydén, H. (1960). The neuron, in The Cell, edited by Brachet, J. and Mirsky, A.E. Vol. 4. Academic Press, New York, pages 215-323.
9. Hillman, H. and Hydén, H. (1965). Characteristics of ATPase activity of isolated neurons of rabbit. Histochemie, 4, 446-450.
10. Hillman, H. and Hydén, H. (1965). Membrane potentials in isolated neurons in vitro from Deiters’ nucleus of rabbit. J. Physiol., 177, 398-410.
11. Sartory, P., Fasham, J. and Hillman, H. (1971). Micrographical observations on the nucleoli in unfixed rabbit Deiters’ neurons. Microscopy, 32, 93-100.
12. Estable, C. and Sotelo, J.R. (1951). Una nueva struttura; el nucleolonema. Publ. Inst. Invest. Sci. Biol. 1, 105-123.
13. Hussain, T., Hillman, H. and Sartory, P. (1974). A nucleolar membrane in neurons. Microscopy, 32, 348-352.
14. Hillman, H. and Jarman, D. (1991). Atlas of the Cellular Structure of the Human Nervous System. Academic Press, London.
15. Chughtai, H., Hillman, H. and Jarman, D. (1987). The effect of haematoxylin and eosin, Palmgren’s and osmic acid procedures on the dimensions and appearances of isolated rabbit medullary neurons. Microscopy, 35, 625-629.
16. Hydén, H. and Pigon, A. (1960). A cytophysiological study of the functional relation between oligodendroglial cells and cells of Deiters’ nucleus. J. Neurochem., 6, 57-72.
17. Hillman, H. Deutsch, K. Allen, J., and Sartory, P. (1977). The histology of neuroglial clumps in rabbit brain. Microscopy, 33, 175-180.
18. Hodgkin, T. and Lister, J.J. (1827). Notes on some microscopic observations of the blood and animal tissues. Phil. Mag.,ii, 130-138.
19. Hillman, H. and Deutsch, K. (1979). Staining of isolated rabbit neurons and neuroglial clumps. Experientia, 35,771-772
20. Hillman, H. (1986). The Cellular Structures of the Mammalian Nervous System. MTP Press, Lancaster, pages 52-91.
21. Peters, A., Palay, S.L. and Webster, H.de F. (1998). The Fine Structure of the Nervous System, Oxford University Press. New York, Frontispiece.
22. Cummins, J. E. and Hydén, H. (1962). Adenosine triphosphate levels and adenosine triphosphatase in neurons, glia, and neuronal membranes of the vestibular nucleus. Biochem. Biophys. Acta, 60, 271 – 283.
23. Ames, 111rd A. and Gurian, B.S. (1960). Measurement of function in an in vitro preparation of mammalian central nervous system. J. Neurophysiol., 23, 676 – 691.
24. Lowenstein, T. (1960). Chemical synthesis of adenosine di– and tri-phosphates containing P.32 Biochem Preparations, 7, 5-22.
25. Berenblum, I. and Chain, E. (1938). Studies on the colorimetric determination of phosphates. Biochem. J., 32, 286 – 288, 298 – 302.
26. Chance, B. (1952). Spectra and reaction kinetics of respiratory pigments of homogenised and intact cells. Nature, 169, 215-221.
27. Shnoll, S.E. (1958). Spontaneous synchronous transitions of actomyosin molecules in solution from one state to another. Voprosi Medicinski Chimii, 4, 443 – 454.
28. Hommes, F.A. (1965). Oscillatory reductions of pyridine nucleotides during aerobic glycolysis in yeasts. Comp. Biochem. Physiol., 14, 231 – 238.
29. Hillman, H. (1966). The effect of visible light on ATP solution. Life Sciences, 5, 589 – 605.
30. Hillman, H. (1991). The Case for New Paradigms in Cell Biology and in Neurobiology. Mellen Press, Lewiston, pages 1 – 5.
31. Feldman, H. and Hillman H. (1969). A clinical description of death in rats and the effect of various conditions on the time until cessation of ventricular contraction, following section between the brain and the spinal chord. Brit. J. of Exp. Path., 50, 158-162.
32. Hillman, H. (1970). The biology of dying and death. Arch. Thanatol., 2,194-198.
33. Andjus, R.K and Smith, A.U. (1955). Reanimation of adult rats from body temperatures of 00 C to +20C. J .Physiol., 128, 446-472.
34. Rogers, P. and Hillman, H. (1970). Increased recovery of rats’ respiration, following profound hypothermia. J. Appl. Physiol., 29, 58-63.
35. Rogers, P. and Hillman, H. (1970). Increased recovery of anaesthetised hypothermic rats induced by intracarotid infusion. Nature, 228, 1314.
36. Hillman, H. (1971) Treatment after exposure to cold. Letter. Lancet, 4th of December, page 1257.
37. Niazi F.A. and Lewis, F.J. (1954). Tolerance of rats to profound hypothermia and simultaneous cardiac standstill. Surgery, 36, 25-32.
38. Niazi F.A. and Lewis, F.J. (1957). Profound hypothermia in the monkey with recovery after long periods of cardiac standstill. J. Appl. Physiol.,10, 137-138.
39. Niazi, F.A. and Lewis, F.J. (1965). Profound hypothermia in the dog. Surg., Gynec., Obstet.102, 98-106.
40. Leonard, C. and Hillman, H. (1972).The degree of recovery and the biochemical exchanges in the brains of rats during cooling and recovery from hypothermic cardiac arrest and recovery of rats. Resuscitation, 1, 335-346.
41. Rogers, P. and Hillman, H. (1972).Biochemical changes in the blood during hypothermic cardiac arrest and recovery of rats. Resuscitation, 1, 25-29.
42. Hillman, H, Loupekine, J. and Fullbrook, P. (1972).The clinical history of cardiac arrest and recovery of anaesthetised hypothermic rats and their reproduction. Resuscitation, 1, 51-60.
43. Hillman, H, Jarman, D. and Brittain, J. (1986). Simple system cuts lambs deaths. Farmers Weekly, 105, 39.
44. Hillman, H. (1984). Abdominal pumping – a promising new technique in: Manni, C. Magalini, S. L. (editors). Emergency and Disaster Medicine, Springer, Berlin, pages 404–410.
45. Ralston, S. H., Babbs, C.F. and Niebauer, M.J. (1982). Cardiopulmonary resuscitation with intermittent abdominal compression in dogs. Anesth., Analg., 61, 645 651.
46. Sherrington, G. S. (1897). In a Textbook of Physiology, 7th edition, edited by Foster, M. and Sherrington, G. S. Vol. 3 Macmillan, London.
47. Held, H. (1897). Beitrage zur Struktur der Nervenzellen und ihrer Fortsatze. Arch Anat., Physiol. Abt., 21, 204 – 294.
48. Auerbach, L. (1898). Nervendingung in Centralorganen. Neurol. Zentralbl., 17, 445 – 454.
49. Hillman, H. (1985). The anatomical synapse in mammals by light and electron microscopy. Medical Hypotheses, 17, 1 – 32.
50. Grey, E. G. (1969). Electron microscopy of excitatory and inhibitory synapses; a brief review, in Akert, K. and Waser, R. (editors). Progress in Brain Research, Mechanism of Synapse Transmission, 31, 141 – 155.
51. Katz, B. (1969). The Release of Neural Transmitter Substances. Sherrington Lecture. Liverpool University Press, Liverpool.
52. Hillman, H. (1991). A re-examination of the vesicle hypothesis of transmission in relation to its applicability to the mammalian central nervous system. Physiol. Chem. Phys. and Med. NMR., 23, 177 – 198.
53. Silver, A. (1974). The Biology of Cholinesterases. North Holland, Amsterdam.
54. Kohn, A. (1992). False Prophets: Fraud and Error in Science and Medicine. Blackwell, Oxford.
55. Broad, W. and Wade, N. (1993). Betrayers of Truth, Fraud and Deceit in Science. Simon Schuster, New York.
56. Lock, S. and Wells, F. (1993). Fraud and Misconduct in Medical Research. British Medical Publishing, London.
57. Hillman, H. (1997). Parafraud in biology. Science and Engineering Ethics, 5, 121–136.
58. Hillman, H. (1996). “What price intellectual honesty?” asks a neurobiologist, in Confronting the Experts, edited by Martin, B. State University of New York Press, Albany, pages 99–130.