Resistance to Infection — The Experimental Approach
The Twenty-fifth Stephen Paget Memorial Lecture was delivered by Professor A. A. Miles. C.B.E., M.D., F.R.C.P., Director of the Lister Institute of Preventive Medicine, on Tuesday, 27th November, 1956, in the Physiology Lecture Theatre, University College, London, W.C.1. The President, The Right Hon. Viscount Waverley, G.C.B., G.C.S.E., G.C.I.E., was in the chair.
The Chairman said that his old friend, Professor Miles, had come to talk on a subject on which he was eminently qualified to speak. "Resistance to Infection: The Experimental Approach." He would call upon him to deliver his lecture.
Resistance to Infection — The Experimental Approach
By PROFESSOR A. A. MILES, C.B.E., M.D., F.R.C.P.
May I first say how sensible I am of the honour done to me by the Research Defence Society in asking me to deliver this Stephen Paget lecture? The subject I have chosen is a vast one, and obviously I can deal only with a little bit of it. And here I should perhaps apologize in advance to those who may have expected that little bit to be a discussion of new researches, because I propose mainly lo look at the science of immunology itself; and to do even that with a critic's eye.
The importance of understanding resistance to infection and of exploiting our understanding needs no emphasis. We are part of a world where all living organisms are woven into a complex web of dependence on each other—a dependence which includes the inter-relations of predatory animals with those they prey upon, and of parasitic organisms with the hosts they inhabit. The balance of nature is largely a question of which species are successful predators and parasites, and which manage to resist or tolerate them. The human species is no exception, and we maintain ourselves in a competitive world by reason, among other things, of what we can do to defend not only ourselves but our cattle and crops, against crippling or fatal infection by microbial parasites.
The ethics of animal experiment
The approach to the problems of resistance to infection with which I am particularly concerned is of course that entailing the use of experimental animals. It is fitting that Stephen Paget lecturers should deal, as they have on many occasions, with the triumphs of experimental analysis that have been of indubitable benefit to mankind. Such benefits are rightly considered to be one of the answers to those who hold that we have no right to do such things. But in dwelling exclusively on these triumphs the proponents of the experimental method may be misunderstood on two counts. First, we may be thought dishonest because we omit to discuss our failures: and second, our legitimate pride in the successes of the experimental approach may appear to be a justification of the means by the end; and to suggest by implication that either we have no moral code or have disregarded a moral code which under other circumstances we should recognize as binding.
The failure to solve an urgent problem by animal experiment is no more than the inevitable reflection of our technical and intellectual shortcomings—either we ask the wrong questions of the experimental system, or we choose a system that could not give us the answer we need. But completely sterile results are rare; most often the experiments which fail to answer the questions asked of them nevertheless give us new facts or reveal new phenomena. These discoveries may immediately prove to be relevant to some other problems. If they do not the experimenter is content to regard them as something whose significance is still in the womb of time. Our critics, on the other hand, see them merely as so many new counters for use in a rather academic game, which, since it cannot be played without animal experiment, is certainly misguided, and may well be disreputable. In what follows, I hope to establish the contemporary importance in researches on resistance to infection of much that seems irrelevant to our critics.
For a moment, I want to examine the second misunderstanding. We do, and we should, justify our use of animals by the results we get. However, if our morality is not to seem merely pragmatic, the primary reasons for using animals at all should be sufficient, irrespective of any outcome; and the reasons for refraining from their use should be secondary to the first principles. I have no time to marshal all the considerations bearing on this point, and must be content to submit that in a final analysis we are justified in using animals because, as the somewhat resolute biological phraseology of my opening remarks was intended to convey, we recognize a clear-cut hierarchy in the animal kingdom—a hierarchy in which, on theistic as well as biological grounds, we unhesitatingly put ourselves at the top; and because we conceive our first duly to be to man.
With these overriding principles established, we may formulate the secondary principle, the principle of discrimination in the use of animals. It arises from our recognition of a kinship with the larger and more highly organized forms of life. It is a kinship that leads us, among other things, to box the ears of a boy we find pulling the wings off a fly, and to practise our cattle breeding and butchery in as humane a manner as is consistent with economic efficiency. But among the many members of the animal kingdom, man awards the degree of kinship in so capricious and various a manner as to leave little ground for supposing that the respect we accord to the living animal is the expression of any underlying ethic. It belongs much more to the domain of aesthetics. This respect, nevertheless, like many things based on aesthetics, is a very real thing to most of us, and is of the highest importance in the formulation of a code of behaviour. We acknowledge it in any consideration of animal experiment not only by avoiding the brutal, but by doing our best to avoid the unnecessary; and our pleasure in demonstrating, by the rewarding result of a humanely-conceived experiment, that we have done both, in no way implies that we subscribe to a moral code we are prepared to sacrifice to expedience. Further discussion of this point is outside the scope of my lecture and would indeed be impertinent, since I have expressed no more than a personal opinion; though I hope it will be acceptable to many of my fellow research workers. I have been explicit about these fundamental principles lest I be thought to have missed the wood in a preoccupation with the trees: because I propose to deal in greater detail with other objections to animal experiment, less fundamental but equally important.
The chief objections, aimed particularly at justification by result, are not primarily ethical. They are made indeed upon legitimate technical grounds, and as such demand a careful technical answer. They are three in number.
The first objection is that animal experiments can reveal nothing that would not emerge from a proper study of natural disease. In other words, observation and anything short of animal experiment are enough for biological and medical science. The second is that the behaviour of animals of one species is not a reliable model of what happens in animals of another. In other words, the argument from an animal to, say, man is misleading because it is biologically invalid. The third I have already mentioned; that the bulk of experiments fail to produce useful solutions of our medical or social problems, and at best provide only facts for the furtherance of various scientific disciplines whose pursuit brings little immediate benefit to mankind. We have then to satisfy ourselves that our experiments are neither unnecessary, irrelevant nor unjustifiable, and I would ask you to bear these three adjectives in mind, as a ground base to my development of the theme of the experimental approach,
the analysis of resistance to infection before animal experiment
Perhaps the most cogent test of necessity would be to ask how far our knowledge of resistance to infection was advanced in the virtual absence of efficient animal experiment. That men and animals vary in their resistance to disease must be an observation of some antiquity. The most dull-witted survivor of an epidemic could hardly fail to realize that he differed from the unlucky ones who died. Those with acuter wits can do much to rationalize these differences. Galen, with the accumulated learning of several hundred years of Greek medicine at his disposal, builds an impressive and exceptionally tidy hypothesis to explain why at epidemic times some go sick and others remain healthy. The chief factor, he says, in the production of disease is the preparation of the body which will suffer it. Let us imagine, for instance, that the atmosphere is carrying divers seeds of pestilence, and that, of the bodies exposed to it. some are choked with excremenii-tious matters apt in themselves to putrefy, that others are void of excrement and pure. Let us further suppose an obstruction of orifices and resultant plethora in the former, likewise a life of luxury, much junketing, drinking, sexual excess and the crudities which must attend on such habits ; in the latter let us suppose cleanliness, freedom from excremcntitious matters, orifices unobstructed and uncompressed, desirable conditions, as we may say. free transpiration, moderate exercise, temperance in diet. Is it not plain that the former class from the first inspiration will receive a beginning of putrefaction and that bad will go to worse, while those which are pure and void of excrement will either escape altogether or suffer so little damage as easily to return into the way of nature? . . . This has been said of one example, but it is a universal truth."
] quote this translation of Galen at length, because at first hearing, this sounds remarkably like good medical sense, if perhaps a little old fashioned in expression. On examination, however, we find it does little more than illustrate the concept of resistance as something modifiable by circumstance. The rest is almost purely notional. The pestilential airs are verbal explanations of the observed prevalence of epidemics in certain climatic conditions ; and although the idea of resistance as the reward of clean living is attractive, its validity in this context is dubious. We do not know the extent to which Galen's idea of clean living may have been derived from the kind of people who did in fact resist epidemic disease—how far, that is, his argument is circular. A more important point is that Galen has no empirical knowledge of his pestilential airs, and assumes that they are ubiquitous : everybody is at equal risk, so that differences in the incidence and severity of the disease appear to be determined exclusively by the condition of each man attacked. What was to prove a more fruitful idea, that the seeds of pestilence might be collected in greater concentration in one place than in another—for example, near or on the diseased person—or that the seeds might differ in their poisonousness or virulence, is lacking. It is stifled, if it ever arose, because Galen's system of interactions of the inward humours with various external epidemic environments covers all the observed facts.
The idea indeed was not seriously entertained for some thirteen centuries, when Girolamo Fracastoro postulated as the cause of infection a living agent, spread by contagion, which multiplied in the diseased body. Bacteriologists are rightly inclined to admire the achievement of the 16th century Fracastoro in making such good sense of the epidemiology of syphilis, but his coniagium vivum nevertheless remains a brilliant guess, and is in fact nearly as notional as Galen's epidemiology. The elucidation of a contagious disease like scabies, where the infecting agent is a readily visible i.isect. brought the doctrine of a microscopic contagium—an invisible ** insect "—a little nearer to reality. Later speculators on the comagium vivum were not slow to point to Leuwenhoek's " little bcasties" as microscopical models for the infective agents they had in mind. But as late as the early 18th century, even really acute-minded descriptions of infection in terms of a contagious agent are wholly speculative. For example. Carlo Cogrossi, in describing what appears to have been an exceptionally fatal outbreak of rinderpest among Italian oxen as an undoubted example of contagium vivum. has to support the idea that the ulcerative lesions observed in the mouth of affected cattle occur there because the hypothetical contagium enters via the throat and nostrils, by what is really an outrageous piece of special pleading. '* That bull which on nuzzling an infected cow suddenly shook his head, shivered and turned with threatening and disdainful mien may perhaps have felt in his nostrils the attack of the ruinous and invisible insects which, invading the mammillary processes, drove him to such furious acts. Those vesicles and tubercles which are found beyond the middle of the tongue in the oxen of the Brescian district support these conjectures."
We can nowadays smile at the explanation, because w'c know' that, unlike the burrowing mites of scabies which Cogrossi probably used as a model, the virus of rinderpest does not advertise its entry by inducing an immediate itch in the skin or mucous membranes. But our right to smile is little more than 50 years old. Not until the early years of this century had bacteriology advanced to the point where it could put guesses of this kind to rigorous test.
The remarkable achievements of bacteriologists from 1850 to 1890 provided facts enough to give reasonable solidity to the notion of the coniagium vivum. but the relation of bacteria to infection was very shakily established. So shakily, that the rigorous logician Charles Creighton saw no reason in 1894 to modify his prejudices to the extent of seriously discussing the germ theory in his massive History of Epidemics in Great Britain. His reference, for example, to a theory of resistance to cholera in terms of an environment that destroys the cholera vibrio is buried in the decent obscurity of a footnote, and given the status of a bit of slightly comic folk lore. The " hypothesis of insect life." he -writes, was '* happily thought of by a working cotton-spinner inManchesterto explain the immunity of the mill-workers. ' I've been thinking. Maister." said a spinner. 4 as how th" cholery comes o'hinsecks that smo' as we corn'd see "em. an' they corn'd live i' factories for th' 'eat and th' ile. Me an" my mates wor speaking' o't last neet' an" we o' on us thowi th' saam thing." " The more intellectually genteel notion that cholera is due to a comma-shaped bacillus belonging to the class of vibrios he dismisses in a stately admonition. " The common microscopic objects uniformly found in the choleraic discharges by later observers have been vibrios, of which half-a-dozen, or perhaps a dozen,'varieties have been distinguished. One of these was somewhat audaciously named the ' cholera germ * or ' comma bacillus of cholera ' by Dr. R. Koch, who went toCalcuttain 1884." Creighton concludes the footnote with what our American friends would call the pay-off line. " All vibrios, which have a corkscrew form when in motion, are apt to assume the comma form when at rest."
criteria of infection and resistance
Cholera was later proved to be an infection with the comma bacillus, but at the time Creighton's admonition was justified. Except by implication. Koch did not in his published works provide an answer to this kind of cold logic until 1910, when he formulated the postulate that must be satisfied to establish a microbe as a cause of. and not merely associated with, an infective disease—namely, the postulate that the microbe must reproduce the disease in the experimental animal. Having isolated a microbe, and established it as the cause of a disease, we have at last a tool for exploring resistance, at least to those infections which fulfil Koch's animal postulate. The postulate, as every medical student knows, cannot be applied in every case, because some microbes— like the malarial parasite and the virus of epidemic jaundice—are so exclusively adapted to life in their human host that no susceptible laboratory animal has yet been found. Nevertheless, the reaction of the animal host in the fully proven infections provides a new postulate which we can apply in these difficult cases. An infecting microbe usually stimulates the host to produce antibodies, which have the property of combining specifically vyith certain constituents (called antigens) of the microbe itself ; and a substantial rise of blood antibody during the infection and a later slow decline is characteristic of many infections. From this we can derive an argument that to some extent replaces Koch's postulate about reproducing the disease in the laboratory : namely, that if during the course of infection there is a substantial new production of antibodies specific for the microbe isolated from the infection, that microbe is the more certainly the cause of the disease.
The point needs emphasizing here, that although in the characteristic antibody response we have a test for determining the microbial cause of a disease, which in difficult cases substitutes for the animal test, its validity is based by analogy on the antibody response observed in those diseases where the microbial cause has with greater certainty been established in the experimental animal. Jf it were not so, this substitute proof of microbial causation would be open to the sort of criticism that Creighton made of Koch's early work on cholera. He would have said that the presence of specific antibody in a disease, like the presence of a microbe, was no more than an association of two things, and certainly no proof that one had been caused by the other.
With the establishment of microbial causes of disease, the study of resistance moved into a new phase. Jt was now possible to investigate infec-lion as the outcome of an interaction of the host and the infecting microbe. On the one hand, we have all the factors determining the resistance of the host, on the other, the factors determining the virulence of the microbe. Until resistance and virulence can be measured, however, we are little better off than Galen was when he analysed resistance in terms of the attack by a pestilential air ; we have only the advantage of a real organism to which we can attribute virulence.
Let me illustrate the difficulty in another way. We observe a patient with a severe infective disease. He may recover : is it because his resistance increases ? He may die : has his resistance gone ? For all we know from observation, these two events could be due respectively to a decrease and an increase ir-virulence of the attacking microbe, wit1 any change in his resistance. We are ir same difficulty with a community. Doe epidemic flare up because resistance drop^ r because the virulence of the prevalent microbe increases ?
Our observations can be likened to watching a tug-of-war in darkness and silence, seeing only the movement of the piece of tape dangling from the middle of the rope. The tape may move towards disease and death on the one side, towards recovery on the other, or the contest may hang for a long time in balance. We know nothing about the contesting teams except that one represents the resistance of the host, tugging towards recovery, and the other the virulence of the microbe. The fortunes of the tape can tell us nothing, because they would change in exactly the same way whether the opposing forces were a couple of five-year-old children or teams of beefy Olympic champions. In other words, severe disease may equally well result from the attack of a feebly virulent organism in a feebly resistant man. and the attack of a highly virulent organism in a highly resistant man. We could discover the strength of the silent and invisible teams in the tug-of-war only by putting at the other end of the rope a team of our own. whose capacity was known and constant. In a like manner, we can measure differences in resistance only in terms of a bacterium of constant virulence. Technically, the measurement is not simple. A culture of an infective bacterium is an ever-changing thing, and unless we know tricks that keep its virulence steady, we can depend on it only for relatively short periods. Furthermore, when several animals, chosen for their apparent similarity, are given a fixed dose of bacteria, not all behave in the same way. Some become severely infected, some lightly, and some are not infected at all. To avoid the day to day changes in virulence of the culture, it is essential lo compare two or more states of resistance at the same time, giving all the animals the same culture : and to avoid the errors that arise because resistance varies from animal to animal, we must in each case lest the average resistance of a large number of animals.
The conclusion is inescapable. Reliable analysis of resistance entails a deliberate exposure to infection of groups of animals. Sometimes we can use man. as when we compare the incidence of infection in vaccinated and unvaccinated persons during a natural epidemic of smallpox. Mostly, however, the infection we wish to study is not naturally prevalent, and the desired experimental procedure is of a kind that could not properly be imposed on a human community.
I have laboured on this elementary exposition of bacteriological method to emphasize the crux of my argument ; that the hypotheses about the microbial causation of infection and about the nature of resistance, both fundamental to our understanding and control of disease, can be substantiated in no other way than by animal experiment, excepting by the corresponding large-scale experiment on a human herd.
Let us now see what kind of information this approach can give us. Recalling Galen's notions about the causation of disease, we may first ask whether in fact an epidemic outbreak is necessarily determined by a change in the resistance of the community at risk ; then turn briefly to resistance in the individual and see how it may be affected by heredity and nutrition ; and lastly consider some problems of explaining resistance in terms of local bodily functions.
The conditions of a natural epidemic can be reproduced experimentally. The late Professor Topley, in a scries of beautiful experiments with populations of mice, obtained answers to a number of the questions raised by observational epidemiology. Epidemics of mouse-typhoid were initiated in communities of mice. When the community was completely isolated, the epidemic ran its course, eventually dying out and leaving a number of apparently healthy survivors. The epidemic could be made more severe by using a mouse-typhoid bacillus of exceptional virulence to initiate it—a not unexpected result, but one which places on a firm footing one intelligent guess about the effect of changes of virulence in changing the severity of the epidemic.
The shifts in population that take place in many natural epidemics were mimicked by introducing fresh mice at regular intervals into an epidemic community. The results were illuminating, because to some extent unexpected. If the epidemic were progressing at a steady rate, with a certain number of deaths, the daily introduction of fresh immigrants increased it. The immigrants suffered heavily—that we should expect—but among the indigenous mice which would undoubtedly have survived the epidemic had the original colony been left in strict quarantine, the disease also spread afresh. Again, when the isolated epidemic community was allowed to reach a stable state in which the infection could be said virtually to have died out, the introduction of uninfected immigrants relighted the epidemic, which then swept through the population with increased fierceness. Here were two results that, as far as could be judged, had nothing to do with changes in the innate resistance of the immigrant mice or in the virulence of the infecting microbe. Simple shifts in the concentration of the mouse community, and new opportunities for the contact of resistant and recovered animals with the unaffected immigrant were sufficient to change the epidemic picture profoundly. It is a result we note in parenthesis that casts doubt on the generally applied principle, thai the best way to prevent an epidemic occurring in a community is to refuse entry to anyone who might already be suffering from it. In some cases it might be much more efficacious to refuse entry to the healthy.
The resistance of each affected animal is presumably varying during the epidemic, and we must distinguish the innate resistance of the animals newly introduced into an infected mouse community and the resistance they develop as a result of exposure to the disease. We have already seen that the resistance of the survivors in a stable post-epidemic population can be overcome when immigrants are admitted. This is the result presumably of new infections and new risks of contagion. The influence of change in the general level of resistance may also be tested by treating the immigrants beforehand with a vaccine that immunizes them against mouse-typhoid. The epidemic flares up as before, but it is less severe and the expectation of life of the newly introduced mice is increased. Here is proof that changes in resistance can modif\ the course of an epidemic. It is not proof incidentally that the vaccine was only feeblx effective. The intensity of risk in Topley's mouse community was great indeed, and a degree of immunization that in these circumstances could only postpone infection was in other experimental circumstances sufficient to protect the animals altogether.
Typhoid fever in man resembles mouse-typhoid in many respects, including a close relationship, bacteriological!)' speaking, between the two causative organisms. We can accordingly regard these results as a corroboration of the practice, derived from other considerations, of immunizing man with a vaccine made from the human typhoid bacillus, in the expectation of its diminishing the risk of typhoid fever during an epidemic outbreak.
In the selection of typhoid vaccines for this purpose, we sometimes assume a closer parallelism of mice and men, arguing that what protects a mouse against the human typhoid bacillus is likely "to protect a man. Until recently the protective power of typhoid vaccines was believed to reside in two bacterial constituents, called the O antigen and the Vi antigen. This belief was in part based on animal tests, which clearly indicated that the Vi antigen is highly protective. Recent trials, however, suggest that in man the Vi antigen has little protective power, and that typhoid vaccines are better if they contain only the 0 aniigen« I mention this discovery because it bears on the objection that animal experiments are irrelevant to human problems. The answer is two-fold. In the first place, experience with other microbes has abundantly justified the assumption that on the whole mammals behave alike in their response to immunizing agents. In the second place, dissimilarities between animals, whether in the restricted field of immunology or elsewhere, are not good enough grounds for a total denial of similarities. "
You may remember the answer given after careful thought by the child who was asked what elephants had which made them different from all other animals; it was "Baby elephants." This goes to the root of the matter, because the genes that determine just how the embryonic elephant will develop are indeed unique. Nevertheless, the building materials and methods of synthesis at the disposal of the genes are not unlimited in their variety ; and being far fewer than the multitudes of different plants and animals, they are shared by all creation, from bacteria upwards. It is not unreasonable, therefore, to expect as many likenesses between animals as there are differences ; and likenesses of this kind are certainly reflected in the man% features common to infective disease among vertebrates. Here is sufficient ground for hoping that generalizing from one animal to another will be fruitful. Torefrain from doing so because it might be invalid would be to make a virtual "appeal to ignorance" : to say. in effect, "this must never be done because some unknown unpredictable factor might intervene to disappoint expectation " ; and thus to indulge in the kind of intellectual nihilism that would be fatal to any human activity, to say nothing of biology.
As regards genetically determined resistance, the results of animal experiment are perhaps even more complex than the problem they were intended to solve. There are good indications in the epidemic history of mixed populations that one race may suffer more than another from a prevalent disease ; though in collecting such statistics it is almost impossible to prove that the hazards to which the two races are exposed are really identical. Experiments confirm our guess that genetic constitution matters. Thus, strains of mice moderately resistant to Salmonella enteritidis infection can by selective in-breeding be separated into a feebly resistant and a strongly resistant strain ; and by back-crossing it is possible to show that the resistance is probably heritable on Mendelian lines. The resistance under genetic control appears to be particular, and not a genera) resistance governed by a single hereditary factor. Thus a strain of mice bred for resistance to infection by Salmonella enteritidis was susceptible to an encephalitis virus, and a second strain, bred from the same parent stock, was resistant to the virus, but susceptible to the Salmonella bacillus. That is to say. different sets of genetic factors determine resistance to different infecting agents.
The genetic factors responsible for resistance even to one kind of infection are far from simple. Thus, two strains of mice were bred, one relatively resistant to mouse typhoid and one relatively susceptible. The susceptible mice on examination were found to have fewerphagocytes in the blood than the resistant mice. Now these phagocytic cells are active against infecting bacteria, and it was reasonable to suppose that mice with the greater number of blood phagocytes owed their immunity'in part to this fact. Mice were accordingly bred to select a strain with high concentration of phagocytes in the blood and one with low ; but unfortunately for the hypothesis, the animals with the smaller concentration of phagocytes proved to be the more resistant to the test infection.
Such studies are in their infancy ; all we know at present is that the genetic constitution of an animal, upon which resistance to all kinds of disease depends, is very complex, and we are far from being able to sort out the genes that determine resistance to infection or to explain their obviously manifold interactions.
resistance and nutrition
In nutrition the experimental approach is illuminating but again not particularly comforting. The traditional association of famine and pestilence epitomizes a very old belief in the influence of diet on resistance to infective disease. But famine comes in the train of much else that disorganizes the life of a community, and although epidemiology and social medicine have substantiated this belief by establishing an association between malnutrition and certain infective diseases like tuberculosis, malnutrition itself is in turn linked with so many inter-related and variable factors in the human environment, and the human herd is so ill-suited to the kind of test that would enable us to estimate the importance of these other factors, that there are few instances where malnutrition could with any confidence be said to be the prime cause of the susceptibility. Some of the conjectures about malnutrition, however, are easily substantiated by animal experiment. Deprivation of the main dietary ingredients—proteins, carbohydrates and fats—predisposes to infection, but for the most part only when the starvation reaches a point where general inanition is imminent. Mankind perhaps needs no experimenter's exhortation not to starve, and will avoid poor living for more obvious reasons than a desire to ward off infection A plentiful diet will presumably keep us healthy : but except in a very broad way we are slill ignorant of the ingredients of the diet responsible for the absence of infection. It is not necessarily enough to ensure that the animals are well set up and robust. When rats, for example, were fed on various diets, good both for their growth and their reproduction—not at all a bad index of suitability—there was no constant association between goodness in this sense and capacity to make the animals resist Salmonella infections or tuberculosis. It may be that the stresses and dangers to which all animals are subject in nature are so various that no single diet is capable of yielding all the benefits we demand of it. Nutritional deficiencies indeed are known in some circumstances to increase resistance. Deficiencies of the vitamins of the B group, for example, often decrease resistance to infection by bacteria and viruses. But some American investigators, having made mice deficient in vitamin Bl and infected them with a poliomyelitis virus, observed that the incubation period of the resulting disease was longer and the incidence of mortality and paralysis lower than in normal mice. This paradoxical result is not entirely inexplicable, because viruses, being largely intracellular parasites, are more intimately concerned with the metabolism taking place in the cells of the animal than bacterial infections are. where the infecting microbes often flourish in extracellular tissue-spaces. A dietary deficiency which impairs the intimate metabolism of the cell might therefore be expected to impair the proliferation of parasites whose fortunes are so closely dependent on the well-being of the cell. But whatever the explanation, the facts are against any easy generalization about diet. There are vndoubted associations of malnutrition with vered resistance to disease but also undob d indications that what is generally good i you may do little for resistance to infection , in certain circumstances it may even pay to be slightly ill-fed.
resistance and localization of infection
My last example comes from the experimental analysis of local resistance. Most of us are personally familiar with that text-book example of a local infection, jthc staphylococcal boil or carbuncle on the hand or forearm which comes to a head, bursts, and is slowly but finally resolved. In less happy cases, the staphylococcus spreads to the lymph nodes in the armpit, there to set up another abscess. It may even invade the blood stream and establish a severe general infection. The simple boil may be regarded as an expression of local resistance, one that in most cases saves the body from a more general invasion, and the walling off of pus in the abscess cavity as a containment of the invaders in the first bridgehead they have established. The presence in the inflammatory focus of fluids containing antibacterial substances, and of phagocytes that can ingest and digest the bacteria, and the ultimate formation of a zone of active phagocytes round the inflamed region, are consistent with this view of a local defensive action. An even more successful defence would be the destruction of the invaders before they could produce any substantial damage. We live in a world swarming with these staphylococci—indeed over half of u>> may carry them in the nose—and we are continually damaging our skin, in a small way. certainly, but enough to let the ubiquitous staphylococci get a foothold. By a natural intrapolation from the course of manifest staphylococcal infection, these numerous infections that never come to anything are presumably dealt with by inflammatory and localizing reactions like those in the manifest disease, but on a microscopic scale. The presumption is indeed extended to cover unsuccessful invasion by all kinds of microbial parasites. But the rapid extermination of a microscopic bridgehead is largely a hypothetical event ; and the participation of the inflammatory reaction rests almost entirely on analogy. The analogy is suspect for another reason ; it is still uncertain whether the inflammation that accompanies so many infections is wholly beneficial to the animal, or merely a reaction to damage done by the infection, after the defence mechanisms have done their best with the invader.
There are few experiments that settle this point decisively, but by an indirect method it can be shown that an early rapid destruction of microbes in fact takes place. Within two to three hours of invasion, the survivors of the original invading force of some bacteria are as few as one in a million. With other bacteria the destruction is less, but nevertheless substantial. For example, only 1 in 100 staphylococci survive to carry on the attack. But this destruction takes place some time before true inflammation starts, and well before any defensive wall has been formed. Indeed, there are indications that a defensive wall to localize the invaders might be a disadvantage to the animal : because when bacteria are carried directly to the blood stream, they are destroyed far more efficiently than in a bridgehead made, for example, in the skin. To explain this aspect of resistance, we must consider far more subtle and complex events than localization by the classic inflammatory reaction, if we are to understand how these defences are maintained in an efficient state, and what derangements lead to their breakdown in cases where the infection is successful.
Sometimes. 1 suspect, the derangements are purely local, and at others an expression of a general change. It would at anv rale be unwise to assume that the local defences always reflected the slate of general resistance in animals, and that what improved the one would necessarily improve the other. The injection of a mouse with a Salmonella bacillus by the intraperitoneal route sets up a generalized infection, and the survival of the mouse indicates a certain degree of, as it were, internal resistance. The mouse is made even more resistant to this test infection by pre-treatment with a vaccine. But when it is challenged by the oral route, where the resistance depends on the integrity of the superficial defences of the lining of the gut. the vaccine is found to have had little protective action. The independence of internal and superficial defences is demonstrable in another way. In the course of a progressive generalized infection of a guinea-pig with a streptococcus, the resistance of the body as a whole is clearly declining. But the superficial resistance of the skin to infection not only does not decline, but may in a few days rise to forty times its original efficacy. Resistance has many facets, and the peculiarities of one facet, the early superficial defences of the body, cannot be ignored in any consideration of the prevention of disease. Most infections creep upon us without any obvious breac' n our body surfaces. If we knew just what .d of local microscopic accident or tempore sloth-fulness on the part of the tissues gave an c , ening for the attack, we might discover important criteria for judging the defensive value of the things we do to keep healthy.
In these brief illustrations of experiments designed to elucidate resistance to infection. 1 have intentionally avoided stressing the immediate benefits to mankind that flow from this approach. In the field of immunology alone there is no lack of benefits 1 might have cited— not only .those of hygiene and prophylactic immunization, but the benefits that have come from a study of the wider implications of the antibody response, ranging from a deeper understanding of the virulence of infecting microbes to the prevention of transfusion accidents and of haemolytic jaundice of the new born. My purpose has been to display what we find at its least impressive, to emphasize the imperfections of the approach and the fact that it creates as many problems as it solves. In spite of this. 1 submit, none of the technical objections to animal experiment—that the experimental approach is unnecessary ; and that the results are either irrelevant to the human problem or cannot be justified by the benefits they bring—can be sustained. As Professor Topley said of the epidemiological problems he tackled by experiment: " It is clear that no amassing of clinical observations, however careful and acute, and no correlation of such observations with environmental factors, however complete the records and statistical analyses, could have solved such problems as these, or have shown us in any detail how and when we might intervene effectively." Many of our observational guesses about things like prophylactic immunization and diet have turned out to be correct, but the complexities of the problem revealed by experiment give us no ground for supposing we should not have gone astray if we had been content merely to act on these guesses. It is in principle arguable that the conclusions of observational biology, substantiated by experimental researches that involved no animals, could in time give us a true picture. But our progress would be agonizingly slow. We could not. and should not, endure the misery of watching humanity suffer from diseases which past experience told us might well be eradicated, solely because we eschewed the use of animals.
It is our duty nevertheless to avoid unnecessary use. When 1 described Professor Topley's experiments as beautiful. J gave them their due as technical and intellectual achievements. But I wished primarily to imply that being designed with such care, and yielding results with such economy, they constitute an ideal example of the application of that secondary principle of discrimination in the use of animals. To my mind he did them so well that they need never be done again ; we can advance unhesitatingly from the point where he left the subject.
As 1 have tried to show, animal experimentation is necessary because we must know facts that otherwise could be obtained only by substituting regimented populations of our fellow men for the animals we use. The contention that what happens in an animal is irrelevant to what happens in man can be made only by insisting on the differences between one animal and another ; and by ignoring the similarities. To do so is lo deny on biological grounds the very kinship that is invoked to forbid the use of animals altogether.
The argument from animal lo man is ofien difficult and sometimes impossible. This is not. however, the important point in the judgement of relevance. Even when the results closely parallel what we have inferred from observation of the human infection, the experimental approach gives us the kind of result whose certainty is better established than that of the observational inference, and which is from the moment of its establishment a cautionary tale that the doctrinaire speculator ignores at his peril. Herein lies the answer to those who demand a justification beyond that of simply adding knowledge to a scientific discipline. If all the investigations 1 have mentioned had given us no obvious guide to the betterment of mankind, they would still have an immediate and important role to play. To anyone with a proper respect for evidence, each new fact is a curb on the excesses of notional biology and medicine. Without knowledge, we must often rely on, and even act upon, notions in our practical affairs ; and the more inadequate the techniques of testing notions and speculations are, the more metaphysical our guides to action. Man is born a metaphysician—usually a bad one ; and a multiplicity of bad metaphysics means a multitude of bad actions. It is the hardest thing in the world to be in sufficient possession of a subject to feel that we are using our knowledge wisely. In our search for the best thing to do, we must not only deny the comforts of unreason to the muddle-headed, but refuse to allow to the metaphysician any priority for the transcendental over Ihe empirical.
Biology from its earliest days has not lacked its acute thinkers, and even with the superior knowledge of the 20th century we can accord to many of them the highest praise for their acumen. But acumen isn't everything. Galen had it. and yet he is a monumental example of acumen overlaid by a notional biology. Just how notional that biology may be is evident in this brief summary of his basic doctrine made by Major Greenwood: "The temperaments depend upon the blending of four elementary qualities, the hot and its opposite the cold, the moist and its opposite tfie dry. Harmoniously blended we have the perfect temperament or eucrasis. and there are eight discords, four when one quality only is in excess, four more when two are in excess. The procatarctic (i.e. the predisposing) factors are manifold, but the constitutions are also referable to four qualities and so may give eight discordant and one harmonious result."Greenwoodwas an exceptionally fair-minded scholar, and gave to Galen all his dues ; so that his comment is one we should ponder on. '* I think." he continues. " a reader will say of this bald description that it sounds rather like a silly Christmas game, and indeed acuteness without wisdom is the silliestthing in the world." Now silliness is a judgment that we can safely make of some aspects of Galen's work, because we are no longer in much danger of failing to do our best about epidemic diseases because someone insists upon applying his principles. But when a silly hypothesis threatens us with a
Dr. Douglas McClean said that he was glad to have the privilege of expressing gratitude to Professor Miles for this instructive, stimulating and entertaining lecture. He also found it extremely chastening. Last year Sir Henry Dale had given a fighting Stephen Paget Lecture ; he drove home the crimes for which the anti-vivisectionists were morally, if not legally, responsible. This time another fighting lecture had been delivered, but Professor Miles seemed to have adopted the motto that offence was the best form of defence and he had shown where they would have been if it had not been for the experimental method and the use of correspondingly silly action, a more damning epithet is required if we are not to fall victims of our own too easy tolerance. If the experimental approach does not solve our urgent problems, we should be more than content to justify the most academic result of animal experiments made by men of good will as a contribution to the wisdom which is the final justification for any science.
experimental animals. They might not have been quite in the picturesque state of mind in which Galen was but he did not think they would have been much further on than Creighton.
Dr. Miles was preaching, of course, to the 100 per cent converted and he hoped that the lecture would be given the maximum amount of publicity in those quarters where it was most useful, particularly in Parliamentary quarters and. he might add. both Houses of Parliament.
The vote of thanks was accorded by acclamation. Those members of the audience not members of the Society then left the meeting SPECIAL GENERAL MEETING
A SPECIAL GENERAL MEETING of the members of the Society was held at the Physiology Lecture Theatre,UniversityCollege.London, on Tuesday. 27th November. 1956, at the conclusion of the twenty-fifth Stephen Paget Memorial Lecture, with the President, the Right Hon. Viscount Waverley. in the chair.
The President said that J.he meeting had been called to consider a Special Resolution which he would ask the Hon. Secretary to read.
The Hon. Secretary (Dr. W. Lane-Petter) read the special resolution as follows :—
That Rule III (2) of the Research Defence
Society shall be amended to read as follows : " The Council shall consist of the President, the Chairman, the Honorary Treasurer, the Honorary Secretary and eighteen members, all of whom shall be elected at the Annual General Meeting." The President said that the change was from fifteen to eighteen members. Was it the pleasure of the meeting that the resolution should be adopted ?
This was agreed to unanimously, and the proceedings of the Special General Meeting concluded.
ANNUAL GENERAL MEETING
T^HE ANNUAL GENERAL MEETING of 1 the Society was held at the Physiology Lecture Theatre, University College, London, on Tuesday, 27th November, 1956. The President, the Right Hon. Viscount Waverley. was in the chair.
The Minutes of the last Annual General
Meeting, published in Conquest in January. 1956. were signed by the President as correct.
The President said that the Annual Report of the Council and the Audited Accounts had been circulated to all members and he would call upon the Honorary Treasurer to move their adoption.