An Introduction to the Study of Experimental Medicine

that a comparative experiment was also required here by another circumstance, viz., that I used the reduction of copper as a test for sugar. This, however, is an empirical characteristic of sugar which might be shown by substances still unknown in the bodily economy. But, even apart from that, I repeat, a comparative experiment would have had to be made as an experimental necessity; for this very case proves that we can never foresee its importance.

So for comparison with the dog fed on sugary soup, I took another dog to which I gave meat to eat, being careful moreover to exclude all sugary or starchy material from its diet ; then I sacrificed the animal during digestion and examined comparatively the blood in its superhepatic veins. Great was my astonishment at finding that the blood of the animal which had not eaten any also contained sugar. We therefore see that comparative experiment led me to the discovery that sugar is constantly present in the blood of the superhepatic veins, no matter what the animal's diet may be. You may imagine that I then abandoned all hypotheses about destruction of sugar, to follow this new and unexpected fact. I first excluded all doubt of its existence by repeated experiments, and I noted that sugar also existed in the blood of fasting animals. But if benefits are linked with comparative experiment, not performing them has its necessary annoyances. This is proved by the following example :

Second example. — Magendie once made investigations on the uses of the cerebrospinal fluid and was led to the conclusion that removing this fluid produces a kind of titubation in animals and a characteristic disturbance in their motions. Indeed, after uncovering the occipito-atloidian membrane, if we pierce it to let the cerebrospinal fluid run out, we notice that the animal is seized with peculiar motor disturbances. Apparently nothing could be simpler or more natural than the influence on their motions of removal of the cerebrospinal fluid ; yet this was an error, and Magendie told me how another experimenter chanced to find it. After cutting the neck muscles, this experimenter was interrupted in his experiment at the moment when he had just laid bare the occipito-atloidian membrane. Now when he came back, to go on with his experiment, he saw that the simple preliminary operation had produced the same titubation, though the cephalorachidian fluid had not been removed. What was merely the result of severing the neck muscles had therefore been attributed to removal of the cerebrospinal fluid. Comparative ex-

periment would obviously have solved the difficulty. In this case, two animals, as we have said, ought to be placed in the same conditions save one, that is, the occipito-atloidian membrane should be laid bare in both animals, and it should be pierced, to let the fluid flow out, in only one of them ; then it would be possible to judge by comparison and thus ascertain the precise part which the removal of the fluid plays in the disturbances. I might quote a great many errors into which able experimenters have fallen by neglecting the precept about comparative experiment. Only, as the examples that I have quoted prove, it is often hard to know in advance whether comparative experiment is necessary or not; and so I repeat that, to avoid all annoyance, we should accept comparative experiment as a veritable command, to be executed even when useless, so as not to be missing when it is necessary. Comparative experiments are sometimes made, now on two animals, or for greater accuracy, on two similar organs in the same animal. Thus, at the time when I wished to judge the influence of certain substances on the glycogen of the liver, I could never find two animals comparable in this respect, even by putting them in exactly similar dietary circumstances, i.e., without food for the same number of days. According to their age, sex, plumpness, etc., animals bear starvation better or worse and destroy more glycogen or less, so that I could never be sure that the differences I found were the result of difference in diet. To remove this source of error, I was forced to make the whole experiment on the same animal, by taking away a preliminary piece of its liver before the dietary injection, and another afterward.

So when we want to see the influence of contraction on the metabolism of the muscle in a frog, we have to compare both members of a single animal, because in this respect two frogs are not always comparable. At the beginning of this chapter, I said that we are often deceived by false values ascribed to words. I wish to explain my idea by examples. First example. — In 1859 I made a report to the Philomathic Society, in which I discussed Brodie's and Magendie's experiments on ligature of the bile duct, and I showed that the divergent results

whicli the two experimenters reached, depended on the fact that one operated on dogs and tied only the bile duct, while the other operated on cats and, without suspecting it, included in his ligature, both the bile duct and a pancreatic duct. Thus I showed the reason for the difference in the results they reached, and I concluded that, in physiology as everywhere else, experiments are rigorous and give identical results whenever we operate in exactly similar conditions.

A propos of this, a member of the society took the floor to attack my conclusions; it was Gerdy, surgeon at the Charite, professor in the faculty of medicine, and known through various works on surgery and physiology. ^^Your anatomical explanation of Brodie's and Magendie^s experiments,'' said he, "is correct, but I cannot accept your general conclusion. You say, in fact, that the results of experiments in physiology are identical ; I deny it. Your conclusion would be correct for inert nature, but it cannot be true for living nature. Whenever life enters into phenomena," he went on, "conditions may be as similar as we please; the results may still be different." To support his opinion, Gerdy cited cases of individuals with the same disease, to whom he had given the same drugs, with different results. He also recalled cases of like operations for the same disease, but followed by cure in one case and death in another. These differences, according to him, all depended on life itself altering the results, though the experimental conditions were the same ; but this could not happen, he thought, in phenomena of inert bodies, into which life does not enter. Opposition to these ideas was prompt and general in the Philomathic Society. Everyone pointed out to Gerdy that his opinions were nothing less than a denial of biological science; and that, in the cases of which he spoke, he completely deceived himself as to the identity of conditions, in this sense, that the diseases which he regarded as similar or identical were not in the least alike, and that he attributed to the influence of life what should be accounted for by our ignorance about phenomena as complex as those of pathology.

Gerdy continued to maintain that life had the effect of altering phenomena so as to make them differ in different individuals, even when the conditions in which they took place were identical. Gerdy believed that one man's vitality was not the saine as another's, and that there must therefore be between individuals differences impossible to define. He would not give up his ideas ; he entrenched himself behind the word vitality and could not be made to understand

that it was only a word, devoid of meaning and corresponding to nothing ; and that saying that something was due to vitality amounted to calling it unknown. In fact, we are often duped by such words as life, death, health, disease, idiosyncrasy. We think we have explained when we say that a phenomenon is due to a vital influence, a morbid influence, or an individual idiosyncrasy. We must really learn, however, that vital phenomenon means only a phenomenon peculiar to living beings, whose cause we do not yet know ; for I think that every phenomenon, called vital to-day, must sooner or later be reduced to definite properties of organized or organic matter. We may, of course, use the expression vitality as chemists use the word affinity, but knowing that fundamentally there are only phenomena and conditions of phenomena which we must learn ; when the conditions necessary to phenomena are known, then occult, vital and mineral forces will disappear.

I am very happy to be in perfect harmony on this point with my colleague and friend, M. Henri Sainte-Claire Deville. This will be seen in the following words spoken by M. Sainte-Claire Deville in explaining his splendid discoveries on the effects of high temperatures to the Chemical Societv of Paris.^ "We must not conceal from ourselves the fact that studying primary causes of the phenomena which we observe and measure has its grave dangers. Escaping exact definition, independent of particular facts, it leads us, much oftener than we think, really to beg the question, and to content ourselves with specious explanations that cannot withstand severe criticism. Chiefly affinity, defined as the force which presides over chemical combinations, has long been and still is an occult cause, a sort of archeus to which we refer all the facts which we do not understand and which we thenceforth consider explained, though they are often only classified and often wrongly classified: in the same way we attribute to catalytic force any number of extremely obscure phenomena which, in my opinion, become more sO' when we refer them en bloc to an entirely unknown cause. In giving them the same name,

• H. Sainte-Claire Deville, Legona su/r la dissociation prononcies devant la 8ociit6 chimique. Paris, 1866. " All this applies to those recently invented forces, to the forces of dissolution and diffusion, to crystalogenic force, to all the special attractive and repelling forces brought in to explain phenomena of heating and superfusion, electric phenomena, etc. we certainly believe we are placing them in the same category. But this classification is not even proved legitimate. What, indeed, could be more arbitrary than setting side by side catalytic phenomena that depend on the action or the presence of platinum sponge and concentrated sulphuric acid, when the platinum and the acid are not, so to speak, parties to the operation ? These phenomena will perhaps later be explained in an essentially different way, according to whether they are produced under the influence of a porous material like platinum sponge or under the influence of a highly energetic chemical agent like concentrated sulphuric acid.

"In our studies, we must therefore lay aside unknown forces, to which we have recourse only because we have not measured their effects. On the contrary, we should direct our attention to the observation and numerical determination of effects, which alone are within our range. By such work, we establish differences and analogies, and new light results from comparisons and measurements. "Thus heat and affinity are constantly face to face in our chemical theories. Affinity completely escapes us, and yet we attribute combination to it, as the effect of an unknown cause. Let us simply study the physical circumstances which accompany a combination, and we shall see how many measurable phenomena, how many curious relations present themselves at every moment. Heat, they say, destroys affinity. Let us patiently study the decomposition of bodies under the influence of heat measured in quantity of work, heat or energy : we shall immediately see how fruitful this study is, and how independent of every unknown force, unknown even from the point of view of the kind of units to which we must refer it for precise or approximate measurement. Especially in this sense affinity, considered as a force, is an occult cause, unless it simply expresses a quality of matter. In that case, it would be used simply to designate the fact that such and such substances can or cannot be combined in such and such definite circumstances."

When a phenomenon takes place outside the living body and does not occur in the organism, that is not because an entity called life prevents the phenomenon from taking place, but because the necessary condition for the phenomenon is not met with inside, as it is outside, the body. Thus it has actually been said that life prevents fibrin from coagulating in living animals' blood vessels, though it does coagulate outside the vessels because life no longer affects it. It

is nothing of the sort ; certain physico-chemical conditions are needed to make fibrin coagulate ; they are harder to produce, but may nevertheless be found in the living, and, as soon as they appear, fibrin coagulates as well inside as outside the organism. The life invoked here is therefore only a physical condition which does or does not exist. I have shown that sugar is produced in the liver more abundantly after death than during life; certain physiologists drew the conclusion that life influences the formation of sugar in the liver; they said that life hinders its formation and death favors it. One is surprised to hear such vitalistic opinions in our day and to see them supported by men who pique themselves on applying to physiology and medicine the accuracy of physical science. I shall later show that here again physical conditions are either present or absent, but nothing else is real ; because again, at the base of all these explanations, only the conditions of phenomena are to be found.

To sum up, we must learn that the words we use to express phenomena whose cause we do not know are nothing in themselves ; and that the moment we grant them any value in criticism or discussion, we abandon experience and fall into scholasticism. In discussing or explaining phenomena, we must be very careful never to abandon observation or put a word in place of a fact. Very often we even expose ourselves to attack, solely because we abandon facts and conclude with a word that goes beyond what we have observed. The following example will prove it clearly.

Second example. — When I made my investigations of pancreatic juice, I noted that this fluid includes a peculiar material, pancreatin, which has characteristics of both albumen and casein. It resembles albumen in being coagulated by heat, but, like casein, differs from it in being precipitated by sulphate of magnesia. Magendie had made experiments, before me, on pancreatic juice, and had said that, according to his tests, pancreatic juice is a fluid containing albumen, while, from my investigations, I concluded that pancreatic juice does not comprise albumen, but does contain pancreatin, which is a material distinct from albumen. I showed my experiments to Ma- gendie, pointing out that we disagreed on the conclusion, but that we nevertheless agreed on the fact that pancreatic juice is coagulated by heat; only that other new characteristics that I had seen prevented my deciding on the presence of albumen. Magendie answered : "This difference between us comes from my having inferred

more than I saw; if I had simply said: pancreatic juice is a liquid coagulated by heat, I should have remained within the facts, and should have been unassailable." This example, which I have always remembered, shows how little value we should ascribe to words apart from the facts they represent. Thus the word albumen means nothing in itself; it merely recalls characteristics and phenomena. By extending this example to medicine, we should see that the words, fever, inflammation, and the names of diseases in general have no meaning at all in themselves.

When we create a word to characterize a phenomenon, we then agree in general on the idea that we wish it to express and the precise meaning we are giving to it ; but with the later progress of science the meaning of the word changes for some people, while for others the word remains in the language with its original meaning. The result is often such discord that men using the same word express very different ideas. Our language, in fact, is only approximate, and even in science it is so indefinite that if we lose sight of phenomena and cling to words, we are speedily outside of reality. We therefore only injure science by arguing in favor of a word which is now merely a source of error, because it no longer expresses the same idea for everyone. Let us therefore conclude that we must always cling to phenomena and see in words only expressions empty of meaning, if the phenomena they should represent are not definite, or if they are absent.

The mind by its very nature has systematic tendencies ; that is why we try to seek agreement about words rather than things. For experimental criticism, this is a false direction which confuses quesr tions and makes us believe in differences of opinion which generally exist only in our way of interpreting phenomena, instead of having some bearing on the existence of facts and on their real importance. Like everyone who has had the good fortune of bringing into science unexpected facts or new ideas, I have been and still am the object of much criticism. Up to this time, I have not answered my opponents, because I have always had investigations on hand so that time and opportunity have been lacking ; but in the remainder of this work, occasion to study them will quite naturally present itself, and by applying the principles of experimental criticism suggested in earlier paragraphs, we shall easily weigh the criticisms in question. Meantime I shall merely say that it is essential to distinguish between two

things in experimental criticism : experimental fact and its interpretation. Science requires us first of all to agree on fact, because that is the basis on which we must reason. As to interpretations and ideas, they may vary, and discussing them is an actual advantage, because such discussion leads us to make other investigations and to undertake new experiments. In physiology, we should therefore never lose sight of the principles of true scientific criticism nor mix them with personalities or artifice. Among the artifices of criticism, many do not concern us because they are extra-scientific ; one of them, however, we must point out. It consists in considering in a piece of work only what is defective and open to attack, while neglecting or concealing what is valid and important. This is the method of false criticism. In science the word criticism is not a synonym for disparagement; criticising means looking for truth by separating the true from the false and distinguishing the good from the bad. While just to men of science, such criticism alone is profitable for science. And this we shall easily show in the particular examples which we shall mention.

Methods of investigation and of scientific criticism cannot vary from one science to another nor, for that matter, in different parts of the same science. It will therefore be easy to show that the rules for physiological investigation, suggested in the last chapter, are absolutely the same as those which should be followed in pathology and therapeutics. Thus methods of investigation of the phenomena of life should be the same in normal as in pathological conditions. This seems to us fundamental in biological science.

As in physiology, so in pathology and in therapeutics, the starting point of scientific investigation is now a casual fact or one occurring by chance, now an hypothesis, i.e., an idea. I have sometimes heard physicians express the opinion that medicine is not a science, because all our knowledge of practical medicine is empirical and bom of chance, while scientific knowledge is deduced with certainty from theories or principles. There is an error here, to which I wish to call attention.

All human knowledge had to begin with casual observations. Man indeed could know things only after seeing them ; and the first time, necessarily, he saw them "by chance ; then he came to conceive ideas about things, to compare old facts and to deduce from them new ones ; in a word, after empirical observation, he was no longer led to find other facts by chance, but by induction. Fundamentally, then, all the sciences began with empiricism, that is to say, observation or chance experience had to form the first period. But empiricism is not a permanent state in any science. In the complex sciences of humanity, empiricism will necessarily govern practice much longer than in simpler sciences. Medical practice to-day is

empirical in most cases; but that does not mean that medicine will never escape from empiricism. The complexity of its phenomena will make it harder to escape ; but that should make us redouble our efforts to enter the scientific path as soon as we can. In a word, empiricism is not the negation of science, as certain physicians seem to think ; it is only its first stage. We must even add that empiricism never wholly disappears from any science. Sciences, in fact, are not lighted up in every portion at once; they develop only a little at a time. In parts of physics and chemistry, empiricism still persists. This is proved every day by chance discoveries, unforeseen by prevailing theories. I therefore conclude that we make discoveries in the sciences only because all are still partially obscure. In medicine more numerous discoveries are still to be made, because almost everywhere empiricism and obscurity prevail. So this very complicated science is proved further behind the times than others ; but that is all.

!New medical observations are generally made by chance; if a patient with a hitherto unknown affection is admitted to a hospital where a physician comes for consultation, surely the physician meets the patient by chance. But a botanist in the field happens on an unfamiliar plant in exactly the same way; and by chance also an astronomer catches sight of a planet, whose existence he did not know of, in the sky. In such circumstances, the physician's originality consists in seeing the fact that chance presents to him and in not letting it escape, and his only merit is accurate observation. I cannot here analyze the characteristics of good medical observation. Reporting instances of chance medical observations would be just as dull. Medical works teem with them; everybody knows them. I shall therefore limit myself to saying in general that, to make a good medical observation, it is not only necessary to have an observing mind, but also to be a physiologist. We shall the better interpret the various meanings of a morbid phenomenon, we shall assign it the proper value, and we shall certainly not fall into the difficulty, with which Sydenham ^ reproached certain physicians, of putting important phenomena of a disease on the same plane as insignificant and accidental facts, like the botanist who described caterpillar bites among the characteristics of a plant. Besides, we must bring to observation of a pathological phenomenon, i.e., a disease, exactly the same state of mind and the same rigor, as to observation of a physio-

logical phenomenon. We must never go beyond facts and must be, as it were, photographers of nature. But once made, every medical observation becomes the starting point, as in physiology, for ideas and hypotheses which experimental physicians go on to investigate through fresh observations of patients or by experiments on animals. We said that, in making physiological investigations, it often happens that a fresh fact arises unsought; that also occurs in pathology. To prove it, I need only cite the recent case of Zenker, who,^ in pursuing his investigations of certain muscular changes in typhoid fever, found trichinae which he was not Igoking for.

Pathological investigation may also take for its starting point a theory, an hypothesis or a preconceived idea. We might easily give examples to prove that absurd ideas, in pathology as in physiolog}% may sometimes lead to useful discoveries, just as it would not be hard to find arguments to prove that even the best accredited theories should be regarded only as temporary, and not as absolute truths to which facts should be bent. Therapeutic investigation conforms to exactly the same rules as physiological and pathological investigation. Everyone knows that the first promoter of therapeutic science was chance, and that only by chance were the effects of most medicines first observed. Physicians have also often been guided in their therapeutic attempts by ideas; and it must also be said that they were often the strangest and most absurd theories or ideas. I need only cite the theories of Paracelsus, who deduced the action of drugs from astrological influences, and recall the ideas of Porta, who assigned medicinal uses to plants, deduced from their resemblances to certain diseased organs; thus carrots cured jaundice; lung-wort, phthisis, etc.^

Summing up, we cannot establish any valid distinction between methods of investigation that should be applied in physiology, in pathology and in hygiene. The method of observation and experiment is still the same, unchangeable in its principles and offering only a few peculiarities in its application, according to the relative complexity of phenomena. We cannot, indeed, find any radical dif- * See Rapport des prix de mMeoine et de chirurgie pour 186 ( Compt. rendua de VAcad. des sciences).

' See Chevreul, Considerations sur Vhistoire de la partie de la midecine qui conceme la prescription des remMes {Jowrnal des savants, 1865). ference in the nature of physiological, pathological and therapeutic phenomena. Since all these phenomena depend on laws peculiar to living matter, they are identical in essence and vary only with the various conditions in which phenomena appear. We shall see later that physiological laws are repeated in pathological phenomena, whence it follows that the foundations of therapeutics must reside in knowledge of the physiological action of morbid causes, of medicines and of poisons; and that is just the same thing.

Criticism of facts gives sciences their true individuality. All scientific criticism should explain facts rationally. If criticism is attributed, on the other hand, to personal feeling, science disappears ; because such criticism rests on a criterion that can neither be proved nor conveyed as scientific truths should be. I have often heard physicians answer, when asked the reason for a diagnosis, "I do not know how I recognize such and such a case, but it is evident"; or when one asks them why they give certain remedies, they answer that they cannot exactly tell, and besides that they need not explain, since they are guided by their medical tact and intuition. It is easy to understand that physicians who reason in that way deny science. But we cannot too strongly protest against such ideas, which are bad, not only because they stifle every germ of science, but also because they especially encourage laziness, ignorance and charlatanism. I entirely understand a physician's saying that he cannot always rationally account for what he is doing, and I accept his conclusion that medical science is still plunged in the shades of empiricism ; but if he goes on to proclaim his medical tact or his intuition as a criterion which he then means to impose on others without further proof, that is wholly antiscientific.